diff --git a/.mailmap b/.mailmap
index a4806f0..9b0d026 100644
--- a/.mailmap
+++ b/.mailmap
@@ -68,6 +68,7 @@
 Kay Sievers <kay.sievers@vrfy.org>
 Kenneth W Chen <kenneth.w.chen@intel.com>
 Koushik <raghavendra.koushik@neterion.com>
+Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
 Leonid I Ananiev <leonid.i.ananiev@intel.com>
 Linas Vepstas <linas@austin.ibm.com>
 Mark Brown <broonie@sirena.org.uk>
@@ -111,3 +112,4 @@
 Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
 Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
 Takashi YOSHII <takashi.yoshii.zj@renesas.com>
+Yusuke Goda <goda.yusuke@renesas.com>
diff --git a/CREDITS b/CREDITS
index 07e32a87..44fce98 100644
--- a/CREDITS
+++ b/CREDITS
@@ -688,10 +688,13 @@
 
 N: Kees Cook
 E: kees@outflux.net
-W: http://outflux.net/
-P: 1024D/17063E6D 9FA3 C49C 23C9 D1BC 2E30  1975 1FFF 4BA9 1706 3E6D
-D: Minor updates to SCSI types, added /proc/pid/maps protection
+E: kees@ubuntu.com
+E: keescook@chromium.org
+W: http://outflux.net/blog/
+P: 4096R/DC6DC026 A5C3 F68F 229D D60F 723E  6E13 8972 F4DF DC6D C026
+D: Various security things, bug fixes, and documentation.
 S: (ask for current address)
+S: Portland, Oregon
 S: USA
 
 N: Robin Cornelius
diff --git a/Documentation/ABI/testing/sysfs-block b/Documentation/ABI/testing/sysfs-block
index 2b5d561..c1eb41c 100644
--- a/Documentation/ABI/testing/sysfs-block
+++ b/Documentation/ABI/testing/sysfs-block
@@ -206,16 +206,3 @@
 		when a discarded area is read the discard_zeroes_data
 		parameter will be set to one. Otherwise it will be 0 and
 		the result of reading a discarded area is undefined.
-What:		/sys/block/<disk>/alias
-Date:		Aug 2011
-Contact:	Nao Nishijima <nao.nishijima.xt@hitachi.com>
-Description:
-		A raw device name of a disk does not always point a same disk
-		each boot-up time. Therefore, users have to use persistent
-		device names, which udev creates when the kernel finds a disk,
-		instead of raw device name. However, kernel doesn't show those
-		persistent names on its messages (e.g. dmesg).
-		This file can store an alias of the disk and it would be
-		appeared in kernel messages if it is set. A disk can have an
-		alias which length is up to 255bytes. Users can use alphabets,
-		numbers, "-" and "_" in alias name. This file is writeonce.
diff --git a/Documentation/ABI/testing/sysfs-bus-rbd b/Documentation/ABI/testing/sysfs-bus-rbd
index fa72ccb..dbedafb 100644
--- a/Documentation/ABI/testing/sysfs-bus-rbd
+++ b/Documentation/ABI/testing/sysfs-bus-rbd
@@ -57,13 +57,6 @@
 
 	 $ echo <snap-name> > /sys/bus/rbd/devices/<dev-id>/snap_create
 
-rollback_snap
-
-	Rolls back data to the specified snapshot. This goes over the entire
-	list of rados blocks and sends a rollback command to each.
-
-	 $ echo <snap-name> > /sys/bus/rbd/devices/<dev-id>/snap_rollback
-
 snap_*
 
 	A directory per each snapshot
diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index c279158..196b8b9 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -32,7 +32,7 @@
       The Linux DRM layer contains code intended to support the needs
       of complex graphics devices, usually containing programmable
       pipelines well suited to 3D graphics acceleration.  Graphics
-      drivers in the kernel can make use of DRM functions to make
+      drivers in the kernel may make use of DRM functions to make
       tasks like memory management, interrupt handling and DMA easier,
       and provide a uniform interface to applications.
     </para>
@@ -57,10 +57,10 @@
       existing drivers.
     </para>
     <para>
-      First, we'll go over some typical driver initialization
+      First, we go over some typical driver initialization
       requirements, like setting up command buffers, creating an
       initial output configuration, and initializing core services.
-      Subsequent sections will cover core internals in more detail,
+      Subsequent sections cover core internals in more detail,
       providing implementation notes and examples.
     </para>
     <para>
@@ -74,7 +74,7 @@
     </para>
     <para>
       The core of every DRM driver is struct drm_driver.  Drivers
-      will typically statically initialize a drm_driver structure,
+      typically statically initialize a drm_driver structure,
       then pass it to drm_init() at load time.
     </para>
 
@@ -88,8 +88,8 @@
     </para>
     <programlisting>
       static struct drm_driver driver = {
-	/* don't use mtrr's here, the Xserver or user space app should
-	 * deal with them for intel hardware.
+	/* Don't use MTRRs here; the Xserver or userspace app should
+	 * deal with them for Intel hardware.
 	 */
 	.driver_features =
 	    DRIVER_USE_AGP | DRIVER_REQUIRE_AGP |
@@ -154,8 +154,8 @@
     </programlisting>
     <para>
       In the example above, taken from the i915 DRM driver, the driver
-      sets several flags indicating what core features it supports.
-      We'll go over the individual callbacks in later sections.  Since
+      sets several flags indicating what core features it supports;
+      we go over the individual callbacks in later sections.  Since
       flags indicate which features your driver supports to the DRM
       core, you need to set most of them prior to calling drm_init().  Some,
       like DRIVER_MODESET can be set later based on user supplied parameters,
@@ -203,8 +203,8 @@
 	<term>DRIVER_HAVE_IRQ</term><term>DRIVER_IRQ_SHARED</term>
 	<listitem>
 	  <para>
-	    DRIVER_HAVE_IRQ indicates whether the driver has a IRQ
-	    handler, DRIVER_IRQ_SHARED indicates whether the device &amp;
+	    DRIVER_HAVE_IRQ indicates whether the driver has an IRQ
+	    handler.  DRIVER_IRQ_SHARED indicates whether the device &amp;
 	    handler support shared IRQs (note that this is required of
 	    PCI drivers).
 	  </para>
@@ -214,8 +214,8 @@
 	<term>DRIVER_DMA_QUEUE</term>
 	<listitem>
 	  <para>
-	    If the driver queues DMA requests and completes them
-	    asynchronously, this flag should be set.  Deprecated.
+	    Should be set if the driver queues DMA requests and completes them
+	    asynchronously.  Deprecated.
 	  </para>
 	</listitem>
       </varlistentry>
@@ -238,7 +238,7 @@
     </variablelist>
     <para>
       In this specific case, the driver requires AGP and supports
-      IRQs.  DMA, as we'll see, is handled by device specific ioctls
+      IRQs.  DMA, as discussed later, is handled by device-specific ioctls
       in this case.  It also supports the kernel mode setting APIs, though
       unlike in the actual i915 driver source, this example unconditionally
       exports KMS capability.
@@ -269,36 +269,34 @@
       initial output configuration.
     </para>
     <para>
-      Note that the tasks performed at driver load time must not
-      conflict with DRM client requirements.  For instance, if user
+      If compatibility is a concern (e.g. with drivers converted over
+      to the new interfaces from the old ones), care must be taken to
+      prevent device initialization and control that is incompatible with
+      currently active userspace drivers.  For instance, if user
       level mode setting drivers are in use, it would be problematic
       to perform output discovery &amp; configuration at load time.
-      Likewise, if pre-memory management aware user level drivers are
+      Likewise, if user-level drivers unaware of memory management are
       in use, memory management and command buffer setup may need to
-      be omitted.  These requirements are driver specific, and care
+      be omitted.  These requirements are driver-specific, and care
       needs to be taken to keep both old and new applications and
       libraries working.  The i915 driver supports the "modeset"
       module parameter to control whether advanced features are
-      enabled at load time or in legacy fashion.  If compatibility is
-      a concern (e.g. with drivers converted over to the new interfaces
-      from the old ones), care must be taken to prevent incompatible
-      device initialization and control with the currently active
-      userspace drivers.
+      enabled at load time or in legacy fashion.
     </para>
 
     <sect2>
       <title>Driver private &amp; performance counters</title>
       <para>
 	The driver private hangs off the main drm_device structure and
-	can be used for tracking various device specific bits of
+	can be used for tracking various device-specific bits of
 	information, like register offsets, command buffer status,
 	register state for suspend/resume, etc.  At load time, a
-	driver can simply allocate one and set drm_device.dev_priv
-	appropriately; at unload the driver can free it and set
-	drm_device.dev_priv to NULL.
+	driver may simply allocate one and set drm_device.dev_priv
+	appropriately; it should be freed and drm_device.dev_priv set
+	to NULL when the driver is unloaded.
       </para>
       <para>
-	The DRM supports several counters which can be used for rough
+	The DRM supports several counters which may be used for rough
 	performance characterization.  Note that the DRM stat counter
 	system is not often used by applications, and supporting
 	additional counters is completely optional.
@@ -307,15 +305,15 @@
 	These interfaces are deprecated and should not be used.  If performance
 	monitoring is desired, the developer should investigate and
 	potentially enhance the kernel perf and tracing infrastructure to export
-	GPU related performance information to performance monitoring
-	tools and applications.
+	GPU related performance information for consumption by performance
+	monitoring tools and applications.
       </para>
     </sect2>
 
     <sect2>
       <title>Configuring the device</title>
       <para>
-	Obviously, device configuration will be device specific.
+	Obviously, device configuration is device-specific.
 	However, there are several common operations: finding a
 	device's PCI resources, mapping them, and potentially setting
 	up an IRQ handler.
@@ -323,10 +321,10 @@
       <para>
 	Finding &amp; mapping resources is fairly straightforward.  The
 	DRM wrapper functions, drm_get_resource_start() and
-	drm_get_resource_len() can be used to find BARs on the given
+	drm_get_resource_len(), may be used to find BARs on the given
 	drm_device struct.  Once those values have been retrieved, the
 	driver load function can call drm_addmap() to create a new
-	mapping for the BAR in question.  Note you'll probably want a
+	mapping for the BAR in question.  Note that you probably want a
 	drm_local_map_t in your driver private structure to track any
 	mappings you create.
 <!-- !Fdrivers/gpu/drm/drm_bufs.c drm_get_resource_* -->
@@ -335,20 +333,20 @@
       <para>
 	if compatibility with other operating systems isn't a concern
 	(DRM drivers can run under various BSD variants and OpenSolaris),
-	native Linux calls can be used for the above, e.g. pci_resource_*
+	native Linux calls may be used for the above, e.g. pci_resource_*
 	and iomap*/iounmap.  See the Linux device driver book for more
 	info.
       </para>
       <para>
-	Once you have a register map, you can use the DRM_READn() and
+	Once you have a register map, you may use the DRM_READn() and
 	DRM_WRITEn() macros to access the registers on your device, or
-	use driver specific versions to offset into your MMIO space
-	relative to a driver specific base pointer (see I915_READ for
-	example).
+	use driver-specific versions to offset into your MMIO space
+	relative to a driver-specific base pointer (see I915_READ for
+	an example).
       </para>
       <para>
 	If your device supports interrupt generation, you may want to
-	setup an interrupt handler at driver load time as well.  This
+	set up an interrupt handler when the driver is loaded.  This
 	is done using the drm_irq_install() function.  If your device
 	supports vertical blank interrupts, it should call
 	drm_vblank_init() to initialize the core vblank handling code before
@@ -357,7 +355,7 @@
       </para>
 <!--!Fdrivers/char/drm/drm_irq.c drm_irq_install-->
       <para>
-	Once your interrupt handler is registered (it'll use your
+	Once your interrupt handler is registered (it uses your
 	drm_driver.irq_handler as the actual interrupt handling
 	function), you can safely enable interrupts on your device,
 	assuming any other state your interrupt handler uses is also
@@ -371,10 +369,10 @@
 	using the pci_map_rom() call, a convenience function that
 	takes care of mapping the actual ROM, whether it has been
 	shadowed into memory (typically at address 0xc0000) or exists
-	on the PCI device in the ROM BAR.  Note that once you've
-	mapped the ROM and extracted any necessary information, be
-	sure to unmap it; on many devices the ROM address decoder is
-	shared with other BARs, so leaving it mapped can cause
+	on the PCI device in the ROM BAR.  Note that after the ROM
+	has been mapped and any necessary information has been extracted,
+	it should be unmapped; on many devices, the ROM address decoder is
+	shared with other BARs, so leaving it mapped could cause
 	undesired behavior like hangs or memory corruption.
 <!--!Fdrivers/pci/rom.c pci_map_rom-->
       </para>
@@ -389,9 +387,9 @@
 	should support a memory manager.
       </para>
       <para>
-	If your driver supports memory management (it should!), you'll
+	If your driver supports memory management (it should!), you
 	need to set that up at load time as well.  How you initialize
-	it depends on which memory manager you're using, TTM or GEM.
+	it depends on which memory manager you're using: TTM or GEM.
       </para>
       <sect3>
 	<title>TTM initialization</title>
@@ -401,7 +399,7 @@
 	  and devices with dedicated video RAM (VRAM), i.e. most discrete
 	  graphics devices.  If your device has dedicated RAM, supporting
 	  TTM is desirable.  TTM also integrates tightly with your
-	  driver specific buffer execution function.  See the radeon
+	  driver-specific buffer execution function.  See the radeon
 	  driver for examples.
 	</para>
 	<para>
@@ -429,21 +427,21 @@
 	  created by the memory manager at runtime.  Your global TTM should
 	  have a type of TTM_GLOBAL_TTM_MEM.  The size field for the global
 	  object should be sizeof(struct ttm_mem_global), and the init and
-	  release hooks should point at your driver specific init and
-	  release routines, which will probably eventually call
-	  ttm_mem_global_init and ttm_mem_global_release respectively.
+	  release hooks should point at your driver-specific init and
+	  release routines, which probably eventually call
+	  ttm_mem_global_init and ttm_mem_global_release, respectively.
 	</para>
 	<para>
 	  Once your global TTM accounting structure is set up and initialized
-	  (done by calling ttm_global_item_ref on the global object you
-	  just created), you'll need to create a buffer object TTM to
+	  by calling ttm_global_item_ref() on it,
+	  you need to create a buffer object TTM to
 	  provide a pool for buffer object allocation by clients and the
 	  kernel itself.  The type of this object should be TTM_GLOBAL_TTM_BO,
 	  and its size should be sizeof(struct ttm_bo_global).  Again,
-	  driver specific init and release functions can be provided,
-	  likely eventually calling ttm_bo_global_init and
-	  ttm_bo_global_release, respectively.  Also like the previous
-	  object, ttm_global_item_ref is used to create an initial reference
+	  driver-specific init and release functions may be provided,
+	  likely eventually calling ttm_bo_global_init() and
+	  ttm_bo_global_release(), respectively.  Also, like the previous
+	  object, ttm_global_item_ref() is used to create an initial reference
 	  count for the TTM, which will call your initialization function.
 	</para>
       </sect3>
@@ -453,27 +451,26 @@
 	  GEM is an alternative to TTM, designed specifically for UMA
 	  devices.  It has simpler initialization and execution requirements
 	  than TTM, but has no VRAM management capability.  Core GEM
-	  initialization is comprised of a basic drm_mm_init call to create
+	  is initialized by calling drm_mm_init() to create
 	  a GTT DRM MM object, which provides an address space pool for
-	  object allocation.  In a KMS configuration, the driver will
-	  need to allocate and initialize a command ring buffer following
-	  basic GEM initialization.  Most UMA devices have a so-called
+	  object allocation.  In a KMS configuration, the driver
+	  needs to allocate and initialize a command ring buffer following
+	  core GEM initialization.  A UMA device usually has what is called a
 	  "stolen" memory region, which provides space for the initial
 	  framebuffer and large, contiguous memory regions required by the
-	  device.  This space is not typically managed by GEM, and must
+	  device.  This space is not typically managed by GEM, and it must
 	  be initialized separately into its own DRM MM object.
 	</para>
 	<para>
-	  Initialization will be driver specific, and will depend on
-	  the architecture of the device.  In the case of Intel
+	  Initialization is driver-specific. In the case of Intel
 	  integrated graphics chips like 965GM, GEM initialization can
 	  be done by calling the internal GEM init function,
 	  i915_gem_do_init().  Since the 965GM is a UMA device
-	  (i.e. it doesn't have dedicated VRAM), GEM will manage
+	  (i.e. it doesn't have dedicated VRAM), GEM manages
 	  making regular RAM available for GPU operations.  Memory set
 	  aside by the BIOS (called "stolen" memory by the i915
-	  driver) will be managed by the DRM memrange allocator; the
-	  rest of the aperture will be managed by GEM.
+	  driver) is managed by the DRM memrange allocator; the
+	  rest of the aperture is managed by GEM.
 	  <programlisting>
 	    /* Basic memrange allocator for stolen space (aka vram) */
 	    drm_memrange_init(&amp;dev_priv->vram, 0, prealloc_size);
@@ -483,7 +480,7 @@
 <!--!Edrivers/char/drm/drm_memrange.c-->
 	</para>
 	<para>
-	  Once the memory manager has been set up, we can allocate the
+	  Once the memory manager has been set up, we may allocate the
 	  command buffer.  In the i915 case, this is also done with a
 	  GEM function, i915_gem_init_ringbuffer().
 	</para>
@@ -493,16 +490,25 @@
     <sect2>
       <title>Output configuration</title>
       <para>
-	The final initialization task is output configuration.  This involves
-	finding and initializing the CRTCs, encoders and connectors
-	for your device, creating an initial configuration and
-	registering a framebuffer console driver.
+	The final initialization task is output configuration.  This involves:
+	<itemizedlist>
+	  <listitem>
+	    Finding and initializing the CRTCs, encoders, and connectors
+	    for the device.
+	  </listitem>
+	  <listitem>
+	    Creating an initial configuration.
+	  </listitem>
+	  <listitem>
+	    Registering a framebuffer console driver.
+	  </listitem>
+	</itemizedlist>
       </para>
       <sect3>
 	<title>Output discovery and initialization</title>
 	<para>
-	  Several core functions exist to create CRTCs, encoders and
-	  connectors, namely drm_crtc_init(), drm_connector_init() and
+	  Several core functions exist to create CRTCs, encoders, and
+	  connectors, namely: drm_crtc_init(), drm_connector_init(), and
 	  drm_encoder_init(), along with several "helper" functions to
 	  perform common tasks.
 	</para>
@@ -555,10 +561,10 @@
 	</programlisting>
 	<para>
 	  In the example above (again, taken from the i915 driver), a
-	  CRT connector and encoder combination is created.  A device
-	  specific i2c bus is also created, for fetching EDID data and
+	  CRT connector and encoder combination is created.  A device-specific
+	  i2c bus is also created for fetching EDID data and
 	  performing monitor detection.  Once the process is complete,
-	  the new connector is registered with sysfs, to make its
+	  the new connector is registered with sysfs to make its
 	  properties available to applications.
 	</para>
 	<sect4>
@@ -567,12 +573,12 @@
 	    Since many PC-class graphics devices have similar display output
 	    designs, the DRM provides a set of helper functions to make
 	    output management easier.  The core helper routines handle
-	    encoder re-routing and disabling of unused functions following
-	    mode set.  Using the helpers is optional, but recommended for
+	    encoder re-routing and the disabling of unused functions following
+	    mode setting.  Using the helpers is optional, but recommended for
 	    devices with PC-style architectures (i.e. a set of display planes
 	    for feeding pixels to encoders which are in turn routed to
 	    connectors).  Devices with more complex requirements needing
-	    finer grained management can opt to use the core callbacks
+	    finer grained management may opt to use the core callbacks
 	    directly.
 	  </para>
 	  <para>
@@ -580,17 +586,25 @@
 	  </para>
 	</sect4>
 	<para>
-	  For each encoder, CRTC and connector, several functions must
-	  be provided, depending on the object type.  Encoder objects
-	  need to provide a DPMS (basically on/off) function, mode fixup
-	  (for converting requested modes into native hardware timings),
-	  and prepare, set and commit functions for use by the core DRM
-	  helper functions.  Connector helpers need to provide mode fetch and
-	  validity functions as well as an encoder matching function for
-	  returning an ideal encoder for a given connector.  The core
-	  connector functions include a DPMS callback, (deprecated)
-	  save/restore routines, detection, mode probing, property handling,
-	  and cleanup functions.
+	  Each encoder object needs to provide:
+	  <itemizedlist>
+	    <listitem>
+	      A DPMS (basically on/off) function.
+	    </listitem>
+	    <listitem>
+	      A mode-fixup function (for converting requested modes into
+	      native hardware timings).
+	    </listitem>
+	    <listitem>
+	      Functions (prepare, set, and commit) for use by the core DRM
+	      helper functions.
+	    </listitem>
+	  </itemizedlist>
+	  Connector helpers need to provide functions (mode-fetch, validity,
+	  and encoder-matching) for returning an ideal encoder for a given
+	  connector.  The core connector functions include a DPMS callback,
+	  save/restore routines (deprecated), detection, mode probing,
+	  property handling, and cleanup functions.
 	</para>
 <!--!Edrivers/char/drm/drm_crtc.h-->
 <!--!Edrivers/char/drm/drm_crtc.c-->
@@ -605,23 +619,34 @@
     <title>VBlank event handling</title>
     <para>
       The DRM core exposes two vertical blank related ioctls:
-      DRM_IOCTL_WAIT_VBLANK and DRM_IOCTL_MODESET_CTL.
+      <variablelist>
+        <varlistentry>
+          <term>DRM_IOCTL_WAIT_VBLANK</term>
+          <listitem>
+            <para>
+              This takes a struct drm_wait_vblank structure as its argument,
+              and it is used to block or request a signal when a specified
+              vblank event occurs.
+            </para>
+          </listitem>
+        </varlistentry>
+        <varlistentry>
+          <term>DRM_IOCTL_MODESET_CTL</term>
+          <listitem>
+            <para>
+              This should be called by application level drivers before and
+              after mode setting, since on many devices the vertical blank
+              counter is reset at that time.  Internally, the DRM snapshots
+              the last vblank count when the ioctl is called with the
+              _DRM_PRE_MODESET command, so that the counter won't go backwards
+              (which is dealt with when _DRM_POST_MODESET is used).
+            </para>
+          </listitem>
+        </varlistentry>
+      </variablelist>
 <!--!Edrivers/char/drm/drm_irq.c-->
     </para>
     <para>
-      DRM_IOCTL_WAIT_VBLANK takes a struct drm_wait_vblank structure
-      as its argument, and is used to block or request a signal when a
-      specified vblank event occurs.
-    </para>
-    <para>
-      DRM_IOCTL_MODESET_CTL should be called by application level
-      drivers before and after mode setting, since on many devices the
-      vertical blank counter will be reset at that time.  Internally,
-      the DRM snapshots the last vblank count when the ioctl is called
-      with the _DRM_PRE_MODESET command so that the counter won't go
-      backwards (which is dealt with when _DRM_POST_MODESET is used).
-    </para>
-    <para>
       To support the functions above, the DRM core provides several
       helper functions for tracking vertical blank counters, and
       requires drivers to provide several callbacks:
@@ -632,24 +657,24 @@
       register.  The enable and disable vblank callbacks should enable
       and disable vertical blank interrupts, respectively.  In the
       absence of DRM clients waiting on vblank events, the core DRM
-      code will use the disable_vblank() function to disable
-      interrupts, which saves power.  They'll be re-enabled again when
+      code uses the disable_vblank() function to disable
+      interrupts, which saves power.  They are re-enabled again when
       a client calls the vblank wait ioctl above.
     </para>
     <para>
-      Devices that don't provide a count register can simply use an
+      A device that doesn't provide a count register may simply use an
       internal atomic counter incremented on every vertical blank
-      interrupt, and can make their enable and disable vblank
-      functions into no-ops.
+      interrupt (and then treat the enable_vblank() and disable_vblank()
+      callbacks as no-ops).
     </para>
   </sect1>
 
   <sect1>
     <title>Memory management</title>
     <para>
-      The memory manager lies at the heart of many DRM operations, and
-      is also required to support advanced client features like OpenGL
-      pbuffers.  The DRM currently contains two memory managers, TTM
+      The memory manager lies at the heart of many DRM operations; it
+      is required to support advanced client features like OpenGL
+      pbuffers.  The DRM currently contains two memory managers: TTM
       and GEM.
     </para>
 
@@ -679,41 +704,46 @@
       <para>
 	GEM-enabled drivers must provide gem_init_object() and
 	gem_free_object() callbacks to support the core memory
-	allocation routines.  They should also provide several driver
-	specific ioctls to support command execution, pinning, buffer
+	allocation routines.  They should also provide several driver-specific
+	ioctls to support command execution, pinning, buffer
 	read &amp; write, mapping, and domain ownership transfers.
       </para>
       <para>
-	On a fundamental level, GEM involves several operations: memory
-	allocation and freeing, command execution, and aperture management
-	at command execution time.  Buffer object allocation is relatively
+	On a fundamental level, GEM involves several operations:
+	<itemizedlist>
+	  <listitem>Memory allocation and freeing</listitem>
+	  <listitem>Command execution</listitem>
+	  <listitem>Aperture management at command execution time</listitem>
+	</itemizedlist>
+	Buffer object allocation is relatively
 	straightforward and largely provided by Linux's shmem layer, which
 	provides memory to back each object.  When mapped into the GTT
 	or used in a command buffer, the backing pages for an object are
 	flushed to memory and marked write combined so as to be coherent
-	with the GPU.  Likewise, when the GPU finishes rendering to an object,
-	if the CPU accesses it, it must be made coherent with the CPU's view
+	with the GPU.  Likewise, if the CPU accesses an object after the GPU
+	has finished rendering to the object, then the object must be made
+	coherent with the CPU's view
 	of memory, usually involving GPU cache flushing of various kinds.
-	This core CPU&lt;-&gt;GPU coherency management is provided by the GEM
-	set domain function, which evaluates an object's current domain and
+	This core CPU&lt;-&gt;GPU coherency management is provided by a
+	device-specific ioctl, which evaluates an object's current domain and
 	performs any necessary flushing or synchronization to put the object
 	into the desired coherency domain (note that the object may be busy,
-	i.e. an active render target; in that case the set domain function
-	will block the client and wait for rendering to complete before
+	i.e. an active render target; in that case, setting the domain
+	blocks the client and waits for rendering to complete before
 	performing any necessary flushing operations).
       </para>
       <para>
 	Perhaps the most important GEM function is providing a command
 	execution interface to clients.  Client programs construct command
-	buffers containing references to previously allocated memory objects
-	and submit them to GEM.  At that point, GEM will take care to bind
+	buffers containing references to previously allocated memory objects,
+	and then submit them to GEM.  At that point, GEM takes care to bind
 	all the objects into the GTT, execute the buffer, and provide
 	necessary synchronization between clients accessing the same buffers.
 	This often involves evicting some objects from the GTT and re-binding
 	others (a fairly expensive operation), and providing relocation
 	support which hides fixed GTT offsets from clients.  Clients must
 	take care not to submit command buffers that reference more objects
-	than can fit in the GTT or GEM will reject them and no rendering
+	than can fit in the GTT; otherwise, GEM will reject them and no rendering
 	will occur.  Similarly, if several objects in the buffer require
 	fence registers to be allocated for correct rendering (e.g. 2D blits
 	on pre-965 chips), care must be taken not to require more fence
@@ -729,7 +759,7 @@
     <title>Output management</title>
     <para>
       At the core of the DRM output management code is a set of
-      structures representing CRTCs, encoders and connectors.
+      structures representing CRTCs, encoders, and connectors.
     </para>
     <para>
       A CRTC is an abstraction representing a part of the chip that
@@ -765,21 +795,19 @@
   <sect1>
     <title>Framebuffer management</title>
     <para>
-      In order to set a mode on a given CRTC, encoder and connector
-      configuration, clients need to provide a framebuffer object which
-      will provide a source of pixels for the CRTC to deliver to the encoder(s)
-      and ultimately the connector(s) in the configuration.  A framebuffer
-      is fundamentally a driver specific memory object, made into an opaque
-      handle by the DRM addfb function.  Once an fb has been created this
-      way it can be passed to the KMS mode setting routines for use in
-      a configuration.
+      Clients need to provide a framebuffer object which provides a source
+      of pixels for a CRTC to deliver to the encoder(s) and ultimately the
+      connector(s). A framebuffer is fundamentally a driver-specific memory
+      object, made into an opaque handle by the DRM's addfb() function.
+      Once a framebuffer has been created this way, it may be passed to the
+      KMS mode setting routines for use in a completed configuration.
     </para>
   </sect1>
 
   <sect1>
     <title>Command submission &amp; fencing</title>
     <para>
-      This should cover a few device specific command submission
+      This should cover a few device-specific command submission
       implementations.
     </para>
   </sect1>
@@ -789,7 +817,7 @@
     <para>
       The DRM core provides some suspend/resume code, but drivers
       wanting full suspend/resume support should provide save() and
-      restore() functions.  These will be called at suspend,
+      restore() functions.  These are called at suspend,
       hibernate, or resume time, and should perform any state save or
       restore required by your device across suspend or hibernate
       states.
@@ -812,8 +840,8 @@
     <para>
       The DRM core exports several interfaces to applications,
       generally intended to be used through corresponding libdrm
-      wrapper functions.  In addition, drivers export device specific
-      interfaces for use by userspace drivers &amp; device aware
+      wrapper functions.  In addition, drivers export device-specific
+      interfaces for use by userspace drivers &amp; device-aware
       applications through ioctls and sysfs files.
     </para>
     <para>
@@ -822,8 +850,8 @@
       management, memory management, and output management.
     </para>
     <para>
-      Cover generic ioctls and sysfs layout here.  Only need high
-      level info, since man pages will cover the rest.
+      Cover generic ioctls and sysfs layout here.  We only need high-level
+      info, since man pages should cover the rest.
     </para>
   </chapter>
 
diff --git a/Documentation/DocBook/uio-howto.tmpl b/Documentation/DocBook/uio-howto.tmpl
index 54883de..ac3d001 100644
--- a/Documentation/DocBook/uio-howto.tmpl
+++ b/Documentation/DocBook/uio-howto.tmpl
@@ -521,6 +521,11 @@
 
 <itemizedlist>
 <listitem><para>
+<varname>const char *name</varname>: Optional. Set this to help identify
+the memory region, it will show up in the corresponding sysfs node.
+</para></listitem>
+
+<listitem><para>
 <varname>int memtype</varname>: Required if the mapping is used. Set this to
 <varname>UIO_MEM_PHYS</varname> if you you have physical memory on your
 card to be mapped. Use <varname>UIO_MEM_LOGICAL</varname> for logical
@@ -553,7 +558,7 @@
 </itemizedlist>
 
 <para>
-Please do not touch the <varname>kobj</varname> element of
+Please do not touch the <varname>map</varname> element of
 <varname>struct uio_mem</varname>! It is used by the UIO framework
 to set up sysfs files for this mapping. Simply leave it alone.
 </para>
diff --git a/Documentation/blockdev/cciss.txt b/Documentation/blockdev/cciss.txt
index 71464e0..b79d0a1 100644
--- a/Documentation/blockdev/cciss.txt
+++ b/Documentation/blockdev/cciss.txt
@@ -98,14 +98,12 @@
 "SCSI support" in your kernel configuration to be able to use SCSI
 tape drives with your Smart Array 5xxx controller.
 
-Additionally, note that the driver will not engage the SCSI core at init 
-time.  The driver must be directed to dynamically engage the SCSI core via 
-the /proc filesystem entry which the "block" side of the driver creates as 
-/proc/driver/cciss/cciss* at runtime.  This is because at driver init time, 
-the SCSI core may not yet be initialized (because the driver is a block 
-driver) and attempting to register it with the SCSI core in such a case 
-would cause a hang.  This is best done via an initialization script 
-(typically in /etc/init.d, but could vary depending on distribution). 
+Additionally, note that the driver will engage the SCSI core at init
+time if any tape drives or medium changers are detected.  The driver may
+also be directed to dynamically engage the SCSI core via the /proc filesystem
+entry which the "block" side of the driver creates as
+/proc/driver/cciss/cciss* at runtime.  This is best done via a script.
+
 For example:
 
 	for x in /proc/driver/cciss/cciss[0-9]*
diff --git a/Documentation/cgroups/freezer-subsystem.txt b/Documentation/cgroups/freezer-subsystem.txt
index c21d777..7e62de1 100644
--- a/Documentation/cgroups/freezer-subsystem.txt
+++ b/Documentation/cgroups/freezer-subsystem.txt
@@ -33,9 +33,9 @@
 
 	From a second, unrelated bash shell:
 	$ kill -SIGSTOP 16690
-	$ kill -SIGCONT 16990
+	$ kill -SIGCONT 16690
 
-	<at this point 16990 exits and causes 16644 to exit too>
+	<at this point 16690 exits and causes 16644 to exit too>
 
 This happens because bash can observe both signals and choose how it
 responds to them.
diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt
index e855278..874921e 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.txt
+++ b/Documentation/devicetree/bindings/vendor-prefixes.txt
@@ -33,6 +33,7 @@
 ramtron	Ramtron International
 samsung	Samsung Semiconductor
 schindler	Schindler
+sil	Silicon Image
 simtek
 sirf	SiRF Technology, Inc.
 stericsson	ST-Ericsson
diff --git a/Documentation/filesystems/btrfs.txt b/Documentation/filesystems/btrfs.txt
index 64087c3..7671352 100644
--- a/Documentation/filesystems/btrfs.txt
+++ b/Documentation/filesystems/btrfs.txt
@@ -63,8 +63,8 @@
 Userspace tools for creating and manipulating Btrfs file systems are
 available from the git repository at the following location:
 
- http://git.kernel.org/?p=linux/kernel/git/mason/btrfs-progs-unstable.git
- git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-progs-unstable.git
+ http://git.kernel.org/?p=linux/kernel/git/mason/btrfs-progs.git
+ git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-progs.git
 
 These include the following tools:
 
diff --git a/Documentation/i2c/ten-bit-addresses b/Documentation/i2c/ten-bit-addresses
index e989070..cdfe139 100644
--- a/Documentation/i2c/ten-bit-addresses
+++ b/Documentation/i2c/ten-bit-addresses
@@ -1,22 +1,24 @@
 The I2C protocol knows about two kinds of device addresses: normal 7 bit
 addresses, and an extended set of 10 bit addresses. The sets of addresses
 do not intersect: the 7 bit address 0x10 is not the same as the 10 bit
-address 0x10 (though a single device could respond to both of them). You
-select a 10 bit address by adding an extra byte after the address
-byte:
-  S Addr7 Rd/Wr ....
-becomes
-  S 11110 Addr10 Rd/Wr
-S is the start bit, Rd/Wr the read/write bit, and if you count the number
-of bits, you will see the there are 8 after the S bit for 7 bit addresses,
-and 16 after the S bit for 10 bit addresses.
+address 0x10 (though a single device could respond to both of them).
 
-WARNING! The current 10 bit address support is EXPERIMENTAL. There are
-several places in the code that will cause SEVERE PROBLEMS with 10 bit
-addresses, even though there is some basic handling and hooks. Also,
-almost no supported adapter handles the 10 bit addresses correctly.
+I2C messages to and from 10-bit address devices have a different format.
+See the I2C specification for the details.
 
-As soon as a real 10 bit address device is spotted 'in the wild', we
-can and will add proper support. Right now, 10 bit address devices
-are defined by the I2C protocol, but we have never seen a single device
-which supports them.
+The current 10 bit address support is minimal. It should work, however
+you can expect some problems along the way:
+* Not all bus drivers support 10-bit addresses. Some don't because the
+  hardware doesn't support them (SMBus doesn't require 10-bit address
+  support for example), some don't because nobody bothered adding the
+  code (or it's there but not working properly.) Software implementation
+  (i2c-algo-bit) is known to work.
+* Some optional features do not support 10-bit addresses. This is the
+  case of automatic detection and instantiation of devices by their,
+  drivers, for example.
+* Many user-space packages (for example i2c-tools) lack support for
+  10-bit addresses.
+
+Note that 10-bit address devices are still pretty rare, so the limitations
+listed above could stay for a long time, maybe even forever if nobody
+needs them to be fixed.
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index a0c5c5f..81c287f 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -315,12 +315,12 @@
 			CPU-intensive style benchmark, and it can vary highly in
 			a microbenchmark depending on workload and compiler.
 
-			1: only for 32-bit processes
-			2: only for 64-bit processes
+			32: only for 32-bit processes
+			64: only for 64-bit processes
 			on: enable for both 32- and 64-bit processes
 			off: disable for both 32- and 64-bit processes
 
-	amd_iommu=	[HW,X86-84]
+	amd_iommu=	[HW,X86-64]
 			Pass parameters to the AMD IOMMU driver in the system.
 			Possible values are:
 			fullflush - enable flushing of IO/TLB entries when
diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index cb7f314..589f2da 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -20,7 +20,7 @@
 	default FALSE
 
 min_pmtu - INTEGER
-	default 562 - minimum discovered Path MTU
+	default 552 - minimum discovered Path MTU
 
 route/max_size - INTEGER
 	Maximum number of routes allowed in the kernel.  Increase
@@ -282,11 +282,11 @@
 	Default: 0 (off)
 
 tcp_max_syn_backlog - INTEGER
-	Maximal number of remembered connection requests, which are
-	still did not receive an acknowledgment from connecting client.
-	Default value is 1024 for systems with more than 128Mb of memory,
-	and 128 for low memory machines. If server suffers of overload,
-	try to increase this number.
+	Maximal number of remembered connection requests, which have not
+	received an acknowledgment from connecting client.
+	The minimal value is 128 for low memory machines, and it will
+	increase in proportion to the memory of machine.
+	If server suffers from overload, try increasing this number.
 
 tcp_max_tw_buckets - INTEGER
 	Maximal number of timewait sockets held by system simultaneously.
diff --git a/Documentation/power/devices.txt b/Documentation/power/devices.txt
index 646a89e..3139fb5 100644
--- a/Documentation/power/devices.txt
+++ b/Documentation/power/devices.txt
@@ -123,9 +123,10 @@
 Subsystem-Level Methods
 -----------------------
 The core methods to suspend and resume devices reside in struct dev_pm_ops
-pointed to by the pm member of struct bus_type, struct device_type and
-struct class.  They are mostly of interest to the people writing infrastructure
-for buses, like PCI or USB, or device type and device class drivers.
+pointed to by the ops member of struct dev_pm_domain, or by the pm member of
+struct bus_type, struct device_type and struct class.  They are mostly of
+interest to the people writing infrastructure for platforms and buses, like PCI
+or USB, or device type and device class drivers.
 
 Bus drivers implement these methods as appropriate for the hardware and the
 drivers using it; PCI works differently from USB, and so on.  Not many people
@@ -139,41 +140,57 @@
 
 /sys/devices/.../power/wakeup files
 -----------------------------------
-All devices in the driver model have two flags to control handling of wakeup
-events (hardware signals that can force the device and/or system out of a low
-power state).  These flags are initialized by bus or device driver code using
+All device objects in the driver model contain fields that control the handling
+of system wakeup events (hardware signals that can force the system out of a
+sleep state).  These fields are initialized by bus or device driver code using
 device_set_wakeup_capable() and device_set_wakeup_enable(), defined in
 include/linux/pm_wakeup.h.
 
-The "can_wakeup" flag just records whether the device (and its driver) can
+The "power.can_wakeup" flag just records whether the device (and its driver) can
 physically support wakeup events.  The device_set_wakeup_capable() routine
-affects this flag.  The "should_wakeup" flag controls whether the device should
-try to use its wakeup mechanism.  device_set_wakeup_enable() affects this flag;
-for the most part drivers should not change its value.  The initial value of
-should_wakeup is supposed to be false for the majority of devices; the major
-exceptions are power buttons, keyboards, and Ethernet adapters whose WoL
-(wake-on-LAN) feature has been set up with ethtool.  It should also default
-to true for devices that don't generate wakeup requests on their own but merely
-forward wakeup requests from one bus to another (like PCI bridges).
+affects this flag.  The "power.wakeup" field is a pointer to an object of type
+struct wakeup_source used for controlling whether or not the device should use
+its system wakeup mechanism and for notifying the PM core of system wakeup
+events signaled by the device.  This object is only present for wakeup-capable
+devices (i.e. devices whose "can_wakeup" flags are set) and is created (or
+removed) by device_set_wakeup_capable().
 
 Whether or not a device is capable of issuing wakeup events is a hardware
 matter, and the kernel is responsible for keeping track of it.  By contrast,
 whether or not a wakeup-capable device should issue wakeup events is a policy
 decision, and it is managed by user space through a sysfs attribute: the
-power/wakeup file.  User space can write the strings "enabled" or "disabled" to
-set or clear the "should_wakeup" flag, respectively.  This file is only present
-for wakeup-capable devices (i.e. devices whose "can_wakeup" flags are set)
-and is created (or removed) by device_set_wakeup_capable().  Reads from the
-file will return the corresponding string.
+"power/wakeup" file.  User space can write the strings "enabled" or "disabled"
+to it to indicate whether or not, respectively, the device is supposed to signal
+system wakeup.  This file is only present if the "power.wakeup" object exists
+for the given device and is created (or removed) along with that object, by
+device_set_wakeup_capable().  Reads from the file will return the corresponding
+string.
 
-The device_may_wakeup() routine returns true only if both flags are set.
+The "power/wakeup" file is supposed to contain the "disabled" string initially
+for the majority of devices; the major exceptions are power buttons, keyboards,
+and Ethernet adapters whose WoL (wake-on-LAN) feature has been set up with
+ethtool.  It should also default to "enabled" for devices that don't generate
+wakeup requests on their own but merely forward wakeup requests from one bus to
+another (like PCI Express ports).
+
+The device_may_wakeup() routine returns true only if the "power.wakeup" object
+exists and the corresponding "power/wakeup" file contains the string "enabled".
 This information is used by subsystems, like the PCI bus type code, to see
 whether or not to enable the devices' wakeup mechanisms.  If device wakeup
 mechanisms are enabled or disabled directly by drivers, they also should use
 device_may_wakeup() to decide what to do during a system sleep transition.
-However for runtime power management, wakeup events should be enabled whenever
-the device and driver both support them, regardless of the should_wakeup flag.
+Device drivers, however, are not supposed to call device_set_wakeup_enable()
+directly in any case.
 
+It ought to be noted that system wakeup is conceptually different from "remote
+wakeup" used by runtime power management, although it may be supported by the
+same physical mechanism.  Remote wakeup is a feature allowing devices in
+low-power states to trigger specific interrupts to signal conditions in which
+they should be put into the full-power state.  Those interrupts may or may not
+be used to signal system wakeup events, depending on the hardware design.  On
+some systems it is impossible to trigger them from system sleep states.  In any
+case, remote wakeup should always be enabled for runtime power management for
+all devices and drivers that support it.
 
 /sys/devices/.../power/control files
 ------------------------------------
@@ -249,20 +266,31 @@
 support all these callbacks and not all drivers use all the callbacks.  The
 various phases always run after tasks have been frozen and before they are
 unfrozen.  Furthermore, the *_noirq phases run at a time when IRQ handlers have
-been disabled (except for those marked with the IRQ_WAKEUP flag).
+been disabled (except for those marked with the IRQF_NO_SUSPEND flag).
 
-All phases use bus, type, or class callbacks (that is, methods defined in
-dev->bus->pm, dev->type->pm, or dev->class->pm).  These callbacks are mutually
-exclusive, so if the device type provides a struct dev_pm_ops object pointed to
-by its pm field (i.e. both dev->type and dev->type->pm are defined), the
-callbacks included in that object (i.e. dev->type->pm) will be used.  Otherwise,
-if the class provides a struct dev_pm_ops object pointed to by its pm field
-(i.e. both dev->class and dev->class->pm are defined), the PM core will use the
-callbacks from that object (i.e. dev->class->pm).  Finally, if the pm fields of
-both the device type and class objects are NULL (or those objects do not exist),
-the callbacks provided by the bus (that is, the callbacks from dev->bus->pm)
-will be used (this allows device types to override callbacks provided by bus
-types or classes if necessary).
+All phases use PM domain, bus, type, or class callbacks (that is, methods
+defined in dev->pm_domain->ops, dev->bus->pm, dev->type->pm, or dev->class->pm).
+These callbacks are regarded by the PM core as mutually exclusive.  Moreover,
+PM domain callbacks always take precedence over bus, type and class callbacks,
+while type callbacks take precedence over bus and class callbacks, and class
+callbacks take precedence over bus callbacks.  To be precise, the following
+rules are used to determine which callback to execute in the given phase:
+
+    1.	If dev->pm_domain is present, the PM core will attempt to execute the
+	callback included in dev->pm_domain->ops.  If that callback is not
+	present, no action will be carried out for the given device.
+
+    2.	Otherwise, if both dev->type and dev->type->pm are present, the callback
+	included in dev->type->pm will be executed.
+
+    3.	Otherwise, if both dev->class and dev->class->pm are present, the
+	callback included in dev->class->pm will be executed.
+
+    4.	Otherwise, if both dev->bus and dev->bus->pm are present, the callback
+	included in dev->bus->pm will be executed.
+
+This allows PM domains and device types to override callbacks provided by bus
+types or device classes if necessary.
 
 These callbacks may in turn invoke device- or driver-specific methods stored in
 dev->driver->pm, but they don't have to.
@@ -283,9 +311,8 @@
 
 	After the prepare callback method returns, no new children may be
 	registered below the device.  The method may also prepare the device or
-	driver in some way for the upcoming system power transition (for
-	example, by allocating additional memory required for this purpose), but
-	it should not put the device into a low-power state.
+	driver in some way for the upcoming system power transition, but it
+	should not put the device into a low-power state.
 
     2.	The suspend methods should quiesce the device to stop it from performing
 	I/O.  They also may save the device registers and put it into the
diff --git a/Documentation/power/runtime_pm.txt b/Documentation/power/runtime_pm.txt
index 5336149..c2ae8bf 100644
--- a/Documentation/power/runtime_pm.txt
+++ b/Documentation/power/runtime_pm.txt
@@ -44,25 +44,33 @@
 };
 
 The ->runtime_suspend(), ->runtime_resume() and ->runtime_idle() callbacks
-are executed by the PM core for either the power domain, or the device type
-(if the device power domain's struct dev_pm_ops does not exist), or the class
-(if the device power domain's and type's struct dev_pm_ops object does not
-exist), or the bus type (if the device power domain's, type's and class'
-struct dev_pm_ops objects do not exist) of the given device, so the priority
-order of callbacks from high to low is that power domain callbacks, device
-type callbacks, class callbacks and bus type callbacks, and the high priority
-one will take precedence over low priority one. The bus type, device type and
-class callbacks are referred to as subsystem-level callbacks in what follows,
-and generally speaking, the power domain callbacks are used for representing
-power domains within a SoC.
+are executed by the PM core for the device's subsystem that may be either of
+the following:
+
+  1. PM domain of the device, if the device's PM domain object, dev->pm_domain,
+     is present.
+
+  2. Device type of the device, if both dev->type and dev->type->pm are present.
+
+  3. Device class of the device, if both dev->class and dev->class->pm are
+     present.
+
+  4. Bus type of the device, if both dev->bus and dev->bus->pm are present.
+
+The PM core always checks which callback to use in the order given above, so the
+priority order of callbacks from high to low is: PM domain, device type, class
+and bus type.  Moreover, the high-priority one will always take precedence over
+a low-priority one.  The PM domain, bus type, device type and class callbacks
+are referred to as subsystem-level callbacks in what follows.
 
 By default, the callbacks are always invoked in process context with interrupts
 enabled.  However, subsystems can use the pm_runtime_irq_safe() helper function
-to tell the PM core that a device's ->runtime_suspend() and ->runtime_resume()
-callbacks should be invoked in atomic context with interrupts disabled.
-This implies that these callback routines must not block or sleep, but it also
-means that the synchronous helper functions listed at the end of Section 4 can
-be used within an interrupt handler or in an atomic context.
+to tell the PM core that their ->runtime_suspend(), ->runtime_resume() and
+->runtime_idle() callbacks may be invoked in atomic context with interrupts
+disabled for a given device.  This implies that the callback routines in
+question must not block or sleep, but it also means that the synchronous helper
+functions listed at the end of Section 4 may be used for that device within an
+interrupt handler or generally in an atomic context.
 
 The subsystem-level suspend callback is _entirely_ _responsible_ for handling
 the suspend of the device as appropriate, which may, but need not include
diff --git a/Documentation/serial/serial-rs485.txt b/Documentation/serial/serial-rs485.txt
index 079cb3d..41c8378 100644
--- a/Documentation/serial/serial-rs485.txt
+++ b/Documentation/serial/serial-rs485.txt
@@ -97,15 +97,23 @@
 
 	struct serial_rs485 rs485conf;
 
-	/* Set RS485 mode: */
+	/* Enable RS485 mode: */
 	rs485conf.flags |= SER_RS485_ENABLED;
 
+	/* Set logical level for RTS pin equal to 1 when sending: */
+	rs485conf.flags |= SER_RS485_RTS_ON_SEND;
+	/* or, set logical level for RTS pin equal to 0 when sending: */
+	rs485conf.flags &= ~(SER_RS485_RTS_ON_SEND);
+
+	/* Set logical level for RTS pin equal to 1 after sending: */
+	rs485conf.flags |= SER_RS485_RTS_AFTER_SEND;
+	/* or, set logical level for RTS pin equal to 0 after sending: */
+	rs485conf.flags &= ~(SER_RS485_RTS_AFTER_SEND);
+
 	/* Set rts delay before send, if needed: */
-	rs485conf.flags |= SER_RS485_RTS_BEFORE_SEND;
 	rs485conf.delay_rts_before_send = ...;
 
 	/* Set rts delay after send, if needed: */
-	rs485conf.flags |= SER_RS485_RTS_AFTER_SEND;
 	rs485conf.delay_rts_after_send = ...;
 
 	/* Set this flag if you want to receive data even whilst sending data */
diff --git a/Documentation/sound/alsa/HD-Audio-Models.txt b/Documentation/sound/alsa/HD-Audio-Models.txt
index 4f34432..edad99a 100644
--- a/Documentation/sound/alsa/HD-Audio-Models.txt
+++ b/Documentation/sound/alsa/HD-Audio-Models.txt
@@ -349,6 +349,7 @@
   ref		Reference board
   mic-ref	Reference board with power management for ports
   dell-s14	Dell laptop
+  dell-vostro-3500	Dell Vostro 3500 laptop
   hp		HP laptops with (inverted) mute-LED
   hp-dv7-4000	HP dv-7 4000
   auto		BIOS setup (default)
diff --git a/Documentation/sound/alsa/HD-Audio.txt b/Documentation/sound/alsa/HD-Audio.txt
index 03e2771..91fee3b 100644
--- a/Documentation/sound/alsa/HD-Audio.txt
+++ b/Documentation/sound/alsa/HD-Audio.txt
@@ -579,7 +579,7 @@
 ~~~~~~~~~~~~~~~~
 The latest development codes for HD-audio are found on sound git tree:
 
-- git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6.git
+- git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git
 
 The master branch or for-next branches can be used as the main
 development branches in general while the HD-audio specific patches
@@ -594,7 +594,7 @@
 install(-modules).  See INSTALL in the package.  The snapshot tarballs
 are found at:
 
-- ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/
+- ftp://ftp.suse.com/pub/people/tiwai/snapshot/
 
 
 Sending a Bug Report
@@ -696,7 +696,7 @@
 
 The hda-verb program is found in the ftp directory:
 
-- ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/misc/
+- ftp://ftp.suse.com/pub/people/tiwai/misc/
 
 Also a git repository is available:
 
@@ -764,7 +764,7 @@
 
 The package is found in:
 
-- ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/misc/
+- ftp://ftp.suse.com/pub/people/tiwai/misc/
 
 A git repository is available:
 
diff --git a/Documentation/sound/alsa/soc/machine.txt b/Documentation/sound/alsa/soc/machine.txt
index 3e2ec9c..d50c14d 100644
--- a/Documentation/sound/alsa/soc/machine.txt
+++ b/Documentation/sound/alsa/soc/machine.txt
@@ -50,8 +50,7 @@
 The machine DAI configuration glues all the codec and CPU DAIs together. It can
 also be used to set up the DAI system clock and for any machine related DAI
 initialisation e.g. the machine audio map can be connected to the codec audio
-map, unconnected codec pins can be set as such. Please see corgi.c, spitz.c
-for examples.
+map, unconnected codec pins can be set as such.
 
 struct snd_soc_dai_link is used to set up each DAI in your machine. e.g.
 
@@ -83,8 +82,7 @@
 The machine driver can optionally extend the codec power map and to become an
 audio power map of the audio subsystem. This allows for automatic power up/down
 of speaker/HP amplifiers, etc. Codec pins can be connected to the machines jack
-sockets in the machine init function. See soc/pxa/spitz.c and dapm.txt for
-details.
+sockets in the machine init function.
 
 
 Machine Controls
diff --git a/Documentation/usb/linux-cdc-acm.inf b/Documentation/usb/linux-cdc-acm.inf
index 37a02ce..f0ffc27 100644
--- a/Documentation/usb/linux-cdc-acm.inf
+++ b/Documentation/usb/linux-cdc-acm.inf
@@ -90,10 +90,10 @@
 [SourceDisksFiles]
 [SourceDisksNames]
 [DeviceList]
-%DESCRIPTION%=DriverInstall, USB\VID_0525&PID_A4A7, USB\VID_1D6B&PID_0104&MI_02
+%DESCRIPTION%=DriverInstall, USB\VID_0525&PID_A4A7, USB\VID_1D6B&PID_0104&MI_02, USB\VID_1D6B&PID_0106&MI_00
 
 [DeviceList.NTamd64]
-%DESCRIPTION%=DriverInstall, USB\VID_0525&PID_A4A7, USB\VID_1D6B&PID_0104&MI_02
+%DESCRIPTION%=DriverInstall, USB\VID_0525&PID_A4A7, USB\VID_1D6B&PID_0104&MI_02, USB\VID_1D6B&PID_0106&MI_00
 
 
 ;------------------------------------------------------------------------------
diff --git a/Kbuild b/Kbuild
index 4caab4f..b8b708a 100644
--- a/Kbuild
+++ b/Kbuild
@@ -92,7 +92,7 @@
 targets += missing-syscalls
 
 quiet_cmd_syscalls = CALL    $<
-      cmd_syscalls = $(CONFIG_SHELL) $< $(CC) $(c_flags)
+      cmd_syscalls = $(CONFIG_SHELL) $< $(CC) $(c_flags) $(missing_syscalls_flags)
 
 missing-syscalls: scripts/checksyscalls.sh $(offsets-file) FORCE
 	$(call cmd,syscalls)
diff --git a/MAINTAINERS b/MAINTAINERS
index 4808256..855afd4 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -511,8 +511,8 @@
 L:	iommu@lists.linux-foundation.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/joro/linux-2.6-iommu.git
 S:	Supported
-F:	arch/x86/kernel/amd_iommu*.c
-F:	arch/x86/include/asm/amd_iommu*.h
+F:	drivers/iommu/amd_iommu*.[ch]
+F:	include/linux/amd-iommu.h
 
 AMD MICROCODE UPDATE SUPPORT
 M:	Andreas Herrmann <andreas.herrmann3@amd.com>
@@ -789,6 +789,7 @@
 S:	Maintained
 T:	git git://git.pengutronix.de/git/imx/linux-2.6.git
 F:	arch/arm/mach-mx*/
+F:	arch/arm/mach-imx/
 F:	arch/arm/plat-mxc/
 
 ARM/FREESCALE IMX51
@@ -804,6 +805,13 @@
 T:	git git://git.linaro.org/people/shawnguo/linux-2.6.git
 F:	arch/arm/mach-imx/*imx6*
 
+ARM/FREESCALE MXS ARM ARCHITECTURE
+M:	Shawn Guo <shawn.guo@linaro.org>
+L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
+S:	Maintained
+T:	git git://git.linaro.org/people/shawnguo/linux-2.6.git
+F:	arch/arm/mach-mxs/
+
 ARM/GLOMATION GESBC9312SX MACHINE SUPPORT
 M:	Lennert Buytenhek <kernel@wantstofly.org>
 L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
@@ -1046,35 +1054,18 @@
 M:	Ben Dooks <ben-linux@fluff.org>
 M:	Kukjin Kim <kgene.kim@samsung.com>
 L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
+L:	linux-samsung-soc@vger.kernel.org (moderated for non-subscribers)
 W:	http://www.fluff.org/ben/linux/
 S:	Maintained
 F:	arch/arm/plat-samsung/
 F:	arch/arm/plat-s3c24xx/
 F:	arch/arm/plat-s5p/
+F:	arch/arm/mach-s3c24*/
+F:	arch/arm/mach-s3c64xx/
 F:	drivers/*/*s3c2410*
 F:	drivers/*/*/*s3c2410*
-
-ARM/S3C2410 ARM ARCHITECTURE
-M:	Ben Dooks <ben-linux@fluff.org>
-L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
-W:	http://www.fluff.org/ben/linux/
-S:	Maintained
-F:	arch/arm/mach-s3c2410/
-
-ARM/S3C244x ARM ARCHITECTURE
-M:	Ben Dooks <ben-linux@fluff.org>
-L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
-W:	http://www.fluff.org/ben/linux/
-S:	Maintained
-F:	arch/arm/mach-s3c2440/
-F:	arch/arm/mach-s3c2443/
-
-ARM/S3C64xx ARM ARCHITECTURE
-M:	Ben Dooks <ben-linux@fluff.org>
-L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
-W:	http://www.fluff.org/ben/linux/
-S:	Maintained
-F:	arch/arm/mach-s3c64xx/
+F:	drivers/spi/spi-s3c*
+F:	sound/soc/samsung/*
 
 ARM/S5P EXYNOS ARM ARCHITECTURES
 M:	Kukjin Kim <kgene.kim@samsung.com>
@@ -1106,6 +1097,7 @@
 ARM/SAMSUNG S5P SERIES Multi Format Codec (MFC) SUPPORT
 M:	Kyungmin Park <kyungmin.park@samsung.com>
 M:	Kamil Debski <k.debski@samsung.com>
+M:     Jeongtae Park <jtp.park@samsung.com>
 L:	linux-arm-kernel@lists.infradead.org
 L:	linux-media@vger.kernel.org
 S:	Maintained
@@ -1788,6 +1780,14 @@
 F:	net/wireless/*
 X:	net/wireless/wext*
 
+CHAR and MISC DRIVERS
+M:	Arnd Bergmann <arnd@arndb.de>
+M:	Greg Kroah-Hartman <greg@kroah.com>
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git
+S:	Maintained
+F:	drivers/char/*
+F:	drivers/misc/*
+
 CHECKPATCH
 M:	Andy Whitcroft <apw@canonical.com>
 S:	Supported
@@ -1926,9 +1926,11 @@
 F:	drivers/connector/
 
 CONTROL GROUPS (CGROUPS)
-M:	Paul Menage <paul@paulmenage.org>
+M:	Tejun Heo <tj@kernel.org>
 M:	Li Zefan <lizf@cn.fujitsu.com>
 L:	containers@lists.linux-foundation.org
+L:	cgroups@vger.kernel.org
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.git
 S:	Maintained
 F:	include/linux/cgroup*
 F:	kernel/cgroup*
@@ -2342,6 +2344,13 @@
 F:	drivers/gpu/drm/i915
 F:	include/drm/i915*
 
+DRM DRIVERS FOR EXYNOS
+M:	Inki Dae <inki.dae@samsung.com>
+L:	dri-devel@lists.freedesktop.org
+S:	Supported
+F:	drivers/gpu/drm/exynos
+F:	include/drm/exynos*
+
 DSCC4 DRIVER
 M:	Francois Romieu <romieu@fr.zoreil.com>
 L:	netdev@vger.kernel.org
@@ -2576,7 +2585,7 @@
 F:	drivers/net/ethernet/i825xx/eexpress.*
 
 ETHERNET BRIDGE
-M:	Stephen Hemminger <shemminger@linux-foundation.org>
+M:	Stephen Hemminger <shemminger@vyatta.com>
 L:	bridge@lists.linux-foundation.org
 L:	netdev@vger.kernel.org
 W:	http://www.linuxfoundation.org/en/Net:Bridge
@@ -3092,6 +3101,7 @@
 
 HIGH-RESOLUTION TIMERS, CLOCKEVENTS, DYNTICKS
 M:	Thomas Gleixner <tglx@linutronix.de>
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers/core
 S:	Maintained
 F:	Documentation/timers/
 F:	kernel/hrtimer.c
@@ -3601,7 +3611,7 @@
 IRQ SUBSYSTEM
 M:	Thomas Gleixner <tglx@linutronix.de>
 S:	Maintained
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git irq/core
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core
 F:	kernel/irq/
 
 ISAPNP
@@ -3710,7 +3720,7 @@
 F:	include/linux/jbd2.h
 
 JSM Neo PCI based serial card
-M:	Breno Leitao <leitao@linux.vnet.ibm.com>
+M:	Lucas Tavares <lucaskt@linux.vnet.ibm.com>
 L:	linux-serial@vger.kernel.org
 S:	Maintained
 F:	drivers/tty/serial/jsm/
@@ -4089,7 +4099,7 @@
 LOCKDEP AND LOCKSTAT
 M:	Peter Zijlstra <peterz@infradead.org>
 M:	Ingo Molnar <mingo@redhat.com>
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/peterz/linux-2.6-lockdep.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git core/locking
 S:	Maintained
 F:	Documentation/lockdep*.txt
 F:	Documentation/lockstat.txt
@@ -4293,9 +4303,11 @@
 F:	mm/
 
 MEMORY RESOURCE CONTROLLER
+M:	Johannes Weiner <hannes@cmpxchg.org>
+M:	Michal Hocko <mhocko@suse.cz>
 M:	Balbir Singh <bsingharora@gmail.com>
-M:	Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
 M:	KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
+L:	cgroups@vger.kernel.org
 L:	linux-mm@kvack.org
 S:	Maintained
 F:	mm/memcontrol.c
@@ -4329,7 +4341,7 @@
 M:	Ralf Baechle <ralf@linux-mips.org>
 L:	linux-mips@linux-mips.org
 W:	http://www.linux-mips.org/
-T:	git git://git.linux-mips.org/pub/scm/linux.git
+T:	git git://git.linux-mips.org/pub/scm/ralf/linux.git
 Q:	http://patchwork.linux-mips.org/project/linux-mips/list/
 S:	Supported
 F:	Documentation/mips/
@@ -4462,7 +4474,7 @@
 F:	drivers/infiniband/hw/nes/
 
 NETEM NETWORK EMULATOR
-M:	Stephen Hemminger <shemminger@linux-foundation.org>
+M:	Stephen Hemminger <shemminger@vyatta.com>
 L:	netem@lists.linux-foundation.org
 S:	Maintained
 F:	net/sched/sch_netem.c
@@ -4939,7 +4951,7 @@
 F:	include/linux/ppdev.h
 
 PARAVIRT_OPS INTERFACE
-M:	Jeremy Fitzhardinge <jeremy@xensource.com>
+M:	Jeremy Fitzhardinge <jeremy@goop.org>
 M:	Chris Wright <chrisw@sous-sol.org>
 M:	Alok Kataria <akataria@vmware.com>
 M:	Rusty Russell <rusty@rustcorp.com.au>
@@ -5075,6 +5087,7 @@
 M:	Paul Mackerras <paulus@samba.org>
 M:	Ingo Molnar <mingo@elte.hu>
 M:	Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git perf/core
 S:	Supported
 F:	kernel/events/*
 F:	include/linux/perf_event.h
@@ -5154,6 +5167,7 @@
 
 POSIX CLOCKS and TIMERS
 M:	Thomas Gleixner <tglx@linutronix.de>
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers/core
 S:	Supported
 F:	fs/timerfd.c
 F:	include/linux/timer*
@@ -5648,7 +5662,6 @@
 F:	include/media/*7146*
 
 SAMSUNG AUDIO (ASoC) DRIVERS
-M:	Jassi Brar <jassisinghbrar@gmail.com>
 M:	Sangbeom Kim <sbkim73@samsung.com>
 L:	alsa-devel@alsa-project.org (moderated for non-subscribers)
 S:	Supported
@@ -5670,6 +5683,7 @@
 TIMEKEEPING, NTP
 M:	John Stultz <johnstul@us.ibm.com>
 M:	Thomas Gleixner <tglx@linutronix.de>
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers/core
 S:	Supported
 F:	include/linux/clocksource.h
 F:	include/linux/time.h
@@ -5694,6 +5708,7 @@
 SCHEDULER
 M:	Ingo Molnar <mingo@elte.hu>
 M:	Peter Zijlstra <peterz@infradead.org>
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched/core
 S:	Maintained
 F:	kernel/sched*
 F:	include/linux/sched.h
@@ -5977,7 +5992,7 @@
 F:	drivers/usb/misc/sisusbvga/
 
 SKGE, SKY2 10/100/1000 GIGABIT ETHERNET DRIVERS
-M:	Stephen Hemminger <shemminger@linux-foundation.org>
+M:	Stephen Hemminger <shemminger@vyatta.com>
 L:	netdev@vger.kernel.org
 S:	Maintained
 F:	drivers/net/ethernet/marvell/sk*
@@ -6122,7 +6137,7 @@
 SOUND - SOC LAYER / DYNAMIC AUDIO POWER MANAGEMENT (ASoC)
 M:	Liam Girdwood <lrg@ti.com>
 M:	Mark Brown <broonie@opensource.wolfsonmicro.com>
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound-2.6.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git
 L:	alsa-devel@alsa-project.org (moderated for non-subscribers)
 W:	http://alsa-project.org/main/index.php/ASoC
 S:	Supported
@@ -6621,7 +6636,7 @@
 M:	Steven Rostedt <rostedt@goodmis.org>
 M:	Frederic Weisbecker <fweisbec@gmail.com>
 M:	Ingo Molnar <mingo@redhat.com>
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git perf/core
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git perf/core
 S:	Maintained
 F:	Documentation/trace/ftrace.txt
 F:	arch/*/*/*/ftrace.h
@@ -7371,7 +7386,7 @@
 M:	Ingo Molnar <mingo@redhat.com>
 M:	"H. Peter Anvin" <hpa@zytor.com>
 M:	x86@kernel.org
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/core
 S:	Maintained
 F:	Documentation/x86/
 F:	arch/x86/
@@ -7391,8 +7406,8 @@
 F:	arch/x86/kernel/cpu/mcheck/*
 
 XEN HYPERVISOR INTERFACE
-M:	Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
 M:	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
+M:	Jeremy Fitzhardinge <jeremy@goop.org>
 L:	xen-devel@lists.xensource.com (moderated for non-subscribers)
 L:	virtualization@lists.linux-foundation.org
 S:	Supported
@@ -7425,7 +7440,8 @@
 
 XFS FILESYSTEM
 P:	Silicon Graphics Inc
-M:	Alex Elder <aelder@sgi.com>
+M:	Ben Myers <bpm@sgi.com>
+M:	Alex Elder <elder@kernel.org>
 M:	xfs-masters@oss.sgi.com
 L:	xfs@oss.sgi.com
 W:	http://oss.sgi.com/projects/xfs
diff --git a/Makefile b/Makefile
index 361e4f0..a43733d 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 3
 PATCHLEVEL = 2
 SUBLEVEL = 0
-EXTRAVERSION = -rc1
+EXTRAVERSION = -rc6
 NAME = Saber-toothed Squirrel
 
 # *DOCUMENTATION*
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 44789ef..776d76b 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -220,8 +220,9 @@
 	  be avoided when possible.
 
 config PHYS_OFFSET
-	hex "Physical address of main memory"
+	hex "Physical address of main memory" if MMU
 	depends on !ARM_PATCH_PHYS_VIRT && !NEED_MACH_MEMORY_H
+	default DRAM_BASE if !MMU
 	help
 	  Please provide the physical address corresponding to the
 	  location of main memory in your system.
@@ -1231,7 +1232,7 @@
 	  capabilities of the processor.
 
 config PL310_ERRATA_588369
-	bool "Clean & Invalidate maintenance operations do not invalidate clean lines"
+	bool "PL310 errata: Clean & Invalidate maintenance operations do not invalidate clean lines"
 	depends on CACHE_L2X0
 	help
 	   The PL310 L2 cache controller implements three types of Clean &
@@ -1256,7 +1257,7 @@
 	  entries regardless of the ASID.
 
 config PL310_ERRATA_727915
-	bool "Background Clean & Invalidate by Way operation can cause data corruption"
+	bool "PL310 errata: Background Clean & Invalidate by Way operation can cause data corruption"
 	depends on CACHE_L2X0
 	help
 	  PL310 implements the Clean & Invalidate by Way L2 cache maintenance
@@ -1289,8 +1290,8 @@
 	  operation is received by a CPU before the ICIALLUIS has completed,
 	  potentially leading to corrupted entries in the cache or TLB.
 
-config ARM_ERRATA_753970
-	bool "ARM errata: cache sync operation may be faulty"
+config PL310_ERRATA_753970
+	bool "PL310 errata: cache sync operation may be faulty"
 	depends on CACHE_PL310
 	help
 	  This option enables the workaround for the 753970 PL310 (r3p0) erratum.
@@ -1352,6 +1353,18 @@
 	  relevant cache maintenance functions and sets a specific bit
 	  in the diagnostic control register of the SCU.
 
+config PL310_ERRATA_769419
+	bool "PL310 errata: no automatic Store Buffer drain"
+	depends on CACHE_L2X0
+	help
+	  On revisions of the PL310 prior to r3p2, the Store Buffer does
+	  not automatically drain. This can cause normal, non-cacheable
+	  writes to be retained when the memory system is idle, leading
+	  to suboptimal I/O performance for drivers using coherent DMA.
+	  This option adds a write barrier to the cpu_idle loop so that,
+	  on systems with an outer cache, the store buffer is drained
+	  explicitly.
+
 endmenu
 
 source "arch/arm/common/Kconfig"
diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile
index 176062a..5df26a9 100644
--- a/arch/arm/boot/Makefile
+++ b/arch/arm/boot/Makefile
@@ -65,6 +65,8 @@
 
 $(obj)/dtbs: $(addprefix $(obj)/, $(dtb-y))
 
+clean-files := *.dtb
+
 quiet_cmd_uimage = UIMAGE  $@
       cmd_uimage = $(CONFIG_SHELL) $(MKIMAGE) -A arm -O linux -T kernel \
 		   -C none -a $(LOADADDR) -e $(STARTADDR) \
diff --git a/arch/arm/boot/dts/tegra-ventana.dts b/arch/arm/boot/dts/tegra-ventana.dts
index 9b29a62..3f9abd6b 100644
--- a/arch/arm/boot/dts/tegra-ventana.dts
+++ b/arch/arm/boot/dts/tegra-ventana.dts
@@ -22,11 +22,10 @@
 	sdhci@c8000400 {
 		cd-gpios = <&gpio 69 0>; /* gpio PI5 */
 		wp-gpios = <&gpio 57 0>; /* gpio PH1 */
-		power-gpios = <&gpio 155 0>; /* gpio PT3 */
+		power-gpios = <&gpio 70 0>; /* gpio PI6 */
 	};
 
 	sdhci@c8000600 {
-		power-gpios = <&gpio 70 0>; /* gpio PI6 */
 		support-8bit;
 	};
 };
diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index 0e6ae47..410a546 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -526,7 +526,8 @@
 		sizeof(u32));
 	BUG_ON(!gic->saved_ppi_conf);
 
-	cpu_pm_register_notifier(&gic_notifier_block);
+	if (gic == &gic_data[0])
+		cpu_pm_register_notifier(&gic_notifier_block);
 }
 #else
 static void __init gic_pm_init(struct gic_chip_data *gic)
@@ -581,13 +582,16 @@
 	 * For primary GICs, skip over SGIs.
 	 * For secondary GICs, skip over PPIs, too.
 	 */
+	domain->hwirq_base = 32;
 	if (gic_nr == 0) {
 		gic_cpu_base_addr = cpu_base;
-		domain->hwirq_base = 16;
-		if (irq_start > 0)
-			irq_start = (irq_start & ~31) + 16;
-	} else
-		domain->hwirq_base = 32;
+
+		if ((irq_start & 31) > 0) {
+			domain->hwirq_base = 16;
+			if (irq_start != -1)
+				irq_start = (irq_start & ~31) + 16;
+		}
+	}
 
 	/*
 	 * Find out how many interrupts are supported.
diff --git a/arch/arm/common/pl330.c b/arch/arm/common/pl330.c
index 7129cfb..f407a6b 100644
--- a/arch/arm/common/pl330.c
+++ b/arch/arm/common/pl330.c
@@ -1211,8 +1211,8 @@
 	ccr |= (rqc->brst_size << CC_SRCBRSTSIZE_SHFT);
 	ccr |= (rqc->brst_size << CC_DSTBRSTSIZE_SHFT);
 
-	ccr |= (rqc->dcctl << CC_SRCCCTRL_SHFT);
-	ccr |= (rqc->scctl << CC_DSTCCTRL_SHFT);
+	ccr |= (rqc->scctl << CC_SRCCCTRL_SHFT);
+	ccr |= (rqc->dcctl << CC_DSTCCTRL_SHFT);
 
 	ccr |= (rqc->swap << CC_SWAP_SHFT);
 
@@ -1623,6 +1623,11 @@
 	return -1;
 }
 
+static bool _chan_ns(const struct pl330_info *pi, int i)
+{
+	return pi->pcfg.irq_ns & (1 << i);
+}
+
 /* Upon success, returns IdentityToken for the
  * allocated channel, NULL otherwise.
  */
@@ -1647,7 +1652,8 @@
 
 	for (i = 0; i < chans; i++) {
 		thrd = &pl330->channels[i];
-		if (thrd->free) {
+		if ((thrd->free) && (!_manager_ns(thrd) ||
+					_chan_ns(pi, i))) {
 			thrd->ev = _alloc_event(thrd);
 			if (thrd->ev >= 0) {
 				thrd->free = false;
diff --git a/arch/arm/configs/at91cap9adk_defconfig b/arch/arm/configs/at91cap9_defconfig
similarity index 93%
rename from arch/arm/configs/at91cap9adk_defconfig
rename to arch/arm/configs/at91cap9_defconfig
index ffb1edd..8826eb2 100644
--- a/arch/arm/configs/at91cap9adk_defconfig
+++ b/arch/arm/configs/at91cap9_defconfig
@@ -38,7 +38,6 @@
 # CONFIG_IPV6 is not set
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_MTD=y
-CONFIG_MTD_PARTITIONS=y
 CONFIG_MTD_CMDLINE_PARTS=y
 CONFIG_MTD_CHAR=y
 CONFIG_MTD_BLOCK=y
@@ -52,16 +51,12 @@
 CONFIG_BLK_DEV_LOOP=y
 CONFIG_BLK_DEV_RAM=y
 CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_ATMEL_SSC=y
 CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 CONFIG_SCSI_MULTI_LUN=y
 CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
 CONFIG_MII=y
 CONFIG_MACB=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
 # CONFIG_INPUT_MOUSEDEV_PSAUX is not set
 CONFIG_INPUT_EVDEV=y
 # CONFIG_INPUT_KEYBOARD is not set
@@ -81,7 +76,6 @@
 CONFIG_WATCHDOG_NOWAYOUT=y
 CONFIG_FB=y
 CONFIG_FB_ATMEL=y
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_LOGO=y
 # CONFIG_LOGO_LINUX_MONO is not set
 # CONFIG_LOGO_LINUX_CLUT224 is not set
@@ -99,7 +93,6 @@
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_DRV_AT91SAM9=y
 CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
 CONFIG_VFAT_FS=y
 CONFIG_TMPFS=y
 CONFIG_JFFS2_FS=y
diff --git a/arch/arm/configs/at91rm9200_defconfig b/arch/arm/configs/at91rm9200_defconfig
index 38cb7c9..bbe4e1a 100644
--- a/arch/arm/configs/at91rm9200_defconfig
+++ b/arch/arm/configs/at91rm9200_defconfig
@@ -5,7 +5,6 @@
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
 CONFIG_BLK_DEV_INITRD=y
 CONFIG_MODULES=y
 CONFIG_MODULE_FORCE_LOAD=y
@@ -56,7 +55,6 @@
 CONFIG_IP_PNP_DHCP=y
 CONFIG_IP_PNP_BOOTP=y
 CONFIG_NET_IPIP=m
-CONFIG_NET_IPGRE=m
 CONFIG_INET_AH=m
 CONFIG_INET_ESP=m
 CONFIG_INET_IPCOMP=m
@@ -75,18 +73,8 @@
 CONFIG_BRIDGE=m
 CONFIG_VLAN_8021Q=m
 CONFIG_BT=m
-CONFIG_BT_L2CAP=m
-CONFIG_BT_SCO=m
-CONFIG_BT_RFCOMM=m
-CONFIG_BT_RFCOMM_TTY=y
-CONFIG_BT_BNEP=m
-CONFIG_BT_BNEP_MC_FILTER=y
-CONFIG_BT_BNEP_PROTO_FILTER=y
-CONFIG_BT_HIDP=m
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_MTD=y
-CONFIG_MTD_CONCAT=y
-CONFIG_MTD_PARTITIONS=y
 CONFIG_MTD_CMDLINE_PARTS=y
 CONFIG_MTD_AFS_PARTS=y
 CONFIG_MTD_CHAR=y
@@ -108,8 +96,6 @@
 CONFIG_BLK_DEV_NBD=y
 CONFIG_BLK_DEV_RAM=y
 CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_ATMEL_TCLIB=y
-CONFIG_EEPROM_LEGACY=m
 CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 CONFIG_BLK_DEV_SR=m
@@ -119,14 +105,23 @@
 # CONFIG_SCSI_LOWLEVEL is not set
 CONFIG_NETDEVICES=y
 CONFIG_TUN=m
+CONFIG_ARM_AT91_ETHER=y
 CONFIG_PHYLIB=y
 CONFIG_DAVICOM_PHY=y
 CONFIG_SMSC_PHY=y
 CONFIG_MICREL_PHY=y
-CONFIG_NET_ETHERNET=y
-CONFIG_ARM_AT91_ETHER=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
+CONFIG_PPP=y
+CONFIG_PPP_BSDCOMP=y
+CONFIG_PPP_DEFLATE=y
+CONFIG_PPP_FILTER=y
+CONFIG_PPP_MPPE=m
+CONFIG_PPP_MULTILINK=y
+CONFIG_PPPOE=m
+CONFIG_PPP_ASYNC=y
+CONFIG_SLIP=m
+CONFIG_SLIP_COMPRESSED=y
+CONFIG_SLIP_SMART=y
+CONFIG_SLIP_MODE_SLIP6=y
 CONFIG_USB_CATC=m
 CONFIG_USB_KAWETH=m
 CONFIG_USB_PEGASUS=m
@@ -139,18 +134,6 @@
 CONFIG_USB_ALI_M5632=y
 CONFIG_USB_AN2720=y
 CONFIG_USB_EPSON2888=y
-CONFIG_PPP=y
-CONFIG_PPP_MULTILINK=y
-CONFIG_PPP_FILTER=y
-CONFIG_PPP_ASYNC=y
-CONFIG_PPP_DEFLATE=y
-CONFIG_PPP_BSDCOMP=y
-CONFIG_PPP_MPPE=m
-CONFIG_PPPOE=m
-CONFIG_SLIP=m
-CONFIG_SLIP_COMPRESSED=y
-CONFIG_SLIP_SMART=y
-CONFIG_SLIP_MODE_SLIP6=y
 # CONFIG_INPUT_MOUSEDEV_PSAUX is not set
 CONFIG_INPUT_MOUSEDEV_SCREEN_X=640
 CONFIG_INPUT_MOUSEDEV_SCREEN_Y=480
@@ -158,9 +141,9 @@
 CONFIG_KEYBOARD_GPIO=y
 # CONFIG_INPUT_MOUSE is not set
 CONFIG_INPUT_TOUCHSCREEN=y
+CONFIG_LEGACY_PTY_COUNT=32
 CONFIG_SERIAL_ATMEL=y
 CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_LEGACY_PTY_COUNT=32
 CONFIG_HW_RANDOM=y
 CONFIG_I2C=y
 CONFIG_I2C_CHARDEV=y
@@ -290,7 +273,6 @@
 CONFIG_NFS_V4=y
 CONFIG_ROOT_NFS=y
 CONFIG_NFSD=y
-CONFIG_SMB_FS=m
 CONFIG_CIFS=m
 CONFIG_PARTITION_ADVANCED=y
 CONFIG_MAC_PARTITION=y
@@ -335,7 +317,6 @@
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_DEBUG_FS=y
 CONFIG_DEBUG_KERNEL=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
 # CONFIG_FTRACE is not set
 CONFIG_CRYPTO_PCBC=y
 CONFIG_CRYPTO_SHA1=y
diff --git a/arch/arm/configs/at91sam9260ek_defconfig b/arch/arm/configs/at91sam9260_defconfig
similarity index 86%
rename from arch/arm/configs/at91sam9260ek_defconfig
rename to arch/arm/configs/at91sam9260_defconfig
index f8a9226..505b376 100644
--- a/arch/arm/configs/at91sam9260ek_defconfig
+++ b/arch/arm/configs/at91sam9260_defconfig
@@ -12,11 +12,23 @@
 # CONFIG_IOSCHED_CFQ is not set
 CONFIG_ARCH_AT91=y
 CONFIG_ARCH_AT91SAM9260=y
+CONFIG_ARCH_AT91SAM9260_SAM9XE=y
 CONFIG_MACH_AT91SAM9260EK=y
+CONFIG_MACH_CAM60=y
+CONFIG_MACH_SAM9_L9260=y
+CONFIG_MACH_AFEB9260=y
+CONFIG_MACH_USB_A9260=y
+CONFIG_MACH_QIL_A9260=y
+CONFIG_MACH_CPU9260=y
+CONFIG_MACH_FLEXIBITY=y
+CONFIG_MACH_SNAPPER_9260=y
+CONFIG_MACH_AT91SAM_DT=y
 CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
 # CONFIG_ARM_THUMB is not set
 CONFIG_ZBOOT_ROM_TEXT=0x0
 CONFIG_ZBOOT_ROM_BSS=0x0
+CONFIG_ARM_APPENDED_DTB=y
+CONFIG_ARM_ATAG_DTB_COMPAT=y
 CONFIG_CMDLINE="mem=64M console=ttyS0,115200 initrd=0x21100000,3145728 root=/dev/ram0 rw"
 CONFIG_FPE_NWFPE=y
 CONFIG_NET=y
@@ -33,12 +45,10 @@
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_BLK_DEV_RAM=y
 CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_ATMEL_SSC=y
 CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 CONFIG_SCSI_MULTI_LUN=y
 CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
 CONFIG_MII=y
 CONFIG_MACB=y
 # CONFIG_INPUT_MOUSEDEV_PSAUX is not set
@@ -55,7 +65,6 @@
 CONFIG_WATCHDOG=y
 CONFIG_WATCHDOG_NOWAYOUT=y
 CONFIG_AT91SAM9X_WATCHDOG=y
-# CONFIG_VGA_CONSOLE is not set
 # CONFIG_USB_HID is not set
 CONFIG_USB=y
 CONFIG_USB_DEVICEFS=y
@@ -71,7 +80,6 @@
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_DRV_AT91SAM9=y
 CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
 CONFIG_VFAT_FS=y
 CONFIG_TMPFS=y
 CONFIG_CRAMFS=y
diff --git a/arch/arm/configs/at91sam9g20ek_defconfig b/arch/arm/configs/at91sam9g20_defconfig
similarity index 90%
rename from arch/arm/configs/at91sam9g20ek_defconfig
rename to arch/arm/configs/at91sam9g20_defconfig
index 9e90e6d..9123568 100644
--- a/arch/arm/configs/at91sam9g20ek_defconfig
+++ b/arch/arm/configs/at91sam9g20_defconfig
@@ -14,6 +14,15 @@
 CONFIG_ARCH_AT91SAM9G20=y
 CONFIG_MACH_AT91SAM9G20EK=y
 CONFIG_MACH_AT91SAM9G20EK_2MMC=y
+CONFIG_MACH_CPU9G20=y
+CONFIG_MACH_ACMENETUSFOXG20=y
+CONFIG_MACH_PORTUXG20=y
+CONFIG_MACH_STAMP9G20=y
+CONFIG_MACH_PCONTROL_G20=y
+CONFIG_MACH_GSIA18S=y
+CONFIG_MACH_USB_A9G20=y
+CONFIG_MACH_SNAPPER_9260=y
+CONFIG_MACH_AT91SAM_DT=y
 CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
 # CONFIG_ARM_THUMB is not set
 CONFIG_AEABI=y
@@ -21,9 +30,10 @@
 CONFIG_LEDS_CPU=y
 CONFIG_ZBOOT_ROM_TEXT=0x0
 CONFIG_ZBOOT_ROM_BSS=0x0
+CONFIG_ARM_APPENDED_DTB=y
+CONFIG_ARM_ATAG_DTB_COMPAT=y
 CONFIG_CMDLINE="mem=64M console=ttyS0,115200 initrd=0x21100000,3145728 root=/dev/ram0 rw"
 CONFIG_FPE_NWFPE=y
-CONFIG_PM=y
 CONFIG_NET=y
 CONFIG_PACKET=y
 CONFIG_UNIX=y
@@ -37,8 +47,6 @@
 # CONFIG_IPV6 is not set
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_MTD=y
-CONFIG_MTD_CONCAT=y
-CONFIG_MTD_PARTITIONS=y
 CONFIG_MTD_CMDLINE_PARTS=y
 CONFIG_MTD_CHAR=y
 CONFIG_MTD_BLOCK=y
@@ -48,17 +56,13 @@
 CONFIG_BLK_DEV_LOOP=y
 CONFIG_BLK_DEV_RAM=y
 CONFIG_BLK_DEV_RAM_SIZE=8192
-CONFIG_ATMEL_SSC=y
 CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 CONFIG_SCSI_MULTI_LUN=y
 # CONFIG_SCSI_LOWLEVEL is not set
 CONFIG_NETDEVICES=y
-CONFIG_NET_ETHERNET=y
 CONFIG_MII=y
 CONFIG_MACB=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
 # CONFIG_INPUT_MOUSEDEV_PSAUX is not set
 CONFIG_INPUT_MOUSEDEV_SCREEN_X=320
 CONFIG_INPUT_MOUSEDEV_SCREEN_Y=240
@@ -66,15 +70,14 @@
 # CONFIG_KEYBOARD_ATKBD is not set
 CONFIG_KEYBOARD_GPIO=y
 # CONFIG_INPUT_MOUSE is not set
+CONFIG_LEGACY_PTY_COUNT=16
 CONFIG_SERIAL_ATMEL=y
 CONFIG_SERIAL_ATMEL_CONSOLE=y
-CONFIG_LEGACY_PTY_COUNT=16
 CONFIG_HW_RANDOM=y
 CONFIG_SPI=y
 CONFIG_SPI_ATMEL=y
 CONFIG_SPI_SPIDEV=y
 # CONFIG_HWMON is not set
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_SOUND=y
 CONFIG_SND=y
 CONFIG_SND_SEQUENCER=y
@@ -82,7 +85,6 @@
 CONFIG_SND_PCM_OSS=y
 CONFIG_SND_SEQUENCER_OSS=y
 # CONFIG_SND_VERBOSE_PROCFS is not set
-CONFIG_SND_AT73C213=y
 CONFIG_USB=y
 CONFIG_USB_DEVICEFS=y
 # CONFIG_USB_DEVICE_CLASS is not set
@@ -105,7 +107,6 @@
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_DRV_AT91SAM9=y
 CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
 CONFIG_MSDOS_FS=y
 CONFIG_VFAT_FS=y
 CONFIG_TMPFS=y
diff --git a/arch/arm/configs/at91sam9g45_defconfig b/arch/arm/configs/at91sam9g45_defconfig
index c5876d2..606d48f 100644
--- a/arch/arm/configs/at91sam9g45_defconfig
+++ b/arch/arm/configs/at91sam9g45_defconfig
@@ -18,6 +18,7 @@
 CONFIG_ARCH_AT91=y
 CONFIG_ARCH_AT91SAM9G45=y
 CONFIG_MACH_AT91SAM9M10G45EK=y
+CONFIG_MACH_AT91SAM_DT=y
 CONFIG_AT91_PROGRAMMABLE_CLOCKS=y
 CONFIG_AT91_SLOW_CLOCK=y
 CONFIG_AEABI=y
@@ -73,11 +74,8 @@
 # CONFIG_SCSI_LOWLEVEL is not set
 CONFIG_NETDEVICES=y
 CONFIG_MII=y
-CONFIG_DAVICOM_PHY=y
-CONFIG_NET_ETHERNET=y
 CONFIG_MACB=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
+CONFIG_DAVICOM_PHY=y
 CONFIG_LIBERTAS_THINFIRM=m
 CONFIG_LIBERTAS_THINFIRM_USB=m
 CONFIG_AT76C50X_USB=m
@@ -131,7 +129,6 @@
 CONFIG_SPI=y
 CONFIG_SPI_ATMEL=y
 # CONFIG_HWMON is not set
-# CONFIG_MFD_SUPPORT is not set
 CONFIG_FB=y
 CONFIG_FB_ATMEL=y
 CONFIG_FB_UDL=m
diff --git a/arch/arm/configs/at91sam9rlek_defconfig b/arch/arm/configs/at91sam9rl_defconfig
similarity index 94%
rename from arch/arm/configs/at91sam9rlek_defconfig
rename to arch/arm/configs/at91sam9rl_defconfig
index 75621e4..ad562ee 100644
--- a/arch/arm/configs/at91sam9rlek_defconfig
+++ b/arch/arm/configs/at91sam9rl_defconfig
@@ -23,8 +23,6 @@
 CONFIG_UNIX=y
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_MTD=y
-CONFIG_MTD_CONCAT=y
-CONFIG_MTD_PARTITIONS=y
 CONFIG_MTD_CMDLINE_PARTS=y
 CONFIG_MTD_CHAR=y
 CONFIG_MTD_BLOCK=y
@@ -35,7 +33,6 @@
 CONFIG_BLK_DEV_RAM=y
 CONFIG_BLK_DEV_RAM_COUNT=4
 CONFIG_BLK_DEV_RAM_SIZE=24576
-CONFIG_ATMEL_SSC=y
 CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 CONFIG_SCSI_MULTI_LUN=y
@@ -62,13 +59,11 @@
 CONFIG_AT91SAM9X_WATCHDOG=y
 CONFIG_FB=y
 CONFIG_FB_ATMEL=y
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_MMC=y
 CONFIG_MMC_AT91=m
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_DRV_AT91SAM9=y
 CONFIG_EXT2_FS=y
-CONFIG_INOTIFY=y
 CONFIG_MSDOS_FS=y
 CONFIG_VFAT_FS=y
 CONFIG_TMPFS=y
diff --git a/arch/arm/configs/ezx_defconfig b/arch/arm/configs/ezx_defconfig
index 227a477..d95763d 100644
--- a/arch/arm/configs/ezx_defconfig
+++ b/arch/arm/configs/ezx_defconfig
@@ -287,7 +287,7 @@
 # CONFIG_USB_DEVICE_CLASS is not set
 CONFIG_USB_OHCI_HCD=y
 CONFIG_USB_GADGET=y
-CONFIG_USB_GADGET_PXA27X=y
+CONFIG_USB_PXA27X=y
 CONFIG_USB_ETH=m
 # CONFIG_USB_ETH_RNDIS is not set
 CONFIG_MMC=y
diff --git a/arch/arm/configs/imote2_defconfig b/arch/arm/configs/imote2_defconfig
index 176ec22..fd996bb 100644
--- a/arch/arm/configs/imote2_defconfig
+++ b/arch/arm/configs/imote2_defconfig
@@ -263,7 +263,7 @@
 # CONFIG_USB_DEVICE_CLASS is not set
 CONFIG_USB_OHCI_HCD=y
 CONFIG_USB_GADGET=y
-CONFIG_USB_GADGET_PXA27X=y
+CONFIG_USB_PXA27X=y
 CONFIG_USB_ETH=m
 # CONFIG_USB_ETH_RNDIS is not set
 CONFIG_MMC=y
diff --git a/arch/arm/configs/magician_defconfig b/arch/arm/configs/magician_defconfig
index a88e64d..443675d 100644
--- a/arch/arm/configs/magician_defconfig
+++ b/arch/arm/configs/magician_defconfig
@@ -132,7 +132,7 @@
 CONFIG_USB_OHCI_HCD=y
 CONFIG_USB_GADGET=y
 CONFIG_USB_GADGET_VBUS_DRAW=500
-CONFIG_USB_GADGET_PXA27X=y
+CONFIG_USB_PXA27X=y
 CONFIG_USB_ETH=m
 # CONFIG_USB_ETH_RNDIS is not set
 CONFIG_USB_GADGETFS=m
diff --git a/arch/arm/configs/omap1_defconfig b/arch/arm/configs/omap1_defconfig
index 7b63462..945a34f 100644
--- a/arch/arm/configs/omap1_defconfig
+++ b/arch/arm/configs/omap1_defconfig
@@ -48,13 +48,7 @@
 CONFIG_MACH_NOKIA770=y
 CONFIG_MACH_AMS_DELTA=y
 CONFIG_MACH_OMAP_GENERIC=y
-CONFIG_OMAP_CLOCKS_SET_BY_BOOTLOADER=y
-CONFIG_OMAP_ARM_216MHZ=y
-CONFIG_OMAP_ARM_195MHZ=y
-CONFIG_OMAP_ARM_192MHZ=y
 CONFIG_OMAP_ARM_182MHZ=y
-CONFIG_OMAP_ARM_168MHZ=y
-# CONFIG_OMAP_ARM_60MHZ is not set
 # CONFIG_ARM_THUMB is not set
 CONFIG_PCCARD=y
 CONFIG_OMAP_CF=y
diff --git a/arch/arm/configs/u300_defconfig b/arch/arm/configs/u300_defconfig
index 4a5a126..374000ec 100644
--- a/arch/arm/configs/u300_defconfig
+++ b/arch/arm/configs/u300_defconfig
@@ -14,8 +14,6 @@
 CONFIG_ARCH_U300=y
 CONFIG_MACH_U300=y
 CONFIG_MACH_U300_BS335=y
-CONFIG_MACH_U300_DUAL_RAM=y
-CONFIG_U300_DEBUG=y
 CONFIG_MACH_U300_SPIDUMMY=y
 CONFIG_NO_HZ=y
 CONFIG_HIGH_RES_TIMERS=y
@@ -26,19 +24,21 @@
 CONFIG_CMDLINE="root=/dev/ram0 rw rootfstype=rootfs console=ttyAMA0,115200n8 lpj=515072"
 CONFIG_CPU_IDLE=y
 CONFIG_FPE_NWFPE=y
-CONFIG_PM=y
 # CONFIG_SUSPEND is not set
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
-# CONFIG_MISC_DEVICES is not set
+CONFIG_MTD=y
+CONFIG_MTD_CMDLINE_PARTS=y
+CONFIG_MTD_NAND=y
+CONFIG_MTD_NAND_FSMC=y
 # CONFIG_INPUT_MOUSEDEV is not set
 CONFIG_INPUT_EVDEV=y
 # CONFIG_KEYBOARD_ATKBD is not set
 # CONFIG_INPUT_MOUSE is not set
 # CONFIG_SERIO is not set
+CONFIG_LEGACY_PTY_COUNT=16
 CONFIG_SERIAL_AMBA_PL011=y
 CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
-CONFIG_LEGACY_PTY_COUNT=16
 # CONFIG_HW_RANDOM is not set
 CONFIG_I2C=y
 # CONFIG_HWMON is not set
@@ -51,6 +51,7 @@
 # CONFIG_HID_SUPPORT is not set
 # CONFIG_USB_SUPPORT is not set
 CONFIG_MMC=y
+CONFIG_MMC_CLKGATE=y
 CONFIG_MMC_ARMMMCI=y
 CONFIG_RTC_CLASS=y
 # CONFIG_RTC_HCTOSYS is not set
@@ -65,10 +66,8 @@
 CONFIG_NLS_ISO8859_1=y
 CONFIG_PRINTK_TIME=y
 CONFIG_DEBUG_FS=y
-CONFIG_DEBUG_KERNEL=y
 # CONFIG_SCHED_DEBUG is not set
 CONFIG_TIMER_STATS=y
 # CONFIG_DEBUG_PREEMPT is not set
 CONFIG_DEBUG_INFO=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
 # CONFIG_CRC32 is not set
diff --git a/arch/arm/configs/u8500_defconfig b/arch/arm/configs/u8500_defconfig
index 97d31a4..2d7b6e7 100644
--- a/arch/arm/configs/u8500_defconfig
+++ b/arch/arm/configs/u8500_defconfig
@@ -10,7 +10,7 @@
 CONFIG_ARCH_U8500=y
 CONFIG_UX500_SOC_DB5500=y
 CONFIG_UX500_SOC_DB8500=y
-CONFIG_MACH_U8500=y
+CONFIG_MACH_HREFV60=y
 CONFIG_MACH_SNOWBALL=y
 CONFIG_MACH_U5500=y
 CONFIG_NO_HZ=y
@@ -24,6 +24,7 @@
 CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
 CONFIG_VFP=y
 CONFIG_NEON=y
+CONFIG_PM_RUNTIME=y
 CONFIG_NET=y
 CONFIG_PACKET=y
 CONFIG_UNIX=y
@@ -41,11 +42,8 @@
 CONFIG_AB8500_PWM=y
 CONFIG_SENSORS_BH1780=y
 CONFIG_NETDEVICES=y
-CONFIG_SMSC_PHY=y
-CONFIG_NET_ETHERNET=y
 CONFIG_SMSC911X=y
-# CONFIG_NETDEV_1000 is not set
-# CONFIG_NETDEV_10000 is not set
+CONFIG_SMSC_PHY=y
 # CONFIG_WLAN is not set
 # CONFIG_INPUT_MOUSEDEV_PSAUX is not set
 CONFIG_INPUT_EVDEV=y
@@ -72,15 +70,12 @@
 CONFIG_SPI_PL022=y
 CONFIG_GPIO_STMPE=y
 CONFIG_GPIO_TC3589X=y
-# CONFIG_HWMON is not set
 CONFIG_MFD_STMPE=y
 CONFIG_MFD_TC3589X=y
+CONFIG_AB5500_CORE=y
 CONFIG_AB8500_CORE=y
 CONFIG_REGULATOR_AB8500=y
 # CONFIG_HID_SUPPORT is not set
-CONFIG_USB_MUSB_HDRC=y
-CONFIG_USB_GADGET_MUSB_HDRC=y
-CONFIG_MUSB_PIO_ONLY=y
 CONFIG_USB_GADGET=y
 CONFIG_AB8500_USB=y
 CONFIG_MMC=y
@@ -97,6 +92,7 @@
 CONFIG_STE_DMA40=y
 CONFIG_STAGING=y
 CONFIG_TOUCHSCREEN_SYNAPTICS_I2C_RMI4=y
+CONFIG_HSEM_U8500=y
 CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 CONFIG_EXT2_FS_POSIX_ACL=y
diff --git a/arch/arm/configs/zeus_defconfig b/arch/arm/configs/zeus_defconfig
index 59577ad..547a3c1 100644
--- a/arch/arm/configs/zeus_defconfig
+++ b/arch/arm/configs/zeus_defconfig
@@ -140,7 +140,7 @@
 CONFIG_USB_SERIAL_GENERIC=y
 CONFIG_USB_SERIAL_MCT_U232=m
 CONFIG_USB_GADGET=m
-CONFIG_USB_GADGET_PXA27X=y
+CONFIG_USB_PXA27X=y
 CONFIG_USB_ETH=m
 CONFIG_USB_GADGETFS=m
 CONFIG_USB_FILE_STORAGE=m
diff --git a/arch/arm/include/asm/hardware/cache-l2x0.h b/arch/arm/include/asm/hardware/cache-l2x0.h
index 1db1143..7df239b 100644
--- a/arch/arm/include/asm/hardware/cache-l2x0.h
+++ b/arch/arm/include/asm/hardware/cache-l2x0.h
@@ -20,6 +20,8 @@
 #ifndef __ASM_ARM_HARDWARE_L2X0_H
 #define __ASM_ARM_HARDWARE_L2X0_H
 
+#include <linux/errno.h>
+
 #define L2X0_CACHE_ID			0x000
 #define L2X0_CACHE_TYPE			0x004
 #define L2X0_CTRL			0x100
diff --git a/arch/arm/include/asm/mach/arch.h b/arch/arm/include/asm/mach/arch.h
index 7d19425..2b0efc3 100644
--- a/arch/arm/include/asm/mach/arch.h
+++ b/arch/arm/include/asm/mach/arch.h
@@ -13,6 +13,7 @@
 struct tag;
 struct meminfo;
 struct sys_timer;
+struct pt_regs;
 
 struct machine_desc {
 	unsigned int		nr;		/* architecture number	*/
diff --git a/arch/arm/include/asm/pmu.h b/arch/arm/include/asm/pmu.h
index 71d99b8..0bda22c 100644
--- a/arch/arm/include/asm/pmu.h
+++ b/arch/arm/include/asm/pmu.h
@@ -55,16 +55,6 @@
 extern void
 release_pmu(enum arm_pmu_type type);
 
-/**
- * init_pmu() - Initialise the PMU.
- *
- * Initialise the system ready for PMU enabling. This should typically set the
- * IRQ affinity and nothing else. The users (oprofile/perf events etc) will do
- * the actual hardware initialisation.
- */
-extern int
-init_pmu(enum arm_pmu_type type);
-
 #else /* CONFIG_CPU_HAS_PMU */
 
 #include <linux/err.h>
diff --git a/arch/arm/include/asm/topology.h b/arch/arm/include/asm/topology.h
index a7e457e..58b8b84 100644
--- a/arch/arm/include/asm/topology.h
+++ b/arch/arm/include/asm/topology.h
@@ -25,7 +25,7 @@
 
 void init_cpu_topology(void);
 void store_cpu_topology(unsigned int cpuid);
-const struct cpumask *cpu_coregroup_mask(unsigned int cpu);
+const struct cpumask *cpu_coregroup_mask(int cpu);
 
 #else
 
diff --git a/arch/arm/include/asm/unistd.h b/arch/arm/include/asm/unistd.h
index c60a294..4a11237 100644
--- a/arch/arm/include/asm/unistd.h
+++ b/arch/arm/include/asm/unistd.h
@@ -402,6 +402,8 @@
 #define __NR_syncfs			(__NR_SYSCALL_BASE+373)
 #define __NR_sendmmsg			(__NR_SYSCALL_BASE+374)
 #define __NR_setns			(__NR_SYSCALL_BASE+375)
+#define __NR_process_vm_readv		(__NR_SYSCALL_BASE+376)
+#define __NR_process_vm_writev		(__NR_SYSCALL_BASE+377)
 
 /*
  * The following SWIs are ARM private.
diff --git a/arch/arm/include/asm/unwind.h b/arch/arm/include/asm/unwind.h
index a5edf42..d1c3f3a 100644
--- a/arch/arm/include/asm/unwind.h
+++ b/arch/arm/include/asm/unwind.h
@@ -30,14 +30,15 @@
 };
 
 struct unwind_idx {
-	unsigned long addr;
+	unsigned long addr_offset;
 	unsigned long insn;
 };
 
 struct unwind_table {
 	struct list_head list;
-	struct unwind_idx *start;
-	struct unwind_idx *stop;
+	const struct unwind_idx *start;
+	const struct unwind_idx *origin;
+	const struct unwind_idx *stop;
 	unsigned long begin_addr;
 	unsigned long end_addr;
 };
@@ -49,15 +50,6 @@
 extern void unwind_table_del(struct unwind_table *tab);
 extern void unwind_backtrace(struct pt_regs *regs, struct task_struct *tsk);
 
-#ifdef CONFIG_ARM_UNWIND
-extern int __init unwind_init(void);
-#else
-static inline int __init unwind_init(void)
-{
-	return 0;
-}
-#endif
-
 #endif	/* !__ASSEMBLY__ */
 
 #ifdef CONFIG_ARM_UNWIND
diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S
index 9943e9e..463ff4a 100644
--- a/arch/arm/kernel/calls.S
+++ b/arch/arm/kernel/calls.S
@@ -385,6 +385,8 @@
 		CALL(sys_syncfs)
 		CALL(sys_sendmmsg)
 /* 375 */	CALL(sys_setns)
+		CALL(sys_process_vm_readv)
+		CALL(sys_process_vm_writev)
 #ifndef syscalls_counted
 .equ syscalls_padding, ((NR_syscalls + 3) & ~3) - NR_syscalls
 #define syscalls_counted
diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
index 9ad50c4..b145f16 100644
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -497,7 +497,7 @@
 	.popsection
 	.pushsection __ex_table,"a"
 	.long	1b, 4b
-#if __LINUX_ARM_ARCH__ >= 7
+#if CONFIG_ARM_THUMB && __LINUX_ARM_ARCH__ >= 6 && CONFIG_CPU_V7
 	.long	2b, 4b
 	.long	3b, 4b
 #endif
diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S
index 566c54c..08c82fd 100644
--- a/arch/arm/kernel/head.S
+++ b/arch/arm/kernel/head.S
@@ -360,7 +360,7 @@
  *  r13 = *virtual* address to jump to upon completion
  */
 __enable_mmu:
-#ifdef CONFIG_ALIGNMENT_TRAP
+#if defined(CONFIG_ALIGNMENT_TRAP) && __LINUX_ARM_ARCH__ < 6
 	orr	r0, r0, #CR_A
 #else
 	bic	r0, r0, #CR_A
diff --git a/arch/arm/kernel/kprobes-arm.c b/arch/arm/kernel/kprobes-arm.c
index 9fe8910..8a30c89 100644
--- a/arch/arm/kernel/kprobes-arm.c
+++ b/arch/arm/kernel/kprobes-arm.c
@@ -519,10 +519,12 @@
 static const union decode_item arm_cccc_0001_____1001_table[] = {
 	/* Synchronization primitives					*/
 
+#if __LINUX_ARM_ARCH__ < 6
+	/* Deprecated on ARMv6 and may be UNDEFINED on v7		*/
 	/* SMP/SWPB		cccc 0001 0x00 xxxx xxxx xxxx 1001 xxxx */
 	DECODE_EMULATEX	(0x0fb000f0, 0x01000090, emulate_rd12rn16rm0_rwflags_nopc,
 						 REGS(NOPC, NOPC, 0, 0, NOPC)),
-
+#endif
 	/* LDREX/STREX{,D,B,H}	cccc 0001 1xxx xxxx xxxx xxxx 1001 xxxx */
 	/* And unallocated instructions...				*/
 	DECODE_END
diff --git a/arch/arm/kernel/kprobes-test-arm.c b/arch/arm/kernel/kprobes-test-arm.c
index fc82de8..ba32b39 100644
--- a/arch/arm/kernel/kprobes-test-arm.c
+++ b/arch/arm/kernel/kprobes-test-arm.c
@@ -427,18 +427,25 @@
 
 	TEST_GROUP("Synchronization primitives")
 
-	/*
-	 * Use hard coded constants for SWP instructions to avoid warnings
-	 * about deprecated instructions.
-	 */
-	TEST_RP( ".word 0xe108e097 @ swp	lr, r",7,VAL2,", [r",8,0,"]")
-	TEST_R(  ".word 0x610d0091 @ swpvs	r0, r",1,VAL1,", [sp]")
-	TEST_RP( ".word 0xe10cd09e @ swp	sp, r",14,VAL2,", [r",12,13*4,"]")
+#if __LINUX_ARM_ARCH__ < 6
+	TEST_RP("swp	lr, r",7,VAL2,", [r",8,0,"]")
+	TEST_R( "swpvs	r0, r",1,VAL1,", [sp]")
+	TEST_RP("swp	sp, r",14,VAL2,", [r",12,13*4,"]")
+#else
+	TEST_UNSUPPORTED(".word 0xe108e097 @ swp	lr, r7, [r8]")
+	TEST_UNSUPPORTED(".word 0x610d0091 @ swpvs	r0, r1, [sp]")
+	TEST_UNSUPPORTED(".word 0xe10cd09e @ swp	sp, r14 [r12]")
+#endif
 	TEST_UNSUPPORTED(".word 0xe102f091 @ swp pc, r1, [r2]")
 	TEST_UNSUPPORTED(".word 0xe102009f @ swp r0, pc, [r2]")
 	TEST_UNSUPPORTED(".word 0xe10f0091 @ swp r0, r1, [pc]")
-	TEST_RP( ".word 0xe148e097 @ swpb	lr, r",7,VAL2,", [r",8,0,"]")
-	TEST_R(  ".word 0x614d0091 @ swpvsb	r0, r",1,VAL1,", [sp]")
+#if __LINUX_ARM_ARCH__ < 6
+	TEST_RP("swpb	lr, r",7,VAL2,", [r",8,0,"]")
+	TEST_R( "swpvsb	r0, r",1,VAL1,", [sp]")
+#else
+	TEST_UNSUPPORTED(".word 0xe148e097 @ swpb	lr, r7, [r8]")
+	TEST_UNSUPPORTED(".word 0x614d0091 @ swpvsb	r0, r1, [sp]")
+#endif
 	TEST_UNSUPPORTED(".word 0xe142f091 @ swpb pc, r1, [r2]")
 
 	TEST_UNSUPPORTED(".word	0xe1100090") /* Unallocated space */
@@ -550,7 +557,7 @@
 	TEST_RPR(  "strccd	r",8, VAL2,", [r",13,0, ", r",12,48,"]")
 	TEST_RPR(  "strd	r",4, VAL1,", [r",2, 24,", r",3, 48,"]!")
 	TEST_RPR(  "strcsd	r",12,VAL2,", [r",11,48,", -r",10,24,"]!")
-	TEST_RPR(  "strd	r",2, VAL1,", [r",3, 24,"], r",4,48,"")
+	TEST_RPR(  "strd	r",2, VAL1,", [r",5, 24,"], r",4,48,"")
 	TEST_RPR(  "strd	r",10,VAL2,", [r",9, 48,"], -r",7,24,"")
 	TEST_UNSUPPORTED(".word 0xe1afc0fa	@ strd r12, [pc, r10]!")
 
diff --git a/arch/arm/kernel/kprobes-test-thumb.c b/arch/arm/kernel/kprobes-test-thumb.c
index 5e726c3..5d8b857 100644
--- a/arch/arm/kernel/kprobes-test-thumb.c
+++ b/arch/arm/kernel/kprobes-test-thumb.c
@@ -222,8 +222,8 @@
 DONT_TEST_IN_ITBLOCK(
 	TEST_BF_R(  "cbnz	r",0,0, ", 2f")
 	TEST_BF_R(  "cbz	r",2,-1,", 2f")
-	TEST_BF_RX( "cbnz	r",4,1, ", 2f",0x20)
-	TEST_BF_RX( "cbz	r",7,0, ", 2f",0x40)
+	TEST_BF_RX( "cbnz	r",4,1, ", 2f", SPACE_0x20)
+	TEST_BF_RX( "cbz	r",7,0, ", 2f", SPACE_0x40)
 )
 	TEST_R("sxth	r0, r",7, HH1,"")
 	TEST_R("sxth	r7, r",0, HH2,"")
@@ -246,7 +246,7 @@
 	TESTCASE_START(code)		\
 	TEST_ARG_PTR(13, offset)	\
 	TEST_ARG_END("")		\
-	TEST_BRANCH_F(code,0)		\
+	TEST_BRANCH_F(code)		\
 	TESTCASE_END
 
 	TEST("push	{r0}")
@@ -319,8 +319,8 @@
 
 	TEST_BF(  "b	2f")
 	TEST_BB(  "b	2b")
-	TEST_BF_X("b	2f", 0x400)
-	TEST_BB_X("b	2b", 0x400)
+	TEST_BF_X("b	2f", SPACE_0x400)
+	TEST_BB_X("b	2b", SPACE_0x400)
 
 	TEST_GROUP("Testing instructions in IT blocks")
 
@@ -746,7 +746,7 @@
 	TEST_BB("bne.w	2b")
 	TEST_BF("bgt.w	2f")
 	TEST_BB("blt.w	2b")
-	TEST_BF_X("bpl.w	2f",0x1000)
+	TEST_BF_X("bpl.w	2f", SPACE_0x1000)
 )
 
 	TEST_UNSUPPORTED("msr	cpsr, r0")
@@ -786,11 +786,11 @@
 
 	TEST_BF(  "b.w	2f")
 	TEST_BB(  "b.w	2b")
-	TEST_BF_X("b.w	2f", 0x1000)
+	TEST_BF_X("b.w	2f", SPACE_0x1000)
 
 	TEST_BF(  "bl.w	2f")
 	TEST_BB(  "bl.w	2b")
-	TEST_BB_X("bl.w	2b", 0x1000)
+	TEST_BB_X("bl.w	2b", SPACE_0x1000)
 
 	TEST_X(	"blx	__dummy_arm_subroutine",
 		".arm				\n\t"
diff --git a/arch/arm/kernel/kprobes-test.h b/arch/arm/kernel/kprobes-test.h
index 0dc5d77..e28a869 100644
--- a/arch/arm/kernel/kprobes-test.h
+++ b/arch/arm/kernel/kprobes-test.h
@@ -149,23 +149,31 @@
 	"1:	"instruction"				\n\t"	\
 	"	nop					\n\t"
 
-#define TEST_BRANCH_F(instruction, xtra_dist)			\
+#define TEST_BRANCH_F(instruction)				\
 	TEST_INSTRUCTION(instruction)				\
-	".if "#xtra_dist"				\n\t"	\
-	"	b	99f				\n\t"	\
-	".space "#xtra_dist"				\n\t"	\
-	".endif						\n\t"	\
 	"	b	99f				\n\t"	\
 	"2:	nop					\n\t"
 
-#define TEST_BRANCH_B(instruction, xtra_dist)			\
+#define TEST_BRANCH_B(instruction)				\
 	"	b	50f				\n\t"	\
 	"	b	99f				\n\t"	\
 	"2:	nop					\n\t"	\
 	"	b	99f				\n\t"	\
-	".if "#xtra_dist"				\n\t"	\
-	".space "#xtra_dist"				\n\t"	\
-	".endif						\n\t"	\
+	TEST_INSTRUCTION(instruction)
+
+#define TEST_BRANCH_FX(instruction, codex)			\
+	TEST_INSTRUCTION(instruction)				\
+	"	b	99f				\n\t"	\
+	codex"						\n\t"	\
+	"	b	99f				\n\t"	\
+	"2:	nop					\n\t"
+
+#define TEST_BRANCH_BX(instruction, codex)			\
+	"	b	50f				\n\t"	\
+	"	b	99f				\n\t"	\
+	"2:	nop					\n\t"	\
+	"	b	99f				\n\t"	\
+	codex"						\n\t"	\
 	TEST_INSTRUCTION(instruction)
 
 #define TESTCASE_END						\
@@ -301,47 +309,60 @@
 	TESTCASE_START(code1 #reg1 code2)	\
 	TEST_ARG_PTR(reg1, val1)		\
 	TEST_ARG_END("")			\
-	TEST_BRANCH_F(code1 #reg1 code2, 0)	\
+	TEST_BRANCH_F(code1 #reg1 code2)	\
 	TESTCASE_END
 
-#define TEST_BF_X(code, xtra_dist)		\
+#define TEST_BF(code)				\
 	TESTCASE_START(code)			\
 	TEST_ARG_END("")			\
-	TEST_BRANCH_F(code, xtra_dist)		\
+	TEST_BRANCH_F(code)			\
 	TESTCASE_END
 
-#define TEST_BB_X(code, xtra_dist)		\
+#define TEST_BB(code)				\
 	TESTCASE_START(code)			\
 	TEST_ARG_END("")			\
-	TEST_BRANCH_B(code, xtra_dist)		\
+	TEST_BRANCH_B(code)			\
 	TESTCASE_END
 
-#define TEST_BF_RX(code1, reg, val, code2, xtra_dist)	\
-	TESTCASE_START(code1 #reg code2)		\
-	TEST_ARG_REG(reg, val)				\
-	TEST_ARG_END("")				\
-	TEST_BRANCH_F(code1 #reg code2, xtra_dist)	\
+#define TEST_BF_R(code1, reg, val, code2)	\
+	TESTCASE_START(code1 #reg code2)	\
+	TEST_ARG_REG(reg, val)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_F(code1 #reg code2)		\
 	TESTCASE_END
 
-#define TEST_BB_RX(code1, reg, val, code2, xtra_dist)	\
-	TESTCASE_START(code1 #reg code2)		\
-	TEST_ARG_REG(reg, val)				\
-	TEST_ARG_END("")				\
-	TEST_BRANCH_B(code1 #reg code2, xtra_dist)	\
+#define TEST_BB_R(code1, reg, val, code2)	\
+	TESTCASE_START(code1 #reg code2)	\
+	TEST_ARG_REG(reg, val)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_B(code1 #reg code2)		\
 	TESTCASE_END
 
-#define TEST_BF(code)	TEST_BF_X(code, 0)
-#define TEST_BB(code)	TEST_BB_X(code, 0)
-
-#define TEST_BF_R(code1, reg, val, code2) TEST_BF_RX(code1, reg, val, code2, 0)
-#define TEST_BB_R(code1, reg, val, code2) TEST_BB_RX(code1, reg, val, code2, 0)
-
 #define TEST_BF_RR(code1, reg1, val1, code2, reg2, val2, code3)	\
 	TESTCASE_START(code1 #reg1 code2 #reg2 code3)		\
 	TEST_ARG_REG(reg1, val1)				\
 	TEST_ARG_REG(reg2, val2)				\
 	TEST_ARG_END("")					\
-	TEST_BRANCH_F(code1 #reg1 code2 #reg2 code3, 0)		\
+	TEST_BRANCH_F(code1 #reg1 code2 #reg2 code3)		\
+	TESTCASE_END
+
+#define TEST_BF_X(code, codex)			\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_FX(code, codex)		\
+	TESTCASE_END
+
+#define TEST_BB_X(code, codex)			\
+	TESTCASE_START(code)			\
+	TEST_ARG_END("")			\
+	TEST_BRANCH_BX(code, codex)		\
+	TESTCASE_END
+
+#define TEST_BF_RX(code1, reg, val, code2, codex)	\
+	TESTCASE_START(code1 #reg code2)		\
+	TEST_ARG_REG(reg, val)				\
+	TEST_ARG_END("")				\
+	TEST_BRANCH_FX(code1 #reg code2, codex)		\
 	TESTCASE_END
 
 #define TEST_X(code, codex)			\
@@ -372,6 +393,25 @@
 	TESTCASE_END
 
 
+/*
+ * Macros for defining space directives spread over multiple lines.
+ * These are required so the compiler guesses better the length of inline asm
+ * code and will spill the literal pool early enough to avoid generating PC
+ * relative loads with out of range offsets.
+ */
+#define TWICE(x)	x x
+#define SPACE_0x8	TWICE(".space 4\n\t")
+#define SPACE_0x10	TWICE(SPACE_0x8)
+#define SPACE_0x20	TWICE(SPACE_0x10)
+#define SPACE_0x40	TWICE(SPACE_0x20)
+#define SPACE_0x80	TWICE(SPACE_0x40)
+#define SPACE_0x100	TWICE(SPACE_0x80)
+#define SPACE_0x200	TWICE(SPACE_0x100)
+#define SPACE_0x400	TWICE(SPACE_0x200)
+#define SPACE_0x800	TWICE(SPACE_0x400)
+#define SPACE_0x1000	TWICE(SPACE_0x800)
+
+
 /* Various values used in test cases... */
 #define N(val)	(val ^ 0xffffffff)
 #define VAL1	0x12345678
diff --git a/arch/arm/kernel/machine_kexec.c b/arch/arm/kernel/machine_kexec.c
index c1b4463..e59bbd4 100644
--- a/arch/arm/kernel/machine_kexec.c
+++ b/arch/arm/kernel/machine_kexec.c
@@ -32,24 +32,6 @@
 
 int machine_kexec_prepare(struct kimage *image)
 {
-	unsigned long page_list;
-	void *reboot_code_buffer;
-	page_list = image->head & PAGE_MASK;
-
-	reboot_code_buffer = page_address(image->control_code_page);
-
-	/* Prepare parameters for reboot_code_buffer*/
-	kexec_start_address = image->start;
-	kexec_indirection_page = page_list;
-	kexec_mach_type = machine_arch_type;
-	kexec_boot_atags = image->start - KEXEC_ARM_ZIMAGE_OFFSET + KEXEC_ARM_ATAGS_OFFSET;
-
-	/* copy our kernel relocation code to the control code page */
-	memcpy(reboot_code_buffer,
-	       relocate_new_kernel, relocate_new_kernel_size);
-
-	flush_icache_range((unsigned long) reboot_code_buffer,
-			   (unsigned long) reboot_code_buffer + KEXEC_CONTROL_PAGE_SIZE);
 	return 0;
 }
 
@@ -100,14 +82,31 @@
 
 void machine_kexec(struct kimage *image)
 {
+	unsigned long page_list;
 	unsigned long reboot_code_buffer_phys;
 	void *reboot_code_buffer;
 
+
+	page_list = image->head & PAGE_MASK;
+
 	/* we need both effective and real address here */
 	reboot_code_buffer_phys =
 	    page_to_pfn(image->control_code_page) << PAGE_SHIFT;
 	reboot_code_buffer = page_address(image->control_code_page);
 
+	/* Prepare parameters for reboot_code_buffer*/
+	kexec_start_address = image->start;
+	kexec_indirection_page = page_list;
+	kexec_mach_type = machine_arch_type;
+	kexec_boot_atags = image->start - KEXEC_ARM_ZIMAGE_OFFSET + KEXEC_ARM_ATAGS_OFFSET;
+
+	/* copy our kernel relocation code to the control code page */
+	memcpy(reboot_code_buffer,
+	       relocate_new_kernel, relocate_new_kernel_size);
+
+
+	flush_icache_range((unsigned long) reboot_code_buffer,
+			   (unsigned long) reboot_code_buffer + KEXEC_CONTROL_PAGE_SIZE);
 	printk(KERN_INFO "Bye!\n");
 
 	if (kexec_reinit)
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 24e2347..88b0941 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -343,19 +343,25 @@
 {
 	struct perf_event *sibling, *leader = event->group_leader;
 	struct pmu_hw_events fake_pmu;
+	DECLARE_BITMAP(fake_used_mask, ARMPMU_MAX_HWEVENTS);
 
-	memset(&fake_pmu, 0, sizeof(fake_pmu));
+	/*
+	 * Initialise the fake PMU. We only need to populate the
+	 * used_mask for the purposes of validation.
+	 */
+	memset(fake_used_mask, 0, sizeof(fake_used_mask));
+	fake_pmu.used_mask = fake_used_mask;
 
 	if (!validate_event(&fake_pmu, leader))
-		return -ENOSPC;
+		return -EINVAL;
 
 	list_for_each_entry(sibling, &leader->sibling_list, group_entry) {
 		if (!validate_event(&fake_pmu, sibling))
-			return -ENOSPC;
+			return -EINVAL;
 	}
 
 	if (!validate_event(&fake_pmu, event))
-		return -ENOSPC;
+		return -EINVAL;
 
 	return 0;
 }
@@ -396,6 +402,9 @@
 	int i, err, irq, irqs;
 	struct platform_device *pmu_device = armpmu->plat_device;
 
+	if (!pmu_device)
+		return -ENODEV;
+
 	err = reserve_pmu(armpmu->type);
 	if (err) {
 		pr_warning("unable to reserve pmu\n");
@@ -631,6 +640,9 @@
 
 static int __devinit armpmu_device_probe(struct platform_device *pdev)
 {
+	if (!cpu_pmu)
+		return -ENODEV;
+
 	cpu_pmu->plat_device = pdev;
 	return 0;
 }
diff --git a/arch/arm/kernel/pmu.c b/arch/arm/kernel/pmu.c
index 2c3407e..2334bf8 100644
--- a/arch/arm/kernel/pmu.c
+++ b/arch/arm/kernel/pmu.c
@@ -33,3 +33,4 @@
 {
 	clear_bit_unlock(type, pmu_lock);
 }
+EXPORT_SYMBOL_GPL(release_pmu);
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 75316f0..3d0c6fb 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -192,6 +192,9 @@
 #endif
 
 			local_irq_disable();
+#ifdef CONFIG_PL310_ERRATA_769419
+			wmb();
+#endif
 			if (hlt_counter) {
 				local_irq_enable();
 				cpu_relax();
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 7e7977a..8fc2c8f 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -461,8 +461,10 @@
 	       cpu_name, read_cpuid_id(), read_cpuid_id() & 15,
 	       proc_arch[cpu_architecture()], cr_alignment);
 
-	sprintf(init_utsname()->machine, "%s%c", list->arch_name, ENDIANNESS);
-	sprintf(elf_platform, "%s%c", list->elf_name, ENDIANNESS);
+	snprintf(init_utsname()->machine, __NEW_UTS_LEN + 1, "%s%c",
+		 list->arch_name, ENDIANNESS);
+	snprintf(elf_platform, ELF_PLATFORM_SIZE, "%s%c",
+		 list->elf_name, ENDIANNESS);
 	elf_hwcap = list->elf_hwcap;
 #ifndef CONFIG_ARM_THUMB
 	elf_hwcap &= ~HWCAP_THUMB;
@@ -893,8 +895,6 @@
 {
 	struct machine_desc *mdesc;
 
-	unwind_init();
-
 	setup_processor();
 	mdesc = setup_machine_fdt(__atags_pointer);
 	if (!mdesc)
@@ -902,6 +902,12 @@
 	machine_desc = mdesc;
 	machine_name = mdesc->name;
 
+#ifdef CONFIG_ZONE_DMA
+	if (mdesc->dma_zone_size) {
+		extern unsigned long arm_dma_zone_size;
+		arm_dma_zone_size = mdesc->dma_zone_size;
+	}
+#endif
 	if (mdesc->soft_reboot)
 		reboot_setup("s");
 
@@ -932,12 +938,6 @@
 
 	tcm_init();
 
-#ifdef CONFIG_ZONE_DMA
-	if (mdesc->dma_zone_size) {
-		extern unsigned long arm_dma_zone_size;
-		arm_dma_zone_size = mdesc->dma_zone_size;
-	}
-#endif
 #ifdef CONFIG_MULTI_IRQ_HANDLER
 	handle_arch_irq = mdesc->handle_irq;
 #endif
diff --git a/arch/arm/kernel/topology.c b/arch/arm/kernel/topology.c
index 1040c00..8200dea 100644
--- a/arch/arm/kernel/topology.c
+++ b/arch/arm/kernel/topology.c
@@ -43,7 +43,7 @@
 
 struct cputopo_arm cpu_topology[NR_CPUS];
 
-const struct cpumask *cpu_coregroup_mask(unsigned int cpu)
+const struct cpumask *cpu_coregroup_mask(int cpu)
 {
 	return &cpu_topology[cpu].core_sibling;
 }
diff --git a/arch/arm/kernel/unwind.c b/arch/arm/kernel/unwind.c
index e7e8365..00df012 100644
--- a/arch/arm/kernel/unwind.c
+++ b/arch/arm/kernel/unwind.c
@@ -67,7 +67,7 @@
 
 struct unwind_ctrl_block {
 	unsigned long vrs[16];		/* virtual register set */
-	unsigned long *insn;		/* pointer to the current instructions word */
+	const unsigned long *insn;	/* pointer to the current instructions word */
 	int entries;			/* number of entries left to interpret */
 	int byte;			/* current byte number in the instructions word */
 };
@@ -83,8 +83,9 @@
 	PC = 15
 };
 
-extern struct unwind_idx __start_unwind_idx[];
-extern struct unwind_idx __stop_unwind_idx[];
+extern const struct unwind_idx __start_unwind_idx[];
+static const struct unwind_idx *__origin_unwind_idx;
+extern const struct unwind_idx __stop_unwind_idx[];
 
 static DEFINE_SPINLOCK(unwind_lock);
 static LIST_HEAD(unwind_tables);
@@ -98,45 +99,99 @@
 })
 
 /*
- * Binary search in the unwind index. The entries entries are
+ * Binary search in the unwind index. The entries are
  * guaranteed to be sorted in ascending order by the linker.
+ *
+ * start = first entry
+ * origin = first entry with positive offset (or stop if there is no such entry)
+ * stop - 1 = last entry
  */
-static struct unwind_idx *search_index(unsigned long addr,
-				       struct unwind_idx *first,
-				       struct unwind_idx *last)
+static const struct unwind_idx *search_index(unsigned long addr,
+				       const struct unwind_idx *start,
+				       const struct unwind_idx *origin,
+				       const struct unwind_idx *stop)
 {
-	pr_debug("%s(%08lx, %p, %p)\n", __func__, addr, first, last);
+	unsigned long addr_prel31;
 
-	if (addr < first->addr) {
-		pr_warning("unwind: Unknown symbol address %08lx\n", addr);
-		return NULL;
-	} else if (addr >= last->addr)
-		return last;
+	pr_debug("%s(%08lx, %p, %p, %p)\n",
+			__func__, addr, start, origin, stop);
 
-	while (first < last - 1) {
-		struct unwind_idx *mid = first + ((last - first + 1) >> 1);
+	/*
+	 * only search in the section with the matching sign. This way the
+	 * prel31 numbers can be compared as unsigned longs.
+	 */
+	if (addr < (unsigned long)start)
+		/* negative offsets: [start; origin) */
+		stop = origin;
+	else
+		/* positive offsets: [origin; stop) */
+		start = origin;
 
-		if (addr < mid->addr)
-			last = mid;
-		else
-			first = mid;
+	/* prel31 for address relavive to start */
+	addr_prel31 = (addr - (unsigned long)start) & 0x7fffffff;
+
+	while (start < stop - 1) {
+		const struct unwind_idx *mid = start + ((stop - start) >> 1);
+
+		/*
+		 * As addr_prel31 is relative to start an offset is needed to
+		 * make it relative to mid.
+		 */
+		if (addr_prel31 - ((unsigned long)mid - (unsigned long)start) <
+				mid->addr_offset)
+			stop = mid;
+		else {
+			/* keep addr_prel31 relative to start */
+			addr_prel31 -= ((unsigned long)mid -
+					(unsigned long)start);
+			start = mid;
+		}
 	}
 
-	return first;
+	if (likely(start->addr_offset <= addr_prel31))
+		return start;
+	else {
+		pr_warning("unwind: Unknown symbol address %08lx\n", addr);
+		return NULL;
+	}
 }
 
-static struct unwind_idx *unwind_find_idx(unsigned long addr)
+static const struct unwind_idx *unwind_find_origin(
+		const struct unwind_idx *start, const struct unwind_idx *stop)
 {
-	struct unwind_idx *idx = NULL;
+	pr_debug("%s(%p, %p)\n", __func__, start, stop);
+	while (start < stop) {
+		const struct unwind_idx *mid = start + ((stop - start) >> 1);
+
+		if (mid->addr_offset >= 0x40000000)
+			/* negative offset */
+			start = mid + 1;
+		else
+			/* positive offset */
+			stop = mid;
+	}
+	pr_debug("%s -> %p\n", __func__, stop);
+	return stop;
+}
+
+static const struct unwind_idx *unwind_find_idx(unsigned long addr)
+{
+	const struct unwind_idx *idx = NULL;
 	unsigned long flags;
 
 	pr_debug("%s(%08lx)\n", __func__, addr);
 
-	if (core_kernel_text(addr))
+	if (core_kernel_text(addr)) {
+		if (unlikely(!__origin_unwind_idx))
+			__origin_unwind_idx =
+				unwind_find_origin(__start_unwind_idx,
+						__stop_unwind_idx);
+
 		/* main unwind table */
 		idx = search_index(addr, __start_unwind_idx,
-				   __stop_unwind_idx - 1);
-	else {
+				   __origin_unwind_idx,
+				   __stop_unwind_idx);
+	} else {
 		/* module unwind tables */
 		struct unwind_table *table;
 
@@ -145,7 +200,8 @@
 			if (addr >= table->begin_addr &&
 			    addr < table->end_addr) {
 				idx = search_index(addr, table->start,
-						   table->stop - 1);
+						   table->origin,
+						   table->stop);
 				/* Move-to-front to exploit common traces */
 				list_move(&table->list, &unwind_tables);
 				break;
@@ -274,7 +330,7 @@
 int unwind_frame(struct stackframe *frame)
 {
 	unsigned long high, low;
-	struct unwind_idx *idx;
+	const struct unwind_idx *idx;
 	struct unwind_ctrl_block ctrl;
 
 	/* only go to a higher address on the stack */
@@ -399,7 +455,6 @@
 				      unsigned long text_size)
 {
 	unsigned long flags;
-	struct unwind_idx *idx;
 	struct unwind_table *tab = kmalloc(sizeof(*tab), GFP_KERNEL);
 
 	pr_debug("%s(%08lx, %08lx, %08lx, %08lx)\n", __func__, start, size,
@@ -408,15 +463,12 @@
 	if (!tab)
 		return tab;
 
-	tab->start = (struct unwind_idx *)start;
-	tab->stop = (struct unwind_idx *)(start + size);
+	tab->start = (const struct unwind_idx *)start;
+	tab->stop = (const struct unwind_idx *)(start + size);
+	tab->origin = unwind_find_origin(tab->start, tab->stop);
 	tab->begin_addr = text_addr;
 	tab->end_addr = text_addr + text_size;
 
-	/* Convert the symbol addresses to absolute values */
-	for (idx = tab->start; idx < tab->stop; idx++)
-		idx->addr = prel31_to_addr(&idx->addr);
-
 	spin_lock_irqsave(&unwind_lock, flags);
 	list_add_tail(&tab->list, &unwind_tables);
 	spin_unlock_irqrestore(&unwind_lock, flags);
@@ -437,16 +489,3 @@
 
 	kfree(tab);
 }
-
-int __init unwind_init(void)
-{
-	struct unwind_idx *idx;
-
-	/* Convert the symbol addresses to absolute values */
-	for (idx = __start_unwind_idx; idx < __stop_unwind_idx; idx++)
-		idx->addr = prel31_to_addr(&idx->addr);
-
-	pr_debug("unwind: ARM stack unwinding initialised\n");
-
-	return 0;
-}
diff --git a/arch/arm/lib/bitops.h b/arch/arm/lib/bitops.h
index 10d868a..d6408d1 100644
--- a/arch/arm/lib/bitops.h
+++ b/arch/arm/lib/bitops.h
@@ -1,5 +1,9 @@
+#include <asm/unwind.h>
+
 #if __LINUX_ARM_ARCH__ >= 6
-	.macro	bitop, instr
+	.macro	bitop, name, instr
+ENTRY(	\name		)
+UNWIND(	.fnstart	)
 	ands	ip, r1, #3
 	strneb	r1, [ip]		@ assert word-aligned
 	mov	r2, #1
@@ -13,9 +17,13 @@
 	cmp	r0, #0
 	bne	1b
 	bx	lr
+UNWIND(	.fnend		)
+ENDPROC(\name		)
 	.endm
 
-	.macro	testop, instr, store
+	.macro	testop, name, instr, store
+ENTRY(	\name		)
+UNWIND(	.fnstart	)
 	ands	ip, r1, #3
 	strneb	r1, [ip]		@ assert word-aligned
 	mov	r2, #1
@@ -34,9 +42,13 @@
 	cmp	r0, #0
 	movne	r0, #1
 2:	bx	lr
+UNWIND(	.fnend		)
+ENDPROC(\name		)
 	.endm
 #else
-	.macro	bitop, instr
+	.macro	bitop, name, instr
+ENTRY(	\name		)
+UNWIND(	.fnstart	)
 	ands	ip, r1, #3
 	strneb	r1, [ip]		@ assert word-aligned
 	and	r2, r0, #31
@@ -49,6 +61,8 @@
 	str	r2, [r1, r0, lsl #2]
 	restore_irqs ip
 	mov	pc, lr
+UNWIND(	.fnend		)
+ENDPROC(\name		)
 	.endm
 
 /**
@@ -59,7 +73,9 @@
  * Note: we can trivially conditionalise the store instruction
  * to avoid dirtying the data cache.
  */
-	.macro	testop, instr, store
+	.macro	testop, name, instr, store
+ENTRY(	\name		)
+UNWIND(	.fnstart	)
 	ands	ip, r1, #3
 	strneb	r1, [ip]		@ assert word-aligned
 	and	r3, r0, #31
@@ -73,5 +89,7 @@
 	moveq	r0, #0
 	restore_irqs ip
 	mov	pc, lr
+UNWIND(	.fnend		)
+ENDPROC(\name		)
 	.endm
 #endif
diff --git a/arch/arm/lib/changebit.S b/arch/arm/lib/changebit.S
index 68ed5b6..f402786 100644
--- a/arch/arm/lib/changebit.S
+++ b/arch/arm/lib/changebit.S
@@ -12,6 +12,4 @@
 #include "bitops.h"
                 .text
 
-ENTRY(_change_bit)
-	bitop	eor
-ENDPROC(_change_bit)
+bitop	_change_bit, eor
diff --git a/arch/arm/lib/clearbit.S b/arch/arm/lib/clearbit.S
index 4c04c3b..f6b75fb 100644
--- a/arch/arm/lib/clearbit.S
+++ b/arch/arm/lib/clearbit.S
@@ -12,6 +12,4 @@
 #include "bitops.h"
                 .text
 
-ENTRY(_clear_bit)
-	bitop	bic
-ENDPROC(_clear_bit)
+bitop	_clear_bit, bic
diff --git a/arch/arm/lib/setbit.S b/arch/arm/lib/setbit.S
index bbee5c6..618feda 100644
--- a/arch/arm/lib/setbit.S
+++ b/arch/arm/lib/setbit.S
@@ -12,6 +12,4 @@
 #include "bitops.h"
 		.text
 
-ENTRY(_set_bit)
-	bitop	orr
-ENDPROC(_set_bit)
+bitop	_set_bit, orr
diff --git a/arch/arm/lib/testchangebit.S b/arch/arm/lib/testchangebit.S
index 15a4d43..4becdc3 100644
--- a/arch/arm/lib/testchangebit.S
+++ b/arch/arm/lib/testchangebit.S
@@ -12,6 +12,4 @@
 #include "bitops.h"
                 .text
 
-ENTRY(_test_and_change_bit)
-	testop	eor, str
-ENDPROC(_test_and_change_bit)
+testop	_test_and_change_bit, eor, str
diff --git a/arch/arm/lib/testclearbit.S b/arch/arm/lib/testclearbit.S
index 521b66b..918841d 100644
--- a/arch/arm/lib/testclearbit.S
+++ b/arch/arm/lib/testclearbit.S
@@ -12,6 +12,4 @@
 #include "bitops.h"
                 .text
 
-ENTRY(_test_and_clear_bit)
-	testop	bicne, strne
-ENDPROC(_test_and_clear_bit)
+testop	_test_and_clear_bit, bicne, strne
diff --git a/arch/arm/lib/testsetbit.S b/arch/arm/lib/testsetbit.S
index 1c98cc2..8d1b2fe 100644
--- a/arch/arm/lib/testsetbit.S
+++ b/arch/arm/lib/testsetbit.S
@@ -12,6 +12,4 @@
 #include "bitops.h"
                 .text
 
-ENTRY(_test_and_set_bit)
-	testop	orreq, streq
-ENDPROC(_test_and_set_bit)
+testop	_test_and_set_bit, orreq, streq
diff --git a/arch/arm/mach-at91/at91cap9_devices.c b/arch/arm/mach-at91/at91cap9_devices.c
index a4401d6..adad70d 100644
--- a/arch/arm/mach-at91/at91cap9_devices.c
+++ b/arch/arm/mach-at91/at91cap9_devices.c
@@ -98,7 +98,7 @@
  *  USB HS Device (Gadget)
  * -------------------------------------------------------------------- */
 
-#if defined(CONFIG_USB_GADGET_ATMEL_USBA) || defined(CONFIG_USB_GADGET_ATMEL_USBA_MODULE)
+#if defined(CONFIG_USB_ATMEL_USBA) || defined(CONFIG_USB_ATMEL_USBA_MODULE)
 
 static struct resource usba_udc_resources[] = {
 	[0] = {
@@ -1021,8 +1021,8 @@
 #if defined(CONFIG_SERIAL_ATMEL)
 static struct resource dbgu_resources[] = {
 	[0] = {
-		.start	= AT91_VA_BASE_SYS + AT91_DBGU,
-		.end	= AT91_VA_BASE_SYS + AT91_DBGU + SZ_512 - 1,
+		.start	= AT91_BASE_SYS + AT91_DBGU,
+		.end	= AT91_BASE_SYS + AT91_DBGU + SZ_512 - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
@@ -1035,7 +1035,6 @@
 static struct atmel_uart_data dbgu_data = {
 	.use_dma_tx	= 0,
 	.use_dma_rx	= 0,		/* DBGU not capable of receive DMA */
-	.regs		= (void __iomem *)(AT91_VA_BASE_SYS + AT91_DBGU),
 };
 
 static u64 dbgu_dmamask = DMA_BIT_MASK(32);
diff --git a/arch/arm/mach-at91/at91rm9200_devices.c b/arch/arm/mach-at91/at91rm9200_devices.c
index 01d8bbd..ad93068 100644
--- a/arch/arm/mach-at91/at91rm9200_devices.c
+++ b/arch/arm/mach-at91/at91rm9200_devices.c
@@ -83,7 +83,7 @@
  *  USB Device (Gadget)
  * -------------------------------------------------------------------- */
 
-#ifdef CONFIG_USB_GADGET_AT91
+#ifdef CONFIG_USB_AT91
 static struct at91_udc_data udc_data;
 
 static struct resource udc_resources[] = {
@@ -877,8 +877,8 @@
 #if defined(CONFIG_SERIAL_ATMEL)
 static struct resource dbgu_resources[] = {
 	[0] = {
-		.start	= AT91_VA_BASE_SYS + AT91_DBGU,
-		.end	= AT91_VA_BASE_SYS + AT91_DBGU + SZ_512 - 1,
+		.start	= AT91_BASE_SYS + AT91_DBGU,
+		.end	= AT91_BASE_SYS + AT91_DBGU + SZ_512 - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
@@ -891,7 +891,6 @@
 static struct atmel_uart_data dbgu_data = {
 	.use_dma_tx	= 0,
 	.use_dma_rx	= 0,		/* DBGU not capable of receive DMA */
-	.regs		= (void __iomem *)(AT91_VA_BASE_SYS + AT91_DBGU),
 };
 
 static u64 dbgu_dmamask = DMA_BIT_MASK(32);
diff --git a/arch/arm/mach-at91/at91sam9260.c b/arch/arm/mach-at91/at91sam9260.c
index b84a9f6..0d20677 100644
--- a/arch/arm/mach-at91/at91sam9260.c
+++ b/arch/arm/mach-at91/at91sam9260.c
@@ -195,9 +195,9 @@
 	CLKDEV_CON_DEV_ID("t0_clk", "atmel_tcb.0", &tc0_clk),
 	CLKDEV_CON_DEV_ID("t1_clk", "atmel_tcb.0", &tc1_clk),
 	CLKDEV_CON_DEV_ID("t2_clk", "atmel_tcb.0", &tc2_clk),
-	CLKDEV_CON_DEV_ID("t3_clk", "atmel_tcb.1", &tc3_clk),
-	CLKDEV_CON_DEV_ID("t4_clk", "atmel_tcb.1", &tc4_clk),
-	CLKDEV_CON_DEV_ID("t5_clk", "atmel_tcb.1", &tc5_clk),
+	CLKDEV_CON_DEV_ID("t0_clk", "atmel_tcb.1", &tc3_clk),
+	CLKDEV_CON_DEV_ID("t1_clk", "atmel_tcb.1", &tc4_clk),
+	CLKDEV_CON_DEV_ID("t2_clk", "atmel_tcb.1", &tc5_clk),
 	CLKDEV_CON_DEV_ID("pclk", "ssc.0", &ssc_clk),
 	/* more usart lookup table for DT entries */
 	CLKDEV_CON_DEV_ID("usart", "fffff200.serial", &mck),
diff --git a/arch/arm/mach-at91/at91sam9260_devices.c b/arch/arm/mach-at91/at91sam9260_devices.c
index 24b6f8c..629fa97 100644
--- a/arch/arm/mach-at91/at91sam9260_devices.c
+++ b/arch/arm/mach-at91/at91sam9260_devices.c
@@ -84,7 +84,7 @@
  *  USB Device (Gadget)
  * -------------------------------------------------------------------- */
 
-#ifdef CONFIG_USB_GADGET_AT91
+#ifdef CONFIG_USB_AT91
 static struct at91_udc_data udc_data;
 
 static struct resource udc_resources[] = {
@@ -837,8 +837,8 @@
 #if defined(CONFIG_SERIAL_ATMEL)
 static struct resource dbgu_resources[] = {
 	[0] = {
-		.start	= AT91_VA_BASE_SYS + AT91_DBGU,
-		.end	= AT91_VA_BASE_SYS + AT91_DBGU + SZ_512 - 1,
+		.start	= AT91_BASE_SYS + AT91_DBGU,
+		.end	= AT91_BASE_SYS + AT91_DBGU + SZ_512 - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
@@ -851,7 +851,6 @@
 static struct atmel_uart_data dbgu_data = {
 	.use_dma_tx	= 0,
 	.use_dma_rx	= 0,		/* DBGU not capable of receive DMA */
-	.regs		= (void __iomem *)(AT91_VA_BASE_SYS + AT91_DBGU),
 };
 
 static u64 dbgu_dmamask = DMA_BIT_MASK(32);
diff --git a/arch/arm/mach-at91/at91sam9261_devices.c b/arch/arm/mach-at91/at91sam9261_devices.c
index 3b70b389..a178b58 100644
--- a/arch/arm/mach-at91/at91sam9261_devices.c
+++ b/arch/arm/mach-at91/at91sam9261_devices.c
@@ -87,7 +87,7 @@
  *  USB Device (Gadget)
  * -------------------------------------------------------------------- */
 
-#ifdef CONFIG_USB_GADGET_AT91
+#ifdef CONFIG_USB_AT91
 static struct at91_udc_data udc_data;
 
 static struct resource udc_resources[] = {
@@ -816,8 +816,8 @@
 #if defined(CONFIG_SERIAL_ATMEL)
 static struct resource dbgu_resources[] = {
 	[0] = {
-		.start	= AT91_VA_BASE_SYS + AT91_DBGU,
-		.end	= AT91_VA_BASE_SYS + AT91_DBGU + SZ_512 - 1,
+		.start	= AT91_BASE_SYS + AT91_DBGU,
+		.end	= AT91_BASE_SYS + AT91_DBGU + SZ_512 - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
@@ -830,7 +830,6 @@
 static struct atmel_uart_data dbgu_data = {
 	.use_dma_tx	= 0,
 	.use_dma_rx	= 0,		/* DBGU not capable of receive DMA */
-	.regs		= (void __iomem *)(AT91_VA_BASE_SYS + AT91_DBGU),
 };
 
 static u64 dbgu_dmamask = DMA_BIT_MASK(32);
diff --git a/arch/arm/mach-at91/at91sam9263_devices.c b/arch/arm/mach-at91/at91sam9263_devices.c
index 3faa1fd..d5fbac9 100644
--- a/arch/arm/mach-at91/at91sam9263_devices.c
+++ b/arch/arm/mach-at91/at91sam9263_devices.c
@@ -92,7 +92,7 @@
  *  USB Device (Gadget)
  * -------------------------------------------------------------------- */
 
-#ifdef CONFIG_USB_GADGET_AT91
+#ifdef CONFIG_USB_AT91
 static struct at91_udc_data udc_data;
 
 static struct resource udc_resources[] = {
@@ -1196,8 +1196,8 @@
 
 static struct resource dbgu_resources[] = {
 	[0] = {
-		.start	= AT91_VA_BASE_SYS + AT91_DBGU,
-		.end	= AT91_VA_BASE_SYS + AT91_DBGU + SZ_512 - 1,
+		.start	= AT91_BASE_SYS + AT91_DBGU,
+		.end	= AT91_BASE_SYS + AT91_DBGU + SZ_512 - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
@@ -1210,7 +1210,6 @@
 static struct atmel_uart_data dbgu_data = {
 	.use_dma_tx	= 0,
 	.use_dma_rx	= 0,		/* DBGU not capable of receive DMA */
-	.regs		= (void __iomem *)(AT91_VA_BASE_SYS + AT91_DBGU),
 };
 
 static u64 dbgu_dmamask = DMA_BIT_MASK(32);
diff --git a/arch/arm/mach-at91/at91sam9g45_devices.c b/arch/arm/mach-at91/at91sam9g45_devices.c
index 000b5e1..09a16d6 100644
--- a/arch/arm/mach-at91/at91sam9g45_devices.c
+++ b/arch/arm/mach-at91/at91sam9g45_devices.c
@@ -197,7 +197,7 @@
  *  USB HS Device (Gadget)
  * -------------------------------------------------------------------- */
 
-#if defined(CONFIG_USB_GADGET_ATMEL_USBA) || defined(CONFIG_USB_GADGET_ATMEL_USBA_MODULE)
+#if defined(CONFIG_USB_ATMEL_USBA) || defined(CONFIG_USB_ATMEL_USBA_MODULE)
 static struct resource usba_udc_resources[] = {
 	[0] = {
 		.start	= AT91SAM9G45_UDPHS_FIFO,
@@ -1332,8 +1332,8 @@
 #if defined(CONFIG_SERIAL_ATMEL)
 static struct resource dbgu_resources[] = {
 	[0] = {
-		.start	= AT91_VA_BASE_SYS + AT91_DBGU,
-		.end	= AT91_VA_BASE_SYS + AT91_DBGU + SZ_512 - 1,
+		.start	= AT91_BASE_SYS + AT91_DBGU,
+		.end	= AT91_BASE_SYS + AT91_DBGU + SZ_512 - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
@@ -1346,7 +1346,6 @@
 static struct atmel_uart_data dbgu_data = {
 	.use_dma_tx	= 0,
 	.use_dma_rx	= 0,
-	.regs		= (void __iomem *)(AT91_VA_BASE_SYS + AT91_DBGU),
 };
 
 static u64 dbgu_dmamask = DMA_BIT_MASK(32);
diff --git a/arch/arm/mach-at91/at91sam9rl_devices.c b/arch/arm/mach-at91/at91sam9rl_devices.c
index 305a851..628eb56 100644
--- a/arch/arm/mach-at91/at91sam9rl_devices.c
+++ b/arch/arm/mach-at91/at91sam9rl_devices.c
@@ -75,7 +75,7 @@
  *  USB HS Device (Gadget)
  * -------------------------------------------------------------------- */
 
-#if defined(CONFIG_USB_GADGET_ATMEL_USBA) || defined(CONFIG_USB_GADGET_ATMEL_USBA_MODULE)
+#if defined(CONFIG_USB_ATMEL_USBA) || defined(CONFIG_USB_ATMEL_USBA_MODULE)
 
 static struct resource usba_udc_resources[] = {
 	[0] = {
@@ -908,8 +908,8 @@
 #if defined(CONFIG_SERIAL_ATMEL)
 static struct resource dbgu_resources[] = {
 	[0] = {
-		.start	= AT91_VA_BASE_SYS + AT91_DBGU,
-		.end	= AT91_VA_BASE_SYS + AT91_DBGU + SZ_512 - 1,
+		.start	= AT91_BASE_SYS + AT91_DBGU,
+		.end	= AT91_BASE_SYS + AT91_DBGU + SZ_512 - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	[1] = {
@@ -922,7 +922,6 @@
 static struct atmel_uart_data dbgu_data = {
 	.use_dma_tx	= 0,
 	.use_dma_rx	= 0,		/* DBGU not capable of receive DMA */
-	.regs		= (void __iomem *)(AT91_VA_BASE_SYS + AT91_DBGU),
 };
 
 static u64 dbgu_dmamask = DMA_BIT_MASK(32);
diff --git a/arch/arm/mach-at91/board-yl-9200.c b/arch/arm/mach-at91/board-yl-9200.c
index 649b052..12a3f95 100644
--- a/arch/arm/mach-at91/board-yl-9200.c
+++ b/arch/arm/mach-at91/board-yl-9200.c
@@ -384,7 +384,7 @@
 #include <video/s1d13xxxfb.h>
 
 
-static void __init yl9200_init_video(void)
+static void yl9200_init_video(void)
 {
 	/* NWAIT Signal */
 	at91_set_A_periph(AT91_PIN_PC6, 0);
diff --git a/arch/arm/mach-at91/include/mach/system_rev.h b/arch/arm/mach-at91/include/mach/system_rev.h
index 8f48660..ec164a4 100644
--- a/arch/arm/mach-at91/include/mach/system_rev.h
+++ b/arch/arm/mach-at91/include/mach/system_rev.h
@@ -19,7 +19,7 @@
 #define BOARD_HAVE_NAND_16BIT	(1 << 31)
 static inline int board_have_nand_16bit(void)
 {
-	return system_rev & BOARD_HAVE_NAND_16BIT;
+	return (system_rev & BOARD_HAVE_NAND_16BIT) ? 1 : 0;
 }
 
 #endif /* __ARCH_SYSTEM_REV_H__ */
diff --git a/arch/arm/mach-at91/include/mach/vmalloc.h b/arch/arm/mach-at91/include/mach/vmalloc.h
index 8eb459f..8e4a1bd 100644
--- a/arch/arm/mach-at91/include/mach/vmalloc.h
+++ b/arch/arm/mach-at91/include/mach/vmalloc.h
@@ -21,6 +21,8 @@
 #ifndef __ASM_ARCH_VMALLOC_H
 #define __ASM_ARCH_VMALLOC_H
 
+#include <mach/hardware.h>
+
 #define VMALLOC_END		(AT91_VIRT_BASE & PGDIR_MASK)
 
 #endif
diff --git a/arch/arm/mach-bcmring/core.c b/arch/arm/mach-bcmring/core.c
index 43eadbcc..430da12 100644
--- a/arch/arm/mach-bcmring/core.c
+++ b/arch/arm/mach-bcmring/core.c
@@ -235,7 +235,7 @@
 	 */
 	bcmring_clocksource_init();
 
-	sp804_clockevents_register(TIMER0_VA_BASE, IRQ_TIMER0, "timer0");
+	sp804_clockevents_init(TIMER0_VA_BASE, IRQ_TIMER0, "timer0");
 }
 
 struct sys_timer bcmring_timer = {
diff --git a/arch/arm/mach-bcmring/dma.c b/arch/arm/mach-bcmring/dma.c
index b52b8de..f4d4d6d 100644
--- a/arch/arm/mach-bcmring/dma.c
+++ b/arch/arm/mach-bcmring/dma.c
@@ -36,6 +36,7 @@
 #include <linux/mm.h>
 #include <linux/pfn.h>
 #include <linux/atomic.h>
+#include <linux/sched.h>
 #include <mach/dma.h>
 
 /* I don't quite understand why dc4 fails when this is set to 1 and DMA is enabled */
diff --git a/arch/arm/mach-davinci/board-da850-evm.c b/arch/arm/mach-davinci/board-da850-evm.c
index 1d7d249..6659a90 100644
--- a/arch/arm/mach-davinci/board-da850-evm.c
+++ b/arch/arm/mach-davinci/board-da850-evm.c
@@ -753,7 +753,7 @@
 	.num_serializer	= ARRAY_SIZE(da850_iis_serializer_direction),
 	.tdm_slots	= 2,
 	.serial_dir	= da850_iis_serializer_direction,
-	.asp_chan_q	= EVENTQ_1,
+	.asp_chan_q	= EVENTQ_0,
 	.version	= MCASP_VERSION_2,
 	.txnumevt	= 1,
 	.rxnumevt	= 1,
diff --git a/arch/arm/mach-davinci/board-dm365-evm.c b/arch/arm/mach-davinci/board-dm365-evm.c
index 1918ae7..46e1f41 100644
--- a/arch/arm/mach-davinci/board-dm365-evm.c
+++ b/arch/arm/mach-davinci/board-dm365-evm.c
@@ -107,7 +107,7 @@
 		/* UBL (a few copies) plus U-Boot */
 		.name		= "bootloader",
 		.offset		= 0,
-		.size		= 28 * NAND_BLOCK_SIZE,
+		.size		= 30 * NAND_BLOCK_SIZE,
 		.mask_flags	= MTD_WRITEABLE, /* force read-only */
 	}, {
 		/* U-Boot environment */
diff --git a/arch/arm/mach-davinci/board-dm646x-evm.c b/arch/arm/mach-davinci/board-dm646x-evm.c
index e574d7f8..635bf77 100644
--- a/arch/arm/mach-davinci/board-dm646x-evm.c
+++ b/arch/arm/mach-davinci/board-dm646x-evm.c
@@ -564,7 +564,7 @@
 	int val;
 	u32 value;
 
-	if (!vpif_vsclkdis_reg || !cpld_client)
+	if (!vpif_vidclkctl_reg || !cpld_client)
 		return -ENXIO;
 
 	val = i2c_smbus_read_byte(cpld_client);
@@ -572,7 +572,7 @@
 		return val;
 
 	spin_lock_irqsave(&vpif_reg_lock, flags);
-	value = __raw_readl(vpif_vsclkdis_reg);
+	value = __raw_readl(vpif_vidclkctl_reg);
 	if (mux_mode) {
 		val &= VPIF_INPUT_TWO_CHANNEL;
 		value |= VIDCH1CLK;
@@ -580,7 +580,7 @@
 		val |= VPIF_INPUT_ONE_CHANNEL;
 		value &= ~VIDCH1CLK;
 	}
-	__raw_writel(value, vpif_vsclkdis_reg);
+	__raw_writel(value, vpif_vidclkctl_reg);
 	spin_unlock_irqrestore(&vpif_reg_lock, flags);
 
 	err = i2c_smbus_write_byte(cpld_client, val);
diff --git a/arch/arm/mach-davinci/dm646x.c b/arch/arm/mach-davinci/dm646x.c
index 0b68ed5..af27c13 100644
--- a/arch/arm/mach-davinci/dm646x.c
+++ b/arch/arm/mach-davinci/dm646x.c
@@ -161,7 +161,6 @@
 	.name = "dsp",
 	.parent = &pll1_sysclk1,
 	.lpsc = DM646X_LPSC_C64X_CPU,
-	.flags = PSC_DSP,
 	.usecount = 1,			/* REVISIT how to disable? */
 };
 
diff --git a/arch/arm/mach-davinci/include/mach/psc.h b/arch/arm/mach-davinci/include/mach/psc.h
index fa59c09..8bc3fc2 100644
--- a/arch/arm/mach-davinci/include/mach/psc.h
+++ b/arch/arm/mach-davinci/include/mach/psc.h
@@ -233,7 +233,7 @@
 #define PTCMD		0x120
 #define PTSTAT		0x128
 #define PDSTAT		0x200
-#define PDCTL1		0x304
+#define PDCTL		0x300
 #define MDSTAT		0x800
 #define MDCTL		0xA00
 
@@ -244,7 +244,10 @@
 #define PSC_STATE_ENABLE	3
 
 #define MDSTAT_STATE_MASK	0x3f
+#define PDSTAT_STATE_MASK	0x1f
 #define MDCTL_FORCE		BIT(31)
+#define PDCTL_NEXT		BIT(1)
+#define PDCTL_EPCGOOD		BIT(8)
 
 #ifndef __ASSEMBLER__
 
diff --git a/arch/arm/mach-davinci/psc.c b/arch/arm/mach-davinci/psc.c
index 1fb6bdf..d7e210f 100644
--- a/arch/arm/mach-davinci/psc.c
+++ b/arch/arm/mach-davinci/psc.c
@@ -52,7 +52,7 @@
 void davinci_psc_config(unsigned int domain, unsigned int ctlr,
 		unsigned int id, bool enable, u32 flags)
 {
-	u32 epcpr, ptcmd, ptstat, pdstat, pdctl1, mdstat, mdctl;
+	u32 epcpr, ptcmd, ptstat, pdstat, pdctl, mdstat, mdctl;
 	void __iomem *psc_base;
 	struct davinci_soc_info *soc_info = &davinci_soc_info;
 	u32 next_state = PSC_STATE_ENABLE;
@@ -79,11 +79,11 @@
 		mdctl |= MDCTL_FORCE;
 	__raw_writel(mdctl, psc_base + MDCTL + 4 * id);
 
-	pdstat = __raw_readl(psc_base + PDSTAT);
-	if ((pdstat & 0x00000001) == 0) {
-		pdctl1 = __raw_readl(psc_base + PDCTL1);
-		pdctl1 |= 0x1;
-		__raw_writel(pdctl1, psc_base + PDCTL1);
+	pdstat = __raw_readl(psc_base + PDSTAT + 4 * domain);
+	if ((pdstat & PDSTAT_STATE_MASK) == 0) {
+		pdctl = __raw_readl(psc_base + PDCTL + 4 * domain);
+		pdctl |= PDCTL_NEXT;
+		__raw_writel(pdctl, psc_base + PDCTL + 4 * domain);
 
 		ptcmd = 1 << domain;
 		__raw_writel(ptcmd, psc_base + PTCMD);
@@ -92,9 +92,9 @@
 			epcpr = __raw_readl(psc_base + EPCPR);
 		} while ((((epcpr >> domain) & 1) == 0));
 
-		pdctl1 = __raw_readl(psc_base + PDCTL1);
-		pdctl1 |= 0x100;
-		__raw_writel(pdctl1, psc_base + PDCTL1);
+		pdctl = __raw_readl(psc_base + PDCTL + 4 * domain);
+		pdctl |= PDCTL_EPCGOOD;
+		__raw_writel(pdctl, psc_base + PDCTL + 4 * domain);
 	} else {
 		ptcmd = 1 << domain;
 		__raw_writel(ptcmd, psc_base + PTCMD);
diff --git a/arch/arm/mach-exynos/cpuidle.c b/arch/arm/mach-exynos/cpuidle.c
index 35f6502..4ebb382 100644
--- a/arch/arm/mach-exynos/cpuidle.c
+++ b/arch/arm/mach-exynos/cpuidle.c
@@ -12,6 +12,8 @@
 #include <linux/init.h>
 #include <linux/cpuidle.h>
 #include <linux/io.h>
+#include <linux/export.h>
+#include <linux/time.h>
 
 #include <asm/proc-fns.h>
 
diff --git a/arch/arm/mach-exynos/mct.c b/arch/arm/mach-exynos/mct.c
index 97343df..85b5527 100644
--- a/arch/arm/mach-exynos/mct.c
+++ b/arch/arm/mach-exynos/mct.c
@@ -44,8 +44,6 @@
 	char name[10];
 };
 
-static DEFINE_PER_CPU(struct mct_clock_event_device, percpu_mct_tick);
-
 static void exynos4_mct_write(unsigned int value, void *addr)
 {
 	void __iomem *stat_addr;
@@ -264,6 +262,9 @@
 }
 
 #ifdef CONFIG_LOCAL_TIMERS
+
+static DEFINE_PER_CPU(struct mct_clock_event_device, percpu_mct_tick);
+
 /* Clock event handling */
 static void exynos4_mct_tick_stop(struct mct_clock_event_device *mevt)
 {
@@ -428,9 +429,13 @@
 
 void local_timer_stop(struct clock_event_device *evt)
 {
+	unsigned int cpu = smp_processor_id();
 	evt->set_mode(CLOCK_EVT_MODE_UNUSED, evt);
 	if (mct_int_type == MCT_INT_SPI)
-		disable_irq(evt->irq);
+		if (cpu == 0)
+			remove_irq(evt->irq, &mct_tick0_event_irq);
+		else
+			remove_irq(evt->irq, &mct_tick1_event_irq);
 	else
 		disable_percpu_irq(IRQ_MCT_LOCALTIMER);
 }
@@ -443,6 +448,7 @@
 
 	clk_rate = clk_get_rate(mct_clk);
 
+#ifdef CONFIG_LOCAL_TIMERS
 	if (mct_int_type == MCT_INT_PPI) {
 		int err;
 
@@ -452,6 +458,7 @@
 		WARN(err, "MCT: can't request IRQ %d (%d)\n",
 		     IRQ_MCT_LOCALTIMER, err);
 	}
+#endif /* CONFIG_LOCAL_TIMERS */
 }
 
 static void __init exynos4_timer_init(void)
diff --git a/arch/arm/mach-highbank/highbank.c b/arch/arm/mach-highbank/highbank.c
index b82dcf0..88660d5 100644
--- a/arch/arm/mach-highbank/highbank.c
+++ b/arch/arm/mach-highbank/highbank.c
@@ -22,6 +22,7 @@
 #include <linux/of_irq.h>
 #include <linux/of_platform.h>
 #include <linux/of_address.h>
+#include <linux/smp.h>
 
 #include <asm/cacheflush.h>
 #include <asm/unified.h>
@@ -72,6 +73,9 @@
 
 void highbank_set_cpu_jump(int cpu, void *jump_addr)
 {
+#ifdef CONFIG_SMP
+	cpu = cpu_logical_map(cpu);
+#endif
 	writel(BSYM(virt_to_phys(jump_addr)), HB_JUMP_TABLE_VIRT(cpu));
 	__cpuc_flush_dcache_area(HB_JUMP_TABLE_VIRT(cpu), 16);
 	outer_clean_range(HB_JUMP_TABLE_PHYS(cpu),
diff --git a/arch/arm/mach-imx/Kconfig b/arch/arm/mach-imx/Kconfig
index 5f7f9c2..c44aa97 100644
--- a/arch/arm/mach-imx/Kconfig
+++ b/arch/arm/mach-imx/Kconfig
@@ -10,11 +10,6 @@
 config HAVE_IMX_SRC
 	bool
 
-#
-# ARCH_MX31 and ARCH_MX35 are left for compatibility
-# Some usages assume that having one of them implies not having (e.g.) ARCH_MX2.
-# To easily distinguish good and reviewed from unreviewed usages new (and IMHO
-# more sensible) names are used: SOC_IMX31 and SOC_IMX35
 config ARCH_MX1
 	bool
 
@@ -27,12 +22,6 @@
 config MACH_MX27
 	bool
 
-config ARCH_MX31
-	bool
-
-config ARCH_MX35
-	bool
-
 config SOC_IMX1
 	bool
 	select ARCH_MX1
@@ -72,7 +61,6 @@
 	select CPU_V6
 	select IMX_HAVE_PLATFORM_MXC_RNGA
 	select ARCH_MXC_AUDMUX_V2
-	select ARCH_MX31
 	select MXC_AVIC
 	select SMP_ON_UP if SMP
 
@@ -82,7 +70,6 @@
 	select ARCH_MXC_IOMUX_V3
 	select ARCH_MXC_AUDMUX_V2
 	select HAVE_EPIT
-	select ARCH_MX35
 	select MXC_AVIC
 	select SMP_ON_UP if SMP
 
diff --git a/arch/arm/mach-imx/Makefile.boot b/arch/arm/mach-imx/Makefile.boot
index 22d8588..cfede57 100644
--- a/arch/arm/mach-imx/Makefile.boot
+++ b/arch/arm/mach-imx/Makefile.boot
@@ -1,22 +1,26 @@
-zreladdr-$(CONFIG_ARCH_MX1)	+= 0x08008000
-params_phys-$(CONFIG_ARCH_MX1)	:= 0x08000100
-initrd_phys-$(CONFIG_ARCH_MX1)	:= 0x08800000
+zreladdr-$(CONFIG_SOC_IMX1)	+= 0x08008000
+params_phys-$(CONFIG_SOC_IMX1)	:= 0x08000100
+initrd_phys-$(CONFIG_SOC_IMX1)	:= 0x08800000
 
-zreladdr-$(CONFIG_MACH_MX21)	+= 0xC0008000
-params_phys-$(CONFIG_MACH_MX21)	:= 0xC0000100
-initrd_phys-$(CONFIG_MACH_MX21)	:= 0xC0800000
+zreladdr-$(CONFIG_SOC_IMX21)	+= 0xC0008000
+params_phys-$(CONFIG_SOC_IMX21)	:= 0xC0000100
+initrd_phys-$(CONFIG_SOC_IMX21)	:= 0xC0800000
 
-zreladdr-$(CONFIG_ARCH_MX25)	+= 0x80008000
-params_phys-$(CONFIG_ARCH_MX25)	:= 0x80000100
-initrd_phys-$(CONFIG_ARCH_MX25)	:= 0x80800000
+zreladdr-$(CONFIG_SOC_IMX25)	+= 0x80008000
+params_phys-$(CONFIG_SOC_IMX25)	:= 0x80000100
+initrd_phys-$(CONFIG_SOC_IMX25)	:= 0x80800000
 
-zreladdr-$(CONFIG_MACH_MX27)	+= 0xA0008000
-params_phys-$(CONFIG_MACH_MX27)	:= 0xA0000100
-initrd_phys-$(CONFIG_MACH_MX27)	:= 0xA0800000
+zreladdr-$(CONFIG_SOC_IMX27)	+= 0xA0008000
+params_phys-$(CONFIG_SOC_IMX27)	:= 0xA0000100
+initrd_phys-$(CONFIG_SOC_IMX27)	:= 0xA0800000
 
-zreladdr-$(CONFIG_ARCH_MX3)	+= 0x80008000
-params_phys-$(CONFIG_ARCH_MX3)	:= 0x80000100
-initrd_phys-$(CONFIG_ARCH_MX3)	:= 0x80800000
+zreladdr-$(CONFIG_SOC_IMX31)	+= 0x80008000
+params_phys-$(CONFIG_SOC_IMX31)	:= 0x80000100
+initrd_phys-$(CONFIG_SOC_IMX31)	:= 0x80800000
+
+zreladdr-$(CONFIG_SOC_IMX35)	+= 0x80008000
+params_phys-$(CONFIG_SOC_IMX35)	:= 0x80000100
+initrd_phys-$(CONFIG_SOC_IMX35)	:= 0x80800000
 
 zreladdr-$(CONFIG_SOC_IMX6Q)	+= 0x10008000
 params_phys-$(CONFIG_SOC_IMX6Q)	:= 0x10000100
diff --git a/arch/arm/mach-imx/clock-imx6q.c b/arch/arm/mach-imx/clock-imx6q.c
index e0b926d..039a7ab 100644
--- a/arch/arm/mach-imx/clock-imx6q.c
+++ b/arch/arm/mach-imx/clock-imx6q.c
@@ -1139,7 +1139,7 @@
 		return -EINVAL;
 
 	max_div = ((d->bm_pred >> d->bp_pred) + 1) *
-		  ((d->bm_pred >> d->bp_pred) + 1);
+		  ((d->bm_podf >> d->bp_podf) + 1);
 
 	div = parent_rate / rate;
 	if (div == 0)
@@ -1953,14 +1953,17 @@
 	imx_map_entry(MX6Q, ANATOP, MT_DEVICE),
 };
 
+void __init imx6q_clock_map_io(void)
+{
+	iotable_init(imx6q_clock_desc, ARRAY_SIZE(imx6q_clock_desc));
+}
+
 int __init mx6q_clocks_init(void)
 {
 	struct device_node *np;
 	void __iomem *base;
 	int i, irq;
 
-	iotable_init(imx6q_clock_desc, ARRAY_SIZE(imx6q_clock_desc));
-
 	/* retrieve the freqency of fixed clocks from device tree */
 	for_each_compatible_node(np, NULL, "fixed-clock") {
 		u32 rate;
@@ -2002,6 +2005,21 @@
 	clk_set_rate(&asrc_serial_clk, 1500000);
 	clk_set_rate(&enfc_clk, 11000000);
 
+	/*
+	 * Before pinctrl API is available, we have to rely on the pad
+	 * configuration set up by bootloader.  For usdhc example here,
+	 * u-boot sets up the pads for 49.5 MHz case, and we have to lower
+	 * the usdhc clock from 198 to 49.5 MHz to match the pad configuration.
+	 *
+	 * FIXME: This is should be removed after pinctrl API is available.
+	 * At that time, usdhc driver can call pinctrl API to change pad
+	 * configuration dynamically per different usdhc clock settings.
+	 */
+	clk_set_rate(&usdhc1_clk, 49500000);
+	clk_set_rate(&usdhc2_clk, 49500000);
+	clk_set_rate(&usdhc3_clk, 49500000);
+	clk_set_rate(&usdhc4_clk, 49500000);
+
 	np = of_find_compatible_node(NULL, NULL, "fsl,imx6q-gpt");
 	base = of_iomap(np, 0);
 	WARN_ON(!base);
diff --git a/arch/arm/mach-imx/mach-imx6q.c b/arch/arm/mach-imx/mach-imx6q.c
index 8bf5fa3..8deb012 100644
--- a/arch/arm/mach-imx/mach-imx6q.c
+++ b/arch/arm/mach-imx/mach-imx6q.c
@@ -34,16 +34,18 @@
 {
 	imx_lluart_map_io();
 	imx_scu_map_io();
+	imx6q_clock_map_io();
 }
 
-static void __init imx6q_gpio_add_irq_domain(struct device_node *np,
+static int __init imx6q_gpio_add_irq_domain(struct device_node *np,
 				struct device_node *interrupt_parent)
 {
-	static int gpio_irq_base = MXC_GPIO_IRQ_START + ARCH_NR_GPIOS -
-				   32 * 7; /* imx6q gets 7 gpio ports */
+	static int gpio_irq_base = MXC_GPIO_IRQ_START + ARCH_NR_GPIOS;
 
+	gpio_irq_base -= 32;
 	irq_domain_add_simple(np, gpio_irq_base);
-	gpio_irq_base += 32;
+
+	return 0;
 }
 
 static const struct of_device_id imx6q_irq_match[] __initconst = {
diff --git a/arch/arm/mach-imx/mm-imx3.c b/arch/arm/mach-imx/mm-imx3.c
index 9f0e82e..31807d2 100644
--- a/arch/arm/mach-imx/mm-imx3.c
+++ b/arch/arm/mach-imx/mm-imx3.c
@@ -33,29 +33,32 @@
 static void imx3_idle(void)
 {
 	unsigned long reg = 0;
-	__asm__ __volatile__(
-		/* disable I and D cache */
-		"mrc p15, 0, %0, c1, c0, 0\n"
-		"bic %0, %0, #0x00001000\n"
-		"bic %0, %0, #0x00000004\n"
-		"mcr p15, 0, %0, c1, c0, 0\n"
-		/* invalidate I cache */
-		"mov %0, #0\n"
-		"mcr p15, 0, %0, c7, c5, 0\n"
-		/* clear and invalidate D cache */
-		"mov %0, #0\n"
-		"mcr p15, 0, %0, c7, c14, 0\n"
-		/* WFI */
-		"mov %0, #0\n"
-		"mcr p15, 0, %0, c7, c0, 4\n"
-		"nop\n" "nop\n" "nop\n" "nop\n"
-		"nop\n" "nop\n" "nop\n"
-		/* enable I and D cache */
-		"mrc p15, 0, %0, c1, c0, 0\n"
-		"orr %0, %0, #0x00001000\n"
-		"orr %0, %0, #0x00000004\n"
-		"mcr p15, 0, %0, c1, c0, 0\n"
-		: "=r" (reg));
+
+	if (!need_resched())
+		__asm__ __volatile__(
+			/* disable I and D cache */
+			"mrc p15, 0, %0, c1, c0, 0\n"
+			"bic %0, %0, #0x00001000\n"
+			"bic %0, %0, #0x00000004\n"
+			"mcr p15, 0, %0, c1, c0, 0\n"
+			/* invalidate I cache */
+			"mov %0, #0\n"
+			"mcr p15, 0, %0, c7, c5, 0\n"
+			/* clear and invalidate D cache */
+			"mov %0, #0\n"
+			"mcr p15, 0, %0, c7, c14, 0\n"
+			/* WFI */
+			"mov %0, #0\n"
+			"mcr p15, 0, %0, c7, c0, 4\n"
+			"nop\n" "nop\n" "nop\n" "nop\n"
+			"nop\n" "nop\n" "nop\n"
+			/* enable I and D cache */
+			"mrc p15, 0, %0, c1, c0, 0\n"
+			"orr %0, %0, #0x00001000\n"
+			"orr %0, %0, #0x00000004\n"
+			"mcr p15, 0, %0, c1, c0, 0\n"
+			: "=r" (reg));
+	local_irq_enable();
 }
 
 static void __iomem *imx3_ioremap(unsigned long phys_addr, size_t size,
@@ -108,6 +111,7 @@
 	l2x0_init(l2x0_base, 0x00030024, 0x00000000);
 }
 
+#ifdef CONFIG_SOC_IMX31
 static struct map_desc mx31_io_desc[] __initdata = {
 	imx_map_entry(MX31, X_MEMC, MT_DEVICE),
 	imx_map_entry(MX31, AVIC, MT_DEVICE_NONSHARED),
@@ -126,33 +130,11 @@
 	iotable_init(mx31_io_desc, ARRAY_SIZE(mx31_io_desc));
 }
 
-static struct map_desc mx35_io_desc[] __initdata = {
-	imx_map_entry(MX35, X_MEMC, MT_DEVICE),
-	imx_map_entry(MX35, AVIC, MT_DEVICE_NONSHARED),
-	imx_map_entry(MX35, AIPS1, MT_DEVICE_NONSHARED),
-	imx_map_entry(MX35, AIPS2, MT_DEVICE_NONSHARED),
-	imx_map_entry(MX35, SPBA0, MT_DEVICE_NONSHARED),
-};
-
-void __init mx35_map_io(void)
-{
-	iotable_init(mx35_io_desc, ARRAY_SIZE(mx35_io_desc));
-}
-
 void __init imx31_init_early(void)
 {
 	mxc_set_cpu_type(MXC_CPU_MX31);
 	mxc_arch_reset_init(MX31_IO_ADDRESS(MX31_WDOG_BASE_ADDR));
-	imx_idle = imx3_idle;
-	imx_ioremap = imx3_ioremap;
-}
-
-void __init imx35_init_early(void)
-{
-	mxc_set_cpu_type(MXC_CPU_MX35);
-	mxc_iomux_v3_init(MX35_IO_ADDRESS(MX35_IOMUXC_BASE_ADDR));
-	mxc_arch_reset_init(MX35_IO_ADDRESS(MX35_WDOG_BASE_ADDR));
-	imx_idle = imx3_idle;
+	pm_idle = imx3_idle;
 	imx_ioremap = imx3_ioremap;
 }
 
@@ -161,11 +143,6 @@
 	mxc_init_irq(MX31_IO_ADDRESS(MX31_AVIC_BASE_ADDR));
 }
 
-void __init mx35_init_irq(void)
-{
-	mxc_init_irq(MX35_IO_ADDRESS(MX35_AVIC_BASE_ADDR));
-}
-
 static struct sdma_script_start_addrs imx31_to1_sdma_script __initdata = {
 	.per_2_per_addr = 1677,
 };
@@ -199,6 +176,35 @@
 
 	imx_add_imx_sdma("imx31-sdma", MX31_SDMA_BASE_ADDR, MX31_INT_SDMA, &imx31_sdma_pdata);
 }
+#endif /* ifdef CONFIG_SOC_IMX31 */
+
+#ifdef CONFIG_SOC_IMX35
+static struct map_desc mx35_io_desc[] __initdata = {
+	imx_map_entry(MX35, X_MEMC, MT_DEVICE),
+	imx_map_entry(MX35, AVIC, MT_DEVICE_NONSHARED),
+	imx_map_entry(MX35, AIPS1, MT_DEVICE_NONSHARED),
+	imx_map_entry(MX35, AIPS2, MT_DEVICE_NONSHARED),
+	imx_map_entry(MX35, SPBA0, MT_DEVICE_NONSHARED),
+};
+
+void __init mx35_map_io(void)
+{
+	iotable_init(mx35_io_desc, ARRAY_SIZE(mx35_io_desc));
+}
+
+void __init imx35_init_early(void)
+{
+	mxc_set_cpu_type(MXC_CPU_MX35);
+	mxc_iomux_v3_init(MX35_IO_ADDRESS(MX35_IOMUXC_BASE_ADDR));
+	mxc_arch_reset_init(MX35_IO_ADDRESS(MX35_WDOG_BASE_ADDR));
+	pm_idle = imx3_idle;
+	imx_ioremap = imx3_ioremap;
+}
+
+void __init mx35_init_irq(void)
+{
+	mxc_init_irq(MX35_IO_ADDRESS(MX35_AVIC_BASE_ADDR));
+}
 
 static struct sdma_script_start_addrs imx35_to1_sdma_script __initdata = {
 	.ap_2_ap_addr = 642,
@@ -254,3 +260,4 @@
 
 	imx_add_imx_sdma("imx35-sdma", MX35_SDMA_BASE_ADDR, MX35_INT_SDMA, &imx35_sdma_pdata);
 }
+#endif /* ifdef CONFIG_SOC_IMX35 */
diff --git a/arch/arm/mach-imx/src.c b/arch/arm/mach-imx/src.c
index 36cacbd..a8e3368 100644
--- a/arch/arm/mach-imx/src.c
+++ b/arch/arm/mach-imx/src.c
@@ -14,6 +14,7 @@
 #include <linux/io.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
+#include <linux/smp.h>
 #include <asm/unified.h>
 
 #define SRC_SCR				0x000
@@ -23,10 +24,15 @@
 
 static void __iomem *src_base;
 
+#ifndef CONFIG_SMP
+#define cpu_logical_map(cpu)		0
+#endif
+
 void imx_enable_cpu(int cpu, bool enable)
 {
 	u32 mask, val;
 
+	cpu = cpu_logical_map(cpu);
 	mask = 1 << (BP_SRC_SCR_CORE1_ENABLE + cpu - 1);
 	val = readl_relaxed(src_base + SRC_SCR);
 	val = enable ? val | mask : val & ~mask;
@@ -35,6 +41,7 @@
 
 void imx_set_cpu_jump(int cpu, void *jump_addr)
 {
+	cpu = cpu_logical_map(cpu);
 	writel_relaxed(BSYM(virt_to_phys(jump_addr)),
 		       src_base + SRC_GPR1 + cpu * 8);
 }
diff --git a/arch/arm/mach-mmp/gplugd.c b/arch/arm/mach-mmp/gplugd.c
index 6915656..4665767 100644
--- a/arch/arm/mach-mmp/gplugd.c
+++ b/arch/arm/mach-mmp/gplugd.c
@@ -182,7 +182,7 @@
 
 	/* on-chip devices */
 	pxa168_add_uart(3);
-	pxa168_add_ssp(0);
+	pxa168_add_ssp(1);
 	pxa168_add_twsi(0, NULL, ARRAY_AND_SIZE(gplugd_i2c_board_info));
 
 	pxa168_add_eth(&gplugd_eth_platform_data);
diff --git a/arch/arm/mach-mmp/include/mach/gpio-pxa.h b/arch/arm/mach-mmp/include/mach/gpio-pxa.h
index d14eeaf..99b4ce1 100644
--- a/arch/arm/mach-mmp/include/mach/gpio-pxa.h
+++ b/arch/arm/mach-mmp/include/mach/gpio-pxa.h
@@ -7,7 +7,7 @@
 #define GPIO_REGS_VIRT	(APB_VIRT_BASE + 0x19000)
 
 #define BANK_OFF(n)	(((n) < 3) ? (n) << 2 : 0x100 + (((n) - 3) << 2))
-#define GPIO_REG(x)	(GPIO_REGS_VIRT + (x))
+#define GPIO_REG(x)	(*(volatile u32 *)(GPIO_REGS_VIRT + (x)))
 
 #define NR_BUILTIN_GPIO		IRQ_GPIO_NUM
 
diff --git a/arch/arm/mach-msm/Makefile b/arch/arm/mach-msm/Makefile
index 4285dfd..4ad3969 100644
--- a/arch/arm/mach-msm/Makefile
+++ b/arch/arm/mach-msm/Makefile
@@ -15,6 +15,8 @@
 obj-$(CONFIG_MSM_SMD) += last_radio_log.o
 obj-$(CONFIG_MSM_SCM) += scm.o scm-boot.o
 
+CFLAGS_scm.o :=$(call as-instr,.arch_extension sec,-DREQUIRES_SEC=1)
+
 obj-$(CONFIG_HOTPLUG_CPU) += hotplug.o
 obj-$(CONFIG_SMP) += headsmp.o platsmp.o
 
diff --git a/arch/arm/mach-msm/board-msm7x30.c b/arch/arm/mach-msm/board-msm7x30.c
index 71de506..db81ed53 100644
--- a/arch/arm/mach-msm/board-msm7x30.c
+++ b/arch/arm/mach-msm/board-msm7x30.c
@@ -42,8 +42,8 @@
 
 extern struct sys_timer msm_timer;
 
-static void __init msm7x30_fixup(struct machine_desc *desc, struct tag *tag,
-			 char **cmdline, struct meminfo *mi)
+static void __init msm7x30_fixup(struct tag *tag, char **cmdline,
+		struct meminfo *mi)
 {
 	for (; tag->hdr.size; tag = tag_next(tag))
 		if (tag->hdr.tag == ATAG_MEM && tag->u.mem.start == 0x200000) {
diff --git a/arch/arm/mach-msm/board-msm8960.c b/arch/arm/mach-msm/board-msm8960.c
index b04468e..6dc1cbd 100644
--- a/arch/arm/mach-msm/board-msm8960.c
+++ b/arch/arm/mach-msm/board-msm8960.c
@@ -32,8 +32,8 @@
 
 #include "devices.h"
 
-static void __init msm8960_fixup(struct machine_desc *desc, struct tag *tag,
-			 char **cmdline, struct meminfo *mi)
+static void __init msm8960_fixup(struct tag *tag, char **cmdline,
+		struct meminfo *mi)
 {
 	for (; tag->hdr.size; tag = tag_next(tag))
 		if (tag->hdr.tag == ATAG_MEM &&
diff --git a/arch/arm/mach-msm/board-msm8x60.c b/arch/arm/mach-msm/board-msm8x60.c
index cf38e22..44bf716 100644
--- a/arch/arm/mach-msm/board-msm8x60.c
+++ b/arch/arm/mach-msm/board-msm8x60.c
@@ -28,8 +28,8 @@
 #include <mach/board.h>
 #include <mach/msm_iomap.h>
 
-static void __init msm8x60_fixup(struct machine_desc *desc, struct tag *tag,
-			 char **cmdline, struct meminfo *mi)
+static void __init msm8x60_fixup(struct tag *tag, char **cmdline,
+		struct meminfo *mi)
 {
 	for (; tag->hdr.size; tag = tag_next(tag))
 		if (tag->hdr.tag == ATAG_MEM &&
diff --git a/arch/arm/mach-msm/devices-iommu.c b/arch/arm/mach-msm/devices-iommu.c
index 24030d0..0fb7a17 100644
--- a/arch/arm/mach-msm/devices-iommu.c
+++ b/arch/arm/mach-msm/devices-iommu.c
@@ -18,6 +18,7 @@
 #include <linux/kernel.h>
 #include <linux/platform_device.h>
 #include <linux/bootmem.h>
+#include <linux/module.h>
 #include <mach/irqs.h>
 #include <mach/iommu.h>
 
diff --git a/arch/arm/mach-msm/scm.c b/arch/arm/mach-msm/scm.c
index 232f97a..bafabb5 100644
--- a/arch/arm/mach-msm/scm.c
+++ b/arch/arm/mach-msm/scm.c
@@ -180,6 +180,9 @@
 			__asmeq("%1", "r0")
 			__asmeq("%2", "r1")
 			__asmeq("%3", "r2")
+#ifdef REQUIRES_SEC
+			".arch_extension sec\n"
+#endif
 			"smc	#0	@ switch to secure world\n"
 			: "=r" (r0)
 			: "r" (r0), "r" (r1), "r" (r2)
diff --git a/arch/arm/mach-mx5/board-mx51_babbage.c b/arch/arm/mach-mx5/board-mx51_babbage.c
index 5c837603..24994bb 100644
--- a/arch/arm/mach-mx5/board-mx51_babbage.c
+++ b/arch/arm/mach-mx5/board-mx51_babbage.c
@@ -362,7 +362,7 @@
 {
 	iomux_v3_cfg_t usbh1stp = MX51_PAD_USBH1_STP__USBH1_STP;
 	iomux_v3_cfg_t power_key = NEW_PAD_CTRL(MX51_PAD_EIM_A27__GPIO2_21,
-		PAD_CTL_SRE_FAST | PAD_CTL_DSE_HIGH | PAD_CTL_PUS_100K_UP);
+		PAD_CTL_SRE_FAST | PAD_CTL_DSE_HIGH);
 
 	imx51_soc_init();
 
diff --git a/arch/arm/mach-mx5/board-mx53_evk.c b/arch/arm/mach-mx5/board-mx53_evk.c
index 6bea31a..64bbfce 100644
--- a/arch/arm/mach-mx5/board-mx53_evk.c
+++ b/arch/arm/mach-mx5/board-mx53_evk.c
@@ -106,7 +106,7 @@
 	gpio_set_value(MX53_EVK_FEC_PHY_RST, 1);
 }
 
-static struct fec_platform_data mx53_evk_fec_pdata = {
+static const struct fec_platform_data mx53_evk_fec_pdata __initconst = {
 	.phy = PHY_INTERFACE_MODE_RMII,
 };
 
diff --git a/arch/arm/mach-mx5/board-mx53_loco.c b/arch/arm/mach-mx5/board-mx53_loco.c
index 7678f77..237bdec 100644
--- a/arch/arm/mach-mx5/board-mx53_loco.c
+++ b/arch/arm/mach-mx5/board-mx53_loco.c
@@ -242,7 +242,7 @@
 	gpio_set_value(LOCO_FEC_PHY_RST, 1);
 }
 
-static struct fec_platform_data mx53_loco_fec_data = {
+static const struct fec_platform_data mx53_loco_fec_data __initconst = {
 	.phy = PHY_INTERFACE_MODE_RMII,
 };
 
diff --git a/arch/arm/mach-mx5/board-mx53_smd.c b/arch/arm/mach-mx5/board-mx53_smd.c
index 59c0845..d42132a 100644
--- a/arch/arm/mach-mx5/board-mx53_smd.c
+++ b/arch/arm/mach-mx5/board-mx53_smd.c
@@ -104,7 +104,7 @@
 	gpio_set_value(SMD_FEC_PHY_RST, 1);
 }
 
-static struct fec_platform_data mx53_smd_fec_data = {
+static const struct fec_platform_data mx53_smd_fec_data __initconst = {
 	.phy = PHY_INTERFACE_MODE_RMII,
 };
 
diff --git a/arch/arm/mach-mx5/clock-mx51-mx53.c b/arch/arm/mach-mx5/clock-mx51-mx53.c
index 2aacf41..4cb2769 100644
--- a/arch/arm/mach-mx5/clock-mx51-mx53.c
+++ b/arch/arm/mach-mx5/clock-mx51-mx53.c
@@ -1281,9 +1281,9 @@
 	NULL,  NULL, &ipg_clk, &gpt_ipg_clk);
 
 DEFINE_CLOCK(pwm1_clk, 0, MXC_CCM_CCGR2, MXC_CCM_CCGRx_CG6_OFFSET,
-	NULL, NULL, &ipg_clk, NULL);
+	NULL, NULL, &ipg_perclk, NULL);
 DEFINE_CLOCK(pwm2_clk, 0, MXC_CCM_CCGR2, MXC_CCM_CCGRx_CG8_OFFSET,
-	NULL, NULL, &ipg_clk, NULL);
+	NULL, NULL, &ipg_perclk, NULL);
 
 /* I2C */
 DEFINE_CLOCK(i2c1_clk, 0, MXC_CCM_CCGR1, MXC_CCM_CCGRx_CG9_OFFSET,
@@ -1634,6 +1634,7 @@
 	return 0;
 }
 
+#ifdef CONFIG_OF
 static void __init clk_get_freq_dt(unsigned long *ckil, unsigned long *osc,
 				   unsigned long *ckih1, unsigned long *ckih2)
 {
@@ -1671,3 +1672,4 @@
 	clk_get_freq_dt(&ckil, &osc, &ckih1, &ckih2);
 	return mx53_clocks_init(ckil, osc, ckih1, ckih2);
 }
+#endif
diff --git a/arch/arm/mach-mx5/cpu.c b/arch/arm/mach-mx5/cpu.c
index 5c53282..5e2e7a8 100644
--- a/arch/arm/mach-mx5/cpu.c
+++ b/arch/arm/mach-mx5/cpu.c
@@ -16,7 +16,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <mach/hardware.h>
-#include <asm/io.h>
+#include <linux/io.h>
 
 static int mx5_cpu_rev = -1;
 
@@ -67,7 +67,8 @@
 	if (!cpu_is_mx51())
 		return 0;
 
-	if (mx51_revision() < IMX_CHIP_REVISION_3_0 && (elf_hwcap & HWCAP_NEON)) {
+	if (mx51_revision() < IMX_CHIP_REVISION_3_0 &&
+			(elf_hwcap & HWCAP_NEON)) {
 		elf_hwcap &= ~HWCAP_NEON;
 		pr_info("Turning off NEON support, detected broken NEON implementation\n");
 	}
diff --git a/arch/arm/mach-mx5/imx51-dt.c b/arch/arm/mach-mx5/imx51-dt.c
index ccc6158..596edd9 100644
--- a/arch/arm/mach-mx5/imx51-dt.c
+++ b/arch/arm/mach-mx5/imx51-dt.c
@@ -44,20 +44,22 @@
 	{ /* sentinel */ }
 };
 
-static void __init imx51_tzic_add_irq_domain(struct device_node *np,
+static int __init imx51_tzic_add_irq_domain(struct device_node *np,
 				struct device_node *interrupt_parent)
 {
 	irq_domain_add_simple(np, 0);
+	return 0;
 }
 
-static void __init imx51_gpio_add_irq_domain(struct device_node *np,
+static int __init imx51_gpio_add_irq_domain(struct device_node *np,
 				struct device_node *interrupt_parent)
 {
-	static int gpio_irq_base = MXC_GPIO_IRQ_START + ARCH_NR_GPIOS -
-				   32 * 4; /* imx51 gets 4 gpio ports */
+	static int gpio_irq_base = MXC_GPIO_IRQ_START + ARCH_NR_GPIOS;
 
+	gpio_irq_base -= 32;
 	irq_domain_add_simple(np, gpio_irq_base);
-	gpio_irq_base += 32;
+
+	return 0;
 }
 
 static const struct of_device_id imx51_irq_match[] __initconst = {
diff --git a/arch/arm/mach-mx5/imx53-dt.c b/arch/arm/mach-mx5/imx53-dt.c
index ccaa0b8..85bfd5f 100644
--- a/arch/arm/mach-mx5/imx53-dt.c
+++ b/arch/arm/mach-mx5/imx53-dt.c
@@ -48,20 +48,22 @@
 	{ /* sentinel */ }
 };
 
-static void __init imx53_tzic_add_irq_domain(struct device_node *np,
+static int __init imx53_tzic_add_irq_domain(struct device_node *np,
 				struct device_node *interrupt_parent)
 {
 	irq_domain_add_simple(np, 0);
+	return 0;
 }
 
-static void __init imx53_gpio_add_irq_domain(struct device_node *np,
+static int __init imx53_gpio_add_irq_domain(struct device_node *np,
 				struct device_node *interrupt_parent)
 {
-	static int gpio_irq_base = MXC_GPIO_IRQ_START + ARCH_NR_GPIOS -
-				   32 * 7; /* imx53 gets 7 gpio ports */
+	static int gpio_irq_base = MXC_GPIO_IRQ_START + ARCH_NR_GPIOS;
 
+	gpio_irq_base -= 32;
 	irq_domain_add_simple(np, gpio_irq_base);
-	gpio_irq_base += 32;
+
+	return 0;
 }
 
 static const struct of_device_id imx53_irq_match[] __initconst = {
diff --git a/arch/arm/mach-mx5/mm.c b/arch/arm/mach-mx5/mm.c
index 26eacc9..df4a508f2 100644
--- a/arch/arm/mach-mx5/mm.c
+++ b/arch/arm/mach-mx5/mm.c
@@ -23,7 +23,9 @@
 
 static void imx5_idle(void)
 {
-	mx5_cpu_lp_set(WAIT_UNCLOCKED_POWER_OFF);
+	if (!need_resched())
+		mx5_cpu_lp_set(WAIT_UNCLOCKED_POWER_OFF);
+	local_irq_enable();
 }
 
 /*
@@ -89,7 +91,7 @@
 	mxc_set_cpu_type(MXC_CPU_MX51);
 	mxc_iomux_v3_init(MX51_IO_ADDRESS(MX51_IOMUXC_BASE_ADDR));
 	mxc_arch_reset_init(MX51_IO_ADDRESS(MX51_WDOG1_BASE_ADDR));
-	imx_idle = imx5_idle;
+	pm_idle = imx5_idle;
 }
 
 void __init imx53_init_early(void)
diff --git a/arch/arm/mach-mxs/clock-mx28.c b/arch/arm/mach-mxs/clock-mx28.c
index 229ae34..da6e4aa 100644
--- a/arch/arm/mach-mxs/clock-mx28.c
+++ b/arch/arm/mach-mxs/clock-mx28.c
@@ -404,7 +404,7 @@
 	reg = __raw_readl(CLKCTRL_BASE_ADDR + HW_CLKCTRL_##dr);		\
 	reg &= ~BM_CLKCTRL_##dr##_DIV;					\
 	reg |= div << BP_CLKCTRL_##dr##_DIV;				\
-	if (reg | (1 << clk->enable_shift)) {				\
+	if (reg & (1 << clk->enable_shift)) {				\
 		pr_err("%s: clock is gated\n", __func__);		\
 		return -EINVAL;						\
 	}								\
diff --git a/arch/arm/mach-mxs/include/mach/mx28.h b/arch/arm/mach-mxs/include/mach/mx28.h
index 75d8611..30c7990 100644
--- a/arch/arm/mach-mxs/include/mach/mx28.h
+++ b/arch/arm/mach-mxs/include/mach/mx28.h
@@ -104,8 +104,8 @@
 #define MX28_INT_CAN1			9
 #define MX28_INT_LRADC_TOUCH		10
 #define MX28_INT_HSADC			13
-#define MX28_INT_IRADC_THRESH0		14
-#define MX28_INT_IRADC_THRESH1		15
+#define MX28_INT_LRADC_THRESH0		14
+#define MX28_INT_LRADC_THRESH1		15
 #define MX28_INT_LRADC_CH0		16
 #define MX28_INT_LRADC_CH1		17
 #define MX28_INT_LRADC_CH2		18
diff --git a/arch/arm/mach-mxs/include/mach/mxs.h b/arch/arm/mach-mxs/include/mach/mxs.h
index 0d2d2b4..bde5f66 100644
--- a/arch/arm/mach-mxs/include/mach/mxs.h
+++ b/arch/arm/mach-mxs/include/mach/mxs.h
@@ -30,6 +30,7 @@
  */
 #define cpu_is_mx23()		(					\
 		machine_is_mx23evk() ||					\
+		machine_is_stmp378x() ||				\
 		0)
 #define cpu_is_mx28()		(					\
 		machine_is_mx28evk() ||					\
diff --git a/arch/arm/mach-mxs/mach-m28evk.c b/arch/arm/mach-mxs/mach-m28evk.c
index 3b1681e..6b00577 100644
--- a/arch/arm/mach-mxs/mach-m28evk.c
+++ b/arch/arm/mach-mxs/mach-m28evk.c
@@ -361,6 +361,6 @@
 MACHINE_START(M28EVK, "DENX M28 EVK")
 	.map_io		= mx28_map_io,
 	.init_irq	= mx28_init_irq,
-	.init_machine	= m28evk_init,
 	.timer		= &m28evk_timer,
+	.init_machine	= m28evk_init,
 MACHINE_END
diff --git a/arch/arm/mach-mxs/mach-mx28evk.c b/arch/arm/mach-mxs/mach-mx28evk.c
index ac2316d..064ec5a 100644
--- a/arch/arm/mach-mxs/mach-mx28evk.c
+++ b/arch/arm/mach-mxs/mach-mx28evk.c
@@ -471,7 +471,8 @@
 			       "mmc0-slot-power");
 	if (ret)
 		pr_warn("failed to request gpio mmc0-slot-power: %d\n", ret);
-	mx28_add_mxs_mmc(0, &mx28evk_mmc_pdata[0]);
+	else
+		mx28_add_mxs_mmc(0, &mx28evk_mmc_pdata[0]);
 
 	ret = gpio_request_one(MX28EVK_MMC1_SLOT_POWER, GPIOF_OUT_INIT_LOW,
 			       "mmc1-slot-power");
@@ -480,7 +481,6 @@
 	else
 		mx28_add_mxs_mmc(1, &mx28evk_mmc_pdata[1]);
 
-	mx28_add_mxs_mmc(1, &mx28evk_mmc_pdata[1]);
 	mx28_add_rtc_stmp3xxx();
 
 	gpio_led_register_device(0, &mx28evk_led_data);
diff --git a/arch/arm/mach-mxs/mach-stmp378x_devb.c b/arch/arm/mach-mxs/mach-stmp378x_devb.c
index 177e531..6834dea 100644
--- a/arch/arm/mach-mxs/mach-stmp378x_devb.c
+++ b/arch/arm/mach-mxs/mach-stmp378x_devb.c
@@ -115,6 +115,6 @@
 MACHINE_START(STMP378X, "STMP378X")
 	.map_io		= mx23_map_io,
 	.init_irq	= mx23_init_irq,
-	.init_machine	= stmp378x_dvb_init,
 	.timer		= &stmp378x_dvb_timer,
+	.init_machine	= stmp378x_dvb_init,
 MACHINE_END
diff --git a/arch/arm/mach-mxs/module-tx28.c b/arch/arm/mach-mxs/module-tx28.c
index 0fcff47..9a7b08b 100644
--- a/arch/arm/mach-mxs/module-tx28.c
+++ b/arch/arm/mach-mxs/module-tx28.c
@@ -66,11 +66,11 @@
 	MX28_PAD_ENET0_CRS__ENET1_RX_EN,
 };
 
-static struct fec_platform_data tx28_fec0_data = {
+static const struct fec_platform_data tx28_fec0_data __initconst = {
 	.phy = PHY_INTERFACE_MODE_RMII,
 };
 
-static struct fec_platform_data tx28_fec1_data = {
+static const struct fec_platform_data tx28_fec1_data __initconst = {
 	.phy = PHY_INTERFACE_MODE_RMII,
 };
 
diff --git a/arch/arm/mach-omap1/Kconfig b/arch/arm/mach-omap1/Kconfig
index e0a0281..73f287d 100644
--- a/arch/arm/mach-omap1/Kconfig
+++ b/arch/arm/mach-omap1/Kconfig
@@ -171,14 +171,6 @@
 comment "OMAP CPU Speed"
 	depends on ARCH_OMAP1
 
-config OMAP_CLOCKS_SET_BY_BOOTLOADER
-	bool "OMAP clocks set by bootloader"
-	depends on ARCH_OMAP1
-	help
-	  Enable this option to prevent the kernel from overriding the clock
-	  frequencies programmed by bootloader for MPU, DSP, MMUs, TC,
-	  internal LCD controller and MPU peripherals.
-
 config OMAP_ARM_216MHZ
 	bool "OMAP ARM 216 MHz CPU (1710 only)"
         depends on ARCH_OMAP1 && ARCH_OMAP16XX
diff --git a/arch/arm/mach-omap1/board-ams-delta.c b/arch/arm/mach-omap1/board-ams-delta.c
index 51bae31..b0f15d2 100644
--- a/arch/arm/mach-omap1/board-ams-delta.c
+++ b/arch/arm/mach-omap1/board-ams-delta.c
@@ -302,8 +302,6 @@
 	omap_cfg_reg(J19_1610_CAM_D6);
 	omap_cfg_reg(J18_1610_CAM_D7);
 
-	iotable_init(ams_delta_io_desc, ARRAY_SIZE(ams_delta_io_desc));
-
 	omap_board_config = ams_delta_config;
 	omap_board_config_size = ARRAY_SIZE(ams_delta_config);
 	omap_serial_init();
@@ -373,10 +371,16 @@
 }
 arch_initcall(ams_delta_modem_init);
 
+static void __init ams_delta_map_io(void)
+{
+	omap15xx_map_io();
+	iotable_init(ams_delta_io_desc, ARRAY_SIZE(ams_delta_io_desc));
+}
+
 MACHINE_START(AMS_DELTA, "Amstrad E3 (Delta)")
 	/* Maintainer: Jonathan McDowell <noodles@earth.li> */
 	.atag_offset	= 0x100,
-	.map_io		= omap15xx_map_io,
+	.map_io		= ams_delta_map_io,
 	.init_early	= omap1_init_early,
 	.reserve	= omap_reserve,
 	.init_irq	= omap1_init_irq,
diff --git a/arch/arm/mach-omap1/clock.h b/arch/arm/mach-omap1/clock.h
index eaf09ef..16b1423 100644
--- a/arch/arm/mach-omap1/clock.h
+++ b/arch/arm/mach-omap1/clock.h
@@ -17,7 +17,8 @@
 
 #include <plat/clock.h>
 
-extern int __init omap1_clk_init(void);
+int omap1_clk_init(void);
+void omap1_clk_late_init(void);
 extern int omap1_clk_enable(struct clk *clk);
 extern void omap1_clk_disable(struct clk *clk);
 extern long omap1_clk_round_rate(struct clk *clk, unsigned long rate);
diff --git a/arch/arm/mach-omap1/clock_data.c b/arch/arm/mach-omap1/clock_data.c
index 92400b9..9ff90a7 100644
--- a/arch/arm/mach-omap1/clock_data.c
+++ b/arch/arm/mach-omap1/clock_data.c
@@ -16,6 +16,8 @@
 
 #include <linux/kernel.h>
 #include <linux/clk.h>
+#include <linux/cpufreq.h>
+#include <linux/delay.h>
 #include <linux/io.h>
 
 #include <asm/mach-types.h>  /* for machine_is_* */
@@ -767,6 +769,15 @@
 	.clk_disable_unused	= omap1_clk_disable_unused,
 };
 
+static void __init omap1_show_rates(void)
+{
+	pr_notice("Clocking rate (xtal/DPLL1/MPU): "
+			"%ld.%01ld/%ld.%01ld/%ld.%01ld MHz\n",
+		ck_ref.rate / 1000000, (ck_ref.rate / 100000) % 10,
+		ck_dpll1.rate / 1000000, (ck_dpll1.rate / 100000) % 10,
+		arm_ck.rate / 1000000, (arm_ck.rate / 100000) % 10);
+}
+
 int __init omap1_clk_init(void)
 {
 	struct omap_clk *c;
@@ -835,9 +846,12 @@
 	/* We want to be in syncronous scalable mode */
 	omap_writew(0x1000, ARM_SYSST);
 
-#ifdef CONFIG_OMAP_CLOCKS_SET_BY_BOOTLOADER
-	/* Use values set by bootloader. Determine PLL rate and recalculate
-	 * dependent clocks as if kernel had changed PLL or divisors.
+
+	/*
+	 * Initially use the values set by bootloader. Determine PLL rate and
+	 * recalculate dependent clocks as if kernel had changed PLL or
+	 * divisors. See also omap1_clk_late_init() that can reprogram dpll1
+	 * after the SRAM is initialized.
 	 */
 	{
 		unsigned pll_ctl_val = omap_readw(DPLL_CTL);
@@ -862,25 +876,10 @@
 			}
 		}
 	}
-#else
-	/* Find the highest supported frequency and enable it */
-	if (omap1_select_table_rate(&virtual_ck_mpu, ~0)) {
-		printk(KERN_ERR "System frequencies not set. Check your config.\n");
-		/* Guess sane values (60MHz) */
-		omap_writew(0x2290, DPLL_CTL);
-		omap_writew(cpu_is_omap7xx() ? 0x3005 : 0x1005, ARM_CKCTL);
-		ck_dpll1.rate = 60000000;
-	}
-#endif
 	propagate_rate(&ck_dpll1);
 	/* Cache rates for clocks connected to ck_ref (not dpll1) */
 	propagate_rate(&ck_ref);
-	printk(KERN_INFO "Clocking rate (xtal/DPLL1/MPU): "
-		"%ld.%01ld/%ld.%01ld/%ld.%01ld MHz\n",
-	       ck_ref.rate / 1000000, (ck_ref.rate / 100000) % 10,
-	       ck_dpll1.rate / 1000000, (ck_dpll1.rate / 100000) % 10,
-	       arm_ck.rate / 1000000, (arm_ck.rate / 100000) % 10);
-
+	omap1_show_rates();
 	if (machine_is_omap_perseus2() || machine_is_omap_fsample()) {
 		/* Select slicer output as OMAP input clock */
 		omap_writew(omap_readw(OMAP7XX_PCC_UPLD_CTRL) & ~0x1,
@@ -925,3 +924,27 @@
 
 	return 0;
 }
+
+#define OMAP1_DPLL1_SANE_VALUE	60000000
+
+void __init omap1_clk_late_init(void)
+{
+	unsigned long rate = ck_dpll1.rate;
+
+	if (rate >= OMAP1_DPLL1_SANE_VALUE)
+		return;
+
+	/* System booting at unusable rate, force reprogramming of DPLL1 */
+	ck_dpll1_p->rate = 0;
+
+	/* Find the highest supported frequency and enable it */
+	if (omap1_select_table_rate(&virtual_ck_mpu, ~0)) {
+		pr_err("System frequencies not set, using default. Check your config.\n");
+		omap_writew(0x2290, DPLL_CTL);
+		omap_writew(cpu_is_omap7xx() ? 0x2005 : 0x0005, ARM_CKCTL);
+		ck_dpll1.rate = OMAP1_DPLL1_SANE_VALUE;
+	}
+	propagate_rate(&ck_dpll1);
+	omap1_show_rates();
+	loops_per_jiffy = cpufreq_scale(loops_per_jiffy, rate, ck_dpll1.rate);
+}
diff --git a/arch/arm/mach-omap1/devices.c b/arch/arm/mach-omap1/devices.c
index 48ef988..475cb2f 100644
--- a/arch/arm/mach-omap1/devices.c
+++ b/arch/arm/mach-omap1/devices.c
@@ -30,6 +30,8 @@
 #include <plat/omap7xx.h>
 #include <plat/mcbsp.h>
 
+#include "clock.h"
+
 /*-------------------------------------------------------------------------*/
 
 #if defined(CONFIG_RTC_DRV_OMAP) || defined(CONFIG_RTC_DRV_OMAP_MODULE)
@@ -293,6 +295,7 @@
 		return -ENODEV;
 
 	omap_sram_init();
+	omap1_clk_late_init();
 
 	/* please keep these calls, and their implementations above,
 	 * in alphabetical order so they're easier to sort through.
diff --git a/arch/arm/mach-omap2/Kconfig b/arch/arm/mach-omap2/Kconfig
index 5034147..e1293aa 100644
--- a/arch/arm/mach-omap2/Kconfig
+++ b/arch/arm/mach-omap2/Kconfig
@@ -334,6 +334,7 @@
 config OMAP3_EMU
 	bool "OMAP3 debugging peripherals"
 	depends on ARCH_OMAP3
+	select ARM_AMBA
 	select OC_ETM
 	help
 	  Say Y here to enable debugging hardware of omap3
diff --git a/arch/arm/mach-omap2/Makefile b/arch/arm/mach-omap2/Makefile
index 69ab1c0..b009f17 100644
--- a/arch/arm/mach-omap2/Makefile
+++ b/arch/arm/mach-omap2/Makefile
@@ -4,7 +4,7 @@
 
 # Common support
 obj-y := id.o io.o control.o mux.o devices.o serial.o gpmc.o timer.o pm.o \
-	 common.o gpio.o dma.o wd_timer.o
+	 common.o gpio.o dma.o wd_timer.o display.o
 
 omap-2-3-common				= irq.o sdrc.o
 hwmod-common				= omap_hwmod.o \
@@ -264,7 +264,4 @@
 obj-y					+= $(smsc911x-m) $(smsc911x-y)
 obj-$(CONFIG_ARCH_OMAP4)		+= hwspinlock.o
 
-disp-$(CONFIG_OMAP2_DSS)		:= display.o
-obj-y					+= $(disp-m) $(disp-y)
-
 obj-y					+= common-board-devices.o twl-common.o
diff --git a/arch/arm/mach-omap2/board-rx51-peripherals.c b/arch/arm/mach-omap2/board-rx51-peripherals.c
index ba1aa07..c15c5c9 100644
--- a/arch/arm/mach-omap2/board-rx51-peripherals.c
+++ b/arch/arm/mach-omap2/board-rx51-peripherals.c
@@ -193,7 +193,7 @@
 static void __init rx51_charger_init(void)
 {
 	WARN_ON(gpio_request_one(RX51_USB_TRANSCEIVER_RST_GPIO,
-		GPIOF_OUT_INIT_LOW, "isp1704_reset"));
+		GPIOF_OUT_INIT_HIGH, "isp1704_reset"));
 
 	platform_device_register(&rx51_charger_device);
 }
diff --git a/arch/arm/mach-omap2/cpuidle34xx.c b/arch/arm/mach-omap2/cpuidle34xx.c
index 1fe35c2..942bb4f 100644
--- a/arch/arm/mach-omap2/cpuidle34xx.c
+++ b/arch/arm/mach-omap2/cpuidle34xx.c
@@ -24,6 +24,7 @@
 
 #include <linux/sched.h>
 #include <linux/cpuidle.h>
+#include <linux/export.h>
 
 #include <plat/prcm.h>
 #include <plat/irqs.h>
diff --git a/arch/arm/mach-omap2/display.c b/arch/arm/mach-omap2/display.c
index adb2756..dce9905 100644
--- a/arch/arm/mach-omap2/display.c
+++ b/arch/arm/mach-omap2/display.c
@@ -27,8 +27,35 @@
 #include <plat/omap_hwmod.h>
 #include <plat/omap_device.h>
 #include <plat/omap-pm.h>
+#include <plat/common.h>
 
 #include "control.h"
+#include "display.h"
+
+#define DISPC_CONTROL		0x0040
+#define DISPC_CONTROL2		0x0238
+#define DISPC_IRQSTATUS		0x0018
+
+#define DSS_SYSCONFIG		0x10
+#define DSS_SYSSTATUS		0x14
+#define DSS_CONTROL		0x40
+#define DSS_SDI_CONTROL		0x44
+#define DSS_PLL_CONTROL		0x48
+
+#define LCD_EN_MASK		(0x1 << 0)
+#define DIGIT_EN_MASK		(0x1 << 1)
+
+#define FRAMEDONE_IRQ_SHIFT	0
+#define EVSYNC_EVEN_IRQ_SHIFT	2
+#define EVSYNC_ODD_IRQ_SHIFT	3
+#define FRAMEDONE2_IRQ_SHIFT	22
+#define FRAMEDONETV_IRQ_SHIFT	24
+
+/*
+ * FRAMEDONE_IRQ_TIMEOUT: how long (in milliseconds) to wait during DISPC
+ *     reset before deciding that something has gone wrong
+ */
+#define FRAMEDONE_IRQ_TIMEOUT		100
 
 static struct platform_device omap_display_device = {
 	.name          = "omapdss",
@@ -172,3 +199,135 @@
 
 	return r;
 }
+
+static void dispc_disable_outputs(void)
+{
+	u32 v, irq_mask = 0;
+	bool lcd_en, digit_en, lcd2_en = false;
+	int i;
+	struct omap_dss_dispc_dev_attr *da;
+	struct omap_hwmod *oh;
+
+	oh = omap_hwmod_lookup("dss_dispc");
+	if (!oh) {
+		WARN(1, "display: could not disable outputs during reset - could not find dss_dispc hwmod\n");
+		return;
+	}
+
+	if (!oh->dev_attr) {
+		pr_err("display: could not disable outputs during reset due to missing dev_attr\n");
+		return;
+	}
+
+	da = (struct omap_dss_dispc_dev_attr *)oh->dev_attr;
+
+	/* store value of LCDENABLE and DIGITENABLE bits */
+	v = omap_hwmod_read(oh, DISPC_CONTROL);
+	lcd_en = v & LCD_EN_MASK;
+	digit_en = v & DIGIT_EN_MASK;
+
+	/* store value of LCDENABLE for LCD2 */
+	if (da->manager_count > 2) {
+		v = omap_hwmod_read(oh, DISPC_CONTROL2);
+		lcd2_en = v & LCD_EN_MASK;
+	}
+
+	if (!(lcd_en | digit_en | lcd2_en))
+		return; /* no managers currently enabled */
+
+	/*
+	 * If any manager was enabled, we need to disable it before
+	 * DSS clocks are disabled or DISPC module is reset
+	 */
+	if (lcd_en)
+		irq_mask |= 1 << FRAMEDONE_IRQ_SHIFT;
+
+	if (digit_en) {
+		if (da->has_framedonetv_irq) {
+			irq_mask |= 1 << FRAMEDONETV_IRQ_SHIFT;
+		} else {
+			irq_mask |= 1 << EVSYNC_EVEN_IRQ_SHIFT |
+				1 << EVSYNC_ODD_IRQ_SHIFT;
+		}
+	}
+
+	if (lcd2_en)
+		irq_mask |= 1 << FRAMEDONE2_IRQ_SHIFT;
+
+	/*
+	 * clear any previous FRAMEDONE, FRAMEDONETV,
+	 * EVSYNC_EVEN/ODD or FRAMEDONE2 interrupts
+	 */
+	omap_hwmod_write(irq_mask, oh, DISPC_IRQSTATUS);
+
+	/* disable LCD and TV managers */
+	v = omap_hwmod_read(oh, DISPC_CONTROL);
+	v &= ~(LCD_EN_MASK | DIGIT_EN_MASK);
+	omap_hwmod_write(v, oh, DISPC_CONTROL);
+
+	/* disable LCD2 manager */
+	if (da->manager_count > 2) {
+		v = omap_hwmod_read(oh, DISPC_CONTROL2);
+		v &= ~LCD_EN_MASK;
+		omap_hwmod_write(v, oh, DISPC_CONTROL2);
+	}
+
+	i = 0;
+	while ((omap_hwmod_read(oh, DISPC_IRQSTATUS) & irq_mask) !=
+	       irq_mask) {
+		i++;
+		if (i > FRAMEDONE_IRQ_TIMEOUT) {
+			pr_err("didn't get FRAMEDONE1/2 or TV interrupt\n");
+			break;
+		}
+		mdelay(1);
+	}
+}
+
+#define MAX_MODULE_SOFTRESET_WAIT	10000
+int omap_dss_reset(struct omap_hwmod *oh)
+{
+	struct omap_hwmod_opt_clk *oc;
+	int c = 0;
+	int i, r;
+
+	if (!(oh->class->sysc->sysc_flags & SYSS_HAS_RESET_STATUS)) {
+		pr_err("dss_core: hwmod data doesn't contain reset data\n");
+		return -EINVAL;
+	}
+
+	for (i = oh->opt_clks_cnt, oc = oh->opt_clks; i > 0; i--, oc++)
+		if (oc->_clk)
+			clk_enable(oc->_clk);
+
+	dispc_disable_outputs();
+
+	/* clear SDI registers */
+	if (cpu_is_omap3430()) {
+		omap_hwmod_write(0x0, oh, DSS_SDI_CONTROL);
+		omap_hwmod_write(0x0, oh, DSS_PLL_CONTROL);
+	}
+
+	/*
+	 * clear DSS_CONTROL register to switch DSS clock sources to
+	 * PRCM clock, if any
+	 */
+	omap_hwmod_write(0x0, oh, DSS_CONTROL);
+
+	omap_test_timeout((omap_hwmod_read(oh, oh->class->sysc->syss_offs)
+				& SYSS_RESETDONE_MASK),
+			MAX_MODULE_SOFTRESET_WAIT, c);
+
+	if (c == MAX_MODULE_SOFTRESET_WAIT)
+		pr_warning("dss_core: waiting for reset to finish failed\n");
+	else
+		pr_debug("dss_core: softreset done\n");
+
+	for (i = oh->opt_clks_cnt, oc = oh->opt_clks; i > 0; i--, oc++)
+		if (oc->_clk)
+			clk_disable(oc->_clk);
+
+	r = (c == MAX_MODULE_SOFTRESET_WAIT) ? -ETIMEDOUT : 0;
+
+	return r;
+}
diff --git a/arch/arm/mach-omap2/display.h b/arch/arm/mach-omap2/display.h
new file mode 100644
index 0000000..b871b017
--- /dev/null
+++ b/arch/arm/mach-omap2/display.h
@@ -0,0 +1,29 @@
+/*
+ * display.h - OMAP2+ integration-specific DSS header
+ *
+ * Copyright (C) 2011 Texas Instruments, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __ARCH_ARM_MACH_OMAP2_DISPLAY_H
+#define __ARCH_ARM_MACH_OMAP2_DISPLAY_H
+
+#include <linux/kernel.h>
+
+struct omap_dss_dispc_dev_attr {
+	u8	manager_count;
+	bool	has_framedonetv_irq;
+};
+
+#endif
diff --git a/arch/arm/mach-omap2/io.h b/arch/arm/mach-omap2/io.h
deleted file mode 100644
index e69de29..0000000
--- a/arch/arm/mach-omap2/io.h
+++ /dev/null
diff --git a/arch/arm/mach-omap2/mcbsp.c b/arch/arm/mach-omap2/mcbsp.c
index 292eee3..28fcb27 100644
--- a/arch/arm/mach-omap2/mcbsp.c
+++ b/arch/arm/mach-omap2/mcbsp.c
@@ -145,6 +145,9 @@
 		pdata->reg_size = 4;
 		pdata->has_ccr = true;
 	}
+	pdata->set_clk_src = omap2_mcbsp_set_clk_src;
+	if (id == 1)
+		pdata->mux_signal = omap2_mcbsp1_mux_rx_clk;
 
 	if (oh->class->rev == MCBSP_CONFIG_TYPE3) {
 		if (id == 2)
@@ -174,9 +177,6 @@
 					name, oh->name);
 		return PTR_ERR(pdev);
 	}
-	pdata->set_clk_src = omap2_mcbsp_set_clk_src;
-	if (id == 1)
-		pdata->mux_signal = omap2_mcbsp1_mux_rx_clk;
 	omap_mcbsp_count++;
 	return 0;
 }
diff --git a/arch/arm/mach-omap2/omap_hwmod.c b/arch/arm/mach-omap2/omap_hwmod.c
index 6b3088d..207a2ff 100644
--- a/arch/arm/mach-omap2/omap_hwmod.c
+++ b/arch/arm/mach-omap2/omap_hwmod.c
@@ -749,7 +749,7 @@
 		ohii = &oh->mpu_irqs[i++];
 	} while (ohii->irq != -1);
 
-	return i;
+	return i-1;
 }
 
 /**
@@ -772,7 +772,7 @@
 		ohdi = &oh->sdma_reqs[i++];
 	} while (ohdi->dma_req != -1);
 
-	return i;
+	return i-1;
 }
 
 /**
@@ -795,7 +795,7 @@
 		mem = &os->addr[i++];
 	} while (mem->pa_start != mem->pa_end);
 
-	return i;
+	return i-1;
 }
 
 /**
diff --git a/arch/arm/mach-omap2/omap_hwmod_2420_data.c b/arch/arm/mach-omap2/omap_hwmod_2420_data.c
index 6d72062..a5409ce 100644
--- a/arch/arm/mach-omap2/omap_hwmod_2420_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_2420_data.c
@@ -875,6 +875,10 @@
 };
 
 static struct omap_hwmod_opt_clk dss_opt_clks[] = {
+	/*
+	 * The DSS HW needs all DSS clocks enabled during reset. The dss_core
+	 * driver does not use these clocks.
+	 */
 	{ .role = "tv_clk", .clk = "dss_54m_fck" },
 	{ .role = "sys_clk", .clk = "dss2_fck" },
 };
@@ -899,7 +903,7 @@
 	.slaves_cnt	= ARRAY_SIZE(omap2420_dss_slaves),
 	.masters	= omap2420_dss_masters,
 	.masters_cnt	= ARRAY_SIZE(omap2420_dss_masters),
-	.flags		= HWMOD_NO_IDLEST,
+	.flags		= HWMOD_NO_IDLEST | HWMOD_CONTROL_OPT_CLKS_IN_RESET,
 };
 
 /* l4_core -> dss_dispc */
@@ -939,6 +943,7 @@
 	.slaves		= omap2420_dss_dispc_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap2420_dss_dispc_slaves),
 	.flags		= HWMOD_NO_IDLEST,
+	.dev_attr	= &omap2_3_dss_dispc_dev_attr
 };
 
 /* l4_core -> dss_rfbi */
@@ -961,6 +966,10 @@
 	&omap2420_l4_core__dss_rfbi,
 };
 
+static struct omap_hwmod_opt_clk dss_rfbi_opt_clks[] = {
+	{ .role = "ick", .clk = "dss_ick" },
+};
+
 static struct omap_hwmod omap2420_dss_rfbi_hwmod = {
 	.name		= "dss_rfbi",
 	.class		= &omap2_rfbi_hwmod_class,
@@ -972,6 +981,8 @@
 			.module_offs = CORE_MOD,
 		},
 	},
+	.opt_clks	= dss_rfbi_opt_clks,
+	.opt_clks_cnt	= ARRAY_SIZE(dss_rfbi_opt_clks),
 	.slaves		= omap2420_dss_rfbi_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap2420_dss_rfbi_slaves),
 	.flags		= HWMOD_NO_IDLEST,
@@ -981,7 +992,7 @@
 static struct omap_hwmod_ocp_if omap2420_l4_core__dss_venc = {
 	.master		= &omap2420_l4_core_hwmod,
 	.slave		= &omap2420_dss_venc_hwmod,
-	.clk		= "dss_54m_fck",
+	.clk		= "dss_ick",
 	.addr		= omap2_dss_venc_addrs,
 	.fw = {
 		.omap2 = {
@@ -1001,7 +1012,7 @@
 static struct omap_hwmod omap2420_dss_venc_hwmod = {
 	.name		= "dss_venc",
 	.class		= &omap2_venc_hwmod_class,
-	.main_clk	= "dss1_fck",
+	.main_clk	= "dss_54m_fck",
 	.prcm		= {
 		.omap2 = {
 			.prcm_reg_id = 1,
diff --git a/arch/arm/mach-omap2/omap_hwmod_2430_data.c b/arch/arm/mach-omap2/omap_hwmod_2430_data.c
index a2580d0..c4f56cb 100644
--- a/arch/arm/mach-omap2/omap_hwmod_2430_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_2430_data.c
@@ -942,6 +942,10 @@
 };
 
 static struct omap_hwmod_opt_clk dss_opt_clks[] = {
+	/*
+	 * The DSS HW needs all DSS clocks enabled during reset. The dss_core
+	 * driver does not use these clocks.
+	 */
 	{ .role = "tv_clk", .clk = "dss_54m_fck" },
 	{ .role = "sys_clk", .clk = "dss2_fck" },
 };
@@ -966,7 +970,7 @@
 	.slaves_cnt	= ARRAY_SIZE(omap2430_dss_slaves),
 	.masters	= omap2430_dss_masters,
 	.masters_cnt	= ARRAY_SIZE(omap2430_dss_masters),
-	.flags		= HWMOD_NO_IDLEST,
+	.flags		= HWMOD_NO_IDLEST | HWMOD_CONTROL_OPT_CLKS_IN_RESET,
 };
 
 /* l4_core -> dss_dispc */
@@ -1000,6 +1004,7 @@
 	.slaves		= omap2430_dss_dispc_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap2430_dss_dispc_slaves),
 	.flags		= HWMOD_NO_IDLEST,
+	.dev_attr	= &omap2_3_dss_dispc_dev_attr
 };
 
 /* l4_core -> dss_rfbi */
@@ -1016,6 +1021,10 @@
 	&omap2430_l4_core__dss_rfbi,
 };
 
+static struct omap_hwmod_opt_clk dss_rfbi_opt_clks[] = {
+	{ .role = "ick", .clk = "dss_ick" },
+};
+
 static struct omap_hwmod omap2430_dss_rfbi_hwmod = {
 	.name		= "dss_rfbi",
 	.class		= &omap2_rfbi_hwmod_class,
@@ -1027,6 +1036,8 @@
 			.module_offs = CORE_MOD,
 		},
 	},
+	.opt_clks	= dss_rfbi_opt_clks,
+	.opt_clks_cnt	= ARRAY_SIZE(dss_rfbi_opt_clks),
 	.slaves		= omap2430_dss_rfbi_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap2430_dss_rfbi_slaves),
 	.flags		= HWMOD_NO_IDLEST,
@@ -1036,7 +1047,7 @@
 static struct omap_hwmod_ocp_if omap2430_l4_core__dss_venc = {
 	.master		= &omap2430_l4_core_hwmod,
 	.slave		= &omap2430_dss_venc_hwmod,
-	.clk		= "dss_54m_fck",
+	.clk		= "dss_ick",
 	.addr		= omap2_dss_venc_addrs,
 	.flags		= OCPIF_SWSUP_IDLE,
 	.user		= OCP_USER_MPU | OCP_USER_SDMA,
@@ -1050,7 +1061,7 @@
 static struct omap_hwmod omap2430_dss_venc_hwmod = {
 	.name		= "dss_venc",
 	.class		= &omap2_venc_hwmod_class,
-	.main_clk	= "dss1_fck",
+	.main_clk	= "dss_54m_fck",
 	.prcm		= {
 		.omap2 = {
 			.prcm_reg_id = 1,
diff --git a/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c b/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c
index c451729..c11273d 100644
--- a/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c
@@ -11,6 +11,7 @@
 #include <plat/omap_hwmod.h>
 #include <plat/serial.h>
 #include <plat/dma.h>
+#include <plat/common.h>
 
 #include <mach/irqs.h>
 
@@ -43,13 +44,15 @@
 	.rev_offs	= 0x0000,
 	.sysc_offs	= 0x0010,
 	.syss_offs	= 0x0014,
-	.sysc_flags	= (SYSC_HAS_SOFTRESET | SYSC_HAS_AUTOIDLE),
+	.sysc_flags	= (SYSC_HAS_SOFTRESET | SYSC_HAS_AUTOIDLE |
+			   SYSS_HAS_RESET_STATUS),
 	.sysc_fields	= &omap_hwmod_sysc_type1,
 };
 
 struct omap_hwmod_class omap2_dss_hwmod_class = {
 	.name	= "dss",
 	.sysc	= &omap2_dss_sysc,
+	.reset	= omap_dss_reset,
 };
 
 /*
diff --git a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
index bc9035e..7f8915a 100644
--- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
@@ -1369,9 +1369,14 @@
 };
 
 static struct omap_hwmod_opt_clk dss_opt_clks[] = {
-	{ .role = "tv_clk", .clk = "dss_tv_fck" },
-	{ .role = "video_clk", .clk = "dss_96m_fck" },
+	/*
+	 * The DSS HW needs all DSS clocks enabled during reset. The dss_core
+	 * driver does not use these clocks.
+	 */
 	{ .role = "sys_clk", .clk = "dss2_alwon_fck" },
+	{ .role = "tv_clk", .clk = "dss_tv_fck" },
+	/* required only on OMAP3430 */
+	{ .role = "tv_dac_clk", .clk = "dss_96m_fck" },
 };
 
 static struct omap_hwmod omap3430es1_dss_core_hwmod = {
@@ -1394,11 +1399,12 @@
 	.slaves_cnt	= ARRAY_SIZE(omap3430es1_dss_slaves),
 	.masters	= omap3xxx_dss_masters,
 	.masters_cnt	= ARRAY_SIZE(omap3xxx_dss_masters),
-	.flags		= HWMOD_NO_IDLEST,
+	.flags		= HWMOD_NO_IDLEST | HWMOD_CONTROL_OPT_CLKS_IN_RESET,
 };
 
 static struct omap_hwmod omap3xxx_dss_core_hwmod = {
 	.name		= "dss_core",
+	.flags		= HWMOD_CONTROL_OPT_CLKS_IN_RESET,
 	.class		= &omap2_dss_hwmod_class,
 	.main_clk	= "dss1_alwon_fck", /* instead of dss_fck */
 	.sdma_reqs	= omap3xxx_dss_sdma_chs,
@@ -1456,6 +1462,7 @@
 	.slaves		= omap3xxx_dss_dispc_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap3xxx_dss_dispc_slaves),
 	.flags		= HWMOD_NO_IDLEST,
+	.dev_attr	= &omap2_3_dss_dispc_dev_attr
 };
 
 /*
@@ -1486,6 +1493,7 @@
 static struct omap_hwmod_ocp_if omap3xxx_l4_core__dss_dsi1 = {
 	.master		= &omap3xxx_l4_core_hwmod,
 	.slave		= &omap3xxx_dss_dsi1_hwmod,
+	.clk		= "dss_ick",
 	.addr		= omap3xxx_dss_dsi1_addrs,
 	.fw = {
 		.omap2 = {
@@ -1502,6 +1510,10 @@
 	&omap3xxx_l4_core__dss_dsi1,
 };
 
+static struct omap_hwmod_opt_clk dss_dsi1_opt_clks[] = {
+	{ .role = "sys_clk", .clk = "dss2_alwon_fck" },
+};
+
 static struct omap_hwmod omap3xxx_dss_dsi1_hwmod = {
 	.name		= "dss_dsi1",
 	.class		= &omap3xxx_dsi_hwmod_class,
@@ -1514,6 +1526,8 @@
 			.module_offs = OMAP3430_DSS_MOD,
 		},
 	},
+	.opt_clks	= dss_dsi1_opt_clks,
+	.opt_clks_cnt	= ARRAY_SIZE(dss_dsi1_opt_clks),
 	.slaves		= omap3xxx_dss_dsi1_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap3xxx_dss_dsi1_slaves),
 	.flags		= HWMOD_NO_IDLEST,
@@ -1540,6 +1554,10 @@
 	&omap3xxx_l4_core__dss_rfbi,
 };
 
+static struct omap_hwmod_opt_clk dss_rfbi_opt_clks[] = {
+	{ .role = "ick", .clk = "dss_ick" },
+};
+
 static struct omap_hwmod omap3xxx_dss_rfbi_hwmod = {
 	.name		= "dss_rfbi",
 	.class		= &omap2_rfbi_hwmod_class,
@@ -1551,6 +1569,8 @@
 			.module_offs = OMAP3430_DSS_MOD,
 		},
 	},
+	.opt_clks	= dss_rfbi_opt_clks,
+	.opt_clks_cnt	= ARRAY_SIZE(dss_rfbi_opt_clks),
 	.slaves		= omap3xxx_dss_rfbi_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap3xxx_dss_rfbi_slaves),
 	.flags		= HWMOD_NO_IDLEST,
@@ -1560,7 +1580,7 @@
 static struct omap_hwmod_ocp_if omap3xxx_l4_core__dss_venc = {
 	.master		= &omap3xxx_l4_core_hwmod,
 	.slave		= &omap3xxx_dss_venc_hwmod,
-	.clk		= "dss_tv_fck",
+	.clk		= "dss_ick",
 	.addr		= omap2_dss_venc_addrs,
 	.fw = {
 		.omap2 = {
@@ -1578,10 +1598,15 @@
 	&omap3xxx_l4_core__dss_venc,
 };
 
+static struct omap_hwmod_opt_clk dss_venc_opt_clks[] = {
+	/* required only on OMAP3430 */
+	{ .role = "tv_dac_clk", .clk = "dss_96m_fck" },
+};
+
 static struct omap_hwmod omap3xxx_dss_venc_hwmod = {
 	.name		= "dss_venc",
 	.class		= &omap2_venc_hwmod_class,
-	.main_clk	= "dss1_alwon_fck",
+	.main_clk	= "dss_tv_fck",
 	.prcm		= {
 		.omap2 = {
 			.prcm_reg_id = 1,
@@ -1589,6 +1614,8 @@
 			.module_offs = OMAP3430_DSS_MOD,
 		},
 	},
+	.opt_clks	= dss_venc_opt_clks,
+	.opt_clks_cnt	= ARRAY_SIZE(dss_venc_opt_clks),
 	.slaves		= omap3xxx_dss_venc_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap3xxx_dss_venc_slaves),
 	.flags		= HWMOD_NO_IDLEST,
diff --git a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c
index 7695e5d..daaf165 100644
--- a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c
@@ -30,6 +30,7 @@
 #include <plat/mmc.h>
 #include <plat/i2c.h>
 #include <plat/dmtimer.h>
+#include <plat/common.h>
 
 #include "omap_hwmod_common_data.h"
 
@@ -1187,6 +1188,7 @@
 static struct omap_hwmod_class omap44xx_dss_hwmod_class = {
 	.name	= "dss",
 	.sysc	= &omap44xx_dss_sysc,
+	.reset	= omap_dss_reset,
 };
 
 /* dss */
@@ -1240,12 +1242,12 @@
 static struct omap_hwmod_opt_clk dss_opt_clks[] = {
 	{ .role = "sys_clk", .clk = "dss_sys_clk" },
 	{ .role = "tv_clk", .clk = "dss_tv_clk" },
-	{ .role = "dss_clk", .clk = "dss_dss_clk" },
-	{ .role = "video_clk", .clk = "dss_48mhz_clk" },
+	{ .role = "hdmi_clk", .clk = "dss_48mhz_clk" },
 };
 
 static struct omap_hwmod omap44xx_dss_hwmod = {
 	.name		= "dss_core",
+	.flags		= HWMOD_CONTROL_OPT_CLKS_IN_RESET,
 	.class		= &omap44xx_dss_hwmod_class,
 	.clkdm_name	= "l3_dss_clkdm",
 	.main_clk	= "dss_dss_clk",
@@ -1325,6 +1327,11 @@
 	{ }
 };
 
+static struct omap_dss_dispc_dev_attr omap44xx_dss_dispc_dev_attr = {
+	.manager_count		= 3,
+	.has_framedonetv_irq	= 1
+};
+
 /* l4_per -> dss_dispc */
 static struct omap_hwmod_ocp_if omap44xx_l4_per__dss_dispc = {
 	.master		= &omap44xx_l4_per_hwmod,
@@ -1340,12 +1347,6 @@
 	&omap44xx_l4_per__dss_dispc,
 };
 
-static struct omap_hwmod_opt_clk dss_dispc_opt_clks[] = {
-	{ .role = "sys_clk", .clk = "dss_sys_clk" },
-	{ .role = "tv_clk", .clk = "dss_tv_clk" },
-	{ .role = "hdmi_clk", .clk = "dss_48mhz_clk" },
-};
-
 static struct omap_hwmod omap44xx_dss_dispc_hwmod = {
 	.name		= "dss_dispc",
 	.class		= &omap44xx_dispc_hwmod_class,
@@ -1359,10 +1360,9 @@
 			.context_offs = OMAP4_RM_DSS_DSS_CONTEXT_OFFSET,
 		},
 	},
-	.opt_clks	= dss_dispc_opt_clks,
-	.opt_clks_cnt	= ARRAY_SIZE(dss_dispc_opt_clks),
 	.slaves		= omap44xx_dss_dispc_slaves,
 	.slaves_cnt	= ARRAY_SIZE(omap44xx_dss_dispc_slaves),
+	.dev_attr	= &omap44xx_dss_dispc_dev_attr
 };
 
 /*
@@ -1624,7 +1624,7 @@
 	.clkdm_name	= "l3_dss_clkdm",
 	.mpu_irqs	= omap44xx_dss_hdmi_irqs,
 	.sdma_reqs	= omap44xx_dss_hdmi_sdma_reqs,
-	.main_clk	= "dss_dss_clk",
+	.main_clk	= "dss_48mhz_clk",
 	.prcm = {
 		.omap4 = {
 			.clkctrl_offs = OMAP4_CM_DSS_DSS_CLKCTRL_OFFSET,
@@ -1785,7 +1785,7 @@
 	.name		= "dss_venc",
 	.class		= &omap44xx_venc_hwmod_class,
 	.clkdm_name	= "l3_dss_clkdm",
-	.main_clk	= "dss_dss_clk",
+	.main_clk	= "dss_tv_clk",
 	.prcm = {
 		.omap4 = {
 			.clkctrl_offs = OMAP4_CM_DSS_DSS_CLKCTRL_OFFSET,
diff --git a/arch/arm/mach-omap2/omap_hwmod_common_data.c b/arch/arm/mach-omap2/omap_hwmod_common_data.c
index de832eb..51e5418 100644
--- a/arch/arm/mach-omap2/omap_hwmod_common_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_common_data.c
@@ -49,3 +49,7 @@
 	.srst_shift	= SYSC_TYPE2_SOFTRESET_SHIFT,
 };
 
+struct omap_dss_dispc_dev_attr omap2_3_dss_dispc_dev_attr = {
+	.manager_count		= 2,
+	.has_framedonetv_irq	= 0
+};
diff --git a/arch/arm/mach-omap2/omap_hwmod_common_data.h b/arch/arm/mach-omap2/omap_hwmod_common_data.h
index 39a7c37..ad5d8f0 100644
--- a/arch/arm/mach-omap2/omap_hwmod_common_data.h
+++ b/arch/arm/mach-omap2/omap_hwmod_common_data.h
@@ -16,6 +16,8 @@
 
 #include <plat/omap_hwmod.h>
 
+#include "display.h"
+
 /* Common address space across OMAP2xxx */
 extern struct omap_hwmod_addr_space omap2xxx_uart1_addr_space[];
 extern struct omap_hwmod_addr_space omap2xxx_uart2_addr_space[];
@@ -111,4 +113,6 @@
 extern struct omap_hwmod_class omap2xxx_mailbox_hwmod_class;
 extern struct omap_hwmod_class omap2xxx_mcspi_class;
 
+extern struct omap_dss_dispc_dev_attr omap2_3_dss_dispc_dev_attr;
+
 #endif
diff --git a/arch/arm/mach-omap2/omap_l3_noc.c b/arch/arm/mach-omap2/omap_l3_noc.c
index 6a66aa5..d15225f 100644
--- a/arch/arm/mach-omap2/omap_l3_noc.c
+++ b/arch/arm/mach-omap2/omap_l3_noc.c
@@ -237,7 +237,7 @@
 static const struct of_device_id l3_noc_match[] = {
 	{.compatible = "ti,omap4-l3-noc", },
 	{},
-}
+};
 MODULE_DEVICE_TABLE(of, l3_noc_match);
 #else
 #define l3_noc_match NULL
diff --git a/arch/arm/mach-omap2/pm.c b/arch/arm/mach-omap2/pm.c
index 1e79bdf..00bff46 100644
--- a/arch/arm/mach-omap2/pm.c
+++ b/arch/arm/mach-omap2/pm.c
@@ -24,6 +24,7 @@
 #include "powerdomain.h"
 #include "clockdomain.h"
 #include "pm.h"
+#include "twl-common.h"
 
 static struct omap_device_pm_latency *pm_lats;
 
@@ -226,11 +227,8 @@
 
 static int __init omap2_common_pm_late_init(void)
 {
-	/* Init the OMAP TWL parameters */
-	omap3_twl_init();
-	omap4_twl_init();
-
 	/* Init the voltage layer */
+	omap_pmic_late_init();
 	omap_voltage_late_init();
 
 	/* Initialize the voltages */
diff --git a/arch/arm/mach-omap2/smartreflex.c b/arch/arm/mach-omap2/smartreflex.c
index 6a4f683..cf246b3 100644
--- a/arch/arm/mach-omap2/smartreflex.c
+++ b/arch/arm/mach-omap2/smartreflex.c
@@ -139,7 +139,7 @@
 		sr_write_reg(sr_info, ERRCONFIG_V1, status);
 	} else if (sr_info->ip_type == SR_TYPE_V2) {
 		/* Read the status bits */
-		sr_read_reg(sr_info, IRQSTATUS);
+		status = sr_read_reg(sr_info, IRQSTATUS);
 
 		/* Clear them by writing back */
 		sr_write_reg(sr_info, IRQSTATUS, status);
diff --git a/arch/arm/mach-omap2/twl-common.c b/arch/arm/mach-omap2/twl-common.c
index 52243577..10b20c6 100644
--- a/arch/arm/mach-omap2/twl-common.c
+++ b/arch/arm/mach-omap2/twl-common.c
@@ -30,6 +30,7 @@
 #include <plat/usb.h>
 
 #include "twl-common.h"
+#include "pm.h"
 
 static struct i2c_board_info __initdata pmic_i2c_board_info = {
 	.addr		= 0x48,
@@ -48,6 +49,16 @@
 	omap_register_i2c_bus(bus, clkrate, &pmic_i2c_board_info, 1);
 }
 
+void __init omap_pmic_late_init(void)
+{
+	/* Init the OMAP TWL parameters (if PMIC has been registerd) */
+	if (!pmic_i2c_board_info.irq)
+		return;
+
+	omap3_twl_init();
+	omap4_twl_init();
+}
+
 #if defined(CONFIG_ARCH_OMAP3)
 static struct twl4030_usb_data omap3_usb_pdata = {
 	.usb_mode	= T2_USB_MODE_ULPI,
diff --git a/arch/arm/mach-omap2/twl-common.h b/arch/arm/mach-omap2/twl-common.h
index 5e83a5b..275dde8 100644
--- a/arch/arm/mach-omap2/twl-common.h
+++ b/arch/arm/mach-omap2/twl-common.h
@@ -1,6 +1,8 @@
 #ifndef __OMAP_PMIC_COMMON__
 #define __OMAP_PMIC_COMMON__
 
+#include <plat/irqs.h>
+
 #define TWL_COMMON_PDATA_USB		(1 << 0)
 #define TWL_COMMON_PDATA_BCI		(1 << 1)
 #define TWL_COMMON_PDATA_MADC		(1 << 2)
@@ -30,6 +32,7 @@
 
 void omap_pmic_init(int bus, u32 clkrate, const char *pmic_type, int pmic_irq,
 		    struct twl4030_platform_data *pmic_data);
+void omap_pmic_late_init(void);
 
 static inline void omap2_pmic_init(const char *pmic_type,
 				   struct twl4030_platform_data *pmic_data)
diff --git a/arch/arm/mach-picoxcell/include/mach/debug-macro.S b/arch/arm/mach-picoxcell/include/mach/debug-macro.S
index 8f2c234..58d4ee3 100644
--- a/arch/arm/mach-picoxcell/include/mach/debug-macro.S
+++ b/arch/arm/mach-picoxcell/include/mach/debug-macro.S
@@ -14,7 +14,7 @@
 
 #define UART_SHIFT 2
 
-		.macro	addruart, rp, rv
+		.macro	addruart, rp, rv, tmp
 		ldr	\rv, =PHYS_TO_IO(PICOXCELL_UART1_BASE)
 		ldr	\rp, =PICOXCELL_UART1_BASE
 		.endm
diff --git a/arch/arm/mach-prima2/pm.c b/arch/arm/mach-prima2/pm.c
index cb53160..26ebb57 100644
--- a/arch/arm/mach-prima2/pm.c
+++ b/arch/arm/mach-prima2/pm.c
@@ -9,6 +9,7 @@
 #include <linux/kernel.h>
 #include <linux/suspend.h>
 #include <linux/slab.h>
+#include <linux/module.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
 #include <linux/of_device.h>
diff --git a/arch/arm/mach-prima2/prima2.c b/arch/arm/mach-prima2/prima2.c
index ef555c0..a12b689 100644
--- a/arch/arm/mach-prima2/prima2.c
+++ b/arch/arm/mach-prima2/prima2.c
@@ -8,6 +8,7 @@
 
 #include <linux/init.h>
 #include <linux/kernel.h>
+#include <asm/sizes.h>
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
 #include <linux/of.h>
diff --git a/arch/arm/mach-pxa/balloon3.c b/arch/arm/mach-pxa/balloon3.c
index fc0b854..4b81f59 100644
--- a/arch/arm/mach-pxa/balloon3.c
+++ b/arch/arm/mach-pxa/balloon3.c
@@ -307,7 +307,7 @@
 /******************************************************************************
  * USB Gadget
  ******************************************************************************/
-#if defined(CONFIG_USB_GADGET_PXA27X)||defined(CONFIG_USB_GADGET_PXA27X_MODULE)
+#if defined(CONFIG_USB_PXA27X)||defined(CONFIG_USB_PXA27X_MODULE)
 static void balloon3_udc_command(int cmd)
 {
 	if (cmd == PXA2XX_UDC_CMD_CONNECT)
diff --git a/arch/arm/mach-pxa/colibri-pxa320.c b/arch/arm/mach-pxa/colibri-pxa320.c
index 692e1ff..d23b92b 100644
--- a/arch/arm/mach-pxa/colibri-pxa320.c
+++ b/arch/arm/mach-pxa/colibri-pxa320.c
@@ -146,7 +146,7 @@
 static inline void __init colibri_pxa320_init_eth(void) {}
 #endif /* CONFIG_AX88796 */
 
-#if defined(CONFIG_USB_GADGET_PXA27X)||defined(CONFIG_USB_GADGET_PXA27X_MODULE)
+#if defined(CONFIG_USB_PXA27X)||defined(CONFIG_USB_PXA27X_MODULE)
 static struct gpio_vbus_mach_info colibri_pxa320_gpio_vbus_info = {
 	.gpio_vbus		= mfp_to_gpio(MFP_PIN_GPIO96),
 	.gpio_pullup		= -1,
diff --git a/arch/arm/mach-pxa/gumstix.c b/arch/arm/mach-pxa/gumstix.c
index 9c8208c..ffdd70d 100644
--- a/arch/arm/mach-pxa/gumstix.c
+++ b/arch/arm/mach-pxa/gumstix.c
@@ -106,7 +106,7 @@
 }
 #endif
 
-#ifdef CONFIG_USB_GADGET_PXA25X
+#ifdef CONFIG_USB_PXA25X
 static struct gpio_vbus_mach_info gumstix_udc_info = {
 	.gpio_vbus		= GPIO_GUMSTIX_USB_GPIOn,
 	.gpio_pullup		= GPIO_GUMSTIX_USB_GPIOx,
diff --git a/arch/arm/mach-pxa/include/mach/palm27x.h b/arch/arm/mach-pxa/include/mach/palm27x.h
index f80bbe2..d4eac3d 100644
--- a/arch/arm/mach-pxa/include/mach/palm27x.h
+++ b/arch/arm/mach-pxa/include/mach/palm27x.h
@@ -37,8 +37,8 @@
 #define palm27x_lcd_init(power, mode)	do {} while (0)
 #endif
 
-#if	defined(CONFIG_USB_GADGET_PXA27X) || \
-	defined(CONFIG_USB_GADGET_PXA27X_MODULE)
+#if	defined(CONFIG_USB_PXA27X) || \
+	defined(CONFIG_USB_PXA27X_MODULE)
 extern void __init palm27x_udc_init(int vbus, int pullup,
 					int vbus_inverted);
 #else
diff --git a/arch/arm/mach-pxa/palm27x.c b/arch/arm/mach-pxa/palm27x.c
index 325c245..fbc10d7 100644
--- a/arch/arm/mach-pxa/palm27x.c
+++ b/arch/arm/mach-pxa/palm27x.c
@@ -164,8 +164,8 @@
 /******************************************************************************
  * USB Gadget
  ******************************************************************************/
-#if	defined(CONFIG_USB_GADGET_PXA27X) || \
-	defined(CONFIG_USB_GADGET_PXA27X_MODULE)
+#if	defined(CONFIG_USB_PXA27X) || \
+	defined(CONFIG_USB_PXA27X_MODULE)
 static struct gpio_vbus_mach_info palm27x_udc_info = {
 	.gpio_vbus_inverted	= 1,
 };
diff --git a/arch/arm/mach-pxa/palmtc.c b/arch/arm/mach-pxa/palmtc.c
index 6ec7cae..2c24c67 100644
--- a/arch/arm/mach-pxa/palmtc.c
+++ b/arch/arm/mach-pxa/palmtc.c
@@ -338,7 +338,7 @@
 /******************************************************************************
  * UDC
  ******************************************************************************/
-#if defined(CONFIG_USB_GADGET_PXA25X)||defined(CONFIG_USB_GADGET_PXA25X_MODULE)
+#if defined(CONFIG_USB_PXA25X)||defined(CONFIG_USB_PXA25X_MODULE)
 static struct gpio_vbus_mach_info palmtc_udc_info = {
 	.gpio_vbus		= GPIO_NR_PALMTC_USB_DETECT_N,
 	.gpio_vbus_inverted	= 1,
diff --git a/arch/arm/mach-pxa/vpac270.c b/arch/arm/mach-pxa/vpac270.c
index a7539a6..ca0c661 100644
--- a/arch/arm/mach-pxa/vpac270.c
+++ b/arch/arm/mach-pxa/vpac270.c
@@ -343,7 +343,7 @@
 /******************************************************************************
  * USB Gadget
  ******************************************************************************/
-#if defined(CONFIG_USB_GADGET_PXA27X)||defined(CONFIG_USB_GADGET_PXA27X_MODULE)
+#if defined(CONFIG_USB_PXA27X)||defined(CONFIG_USB_PXA27X_MODULE)
 static struct gpio_vbus_mach_info vpac270_gpio_vbus_info = {
 	.gpio_vbus		= GPIO41_VPAC270_UDC_DETECT,
 	.gpio_pullup		= -1,
diff --git a/arch/arm/mach-s3c64xx/dev-spi.c b/arch/arm/mach-s3c64xx/dev-spi.c
index 5e6b420..3341fd1 100644
--- a/arch/arm/mach-s3c64xx/dev-spi.c
+++ b/arch/arm/mach-s3c64xx/dev-spi.c
@@ -10,6 +10,7 @@
 
 #include <linux/kernel.h>
 #include <linux/string.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 #include <linux/dma-mapping.h>
 #include <linux/gpio.h>
diff --git a/arch/arm/mach-s3c64xx/mach-crag6410-module.c b/arch/arm/mach-s3c64xx/mach-crag6410-module.c
index 6666856..f208154 100644
--- a/arch/arm/mach-s3c64xx/mach-crag6410-module.c
+++ b/arch/arm/mach-s3c64xx/mach-crag6410-module.c
@@ -8,7 +8,7 @@
  * published by the Free Software Foundation.
  */
 
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/interrupt.h>
 #include <linux/i2c.h>
 
diff --git a/arch/arm/mach-s3c64xx/s3c6400.c b/arch/arm/mach-s3c64xx/s3c6400.c
index 7a3bc32..51c00f2 100644
--- a/arch/arm/mach-s3c64xx/s3c6400.c
+++ b/arch/arm/mach-s3c64xx/s3c6400.c
@@ -70,7 +70,7 @@
 	s3c64xx_init_irq(~0 & ~(0xf << 5), ~0);
 }
 
-struct sysdev_class s3c6400_sysclass = {
+static struct sysdev_class s3c6400_sysclass = {
 	.name	= "s3c6400-core",
 };
 
diff --git a/arch/arm/mach-s3c64xx/setup-fb-24bpp.c b/arch/arm/mach-s3c64xx/setup-fb-24bpp.c
index 83d2afb..2cf8002 100644
--- a/arch/arm/mach-s3c64xx/setup-fb-24bpp.c
+++ b/arch/arm/mach-s3c64xx/setup-fb-24bpp.c
@@ -20,7 +20,7 @@
 #include <plat/fb.h>
 #include <plat/gpio-cfg.h>
 
-extern void s3c64xx_fb_gpio_setup_24bpp(void)
+void s3c64xx_fb_gpio_setup_24bpp(void)
 {
 	s3c_gpio_cfgrange_nopull(S3C64XX_GPI(0), 16, S3C_GPIO_SFN(2));
 	s3c_gpio_cfgrange_nopull(S3C64XX_GPJ(0), 12, S3C_GPIO_SFN(2));
diff --git a/arch/arm/mach-s5pv210/mach-smdkv210.c b/arch/arm/mach-s5pv210/mach-smdkv210.c
index a9106c3..8662ef6 100644
--- a/arch/arm/mach-s5pv210/mach-smdkv210.c
+++ b/arch/arm/mach-s5pv210/mach-smdkv210.c
@@ -273,6 +273,7 @@
 
 static struct platform_pwm_backlight_data smdkv210_bl_data = {
 	.pwm_id = 3,
+	.pwm_period_ns = 1000,
 };
 
 static void __init smdkv210_map_io(void)
diff --git a/arch/arm/mach-sa1100/Makefile.boot b/arch/arm/mach-sa1100/Makefile.boot
index 5a616f6..f7951aa 100644
--- a/arch/arm/mach-sa1100/Makefile.boot
+++ b/arch/arm/mach-sa1100/Makefile.boot
@@ -1,5 +1,5 @@
-ifeq ($(CONFIG_ARCH_SA1100),y)
-   zreladdr-$(CONFIG_SA1111)		+= 0xc0208000
+ifeq ($(CONFIG_SA1111),y)
+   zreladdr-y	+= 0xc0208000
 else
    zreladdr-y	+= 0xc0008000
 endif
diff --git a/arch/arm/mach-shmobile/Makefile b/arch/arm/mach-shmobile/Makefile
index 2aec2f7..737bdc6 100644
--- a/arch/arm/mach-shmobile/Makefile
+++ b/arch/arm/mach-shmobile/Makefile
@@ -3,7 +3,7 @@
 #
 
 # Common objects
-obj-y				:= timer.o console.o clock.o pm_runtime.o
+obj-y				:= timer.o console.o clock.o
 
 # CPU objects
 obj-$(CONFIG_ARCH_SH7367)	+= setup-sh7367.o clock-sh7367.o intc-sh7367.o
diff --git a/arch/arm/mach-shmobile/board-ag5evm.c b/arch/arm/mach-shmobile/board-ag5evm.c
index 83624e2..b862e9f 100644
--- a/arch/arm/mach-shmobile/board-ag5evm.c
+++ b/arch/arm/mach-shmobile/board-ag5evm.c
@@ -515,14 +515,14 @@
 	/* enable MMCIF */
 	gpio_request(GPIO_FN_MMCCLK0, NULL);
 	gpio_request(GPIO_FN_MMCCMD0_PU, NULL);
-	gpio_request(GPIO_FN_MMCD0_0, NULL);
-	gpio_request(GPIO_FN_MMCD0_1, NULL);
-	gpio_request(GPIO_FN_MMCD0_2, NULL);
-	gpio_request(GPIO_FN_MMCD0_3, NULL);
-	gpio_request(GPIO_FN_MMCD0_4, NULL);
-	gpio_request(GPIO_FN_MMCD0_5, NULL);
-	gpio_request(GPIO_FN_MMCD0_6, NULL);
-	gpio_request(GPIO_FN_MMCD0_7, NULL);
+	gpio_request(GPIO_FN_MMCD0_0_PU, NULL);
+	gpio_request(GPIO_FN_MMCD0_1_PU, NULL);
+	gpio_request(GPIO_FN_MMCD0_2_PU, NULL);
+	gpio_request(GPIO_FN_MMCD0_3_PU, NULL);
+	gpio_request(GPIO_FN_MMCD0_4_PU, NULL);
+	gpio_request(GPIO_FN_MMCD0_5_PU, NULL);
+	gpio_request(GPIO_FN_MMCD0_6_PU, NULL);
+	gpio_request(GPIO_FN_MMCD0_7_PU, NULL);
 	gpio_request(GPIO_PORT208, NULL); /* Reset */
 	gpio_direction_output(GPIO_PORT208, 1);
 
diff --git a/arch/arm/mach-shmobile/board-ap4evb.c b/arch/arm/mach-shmobile/board-ap4evb.c
index a3aa0f6..4c865ec 100644
--- a/arch/arm/mach-shmobile/board-ap4evb.c
+++ b/arch/arm/mach-shmobile/board-ap4evb.c
@@ -201,7 +201,7 @@
 static struct resource nor_flash_resources[] = {
 	[0]	= {
 		.start	= 0x20000000, /* CS0 shadow instead of regular CS0 */
-		.end	= 0x28000000 - 1, /* needed by USB MASK ROM boot */		
+		.end	= 0x28000000 - 1, /* needed by USB MASK ROM boot */
 		.flags	= IORESOURCE_MEM,
 	}
 };
diff --git a/arch/arm/mach-shmobile/board-kota2.c b/arch/arm/mach-shmobile/board-kota2.c
index adc7312..bd9a784 100644
--- a/arch/arm/mach-shmobile/board-kota2.c
+++ b/arch/arm/mach-shmobile/board-kota2.c
@@ -48,6 +48,7 @@
 #include <asm/hardware/cache-l2x0.h>
 #include <asm/traps.h>
 
+/* SMSC 9220 */
 static struct resource smsc9220_resources[] = {
 	[0] = {
 		.start		= 0x14000000, /* CS5A */
@@ -77,6 +78,7 @@
 	.num_resources	= ARRAY_SIZE(smsc9220_resources),
 };
 
+/* KEYSC */
 static struct sh_keysc_info keysc_platdata = {
 	.mode		= SH_KEYSC_MODE_6,
 	.scan_timing	= 3,
@@ -120,6 +122,7 @@
 	},
 };
 
+/* GPIO KEY */
 #define GPIO_KEY(c, g, d) { .code = c, .gpio = g, .desc = d, .active_low = 1 }
 
 static struct gpio_keys_button gpio_buttons[] = {
@@ -150,6 +153,7 @@
 	},
 };
 
+/* GPIO LED */
 #define GPIO_LED(n, g) { .name = n, .gpio = g }
 
 static struct gpio_led gpio_leds[] = {
@@ -175,6 +179,7 @@
 	},
 };
 
+/* MMCIF */
 static struct resource mmcif_resources[] = {
 	[0] = {
 		.name   = "MMCIF",
@@ -207,6 +212,7 @@
 	.resource       = mmcif_resources,
 };
 
+/* SDHI0 */
 static struct sh_mobile_sdhi_info sdhi0_info = {
 	.tmio_caps      = MMC_CAP_SD_HIGHSPEED,
 	.tmio_flags     = TMIO_MMC_WRPROTECT_DISABLE | TMIO_MMC_HAS_IDLE_WAIT,
@@ -243,6 +249,7 @@
 	},
 };
 
+/* SDHI1 */
 static struct sh_mobile_sdhi_info sdhi1_info = {
 	.tmio_caps      = MMC_CAP_NONREMOVABLE | MMC_CAP_SDIO_IRQ,
 	.tmio_flags     = TMIO_MMC_WRPROTECT_DISABLE | TMIO_MMC_HAS_IDLE_WAIT,
diff --git a/arch/arm/mach-shmobile/clock-sh7372.c b/arch/arm/mach-shmobile/clock-sh7372.c
index 6697592..995a9c3 100644
--- a/arch/arm/mach-shmobile/clock-sh7372.c
+++ b/arch/arm/mach-shmobile/clock-sh7372.c
@@ -476,7 +476,7 @@
 	.disable	= fsidiv_disable,
 };
 
-static struct clk_mapping sh7372_fsidiva_clk_mapping = {
+static struct clk_mapping fsidiva_clk_mapping = {
 	.phys	= FSIDIVA,
 	.len	= 8,
 };
@@ -484,10 +484,10 @@
 struct clk sh7372_fsidiva_clk = {
 	.ops		= &fsidiv_clk_ops,
 	.parent		= &div6_reparent_clks[DIV6_FSIA], /* late install */
-	.mapping	= &sh7372_fsidiva_clk_mapping,
+	.mapping	= &fsidiva_clk_mapping,
 };
 
-static struct clk_mapping sh7372_fsidivb_clk_mapping = {
+static struct clk_mapping fsidivb_clk_mapping = {
 	.phys	= FSIDIVB,
 	.len	= 8,
 };
@@ -495,7 +495,7 @@
 struct clk sh7372_fsidivb_clk = {
 	.ops		= &fsidiv_clk_ops,
 	.parent		= &div6_reparent_clks[DIV6_FSIB],  /* late install */
-	.mapping	= &sh7372_fsidivb_clk_mapping,
+	.mapping	= &fsidivb_clk_mapping,
 };
 
 static struct clk *late_main_clks[] = {
diff --git a/arch/arm/mach-shmobile/cpuidle.c b/arch/arm/mach-shmobile/cpuidle.c
index 2e44f11..1b23342 100644
--- a/arch/arm/mach-shmobile/cpuidle.c
+++ b/arch/arm/mach-shmobile/cpuidle.c
@@ -26,65 +26,59 @@
 };
 
 static int shmobile_cpuidle_enter(struct cpuidle_device *dev,
-				  struct cpuidle_state *state)
+				  struct cpuidle_driver *drv,
+				  int index)
 {
 	ktime_t before, after;
-	int requested_state = state - &dev->states[0];
 
-	dev->last_state = &dev->states[requested_state];
 	before = ktime_get();
 
 	local_irq_disable();
 	local_fiq_disable();
 
-	shmobile_cpuidle_modes[requested_state]();
+	shmobile_cpuidle_modes[index]();
 
 	local_irq_enable();
 	local_fiq_enable();
 
 	after = ktime_get();
-	return ktime_to_ns(ktime_sub(after, before)) >> 10;
+	dev->last_residency = ktime_to_ns(ktime_sub(after, before)) >> 10;
+
+	return index;
 }
 
 static struct cpuidle_device shmobile_cpuidle_dev;
 static struct cpuidle_driver shmobile_cpuidle_driver = {
 	.name =		"shmobile_cpuidle",
 	.owner =	THIS_MODULE,
+	.states[0] = {
+		.name = "C1",
+		.desc = "WFI",
+		.exit_latency = 1,
+		.target_residency = 1 * 2,
+		.flags = CPUIDLE_FLAG_TIME_VALID,
+	},
+	.safe_state_index = 0, /* C1 */
+	.state_count = 1,
 };
 
-void (*shmobile_cpuidle_setup)(struct cpuidle_device *dev);
+void (*shmobile_cpuidle_setup)(struct cpuidle_driver *drv);
 
 static int shmobile_cpuidle_init(void)
 {
 	struct cpuidle_device *dev = &shmobile_cpuidle_dev;
-	struct cpuidle_state *state;
+	struct cpuidle_driver *drv = &shmobile_cpuidle_driver;
 	int i;
 
-	cpuidle_register_driver(&shmobile_cpuidle_driver);
-
-	for (i = 0; i < CPUIDLE_STATE_MAX; i++) {
-		dev->states[i].name[0] = '\0';
-		dev->states[i].desc[0] = '\0';
-		dev->states[i].enter = shmobile_cpuidle_enter;
-	}
-
-	i = CPUIDLE_DRIVER_STATE_START;
-
-	state = &dev->states[i++];
-	snprintf(state->name, CPUIDLE_NAME_LEN, "C1");
-	strncpy(state->desc, "WFI", CPUIDLE_DESC_LEN);
-	state->exit_latency = 1;
-	state->target_residency = 1 * 2;
-	state->power_usage = 3;
-	state->flags = 0;
-	state->flags |= CPUIDLE_FLAG_TIME_VALID;
-
-	dev->safe_state = state;
-	dev->state_count = i;
+	for (i = 0; i < CPUIDLE_STATE_MAX; i++)
+		drv->states[i].enter = shmobile_cpuidle_enter;
 
 	if (shmobile_cpuidle_setup)
-		shmobile_cpuidle_setup(dev);
+		shmobile_cpuidle_setup(drv);
 
+	cpuidle_register_driver(drv);
+
+	dev->state_count = drv->state_count;
 	cpuidle_register_device(dev);
 
 	return 0;
diff --git a/arch/arm/mach-shmobile/include/mach/common.h b/arch/arm/mach-shmobile/include/mach/common.h
index c0cdbf9..834bd6c 100644
--- a/arch/arm/mach-shmobile/include/mach/common.h
+++ b/arch/arm/mach-shmobile/include/mach/common.h
@@ -9,9 +9,9 @@
 extern void shmobile_handle_irq_intc(struct pt_regs *);
 extern void shmobile_handle_irq_gic(struct pt_regs *);
 extern struct platform_suspend_ops shmobile_suspend_ops;
-struct cpuidle_device;
+struct cpuidle_driver;
 extern void (*shmobile_cpuidle_modes[])(void);
-extern void (*shmobile_cpuidle_setup)(struct cpuidle_device *dev);
+extern void (*shmobile_cpuidle_setup)(struct cpuidle_driver *drv);
 
 extern void sh7367_init_irq(void);
 extern void sh7367_add_early_devices(void);
diff --git a/arch/arm/mach-shmobile/include/mach/sh73a0.h b/arch/arm/mach-shmobile/include/mach/sh73a0.h
index 18ae6a9..881d515 100644
--- a/arch/arm/mach-shmobile/include/mach/sh73a0.h
+++ b/arch/arm/mach-shmobile/include/mach/sh73a0.h
@@ -470,6 +470,14 @@
 	GPIO_FN_SDHICMD2_PU,
 	GPIO_FN_MMCCMD0_PU,
 	GPIO_FN_MMCCMD1_PU,
+	GPIO_FN_MMCD0_0_PU,
+	GPIO_FN_MMCD0_1_PU,
+	GPIO_FN_MMCD0_2_PU,
+	GPIO_FN_MMCD0_3_PU,
+	GPIO_FN_MMCD0_4_PU,
+	GPIO_FN_MMCD0_5_PU,
+	GPIO_FN_MMCD0_6_PU,
+	GPIO_FN_MMCD0_7_PU,
 	GPIO_FN_FSIACK_PU,
 	GPIO_FN_FSIAILR_PU,
 	GPIO_FN_FSIAIBT_PU,
diff --git a/arch/arm/mach-shmobile/pfc-sh7367.c b/arch/arm/mach-shmobile/pfc-sh7367.c
index 128555e..e6e5246 100644
--- a/arch/arm/mach-shmobile/pfc-sh7367.c
+++ b/arch/arm/mach-shmobile/pfc-sh7367.c
@@ -21,68 +21,49 @@
 #include <linux/gpio.h>
 #include <mach/sh7367.h>
 
-#define _1(fn, pfx, sfx) fn(pfx, sfx)
-
-#define _10(fn, pfx, sfx)				\
-	_1(fn, pfx##0, sfx), _1(fn, pfx##1, sfx),	\
-	_1(fn, pfx##2, sfx), _1(fn, pfx##3, sfx),	\
-	_1(fn, pfx##4, sfx), _1(fn, pfx##5, sfx),	\
-	_1(fn, pfx##6, sfx), _1(fn, pfx##7, sfx),	\
-	_1(fn, pfx##8, sfx), _1(fn, pfx##9, sfx)
-
-#define _90(fn, pfx, sfx)				\
-	_10(fn, pfx##1, sfx), _10(fn, pfx##2, sfx),	\
-	_10(fn, pfx##3, sfx), _10(fn, pfx##4, sfx),	\
-	_10(fn, pfx##5, sfx), _10(fn, pfx##6, sfx),	\
-	_10(fn, pfx##7, sfx), _10(fn, pfx##8, sfx),	\
-	_10(fn, pfx##9, sfx)
-
-#define _273(fn, pfx, sfx)		\
-	_10(fn, pfx, sfx), _90(fn, pfx, sfx),		\
-	_10(fn, pfx##10, sfx), _90(fn, pfx##1, sfx),	\
-	_10(fn, pfx##20, sfx), _10(fn, pfx##21, sfx),	\
-	_10(fn, pfx##22, sfx), _10(fn, pfx##23, sfx),	\
-	_10(fn, pfx##24, sfx), _10(fn, pfx##25, sfx),	\
-	_10(fn, pfx##26, sfx), _1(fn, pfx##270, sfx),	\
-	_1(fn, pfx##271, sfx), _1(fn, pfx##272, sfx)
-
-#define _PORT(pfx, sfx) pfx##_##sfx
-#define PORT_273(str) _273(_PORT, PORT, str)
+#define CPU_ALL_PORT(fn, pfx, sfx)				\
+	PORT_10(fn, pfx, sfx), PORT_90(fn, pfx, sfx),		\
+	PORT_10(fn, pfx##10, sfx), PORT_90(fn, pfx##1, sfx),	\
+	PORT_10(fn, pfx##20, sfx), PORT_10(fn, pfx##21, sfx),	\
+	PORT_10(fn, pfx##22, sfx), PORT_10(fn, pfx##23, sfx),	\
+	PORT_10(fn, pfx##24, sfx), PORT_10(fn, pfx##25, sfx),	\
+	PORT_10(fn, pfx##26, sfx), PORT_1(fn, pfx##270, sfx),	\
+	PORT_1(fn, pfx##271, sfx), PORT_1(fn, pfx##272, sfx)
 
 enum {
 	PINMUX_RESERVED = 0,
 
 	PINMUX_DATA_BEGIN,
-	PORT_273(DATA), /* PORT0_DATA -> PORT272_DATA */
+	PORT_ALL(DATA), /* PORT0_DATA -> PORT272_DATA */
 	PINMUX_DATA_END,
 
 	PINMUX_INPUT_BEGIN,
-	PORT_273(IN), /* PORT0_IN -> PORT272_IN */
+	PORT_ALL(IN), /* PORT0_IN -> PORT272_IN */
 	PINMUX_INPUT_END,
 
 	PINMUX_INPUT_PULLUP_BEGIN,
-	PORT_273(IN_PU), /* PORT0_IN_PU -> PORT272_IN_PU */
+	PORT_ALL(IN_PU), /* PORT0_IN_PU -> PORT272_IN_PU */
 	PINMUX_INPUT_PULLUP_END,
 
 	PINMUX_INPUT_PULLDOWN_BEGIN,
-	PORT_273(IN_PD), /* PORT0_IN_PD -> PORT272_IN_PD */
+	PORT_ALL(IN_PD), /* PORT0_IN_PD -> PORT272_IN_PD */
 	PINMUX_INPUT_PULLDOWN_END,
 
 	PINMUX_OUTPUT_BEGIN,
-	PORT_273(OUT), /* PORT0_OUT -> PORT272_OUT */
+	PORT_ALL(OUT), /* PORT0_OUT -> PORT272_OUT */
 	PINMUX_OUTPUT_END,
 
 	PINMUX_FUNCTION_BEGIN,
-	PORT_273(FN_IN), /* PORT0_FN_IN -> PORT272_FN_IN */
-	PORT_273(FN_OUT), /* PORT0_FN_OUT -> PORT272_FN_OUT */
-	PORT_273(FN0), /* PORT0_FN0 -> PORT272_FN0 */
-	PORT_273(FN1), /* PORT0_FN1 -> PORT272_FN1 */
-	PORT_273(FN2), /* PORT0_FN2 -> PORT272_FN2 */
-	PORT_273(FN3), /* PORT0_FN3 -> PORT272_FN3 */
-	PORT_273(FN4), /* PORT0_FN4 -> PORT272_FN4 */
-	PORT_273(FN5), /* PORT0_FN5 -> PORT272_FN5 */
-	PORT_273(FN6), /* PORT0_FN6 -> PORT272_FN6 */
-	PORT_273(FN7), /* PORT0_FN7 -> PORT272_FN7 */
+	PORT_ALL(FN_IN), /* PORT0_FN_IN -> PORT272_FN_IN */
+	PORT_ALL(FN_OUT), /* PORT0_FN_OUT -> PORT272_FN_OUT */
+	PORT_ALL(FN0), /* PORT0_FN0 -> PORT272_FN0 */
+	PORT_ALL(FN1), /* PORT0_FN1 -> PORT272_FN1 */
+	PORT_ALL(FN2), /* PORT0_FN2 -> PORT272_FN2 */
+	PORT_ALL(FN3), /* PORT0_FN3 -> PORT272_FN3 */
+	PORT_ALL(FN4), /* PORT0_FN4 -> PORT272_FN4 */
+	PORT_ALL(FN5), /* PORT0_FN5 -> PORT272_FN5 */
+	PORT_ALL(FN6), /* PORT0_FN6 -> PORT272_FN6 */
+	PORT_ALL(FN7), /* PORT0_FN7 -> PORT272_FN7 */
 
 	MSELBCR_MSEL2_1, MSELBCR_MSEL2_0,
 	PINMUX_FUNCTION_END,
@@ -327,41 +308,6 @@
 	PINMUX_MARK_END,
 };
 
-#define PORT_DATA_I(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_IN)
-
-#define PORT_DATA_I_PD(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, \
-		    PORT##nr##_IN, PORT##nr##_IN_PD)
-
-#define PORT_DATA_I_PU(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, \
-		    PORT##nr##_IN, PORT##nr##_IN_PU)
-
-#define PORT_DATA_I_PU_PD(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, \
-		    PORT##nr##_IN, PORT##nr##_IN_PD, PORT##nr##_IN_PU)
-
-#define PORT_DATA_O(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT)
-
-#define PORT_DATA_IO(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT, \
-		    PORT##nr##_IN)
-
-#define PORT_DATA_IO_PD(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT, \
-		    PORT##nr##_IN, PORT##nr##_IN_PD)
-
-#define PORT_DATA_IO_PU(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT, \
-		    PORT##nr##_IN, PORT##nr##_IN_PU)
-
-#define PORT_DATA_IO_PU_PD(nr) \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT, \
-		    PORT##nr##_IN, PORT##nr##_IN_PD, PORT##nr##_IN_PU)
-
-
 static pinmux_enum_t pinmux_data[] = {
 
 	/* specify valid pin states for each pin in GPIO mode */
@@ -1098,13 +1044,9 @@
 	PINMUX_DATA(DIVLOCK_MARK, PORT272_FN1),
 };
 
-#define _GPIO_PORT(pfx, sfx) PINMUX_GPIO(GPIO_PORT##pfx, PORT##pfx##_DATA)
-#define GPIO_PORT_273() _273(_GPIO_PORT, , unused)
-#define GPIO_FN(str) PINMUX_GPIO(GPIO_FN_##str, str##_MARK)
-
 static struct pinmux_gpio pinmux_gpios[] = {
 	/* 49-1 -> 49-6 (GPIO) */
-	GPIO_PORT_273(),
+	GPIO_PORT_ALL(),
 
 	/* Special Pull-up / Pull-down Functions */
 	GPIO_FN(PORT48_KEYIN0_PU), GPIO_FN(PORT49_KEYIN1_PU),
@@ -1345,22 +1287,6 @@
 	GPIO_FN(DIVLOCK),
 };
 
-/* helper for top 4 bits in PORTnCR */
-#define PCRH(in, in_pd, in_pu, out)		\
-	0, (out), (in), 0,			\
-	0, 0, 0, 0,				\
-	0, 0, (in_pd), 0,			\
-	0, 0, (in_pu), 0
-
-#define PORTCR(nr, reg)						\
-	{ PINMUX_CFG_REG("PORT" nr "CR", reg, 8, 4) {		\
-		PCRH(PORT##nr##_IN, PORT##nr##_IN_PD,		\
-		     PORT##nr##_IN_PU, PORT##nr##_OUT),		\
-		PORT##nr##_FN0, PORT##nr##_FN1, PORT##nr##_FN2,	\
-		PORT##nr##_FN3,	PORT##nr##_FN4, PORT##nr##_FN5,	\
-		PORT##nr##_FN6, PORT##nr##_FN7 }		\
-	}
-
 static struct pinmux_cfg_reg pinmux_config_regs[] = {
 	PORTCR(0, 0xe6050000), /* PORT0CR */
 	PORTCR(1, 0xe6050001), /* PORT1CR */
diff --git a/arch/arm/mach-shmobile/pfc-sh7372.c b/arch/arm/mach-shmobile/pfc-sh7372.c
index 9c265da..1bd6585 100644
--- a/arch/arm/mach-shmobile/pfc-sh7372.c
+++ b/arch/arm/mach-shmobile/pfc-sh7372.c
@@ -25,27 +25,13 @@
 #include <linux/gpio.h>
 #include <mach/sh7372.h>
 
-#define _1(fn, pfx, sfx) fn(pfx, sfx)
-
-#define _10(fn, pfx, sfx)				\
-	_1(fn, pfx##0, sfx), _1(fn, pfx##1, sfx),	\
-	_1(fn, pfx##2, sfx), _1(fn, pfx##3, sfx),	\
-	_1(fn, pfx##4, sfx), _1(fn, pfx##5, sfx),	\
-	_1(fn, pfx##6, sfx), _1(fn, pfx##7, sfx),	\
-	_1(fn, pfx##8, sfx), _1(fn, pfx##9, sfx)
-
-#define _80(fn, pfx, sfx)				\
-	_10(fn, pfx##1, sfx),	_10(fn, pfx##2, sfx),	\
-	_10(fn, pfx##3, sfx),	_10(fn, pfx##4, sfx),	\
-	_10(fn, pfx##5, sfx),	_10(fn, pfx##6, sfx),	\
-	_10(fn, pfx##7, sfx),	_10(fn, pfx##8, sfx)
-
-#define _190(fn, pfx, sfx) \
-	_10(fn, pfx, sfx), _80(fn, pfx, sfx), _10(fn, pfx##9, sfx), \
-	_10(fn, pfx##10, sfx), _80(fn, pfx##1, sfx), _1(fn, pfx##190, sfx)
-
-#define _PORT(pfx, sfx) pfx##_##sfx
-#define PORT_ALL(str) _190(_PORT, PORT, str)
+#define CPU_ALL_PORT(fn, pfx, sfx) \
+	PORT_10(fn, pfx, sfx),		PORT_90(fn, pfx, sfx), \
+	PORT_10(fn, pfx##10, sfx),	PORT_10(fn, pfx##11, sfx), \
+	PORT_10(fn, pfx##12, sfx),	PORT_10(fn, pfx##13, sfx), \
+	PORT_10(fn, pfx##14, sfx),	PORT_10(fn, pfx##15, sfx), \
+	PORT_10(fn, pfx##16, sfx),	PORT_10(fn, pfx##17, sfx), \
+	PORT_10(fn, pfx##18, sfx),	PORT_1(fn, pfx##190, sfx)
 
 enum {
 	PINMUX_RESERVED = 0,
@@ -381,108 +367,124 @@
 	PINMUX_MARK_END,
 };
 
-/* PORT_DATA_I_PD(nr) */
-#define _I___D(nr)			     \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, \
-		    PORT##nr##_IN, PORT##nr##_IN_PD)
-
-/* PORT_DATA_I_PU(nr) */
-#define _I__U_(nr)			     \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, \
-		    PORT##nr##_IN, PORT##nr##_IN_PU)
-
-/* PORT_DATA_I_PU_PD(nr) */
-#define _I__UD(nr)			     \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, \
-		    PORT##nr##_IN, PORT##nr##_IN_PD, PORT##nr##_IN_PU)
-
-/* PORT_DATA_O(nr) */
-#define __O___(nr)							\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT)
-
-/* PORT_DATA_IO(nr) */
-#define _IO___(nr)				     \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT, \
-		    PORT##nr##_IN)
-
-/* PORT_DATA_IO_PD(nr) */
-#define _IO__D(nr)					     \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT, \
-		    PORT##nr##_IN, PORT##nr##_IN_PD)
-
-/* PORT_DATA_IO_PU(nr) */
-#define _IO_U_(nr)					     \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT, \
-		    PORT##nr##_IN, PORT##nr##_IN_PU)
-
-/* PORT_DATA_IO_PU_PD(nr) */
-#define _IO_UD(nr)					     \
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT, \
-		    PORT##nr##_IN, PORT##nr##_IN_PD, PORT##nr##_IN_PU)
-
-
 static pinmux_enum_t pinmux_data[] = {
 
 	/* specify valid pin states for each pin in GPIO mode */
+	PORT_DATA_IO_PD(0),		PORT_DATA_IO_PD(1),
+	PORT_DATA_O(2),			PORT_DATA_I_PD(3),
+	PORT_DATA_I_PD(4),		PORT_DATA_I_PD(5),
+	PORT_DATA_IO_PU_PD(6),		PORT_DATA_I_PD(7),
+	PORT_DATA_IO_PD(8),		PORT_DATA_O(9),
 
-	_IO__D(0), _IO__D(1), __O___(2), _I___D(3), _I___D(4),
-	_I___D(5), _IO_UD(6), _I___D(7), _IO__D(8), __O___(9),
+	PORT_DATA_O(10),		PORT_DATA_O(11),
+	PORT_DATA_IO_PU_PD(12),		PORT_DATA_IO_PD(13),
+	PORT_DATA_IO_PD(14),		PORT_DATA_O(15),
+	PORT_DATA_IO_PD(16),		PORT_DATA_IO_PD(17),
+	PORT_DATA_I_PD(18),		PORT_DATA_IO(19),
 
-	__O___(10), __O___(11), _IO_UD(12), _IO__D(13), _IO__D(14),
-	__O___(15), _IO__D(16), _IO__D(17), _I___D(18), _IO___(19),
+	PORT_DATA_IO(20),		PORT_DATA_IO(21),
+	PORT_DATA_IO(22),		PORT_DATA_IO(23),
+	PORT_DATA_IO(24),		PORT_DATA_IO(25),
+	PORT_DATA_IO(26),		PORT_DATA_IO(27),
+	PORT_DATA_IO(28),		PORT_DATA_IO(29),
 
-	_IO___(20), _IO___(21), _IO___(22), _IO___(23), _IO___(24),
-	_IO___(25), _IO___(26), _IO___(27), _IO___(28), _IO___(29),
+	PORT_DATA_IO(30),		PORT_DATA_IO(31),
+	PORT_DATA_IO(32),		PORT_DATA_IO(33),
+	PORT_DATA_IO(34),		PORT_DATA_IO(35),
+	PORT_DATA_IO(36),		PORT_DATA_IO(37),
+	PORT_DATA_IO(38),		PORT_DATA_IO(39),
 
-	_IO___(30), _IO___(31), _IO___(32), _IO___(33), _IO___(34),
-	_IO___(35), _IO___(36), _IO___(37), _IO___(38), _IO___(39),
+	PORT_DATA_IO(40),		PORT_DATA_IO(41),
+	PORT_DATA_IO(42),		PORT_DATA_IO(43),
+	PORT_DATA_IO(44),		PORT_DATA_IO(45),
+	PORT_DATA_IO_PU(46),		PORT_DATA_IO_PU(47),
+	PORT_DATA_IO_PU(48),		PORT_DATA_IO_PU(49),
 
-	_IO___(40), _IO___(41), _IO___(42), _IO___(43), _IO___(44),
-	_IO___(45), _IO_U_(46), _IO_U_(47), _IO_U_(48), _IO_U_(49),
+	PORT_DATA_IO_PU(50),		PORT_DATA_IO_PU(51),
+	PORT_DATA_IO_PU(52),		PORT_DATA_IO_PU(53),
+	PORT_DATA_IO_PU(54),		PORT_DATA_IO_PU(55),
+	PORT_DATA_IO_PU(56),		PORT_DATA_IO_PU(57),
+	PORT_DATA_IO_PU(58),		PORT_DATA_IO_PU(59),
 
-	_IO_U_(50), _IO_U_(51), _IO_U_(52), _IO_U_(53), _IO_U_(54),
-	_IO_U_(55), _IO_U_(56), _IO_U_(57), _IO_U_(58), _IO_U_(59),
+	PORT_DATA_IO_PU(60),		PORT_DATA_IO_PU(61),
+	PORT_DATA_IO(62),		PORT_DATA_O(63),
+	PORT_DATA_O(64),		PORT_DATA_IO_PU(65),
+	PORT_DATA_O(66),		PORT_DATA_IO_PU(67),  /*66?*/
+	PORT_DATA_O(68),		PORT_DATA_IO(69),
 
-	_IO_U_(60), _IO_U_(61), _IO___(62), __O___(63), __O___(64),
-	_IO_U_(65), __O___(66), _IO_U_(67), __O___(68), _IO___(69), /*66?*/
+	PORT_DATA_IO(70),		PORT_DATA_IO(71),
+	PORT_DATA_O(72),		PORT_DATA_I_PU(73),
+	PORT_DATA_I_PU_PD(74),		PORT_DATA_IO_PU_PD(75),
+	PORT_DATA_IO_PU_PD(76),		PORT_DATA_IO_PU_PD(77),
+	PORT_DATA_IO_PU_PD(78),		PORT_DATA_IO_PU_PD(79),
 
-	_IO___(70), _IO___(71), __O___(72), _I__U_(73), _I__UD(74),
-	_IO_UD(75), _IO_UD(76), _IO_UD(77), _IO_UD(78), _IO_UD(79),
+	PORT_DATA_IO_PU_PD(80),		PORT_DATA_IO_PU_PD(81),
+	PORT_DATA_IO_PU_PD(82),		PORT_DATA_IO_PU_PD(83),
+	PORT_DATA_IO_PU_PD(84),		PORT_DATA_IO_PU_PD(85),
+	PORT_DATA_IO_PU_PD(86),		PORT_DATA_IO_PU_PD(87),
+	PORT_DATA_IO_PU_PD(88),		PORT_DATA_IO_PU_PD(89),
 
-	_IO_UD(80), _IO_UD(81), _IO_UD(82), _IO_UD(83), _IO_UD(84),
-	_IO_UD(85), _IO_UD(86), _IO_UD(87), _IO_UD(88), _IO_UD(89),
+	PORT_DATA_IO_PU_PD(90),		PORT_DATA_IO_PU_PD(91),
+	PORT_DATA_IO_PU_PD(92),		PORT_DATA_IO_PU_PD(93),
+	PORT_DATA_IO_PU_PD(94),		PORT_DATA_IO_PU_PD(95),
+	PORT_DATA_IO_PU(96),		PORT_DATA_IO_PU_PD(97),
+	PORT_DATA_IO_PU_PD(98),		PORT_DATA_O(99), /*99?*/
 
-	_IO_UD(90), _IO_UD(91), _IO_UD(92), _IO_UD(93), _IO_UD(94),
-	_IO_UD(95), _IO_U_(96), _IO_UD(97), _IO_UD(98), __O___(99), /*99?*/
+	PORT_DATA_IO_PD(100),		PORT_DATA_IO_PD(101),
+	PORT_DATA_IO_PD(102),		PORT_DATA_IO_PD(103),
+	PORT_DATA_IO_PD(104),		PORT_DATA_IO_PD(105),
+	PORT_DATA_IO_PU(106),		PORT_DATA_IO_PU(107),
+	PORT_DATA_IO_PU(108),		PORT_DATA_IO_PU(109),
 
-	_IO__D(100), _IO__D(101), _IO__D(102), _IO__D(103), _IO__D(104),
-	_IO__D(105), _IO_U_(106), _IO_U_(107), _IO_U_(108), _IO_U_(109),
+	PORT_DATA_IO_PU(110),		PORT_DATA_IO_PU(111),
+	PORT_DATA_IO_PD(112),		PORT_DATA_IO_PD(113),
+	PORT_DATA_IO_PU(114),		PORT_DATA_IO_PU(115),
+	PORT_DATA_IO_PU(116),		PORT_DATA_IO_PU(117),
+	PORT_DATA_IO_PU(118),		PORT_DATA_IO_PU(119),
 
-	_IO_U_(110), _IO_U_(111), _IO__D(112), _IO__D(113), _IO_U_(114),
-	_IO_U_(115), _IO_U_(116), _IO_U_(117), _IO_U_(118), _IO_U_(119),
+	PORT_DATA_IO_PU(120),		PORT_DATA_IO_PD(121),
+	PORT_DATA_IO_PD(122),		PORT_DATA_IO_PD(123),
+	PORT_DATA_IO_PD(124),		PORT_DATA_IO_PD(125),
+	PORT_DATA_IO_PD(126),		PORT_DATA_IO_PD(127),
+	PORT_DATA_IO_PD(128),		PORT_DATA_IO_PU_PD(129),
 
-	_IO_U_(120), _IO__D(121), _IO__D(122), _IO__D(123), _IO__D(124),
-	_IO__D(125), _IO__D(126), _IO__D(127), _IO__D(128), _IO_UD(129),
+	PORT_DATA_IO_PU_PD(130),	PORT_DATA_IO_PU_PD(131),
+	PORT_DATA_IO_PU_PD(132),	PORT_DATA_IO_PU_PD(133),
+	PORT_DATA_IO_PU_PD(134),	PORT_DATA_IO_PU_PD(135),
+	PORT_DATA_IO_PD(136),		PORT_DATA_IO_PD(137),
+	PORT_DATA_IO_PD(138),		PORT_DATA_IO_PD(139),
 
-	_IO_UD(130), _IO_UD(131), _IO_UD(132), _IO_UD(133), _IO_UD(134),
-	_IO_UD(135), _IO__D(136), _IO__D(137), _IO__D(138), _IO__D(139),
+	PORT_DATA_IO_PD(140),		PORT_DATA_IO_PD(141),
+	PORT_DATA_IO_PD(142),		PORT_DATA_IO_PU_PD(143),
+	PORT_DATA_IO_PD(144),		PORT_DATA_IO_PD(145),
+	PORT_DATA_IO_PD(146),		PORT_DATA_IO_PD(147),
+	PORT_DATA_IO_PD(148),		PORT_DATA_IO_PD(149),
 
-	_IO__D(140), _IO__D(141), _IO__D(142), _IO_UD(143), _IO__D(144),
-	_IO__D(145), _IO__D(146), _IO__D(147), _IO__D(148), _IO__D(149),
+	PORT_DATA_IO_PD(150),		PORT_DATA_IO_PD(151),
+	PORT_DATA_IO_PU_PD(152),	PORT_DATA_I_PD(153),
+	PORT_DATA_IO_PU_PD(154),	PORT_DATA_I_PD(155),
+	PORT_DATA_IO_PD(156),		PORT_DATA_IO_PD(157),
+	PORT_DATA_I_PD(158),		PORT_DATA_IO_PD(159),
 
-	_IO__D(150), _IO__D(151), _IO_UD(152), _I___D(153), _IO_UD(154),
-	_I___D(155), _IO__D(156), _IO__D(157), _I___D(158), _IO__D(159),
+	PORT_DATA_O(160),		PORT_DATA_IO_PD(161),
+	PORT_DATA_IO_PD(162),		PORT_DATA_IO_PD(163),
+	PORT_DATA_I_PD(164),		PORT_DATA_IO_PD(165),
+	PORT_DATA_I_PD(166),		PORT_DATA_I_PD(167),
+	PORT_DATA_I_PD(168),		PORT_DATA_I_PD(169),
 
-	__O___(160), _IO__D(161), _IO__D(162), _IO__D(163), _I___D(164),
-	_IO__D(165), _I___D(166), _I___D(167), _I___D(168), _I___D(169),
+	PORT_DATA_I_PD(170),		PORT_DATA_O(171),
+	PORT_DATA_IO_PU_PD(172),	PORT_DATA_IO_PU_PD(173),
+	PORT_DATA_IO_PU_PD(174),	PORT_DATA_IO_PU_PD(175),
+	PORT_DATA_IO_PU_PD(176),	PORT_DATA_IO_PU_PD(177),
+	PORT_DATA_IO_PU_PD(178),	PORT_DATA_O(179),
 
-	_I___D(170), __O___(171), _IO_UD(172), _IO_UD(173), _IO_UD(174),
-	_IO_UD(175), _IO_UD(176), _IO_UD(177), _IO_UD(178), __O___(179),
+	PORT_DATA_IO_PU_PD(180),	PORT_DATA_IO_PU_PD(181),
+	PORT_DATA_IO_PU_PD(182),	PORT_DATA_IO_PU_PD(183),
+	PORT_DATA_IO_PU_PD(184),	PORT_DATA_O(185),
+	PORT_DATA_IO_PU_PD(186),	PORT_DATA_IO_PU_PD(187),
+	PORT_DATA_IO_PU_PD(188),	PORT_DATA_IO_PU_PD(189),
 
-	_IO_UD(180), _IO_UD(181), _IO_UD(182), _IO_UD(183), _IO_UD(184),
-	__O___(185), _IO_UD(186), _IO_UD(187), _IO_UD(188), _IO_UD(189),
-
-	_IO_UD(190),
+	PORT_DATA_IO_PU_PD(190),
 
 	/* IRQ */
 	PINMUX_DATA(IRQ0_6_MARK,	PORT6_FN0, 	MSEL1CR_0_0),
@@ -926,10 +928,6 @@
 	PINMUX_DATA(MFIv4_MARK,		MSEL4CR_6_1),
 };
 
-#define _GPIO_PORT(pfx, sfx) PINMUX_GPIO(GPIO_PORT##pfx, PORT##pfx##_DATA)
-#define GPIO_PORT_ALL() _190(_GPIO_PORT, , unused)
-#define GPIO_FN(str) PINMUX_GPIO(GPIO_FN_##str, str##_MARK)
-
 static struct pinmux_gpio pinmux_gpios[] = {
 
 	/* PORT */
@@ -1201,22 +1199,6 @@
 	GPIO_FN(SDENC_DV_CLKI),
 };
 
-/* helper for top 4 bits in PORTnCR */
-#define PCRH(in, in_pd, in_pu, out)		\
-	0, (out), (in), 0,			\
-	0, 0, 0, 0,				\
-	0, 0, (in_pd), 0,			\
-	0, 0, (in_pu), 0
-
-#define PORTCR(nr, reg)						\
-	{ PINMUX_CFG_REG("PORT" nr "CR", reg, 8, 4) {		\
-		PCRH(PORT##nr##_IN, PORT##nr##_IN_PD,		\
-		     PORT##nr##_IN_PU, PORT##nr##_OUT),		\
-		PORT##nr##_FN0, PORT##nr##_FN1, PORT##nr##_FN2,	\
-		PORT##nr##_FN3,	PORT##nr##_FN4, PORT##nr##_FN5,	\
-		PORT##nr##_FN6, PORT##nr##_FN7 }		\
-	}
-
 static struct pinmux_cfg_reg pinmux_config_regs[] = {
 	PORTCR(0,	0xE6051000), /* PORT0CR */
 	PORTCR(1,	0xE6051001), /* PORT1CR */
diff --git a/arch/arm/mach-shmobile/pfc-sh7377.c b/arch/arm/mach-shmobile/pfc-sh7377.c
index 613e684..2f10511 100644
--- a/arch/arm/mach-shmobile/pfc-sh7377.c
+++ b/arch/arm/mach-shmobile/pfc-sh7377.c
@@ -22,84 +22,65 @@
 #include <linux/gpio.h>
 #include <mach/sh7377.h>
 
-#define _1(fn, pfx, sfx) fn(pfx, sfx)
-
-#define _10(fn, pfx, sfx)				\
-	_1(fn, pfx##0, sfx), _1(fn, pfx##1, sfx),	\
-	_1(fn, pfx##2, sfx), _1(fn, pfx##3, sfx),	\
-	_1(fn, pfx##4, sfx), _1(fn, pfx##5, sfx),	\
-	_1(fn, pfx##6, sfx), _1(fn, pfx##7, sfx),	\
-	_1(fn, pfx##8, sfx), _1(fn, pfx##9, sfx)
-
-#define _90(fn, pfx, sfx)				\
-	_10(fn, pfx##1, sfx), _10(fn, pfx##2, sfx),	\
-	_10(fn, pfx##3, sfx), _10(fn, pfx##4, sfx),	\
-	_10(fn, pfx##5, sfx), _10(fn, pfx##6, sfx),	\
-	_10(fn, pfx##7, sfx), _10(fn, pfx##8, sfx),	\
-	_10(fn, pfx##9, sfx)
-
-#define _265(fn, pfx, sfx)				\
-	_10(fn, pfx, sfx), _90(fn, pfx, sfx),		\
-	_10(fn, pfx##10, sfx),				\
-	_1(fn, pfx##110, sfx), _1(fn, pfx##111, sfx),	\
-	_1(fn, pfx##112, sfx), _1(fn, pfx##113, sfx),	\
-	_1(fn, pfx##114, sfx), _1(fn, pfx##115, sfx),	\
-	_1(fn, pfx##116, sfx), _1(fn, pfx##117, sfx),	\
-	_1(fn, pfx##118, sfx),				\
-	_1(fn, pfx##128, sfx), _1(fn, pfx##129, sfx),	\
-	_10(fn, pfx##13, sfx), _10(fn, pfx##14, sfx),	\
-	_10(fn, pfx##15, sfx),				\
-	_1(fn, pfx##160, sfx), _1(fn, pfx##161, sfx),	\
-	_1(fn, pfx##162, sfx), _1(fn, pfx##163, sfx),	\
-	_1(fn, pfx##164, sfx),				\
-	_1(fn, pfx##192, sfx), _1(fn, pfx##193, sfx),	\
-	_1(fn, pfx##194, sfx), _1(fn, pfx##195, sfx),	\
-	_1(fn, pfx##196, sfx), _1(fn, pfx##197, sfx),	\
-	_1(fn, pfx##198, sfx), _1(fn, pfx##199, sfx),	\
-	_10(fn, pfx##20, sfx), _10(fn, pfx##21, sfx),	\
-	_10(fn, pfx##22, sfx), _10(fn, pfx##23, sfx),	\
-	_10(fn, pfx##24, sfx), _10(fn, pfx##25, sfx),	\
-	_1(fn, pfx##260, sfx), _1(fn, pfx##261, sfx),	\
-	_1(fn, pfx##262, sfx), _1(fn, pfx##263, sfx),	\
-	_1(fn, pfx##264, sfx)
-
-#define _PORT(pfx, sfx) pfx##_##sfx
-#define PORT_265(str) _265(_PORT, PORT, str)
+#define CPU_ALL_PORT(fn, pfx, sfx)				\
+	PORT_10(fn, pfx, sfx), PORT_90(fn, pfx, sfx),		\
+	PORT_10(fn, pfx##10, sfx),				\
+	PORT_1(fn, pfx##110, sfx), PORT_1(fn, pfx##111, sfx),	\
+	PORT_1(fn, pfx##112, sfx), PORT_1(fn, pfx##113, sfx),	\
+	PORT_1(fn, pfx##114, sfx), PORT_1(fn, pfx##115, sfx),	\
+	PORT_1(fn, pfx##116, sfx), PORT_1(fn, pfx##117, sfx),	\
+	PORT_1(fn, pfx##118, sfx),				\
+	PORT_1(fn, pfx##128, sfx), PORT_1(fn, pfx##129, sfx),	\
+	PORT_10(fn, pfx##13, sfx), PORT_10(fn, pfx##14, sfx),	\
+	PORT_10(fn, pfx##15, sfx),				\
+	PORT_1(fn, pfx##160, sfx), PORT_1(fn, pfx##161, sfx),	\
+	PORT_1(fn, pfx##162, sfx), PORT_1(fn, pfx##163, sfx),	\
+	PORT_1(fn, pfx##164, sfx),				\
+	PORT_1(fn, pfx##192, sfx), PORT_1(fn, pfx##193, sfx),	\
+	PORT_1(fn, pfx##194, sfx), PORT_1(fn, pfx##195, sfx),	\
+	PORT_1(fn, pfx##196, sfx), PORT_1(fn, pfx##197, sfx),	\
+	PORT_1(fn, pfx##198, sfx), PORT_1(fn, pfx##199, sfx),	\
+	PORT_10(fn, pfx##20, sfx), PORT_10(fn, pfx##21, sfx),	\
+	PORT_10(fn, pfx##22, sfx), PORT_10(fn, pfx##23, sfx),	\
+	PORT_10(fn, pfx##24, sfx), PORT_10(fn, pfx##25, sfx),	\
+	PORT_1(fn, pfx##260, sfx), PORT_1(fn, pfx##261, sfx),	\
+	PORT_1(fn, pfx##262, sfx), PORT_1(fn, pfx##263, sfx),	\
+	PORT_1(fn, pfx##264, sfx)
 
 enum {
 	PINMUX_RESERVED = 0,
 
 	PINMUX_DATA_BEGIN,
-	PORT_265(DATA), /* PORT0_DATA -> PORT264_DATA */
+	PORT_ALL(DATA), /* PORT0_DATA -> PORT264_DATA */
 	PINMUX_DATA_END,
 
 	PINMUX_INPUT_BEGIN,
-	PORT_265(IN), /* PORT0_IN -> PORT264_IN */
+	PORT_ALL(IN), /* PORT0_IN -> PORT264_IN */
 	PINMUX_INPUT_END,
 
 	PINMUX_INPUT_PULLUP_BEGIN,
-	PORT_265(IN_PU), /* PORT0_IN_PU -> PORT264_IN_PU */
+	PORT_ALL(IN_PU), /* PORT0_IN_PU -> PORT264_IN_PU */
 	PINMUX_INPUT_PULLUP_END,
 
 	PINMUX_INPUT_PULLDOWN_BEGIN,
-	PORT_265(IN_PD), /* PORT0_IN_PD -> PORT264_IN_PD */
+	PORT_ALL(IN_PD), /* PORT0_IN_PD -> PORT264_IN_PD */
 	PINMUX_INPUT_PULLDOWN_END,
 
 	PINMUX_OUTPUT_BEGIN,
-	PORT_265(OUT), /* PORT0_OUT -> PORT264_OUT */
+	PORT_ALL(OUT), /* PORT0_OUT -> PORT264_OUT */
 	PINMUX_OUTPUT_END,
 
 	PINMUX_FUNCTION_BEGIN,
-	PORT_265(FN_IN), /* PORT0_FN_IN -> PORT264_FN_IN */
-	PORT_265(FN_OUT), /* PORT0_FN_OUT -> PORT264_FN_OUT */
-	PORT_265(FN0), /* PORT0_FN0 -> PORT264_FN0 */
-	PORT_265(FN1), /* PORT0_FN1 -> PORT264_FN1 */
-	PORT_265(FN2), /* PORT0_FN2 -> PORT264_FN2 */
-	PORT_265(FN3), /* PORT0_FN3 -> PORT264_FN3 */
-	PORT_265(FN4), /* PORT0_FN4 -> PORT264_FN4 */
-	PORT_265(FN5), /* PORT0_FN5 -> PORT264_FN5 */
-	PORT_265(FN6), /* PORT0_FN6 -> PORT264_FN6 */
-	PORT_265(FN7), /* PORT0_FN7 -> PORT264_FN7 */
+	PORT_ALL(FN_IN), /* PORT0_FN_IN -> PORT264_FN_IN */
+	PORT_ALL(FN_OUT), /* PORT0_FN_OUT -> PORT264_FN_OUT */
+	PORT_ALL(FN0), /* PORT0_FN0 -> PORT264_FN0 */
+	PORT_ALL(FN1), /* PORT0_FN1 -> PORT264_FN1 */
+	PORT_ALL(FN2), /* PORT0_FN2 -> PORT264_FN2 */
+	PORT_ALL(FN3), /* PORT0_FN3 -> PORT264_FN3 */
+	PORT_ALL(FN4), /* PORT0_FN4 -> PORT264_FN4 */
+	PORT_ALL(FN5), /* PORT0_FN5 -> PORT264_FN5 */
+	PORT_ALL(FN6), /* PORT0_FN6 -> PORT264_FN6 */
+	PORT_ALL(FN7), /* PORT0_FN7 -> PORT264_FN7 */
 
 	MSELBCR_MSEL17_1, MSELBCR_MSEL17_0,
 	MSELBCR_MSEL16_1, MSELBCR_MSEL16_0,
@@ -360,45 +341,6 @@
 	PINMUX_MARK_END,
 };
 
-#define PORT_DATA_I(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_IN)
-
-#define PORT_DATA_I_PD(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_IN, PORT##nr##_IN_PD)
-
-#define PORT_DATA_I_PU(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_IN, PORT##nr##_IN_PU)
-
-#define PORT_DATA_I_PU_PD(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_IN, PORT##nr##_IN_PD,	\
-				PORT##nr##_IN_PU)
-
-#define PORT_DATA_O(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT)
-
-#define PORT_DATA_IO(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT, PORT##nr##_IN)
-
-#define PORT_DATA_IO_PD(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT, PORT##nr##_IN,		\
-				PORT##nr##_IN_PD)
-
-#define PORT_DATA_IO_PU(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT, PORT##nr##_IN,		\
-				PORT##nr##_IN_PU)
-
-#define PORT_DATA_IO_PU_PD(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT, PORT##nr##_IN,		\
-				PORT##nr##_IN_PD, PORT##nr##_IN_PU)
-
 static pinmux_enum_t pinmux_data[] = {
 	/* specify valid pin states for each pin in GPIO mode */
 	/* 55-1 (GPIO) */
@@ -1078,13 +1020,9 @@
 	PINMUX_DATA(RESETOUTS_MARK, PORT264_FN1),
 };
 
-#define _GPIO_PORT(pfx, sfx) PINMUX_GPIO(GPIO_PORT##pfx, PORT##pfx##_DATA)
-#define GPIO_PORT_265() _265(_GPIO_PORT, , unused)
-#define GPIO_FN(str) PINMUX_GPIO(GPIO_FN_##str, str##_MARK)
-
 static struct pinmux_gpio pinmux_gpios[] = {
 	/* 55-1 -> 55-5 (GPIO) */
-	GPIO_PORT_265(),
+	GPIO_PORT_ALL(),
 
 	/* Special Pull-up / Pull-down Functions */
 	GPIO_FN(PORT66_KEYIN0_PU), GPIO_FN(PORT67_KEYIN1_PU),
@@ -1362,23 +1300,6 @@
 	GPIO_FN(RESETOUTS),
 };
 
-/* helper for top 4 bits in PORTnCR */
-#define PCRH(in, in_pd, in_pu, out)	\
-	0, (out), (in), 0,	\
-		0, 0, 0, 0,	\
-		0, 0, (in_pd), 0,	\
-		0, 0, (in_pu), 0
-
-#define PORTCR(nr, reg)	\
-	{ PINMUX_CFG_REG("PORT" nr "CR", reg, 8, 4) {	\
-			PCRH(PORT##nr##_IN, PORT##nr##_IN_PD,	\
-				 PORT##nr##_IN_PU, PORT##nr##_OUT),	\
-				PORT##nr##_FN0, PORT##nr##_FN1,	\
-				PORT##nr##_FN2, PORT##nr##_FN3,	\
-				PORT##nr##_FN4, PORT##nr##_FN5,	\
-				PORT##nr##_FN6, PORT##nr##_FN7 }	\
-	}
-
 static struct pinmux_cfg_reg pinmux_config_regs[] = {
 	PORTCR(0, 0xe6050000), /* PORT0CR */
 	PORTCR(1, 0xe6050001), /* PORT1CR */
diff --git a/arch/arm/mach-shmobile/pfc-sh73a0.c b/arch/arm/mach-shmobile/pfc-sh73a0.c
index 5abe02f..e05634c 100644
--- a/arch/arm/mach-shmobile/pfc-sh73a0.c
+++ b/arch/arm/mach-shmobile/pfc-sh73a0.c
@@ -24,83 +24,71 @@
 #include <mach/sh73a0.h>
 #include <mach/irqs.h>
 
-#define _1(fn, pfx, sfx) fn(pfx, sfx)
-
-#define _10(fn, pfx, sfx)				\
-	_1(fn, pfx##0, sfx), _1(fn, pfx##1, sfx),	\
-	_1(fn, pfx##2, sfx), _1(fn, pfx##3, sfx),	\
-	_1(fn, pfx##4, sfx), _1(fn, pfx##5, sfx),	\
-	_1(fn, pfx##6, sfx), _1(fn, pfx##7, sfx),	\
-	_1(fn, pfx##8, sfx), _1(fn, pfx##9, sfx)
-
-#define _310(fn, pfx, sfx)				\
-	_10(fn, pfx,    sfx), _10(fn, pfx##1, sfx),	\
-	_10(fn, pfx##2, sfx), _10(fn, pfx##3, sfx),	\
-	_10(fn, pfx##4, sfx), _10(fn, pfx##5, sfx),	\
-	_10(fn, pfx##6, sfx), _10(fn, pfx##7, sfx),	\
-	_10(fn, pfx##8, sfx), _10(fn, pfx##9, sfx),	\
-	_10(fn, pfx##10, sfx),				\
-	_1(fn, pfx##110, sfx), _1(fn, pfx##111, sfx),	\
-	_1(fn, pfx##112, sfx), _1(fn, pfx##113, sfx),	\
-	_1(fn, pfx##114, sfx), _1(fn, pfx##115, sfx),	\
-	_1(fn, pfx##116, sfx), _1(fn, pfx##117, sfx),	\
-	_1(fn, pfx##118, sfx),				\
-	_1(fn, pfx##128, sfx), _1(fn, pfx##129, sfx),	\
-	_10(fn, pfx##13, sfx), _10(fn, pfx##14, sfx),	\
-	_10(fn, pfx##15, sfx),				\
-	_1(fn, pfx##160, sfx), _1(fn, pfx##161, sfx),	\
-	_1(fn, pfx##162, sfx), _1(fn, pfx##163, sfx),	\
-	_1(fn, pfx##164, sfx),				\
-	_1(fn, pfx##192, sfx), _1(fn, pfx##193, sfx),	\
-	_1(fn, pfx##194, sfx), _1(fn, pfx##195, sfx),	\
-	_1(fn, pfx##196, sfx), _1(fn, pfx##197, sfx),	\
-	_1(fn, pfx##198, sfx), _1(fn, pfx##199, sfx),	\
-	_10(fn, pfx##20, sfx), _10(fn, pfx##21, sfx),	\
-	_10(fn, pfx##22, sfx), _10(fn, pfx##23, sfx),	\
-	_10(fn, pfx##24, sfx), _10(fn, pfx##25, sfx),	\
-	_10(fn, pfx##26, sfx), _10(fn, pfx##27, sfx),	\
-	_1(fn, pfx##280, sfx), _1(fn, pfx##281, sfx),	\
-	_1(fn, pfx##282, sfx),				\
-	_1(fn, pfx##288, sfx), _1(fn, pfx##289, sfx),	\
-	_10(fn, pfx##29, sfx), _10(fn, pfx##30, sfx)
-
-#define _PORT(pfx, sfx) pfx##_##sfx
-#define PORT_310(str) _310(_PORT, PORT, str)
+#define CPU_ALL_PORT(fn, pfx, sfx)				\
+	PORT_10(fn, pfx,    sfx), PORT_10(fn, pfx##1, sfx),	\
+	PORT_10(fn, pfx##2, sfx), PORT_10(fn, pfx##3, sfx),	\
+	PORT_10(fn, pfx##4, sfx), PORT_10(fn, pfx##5, sfx),	\
+	PORT_10(fn, pfx##6, sfx), PORT_10(fn, pfx##7, sfx),	\
+	PORT_10(fn, pfx##8, sfx), PORT_10(fn, pfx##9, sfx),	\
+	PORT_10(fn, pfx##10, sfx),				\
+	PORT_1(fn, pfx##110, sfx), PORT_1(fn, pfx##111, sfx),	\
+	PORT_1(fn, pfx##112, sfx), PORT_1(fn, pfx##113, sfx),	\
+	PORT_1(fn, pfx##114, sfx), PORT_1(fn, pfx##115, sfx),	\
+	PORT_1(fn, pfx##116, sfx), PORT_1(fn, pfx##117, sfx),	\
+	PORT_1(fn, pfx##118, sfx),				\
+	PORT_1(fn, pfx##128, sfx), PORT_1(fn, pfx##129, sfx),	\
+	PORT_10(fn, pfx##13, sfx), PORT_10(fn, pfx##14, sfx),	\
+	PORT_10(fn, pfx##15, sfx),				\
+	PORT_1(fn, pfx##160, sfx), PORT_1(fn, pfx##161, sfx),	\
+	PORT_1(fn, pfx##162, sfx), PORT_1(fn, pfx##163, sfx),	\
+	PORT_1(fn, pfx##164, sfx),				\
+	PORT_1(fn, pfx##192, sfx), PORT_1(fn, pfx##193, sfx),	\
+	PORT_1(fn, pfx##194, sfx), PORT_1(fn, pfx##195, sfx),	\
+	PORT_1(fn, pfx##196, sfx), PORT_1(fn, pfx##197, sfx),	\
+	PORT_1(fn, pfx##198, sfx), PORT_1(fn, pfx##199, sfx),	\
+	PORT_10(fn, pfx##20, sfx), PORT_10(fn, pfx##21, sfx),	\
+	PORT_10(fn, pfx##22, sfx), PORT_10(fn, pfx##23, sfx),	\
+	PORT_10(fn, pfx##24, sfx), PORT_10(fn, pfx##25, sfx),	\
+	PORT_10(fn, pfx##26, sfx), PORT_10(fn, pfx##27, sfx),	\
+	PORT_1(fn, pfx##280, sfx), PORT_1(fn, pfx##281, sfx),	\
+	PORT_1(fn, pfx##282, sfx),				\
+	PORT_1(fn, pfx##288, sfx), PORT_1(fn, pfx##289, sfx),	\
+	PORT_10(fn, pfx##29, sfx), PORT_10(fn, pfx##30, sfx)
 
 enum {
 	PINMUX_RESERVED = 0,
 
 	PINMUX_DATA_BEGIN,
-	PORT_310(DATA),			/* PORT0_DATA -> PORT309_DATA */
+	PORT_ALL(DATA),			/* PORT0_DATA -> PORT309_DATA */
 	PINMUX_DATA_END,
 
 	PINMUX_INPUT_BEGIN,
-	PORT_310(IN),			/* PORT0_IN -> PORT309_IN */
+	PORT_ALL(IN),			/* PORT0_IN -> PORT309_IN */
 	PINMUX_INPUT_END,
 
 	PINMUX_INPUT_PULLUP_BEGIN,
-	PORT_310(IN_PU),		/* PORT0_IN_PU -> PORT309_IN_PU */
+	PORT_ALL(IN_PU),		/* PORT0_IN_PU -> PORT309_IN_PU */
 	PINMUX_INPUT_PULLUP_END,
 
 	PINMUX_INPUT_PULLDOWN_BEGIN,
-	PORT_310(IN_PD),		/* PORT0_IN_PD -> PORT309_IN_PD */
+	PORT_ALL(IN_PD),		/* PORT0_IN_PD -> PORT309_IN_PD */
 	PINMUX_INPUT_PULLDOWN_END,
 
 	PINMUX_OUTPUT_BEGIN,
-	PORT_310(OUT),			/* PORT0_OUT -> PORT309_OUT */
+	PORT_ALL(OUT),			/* PORT0_OUT -> PORT309_OUT */
 	PINMUX_OUTPUT_END,
 
 	PINMUX_FUNCTION_BEGIN,
-	PORT_310(FN_IN),		/* PORT0_FN_IN -> PORT309_FN_IN */
-	PORT_310(FN_OUT),		/* PORT0_FN_OUT -> PORT309_FN_OUT */
-	PORT_310(FN0),			/* PORT0_FN0 -> PORT309_FN0 */
-	PORT_310(FN1),			/* PORT0_FN1 -> PORT309_FN1 */
-	PORT_310(FN2),			/* PORT0_FN2 -> PORT309_FN2 */
-	PORT_310(FN3),			/* PORT0_FN3 -> PORT309_FN3 */
-	PORT_310(FN4),			/* PORT0_FN4 -> PORT309_FN4 */
-	PORT_310(FN5),			/* PORT0_FN5 -> PORT309_FN5 */
-	PORT_310(FN6),			/* PORT0_FN6 -> PORT309_FN6 */
-	PORT_310(FN7),			/* PORT0_FN7 -> PORT309_FN7 */
+	PORT_ALL(FN_IN),		/* PORT0_FN_IN -> PORT309_FN_IN */
+	PORT_ALL(FN_OUT),		/* PORT0_FN_OUT -> PORT309_FN_OUT */
+	PORT_ALL(FN0),			/* PORT0_FN0 -> PORT309_FN0 */
+	PORT_ALL(FN1),			/* PORT0_FN1 -> PORT309_FN1 */
+	PORT_ALL(FN2),			/* PORT0_FN2 -> PORT309_FN2 */
+	PORT_ALL(FN3),			/* PORT0_FN3 -> PORT309_FN3 */
+	PORT_ALL(FN4),			/* PORT0_FN4 -> PORT309_FN4 */
+	PORT_ALL(FN5),			/* PORT0_FN5 -> PORT309_FN5 */
+	PORT_ALL(FN6),			/* PORT0_FN6 -> PORT309_FN6 */
+	PORT_ALL(FN7),			/* PORT0_FN7 -> PORT309_FN7 */
 
 	MSEL2CR_MSEL19_0, MSEL2CR_MSEL19_1,
 	MSEL2CR_MSEL18_0, MSEL2CR_MSEL18_1,
@@ -508,6 +496,14 @@
 	SDHICMD2_PU_MARK,
 	MMCCMD0_PU_MARK,
 	MMCCMD1_PU_MARK,
+	MMCD0_0_PU_MARK,
+	MMCD0_1_PU_MARK,
+	MMCD0_2_PU_MARK,
+	MMCD0_3_PU_MARK,
+	MMCD0_4_PU_MARK,
+	MMCD0_5_PU_MARK,
+	MMCD0_6_PU_MARK,
+	MMCD0_7_PU_MARK,
 	FSIBISLD_PU_MARK,
 	FSIACK_PU_MARK,
 	FSIAILR_PU_MARK,
@@ -517,45 +513,6 @@
 	PINMUX_MARK_END,
 };
 
-#define PORT_DATA_I(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_IN)
-
-#define PORT_DATA_I_PD(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_IN, PORT##nr##_IN_PD)
-
-#define PORT_DATA_I_PU(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_IN, PORT##nr##_IN_PU)
-
-#define PORT_DATA_I_PU_PD(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_IN, PORT##nr##_IN_PD,	\
-				PORT##nr##_IN_PU)
-
-#define PORT_DATA_O(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT)
-
-#define PORT_DATA_IO(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT, PORT##nr##_IN)
-
-#define PORT_DATA_IO_PD(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT, PORT##nr##_IN,		\
-				PORT##nr##_IN_PD)
-
-#define PORT_DATA_IO_PU(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT, PORT##nr##_IN,		\
-				PORT##nr##_IN_PU)
-
-#define PORT_DATA_IO_PU_PD(nr)	\
-	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
-				PORT##nr##_OUT, PORT##nr##_IN,		\
-				PORT##nr##_IN_PD, PORT##nr##_IN_PU)
-
 static pinmux_enum_t pinmux_data[] = {
 	/* specify valid pin states for each pin in GPIO mode */
 
@@ -1561,6 +1518,24 @@
 		MSEL4CR_MSEL15_0),
 	PINMUX_DATA(MMCCMD1_PU_MARK, PORT297_FN2, PORT297_IN_PU,
 		MSEL4CR_MSEL15_1),
+
+	PINMUX_DATA(MMCD0_0_PU_MARK,
+		    PORT271_FN1, PORT271_IN_PU, MSEL4CR_MSEL15_0),
+	PINMUX_DATA(MMCD0_1_PU_MARK,
+		    PORT272_FN1, PORT272_IN_PU, MSEL4CR_MSEL15_0),
+	PINMUX_DATA(MMCD0_2_PU_MARK,
+		    PORT273_FN1, PORT273_IN_PU, MSEL4CR_MSEL15_0),
+	PINMUX_DATA(MMCD0_3_PU_MARK,
+		    PORT274_FN1, PORT274_IN_PU, MSEL4CR_MSEL15_0),
+	PINMUX_DATA(MMCD0_4_PU_MARK,
+		    PORT275_FN1, PORT275_IN_PU, MSEL4CR_MSEL15_0),
+	PINMUX_DATA(MMCD0_5_PU_MARK,
+		    PORT276_FN1, PORT276_IN_PU, MSEL4CR_MSEL15_0),
+	PINMUX_DATA(MMCD0_6_PU_MARK,
+		    PORT277_FN1, PORT277_IN_PU, MSEL4CR_MSEL15_0),
+	PINMUX_DATA(MMCD0_7_PU_MARK,
+		    PORT278_FN1, PORT278_IN_PU, MSEL4CR_MSEL15_0),
+
 	PINMUX_DATA(FSIBISLD_PU_MARK, PORT39_FN1, PORT39_IN_PU),
 	PINMUX_DATA(FSIACK_PU_MARK, PORT49_FN1, PORT49_IN_PU),
 	PINMUX_DATA(FSIAILR_PU_MARK, PORT50_FN5, PORT50_IN_PU),
@@ -1568,12 +1543,8 @@
 	PINMUX_DATA(FSIAISLD_PU_MARK, PORT55_FN1, PORT55_IN_PU),
 };
 
-#define _GPIO_PORT(pfx, sfx) PINMUX_GPIO(GPIO_PORT##pfx, PORT##pfx##_DATA)
-#define GPIO_PORT_310() _310(_GPIO_PORT, , unused)
-#define GPIO_FN(str) PINMUX_GPIO(GPIO_FN_##str, str##_MARK)
-
 static struct pinmux_gpio pinmux_gpios[] = {
-	GPIO_PORT_310(),
+	GPIO_PORT_ALL(),
 
 	/* Table 25-1 (Functions 0-7) */
 	GPIO_FN(VBUS_0),
@@ -2236,24 +2207,20 @@
 	GPIO_FN(SDHICMD2_PU),
 	GPIO_FN(MMCCMD0_PU),
 	GPIO_FN(MMCCMD1_PU),
+	GPIO_FN(MMCD0_0_PU),
+	GPIO_FN(MMCD0_1_PU),
+	GPIO_FN(MMCD0_2_PU),
+	GPIO_FN(MMCD0_3_PU),
+	GPIO_FN(MMCD0_4_PU),
+	GPIO_FN(MMCD0_5_PU),
+	GPIO_FN(MMCD0_6_PU),
+	GPIO_FN(MMCD0_7_PU),
 	GPIO_FN(FSIACK_PU),
 	GPIO_FN(FSIAILR_PU),
 	GPIO_FN(FSIAIBT_PU),
 	GPIO_FN(FSIAISLD_PU),
 };
 
-#define PORTCR(nr, reg)	\
-	{ PINMUX_CFG_REG("PORT" nr "CR", reg, 8, 4) { \
-		0, \
-		/*0001*/ PORT##nr##_OUT , \
-		/*0010*/ PORT##nr##_IN , 0, 0, 0, 0, 0, 0, 0, \
-		/*1010*/ PORT##nr##_IN_PD, 0, 0, 0, \
-		/*1110*/ PORT##nr##_IN_PU, 0, \
-		PORT##nr##_FN0, PORT##nr##_FN1,	PORT##nr##_FN2, \
-		PORT##nr##_FN3,	PORT##nr##_FN4, PORT##nr##_FN5, \
-		PORT##nr##_FN6, PORT##nr##_FN7, 0, 0, 0, 0, 0, 0, 0, 0 } \
-	}
-
 static struct pinmux_cfg_reg pinmux_config_regs[] = {
 	PORTCR(0, 0xe6050000), /* PORT0CR */
 	PORTCR(1, 0xe6050001), /* PORT1CR */
diff --git a/arch/arm/mach-shmobile/pm-sh7372.c b/arch/arm/mach-shmobile/pm-sh7372.c
index 7961273..34bbcbf 100644
--- a/arch/arm/mach-shmobile/pm-sh7372.c
+++ b/arch/arm/mach-shmobile/pm-sh7372.c
@@ -20,6 +20,7 @@
 #include <linux/delay.h>
 #include <linux/irq.h>
 #include <linux/bitrev.h>
+#include <linux/console.h>
 #include <asm/system.h>
 #include <asm/io.h>
 #include <asm/tlbflush.h>
@@ -106,9 +107,8 @@
 	return 0;
 }
 
-static int pd_power_up(struct generic_pm_domain *genpd)
+static int __pd_power_up(struct sh7372_pm_domain *sh7372_pd, bool do_resume)
 {
-	struct sh7372_pm_domain *sh7372_pd = to_sh7372_pd(genpd);
 	unsigned int mask = 1 << sh7372_pd->bit_shift;
 	unsigned int retry_count;
 	int ret = 0;
@@ -123,13 +123,13 @@
 
 	for (retry_count = 2 * PSTR_RETRIES; retry_count; retry_count--) {
 		if (!(__raw_readl(SWUCR) & mask))
-			goto out;
+			break;
 		if (retry_count > PSTR_RETRIES)
 			udelay(PSTR_DELAY_US);
 		else
 			cpu_relax();
 	}
-	if (__raw_readl(SWUCR) & mask)
+	if (!retry_count)
 		ret = -EIO;
 
 	if (!sh7372_pd->no_debug)
@@ -137,12 +137,17 @@
 			 mask, __raw_readl(PSTR));
 
  out:
-	if (ret == 0 && sh7372_pd->resume)
+	if (ret == 0 && sh7372_pd->resume && do_resume)
 		sh7372_pd->resume();
 
 	return ret;
 }
 
+static int pd_power_up(struct generic_pm_domain *genpd)
+{
+	 return __pd_power_up(to_sh7372_pd(genpd), true);
+}
+
 static void sh7372_a4r_suspend(void)
 {
 	sh7372_intcs_suspend();
@@ -174,7 +179,7 @@
 	genpd->active_wakeup = pd_active_wakeup;
 	genpd->power_off = pd_power_down;
 	genpd->power_on = pd_power_up;
-	genpd->power_on(&sh7372_pd->genpd);
+	__pd_power_up(sh7372_pd, false);
 }
 
 void sh7372_add_device_to_domain(struct sh7372_pm_domain *sh7372_pd,
@@ -227,11 +232,23 @@
 	.no_debug = true,
 };
 
+static void sh7372_a3sp_init(void)
+{
+	/* serial consoles make use of SCIF hardware located in A3SP,
+	 * keep such power domain on if "no_console_suspend" is set.
+	 */
+	sh7372_a3sp.stay_on = !console_suspend_enabled;
+}
+
 struct sh7372_pm_domain sh7372_a3sg = {
 	.bit_shift = 13,
 };
 
-#endif /* CONFIG_PM */
+#else /* !CONFIG_PM */
+
+static inline void sh7372_a3sp_init(void) {}
+
+#endif /* !CONFIG_PM */
 
 #if defined(CONFIG_SUSPEND) || defined(CONFIG_CPU_IDLE)
 static int sh7372_do_idle_core_standby(unsigned long unused)
@@ -402,22 +419,18 @@
 
 #ifdef CONFIG_CPU_IDLE
 
-static void sh7372_cpuidle_setup(struct cpuidle_device *dev)
+static void sh7372_cpuidle_setup(struct cpuidle_driver *drv)
 {
-	struct cpuidle_state *state;
-	int i = dev->state_count;
+	struct cpuidle_state *state = &drv->states[drv->state_count];
 
-	state = &dev->states[i];
 	snprintf(state->name, CPUIDLE_NAME_LEN, "C2");
 	strncpy(state->desc, "Core Standby Mode", CPUIDLE_DESC_LEN);
 	state->exit_latency = 10;
 	state->target_residency = 20 + 10;
-	state->power_usage = 1; /* perhaps not */
-	state->flags = 0;
-	state->flags |= CPUIDLE_FLAG_TIME_VALID;
-	shmobile_cpuidle_modes[i] = sh7372_enter_core_standby;
+	state->flags = CPUIDLE_FLAG_TIME_VALID;
+	shmobile_cpuidle_modes[drv->state_count] = sh7372_enter_core_standby;
 
-	dev->state_count = i + 1;
+	drv->state_count++;
 }
 
 static void sh7372_cpuidle_init(void)
@@ -469,6 +482,8 @@
 	/* do not convert A3SM, A3SP, A3SG, A4R power down into A4S */
 	__raw_writel(0, PDNSEL);
 
+	sh7372_a3sp_init();
+
 	sh7372_suspend_init();
 	sh7372_cpuidle_init();
 }
diff --git a/arch/arm/mach-tegra/board-dt.c b/arch/arm/mach-tegra/board-dt.c
index d368f8d..74743ad 100644
--- a/arch/arm/mach-tegra/board-dt.c
+++ b/arch/arm/mach-tegra/board-dt.c
@@ -101,6 +101,13 @@
 
 	tegra_clk_init_from_table(tegra_dt_clk_init_table);
 
+	/*
+	 * Finished with the static registrations now; fill in the missing
+	 * devices
+	 */
+	of_platform_populate(NULL, tegra_dt_match_table,
+				tegra20_auxdata_lookup, NULL);
+
 	for (i = 0; i < ARRAY_SIZE(pinmux_configs); i++) {
 		if (of_machine_is_compatible(pinmux_configs[i].machine)) {
 			pinmux_configs[i].init();
@@ -110,12 +117,6 @@
 
 	WARN(i == ARRAY_SIZE(pinmux_configs),
 		"Unknown platform! Pinmuxing not initialized\n");
-
-	/*
-	 * Finished with the static registrations now; fill in the missing
-	 * devices
-	 */
-	of_platform_populate(NULL, tegra_dt_match_table, tegra20_auxdata_lookup, NULL);
 }
 
 static const char * tegra_dt_board_compat[] = {
diff --git a/arch/arm/mach-tegra/board-harmony-pinmux.c b/arch/arm/mach-tegra/board-harmony-pinmux.c
index e99b456..7a4a26d 100644
--- a/arch/arm/mach-tegra/board-harmony-pinmux.c
+++ b/arch/arm/mach-tegra/board-harmony-pinmux.c
@@ -16,6 +16,8 @@
 
 #include <linux/kernel.h>
 #include <linux/gpio.h>
+#include <linux/of.h>
+
 #include <mach/pinmux.h>
 
 #include "gpio-names.h"
@@ -161,7 +163,9 @@
 
 void harmony_pinmux_init(void)
 {
-	platform_add_devices(pinmux_devices, ARRAY_SIZE(pinmux_devices));
+	if (!of_machine_is_compatible("nvidia,tegra20"))
+		platform_add_devices(pinmux_devices,
+					ARRAY_SIZE(pinmux_devices));
 
 	tegra_pinmux_config_table(harmony_pinmux, ARRAY_SIZE(harmony_pinmux));
 
diff --git a/arch/arm/mach-tegra/board-paz00-pinmux.c b/arch/arm/mach-tegra/board-paz00-pinmux.c
index fb20894..be30e21 100644
--- a/arch/arm/mach-tegra/board-paz00-pinmux.c
+++ b/arch/arm/mach-tegra/board-paz00-pinmux.c
@@ -16,6 +16,8 @@
 
 #include <linux/kernel.h>
 #include <linux/gpio.h>
+#include <linux/of.h>
+
 #include <mach/pinmux.h>
 
 #include "gpio-names.h"
@@ -158,7 +160,9 @@
 
 void paz00_pinmux_init(void)
 {
-	platform_add_devices(pinmux_devices, ARRAY_SIZE(pinmux_devices));
+	if (!of_machine_is_compatible("nvidia,tegra20"))
+		platform_add_devices(pinmux_devices,
+					ARRAY_SIZE(pinmux_devices));
 
 	tegra_pinmux_config_table(paz00_pinmux, ARRAY_SIZE(paz00_pinmux));
 
diff --git a/arch/arm/mach-tegra/board-seaboard-pinmux.c b/arch/arm/mach-tegra/board-seaboard-pinmux.c
index fbce31d..b1c2972 100644
--- a/arch/arm/mach-tegra/board-seaboard-pinmux.c
+++ b/arch/arm/mach-tegra/board-seaboard-pinmux.c
@@ -16,6 +16,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/gpio.h>
+#include <linux/of.h>
 
 #include <mach/pinmux.h>
 #include <mach/pinmux-t2.h>
@@ -191,6 +192,7 @@
 	{ .gpio = TEGRA_GPIO_SD2_POWER,		.enable = true },
 	{ .gpio = TEGRA_GPIO_LIDSWITCH,		.enable = true },
 	{ .gpio = TEGRA_GPIO_POWERKEY,		.enable = true },
+	{ .gpio = TEGRA_GPIO_HP_DET,		.enable = true },
 	{ .gpio = TEGRA_GPIO_ISL29018_IRQ,	.enable = true },
 	{ .gpio = TEGRA_GPIO_CDC_IRQ,		.enable = true },
 	{ .gpio = TEGRA_GPIO_USB1,		.enable = true },
@@ -218,7 +220,9 @@
 
 void __init seaboard_common_pinmux_init(void)
 {
-	platform_add_devices(pinmux_devices, ARRAY_SIZE(pinmux_devices));
+	if (!of_machine_is_compatible("nvidia,tegra20"))
+		platform_add_devices(pinmux_devices,
+					ARRAY_SIZE(pinmux_devices));
 
 	tegra_pinmux_config_table(seaboard_pinmux, ARRAY_SIZE(seaboard_pinmux));
 
diff --git a/arch/arm/mach-tegra/board-trimslice-pinmux.c b/arch/arm/mach-tegra/board-trimslice-pinmux.c
index 4969dd2..7ab719d 100644
--- a/arch/arm/mach-tegra/board-trimslice-pinmux.c
+++ b/arch/arm/mach-tegra/board-trimslice-pinmux.c
@@ -16,6 +16,7 @@
 #include <linux/gpio.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
+#include <linux/of.h>
 
 #include <mach/pinmux.h>
 
@@ -157,7 +158,9 @@
 
 void __init trimslice_pinmux_init(void)
 {
-	platform_add_devices(pinmux_devices, ARRAY_SIZE(pinmux_devices));
+	if (!of_machine_is_compatible("nvidia,tegra20"))
+		platform_add_devices(pinmux_devices,
+					ARRAY_SIZE(pinmux_devices));
 	tegra_pinmux_config_table(trimslice_pinmux, ARRAY_SIZE(trimslice_pinmux));
 	tegra_gpio_config(gpio_table, ARRAY_SIZE(gpio_table));
 }
diff --git a/arch/arm/mach-w90x900/dev.c b/arch/arm/mach-w90x900/dev.c
index 7a1fa6a..5b0c38a 100644
--- a/arch/arm/mach-w90x900/dev.c
+++ b/arch/arm/mach-w90x900/dev.c
@@ -422,7 +422,7 @@
 
 /* LCD controller*/
 
-static struct nuc900fb_display __initdata nuc900_lcd_info[] = {
+static struct nuc900fb_display nuc900_lcd_info[] = {
 	/* Giantplus Technology GPM1040A0 320x240 Color TFT LCD */
 	[0] = {
 		.type		= LCM_DCCS_VA_SRC_RGB565,
@@ -445,7 +445,7 @@
 	},
 };
 
-static struct nuc900fb_mach_info nuc900_fb_info __initdata = {
+static struct nuc900fb_mach_info nuc900_fb_info = {
 #if defined(CONFIG_GPM1040A0_320X240)
 	.displays		= &nuc900_lcd_info[0],
 #else
diff --git a/arch/arm/mach-w90x900/include/mach/mfp.h b/arch/arm/mach-w90x900/include/mach/mfp.h
index 94c0e71..23ef1f5 100644
--- a/arch/arm/mach-w90x900/include/mach/mfp.h
+++ b/arch/arm/mach-w90x900/include/mach/mfp.h
@@ -19,6 +19,7 @@
 extern void mfp_set_groupf(struct device *dev);
 extern void mfp_set_groupc(struct device *dev);
 extern void mfp_set_groupi(struct device *dev);
-extern void mfp_set_groupg(struct device *dev);
+extern void mfp_set_groupg(struct device *dev, const char *subname);
+extern void mfp_set_groupd(struct device *dev, const char *subname);
 
 #endif /* __ASM_ARCH_MFP_H */
diff --git a/arch/arm/mach-w90x900/include/mach/nuc900_spi.h b/arch/arm/mach-w90x900/include/mach/nuc900_spi.h
index bd94819..2c4e0c1 100644
--- a/arch/arm/mach-w90x900/include/mach/nuc900_spi.h
+++ b/arch/arm/mach-w90x900/include/mach/nuc900_spi.h
@@ -14,7 +14,7 @@
 #ifndef __ASM_ARCH_SPI_H
 #define __ASM_ARCH_SPI_H
 
-extern void mfp_set_groupg(struct device *dev);
+extern void mfp_set_groupg(struct device *dev, const char *subname);
 
 struct nuc900_spi_info {
 	unsigned int num_cs;
diff --git a/arch/arm/mach-w90x900/mfp.c b/arch/arm/mach-w90x900/mfp.c
index fb7fb62..9dd7461 100644
--- a/arch/arm/mach-w90x900/mfp.c
+++ b/arch/arm/mach-w90x900/mfp.c
@@ -26,10 +26,8 @@
 #define REG_MFSEL	(W90X900_VA_GCR + 0xC)
 
 #define GPSELF		(0x01 << 1)
-
 #define GPSELC		(0x03 << 2)
-#define ENKPI		(0x02 << 2)
-#define ENNAND		(0x01 << 2)
+#define GPSELD		(0x0f << 4)
 
 #define GPSELEI0	(0x01 << 26)
 #define GPSELEI1	(0x01 << 27)
@@ -37,11 +35,16 @@
 #define GPIOG0TO1	(0x03 << 14)
 #define GPIOG2TO3	(0x03 << 16)
 #define GPIOG22TO23	(0x03 << 22)
+#define GPIOG18TO20	(0x07 << 18)
 
 #define ENSPI		(0x0a << 14)
 #define ENI2C0		(0x01 << 14)
 #define ENI2C1		(0x01 << 16)
 #define ENAC97		(0x02 << 22)
+#define ENSD1		(0x02 << 18)
+#define ENSD0		(0x0a << 4)
+#define ENKPI		(0x02 << 2)
+#define ENNAND		(0x01 << 2)
 
 static DEFINE_MUTEX(mfp_mutex);
 
@@ -127,16 +130,19 @@
 }
 EXPORT_SYMBOL(mfp_set_groupi);
 
-void mfp_set_groupg(struct device *dev)
+void mfp_set_groupg(struct device *dev, const char *subname)
 {
 	unsigned long mfpen;
 	const char *dev_id;
 
-	BUG_ON(!dev);
+	BUG_ON((!dev) && (!subname));
 
 	mutex_lock(&mfp_mutex);
 
-	dev_id = dev_name(dev);
+	if (subname != NULL)
+		dev_id = subname;
+	else
+		dev_id = dev_name(dev);
 
 	mfpen = __raw_readl(REG_MFSEL);
 
@@ -152,6 +158,9 @@
 	} else if (strcmp(dev_id, "nuc900-audio") == 0) {
 		mfpen &= ~(GPIOG22TO23);
 		mfpen |= ENAC97;/*enable AC97*/
+	} else if (strcmp(dev_id, "nuc900-mmc-port1") == 0) {
+		mfpen &= ~(GPIOG18TO20);
+		mfpen |= (ENSD1 | 0x01);/*enable sd1*/
 	} else {
 		mfpen &= ~(GPIOG0TO1 | GPIOG2TO3);/*GPIOG[3:0]*/
 	}
@@ -162,3 +171,30 @@
 }
 EXPORT_SYMBOL(mfp_set_groupg);
 
+void mfp_set_groupd(struct device *dev, const char *subname)
+{
+	unsigned long mfpen;
+	const char *dev_id;
+
+	BUG_ON((!dev) && (!subname));
+
+	mutex_lock(&mfp_mutex);
+
+	if (subname != NULL)
+		dev_id = subname;
+	else
+		dev_id = dev_name(dev);
+
+	mfpen = __raw_readl(REG_MFSEL);
+
+	if (strcmp(dev_id, "nuc900-mmc-port0") == 0) {
+		mfpen &= ~GPSELD;/*enable sd0*/
+		mfpen |= ENSD0;
+	} else
+		mfpen &= (~GPSELD);
+
+	__raw_writel(mfpen, REG_MFSEL);
+
+	mutex_unlock(&mfp_mutex);
+}
+EXPORT_SYMBOL(mfp_set_groupd);
diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c
index 8ac9e9f..b1e192b 100644
--- a/arch/arm/mm/cache-l2x0.c
+++ b/arch/arm/mm/cache-l2x0.c
@@ -61,7 +61,7 @@
 {
 	void __iomem *base = l2x0_base;
 
-#ifdef CONFIG_ARM_ERRATA_753970
+#ifdef CONFIG_PL310_ERRATA_753970
 	/* write to an unmmapped register */
 	writel_relaxed(0, base + L2X0_DUMMY_REG);
 #else
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index e4e7f6c..1aa664a 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -168,7 +168,7 @@
 	pte_t *pte;
 	int i = 0;
 	unsigned long base = consistent_base;
-	unsigned long num_ptes = (CONSISTENT_END - base) >> PGDIR_SHIFT;
+	unsigned long num_ptes = (CONSISTENT_END - base) >> PMD_SHIFT;
 
 	consistent_pte = kmalloc(num_ptes * sizeof(pte_t), GFP_KERNEL);
 	if (!consistent_pte) {
@@ -332,6 +332,15 @@
 	struct page *page;
 	void *addr;
 
+	/*
+	 * Following is a work-around (a.k.a. hack) to prevent pages
+	 * with __GFP_COMP being passed to split_page() which cannot
+	 * handle them.  The real problem is that this flag probably
+	 * should be 0 on ARM as it is not supported on this
+	 * platform; see CONFIG_HUGETLBFS.
+	 */
+	gfp &= ~(__GFP_COMP);
+
 	*handle = ~0;
 	size = PAGE_ALIGN(size);
 
diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c
index 74be05f..44b628e 100644
--- a/arch/arm/mm/mmap.c
+++ b/arch/arm/mm/mmap.c
@@ -9,8 +9,7 @@
 #include <linux/io.h>
 #include <linux/personality.h>
 #include <linux/random.h>
-#include <asm/cputype.h>
-#include <asm/system.h>
+#include <asm/cachetype.h>
 
 #define COLOUR_ALIGN(addr,pgoff)		\
 	((((addr)+SHMLBA-1)&~(SHMLBA-1)) +	\
@@ -32,25 +31,15 @@
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned long start_addr;
-#if defined(CONFIG_CPU_V6) || defined(CONFIG_CPU_V6K)
-	unsigned int cache_type;
-	int do_align = 0, aliasing = 0;
+	int do_align = 0;
+	int aliasing = cache_is_vipt_aliasing();
 
 	/*
 	 * We only need to do colour alignment if either the I or D
-	 * caches alias.  This is indicated by bits 9 and 21 of the
-	 * cache type register.
+	 * caches alias.
 	 */
-	cache_type = read_cpuid_cachetype();
-	if (cache_type != read_cpuid_id()) {
-		aliasing = (cache_type | cache_type >> 12) & (1 << 11);
-		if (aliasing)
-			do_align = filp || flags & MAP_SHARED;
-	}
-#else
-#define do_align 0
-#define aliasing 0
-#endif
+	if (aliasing)
+		do_align = filp || (flags & MAP_SHARED);
 
 	/*
 	 * We enforce the MAP_FIXED case.
diff --git a/arch/arm/plat-mxc/Kconfig b/arch/arm/plat-mxc/Kconfig
index a08a951..b3a1f2b3 100644
--- a/arch/arm/plat-mxc/Kconfig
+++ b/arch/arm/plat-mxc/Kconfig
@@ -10,7 +10,7 @@
 
 config ARCH_IMX_V4_V5
 	bool "i.MX1, i.MX21, i.MX25, i.MX27"
-	select AUTO_ZRELADDR
+	select AUTO_ZRELADDR if !ZBOOT_ROM
 	select ARM_PATCH_PHYS_VIRT
 	help
 	  This enables support for systems based on the Freescale i.MX ARMv4
@@ -26,7 +26,7 @@
 
 config ARCH_MX5
 	bool "i.MX50, i.MX51, i.MX53"
-	select AUTO_ZRELADDR
+	select AUTO_ZRELADDR if !ZBOOT_ROM
 	select ARM_PATCH_PHYS_VIRT
 	help
 	  This enables support for machines using Freescale's i.MX50 and i.MX53
diff --git a/arch/arm/plat-mxc/avic.c b/arch/arm/plat-mxc/avic.c
index 8875fb4..55f1569 100644
--- a/arch/arm/plat-mxc/avic.c
+++ b/arch/arm/plat-mxc/avic.c
@@ -22,6 +22,7 @@
 #include <linux/io.h>
 #include <mach/common.h>
 #include <asm/mach/irq.h>
+#include <asm/exception.h>
 #include <mach/hardware.h>
 
 #include "irq-common.h"
diff --git a/arch/arm/plat-mxc/cpufreq.c b/arch/arm/plat-mxc/cpufreq.c
index 74aac96..adbff70 100644
--- a/arch/arm/plat-mxc/cpufreq.c
+++ b/arch/arm/plat-mxc/cpufreq.c
@@ -17,6 +17,7 @@
  * the CPU clock speed on the fly.
  */
 
+#include <linux/module.h>
 #include <linux/cpufreq.h>
 #include <linux/clk.h>
 #include <linux/err.h>
diff --git a/arch/arm/plat-mxc/gic.c b/arch/arm/plat-mxc/gic.c
index b3b8eed..12f8f81 100644
--- a/arch/arm/plat-mxc/gic.c
+++ b/arch/arm/plat-mxc/gic.c
@@ -28,21 +28,14 @@
 		if (irqnr == 1023)
 			break;
 
-		if (irqnr > 29 && irqnr < 1021)
+		if (irqnr > 15 && irqnr < 1021)
 			handle_IRQ(irqnr, regs);
 #ifdef CONFIG_SMP
-		else if (irqnr < 16) {
+		else {
 			writel_relaxed(irqstat, gic_cpu_base_addr +
 						GIC_CPU_EOI);
 			handle_IPI(irqnr, regs);
 		}
 #endif
-#ifdef CONFIG_LOCAL_TIMERS
-		else if (irqnr == 29) {
-			writel_relaxed(irqstat, gic_cpu_base_addr +
-						GIC_CPU_EOI);
-			handle_local_timer(regs);
-		}
-#endif
 	} while (1);
 }
diff --git a/arch/arm/plat-mxc/include/mach/common.h b/arch/arm/plat-mxc/include/mach/common.h
index 83b745a..c75f254 100644
--- a/arch/arm/plat-mxc/include/mach/common.h
+++ b/arch/arm/plat-mxc/include/mach/common.h
@@ -85,7 +85,6 @@
 };
 
 extern void mx5_cpu_lp_set(enum mxc_cpu_pwr_mode mode);
-extern void (*imx_idle)(void);
 extern void imx_print_silicon_rev(const char *cpu, int srev);
 
 void avic_handle_irq(struct pt_regs *);
@@ -133,4 +132,5 @@
 extern void imx53_smd_common_init(void);
 extern int imx6q_set_lpm(enum mxc_cpu_pwr_mode mode);
 extern void imx6q_pm_init(void);
+extern void imx6q_clock_map_io(void);
 #endif
diff --git a/arch/arm/plat-mxc/include/mach/entry-macro.S b/arch/arm/plat-mxc/include/mach/entry-macro.S
index 9fe0dfc..ca5cf26 100644
--- a/arch/arm/plat-mxc/include/mach/entry-macro.S
+++ b/arch/arm/plat-mxc/include/mach/entry-macro.S
@@ -25,6 +25,3 @@
 
 	.macro test_for_ipi, irqnr, irqstat, base, tmp
 	.endm
-
-	.macro test_for_ltirq, irqnr, irqstat, base, tmp
-	.endm
diff --git a/arch/arm/plat-mxc/include/mach/mxc.h b/arch/arm/plat-mxc/include/mach/mxc.h
index 00a7819..a4d36d601 100644
--- a/arch/arm/plat-mxc/include/mach/mxc.h
+++ b/arch/arm/plat-mxc/include/mach/mxc.h
@@ -50,20 +50,6 @@
 #define IMX_CHIP_REVISION_3_3		0x33
 #define IMX_CHIP_REVISION_UNKNOWN	0xff
 
-#define IMX_CHIP_REVISION_1_0_STRING		"1.0"
-#define IMX_CHIP_REVISION_1_1_STRING		"1.1"
-#define IMX_CHIP_REVISION_1_2_STRING		"1.2"
-#define IMX_CHIP_REVISION_1_3_STRING		"1.3"
-#define IMX_CHIP_REVISION_2_0_STRING		"2.0"
-#define IMX_CHIP_REVISION_2_1_STRING		"2.1"
-#define IMX_CHIP_REVISION_2_2_STRING		"2.2"
-#define IMX_CHIP_REVISION_2_3_STRING		"2.3"
-#define IMX_CHIP_REVISION_3_0_STRING		"3.0"
-#define IMX_CHIP_REVISION_3_1_STRING		"3.1"
-#define IMX_CHIP_REVISION_3_2_STRING		"3.2"
-#define IMX_CHIP_REVISION_3_3_STRING		"3.3"
-#define IMX_CHIP_REVISION_UNKNOWN_STRING	"unknown"
-
 #ifndef __ASSEMBLY__
 extern unsigned int __mxc_cpu_type;
 #endif
diff --git a/arch/arm/plat-mxc/include/mach/system.h b/arch/arm/plat-mxc/include/mach/system.h
index cf88b35..b9895d2 100644
--- a/arch/arm/plat-mxc/include/mach/system.h
+++ b/arch/arm/plat-mxc/include/mach/system.h
@@ -17,14 +17,9 @@
 #ifndef __ASM_ARCH_MXC_SYSTEM_H__
 #define __ASM_ARCH_MXC_SYSTEM_H__
 
-extern void (*imx_idle)(void);
-
 static inline void arch_idle(void)
 {
-	if (imx_idle != NULL)
-		(imx_idle)();
-	else
-		cpu_do_idle();
+	cpu_do_idle();
 }
 
 void arch_reset(char mode, const char *cmd);
diff --git a/arch/arm/plat-mxc/pwm.c b/arch/arm/plat-mxc/pwm.c
index 42d74ea..845de59 100644
--- a/arch/arm/plat-mxc/pwm.c
+++ b/arch/arm/plat-mxc/pwm.c
@@ -32,6 +32,9 @@
 #define MX3_PWMSAR                0x0C    /* PWM Sample Register */
 #define MX3_PWMPR                 0x10    /* PWM Period Register */
 #define MX3_PWMCR_PRESCALER(x)    (((x - 1) & 0xFFF) << 4)
+#define MX3_PWMCR_DOZEEN                (1 << 24)
+#define MX3_PWMCR_WAITEN                (1 << 23)
+#define MX3_PWMCR_DBGEN			(1 << 22)
 #define MX3_PWMCR_CLKSRC_IPG_HIGH (2 << 16)
 #define MX3_PWMCR_CLKSRC_IPG      (1 << 16)
 #define MX3_PWMCR_EN              (1 << 0)
@@ -77,7 +80,9 @@
 		writel(duty_cycles, pwm->mmio_base + MX3_PWMSAR);
 		writel(period_cycles, pwm->mmio_base + MX3_PWMPR);
 
-		cr = MX3_PWMCR_PRESCALER(prescale) | MX3_PWMCR_EN;
+		cr = MX3_PWMCR_PRESCALER(prescale) |
+			MX3_PWMCR_DOZEEN | MX3_PWMCR_WAITEN |
+			MX3_PWMCR_DBGEN | MX3_PWMCR_EN;
 
 		if (cpu_is_mx25())
 			cr |= MX3_PWMCR_CLKSRC_IPG;
diff --git a/arch/arm/plat-mxc/system.c b/arch/arm/plat-mxc/system.c
index 9dad8dc..d65fb31 100644
--- a/arch/arm/plat-mxc/system.c
+++ b/arch/arm/plat-mxc/system.c
@@ -21,6 +21,7 @@
 #include <linux/io.h>
 #include <linux/err.h>
 #include <linux/delay.h>
+#include <linux/module.h>
 
 #include <mach/hardware.h>
 #include <mach/common.h>
@@ -28,8 +29,8 @@
 #include <asm/system.h>
 #include <asm/mach-types.h>
 
-void (*imx_idle)(void) = NULL;
 void __iomem *(*imx_ioremap)(unsigned long, size_t, unsigned int) = NULL;
+EXPORT_SYMBOL_GPL(imx_ioremap);
 
 static void __iomem *wdog_base;
 
diff --git a/arch/arm/plat-mxc/tzic.c b/arch/arm/plat-mxc/tzic.c
index e993a184..a3c164c 100644
--- a/arch/arm/plat-mxc/tzic.c
+++ b/arch/arm/plat-mxc/tzic.c
@@ -17,6 +17,7 @@
 #include <linux/io.h>
 
 #include <asm/mach/irq.h>
+#include <asm/exception.h>
 
 #include <mach/hardware.h>
 #include <mach/common.h>
diff --git a/arch/arm/plat-omap/include/plat/clock.h b/arch/arm/plat-omap/include/plat/clock.h
index 197ca03..eb73ab4 100644
--- a/arch/arm/plat-omap/include/plat/clock.h
+++ b/arch/arm/plat-omap/include/plat/clock.h
@@ -165,8 +165,8 @@
 	u8			auto_recal_bit;
 	u8			recal_en_bit;
 	u8			recal_st_bit;
-	u8			flags;
 #  endif
+	u8			flags;
 };
 
 #endif
diff --git a/arch/arm/plat-omap/include/plat/common.h b/arch/arm/plat-omap/include/plat/common.h
index c50df48..3ff3e36 100644
--- a/arch/arm/plat-omap/include/plat/common.h
+++ b/arch/arm/plat-omap/include/plat/common.h
@@ -30,6 +30,7 @@
 #include <linux/delay.h>
 
 #include <plat/i2c.h>
+#include <plat/omap_hwmod.h>
 
 struct sys_timer;
 
@@ -55,6 +56,8 @@
 void ti816x_init_early(void);
 void omap4430_init_early(void);
 
+extern int omap_dss_reset(struct omap_hwmod *);
+
 void omap_sram_init(void);
 
 /*
diff --git a/arch/arm/plat-s3c24xx/cpu-freq-debugfs.c b/arch/arm/plat-s3c24xx/cpu-freq-debugfs.c
index a927666..c7adad0 100644
--- a/arch/arm/plat-s3c24xx/cpu-freq-debugfs.c
+++ b/arch/arm/plat-s3c24xx/cpu-freq-debugfs.c
@@ -12,7 +12,7 @@
 */
 
 #include <linux/init.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/interrupt.h>
 #include <linux/ioport.h>
 #include <linux/cpufreq.h>
diff --git a/arch/arm/plat-s5p/sysmmu.c b/arch/arm/plat-s5p/sysmmu.c
index e1cbc72..c8bec9c 100644
--- a/arch/arm/plat-s5p/sysmmu.c
+++ b/arch/arm/plat-s5p/sysmmu.c
@@ -11,6 +11,7 @@
 #include <linux/io.h>
 #include <linux/interrupt.h>
 #include <linux/platform_device.h>
+#include <linux/export.h>
 
 #include <asm/pgtable.h>
 
diff --git a/arch/arm/plat-samsung/dev-backlight.c b/arch/arm/plat-samsung/dev-backlight.c
index e657305..a976c02 100644
--- a/arch/arm/plat-samsung/dev-backlight.c
+++ b/arch/arm/plat-samsung/dev-backlight.c
@@ -15,7 +15,6 @@
 #include <linux/slab.h>
 #include <linux/io.h>
 #include <linux/pwm_backlight.h>
-#include <linux/slab.h>
 
 #include <plat/devs.h>
 #include <plat/gpio-cfg.h>
diff --git a/arch/arm/plat-samsung/include/plat/gpio-cfg.h b/arch/arm/plat-samsung/include/plat/gpio-cfg.h
index d48245b..df8155b 100644
--- a/arch/arm/plat-samsung/include/plat/gpio-cfg.h
+++ b/arch/arm/plat-samsung/include/plat/gpio-cfg.h
@@ -24,6 +24,8 @@
 #ifndef __PLAT_GPIO_CFG_H
 #define __PLAT_GPIO_CFG_H __FILE__
 
+#include<linux/types.h>
+
 typedef unsigned int __bitwise__ samsung_gpio_pull_t;
 typedef unsigned int __bitwise__ s5p_gpio_drvstr_t;
 
diff --git a/arch/arm/plat-samsung/pd.c b/arch/arm/plat-samsung/pd.c
index efe1d56..312b510 100644
--- a/arch/arm/plat-samsung/pd.c
+++ b/arch/arm/plat-samsung/pd.c
@@ -11,7 +11,7 @@
 */
 
 #include <linux/init.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 #include <linux/err.h>
 #include <linux/pm_runtime.h>
diff --git a/arch/arm/plat-samsung/pwm.c b/arch/arm/plat-samsung/pwm.c
index dc1185d..c559d84 100644
--- a/arch/arm/plat-samsung/pwm.c
+++ b/arch/arm/plat-samsung/pwm.c
@@ -11,7 +11,7 @@
  * the Free Software Foundation; either version 2 of the License.
 */
 
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/kernel.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
diff --git a/arch/arm/tools/mach-types b/arch/arm/tools/mach-types
index 5bdeef9..ccbe16f 100644
--- a/arch/arm/tools/mach-types
+++ b/arch/arm/tools/mach-types
@@ -1123,5 +1123,6 @@
 thales_adc		MACH_THALES_ADC		THALES_ADC		3492
 ubisys_p9d_evp		MACH_UBISYS_P9D_EVP	UBISYS_P9D_EVP		3493
 atdgp318		MACH_ATDGP318		ATDGP318		3494
+m28evk			MACH_M28EVK		M28EVK			3613
 smdk4212		MACH_SMDK4212		SMDK4212		3638
 smdk4412		MACH_SMDK4412		SMDK4412		3765
diff --git a/arch/blackfin/include/asm/bfin_serial.h b/arch/blackfin/include/asm/bfin_serial.h
index 7fd0ec7..ecacdf3 100644
--- a/arch/blackfin/include/asm/bfin_serial.h
+++ b/arch/blackfin/include/asm/bfin_serial.h
@@ -32,6 +32,8 @@
 struct bfin_serial_port {
 	struct uart_port port;
 	unsigned int old_status;
+	int tx_irq;
+	int rx_irq;
 	int status_irq;
 #ifndef BFIN_UART_BF54X_STYLE
 	unsigned int lsr;
diff --git a/arch/blackfin/mach-bf518/boards/ezbrd.c b/arch/blackfin/mach-bf518/boards/ezbrd.c
index 1082e49..d1c0c0c 100644
--- a/arch/blackfin/mach-bf518/boards/ezbrd.c
+++ b/arch/blackfin/mach-bf518/boards/ezbrd.c
@@ -373,8 +373,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -416,8 +421,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf518/boards/tcm-bf518.c b/arch/blackfin/mach-bf518/boards/tcm-bf518.c
index 55c1279..5470bf8 100644
--- a/arch/blackfin/mach-bf518/boards/tcm-bf518.c
+++ b/arch/blackfin/mach-bf518/boards/tcm-bf518.c
@@ -309,8 +309,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -352,8 +357,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf527/boards/ad7160eval.c b/arch/blackfin/mach-bf527/boards/ad7160eval.c
index 8d65d47..5bc6938 100644
--- a/arch/blackfin/mach-bf527/boards/ad7160eval.c
+++ b/arch/blackfin/mach-bf527/boards/ad7160eval.c
@@ -381,8 +381,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -424,8 +429,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf527/boards/cm_bf527.c b/arch/blackfin/mach-bf527/boards/cm_bf527.c
index 6410fc1..cd28969 100644
--- a/arch/blackfin/mach-bf527/boards/cm_bf527.c
+++ b/arch/blackfin/mach-bf527/boards/cm_bf527.c
@@ -8,6 +8,7 @@
  */
 
 #include <linux/device.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
@@ -539,8 +540,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -582,8 +588,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -801,7 +812,6 @@
 #if defined(CONFIG_KEYBOARD_GPIO) || defined(CONFIG_KEYBOARD_GPIO_MODULE)
 #include <linux/input.h>
 #include <linux/gpio_keys.h>
-#include <linux/export.h>
 
 static struct gpio_keys_button bfin_gpio_keys_table[] = {
 	{BTN_0, GPIO_PF14, 1, "gpio-keys: BTN0"},
diff --git a/arch/blackfin/mach-bf527/boards/ezbrd.c b/arch/blackfin/mach-bf527/boards/ezbrd.c
index 64f7278..9f792ea 100644
--- a/arch/blackfin/mach-bf527/boards/ezbrd.c
+++ b/arch/blackfin/mach-bf527/boards/ezbrd.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/device.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
@@ -417,8 +418,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -460,8 +466,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -674,7 +685,6 @@
 #if defined(CONFIG_KEYBOARD_GPIO) || defined(CONFIG_KEYBOARD_GPIO_MODULE)
 #include <linux/input.h>
 #include <linux/gpio_keys.h>
-#include <linux/export.h>
 
 static struct gpio_keys_button bfin_gpio_keys_table[] = {
 	{BTN_0, GPIO_PG0, 1, "gpio-keys: BTN0"},
diff --git a/arch/blackfin/mach-bf527/boards/ezkit.c b/arch/blackfin/mach-bf527/boards/ezkit.c
index e4c6a12..3ecafff 100644
--- a/arch/blackfin/mach-bf527/boards/ezkit.c
+++ b/arch/blackfin/mach-bf527/boards/ezkit.c
@@ -711,8 +711,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -754,8 +759,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf527/boards/tll6527m.c b/arch/blackfin/mach-bf527/boards/tll6527m.c
index 76dbc03..3a92c43 100644
--- a/arch/blackfin/mach-bf527/boards/tll6527m.c
+++ b/arch/blackfin/mach-bf527/boards/tll6527m.c
@@ -496,8 +496,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -540,8 +545,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf533/boards/H8606.c b/arch/blackfin/mach-bf533/boards/H8606.c
index 5da5787..47cadd3 100644
--- a/arch/blackfin/mach-bf533/boards/H8606.c
+++ b/arch/blackfin/mach-bf533/boards/H8606.c
@@ -238,8 +238,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX + 1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf533/boards/blackstamp.c b/arch/blackfin/mach-bf533/boards/blackstamp.c
index b0ec825..18817d5 100644
--- a/arch/blackfin/mach-bf533/boards/blackstamp.c
+++ b/arch/blackfin/mach-bf533/boards/blackstamp.c
@@ -193,8 +193,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX + 1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf533/boards/cm_bf533.c b/arch/blackfin/mach-bf533/boards/cm_bf533.c
index 14f54a3..2c8f30e 100644
--- a/arch/blackfin/mach-bf533/boards/cm_bf533.c
+++ b/arch/blackfin/mach-bf533/boards/cm_bf533.c
@@ -221,8 +221,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX + 1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf533/boards/ezkit.c b/arch/blackfin/mach-bf533/boards/ezkit.c
index ecd2801..144556e 100644
--- a/arch/blackfin/mach-bf533/boards/ezkit.c
+++ b/arch/blackfin/mach-bf533/boards/ezkit.c
@@ -292,8 +292,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX + 1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf533/boards/ip0x.c b/arch/blackfin/mach-bf533/boards/ip0x.c
index fbee77f..b597d4e 100644
--- a/arch/blackfin/mach-bf533/boards/ip0x.c
+++ b/arch/blackfin/mach-bf533/boards/ip0x.c
@@ -151,8 +151,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX + 1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf533/boards/stamp.c b/arch/blackfin/mach-bf533/boards/stamp.c
index 964a8e5..2afd02e 100644
--- a/arch/blackfin/mach-bf533/boards/stamp.c
+++ b/arch/blackfin/mach-bf533/boards/stamp.c
@@ -298,8 +298,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX + 1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf537/boards/cm_bf537e.c b/arch/blackfin/mach-bf537/boards/cm_bf537e.c
index 1471c51..604a430 100644
--- a/arch/blackfin/mach-bf537/boards/cm_bf537e.c
+++ b/arch/blackfin/mach-bf537/boards/cm_bf537e.c
@@ -8,6 +8,7 @@
  */
 
 #include <linux/device.h>
+#include <linux/export.h>
 #include <linux/etherdevice.h>
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
@@ -305,8 +306,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -366,8 +372,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -569,7 +580,6 @@
 
 #if defined(CONFIG_BFIN_MAC) || defined(CONFIG_BFIN_MAC_MODULE)
 #include <linux/bfin_mac.h>
-#include <linux/export.h>
 static const unsigned short bfin_mac_peripherals[] = P_MII0;
 
 static struct bfin_phydev_platform_data bfin_phydev_data[] = {
diff --git a/arch/blackfin/mach-bf537/boards/cm_bf537u.c b/arch/blackfin/mach-bf537/boards/cm_bf537u.c
index 47cf37d..d916b46 100644
--- a/arch/blackfin/mach-bf537/boards/cm_bf537u.c
+++ b/arch/blackfin/mach-bf537/boards/cm_bf537u.c
@@ -9,6 +9,7 @@
 
 #include <linux/device.h>
 #include <linux/etherdevice.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
@@ -306,8 +307,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -349,8 +355,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -534,7 +545,6 @@
 
 #if defined(CONFIG_BFIN_MAC) || defined(CONFIG_BFIN_MAC_MODULE)
 #include <linux/bfin_mac.h>
-#include <linux/export.h>
 static const unsigned short bfin_mac_peripherals[] = P_MII0;
 
 static struct bfin_phydev_platform_data bfin_phydev_data[] = {
diff --git a/arch/blackfin/mach-bf537/boards/dnp5370.c b/arch/blackfin/mach-bf537/boards/dnp5370.c
index 33e69e4..5f30722 100644
--- a/arch/blackfin/mach-bf537/boards/dnp5370.c
+++ b/arch/blackfin/mach-bf537/boards/dnp5370.c
@@ -12,6 +12,7 @@
  */
 
 #include <linux/device.h>
+#include <linux/export.h>
 #include <linux/kernel.h>
 #include <linux/platform_device.h>
 #include <linux/io.h>
@@ -49,7 +50,6 @@
 
 #if defined(CONFIG_BFIN_MAC) || defined(CONFIG_BFIN_MAC_MODULE)
 #include <linux/bfin_mac.h>
-#include <linux/export.h>
 static const unsigned short bfin_mac_peripherals[] = P_RMII0;
 
 static struct bfin_phydev_platform_data bfin_phydev_data[] = {
@@ -237,8 +237,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -281,8 +286,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end   = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end   = IRQ_UART1_RX+1,
+		.end   = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf537/boards/minotaur.c b/arch/blackfin/mach-bf537/boards/minotaur.c
index c62f9dc..3901dd0 100644
--- a/arch/blackfin/mach-bf537/boards/minotaur.c
+++ b/arch/blackfin/mach-bf537/boards/minotaur.c
@@ -240,8 +240,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -283,8 +288,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf537/boards/pnav10.c b/arch/blackfin/mach-bf537/boards/pnav10.c
index 3099e91..aebd31c 100644
--- a/arch/blackfin/mach-bf537/boards/pnav10.c
+++ b/arch/blackfin/mach-bf537/boards/pnav10.c
@@ -8,6 +8,7 @@
 
 #include <linux/device.h>
 #include <linux/etherdevice.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
@@ -309,8 +310,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -352,8 +358,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf537/boards/stamp.c b/arch/blackfin/mach-bf537/boards/stamp.c
index 27f955d..7fbb0bb 100644
--- a/arch/blackfin/mach-bf537/boards/stamp.c
+++ b/arch/blackfin/mach-bf537/boards/stamp.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/device.h>
+#include <linux/export.h>
 #include <linux/kernel.h>
 #include <linux/platform_device.h>
 #include <linux/io.h>
@@ -1566,8 +1567,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -1621,8 +1627,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -1992,7 +2003,6 @@
 
 #if defined(CONFIG_BACKLIGHT_ADP8860) || defined(CONFIG_BACKLIGHT_ADP8860_MODULE)
 #include <linux/i2c/adp8860.h>
-#include <linux/export.h>
 static struct led_info adp8860_leds[] = {
 	{
 		.name = "adp8860-led7",
diff --git a/arch/blackfin/mach-bf537/boards/tcm_bf537.c b/arch/blackfin/mach-bf537/boards/tcm_bf537.c
index 8418030..6917ce2 100644
--- a/arch/blackfin/mach-bf537/boards/tcm_bf537.c
+++ b/arch/blackfin/mach-bf537/boards/tcm_bf537.c
@@ -9,6 +9,7 @@
 
 #include <linux/device.h>
 #include <linux/etherdevice.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
@@ -306,8 +307,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -349,8 +355,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -536,7 +547,6 @@
 
 #if defined(CONFIG_BFIN_MAC) || defined(CONFIG_BFIN_MAC_MODULE)
 #include <linux/bfin_mac.h>
-#include <linux/export.h>
 static const unsigned short bfin_mac_peripherals[] = P_MII0;
 
 static struct bfin_phydev_platform_data bfin_phydev_data[] = {
diff --git a/arch/blackfin/mach-bf538/boards/ezkit.c b/arch/blackfin/mach-bf538/boards/ezkit.c
index 629f3c3..8356eb5 100644
--- a/arch/blackfin/mach-bf538/boards/ezkit.c
+++ b/arch/blackfin/mach-bf538/boards/ezkit.c
@@ -49,8 +49,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -104,8 +109,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -147,8 +157,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART2_TX,
+		.end = IRQ_UART2_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART2_RX,
-		.end = IRQ_UART2_RX+1,
+		.end = IRQ_UART2_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf548/boards/cm_bf548.c b/arch/blackfin/mach-bf548/boards/cm_bf548.c
index 212b9e0..0350eac 100644
--- a/arch/blackfin/mach-bf548/boards/cm_bf548.c
+++ b/arch/blackfin/mach-bf548/boards/cm_bf548.c
@@ -135,8 +135,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -178,8 +183,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -237,8 +247,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART2_TX,
+		.end = IRQ_UART2_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART2_RX,
-		.end = IRQ_UART2_RX+1,
+		.end = IRQ_UART2_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -280,8 +295,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART3_TX,
+		.end = IRQ_UART3_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART3_RX,
-		.end = IRQ_UART3_RX+1,
+		.end = IRQ_UART3_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf548/boards/ezkit.c b/arch/blackfin/mach-bf548/boards/ezkit.c
index cd9cbb6..bb868ac 100644
--- a/arch/blackfin/mach-bf548/boards/ezkit.c
+++ b/arch/blackfin/mach-bf548/boards/ezkit.c
@@ -241,8 +241,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART0_TX,
+		.end = IRQ_UART0_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART0_RX,
-		.end = IRQ_UART0_RX+1,
+		.end = IRQ_UART0_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -284,8 +289,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART1_TX,
+		.end = IRQ_UART1_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART1_RX,
-		.end = IRQ_UART1_RX+1,
+		.end = IRQ_UART1_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -343,8 +353,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART2_TX,
+		.end = IRQ_UART2_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART2_RX,
-		.end = IRQ_UART2_RX+1,
+		.end = IRQ_UART2_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
@@ -386,8 +401,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART3_TX,
+		.end = IRQ_UART3_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART3_RX,
-		.end = IRQ_UART3_RX+1,
+		.end = IRQ_UART3_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf561/boards/acvilon.c b/arch/blackfin/mach-bf561/boards/acvilon.c
index 972e134..b1b7339 100644
--- a/arch/blackfin/mach-bf561/boards/acvilon.c
+++ b/arch/blackfin/mach-bf561/boards/acvilon.c
@@ -203,8 +203,13 @@
 	 .flags = IORESOURCE_MEM,
 	 },
 	{
+	 .start = IRQ_UART_TX,
+	 .end = IRQ_UART_TX,
+	 .flags = IORESOURCE_IRQ,
+	 },
+	{
 	 .start = IRQ_UART_RX,
-	 .end = IRQ_UART_RX + 1,
+	 .end = IRQ_UART_RX,
 	 .flags = IORESOURCE_IRQ,
 	 },
 	{
diff --git a/arch/blackfin/mach-bf561/boards/cm_bf561.c b/arch/blackfin/mach-bf561/boards/cm_bf561.c
index c1b72f2..c017cf0 100644
--- a/arch/blackfin/mach-bf561/boards/cm_bf561.c
+++ b/arch/blackfin/mach-bf561/boards/cm_bf561.c
@@ -277,8 +277,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART_TX,
+		.end = IRQ_UART_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART_RX,
-		.end = IRQ_UART_RX+1,
+		.end = IRQ_UART_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf561/boards/ezkit.c b/arch/blackfin/mach-bf561/boards/ezkit.c
index 9490dc8..27f22ed 100644
--- a/arch/blackfin/mach-bf561/boards/ezkit.c
+++ b/arch/blackfin/mach-bf561/boards/ezkit.c
@@ -172,8 +172,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART_TX,
+		.end = IRQ_UART_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART_RX,
-		.end = IRQ_UART_RX+1,
+		.end = IRQ_UART_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/blackfin/mach-bf561/boards/tepla.c b/arch/blackfin/mach-bf561/boards/tepla.c
index bb056e6..1a57bc9 100644
--- a/arch/blackfin/mach-bf561/boards/tepla.c
+++ b/arch/blackfin/mach-bf561/boards/tepla.c
@@ -51,8 +51,13 @@
 		.flags = IORESOURCE_MEM,
 	},
 	{
+		.start = IRQ_UART_TX,
+		.end = IRQ_UART_TX,
+		.flags = IORESOURCE_IRQ,
+	},
+	{
 		.start = IRQ_UART_RX,
-		.end = IRQ_UART_RX+1,
+		.end = IRQ_UART_RX,
 		.flags = IORESOURCE_IRQ,
 	},
 	{
diff --git a/arch/cris/arch-v10/drivers/Kconfig b/arch/cris/arch-v10/drivers/Kconfig
index 32d9086..5f2cdb3 100644
--- a/arch/cris/arch-v10/drivers/Kconfig
+++ b/arch/cris/arch-v10/drivers/Kconfig
@@ -3,7 +3,7 @@
 config ETRAX_ETHERNET
 	bool "Ethernet support"
 	depends on ETRAX_ARCH_V10
-	select NET_ETHERNET
+	select ETHERNET
 	select NET_CORE
 	select MII
 	help
diff --git a/arch/cris/arch-v32/drivers/Kconfig b/arch/cris/arch-v32/drivers/Kconfig
index e47e9c3..de43aad 100644
--- a/arch/cris/arch-v32/drivers/Kconfig
+++ b/arch/cris/arch-v32/drivers/Kconfig
@@ -3,7 +3,7 @@
 config ETRAX_ETHERNET
 	bool "Ethernet support"
 	depends on ETRAX_ARCH_V32
-	select NET_ETHERNET
+	select ETHERNET
 	select NET_CORE
 	select MII
 	help
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index 6c28582..361d540 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -4,8 +4,8 @@
 	select HAVE_IDE
 	select HAVE_AOUT if MMU
 	select GENERIC_ATOMIC64 if MMU
-	select HAVE_GENERIC_HARDIRQS if !MMU
-	select GENERIC_IRQ_SHOW if !MMU
+	select HAVE_GENERIC_HARDIRQS
+	select GENERIC_IRQ_SHOW
 	select ARCH_HAVE_NMI_SAFE_CMPXCHG if RMW_INSNS
 
 config RWSEM_GENERIC_SPINLOCK
diff --git a/arch/m68k/Kconfig.bus b/arch/m68k/Kconfig.bus
index 8294f0c..3adb499 100644
--- a/arch/m68k/Kconfig.bus
+++ b/arch/m68k/Kconfig.bus
@@ -2,6 +2,15 @@
 
 comment "Bus Support"
 
+config DIO
+	bool "DIO bus support"
+	depends on HP300
+	default y
+	help
+	  Say Y here to enable support for the "DIO" expansion bus used in
+	  HP300 machines. If you are using such a system you almost certainly
+	  want this.
+
 config NUBUS
 	bool
 	depends on MAC
diff --git a/arch/m68k/Kconfig.devices b/arch/m68k/Kconfig.devices
index d214034..6033f5d 100644
--- a/arch/m68k/Kconfig.devices
+++ b/arch/m68k/Kconfig.devices
@@ -24,6 +24,37 @@
 	  including the model, CPU, MMU, clock speed, BogoMIPS rating,
 	  and memory size.
 
+config NATFEAT
+	bool "ARAnyM emulator support"
+	depends on ATARI
+	help
+	  This option enables support for ARAnyM native features, such as
+	  access to a disk image as /dev/hda.
+
+config NFBLOCK
+	tristate "NatFeat block device support"
+	depends on BLOCK && NATFEAT
+	help
+	  Say Y to include support for the ARAnyM NatFeat block device
+	  which allows direct access to the hard drives without using
+	  the hardware emulation.
+
+config NFCON
+	tristate "NatFeat console driver"
+	depends on NATFEAT
+	help
+	  Say Y to include support for the ARAnyM NatFeat console driver
+	  which allows the console output to be redirected to the stderr
+	  output of ARAnyM.
+
+config NFETH
+	tristate "NatFeat Ethernet support"
+	depends on ETHERNET && NATFEAT
+	help
+	  Say Y to include support for the ARAnyM NatFeat network device
+	  which will emulate a regular ethernet device while presenting an
+	  ethertap device to the host system.
+
 endmenu
 
 menu "Character devices"
diff --git a/arch/m68k/amiga/amiints.c b/arch/m68k/amiga/amiints.c
index c5b5212..47b5f90 100644
--- a/arch/m68k/amiga/amiints.c
+++ b/arch/m68k/amiga/amiints.c
@@ -1,43 +1,15 @@
 /*
- * linux/arch/m68k/amiga/amiints.c -- Amiga Linux interrupt handling code
+ * Amiga Linux interrupt handling code
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file COPYING in the main directory of this archive
  * for more details.
- *
- * 11/07/96: rewritten interrupt handling, irq lists are exists now only for
- *           this sources where it makes sense (VERTB/PORTS/EXTER) and you must
- *           be careful that dev_id for this sources is unique since this the
- *           only possibility to distinguish between different handlers for
- *           free_irq. irq lists also have different irq flags:
- *           - IRQ_FLG_FAST: handler is inserted at top of list (after other
- *                           fast handlers)
- *           - IRQ_FLG_SLOW: handler is inserted at bottom of list and before
- *                           they're executed irq level is set to the previous
- *                           one, but handlers don't need to be reentrant, if
- *                           reentrance occurred, slow handlers will be just
- *                           called again.
- *           The whole interrupt handling for CIAs is moved to cia.c
- *           /Roman Zippel
- *
- * 07/08/99: rewamp of the interrupt handling - we now have two types of
- *           interrupts, normal and fast handlers, fast handlers being
- *           marked with IRQF_DISABLED and runs with all other interrupts
- *           disabled. Normal interrupts disable their own source but
- *           run with all other interrupt sources enabled.
- *           PORTS and EXTER interrupts are always shared even if the
- *           drivers do not explicitly mark this when calling
- *           request_irq which they really should do.
- *           This is similar to the way interrupts are handled on all
- *           other architectures and makes a ton of sense besides
- *           having the advantage of making it easier to share
- *           drivers.
- *           /Jes
  */
 
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/errno.h>
+#include <linux/irq.h>
 
 #include <asm/irq.h>
 #include <asm/traps.h>
@@ -45,20 +17,128 @@
 #include <asm/amigaints.h>
 #include <asm/amipcmcia.h>
 
-static void amiga_enable_irq(unsigned int irq);
-static void amiga_disable_irq(unsigned int irq);
-static irqreturn_t ami_int1(int irq, void *dev_id);
-static irqreturn_t ami_int3(int irq, void *dev_id);
-static irqreturn_t ami_int4(int irq, void *dev_id);
-static irqreturn_t ami_int5(int irq, void *dev_id);
 
-static struct irq_controller amiga_irq_controller = {
+/*
+ * Enable/disable a particular machine specific interrupt source.
+ * Note that this may affect other interrupts in case of a shared interrupt.
+ * This function should only be called for a _very_ short time to change some
+ * internal data, that may not be changed by the interrupt at the same time.
+ */
+
+static void amiga_irq_enable(struct irq_data *data)
+{
+	amiga_custom.intena = IF_SETCLR | (1 << (data->irq - IRQ_USER));
+}
+
+static void amiga_irq_disable(struct irq_data *data)
+{
+	amiga_custom.intena = 1 << (data->irq - IRQ_USER);
+}
+
+static struct irq_chip amiga_irq_chip = {
 	.name		= "amiga",
-	.lock		= __SPIN_LOCK_UNLOCKED(amiga_irq_controller.lock),
-	.enable		= amiga_enable_irq,
-	.disable	= amiga_disable_irq,
+	.irq_enable	= amiga_irq_enable,
+	.irq_disable	= amiga_irq_disable,
 };
 
+
+/*
+ * The builtin Amiga hardware interrupt handlers.
+ */
+
+static void ami_int1(unsigned int irq, struct irq_desc *desc)
+{
+	unsigned short ints = amiga_custom.intreqr & amiga_custom.intenar;
+
+	/* if serial transmit buffer empty, interrupt */
+	if (ints & IF_TBE) {
+		amiga_custom.intreq = IF_TBE;
+		generic_handle_irq(IRQ_AMIGA_TBE);
+	}
+
+	/* if floppy disk transfer complete, interrupt */
+	if (ints & IF_DSKBLK) {
+		amiga_custom.intreq = IF_DSKBLK;
+		generic_handle_irq(IRQ_AMIGA_DSKBLK);
+	}
+
+	/* if software interrupt set, interrupt */
+	if (ints & IF_SOFT) {
+		amiga_custom.intreq = IF_SOFT;
+		generic_handle_irq(IRQ_AMIGA_SOFT);
+	}
+}
+
+static void ami_int3(unsigned int irq, struct irq_desc *desc)
+{
+	unsigned short ints = amiga_custom.intreqr & amiga_custom.intenar;
+
+	/* if a blitter interrupt */
+	if (ints & IF_BLIT) {
+		amiga_custom.intreq = IF_BLIT;
+		generic_handle_irq(IRQ_AMIGA_BLIT);
+	}
+
+	/* if a copper interrupt */
+	if (ints & IF_COPER) {
+		amiga_custom.intreq = IF_COPER;
+		generic_handle_irq(IRQ_AMIGA_COPPER);
+	}
+
+	/* if a vertical blank interrupt */
+	if (ints & IF_VERTB) {
+		amiga_custom.intreq = IF_VERTB;
+		generic_handle_irq(IRQ_AMIGA_VERTB);
+	}
+}
+
+static void ami_int4(unsigned int irq, struct irq_desc *desc)
+{
+	unsigned short ints = amiga_custom.intreqr & amiga_custom.intenar;
+
+	/* if audio 0 interrupt */
+	if (ints & IF_AUD0) {
+		amiga_custom.intreq = IF_AUD0;
+		generic_handle_irq(IRQ_AMIGA_AUD0);
+	}
+
+	/* if audio 1 interrupt */
+	if (ints & IF_AUD1) {
+		amiga_custom.intreq = IF_AUD1;
+		generic_handle_irq(IRQ_AMIGA_AUD1);
+	}
+
+	/* if audio 2 interrupt */
+	if (ints & IF_AUD2) {
+		amiga_custom.intreq = IF_AUD2;
+		generic_handle_irq(IRQ_AMIGA_AUD2);
+	}
+
+	/* if audio 3 interrupt */
+	if (ints & IF_AUD3) {
+		amiga_custom.intreq = IF_AUD3;
+		generic_handle_irq(IRQ_AMIGA_AUD3);
+	}
+}
+
+static void ami_int5(unsigned int irq, struct irq_desc *desc)
+{
+	unsigned short ints = amiga_custom.intreqr & amiga_custom.intenar;
+
+	/* if serial receive buffer full interrupt */
+	if (ints & IF_RBF) {
+		/* acknowledge of IF_RBF must be done by the serial interrupt */
+		generic_handle_irq(IRQ_AMIGA_RBF);
+	}
+
+	/* if a disk sync interrupt */
+	if (ints & IF_DSKSYN) {
+		amiga_custom.intreq = IF_DSKSYN;
+		generic_handle_irq(IRQ_AMIGA_DSKSYN);
+	}
+}
+
+
 /*
  * void amiga_init_IRQ(void)
  *
@@ -72,16 +152,13 @@
 
 void __init amiga_init_IRQ(void)
 {
-	if (request_irq(IRQ_AUTO_1, ami_int1, 0, "int1", NULL))
-		pr_err("Couldn't register int%d\n", 1);
-	if (request_irq(IRQ_AUTO_3, ami_int3, 0, "int3", NULL))
-		pr_err("Couldn't register int%d\n", 3);
-	if (request_irq(IRQ_AUTO_4, ami_int4, 0, "int4", NULL))
-		pr_err("Couldn't register int%d\n", 4);
-	if (request_irq(IRQ_AUTO_5, ami_int5, 0, "int5", NULL))
-		pr_err("Couldn't register int%d\n", 5);
+	m68k_setup_irq_controller(&amiga_irq_chip, handle_simple_irq, IRQ_USER,
+				  AMI_STD_IRQS);
 
-	m68k_setup_irq_controller(&amiga_irq_controller, IRQ_USER, AMI_STD_IRQS);
+	irq_set_chained_handler(IRQ_AUTO_1, ami_int1);
+	irq_set_chained_handler(IRQ_AUTO_3, ami_int3);
+	irq_set_chained_handler(IRQ_AUTO_4, ami_int4);
+	irq_set_chained_handler(IRQ_AUTO_5, ami_int5);
 
 	/* turn off PCMCIA interrupts */
 	if (AMIGAHW_PRESENT(PCMCIA))
@@ -95,120 +172,3 @@
 	cia_init_IRQ(&ciaa_base);
 	cia_init_IRQ(&ciab_base);
 }
-
-/*
- * Enable/disable a particular machine specific interrupt source.
- * Note that this may affect other interrupts in case of a shared interrupt.
- * This function should only be called for a _very_ short time to change some
- * internal data, that may not be changed by the interrupt at the same time.
- */
-
-static void amiga_enable_irq(unsigned int irq)
-{
-	amiga_custom.intena = IF_SETCLR | (1 << (irq - IRQ_USER));
-}
-
-static void amiga_disable_irq(unsigned int irq)
-{
-	amiga_custom.intena = 1 << (irq - IRQ_USER);
-}
-
-/*
- * The builtin Amiga hardware interrupt handlers.
- */
-
-static irqreturn_t ami_int1(int irq, void *dev_id)
-{
-	unsigned short ints = amiga_custom.intreqr & amiga_custom.intenar;
-
-	/* if serial transmit buffer empty, interrupt */
-	if (ints & IF_TBE) {
-		amiga_custom.intreq = IF_TBE;
-		m68k_handle_int(IRQ_AMIGA_TBE);
-	}
-
-	/* if floppy disk transfer complete, interrupt */
-	if (ints & IF_DSKBLK) {
-		amiga_custom.intreq = IF_DSKBLK;
-		m68k_handle_int(IRQ_AMIGA_DSKBLK);
-	}
-
-	/* if software interrupt set, interrupt */
-	if (ints & IF_SOFT) {
-		amiga_custom.intreq = IF_SOFT;
-		m68k_handle_int(IRQ_AMIGA_SOFT);
-	}
-	return IRQ_HANDLED;
-}
-
-static irqreturn_t ami_int3(int irq, void *dev_id)
-{
-	unsigned short ints = amiga_custom.intreqr & amiga_custom.intenar;
-
-	/* if a blitter interrupt */
-	if (ints & IF_BLIT) {
-		amiga_custom.intreq = IF_BLIT;
-		m68k_handle_int(IRQ_AMIGA_BLIT);
-	}
-
-	/* if a copper interrupt */
-	if (ints & IF_COPER) {
-		amiga_custom.intreq = IF_COPER;
-		m68k_handle_int(IRQ_AMIGA_COPPER);
-	}
-
-	/* if a vertical blank interrupt */
-	if (ints & IF_VERTB) {
-		amiga_custom.intreq = IF_VERTB;
-		m68k_handle_int(IRQ_AMIGA_VERTB);
-	}
-	return IRQ_HANDLED;
-}
-
-static irqreturn_t ami_int4(int irq, void *dev_id)
-{
-	unsigned short ints = amiga_custom.intreqr & amiga_custom.intenar;
-
-	/* if audio 0 interrupt */
-	if (ints & IF_AUD0) {
-		amiga_custom.intreq = IF_AUD0;
-		m68k_handle_int(IRQ_AMIGA_AUD0);
-	}
-
-	/* if audio 1 interrupt */
-	if (ints & IF_AUD1) {
-		amiga_custom.intreq = IF_AUD1;
-		m68k_handle_int(IRQ_AMIGA_AUD1);
-	}
-
-	/* if audio 2 interrupt */
-	if (ints & IF_AUD2) {
-		amiga_custom.intreq = IF_AUD2;
-		m68k_handle_int(IRQ_AMIGA_AUD2);
-	}
-
-	/* if audio 3 interrupt */
-	if (ints & IF_AUD3) {
-		amiga_custom.intreq = IF_AUD3;
-		m68k_handle_int(IRQ_AMIGA_AUD3);
-	}
-	return IRQ_HANDLED;
-}
-
-static irqreturn_t ami_int5(int irq, void *dev_id)
-{
-	unsigned short ints = amiga_custom.intreqr & amiga_custom.intenar;
-
-	/* if serial receive buffer full interrupt */
-	if (ints & IF_RBF) {
-		/* acknowledge of IF_RBF must be done by the serial interrupt */
-		m68k_handle_int(IRQ_AMIGA_RBF);
-	}
-
-	/* if a disk sync interrupt */
-	if (ints & IF_DSKSYN) {
-		amiga_custom.intreq = IF_DSKSYN;
-		m68k_handle_int(IRQ_AMIGA_DSKSYN);
-	}
-	return IRQ_HANDLED;
-}
diff --git a/arch/m68k/amiga/cia.c b/arch/m68k/amiga/cia.c
index ecd0f7c..18c0e29 100644
--- a/arch/m68k/amiga/cia.c
+++ b/arch/m68k/amiga/cia.c
@@ -93,13 +93,14 @@
 	amiga_custom.intreq = base->int_mask;
 	for (; ints; mach_irq++, ints >>= 1) {
 		if (ints & 1)
-			m68k_handle_int(mach_irq);
+			generic_handle_irq(mach_irq);
 	}
 	return IRQ_HANDLED;
 }
 
-static void cia_enable_irq(unsigned int irq)
+static void cia_irq_enable(struct irq_data *data)
 {
+	unsigned int irq = data->irq;
 	unsigned char mask;
 
 	if (irq >= IRQ_AMIGA_CIAB) {
@@ -113,19 +114,20 @@
 	}
 }
 
-static void cia_disable_irq(unsigned int irq)
+static void cia_irq_disable(struct irq_data *data)
 {
+	unsigned int irq = data->irq;
+
 	if (irq >= IRQ_AMIGA_CIAB)
 		cia_able_irq(&ciab_base, 1 << (irq - IRQ_AMIGA_CIAB));
 	else
 		cia_able_irq(&ciaa_base, 1 << (irq - IRQ_AMIGA_CIAA));
 }
 
-static struct irq_controller cia_irq_controller = {
+static struct irq_chip cia_irq_chip = {
 	.name		= "cia",
-	.lock		= __SPIN_LOCK_UNLOCKED(cia_irq_controller.lock),
-	.enable		= cia_enable_irq,
-	.disable	= cia_disable_irq,
+	.irq_enable	= cia_irq_enable,
+	.irq_disable	= cia_irq_disable,
 };
 
 /*
@@ -134,9 +136,9 @@
  * into this chain.
  */
 
-static void auto_enable_irq(unsigned int irq)
+static void auto_irq_enable(struct irq_data *data)
 {
-	switch (irq) {
+	switch (data->irq) {
 	case IRQ_AUTO_2:
 		amiga_custom.intena = IF_SETCLR | IF_PORTS;
 		break;
@@ -146,9 +148,9 @@
 	}
 }
 
-static void auto_disable_irq(unsigned int irq)
+static void auto_irq_disable(struct irq_data *data)
 {
-	switch (irq) {
+	switch (data->irq) {
 	case IRQ_AUTO_2:
 		amiga_custom.intena = IF_PORTS;
 		break;
@@ -158,24 +160,25 @@
 	}
 }
 
-static struct irq_controller auto_irq_controller = {
+static struct irq_chip auto_irq_chip = {
 	.name		= "auto",
-	.lock		= __SPIN_LOCK_UNLOCKED(auto_irq_controller.lock),
-	.enable		= auto_enable_irq,
-	.disable	= auto_disable_irq,
+	.irq_enable	= auto_irq_enable,
+	.irq_disable	= auto_irq_disable,
 };
 
 void __init cia_init_IRQ(struct ciabase *base)
 {
-	m68k_setup_irq_controller(&cia_irq_controller, base->cia_irq, CIA_IRQS);
+	m68k_setup_irq_controller(&cia_irq_chip, handle_simple_irq,
+				  base->cia_irq, CIA_IRQS);
 
 	/* clear any pending interrupt and turn off all interrupts */
 	cia_set_irq(base, CIA_ICR_ALL);
 	cia_able_irq(base, CIA_ICR_ALL);
 
 	/* override auto int and install CIA handler */
-	m68k_setup_irq_controller(&auto_irq_controller, base->handler_irq, 1);
-	m68k_irq_startup(base->handler_irq);
+	m68k_setup_irq_controller(&auto_irq_chip, handle_simple_irq,
+				  base->handler_irq, 1);
+	m68k_irq_startup_irq(base->handler_irq);
 	if (request_irq(base->handler_irq, cia_handler, IRQF_SHARED,
 			base->name, base))
 		pr_err("Couldn't register %s interrupt\n", base->name);
diff --git a/arch/m68k/apollo/dn_ints.c b/arch/m68k/apollo/dn_ints.c
index 5d47f3a..17be1e7 100644
--- a/arch/m68k/apollo/dn_ints.c
+++ b/arch/m68k/apollo/dn_ints.c
@@ -1,19 +1,13 @@
 #include <linux/interrupt.h>
+#include <linux/irq.h>
 
-#include <asm/irq.h>
 #include <asm/traps.h>
 #include <asm/apollohw.h>
 
-void dn_process_int(unsigned int irq, struct pt_regs *fp)
+unsigned int apollo_irq_startup(struct irq_data *data)
 {
-	__m68k_handle_int(irq, fp);
+	unsigned int irq = data->irq;
 
-	*(volatile unsigned char *)(pica)=0x20;
-	*(volatile unsigned char *)(picb)=0x20;
-}
-
-int apollo_irq_startup(unsigned int irq)
-{
 	if (irq < 8)
 		*(volatile unsigned char *)(pica+1) &= ~(1 << irq);
 	else
@@ -21,24 +15,33 @@
 	return 0;
 }
 
-void apollo_irq_shutdown(unsigned int irq)
+void apollo_irq_shutdown(struct irq_data *data)
 {
+	unsigned int irq = data->irq;
+
 	if (irq < 8)
 		*(volatile unsigned char *)(pica+1) |= (1 << irq);
 	else
 		*(volatile unsigned char *)(picb+1) |= (1 << (irq - 8));
 }
 
-static struct irq_controller apollo_irq_controller = {
+void apollo_irq_eoi(struct irq_data *data)
+{
+	*(volatile unsigned char *)(pica) = 0x20;
+	*(volatile unsigned char *)(picb) = 0x20;
+}
+
+static struct irq_chip apollo_irq_chip = {
 	.name           = "apollo",
-	.lock           = __SPIN_LOCK_UNLOCKED(apollo_irq_controller.lock),
-	.startup        = apollo_irq_startup,
-	.shutdown       = apollo_irq_shutdown,
+	.irq_startup    = apollo_irq_startup,
+	.irq_shutdown   = apollo_irq_shutdown,
+	.irq_eoi	= apollo_irq_eoi,
 };
 
 
 void __init dn_init_IRQ(void)
 {
-	m68k_setup_user_interrupt(VEC_USER + 96, 16, dn_process_int);
-	m68k_setup_irq_controller(&apollo_irq_controller, IRQ_APOLLO, 16);
+	m68k_setup_user_interrupt(VEC_USER + 96, 16);
+	m68k_setup_irq_controller(&apollo_irq_chip, handle_fasteoi_irq,
+				  IRQ_APOLLO, 16);
 }
diff --git a/arch/m68k/atari/ataints.c b/arch/m68k/atari/ataints.c
index 26a804e..6d196da 100644
--- a/arch/m68k/atari/ataints.c
+++ b/arch/m68k/atari/ataints.c
@@ -60,244 +60,8 @@
  * <asm/atariints.h>): Autovector interrupts are 1..7, then follow ST-MFP,
  * TT-MFP, SCC, and finally VME interrupts. Vector numbers for the latter can
  * be allocated by atari_register_vme_int().
- *
- * Each interrupt can be of three types:
- *
- *  - SLOW: The handler runs with all interrupts enabled, except the one it
- *    was called by (to avoid reentering). This should be the usual method.
- *    But it is currently possible only for MFP ints, since only the MFP
- *    offers an easy way to mask interrupts.
- *
- *  - FAST: The handler runs with all interrupts disabled. This should be used
- *    only for really fast handlers, that just do actions immediately
- *    necessary, and let the rest do a bottom half or task queue.
- *
- *  - PRIORITIZED: The handler can be interrupted by higher-level ints
- *    (greater IPL, no MFP priorities!). This is the method of choice for ints
- *    which should be slow, but are not from a MFP.
- *
- * The feature of more than one handler for one int source is still there, but
- * only applicable if all handers are of the same type. To not slow down
- * processing of ints with only one handler by the chaining feature, the list
- * calling function atari_call_irq_list() is only plugged in at the time the
- * second handler is registered.
- *
- * Implementation notes: For fast-as-possible int handling, there are separate
- * entry points for each type (slow/fast/prio). The assembler handler calls
- * the irq directly in the usual case, no C wrapper is involved. In case of
- * multiple handlers, atari_call_irq_list() is registered as handler and calls
- * in turn the real irq's. To ease access from assembler level to the irq
- * function pointer and accompanying data, these two are stored in a separate
- * array, irq_handler[]. The rest of data (type, name) are put into a second
- * array, irq_param, that is accessed from C only. For each slow interrupt (32
- * in all) there are separate handler functions, which makes it possible to
- * hard-code the MFP register address and value, are necessary to mask the
- * int. If there'd be only one generic function, lots of calculations would be
- * needed to determine MFP register and int mask from the vector number :-(
- *
- * Furthermore, slow ints may not lower the IPL below its previous value
- * (before the int happened). This is needed so that an int of class PRIO, on
- * that this int may be stacked, cannot be reentered. This feature is
- * implemented as follows: If the stack frame format is 1 (throwaway), the int
- * is not stacked, and the IPL is anded with 0xfbff, resulting in a new level
- * 2, which still blocks the HSYNC, but no interrupts of interest. If the
- * frame format is 0, the int is nested, and the old IPL value can be found in
- * the sr copy in the frame.
  */
 
-#if 0
-
-#define	NUM_INT_SOURCES	(8 + NUM_ATARI_SOURCES)
-
-typedef void (*asm_irq_handler)(void);
-
-struct irqhandler {
-	irqreturn_t (*handler)(int, void *, struct pt_regs *);
-	void	*dev_id;
-};
-
-struct irqparam {
-	unsigned long	flags;
-	const char	*devname;
-};
-
-/*
- * Array with irq's and their parameter data. This array is accessed from low
- * level assembler code, so an element size of 8 allows usage of index scaling
- * addressing mode.
- */
-static struct irqhandler irq_handler[NUM_INT_SOURCES];
-
-/*
- * This array hold the rest of parameters of int handlers: type
- * (slow,fast,prio) and the name of the handler. These values are only
- * accessed from C
- */
-static struct irqparam irq_param[NUM_INT_SOURCES];
-
-/* check for valid int number (complex, sigh...) */
-#define	IS_VALID_INTNO(n)											\
-	((n) > 0 &&														\
-	 /* autovec and ST-MFP ok anyway */								\
-	 (((n) < TTMFP_SOURCE_BASE) ||									\
-	  /* TT-MFP ok if present */									\
-	  ((n) >= TTMFP_SOURCE_BASE && (n) < SCC_SOURCE_BASE &&			\
-	   ATARIHW_PRESENT(TT_MFP)) ||									\
-	  /* SCC ok if present and number even */						\
-	  ((n) >= SCC_SOURCE_BASE && (n) < VME_SOURCE_BASE &&			\
-	   !((n) & 1) && ATARIHW_PRESENT(SCC)) ||						\
-	  /* greater numbers ok if they are registered VME vectors */		\
-	  ((n) >= VME_SOURCE_BASE && (n) < VME_SOURCE_BASE + VME_MAX_SOURCES && \
-		  free_vme_vec_bitmap & (1 << ((n) - VME_SOURCE_BASE)))))
-
-
-/*
- * Here start the assembler entry points for interrupts
- */
-
-#define IRQ_NAME(nr) atari_slow_irq_##nr##_handler(void)
-
-#define	BUILD_SLOW_IRQ(n)						   \
-asmlinkage void IRQ_NAME(n);						   \
-/* Dummy function to allow asm with operands.  */			   \
-void atari_slow_irq_##n##_dummy (void) {				   \
-__asm__ (__ALIGN_STR "\n"						   \
-"atari_slow_irq_" #n "_handler:\t"					   \
-"	addl	%6,%5\n"	/* preempt_count() += HARDIRQ_OFFSET */	   \
-	SAVE_ALL_INT "\n"						   \
-	GET_CURRENT(%%d0) "\n"						   \
-"	andb	#~(1<<(%c3&7)),%a4:w\n"	/* mask this interrupt */	   \
-	/* get old IPL from stack frame */				   \
-"	bfextu	%%sp@(%c2){#5,#3},%%d0\n"				   \
-"	movew	%%sr,%%d1\n"						   \
-"	bfins	%%d0,%%d1{#21,#3}\n"					   \
-"	movew	%%d1,%%sr\n"		/* set IPL = previous value */	   \
-"	addql	#1,%a0\n"						   \
-"	lea	%a1,%%a0\n"						   \
-"	pea	%%sp@\n"		/* push addr of frame */	   \
-"	movel	%%a0@(4),%%sp@-\n"	/* push handler data */		   \
-"	pea	(%c3+8)\n"		/* push int number */		   \
-"	movel	%%a0@,%%a0\n"						   \
-"	jbsr	%%a0@\n"		/* call the handler */		   \
-"	addql	#8,%%sp\n"						   \
-"	addql	#4,%%sp\n"						   \
-"	orw	#0x0600,%%sr\n"						   \
-"	andw	#0xfeff,%%sr\n"		/* set IPL = 6 again */		   \
-"	orb	#(1<<(%c3&7)),%a4:w\n"	/* now unmask the int again */	   \
-"	jbra	ret_from_interrupt\n"					   \
-	 : : "i" (&kstat_cpu(0).irqs[n+8]), "i" (&irq_handler[n+8]),	   \
-	     "n" (PT_OFF_SR), "n" (n),					   \
-	     "i" (n & 8 ? (n & 16 ? &tt_mfp.int_mk_a : &st_mfp.int_mk_a)   \
-		        : (n & 16 ? &tt_mfp.int_mk_b : &st_mfp.int_mk_b)), \
-	     "m" (preempt_count()), "di" (HARDIRQ_OFFSET)		   \
-);									   \
-	for (;;);			/* fake noreturn */		   \
-}
-
-BUILD_SLOW_IRQ(0);
-BUILD_SLOW_IRQ(1);
-BUILD_SLOW_IRQ(2);
-BUILD_SLOW_IRQ(3);
-BUILD_SLOW_IRQ(4);
-BUILD_SLOW_IRQ(5);
-BUILD_SLOW_IRQ(6);
-BUILD_SLOW_IRQ(7);
-BUILD_SLOW_IRQ(8);
-BUILD_SLOW_IRQ(9);
-BUILD_SLOW_IRQ(10);
-BUILD_SLOW_IRQ(11);
-BUILD_SLOW_IRQ(12);
-BUILD_SLOW_IRQ(13);
-BUILD_SLOW_IRQ(14);
-BUILD_SLOW_IRQ(15);
-BUILD_SLOW_IRQ(16);
-BUILD_SLOW_IRQ(17);
-BUILD_SLOW_IRQ(18);
-BUILD_SLOW_IRQ(19);
-BUILD_SLOW_IRQ(20);
-BUILD_SLOW_IRQ(21);
-BUILD_SLOW_IRQ(22);
-BUILD_SLOW_IRQ(23);
-BUILD_SLOW_IRQ(24);
-BUILD_SLOW_IRQ(25);
-BUILD_SLOW_IRQ(26);
-BUILD_SLOW_IRQ(27);
-BUILD_SLOW_IRQ(28);
-BUILD_SLOW_IRQ(29);
-BUILD_SLOW_IRQ(30);
-BUILD_SLOW_IRQ(31);
-
-asm_irq_handler slow_handlers[32] = {
-	[0]	= atari_slow_irq_0_handler,
-	[1]	= atari_slow_irq_1_handler,
-	[2]	= atari_slow_irq_2_handler,
-	[3]	= atari_slow_irq_3_handler,
-	[4]	= atari_slow_irq_4_handler,
-	[5]	= atari_slow_irq_5_handler,
-	[6]	= atari_slow_irq_6_handler,
-	[7]	= atari_slow_irq_7_handler,
-	[8]	= atari_slow_irq_8_handler,
-	[9]	= atari_slow_irq_9_handler,
-	[10]	= atari_slow_irq_10_handler,
-	[11]	= atari_slow_irq_11_handler,
-	[12]	= atari_slow_irq_12_handler,
-	[13]	= atari_slow_irq_13_handler,
-	[14]	= atari_slow_irq_14_handler,
-	[15]	= atari_slow_irq_15_handler,
-	[16]	= atari_slow_irq_16_handler,
-	[17]	= atari_slow_irq_17_handler,
-	[18]	= atari_slow_irq_18_handler,
-	[19]	= atari_slow_irq_19_handler,
-	[20]	= atari_slow_irq_20_handler,
-	[21]	= atari_slow_irq_21_handler,
-	[22]	= atari_slow_irq_22_handler,
-	[23]	= atari_slow_irq_23_handler,
-	[24]	= atari_slow_irq_24_handler,
-	[25]	= atari_slow_irq_25_handler,
-	[26]	= atari_slow_irq_26_handler,
-	[27]	= atari_slow_irq_27_handler,
-	[28]	= atari_slow_irq_28_handler,
-	[29]	= atari_slow_irq_29_handler,
-	[30]	= atari_slow_irq_30_handler,
-	[31]	= atari_slow_irq_31_handler
-};
-
-asmlinkage void atari_fast_irq_handler( void );
-asmlinkage void atari_prio_irq_handler( void );
-
-/* Dummy function to allow asm with operands.  */
-void atari_fast_prio_irq_dummy (void) {
-__asm__ (__ALIGN_STR "\n"
-"atari_fast_irq_handler:\n\t"
-	"orw	#0x700,%%sr\n"		/* disable all interrupts */
-"atari_prio_irq_handler:\n\t"
-	"addl	%3,%2\n\t"		/* preempt_count() += HARDIRQ_OFFSET */
-	SAVE_ALL_INT "\n\t"
-	GET_CURRENT(%%d0) "\n\t"
-	/* get vector number from stack frame and convert to source */
-	"bfextu	%%sp@(%c1){#4,#10},%%d0\n\t"
-	"subw	#(0x40-8),%%d0\n\t"
-	"jpl	1f\n\t"
-	"addw	#(0x40-8-0x18),%%d0\n"
-    "1:\tlea	%a0,%%a0\n\t"
-	"addql	#1,%%a0@(%%d0:l:4)\n\t"
-	"lea	irq_handler,%%a0\n\t"
-	"lea	%%a0@(%%d0:l:8),%%a0\n\t"
-	"pea	%%sp@\n\t"		/* push frame address */
-	"movel	%%a0@(4),%%sp@-\n\t"	/* push handler data */
-	"movel	%%d0,%%sp@-\n\t"	/* push int number */
-	"movel	%%a0@,%%a0\n\t"
-	"jsr	%%a0@\n\t"		/* and call the handler */
-	"addql	#8,%%sp\n\t"
-	"addql	#4,%%sp\n\t"
-	"jbra	ret_from_interrupt"
-	 : : "i" (&kstat_cpu(0).irqs), "n" (PT_OFF_FORMATVEC),
-	     "m" (preempt_count()), "di" (HARDIRQ_OFFSET)
-);
-	for (;;);
-}
-#endif
-
 /*
  * Bitmap for free interrupt vector numbers
  * (new vectors starting from 0x70 can be allocated by
@@ -320,31 +84,44 @@
 
 extern int atari_SCC_reset_done;
 
-static int atari_startup_irq(unsigned int irq)
+static unsigned int atari_irq_startup(struct irq_data *data)
 {
-	m68k_irq_startup(irq);
+	unsigned int irq = data->irq;
+
+	m68k_irq_startup(data);
 	atari_turnon_irq(irq);
 	atari_enable_irq(irq);
 	return 0;
 }
 
-static void atari_shutdown_irq(unsigned int irq)
+static void atari_irq_shutdown(struct irq_data *data)
 {
+	unsigned int irq = data->irq;
+
 	atari_disable_irq(irq);
 	atari_turnoff_irq(irq);
-	m68k_irq_shutdown(irq);
+	m68k_irq_shutdown(data);
 
 	if (irq == IRQ_AUTO_4)
 	    vectors[VEC_INT4] = falcon_hblhandler;
 }
 
-static struct irq_controller atari_irq_controller = {
+static void atari_irq_enable(struct irq_data *data)
+{
+	atari_enable_irq(data->irq);
+}
+
+static void atari_irq_disable(struct irq_data *data)
+{
+	atari_disable_irq(data->irq);
+}
+
+static struct irq_chip atari_irq_chip = {
 	.name		= "atari",
-	.lock		= __SPIN_LOCK_UNLOCKED(atari_irq_controller.lock),
-	.startup	= atari_startup_irq,
-	.shutdown	= atari_shutdown_irq,
-	.enable		= atari_enable_irq,
-	.disable	= atari_disable_irq,
+	.irq_startup	= atari_irq_startup,
+	.irq_shutdown	= atari_irq_shutdown,
+	.irq_enable	= atari_irq_enable,
+	.irq_disable	= atari_irq_disable,
 };
 
 /*
@@ -360,8 +137,9 @@
 
 void __init atari_init_IRQ(void)
 {
-	m68k_setup_user_interrupt(VEC_USER, NUM_ATARI_SOURCES - IRQ_USER, NULL);
-	m68k_setup_irq_controller(&atari_irq_controller, 1, NUM_ATARI_SOURCES - 1);
+	m68k_setup_user_interrupt(VEC_USER, NUM_ATARI_SOURCES - IRQ_USER);
+	m68k_setup_irq_controller(&atari_irq_chip, handle_simple_irq, 1,
+				  NUM_ATARI_SOURCES - 1);
 
 	/* Initialize the MFP(s) */
 
diff --git a/arch/m68k/bvme6000/config.c b/arch/m68k/bvme6000/config.c
index 1edd950..8128647 100644
--- a/arch/m68k/bvme6000/config.c
+++ b/arch/m68k/bvme6000/config.c
@@ -86,7 +86,7 @@
  */
 static void __init bvme6000_init_IRQ(void)
 {
-	m68k_setup_user_interrupt(VEC_USER, 192, NULL);
+	m68k_setup_user_interrupt(VEC_USER, 192);
 }
 
 void __init config_bvme6000(void)
diff --git a/arch/m68k/hp300/time.c b/arch/m68k/hp300/time.c
index f6312c7..c87fe69 100644
--- a/arch/m68k/hp300/time.c
+++ b/arch/m68k/hp300/time.c
@@ -70,7 +70,7 @@
 
   asm volatile(" movpw %0,%1@(5)" : : "d" (INTVAL), "a" (CLOCKBASE));
 
-  if (request_irq(IRQ_AUTO_6, hp300_tick, IRQ_FLG_STD, "timer tick", vector))
+  if (request_irq(IRQ_AUTO_6, hp300_tick, 0, "timer tick", vector))
     pr_err("Couldn't register timer interrupt\n");
 
   out_8(CLOCKBASE + CLKCR2, 0x1);		/* select CR1 */
diff --git a/arch/m68k/include/asm/hardirq.h b/arch/m68k/include/asm/hardirq.h
index 870e534..db30ed2 100644
--- a/arch/m68k/include/asm/hardirq.h
+++ b/arch/m68k/include/asm/hardirq.h
@@ -18,6 +18,11 @@
 
 #ifdef CONFIG_MMU
 
+static inline void ack_bad_irq(unsigned int irq)
+{
+	pr_crit("unexpected IRQ trap at vector %02x\n", irq);
+}
+
 /* entry.S is sensitive to the offsets of these fields */
 typedef struct {
 	unsigned int __softirq_pending;
diff --git a/arch/m68k/include/asm/irq.h b/arch/m68k/include/asm/irq.h
index 69ed0d74..6198df5 100644
--- a/arch/m68k/include/asm/irq.h
+++ b/arch/m68k/include/asm/irq.h
@@ -27,11 +27,6 @@
 
 #ifdef CONFIG_MMU
 
-#include <linux/linkage.h>
-#include <linux/hardirq.h>
-#include <linux/irqreturn.h>
-#include <linux/spinlock_types.h>
-
 /*
  * Interrupt source definitions
  * General interrupt sources are the level 1-7.
@@ -54,10 +49,6 @@
 
 #define IRQ_USER	8
 
-extern unsigned int irq_canonicalize(unsigned int irq);
-
-struct pt_regs;
-
 /*
  * various flags for request_irq() - the Amiga now uses the standard
  * mechanism like all other architectures - IRQF_DISABLED and
@@ -71,57 +62,27 @@
 #define IRQ_FLG_STD	(0x8000)	/* internally used		*/
 #endif
 
-/*
- * This structure is used to chain together the ISRs for a particular
- * interrupt source (if it supports chaining).
- */
-typedef struct irq_node {
-	irqreturn_t	(*handler)(int, void *);
-	void		*dev_id;
-	struct irq_node *next;
-	unsigned long	flags;
-	const char	*devname;
-} irq_node_t;
+struct irq_data;
+struct irq_chip;
+struct irq_desc;
+extern unsigned int m68k_irq_startup(struct irq_data *data);
+extern unsigned int m68k_irq_startup_irq(unsigned int irq);
+extern void m68k_irq_shutdown(struct irq_data *data);
+extern void m68k_setup_auto_interrupt(void (*handler)(unsigned int,
+						      struct pt_regs *));
+extern void m68k_setup_user_interrupt(unsigned int vec, unsigned int cnt);
+extern void m68k_setup_irq_controller(struct irq_chip *,
+				      void (*handle)(unsigned int irq,
+						     struct irq_desc *desc),
+				      unsigned int irq, unsigned int cnt);
 
-/*
- * This structure has only 4 elements for speed reasons
- */
-struct irq_handler {
-	int		(*handler)(int, void *);
-	unsigned long	flags;
-	void		*dev_id;
-	const char	*devname;
-};
-
-struct irq_controller {
-	const char *name;
-	spinlock_t lock;
-	int (*startup)(unsigned int irq);
-	void (*shutdown)(unsigned int irq);
-	void (*enable)(unsigned int irq);
-	void (*disable)(unsigned int irq);
-};
-
-extern int m68k_irq_startup(unsigned int);
-extern void m68k_irq_shutdown(unsigned int);
-
-/*
- * This function returns a new irq_node_t
- */
-extern irq_node_t *new_irq_node(void);
-
-extern void m68k_setup_auto_interrupt(void (*handler)(unsigned int, struct pt_regs *));
-extern void m68k_setup_user_interrupt(unsigned int vec, unsigned int cnt,
-				      void (*handler)(unsigned int, struct pt_regs *));
-extern void m68k_setup_irq_controller(struct irq_controller *, unsigned int, unsigned int);
-
-asmlinkage void m68k_handle_int(unsigned int);
-asmlinkage void __m68k_handle_int(unsigned int, struct pt_regs *);
+extern unsigned int irq_canonicalize(unsigned int irq);
 
 #else
 #define irq_canonicalize(irq)  (irq)
 #endif /* CONFIG_MMU */
 
 asmlinkage void do_IRQ(int irq, struct pt_regs *regs);
+extern atomic_t irq_err_count;
 
 #endif /* _M68K_IRQ_H_ */
diff --git a/arch/m68k/include/asm/macintosh.h b/arch/m68k/include/asm/macintosh.h
index c2a1c5e..12ebe43 100644
--- a/arch/m68k/include/asm/macintosh.h
+++ b/arch/m68k/include/asm/macintosh.h
@@ -12,6 +12,8 @@
 extern void mac_poweroff(void);
 extern void mac_init_IRQ(void);
 extern int mac_irq_pending(unsigned int);
+extern void mac_irq_enable(struct irq_data *data);
+extern void mac_irq_disable(struct irq_data *data);
 
 /*
  *	Floppy driver magic hook - probably shouldn't be here
diff --git a/arch/m68k/include/asm/q40ints.h b/arch/m68k/include/asm/q40ints.h
index 3d970af..22f12c9 100644
--- a/arch/m68k/include/asm/q40ints.h
+++ b/arch/m68k/include/asm/q40ints.h
@@ -24,6 +24,3 @@
 #define Q40_IRQ10_MASK       (1<<5)
 #define Q40_IRQ14_MASK       (1<<6)
 #define Q40_IRQ15_MASK       (1<<7)
-
-extern unsigned long q40_probe_irq_on (void);
-extern int q40_probe_irq_off (unsigned long irqs);
diff --git a/arch/m68k/include/asm/unistd.h b/arch/m68k/include/asm/unistd.h
index 43f984e..303192f 100644
--- a/arch/m68k/include/asm/unistd.h
+++ b/arch/m68k/include/asm/unistd.h
@@ -350,10 +350,12 @@
 #define __NR_clock_adjtime	342
 #define __NR_syncfs		343
 #define __NR_setns		344
+#define __NR_process_vm_readv	345
+#define __NR_process_vm_writev	346
 
 #ifdef __KERNEL__
 
-#define NR_syscalls		345
+#define NR_syscalls		347
 
 #define __ARCH_WANT_IPC_PARSE_VERSION
 #define __ARCH_WANT_OLD_READDIR
diff --git a/arch/m68k/kernel/Makefile b/arch/m68k/kernel/Makefile
index e7f0f2e..c569619 100644
--- a/arch/m68k/kernel/Makefile
+++ b/arch/m68k/kernel/Makefile
@@ -6,16 +6,15 @@
 extra-$(CONFIG_SUN3)	:= sun3-head.o
 extra-y			+= vmlinux.lds
 
-obj-y	:= entry.o m68k_ksyms.o module.o process.o ptrace.o setup.o signal.o \
-	   sys_m68k.o syscalltable.o time.o traps.o
+obj-y	:= entry.o irq.o m68k_ksyms.o module.o process.o ptrace.o setup.o \
+	   signal.o sys_m68k.o syscalltable.o time.o traps.o
 
-obj-$(CONFIG_MMU)	+= ints.o devres.o vectors.o
-devres-$(CONFIG_MMU)	= ../../../kernel/irq/devres.o
+obj-$(CONFIG_MMU)	+= ints.o vectors.o
 
 ifndef CONFIG_MMU_SUN3
 obj-y			+= dma.o
 endif
 ifndef CONFIG_MMU
-obj-y			+= init_task.o irq.o
+obj-y			+= init_task.o
 endif
 
diff --git a/arch/m68k/kernel/entry_mm.S b/arch/m68k/kernel/entry_mm.S
index bd0ec05..c713f51 100644
--- a/arch/m68k/kernel/entry_mm.S
+++ b/arch/m68k/kernel/entry_mm.S
@@ -48,7 +48,7 @@
 .globl sys_fork, sys_clone, sys_vfork
 .globl ret_from_interrupt, bad_interrupt
 .globl auto_irqhandler_fixup
-.globl user_irqvec_fixup, user_irqhandler_fixup
+.globl user_irqvec_fixup
 
 .text
 ENTRY(buserr)
@@ -207,7 +207,7 @@
 	movel	%sp,%sp@-
 	movel	%d0,%sp@-		|  put vector # on stack
 auto_irqhandler_fixup = . + 2
-	jsr	__m68k_handle_int	|  process the IRQ
+	jsr	do_IRQ			|  process the IRQ
 	addql	#8,%sp			|  pop parameters off stack
 
 ret_from_interrupt:
@@ -240,8 +240,7 @@
 
 	movel	%sp,%sp@-
 	movel	%d0,%sp@-		|  put vector # on stack
-user_irqhandler_fixup = . + 2
-	jsr	__m68k_handle_int	|  process the IRQ
+	jsr	do_IRQ			|  process the IRQ
 	addql	#8,%sp			|  pop parameters off stack
 
 	subqb	#1,%curptr@(TASK_INFO+TINFO_PREEMPT+1)
diff --git a/arch/m68k/kernel/ints.c b/arch/m68k/kernel/ints.c
index 761ee04..74fefac 100644
--- a/arch/m68k/kernel/ints.c
+++ b/arch/m68k/kernel/ints.c
@@ -4,25 +4,6 @@
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file COPYING in the main directory of this archive
  * for more details.
- *
- * 07/03/96: Timer initialization, and thus mach_sched_init(),
- *           removed from request_irq() and moved to init_time().
- *           We should therefore consider renaming our add_isr() and
- *           remove_isr() to request_irq() and free_irq()
- *           respectively, so they are compliant with the other
- *           architectures.                                     /Jes
- * 11/07/96: Changed all add_/remove_isr() to request_/free_irq() calls.
- *           Removed irq list support, if any machine needs an irq server
- *           it must implement this itself (as it's already done), instead
- *           only default handler are used with mach_default_handler.
- *           request_irq got some flags different from other architectures:
- *           - IRQ_FLG_REPLACE : Replace an existing handler (the default one
- *                               can be replaced without this flag)
- *           - IRQ_FLG_LOCK : handler can't be replaced
- *           There are other machine depending flags, see there
- *           If you want to replace a default handler you should know what
- *           you're doing, since it might handle different other irq sources
- *           which must be served                               /Roman Zippel
  */
 
 #include <linux/module.h>
@@ -47,33 +28,22 @@
 #endif
 
 extern u32 auto_irqhandler_fixup[];
-extern u32 user_irqhandler_fixup[];
 extern u16 user_irqvec_fixup[];
 
-/* table for system interrupt handlers */
-static struct irq_node *irq_list[NR_IRQS];
-static struct irq_controller *irq_controller[NR_IRQS];
-static int irq_depth[NR_IRQS];
-
 static int m68k_first_user_vec;
 
-static struct irq_controller auto_irq_controller = {
+static struct irq_chip auto_irq_chip = {
 	.name		= "auto",
-	.lock		= __SPIN_LOCK_UNLOCKED(auto_irq_controller.lock),
-	.startup	= m68k_irq_startup,
-	.shutdown	= m68k_irq_shutdown,
+	.irq_startup	= m68k_irq_startup,
+	.irq_shutdown	= m68k_irq_shutdown,
 };
 
-static struct irq_controller user_irq_controller = {
+static struct irq_chip user_irq_chip = {
 	.name		= "user",
-	.lock		= __SPIN_LOCK_UNLOCKED(user_irq_controller.lock),
-	.startup	= m68k_irq_startup,
-	.shutdown	= m68k_irq_shutdown,
+	.irq_startup	= m68k_irq_startup,
+	.irq_shutdown	= m68k_irq_shutdown,
 };
 
-#define NUM_IRQ_NODES 100
-static irq_node_t nodes[NUM_IRQ_NODES];
-
 /*
  * void init_IRQ(void)
  *
@@ -96,7 +66,7 @@
 	}
 
 	for (i = IRQ_AUTO_1; i <= IRQ_AUTO_7; i++)
-		irq_controller[i] = &auto_irq_controller;
+		irq_set_chip_and_handler(i, &auto_irq_chip, handle_simple_irq);
 
 	mach_init_IRQ();
 }
@@ -106,7 +76,7 @@
  * @handler: called from auto vector interrupts
  *
  * setup the handler to be called from auto vector interrupts instead of the
- * standard __m68k_handle_int(), it will be called with irq numbers in the range
+ * standard do_IRQ(), it will be called with irq numbers in the range
  * from IRQ_AUTO_1 - IRQ_AUTO_7.
  */
 void __init m68k_setup_auto_interrupt(void (*handler)(unsigned int, struct pt_regs *))
@@ -120,217 +90,49 @@
  * m68k_setup_user_interrupt
  * @vec: first user vector interrupt to handle
  * @cnt: number of active user vector interrupts
- * @handler: called from user vector interrupts
  *
  * setup user vector interrupts, this includes activating the specified range
  * of interrupts, only then these interrupts can be requested (note: this is
- * different from auto vector interrupts). An optional handler can be installed
- * to be called instead of the default __m68k_handle_int(), it will be called
- * with irq numbers starting from IRQ_USER.
+ * different from auto vector interrupts).
  */
-void __init m68k_setup_user_interrupt(unsigned int vec, unsigned int cnt,
-				      void (*handler)(unsigned int, struct pt_regs *))
+void __init m68k_setup_user_interrupt(unsigned int vec, unsigned int cnt)
 {
 	int i;
 
 	BUG_ON(IRQ_USER + cnt > NR_IRQS);
 	m68k_first_user_vec = vec;
 	for (i = 0; i < cnt; i++)
-		irq_controller[IRQ_USER + i] = &user_irq_controller;
+		irq_set_chip(IRQ_USER + i, &user_irq_chip);
 	*user_irqvec_fixup = vec - IRQ_USER;
-	if (handler)
-		*user_irqhandler_fixup = (u32)handler;
 	flush_icache();
 }
 
 /**
  * m68k_setup_irq_controller
- * @contr: irq controller which controls specified irq
+ * @chip: irq chip which controls specified irq
+ * @handle: flow handler which handles specified irq
  * @irq: first irq to be managed by the controller
+ * @cnt: number of irqs to be managed by the controller
  *
  * Change the controller for the specified range of irq, which will be used to
  * manage these irq. auto/user irq already have a default controller, which can
  * be changed as well, but the controller probably should use m68k_irq_startup/
  * m68k_irq_shutdown.
  */
-void m68k_setup_irq_controller(struct irq_controller *contr, unsigned int irq,
+void m68k_setup_irq_controller(struct irq_chip *chip,
+			       irq_flow_handler_t handle, unsigned int irq,
 			       unsigned int cnt)
 {
 	int i;
 
-	for (i = 0; i < cnt; i++)
-		irq_controller[irq + i] = contr;
+	for (i = 0; i < cnt; i++) {
+		irq_set_chip(irq + i, chip);
+		if (handle)
+			irq_set_handler(irq + i, handle);
+	}
 }
 
-irq_node_t *new_irq_node(void)
-{
-	irq_node_t *node;
-	short i;
-
-	for (node = nodes, i = NUM_IRQ_NODES-1; i >= 0; node++, i--) {
-		if (!node->handler) {
-			memset(node, 0, sizeof(*node));
-			return node;
-		}
-	}
-
-	printk ("new_irq_node: out of nodes\n");
-	return NULL;
-}
-
-int setup_irq(unsigned int irq, struct irq_node *node)
-{
-	struct irq_controller *contr;
-	struct irq_node **prev;
-	unsigned long flags;
-
-	if (irq >= NR_IRQS || !(contr = irq_controller[irq])) {
-		printk("%s: Incorrect IRQ %d from %s\n",
-		       __func__, irq, node->devname);
-		return -ENXIO;
-	}
-
-	spin_lock_irqsave(&contr->lock, flags);
-
-	prev = irq_list + irq;
-	if (*prev) {
-		/* Can't share interrupts unless both agree to */
-		if (!((*prev)->flags & node->flags & IRQF_SHARED)) {
-			spin_unlock_irqrestore(&contr->lock, flags);
-			return -EBUSY;
-		}
-		while (*prev)
-			prev = &(*prev)->next;
-	}
-
-	if (!irq_list[irq]) {
-		if (contr->startup)
-			contr->startup(irq);
-		else
-			contr->enable(irq);
-	}
-	node->next = NULL;
-	*prev = node;
-
-	spin_unlock_irqrestore(&contr->lock, flags);
-
-	return 0;
-}
-
-int request_irq(unsigned int irq,
-		irq_handler_t handler,
-		unsigned long flags, const char *devname, void *dev_id)
-{
-	struct irq_node *node;
-	int res;
-
-	node = new_irq_node();
-	if (!node)
-		return -ENOMEM;
-
-	node->handler = handler;
-	node->flags   = flags;
-	node->dev_id  = dev_id;
-	node->devname = devname;
-
-	res = setup_irq(irq, node);
-	if (res)
-		node->handler = NULL;
-
-	return res;
-}
-
-EXPORT_SYMBOL(request_irq);
-
-void free_irq(unsigned int irq, void *dev_id)
-{
-	struct irq_controller *contr;
-	struct irq_node **p, *node;
-	unsigned long flags;
-
-	if (irq >= NR_IRQS || !(contr = irq_controller[irq])) {
-		printk("%s: Incorrect IRQ %d\n", __func__, irq);
-		return;
-	}
-
-	spin_lock_irqsave(&contr->lock, flags);
-
-	p = irq_list + irq;
-	while ((node = *p)) {
-		if (node->dev_id == dev_id)
-			break;
-		p = &node->next;
-	}
-
-	if (node) {
-		*p = node->next;
-		node->handler = NULL;
-	} else
-		printk("%s: Removing probably wrong IRQ %d\n",
-		       __func__, irq);
-
-	if (!irq_list[irq]) {
-		if (contr->shutdown)
-			contr->shutdown(irq);
-		else
-			contr->disable(irq);
-	}
-
-	spin_unlock_irqrestore(&contr->lock, flags);
-}
-
-EXPORT_SYMBOL(free_irq);
-
-void enable_irq(unsigned int irq)
-{
-	struct irq_controller *contr;
-	unsigned long flags;
-
-	if (irq >= NR_IRQS || !(contr = irq_controller[irq])) {
-		printk("%s: Incorrect IRQ %d\n",
-		       __func__, irq);
-		return;
-	}
-
-	spin_lock_irqsave(&contr->lock, flags);
-	if (irq_depth[irq]) {
-		if (!--irq_depth[irq]) {
-			if (contr->enable)
-				contr->enable(irq);
-		}
-	} else
-		WARN_ON(1);
-	spin_unlock_irqrestore(&contr->lock, flags);
-}
-
-EXPORT_SYMBOL(enable_irq);
-
-void disable_irq(unsigned int irq)
-{
-	struct irq_controller *contr;
-	unsigned long flags;
-
-	if (irq >= NR_IRQS || !(contr = irq_controller[irq])) {
-		printk("%s: Incorrect IRQ %d\n",
-		       __func__, irq);
-		return;
-	}
-
-	spin_lock_irqsave(&contr->lock, flags);
-	if (!irq_depth[irq]++) {
-		if (contr->disable)
-			contr->disable(irq);
-	}
-	spin_unlock_irqrestore(&contr->lock, flags);
-}
-
-EXPORT_SYMBOL(disable_irq);
-
-void disable_irq_nosync(unsigned int irq) __attribute__((alias("disable_irq")));
-
-EXPORT_SYMBOL(disable_irq_nosync);
-
-int m68k_irq_startup(unsigned int irq)
+unsigned int m68k_irq_startup_irq(unsigned int irq)
 {
 	if (irq <= IRQ_AUTO_7)
 		vectors[VEC_SPUR + irq] = auto_inthandler;
@@ -339,8 +141,15 @@
 	return 0;
 }
 
-void m68k_irq_shutdown(unsigned int irq)
+unsigned int m68k_irq_startup(struct irq_data *data)
 {
+	return m68k_irq_startup_irq(data->irq);
+}
+
+void m68k_irq_shutdown(struct irq_data *data)
+{
+	unsigned int irq = data->irq;
+
 	if (irq <= IRQ_AUTO_7)
 		vectors[VEC_SPUR + irq] = bad_inthandler;
 	else
@@ -348,33 +157,6 @@
 }
 
 
-/*
- * Do we need these probe functions on the m68k?
- *
- *  ... may be useful with ISA devices
- */
-unsigned long probe_irq_on (void)
-{
-#ifdef CONFIG_Q40
-	if (MACH_IS_Q40)
-		return q40_probe_irq_on();
-#endif
-	return 0;
-}
-
-EXPORT_SYMBOL(probe_irq_on);
-
-int probe_irq_off (unsigned long irqs)
-{
-#ifdef CONFIG_Q40
-	if (MACH_IS_Q40)
-		return q40_probe_irq_off(irqs);
-#endif
-	return 0;
-}
-
-EXPORT_SYMBOL(probe_irq_off);
-
 unsigned int irq_canonicalize(unsigned int irq)
 {
 #ifdef CONFIG_Q40
@@ -386,52 +168,9 @@
 
 EXPORT_SYMBOL(irq_canonicalize);
 
-asmlinkage void m68k_handle_int(unsigned int irq)
-{
-	struct irq_node *node;
-	kstat_cpu(0).irqs[irq]++;
-	node = irq_list[irq];
-	do {
-		node->handler(irq, node->dev_id);
-		node = node->next;
-	} while (node);
-}
-
-asmlinkage void __m68k_handle_int(unsigned int irq, struct pt_regs *regs)
-{
-	struct pt_regs *old_regs;
-	old_regs = set_irq_regs(regs);
-	m68k_handle_int(irq);
-	set_irq_regs(old_regs);
-}
 
 asmlinkage void handle_badint(struct pt_regs *regs)
 {
-	kstat_cpu(0).irqs[0]++;
-	printk("unexpected interrupt from %u\n", regs->vector);
+	atomic_inc(&irq_err_count);
+	pr_warn("unexpected interrupt from %u\n", regs->vector);
 }
-
-int show_interrupts(struct seq_file *p, void *v)
-{
-	struct irq_controller *contr;
-	struct irq_node *node;
-	int i = *(loff_t *) v;
-
-	/* autovector interrupts */
-	if (irq_list[i]) {
-		contr = irq_controller[i];
-		node = irq_list[i];
-		seq_printf(p, "%-8s %3u: %10u %s", contr->name, i, kstat_cpu(0).irqs[i], node->devname);
-		while ((node = node->next))
-			seq_printf(p, ", %s", node->devname);
-		seq_puts(p, "\n");
-	}
-	return 0;
-}
-
-#ifdef CONFIG_PROC_FS
-void init_irq_proc(void)
-{
-	/* Insert /proc/irq driver here */
-}
-#endif
diff --git a/arch/m68k/kernel/syscalltable.S b/arch/m68k/kernel/syscalltable.S
index c468f2e..ce827b3 100644
--- a/arch/m68k/kernel/syscalltable.S
+++ b/arch/m68k/kernel/syscalltable.S
@@ -365,4 +365,6 @@
 	.long sys_clock_adjtime
 	.long sys_syncfs
 	.long sys_setns
+	.long sys_process_vm_readv	/* 345 */
+	.long sys_process_vm_writev
 
diff --git a/arch/m68k/mac/baboon.c b/arch/m68k/mac/baboon.c
index 2a96beb..b403924 100644
--- a/arch/m68k/mac/baboon.c
+++ b/arch/m68k/mac/baboon.c
@@ -11,6 +11,7 @@
 #include <linux/mm.h>
 #include <linux/delay.h>
 #include <linux/init.h>
+#include <linux/irq.h>
 
 #include <asm/traps.h>
 #include <asm/bootinfo.h>
@@ -20,9 +21,6 @@
 
 /* #define DEBUG_IRQS */
 
-extern void mac_enable_irq(unsigned int);
-extern void mac_disable_irq(unsigned int);
-
 int baboon_present;
 static volatile struct baboon *baboon;
 static unsigned char baboon_disabled;
@@ -53,7 +51,7 @@
  * Baboon interrupt handler. This works a lot like a VIA.
  */
 
-static irqreturn_t baboon_irq(int irq, void *dev_id)
+static void baboon_irq(unsigned int irq, struct irq_desc *desc)
 {
 	int irq_bit, irq_num;
 	unsigned char events;
@@ -64,15 +62,16 @@
 		(uint) baboon->mb_status);
 #endif
 
-	if (!(events = baboon->mb_ifr & 0x07))
-		return IRQ_NONE;
+	events = baboon->mb_ifr & 0x07;
+	if (!events)
+		return;
 
 	irq_num = IRQ_BABOON_0;
 	irq_bit = 1;
 	do {
 	        if (events & irq_bit) {
 			baboon->mb_ifr &= ~irq_bit;
-			m68k_handle_int(irq_num);
+			generic_handle_irq(irq_num);
 		}
 		irq_bit <<= 1;
 		irq_num++;
@@ -82,7 +81,6 @@
 	/* for now we need to smash all interrupts */
 	baboon->mb_ifr &= ~events;
 #endif
-	return IRQ_HANDLED;
 }
 
 /*
@@ -92,8 +90,7 @@
 void __init baboon_register_interrupts(void)
 {
 	baboon_disabled = 0;
-	if (request_irq(IRQ_NUBUS_C, baboon_irq, 0, "baboon", (void *)baboon))
-		pr_err("Couldn't register baboon interrupt\n");
+	irq_set_chained_handler(IRQ_NUBUS_C, baboon_irq);
 }
 
 /*
@@ -111,7 +108,7 @@
 
 	baboon_disabled &= ~(1 << irq_idx);
 	if (!baboon_disabled)
-		mac_enable_irq(IRQ_NUBUS_C);
+		mac_irq_enable(irq_get_irq_data(IRQ_NUBUS_C));
 }
 
 void baboon_irq_disable(int irq)
@@ -124,7 +121,7 @@
 
 	baboon_disabled |= 1 << irq_idx;
 	if (baboon_disabled)
-		mac_disable_irq(IRQ_NUBUS_C);
+		mac_irq_disable(irq_get_irq_data(IRQ_NUBUS_C));
 }
 
 void baboon_irq_clear(int irq)
diff --git a/arch/m68k/mac/iop.c b/arch/m68k/mac/iop.c
index 1ad4e9d..a5462cc 100644
--- a/arch/m68k/mac/iop.c
+++ b/arch/m68k/mac/iop.c
@@ -305,15 +305,13 @@
 {
 	if (iop_ism_present) {
 		if (oss_present) {
-			if (request_irq(OSS_IRQLEV_IOPISM, iop_ism_irq,
-					IRQ_FLG_LOCK, "ISM IOP",
-					(void *) IOP_NUM_ISM))
+			if (request_irq(OSS_IRQLEV_IOPISM, iop_ism_irq, 0,
+					"ISM IOP", (void *)IOP_NUM_ISM))
 				pr_err("Couldn't register ISM IOP interrupt\n");
 			oss_irq_enable(IRQ_MAC_ADB);
 		} else {
-			if (request_irq(IRQ_VIA2_0, iop_ism_irq,
-					IRQ_FLG_LOCK|IRQ_FLG_FAST, "ISM IOP",
-					(void *) IOP_NUM_ISM))
+			if (request_irq(IRQ_VIA2_0, iop_ism_irq, 0, "ISM IOP",
+					(void *)IOP_NUM_ISM))
 				pr_err("Couldn't register ISM IOP interrupt\n");
 		}
 		if (!iop_alive(iop_base[IOP_NUM_ISM])) {
diff --git a/arch/m68k/mac/macints.c b/arch/m68k/mac/macints.c
index f92190c..ba220b7 100644
--- a/arch/m68k/mac/macints.c
+++ b/arch/m68k/mac/macints.c
@@ -190,14 +190,10 @@
 
 /* #define DEBUG_MACINTS */
 
-void mac_enable_irq(unsigned int irq);
-void mac_disable_irq(unsigned int irq);
-
-static struct irq_controller mac_irq_controller = {
+static struct irq_chip mac_irq_chip = {
 	.name		= "mac",
-	.lock		= __SPIN_LOCK_UNLOCKED(mac_irq_controller.lock),
-	.enable		= mac_enable_irq,
-	.disable	= mac_disable_irq,
+	.irq_enable	= mac_irq_enable,
+	.irq_disable	= mac_irq_disable,
 };
 
 void __init mac_init_IRQ(void)
@@ -205,7 +201,7 @@
 #ifdef DEBUG_MACINTS
 	printk("mac_init_IRQ(): Setting things up...\n");
 #endif
-	m68k_setup_irq_controller(&mac_irq_controller, IRQ_USER,
+	m68k_setup_irq_controller(&mac_irq_chip, handle_simple_irq, IRQ_USER,
 				  NUM_MAC_SOURCES - IRQ_USER);
 	/* Make sure the SONIC interrupt is cleared or things get ugly */
 #ifdef SHUTUP_SONIC
@@ -241,16 +237,17 @@
 }
 
 /*
- *  mac_enable_irq - enable an interrupt source
- * mac_disable_irq - disable an interrupt source
+ *  mac_irq_enable - enable an interrupt source
+ * mac_irq_disable - disable an interrupt source
  *   mac_clear_irq - clears a pending interrupt
- * mac_pending_irq - Returns the pending status of an IRQ (nonzero = pending)
+ * mac_irq_pending - returns the pending status of an IRQ (nonzero = pending)
  *
  * These routines are just dispatchers to the VIA/OSS/PSC routines.
  */
 
-void mac_enable_irq(unsigned int irq)
+void mac_irq_enable(struct irq_data *data)
 {
+	int irq = data->irq;
 	int irq_src = IRQ_SRC(irq);
 
 	switch(irq_src) {
@@ -283,8 +280,9 @@
 	}
 }
 
-void mac_disable_irq(unsigned int irq)
+void mac_irq_disable(struct irq_data *data)
 {
+	int irq = data->irq;
 	int irq_src = IRQ_SRC(irq);
 
 	switch(irq_src) {
diff --git a/arch/m68k/mac/oss.c b/arch/m68k/mac/oss.c
index a9c0f5a..a4c82da 100644
--- a/arch/m68k/mac/oss.c
+++ b/arch/m68k/mac/oss.c
@@ -19,6 +19,7 @@
 #include <linux/mm.h>
 #include <linux/delay.h>
 #include <linux/init.h>
+#include <linux/irq.h>
 
 #include <asm/bootinfo.h>
 #include <asm/macintosh.h>
@@ -29,10 +30,7 @@
 int oss_present;
 volatile struct mac_oss *oss;
 
-static irqreturn_t oss_irq(int, void *);
-static irqreturn_t oss_nubus_irq(int, void *);
-
-extern irqreturn_t via1_irq(int, void *);
+extern void via1_irq(unsigned int irq, struct irq_desc *desc);
 
 /*
  * Initialize the OSS
@@ -60,26 +58,6 @@
 }
 
 /*
- * Register the OSS and NuBus interrupt dispatchers.
- */
-
-void __init oss_register_interrupts(void)
-{
-	if (request_irq(OSS_IRQLEV_SCSI, oss_irq, IRQ_FLG_LOCK,
-			"scsi", (void *) oss))
-		pr_err("Couldn't register %s interrupt\n", "scsi");
-	if (request_irq(OSS_IRQLEV_NUBUS, oss_nubus_irq, IRQ_FLG_LOCK,
-			"nubus", (void *) oss))
-		pr_err("Couldn't register %s interrupt\n", "nubus");
-	if (request_irq(OSS_IRQLEV_SOUND, oss_irq, IRQ_FLG_LOCK,
-			"sound", (void *) oss))
-		pr_err("Couldn't register %s interrupt\n", "sound");
-	if (request_irq(OSS_IRQLEV_VIA1, via1_irq, IRQ_FLG_LOCK,
-			"via1", (void *) via1))
-		pr_err("Couldn't register %s interrupt\n", "via1");
-}
-
-/*
  * Initialize OSS for Nubus access
  */
 
@@ -92,17 +70,17 @@
  * and SCSI; everything else is routed to its own autovector IRQ.
  */
 
-static irqreturn_t oss_irq(int irq, void *dev_id)
+static void oss_irq(unsigned int irq, struct irq_desc *desc)
 {
 	int events;
 
 	events = oss->irq_pending & (OSS_IP_SOUND|OSS_IP_SCSI);
 	if (!events)
-		return IRQ_NONE;
+		return;
 
 #ifdef DEBUG_IRQS
 	if ((console_loglevel == 10) && !(events & OSS_IP_SCSI)) {
-		printk("oss_irq: irq %d events = 0x%04X\n", irq,
+		printk("oss_irq: irq %u events = 0x%04X\n", irq,
 			(int) oss->irq_pending);
 	}
 #endif
@@ -113,11 +91,10 @@
 		/* FIXME: call sound handler */
 	} else if (events & OSS_IP_SCSI) {
 		oss->irq_pending &= ~OSS_IP_SCSI;
-		m68k_handle_int(IRQ_MAC_SCSI);
+		generic_handle_irq(IRQ_MAC_SCSI);
 	} else {
 		/* FIXME: error check here? */
 	}
-	return IRQ_HANDLED;
 }
 
 /*
@@ -126,13 +103,13 @@
  * Unlike the VIA/RBV this is on its own autovector interrupt level.
  */
 
-static irqreturn_t oss_nubus_irq(int irq, void *dev_id)
+static void oss_nubus_irq(unsigned int irq, struct irq_desc *desc)
 {
 	int events, irq_bit, i;
 
 	events = oss->irq_pending & OSS_IP_NUBUS;
 	if (!events)
-		return IRQ_NONE;
+		return;
 
 #ifdef DEBUG_NUBUS_INT
 	if (console_loglevel > 7) {
@@ -148,10 +125,21 @@
 		irq_bit >>= 1;
 		if (events & irq_bit) {
 			oss->irq_pending &= ~irq_bit;
-			m68k_handle_int(NUBUS_SOURCE_BASE + i);
+			generic_handle_irq(NUBUS_SOURCE_BASE + i);
 		}
 	} while(events & (irq_bit - 1));
-	return IRQ_HANDLED;
+}
+
+/*
+ * Register the OSS and NuBus interrupt dispatchers.
+ */
+
+void __init oss_register_interrupts(void)
+{
+	irq_set_chained_handler(OSS_IRQLEV_SCSI, oss_irq);
+	irq_set_chained_handler(OSS_IRQLEV_NUBUS, oss_nubus_irq);
+	irq_set_chained_handler(OSS_IRQLEV_SOUND, oss_irq);
+	irq_set_chained_handler(OSS_IRQLEV_VIA1, via1_irq);
 }
 
 /*
diff --git a/arch/m68k/mac/psc.c b/arch/m68k/mac/psc.c
index a4c3eb6..e6c2d20 100644
--- a/arch/m68k/mac/psc.c
+++ b/arch/m68k/mac/psc.c
@@ -18,6 +18,7 @@
 #include <linux/mm.h>
 #include <linux/delay.h>
 #include <linux/init.h>
+#include <linux/irq.h>
 
 #include <asm/traps.h>
 #include <asm/bootinfo.h>
@@ -30,8 +31,6 @@
 int psc_present;
 volatile __u8 *psc;
 
-irqreturn_t psc_irq(int, void *);
-
 /*
  * Debugging dump, used in various places to see what's going on.
  */
@@ -112,52 +111,52 @@
 }
 
 /*
- * Register the PSC interrupt dispatchers for autovector interrupts 3-6.
- */
-
-void __init psc_register_interrupts(void)
-{
-	if (request_irq(IRQ_AUTO_3, psc_irq, 0, "psc3", (void *) 0x30))
-		pr_err("Couldn't register psc%d interrupt\n", 3);
-	if (request_irq(IRQ_AUTO_4, psc_irq, 0, "psc4", (void *) 0x40))
-		pr_err("Couldn't register psc%d interrupt\n", 4);
-	if (request_irq(IRQ_AUTO_5, psc_irq, 0, "psc5", (void *) 0x50))
-		pr_err("Couldn't register psc%d interrupt\n", 5);
-	if (request_irq(IRQ_AUTO_6, psc_irq, 0, "psc6", (void *) 0x60))
-		pr_err("Couldn't register psc%d interrupt\n", 6);
-}
-
-/*
  * PSC interrupt handler. It's a lot like the VIA interrupt handler.
  */
 
-irqreturn_t psc_irq(int irq, void *dev_id)
+static void psc_irq(unsigned int irq, struct irq_desc *desc)
 {
-	int pIFR	= pIFRbase + ((int) dev_id);
-	int pIER	= pIERbase + ((int) dev_id);
+	unsigned int offset = (unsigned int)irq_desc_get_handler_data(desc);
+	int pIFR	= pIFRbase + offset;
+	int pIER	= pIERbase + offset;
 	int irq_num;
 	unsigned char irq_bit, events;
 
 #ifdef DEBUG_IRQS
-	printk("psc_irq: irq %d pIFR = 0x%02X pIER = 0x%02X\n",
+	printk("psc_irq: irq %u pIFR = 0x%02X pIER = 0x%02X\n",
 		irq, (int) psc_read_byte(pIFR), (int) psc_read_byte(pIER));
 #endif
 
 	events = psc_read_byte(pIFR) & psc_read_byte(pIER) & 0xF;
 	if (!events)
-		return IRQ_NONE;
+		return;
 
 	irq_num = irq << 3;
 	irq_bit = 1;
 	do {
 		if (events & irq_bit) {
 			psc_write_byte(pIFR, irq_bit);
-			m68k_handle_int(irq_num);
+			generic_handle_irq(irq_num);
 		}
 		irq_num++;
 		irq_bit <<= 1;
 	} while (events >= irq_bit);
-	return IRQ_HANDLED;
+}
+
+/*
+ * Register the PSC interrupt dispatchers for autovector interrupts 3-6.
+ */
+
+void __init psc_register_interrupts(void)
+{
+	irq_set_chained_handler(IRQ_AUTO_3, psc_irq);
+	irq_set_handler_data(IRQ_AUTO_3, (void *)0x30);
+	irq_set_chained_handler(IRQ_AUTO_4, psc_irq);
+	irq_set_handler_data(IRQ_AUTO_4, (void *)0x40);
+	irq_set_chained_handler(IRQ_AUTO_5, psc_irq);
+	irq_set_handler_data(IRQ_AUTO_5, (void *)0x50);
+	irq_set_chained_handler(IRQ_AUTO_6, psc_irq);
+	irq_set_handler_data(IRQ_AUTO_6, (void *)0x60);
 }
 
 void psc_irq_enable(int irq) {
diff --git a/arch/m68k/mac/via.c b/arch/m68k/mac/via.c
index e71166d..f1600ad 100644
--- a/arch/m68k/mac/via.c
+++ b/arch/m68k/mac/via.c
@@ -28,6 +28,7 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/irq.h>
 
 #include <asm/bootinfo.h>
 #include <asm/macintosh.h>
@@ -77,9 +78,6 @@
 static u8 nubus_disabled;
 
 void via_debug_dump(void);
-irqreturn_t via1_irq(int, void *);
-irqreturn_t via2_irq(int, void *);
-irqreturn_t via_nubus_irq(int, void *);
 void via_irq_enable(int irq);
 void via_irq_disable(int irq);
 void via_irq_clear(int irq);
@@ -281,40 +279,11 @@
 	via1[vT1CL] = MAC_CLOCK_LOW;
 	via1[vT1CH] = MAC_CLOCK_HIGH;
 
-	if (request_irq(IRQ_MAC_TIMER_1, func, IRQ_FLG_LOCK, "timer", func))
+	if (request_irq(IRQ_MAC_TIMER_1, func, 0, "timer", func))
 		pr_err("Couldn't register %s interrupt\n", "timer");
 }
 
 /*
- * Register the interrupt dispatchers for VIA or RBV machines only.
- */
-
-void __init via_register_interrupts(void)
-{
-	if (via_alt_mapping) {
-		if (request_irq(IRQ_AUTO_1, via1_irq,
-				IRQ_FLG_LOCK|IRQ_FLG_FAST, "software",
-				(void *) via1))
-			pr_err("Couldn't register %s interrupt\n", "software");
-		if (request_irq(IRQ_AUTO_6, via1_irq,
-				IRQ_FLG_LOCK|IRQ_FLG_FAST, "via1",
-				(void *) via1))
-			pr_err("Couldn't register %s interrupt\n", "via1");
-	} else {
-		if (request_irq(IRQ_AUTO_1, via1_irq,
-				IRQ_FLG_LOCK|IRQ_FLG_FAST, "via1",
-				(void *) via1))
-			pr_err("Couldn't register %s interrupt\n", "via1");
-	}
-	if (request_irq(IRQ_AUTO_2, via2_irq, IRQ_FLG_LOCK|IRQ_FLG_FAST,
-			"via2", (void *) via2))
-		pr_err("Couldn't register %s interrupt\n", "via2");
-	if (request_irq(IRQ_MAC_NUBUS, via_nubus_irq,
-			IRQ_FLG_LOCK|IRQ_FLG_FAST, "nubus", (void *) via2))
-		pr_err("Couldn't register %s interrupt\n", "nubus");
-}
-
-/*
  * Debugging dump, used in various places to see what's going on.
  */
 
@@ -446,48 +415,46 @@
  * via6522.c :-), disable/pending masks added.
  */
 
-irqreturn_t via1_irq(int irq, void *dev_id)
+void via1_irq(unsigned int irq, struct irq_desc *desc)
 {
 	int irq_num;
 	unsigned char irq_bit, events;
 
 	events = via1[vIFR] & via1[vIER] & 0x7F;
 	if (!events)
-		return IRQ_NONE;
+		return;
 
 	irq_num = VIA1_SOURCE_BASE;
 	irq_bit = 1;
 	do {
 		if (events & irq_bit) {
 			via1[vIFR] = irq_bit;
-			m68k_handle_int(irq_num);
+			generic_handle_irq(irq_num);
 		}
 		++irq_num;
 		irq_bit <<= 1;
 	} while (events >= irq_bit);
-	return IRQ_HANDLED;
 }
 
-irqreturn_t via2_irq(int irq, void *dev_id)
+static void via2_irq(unsigned int irq, struct irq_desc *desc)
 {
 	int irq_num;
 	unsigned char irq_bit, events;
 
 	events = via2[gIFR] & via2[gIER] & 0x7F;
 	if (!events)
-		return IRQ_NONE;
+		return;
 
 	irq_num = VIA2_SOURCE_BASE;
 	irq_bit = 1;
 	do {
 		if (events & irq_bit) {
 			via2[gIFR] = irq_bit | rbv_clear;
-			m68k_handle_int(irq_num);
+			generic_handle_irq(irq_num);
 		}
 		++irq_num;
 		irq_bit <<= 1;
 	} while (events >= irq_bit);
-	return IRQ_HANDLED;
 }
 
 /*
@@ -495,7 +462,7 @@
  * VIA2 dispatcher as a fast interrupt handler.
  */
 
-irqreturn_t via_nubus_irq(int irq, void *dev_id)
+void via_nubus_irq(unsigned int irq, struct irq_desc *desc)
 {
 	int slot_irq;
 	unsigned char slot_bit, events;
@@ -506,7 +473,7 @@
 	else
 		events &= ~via2[vDirA];
 	if (!events)
-		return IRQ_NONE;
+		return;
 
 	do {
 		slot_irq = IRQ_NUBUS_F;
@@ -514,7 +481,7 @@
 		do {
 			if (events & slot_bit) {
 				events &= ~slot_bit;
-				m68k_handle_int(slot_irq);
+				generic_handle_irq(slot_irq);
 			}
 			--slot_irq;
 			slot_bit >>= 1;
@@ -528,7 +495,24 @@
 		else
 			events &= ~via2[vDirA];
 	} while (events);
-	return IRQ_HANDLED;
+}
+
+/*
+ * Register the interrupt dispatchers for VIA or RBV machines only.
+ */
+
+void __init via_register_interrupts(void)
+{
+	if (via_alt_mapping) {
+		/* software interrupt */
+		irq_set_chained_handler(IRQ_AUTO_1, via1_irq);
+		/* via1 interrupt */
+		irq_set_chained_handler(IRQ_AUTO_6, via1_irq);
+	} else {
+		irq_set_chained_handler(IRQ_AUTO_1, via1_irq);
+	}
+	irq_set_chained_handler(IRQ_AUTO_2, via2_irq);
+	irq_set_chained_handler(IRQ_MAC_NUBUS, via_nubus_irq);
 }
 
 void via_irq_enable(int irq) {
diff --git a/arch/m68k/mvme147/config.c b/arch/m68k/mvme147/config.c
index 6cb9c3a..5de924e 100644
--- a/arch/m68k/mvme147/config.c
+++ b/arch/m68k/mvme147/config.c
@@ -81,7 +81,7 @@
 
 void __init mvme147_init_IRQ(void)
 {
-	m68k_setup_user_interrupt(VEC_USER, 192, NULL);
+	m68k_setup_user_interrupt(VEC_USER, 192);
 }
 
 void __init config_mvme147(void)
@@ -114,8 +114,7 @@
 void mvme147_sched_init (irq_handler_t timer_routine)
 {
 	tick_handler = timer_routine;
-	if (request_irq(PCC_IRQ_TIMER1, mvme147_timer_int, IRQ_FLG_REPLACE,
-			"timer 1", NULL))
+	if (request_irq(PCC_IRQ_TIMER1, mvme147_timer_int, 0, "timer 1", NULL))
 		pr_err("Couldn't register timer interrupt\n");
 
 	/* Init the clock with a value */
diff --git a/arch/m68k/mvme16x/config.c b/arch/m68k/mvme16x/config.c
index 0b28e26..31a66d9 100644
--- a/arch/m68k/mvme16x/config.c
+++ b/arch/m68k/mvme16x/config.c
@@ -117,7 +117,7 @@
 
 static void __init mvme16x_init_IRQ (void)
 {
-	m68k_setup_user_interrupt(VEC_USER, 192, NULL);
+	m68k_setup_user_interrupt(VEC_USER, 192);
 }
 
 #define pcc2chip	((volatile u_char *)0xfff42000)
diff --git a/arch/m68k/q40/q40ints.c b/arch/m68k/q40/q40ints.c
index 9f0e3d5..2b88849 100644
--- a/arch/m68k/q40/q40ints.c
+++ b/arch/m68k/q40/q40ints.c
@@ -15,10 +15,10 @@
 #include <linux/kernel.h>
 #include <linux/errno.h>
 #include <linux/interrupt.h>
+#include <linux/irq.h>
 
 #include <asm/ptrace.h>
 #include <asm/system.h>
-#include <asm/irq.h>
 #include <asm/traps.h>
 
 #include <asm/q40_master.h>
@@ -35,35 +35,36 @@
 */
 
 static void q40_irq_handler(unsigned int, struct pt_regs *fp);
-static void q40_enable_irq(unsigned int);
-static void q40_disable_irq(unsigned int);
+static void q40_irq_enable(struct irq_data *data);
+static void q40_irq_disable(struct irq_data *data);
 
 unsigned short q40_ablecount[35];
 unsigned short q40_state[35];
 
-static int q40_irq_startup(unsigned int irq)
+static unsigned int q40_irq_startup(struct irq_data *data)
 {
+	unsigned int irq = data->irq;
+
 	/* test for ISA ints not implemented by HW */
 	switch (irq) {
 	case 1: case 2: case 8: case 9:
 	case 11: case 12: case 13:
 		printk("%s: ISA IRQ %d not implemented by HW\n", __func__, irq);
-		return -ENXIO;
+		/* FIXME return -ENXIO; */
 	}
 	return 0;
 }
 
-static void q40_irq_shutdown(unsigned int irq)
+static void q40_irq_shutdown(struct irq_data *data)
 {
 }
 
-static struct irq_controller q40_irq_controller = {
+static struct irq_chip q40_irq_chip = {
 	.name		= "q40",
-	.lock		= __SPIN_LOCK_UNLOCKED(q40_irq_controller.lock),
-	.startup	= q40_irq_startup,
-	.shutdown	= q40_irq_shutdown,
-	.enable		= q40_enable_irq,
-	.disable	= q40_disable_irq,
+	.irq_startup	= q40_irq_startup,
+	.irq_shutdown	= q40_irq_shutdown,
+	.irq_enable	= q40_irq_enable,
+	.irq_disable	= q40_irq_disable,
 };
 
 /*
@@ -81,13 +82,14 @@
 
 void __init q40_init_IRQ(void)
 {
-	m68k_setup_irq_controller(&q40_irq_controller, 1, Q40_IRQ_MAX);
+	m68k_setup_irq_controller(&q40_irq_chip, handle_simple_irq, 1,
+				  Q40_IRQ_MAX);
 
 	/* setup handler for ISA ints */
 	m68k_setup_auto_interrupt(q40_irq_handler);
 
-	m68k_irq_startup(IRQ_AUTO_2);
-	m68k_irq_startup(IRQ_AUTO_4);
+	m68k_irq_startup_irq(IRQ_AUTO_2);
+	m68k_irq_startup_irq(IRQ_AUTO_4);
 
 	/* now enable some ints.. */
 	master_outb(1, EXT_ENABLE_REG);  /* ISA IRQ 5-15 */
@@ -218,11 +220,11 @@
 	switch (irq) {
 	case 4:
 	case 6:
-		__m68k_handle_int(Q40_IRQ_SAMPLE, fp);
+		do_IRQ(Q40_IRQ_SAMPLE, fp);
 		return;
 	}
 	if (mir & Q40_IRQ_FRAME_MASK) {
-		__m68k_handle_int(Q40_IRQ_FRAME, fp);
+		do_IRQ(Q40_IRQ_FRAME, fp);
 		master_outb(-1, FRAME_CLEAR_REG);
 	}
 	if ((mir & Q40_IRQ_SER_MASK) || (mir & Q40_IRQ_EXT_MASK)) {
@@ -257,7 +259,7 @@
 					goto iirq;
 				}
 				q40_state[irq] |= IRQ_INPROGRESS;
-				__m68k_handle_int(irq, fp);
+				do_IRQ(irq, fp);
 				q40_state[irq] &= ~IRQ_INPROGRESS;
 
 				/* naively enable everything, if that fails than    */
@@ -288,25 +290,29 @@
 	mir = master_inb(IIRQ_REG);
 	/* should test whether keyboard irq is really enabled, doing it in defhand */
 	if (mir & Q40_IRQ_KEYB_MASK)
-		__m68k_handle_int(Q40_IRQ_KEYBOARD, fp);
+		do_IRQ(Q40_IRQ_KEYBOARD, fp);
 
 	return;
 }
 
-void q40_enable_irq(unsigned int irq)
+void q40_irq_enable(struct irq_data *data)
 {
+	unsigned int irq = data->irq;
+
 	if (irq >= 5 && irq <= 15) {
 		mext_disabled--;
 		if (mext_disabled > 0)
-			printk("q40_enable_irq : nested disable/enable\n");
+			printk("q40_irq_enable : nested disable/enable\n");
 		if (mext_disabled == 0)
 			master_outb(1, EXT_ENABLE_REG);
 	}
 }
 
 
-void q40_disable_irq(unsigned int irq)
+void q40_irq_disable(struct irq_data *data)
 {
+	unsigned int irq = data->irq;
+
 	/* disable ISA iqs : only do something if the driver has been
 	 * verified to be Q40 "compatible" - right now IDE, NE2K
 	 * Any driver should not attempt to sleep across disable_irq !!
@@ -319,13 +325,3 @@
 			printk("disable_irq nesting count %d\n",mext_disabled);
 	}
 }
-
-unsigned long q40_probe_irq_on(void)
-{
-	printk("irq probing not working - reconfigure the driver to avoid this\n");
-	return -1;
-}
-int q40_probe_irq_off(unsigned long irqs)
-{
-	return -1;
-}
diff --git a/arch/m68k/sun3/sun3ints.c b/arch/m68k/sun3/sun3ints.c
index 6464ad3..78b60f5 100644
--- a/arch/m68k/sun3/sun3ints.c
+++ b/arch/m68k/sun3/sun3ints.c
@@ -51,25 +51,29 @@
 
 static irqreturn_t sun3_int7(int irq, void *dev_id)
 {
-	*sun3_intreg |=  (1 << irq);
-	if (!(kstat_cpu(0).irqs[irq] % 2000))
-		sun3_leds(led_pattern[(kstat_cpu(0).irqs[irq] % 16000) / 2000]);
+	unsigned int cnt;
+
+	cnt = kstat_irqs_cpu(irq, 0);
+	if (!(cnt % 2000))
+		sun3_leds(led_pattern[cnt % 16000 / 2000]);
 	return IRQ_HANDLED;
 }
 
 static irqreturn_t sun3_int5(int irq, void *dev_id)
 {
+	unsigned int cnt;
+
 #ifdef CONFIG_SUN3
 	intersil_clear();
 #endif
-        *sun3_intreg |=  (1 << irq);
 #ifdef CONFIG_SUN3
 	intersil_clear();
 #endif
 	xtime_update(1);
 	update_process_times(user_mode(get_irq_regs()));
-        if (!(kstat_cpu(0).irqs[irq] % 20))
-                sun3_leds(led_pattern[(kstat_cpu(0).irqs[irq] % 160) / 20]);
+	cnt = kstat_irqs_cpu(irq, 0);
+	if (!(cnt % 20))
+		sun3_leds(led_pattern[cnt % 160 / 20]);
 	return IRQ_HANDLED;
 }
 
@@ -79,29 +83,33 @@
 	return IRQ_HANDLED;
 }
 
-static void sun3_inthandle(unsigned int irq, struct pt_regs *fp)
+static void sun3_irq_enable(struct irq_data *data)
 {
-        *sun3_intreg &= ~(1 << irq);
+    sun3_enable_irq(data->irq);
+};
 
-	__m68k_handle_int(irq, fp);
-}
+static void sun3_irq_disable(struct irq_data *data)
+{
+    sun3_disable_irq(data->irq);
+};
 
-static struct irq_controller sun3_irq_controller = {
+static struct irq_chip sun3_irq_chip = {
 	.name		= "sun3",
-	.lock		= __SPIN_LOCK_UNLOCKED(sun3_irq_controller.lock),
-	.startup	= m68k_irq_startup,
-	.shutdown	= m68k_irq_shutdown,
-	.enable		= sun3_enable_irq,
-	.disable	= sun3_disable_irq,
+	.irq_startup	= m68k_irq_startup,
+	.irq_shutdown	= m68k_irq_shutdown,
+	.irq_enable	= sun3_irq_enable,
+	.irq_disable	= sun3_irq_disable,
+	.irq_mask	= sun3_irq_disable,
+	.irq_unmask	= sun3_irq_enable,
 };
 
 void __init sun3_init_IRQ(void)
 {
 	*sun3_intreg = 1;
 
-	m68k_setup_auto_interrupt(sun3_inthandle);
-	m68k_setup_irq_controller(&sun3_irq_controller, IRQ_AUTO_1, 7);
-	m68k_setup_user_interrupt(VEC_USER, 128, NULL);
+	m68k_setup_irq_controller(&sun3_irq_chip, handle_level_irq, IRQ_AUTO_1,
+				  7);
+	m68k_setup_user_interrupt(VEC_USER, 128);
 
 	if (request_irq(IRQ_AUTO_5, sun3_int5, 0, "int5", NULL))
 		pr_err("Couldn't register %s interrupt\n", "int5");
diff --git a/arch/microblaze/include/asm/namei.h b/arch/microblaze/include/asm/namei.h
deleted file mode 100644
index 61d60b8..0000000
--- a/arch/microblaze/include/asm/namei.h
+++ /dev/null
@@ -1,22 +0,0 @@
-/*
- * Copyright (C) 2006 Atmark Techno, Inc.
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License. See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-
-#ifndef _ASM_MICROBLAZE_NAMEI_H
-#define _ASM_MICROBLAZE_NAMEI_H
-
-#ifdef __KERNEL__
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-#define __emul_prefix() NULL
-
-#endif /* __KERNEL__ */
-
-#endif /* _ASM_MICROBLAZE_NAMEI_H */
diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index 9b4cb00..0be3186 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -286,11 +286,11 @@
 archprepare:
 ifdef CONFIG_MIPS32_N32
 	@echo '  Checking missing-syscalls for N32'
-	$(Q)$(MAKE) $(build)=. missing-syscalls ccflags-y="-mabi=n32"
+	$(Q)$(MAKE) $(build)=. missing-syscalls missing_syscalls_flags="-mabi=n32"
 endif
 ifdef CONFIG_MIPS32_O32
 	@echo '  Checking missing-syscalls for O32'
-	$(Q)$(MAKE) $(build)=. missing-syscalls ccflags-y="-mabi=32"
+	$(Q)$(MAKE) $(build)=. missing-syscalls missing_syscalls_flags="-mabi=32"
 endif
 
 install:
diff --git a/arch/mips/cavium-octeon/flash_setup.c b/arch/mips/cavium-octeon/flash_setup.c
index 975c203..0a430e0 100644
--- a/arch/mips/cavium-octeon/flash_setup.c
+++ b/arch/mips/cavium-octeon/flash_setup.c
@@ -17,8 +17,6 @@
 
 static struct map_info flash_map;
 static struct mtd_info *mymtd;
-static int nr_parts;
-static struct mtd_partition *parts;
 static const char *part_probe_types[] = {
 	"cmdlinepart",
 #ifdef CONFIG_MTD_REDBOOT_PARTS
@@ -61,11 +59,8 @@
 		mymtd = do_map_probe("cfi_probe", &flash_map);
 		if (mymtd) {
 			mymtd->owner = THIS_MODULE;
-
-			nr_parts = parse_mtd_partitions(mymtd,
-							part_probe_types,
-							&parts, 0);
-			mtd_device_register(mymtd, parts, nr_parts);
+			mtd_device_parse_register(mymtd, part_probe_types,
+						  0, NULL, 0);
 		} else {
 			pr_err("Failed to register MTD device for flash\n");
 		}
diff --git a/arch/mips/cavium-octeon/smp.c b/arch/mips/cavium-octeon/smp.c
index 8b60642..efcfff4 100644
--- a/arch/mips/cavium-octeon/smp.c
+++ b/arch/mips/cavium-octeon/smp.c
@@ -207,8 +207,9 @@
 	 * the other bits alone.
 	 */
 	cvmx_write_csr(CVMX_CIU_MBOX_CLRX(cvmx_get_core_num()), 0xffff);
-	if (request_irq(OCTEON_IRQ_MBOX0, mailbox_interrupt, IRQF_DISABLED,
-			"SMP-IPI", mailbox_interrupt)) {
+	if (request_irq(OCTEON_IRQ_MBOX0, mailbox_interrupt,
+			IRQF_PERCPU | IRQF_NO_THREAD, "SMP-IPI",
+			mailbox_interrupt)) {
 		panic("Cannot request_irq(OCTEON_IRQ_MBOX0)\n");
 	}
 }
diff --git a/arch/mips/emma/common/prom.c b/arch/mips/emma/common/prom.c
index 708f087..cae4225 100644
--- a/arch/mips/emma/common/prom.c
+++ b/arch/mips/emma/common/prom.c
@@ -50,7 +50,7 @@
 
 	/* arg[0] is "g", the rest is boot parameters */
 	for (i = 1; i < argc; i++) {
-		if (strlen(arcs_cmdline) + strlen(arg[i] + 1)
+		if (strlen(arcs_cmdline) + strlen(arg[i]) + 1
 		    >= sizeof(arcs_cmdline))
 			break;
 		strcat(arcs_cmdline, arg[i]);
diff --git a/arch/mips/include/asm/mach-bcm47xx/gpio.h b/arch/mips/include/asm/mach-bcm47xx/gpio.h
index 76961ca..2ef17e8 100644
--- a/arch/mips/include/asm/mach-bcm47xx/gpio.h
+++ b/arch/mips/include/asm/mach-bcm47xx/gpio.h
@@ -36,6 +36,8 @@
 	return -EINVAL;
 }
 
+#define gpio_get_value_cansleep	gpio_get_value
+
 static inline void gpio_set_value(unsigned gpio, int value)
 {
 	switch (bcm47xx_bus_type) {
@@ -54,6 +56,19 @@
 	}
 }
 
+#define gpio_set_value_cansleep gpio_set_value
+
+static inline int gpio_cansleep(unsigned gpio)
+{
+	return 0;
+}
+
+static inline int gpio_is_valid(unsigned gpio)
+{
+	return gpio < (BCM47XX_EXTIF_GPIO_LINES + BCM47XX_CHIPCO_GPIO_LINES);
+}
+
+
 static inline int gpio_direction_input(unsigned gpio)
 {
 	switch (bcm47xx_bus_type) {
@@ -137,7 +152,4 @@
 }
 
 
-/* cansleep wrappers */
-#include <asm-generic/gpio.h>
-
 #endif /* __BCM47XX_GPIO_H */
diff --git a/arch/mips/include/asm/unistd.h b/arch/mips/include/asm/unistd.h
index ecea787..d8dad53 100644
--- a/arch/mips/include/asm/unistd.h
+++ b/arch/mips/include/asm/unistd.h
@@ -365,16 +365,18 @@
 #define __NR_syncfs			(__NR_Linux + 342)
 #define __NR_sendmmsg			(__NR_Linux + 343)
 #define __NR_setns			(__NR_Linux + 344)
+#define __NR_process_vm_readv		(__NR_Linux + 345)
+#define __NR_process_vm_writev		(__NR_Linux + 346)
 
 /*
  * Offset of the last Linux o32 flavoured syscall
  */
-#define __NR_Linux_syscalls		344
+#define __NR_Linux_syscalls		346
 
 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */
 
 #define __NR_O32_Linux			4000
-#define __NR_O32_Linux_syscalls		344
+#define __NR_O32_Linux_syscalls		346
 
 #if _MIPS_SIM == _MIPS_SIM_ABI64
 
@@ -686,16 +688,18 @@
 #define __NR_syncfs			(__NR_Linux + 301)
 #define __NR_sendmmsg			(__NR_Linux + 302)
 #define __NR_setns			(__NR_Linux + 303)
+#define __NR_process_vm_readv		(__NR_Linux + 304)
+#define __NR_process_vm_writev		(__NR_Linux + 305)
 
 /*
  * Offset of the last Linux 64-bit flavoured syscall
  */
-#define __NR_Linux_syscalls		303
+#define __NR_Linux_syscalls		305
 
 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */
 
 #define __NR_64_Linux			5000
-#define __NR_64_Linux_syscalls		303
+#define __NR_64_Linux_syscalls		305
 
 #if _MIPS_SIM == _MIPS_SIM_NABI32
 
@@ -1012,16 +1016,18 @@
 #define __NR_syncfs			(__NR_Linux + 306)
 #define __NR_sendmmsg			(__NR_Linux + 307)
 #define __NR_setns			(__NR_Linux + 308)
+#define __NR_process_vm_readv		(__NR_Linux + 309)
+#define __NR_process_vm_writev		(__NR_Linux + 310)
 
 /*
  * Offset of the last N32 flavoured syscall
  */
-#define __NR_Linux_syscalls		308
+#define __NR_Linux_syscalls		310
 
 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */
 
 #define __NR_N32_Linux			6000
-#define __NR_N32_Linux_syscalls		308
+#define __NR_N32_Linux_syscalls		310
 
 #ifdef __KERNEL__
 
diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
index 98c5a97..e2d8e19 100644
--- a/arch/mips/kernel/cevt-r4k.c
+++ b/arch/mips/kernel/cevt-r4k.c
@@ -103,19 +103,10 @@
 
 /*
  * Compare interrupt can be routed and latched outside the core,
- * so a single execution hazard barrier may not be enough to give
- * it time to clear as seen in the Cause register.  4 time the
- * pipeline depth seems reasonably conservative, and empirically
- * works better in configurations with high CPU/bus clock ratios.
+ * so wait up to worst case number of cycle counter ticks for timer interrupt
+ * changes to propagate to the cause register.
  */
-
-#define compare_change_hazard() \
-	do { \
-		irq_disable_hazard(); \
-		irq_disable_hazard(); \
-		irq_disable_hazard(); \
-		irq_disable_hazard(); \
-	} while (0)
+#define COMPARE_INT_SEEN_TICKS 50
 
 int c0_compare_int_usable(void)
 {
@@ -126,8 +117,12 @@
 	 * IP7 already pending?  Try to clear it by acking the timer.
 	 */
 	if (c0_compare_int_pending()) {
-		write_c0_compare(read_c0_count());
-		compare_change_hazard();
+		cnt = read_c0_count();
+		write_c0_compare(cnt);
+		back_to_back_c0_hazard();
+		while (read_c0_count() < (cnt  + COMPARE_INT_SEEN_TICKS))
+			if (!c0_compare_int_pending())
+				break;
 		if (c0_compare_int_pending())
 			return 0;
 	}
@@ -136,7 +131,7 @@
 		cnt = read_c0_count();
 		cnt += delta;
 		write_c0_compare(cnt);
-		compare_change_hazard();
+		back_to_back_c0_hazard();
 		if ((int)(read_c0_count() - cnt) < 0)
 		    break;
 		/* increase delta if the timer was already expired */
@@ -145,12 +140,17 @@
 	while ((int)(read_c0_count() - cnt) <= 0)
 		;	/* Wait for expiry  */
 
-	compare_change_hazard();
+	while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS))
+		if (c0_compare_int_pending())
+			break;
 	if (!c0_compare_int_pending())
 		return 0;
-
-	write_c0_compare(read_c0_count());
-	compare_change_hazard();
+	cnt = read_c0_count();
+	write_c0_compare(cnt);
+	back_to_back_c0_hazard();
+	while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS))
+		if (!c0_compare_int_pending())
+			break;
 	if (c0_compare_int_pending())
 		return 0;
 
diff --git a/arch/mips/kernel/cpufreq/loongson2_clock.c b/arch/mips/kernel/cpufreq/loongson2_clock.c
index cefc6e2..5426779 100644
--- a/arch/mips/kernel/cpufreq/loongson2_clock.c
+++ b/arch/mips/kernel/cpufreq/loongson2_clock.c
@@ -7,6 +7,7 @@
  * for more details.
  */
 
+#include <linux/module.h>
 #include <linux/cpufreq.h>
 #include <linux/platform_device.h>
 
diff --git a/arch/mips/kernel/perf_event_mipsxx.c b/arch/mips/kernel/perf_event_mipsxx.c
index 4f2971b..315fc0b 100644
--- a/arch/mips/kernel/perf_event_mipsxx.c
+++ b/arch/mips/kernel/perf_event_mipsxx.c
@@ -623,7 +623,7 @@
 	if (!atomic_inc_not_zero(&active_events)) {
 		if (atomic_read(&active_events) > MIPS_MAX_HWEVENTS) {
 			atomic_dec(&active_events);
-			return -ENOSPC;
+			return -EINVAL;
 		}
 
 		mutex_lock(&pmu_reserve_mutex);
@@ -732,15 +732,15 @@
 	memset(&fake_cpuc, 0, sizeof(fake_cpuc));
 
 	if (!validate_event(&fake_cpuc, leader))
-		return -ENOSPC;
+		return -EINVAL;
 
 	list_for_each_entry(sibling, &leader->sibling_list, group_entry) {
 		if (!validate_event(&fake_cpuc, sibling))
-			return -ENOSPC;
+			return -EINVAL;
 	}
 
 	if (!validate_event(&fake_cpuc, event))
-		return -ENOSPC;
+		return -EINVAL;
 
 	return 0;
 }
diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index 4792065..a632bc1 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -591,6 +591,8 @@
 	sys	sys_syncfs		1
 	sys	sys_sendmmsg		4
 	sys	sys_setns		2
+	sys	sys_process_vm_readv	6	/* 4345 */
+	sys	sys_process_vm_writev	6
 	.endm
 
 	/* We pre-compute the number of _instruction_ bytes needed to
diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S
index fb7334b..3b5a5e9 100644
--- a/arch/mips/kernel/scall64-64.S
+++ b/arch/mips/kernel/scall64-64.S
@@ -430,4 +430,6 @@
 	PTR	sys_syncfs
 	PTR	sys_sendmmsg
 	PTR	sys_setns
+	PTR	sys_process_vm_readv
+	PTR	sys_process_vm_writev		/* 5305 */
 	.size	sys_call_table,.-sys_call_table
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 6de1f59..6be6f70 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -430,4 +430,6 @@
 	PTR	sys_syncfs
 	PTR	compat_sys_sendmmsg
 	PTR	sys_setns
+	PTR	compat_sys_process_vm_readv
+	PTR	compat_sys_process_vm_writev	/* 6310 */
 	.size	sysn32_call_table,.-sysn32_call_table
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index 1d81316..5422855 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -548,4 +548,6 @@
 	PTR	sys_syncfs
 	PTR	compat_sys_sendmmsg
 	PTR	sys_setns
+	PTR	compat_sys_process_vm_readv	/* 4345 */
+	PTR	compat_sys_process_vm_writev
 	.size	sys_call_table,.-sys_call_table
diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 261ccbc..5c8a49d 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -1596,7 +1596,8 @@
 	}
 #endif /* CONFIG_MIPS_MT_SMTC */
 
-	cpu_data[cpu].asid_cache = ASID_FIRST_VERSION;
+	if (!cpu_data[cpu].asid_cache)
+		cpu_data[cpu].asid_cache = ASID_FIRST_VERSION;
 
 	atomic_inc(&init_mm.mm_count);
 	current->active_mm = &init_mm;
diff --git a/arch/mips/lantiq/clk.c b/arch/mips/lantiq/clk.c
index 7e9c0ff..77ed70f 100644
--- a/arch/mips/lantiq/clk.c
+++ b/arch/mips/lantiq/clk.c
@@ -7,7 +7,7 @@
  * Copyright (C) 2010 John Crispin <blogic@openwrt.org>
  */
 #include <linux/io.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
diff --git a/arch/mips/lantiq/devices.c b/arch/mips/lantiq/devices.c
index 44a3677..de1cb2b 100644
--- a/arch/mips/lantiq/devices.c
+++ b/arch/mips/lantiq/devices.c
@@ -7,7 +7,7 @@
  */
 
 #include <linux/init.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/types.h>
 #include <linux/string.h>
 #include <linux/kernel.h>
diff --git a/arch/mips/lantiq/prom.c b/arch/mips/lantiq/prom.c
index 56ba007..e34fcfd 100644
--- a/arch/mips/lantiq/prom.c
+++ b/arch/mips/lantiq/prom.c
@@ -6,7 +6,7 @@
  * Copyright (C) 2010 John Crispin <blogic@openwrt.org>
  */
 
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/clk.h>
 #include <asm/bootinfo.h>
 #include <asm/time.h>
diff --git a/arch/mips/lantiq/setup.c b/arch/mips/lantiq/setup.c
index 9b8af77..1ff6c9d 100644
--- a/arch/mips/lantiq/setup.c
+++ b/arch/mips/lantiq/setup.c
@@ -7,7 +7,7 @@
  */
 
 #include <linux/kernel.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/io.h>
 #include <linux/ioport.h>
 #include <asm/bootinfo.h>
diff --git a/arch/mips/lantiq/xway/clk-ase.c b/arch/mips/lantiq/xway/clk-ase.c
index 22d823a..6522583 100644
--- a/arch/mips/lantiq/xway/clk-ase.c
+++ b/arch/mips/lantiq/xway/clk-ase.c
@@ -7,7 +7,7 @@
  */
 
 #include <linux/io.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/init.h>
 #include <linux/clk.h>
 
diff --git a/arch/mips/lantiq/xway/clk-xway.c b/arch/mips/lantiq/xway/clk-xway.c
index ddd3959..696b1a3 100644
--- a/arch/mips/lantiq/xway/clk-xway.c
+++ b/arch/mips/lantiq/xway/clk-xway.c
@@ -7,7 +7,7 @@
  */
 
 #include <linux/io.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/init.h>
 #include <linux/clk.h>
 
diff --git a/arch/mips/lantiq/xway/devices.c b/arch/mips/lantiq/xway/devices.c
index d0e32ab..d614aa7 100644
--- a/arch/mips/lantiq/xway/devices.c
+++ b/arch/mips/lantiq/xway/devices.c
@@ -7,7 +7,7 @@
  */
 
 #include <linux/init.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/types.h>
 #include <linux/string.h>
 #include <linux/mtd/physmap.h>
diff --git a/arch/mips/lantiq/xway/dma.c b/arch/mips/lantiq/xway/dma.c
index 4278a45..cbb6ae5 100644
--- a/arch/mips/lantiq/xway/dma.c
+++ b/arch/mips/lantiq/xway/dma.c
@@ -19,6 +19,7 @@
 #include <linux/platform_device.h>
 #include <linux/io.h>
 #include <linux/dma-mapping.h>
+#include <linux/export.h>
 
 #include <lantiq_soc.h>
 #include <xway_dma.h>
diff --git a/arch/mips/lantiq/xway/gpio.c b/arch/mips/lantiq/xway/gpio.c
index a321451..d2fa98f 100644
--- a/arch/mips/lantiq/xway/gpio.c
+++ b/arch/mips/lantiq/xway/gpio.c
@@ -7,7 +7,7 @@
  */
 
 #include <linux/slab.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 #include <linux/gpio.h>
 #include <linux/ioport.h>
diff --git a/arch/mips/lantiq/xway/gpio_ebu.c b/arch/mips/lantiq/xway/gpio_ebu.c
index a479355..b91c7f1 100644
--- a/arch/mips/lantiq/xway/gpio_ebu.c
+++ b/arch/mips/lantiq/xway/gpio_ebu.c
@@ -7,7 +7,7 @@
  */
 
 #include <linux/init.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/types.h>
 #include <linux/platform_device.h>
 #include <linux/mutex.h>
diff --git a/arch/mips/lantiq/xway/gpio_stp.c b/arch/mips/lantiq/xway/gpio_stp.c
index 67d59d6..ff9991c 100644
--- a/arch/mips/lantiq/xway/gpio_stp.c
+++ b/arch/mips/lantiq/xway/gpio_stp.c
@@ -9,7 +9,7 @@
 
 #include <linux/slab.h>
 #include <linux/init.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/types.h>
 #include <linux/platform_device.h>
 #include <linux/mutex.h>
diff --git a/arch/mips/lantiq/xway/prom-ase.c b/arch/mips/lantiq/xway/prom-ase.c
index abe49f4..ae4959a 100644
--- a/arch/mips/lantiq/xway/prom-ase.c
+++ b/arch/mips/lantiq/xway/prom-ase.c
@@ -6,7 +6,7 @@
  *  Copyright (C) 2010 John Crispin <blogic@openwrt.org>
  */
 
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/clk.h>
 #include <asm/bootinfo.h>
 #include <asm/time.h>
diff --git a/arch/mips/lantiq/xway/prom-xway.c b/arch/mips/lantiq/xway/prom-xway.c
index 1686692a..2228133 100644
--- a/arch/mips/lantiq/xway/prom-xway.c
+++ b/arch/mips/lantiq/xway/prom-xway.c
@@ -6,7 +6,7 @@
  *  Copyright (C) 2010 John Crispin <blogic@openwrt.org>
  */
 
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/clk.h>
 #include <asm/bootinfo.h>
 #include <asm/time.h>
diff --git a/arch/mips/lantiq/xway/reset.c b/arch/mips/lantiq/xway/reset.c
index a1be36d..3d41f0b 100644
--- a/arch/mips/lantiq/xway/reset.c
+++ b/arch/mips/lantiq/xway/reset.c
@@ -10,7 +10,7 @@
 #include <linux/io.h>
 #include <linux/ioport.h>
 #include <linux/pm.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <asm/reboot.h>
 
 #include <lantiq_soc.h>
diff --git a/arch/mips/nxp/pnx8550/common/pci.c b/arch/mips/nxp/pnx8550/common/pci.c
deleted file mode 100644
index 98e86dd..0000000
--- a/arch/mips/nxp/pnx8550/common/pci.c
+++ /dev/null
@@ -1,134 +0,0 @@
-/*
- *
- * BRIEF MODULE DESCRIPTION
- *
- * Author: source@mvista.com
- *
- *  This program is free software; you can distribute it and/or modify it
- *  under the terms of the GNU General Public License (Version 2) as
- *  published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
- *  for more details.
- *
- *  You should have received a copy of the GNU General Public License along
- *  with this program; if not, write to the Free Software Foundation, Inc.,
- *  59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
- */
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <linux/kernel.h>
-#include <linux/init.h>
-
-#include <pci.h>
-#include <glb.h>
-#include <nand.h>
-
-static struct resource pci_io_resource = {
-	.start	= PNX8550_PCIIO + 0x1000,	/* reserve regacy I/O space */
-	.end	= PNX8550_PCIIO + PNX8550_PCIIO_SIZE,
-	.name	= "pci IO space",
-	.flags	= IORESOURCE_IO
-};
-
-static struct resource pci_mem_resource = {
-	.start	= PNX8550_PCIMEM,
-	.end	= PNX8550_PCIMEM + PNX8550_PCIMEM_SIZE - 1,
-	.name	= "pci memory space",
-	.flags	= IORESOURCE_MEM
-};
-
-extern struct pci_ops pnx8550_pci_ops;
-
-static struct pci_controller pnx8550_controller = {
-	.pci_ops	= &pnx8550_pci_ops,
-	.io_map_base	= PNX8550_PORT_BASE,
-	.io_resource	= &pci_io_resource,
-	.mem_resource	= &pci_mem_resource,
-};
-
-/* Return the total size of DRAM-memory, (RANK0 + RANK1) */
-static inline unsigned long get_system_mem_size(void)
-{
-	/* Read IP2031_RANK0_ADDR_LO */
-	unsigned long dram_r0_lo = inl(PCI_BASE | 0x65010);
-	/* Read IP2031_RANK1_ADDR_HI */
-	unsigned long dram_r1_hi = inl(PCI_BASE | 0x65018);
-
-	return dram_r1_hi - dram_r0_lo + 1;
-}
-
-static int __init pnx8550_pci_setup(void)
-{
-	int pci_mem_code;
-	int mem_size = get_system_mem_size() >> 20;
-
-	/* Clear the Global 2 Register, PCI Inta Output Enable Registers
-	   Bit 1:Enable DAC Powerdown
-	  -> 0:DACs are enabled and are working normally
-	     1:DACs are powerdown
-	   Bit 0:Enable of PCI inta output
-	  -> 0 = Disable PCI inta output
-	     1 = Enable PCI inta output
-	*/
-	PNX8550_GLB2_ENAB_INTA_O = 0;
-
-	/* Calc the PCI mem size code */
-	if (mem_size >= 128)
-		pci_mem_code = SIZE_128M;
-	else if (mem_size >= 64)
-		pci_mem_code = SIZE_64M;
-	else if (mem_size >= 32)
-		pci_mem_code = SIZE_32M;
-	else
-		pci_mem_code = SIZE_16M;
-
-	/* Set PCI_XIO registers */
-	outl(pci_mem_resource.start, PCI_BASE | PCI_BASE1_LO);
-	outl(pci_mem_resource.end + 1, PCI_BASE | PCI_BASE1_HI);
-	outl(pci_io_resource.start, PCI_BASE | PCI_BASE2_LO);
-	outl(pci_io_resource.end, PCI_BASE | PCI_BASE2_HI);
-
-	/* Send memory transaction via PCI_BASE2 */
-	outl(0x00000001, PCI_BASE | PCI_IO);
-
-	/* Unlock the setup register */
-	outl(0xca, PCI_BASE | PCI_UNLOCKREG);
-
-	/*
-	 * BAR0 of PNX8550 (pci base 10) must be zero in order for ide
-	 * to work, and in order for bus_to_baddr to work without any
-	 * hacks.
-	 */
-	outl(0x00000000, PCI_BASE | PCI_BASE10);
-
-	/*
-	 *These two bars are set by default or the boot code.
-	 * However, it's safer to set them here so we're not boot
-	 * code dependent.
-	 */
-	outl(0x1be00000, PCI_BASE | PCI_BASE14);  /* PNX MMIO */
-	outl(PNX8550_NAND_BASE_ADDR, PCI_BASE | PCI_BASE18);  /* XIO      */
-
-	outl(PCI_EN_TA |
-	     PCI_EN_PCI2MMI |
-	     PCI_EN_XIO |
-	     PCI_SETUP_BASE18_SIZE(SIZE_32M) |
-	     PCI_SETUP_BASE18_EN |
-	     PCI_SETUP_BASE14_EN |
-	     PCI_SETUP_BASE10_PREF |
-	     PCI_SETUP_BASE10_SIZE(pci_mem_code) |
-	     PCI_SETUP_CFGMANAGE_EN |
-	     PCI_SETUP_PCIARB_EN,
-	     PCI_BASE |
-	     PCI_SETUP);	/* PCI_SETUP */
-	outl(0x00000000, PCI_BASE | PCI_CTRL);	/* PCI_CONTROL */
-
-	register_pci_controller(&pnx8550_controller);
-
-	return 0;
-}
-
-arch_initcall(pnx8550_pci_setup);
diff --git a/arch/mips/nxp/pnx8550/common/setup.c b/arch/mips/nxp/pnx8550/common/setup.c
deleted file mode 100644
index 71adac3..0000000
--- a/arch/mips/nxp/pnx8550/common/setup.c
+++ /dev/null
@@ -1,143 +0,0 @@
-/*
- *
- * 2.6 port, Embedded Alley Solutions, Inc
- *
- *  Based on Per Hallsmark, per.hallsmark@mvista.com
- *
- *  This program is free software; you can distribute it and/or modify it
- *  under the terms of the GNU General Public License (Version 2) as
- *  published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
- *  for more details.
- *
- *  You should have received a copy of the GNU General Public License along
- *  with this program; if not, write to the Free Software Foundation, Inc.,
- *  59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
- */
-#include <linux/init.h>
-#include <linux/sched.h>
-#include <linux/ioport.h>
-#include <linux/irq.h>
-#include <linux/mm.h>
-#include <linux/delay.h>
-#include <linux/interrupt.h>
-#include <linux/serial_pnx8xxx.h>
-#include <linux/pm.h>
-
-#include <asm/cpu.h>
-#include <asm/bootinfo.h>
-#include <asm/irq.h>
-#include <asm/mipsregs.h>
-#include <asm/reboot.h>
-#include <asm/pgtable.h>
-#include <asm/time.h>
-
-#include <glb.h>
-#include <int.h>
-#include <pci.h>
-#include <uart.h>
-#include <nand.h>
-
-extern void __init board_setup(void);
-extern void pnx8550_machine_restart(char *);
-extern void pnx8550_machine_halt(void);
-extern void pnx8550_machine_power_off(void);
-extern struct resource ioport_resource;
-extern struct resource iomem_resource;
-extern char *prom_getcmdline(void);
-
-struct resource standard_io_resources[] = {
-	{
-		.start	= 0x00,
-		.end	= 0x1f,
-		.name	= "dma1",
-		.flags	= IORESOURCE_BUSY
-	}, {
-		.start	= 0x40,
-		.end	= 0x5f,
-		.name	= "timer",
-		.flags	= IORESOURCE_BUSY
-	}, {
-		.start	= 0x80,
-		.end	= 0x8f,
-		.name	= "dma page reg",
-		.flags	= IORESOURCE_BUSY
-	}, {
-		.start	= 0xc0,
-		.end	= 0xdf,
-		.name	= "dma2",
-		.flags	= IORESOURCE_BUSY
-	},
-};
-
-#define STANDARD_IO_RESOURCES ARRAY_SIZE(standard_io_resources)
-
-extern struct resource pci_io_resource;
-extern struct resource pci_mem_resource;
-
-/* Return the total size of DRAM-memory, (RANK0 + RANK1) */
-unsigned long get_system_mem_size(void)
-{
-	/* Read IP2031_RANK0_ADDR_LO */
-	unsigned long dram_r0_lo = inl(PCI_BASE | 0x65010);
-	/* Read IP2031_RANK1_ADDR_HI */
-	unsigned long dram_r1_hi = inl(PCI_BASE | 0x65018);
-
-	return dram_r1_hi - dram_r0_lo + 1;
-}
-
-int pnx8550_console_port = -1;
-
-void __init plat_mem_setup(void)
-{
-	int i;
-	char* argptr;
-
-	board_setup();  /* board specific setup */
-
-        _machine_restart = pnx8550_machine_restart;
-        _machine_halt = pnx8550_machine_halt;
-        pm_power_off = pnx8550_machine_power_off;
-
-	/* Clear the Global 2 Register, PCI Inta Output Enable Registers
-	   Bit 1:Enable DAC Powerdown
-	  -> 0:DACs are enabled and are working normally
-	     1:DACs are powerdown
-	   Bit 0:Enable of PCI inta output
-	  -> 0 = Disable PCI inta output
-	     1 = Enable PCI inta output
-	*/
-	PNX8550_GLB2_ENAB_INTA_O = 0;
-
-	/* IO/MEM resources. */
-	set_io_port_base(PNX8550_PORT_BASE);
-	ioport_resource.start = 0;
-	ioport_resource.end = ~0;
-	iomem_resource.start = 0;
-	iomem_resource.end = ~0;
-
-	/* Request I/O space for devices on this board */
-	for (i = 0; i < STANDARD_IO_RESOURCES; i++)
-		request_resource(&ioport_resource, standard_io_resources + i);
-
-	/* Place the Mode Control bit for GPIO pin 16 in primary function */
-	/* Pin 16 is used by UART1, UA1_TX                                */
-	outl((PNX8550_GPIO_MODE_PRIMOP << PNX8550_GPIO_MC_16_BIT) |
-			(PNX8550_GPIO_MODE_PRIMOP << PNX8550_GPIO_MC_17_BIT),
-			PNX8550_GPIO_MC1);
-
-	argptr = prom_getcmdline();
-	if ((argptr = strstr(argptr, "console=ttyS")) != NULL) {
-		argptr += strlen("console=ttyS");
-		pnx8550_console_port = *argptr == '0' ? 0 : 1;
-
-		/* We must initialize the UART (console) before early printk */
-		/* Set LCR to 8-bit and BAUD to 38400 (no 5)                */
-		ip3106_lcr(UART_BASE, pnx8550_console_port) =
-			PNX8XXX_UART_LCR_8BIT;
-		ip3106_baud(UART_BASE, pnx8550_console_port) = 5;
-	}
-}
diff --git a/arch/mips/pci/pci-alchemy.c b/arch/mips/pci/pci-alchemy.c
index 4ee5710..b5ce041 100644
--- a/arch/mips/pci/pci-alchemy.c
+++ b/arch/mips/pci/pci-alchemy.c
@@ -7,6 +7,7 @@
  * Support for all devices (greater than 16) added by David Gathright.
  */
 
+#include <linux/export.h>
 #include <linux/types.h>
 #include <linux/pci.h>
 #include <linux/platform_device.h>
diff --git a/arch/mips/pci/pci-lantiq.c b/arch/mips/pci/pci-lantiq.c
index 8656388..be1e1af 100644
--- a/arch/mips/pci/pci-lantiq.c
+++ b/arch/mips/pci/pci-lantiq.c
@@ -13,6 +13,7 @@
 #include <linux/delay.h>
 #include <linux/mm.h>
 #include <linux/vmalloc.h>
+#include <linux/export.h>
 #include <linux/platform_device.h>
 
 #include <asm/pci.h>
diff --git a/arch/mips/pmc-sierra/yosemite/prom.c b/arch/mips/pmc-sierra/yosemite/prom.c
index cf4c8687..dcc926e 100644
--- a/arch/mips/pmc-sierra/yosemite/prom.c
+++ b/arch/mips/pmc-sierra/yosemite/prom.c
@@ -102,7 +102,7 @@
 
 	/* Get the boot parameters */
 	for (i = 1; i < argc; i++) {
-		if (strlen(arcs_cmdline) + strlen(arg[i] + 1) >=
+		if (strlen(arcs_cmdline) + strlen(arg[i]) + 1 >=
 		    sizeof(arcs_cmdline))
 			break;
 
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index b177caa..951e18f 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -345,7 +345,7 @@
 
 config KEXEC
 	bool "kexec system call (EXPERIMENTAL)"
-	depends on (PPC_BOOK3S || FSL_BOOKE || (44x && !SMP && !47x)) && EXPERIMENTAL
+	depends on (PPC_BOOK3S || FSL_BOOKE || (44x && !SMP && !PPC_47x)) && EXPERIMENTAL
 	help
 	  kexec is a system call that implements the ability to shutdown your
 	  current kernel, and to start another kernel.  It is like a reboot
diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile
index 57af16e..70ba0c0 100644
--- a/arch/powerpc/Makefile
+++ b/arch/powerpc/Makefile
@@ -255,12 +255,6 @@
 		echo 'disable kernel modules' ; \
 		false ; \
 	fi
-	@if ! /bin/echo dssall | $(AS) -many -o $(TOUT) >/dev/null 2>&1 ; then \
-		echo -n '*** ${VERSION}.${PATCHLEVEL} kernels no longer build ' ; \
-		echo 'correctly with old versions of binutils.' ; \
-		echo '*** Please upgrade your binutils to 2.12.1 or newer' ; \
-		false ; \
-	fi
 
 CLEAN_FILES += $(TOUT)
 
diff --git a/arch/powerpc/boot/dts/charon.dts b/arch/powerpc/boot/dts/charon.dts
new file mode 100644
index 0000000..0e00e50
--- /dev/null
+++ b/arch/powerpc/boot/dts/charon.dts
@@ -0,0 +1,236 @@
+/*
+ * charon board Device Tree Source
+ *
+ * Copyright (C) 2007 Semihalf
+ * Marian Balakowicz <m8@semihalf.com>
+ *
+ * Copyright (C) 2010 DENX Software Engineering GmbH
+ * Heiko Schocher <hs@denx.de>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+/dts-v1/;
+
+/ {
+	model = "anon,charon";
+	compatible = "anon,charon";
+	#address-cells = <1>;
+	#size-cells = <1>;
+	interrupt-parent = <&mpc5200_pic>;
+
+	cpus {
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		PowerPC,5200@0 {
+			device_type = "cpu";
+			reg = <0>;
+			d-cache-line-size = <32>;
+			i-cache-line-size = <32>;
+			d-cache-size = <0x4000>;	// L1, 16K
+			i-cache-size = <0x4000>;	// L1, 16K
+			timebase-frequency = <0>;	// from bootloader
+			bus-frequency = <0>;		// from bootloader
+			clock-frequency = <0>;		// from bootloader
+		};
+	};
+
+	memory {
+		device_type = "memory";
+		reg = <0x00000000 0x08000000>;	// 128MB
+	};
+
+	soc5200@f0000000 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		compatible = "fsl,mpc5200-immr";
+		ranges = <0 0xf0000000 0x0000c000>;
+		reg = <0xf0000000 0x00000100>;
+		bus-frequency = <0>;		// from bootloader
+		system-frequency = <0>;		// from bootloader
+
+		cdm@200 {
+			compatible = "fsl,mpc5200-cdm";
+			reg = <0x200 0x38>;
+		};
+
+		mpc5200_pic: interrupt-controller@500 {
+			// 5200 interrupts are encoded into two levels;
+			interrupt-controller;
+			#interrupt-cells = <3>;
+			compatible = "fsl,mpc5200-pic";
+			reg = <0x500 0x80>;
+		};
+
+		timer@600 {	// General Purpose Timer
+			compatible = "fsl,mpc5200-gpt";
+			reg = <0x600 0x10>;
+			interrupts = <1 9 0>;
+			fsl,has-wdt;
+		};
+
+		can@900 {
+			compatible = "fsl,mpc5200-mscan";
+			interrupts = <2 17 0>;
+			reg = <0x900 0x80>;
+		};
+
+		can@980 {
+			compatible = "fsl,mpc5200-mscan";
+			interrupts = <2 18 0>;
+			reg = <0x980 0x80>;
+		};
+
+		gpio_simple: gpio@b00 {
+			compatible = "fsl,mpc5200-gpio";
+			reg = <0xb00 0x40>;
+			interrupts = <1 7 0>;
+			gpio-controller;
+			#gpio-cells = <2>;
+		};
+
+		usb@1000 {
+			compatible = "fsl,mpc5200-ohci","ohci-be";
+			reg = <0x1000 0xff>;
+			interrupts = <2 6 0>;
+		};
+
+		dma-controller@1200 {
+			device_type = "dma-controller";
+			compatible = "fsl,mpc5200-bestcomm";
+			reg = <0x1200 0x80>;
+			interrupts = <3 0 0  3 1 0  3 2 0  3 3 0
+			              3 4 0  3 5 0  3 6 0  3 7 0
+			              3 8 0  3 9 0  3 10 0  3 11 0
+			              3 12 0  3 13 0  3 14 0  3 15 0>;
+		};
+
+		xlb@1f00 {
+			compatible = "fsl,mpc5200-xlb";
+			reg = <0x1f00 0x100>;
+		};
+
+		serial@2000 {		// PSC1
+			compatible = "fsl,mpc5200-psc-uart";
+			reg = <0x2000 0x100>;
+			interrupts = <2 1 0>;
+		};
+
+		serial@2400 {		// PSC3
+			compatible = "fsl,mpc5200-psc-uart";
+			reg = <0x2400 0x100>;
+			interrupts = <2 3 0>;
+		};
+
+		ethernet@3000 {
+			compatible = "fsl,mpc5200-fec";
+			reg = <0x3000 0x400>;
+			local-mac-address = [ 00 00 00 00 00 00 ];
+			interrupts = <2 5 0>;
+			fixed-link = <1 1 100 0 0>;
+		};
+
+		mdio@3000 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			compatible = "fsl,mpc5200-mdio";
+			reg = <0x3000 0x400>;       // fec range, since we need to setup fec interrupts
+			interrupts = <2 5 0>;   // these are for "mii command finished", not link changes & co.
+		};
+
+		ata@3a00 {
+			compatible = "fsl,mpc5200-ata";
+			reg = <0x3a00 0x100>;
+			interrupts = <2 7 0>;
+		};
+
+		i2c@3d00 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			compatible = "fsl,mpc5200-i2c","fsl-i2c";
+			reg = <0x3d00 0x40>;
+			interrupts = <2 15 0>;
+		};
+
+
+		i2c@3d40 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			compatible = "fsl,mpc5200-i2c","fsl-i2c";
+			reg = <0x3d40 0x40>;
+			interrupts = <2 16 0>;
+
+			dtt@28 {
+				compatible = "national,lm80";
+				reg = <0x28>;
+			};
+
+			rtc@68 {
+				compatible = "dallas,ds1374";
+				reg = <0x68>;
+			};
+		};
+
+		sram@8000 {
+			compatible = "fsl,mpc5200-sram";
+			reg = <0x8000 0x4000>;
+		};
+	};
+
+	localbus {
+		compatible = "fsl,mpc5200-lpb","simple-bus";
+		#address-cells = <2>;
+		#size-cells = <1>;
+		ranges = <	0 0 0xfc000000 0x02000000
+				1 0 0xe0000000 0x04000000 // CS1 range, SM501
+				3 0 0xe8000000 0x00080000>;
+
+		flash@0,0 {
+			compatible = "cfi-flash";
+			reg = <0 0 0x02000000>;
+			bank-width = <4>;
+			device-width = <2>;
+			#size-cells = <1>;
+			#address-cells = <1>;
+		};
+
+		display@1,0 {
+			compatible = "smi,sm501";
+			reg = <1 0x00000000 0x00800000
+			       1 0x03e00000 0x00200000>;
+			mode = "640x480-32@60";
+			interrupts = <1 1 3>;
+			little-endian;
+		};
+
+		mram0@3,0 {
+			compatible = "mtd-ram";
+			reg = <3 0x00000 0x80000>;
+			bank-width = <1>;
+		};
+	};
+
+	pci@f0000d00 {
+		#interrupt-cells = <1>;
+		#size-cells = <2>;
+		#address-cells = <3>;
+		device_type = "pci";
+		compatible = "fsl,mpc5200-pci";
+		reg = <0xf0000d00 0x100>;
+		interrupt-map-mask = <0xf800 0 0 7>;
+		interrupt-map = <0xc000 0 0 1 &mpc5200_pic 0 0 3
+				 0xc000 0 0 2 &mpc5200_pic 0 0 3
+				 0xc000 0 0 3 &mpc5200_pic 0 0 3
+				 0xc000 0 0 4 &mpc5200_pic 0 0 3>;
+		clock-frequency = <0>; // From boot loader
+		interrupts = <2 8 0 2 9 0 2 10 0>;
+		bus-range = <0 0>;
+		ranges = <0x42000000 0 0x80000000 0x80000000 0 0x10000000
+			  0x02000000 0 0x90000000 0x90000000 0 0x10000000
+			  0x01000000 0 0x00000000 0xa0000000 0 0x01000000>;
+	};
+};
diff --git a/arch/powerpc/boot/dts/p1023rds.dts b/arch/powerpc/boot/dts/p1023rds.dts
index d9b7767..d3b4782 100644
--- a/arch/powerpc/boot/dts/p1023rds.dts
+++ b/arch/powerpc/boot/dts/p1023rds.dts
@@ -449,6 +449,7 @@
 			interrupt-parent = <&mpic>;
 			interrupts = <16 2>;
 			interrupt-map-mask = <0xf800 0 0 7>;
+			/* IRQ[0:3] are pulled up on board, set to active-low */
 			interrupt-map = <
 				/* IDSEL 0x0 */
 				0000 0 0 1 &mpic 0 1
@@ -488,11 +489,15 @@
 			interrupt-parent = <&mpic>;
 			interrupts = <16 2>;
 			interrupt-map-mask = <0xf800 0 0 7>;
+			/*
+			 * IRQ[4:6] only for PCIe, set to active-high,
+			 * IRQ[7] is pulled up on board, set to active-low
+			 */
 			interrupt-map = <
 				/* IDSEL 0x0 */
-				0000 0 0 1 &mpic 4 1
-				0000 0 0 2 &mpic 5 1
-				0000 0 0 3 &mpic 6 1
+				0000 0 0 1 &mpic 4 2
+				0000 0 0 2 &mpic 5 2
+				0000 0 0 3 &mpic 6 2
 				0000 0 0 4 &mpic 7 1
 				>;
 			ranges = <0x2000000 0x0 0xa0000000
@@ -527,12 +532,16 @@
 			interrupt-parent = <&mpic>;
 			interrupts = <16 2>;
 			interrupt-map-mask = <0xf800 0 0 7>;
+			/*
+			 * IRQ[8:10] are pulled up on board, set to active-low
+			 * IRQ[11] only for PCIe, set to active-high,
+			 */
 			interrupt-map = <
 				/* IDSEL 0x0 */
 				0000 0 0 1 &mpic 8 1
 				0000 0 0 2 &mpic 9 1
 				0000 0 0 3 &mpic 10 1
-				0000 0 0 4 &mpic 11 1
+				0000 0 0 4 &mpic 11 2
 				>;
 			ranges = <0x2000000 0x0 0x80000000
 				  0x2000000 0x0 0x80000000
diff --git a/arch/powerpc/configs/52xx/tqm5200_defconfig b/arch/powerpc/configs/52xx/tqm5200_defconfig
index 959cd2c..716a37b 100644
--- a/arch/powerpc/configs/52xx/tqm5200_defconfig
+++ b/arch/powerpc/configs/52xx/tqm5200_defconfig
@@ -1,9 +1,10 @@
 CONFIG_EXPERIMENTAL=y
 CONFIG_SYSVIPC=y
+CONFIG_SPARSE_IRQ=y
 CONFIG_LOG_BUF_SHIFT=14
 CONFIG_BLK_DEV_INITRD=y
 # CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
-CONFIG_EXPERT=y
+CONFIG_EMBEDDED=y
 # CONFIG_SYSCTL_SYSCALL is not set
 # CONFIG_KALLSYMS is not set
 # CONFIG_EPOLL is not set
@@ -17,7 +18,6 @@
 CONFIG_PPC_MPC5200_BUGFIX=y
 # CONFIG_PPC_PMAC is not set
 CONFIG_PPC_BESTCOMM=y
-CONFIG_SPARSE_IRQ=y
 CONFIG_PM=y
 # CONFIG_PCI is not set
 CONFIG_NET=y
@@ -38,17 +38,18 @@
 CONFIG_MTD_CONCAT=y
 CONFIG_MTD_PARTITIONS=y
 CONFIG_MTD_CMDLINE_PARTS=y
+CONFIG_MTD_OF_PARTS=y
 CONFIG_MTD_CHAR=y
 CONFIG_MTD_BLOCK=y
 CONFIG_MTD_CFI=y
 CONFIG_MTD_CFI_AMDSTD=y
 CONFIG_MTD_ROM=y
 CONFIG_MTD_PHYSMAP_OF=y
+CONFIG_MTD_PLATRAM=y
 CONFIG_PROC_DEVICETREE=y
 CONFIG_BLK_DEV_LOOP=y
 CONFIG_BLK_DEV_RAM=y
 CONFIG_BLK_DEV_RAM_SIZE=32768
-# CONFIG_MISC_DEVICES is not set
 CONFIG_BLK_DEV_SD=y
 CONFIG_CHR_DEV_SG=y
 CONFIG_ATA=y
@@ -56,13 +57,11 @@
 CONFIG_PATA_PLATFORM=y
 CONFIG_NETDEVICES=y
 CONFIG_LXT_PHY=y
+CONFIG_FIXED_PHY=y
 CONFIG_NET_ETHERNET=y
 CONFIG_FEC_MPC52xx=y
 # CONFIG_NETDEV_1000 is not set
 # CONFIG_NETDEV_10000 is not set
-# CONFIG_INPUT is not set
-# CONFIG_SERIO is not set
-# CONFIG_VT is not set
 CONFIG_SERIAL_MPC52xx=y
 CONFIG_SERIAL_MPC52xx_CONSOLE=y
 CONFIG_SERIAL_MPC52xx_CONSOLE_BAUD=115200
@@ -70,7 +69,13 @@
 CONFIG_I2C=y
 CONFIG_I2C_CHARDEV=y
 CONFIG_I2C_MPC=y
+CONFIG_SENSORS_LM80=y
 CONFIG_WATCHDOG=y
+CONFIG_MFD_SM501=y
+CONFIG_FB=y
+CONFIG_FB_FOREIGN_ENDIAN=y
+CONFIG_FB_SM501=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
 CONFIG_USB=y
 CONFIG_USB_DEVICEFS=y
 # CONFIG_USB_DEVICE_CLASS is not set
@@ -80,10 +85,10 @@
 CONFIG_USB_STORAGE=y
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_DRV_DS1307=y
+CONFIG_RTC_DRV_DS1374=y
 CONFIG_EXT2_FS=y
 CONFIG_EXT3_FS=y
 # CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
-CONFIG_INOTIFY=y
 CONFIG_MSDOS_FS=y
 CONFIG_VFAT_FS=y
 CONFIG_PROC_KCORE=y
@@ -102,7 +107,6 @@
 CONFIG_DETECT_HUNG_TASK=y
 # CONFIG_DEBUG_BUGVERBOSE is not set
 CONFIG_DEBUG_INFO=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
 CONFIG_CRYPTO_ECB=y
 CONFIG_CRYPTO_PCBC=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
diff --git a/arch/powerpc/configs/ppc44x_defconfig b/arch/powerpc/configs/ppc44x_defconfig
index 6cdf1c0..3b98d73 100644
--- a/arch/powerpc/configs/ppc44x_defconfig
+++ b/arch/powerpc/configs/ppc44x_defconfig
@@ -52,6 +52,8 @@
 CONFIG_MTD_JEDECPROBE=y
 CONFIG_MTD_CFI_AMDSTD=y
 CONFIG_MTD_PHYSMAP_OF=y
+CONFIG_MTD_NAND=m
+CONFIG_MTD_NAND_NDFC=m
 CONFIG_MTD_UBI=m
 CONFIG_MTD_UBI_GLUEBI=m
 CONFIG_PROC_DEVICETREE=y
diff --git a/arch/powerpc/configs/ppc64_defconfig b/arch/powerpc/configs/ppc64_defconfig
index 84a685a5..535711f 100644
--- a/arch/powerpc/configs/ppc64_defconfig
+++ b/arch/powerpc/configs/ppc64_defconfig
@@ -485,3 +485,7 @@
 CONFIG_CRYPTO_LZO=m
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 # CONFIG_CRYPTO_HW is not set
+CONFIG_VIRTUALIZATION=y
+CONFIG_KVM_BOOK3S_64=m
+CONFIG_KVM_BOOK3S_64_HV=y
+CONFIG_VHOST_NET=m
diff --git a/arch/powerpc/configs/pseries_defconfig b/arch/powerpc/configs/pseries_defconfig
index 96a58b7..a72f241 100644
--- a/arch/powerpc/configs/pseries_defconfig
+++ b/arch/powerpc/configs/pseries_defconfig
@@ -362,3 +362,7 @@
 CONFIG_CRYPTO_LZO=m
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 # CONFIG_CRYPTO_HW is not set
+CONFIG_VIRTUALIZATION=y
+CONFIG_KVM_BOOK3S_64=m
+CONFIG_KVM_BOOK3S_64_HV=y
+CONFIG_VHOST_NET=m
diff --git a/arch/powerpc/include/asm/atomic.h b/arch/powerpc/include/asm/atomic.h
index e2a4c26..02e41b5 100644
--- a/arch/powerpc/include/asm/atomic.h
+++ b/arch/powerpc/include/asm/atomic.h
@@ -49,13 +49,13 @@
 	int t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	lwarx	%0,0,%2		# atomic_add_return\n\
 	add	%0,%1,%0\n"
 	PPC405_ERR77(0,%2)
 "	stwcx.	%0,0,%2 \n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (t)
 	: "r" (a), "r" (&v->counter)
 	: "cc", "memory");
@@ -85,13 +85,13 @@
 	int t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	lwarx	%0,0,%2		# atomic_sub_return\n\
 	subf	%0,%1,%0\n"
 	PPC405_ERR77(0,%2)
 "	stwcx.	%0,0,%2 \n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (t)
 	: "r" (a), "r" (&v->counter)
 	: "cc", "memory");
@@ -119,13 +119,13 @@
 	int t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	lwarx	%0,0,%1		# atomic_inc_return\n\
 	addic	%0,%0,1\n"
 	PPC405_ERR77(0,%1)
 "	stwcx.	%0,0,%1 \n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (t)
 	: "r" (&v->counter)
 	: "cc", "xer", "memory");
@@ -163,13 +163,13 @@
 	int t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	lwarx	%0,0,%1		# atomic_dec_return\n\
 	addic	%0,%0,-1\n"
 	PPC405_ERR77(0,%1)
 "	stwcx.	%0,0,%1\n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (t)
 	: "r" (&v->counter)
 	: "cc", "xer", "memory");
@@ -194,7 +194,7 @@
 	int t;
 
 	__asm__ __volatile__ (
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	lwarx	%0,0,%1		# __atomic_add_unless\n\
 	cmpw	0,%0,%3 \n\
 	beq-	2f \n\
@@ -202,7 +202,7 @@
 	PPC405_ERR77(0,%2)
 "	stwcx.	%0,0,%1 \n\
 	bne-	1b \n"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 "	subf	%0,%2,%0 \n\
 2:"
 	: "=&r" (t)
@@ -226,7 +226,7 @@
 	int t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	lwarx	%0,0,%1		# atomic_dec_if_positive\n\
 	cmpwi	%0,1\n\
 	addi	%0,%0,-1\n\
@@ -234,7 +234,7 @@
 	PPC405_ERR77(0,%1)
 "	stwcx.	%0,0,%1\n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	"\n\
 2:"	: "=&b" (t)
 	: "r" (&v->counter)
@@ -285,12 +285,12 @@
 	long t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	ldarx	%0,0,%2		# atomic64_add_return\n\
 	add	%0,%1,%0\n\
 	stdcx.	%0,0,%2 \n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (t)
 	: "r" (a), "r" (&v->counter)
 	: "cc", "memory");
@@ -319,12 +319,12 @@
 	long t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	ldarx	%0,0,%2		# atomic64_sub_return\n\
 	subf	%0,%1,%0\n\
 	stdcx.	%0,0,%2 \n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (t)
 	: "r" (a), "r" (&v->counter)
 	: "cc", "memory");
@@ -351,12 +351,12 @@
 	long t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	ldarx	%0,0,%1		# atomic64_inc_return\n\
 	addic	%0,%0,1\n\
 	stdcx.	%0,0,%1 \n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (t)
 	: "r" (&v->counter)
 	: "cc", "xer", "memory");
@@ -393,12 +393,12 @@
 	long t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	ldarx	%0,0,%1		# atomic64_dec_return\n\
 	addic	%0,%0,-1\n\
 	stdcx.	%0,0,%1\n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (t)
 	: "r" (&v->counter)
 	: "cc", "xer", "memory");
@@ -418,13 +418,13 @@
 	long t;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	ldarx	%0,0,%1		# atomic64_dec_if_positive\n\
 	addic.	%0,%0,-1\n\
 	blt-	2f\n\
 	stdcx.	%0,0,%1\n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	"\n\
 2:"	: "=&r" (t)
 	: "r" (&v->counter)
@@ -450,14 +450,14 @@
 	long t;
 
 	__asm__ __volatile__ (
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	ldarx	%0,0,%1		# __atomic_add_unless\n\
 	cmpd	0,%0,%3 \n\
 	beq-	2f \n\
 	add	%0,%2,%0 \n"
 "	stdcx.	%0,0,%1 \n\
 	bne-	1b \n"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 "	subf	%0,%2,%0 \n\
 2:"
 	: "=&r" (t)
diff --git a/arch/powerpc/include/asm/bitops.h b/arch/powerpc/include/asm/bitops.h
index e137afc..efdc926 100644
--- a/arch/powerpc/include/asm/bitops.h
+++ b/arch/powerpc/include/asm/bitops.h
@@ -124,14 +124,14 @@
 	return (old & mask);				\
 }
 
-DEFINE_TESTOP(test_and_set_bits, or, PPC_RELEASE_BARRIER,
-	      PPC_ACQUIRE_BARRIER, 0)
+DEFINE_TESTOP(test_and_set_bits, or, PPC_ATOMIC_ENTRY_BARRIER,
+	      PPC_ATOMIC_EXIT_BARRIER, 0)
 DEFINE_TESTOP(test_and_set_bits_lock, or, "",
 	      PPC_ACQUIRE_BARRIER, 1)
-DEFINE_TESTOP(test_and_clear_bits, andc, PPC_RELEASE_BARRIER,
-	      PPC_ACQUIRE_BARRIER, 0)
-DEFINE_TESTOP(test_and_change_bits, xor, PPC_RELEASE_BARRIER,
-	      PPC_ACQUIRE_BARRIER, 0)
+DEFINE_TESTOP(test_and_clear_bits, andc, PPC_ATOMIC_ENTRY_BARRIER,
+	      PPC_ATOMIC_EXIT_BARRIER, 0)
+DEFINE_TESTOP(test_and_change_bits, xor, PPC_ATOMIC_ENTRY_BARRIER,
+	      PPC_ATOMIC_EXIT_BARRIER, 0)
 
 static __inline__ int test_and_set_bit(unsigned long nr,
 				       volatile unsigned long *addr)
diff --git a/arch/powerpc/include/asm/floppy.h b/arch/powerpc/include/asm/floppy.h
index 24bd34c..936a904 100644
--- a/arch/powerpc/include/asm/floppy.h
+++ b/arch/powerpc/include/asm/floppy.h
@@ -108,10 +108,10 @@
 {
 	if (can_use_virtual_dma)
 		return request_irq(FLOPPY_IRQ, floppy_hardint,
-				   IRQF_DISABLED, "floppy", NULL);
+				   0, "floppy", NULL);
 	else
 		return request_irq(FLOPPY_IRQ, floppy_interrupt,
-				   IRQF_DISABLED, "floppy", NULL);
+				   0, "floppy", NULL);
 }
 
 static int vdma_dma_setup(char *addr, unsigned long size, int mode, int io)
diff --git a/arch/powerpc/include/asm/futex.h b/arch/powerpc/include/asm/futex.h
index c94e4a3..2a9cf84 100644
--- a/arch/powerpc/include/asm/futex.h
+++ b/arch/powerpc/include/asm/futex.h
@@ -11,12 +11,13 @@
 
 #define __futex_atomic_op(insn, ret, oldval, uaddr, oparg) \
   __asm__ __volatile ( \
-	PPC_RELEASE_BARRIER \
+	PPC_ATOMIC_ENTRY_BARRIER \
 "1:	lwarx	%0,0,%2\n" \
 	insn \
 	PPC405_ERR77(0, %2) \
 "2:	stwcx.	%1,0,%2\n" \
 	"bne-	1b\n" \
+	PPC_ATOMIC_EXIT_BARRIER \
 	"li	%1,0\n" \
 "3:	.section .fixup,\"ax\"\n" \
 "4:	li	%1,%3\n" \
@@ -92,14 +93,14 @@
 		return -EFAULT;
 
         __asm__ __volatile__ (
-        PPC_RELEASE_BARRIER
+        PPC_ATOMIC_ENTRY_BARRIER
 "1:     lwarx   %1,0,%3         # futex_atomic_cmpxchg_inatomic\n\
         cmpw    0,%1,%4\n\
         bne-    3f\n"
         PPC405_ERR77(0,%3)
 "2:     stwcx.  %5,0,%3\n\
         bne-    1b\n"
-        PPC_ACQUIRE_BARRIER
+        PPC_ATOMIC_EXIT_BARRIER
 "3:	.section .fixup,\"ax\"\n\
 4:	li	%0,%6\n\
 	b	3b\n\
diff --git a/arch/powerpc/include/asm/kvm.h b/arch/powerpc/include/asm/kvm.h
index 08fe69e..0ad432b 100644
--- a/arch/powerpc/include/asm/kvm.h
+++ b/arch/powerpc/include/asm/kvm.h
@@ -149,12 +149,6 @@
 #define KVM_SREGS_E_UPDATE_DBSR		(1 << 3)
 
 /*
- * Book3S special bits to indicate contents in the struct by maintaining
- * backwards compatibility with older structs. If adding a new field,
- * please make sure to add a flag for that new field */
-#define KVM_SREGS_S_HIOR		(1 << 0)
-
-/*
  * In KVM_SET_SREGS, reserved/pad fields must be left untouched from a
  * previous KVM_GET_REGS.
  *
@@ -179,8 +173,6 @@
 				__u64 ibat[8]; 
 				__u64 dbat[8]; 
 			} ppc32;
-			__u64 flags; /* KVM_SREGS_S_ */
-			__u64 hior;
 		} s;
 		struct {
 			union {
diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
index a384ffd..d4df013 100644
--- a/arch/powerpc/include/asm/kvm_book3s.h
+++ b/arch/powerpc/include/asm/kvm_book3s.h
@@ -90,8 +90,6 @@
 #endif
 	int context_id[SID_CONTEXTS];
 
-	bool hior_sregs;		/* HIOR is set by SREGS, not PVR */
-
 	struct hlist_head hpte_hash_pte[HPTEG_HASH_NUM_PTE];
 	struct hlist_head hpte_hash_pte_long[HPTEG_HASH_NUM_PTE_LONG];
 	struct hlist_head hpte_hash_vpte[HPTEG_HASH_NUM_VPTE];
diff --git a/arch/powerpc/include/asm/lv1call.h b/arch/powerpc/include/asm/lv1call.h
index 9cd5fc8..f77c708 100644
--- a/arch/powerpc/include/asm/lv1call.h
+++ b/arch/powerpc/include/asm/lv1call.h
@@ -316,7 +316,7 @@
 LV1_CALL(gpu_context_iomap,                             5, 0, 221 )
 LV1_CALL(gpu_context_attribute,                         6, 0, 225 )
 LV1_CALL(gpu_context_intr,                              1, 1, 227 )
-LV1_CALL(gpu_attribute,                                 5, 0, 228 )
+LV1_CALL(gpu_attribute,                                 3, 0, 228 )
 LV1_CALL(get_rtc,                                       0, 2, 232 )
 LV1_CALL(set_ppe_periodic_tracer_frequency,             1, 0, 240 )
 LV1_CALL(start_ppe_periodic_tracer,                     5, 0, 241 )
diff --git a/arch/powerpc/include/asm/reg_booke.h b/arch/powerpc/include/asm/reg_booke.h
index 28cdbd9..03c48e8 100644
--- a/arch/powerpc/include/asm/reg_booke.h
+++ b/arch/powerpc/include/asm/reg_booke.h
@@ -31,7 +31,7 @@
 
 #define MSR_		MSR_ME | MSR_CE
 #define MSR_KERNEL	MSR_ | MSR_64BIT
-#define MSR_USER32	MSR_ | MSR_PR | MSR_EE | MSR_DE
+#define MSR_USER32	MSR_ | MSR_PR | MSR_EE
 #define MSR_USER64	MSR_USER32 | MSR_64BIT
 #elif defined (CONFIG_40x)
 #define MSR_KERNEL	(MSR_ME|MSR_RI|MSR_IR|MSR_DR|MSR_CE)
diff --git a/arch/powerpc/include/asm/sections.h b/arch/powerpc/include/asm/sections.h
index 6fbce72..a0f358d 100644
--- a/arch/powerpc/include/asm/sections.h
+++ b/arch/powerpc/include/asm/sections.h
@@ -8,7 +8,7 @@
 
 #ifdef __powerpc64__
 
-extern char _end[];
+extern char __end_interrupts[];
 
 static inline int in_kernel_text(unsigned long addr)
 {
diff --git a/arch/powerpc/include/asm/synch.h b/arch/powerpc/include/asm/synch.h
index d7cab44..e682a71 100644
--- a/arch/powerpc/include/asm/synch.h
+++ b/arch/powerpc/include/asm/synch.h
@@ -13,6 +13,7 @@
 extern unsigned int __start___lwsync_fixup, __stop___lwsync_fixup;
 extern void do_lwsync_fixups(unsigned long value, void *fixup_start,
 			     void *fixup_end);
+extern void do_final_fixups(void);
 
 static inline void eieio(void)
 {
@@ -41,11 +42,15 @@
 	START_LWSYNC_SECTION(97);			\
 	isync;						\
 	MAKE_LWSYNC_SECTION_ENTRY(97, __lwsync_fixup);
-#define PPC_ACQUIRE_BARRIER	"\n" stringify_in_c(__PPC_ACQUIRE_BARRIER)
-#define PPC_RELEASE_BARRIER	stringify_in_c(LWSYNC) "\n"
+#define PPC_ACQUIRE_BARRIER	 "\n" stringify_in_c(__PPC_ACQUIRE_BARRIER)
+#define PPC_RELEASE_BARRIER	 stringify_in_c(LWSYNC) "\n"
+#define PPC_ATOMIC_ENTRY_BARRIER "\n" stringify_in_c(LWSYNC) "\n"
+#define PPC_ATOMIC_EXIT_BARRIER	 "\n" stringify_in_c(sync) "\n"
 #else
 #define PPC_ACQUIRE_BARRIER
 #define PPC_RELEASE_BARRIER
+#define PPC_ATOMIC_ENTRY_BARRIER
+#define PPC_ATOMIC_EXIT_BARRIER
 #endif
 
 #endif /* __KERNEL__ */
diff --git a/arch/powerpc/include/asm/xics.h b/arch/powerpc/include/asm/xics.h
index bd6c401..c48de98 100644
--- a/arch/powerpc/include/asm/xics.h
+++ b/arch/powerpc/include/asm/xics.h
@@ -15,8 +15,8 @@
 #define	DEFAULT_PRIORITY	5
 
 /*
- * Mark IPIs as higher priority so we can take them inside interrupts that
- * arent marked IRQF_DISABLED
+ * Mark IPIs as higher priority so we can take them inside interrupts
+ * FIXME: still true now?
  */
 #define IPI_PRIORITY		4
 
diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
index 56212bc..4f80cf1 100644
--- a/arch/powerpc/kernel/entry_32.S
+++ b/arch/powerpc/kernel/entry_32.S
@@ -215,7 +215,22 @@
 	stw	r9,8(r1)
 	stw	r11,12(r1)
 	stw	r3,ORIG_GPR3(r1)
+	/*
+	 * The trace_hardirqs_off will use CALLER_ADDR0 and CALLER_ADDR1.
+	 * If from user mode there is only one stack frame on the stack, and
+	 * accessing CALLER_ADDR1 will cause oops. So we need create a dummy
+	 * stack frame to make trace_hardirqs_off happy.
+	 */
+	andi.	r12,r12,MSR_PR
+	beq	11f
+	stwu	r1,-16(r1)
 	bl	trace_hardirqs_off
+	addi	r1,r1,16
+	b	12f
+
+11:
+	bl	trace_hardirqs_off
+12:
 	lwz	r0,GPR0(r1)
 	lwz	r3,ORIG_GPR3(r1)
 	lwz	r4,GPR4(r1)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index a54d92f..cf9c69b 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -267,7 +267,7 @@
 
 #ifdef CONFIG_CBE_RAS
 	STD_EXCEPTION_HV(0x1200, 0x1202, cbe_system_error)
-	KVM_HANDLER_PR_SKIP(PACA_EXGEN, EXC_HV, 0x1202)
+	KVM_HANDLER_SKIP(PACA_EXGEN, EXC_HV, 0x1202)
 #endif /* CONFIG_CBE_RAS */
 
 	STD_EXCEPTION_PSERIES(0x1300, 0x1300, instruction_breakpoint)
@@ -275,7 +275,7 @@
 
 #ifdef CONFIG_CBE_RAS
 	STD_EXCEPTION_HV(0x1600, 0x1602, cbe_maintenance)
-	KVM_HANDLER_PR_SKIP(PACA_EXGEN, EXC_HV, 0x1602)
+	KVM_HANDLER_SKIP(PACA_EXGEN, EXC_HV, 0x1602)
 #endif /* CONFIG_CBE_RAS */
 
 	STD_EXCEPTION_PSERIES(0x1700, 0x1700, altivec_assist)
@@ -283,7 +283,7 @@
 
 #ifdef CONFIG_CBE_RAS
 	STD_EXCEPTION_HV(0x1800, 0x1802, cbe_thermal)
-	KVM_HANDLER_PR_SKIP(PACA_EXGEN, EXC_HV, 0x1802)
+	KVM_HANDLER_SKIP(PACA_EXGEN, EXC_HV, 0x1802)
 #endif /* CONFIG_CBE_RAS */
 
 	. = 0x3000
diff --git a/arch/powerpc/kernel/jump_label.c b/arch/powerpc/kernel/jump_label.c
index 368d158..a1ed8a8 100644
--- a/arch/powerpc/kernel/jump_label.c
+++ b/arch/powerpc/kernel/jump_label.c
@@ -11,6 +11,7 @@
 #include <linux/jump_label.h>
 #include <asm/code-patching.h>
 
+#ifdef HAVE_JUMP_LABEL
 void arch_jump_label_transform(struct jump_entry *entry,
 			       enum jump_label_type type)
 {
@@ -21,3 +22,4 @@
 	else
 		patch_instruction(addr, PPC_INST_NOP);
 }
+#endif
diff --git a/arch/powerpc/kernel/kvm.c b/arch/powerpc/kernel/kvm.c
index 35f2764..2985338 100644
--- a/arch/powerpc/kernel/kvm.c
+++ b/arch/powerpc/kernel/kvm.c
@@ -132,7 +132,6 @@
 	/* On relocatable kernels interrupts handlers and our code
 	   can be in different regions, so we don't patch them */
 
-	extern u32 __end_interrupts;
 	if ((ulong)inst < (ulong)&__end_interrupts)
 		return;
 #endif
diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S
index f7d760a..7cd07b4 100644
--- a/arch/powerpc/kernel/misc_32.S
+++ b/arch/powerpc/kernel/misc_32.S
@@ -738,7 +738,7 @@
 	mr      r5, r31
 
 	li	r0, 0
-#elif defined(CONFIG_44x)  && !defined(CONFIG_47x)
+#elif defined(CONFIG_44x)  && !defined(CONFIG_PPC_47x)
 
 /*
  * Code for setting up 1:1 mapping for PPC440x for KEXEC
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index 9054ca9..6457574 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -486,28 +486,6 @@
 	new_thread = &new->thread;
 	old_thread = &current->thread;
 
-#if defined(CONFIG_PPC_BOOK3E_64)
-	/* XXX Current Book3E code doesn't deal with kernel side DBCR0,
-	 * we always hold the user values, so we set it now.
-	 *
-	 * However, we ensure the kernel MSR:DE is appropriately cleared too
-	 * to avoid spurrious single step exceptions in the kernel.
-	 *
-	 * This will have to change to merge with the ppc32 code at some point,
-	 * but I don't like much what ppc32 is doing today so there's some
-	 * thinking needed there
-	 */
-	if ((new_thread->dbcr0 | old_thread->dbcr0) & DBCR0_IDM) {
-		u32 dbcr0;
-
-		mtmsr(mfmsr() & ~MSR_DE);
-		isync();
-		dbcr0 = mfspr(SPRN_DBCR0);
-		dbcr0 = (dbcr0 & DBCR0_EDM) | new_thread->dbcr0;
-		mtspr(SPRN_DBCR0, dbcr0);
-	}
-#endif /* CONFIG_PPC64_BOOK3E */
-
 #ifdef CONFIG_PPC64
 	/*
 	 * Collect processor utilization data per process
@@ -657,7 +635,7 @@
 	if ((regs->trap != 0xc00) && cpu_has_feature(CPU_FTR_CFAR))
 		printk("CFAR: "REG"\n", regs->orig_gpr3);
 	if (trap == 0x300 || trap == 0x600)
-#ifdef CONFIG_PPC_ADV_DEBUG_REGS
+#if defined(CONFIG_4xx) || defined(CONFIG_BOOKE)
 		printk("DEAR: "REG", ESR: "REG"\n", regs->dar, regs->dsisr);
 #else
 		printk("DAR: "REG", DSISR: %08lx\n", regs->dar, regs->dsisr);
diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index b4fa661..cc58486 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -1579,10 +1579,8 @@
 		return;
 
 	base = alloc_down(size, PAGE_SIZE, 0);
-	if (base == 0) {
-		prom_printf("RTAS allocation failed !\n");
-		return;
-	}
+	if (base == 0)
+		prom_panic("Could not allocate memory for RTAS\n");
 
 	rtas_inst = call_prom("open", 1, 1, ADDR("/rtas"));
 	if (!IHANDLE_VALID(rtas_inst)) {
diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index c1ce863..ac76108 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -107,6 +107,8 @@
 			 PTRRELOC(&__start___lwsync_fixup),
 			 PTRRELOC(&__stop___lwsync_fixup));
 
+	do_final_fixups();
+
 	return KERNELBASE + offset;
 }
 
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 1a9dea8..fb9bb46 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -359,6 +359,7 @@
 			  &__start___fw_ftr_fixup, &__stop___fw_ftr_fixup);
 	do_lwsync_fixups(cur_cpu_spec->cpu_features,
 			 &__start___lwsync_fixup, &__stop___lwsync_fixup);
+	do_final_fixups();
 
 	/*
 	 * Unflatten the device-tree passed by prom_init or kexec
diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c
index 78b76dc..836a5a1 100644
--- a/arch/powerpc/kernel/signal_32.c
+++ b/arch/powerpc/kernel/signal_32.c
@@ -97,7 +97,7 @@
 	compat_sigset_t	cset;
 
 	switch (_NSIG_WORDS) {
-	case 4: cset.sig[5] = set->sig[3] & 0xffffffffull;
+	case 4: cset.sig[6] = set->sig[3] & 0xffffffffull;
 		cset.sig[7] = set->sig[3] >> 32;
 	case 3: cset.sig[4] = set->sig[2] & 0xffffffffull;
 		cset.sig[5] = set->sig[2] >> 32;
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 25ddbfc..6df7090 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -187,7 +187,7 @@
 		return 1;
 	}
 #endif
-	err = request_irq(virq, smp_ipi_action[msg], IRQF_DISABLED|IRQF_PERCPU,
+	err = request_irq(virq, smp_ipi_action[msg], IRQF_PERCPU,
 			  smp_ipi_name[msg], 0);
 	WARN(err < 0, "unable to request_irq %d for %s (rc %d)\n",
 		virq, smp_ipi_name[msg], err);
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index 4e59082..5459d14 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -1298,14 +1298,12 @@
 
 		if (user_mode(regs)) {
 			current->thread.dbcr0 &= ~DBCR0_IC;
-#ifdef CONFIG_PPC_ADV_DEBUG_REGS
 			if (DBCR_ACTIVE_EVENTS(current->thread.dbcr0,
 					       current->thread.dbcr1))
 				regs->msr |= MSR_DE;
 			else
 				/* Make sure the IDM bit is off */
 				current->thread.dbcr0 &= ~DBCR0_IDM;
-#endif
 		}
 
 		_exception(SIGTRAP, regs, TRAP_TRACE, regs->nip);
diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c
index 0cdbc07..0cb137a 100644
--- a/arch/powerpc/kvm/book3s_hv.c
+++ b/arch/powerpc/kvm/book3s_hv.c
@@ -44,6 +44,7 @@
 #include <asm/processor.h>
 #include <asm/cputhreads.h>
 #include <asm/page.h>
+#include <asm/hvcall.h>
 #include <linux/gfp.h>
 #include <linux/sched.h>
 #include <linux/vmalloc.h>
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index f422231..44d8829 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1263,7 +1263,7 @@
 	addi	r6,r5,VCORE_NAPPING_THREADS
 31:	lwarx	r4,0,r6
 	or	r4,r4,r0
-	popcntw	r7,r4
+	PPC_POPCNTW(r7,r4)
 	cmpw	r7,r8
 	bge	2f
 	stwcx.	r4,0,r6
diff --git a/arch/powerpc/kvm/book3s_pr.c b/arch/powerpc/kvm/book3s_pr.c
index bc4d50d..3c791e1 100644
--- a/arch/powerpc/kvm/book3s_pr.c
+++ b/arch/powerpc/kvm/book3s_pr.c
@@ -151,16 +151,14 @@
 #ifdef CONFIG_PPC_BOOK3S_64
 	if ((pvr >= 0x330000) && (pvr < 0x70330000)) {
 		kvmppc_mmu_book3s_64_init(vcpu);
-		if (!to_book3s(vcpu)->hior_sregs)
-			to_book3s(vcpu)->hior = 0xfff00000;
+		to_book3s(vcpu)->hior = 0xfff00000;
 		to_book3s(vcpu)->msr_mask = 0xffffffffffffffffULL;
 		vcpu->arch.cpu_type = KVM_CPU_3S_64;
 	} else
 #endif
 	{
 		kvmppc_mmu_book3s_32_init(vcpu);
-		if (!to_book3s(vcpu)->hior_sregs)
-			to_book3s(vcpu)->hior = 0;
+		to_book3s(vcpu)->hior = 0;
 		to_book3s(vcpu)->msr_mask = 0xffffffffULL;
 		vcpu->arch.cpu_type = KVM_CPU_3S_32;
 	}
@@ -797,9 +795,6 @@
 		}
 	}
 
-	if (sregs->u.s.flags & KVM_SREGS_S_HIOR)
-		sregs->u.s.hior = to_book3s(vcpu)->hior;
-
 	return 0;
 }
 
@@ -836,11 +831,6 @@
 	/* Flush the MMU after messing with the segments */
 	kvmppc_mmu_pte_flush(vcpu, 0, 0);
 
-	if (sregs->u.s.flags & KVM_SREGS_S_HIOR) {
-		to_book3s(vcpu)->hior_sregs = true;
-		to_book3s(vcpu)->hior = sregs->u.s.hior;
-	}
-
 	return 0;
 }
 
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index efbf9ad..607fbdf 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -208,7 +208,6 @@
 	case KVM_CAP_PPC_BOOKE_SREGS:
 #else
 	case KVM_CAP_PPC_SEGSTATE:
-	case KVM_CAP_PPC_HIOR:
 	case KVM_CAP_PPC_PAPR:
 #endif
 	case KVM_CAP_PPC_UNSET_IRQ:
diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c
index 0d08d01..7a8a748 100644
--- a/arch/powerpc/lib/feature-fixups.c
+++ b/arch/powerpc/lib/feature-fixups.c
@@ -18,6 +18,8 @@
 #include <linux/init.h>
 #include <asm/cputable.h>
 #include <asm/code-patching.h>
+#include <asm/page.h>
+#include <asm/sections.h>
 
 
 struct fixup_entry {
@@ -128,6 +130,27 @@
 	}
 }
 
+void do_final_fixups(void)
+{
+#if defined(CONFIG_PPC64) && defined(CONFIG_RELOCATABLE)
+	int *src, *dest;
+	unsigned long length;
+
+	if (PHYSICAL_START == 0)
+		return;
+
+	src = (int *)(KERNELBASE + PHYSICAL_START);
+	dest = (int *)KERNELBASE;
+	length = (__end_interrupts - _stext) / sizeof(int);
+
+	while (length--) {
+		patch_instruction(dest, *src);
+		src++;
+		dest++;
+	}
+#endif
+}
+
 #ifdef CONFIG_FTR_FIXUP_SELFTEST
 
 #define check(x)	\
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index 5964371..8558b57 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -15,6 +15,7 @@
 #include <linux/of_fdt.h>
 #include <linux/memblock.h>
 #include <linux/bootmem.h>
+#include <linux/moduleparam.h>
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 #include <asm/tlb.h>
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 16da595..2dd6bdd 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -34,6 +34,7 @@
 #include <linux/suspend.h>
 #include <linux/memblock.h>
 #include <linux/hugetlb.h>
+#include <linux/slab.h>
 
 #include <asm/pgalloc.h>
 #include <asm/prom.h>
@@ -555,3 +556,32 @@
 		book3e_hugetlb_preload(vma->vm_mm, address, *ptep);
 #endif
 }
+
+/*
+ * System memory should not be in /proc/iomem but various tools expect it
+ * (eg kdump).
+ */
+static int add_system_ram_resources(void)
+{
+	struct memblock_region *reg;
+
+	for_each_memblock(memory, reg) {
+		struct resource *res;
+		unsigned long base = reg->base;
+		unsigned long size = reg->size;
+
+		res = kzalloc(sizeof(struct resource), GFP_KERNEL);
+		WARN_ON(!res);
+
+		if (res) {
+			res->name = "System RAM";
+			res->start = base;
+			res->end = base + size - 1;
+			res->flags = IORESOURCE_MEM;
+			WARN_ON(request_resource(&iomem_resource, res) < 0);
+		}
+	}
+
+	return 0;
+}
+subsys_initcall(add_system_ram_resources);
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index c7dd4de..b22a83a 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -315,7 +315,10 @@
 	struct device_node *root;
 	const char *vec5;
 
-	root = of_find_node_by_path("/rtas");
+	if (firmware_has_feature(FW_FEATURE_OPAL))
+		root = of_find_node_by_path("/ibm,opal");
+	else
+		root = of_find_node_by_path("/rtas");
 	if (!root)
 		root = of_find_node_by_path("/");
 
@@ -344,12 +347,19 @@
 
 #define VEC5_AFFINITY_BYTE	5
 #define VEC5_AFFINITY		0x80
-	chosen = of_find_node_by_path("/chosen");
-	if (chosen) {
-		vec5 = of_get_property(chosen, "ibm,architecture-vec-5", NULL);
-		if (vec5 && (vec5[VEC5_AFFINITY_BYTE] & VEC5_AFFINITY)) {
-			dbg("Using form 1 affinity\n");
-			form1_affinity = 1;
+
+	if (firmware_has_feature(FW_FEATURE_OPAL))
+		form1_affinity = 1;
+	else {
+		chosen = of_find_node_by_path("/chosen");
+		if (chosen) {
+			vec5 = of_get_property(chosen,
+					       "ibm,architecture-vec-5", NULL);
+			if (vec5 && (vec5[VEC5_AFFINITY_BYTE] &
+							VEC5_AFFINITY)) {
+				dbg("Using form 1 affinity\n");
+				form1_affinity = 1;
+			}
 		}
 	}
 
diff --git a/arch/powerpc/platforms/52xx/mpc5200_simple.c b/arch/powerpc/platforms/52xx/mpc5200_simple.c
index e36d6e2..846b789 100644
--- a/arch/powerpc/platforms/52xx/mpc5200_simple.c
+++ b/arch/powerpc/platforms/52xx/mpc5200_simple.c
@@ -50,6 +50,7 @@
 
 /* list of the supported boards */
 static const char *board[] __initdata = {
+	"anon,charon",
 	"intercontrol,digsy-mtc",
 	"manroland,mucmc52",
 	"manroland,uc101",
diff --git a/arch/powerpc/platforms/85xx/Kconfig b/arch/powerpc/platforms/85xx/Kconfig
index 45023e2..d7946be2 100644
--- a/arch/powerpc/platforms/85xx/Kconfig
+++ b/arch/powerpc/platforms/85xx/Kconfig
@@ -203,7 +203,7 @@
 	select PPC_E500MC
 	select PHYS_64BIT
 	select SWIOTLB
-	select MPC8xxx_GPIO
+	select GPIO_MPC8XXX
 	select HAS_RAPIDIO
 	select PPC_EPAPR_HV_PIC
 	help
diff --git a/arch/powerpc/platforms/85xx/p3060_qds.c b/arch/powerpc/platforms/85xx/p3060_qds.c
index 01dcf44..081cf4a 100644
--- a/arch/powerpc/platforms/85xx/p3060_qds.c
+++ b/arch/powerpc/platforms/85xx/p3060_qds.c
@@ -70,7 +70,7 @@
 	.power_save		= e500_idle,
 };
 
-machine_device_initcall(p3060_qds, declare_of_platform_devices);
+machine_device_initcall(p3060_qds, corenet_ds_publish_devices);
 
 #ifdef CONFIG_SWIOTLB
 machine_arch_initcall(p3060_qds, swiotlb_setup_bus_notifier);
diff --git a/arch/powerpc/platforms/Kconfig b/arch/powerpc/platforms/Kconfig
index e458872..3fe6d92 100644
--- a/arch/powerpc/platforms/Kconfig
+++ b/arch/powerpc/platforms/Kconfig
@@ -347,7 +347,7 @@
 
 config MCU_MPC8349EMITX
 	bool "MPC8349E-mITX MCU driver"
-	depends on I2C && PPC_83xx
+	depends on I2C=y && PPC_83xx
 	select GENERIC_GPIO
 	select ARCH_REQUIRE_GPIOLIB
 	help
diff --git a/arch/powerpc/platforms/cell/beat.c b/arch/powerpc/platforms/cell/beat.c
index 232fc38..852592b 100644
--- a/arch/powerpc/platforms/cell/beat.c
+++ b/arch/powerpc/platforms/cell/beat.c
@@ -230,7 +230,7 @@
 		}
 		ev->virq = virq;
 
-		rc = request_irq(virq, ev->handler, IRQF_DISABLED,
+		rc = request_irq(virq, ev->handler, 0,
 				      ev->typecode, NULL);
 		if (rc != 0) {
 			printk(KERN_ERR "Beat: failed to request virtual IRQ"
diff --git a/arch/powerpc/platforms/cell/celleb_scc_pciex.c b/arch/powerpc/platforms/cell/celleb_scc_pciex.c
index ae790ac..14be2bd 100644
--- a/arch/powerpc/platforms/cell/celleb_scc_pciex.c
+++ b/arch/powerpc/platforms/cell/celleb_scc_pciex.c
@@ -514,7 +514,7 @@
 	virq = irq_create_of_mapping(oirq.controller, oirq.specifier,
 				     oirq.size);
 	if (request_irq(virq, pciex_handle_internal_irq,
-			IRQF_DISABLED, "pciex", (void *)phb)) {
+			0, "pciex", (void *)phb)) {
 		pr_err("PCIEXC:Failed to request irq\n");
 		goto error;
 	}
diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c
index fc46fca..592c3d5 100644
--- a/arch/powerpc/platforms/cell/iommu.c
+++ b/arch/powerpc/platforms/cell/iommu.c
@@ -412,8 +412,7 @@
 			IIC_IRQ_IOEX_ATI | (iommu->nid << IIC_IRQ_NODE_SHIFT));
 	BUG_ON(virq == NO_IRQ);
 
-	ret = request_irq(virq, ioc_interrupt, IRQF_DISABLED,
-			iommu->name, iommu);
+	ret = request_irq(virq, ioc_interrupt, 0, iommu->name, iommu);
 	BUG_ON(ret);
 
 	/* set the IOC segment table origin register (and turn on the iommu) */
diff --git a/arch/powerpc/platforms/cell/pmu.c b/arch/powerpc/platforms/cell/pmu.c
index 1acf360..59c1a16 100644
--- a/arch/powerpc/platforms/cell/pmu.c
+++ b/arch/powerpc/platforms/cell/pmu.c
@@ -392,7 +392,7 @@
 		}
 
 		rc = request_irq(irq, cbe_pm_irq,
-				 IRQF_DISABLED, "cbe-pmu-0", NULL);
+				 0, "cbe-pmu-0", NULL);
 		if (rc) {
 			printk("ERROR: Request for irq on node %d failed\n",
 			       node);
diff --git a/arch/powerpc/platforms/cell/spu_base.c b/arch/powerpc/platforms/cell/spu_base.c
index 3675da7..e94d3ec 100644
--- a/arch/powerpc/platforms/cell/spu_base.c
+++ b/arch/powerpc/platforms/cell/spu_base.c
@@ -442,8 +442,7 @@
 		snprintf(spu->irq_c0, sizeof (spu->irq_c0), "spe%02d.0",
 			 spu->number);
 		ret = request_irq(spu->irqs[0], spu_irq_class_0,
-				  IRQF_DISABLED,
-				  spu->irq_c0, spu);
+				  0, spu->irq_c0, spu);
 		if (ret)
 			goto bail0;
 	}
@@ -451,8 +450,7 @@
 		snprintf(spu->irq_c1, sizeof (spu->irq_c1), "spe%02d.1",
 			 spu->number);
 		ret = request_irq(spu->irqs[1], spu_irq_class_1,
-				  IRQF_DISABLED,
-				  spu->irq_c1, spu);
+				  0, spu->irq_c1, spu);
 		if (ret)
 			goto bail1;
 	}
@@ -460,8 +458,7 @@
 		snprintf(spu->irq_c2, sizeof (spu->irq_c2), "spe%02d.2",
 			 spu->number);
 		ret = request_irq(spu->irqs[2], spu_irq_class_2,
-				  IRQF_DISABLED,
-				  spu->irq_c2, spu);
+				  0, spu->irq_c2, spu);
 		if (ret)
 			goto bail2;
 	}
diff --git a/arch/powerpc/platforms/powermac/pic.c b/arch/powerpc/platforms/powermac/pic.c
index cb40e92..901bfbd 100644
--- a/arch/powerpc/platforms/powermac/pic.c
+++ b/arch/powerpc/platforms/powermac/pic.c
@@ -272,7 +272,6 @@
 
 static struct irqaction gatwick_cascade_action = {
 	.handler	= gatwick_action,
-	.flags		= IRQF_DISABLED,
 	.name		= "cascade",
 };
 
diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index 9a521dc..9b6a820 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -200,7 +200,7 @@
 
 	if (psurge_secondary_virq)
 		rc = request_irq(psurge_secondary_virq, psurge_ipi_intr,
-			IRQF_DISABLED|IRQF_PERCPU, "IPI", NULL);
+			IRQF_PERCPU, "IPI", NULL);
 
 	if (rc)
 		pr_err("Failed to setup secondary cpu IPI\n");
@@ -408,7 +408,7 @@
 
 static struct irqaction psurge_irqaction = {
 	.handler = psurge_ipi_intr,
-	.flags = IRQF_DISABLED|IRQF_PERCPU,
+	.flags = IRQF_PERCPU,
 	.name = "primary IPI",
 };
 
diff --git a/arch/powerpc/platforms/ps3/device-init.c b/arch/powerpc/platforms/ps3/device-init.c
index 6c4b583..3f175e8 100644
--- a/arch/powerpc/platforms/ps3/device-init.c
+++ b/arch/powerpc/platforms/ps3/device-init.c
@@ -825,7 +825,7 @@
 
 	spin_lock_init(&dev.lock);
 
-	res = request_irq(irq, ps3_notification_interrupt, IRQF_DISABLED,
+	res = request_irq(irq, ps3_notification_interrupt, 0,
 			  "ps3_notification", &dev);
 	if (res) {
 		pr_err("%s:%u: request_irq failed %d\n", __func__, __LINE__,
diff --git a/arch/powerpc/platforms/ps3/interrupt.c b/arch/powerpc/platforms/ps3/interrupt.c
index 404bc52..1d6f4f4 100644
--- a/arch/powerpc/platforms/ps3/interrupt.c
+++ b/arch/powerpc/platforms/ps3/interrupt.c
@@ -88,6 +88,7 @@
 	struct ps3_bmp bmp __attribute__ ((aligned (PS3_BMP_MINALIGN)));
 	u64 ppe_id;
 	u64 thread_id;
+	unsigned long ipi_mask;
 };
 
 static DEFINE_PER_CPU(struct ps3_private, ps3_private);
@@ -144,7 +145,11 @@
 static void ps3_chip_eoi(struct irq_data *d)
 {
 	const struct ps3_private *pd = irq_data_get_irq_chip_data(d);
-	lv1_end_of_interrupt_ext(pd->ppe_id, pd->thread_id, d->irq);
+
+	/* non-IPIs are EOIed here. */
+
+	if (!test_bit(63 - d->irq, &pd->ipi_mask))
+		lv1_end_of_interrupt_ext(pd->ppe_id, pd->thread_id, d->irq);
 }
 
 /**
@@ -691,6 +696,16 @@
 		cpu, virq, pd->bmp.ipi_debug_brk_mask);
 }
 
+void __init ps3_register_ipi_irq(unsigned int cpu, unsigned int virq)
+{
+	struct ps3_private *pd = &per_cpu(ps3_private, cpu);
+
+	set_bit(63 - virq, &pd->ipi_mask);
+
+	DBG("%s:%d: cpu %u, virq %u, ipi_mask %lxh\n", __func__, __LINE__,
+		cpu, virq, pd->ipi_mask);
+}
+
 static unsigned int ps3_get_irq(void)
 {
 	struct ps3_private *pd = &__get_cpu_var(ps3_private);
@@ -720,6 +735,12 @@
 		BUG();
 	}
 #endif
+
+	/* IPIs are EOIed here. */
+
+	if (test_bit(63 - plug, &pd->ipi_mask))
+		lv1_end_of_interrupt_ext(pd->ppe_id, pd->thread_id, plug);
+
 	return plug;
 }
 
diff --git a/arch/powerpc/platforms/ps3/platform.h b/arch/powerpc/platforms/ps3/platform.h
index 9a196a8..1a633ed 100644
--- a/arch/powerpc/platforms/ps3/platform.h
+++ b/arch/powerpc/platforms/ps3/platform.h
@@ -43,6 +43,7 @@
 void ps3_init_IRQ(void);
 void ps3_shutdown_IRQ(int cpu);
 void __init ps3_register_ipi_debug_brk(unsigned int cpu, unsigned int virq);
+void __init ps3_register_ipi_irq(unsigned int cpu, unsigned int virq);
 
 /* smp */
 
diff --git a/arch/powerpc/platforms/ps3/repository.c b/arch/powerpc/platforms/ps3/repository.c
index 5e304c2..ca40f6a 100644
--- a/arch/powerpc/platforms/ps3/repository.c
+++ b/arch/powerpc/platforms/ps3/repository.c
@@ -184,7 +184,7 @@
 	enum ps3_bus_type *bus_type)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_PME,
 		make_first_field("bus", bus_index),
@@ -199,7 +199,7 @@
 	unsigned int *num_dev)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_PME,
 		make_first_field("bus", bus_index),
@@ -239,7 +239,7 @@
 	unsigned int dev_index, enum ps3_dev_type *dev_type)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_PME,
 		make_first_field("bus", bus_index),
@@ -256,8 +256,8 @@
 	enum ps3_interrupt_type *intr_type, unsigned int *interrupt_id)
 {
 	int result;
-	u64 v1;
-	u64 v2;
+	u64 v1 = 0;
+	u64 v2 = 0;
 
 	result = read_node(PS3_LPAR_ID_PME,
 		make_first_field("bus", bus_index),
@@ -275,7 +275,7 @@
 	enum ps3_reg_type *reg_type)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_PME,
 		make_first_field("bus", bus_index),
@@ -615,7 +615,7 @@
 	unsigned int dev_index, unsigned int *num_regions)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_PME,
 		make_first_field("bus", bus_index),
@@ -631,7 +631,7 @@
 	unsigned int *region_id)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_PME,
 	    make_first_field("bus", bus_index),
@@ -786,7 +786,7 @@
 int ps3_repository_read_num_spu_reserved(unsigned int *num_spu_reserved)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_CURRENT,
 		make_first_field("bi", 0),
@@ -805,7 +805,7 @@
 int ps3_repository_read_num_spu_resource_id(unsigned int *num_resource_id)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_CURRENT,
 		make_first_field("bi", 0),
@@ -827,8 +827,8 @@
 	enum ps3_spu_resource_type *resource_type, unsigned int *resource_id)
 {
 	int result;
-	u64 v1;
-	u64 v2;
+	u64 v1 = 0;
+	u64 v2 = 0;
 
 	result = read_node(PS3_LPAR_ID_CURRENT,
 		make_first_field("bi", 0),
@@ -854,7 +854,7 @@
 int ps3_repository_read_boot_dat_size(unsigned int *size)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_CURRENT,
 		make_first_field("bi", 0),
@@ -869,7 +869,7 @@
 int ps3_repository_read_vuart_av_port(unsigned int *port)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_CURRENT,
 		make_first_field("bi", 0),
@@ -884,7 +884,7 @@
 int ps3_repository_read_vuart_sysmgr_port(unsigned int *port)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_CURRENT,
 		make_first_field("bi", 0),
@@ -919,7 +919,7 @@
 int ps3_repository_read_num_be(unsigned int *num_be)
 {
 	int result;
-	u64 v1;
+	u64 v1 = 0;
 
 	result = read_node(PS3_LPAR_ID_PME,
 		make_first_field("ben", 0),
diff --git a/arch/powerpc/platforms/ps3/smp.c b/arch/powerpc/platforms/ps3/smp.c
index 4c44794..efc1cd8 100644
--- a/arch/powerpc/platforms/ps3/smp.c
+++ b/arch/powerpc/platforms/ps3/smp.c
@@ -59,46 +59,49 @@
 
 static int ps3_smp_probe(void)
 {
-	return 2;
-}
+	int cpu;
 
-static void __init ps3_smp_setup_cpu(int cpu)
-{
-	int result;
-	unsigned int *virqs = per_cpu(ps3_ipi_virqs, cpu);
-	int i;
+	for (cpu = 0; cpu < 2; cpu++) {
+		int result;
+		unsigned int *virqs = per_cpu(ps3_ipi_virqs, cpu);
+		int i;
 
-	DBG(" -> %s:%d: (%d)\n", __func__, __LINE__, cpu);
+		DBG(" -> %s:%d: (%d)\n", __func__, __LINE__, cpu);
 
-	/*
-	 * Check assumptions on ps3_ipi_virqs[] indexing. If this
-	 * check fails, then a different mapping of PPC_MSG_
-	 * to index needs to be setup.
-	 */
+		/*
+		* Check assumptions on ps3_ipi_virqs[] indexing. If this
+		* check fails, then a different mapping of PPC_MSG_
+		* to index needs to be setup.
+		*/
 
-	BUILD_BUG_ON(PPC_MSG_CALL_FUNCTION    != 0);
-	BUILD_BUG_ON(PPC_MSG_RESCHEDULE       != 1);
-	BUILD_BUG_ON(PPC_MSG_CALL_FUNC_SINGLE != 2);
-	BUILD_BUG_ON(PPC_MSG_DEBUGGER_BREAK   != 3);
+		BUILD_BUG_ON(PPC_MSG_CALL_FUNCTION    != 0);
+		BUILD_BUG_ON(PPC_MSG_RESCHEDULE       != 1);
+		BUILD_BUG_ON(PPC_MSG_CALL_FUNC_SINGLE != 2);
+		BUILD_BUG_ON(PPC_MSG_DEBUGGER_BREAK   != 3);
 
-	for (i = 0; i < MSG_COUNT; i++) {
-		result = ps3_event_receive_port_setup(cpu, &virqs[i]);
+		for (i = 0; i < MSG_COUNT; i++) {
+			result = ps3_event_receive_port_setup(cpu, &virqs[i]);
 
-		if (result)
-			continue;
+			if (result)
+				continue;
 
-		DBG("%s:%d: (%d, %d) => virq %u\n",
-			__func__, __LINE__, cpu, i, virqs[i]);
+			DBG("%s:%d: (%d, %d) => virq %u\n",
+				__func__, __LINE__, cpu, i, virqs[i]);
 
-		result = smp_request_message_ipi(virqs[i], i);
+			result = smp_request_message_ipi(virqs[i], i);
 
-		if (result)
-			virqs[i] = NO_IRQ;
+			if (result)
+				virqs[i] = NO_IRQ;
+			else
+				ps3_register_ipi_irq(cpu, virqs[i]);
+		}
+
+		ps3_register_ipi_debug_brk(cpu, virqs[PPC_MSG_DEBUGGER_BREAK]);
+
+		DBG(" <- %s:%d: (%d)\n", __func__, __LINE__, cpu);
 	}
 
-	ps3_register_ipi_debug_brk(cpu, virqs[PPC_MSG_DEBUGGER_BREAK]);
-
-	DBG(" <- %s:%d: (%d)\n", __func__, __LINE__, cpu);
+	return 2;
 }
 
 void ps3_smp_cleanup_cpu(int cpu)
@@ -121,7 +124,6 @@
 	.probe		= ps3_smp_probe,
 	.message_pass	= ps3_smp_message_pass,
 	.kick_cpu	= smp_generic_kick_cpu,
-	.setup_cpu	= ps3_smp_setup_cpu,
 };
 
 void smp_init_ps3(void)
diff --git a/arch/powerpc/sysdev/ehv_pic.c b/arch/powerpc/sysdev/ehv_pic.c
index af1a5df..b6731e4 100644
--- a/arch/powerpc/sysdev/ehv_pic.c
+++ b/arch/powerpc/sysdev/ehv_pic.c
@@ -280,6 +280,7 @@
 
 	if (!ehv_pic->irqhost) {
 		of_node_put(np);
+		kfree(ehv_pic);
 		return;
 	}
 
diff --git a/arch/powerpc/sysdev/fsl_lbc.c b/arch/powerpc/sysdev/fsl_lbc.c
index c4d96fa3..d5c3c90 100644
--- a/arch/powerpc/sysdev/fsl_lbc.c
+++ b/arch/powerpc/sysdev/fsl_lbc.c
@@ -328,6 +328,7 @@
 err:
 	iounmap(fsl_lbc_ctrl_dev->regs);
 	kfree(fsl_lbc_ctrl_dev);
+	fsl_lbc_ctrl_dev = NULL;
 	return ret;
 }
 
diff --git a/arch/powerpc/sysdev/fsl_rio.c b/arch/powerpc/sysdev/fsl_rio.c
index de170fd..22ffccd 100644
--- a/arch/powerpc/sysdev/fsl_rio.c
+++ b/arch/powerpc/sysdev/fsl_rio.c
@@ -23,7 +23,7 @@
  */
 
 #include <linux/init.h>
-#include <linux/export.h>
+#include <linux/module.h>
 #include <linux/types.h>
 #include <linux/dma-mapping.h>
 #include <linux/interrupt.h>
diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index 0842c6f..8c7e852 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -800,8 +800,6 @@
 	 * IPIs are marked IRQ_PER_CPU. This has the side effect of
 	 * preventing the IRQ_PENDING/IRQ_INPROGRESS logic from
 	 * applying to them. We EOI them late to avoid re-entering.
-	 * We mark IPI's with IRQF_DISABLED as they must run with
-	 * irqs disabled.
 	 */
 	mpic_eoi(mpic);
 }
diff --git a/arch/powerpc/sysdev/ppc4xx_soc.c b/arch/powerpc/sysdev/ppc4xx_soc.c
index d3d6ce3..0debcc3 100644
--- a/arch/powerpc/sysdev/ppc4xx_soc.c
+++ b/arch/powerpc/sysdev/ppc4xx_soc.c
@@ -115,7 +115,7 @@
 	}
 
 	/* Install error handler */
-	if (request_irq(irq, l2c_error_handler, IRQF_DISABLED, "L2C", 0) < 0) {
+	if (request_irq(irq, l2c_error_handler, 0, "L2C", 0) < 0) {
 		printk(KERN_ERR "Cannot install L2C error handler"
 		       ", cache is not enabled\n");
 		of_node_put(np);
diff --git a/arch/powerpc/sysdev/qe_lib/qe.c b/arch/powerpc/sysdev/qe_lib/qe.c
index 3363fbc..ceb09cb 100644
--- a/arch/powerpc/sysdev/qe_lib/qe.c
+++ b/arch/powerpc/sysdev/qe_lib/qe.c
@@ -216,7 +216,7 @@
 	/* Errata QE_General4, which affects some MPC832x and MPC836x SOCs, says
 	   that the BRG divisor must be even if you're not using divide-by-16
 	   mode. */
-	if (!div16 && (divisor & 1))
+	if (!div16 && (divisor & 1) && (divisor > 3))
 		divisor++;
 
 	tempval = ((divisor - 1) << QE_BRGC_DIVISOR_SHIFT) |
diff --git a/arch/powerpc/sysdev/xics/xics-common.c b/arch/powerpc/sysdev/xics/xics-common.c
index 3d93a8d..63762c6 100644
--- a/arch/powerpc/sysdev/xics/xics-common.c
+++ b/arch/powerpc/sysdev/xics/xics-common.c
@@ -134,11 +134,10 @@
 	BUG_ON(ipi == NO_IRQ);
 
 	/*
-	 * IPIs are marked IRQF_DISABLED as they must run with irqs
-	 * disabled, and PERCPU.  The handler was set in map.
+	 * IPIs are marked IRQF_PERCPU. The handler was set in map.
 	 */
 	BUG_ON(request_irq(ipi, icp_ops->ipi_action,
-			   IRQF_DISABLED|IRQF_PERCPU, "IPI", NULL));
+			   IRQF_PERCPU, "IPI", NULL));
 }
 
 int __init xics_smp_probe(void)
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index a9fbd43..373679b 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -572,6 +572,7 @@
 config CRASH_DUMP
 	bool "kernel crash dumps"
 	depends on 64BIT
+	select KEXEC
 	help
 	  Generate crash dump after being started by kexec.
 	  Crash dump kernels are loaded in the main kernel with kexec-tools
diff --git a/arch/s390/crypto/crypt_s390.h b/arch/s390/crypto/crypt_s390.h
index 4967677..ffd1ac2 100644
--- a/arch/s390/crypto/crypt_s390.h
+++ b/arch/s390/crypto/crypt_s390.h
@@ -368,9 +368,12 @@
 
 	if (facility_mask & CRYPT_S390_MSA && !test_facility(17))
 		return 0;
-	if (facility_mask & CRYPT_S390_MSA3 && !test_facility(76))
+
+	if (facility_mask & CRYPT_S390_MSA3 &&
+	    (!test_facility(2) || !test_facility(76)))
 		return 0;
-	if (facility_mask & CRYPT_S390_MSA4 && !test_facility(77))
+	if (facility_mask & CRYPT_S390_MSA4 &&
+	    (!test_facility(2) || !test_facility(77)))
 		return 0;
 
 	switch (func & CRYPT_S390_OP_MASK) {
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 24e1847..b0c235c 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -47,7 +47,7 @@
 #define KVM_HPAGE_MASK(x)	(~(KVM_HPAGE_SIZE(x) - 1))
 #define KVM_PAGES_PER_HPAGE(x)	(KVM_HPAGE_SIZE(x) / PAGE_SIZE)
 
-#define CPUSTAT_HOST       0x80000000
+#define CPUSTAT_STOPPED    0x80000000
 #define CPUSTAT_WAIT       0x10000000
 #define CPUSTAT_ECALL_PEND 0x08000000
 #define CPUSTAT_STOP_INT   0x04000000
@@ -139,6 +139,7 @@
 	u32 instruction_stfl;
 	u32 instruction_tprot;
 	u32 instruction_sigp_sense;
+	u32 instruction_sigp_sense_running;
 	u32 instruction_sigp_external_call;
 	u32 instruction_sigp_emergency;
 	u32 instruction_sigp_stop;
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 34ede0e..4f289ff 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -593,14 +593,16 @@
 	unsigned long address, bits;
 	unsigned char skey;
 
+	if (!pte_present(*ptep))
+		return pgste;
 	address = pte_val(*ptep) & PAGE_MASK;
 	skey = page_get_storage_key(address);
 	bits = skey & (_PAGE_CHANGED | _PAGE_REFERENCED);
 	/* Clear page changed & referenced bit in the storage key */
-	if (bits) {
-		skey ^= bits;
-		page_set_storage_key(address, skey, 1);
-	}
+	if (bits & _PAGE_CHANGED)
+		page_set_storage_key(address, skey ^ bits, 1);
+	else if (bits)
+		page_reset_referenced(address);
 	/* Transfer page changed & referenced bit to guest bits in pgste */
 	pgste_val(pgste) |= bits << 48;		/* RCP_GR_BIT & RCP_GC_BIT */
 	/* Get host changed & referenced bits from pgste */
@@ -625,6 +627,8 @@
 #ifdef CONFIG_PGSTE
 	int young;
 
+	if (!pte_present(*ptep))
+		return pgste;
 	young = page_reset_referenced(pte_val(*ptep) & PAGE_MASK);
 	/* Transfer page referenced bit to pte software bit (host view) */
 	if (young || (pgste_val(pgste) & RCP_HR_BIT))
@@ -638,13 +642,15 @@
 
 }
 
-static inline void pgste_set_pte(pte_t *ptep, pgste_t pgste)
+static inline void pgste_set_pte(pte_t *ptep, pgste_t pgste, pte_t entry)
 {
 #ifdef CONFIG_PGSTE
 	unsigned long address;
 	unsigned long okey, nkey;
 
-	address = pte_val(*ptep) & PAGE_MASK;
+	if (!pte_present(entry))
+		return;
+	address = pte_val(entry) & PAGE_MASK;
 	okey = nkey = page_get_storage_key(address);
 	nkey &= ~(_PAGE_ACC_BITS | _PAGE_FP_BIT);
 	/* Set page access key and fetch protection bit from pgste */
@@ -712,7 +718,7 @@
 
 	if (mm_has_pgste(mm)) {
 		pgste = pgste_get_lock(ptep);
-		pgste_set_pte(ptep, pgste);
+		pgste_set_pte(ptep, pgste, entry);
 		*ptep = entry;
 		pgste_set_unlock(ptep, pgste);
 	} else
diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h
index 5a09971..097183c 100644
--- a/arch/s390/include/asm/setup.h
+++ b/arch/s390/include/asm/setup.h
@@ -82,6 +82,7 @@
 #define MACHINE_FLAG_LPAR	(1UL << 12)
 #define MACHINE_FLAG_SPP	(1UL << 13)
 #define MACHINE_FLAG_TOPOLOGY	(1UL << 14)
+#define MACHINE_FLAG_STCKF	(1UL << 15)
 
 #define MACHINE_IS_VM		(S390_lowcore.machine_flags & MACHINE_FLAG_VM)
 #define MACHINE_IS_KVM		(S390_lowcore.machine_flags & MACHINE_FLAG_KVM)
@@ -100,6 +101,7 @@
 #define MACHINE_HAS_PFMF	(0)
 #define MACHINE_HAS_SPP		(0)
 #define MACHINE_HAS_TOPOLOGY	(0)
+#define MACHINE_HAS_STCKF	(0)
 #else /* __s390x__ */
 #define MACHINE_HAS_IEEE	(1)
 #define MACHINE_HAS_CSP		(1)
@@ -111,6 +113,7 @@
 #define MACHINE_HAS_PFMF	(S390_lowcore.machine_flags & MACHINE_FLAG_PFMF)
 #define MACHINE_HAS_SPP		(S390_lowcore.machine_flags & MACHINE_FLAG_SPP)
 #define MACHINE_HAS_TOPOLOGY	(S390_lowcore.machine_flags & MACHINE_FLAG_TOPOLOGY)
+#define MACHINE_HAS_STCKF	(S390_lowcore.machine_flags & MACHINE_FLAG_STCKF)
 #endif /* __s390x__ */
 
 #define ZFCPDUMP_HSA_SIZE	(32UL<<20)
diff --git a/arch/s390/include/asm/timex.h b/arch/s390/include/asm/timex.h
index d610bef..c447a27 100644
--- a/arch/s390/include/asm/timex.h
+++ b/arch/s390/include/asm/timex.h
@@ -90,7 +90,7 @@
 {
 	unsigned long long clk;
 
-	if (test_facility(25))
+	if (MACHINE_HAS_STCKF)
 		asm volatile(".insn	s,0xb27c0000,%0" : "=Q" (clk) : : "cc");
 	else
 		clk = get_clock();
diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h
index 404bdb96..58de4c9 100644
--- a/arch/s390/include/asm/unistd.h
+++ b/arch/s390/include/asm/unistd.h
@@ -277,7 +277,9 @@
 #define __NR_clock_adjtime	337
 #define __NR_syncfs		338
 #define __NR_setns		339
-#define NR_syscalls 340
+#define __NR_process_vm_readv	340
+#define __NR_process_vm_writev	341
+#define NR_syscalls 342
 
 /* 
  * There are some system calls that are not present on 64 bit, some
diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index 5006a1d..18c51df 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -1627,3 +1627,23 @@
 	lgfr	%r2,%r2			# int
 	lgfr	%r3,%r3			# int
 	jg	sys_setns
+
+ENTRY(compat_sys_process_vm_readv_wrapper)
+	lgfr	%r2,%r2			# compat_pid_t
+	llgtr	%r3,%r3			# struct compat_iovec __user *
+	llgfr	%r4,%r4			# unsigned long
+	llgtr	%r5,%r5			# struct compat_iovec __user *
+	llgfr	%r6,%r6			# unsigned long
+	llgf	%r0,164(%r15)		# unsigned long
+	stg	%r0,160(%r15)
+	jg	sys_process_vm_readv
+
+ENTRY(compat_sys_process_vm_writev_wrapper)
+	lgfr	%r2,%r2			# compat_pid_t
+	llgtr	%r3,%r3			# struct compat_iovec __user *
+	llgfr	%r4,%r4			# unsigned long
+	llgtr	%r5,%r5			# struct compat_iovec __user *
+	llgfr	%r6,%r6			# unsigned long
+	llgf	%r0,164(%r15)		# unsigned long
+	stg	%r0,160(%r15)
+	jg	sys_process_vm_writev
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index 37394b3..c9ffe00 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -390,6 +390,8 @@
 		S390_lowcore.machine_flags |= MACHINE_FLAG_MVCOS;
 	if (test_facility(40))
 		S390_lowcore.machine_flags |= MACHINE_FLAG_SPP;
+	if (test_facility(25))
+		S390_lowcore.machine_flags |= MACHINE_FLAG_STCKF;
 #endif
 }
 
diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index 450931a..573bc29 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -296,13 +296,6 @@
 		     ((data & PSW_MASK_EA) && !(data & PSW_MASK_BA))))
 			/* Invalid psw mask. */
 			return -EINVAL;
-		if (addr == (addr_t) &dummy->regs.psw.addr)
-			/*
-			 * The debugger changed the instruction address,
-			 * reset system call restart, see signal.c:do_signal
-			 */
-			task_thread_info(child)->system_call = 0;
-
 		*(addr_t *)((addr_t) &task_pt_regs(child)->psw + addr) = data;
 
 	} else if (addr < (addr_t) (&dummy->regs.orig_gpr2)) {
@@ -614,11 +607,6 @@
 			/* Transfer 31 bit amode bit to psw mask. */
 			regs->psw.mask = (regs->psw.mask & ~PSW_MASK_BA) |
 				(__u64)(tmp & PSW32_ADDR_AMODE);
-			/*
-			 * The debugger changed the instruction address,
-			 * reset system call restart, see signal.c:do_signal
-			 */
-			task_thread_info(child)->system_call = 0;
 		} else {
 			/* gpr 0-15 */
 			*(__u32*)((addr_t) &regs->psw + addr*2 + 4) = tmp;
@@ -905,6 +893,14 @@
 	return 0;
 }
 
+static int s390_last_break_set(struct task_struct *target,
+			       const struct user_regset *regset,
+			       unsigned int pos, unsigned int count,
+			       const void *kbuf, const void __user *ubuf)
+{
+	return 0;
+}
+
 #endif
 
 static int s390_system_call_get(struct task_struct *target,
@@ -951,6 +947,7 @@
 		.size = sizeof(long),
 		.align = sizeof(long),
 		.get = s390_last_break_get,
+		.set = s390_last_break_set,
 	},
 #endif
 	[REGSET_SYSTEM_CALL] = {
@@ -1116,6 +1113,14 @@
 	return 0;
 }
 
+static int s390_compat_last_break_set(struct task_struct *target,
+				      const struct user_regset *regset,
+				      unsigned int pos, unsigned int count,
+				      const void *kbuf, const void __user *ubuf)
+{
+	return 0;
+}
+
 static const struct user_regset s390_compat_regsets[] = {
 	[REGSET_GENERAL] = {
 		.core_note_type = NT_PRSTATUS,
@@ -1139,6 +1144,7 @@
 		.size = sizeof(long),
 		.align = sizeof(long),
 		.get = s390_compat_last_break_get,
+		.set = s390_compat_last_break_set,
 	},
 	[REGSET_SYSTEM_CALL] = {
 		.core_note_type = NT_S390_SYSTEM_CALL,
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 8ac6bfa..e54c4ff 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -211,6 +211,8 @@
 
 	if (ipl_info.type != IPL_TYPE_FCP_DUMP)
 		return;
+	if (OLDMEM_BASE)
+		return;
 	if (console_devno != -1)
 		sprintf(str, " cio_ignore=all,!0.0.%04x,!0.0.%04x",
 			ipl_info.data.fcp.dev_id.devno, console_devno);
@@ -482,7 +484,7 @@
 
 
 #ifdef CONFIG_ZFCPDUMP
-	if (ipl_info.type == IPL_TYPE_FCP_DUMP) {
+	if (ipl_info.type == IPL_TYPE_FCP_DUMP && !OLDMEM_BASE) {
 		memory_end = ZFCPDUMP_HSA_SIZE;
 		memory_end_set = 1;
 	}
@@ -577,7 +579,7 @@
 		*msg = "first memory chunk must be at least crashkernel size";
 		return 0;
 	}
-	if (is_kdump_kernel() && (crash_size == OLDMEM_SIZE))
+	if (OLDMEM_BASE && crash_size == OLDMEM_SIZE)
 		return OLDMEM_BASE;
 
 	for (i = MEMORY_CHUNKS - 1; i >= 0; i--) {
diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c
index 05a85bc..7f6f9f3 100644
--- a/arch/s390/kernel/signal.c
+++ b/arch/s390/kernel/signal.c
@@ -460,9 +460,9 @@
 						     regs->svc_code >> 16);
 				break;
 			}
-			/* No longer in a system call */
-			clear_thread_flag(TIF_SYSCALL);
 		}
+		/* No longer in a system call */
+		clear_thread_flag(TIF_SYSCALL);
 
 		if ((is_compat_task() ?
 		     handle_signal32(signr, &ka, &info, oldset, regs) :
@@ -486,6 +486,7 @@
 	}
 
 	/* No handlers present - check for system call restart */
+	clear_thread_flag(TIF_SYSCALL);
 	if (current_thread_info()->system_call) {
 		regs->svc_code = current_thread_info()->system_call;
 		switch (regs->gprs[2]) {
@@ -500,9 +501,6 @@
 			regs->gprs[2] = regs->orig_gpr2;
 			set_thread_flag(TIF_SYSCALL);
 			break;
-		default:
-			clear_thread_flag(TIF_SYSCALL);
-			break;
 		}
 	}
 
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 73eb08c..bcab2f0 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -348,3 +348,5 @@
 SYSCALL(sys_clock_adjtime,sys_clock_adjtime,compat_sys_clock_adjtime_wrapper)
 SYSCALL(sys_syncfs,sys_syncfs,sys_syncfs_wrapper)
 SYSCALL(sys_setns,sys_setns,sys_setns_wrapper)
+SYSCALL(sys_process_vm_readv,sys_process_vm_readv,compat_sys_process_vm_readv_wrapper) /* 340 */
+SYSCALL(sys_process_vm_writev,sys_process_vm_writev,compat_sys_process_vm_writev_wrapper)
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index 77b8942..fdb5b8c 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -68,8 +68,10 @@
 	return mask;
 }
 
-static void add_cpus_to_mask(struct topology_cpu *tl_cpu,
-			     struct mask_info *book, struct mask_info *core)
+static struct mask_info *add_cpus_to_mask(struct topology_cpu *tl_cpu,
+					  struct mask_info *book,
+					  struct mask_info *core,
+					  int z10)
 {
 	unsigned int cpu;
 
@@ -88,10 +90,16 @@
 			cpu_book_id[lcpu] = book->id;
 #endif
 			cpumask_set_cpu(lcpu, &core->mask);
-			cpu_core_id[lcpu] = core->id;
+			if (z10) {
+				cpu_core_id[lcpu] = rcpu;
+				core = core->next;
+			} else {
+				cpu_core_id[lcpu] = core->id;
+			}
 			smp_cpu_polarization[lcpu] = tl_cpu->pp;
 		}
 	}
+	return core;
 }
 
 static void clear_masks(void)
@@ -123,18 +131,41 @@
 {
 #ifdef CONFIG_SCHED_BOOK
 	struct mask_info *book = &book_info;
+	struct cpuid cpu_id;
 #else
 	struct mask_info *book = NULL;
 #endif
 	struct mask_info *core = &core_info;
 	union topology_entry *tle, *end;
+	int z10 = 0;
 
-
+#ifdef CONFIG_SCHED_BOOK
+	get_cpu_id(&cpu_id);
+	z10 = cpu_id.machine == 0x2097 || cpu_id.machine == 0x2098;
+#endif
 	spin_lock_irq(&topology_lock);
 	clear_masks();
 	tle = info->tle;
 	end = (union topology_entry *)((unsigned long)info + info->length);
 	while (tle < end) {
+#ifdef CONFIG_SCHED_BOOK
+		if (z10) {
+			switch (tle->nl) {
+			case 1:
+				book = book->next;
+				book->id = tle->container.id;
+				break;
+			case 0:
+				core = add_cpus_to_mask(&tle->cpu, book, core, z10);
+				break;
+			default:
+				clear_masks();
+				goto out;
+			}
+			tle = next_tle(tle);
+			continue;
+		}
+#endif
 		switch (tle->nl) {
 #ifdef CONFIG_SCHED_BOOK
 		case 2:
@@ -147,7 +178,7 @@
 			core->id = tle->container.id;
 			break;
 		case 0:
-			add_cpus_to_mask(&tle->cpu, book, core);
+			add_cpus_to_mask(&tle->cpu, book, core, z10);
 			break;
 		default:
 			clear_masks();
@@ -328,8 +359,8 @@
 	for (i = 0; i < TOPOLOGY_NR_MAG; i++)
 		printk(" %d", info->mag[i]);
 	printk(" / %d\n", info->mnest);
-	alloc_masks(info, &core_info, 2);
+	alloc_masks(info, &core_info, 1);
 #ifdef CONFIG_SCHED_BOOK
-	alloc_masks(info, &book_info, 3);
+	alloc_masks(info, &book_info, 2);
 #endif
 }
diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S
index 56fe6bc..e4c79eb 100644
--- a/arch/s390/kernel/vmlinux.lds.S
+++ b/arch/s390/kernel/vmlinux.lds.S
@@ -43,6 +43,8 @@
 
 	NOTES :text :note
 
+	.dummy : { *(.dummy) } :data
+
 	RODATA
 
 #ifdef CONFIG_SHARED_KERNEL
diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
index 87cedd6..8943e82 100644
--- a/arch/s390/kvm/diag.c
+++ b/arch/s390/kvm/diag.c
@@ -70,7 +70,7 @@
 		return -EOPNOTSUPP;
 	}
 
-	atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
+	atomic_set_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
 	vcpu->run->s390_reset_flags |= KVM_S390_RESET_SUBSYSTEM;
 	vcpu->run->s390_reset_flags |= KVM_S390_RESET_IPL;
 	vcpu->run->s390_reset_flags |= KVM_S390_RESET_CPU_INIT;
diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index c7c5189..0243454 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -132,7 +132,6 @@
 	int rc = 0;
 
 	vcpu->stat.exit_stop_request++;
-	atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
 	spin_lock_bh(&vcpu->arch.local_int.lock);
 	if (vcpu->arch.local_int.action_bits & ACTION_STORE_ON_STOP) {
 		vcpu->arch.local_int.action_bits &= ~ACTION_STORE_ON_STOP;
@@ -149,6 +148,8 @@
 	}
 
 	if (vcpu->arch.local_int.action_bits & ACTION_STOP_ON_STOP) {
+		atomic_set_mask(CPUSTAT_STOPPED,
+				&vcpu->arch.sie_block->cpuflags);
 		vcpu->arch.local_int.action_bits &= ~ACTION_STOP_ON_STOP;
 		VCPU_EVENT(vcpu, 3, "%s", "cpu stopped");
 		rc = -EOPNOTSUPP;
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index 87c1670..278ee00 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -252,6 +252,7 @@
 			offsetof(struct _lowcore, restart_psw), sizeof(psw_t));
 		if (rc == -EFAULT)
 			exception = 1;
+		atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
 		break;
 
 	case KVM_S390_PROGRAM_INT:
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 0bd3bea..d1c44573 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -65,6 +65,7 @@
 	{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
 	{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
 	{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
+	{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
 	{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
 	{ "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) },
 	{ "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) },
@@ -127,6 +128,7 @@
 	switch (ext) {
 	case KVM_CAP_S390_PSW:
 	case KVM_CAP_S390_GMAP:
+	case KVM_CAP_SYNC_MMU:
 		r = 1;
 		break;
 	default:
@@ -270,10 +272,12 @@
 	restore_fp_regs(&vcpu->arch.guest_fpregs);
 	restore_access_regs(vcpu->arch.guest_acrs);
 	gmap_enable(vcpu->arch.gmap);
+	atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
 }
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
+	atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
 	gmap_disable(vcpu->arch.gmap);
 	save_fp_regs(&vcpu->arch.guest_fpregs);
 	save_access_regs(vcpu->arch.guest_acrs);
@@ -301,7 +305,9 @@
 
 int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 {
-	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH | CPUSTAT_SM);
+	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH |
+						    CPUSTAT_SM |
+						    CPUSTAT_STOPPED);
 	vcpu->arch.sie_block->ecb   = 6;
 	vcpu->arch.sie_block->eca   = 0xC1002001U;
 	vcpu->arch.sie_block->fac   = (int) (long) facilities;
@@ -428,7 +434,7 @@
 {
 	int rc = 0;
 
-	if (atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_RUNNING)
+	if (!(atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_STOPPED))
 		rc = -EBUSY;
 	else {
 		vcpu->run->psw_mask = psw.mask;
@@ -501,7 +507,7 @@
 	if (vcpu->sigset_active)
 		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);
 
-	atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
+	atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
 
 	BUG_ON(vcpu->kvm->arch.float_int.local_int[vcpu->vcpu_id] == NULL);
 
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index 3916263..d026389 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -336,6 +336,7 @@
 	u64 address1 = disp1 + base1 ? vcpu->arch.guest_gprs[base1] : 0;
 	u64 address2 = disp2 + base2 ? vcpu->arch.guest_gprs[base2] : 0;
 	struct vm_area_struct *vma;
+	unsigned long user_address;
 
 	vcpu->stat.instruction_tprot++;
 
@@ -349,9 +350,14 @@
 		return -EOPNOTSUPP;
 
 
+	/* we must resolve the address without holding the mmap semaphore.
+	 * This is ok since the userspace hypervisor is not supposed to change
+	 * the mapping while the guest queries the memory. Otherwise the guest
+	 * might crash or get wrong info anyway. */
+	user_address = (unsigned long) __guestaddr_to_user(vcpu, address1);
+
 	down_read(&current->mm->mmap_sem);
-	vma = find_vma(current->mm,
-			(unsigned long) __guestaddr_to_user(vcpu, address1));
+	vma = find_vma(current->mm, user_address);
 	if (!vma) {
 		up_read(&current->mm->mmap_sem);
 		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c
index f815118..0a7941d 100644
--- a/arch/s390/kvm/sigp.c
+++ b/arch/s390/kvm/sigp.c
@@ -31,9 +31,11 @@
 #define SIGP_SET_PREFIX        0x0d
 #define SIGP_STORE_STATUS_ADDR 0x0e
 #define SIGP_SET_ARCH          0x12
+#define SIGP_SENSE_RUNNING     0x15
 
 /* cpu status bits */
 #define SIGP_STAT_EQUIPMENT_CHECK   0x80000000UL
+#define SIGP_STAT_NOT_RUNNING	    0x00000400UL
 #define SIGP_STAT_INCORRECT_STATE   0x00000200UL
 #define SIGP_STAT_INVALID_PARAMETER 0x00000100UL
 #define SIGP_STAT_EXT_CALL_PENDING  0x00000080UL
@@ -57,8 +59,8 @@
 	spin_lock(&fi->lock);
 	if (fi->local_int[cpu_addr] == NULL)
 		rc = 3; /* not operational */
-	else if (atomic_read(fi->local_int[cpu_addr]->cpuflags)
-		 & CPUSTAT_RUNNING) {
+	else if (!(atomic_read(fi->local_int[cpu_addr]->cpuflags)
+		  & CPUSTAT_STOPPED)) {
 		*reg &= 0xffffffff00000000UL;
 		rc = 1; /* status stored */
 	} else {
@@ -251,7 +253,7 @@
 
 	spin_lock_bh(&li->lock);
 	/* cpu must be in stopped state */
-	if (atomic_read(li->cpuflags) & CPUSTAT_RUNNING) {
+	if (!(atomic_read(li->cpuflags) & CPUSTAT_STOPPED)) {
 		rc = 1; /* incorrect state */
 		*reg &= SIGP_STAT_INCORRECT_STATE;
 		kfree(inti);
@@ -275,6 +277,38 @@
 	return rc;
 }
 
+static int __sigp_sense_running(struct kvm_vcpu *vcpu, u16 cpu_addr,
+				unsigned long *reg)
+{
+	int rc;
+	struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int;
+
+	if (cpu_addr >= KVM_MAX_VCPUS)
+		return 3; /* not operational */
+
+	spin_lock(&fi->lock);
+	if (fi->local_int[cpu_addr] == NULL)
+		rc = 3; /* not operational */
+	else {
+		if (atomic_read(fi->local_int[cpu_addr]->cpuflags)
+		    & CPUSTAT_RUNNING) {
+			/* running */
+			rc = 1;
+		} else {
+			/* not running */
+			*reg &= 0xffffffff00000000UL;
+			*reg |= SIGP_STAT_NOT_RUNNING;
+			rc = 0;
+		}
+	}
+	spin_unlock(&fi->lock);
+
+	VCPU_EVENT(vcpu, 4, "sensed running status of cpu %x rc %x", cpu_addr,
+		   rc);
+
+	return rc;
+}
+
 int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu)
 {
 	int r1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4;
@@ -331,6 +365,11 @@
 		rc = __sigp_set_prefix(vcpu, cpu_addr, parameter,
 				       &vcpu->arch.guest_gprs[r1]);
 		break;
+	case SIGP_SENSE_RUNNING:
+		vcpu->stat.instruction_sigp_sense_running++;
+		rc = __sigp_sense_running(vcpu, cpu_addr,
+					  &vcpu->arch.guest_gprs[r1]);
+		break;
 	case SIGP_RESTART:
 		vcpu->stat.instruction_sigp_restart++;
 		/* user space must know about restart */
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index 1766def..a9a3018 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -587,8 +587,13 @@
 		} else {
 			/* Completion interrupt was faster than initial
 			 * interrupt. Set pfault_wait to -1 so the initial
-			 * interrupt doesn't put the task to sleep. */
-			tsk->thread.pfault_wait = -1;
+			 * interrupt doesn't put the task to sleep.
+			 * If the task is not running, ignore the completion
+			 * interrupt since it must be a leftover of a PFAULT
+			 * CANCEL operation which didn't remove all pending
+			 * completion interrupts. */
+			if (tsk->state == TASK_RUNNING)
+				tsk->thread.pfault_wait = -1;
 		}
 		put_task_struct(tsk);
 	} else {
diff --git a/arch/sh/include/asm/page.h b/arch/sh/include/asm/page.h
index 0dca9a5..15d9703 100644
--- a/arch/sh/include/asm/page.h
+++ b/arch/sh/include/asm/page.h
@@ -151,8 +151,13 @@
 #endif /* !__ASSEMBLY__ */
 
 #ifdef CONFIG_UNCACHED_MAPPING
+#if defined(CONFIG_29BIT)
+#define UNCAC_ADDR(addr)	P2SEGADDR(addr)
+#define CAC_ADDR(addr)		P1SEGADDR(addr)
+#else
 #define UNCAC_ADDR(addr)	((addr) - PAGE_OFFSET + uncached_start)
 #define CAC_ADDR(addr)		((addr) - uncached_start + PAGE_OFFSET)
+#endif
 #else
 #define UNCAC_ADDR(addr)	((addr))
 #define CAC_ADDR(addr)		((addr))
diff --git a/arch/sh/include/asm/unistd_32.h b/arch/sh/include/asm/unistd_32.h
index 3432008..152b862 100644
--- a/arch/sh/include/asm/unistd_32.h
+++ b/arch/sh/include/asm/unistd_32.h
@@ -375,8 +375,10 @@
 #define __NR_syncfs		362
 #define __NR_sendmmsg		363
 #define __NR_setns		364
+#define __NR_process_vm_readv	365
+#define __NR_process_vm_writev	366
 
-#define NR_syscalls 365
+#define NR_syscalls 367
 
 #ifdef __KERNEL__
 
diff --git a/arch/sh/include/asm/unistd_64.h b/arch/sh/include/asm/unistd_64.h
index ec98986..c330c23 100644
--- a/arch/sh/include/asm/unistd_64.h
+++ b/arch/sh/include/asm/unistd_64.h
@@ -396,10 +396,12 @@
 #define __NR_syncfs		373
 #define __NR_sendmmsg		374
 #define __NR_setns		375
+#define __NR_process_vm_readv	376
+#define __NR_process_vm_writev	377
 
 #ifdef __KERNEL__
 
-#define NR_syscalls 376
+#define NR_syscalls 378
 
 #define __ARCH_WANT_IPC_PARSE_VERSION
 #define __ARCH_WANT_OLD_READDIR
diff --git a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
index a43124e..0bd744f 100644
--- a/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
+++ b/arch/sh/kernel/cpu/sh2a/setup-sh7203.c
@@ -176,10 +176,12 @@
 static struct plat_sci_port scif0_platform_data = {
 	.mapbase	= 0xfffe8000,
 	.flags		= UPF_BOOT_AUTOCONF,
-	.scscr		= SCSCR_RE | SCSCR_TE | SCSCR_REIE,
+	.scscr		= SCSCR_RIE | SCSCR_TIE | SCSCR_RE | SCSCR_TE |
+			  SCSCR_REIE,
 	.scbrr_algo_id	= SCBRR_ALGO_2,
 	.type		= PORT_SCIF,
 	.irqs		=  { 192, 192, 192, 192 },
+	.regtype	= SCIx_SH2_SCIF_FIFODATA_REGTYPE,
 };
 
 static struct platform_device scif0_device = {
@@ -193,10 +195,12 @@
 static struct plat_sci_port scif1_platform_data = {
 	.mapbase	= 0xfffe8800,
 	.flags		= UPF_BOOT_AUTOCONF,
-	.scscr		= SCSCR_RE | SCSCR_TE | SCSCR_REIE,
+	.scscr		= SCSCR_RIE | SCSCR_TIE | SCSCR_RE | SCSCR_TE |
+			  SCSCR_REIE,
 	.scbrr_algo_id	= SCBRR_ALGO_2,
 	.type		= PORT_SCIF,
 	.irqs		=  { 196, 196, 196, 196 },
+	.regtype	= SCIx_SH2_SCIF_FIFODATA_REGTYPE,
 };
 
 static struct platform_device scif1_device = {
@@ -210,10 +214,12 @@
 static struct plat_sci_port scif2_platform_data = {
 	.mapbase	= 0xfffe9000,
 	.flags		= UPF_BOOT_AUTOCONF,
-	.scscr		= SCSCR_RE | SCSCR_TE | SCSCR_REIE,
+	.scscr		= SCSCR_RIE | SCSCR_TIE | SCSCR_RE | SCSCR_TE |
+			  SCSCR_REIE,
 	.scbrr_algo_id	= SCBRR_ALGO_2,
 	.type		= PORT_SCIF,
 	.irqs		=  { 200, 200, 200, 200 },
+	.regtype	= SCIx_SH2_SCIF_FIFODATA_REGTYPE,
 };
 
 static struct platform_device scif2_device = {
@@ -227,10 +233,12 @@
 static struct plat_sci_port scif3_platform_data = {
 	.mapbase	= 0xfffe9800,
 	.flags		= UPF_BOOT_AUTOCONF,
-	.scscr		= SCSCR_RE | SCSCR_TE | SCSCR_REIE,
+	.scscr		= SCSCR_RIE | SCSCR_TIE | SCSCR_RE | SCSCR_TE |
+			  SCSCR_REIE,
 	.scbrr_algo_id	= SCBRR_ALGO_2,
 	.type		= PORT_SCIF,
 	.irqs		=  { 204, 204, 204, 204 },
+	.regtype	= SCIx_SH2_SCIF_FIFODATA_REGTYPE,
 };
 
 static struct platform_device scif3_device = {
diff --git a/arch/sh/kernel/syscalls_32.S b/arch/sh/kernel/syscalls_32.S
index 293e39c..ee56a9b 100644
--- a/arch/sh/kernel/syscalls_32.S
+++ b/arch/sh/kernel/syscalls_32.S
@@ -382,3 +382,5 @@
 	.long sys_syncfs
 	.long sys_sendmmsg
 	.long sys_setns
+	.long sys_process_vm_readv	/* 365 */
+	.long sys_process_vm_writev
diff --git a/arch/sh/kernel/syscalls_64.S b/arch/sh/kernel/syscalls_64.S
index ceb34b9..9af7de2 100644
--- a/arch/sh/kernel/syscalls_64.S
+++ b/arch/sh/kernel/syscalls_64.S
@@ -402,3 +402,5 @@
 	.long sys_syncfs
 	.long sys_sendmmsg
 	.long sys_setns			/* 375 */
+	.long sys_process_vm_readv
+	.long sys_process_vm_writev
diff --git a/arch/sparc/include/asm/pgtable_32.h b/arch/sparc/include/asm/pgtable_32.h
index 5b31a8e..a790cc6 100644
--- a/arch/sparc/include/asm/pgtable_32.h
+++ b/arch/sparc/include/asm/pgtable_32.h
@@ -431,10 +431,6 @@
 #define kern_addr_valid(addr) \
 	(test_bit(__pa((unsigned long)(addr))>>20, sparc_valid_addr_bitmap))
 
-extern int io_remap_pfn_range(struct vm_area_struct *vma,
-			      unsigned long from, unsigned long pfn,
-			      unsigned long size, pgprot_t prot);
-
 /*
  * For sparc32&64, the pfn in io_remap_pfn_range() carries <iospace> in
  * its high 4 bits.  These macros/functions put it there or get it from there.
@@ -443,6 +439,22 @@
 #define GET_IOSPACE(pfn)		(pfn >> (BITS_PER_LONG - 4))
 #define GET_PFN(pfn)			(pfn & 0x0fffffffUL)
 
+extern int remap_pfn_range(struct vm_area_struct *, unsigned long, unsigned long,
+			   unsigned long, pgprot_t);
+
+static inline int io_remap_pfn_range(struct vm_area_struct *vma,
+				     unsigned long from, unsigned long pfn,
+				     unsigned long size, pgprot_t prot)
+{
+	unsigned long long offset, space, phys_base;
+
+	offset = ((unsigned long long) GET_PFN(pfn)) << PAGE_SHIFT;
+	space = GET_IOSPACE(pfn);
+	phys_base = offset | (space << 32ULL);
+
+	return remap_pfn_range(vma, from, phys_base >> PAGE_SHIFT, size, prot);
+}
+
 #define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
 #define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
 ({									  \
diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h
index adf8932..38ebb2c 100644
--- a/arch/sparc/include/asm/pgtable_64.h
+++ b/arch/sparc/include/asm/pgtable_64.h
@@ -757,10 +757,6 @@
 
 extern int page_in_phys_avail(unsigned long paddr);
 
-extern int io_remap_pfn_range(struct vm_area_struct *vma, unsigned long from,
-			       unsigned long pfn,
-			       unsigned long size, pgprot_t prot);
-
 /*
  * For sparc32&64, the pfn in io_remap_pfn_range() carries <iospace> in
  * its high 4 bits.  These macros/functions put it there or get it from there.
@@ -769,6 +765,22 @@
 #define GET_IOSPACE(pfn)		(pfn >> (BITS_PER_LONG - 4))
 #define GET_PFN(pfn)			(pfn & 0x0fffffffffffffffUL)
 
+extern int remap_pfn_range(struct vm_area_struct *, unsigned long, unsigned long,
+			   unsigned long, pgprot_t);
+
+static inline int io_remap_pfn_range(struct vm_area_struct *vma,
+				     unsigned long from, unsigned long pfn,
+				     unsigned long size, pgprot_t prot)
+{
+	unsigned long offset = GET_PFN(pfn) << PAGE_SHIFT;
+	int space = GET_IOSPACE(pfn);
+	unsigned long phys_base;
+
+	phys_base = offset | (((unsigned long) space) << 32UL);
+
+	return remap_pfn_range(vma, from, phys_base >> PAGE_SHIFT, size, prot);
+}
+
 #include <asm-generic/pgtable.h>
 
 /* We provide our own get_unmapped_area to cope with VA holes and
diff --git a/arch/sparc/include/asm/unistd.h b/arch/sparc/include/asm/unistd.h
index 6260d5d..c7cb0af 100644
--- a/arch/sparc/include/asm/unistd.h
+++ b/arch/sparc/include/asm/unistd.h
@@ -406,8 +406,10 @@
 #define __NR_syncfs		335
 #define __NR_sendmmsg		336
 #define __NR_setns		337
+#define __NR_process_vm_readv	338
+#define __NR_process_vm_writev	339
 
-#define NR_syscalls		338
+#define NR_syscalls		340
 
 #ifdef __32bit_syscall_numbers__
 /* Sparc 32-bit only has the "setresuid32", "getresuid32" variants,
diff --git a/arch/sparc/kernel/ds.c b/arch/sparc/kernel/ds.c
index 7429b47..381edcd 100644
--- a/arch/sparc/kernel/ds.c
+++ b/arch/sparc/kernel/ds.c
@@ -1181,13 +1181,11 @@
 
 	dp->rcv_buf_len = 4096;
 
-	dp->ds_states = kzalloc(sizeof(ds_states_template),
-				GFP_KERNEL);
+	dp->ds_states = kmemdup(ds_states_template,
+				sizeof(ds_states_template), GFP_KERNEL);
 	if (!dp->ds_states)
 		goto out_free_rcv_buf;
 
-	memcpy(dp->ds_states, ds_states_template,
-	       sizeof(ds_states_template));
 	dp->num_ds_states = ARRAY_SIZE(ds_states_template);
 
 	for (i = 0; i < dp->num_ds_states; i++)
diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h
index e27f8ea..0c218e4 100644
--- a/arch/sparc/kernel/entry.h
+++ b/arch/sparc/kernel/entry.h
@@ -42,6 +42,9 @@
 extern void fpload(unsigned long *fpregs, unsigned long *fsr);
 
 #else /* CONFIG_SPARC32 */
+
+#include <asm/trap_block.h>
+
 struct popc_3insn_patch_entry {
 	unsigned int	addr;
 	unsigned int	insns[3];
@@ -57,6 +60,10 @@
 	__popc_6insn_patch_end;
 
 extern void __init per_cpu_patch(void);
+extern void sun4v_patch_1insn_range(struct sun4v_1insn_patch_entry *,
+				    struct sun4v_1insn_patch_entry *);
+extern void sun4v_patch_2insn_range(struct sun4v_2insn_patch_entry *,
+				    struct sun4v_2insn_patch_entry *);
 extern void __init sun4v_patch(void);
 extern void __init boot_cpu_id_too_large(int cpu);
 extern unsigned int dcache_parity_tl1_occurred;
diff --git a/arch/sparc/kernel/module.c b/arch/sparc/kernel/module.c
index da0c6c7..e551987 100644
--- a/arch/sparc/kernel/module.c
+++ b/arch/sparc/kernel/module.c
@@ -17,6 +17,8 @@
 #include <asm/processor.h>
 #include <asm/spitfire.h>
 
+#include "entry.h"
+
 #ifdef CONFIG_SPARC64
 
 #include <linux/jump_label.h>
@@ -203,6 +205,29 @@
 }
 
 #ifdef CONFIG_SPARC64
+static void do_patch_sections(const Elf_Ehdr *hdr,
+			      const Elf_Shdr *sechdrs)
+{
+	const Elf_Shdr *s, *sun4v_1insn = NULL, *sun4v_2insn = NULL;
+	char *secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset;
+
+	for (s = sechdrs; s < sechdrs + hdr->e_shnum; s++) {
+		if (!strcmp(".sun4v_1insn_patch", secstrings + s->sh_name))
+			sun4v_1insn = s;
+		if (!strcmp(".sun4v_2insn_patch", secstrings + s->sh_name))
+			sun4v_2insn = s;
+	}
+
+	if (sun4v_1insn && tlb_type == hypervisor) {
+		void *p = (void *) sun4v_1insn->sh_addr;
+		sun4v_patch_1insn_range(p, p + sun4v_1insn->sh_size);
+	}
+	if (sun4v_2insn && tlb_type == hypervisor) {
+		void *p = (void *) sun4v_2insn->sh_addr;
+		sun4v_patch_2insn_range(p, p + sun4v_2insn->sh_size);
+	}
+}
+
 int module_finalize(const Elf_Ehdr *hdr,
 		    const Elf_Shdr *sechdrs,
 		    struct module *me)
@@ -210,6 +235,8 @@
 	/* make jump label nops */
 	jump_label_apply_nops(me);
 
+	do_patch_sections(hdr, sechdrs);
+
 	/* Cheetah's I-cache is fully coherent.  */
 	if (tlb_type == spitfire) {
 		unsigned long va;
diff --git a/arch/sparc/kernel/prom_common.c b/arch/sparc/kernel/prom_common.c
index 4661480..741df91 100644
--- a/arch/sparc/kernel/prom_common.c
+++ b/arch/sparc/kernel/prom_common.c
@@ -58,12 +58,10 @@
 	void *new_val;
 	int err;
 
-	new_val = kmalloc(len, GFP_KERNEL);
+	new_val = kmemdup(val, len, GFP_KERNEL);
 	if (!new_val)
 		return -ENOMEM;
 
-	memcpy(new_val, val, len);
-
 	err = -ENODEV;
 
 	mutex_lock(&of_set_property_mutex);
diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c
index c965595a..a854a1c 100644
--- a/arch/sparc/kernel/setup_64.c
+++ b/arch/sparc/kernel/setup_64.c
@@ -234,40 +234,50 @@
 	}
 }
 
+void sun4v_patch_1insn_range(struct sun4v_1insn_patch_entry *start,
+			     struct sun4v_1insn_patch_entry *end)
+{
+	while (start < end) {
+		unsigned long addr = start->addr;
+
+		*(unsigned int *) (addr +  0) = start->insn;
+		wmb();
+		__asm__ __volatile__("flush	%0" : : "r" (addr +  0));
+
+		start++;
+	}
+}
+
+void sun4v_patch_2insn_range(struct sun4v_2insn_patch_entry *start,
+			     struct sun4v_2insn_patch_entry *end)
+{
+	while (start < end) {
+		unsigned long addr = start->addr;
+
+		*(unsigned int *) (addr +  0) = start->insns[0];
+		wmb();
+		__asm__ __volatile__("flush	%0" : : "r" (addr +  0));
+
+		*(unsigned int *) (addr +  4) = start->insns[1];
+		wmb();
+		__asm__ __volatile__("flush	%0" : : "r" (addr +  4));
+
+		start++;
+	}
+}
+
 void __init sun4v_patch(void)
 {
 	extern void sun4v_hvapi_init(void);
-	struct sun4v_1insn_patch_entry *p1;
-	struct sun4v_2insn_patch_entry *p2;
 
 	if (tlb_type != hypervisor)
 		return;
 
-	p1 = &__sun4v_1insn_patch;
-	while (p1 < &__sun4v_1insn_patch_end) {
-		unsigned long addr = p1->addr;
+	sun4v_patch_1insn_range(&__sun4v_1insn_patch,
+				&__sun4v_1insn_patch_end);
 
-		*(unsigned int *) (addr +  0) = p1->insn;
-		wmb();
-		__asm__ __volatile__("flush	%0" : : "r" (addr +  0));
-
-		p1++;
-	}
-
-	p2 = &__sun4v_2insn_patch;
-	while (p2 < &__sun4v_2insn_patch_end) {
-		unsigned long addr = p2->addr;
-
-		*(unsigned int *) (addr +  0) = p2->insns[0];
-		wmb();
-		__asm__ __volatile__("flush	%0" : : "r" (addr +  0));
-
-		*(unsigned int *) (addr +  4) = p2->insns[1];
-		wmb();
-		__asm__ __volatile__("flush	%0" : : "r" (addr +  4));
-
-		p2++;
-	}
+	sun4v_patch_2insn_range(&__sun4v_2insn_patch,
+				&__sun4v_2insn_patch_end);
 
 	sun4v_hvapi_init();
 }
diff --git a/arch/sparc/kernel/signal32.c b/arch/sparc/kernel/signal32.c
index 2caa556..023b886 100644
--- a/arch/sparc/kernel/signal32.c
+++ b/arch/sparc/kernel/signal32.c
@@ -822,21 +822,23 @@
  * want to handle. Thus you cannot kill init even with a SIGKILL even by
  * mistake.
  */
-void do_signal32(sigset_t *oldset, struct pt_regs * regs,
-		 int restart_syscall, unsigned long orig_i0)
+void do_signal32(sigset_t *oldset, struct pt_regs * regs)
 {
 	struct k_sigaction ka;
+	unsigned long orig_i0;
+	int restart_syscall;
 	siginfo_t info;
 	int signr;
 	
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 
-	/* If the debugger messes with the program counter, it clears
-	 * the "in syscall" bit, directing us to not perform a syscall
-	 * restart.
-	 */
-	if (restart_syscall && !pt_regs_is_syscall(regs))
-		restart_syscall = 0;
+	restart_syscall = 0;
+	orig_i0 = 0;
+	if (pt_regs_is_syscall(regs) &&
+	    (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY))) {
+		restart_syscall = 1;
+		orig_i0 = regs->u_regs[UREG_G6];
+	}
 
 	if (signr > 0) {
 		if (restart_syscall)
diff --git a/arch/sparc/kernel/signal_32.c b/arch/sparc/kernel/signal_32.c
index 8ce247a..d54c6e5 100644
--- a/arch/sparc/kernel/signal_32.c
+++ b/arch/sparc/kernel/signal_32.c
@@ -519,10 +519,26 @@
 	siginfo_t info;
 	int signr;
 
+	/* It's a lot of work and synchronization to add a new ptrace
+	 * register for GDB to save and restore in order to get
+	 * orig_i0 correct for syscall restarts when debugging.
+	 *
+	 * Although it should be the case that most of the global
+	 * registers are volatile across a system call, glibc already
+	 * depends upon that fact that we preserve them.  So we can't
+	 * just use any global register to save away the orig_i0 value.
+	 *
+	 * In particular %g2, %g3, %g4, and %g5 are all assumed to be
+	 * preserved across a system call trap by various pieces of
+	 * code in glibc.
+	 *
+	 * %g7 is used as the "thread register".   %g6 is not used in
+	 * any fixed manner.  %g6 is used as a scratch register and
+	 * a compiler temporary, but it's value is never used across
+	 * a system call.  Therefore %g6 is usable for orig_i0 storage.
+	 */
 	if (pt_regs_is_syscall(regs) && (regs->psr & PSR_C))
-		restart_syscall = 1;
-	else
-		restart_syscall = 0;
+		regs->u_regs[UREG_G6] = orig_i0;
 
 	if (test_thread_flag(TIF_RESTORE_SIGMASK))
 		oldset = &current->saved_sigmask;
@@ -535,8 +551,12 @@
 	 * the software "in syscall" bit, directing us to not perform
 	 * a syscall restart.
 	 */
-	if (restart_syscall && !pt_regs_is_syscall(regs))
-		restart_syscall = 0;
+	restart_syscall = 0;
+	if (pt_regs_is_syscall(regs) && (regs->psr & PSR_C)) {
+		restart_syscall = 1;
+		orig_i0 = regs->u_regs[UREG_G6];
+	}
+
 
 	if (signr > 0) {
 		if (restart_syscall)
diff --git a/arch/sparc/kernel/signal_64.c b/arch/sparc/kernel/signal_64.c
index a2b8159..f0836cd 100644
--- a/arch/sparc/kernel/signal_64.c
+++ b/arch/sparc/kernel/signal_64.c
@@ -529,11 +529,27 @@
 	siginfo_t info;
 	int signr;
 	
+	/* It's a lot of work and synchronization to add a new ptrace
+	 * register for GDB to save and restore in order to get
+	 * orig_i0 correct for syscall restarts when debugging.
+	 *
+	 * Although it should be the case that most of the global
+	 * registers are volatile across a system call, glibc already
+	 * depends upon that fact that we preserve them.  So we can't
+	 * just use any global register to save away the orig_i0 value.
+	 *
+	 * In particular %g2, %g3, %g4, and %g5 are all assumed to be
+	 * preserved across a system call trap by various pieces of
+	 * code in glibc.
+	 *
+	 * %g7 is used as the "thread register".   %g6 is not used in
+	 * any fixed manner.  %g6 is used as a scratch register and
+	 * a compiler temporary, but it's value is never used across
+	 * a system call.  Therefore %g6 is usable for orig_i0 storage.
+	 */
 	if (pt_regs_is_syscall(regs) &&
-	    (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY))) {
-		restart_syscall = 1;
-	} else
-		restart_syscall = 0;
+	    (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY)))
+		regs->u_regs[UREG_G6] = orig_i0;
 
 	if (current_thread_info()->status & TS_RESTORE_SIGMASK)
 		oldset = &current->saved_sigmask;
@@ -542,22 +558,20 @@
 
 #ifdef CONFIG_COMPAT
 	if (test_thread_flag(TIF_32BIT)) {
-		extern void do_signal32(sigset_t *, struct pt_regs *,
-					int restart_syscall,
-					unsigned long orig_i0);
-		do_signal32(oldset, regs, restart_syscall, orig_i0);
+		extern void do_signal32(sigset_t *, struct pt_regs *);
+		do_signal32(oldset, regs);
 		return;
 	}
 #endif	
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 
-	/* If the debugger messes with the program counter, it clears
-	 * the software "in syscall" bit, directing us to not perform
-	 * a syscall restart.
-	 */
-	if (restart_syscall && !pt_regs_is_syscall(regs))
-		restart_syscall = 0;
+	restart_syscall = 0;
+	if (pt_regs_is_syscall(regs) &&
+	    (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY))) {
+		restart_syscall = 1;
+		orig_i0 = regs->u_regs[UREG_G6];
+	}
 
 	if (signr > 0) {
 		if (restart_syscall)
diff --git a/arch/sparc/kernel/sigutil_64.c b/arch/sparc/kernel/sigutil_64.c
index e7dc508..b19570d 100644
--- a/arch/sparc/kernel/sigutil_64.c
+++ b/arch/sparc/kernel/sigutil_64.c
@@ -2,6 +2,7 @@
 #include <linux/types.h>
 #include <linux/thread_info.h>
 #include <linux/uaccess.h>
+#include <linux/errno.h>
 
 #include <asm/sigcontext.h>
 #include <asm/fpumacro.h>
diff --git a/arch/sparc/kernel/systbls_32.S b/arch/sparc/kernel/systbls_32.S
index 09d8ec4..63402f9 100644
--- a/arch/sparc/kernel/systbls_32.S
+++ b/arch/sparc/kernel/systbls_32.S
@@ -84,4 +84,4 @@
 /*320*/	.long sys_dup3, sys_pipe2, sys_inotify_init1, sys_accept4, sys_preadv
 /*325*/	.long sys_pwritev, sys_rt_tgsigqueueinfo, sys_perf_event_open, sys_recvmmsg, sys_fanotify_init
 /*330*/	.long sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime
-/*335*/	.long sys_syncfs, sys_sendmmsg, sys_setns
+/*335*/	.long sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev
diff --git a/arch/sparc/kernel/systbls_64.S b/arch/sparc/kernel/systbls_64.S
index edbec45..db86b1a 100644
--- a/arch/sparc/kernel/systbls_64.S
+++ b/arch/sparc/kernel/systbls_64.S
@@ -85,7 +85,7 @@
 /*320*/	.word sys_dup3, sys_pipe2, sys_inotify_init1, sys_accept4, compat_sys_preadv
 	.word compat_sys_pwritev, compat_sys_rt_tgsigqueueinfo, sys_perf_event_open, compat_sys_recvmmsg, sys_fanotify_init
 /*330*/	.word sys32_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, compat_sys_open_by_handle_at, compat_sys_clock_adjtime
-	.word sys_syncfs, compat_sys_sendmmsg, sys_setns
+	.word sys_syncfs, compat_sys_sendmmsg, sys_setns, compat_sys_process_vm_readv, compat_sys_process_vm_writev
 
 #endif /* CONFIG_COMPAT */
 
@@ -162,4 +162,4 @@
 /*320*/	.word sys_dup3, sys_pipe2, sys_inotify_init1, sys_accept4, sys_preadv
 	.word sys_pwritev, sys_rt_tgsigqueueinfo, sys_perf_event_open, sys_recvmmsg, sys_fanotify_init
 /*330*/	.word sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime
-	.word sys_syncfs, sys_sendmmsg, sys_setns
+	.word sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev
diff --git a/arch/sparc/mm/Makefile b/arch/sparc/mm/Makefile
index e3cda21..301421c 100644
--- a/arch/sparc/mm/Makefile
+++ b/arch/sparc/mm/Makefile
@@ -8,7 +8,6 @@
 obj-y                   += fault_$(BITS).o
 obj-y                   += init_$(BITS).o
 obj-$(CONFIG_SPARC32)   += loadmmu.o
-obj-y                   += generic_$(BITS).o
 obj-$(CONFIG_SPARC32)   += extable.o btfixup.o srmmu.o iommu.o io-unit.o
 obj-$(CONFIG_SPARC32)   += hypersparc.o viking.o tsunami.o swift.o
 obj-$(CONFIG_SPARC_LEON)+= leon_mm.o
diff --git a/arch/sparc/mm/btfixup.c b/arch/sparc/mm/btfixup.c
index 5175ac2..8a7f817 100644
--- a/arch/sparc/mm/btfixup.c
+++ b/arch/sparc/mm/btfixup.c
@@ -302,8 +302,7 @@
 				case 'i':	/* INT */
 					if ((insn & 0xc1c00000) == 0x01000000) /* %HI */
 						set_addr(addr, q[1], fmangled, (insn & 0xffc00000) | (p[1] >> 10));
-					else if ((insn & 0x80002000) == 0x80002000 &&
-					         (insn & 0x01800000) != 0x01800000) /* %LO */
+					else if ((insn & 0x80002000) == 0x80002000) /* %LO */
 						set_addr(addr, q[1], fmangled, (insn & 0xffffe000) | (p[1] & 0x3ff));
 					else {
 						prom_printf(insn_i, p, addr, insn);
diff --git a/arch/sparc/mm/generic_32.c b/arch/sparc/mm/generic_32.c
deleted file mode 100644
index 6ca39a6..0000000
--- a/arch/sparc/mm/generic_32.c
+++ /dev/null
@@ -1,99 +0,0 @@
-/*
- * generic.c: Generic Sparc mm routines that are not dependent upon
- *            MMU type but are Sparc specific.
- *
- * Copyright (C) 1996 David S. Miller (davem@caip.rutgers.edu)
- */
-
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/pagemap.h>
-#include <linux/export.h>
-
-#include <asm/pgalloc.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-#include <asm/cacheflush.h>
-#include <asm/tlbflush.h>
-
-/* Remap IO memory, the same way as remap_pfn_range(), but use
- * the obio memory space.
- *
- * They use a pgprot that sets PAGE_IO and does not check the
- * mem_map table as this is independent of normal memory.
- */
-static inline void io_remap_pte_range(struct mm_struct *mm, pte_t * pte, unsigned long address, unsigned long size,
-	unsigned long offset, pgprot_t prot, int space)
-{
-	unsigned long end;
-
-	address &= ~PMD_MASK;
-	end = address + size;
-	if (end > PMD_SIZE)
-		end = PMD_SIZE;
-	do {
-		set_pte_at(mm, address, pte, mk_pte_io(offset, prot, space));
-		address += PAGE_SIZE;
-		offset += PAGE_SIZE;
-		pte++;
-	} while (address < end);
-}
-
-static inline int io_remap_pmd_range(struct mm_struct *mm, pmd_t * pmd, unsigned long address, unsigned long size,
-	unsigned long offset, pgprot_t prot, int space)
-{
-	unsigned long end;
-
-	address &= ~PGDIR_MASK;
-	end = address + size;
-	if (end > PGDIR_SIZE)
-		end = PGDIR_SIZE;
-	offset -= address;
-	do {
-		pte_t *pte = pte_alloc_map(mm, NULL, pmd, address);
-		if (!pte)
-			return -ENOMEM;
-		io_remap_pte_range(mm, pte, address, end - address, address + offset, prot, space);
-		address = (address + PMD_SIZE) & PMD_MASK;
-		pmd++;
-	} while (address < end);
-	return 0;
-}
-
-int io_remap_pfn_range(struct vm_area_struct *vma, unsigned long from,
-		       unsigned long pfn, unsigned long size, pgprot_t prot)
-{
-	int error = 0;
-	pgd_t * dir;
-	unsigned long beg = from;
-	unsigned long end = from + size;
-	struct mm_struct *mm = vma->vm_mm;
-	int space = GET_IOSPACE(pfn);
-	unsigned long offset = GET_PFN(pfn) << PAGE_SHIFT;
-
-	/* See comment in mm/memory.c remap_pfn_range */
-	vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;
-	vma->vm_pgoff = (offset >> PAGE_SHIFT) |
-		((unsigned long)space << 28UL);
-
-	offset -= from;
-	dir = pgd_offset(mm, from);
-	flush_cache_range(vma, beg, end);
-
-	while (from < end) {
-		pmd_t *pmd = pmd_alloc(mm, dir, from);
-		error = -ENOMEM;
-		if (!pmd)
-			break;
-		error = io_remap_pmd_range(mm, pmd, from, end - from, offset + from, prot, space);
-		if (error)
-			break;
-		from = (from + PGDIR_SIZE) & PGDIR_MASK;
-		dir++;
-	}
-
-	flush_tlb_range(vma, beg, end);
-	return error;
-}
-EXPORT_SYMBOL(io_remap_pfn_range);
diff --git a/arch/sparc/mm/generic_64.c b/arch/sparc/mm/generic_64.c
deleted file mode 100644
index 9b357dd..0000000
--- a/arch/sparc/mm/generic_64.c
+++ /dev/null
@@ -1,165 +0,0 @@
-/*
- * generic.c: Generic Sparc mm routines that are not dependent upon
- *            MMU type but are Sparc specific.
- *
- * Copyright (C) 1996 David S. Miller (davem@caip.rutgers.edu)
- */
-
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/export.h>
-#include <linux/pagemap.h>
-
-#include <asm/pgalloc.h>
-#include <asm/pgtable.h>
-#include <asm/page.h>
-#include <asm/tlbflush.h>
-
-/* Remap IO memory, the same way as remap_pfn_range(), but use
- * the obio memory space.
- *
- * They use a pgprot that sets PAGE_IO and does not check the
- * mem_map table as this is independent of normal memory.
- */
-static inline void io_remap_pte_range(struct mm_struct *mm, pte_t * pte,
-				      unsigned long address,
-				      unsigned long size,
-				      unsigned long offset, pgprot_t prot,
-				      int space)
-{
-	unsigned long end;
-
-	/* clear hack bit that was used as a write_combine side-effect flag */
-	offset &= ~0x1UL;
-	address &= ~PMD_MASK;
-	end = address + size;
-	if (end > PMD_SIZE)
-		end = PMD_SIZE;
-	do {
-		pte_t entry;
-		unsigned long curend = address + PAGE_SIZE;
-		
-		entry = mk_pte_io(offset, prot, space, PAGE_SIZE);
-		if (!(address & 0xffff)) {
-			if (PAGE_SIZE < (4 * 1024 * 1024) &&
-			    !(address & 0x3fffff) &&
-			    !(offset & 0x3ffffe) &&
-			    end >= address + 0x400000) {
-				entry = mk_pte_io(offset, prot, space,
-						  4 * 1024 * 1024);
-				curend = address + 0x400000;
-				offset += 0x400000;
-			} else if (PAGE_SIZE < (512 * 1024) &&
-				   !(address & 0x7ffff) &&
-				   !(offset & 0x7fffe) &&
-				   end >= address + 0x80000) {
-				entry = mk_pte_io(offset, prot, space,
-						  512 * 1024 * 1024);
-				curend = address + 0x80000;
-				offset += 0x80000;
-			} else if (PAGE_SIZE < (64 * 1024) &&
-				   !(offset & 0xfffe) &&
-				   end >= address + 0x10000) {
-				entry = mk_pte_io(offset, prot, space,
-						  64 * 1024);
-				curend = address + 0x10000;
-				offset += 0x10000;
-			} else
-				offset += PAGE_SIZE;
-		} else
-			offset += PAGE_SIZE;
-
-		if (pte_write(entry))
-			entry = pte_mkdirty(entry);
-		do {
-			BUG_ON(!pte_none(*pte));
-			set_pte_at(mm, address, pte, entry);
-			address += PAGE_SIZE;
-			pte_val(entry) += PAGE_SIZE;
-			pte++;
-		} while (address < curend);
-	} while (address < end);
-}
-
-static inline int io_remap_pmd_range(struct mm_struct *mm, pmd_t * pmd, unsigned long address, unsigned long size,
-	unsigned long offset, pgprot_t prot, int space)
-{
-	unsigned long end;
-
-	address &= ~PGDIR_MASK;
-	end = address + size;
-	if (end > PGDIR_SIZE)
-		end = PGDIR_SIZE;
-	offset -= address;
-	do {
-		pte_t *pte = pte_alloc_map(mm, NULL, pmd, address);
-		if (!pte)
-			return -ENOMEM;
-		io_remap_pte_range(mm, pte, address, end - address, address + offset, prot, space);
-		pte_unmap(pte);
-		address = (address + PMD_SIZE) & PMD_MASK;
-		pmd++;
-	} while (address < end);
-	return 0;
-}
-
-static inline int io_remap_pud_range(struct mm_struct *mm, pud_t * pud, unsigned long address, unsigned long size,
-	unsigned long offset, pgprot_t prot, int space)
-{
-	unsigned long end;
-
-	address &= ~PUD_MASK;
-	end = address + size;
-	if (end > PUD_SIZE)
-		end = PUD_SIZE;
-	offset -= address;
-	do {
-		pmd_t *pmd = pmd_alloc(mm, pud, address);
-		if (!pud)
-			return -ENOMEM;
-		io_remap_pmd_range(mm, pmd, address, end - address, address + offset, prot, space);
-		address = (address + PUD_SIZE) & PUD_MASK;
-		pud++;
-	} while (address < end);
-	return 0;
-}
-
-int io_remap_pfn_range(struct vm_area_struct *vma, unsigned long from,
-		unsigned long pfn, unsigned long size, pgprot_t prot)
-{
-	int error = 0;
-	pgd_t * dir;
-	unsigned long beg = from;
-	unsigned long end = from + size;
-	struct mm_struct *mm = vma->vm_mm;
-	int space = GET_IOSPACE(pfn);
-	unsigned long offset = GET_PFN(pfn) << PAGE_SHIFT;
-	unsigned long phys_base;
-
-	phys_base = offset | (((unsigned long) space) << 32UL);
-
-	/* See comment in mm/memory.c remap_pfn_range */
-	vma->vm_flags |= VM_IO | VM_RESERVED | VM_PFNMAP;
-	vma->vm_pgoff = phys_base >> PAGE_SHIFT;
-
-	offset -= from;
-	dir = pgd_offset(mm, from);
-	flush_cache_range(vma, beg, end);
-
-	while (from < end) {
-		pud_t *pud = pud_alloc(mm, dir, from);
-		error = -ENOMEM;
-		if (!pud)
-			break;
-		error = io_remap_pud_range(mm, pud, from, end - from, offset + from, prot, space);
-		if (error)
-			break;
-		from = (from + PGDIR_SIZE) & PGDIR_MASK;
-		dir++;
-	}
-
-	flush_tlb_range(vma, beg, end);
-	return error;
-}
-EXPORT_SYMBOL(io_remap_pfn_range);
diff --git a/arch/tile/include/asm/irq.h b/arch/tile/include/asm/irq.h
index 94e9a51..f80f8ce 100644
--- a/arch/tile/include/asm/irq.h
+++ b/arch/tile/include/asm/irq.h
@@ -74,16 +74,6 @@
  */
 void tile_irq_activate(unsigned int irq, int tile_irq_type);
 
-/*
- * For onboard, non-PCI (e.g. TILE_IRQ_PERCPU) devices, drivers know
- * how to use enable/disable_percpu_irq() to manage interrupts on each
- * core.  We can't use the generic enable/disable_irq() because they
- * use a single reference count per irq, rather than per cpu per irq.
- */
-void enable_percpu_irq(unsigned int irq);
-void disable_percpu_irq(unsigned int irq);
-
-
 void setup_irq_regs(void);
 
 #endif /* _ASM_TILE_IRQ_H */
diff --git a/arch/tile/kernel/irq.c b/arch/tile/kernel/irq.c
index aa0134d..02e6280 100644
--- a/arch/tile/kernel/irq.c
+++ b/arch/tile/kernel/irq.c
@@ -152,14 +152,13 @@
  * Remove an irq from the disabled mask.  If we're in an interrupt
  * context, defer enabling the HW interrupt until we leave.
  */
-void enable_percpu_irq(unsigned int irq)
+static void tile_irq_chip_enable(struct irq_data *d)
 {
-	get_cpu_var(irq_disable_mask) &= ~(1UL << irq);
+	get_cpu_var(irq_disable_mask) &= ~(1UL << d->irq);
 	if (__get_cpu_var(irq_depth) == 0)
-		unmask_irqs(1UL << irq);
+		unmask_irqs(1UL << d->irq);
 	put_cpu_var(irq_disable_mask);
 }
-EXPORT_SYMBOL(enable_percpu_irq);
 
 /*
  * Add an irq to the disabled mask.  We disable the HW interrupt
@@ -167,13 +166,12 @@
  * in an interrupt context, the return path is careful to avoid
  * unmasking a newly disabled interrupt.
  */
-void disable_percpu_irq(unsigned int irq)
+static void tile_irq_chip_disable(struct irq_data *d)
 {
-	get_cpu_var(irq_disable_mask) |= (1UL << irq);
-	mask_irqs(1UL << irq);
+	get_cpu_var(irq_disable_mask) |= (1UL << d->irq);
+	mask_irqs(1UL << d->irq);
 	put_cpu_var(irq_disable_mask);
 }
-EXPORT_SYMBOL(disable_percpu_irq);
 
 /* Mask an interrupt. */
 static void tile_irq_chip_mask(struct irq_data *d)
@@ -209,6 +207,8 @@
 
 static struct irq_chip tile_irq_chip = {
 	.name = "tile_irq_chip",
+	.irq_enable = tile_irq_chip_enable,
+	.irq_disable = tile_irq_chip_disable,
 	.irq_ack = tile_irq_chip_ack,
 	.irq_eoi = tile_irq_chip_eoi,
 	.irq_mask = tile_irq_chip_mask,
diff --git a/arch/tile/kernel/pci-dma.c b/arch/tile/kernel/pci-dma.c
index 658f2ce..b3ed19f 100644
--- a/arch/tile/kernel/pci-dma.c
+++ b/arch/tile/kernel/pci-dma.c
@@ -15,6 +15,7 @@
 #include <linux/mm.h>
 #include <linux/dma-mapping.h>
 #include <linux/vmalloc.h>
+#include <linux/export.h>
 #include <asm/tlbflush.h>
 #include <asm/homecache.h>
 
diff --git a/arch/tile/kernel/pci.c b/arch/tile/kernel/pci.c
index 2a8014c..9d610d3 100644
--- a/arch/tile/kernel/pci.c
+++ b/arch/tile/kernel/pci.c
@@ -24,6 +24,7 @@
 #include <linux/irq.h>
 #include <linux/io.h>
 #include <linux/uaccess.h>
+#include <linux/export.h>
 
 #include <asm/processor.h>
 #include <asm/sections.h>
diff --git a/arch/tile/kernel/sysfs.c b/arch/tile/kernel/sysfs.c
index b671a86..6029082 100644
--- a/arch/tile/kernel/sysfs.c
+++ b/arch/tile/kernel/sysfs.c
@@ -18,6 +18,7 @@
 #include <linux/cpu.h>
 #include <linux/slab.h>
 #include <linux/smp.h>
+#include <linux/stat.h>
 #include <hv/hypervisor.h>
 
 /* Return a string queried from the hypervisor, truncated to page size. */
diff --git a/arch/tile/lib/exports.c b/arch/tile/lib/exports.c
index a87d2a8..2a81d32 100644
--- a/arch/tile/lib/exports.c
+++ b/arch/tile/lib/exports.c
@@ -39,6 +39,9 @@
 EXPORT_SYMBOL(current_text_addr);
 EXPORT_SYMBOL(dump_stack);
 
+/* arch/tile/kernel/head.S */
+EXPORT_SYMBOL(empty_zero_page);
+
 /* arch/tile/lib/, various memcpy files */
 EXPORT_SYMBOL(memcpy);
 EXPORT_SYMBOL(__copy_to_user_inatomic);
diff --git a/arch/tile/mm/homecache.c b/arch/tile/mm/homecache.c
index cbe6f4f..1cc6ae4 100644
--- a/arch/tile/mm/homecache.c
+++ b/arch/tile/mm/homecache.c
@@ -449,9 +449,12 @@
 	VM_BUG_ON(!virt_addr_valid((void *)addr));
 	page = virt_to_page((void *)addr);
 	if (put_page_testzero(page)) {
-		int pages = (1 << order);
 		homecache_change_page_home(page, order, initial_page_home());
-		while (pages--)
-			__free_page(page++);
+		if (order == 0) {
+			free_hot_cold_page(page, 0);
+		} else {
+			init_page_count(page);
+			__free_pages(page, order);
+		}
 	}
 }
diff --git a/arch/unicore32/Kconfig b/arch/unicore32/Kconfig
index e57dcce..942ed617 100644
--- a/arch/unicore32/Kconfig
+++ b/arch/unicore32/Kconfig
@@ -237,13 +237,13 @@
 
 config I2C_BATTERY_BQ27200
 	tristate "I2C Battery BQ27200 Support"
-	select PUV3_I2C
+	select I2C_PUV3
 	select POWER_SUPPLY
 	select BATTERY_BQ27x00
 
 config I2C_EEPROM_AT24
 	tristate "I2C EEPROMs AT24 support"
-	select PUV3_I2C
+	select I2C_PUV3
 	select MISC_DEVICES
 	select EEPROM_AT24
 
diff --git a/arch/unicore32/Kconfig.debug b/arch/unicore32/Kconfig.debug
index ae2ec33..1a36262 100644
--- a/arch/unicore32/Kconfig.debug
+++ b/arch/unicore32/Kconfig.debug
@@ -44,18 +44,4 @@
 	  Say Y here if you want the debug print routines to direct their
 	  output to the UniCore On-Chip-Debugger channel using CP #1.
 
-config DEBUG_OCD_BREAKPOINT
-	bool "Breakpoint support via On-Chip-Debugger"
-	depends on DEBUG_OCD
-
-config DEBUG_UART
-	int "Kernel low-level debugging messages via serial port"
-	depends on DEBUG_LL
-	range 0 1
-	default "0"
-	help
-	  Choice for UART for kernel low-level using PKUnity UARTS,
-	  should be between zero and one. The port must have been
-	  initialised by the boot-loader before use.
-
 endmenu
diff --git a/arch/unicore32/boot/compressed/Makefile b/arch/unicore32/boot/compressed/Makefile
index b0954a2..950a9af 100644
--- a/arch/unicore32/boot/compressed/Makefile
+++ b/arch/unicore32/boot/compressed/Makefile
@@ -10,8 +10,8 @@
 # Copyright (C) 2001~2010 GUAN Xue-tao
 #
 
-EXTRA_CFLAGS	:= -fpic -fno-builtin
-EXTRA_AFLAGS	:= -Wa,-march=all
+ccflags-y	:= -fpic -fno-builtin
+asflags-y	:= -Wa,-march=all
 
 OBJS		:= misc.o
 
diff --git a/arch/unicore32/include/asm/bitops.h b/arch/unicore32/include/asm/bitops.h
index 1628a63..401f597 100644
--- a/arch/unicore32/include/asm/bitops.h
+++ b/arch/unicore32/include/asm/bitops.h
@@ -13,12 +13,6 @@
 #ifndef __UNICORE_BITOPS_H__
 #define __UNICORE_BITOPS_H__
 
-#define find_next_bit		__uc32_find_next_bit
-#define find_next_zero_bit	__uc32_find_next_zero_bit
-
-#define find_first_bit		__uc32_find_first_bit
-#define find_first_zero_bit	__uc32_find_first_zero_bit
-
 #define _ASM_GENERIC_BITOPS_FLS_H_
 #define _ASM_GENERIC_BITOPS___FLS_H_
 #define _ASM_GENERIC_BITOPS_FFS_H_
@@ -44,4 +38,10 @@
 
 #include <asm-generic/bitops.h>
 
+/* following definitions: to avoid using codes in lib/find_*.c */
+#define find_next_bit		find_next_bit
+#define find_next_zero_bit	find_next_zero_bit
+#define find_first_bit		find_first_bit
+#define find_first_zero_bit	find_first_zero_bit
+
 #endif /* __UNICORE_BITOPS_H__ */
diff --git a/arch/unicore32/include/asm/processor.h b/arch/unicore32/include/asm/processor.h
index e11cb07..f0d780a 100644
--- a/arch/unicore32/include/asm/processor.h
+++ b/arch/unicore32/include/asm/processor.h
@@ -53,7 +53,6 @@
 #define start_thread(regs, pc, sp)					\
 ({									\
 	unsigned long *stack = (unsigned long *)sp;			\
-	set_fs(USER_DS);						\
 	memset(regs->uregs, 0, sizeof(regs->uregs));			\
 	regs->UCreg_asr = USER_MODE;					\
 	regs->UCreg_pc = pc & ~1;	/* pc */                        \
diff --git a/arch/unicore32/kernel/ksyms.c b/arch/unicore32/kernel/ksyms.c
index a897080..d98bd81 100644
--- a/arch/unicore32/kernel/ksyms.c
+++ b/arch/unicore32/kernel/ksyms.c
@@ -24,8 +24,8 @@
 
 #include "ksyms.h"
 
-EXPORT_SYMBOL(__uc32_find_next_zero_bit);
-EXPORT_SYMBOL(__uc32_find_next_bit);
+EXPORT_SYMBOL(find_next_zero_bit);
+EXPORT_SYMBOL(find_next_bit);
 
 EXPORT_SYMBOL(__backtrace);
 
diff --git a/arch/unicore32/lib/findbit.S b/arch/unicore32/lib/findbit.S
index c360ce9..c777462 100644
--- a/arch/unicore32/lib/findbit.S
+++ b/arch/unicore32/lib/findbit.S
@@ -17,7 +17,7 @@
  * Purpose  : Find a 'zero' bit
  * Prototype: int find_first_zero_bit(void *addr, unsigned int maxbit);
  */
-__uc32_find_first_zero_bit:
+ENTRY(find_first_zero_bit)
 		cxor.a	r1, #0
 		beq	3f
 		mov	r2, #0
@@ -29,13 +29,14 @@
 		bub	1b
 3:		mov	r0, r1			@ no free bits
 		mov	pc, lr
+ENDPROC(find_first_zero_bit)
 
 /*
  * Purpose  : Find next 'zero' bit
  * Prototype: int find_next_zero_bit
  *		(void *addr, unsigned int maxbit, int offset)
  */
-ENTRY(__uc32_find_next_zero_bit)
+ENTRY(find_next_zero_bit)
 		cxor.a	r1, #0
 		beq	3b
 		and.a	ip, r2, #7
@@ -47,14 +48,14 @@
 		or	r2, r2, #7		@ if zero, then no bits here
 		add	r2, r2, #1		@ align bit pointer
 		b	2b			@ loop for next bit
-ENDPROC(__uc32_find_next_zero_bit)
+ENDPROC(find_next_zero_bit)
 
 /*
  * Purpose  : Find a 'one' bit
  * Prototype: int find_first_bit
  *		(const unsigned long *addr, unsigned int maxbit);
  */
-__uc32_find_first_bit:
+ENTRY(find_first_bit)
 		cxor.a	r1, #0
 		beq	3f
 		mov	r2, #0
@@ -66,13 +67,14 @@
 		bub	1b
 3:		mov	r0, r1			@ no free bits
 		mov	pc, lr
+ENDPROC(find_first_bit)
 
 /*
  * Purpose  : Find next 'one' bit
  * Prototype: int find_next_zero_bit
  *		(void *addr, unsigned int maxbit, int offset)
  */
-ENTRY(__uc32_find_next_bit)
+ENTRY(find_next_bit)
 		cxor.a	r1, #0
 		beq	3b
 		and.a	ip, r2, #7
@@ -83,7 +85,7 @@
 		or	r2, r2, #7		@ if zero, then no bits here
 		add	r2, r2, #1		@ align bit pointer
 		b	2b			@ loop for next bit
-ENDPROC(__uc32_find_next_bit)
+ENDPROC(find_next_bit)
 
 /*
  * One or more bits in the LSB of r3 are assumed to be set.
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index cb9a104..efb4294 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -390,7 +390,7 @@
 	  This option compiles in support for the CE4100 SOC for settop
 	  boxes and media devices.
 
-config X86_INTEL_MID
+config X86_WANT_INTEL_MID
 	bool "Intel MID platform support"
 	depends on X86_32
 	depends on X86_EXTENDED_PLATFORM
@@ -399,7 +399,10 @@
 	  systems which do not have the PCI legacy interfaces (Moorestown,
 	  Medfield). If you are building for a PC class system say N here.
 
-if X86_INTEL_MID
+if X86_WANT_INTEL_MID
+
+config X86_INTEL_MID
+	bool
 
 config X86_MRST
        bool "Moorestown MID platform"
@@ -411,6 +414,7 @@
 	select SPI
 	select INTEL_SCU_IPC
 	select X86_PLATFORM_DEVICES
+	select X86_INTEL_MID
 	---help---
 	  Moorestown is Intel's Low Power Intel Architecture (LPIA) based Moblin
 	  Internet Device(MID) platform. Moorestown consists of two chips:
diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index 9b7273c..1a6c09a 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -49,6 +49,7 @@
 extern int local_apic_timer_c2_ok;
 
 extern int disable_apic;
+extern unsigned int lapic_timer_frequency;
 
 #ifdef CONFIG_SMP
 extern void __inquire_remote_apic(int apicid);
diff --git a/arch/x86/include/asm/intel_scu_ipc.h b/arch/x86/include/asm/intel_scu_ipc.h
index 4420993..925b605 100644
--- a/arch/x86/include/asm/intel_scu_ipc.h
+++ b/arch/x86/include/asm/intel_scu_ipc.h
@@ -3,11 +3,15 @@
 
 #include <linux/notifier.h>
 
-#define IPCMSG_VRTC	0xFA	 /* Set vRTC device */
+#define IPCMSG_WARM_RESET	0xF0
+#define IPCMSG_COLD_RESET	0xF1
+#define IPCMSG_SOFT_RESET	0xF2
+#define IPCMSG_COLD_BOOT	0xF3
 
-/* Command id associated with message IPCMSG_VRTC */
-#define IPC_CMD_VRTC_SETTIME      1 /* Set time */
-#define IPC_CMD_VRTC_SETALARM     2 /* Set alarm */
+#define IPCMSG_VRTC		0xFA	 /* Set vRTC device */
+	/* Command id associated with message IPCMSG_VRTC */
+	#define IPC_CMD_VRTC_SETTIME      1 /* Set time */
+	#define IPC_CMD_VRTC_SETALARM     2 /* Set alarm */
 
 /* Read single register */
 int intel_scu_ipc_ioread8(u16 addr, u8 *data);
diff --git a/arch/x86/include/asm/mach_traps.h b/arch/x86/include/asm/mach_traps.h
index 72a8b52..a01e7ec7 100644
--- a/arch/x86/include/asm/mach_traps.h
+++ b/arch/x86/include/asm/mach_traps.h
@@ -17,7 +17,7 @@
 #define NMI_REASON_CLEAR_IOCHK	0x08
 #define NMI_REASON_CLEAR_MASK	0x0f
 
-static inline unsigned char get_nmi_reason(void)
+static inline unsigned char default_get_nmi_reason(void)
 {
 	return inb(NMI_REASON_PORT);
 }
diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index c9321f3..0e8ae57 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -201,7 +201,10 @@
 void mce_notify_process(void);
 
 DECLARE_PER_CPU(struct mce, injectm);
-extern struct file_operations mce_chrdev_ops;
+
+extern void register_mce_write_callback(ssize_t (*)(struct file *filp,
+				    const char __user *ubuf,
+				    size_t usize, loff_t *off));
 
 /*
  * Exception handler
diff --git a/arch/x86/include/asm/mrst.h b/arch/x86/include/asm/mrst.h
index 719f00b..93f7909 100644
--- a/arch/x86/include/asm/mrst.h
+++ b/arch/x86/include/asm/mrst.h
@@ -31,11 +31,20 @@
 };
 
 extern enum mrst_cpu_type __mrst_cpu_chip;
+
+#ifdef CONFIG_X86_INTEL_MID
+
 static inline enum mrst_cpu_type mrst_identify_cpu(void)
 {
 	return __mrst_cpu_chip;
 }
 
+#else /* !CONFIG_X86_INTEL_MID */
+
+#define mrst_identify_cpu()    (0)
+
+#endif /* !CONFIG_X86_INTEL_MID */
+
 enum mrst_timer_options {
 	MRST_TIMER_DEFAULT,
 	MRST_TIMER_APBT_ONLY,
@@ -44,6 +53,13 @@
 
 extern enum mrst_timer_options mrst_timer_options;
 
+/*
+ * Penwell uses spread spectrum clock, so the freq number is not exactly
+ * the same as reported by MSR based on SDM.
+ */
+#define PENWELL_FSB_FREQ_83SKU         83200
+#define PENWELL_FSB_FREQ_100SKU        99840
+
 #define SFI_MTMR_MAX_NUM 8
 #define SFI_MRTC_MAX	8
 
diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h
index 084ef95..95203d4 100644
--- a/arch/x86/include/asm/msr.h
+++ b/arch/x86/include/asm/msr.h
@@ -169,7 +169,14 @@
 	return native_write_msr_safe(msr, low, high);
 }
 
-/* rdmsr with exception handling */
+/*
+ * rdmsr with exception handling.
+ *
+ * Please note that the exception handling works only after we've
+ * switched to the "smart" #GP handler in trap_init() which knows about
+ * exception tables - using this macro earlier than that causes machine
+ * hangs on boxes which do not implement the @msr in the first argument.
+ */
 #define rdmsr_safe(msr, p1, p2)					\
 ({								\
 	int __err;						\
diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h
index c2ff2a1..2d2f01c 100644
--- a/arch/x86/include/asm/system.h
+++ b/arch/x86/include/asm/system.h
@@ -401,6 +401,7 @@
 extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
 
 void default_idle(void);
+bool set_pm_idle_to_default(void);
 
 void stop_this_cpu(void *dummy);
 
diff --git a/arch/x86/include/asm/timer.h b/arch/x86/include/asm/timer.h
index fa7b917..431793e 100644
--- a/arch/x86/include/asm/timer.h
+++ b/arch/x86/include/asm/timer.h
@@ -32,6 +32,22 @@
  *  (mathieu.desnoyers@polymtl.ca)
  *
  *			-johnstul@us.ibm.com "math is hard, lets go shopping!"
+ *
+ * In:
+ *
+ * ns = cycles * cyc2ns_scale / SC
+ *
+ * Although we may still have enough bits to store the value of ns,
+ * in some cases, we may not have enough bits to store cycles * cyc2ns_scale,
+ * leading to an incorrect result.
+ *
+ * To avoid this, we can decompose 'cycles' into quotient and remainder
+ * of division by SC.  Then,
+ *
+ * ns = (quot * SC + rem) * cyc2ns_scale / SC
+ *    = quot * cyc2ns_scale + (rem * cyc2ns_scale) / SC
+ *
+ *			- sqazi@google.com
  */
 
 DECLARE_PER_CPU(unsigned long, cyc2ns);
@@ -41,9 +57,14 @@
 
 static inline unsigned long long __cycles_2_ns(unsigned long long cyc)
 {
+	unsigned long long quot;
+	unsigned long long rem;
 	int cpu = smp_processor_id();
 	unsigned long long ns = per_cpu(cyc2ns_offset, cpu);
-	ns += cyc * per_cpu(cyc2ns, cpu) >> CYC2NS_SCALE_FACTOR;
+	quot = (cyc >> CYC2NS_SCALE_FACTOR);
+	rem = cyc & ((1ULL << CYC2NS_SCALE_FACTOR) - 1);
+	ns += quot * per_cpu(cyc2ns, cpu) +
+		((rem * per_cpu(cyc2ns, cpu)) >> CYC2NS_SCALE_FACTOR);
 	return ns;
 }
 
diff --git a/arch/x86/include/asm/uv/uv_mmrs.h b/arch/x86/include/asm/uv/uv_mmrs.h
index 10474fb..cf1d736 100644
--- a/arch/x86/include/asm/uv/uv_mmrs.h
+++ b/arch/x86/include/asm/uv/uv_mmrs.h
@@ -57,6 +57,7 @@
 
 #define UV1_HUB_PART_NUMBER	0x88a5
 #define UV2_HUB_PART_NUMBER	0x8eb8
+#define UV2_HUB_PART_NUMBER_X	0x1111
 
 /* Compat: if this #define is present, UV headers support UV2 */
 #define UV2_HUB_IS_SUPPORTED	1
diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h
index d3d8590..1971e65 100644
--- a/arch/x86/include/asm/x86_init.h
+++ b/arch/x86/include/asm/x86_init.h
@@ -152,6 +152,7 @@
 /**
  * struct x86_platform_ops - platform specific runtime functions
  * @calibrate_tsc:		calibrate TSC
+ * @wallclock_init:		init the wallclock device
  * @get_wallclock:		get time from HW clock like RTC etc.
  * @set_wallclock:		set time back to HW clock
  * @is_untracked_pat_range	exclude from PAT logic
@@ -160,11 +161,13 @@
  */
 struct x86_platform_ops {
 	unsigned long (*calibrate_tsc)(void);
+	void (*wallclock_init)(void);
 	unsigned long (*get_wallclock)(void);
 	int (*set_wallclock)(unsigned long nowtime);
 	void (*iommu_shutdown)(void);
 	bool (*is_untracked_pat_range)(u64 start, u64 end);
 	void (*nmi_init)(void);
+	unsigned char (*get_nmi_reason)(void);
 	int (*i8042_detect)(void);
 };
 
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index c638228..1f84794 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -738,5 +738,5 @@
 
 	atomic_set(&stop_machine_first, 1);
 	wrote_text = 0;
-	__stop_machine(stop_machine_text_poke, (void *)&tpp, NULL);
+	__stop_machine(stop_machine_text_poke, (void *)&tpp, cpu_online_mask);
 }
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index a2fd72e..f98d84c 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -186,7 +186,7 @@
 	.flags = IORESOURCE_MEM | IORESOURCE_BUSY,
 };
 
-static unsigned int calibration_result;
+unsigned int lapic_timer_frequency = 0;
 
 static void apic_pm_activate(void);
 
@@ -454,7 +454,7 @@
 	switch (mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
 	case CLOCK_EVT_MODE_ONESHOT:
-		__setup_APIC_LVTT(calibration_result,
+		__setup_APIC_LVTT(lapic_timer_frequency,
 				  mode != CLOCK_EVT_MODE_PERIODIC, 1);
 		break;
 	case CLOCK_EVT_MODE_UNUSED:
@@ -638,6 +638,25 @@
 	long delta, deltatsc;
 	int pm_referenced = 0;
 
+	/**
+	 * check if lapic timer has already been calibrated by platform
+	 * specific routine, such as tsc calibration code. if so, we just fill
+	 * in the clockevent structure and return.
+	 */
+
+	if (lapic_timer_frequency) {
+		apic_printk(APIC_VERBOSE, "lapic timer already calibrated %d\n",
+				lapic_timer_frequency);
+		lapic_clockevent.mult = div_sc(lapic_timer_frequency/APIC_DIVISOR,
+					TICK_NSEC, lapic_clockevent.shift);
+		lapic_clockevent.max_delta_ns =
+			clockevent_delta2ns(0x7FFFFF, &lapic_clockevent);
+		lapic_clockevent.min_delta_ns =
+			clockevent_delta2ns(0xF, &lapic_clockevent);
+		lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY;
+		return 0;
+	}
+
 	local_irq_disable();
 
 	/* Replace the global interrupt handler */
@@ -679,12 +698,12 @@
 	lapic_clockevent.min_delta_ns =
 		clockevent_delta2ns(0xF, &lapic_clockevent);
 
-	calibration_result = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS;
+	lapic_timer_frequency = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS;
 
 	apic_printk(APIC_VERBOSE, "..... delta %ld\n", delta);
 	apic_printk(APIC_VERBOSE, "..... mult: %u\n", lapic_clockevent.mult);
 	apic_printk(APIC_VERBOSE, "..... calibration result: %u\n",
-		    calibration_result);
+		    lapic_timer_frequency);
 
 	if (cpu_has_tsc) {
 		apic_printk(APIC_VERBOSE, "..... CPU clock speed is "
@@ -695,13 +714,13 @@
 
 	apic_printk(APIC_VERBOSE, "..... host bus clock speed is "
 		    "%u.%04u MHz.\n",
-		    calibration_result / (1000000 / HZ),
-		    calibration_result % (1000000 / HZ));
+		    lapic_timer_frequency / (1000000 / HZ),
+		    lapic_timer_frequency % (1000000 / HZ));
 
 	/*
 	 * Do a sanity check on the APIC calibration result
 	 */
-	if (calibration_result < (1000000 / HZ)) {
+	if (lapic_timer_frequency < (1000000 / HZ)) {
 		local_irq_enable();
 		pr_warning("APIC frequency too slow, disabling apic timer\n");
 		return -1;
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 3c31fa9..6d939d7 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -193,10 +193,8 @@
 	struct irq_cfg *cfg;
 	int count, node, i;
 
-	if (!legacy_pic->nr_legacy_irqs) {
-		nr_irqs_gsi = 0;
+	if (!legacy_pic->nr_legacy_irqs)
 		io_apic_irqs = ~0UL;
-	}
 
 	for (i = 0; i < nr_ioapics; i++) {
 		ioapics[i].saved_registers =
@@ -1696,6 +1694,7 @@
 	int ioapic_idx;
 	struct irq_cfg *cfg;
 	unsigned int irq;
+	struct irq_chip *chip;
 
 	printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
 	for (ioapic_idx = 0; ioapic_idx < nr_ioapics; ioapic_idx++)
@@ -1716,6 +1715,10 @@
 	for_each_active_irq(irq) {
 		struct irq_pin_list *entry;
 
+		chip = irq_get_chip(irq);
+		if (chip != &ioapic_chip)
+			continue;
+
 		cfg = irq_get_chip_data(irq);
 		if (!cfg)
 			continue;
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index 62ae300..9d59bba 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -93,6 +93,8 @@
 
 	if (node_id.s.part_number == UV2_HUB_PART_NUMBER)
 		uv_min_hub_revision_id += UV2_HUB_REVISION_BASE - 1;
+	if (node_id.s.part_number == UV2_HUB_PART_NUMBER_X)
+		uv_min_hub_revision_id += UV2_HUB_REVISION_BASE - 1;
 
 	uv_hub_info->hub_revision = uv_min_hub_revision_id;
 	pnode = (node_id.s.node_id >> 1) & ((1 << m_n_config.s.n_skt) - 1);
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index c7e46cb..0bab2b1 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -442,8 +442,6 @@
 
 static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
 {
-	u32 dummy;
-
 	early_init_amd_mc(c);
 
 	/*
@@ -473,12 +471,12 @@
 			set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
 	}
 #endif
-
-	rdmsr_safe(MSR_AMD64_PATCH_LEVEL, &c->microcode, &dummy);
 }
 
 static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 {
+	u32 dummy;
+
 #ifdef CONFIG_SMP
 	unsigned long long value;
 
@@ -657,6 +655,8 @@
 			checking_wrmsrl(MSR_AMD64_MCx_MASK(4), mask);
 		}
 	}
+
+	rdmsr_safe(MSR_AMD64_PATCH_LEVEL, &c->microcode, &dummy);
 }
 
 #ifdef CONFIG_X86_32
diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index 6199232..319882e 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -208,7 +208,7 @@
 	if (!alloc_cpumask_var(&mce_inject_cpumask, GFP_KERNEL))
 		return -ENOMEM;
 	printk(KERN_INFO "Machine check injector initialized\n");
-	mce_chrdev_ops.write = mce_write;
+	register_mce_write_callback(mce_write);
 	register_nmi_handler(NMI_LOCAL, mce_raise_notify, 0,
 				"mce_notify");
 	return 0;
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 362056a..2af127d 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -1634,16 +1634,35 @@
 	}
 }
 
-/* Modified in mce-inject.c, so not static or const */
-struct file_operations mce_chrdev_ops = {
+static ssize_t (*mce_write)(struct file *filp, const char __user *ubuf,
+			    size_t usize, loff_t *off);
+
+void register_mce_write_callback(ssize_t (*fn)(struct file *filp,
+			     const char __user *ubuf,
+			     size_t usize, loff_t *off))
+{
+	mce_write = fn;
+}
+EXPORT_SYMBOL_GPL(register_mce_write_callback);
+
+ssize_t mce_chrdev_write(struct file *filp, const char __user *ubuf,
+			 size_t usize, loff_t *off)
+{
+	if (mce_write)
+		return mce_write(filp, ubuf, usize, off);
+	else
+		return -EINVAL;
+}
+
+static const struct file_operations mce_chrdev_ops = {
 	.open			= mce_chrdev_open,
 	.release		= mce_chrdev_release,
 	.read			= mce_chrdev_read,
+	.write			= mce_chrdev_write,
 	.poll			= mce_chrdev_poll,
 	.unlocked_ioctl		= mce_chrdev_ioctl,
 	.llseek			= no_llseek,
 };
-EXPORT_SYMBOL_GPL(mce_chrdev_ops);
 
 static struct miscdevice mce_chrdev_device = {
 	MISC_MCELOG_MINOR,
diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index a71efcdb..97b2635 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -547,6 +547,7 @@
 
 		if (tmp != mask_lo) {
 			printk(KERN_WARNING "mtrr: your BIOS has configured an incorrect mask, fixing it.\n");
+			add_taint(TAINT_FIRMWARE_WORKAROUND);
 			mask_lo = tmp;
 		}
 	}
@@ -693,6 +694,7 @@
 
 	/* Disable MTRRs, and set the default type to uncached */
 	mtrr_wrmsr(MSR_MTRRdefType, deftype_lo & ~0xcff, deftype_hi);
+	wbinvd();
 }
 
 static void post_set(void) __releases(set_atomicity_lock)
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 6408910..2bda212 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -312,12 +312,8 @@
 			return -EOPNOTSUPP;
 	}
 
-	/*
-	 * Do not allow config1 (extended registers) to propagate,
-	 * there's no sane user-space generalization yet:
-	 */
 	if (attr->type == PERF_TYPE_RAW)
-		return 0;
+		return x86_pmu_extra_regs(event->attr.config, event);
 
 	if (attr->type == PERF_TYPE_HW_CACHE)
 		return set_ext_hw_attr(hwc, event);
@@ -588,7 +584,7 @@
 				x86_pmu.put_event_constraints(cpuc, cpuc->event_list[i]);
 		}
 	}
-	return num ? -ENOSPC : 0;
+	return num ? -EINVAL : 0;
 }
 
 /*
@@ -607,7 +603,7 @@
 
 	if (is_x86_event(leader)) {
 		if (n >= max_count)
-			return -ENOSPC;
+			return -EINVAL;
 		cpuc->event_list[n] = leader;
 		n++;
 	}
@@ -620,7 +616,7 @@
 			continue;
 
 		if (n >= max_count)
-			return -ENOSPC;
+			return -EINVAL;
 
 		cpuc->event_list[n] = event;
 		n++;
@@ -1316,7 +1312,7 @@
 	c = x86_pmu.get_event_constraints(fake_cpuc, event);
 
 	if (!c || !c->weight)
-		ret = -ENOSPC;
+		ret = -EINVAL;
 
 	if (x86_pmu.put_event_constraints)
 		x86_pmu.put_event_constraints(fake_cpuc, event);
@@ -1341,7 +1337,7 @@
 {
 	struct perf_event *leader = event->group_leader;
 	struct cpu_hw_events *fake_cpuc;
-	int ret = -ENOSPC, n;
+	int ret = -EINVAL, n;
 
 	fake_cpuc = allocate_fake_cpuc();
 	if (IS_ERR(fake_cpuc))
diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
index ab6343d..3b8a2d3 100644
--- a/arch/x86/kernel/cpu/perf_event_amd_ibs.c
+++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
@@ -199,8 +199,7 @@
 		goto out;
 	}
 
-	pr_err(FW_BUG "using offset %d for IBS interrupts\n", offset);
-	pr_err(FW_BUG "workaround enabled for IBS LVT offset\n");
+	pr_info("IBS: LVT offset %d assigned\n", offset);
 
 	return 0;
 out:
@@ -265,19 +264,23 @@
 static __init int amd_ibs_init(void)
 {
 	u32 caps;
-	int ret;
+	int ret = -EINVAL;
 
 	caps = __get_ibs_caps();
 	if (!caps)
 		return -ENODEV;	/* ibs not supported by the cpu */
 
-	if (!ibs_eilvt_valid()) {
-		ret = force_ibs_eilvt_setup();
-		if (ret) {
-			pr_err("Failed to setup IBS, %d\n", ret);
-			return ret;
-		}
-	}
+	/*
+	 * Force LVT offset assignment for family 10h: The offsets are
+	 * not assigned by the BIOS for this family, so the OS is
+	 * responsible for doing it. If the OS assignment fails, fall
+	 * back to BIOS settings and try to setup this.
+	 */
+	if (boot_cpu_data.x86 == 0x10)
+		force_ibs_eilvt_setup();
+
+	if (!ibs_eilvt_valid())
+		goto out;
 
 	get_online_cpus();
 	ibs_caps = caps;
@@ -287,7 +290,11 @@
 	smp_call_function(setup_APIC_ibs, NULL, 1);
 	put_online_cpus();
 
-	return perf_event_ibs_init();
+	ret = perf_event_ibs_init();
+out:
+	if (ret)
+		pr_err("Failed to setup IBS, %d\n", ret);
+	return ret;
 }
 
 /* Since we need the pci subsystem to init ibs we can't do this earlier: */
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index 2be5ebe..8d601b1 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -1545,6 +1545,13 @@
 	x86_pmu.pebs_constraints = NULL;
 }
 
+static void intel_sandybridge_quirks(void)
+{
+	printk(KERN_WARNING "PEBS disabled due to CPU errata.\n");
+	x86_pmu.pebs = 0;
+	x86_pmu.pebs_constraints = NULL;
+}
+
 __init int intel_pmu_init(void)
 {
 	union cpuid10_edx edx;
@@ -1694,6 +1701,7 @@
 		break;
 
 	case 42: /* SandyBridge */
+		x86_pmu.quirks = intel_sandybridge_quirks;
 	case 45: /* SandyBridge, "Romely-EP" */
 		memcpy(hw_cache_event_ids, snb_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index c0d238f..73da6b6 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -493,6 +493,7 @@
 	unsigned long from = cpuc->lbr_entries[0].from;
 	unsigned long old_to, to = cpuc->lbr_entries[0].to;
 	unsigned long ip = regs->ip;
+	int is_64bit = 0;
 
 	/*
 	 * We don't need to fixup if the PEBS assist is fault like
@@ -544,7 +545,10 @@
 		} else
 			kaddr = (void *)to;
 
-		kernel_insn_init(&insn, kaddr);
+#ifdef CONFIG_X86_64
+		is_64bit = kernel_ip(to) || !test_thread_flag(TIF_IA32);
+#endif
+		insn_init(&insn, kaddr, is_64bit);
 		insn_get_length(&insn);
 		to += insn.length;
 	} while (to < ip);
diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c
index 492bf13..ef484d9 100644
--- a/arch/x86/kernel/cpu/perf_event_p4.c
+++ b/arch/x86/kernel/cpu/perf_event_p4.c
@@ -1268,7 +1268,7 @@
 	}
 
 done:
-	return num ? -ENOSPC : 0;
+	return num ? -EINVAL : 0;
 }
 
 static __initconst const struct x86_pmu p4_pmu = {
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index b946a9e..1bb0bf4 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -1049,6 +1049,14 @@
 }
 EXPORT_SYMBOL_GPL(hpet_rtc_timer_init);
 
+static void hpet_disable_rtc_channel(void)
+{
+	unsigned long cfg;
+	cfg = hpet_readl(HPET_T1_CFG);
+	cfg &= ~HPET_TN_ENABLE;
+	hpet_writel(cfg, HPET_T1_CFG);
+}
+
 /*
  * The functions below are called from rtc driver.
  * Return 0 if HPET is not being used.
@@ -1060,6 +1068,9 @@
 		return 0;
 
 	hpet_rtc_flags &= ~bit_mask;
+	if (unlikely(!hpet_rtc_flags))
+		hpet_disable_rtc_channel();
+
 	return 1;
 }
 EXPORT_SYMBOL_GPL(hpet_mask_rtc_irq_bit);
@@ -1125,15 +1136,11 @@
 
 static void hpet_rtc_timer_reinit(void)
 {
-	unsigned int cfg, delta;
+	unsigned int delta;
 	int lost_ints = -1;
 
-	if (unlikely(!hpet_rtc_flags)) {
-		cfg = hpet_readl(HPET_T1_CFG);
-		cfg &= ~HPET_TN_ENABLE;
-		hpet_writel(cfg, HPET_T1_CFG);
-		return;
-	}
+	if (unlikely(!hpet_rtc_flags))
+		hpet_disable_rtc_channel();
 
 	if (!(hpet_rtc_flags & RTC_PIE) || hpet_pie_limit)
 		delta = hpet_default_delta;
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index acf8fbf..69bca46 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -38,6 +38,9 @@
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 	u64 curbase = (u64)task_stack_page(current);
 
+	if (user_mode_vm(regs))
+		return;
+
 	WARN_ONCE(regs->sp >= curbase &&
 		  regs->sp <= curbase + THREAD_SIZE &&
 		  regs->sp <  curbase + sizeof(struct thread_info) +
diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index c1a0188..44842d7 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -74,9 +74,10 @@
 	struct pvclock_vcpu_time_info *src;
 	cycle_t ret;
 
-	src = &get_cpu_var(hv_clock);
+	preempt_disable_notrace();
+	src = &__get_cpu_var(hv_clock);
 	ret = pvclock_clocksource_read(src);
-	put_cpu_var(hv_clock);
+	preempt_enable_notrace();
 	return ret;
 }
 
diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c
index f2d2a66..9d46f5e 100644
--- a/arch/x86/kernel/microcode_core.c
+++ b/arch/x86/kernel/microcode_core.c
@@ -256,7 +256,7 @@
 	return 0;
 }
 
-static void microcode_dev_exit(void)
+static void __exit microcode_dev_exit(void)
 {
 	misc_deregister(&microcode_dev);
 }
@@ -519,10 +519,8 @@
 
 	microcode_pdev = platform_device_register_simple("microcode", -1,
 							 NULL, 0);
-	if (IS_ERR(microcode_pdev)) {
-		microcode_dev_exit();
+	if (IS_ERR(microcode_pdev))
 		return PTR_ERR(microcode_pdev);
-	}
 
 	get_online_cpus();
 	mutex_lock(&microcode_mutex);
@@ -532,14 +530,12 @@
 	mutex_unlock(&microcode_mutex);
 	put_online_cpus();
 
-	if (error) {
-		platform_device_unregister(microcode_pdev);
-		return error;
-	}
+	if (error)
+		goto out_pdev;
 
 	error = microcode_dev_init();
 	if (error)
-		return error;
+		goto out_sysdev_driver;
 
 	register_syscore_ops(&mc_syscore_ops);
 	register_hotcpu_notifier(&mc_cpu_notifier);
@@ -548,6 +544,20 @@
 		" <tigran@aivazian.fsnet.co.uk>, Peter Oruba\n");
 
 	return 0;
+
+out_sysdev_driver:
+	get_online_cpus();
+	mutex_lock(&microcode_mutex);
+
+	sysdev_driver_unregister(&cpu_sysdev_class, &mc_sysdev_driver);
+
+	mutex_unlock(&microcode_mutex);
+	put_online_cpus();
+
+out_pdev:
+	platform_device_unregister(microcode_pdev);
+	return error;
+
 }
 module_init(microcode_init);
 
diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index 9103b89..0741b062 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -95,8 +95,8 @@
 	}
 #endif
 
+	set_bit(m->busid, mp_bus_not_pci);
 	if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA) - 1) == 0) {
-		set_bit(m->busid, mp_bus_not_pci);
 #if defined(CONFIG_EISA) || defined(CONFIG_MCA)
 		mp_bus_id_to_type[m->busid] = MP_BUS_ISA;
 #endif
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index b9c8628..e88f37b 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -29,6 +29,7 @@
 #include <asm/traps.h>
 #include <asm/mach_traps.h>
 #include <asm/nmi.h>
+#include <asm/x86_init.h>
 
 #define NMI_MAX_NAMELEN	16
 struct nmiaction {
@@ -348,7 +349,7 @@
 
 	/* Non-CPU-specific NMI: NMI sources can be processed on any CPU */
 	raw_spin_lock(&nmi_reason_lock);
-	reason = get_nmi_reason();
+	reason = x86_platform.get_nmi_reason();
 
 	if (reason & NMI_REASON_MASK) {
 		if (reason & NMI_REASON_SERR)
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index b9b3b1a..ee5d4fb 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -403,6 +403,14 @@
 EXPORT_SYMBOL(default_idle);
 #endif
 
+bool set_pm_idle_to_default(void)
+{
+	bool ret = !!pm_idle;
+
+	pm_idle = default_idle;
+
+	return ret;
+}
 void stop_this_cpu(void *dummy)
 {
 	local_irq_disable();
diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c
index b78643d..03920a1 100644
--- a/arch/x86/kernel/quirks.c
+++ b/arch/x86/kernel/quirks.c
@@ -553,4 +553,17 @@
 			quirk_amd_nb_node);
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_LINK,
 			quirk_amd_nb_node);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F0,
+			quirk_amd_nb_node);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F1,
+			quirk_amd_nb_node);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F2,
+			quirk_amd_nb_node);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F3,
+			quirk_amd_nb_node);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F4,
+			quirk_amd_nb_node);
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F5,
+			quirk_amd_nb_node);
+
 #endif
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index e334be1..37a458b 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -124,7 +124,7 @@
  */
 
 /*
- * Some machines require the "reboot=b"  commandline option,
+ * Some machines require the "reboot=b" or "reboot=k"  commandline options,
  * this quirk makes that automatic.
  */
 static int __init set_bios_reboot(const struct dmi_system_id *d)
@@ -136,6 +136,15 @@
 	return 0;
 }
 
+static int __init set_kbd_reboot(const struct dmi_system_id *d)
+{
+	if (reboot_type != BOOT_KBD) {
+		reboot_type = BOOT_KBD;
+		printk(KERN_INFO "%s series board detected. Selecting KBD-method for reboot.\n", d->ident);
+	}
+	return 0;
+}
+
 static struct dmi_system_id __initdata reboot_dmi_table[] = {
 	{	/* Handle problems with rebooting on Dell E520's */
 		.callback = set_bios_reboot,
@@ -295,7 +304,7 @@
 		},
 	},
 	{ /* Handle reboot issue on Acer Aspire one */
-		.callback = set_bios_reboot,
+		.callback = set_kbd_reboot,
 		.ident = "Acer Aspire One A110",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
@@ -443,6 +452,14 @@
 			DMI_MATCH(DMI_PRODUCT_NAME, "Latitude E6420"),
 		},
 	},
+	{	/* Handle problems with rebooting on the OptiPlex 990. */
+		.callback = set_pci_reboot,
+		.ident = "Dell OptiPlex 990",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 990"),
+		},
+	},
 	{ }
 };
 
diff --git a/arch/x86/kernel/rtc.c b/arch/x86/kernel/rtc.c
index 348ce01..af6db6e 100644
--- a/arch/x86/kernel/rtc.c
+++ b/arch/x86/kernel/rtc.c
@@ -12,6 +12,7 @@
 #include <asm/vsyscall.h>
 #include <asm/x86_init.h>
 #include <asm/time.h>
+#include <asm/mrst.h>
 
 #ifdef CONFIG_X86_32
 /*
@@ -242,6 +243,10 @@
 	if (of_have_populated_dt())
 		return 0;
 
+	/* Intel MID platforms don't have ioport rtc */
+	if (mrst_identify_cpu())
+		return -ENODEV;
+
 	platform_device_register(&rtc_device);
 	dev_info(&rtc_device.dev,
 		 "registered platform RTC device (no PNP device found)\n");
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index afaf384..cf0ef98 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -1045,6 +1045,8 @@
 
 	x86_init.timers.wallclock_init();
 
+	x86_platform.wallclock_init();
+
 	mcheck_init();
 
 	arch_init_ideal_nops();
diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c
index 6f164bd..c1d6cd5 100644
--- a/arch/x86/kernel/x86_init.c
+++ b/arch/x86/kernel/x86_init.c
@@ -21,12 +21,14 @@
 #include <asm/pat.h>
 #include <asm/tsc.h>
 #include <asm/iommu.h>
+#include <asm/mach_traps.h>
 
 void __cpuinit x86_init_noop(void) { }
 void __init x86_init_uint_noop(unsigned int unused) { }
 void __init x86_init_pgd_noop(pgd_t *unused) { }
 int __init iommu_init_noop(void) { return 0; }
 void iommu_shutdown_noop(void) { }
+void wallclock_init_noop(void) { }
 
 /*
  * The platform setup functions are preset with the default functions
@@ -97,11 +99,13 @@
 
 struct x86_platform_ops x86_platform = {
 	.calibrate_tsc			= native_calibrate_tsc,
+	.wallclock_init			= wallclock_init_noop,
 	.get_wallclock			= mach_get_cmos_time,
 	.set_wallclock			= mach_set_rtc_mmss,
 	.iommu_shutdown			= iommu_shutdown_noop,
 	.is_untracked_pat_range		= is_ISA_range,
 	.nmi_init			= default_nmi_init,
+	.get_nmi_reason			= default_get_nmi_reason,
 	.i8042_detect			= default_i8042_detect
 };
 
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index a0d6bd9..579a0b5 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -39,6 +39,7 @@
 #include <asm/mce.h>
 #include <asm/i387.h>
 #include <asm/xcr.h>
+#include <asm/perf_event.h>
 
 #include "trace.h"
 
@@ -118,7 +119,7 @@
 static int ple_window = KVM_VMX_DEFAULT_PLE_WINDOW;
 module_param(ple_window, int, S_IRUGO);
 
-#define NR_AUTOLOAD_MSRS 1
+#define NR_AUTOLOAD_MSRS 8
 #define VMCS02_POOL_SIZE 1
 
 struct vmcs {
@@ -622,6 +623,7 @@
 static unsigned long *vmx_msr_bitmap_longmode;
 
 static bool cpu_has_load_ia32_efer;
+static bool cpu_has_load_perf_global_ctrl;
 
 static DECLARE_BITMAP(vmx_vpid_bitmap, VMX_NR_VPIDS);
 static DEFINE_SPINLOCK(vmx_vpid_lock);
@@ -1191,15 +1193,34 @@
 	vmcs_write32(EXCEPTION_BITMAP, eb);
 }
 
+static void clear_atomic_switch_msr_special(unsigned long entry,
+		unsigned long exit)
+{
+	vmcs_clear_bits(VM_ENTRY_CONTROLS, entry);
+	vmcs_clear_bits(VM_EXIT_CONTROLS, exit);
+}
+
 static void clear_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr)
 {
 	unsigned i;
 	struct msr_autoload *m = &vmx->msr_autoload;
 
-	if (msr == MSR_EFER && cpu_has_load_ia32_efer) {
-		vmcs_clear_bits(VM_ENTRY_CONTROLS, VM_ENTRY_LOAD_IA32_EFER);
-		vmcs_clear_bits(VM_EXIT_CONTROLS, VM_EXIT_LOAD_IA32_EFER);
-		return;
+	switch (msr) {
+	case MSR_EFER:
+		if (cpu_has_load_ia32_efer) {
+			clear_atomic_switch_msr_special(VM_ENTRY_LOAD_IA32_EFER,
+					VM_EXIT_LOAD_IA32_EFER);
+			return;
+		}
+		break;
+	case MSR_CORE_PERF_GLOBAL_CTRL:
+		if (cpu_has_load_perf_global_ctrl) {
+			clear_atomic_switch_msr_special(
+					VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL,
+					VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL);
+			return;
+		}
+		break;
 	}
 
 	for (i = 0; i < m->nr; ++i)
@@ -1215,25 +1236,55 @@
 	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, m->nr);
 }
 
+static void add_atomic_switch_msr_special(unsigned long entry,
+		unsigned long exit, unsigned long guest_val_vmcs,
+		unsigned long host_val_vmcs, u64 guest_val, u64 host_val)
+{
+	vmcs_write64(guest_val_vmcs, guest_val);
+	vmcs_write64(host_val_vmcs, host_val);
+	vmcs_set_bits(VM_ENTRY_CONTROLS, entry);
+	vmcs_set_bits(VM_EXIT_CONTROLS, exit);
+}
+
 static void add_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr,
 				  u64 guest_val, u64 host_val)
 {
 	unsigned i;
 	struct msr_autoload *m = &vmx->msr_autoload;
 
-	if (msr == MSR_EFER && cpu_has_load_ia32_efer) {
-		vmcs_write64(GUEST_IA32_EFER, guest_val);
-		vmcs_write64(HOST_IA32_EFER, host_val);
-		vmcs_set_bits(VM_ENTRY_CONTROLS, VM_ENTRY_LOAD_IA32_EFER);
-		vmcs_set_bits(VM_EXIT_CONTROLS, VM_EXIT_LOAD_IA32_EFER);
-		return;
+	switch (msr) {
+	case MSR_EFER:
+		if (cpu_has_load_ia32_efer) {
+			add_atomic_switch_msr_special(VM_ENTRY_LOAD_IA32_EFER,
+					VM_EXIT_LOAD_IA32_EFER,
+					GUEST_IA32_EFER,
+					HOST_IA32_EFER,
+					guest_val, host_val);
+			return;
+		}
+		break;
+	case MSR_CORE_PERF_GLOBAL_CTRL:
+		if (cpu_has_load_perf_global_ctrl) {
+			add_atomic_switch_msr_special(
+					VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL,
+					VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL,
+					GUEST_IA32_PERF_GLOBAL_CTRL,
+					HOST_IA32_PERF_GLOBAL_CTRL,
+					guest_val, host_val);
+			return;
+		}
+		break;
 	}
 
 	for (i = 0; i < m->nr; ++i)
 		if (m->guest[i].index == msr)
 			break;
 
-	if (i == m->nr) {
+	if (i == NR_AUTOLOAD_MSRS) {
+		printk_once(KERN_WARNING"Not enough mst switch entries. "
+				"Can't add msr %x\n", msr);
+		return;
+	} else if (i == m->nr) {
 		++m->nr;
 		vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, m->nr);
 		vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, m->nr);
@@ -2455,6 +2506,42 @@
 		&& allow_1_setting(MSR_IA32_VMX_EXIT_CTLS,
 				   VM_EXIT_LOAD_IA32_EFER);
 
+	cpu_has_load_perf_global_ctrl =
+		allow_1_setting(MSR_IA32_VMX_ENTRY_CTLS,
+				VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL)
+		&& allow_1_setting(MSR_IA32_VMX_EXIT_CTLS,
+				   VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL);
+
+	/*
+	 * Some cpus support VM_ENTRY_(LOAD|SAVE)_IA32_PERF_GLOBAL_CTRL
+	 * but due to arrata below it can't be used. Workaround is to use
+	 * msr load mechanism to switch IA32_PERF_GLOBAL_CTRL.
+	 *
+	 * VM Exit May Incorrectly Clear IA32_PERF_GLOBAL_CTRL [34:32]
+	 *
+	 * AAK155             (model 26)
+	 * AAP115             (model 30)
+	 * AAT100             (model 37)
+	 * BC86,AAY89,BD102   (model 44)
+	 * BA97               (model 46)
+	 *
+	 */
+	if (cpu_has_load_perf_global_ctrl && boot_cpu_data.x86 == 0x6) {
+		switch (boot_cpu_data.x86_model) {
+		case 26:
+		case 30:
+		case 37:
+		case 44:
+		case 46:
+			cpu_has_load_perf_global_ctrl = false;
+			printk_once(KERN_WARNING"kvm: VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL "
+					"does not work properly. Using workaround\n");
+			break;
+		default:
+			break;
+		}
+	}
+
 	return 0;
 }
 
@@ -5968,6 +6055,24 @@
 	vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, 0);
 }
 
+static void atomic_switch_perf_msrs(struct vcpu_vmx *vmx)
+{
+	int i, nr_msrs;
+	struct perf_guest_switch_msr *msrs;
+
+	msrs = perf_guest_get_msrs(&nr_msrs);
+
+	if (!msrs)
+		return;
+
+	for (i = 0; i < nr_msrs; i++)
+		if (msrs[i].host == msrs[i].guest)
+			clear_atomic_switch_msr(vmx, msrs[i].msr);
+		else
+			add_atomic_switch_msr(vmx, msrs[i].msr, msrs[i].guest,
+					msrs[i].host);
+}
+
 #ifdef CONFIG_X86_64
 #define R "r"
 #define Q "q"
@@ -6017,6 +6122,8 @@
 	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
 		vmx_set_interrupt_shadow(vcpu, 0);
 
+	atomic_switch_perf_msrs(vmx);
+
 	vmx->__launched = vmx->loaded_vmcs->launched;
 	asm(
 		/* Store host registers */
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
index ea30585..dd74e46 100644
--- a/arch/x86/mm/gup.c
+++ b/arch/x86/mm/gup.c
@@ -201,6 +201,8 @@
 	do {
 		VM_BUG_ON(compound_head(page) != head);
 		pages[*nr] = page;
+		if (PageTail(page))
+			get_huge_page_tail(page);
 		(*nr)++;
 		page++;
 		refs++;
diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
index b499626..f4f29b1 100644
--- a/arch/x86/mm/highmem_32.c
+++ b/arch/x86/mm/highmem_32.c
@@ -45,6 +45,7 @@
 	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
 	BUG_ON(!pte_none(*(kmap_pte-idx)));
 	set_pte(kmap_pte-idx, mk_pte(page, prot));
+	arch_flush_lazy_mmu_mode();
 
 	return (void *)vaddr;
 }
@@ -88,6 +89,7 @@
 		 */
 		kpte_clear_flush(kmap_pte-idx, vaddr);
 		kmap_atomic_idx_pop();
+		arch_flush_lazy_mmu_mode();
 	}
 #ifdef CONFIG_DEBUG_HIGHMEM
 	else {
diff --git a/arch/x86/oprofile/init.c b/arch/x86/oprofile/init.c
index cdfe4c5..f148cf6 100644
--- a/arch/x86/oprofile/init.c
+++ b/arch/x86/oprofile/init.c
@@ -21,6 +21,7 @@
 extern void op_nmi_exit(void);
 extern void x86_backtrace(struct pt_regs * const regs, unsigned int depth);
 
+static int nmi_timer;
 
 int __init oprofile_arch_init(struct oprofile_operations *ops)
 {
@@ -31,8 +32,9 @@
 #ifdef CONFIG_X86_LOCAL_APIC
 	ret = op_nmi_init(ops);
 #endif
+	nmi_timer = (ret != 0);
 #ifdef CONFIG_X86_IO_APIC
-	if (ret < 0)
+	if (nmi_timer)
 		ret = op_nmi_timer_init(ops);
 #endif
 	ops->backtrace = x86_backtrace;
@@ -44,6 +46,7 @@
 void oprofile_arch_exit(void)
 {
 #ifdef CONFIG_X86_LOCAL_APIC
-	op_nmi_exit();
+	if (!nmi_timer)
+		op_nmi_exit();
 #endif
 }
diff --git a/arch/x86/platform/ce4100/ce4100.c b/arch/x86/platform/ce4100/ce4100.c
index 28071bb..4c61b52 100644
--- a/arch/x86/platform/ce4100/ce4100.c
+++ b/arch/x86/platform/ce4100/ce4100.c
@@ -109,7 +109,7 @@
 }
 
 #else
-static inline void sdv_serial_fixup(void);
+static inline void sdv_serial_fixup(void) {};
 #endif
 
 static void __init sdv_arch_setup(void)
diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c
index e36bf71..40e4469 100644
--- a/arch/x86/platform/efi/efi_32.c
+++ b/arch/x86/platform/efi/efi_32.c
@@ -39,43 +39,14 @@
  */
 
 static unsigned long efi_rt_eflags;
-static pgd_t efi_bak_pg_dir_pointer[2];
 
 void efi_call_phys_prelog(void)
 {
-	unsigned long cr4;
-	unsigned long temp;
 	struct desc_ptr gdt_descr;
 
 	local_irq_save(efi_rt_eflags);
 
-	/*
-	 * If I don't have PAE, I should just duplicate two entries in page
-	 * directory. If I have PAE, I just need to duplicate one entry in
-	 * page directory.
-	 */
-	cr4 = read_cr4_safe();
-
-	if (cr4 & X86_CR4_PAE) {
-		efi_bak_pg_dir_pointer[0].pgd =
-		    swapper_pg_dir[pgd_index(0)].pgd;
-		swapper_pg_dir[0].pgd =
-		    swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd;
-	} else {
-		efi_bak_pg_dir_pointer[0].pgd =
-		    swapper_pg_dir[pgd_index(0)].pgd;
-		efi_bak_pg_dir_pointer[1].pgd =
-		    swapper_pg_dir[pgd_index(0x400000)].pgd;
-		swapper_pg_dir[pgd_index(0)].pgd =
-		    swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd;
-		temp = PAGE_OFFSET + 0x400000;
-		swapper_pg_dir[pgd_index(0x400000)].pgd =
-		    swapper_pg_dir[pgd_index(temp)].pgd;
-	}
-
-	/*
-	 * After the lock is released, the original page table is restored.
-	 */
+	load_cr3(initial_page_table);
 	__flush_tlb_all();
 
 	gdt_descr.address = __pa(get_cpu_gdt_table(0));
@@ -85,28 +56,13 @@
 
 void efi_call_phys_epilog(void)
 {
-	unsigned long cr4;
 	struct desc_ptr gdt_descr;
 
 	gdt_descr.address = (unsigned long)get_cpu_gdt_table(0);
 	gdt_descr.size = GDT_SIZE - 1;
 	load_gdt(&gdt_descr);
 
-	cr4 = read_cr4_safe();
-
-	if (cr4 & X86_CR4_PAE) {
-		swapper_pg_dir[pgd_index(0)].pgd =
-		    efi_bak_pg_dir_pointer[0].pgd;
-	} else {
-		swapper_pg_dir[pgd_index(0)].pgd =
-		    efi_bak_pg_dir_pointer[0].pgd;
-		swapper_pg_dir[pgd_index(0x400000)].pgd =
-		    efi_bak_pg_dir_pointer[1].pgd;
-	}
-
-	/*
-	 * After the lock is released, the original page table is restored.
-	 */
+	load_cr3(swapper_pg_dir);
 	__flush_tlb_all();
 
 	local_irq_restore(efi_rt_eflags);
diff --git a/arch/x86/platform/mrst/mrst.c b/arch/x86/platform/mrst/mrst.c
index 6ed7afd..ad4ec1c 100644
--- a/arch/x86/platform/mrst/mrst.c
+++ b/arch/x86/platform/mrst/mrst.c
@@ -76,6 +76,20 @@
 EXPORT_SYMBOL_GPL(sfi_mrtc_array);
 int sfi_mrtc_num;
 
+static void mrst_power_off(void)
+{
+	if (__mrst_cpu_chip == MRST_CPU_CHIP_LINCROFT)
+		intel_scu_ipc_simple_command(IPCMSG_COLD_RESET, 1);
+}
+
+static void mrst_reboot(void)
+{
+	if (__mrst_cpu_chip == MRST_CPU_CHIP_LINCROFT)
+		intel_scu_ipc_simple_command(IPCMSG_COLD_RESET, 0);
+	else
+		intel_scu_ipc_simple_command(IPCMSG_COLD_BOOT, 0);
+}
+
 /* parse all the mtimer info to a static mtimer array */
 static int __init sfi_parse_mtmr(struct sfi_table_header *table)
 {
@@ -187,11 +201,34 @@
 static unsigned long __init mrst_calibrate_tsc(void)
 {
 	unsigned long flags, fast_calibrate;
+	if (__mrst_cpu_chip == MRST_CPU_CHIP_PENWELL) {
+		u32 lo, hi, ratio, fsb;
 
-	local_irq_save(flags);
-	fast_calibrate = apbt_quick_calibrate();
-	local_irq_restore(flags);
-
+		rdmsr(MSR_IA32_PERF_STATUS, lo, hi);
+		pr_debug("IA32 perf status is 0x%x, 0x%0x\n", lo, hi);
+		ratio = (hi >> 8) & 0x1f;
+		pr_debug("ratio is %d\n", ratio);
+		if (!ratio) {
+			pr_err("read a zero ratio, should be incorrect!\n");
+			pr_err("force tsc ratio to 16 ...\n");
+			ratio = 16;
+		}
+		rdmsr(MSR_FSB_FREQ, lo, hi);
+		if ((lo & 0x7) == 0x7)
+			fsb = PENWELL_FSB_FREQ_83SKU;
+		else
+			fsb = PENWELL_FSB_FREQ_100SKU;
+		fast_calibrate = ratio * fsb;
+		pr_debug("read penwell tsc %lu khz\n", fast_calibrate);
+		lapic_timer_frequency = fsb * 1000 / HZ;
+		/* mark tsc clocksource as reliable */
+		set_cpu_cap(&boot_cpu_data, X86_FEATURE_TSC_RELIABLE);
+	} else {
+		local_irq_save(flags);
+		fast_calibrate = apbt_quick_calibrate();
+		local_irq_restore(flags);
+	}
+	
 	if (fast_calibrate)
 		return fast_calibrate;
 
@@ -242,15 +279,15 @@
 	return 0;
 }
 
-/* Reboot and power off are handled by the SCU on a MID device */
-static void mrst_power_off(void)
+/*
+ * Moorestown does not have external NMI source nor port 0x61 to report
+ * NMI status. The possible NMI sources are from pmu as a result of NMI
+ * watchdog or lock debug. Reading io port 0x61 results in 0xff which
+ * misled NMI handler.
+ */
+static unsigned char mrst_get_nmi_reason(void)
 {
-	intel_scu_ipc_simple_command(0xf1, 1);
-}
-
-static void mrst_reboot(void)
-{
-	intel_scu_ipc_simple_command(0xf1, 0);
+	return 0;
 }
 
 /*
@@ -274,6 +311,8 @@
 	x86_platform.calibrate_tsc = mrst_calibrate_tsc;
 	x86_platform.i8042_detect = mrst_i8042_detect;
 	x86_init.timers.wallclock_init = mrst_rtc_init;
+	x86_platform.get_nmi_reason = mrst_get_nmi_reason;
+
 	x86_init.pci.init = pci_mrst_init;
 	x86_init.pci.fixup_irqs = x86_init_noop;
 
@@ -448,6 +487,46 @@
 	return max7315;
 }
 
+static void *tca6416_platform_data(void *info)
+{
+	static struct pca953x_platform_data tca6416;
+	struct i2c_board_info *i2c_info = info;
+	int gpio_base, intr;
+	char base_pin_name[SFI_NAME_LEN + 1];
+	char intr_pin_name[SFI_NAME_LEN + 1];
+
+	strcpy(i2c_info->type, "tca6416");
+	strcpy(base_pin_name, "tca6416_base");
+	strcpy(intr_pin_name, "tca6416_int");
+
+	gpio_base = get_gpio_by_name(base_pin_name);
+	intr = get_gpio_by_name(intr_pin_name);
+
+	if (gpio_base == -1)
+		return NULL;
+	tca6416.gpio_base = gpio_base;
+	if (intr != -1) {
+		i2c_info->irq = intr + MRST_IRQ_OFFSET;
+		tca6416.irq_base = gpio_base + MRST_IRQ_OFFSET;
+	} else {
+		i2c_info->irq = -1;
+		tca6416.irq_base = -1;
+	}
+	return &tca6416;
+}
+
+static void *mpu3050_platform_data(void *info)
+{
+	struct i2c_board_info *i2c_info = info;
+	int intr = get_gpio_by_name("mpu3050_int");
+
+	if (intr == -1)
+		return NULL;
+
+	i2c_info->irq = intr + MRST_IRQ_OFFSET;
+	return NULL;
+}
+
 static void __init *emc1403_platform_data(void *info)
 {
 	static short intr2nd_pdata;
@@ -608,13 +687,17 @@
 }
 
 static const struct devs_id __initconst device_ids[] = {
+	{"bma023", SFI_DEV_TYPE_I2C, 1, &no_platform_data},
 	{"pmic_gpio", SFI_DEV_TYPE_SPI, 1, &pmic_gpio_platform_data},
+	{"pmic_gpio", SFI_DEV_TYPE_IPC, 1, &pmic_gpio_platform_data},
 	{"spi_max3111", SFI_DEV_TYPE_SPI, 0, &max3111_platform_data},
 	{"i2c_max7315", SFI_DEV_TYPE_I2C, 1, &max7315_platform_data},
 	{"i2c_max7315_2", SFI_DEV_TYPE_I2C, 1, &max7315_platform_data},
+	{"tca6416", SFI_DEV_TYPE_I2C, 1, &tca6416_platform_data},
 	{"emc1403", SFI_DEV_TYPE_I2C, 1, &emc1403_platform_data},
 	{"i2c_accel", SFI_DEV_TYPE_I2C, 0, &lis331dl_platform_data},
 	{"pmic_audio", SFI_DEV_TYPE_IPC, 1, &no_platform_data},
+	{"mpu3050", SFI_DEV_TYPE_I2C, 1, &mpu3050_platform_data},
 
 	/* MSIC subdevices */
 	{"msic_battery", SFI_DEV_TYPE_IPC, 1, &msic_battery_platform_data},
diff --git a/arch/x86/platform/mrst/vrtc.c b/arch/x86/platform/mrst/vrtc.c
index a8ac6f1..225bd0f 100644
--- a/arch/x86/platform/mrst/vrtc.c
+++ b/arch/x86/platform/mrst/vrtc.c
@@ -76,8 +76,8 @@
 
 	spin_unlock_irqrestore(&rtc_lock, flags);
 
-	/* vRTC YEAR reg contains the offset to 1960 */
-	year += 1960;
+	/* vRTC YEAR reg contains the offset to 1972 */
+	year += 1972;
 
 	printk(KERN_INFO "vRTC: sec: %d min: %d hour: %d day: %d "
 		"mon: %d year: %d\n", sec, min, hour, mday, mon, year);
diff --git a/arch/x86/um/asm/processor.h b/arch/x86/um/asm/processor.h
index 118c143..2c32df6 100644
--- a/arch/x86/um/asm/processor.h
+++ b/arch/x86/um/asm/processor.h
@@ -11,7 +11,7 @@
 #endif
 
 #define KSTK_EIP(tsk) KSTK_REG(tsk, HOST_IP)
-#define KSTK_ESP(tsk) KSTK_REG(tsk, HOST_IP)
+#define KSTK_ESP(tsk) KSTK_REG(tsk, HOST_SP)
 #define KSTK_EBP(tsk) KSTK_REG(tsk, HOST_BP)
 
 #define ARCH_IS_STACKGROW(address) \
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index da8afd5..1f92865 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1356,7 +1356,7 @@
 	int cpu = (long)hcpu;
 	switch (action) {
 	case CPU_UP_PREPARE:
-		per_cpu(xen_vcpu, cpu) = &HYPERVISOR_shared_info->vcpu_info[cpu];
+		xen_vcpu_setup(cpu);
 		if (xen_have_vector_callback)
 			xen_init_lock_cpu(cpu);
 		break;
@@ -1386,7 +1386,6 @@
 	xen_hvm_smp_init();
 	register_cpu_notifier(&xen_hvm_cpu_notifier);
 	xen_unplug_emulated_devices();
-	have_vcpu_info_placement = 0;
 	x86_init.irqs.intr_init = xen_init_IRQ;
 	xen_hvm_init_time_ops();
 	xen_hvm_init_mmu_ops();
diff --git a/arch/x86/xen/grant-table.c b/arch/x86/xen/grant-table.c
index 6bbfd7a..5a40d24 100644
--- a/arch/x86/xen/grant-table.c
+++ b/arch/x86/xen/grant-table.c
@@ -71,7 +71,7 @@
 
 	if (shared == NULL) {
 		struct vm_struct *area =
-			alloc_vm_area(PAGE_SIZE * max_nr_gframes);
+			alloc_vm_area(PAGE_SIZE * max_nr_gframes, NULL);
 		BUG_ON(area == NULL);
 		shared = area->addr;
 		*__shared = shared;
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index 38d0af4..b2c7179 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -173,9 +173,21 @@
 	domid_t domid = DOMID_SELF;
 	int ret;
 
-	ret = HYPERVISOR_memory_op(XENMEM_maximum_reservation, &domid);
-	if (ret > 0)
-		max_pages = ret;
+	/*
+	 * For the initial domain we use the maximum reservation as
+	 * the maximum page.
+	 *
+	 * For guest domains the current maximum reservation reflects
+	 * the current maximum rather than the static maximum. In this
+	 * case the e820 map provided to us will cover the static
+	 * maximum region.
+	 */
+	if (xen_initial_domain()) {
+		ret = HYPERVISOR_memory_op(XENMEM_maximum_reservation, &domid);
+		if (ret > 0)
+			max_pages = ret;
+	}
+
 	return min(max_pages, MAX_DOMAIN_PAGES);
 }
 
@@ -410,6 +422,6 @@
 #endif
 	disable_cpuidle();
 	boot_option_idle_override = IDLE_HALT;
-
+	WARN_ON(set_pm_idle_to_default());
 	fiddle_vdso();
 }
diff --git a/block/blk-core.c b/block/blk-core.c
index f43c8a5..15de223 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -366,7 +366,14 @@
 		if (drain_all)
 			blk_throtl_drain(q);
 
-		__blk_run_queue(q);
+		/*
+		 * This function might be called on a queue which failed
+		 * driver init after queue creation.  Some drivers
+		 * (e.g. fd) get unhappy in such cases.  Kick queue iff
+		 * dispatch queue has something on it.
+		 */
+		if (!list_empty(&q->queue_head))
+			__blk_run_queue(q);
 
 		if (drain_all)
 			nr_rqs = q->rq.count[0] + q->rq.count[1];
@@ -467,6 +474,7 @@
 	q->backing_dev_info.state = 0;
 	q->backing_dev_info.capabilities = BDI_CAP_MAP_COPY;
 	q->backing_dev_info.name = "block";
+	q->node = node_id;
 
 	err = bdi_init(&q->backing_dev_info);
 	if (err) {
@@ -551,7 +559,7 @@
 	if (!uninit_q)
 		return NULL;
 
-	q = blk_init_allocated_queue_node(uninit_q, rfn, lock, node_id);
+	q = blk_init_allocated_queue(uninit_q, rfn, lock);
 	if (!q)
 		blk_cleanup_queue(uninit_q);
 
@@ -563,18 +571,9 @@
 blk_init_allocated_queue(struct request_queue *q, request_fn_proc *rfn,
 			 spinlock_t *lock)
 {
-	return blk_init_allocated_queue_node(q, rfn, lock, -1);
-}
-EXPORT_SYMBOL(blk_init_allocated_queue);
-
-struct request_queue *
-blk_init_allocated_queue_node(struct request_queue *q, request_fn_proc *rfn,
-			      spinlock_t *lock, int node_id)
-{
 	if (!q)
 		return NULL;
 
-	q->node = node_id;
 	if (blk_init_free_list(q))
 		return NULL;
 
@@ -604,7 +603,7 @@
 
 	return NULL;
 }
-EXPORT_SYMBOL(blk_init_allocated_queue_node);
+EXPORT_SYMBOL(blk_init_allocated_queue);
 
 int blk_get_queue(struct request_queue *q)
 {
@@ -1379,15 +1378,19 @@
 		 */
 		if (list_empty(&plug->list))
 			trace_block_plug(q);
-		else if (!plug->should_sort) {
-			struct request *__rq;
+		else {
+			if (!plug->should_sort) {
+				struct request *__rq;
 
-			__rq = list_entry_rq(plug->list.prev);
-			if (__rq->q != q)
-				plug->should_sort = 1;
+				__rq = list_entry_rq(plug->list.prev);
+				if (__rq->q != q)
+					plug->should_sort = 1;
+			}
+			if (request_count >= BLK_MAX_REQUEST_COUNT) {
+				blk_flush_plug_list(plug, false);
+				trace_block_plug(q);
+			}
 		}
-		if (request_count >= BLK_MAX_REQUEST_COUNT)
-			blk_flush_plug_list(plug, false);
 		list_add_tail(&req->queuelist, &plug->list);
 		drive_stat_acct(req, 1);
 	} else {
diff --git a/block/blk-map.c b/block/blk-map.c
index e663ac2..164cd00 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -204,10 +204,11 @@
 		if (!iov[i].iov_len)
 			return -EINVAL;
 
-		if (uaddr & queue_dma_alignment(q)) {
+		/*
+		 * Keep going so we check length of all segments
+		 */
+		if (uaddr & queue_dma_alignment(q))
 			unaligned = 1;
-			break;
-		}
 	}
 
 	if (unaligned || (q->dma_pad_mask & len) || map_data)
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 16ace89..4c12869 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -3184,7 +3184,7 @@
 		}
 	}
 
-	if (ret)
+	if (ret && ret != -EEXIST)
 		printk(KERN_ERR "cfq: cic link failed!\n");
 
 	return ret;
@@ -3200,6 +3200,7 @@
 {
 	struct io_context *ioc = NULL;
 	struct cfq_io_context *cic;
+	int ret;
 
 	might_sleep_if(gfp_mask & __GFP_WAIT);
 
@@ -3207,6 +3208,7 @@
 	if (!ioc)
 		return NULL;
 
+retry:
 	cic = cfq_cic_lookup(cfqd, ioc);
 	if (cic)
 		goto out;
@@ -3215,7 +3217,12 @@
 	if (cic == NULL)
 		goto err;
 
-	if (cfq_cic_link(cfqd, ioc, cic, gfp_mask))
+	ret = cfq_cic_link(cfqd, ioc, cic, gfp_mask);
+	if (ret == -EEXIST) {
+		/* someone has linked cic to ioc already */
+		cfq_cic_free(cic);
+		goto retry;
+	} else if (ret)
 		goto err_free;
 
 out:
@@ -4036,6 +4043,11 @@
 
 	if (blkio_alloc_blkg_stats(&cfqg->blkg)) {
 		kfree(cfqg);
+
+		spin_lock(&cic_index_lock);
+		ida_remove(&cic_index_ida, cfqd->cic_index);
+		spin_unlock(&cic_index_lock);
+
 		kfree(cfqd);
 		return NULL;
 	}
diff --git a/block/genhd.c b/block/genhd.c
index 9253839..02e9fca 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -19,7 +19,6 @@
 #include <linux/mutex.h>
 #include <linux/idr.h>
 #include <linux/log2.h>
-#include <linux/ctype.h>
 
 #include "blk.h"
 
@@ -916,74 +915,6 @@
 
 subsys_initcall(genhd_device_init);
 
-static ssize_t alias_show(struct device *dev,
-			       struct device_attribute *attr, char *buf)
-{
-	struct gendisk *disk = dev_to_disk(dev);
-	ssize_t ret = 0;
-
-	if (disk->alias)
-		ret = snprintf(buf, ALIAS_LEN, "%s\n", disk->alias);
-	return ret;
-}
-
-static ssize_t alias_store(struct device *dev, struct device_attribute *attr,
-			   const char *buf, size_t count)
-{
-	struct gendisk *disk = dev_to_disk(dev);
-	char *alias;
-	char *envp[] = { NULL, NULL };
-	unsigned char c;
-	int i;
-	ssize_t ret = count;
-
-	if (!count)
-		return -EINVAL;
-
-	if (count >= ALIAS_LEN) {
-		printk(KERN_ERR "alias: alias is too long\n");
-		return -EINVAL;
-	}
-
-	/* Validation check */
-	for (i = 0; i < count; i++) {
-		c = buf[i];
-		if (i == count - 1 && c == '\n')
-			break;
-		if (!isalnum(c) && c != '_' && c != '-') {
-			printk(KERN_ERR "alias: invalid alias\n");
-			return -EINVAL;
-		}
-	}
-
-	if (disk->alias) {
-		printk(KERN_INFO "alias: %s is already assigned (%s)\n",
-		       disk->disk_name, disk->alias);
-		return -EINVAL;
-	}
-
-	alias = kasprintf(GFP_KERNEL, "%s", buf);
-	if (!alias)
-		return -ENOMEM;
-
-	if (alias[count - 1] == '\n')
-		alias[count - 1] = '\0';
-
-	envp[0] = kasprintf(GFP_KERNEL, "ALIAS=%s", alias);
-	if (!envp[0]) {
-		kfree(alias);
-		return -ENOMEM;
-	}
-
-	disk->alias = alias;
-	printk(KERN_INFO "alias: assigned %s to %s\n", alias, disk->disk_name);
-
-	kobject_uevent_env(&dev->kobj, KOBJ_ADD, envp);
-
-	kfree(envp[0]);
-	return ret;
-}
-
 static ssize_t disk_range_show(struct device *dev,
 			       struct device_attribute *attr, char *buf)
 {
@@ -1043,7 +974,6 @@
 	return sprintf(buf, "%d\n", queue_discard_alignment(disk->queue));
 }
 
-static DEVICE_ATTR(alias, S_IRUGO|S_IWUSR, alias_show, alias_store);
 static DEVICE_ATTR(range, S_IRUGO, disk_range_show, NULL);
 static DEVICE_ATTR(ext_range, S_IRUGO, disk_ext_range_show, NULL);
 static DEVICE_ATTR(removable, S_IRUGO, disk_removable_show, NULL);
@@ -1066,7 +996,6 @@
 #endif
 
 static struct attribute *disk_attrs[] = {
-	&dev_attr_alias.attr,
 	&dev_attr_range.attr,
 	&dev_attr_ext_range.attr,
 	&dev_attr_removable.attr,
diff --git a/crypto/ablkcipher.c b/crypto/ablkcipher.c
index a816f24..a0f768c 100644
--- a/crypto/ablkcipher.c
+++ b/crypto/ablkcipher.c
@@ -383,6 +383,7 @@
 	return 0;
 }
 
+#ifdef CONFIG_NET
 static int crypto_ablkcipher_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_blkcipher rblkcipher;
@@ -404,6 +405,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_ablkcipher_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 static void crypto_ablkcipher_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
@@ -457,6 +464,7 @@
 	return 0;
 }
 
+#ifdef CONFIG_NET
 static int crypto_givcipher_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_blkcipher rblkcipher;
@@ -478,6 +486,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_givcipher_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 static void crypto_givcipher_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
diff --git a/crypto/aead.c b/crypto/aead.c
index 701556f..04add3dc 100644
--- a/crypto/aead.c
+++ b/crypto/aead.c
@@ -111,6 +111,7 @@
 	return 0;
 }
 
+#ifdef CONFIG_NET
 static int crypto_aead_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_aead raead;
@@ -132,6 +133,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_aead_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 static void crypto_aead_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
@@ -190,6 +197,7 @@
 	return 0;
 }
 
+#ifdef CONFIG_NET
 static int crypto_nivaead_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_aead raead;
@@ -210,6 +218,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_nivaead_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 
 static void crypto_nivaead_show(struct seq_file *m, struct crypto_alg *alg)
diff --git a/crypto/ahash.c b/crypto/ahash.c
index a3e6ef9..ac93c99 100644
--- a/crypto/ahash.c
+++ b/crypto/ahash.c
@@ -399,6 +399,7 @@
 	return sizeof(struct crypto_shash *);
 }
 
+#ifdef CONFIG_NET
 static int crypto_ahash_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_hash rhash;
@@ -416,6 +417,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_ahash_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 static void crypto_ahash_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
diff --git a/crypto/blkcipher.c b/crypto/blkcipher.c
index 2572d26..1e61d1a 100644
--- a/crypto/blkcipher.c
+++ b/crypto/blkcipher.c
@@ -494,6 +494,7 @@
 		return crypto_init_blkcipher_ops_async(tfm);
 }
 
+#ifdef CONFIG_NET
 static int crypto_blkcipher_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_blkcipher rblkcipher;
@@ -515,6 +516,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_blkcipher_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 static void crypto_blkcipher_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
diff --git a/crypto/crypto_user.c b/crypto/crypto_user.c
index 2abca78..0605a2b 100644
--- a/crypto/crypto_user.c
+++ b/crypto/crypto_user.c
@@ -44,9 +44,6 @@
 
 	down_read(&crypto_alg_sem);
 
-	if (list_empty(&crypto_alg_list))
-		return NULL;
-
 	list_for_each_entry(q, &crypto_alg_list, cra_list) {
 		int match = 0;
 
diff --git a/crypto/pcompress.c b/crypto/pcompress.c
index fefda78..2e458e5 100644
--- a/crypto/pcompress.c
+++ b/crypto/pcompress.c
@@ -48,6 +48,7 @@
 	return 0;
 }
 
+#ifdef CONFIG_NET
 static int crypto_pcomp_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_comp rpcomp;
@@ -62,6 +63,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_pcomp_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 static void crypto_pcomp_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
diff --git a/crypto/rng.c b/crypto/rng.c
index feb7de0..64f864f 100644
--- a/crypto/rng.c
+++ b/crypto/rng.c
@@ -60,6 +60,7 @@
 	return 0;
 }
 
+#ifdef CONFIG_NET
 static int crypto_rng_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_rng rrng;
@@ -76,6 +77,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_rng_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 static void crypto_rng_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
diff --git a/crypto/shash.c b/crypto/shash.c
index ea8a9c6..91009127 100644
--- a/crypto/shash.c
+++ b/crypto/shash.c
@@ -524,6 +524,7 @@
 	return alg->cra_ctxsize;
 }
 
+#ifdef CONFIG_NET
 static int crypto_shash_report(struct sk_buff *skb, struct crypto_alg *alg)
 {
 	struct crypto_report_hash rhash;
@@ -541,6 +542,12 @@
 nla_put_failure:
 	return -EMSGSIZE;
 }
+#else
+static int crypto_shash_report(struct sk_buff *skb, struct crypto_alg *alg)
+{
+	return -ENOSYS;
+}
+#endif
 
 static void crypto_shash_show(struct seq_file *m, struct crypto_alg *alg)
 	__attribute__ ((unused));
diff --git a/drivers/acpi/apei/erst.c b/drivers/acpi/apei/erst.c
index 1274080..631b947 100644
--- a/drivers/acpi/apei/erst.c
+++ b/drivers/acpi/apei/erst.c
@@ -932,7 +932,8 @@
 static int erst_open_pstore(struct pstore_info *psi);
 static int erst_close_pstore(struct pstore_info *psi);
 static ssize_t erst_reader(u64 *id, enum pstore_type_id *type,
-			   struct timespec *time, struct pstore_info *psi);
+			   struct timespec *time, char **buf,
+			   struct pstore_info *psi);
 static int erst_writer(enum pstore_type_id type, u64 *id, unsigned int part,
 		       size_t size, struct pstore_info *psi);
 static int erst_clearer(enum pstore_type_id type, u64 id,
@@ -986,17 +987,23 @@
 }
 
 static ssize_t erst_reader(u64 *id, enum pstore_type_id *type,
-			   struct timespec *time, struct pstore_info *psi)
+			   struct timespec *time, char **buf,
+			   struct pstore_info *psi)
 {
 	int rc;
 	ssize_t len = 0;
 	u64 record_id;
-	struct cper_pstore_record *rcd = (struct cper_pstore_record *)
-					(erst_info.buf - sizeof(*rcd));
+	struct cper_pstore_record *rcd;
+	size_t rcd_len = sizeof(*rcd) + erst_info.bufsize;
 
 	if (erst_disable)
 		return -ENODEV;
 
+	rcd = kmalloc(rcd_len, GFP_KERNEL);
+	if (!rcd) {
+		rc = -ENOMEM;
+		goto out;
+	}
 skip:
 	rc = erst_get_record_id_next(&reader_pos, &record_id);
 	if (rc)
@@ -1004,22 +1011,27 @@
 
 	/* no more record */
 	if (record_id == APEI_ERST_INVALID_RECORD_ID) {
-		rc = -1;
+		rc = -EINVAL;
 		goto out;
 	}
 
-	len = erst_read(record_id, &rcd->hdr, sizeof(*rcd) +
-			erst_info.bufsize);
+	len = erst_read(record_id, &rcd->hdr, rcd_len);
 	/* The record may be cleared by others, try read next record */
 	if (len == -ENOENT)
 		goto skip;
-	else if (len < 0) {
-		rc = -1;
+	else if (len < sizeof(*rcd)) {
+		rc = -EIO;
 		goto out;
 	}
 	if (uuid_le_cmp(rcd->hdr.creator_id, CPER_CREATOR_PSTORE) != 0)
 		goto skip;
 
+	*buf = kmalloc(len, GFP_KERNEL);
+	if (*buf == NULL) {
+		rc = -ENOMEM;
+		goto out;
+	}
+	memcpy(*buf, rcd->data, len - sizeof(*rcd));
 	*id = record_id;
 	if (uuid_le_cmp(rcd->sec_hdr.section_type,
 			CPER_SECTION_TYPE_DMESG) == 0)
@@ -1037,6 +1049,7 @@
 	time->tv_nsec = 0;
 
 out:
+	kfree(rcd);
 	return (rc < 0) ? rc : (len - sizeof(*rcd));
 }
 
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 73b2909..0e8e2de 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -224,7 +224,6 @@
 /*
  * Suspend / resume control
  */
-static int acpi_idle_suspend;
 static u32 saved_bm_rld;
 
 static void acpi_idle_bm_rld_save(void)
@@ -243,21 +242,13 @@
 
 int acpi_processor_suspend(struct acpi_device * device, pm_message_t state)
 {
-	if (acpi_idle_suspend == 1)
-		return 0;
-
 	acpi_idle_bm_rld_save();
-	acpi_idle_suspend = 1;
 	return 0;
 }
 
 int acpi_processor_resume(struct acpi_device * device)
 {
-	if (acpi_idle_suspend == 0)
-		return 0;
-
 	acpi_idle_bm_rld_restore();
-	acpi_idle_suspend = 0;
 	return 0;
 }
 
@@ -763,13 +754,6 @@
 
 	local_irq_disable();
 
-	/* Do not access any ACPI IO ports in suspend path */
-	if (acpi_idle_suspend) {
-		local_irq_enable();
-		cpu_relax();
-		return -EINVAL;
-	}
-
 	lapic_timer_state_broadcast(pr, cx, 1);
 	kt1 = ktime_get_real();
 	acpi_idle_do_entry(cx);
@@ -810,13 +794,6 @@
 
 	local_irq_disable();
 
-	if (acpi_idle_suspend) {
-		local_irq_enable();
-		cpu_relax();
-		return -EINVAL;
-	}
-
-
 	if (cx->entry_method != ACPI_CSTATE_FFH) {
 		current_thread_info()->status &= ~TS_POLLING;
 		/*
@@ -895,12 +872,6 @@
 	if (unlikely(!pr))
 		return -EINVAL;
 
-
-	if (acpi_idle_suspend) {
-		cpu_relax();
-		return -EINVAL;
-	}
-
 	if (!cx->bm_sts_skip && acpi_idle_bm_check()) {
 		if (drv->safe_state_index >= 0) {
 			return drv->states[drv->safe_state_index].enter(dev,
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index fb7b90b..cf26222 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -390,6 +390,9 @@
 	/* Promise */
 	{ PCI_VDEVICE(PROMISE, 0x3f20), board_ahci },	/* PDC42819 */
 
+	/* Asmedia */
+	{ PCI_VDEVICE(ASMEDIA, 0x0612), board_ahci },	/* ASM1061 */
+
 	/* Generic, PCI class code for AHCI */
 	{ PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
 	  PCI_CLASS_STORAGE_SATA_AHCI, 0xffffff, board_ahci },
diff --git a/drivers/ata/ahci_platform.c b/drivers/ata/ahci_platform.c
index 004f2ce..43b8758 100644
--- a/drivers/ata/ahci_platform.c
+++ b/drivers/ata/ahci_platform.c
@@ -65,9 +65,9 @@
 static int __init ahci_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
-	struct ahci_platform_data *pdata = dev->platform_data;
+	struct ahci_platform_data *pdata = dev_get_platdata(dev);
 	const struct platform_device_id *id = platform_get_device_id(pdev);
-	struct ata_port_info pi = ahci_port_info[id->driver_data];
+	struct ata_port_info pi = ahci_port_info[id ? id->driver_data : 0];
 	const struct ata_port_info *ppi[] = { &pi, NULL };
 	struct ahci_host_priv *hpriv;
 	struct ata_host *host;
@@ -191,7 +191,7 @@
 static int __devexit ahci_remove(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
-	struct ahci_platform_data *pdata = dev->platform_data;
+	struct ahci_platform_data *pdata = dev_get_platdata(dev);
 	struct ata_host *host = dev_get_drvdata(dev);
 
 	ata_host_detach(host);
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index f22957c..a9b2820 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2883,7 +2883,7 @@
 	    sata_scr_read(link, SCR_STATUS, &sstatus))
 		rc = -ERESTART;
 
-	if (rc == -ERESTART || try >= max_tries) {
+	if (try >= max_tries) {
 		/*
 		 * Thaw host port even if reset failed, so that the port
 		 * can be retried on the next phy event.  This risks
@@ -2909,6 +2909,16 @@
 		ata_eh_acquire(ap);
 	}
 
+	/*
+	 * While disks spinup behind PMP, some controllers fail sending SRST.
+	 * They need to be reset - as well as the PMP - before retrying.
+	 */
+	if (rc == -ERESTART) {
+		if (ata_is_host_link(link))
+			ata_eh_thaw_port(ap);
+		goto out;
+	}
+
 	if (try == max_tries - 1) {
 		sata_down_spd_limit(link, 0);
 		if (slave)
diff --git a/drivers/ata/libata-pmp.c b/drivers/ata/libata-pmp.c
index 104462d..21b80c5 100644
--- a/drivers/ata/libata-pmp.c
+++ b/drivers/ata/libata-pmp.c
@@ -389,12 +389,9 @@
 			/* link reports offline after LPM */
 			link->flags |= ATA_LFLAG_NO_LPM;
 
-			/* Class code report is unreliable and SRST
-			 * times out under certain configurations.
-			 */
+			/* Class code report is unreliable. */
 			if (link->pmp < 5)
-				link->flags |= ATA_LFLAG_NO_SRST |
-					       ATA_LFLAG_ASSUME_ATA;
+				link->flags |= ATA_LFLAG_ASSUME_ATA;
 
 			/* port 5 is for SEMB device and it doesn't like SRST */
 			if (link->pmp == 5)
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 72a9770..2a5412e 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -1217,6 +1217,10 @@
 
 /**
  *	__ata_change_queue_depth - helper for ata_scsi_change_queue_depth
+ *	@ap: ATA port to which the device change the queue depth
+ *	@sdev: SCSI device to configure queue depth for
+ *	@queue_depth: new queue depth
+ *	@reason: calling context
  *
  *	libsas and libata have different approaches for associating a sdev to
  *	its ata_port.
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 63d5327..4cadfa2 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -2533,10 +2533,12 @@
 	if (rc)
 		goto out;
 
+#ifdef CONFIG_ATA_BMDMA
 	if (bmdma)
 		/* prepare and activate BMDMA host */
 		rc = ata_pci_bmdma_prepare_host(pdev, ppi, &host);
 	else
+#endif
 		/* prepare and activate SFF host */
 		rc = ata_pci_sff_prepare_host(pdev, ppi, &host);
 	if (rc)
@@ -2544,10 +2546,12 @@
 	host->private_data = host_priv;
 	host->flags |= hflags;
 
+#ifdef CONFIG_ATA_BMDMA
 	if (bmdma) {
 		pci_set_master(pdev);
 		rc = ata_pci_sff_activate_host(host, ata_bmdma_interrupt, sht);
 	} else
+#endif
 		rc = ata_pci_sff_activate_host(host, ata_sff_interrupt, sht);
 out:
 	if (rc == 0)
diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c
index a72ab0d..2a472c5 100644
--- a/drivers/ata/pata_of_platform.c
+++ b/drivers/ata/pata_of_platform.c
@@ -52,7 +52,7 @@
 	}
 
 	ret = of_irq_to_resource(dn, 0, &irq_res);
-	if (ret == NO_IRQ)
+	if (!ret)
 		irq_res.start = irq_res.end = 0;
 	else
 		irq_res.flags = 0;
diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c
index 447d9c0..95ec435 100644
--- a/drivers/ata/sata_sis.c
+++ b/drivers/ata/sata_sis.c
@@ -104,7 +104,7 @@
 };
 
 MODULE_AUTHOR("Uwe Koziolek");
-MODULE_DESCRIPTION("low-level driver for Silicon Integratad Systems SATA controller");
+MODULE_DESCRIPTION("low-level driver for Silicon Integrated Systems SATA controller");
 MODULE_LICENSE("GPL");
 MODULE_DEVICE_TABLE(pci, sis_pci_tbl);
 MODULE_VERSION(DRV_VERSION);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 82c86545..919daa7 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -22,6 +22,7 @@
 #include <linux/kallsyms.h>
 #include <linux/mutex.h>
 #include <linux/async.h>
+#include <linux/pm_runtime.h>
 
 #include "base.h"
 #include "power/power.h"
@@ -1743,6 +1744,10 @@
 		list_del_init(&dev->kobj.entry);
 		spin_unlock(&devices_kset->list_lock);
 
+		/* Don't allow any more runtime suspends */
+		pm_runtime_get_noresume(dev);
+		pm_runtime_barrier(dev);
+
 		if (dev->bus && dev->bus->shutdown) {
 			dev_dbg(dev, "shutdown\n");
 			dev->bus->shutdown(dev);
diff --git a/drivers/base/node.c b/drivers/base/node.c
index 793f796..5693ece 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -127,12 +127,13 @@
 		       nid, K(node_page_state(nid, NR_WRITEBACK)),
 		       nid, K(node_page_state(nid, NR_FILE_PAGES)),
 		       nid, K(node_page_state(nid, NR_FILE_MAPPED)),
-		       nid, K(node_page_state(nid, NR_ANON_PAGES)
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
+		       nid, K(node_page_state(nid, NR_ANON_PAGES)
 			+ node_page_state(nid, NR_ANON_TRANSPARENT_HUGEPAGES) *
-			HPAGE_PMD_NR
+			HPAGE_PMD_NR),
+#else
+		       nid, K(node_page_state(nid, NR_ANON_PAGES)),
 #endif
-		       ),
 		       nid, K(node_page_state(nid, NR_SHMEM)),
 		       nid, node_page_state(nid, NR_KERNEL_STACK) *
 				THREAD_SIZE / 1024,
@@ -143,13 +144,14 @@
 		       nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE) +
 				node_page_state(nid, NR_SLAB_UNRECLAIMABLE)),
 		       nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE)),
-		       nid, K(node_page_state(nid, NR_SLAB_UNRECLAIMABLE))
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
+		       nid, K(node_page_state(nid, NR_SLAB_UNRECLAIMABLE))
 			, nid,
 			K(node_page_state(nid, NR_ANON_TRANSPARENT_HUGEPAGES) *
-			HPAGE_PMD_NR)
+			HPAGE_PMD_NR));
+#else
+		       nid, K(node_page_state(nid, NR_SLAB_UNRECLAIMABLE)));
 #endif
-		       );
 	n += hugetlb_report_node_meminfo(nid, buf + n);
 	return n;
 }
diff --git a/drivers/base/power/clock_ops.c b/drivers/base/power/clock_ops.c
index 5f0f85d..428e55e 100644
--- a/drivers/base/power/clock_ops.c
+++ b/drivers/base/power/clock_ops.c
@@ -229,7 +229,8 @@
 
 	list_for_each_entry_reverse(ce, &psd->clock_list, node) {
 		if (ce->status < PCE_STATUS_ERROR) {
-			clk_disable(ce->clk);
+			if (ce->status == PCE_STATUS_ENABLED)
+				clk_disable(ce->clk);
 			ce->status = PCE_STATUS_ACQUIRED;
 		}
 	}
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 7fa0984..c3d2dfc 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -920,7 +920,8 @@
  End:
 	if (!error) {
 		dev->power.is_suspended = true;
-		if (dev->power.wakeup_path && dev->parent)
+		if (dev->power.wakeup_path
+		    && dev->parent && !dev->parent->power.ignore_children)
 			dev->parent->power.wakeup_path = true;
 	}
 
diff --git a/drivers/base/power/opp.c b/drivers/base/power/opp.c
index 434a6c0..95706fa 100644
--- a/drivers/base/power/opp.c
+++ b/drivers/base/power/opp.c
@@ -669,7 +669,7 @@
 	struct device_opp *dev_opp = find_device_opp(dev);
 
 	if (IS_ERR(dev_opp))
-		return ERR_PTR(PTR_ERR(dev_opp)); /* matching type */
+		return ERR_CAST(dev_opp); /* matching type */
 
 	return &dev_opp->head;
 }
diff --git a/drivers/base/power/qos.c b/drivers/base/power/qos.c
index 30a94ea..86de6c5 100644
--- a/drivers/base/power/qos.c
+++ b/drivers/base/power/qos.c
@@ -212,11 +212,9 @@
 	if (!dev || !req) /*guard against callers passing in null */
 		return -EINVAL;
 
-	if (dev_pm_qos_request_active(req)) {
-		WARN(1, KERN_ERR "dev_pm_qos_add_request() called for already "
-			"added request\n");
+	if (WARN(dev_pm_qos_request_active(req),
+		 "%s() called for already added request\n", __func__))
 		return -EINVAL;
-	}
 
 	req->dev = dev;
 
@@ -271,11 +269,9 @@
 	if (!req) /*guard against callers passing in null */
 		return -EINVAL;
 
-	if (!dev_pm_qos_request_active(req)) {
-		WARN(1, KERN_ERR "dev_pm_qos_update_request() called for "
-			"unknown object\n");
+	if (WARN(!dev_pm_qos_request_active(req),
+		 "%s() called for unknown object\n", __func__))
 		return -EINVAL;
-	}
 
 	mutex_lock(&dev_pm_qos_mtx);
 
@@ -312,11 +308,9 @@
 	if (!req) /*guard against callers passing in null */
 		return -EINVAL;
 
-	if (!dev_pm_qos_request_active(req)) {
-		WARN(1, KERN_ERR "dev_pm_qos_remove_request() called for "
-			"unknown object\n");
+	if (WARN(!dev_pm_qos_request_active(req),
+		 "%s() called for unknown object\n", __func__))
 		return -EINVAL;
-	}
 
 	mutex_lock(&dev_pm_qos_mtx);
 
diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index 486f94e..587cce5 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -24,6 +24,7 @@
 #include <linux/interrupt.h>
 #include <linux/types.h>
 #include <linux/pci.h>
+#include <linux/pci-aspm.h>
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/delay.h>
@@ -2600,6 +2601,8 @@
 			c->Request.Timeout = 0;
 			c->Request.CDB[0] = BMIC_WRITE;
 			c->Request.CDB[6] = BMIC_CACHE_FLUSH;
+			c->Request.CDB[7] = (size >> 8) & 0xFF;
+			c->Request.CDB[8] = size & 0xFF;
 			break;
 		case TEST_UNIT_READY:
 			c->Request.CDBLen = 6;
@@ -4319,6 +4322,10 @@
 		dev_warn(&h->pdev->dev, "controller appears to be disabled\n");
 		return -ENODEV;
 	}
+
+	pci_disable_link_state(h->pdev, PCIE_LINK_STATE_L0S |
+				PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_CLKPM);
+
 	err = pci_enable_device(h->pdev);
 	if (err) {
 		dev_warn(&h->pdev->dev, "Unable to Enable PCI device\n");
@@ -4875,7 +4882,7 @@
 {
 	if (h->msix_vector || h->msi_vector) {
 		if (!request_irq(h->intr[h->intr_mode], msixhandler,
-				IRQF_DISABLED, h->devname, h))
+				0, h->devname, h))
 			return 0;
 		dev_err(&h->pdev->dev, "Unable to get msi irq %d"
 			" for %s\n", h->intr[h->intr_mode],
@@ -4884,7 +4891,7 @@
 	}
 
 	if (!request_irq(h->intr[h->intr_mode], intxhandler,
-			IRQF_DISABLED, h->devname, h))
+			IRQF_SHARED, h->devname, h))
 		return 0;
 	dev_err(&h->pdev->dev, "Unable to get irq %d for %s\n",
 		h->intr[h->intr_mode], h->devname);
@@ -5158,6 +5165,7 @@
 	h->cciss_max_sectors = 8192;
 
 	rebuild_lun_table(h, 1, 0);
+	cciss_engage_scsi(h);
 	h->busy_initializing = 0;
 	return 1;
 
diff --git a/drivers/block/cciss_scsi.c b/drivers/block/cciss_scsi.c
index 951a4e3..e820b68 100644
--- a/drivers/block/cciss_scsi.c
+++ b/drivers/block/cciss_scsi.c
@@ -1720,5 +1720,6 @@
 /* If no tape support, then these become defined out of existence */
 
 #define cciss_scsi_setup(cntl_num)
+#define cciss_engage_scsi(h)
 
 #endif /* CONFIG_CISS_SCSI_TAPE */
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 3d80682..1e888c9 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -161,17 +161,19 @@
 	&xor_funcs
 };
 
-static loff_t get_loop_size(struct loop_device *lo, struct file *file)
+static loff_t get_size(loff_t offset, loff_t sizelimit, struct file *file)
 {
-	loff_t size, offset, loopsize;
+	loff_t size, loopsize;
 
 	/* Compute loopsize in bytes */
 	size = i_size_read(file->f_mapping->host);
-	offset = lo->lo_offset;
 	loopsize = size - offset;
-	if (lo->lo_sizelimit > 0 && lo->lo_sizelimit < loopsize)
-		loopsize = lo->lo_sizelimit;
+	/* offset is beyond i_size, wierd but possible */
+	if (loopsize < 0)
+		return 0;
 
+	if (sizelimit > 0 && sizelimit < loopsize)
+		loopsize = sizelimit;
 	/*
 	 * Unfortunately, if we want to do I/O on the device,
 	 * the number of 512-byte sectors has to fit into a sector_t.
@@ -179,17 +181,25 @@
 	return loopsize >> 9;
 }
 
-static int
-figure_loop_size(struct loop_device *lo)
+static loff_t get_loop_size(struct loop_device *lo, struct file *file)
 {
-	loff_t size = get_loop_size(lo, lo->lo_backing_file);
+	return get_size(lo->lo_offset, lo->lo_sizelimit, file);
+}
+
+static int
+figure_loop_size(struct loop_device *lo, loff_t offset, loff_t sizelimit)
+{
+	loff_t size = get_size(offset, sizelimit, lo->lo_backing_file);
 	sector_t x = (sector_t)size;
 
 	if (unlikely((loff_t)x != size))
 		return -EFBIG;
-
+	if (lo->lo_offset != offset)
+		lo->lo_offset = offset;
+	if (lo->lo_sizelimit != sizelimit)
+		lo->lo_sizelimit = sizelimit;
 	set_capacity(lo->lo_disk, x);
-	return 0;					
+	return 0;
 }
 
 static inline int
@@ -372,7 +382,8 @@
 
 	if (retval < 0)
 		return retval;
-
+	if (retval != bvec->bv_len)
+		return -EIO;
 	return 0;
 }
 
@@ -411,7 +422,7 @@
 
 		/*
 		 * We use punch hole to reclaim the free space used by the
-		 * image a.k.a. discard. However we do support discard if
+		 * image a.k.a. discard. However we do not support discard if
 		 * encryption is enabled, because it may give an attacker
 		 * useful information.
 		 */
@@ -786,7 +797,7 @@
 	}
 
 	q->limits.discard_granularity = inode->i_sb->s_blocksize;
-	q->limits.discard_alignment = inode->i_sb->s_blocksize;
+	q->limits.discard_alignment = 0;
 	q->limits.max_discard_sectors = UINT_MAX >> 9;
 	q->limits.discard_zeroes_data = 1;
 	queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, q);
@@ -1058,9 +1069,7 @@
 
 	if (lo->lo_offset != info->lo_offset ||
 	    lo->lo_sizelimit != info->lo_sizelimit) {
-		lo->lo_offset = info->lo_offset;
-		lo->lo_sizelimit = info->lo_sizelimit;
-		if (figure_loop_size(lo))
+		if (figure_loop_size(lo, info->lo_offset, info->lo_sizelimit))
 			return -EFBIG;
 	}
 	loop_config_discard(lo);
@@ -1246,7 +1255,7 @@
 	err = -ENXIO;
 	if (unlikely(lo->lo_state != Lo_bound))
 		goto out;
-	err = figure_loop_size(lo);
+	err = figure_loop_size(lo, lo->lo_offset, lo->lo_sizelimit);
 	if (unlikely(err))
 		goto out;
 	sec = get_capacity(lo->lo_disk);
@@ -1284,13 +1293,19 @@
 			goto out_unlocked;
 		break;
 	case LOOP_SET_STATUS:
-		err = loop_set_status_old(lo, (struct loop_info __user *) arg);
+		err = -EPERM;
+		if ((mode & FMODE_WRITE) || capable(CAP_SYS_ADMIN))
+			err = loop_set_status_old(lo,
+					(struct loop_info __user *)arg);
 		break;
 	case LOOP_GET_STATUS:
 		err = loop_get_status_old(lo, (struct loop_info __user *) arg);
 		break;
 	case LOOP_SET_STATUS64:
-		err = loop_set_status64(lo, (struct loop_info64 __user *) arg);
+		err = -EPERM;
+		if ((mode & FMODE_WRITE) || capable(CAP_SYS_ADMIN))
+			err = loop_set_status64(lo,
+					(struct loop_info64 __user *) arg);
 		break;
 	case LOOP_GET_STATUS64:
 		err = loop_get_status64(lo, (struct loop_info64 __user *) arg);
diff --git a/drivers/block/paride/pg.c b/drivers/block/paride/pg.c
index 6b9a200..a79fb4f 100644
--- a/drivers/block/paride/pg.c
+++ b/drivers/block/paride/pg.c
@@ -630,6 +630,7 @@
 		if (dev->status & 0x10)
 			return -ETIME;
 
+	memset(&hdr, 0, sizeof(hdr));
 	hdr.magic = PG_MAGIC;
 	hdr.dlen = dev->dlen;
 	copy = 0;
diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c
index 65cc424..148ab94 100644
--- a/drivers/block/rbd.c
+++ b/drivers/block/rbd.c
@@ -183,10 +183,6 @@
 
 static int __rbd_init_snaps_header(struct rbd_device *rbd_dev);
 static void rbd_dev_release(struct device *dev);
-static ssize_t rbd_snap_rollback(struct device *dev,
-				 struct device_attribute *attr,
-				 const char *buf,
-				 size_t size);
 static ssize_t rbd_snap_add(struct device *dev,
 			    struct device_attribute *attr,
 			    const char *buf,
@@ -461,6 +457,10 @@
 	u32 snap_count = le32_to_cpu(ondisk->snap_count);
 	int ret = -ENOMEM;
 
+	if (memcmp(ondisk, RBD_HEADER_TEXT, sizeof(RBD_HEADER_TEXT))) {
+		return -ENXIO;
+	}
+
 	init_rwsem(&header->snap_rwsem);
 	header->snap_names_len = le64_to_cpu(ondisk->snap_names_len);
 	header->snapc = kmalloc(sizeof(struct ceph_snap_context) +
@@ -1356,32 +1356,6 @@
 }
 
 /*
- * Request sync osd rollback
- */
-static int rbd_req_sync_rollback_obj(struct rbd_device *dev,
-				     u64 snapid,
-				     const char *obj)
-{
-	struct ceph_osd_req_op *ops;
-	int ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_ROLLBACK, 0);
-	if (ret < 0)
-		return ret;
-
-	ops[0].snap.snapid = snapid;
-
-	ret = rbd_req_sync_op(dev, NULL,
-			       CEPH_NOSNAP,
-			       0,
-			       CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK,
-			       ops,
-			       1, obj, 0, 0, NULL, NULL, NULL);
-
-	rbd_destroy_ops(ops);
-
-	return ret;
-}
-
-/*
  * Request sync osd read
  */
 static int rbd_req_sync_exec(struct rbd_device *dev,
@@ -1610,8 +1584,13 @@
 			goto out_dh;
 
 		rc = rbd_header_from_disk(header, dh, snap_count, GFP_KERNEL);
-		if (rc < 0)
+		if (rc < 0) {
+			if (rc == -ENXIO) {
+				pr_warning("unrecognized header format"
+					   " for image %s", rbd_dev->obj);
+			}
 			goto out_dh;
+		}
 
 		if (snap_count != header->total_snaps) {
 			snap_count = header->total_snaps;
@@ -1882,7 +1861,6 @@
 static DEVICE_ATTR(refresh, S_IWUSR, NULL, rbd_image_refresh);
 static DEVICE_ATTR(current_snap, S_IRUGO, rbd_snap_show, NULL);
 static DEVICE_ATTR(create_snap, S_IWUSR, NULL, rbd_snap_add);
-static DEVICE_ATTR(rollback_snap, S_IWUSR, NULL, rbd_snap_rollback);
 
 static struct attribute *rbd_attrs[] = {
 	&dev_attr_size.attr,
@@ -1893,7 +1871,6 @@
 	&dev_attr_current_snap.attr,
 	&dev_attr_refresh.attr,
 	&dev_attr_create_snap.attr,
-	&dev_attr_rollback_snap.attr,
 	NULL
 };
 
@@ -2424,64 +2401,6 @@
 	return ret;
 }
 
-static ssize_t rbd_snap_rollback(struct device *dev,
-				 struct device_attribute *attr,
-				 const char *buf,
-				 size_t count)
-{
-	struct rbd_device *rbd_dev = dev_to_rbd(dev);
-	int ret;
-	u64 snapid;
-	u64 cur_ofs;
-	char *seg_name = NULL;
-	char *snap_name = kmalloc(count + 1, GFP_KERNEL);
-	ret = -ENOMEM;
-	if (!snap_name)
-		return ret;
-
-	/* parse snaps add command */
-	snprintf(snap_name, count, "%s", buf);
-	seg_name = kmalloc(RBD_MAX_SEG_NAME_LEN + 1, GFP_NOIO);
-	if (!seg_name)
-		goto done;
-
-	mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
-
-	ret = snap_by_name(&rbd_dev->header, snap_name, &snapid, NULL);
-	if (ret < 0)
-		goto done_unlock;
-
-	dout("snapid=%lld\n", snapid);
-
-	cur_ofs = 0;
-	while (cur_ofs < rbd_dev->header.image_size) {
-		cur_ofs += rbd_get_segment(&rbd_dev->header,
-					   rbd_dev->obj,
-					   cur_ofs, (u64)-1,
-					   seg_name, NULL);
-		dout("seg_name=%s\n", seg_name);
-
-		ret = rbd_req_sync_rollback_obj(rbd_dev, snapid, seg_name);
-		if (ret < 0)
-			pr_warning("could not roll back obj %s err=%d\n",
-				   seg_name, ret);
-	}
-
-	ret = __rbd_update_snaps(rbd_dev);
-	if (ret < 0)
-		goto done_unlock;
-
-	ret = count;
-
-done_unlock:
-	mutex_unlock(&ctl_mutex);
-done:
-	kfree(seg_name);
-	kfree(snap_name);
-
-	return ret;
-}
-
 static struct bus_attribute rbd_bus_attrs[] = {
 	__ATTR(add, S_IWUSR, NULL, rbd_add),
 	__ATTR(remove, S_IWUSR, NULL, rbd_remove),
diff --git a/drivers/block/swim3.c b/drivers/block/swim3.c
index ae3e167..89ddab1 100644
--- a/drivers/block/swim3.c
+++ b/drivers/block/swim3.c
@@ -16,6 +16,8 @@
  * handle GCR disks
  */
 
+#undef DEBUG
+
 #include <linux/stddef.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
@@ -36,13 +38,11 @@
 #include <asm/machdep.h>
 #include <asm/pmac_feature.h>
 
-static DEFINE_MUTEX(swim3_mutex);
-static struct request_queue *swim3_queue;
-static struct gendisk *disks[2];
-static struct request *fd_req;
-
 #define MAX_FLOPPIES	2
 
+static DEFINE_MUTEX(swim3_mutex);
+static struct gendisk *disks[MAX_FLOPPIES];
+
 enum swim_state {
 	idle,
 	locating,
@@ -177,7 +177,6 @@
 
 struct floppy_state {
 	enum swim_state	state;
-	spinlock_t lock;
 	struct swim3 __iomem *swim3;	/* hardware registers */
 	struct dbdma_regs __iomem *dma;	/* DMA controller registers */
 	int	swim3_intr;	/* interrupt number for SWIM3 */
@@ -204,8 +203,20 @@
 	int	wanted;
 	struct macio_dev *mdev;
 	char	dbdma_cmd_space[5 * sizeof(struct dbdma_cmd)];
+	int	index;
+	struct request *cur_req;
 };
 
+#define swim3_err(fmt, arg...)	dev_err(&fs->mdev->ofdev.dev, "[fd%d] " fmt, fs->index, arg)
+#define swim3_warn(fmt, arg...)	dev_warn(&fs->mdev->ofdev.dev, "[fd%d] " fmt, fs->index, arg)
+#define swim3_info(fmt, arg...)	dev_info(&fs->mdev->ofdev.dev, "[fd%d] " fmt, fs->index, arg)
+
+#ifdef DEBUG
+#define swim3_dbg(fmt, arg...)	dev_dbg(&fs->mdev->ofdev.dev, "[fd%d] " fmt, fs->index, arg)
+#else
+#define swim3_dbg(fmt, arg...)	do { } while(0)
+#endif
+
 static struct floppy_state floppy_states[MAX_FLOPPIES];
 static int floppy_count = 0;
 static DEFINE_SPINLOCK(swim3_lock);
@@ -224,17 +235,8 @@
 	0, 0, 0, 0, 0, 0
 };
 
-static void swim3_select(struct floppy_state *fs, int sel);
-static void swim3_action(struct floppy_state *fs, int action);
-static int swim3_readbit(struct floppy_state *fs, int bit);
-static void do_fd_request(struct request_queue * q);
-static void start_request(struct floppy_state *fs);
-static void set_timeout(struct floppy_state *fs, int nticks,
-			void (*proc)(unsigned long));
-static void scan_track(struct floppy_state *fs);
 static void seek_track(struct floppy_state *fs, int n);
 static void init_dma(struct dbdma_cmd *cp, int cmd, void *buf, int count);
-static void setup_transfer(struct floppy_state *fs);
 static void act(struct floppy_state *fs);
 static void scan_timeout(unsigned long data);
 static void seek_timeout(unsigned long data);
@@ -254,20 +256,23 @@
 					unsigned int clearing);
 static int floppy_revalidate(struct gendisk *disk);
 
-static bool swim3_end_request(int err, unsigned int nr_bytes)
+static bool swim3_end_request(struct floppy_state *fs, int err, unsigned int nr_bytes)
 {
-	if (__blk_end_request(fd_req, err, nr_bytes))
+	struct request *req = fs->cur_req;
+	int rc;
+
+	swim3_dbg("  end request, err=%d nr_bytes=%d, cur_req=%p\n",
+		  err, nr_bytes, req);
+
+	if (err)
+		nr_bytes = blk_rq_cur_bytes(req);
+	rc = __blk_end_request(req, err, nr_bytes);
+	if (rc)
 		return true;
-
-	fd_req = NULL;
+	fs->cur_req = NULL;
 	return false;
 }
 
-static bool swim3_end_request_cur(int err)
-{
-	return swim3_end_request(err, blk_rq_cur_bytes(fd_req));
-}
-
 static void swim3_select(struct floppy_state *fs, int sel)
 {
 	struct swim3 __iomem *sw = fs->swim3;
@@ -303,50 +308,53 @@
 	return (stat & DATA) == 0;
 }
 
-static void do_fd_request(struct request_queue * q)
-{
-	int i;
-
-	for(i=0; i<floppy_count; i++) {
-		struct floppy_state *fs = &floppy_states[i];
-		if (fs->mdev->media_bay &&
-		    check_media_bay(fs->mdev->media_bay) != MB_FD)
-			continue;
-		start_request(fs);
-	}
-}
-
 static void start_request(struct floppy_state *fs)
 {
 	struct request *req;
 	unsigned long x;
 
+	swim3_dbg("start request, initial state=%d\n", fs->state);
+
 	if (fs->state == idle && fs->wanted) {
 		fs->state = available;
 		wake_up(&fs->wait);
 		return;
 	}
 	while (fs->state == idle) {
-		if (!fd_req) {
-			fd_req = blk_fetch_request(swim3_queue);
-			if (!fd_req)
+		swim3_dbg("start request, idle loop, cur_req=%p\n", fs->cur_req);
+		if (!fs->cur_req) {
+			fs->cur_req = blk_fetch_request(disks[fs->index]->queue);
+			swim3_dbg("  fetched request %p\n", fs->cur_req);
+			if (!fs->cur_req)
 				break;
 		}
-		req = fd_req;
-#if 0
-		printk("do_fd_req: dev=%s cmd=%d sec=%ld nr_sec=%u buf=%p\n",
-		       req->rq_disk->disk_name, req->cmd,
-		       (long)blk_rq_pos(req), blk_rq_sectors(req), req->buffer);
-		printk("           errors=%d current_nr_sectors=%u\n",
-		       req->errors, blk_rq_cur_sectors(req));
+		req = fs->cur_req;
+
+		if (fs->mdev->media_bay &&
+		    check_media_bay(fs->mdev->media_bay) != MB_FD) {
+			swim3_dbg("%s", "  media bay absent, dropping req\n");
+			swim3_end_request(fs, -ENODEV, 0);
+			continue;
+		}
+
+#if 0 /* This is really too verbose */
+		swim3_dbg("do_fd_req: dev=%s cmd=%d sec=%ld nr_sec=%u buf=%p\n",
+			  req->rq_disk->disk_name, req->cmd,
+			  (long)blk_rq_pos(req), blk_rq_sectors(req),
+			  req->buffer);
+		swim3_dbg("           errors=%d current_nr_sectors=%u\n",
+			  req->errors, blk_rq_cur_sectors(req));
 #endif
 
 		if (blk_rq_pos(req) >= fs->total_secs) {
-			swim3_end_request_cur(-EIO);
+			swim3_dbg("  pos out of bounds (%ld, max is %ld)\n",
+				  (long)blk_rq_pos(req), (long)fs->total_secs);
+			swim3_end_request(fs, -EIO, 0);
 			continue;
 		}
 		if (fs->ejected) {
-			swim3_end_request_cur(-EIO);
+			swim3_dbg("%s", "  disk ejected\n");
+			swim3_end_request(fs, -EIO, 0);
 			continue;
 		}
 
@@ -354,7 +362,8 @@
 			if (fs->write_prot < 0)
 				fs->write_prot = swim3_readbit(fs, WRITE_PROT);
 			if (fs->write_prot) {
-				swim3_end_request_cur(-EIO);
+				swim3_dbg("%s", "  try to write, disk write protected\n");
+				swim3_end_request(fs, -EIO, 0);
 				continue;
 			}
 		}
@@ -369,7 +378,6 @@
 		x = ((long)blk_rq_pos(req)) % fs->secpercyl;
 		fs->head = x / fs->secpertrack;
 		fs->req_sector = x % fs->secpertrack + 1;
-		fd_req = req;
 		fs->state = do_transfer;
 		fs->retries = 0;
 
@@ -377,12 +385,14 @@
 	}
 }
 
+static void do_fd_request(struct request_queue * q)
+{
+	start_request(q->queuedata);
+}
+
 static void set_timeout(struct floppy_state *fs, int nticks,
 			void (*proc)(unsigned long))
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&fs->lock, flags);
 	if (fs->timeout_pending)
 		del_timer(&fs->timeout);
 	fs->timeout.expires = jiffies + nticks;
@@ -390,7 +400,6 @@
 	fs->timeout.data = (unsigned long) fs;
 	add_timer(&fs->timeout);
 	fs->timeout_pending = 1;
-	spin_unlock_irqrestore(&fs->lock, flags);
 }
 
 static inline void scan_track(struct floppy_state *fs)
@@ -442,40 +451,45 @@
 	struct swim3 __iomem *sw = fs->swim3;
 	struct dbdma_cmd *cp = fs->dma_cmd;
 	struct dbdma_regs __iomem *dr = fs->dma;
+	struct request *req = fs->cur_req;
 
-	if (blk_rq_cur_sectors(fd_req) <= 0) {
-		printk(KERN_ERR "swim3: transfer 0 sectors?\n");
+	if (blk_rq_cur_sectors(req) <= 0) {
+		swim3_warn("%s", "Transfer 0 sectors ?\n");
 		return;
 	}
-	if (rq_data_dir(fd_req) == WRITE)
+	if (rq_data_dir(req) == WRITE)
 		n = 1;
 	else {
 		n = fs->secpertrack - fs->req_sector + 1;
-		if (n > blk_rq_cur_sectors(fd_req))
-			n = blk_rq_cur_sectors(fd_req);
+		if (n > blk_rq_cur_sectors(req))
+			n = blk_rq_cur_sectors(req);
 	}
+
+	swim3_dbg("  setup xfer at sect %d (of %d) head %d for %d\n",
+		  fs->req_sector, fs->secpertrack, fs->head, n);
+
 	fs->scount = n;
 	swim3_select(fs, fs->head? READ_DATA_1: READ_DATA_0);
 	out_8(&sw->sector, fs->req_sector);
 	out_8(&sw->nsect, n);
 	out_8(&sw->gap3, 0);
 	out_le32(&dr->cmdptr, virt_to_bus(cp));
-	if (rq_data_dir(fd_req) == WRITE) {
+	if (rq_data_dir(req) == WRITE) {
 		/* Set up 3 dma commands: write preamble, data, postamble */
 		init_dma(cp, OUTPUT_MORE, write_preamble, sizeof(write_preamble));
 		++cp;
-		init_dma(cp, OUTPUT_MORE, fd_req->buffer, 512);
+		init_dma(cp, OUTPUT_MORE, req->buffer, 512);
 		++cp;
 		init_dma(cp, OUTPUT_LAST, write_postamble, sizeof(write_postamble));
 	} else {
-		init_dma(cp, INPUT_LAST, fd_req->buffer, n * 512);
+		init_dma(cp, INPUT_LAST, req->buffer, n * 512);
 	}
 	++cp;
 	out_le16(&cp->command, DBDMA_STOP);
 	out_8(&sw->control_bic, DO_ACTION | WRITE_SECTORS);
 	in_8(&sw->error);
 	out_8(&sw->control_bic, DO_ACTION | WRITE_SECTORS);
-	if (rq_data_dir(fd_req) == WRITE)
+	if (rq_data_dir(req) == WRITE)
 		out_8(&sw->control_bis, WRITE_SECTORS);
 	in_8(&sw->intr);
 	out_le32(&dr->control, (RUN << 16) | RUN);
@@ -488,12 +502,16 @@
 static void act(struct floppy_state *fs)
 {
 	for (;;) {
+		swim3_dbg("  act loop, state=%d, req_cyl=%d, cur_cyl=%d\n",
+			  fs->state, fs->req_cyl, fs->cur_cyl);
+
 		switch (fs->state) {
 		case idle:
 			return;		/* XXX shouldn't get here */
 
 		case locating:
 			if (swim3_readbit(fs, TRACK_ZERO)) {
+				swim3_dbg("%s", "    locate track 0\n");
 				fs->cur_cyl = 0;
 				if (fs->req_cyl == 0)
 					fs->state = do_transfer;
@@ -511,7 +529,7 @@
 				break;
 			}
 			if (fs->req_cyl == fs->cur_cyl) {
-				printk("whoops, seeking 0\n");
+				swim3_warn("%s", "Whoops, seeking 0\n");
 				fs->state = do_transfer;
 				break;
 			}
@@ -527,7 +545,9 @@
 		case do_transfer:
 			if (fs->cur_cyl != fs->req_cyl) {
 				if (fs->retries > 5) {
-					swim3_end_request_cur(-EIO);
+					swim3_err("Wrong cylinder in transfer, want: %d got %d\n",
+						  fs->req_cyl, fs->cur_cyl);
+					swim3_end_request(fs, -EIO, 0);
 					fs->state = idle;
 					return;
 				}
@@ -542,7 +562,7 @@
 			return;
 
 		default:
-			printk(KERN_ERR"swim3: unknown state %d\n", fs->state);
+			swim3_err("Unknown state %d\n", fs->state);
 			return;
 		}
 	}
@@ -552,59 +572,75 @@
 {
 	struct floppy_state *fs = (struct floppy_state *) data;
 	struct swim3 __iomem *sw = fs->swim3;
+	unsigned long flags;
 
+	swim3_dbg("* scan timeout, state=%d\n", fs->state);
+
+	spin_lock_irqsave(&swim3_lock, flags);
 	fs->timeout_pending = 0;
 	out_8(&sw->control_bic, DO_ACTION | WRITE_SECTORS);
 	out_8(&sw->select, RELAX);
 	out_8(&sw->intr_enable, 0);
 	fs->cur_cyl = -1;
 	if (fs->retries > 5) {
-		swim3_end_request_cur(-EIO);
+		swim3_end_request(fs, -EIO, 0);
 		fs->state = idle;
 		start_request(fs);
 	} else {
 		fs->state = jogging;
 		act(fs);
 	}
+	spin_unlock_irqrestore(&swim3_lock, flags);
 }
 
 static void seek_timeout(unsigned long data)
 {
 	struct floppy_state *fs = (struct floppy_state *) data;
 	struct swim3 __iomem *sw = fs->swim3;
+	unsigned long flags;
 
+	swim3_dbg("* seek timeout, state=%d\n", fs->state);
+
+	spin_lock_irqsave(&swim3_lock, flags);
 	fs->timeout_pending = 0;
 	out_8(&sw->control_bic, DO_SEEK);
 	out_8(&sw->select, RELAX);
 	out_8(&sw->intr_enable, 0);
-	printk(KERN_ERR "swim3: seek timeout\n");
-	swim3_end_request_cur(-EIO);
+	swim3_err("%s", "Seek timeout\n");
+	swim3_end_request(fs, -EIO, 0);
 	fs->state = idle;
 	start_request(fs);
+	spin_unlock_irqrestore(&swim3_lock, flags);
 }
 
 static void settle_timeout(unsigned long data)
 {
 	struct floppy_state *fs = (struct floppy_state *) data;
 	struct swim3 __iomem *sw = fs->swim3;
+	unsigned long flags;
 
+	swim3_dbg("* settle timeout, state=%d\n", fs->state);
+
+	spin_lock_irqsave(&swim3_lock, flags);
 	fs->timeout_pending = 0;
 	if (swim3_readbit(fs, SEEK_COMPLETE)) {
 		out_8(&sw->select, RELAX);
 		fs->state = locating;
 		act(fs);
-		return;
+		goto unlock;
 	}
 	out_8(&sw->select, RELAX);
 	if (fs->settle_time < 2*HZ) {
 		++fs->settle_time;
 		set_timeout(fs, 1, settle_timeout);
-		return;
+		goto unlock;
 	}
-	printk(KERN_ERR "swim3: seek settle timeout\n");
-	swim3_end_request_cur(-EIO);
+	swim3_err("%s", "Seek settle timeout\n");
+	swim3_end_request(fs, -EIO, 0);
 	fs->state = idle;
 	start_request(fs);
+ unlock:
+	spin_unlock_irqrestore(&swim3_lock, flags);
 }
 
 static void xfer_timeout(unsigned long data)
@@ -612,8 +648,12 @@
 	struct floppy_state *fs = (struct floppy_state *) data;
 	struct swim3 __iomem *sw = fs->swim3;
 	struct dbdma_regs __iomem *dr = fs->dma;
+	unsigned long flags;
 	int n;
 
+	swim3_dbg("* xfer timeout, state=%d\n", fs->state);
+
+	spin_lock_irqsave(&swim3_lock, flags);
 	fs->timeout_pending = 0;
 	out_le32(&dr->control, RUN << 16);
 	/* We must wait a bit for dbdma to stop */
@@ -622,12 +662,13 @@
 	out_8(&sw->intr_enable, 0);
 	out_8(&sw->control_bic, WRITE_SECTORS | DO_ACTION);
 	out_8(&sw->select, RELAX);
-	printk(KERN_ERR "swim3: timeout %sing sector %ld\n",
-	       (rq_data_dir(fd_req)==WRITE? "writ": "read"),
-	       (long)blk_rq_pos(fd_req));
-	swim3_end_request_cur(-EIO);
+	swim3_err("Timeout %sing sector %ld\n",
+	       (rq_data_dir(fs->cur_req)==WRITE? "writ": "read"),
+	       (long)blk_rq_pos(fs->cur_req));
+	swim3_end_request(fs, -EIO, 0);
 	fs->state = idle;
 	start_request(fs);
+	spin_unlock_irqrestore(&swim3_lock, flags);
 }
 
 static irqreturn_t swim3_interrupt(int irq, void *dev_id)
@@ -638,12 +679,17 @@
 	int stat, resid;
 	struct dbdma_regs __iomem *dr;
 	struct dbdma_cmd *cp;
+	unsigned long flags;
+	struct request *req = fs->cur_req;
 
+	swim3_dbg("* interrupt, state=%d\n", fs->state);
+
+	spin_lock_irqsave(&swim3_lock, flags);
 	intr = in_8(&sw->intr);
 	err = (intr & ERROR_INTR)? in_8(&sw->error): 0;
 	if ((intr & ERROR_INTR) && fs->state != do_transfer)
-		printk(KERN_ERR "swim3_interrupt, state=%d, dir=%x, intr=%x, err=%x\n",
-		       fs->state, rq_data_dir(fd_req), intr, err);
+		swim3_err("Non-transfer error interrupt: state=%d, dir=%x, intr=%x, err=%x\n",
+			  fs->state, rq_data_dir(req), intr, err);
 	switch (fs->state) {
 	case locating:
 		if (intr & SEEN_SECTOR) {
@@ -653,10 +699,10 @@
 			del_timer(&fs->timeout);
 			fs->timeout_pending = 0;
 			if (sw->ctrack == 0xff) {
-				printk(KERN_ERR "swim3: seen sector but cyl=ff?\n");
+				swim3_err("%s", "Seen sector but cyl=ff?\n");
 				fs->cur_cyl = -1;
 				if (fs->retries > 5) {
-					swim3_end_request_cur(-EIO);
+					swim3_end_request(fs, -EIO, 0);
 					fs->state = idle;
 					start_request(fs);
 				} else {
@@ -668,8 +714,8 @@
 			fs->cur_cyl = sw->ctrack;
 			fs->cur_sector = sw->csect;
 			if (fs->expect_cyl != -1 && fs->expect_cyl != fs->cur_cyl)
-				printk(KERN_ERR "swim3: expected cyl %d, got %d\n",
-				       fs->expect_cyl, fs->cur_cyl);
+				swim3_err("Expected cyl %d, got %d\n",
+					  fs->expect_cyl, fs->cur_cyl);
 			fs->state = do_transfer;
 			act(fs);
 		}
@@ -704,7 +750,7 @@
 		fs->timeout_pending = 0;
 		dr = fs->dma;
 		cp = fs->dma_cmd;
-		if (rq_data_dir(fd_req) == WRITE)
+		if (rq_data_dir(req) == WRITE)
 			++cp;
 		/*
 		 * Check that the main data transfer has finished.
@@ -729,31 +775,32 @@
 		if (intr & ERROR_INTR) {
 			n = fs->scount - 1 - resid / 512;
 			if (n > 0) {
-				blk_update_request(fd_req, 0, n << 9);
+				blk_update_request(req, 0, n << 9);
 				fs->req_sector += n;
 			}
 			if (fs->retries < 5) {
 				++fs->retries;
 				act(fs);
 			} else {
-				printk("swim3: error %sing block %ld (err=%x)\n",
-				       rq_data_dir(fd_req) == WRITE? "writ": "read",
-				       (long)blk_rq_pos(fd_req), err);
-				swim3_end_request_cur(-EIO);
+				swim3_err("Error %sing block %ld (err=%x)\n",
+				       rq_data_dir(req) == WRITE? "writ": "read",
+				       (long)blk_rq_pos(req), err);
+				swim3_end_request(fs, -EIO, 0);
 				fs->state = idle;
 			}
 		} else {
 			if ((stat & ACTIVE) == 0 || resid != 0) {
 				/* musta been an error */
-				printk(KERN_ERR "swim3: fd dma: stat=%x resid=%d\n", stat, resid);
-				printk(KERN_ERR "  state=%d, dir=%x, intr=%x, err=%x\n",
-				       fs->state, rq_data_dir(fd_req), intr, err);
-				swim3_end_request_cur(-EIO);
+				swim3_err("fd dma error: stat=%x resid=%d\n", stat, resid);
+				swim3_err("  state=%d, dir=%x, intr=%x, err=%x\n",
+					  fs->state, rq_data_dir(req), intr, err);
+				swim3_end_request(fs, -EIO, 0);
 				fs->state = idle;
 				start_request(fs);
 				break;
 			}
-			if (swim3_end_request(0, fs->scount << 9)) {
+			fs->retries = 0;
+			if (swim3_end_request(fs, 0, fs->scount << 9)) {
 				fs->req_sector += fs->scount;
 				if (fs->req_sector > fs->secpertrack) {
 					fs->req_sector -= fs->secpertrack;
@@ -770,8 +817,9 @@
 			start_request(fs);
 		break;
 	default:
-		printk(KERN_ERR "swim3: don't know what to do in state %d\n", fs->state);
+		swim3_err("Don't know what to do in state %d\n", fs->state);
 	}
+	spin_unlock_irqrestore(&swim3_lock, flags);
 	return IRQ_HANDLED;
 }
 
@@ -781,26 +829,31 @@
 }
 */
 
+/* Called under the mutex to grab exclusive access to a drive */
 static int grab_drive(struct floppy_state *fs, enum swim_state state,
 		      int interruptible)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&fs->lock, flags);
-	if (fs->state != idle) {
+	swim3_dbg("%s", "-> grab drive\n");
+
+	spin_lock_irqsave(&swim3_lock, flags);
+	if (fs->state != idle && fs->state != available) {
 		++fs->wanted;
 		while (fs->state != available) {
+			spin_unlock_irqrestore(&swim3_lock, flags);
 			if (interruptible && signal_pending(current)) {
 				--fs->wanted;
-				spin_unlock_irqrestore(&fs->lock, flags);
 				return -EINTR;
 			}
 			interruptible_sleep_on(&fs->wait);
+			spin_lock_irqsave(&swim3_lock, flags);
 		}
 		--fs->wanted;
 	}
 	fs->state = state;
-	spin_unlock_irqrestore(&fs->lock, flags);
+	spin_unlock_irqrestore(&swim3_lock, flags);
+
 	return 0;
 }
 
@@ -808,10 +861,12 @@
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&fs->lock, flags);
+	swim3_dbg("%s", "-> release drive\n");
+
+	spin_lock_irqsave(&swim3_lock, flags);
 	fs->state = idle;
 	start_request(fs);
-	spin_unlock_irqrestore(&fs->lock, flags);
+	spin_unlock_irqrestore(&swim3_lock, flags);
 }
 
 static int fd_eject(struct floppy_state *fs)
@@ -966,6 +1021,7 @@
 {
 	struct floppy_state *fs = disk->private_data;
 	struct swim3 __iomem *sw = fs->swim3;
+
 	mutex_lock(&swim3_mutex);
 	if (fs->ref_count > 0 && --fs->ref_count == 0) {
 		swim3_action(fs, MOTOR_OFF);
@@ -1031,30 +1087,48 @@
 	.revalidate_disk= floppy_revalidate,
 };
 
+static void swim3_mb_event(struct macio_dev* mdev, int mb_state)
+{
+	struct floppy_state *fs = macio_get_drvdata(mdev);
+	struct swim3 __iomem *sw = fs->swim3;
+
+	if (!fs)
+		return;
+	if (mb_state != MB_FD)
+		return;
+
+	/* Clear state */
+	out_8(&sw->intr_enable, 0);
+	in_8(&sw->intr);
+	in_8(&sw->error);
+}
+
 static int swim3_add_device(struct macio_dev *mdev, int index)
 {
 	struct device_node *swim = mdev->ofdev.dev.of_node;
 	struct floppy_state *fs = &floppy_states[index];
 	int rc = -EBUSY;
 
+	/* Do this first for message macros */
+	memset(fs, 0, sizeof(*fs));
+	fs->mdev = mdev;
+	fs->index = index;
+
 	/* Check & Request resources */
 	if (macio_resource_count(mdev) < 2) {
-		printk(KERN_WARNING "ifd%d: no address for %s\n",
-		       index, swim->full_name);
+		swim3_err("%s", "No address in device-tree\n");
 		return -ENXIO;
 	}
-	if (macio_irq_count(mdev) < 2) {
-		printk(KERN_WARNING "fd%d: no intrs for device %s\n",
-			index, swim->full_name);
+	if (macio_irq_count(mdev) < 1) {
+		swim3_err("%s", "No interrupt in device-tree\n");
+		return -ENXIO;
 	}
 	if (macio_request_resource(mdev, 0, "swim3 (mmio)")) {
-		printk(KERN_ERR "fd%d: can't request mmio resource for %s\n",
-		       index, swim->full_name);
+		swim3_err("%s", "Can't request mmio resource\n");
 		return -EBUSY;
 	}
 	if (macio_request_resource(mdev, 1, "swim3 (dma)")) {
-		printk(KERN_ERR "fd%d: can't request dma resource for %s\n",
-		       index, swim->full_name);
+		swim3_err("%s", "Can't request dma resource\n");
 		macio_release_resource(mdev, 0);
 		return -EBUSY;
 	}
@@ -1063,22 +1137,18 @@
 	if (mdev->media_bay == NULL)
 		pmac_call_feature(PMAC_FTR_SWIM3_ENABLE, swim, 0, 1);
 	
-	memset(fs, 0, sizeof(*fs));
-	spin_lock_init(&fs->lock);
 	fs->state = idle;
 	fs->swim3 = (struct swim3 __iomem *)
 		ioremap(macio_resource_start(mdev, 0), 0x200);
 	if (fs->swim3 == NULL) {
-		printk("fd%d: couldn't map registers for %s\n",
-		       index, swim->full_name);
+		swim3_err("%s", "Couldn't map mmio registers\n");
 		rc = -ENOMEM;
 		goto out_release;
 	}
 	fs->dma = (struct dbdma_regs __iomem *)
 		ioremap(macio_resource_start(mdev, 1), 0x200);
 	if (fs->dma == NULL) {
-		printk("fd%d: couldn't map DMA for %s\n",
-		       index, swim->full_name);
+		swim3_err("%s", "Couldn't map dma registers\n");
 		iounmap(fs->swim3);
 		rc = -ENOMEM;
 		goto out_release;
@@ -1090,31 +1160,25 @@
 	fs->secpercyl = 36;
 	fs->secpertrack = 18;
 	fs->total_secs = 2880;
-	fs->mdev = mdev;
 	init_waitqueue_head(&fs->wait);
 
 	fs->dma_cmd = (struct dbdma_cmd *) DBDMA_ALIGN(fs->dbdma_cmd_space);
 	memset(fs->dma_cmd, 0, 2 * sizeof(struct dbdma_cmd));
 	st_le16(&fs->dma_cmd[1].command, DBDMA_STOP);
 
+	if (mdev->media_bay == NULL || check_media_bay(mdev->media_bay) == MB_FD)
+		swim3_mb_event(mdev, MB_FD);
+
 	if (request_irq(fs->swim3_intr, swim3_interrupt, 0, "SWIM3", fs)) {
-		printk(KERN_ERR "fd%d: couldn't request irq %d for %s\n",
-		       index, fs->swim3_intr, swim->full_name);
+		swim3_err("%s", "Couldn't request interrupt\n");
 		pmac_call_feature(PMAC_FTR_SWIM3_ENABLE, swim, 0, 0);
 		goto out_unmap;
 		return -EBUSY;
 	}
-/*
-	if (request_irq(fs->dma_intr, fd_dma_interrupt, 0, "SWIM3-dma", fs)) {
-		printk(KERN_ERR "Couldn't get irq %d for SWIM3 DMA",
-		       fs->dma_intr);
-		return -EBUSY;
-	}
-*/
 
 	init_timer(&fs->timeout);
 
-	printk(KERN_INFO "fd%d: SWIM3 floppy controller %s\n", floppy_count,
+	swim3_info("SWIM3 floppy controller %s\n",
 		mdev->media_bay ? "in media bay" : "");
 
 	return 0;
@@ -1132,41 +1196,42 @@
 
 static int __devinit swim3_attach(struct macio_dev *mdev, const struct of_device_id *match)
 {
-	int i, rc;
 	struct gendisk *disk;
+	int index, rc;
+
+	index = floppy_count++;
+	if (index >= MAX_FLOPPIES)
+		return -ENXIO;
 
 	/* Add the drive */
-	rc = swim3_add_device(mdev, floppy_count);
+	rc = swim3_add_device(mdev, index);
 	if (rc)
 		return rc;
+	/* Now register that disk. Same comment about failure handling */
+	disk = disks[index] = alloc_disk(1);
+	if (disk == NULL)
+		return -ENOMEM;
+	disk->queue = blk_init_queue(do_fd_request, &swim3_lock);
+	if (disk->queue == NULL) {
+		put_disk(disk);
+		return -ENOMEM;
+	}
+	disk->queue->queuedata = &floppy_states[index];
 
-	/* Now create the queue if not there yet */
-	if (swim3_queue == NULL) {
+	if (index == 0) {
 		/* If we failed, there isn't much we can do as the driver is still
 		 * too dumb to remove the device, just bail out
 		 */
 		if (register_blkdev(FLOPPY_MAJOR, "fd"))
 			return 0;
-		swim3_queue = blk_init_queue(do_fd_request, &swim3_lock);
-		if (swim3_queue == NULL) {
-			unregister_blkdev(FLOPPY_MAJOR, "fd");
-			return 0;
-		}
 	}
 
-	/* Now register that disk. Same comment about failure handling */
-	i = floppy_count++;
-	disk = disks[i] = alloc_disk(1);
-	if (disk == NULL)
-		return 0;
-
 	disk->major = FLOPPY_MAJOR;
-	disk->first_minor = i;
+	disk->first_minor = index;
 	disk->fops = &floppy_fops;
-	disk->private_data = &floppy_states[i];
-	disk->queue = swim3_queue;
+	disk->private_data = &floppy_states[index];
 	disk->flags |= GENHD_FL_REMOVABLE;
-	sprintf(disk->disk_name, "fd%d", i);
+	sprintf(disk->disk_name, "fd%d", index);
 	set_capacity(disk, 2880);
 	add_disk(disk);
 
@@ -1194,6 +1259,9 @@
 		.of_match_table	= swim3_match,
 	},
 	.probe		= swim3_attach,
+#ifdef CONFIG_PMAC_MEDIABAY
+	.mediabay_event	= swim3_mb_event,
+#endif
 #if 0
 	.suspend	= swim3_suspend,
 	.resume		= swim3_resume,
diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
index 11b41fd..5ccf142 100644
--- a/drivers/bluetooth/Kconfig
+++ b/drivers/bluetooth/Kconfig
@@ -188,7 +188,7 @@
 	  The core driver to support Marvell Bluetooth devices.
 
 	  This driver is required if you want to support
-	  Marvell Bluetooth devices, such as 8688/8787.
+	  Marvell Bluetooth devices, such as 8688/8787/8797.
 
 	  Say Y here to compile Marvell Bluetooth driver
 	  into the kernel or say M to compile it as module.
@@ -201,8 +201,8 @@
 	  The driver for Marvell Bluetooth chipsets with SDIO interface.
 
 	  This driver is required if you want to use Marvell Bluetooth
-	  devices with SDIO interface. Currently SD8688/SD8787 chipsets are
-	  supported.
+	  devices with SDIO interface. Currently SD8688/SD8787/SD8797
+	  chipsets are supported.
 
 	  Say Y here to compile support for Marvell BT-over-SDIO driver
 	  into the kernel or say M to compile it as module.
diff --git a/drivers/bluetooth/btmrvl_sdio.c b/drivers/bluetooth/btmrvl_sdio.c
index 9ef4816..27b74b0 100644
--- a/drivers/bluetooth/btmrvl_sdio.c
+++ b/drivers/bluetooth/btmrvl_sdio.c
@@ -65,7 +65,7 @@
 	.io_port_1 = 0x01,
 	.io_port_2 = 0x02,
 };
-static const struct btmrvl_sdio_card_reg btmrvl_reg_8787 = {
+static const struct btmrvl_sdio_card_reg btmrvl_reg_87xx = {
 	.cfg = 0x00,
 	.host_int_mask = 0x02,
 	.host_intstatus = 0x03,
@@ -92,7 +92,14 @@
 static const struct btmrvl_sdio_device btmrvl_sdio_sd8787 = {
 	.helper		= NULL,
 	.firmware	= "mrvl/sd8787_uapsta.bin",
-	.reg		= &btmrvl_reg_8787,
+	.reg		= &btmrvl_reg_87xx,
+	.sd_blksz_fw_dl	= 256,
+};
+
+static const struct btmrvl_sdio_device btmrvl_sdio_sd8797 = {
+	.helper		= NULL,
+	.firmware	= "mrvl/sd8797_uapsta.bin",
+	.reg		= &btmrvl_reg_87xx,
 	.sd_blksz_fw_dl	= 256,
 };
 
@@ -103,6 +110,9 @@
 	/* Marvell SD8787 Bluetooth device */
 	{ SDIO_DEVICE(SDIO_VENDOR_ID_MARVELL, 0x911A),
 			.driver_data = (unsigned long) &btmrvl_sdio_sd8787 },
+	/* Marvell SD8797 Bluetooth device */
+	{ SDIO_DEVICE(SDIO_VENDOR_ID_MARVELL, 0x912A),
+			.driver_data = (unsigned long) &btmrvl_sdio_sd8797 },
 
 	{ }	/* Terminating entry */
 };
@@ -1076,3 +1086,4 @@
 MODULE_FIRMWARE("sd8688_helper.bin");
 MODULE_FIRMWARE("sd8688.bin");
 MODULE_FIRMWARE("mrvl/sd8787_uapsta.bin");
+MODULE_FIRMWARE("mrvl/sd8797_uapsta.bin");
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index f9b7260..eabc437 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -100,6 +100,9 @@
 	/* Canyon CN-BTU1 with HID interfaces */
 	{ USB_DEVICE(0x0c10, 0x0000) },
 
+	/* Broadcom BCM20702A0 */
+	{ USB_DEVICE(0x413c, 0x8197) },
+
 	{ }	/* Terminating entry */
 };
 
@@ -774,9 +777,8 @@
 		usb_mark_last_busy(data->udev);
 	}
 
-	usb_free_urb(urb);
-
 done:
+	usb_free_urb(urb);
 	return err;
 }
 
diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index 66cd0b8..c92424c 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -1186,10 +1186,11 @@
 /* Certain Gen5 chipsets require require idling the GPU before
  * unmapping anything from the GTT when VT-d is enabled.
  */
-extern int intel_iommu_gfx_mapped;
 static inline int needs_idle_maps(void)
 {
+#ifdef CONFIG_INTEL_IOMMU
 	const unsigned short gpu_devid = intel_private.pcidev->device;
+	extern int intel_iommu_gfx_mapped;
 
 	/* Query intel_iommu to see if we need the workaround. Presumably that
 	 * was loaded first.
@@ -1198,7 +1199,7 @@
 	     gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG) &&
 	     intel_iommu_gfx_mapped)
 		return 1;
-
+#endif
 	return 0;
 }
 
@@ -1236,7 +1237,7 @@
 		intel_private.gtt_bus_addr = reg_addr + gtt_offset;
 	}
 
-	if (needs_idle_maps());
+	if (needs_idle_maps())
 		intel_private.base.do_idle_maps = 1;
 
 	intel_i9xx_setup_flush();
diff --git a/drivers/char/random.c b/drivers/char/random.c
index 63e19ba..6035ab8 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -941,7 +941,7 @@
 		if (!arch_get_random_long(&v))
 			break;
 		
-		memcpy(buf, &v, chunk);
+		memcpy(p, &v, chunk);
 		p += chunk;
 		nbytes -= chunk;
 	}
diff --git a/drivers/cpufreq/db8500-cpufreq.c b/drivers/cpufreq/db8500-cpufreq.c
index edaa987..f500201 100644
--- a/drivers/cpufreq/db8500-cpufreq.c
+++ b/drivers/cpufreq/db8500-cpufreq.c
@@ -109,7 +109,7 @@
 
 static int __cpuinit db8500_cpufreq_init(struct cpufreq_policy *policy)
 {
-	int res;
+	int i, res;
 
 	BUILD_BUG_ON(ARRAY_SIZE(idx2opp) + 1 != ARRAY_SIZE(freq_table));
 
@@ -120,8 +120,8 @@
 			freq_table[3].frequency = 1000000;
 	}
 	pr_info("db8500-cpufreq : Available frequencies:\n");
-	while (freq_table[i].frequency != CPUFREQ_TABLE_END)
-		pr_info("  %d Mhz\n", freq_table[i++].frequency/1000);
+	for (i = 0; freq_table[i].frequency != CPUFREQ_TABLE_END; i++)
+		pr_info("  %d Mhz\n", freq_table[i].frequency/1000);
 
 	/* get policy fields based on the table */
 	res = cpufreq_frequency_table_cpuinfo(policy, freq_table);
diff --git a/drivers/crypto/mv_cesa.c b/drivers/crypto/mv_cesa.c
index 5c6f56f..dcd8bab 100644
--- a/drivers/crypto/mv_cesa.c
+++ b/drivers/crypto/mv_cesa.c
@@ -343,11 +343,13 @@
 		else
 			op.config |= CFG_MID_FRAG;
 
-		writel(req_ctx->state[0], cpg->reg + DIGEST_INITIAL_VAL_A);
-		writel(req_ctx->state[1], cpg->reg + DIGEST_INITIAL_VAL_B);
-		writel(req_ctx->state[2], cpg->reg + DIGEST_INITIAL_VAL_C);
-		writel(req_ctx->state[3], cpg->reg + DIGEST_INITIAL_VAL_D);
-		writel(req_ctx->state[4], cpg->reg + DIGEST_INITIAL_VAL_E);
+		if (first_block) {
+			writel(req_ctx->state[0], cpg->reg + DIGEST_INITIAL_VAL_A);
+			writel(req_ctx->state[1], cpg->reg + DIGEST_INITIAL_VAL_B);
+			writel(req_ctx->state[2], cpg->reg + DIGEST_INITIAL_VAL_C);
+			writel(req_ctx->state[3], cpg->reg + DIGEST_INITIAL_VAL_D);
+			writel(req_ctx->state[4], cpg->reg + DIGEST_INITIAL_VAL_E);
+		}
 	}
 
 	memcpy(cpg->sram + SRAM_CONFIG, &op, sizeof(struct sec_accel_config));
diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig
index 643b055..8f04910 100644
--- a/drivers/devfreq/Kconfig
+++ b/drivers/devfreq/Kconfig
@@ -1,36 +1,29 @@
-config ARCH_HAS_DEVFREQ
-	bool
-	depends on ARCH_HAS_OPP
-	help
-	  Denotes that the architecture supports DEVFREQ. If the architecture
-	  supports multiple OPP entries per device and the frequency of the
-	  devices with OPPs may be altered dynamically, the architecture
-	  supports DEVFREQ.
-
 menuconfig PM_DEVFREQ
 	bool "Generic Dynamic Voltage and Frequency Scaling (DVFS) support"
-	depends on PM_OPP && ARCH_HAS_DEVFREQ
 	help
-	  With OPP support, a device may have a list of frequencies and
-	  voltages available. DEVFREQ, a generic DVFS framework can be
-	  registered for a device with OPP support in order to let the
-	  governor provided to DEVFREQ choose an operating frequency
-	  based on the OPP's list and the policy given with DEVFREQ.
+	  A device may have a list of frequencies and voltages available.
+	  devfreq, a generic DVFS framework can be registered for a device
+	  in order to let the governor provided to devfreq choose an
+	  operating frequency based on the device driver's policy.
 
-	  Each device may have its own governor and policy. DEVFREQ can
+	  Each device may have its own governor and policy. Devfreq can
 	  reevaluate the device state periodically and/or based on the
-	  OPP list changes (each frequency/voltage pair in OPP may be
-	  disabled or enabled).
+	  notification to "nb", a notifier block, of devfreq.
 
-	  Like some CPUs with CPUFREQ, a device may have multiple clocks.
+	  Like some CPUs with CPUfreq, a device may have multiple clocks.
 	  However, because the clock frequencies of a single device are
-	  determined by the single device's state, an instance of DEVFREQ
+	  determined by the single device's state, an instance of devfreq
 	  is attached to a single device and returns a "representative"
-	  clock frequency from the OPP of the device, which is also attached
-	  to a device by 1-to-1. The device registering DEVFREQ takes the
-	  responsiblity to "interpret" the frequency listed in OPP and
+	  clock frequency of the device, which is also attached
+	  to a device by 1-to-1. The device registering devfreq takes the
+	  responsiblity to "interpret" the representative frequency and
 	  to set its every clock accordingly with the "target" callback
-	  given to DEVFREQ.
+	  given to devfreq.
+
+	  When OPP is used with the devfreq device, it is recommended to
+	  register devfreq's nb to the OPP's notifier head.  If OPP is
+	  used with the devfreq device, you may use OPP helper
+	  functions defined in devfreq.h.
 
 if PM_DEVFREQ
 
diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c
index 5d15b81..59d24e9 100644
--- a/drivers/devfreq/devfreq.c
+++ b/drivers/devfreq/devfreq.c
@@ -15,7 +15,9 @@
 #include <linux/errno.h>
 #include <linux/err.h>
 #include <linux/init.h>
+#include <linux/module.h>
 #include <linux/slab.h>
+#include <linux/stat.h>
 #include <linux/opp.h>
 #include <linux/devfreq.h>
 #include <linux/workqueue.h>
@@ -416,10 +418,14 @@
  */
 int devfreq_remove_device(struct devfreq *devfreq)
 {
+	bool central_polling;
+
 	if (!devfreq)
 		return -EINVAL;
 
-	if (!devfreq->governor->no_central_polling) {
+	central_polling = !devfreq->governor->no_central_polling;
+
+	if (central_polling) {
 		mutex_lock(&devfreq_list_lock);
 		while (wait_remove_device == devfreq) {
 			mutex_unlock(&devfreq_list_lock);
@@ -431,7 +437,7 @@
 	mutex_lock(&devfreq->lock);
 	_remove_devfreq(devfreq, false); /* it unlocks devfreq->lock */
 
-	if (!devfreq->governor->no_central_polling)
+	if (central_polling)
 		mutex_unlock(&devfreq_list_lock);
 
 	return 0;
diff --git a/drivers/edac/mpc85xx_edac.c b/drivers/edac/mpc85xx_edac.c
index 8af8e864..73464a6 100644
--- a/drivers/edac/mpc85xx_edac.c
+++ b/drivers/edac/mpc85xx_edac.c
@@ -1128,7 +1128,7 @@
 	{ .compatible = "fsl,p1020-memory-controller", },
 	{ .compatible = "fsl,p1021-memory-controller", },
 	{ .compatible = "fsl,p2020-memory-controller", },
-	{ .compatible = "fsl,p4080-memory-controller", },
+	{ .compatible = "fsl,qoriq-memory-controller", },
 	{},
 };
 MODULE_DEVICE_TABLE(of, mpc85xx_mc_err_of_match);
diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
index bcb1126..153980b 100644
--- a/drivers/firmware/dmi_scan.c
+++ b/drivers/firmware/dmi_scan.c
@@ -585,14 +585,12 @@
 }
 
 /**
- *	dmi_name_in_vendors - Check if string is anywhere in the DMI vendor information.
+ *	dmi_name_in_vendors - Check if string is in the DMI system or board vendor name
  *	@str: 	Case sensitive Name
  */
 int dmi_name_in_vendors(const char *str)
 {
-	static int fields[] = { DMI_BIOS_VENDOR, DMI_BIOS_VERSION, DMI_SYS_VENDOR,
-				DMI_PRODUCT_NAME, DMI_PRODUCT_VERSION, DMI_BOARD_VENDOR,
-				DMI_BOARD_NAME, DMI_BOARD_VERSION, DMI_NONE };
+	static int fields[] = { DMI_SYS_VENDOR, DMI_BOARD_VENDOR, DMI_NONE };
 	int i;
 	for (i = 0; fields[i] != DMI_NONE; i++) {
 		int f = fields[i];
diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c
index 8370f72..b0a8117 100644
--- a/drivers/firmware/efivars.c
+++ b/drivers/firmware/efivars.c
@@ -457,7 +457,8 @@
 }
 
 static ssize_t efi_pstore_read(u64 *id, enum pstore_type_id *type,
-			       struct timespec *timespec, struct pstore_info *psi)
+			       struct timespec *timespec,
+			       char **buf, struct pstore_info *psi)
 {
 	efi_guid_t vendor = LINUX_EFI_CRASH_GUID;
 	struct efivars *efivars = psi->data;
@@ -478,7 +479,11 @@
 				timespec->tv_nsec = 0;
 				get_var_data_locked(efivars, &efivars->walk_entry->var);
 				size = efivars->walk_entry->var.DataSize;
-				memcpy(psi->buf, efivars->walk_entry->var.Data, size);
+				*buf = kmalloc(size, GFP_KERNEL);
+				if (*buf == NULL)
+					return -ENOMEM;
+				memcpy(*buf, efivars->walk_entry->var.Data,
+				       size);
 				efivars->walk_entry = list_entry(efivars->walk_entry->list.next,
 					           struct efivar_entry, list);
 				return size;
@@ -576,7 +581,8 @@
 }
 
 static ssize_t efi_pstore_read(u64 *id, enum pstore_type_id *type,
-			       struct timespec *time, struct pstore_info *psi)
+			       struct timespec *timespec,
+			       char **buf, struct pstore_info *psi)
 {
 	return -1;
 }
diff --git a/drivers/firmware/iscsi_ibft.c b/drivers/firmware/iscsi_ibft.c
index c811cb1..2cce44a 100644
--- a/drivers/firmware/iscsi_ibft.c
+++ b/drivers/firmware/iscsi_ibft.c
@@ -746,6 +746,37 @@
 	ibft_cleanup();
 }
 
+#ifdef CONFIG_ACPI
+static const struct {
+	char *sign;
+} ibft_signs[] = {
+	/*
+	 * One spec says "IBFT", the other says "iBFT". We have to check
+	 * for both.
+	 */
+	{ ACPI_SIG_IBFT },
+	{ "iBFT" },
+};
+
+static void __init acpi_find_ibft_region(void)
+{
+	int i;
+	struct acpi_table_header *table = NULL;
+
+	if (acpi_disabled)
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(ibft_signs) && !ibft_addr; i++) {
+		acpi_get_table(ibft_signs[i].sign, 0, &table);
+		ibft_addr = (struct acpi_table_ibft *)table;
+	}
+}
+#else
+static void __init acpi_find_ibft_region(void)
+{
+}
+#endif
+
 /*
  * ibft_init() - creates sysfs tree entries for the iBFT data.
  */
@@ -753,9 +784,16 @@
 {
 	int rc = 0;
 
+	/*
+	   As on UEFI systems the setup_arch()/find_ibft_region()
+	   is called before ACPI tables are parsed and it only does
+	   legacy finding.
+	*/
+	if (!ibft_addr)
+		acpi_find_ibft_region();
+
 	if (ibft_addr) {
-		printk(KERN_INFO "iBFT detected at 0x%llx.\n",
-		       (u64)isa_virt_to_bus(ibft_addr));
+		pr_info("iBFT detected.\n");
 
 		rc = ibft_check_device();
 		if (rc)
diff --git a/drivers/firmware/iscsi_ibft_find.c b/drivers/firmware/iscsi_ibft_find.c
index bfe7232..4da4eb9a 100644
--- a/drivers/firmware/iscsi_ibft_find.c
+++ b/drivers/firmware/iscsi_ibft_find.c
@@ -45,13 +45,6 @@
 static const struct {
 	char *sign;
 } ibft_signs[] = {
-#ifdef CONFIG_ACPI
-	/*
-	 * One spec says "IBFT", the other says "iBFT". We have to check
-	 * for both.
-	 */
-	{ ACPI_SIG_IBFT },
-#endif
 	{ "iBFT" },
 	{ "BIFT" },	/* Broadcom iSCSI Offload */
 };
@@ -62,14 +55,6 @@
 #define VGA_MEM 0xA0000 /* VGA buffer */
 #define VGA_SIZE 0x20000 /* 128kB */
 
-#ifdef CONFIG_ACPI
-static int __init acpi_find_ibft(struct acpi_table_header *header)
-{
-	ibft_addr = (struct acpi_table_ibft *)header;
-	return 0;
-}
-#endif /* CONFIG_ACPI */
-
 static int __init find_ibft_in_mem(void)
 {
 	unsigned long pos;
@@ -94,6 +79,7 @@
 				 * the table cannot be valid. */
 				if (pos + len <= (IBFT_END-1)) {
 					ibft_addr = (struct acpi_table_ibft *)virt;
+					pr_info("iBFT found at 0x%lx.\n", pos);
 					goto done;
 				}
 			}
@@ -108,20 +94,12 @@
  */
 unsigned long __init find_ibft_region(unsigned long *sizep)
 {
-#ifdef CONFIG_ACPI
-	int i;
-#endif
 	ibft_addr = NULL;
 
-#ifdef CONFIG_ACPI
-	for (i = 0; i < ARRAY_SIZE(ibft_signs) && !ibft_addr; i++)
-		acpi_table_parse(ibft_signs[i].sign, acpi_find_ibft);
-#endif /* CONFIG_ACPI */
-
 	/* iBFT 1.03 section 1.4.3.1 mandates that UEFI machines will
 	 * only use ACPI for this */
 
-	if (!ibft_addr && !efi_enabled)
+	if (!efi_enabled)
 		find_ibft_in_mem();
 
 	if (ibft_addr) {
diff --git a/drivers/firmware/sigma.c b/drivers/firmware/sigma.c
index f10fc52..1eedb6f 100644
--- a/drivers/firmware/sigma.c
+++ b/drivers/firmware/sigma.c
@@ -14,13 +14,34 @@
 #include <linux/module.h>
 #include <linux/sigma.h>
 
-/* Return: 0==OK, <0==error, =1 ==no more actions */
-static int
-process_sigma_action(struct i2c_client *client, struct sigma_firmware *ssfw)
+static size_t sigma_action_size(struct sigma_action *sa)
 {
-	struct sigma_action *sa = (void *)(ssfw->fw->data + ssfw->pos);
+	size_t payload = 0;
+
+	switch (sa->instr) {
+	case SIGMA_ACTION_WRITEXBYTES:
+	case SIGMA_ACTION_WRITESINGLE:
+	case SIGMA_ACTION_WRITESAFELOAD:
+		payload = sigma_action_len(sa);
+		break;
+	default:
+		break;
+	}
+
+	payload = ALIGN(payload, 2);
+
+	return payload + sizeof(struct sigma_action);
+}
+
+/*
+ * Returns a negative error value in case of an error, 0 if processing of
+ * the firmware should be stopped after this action, 1 otherwise.
+ */
+static int
+process_sigma_action(struct i2c_client *client, struct sigma_action *sa)
+{
 	size_t len = sigma_action_len(sa);
-	int ret = 0;
+	int ret;
 
 	pr_debug("%s: instr:%i addr:%#x len:%zu\n", __func__,
 		sa->instr, sa->addr, len);
@@ -29,44 +50,50 @@
 	case SIGMA_ACTION_WRITEXBYTES:
 	case SIGMA_ACTION_WRITESINGLE:
 	case SIGMA_ACTION_WRITESAFELOAD:
-		if (ssfw->fw->size < ssfw->pos + len)
-			return -EINVAL;
 		ret = i2c_master_send(client, (void *)&sa->addr, len);
 		if (ret < 0)
 			return -EINVAL;
 		break;
-
 	case SIGMA_ACTION_DELAY:
-		ret = 0;
 		udelay(len);
 		len = 0;
 		break;
-
 	case SIGMA_ACTION_END:
-		return 1;
-
+		return 0;
 	default:
 		return -EINVAL;
 	}
 
-	/* when arrive here ret=0 or sent data */
-	ssfw->pos += sigma_action_size(sa, len);
-	return ssfw->pos == ssfw->fw->size;
+	return 1;
 }
 
 static int
 process_sigma_actions(struct i2c_client *client, struct sigma_firmware *ssfw)
 {
-	pr_debug("%s: processing %p\n", __func__, ssfw);
+	struct sigma_action *sa;
+	size_t size;
+	int ret;
 
-	while (1) {
-		int ret = process_sigma_action(client, ssfw);
+	while (ssfw->pos + sizeof(*sa) <= ssfw->fw->size) {
+		sa = (struct sigma_action *)(ssfw->fw->data + ssfw->pos);
+
+		size = sigma_action_size(sa);
+		ssfw->pos += size;
+		if (ssfw->pos > ssfw->fw->size || size == 0)
+			break;
+
+		ret = process_sigma_action(client, sa);
+
 		pr_debug("%s: action returned %i\n", __func__, ret);
-		if (ret == 1)
-			return 0;
-		else if (ret)
+
+		if (ret <= 0)
 			return ret;
 	}
+
+	if (ssfw->pos != ssfw->fw->size)
+		return -EINVAL;
+
+	return 0;
 }
 
 int process_sigma_firmware(struct i2c_client *client, const char *name)
@@ -89,16 +116,24 @@
 
 	/* then verify the header */
 	ret = -EINVAL;
-	if (fw->size < sizeof(*ssfw_head))
+
+	/*
+	 * Reject too small or unreasonable large files. The upper limit has been
+	 * chosen a bit arbitrarily, but it should be enough for all practical
+	 * purposes and having the limit makes it easier to avoid integer
+	 * overflows later in the loading process.
+	 */
+	if (fw->size < sizeof(*ssfw_head) || fw->size >= 0x4000000)
 		goto done;
 
 	ssfw_head = (void *)fw->data;
 	if (memcmp(ssfw_head->magic, SIGMA_MAGIC, ARRAY_SIZE(ssfw_head->magic)))
 		goto done;
 
-	crc = crc32(0, fw->data, fw->size);
+	crc = crc32(0, fw->data + sizeof(*ssfw_head),
+			fw->size - sizeof(*ssfw_head));
 	pr_debug("%s: crc=%x\n", __func__, crc);
-	if (crc != ssfw_head->crc)
+	if (crc != le32_to_cpu(ssfw_head->crc))
 		goto done;
 
 	ssfw.pos = sizeof(*ssfw_head);
diff --git a/drivers/gpio/Makefile b/drivers/gpio/Makefile
index dbcb0bc..4e018d6 100644
--- a/drivers/gpio/Makefile
+++ b/drivers/gpio/Makefile
@@ -18,7 +18,7 @@
 obj-$(CONFIG_GPIO_EP93XX)	+= gpio-ep93xx.o
 obj-$(CONFIG_GPIO_IT8761E)	+= gpio-it8761e.o
 obj-$(CONFIG_GPIO_JANZ_TTL)	+= gpio-janz-ttl.o
-obj-$(CONFIG_MACH_KS8695)	+= gpio-ks8695.o
+obj-$(CONFIG_ARCH_KS8695)	+= gpio-ks8695.o
 obj-$(CONFIG_GPIO_LANGWELL)	+= gpio-langwell.o
 obj-$(CONFIG_ARCH_LPC32XX)	+= gpio-lpc32xx.o
 obj-$(CONFIG_GPIO_MAX730X)	+= gpio-max730x.o
diff --git a/drivers/gpio/gpio-da9052.c b/drivers/gpio/gpio-da9052.c
index 038f5eb8..f8ce29e 100644
--- a/drivers/gpio/gpio-da9052.c
+++ b/drivers/gpio/gpio-da9052.c
@@ -22,7 +22,6 @@
 #include <linux/mfd/da9052/da9052.h>
 #include <linux/mfd/da9052/reg.h>
 #include <linux/mfd/da9052/pdata.h>
-#include <linux/mfd/da9052/gpio.h>
 
 #define DA9052_INPUT				1
 #define DA9052_OUTPUT_OPENDRAIN		2
@@ -43,6 +42,9 @@
 #define DA9052_GPIO_MASK_UPPER_NIBBLE		0xF0
 #define DA9052_GPIO_MASK_LOWER_NIBBLE		0x0F
 #define DA9052_GPIO_NIBBLE_SHIFT		4
+#define DA9052_IRQ_GPI0			16
+#define DA9052_GPIO_ODD_SHIFT			7
+#define DA9052_GPIO_EVEN_SHIFT			3
 
 struct da9052_gpio {
 	struct da9052 *da9052;
@@ -104,33 +106,26 @@
 static void da9052_gpio_set(struct gpio_chip *gc, unsigned offset, int value)
 {
 	struct da9052_gpio *gpio = to_da9052_gpio(gc);
-	unsigned char register_value = 0;
 	int ret;
 
 	if (da9052_gpio_port_odd(offset)) {
-		if (value) {
-			register_value = DA9052_GPIO_ODD_PORT_MODE;
 			ret = da9052_reg_update(gpio->da9052, (offset >> 1) +
 						DA9052_GPIO_0_1_REG,
 						DA9052_GPIO_ODD_PORT_MODE,
-						register_value);
+						value << DA9052_GPIO_ODD_SHIFT);
 			if (ret != 0)
 				dev_err(gpio->da9052->dev,
 					"Failed to updated gpio odd reg,%d",
 					ret);
-		}
 	} else {
-		if (value) {
-			register_value = DA9052_GPIO_EVEN_PORT_MODE;
 			ret = da9052_reg_update(gpio->da9052, (offset >> 1) +
 						DA9052_GPIO_0_1_REG,
 						DA9052_GPIO_EVEN_PORT_MODE,
-						register_value);
+						value << DA9052_GPIO_EVEN_SHIFT);
 			if (ret != 0)
 				dev_err(gpio->da9052->dev,
 					"Failed to updated gpio even reg,%d",
 					ret);
-		}
 	}
 }
 
@@ -201,9 +196,9 @@
 	.direction_input = da9052_gpio_direction_input,
 	.direction_output = da9052_gpio_direction_output,
 	.to_irq = da9052_gpio_to_irq,
-	.can_sleep = 1;
-	.ngpio = 16;
-	.base = -1;
+	.can_sleep = 1,
+	.ngpio = 16,
+	.base = -1,
 };
 
 static int __devinit da9052_gpio_probe(struct platform_device *pdev)
diff --git a/drivers/gpio/gpio-ml-ioh.c b/drivers/gpio/gpio-ml-ioh.c
index ea8e738..461958f 100644
--- a/drivers/gpio/gpio-ml-ioh.c
+++ b/drivers/gpio/gpio-ml-ioh.c
@@ -332,6 +332,34 @@
 		  &chip->reg->regs[chip->ch].imask);
 }
 
+static void ioh_irq_disable(struct irq_data *d)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	struct ioh_gpio *chip = gc->private;
+	unsigned long flags;
+	u32 ien;
+
+	spin_lock_irqsave(&chip->spinlock, flags);
+	ien = ioread32(&chip->reg->regs[chip->ch].ien);
+	ien &= ~(1 << (d->irq - chip->irq_base));
+	iowrite32(ien, &chip->reg->regs[chip->ch].ien);
+	spin_unlock_irqrestore(&chip->spinlock, flags);
+}
+
+static void ioh_irq_enable(struct irq_data *d)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	struct ioh_gpio *chip = gc->private;
+	unsigned long flags;
+	u32 ien;
+
+	spin_lock_irqsave(&chip->spinlock, flags);
+	ien = ioread32(&chip->reg->regs[chip->ch].ien);
+	ien |= 1 << (d->irq - chip->irq_base);
+	iowrite32(ien, &chip->reg->regs[chip->ch].ien);
+	spin_unlock_irqrestore(&chip->spinlock, flags);
+}
+
 static irqreturn_t ioh_gpio_handler(int irq, void *dev_id)
 {
 	struct ioh_gpio *chip = dev_id;
@@ -339,7 +367,7 @@
 	int i, j;
 	int ret = IRQ_NONE;
 
-	for (i = 0; i < 8; i++) {
+	for (i = 0; i < 8; i++, chip++) {
 		reg_val = ioread32(&chip->reg->regs[i].istatus);
 		for (j = 0; j < num_ports[i]; j++) {
 			if (reg_val & BIT(j)) {
@@ -370,6 +398,8 @@
 	ct->chip.irq_mask = ioh_irq_mask;
 	ct->chip.irq_unmask = ioh_irq_unmask;
 	ct->chip.irq_set_type = ioh_irq_type;
+	ct->chip.irq_disable = ioh_irq_disable;
+	ct->chip.irq_enable = ioh_irq_enable;
 
 	irq_setup_generic_chip(gc, IRQ_MSK(num), IRQ_GC_INIT_MASK_CACHE,
 			       IRQ_NOREQUEST | IRQ_NOPROBE, 0);
diff --git a/drivers/gpio/gpio-mpc8xxx.c b/drivers/gpio/gpio-mpc8xxx.c
index ec3fcf0..5cd04b6 100644
--- a/drivers/gpio/gpio-mpc8xxx.c
+++ b/drivers/gpio/gpio-mpc8xxx.c
@@ -132,6 +132,15 @@
 	return 0;
 }
 
+static int mpc5121_gpio_dir_out(struct gpio_chip *gc, unsigned int gpio, int val)
+{
+	/* GPIO 28..31 are input only on MPC5121 */
+	if (gpio >= 28)
+		return -EINVAL;
+
+	return mpc8xxx_gpio_dir_out(gc, gpio, val);
+}
+
 static int mpc8xxx_gpio_to_irq(struct gpio_chip *gc, unsigned offset)
 {
 	struct of_mm_gpio_chip *mm = to_of_mm_gpio_chip(gc);
@@ -340,11 +349,10 @@
 	mm_gc->save_regs = mpc8xxx_gpio_save_regs;
 	gc->ngpio = MPC8XXX_GPIO_PINS;
 	gc->direction_input = mpc8xxx_gpio_dir_in;
-	gc->direction_output = mpc8xxx_gpio_dir_out;
-	if (of_device_is_compatible(np, "fsl,mpc8572-gpio"))
-		gc->get = mpc8572_gpio_get;
-	else
-		gc->get = mpc8xxx_gpio_get;
+	gc->direction_output = of_device_is_compatible(np, "fsl,mpc5121-gpio") ?
+		mpc5121_gpio_dir_out : mpc8xxx_gpio_dir_out;
+	gc->get = of_device_is_compatible(np, "fsl,mpc8572-gpio") ?
+		mpc8572_gpio_get : mpc8xxx_gpio_get;
 	gc->set = mpc8xxx_gpio_set;
 	gc->to_irq = mpc8xxx_gpio_to_irq;
 
diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index 0e49d87..0b05629 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -148,13 +148,17 @@
 	return (__raw_readl(reg) & GPIO_BIT(bank, gpio)) != 0;
 }
 
-#define MOD_REG_BIT(reg, bit_mask, set)	\
-do {	\
-	int l = __raw_readl(base + reg); \
-	if (set) l |= bit_mask; \
-	else l &= ~bit_mask; \
-	__raw_writel(l, base + reg); \
-} while(0)
+static inline void _gpio_rmw(void __iomem *base, u32 reg, u32 mask, bool set)
+{
+	int l = __raw_readl(base + reg);
+
+	if (set) 
+		l |= mask;
+	else
+		l &= ~mask;
+
+	__raw_writel(l, base + reg);
+}
 
 /**
  * _set_gpio_debounce - low level gpio debounce time
@@ -210,28 +214,28 @@
 	u32 gpio_bit = 1 << gpio;
 
 	if (cpu_is_omap44xx()) {
-		MOD_REG_BIT(OMAP4_GPIO_LEVELDETECT0, gpio_bit,
-			trigger & IRQ_TYPE_LEVEL_LOW);
-		MOD_REG_BIT(OMAP4_GPIO_LEVELDETECT1, gpio_bit,
-			trigger & IRQ_TYPE_LEVEL_HIGH);
-		MOD_REG_BIT(OMAP4_GPIO_RISINGDETECT, gpio_bit,
-			trigger & IRQ_TYPE_EDGE_RISING);
-		MOD_REG_BIT(OMAP4_GPIO_FALLINGDETECT, gpio_bit,
-			trigger & IRQ_TYPE_EDGE_FALLING);
+		_gpio_rmw(base, OMAP4_GPIO_LEVELDETECT0, gpio_bit,
+			  trigger & IRQ_TYPE_LEVEL_LOW);
+		_gpio_rmw(base, OMAP4_GPIO_LEVELDETECT1, gpio_bit,
+			  trigger & IRQ_TYPE_LEVEL_HIGH);
+		_gpio_rmw(base, OMAP4_GPIO_RISINGDETECT, gpio_bit,
+			  trigger & IRQ_TYPE_EDGE_RISING);
+		_gpio_rmw(base, OMAP4_GPIO_FALLINGDETECT, gpio_bit,
+			  trigger & IRQ_TYPE_EDGE_FALLING);
 	} else {
-		MOD_REG_BIT(OMAP24XX_GPIO_LEVELDETECT0, gpio_bit,
-			trigger & IRQ_TYPE_LEVEL_LOW);
-		MOD_REG_BIT(OMAP24XX_GPIO_LEVELDETECT1, gpio_bit,
-			trigger & IRQ_TYPE_LEVEL_HIGH);
-		MOD_REG_BIT(OMAP24XX_GPIO_RISINGDETECT, gpio_bit,
-			trigger & IRQ_TYPE_EDGE_RISING);
-		MOD_REG_BIT(OMAP24XX_GPIO_FALLINGDETECT, gpio_bit,
-			trigger & IRQ_TYPE_EDGE_FALLING);
+		_gpio_rmw(base, OMAP24XX_GPIO_LEVELDETECT0, gpio_bit,
+			  trigger & IRQ_TYPE_LEVEL_LOW);
+		_gpio_rmw(base, OMAP24XX_GPIO_LEVELDETECT1, gpio_bit,
+			  trigger & IRQ_TYPE_LEVEL_HIGH);
+		_gpio_rmw(base, OMAP24XX_GPIO_RISINGDETECT, gpio_bit,
+			  trigger & IRQ_TYPE_EDGE_RISING);
+		_gpio_rmw(base, OMAP24XX_GPIO_FALLINGDETECT, gpio_bit,
+			  trigger & IRQ_TYPE_EDGE_FALLING);
 	}
 	if (likely(!(bank->non_wakeup_gpios & gpio_bit))) {
 		if (cpu_is_omap44xx()) {
-			MOD_REG_BIT(OMAP4_GPIO_IRQWAKEN0, gpio_bit,
-				trigger != 0);
+			_gpio_rmw(base, OMAP4_GPIO_IRQWAKEN0, gpio_bit,
+				  trigger != 0);
 		} else {
 			/*
 			 * GPIO wakeup request can only be generated on edge
@@ -1086,6 +1090,11 @@
 
 	gc = irq_alloc_generic_chip("MPUIO", 1, irq_start, bank->base,
 				    handle_simple_irq);
+	if (!gc) {
+		dev_err(bank->dev, "Memory alloc failed for gc\n");
+		return;
+	}
+
 	ct = gc->chip_types;
 
 	/* NOTE: No ack required, reading IRQ status clears it. */
diff --git a/drivers/gpio/gpio-pca953x.c b/drivers/gpio/gpio-pca953x.c
index 0550dcb..d3f3e8f 100644
--- a/drivers/gpio/gpio-pca953x.c
+++ b/drivers/gpio/gpio-pca953x.c
@@ -546,7 +546,7 @@
  * Translate OpenFirmware node properties into platform_data
  * WARNING: This is DEPRECATED and will be removed eventually!
  */
-void
+static void
 pca953x_get_alt_pdata(struct i2c_client *client, int *gpio_base, int *invert)
 {
 	struct device_node *node;
@@ -574,7 +574,7 @@
 		*invert = *val;
 }
 #else
-void
+static void
 pca953x_get_alt_pdata(struct i2c_client *client, int *gpio_base, int *invert)
 {
 	*gpio_base = -1;
@@ -596,9 +596,6 @@
 
 	/* set platform specific polarity inversion */
 	ret = pca953x_write_reg(chip, PCA953X_INVERT, invert);
-	if (ret)
-		goto out;
-	return 0;
 out:
 	return ret;
 }
@@ -640,7 +637,7 @@
 	struct pca953x_platform_data *pdata;
 	struct pca953x_chip *chip;
 	int irq_base=0, invert=0;
-	int ret = 0;
+	int ret;
 
 	chip = kzalloc(sizeof(struct pca953x_chip), GFP_KERNEL);
 	if (chip == NULL)
@@ -673,10 +670,10 @@
 	pca953x_setup_gpio(chip, id->driver_data & PCA_GPIO_MASK);
 
 	if (chip->chip_type == PCA953X_TYPE)
-		device_pca953x_init(chip, invert);
-	else if (chip->chip_type == PCA957X_TYPE)
-		device_pca957x_init(chip, invert);
+		ret = device_pca953x_init(chip, invert);
 	else
+		ret = device_pca957x_init(chip, invert);
+	if (ret)
 		goto out_failed;
 
 	ret = pca953x_irq_setup(chip, id, irq_base);
diff --git a/drivers/gpio/gpio-pl061.c b/drivers/gpio/gpio-pl061.c
index 093c90b..4102f63 100644
--- a/drivers/gpio/gpio-pl061.c
+++ b/drivers/gpio/gpio-pl061.c
@@ -238,10 +238,6 @@
 	int ret, irq, i;
 	static DECLARE_BITMAP(init_irq, NR_IRQS);
 
-	pdata = dev->dev.platform_data;
-	if (pdata == NULL)
-		return -ENODEV;
-
 	chip = kzalloc(sizeof(*chip), GFP_KERNEL);
 	if (chip == NULL)
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 4507904..2418429 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -9,7 +9,6 @@
 	depends on (AGP || AGP=n) && !EMULATED_CMPXCHG && MMU
 	select I2C
 	select I2C_ALGOBIT
-	select SLOW_WORK
 	help
 	  Kernel-level support for the Direct Rendering Infrastructure (DRI)
 	  introduced in XFree86 4.0. If you say Y here, you need to select
@@ -96,6 +95,7 @@
 	select FB_CFB_IMAGEBLIT
 	# i915 depends on ACPI_VIDEO when ACPI is enabled
 	# but for select to work, need to select ACPI_VIDEO's dependencies, ick
+	select BACKLIGHT_LCD_SUPPORT if ACPI
 	select BACKLIGHT_CLASS_DEVICE if ACPI
 	select VIDEO_OUTPUT_CONTROL if ACPI
 	select INPUT if ACPI
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 6307486..0cde1b8 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -9,7 +9,7 @@
 		drm_drv.o drm_fops.o drm_gem.o drm_ioctl.o drm_irq.o \
 		drm_lock.o drm_memory.o drm_proc.o drm_stub.o drm_vm.o \
 		drm_agpsupport.o drm_scatter.o ati_pcigart.o drm_pci.o \
-		drm_platform.o drm_sysfs.o drm_hashtab.o drm_sman.o drm_mm.o \
+		drm_platform.o drm_sysfs.o drm_hashtab.o drm_mm.o \
 		drm_crtc.o drm_modes.o drm_edid.o \
 		drm_info.o drm_debugfs.o drm_encoder_slave.o \
 		drm_trace_points.o drm_global.o drm_usb.o
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 2556589..f259a25 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -554,7 +554,7 @@
 int drm_plane_init(struct drm_device *dev, struct drm_plane *plane,
 		   unsigned long possible_crtcs,
 		   const struct drm_plane_funcs *funcs,
-		   uint32_t *formats, uint32_t format_count)
+		   const uint32_t *formats, uint32_t format_count)
 {
 	mutex_lock(&dev->mode_config.mutex);
 
@@ -566,6 +566,7 @@
 	if (!plane->format_types) {
 		DRM_DEBUG_KMS("out of memory when allocating plane\n");
 		drm_mode_object_put(dev, &plane->base);
+		mutex_unlock(&dev->mode_config.mutex);
 		return -ENOMEM;
 	}
 
@@ -1440,7 +1441,7 @@
 	 */
 	if ((out_resp->count_modes >= mode_count) && mode_count) {
 		copied = 0;
-		mode_ptr = (struct drm_mode_modeinfo *)(unsigned long)out_resp->modes_ptr;
+		mode_ptr = (struct drm_mode_modeinfo __user *)(unsigned long)out_resp->modes_ptr;
 		list_for_each_entry(mode, &connector->modes, head) {
 			drm_crtc_convert_to_umode(&u_mode, mode);
 			if (copy_to_user(mode_ptr + copied,
@@ -1455,8 +1456,8 @@
 
 	if ((out_resp->count_props >= props_count) && props_count) {
 		copied = 0;
-		prop_ptr = (uint32_t *)(unsigned long)(out_resp->props_ptr);
-		prop_values = (uint64_t *)(unsigned long)(out_resp->prop_values_ptr);
+		prop_ptr = (uint32_t __user *)(unsigned long)(out_resp->props_ptr);
+		prop_values = (uint64_t __user *)(unsigned long)(out_resp->prop_values_ptr);
 		for (i = 0; i < DRM_CONNECTOR_MAX_PROPERTY; i++) {
 			if (connector->property_ids[i] != 0) {
 				if (put_user(connector->property_ids[i],
@@ -1478,7 +1479,7 @@
 
 	if ((out_resp->count_encoders >= encoders_count) && encoders_count) {
 		copied = 0;
-		encoder_ptr = (uint32_t *)(unsigned long)(out_resp->encoders_ptr);
+		encoder_ptr = (uint32_t __user *)(unsigned long)(out_resp->encoders_ptr);
 		for (i = 0; i < DRM_CONNECTOR_MAX_ENCODER; i++) {
 			if (connector->encoder_ids[i] != 0) {
 				if (put_user(connector->encoder_ids[i],
@@ -1560,7 +1561,7 @@
 	 */
 	if (config->num_plane &&
 	    (plane_resp->count_planes >= config->num_plane)) {
-		plane_ptr = (uint32_t *)(unsigned long)plane_resp->plane_id_ptr;
+		plane_ptr = (uint32_t __user *)(unsigned long)plane_resp->plane_id_ptr;
 
 		list_for_each_entry(plane, &config->plane_list, head) {
 			if (put_user(plane->base.id, plane_ptr + copied)) {
@@ -1627,7 +1628,7 @@
 	 */
 	if (plane->format_count &&
 	    (plane_resp->count_format_types >= plane->format_count)) {
-		format_ptr = (uint32_t *)(unsigned long)plane_resp->format_type_ptr;
+		format_ptr = (uint32_t __user *)(unsigned long)plane_resp->format_type_ptr;
 		if (copy_to_user(format_ptr,
 				 plane->format_types,
 				 sizeof(uint32_t) * plane->format_count)) {
@@ -1660,6 +1661,8 @@
 	struct drm_crtc *crtc;
 	struct drm_framebuffer *fb;
 	int ret = 0;
+	unsigned int fb_width, fb_height;
+	int i;
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
@@ -1683,6 +1686,8 @@
 	/* No fb means shut it down */
 	if (!plane_req->fb_id) {
 		plane->funcs->disable_plane(plane);
+		plane->crtc = NULL;
+		plane->fb = NULL;
 		goto out;
 	}
 
@@ -1706,6 +1711,50 @@
 	}
 	fb = obj_to_fb(obj);
 
+	/* Check whether this plane supports the fb pixel format. */
+	for (i = 0; i < plane->format_count; i++)
+		if (fb->pixel_format == plane->format_types[i])
+			break;
+	if (i == plane->format_count) {
+		DRM_DEBUG_KMS("Invalid pixel format 0x%08x\n", fb->pixel_format);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	fb_width = fb->width << 16;
+	fb_height = fb->height << 16;
+
+	/* Make sure source coordinates are inside the fb. */
+	if (plane_req->src_w > fb_width ||
+	    plane_req->src_x > fb_width - plane_req->src_w ||
+	    plane_req->src_h > fb_height ||
+	    plane_req->src_y > fb_height - plane_req->src_h) {
+		DRM_DEBUG_KMS("Invalid source coordinates "
+			      "%u.%06ux%u.%06u+%u.%06u+%u.%06u\n",
+			      plane_req->src_w >> 16,
+			      ((plane_req->src_w & 0xffff) * 15625) >> 10,
+			      plane_req->src_h >> 16,
+			      ((plane_req->src_h & 0xffff) * 15625) >> 10,
+			      plane_req->src_x >> 16,
+			      ((plane_req->src_x & 0xffff) * 15625) >> 10,
+			      plane_req->src_y >> 16,
+			      ((plane_req->src_y & 0xffff) * 15625) >> 10);
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	/* Give drivers some help against integer overflows */
+	if (plane_req->crtc_w > INT_MAX ||
+	    plane_req->crtc_x > INT_MAX - (int32_t) plane_req->crtc_w ||
+	    plane_req->crtc_h > INT_MAX ||
+	    plane_req->crtc_y > INT_MAX - (int32_t) plane_req->crtc_h) {
+		DRM_DEBUG_KMS("Invalid CRTC coordinates %ux%u+%d+%d\n",
+			      plane_req->crtc_w, plane_req->crtc_h,
+			      plane_req->crtc_x, plane_req->crtc_y);
+		ret = -ERANGE;
+		goto out;
+	}
+
 	ret = plane->funcs->update_plane(plane, crtc, fb,
 					 plane_req->crtc_x, plane_req->crtc_y,
 					 plane_req->crtc_w, plane_req->crtc_h,
@@ -1828,7 +1877,7 @@
 		}
 
 		for (i = 0; i < crtc_req->count_connectors; i++) {
-			set_connectors_ptr = (uint32_t *)(unsigned long)crtc_req->set_connectors_ptr;
+			set_connectors_ptr = (uint32_t __user *)(unsigned long)crtc_req->set_connectors_ptr;
 			if (get_user(out_id, &set_connectors_ptr[i])) {
 				ret = -EFAULT;
 				goto out;
@@ -2012,6 +2061,75 @@
 	return ret;
 }
 
+static int format_check(struct drm_mode_fb_cmd2 *r)
+{
+	uint32_t format = r->pixel_format & ~DRM_FORMAT_BIG_ENDIAN;
+
+	switch (format) {
+	case DRM_FORMAT_C8:
+	case DRM_FORMAT_RGB332:
+	case DRM_FORMAT_BGR233:
+	case DRM_FORMAT_XRGB4444:
+	case DRM_FORMAT_XBGR4444:
+	case DRM_FORMAT_RGBX4444:
+	case DRM_FORMAT_BGRX4444:
+	case DRM_FORMAT_ARGB4444:
+	case DRM_FORMAT_ABGR4444:
+	case DRM_FORMAT_RGBA4444:
+	case DRM_FORMAT_BGRA4444:
+	case DRM_FORMAT_XRGB1555:
+	case DRM_FORMAT_XBGR1555:
+	case DRM_FORMAT_RGBX5551:
+	case DRM_FORMAT_BGRX5551:
+	case DRM_FORMAT_ARGB1555:
+	case DRM_FORMAT_ABGR1555:
+	case DRM_FORMAT_RGBA5551:
+	case DRM_FORMAT_BGRA5551:
+	case DRM_FORMAT_RGB565:
+	case DRM_FORMAT_BGR565:
+	case DRM_FORMAT_RGB888:
+	case DRM_FORMAT_BGR888:
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_XBGR8888:
+	case DRM_FORMAT_RGBX8888:
+	case DRM_FORMAT_BGRX8888:
+	case DRM_FORMAT_ARGB8888:
+	case DRM_FORMAT_ABGR8888:
+	case DRM_FORMAT_RGBA8888:
+	case DRM_FORMAT_BGRA8888:
+	case DRM_FORMAT_XRGB2101010:
+	case DRM_FORMAT_XBGR2101010:
+	case DRM_FORMAT_RGBX1010102:
+	case DRM_FORMAT_BGRX1010102:
+	case DRM_FORMAT_ARGB2101010:
+	case DRM_FORMAT_ABGR2101010:
+	case DRM_FORMAT_RGBA1010102:
+	case DRM_FORMAT_BGRA1010102:
+	case DRM_FORMAT_YUYV:
+	case DRM_FORMAT_YVYU:
+	case DRM_FORMAT_UYVY:
+	case DRM_FORMAT_VYUY:
+	case DRM_FORMAT_AYUV:
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV21:
+	case DRM_FORMAT_NV16:
+	case DRM_FORMAT_NV61:
+	case DRM_FORMAT_YUV410:
+	case DRM_FORMAT_YVU410:
+	case DRM_FORMAT_YUV411:
+	case DRM_FORMAT_YVU411:
+	case DRM_FORMAT_YUV420:
+	case DRM_FORMAT_YVU420:
+	case DRM_FORMAT_YUV422:
+	case DRM_FORMAT_YVU422:
+	case DRM_FORMAT_YUV444:
+	case DRM_FORMAT_YVU444:
+		return 0;
+	default:
+		return -EINVAL;
+	}
+}
+
 /**
  * drm_mode_addfb2 - add an FB to the graphics configuration
  * @inode: inode from the ioctl
@@ -2051,6 +2169,12 @@
 		return -EINVAL;
 	}
 
+	ret = format_check(r);
+	if (ret) {
+		DRM_ERROR("bad framebuffer format 0x%08x\n", r->pixel_format);
+		return ret;
+	}
+
 	mutex_lock(&dev->mode_config.mutex);
 
 	fb = dev->mode_config.funcs->fb_create(dev, file_priv, r);
@@ -2168,7 +2292,7 @@
 	r->width = fb->width;
 	r->depth = fb->depth;
 	r->bpp = fb->bits_per_pixel;
-	r->pitch = fb->pitch;
+	r->pitch = fb->pitches[0];
 	fb->funcs->create_handle(fb, file_priv, &r->handle);
 
 out:
@@ -2200,7 +2324,7 @@
 	fb = obj_to_fb(obj);
 
 	num_clips = r->num_clips;
-	clips_ptr = (struct drm_clip_rect *)(unsigned long)r->clips_ptr;
+	clips_ptr = (struct drm_clip_rect __user *)(unsigned long)r->clips_ptr;
 
 	if (!num_clips != !clips_ptr) {
 		ret = -EINVAL;
@@ -2216,6 +2340,10 @@
 	}
 
 	if (num_clips && clips_ptr) {
+		if (num_clips < 0 || num_clips > DRM_MODE_FB_DIRTY_MAX_CLIPS) {
+			ret = -EINVAL;
+			goto out_err1;
+		}
 		clips = kzalloc(num_clips * sizeof(*clips), GFP_KERNEL);
 		if (!clips) {
 			ret = -ENOMEM;
@@ -2461,8 +2589,10 @@
 	property->num_values = num_values;
 	INIT_LIST_HEAD(&property->enum_blob_list);
 
-	if (name)
+	if (name) {
 		strncpy(property->name, name, DRM_PROP_NAME_LEN);
+		property->name[DRM_PROP_NAME_LEN-1] = '\0';
+	}
 
 	list_add_tail(&property->head, &dev->mode_config.property_list);
 	return property;
@@ -2590,7 +2720,7 @@
 	struct drm_property_enum *prop_enum;
 	struct drm_mode_property_enum __user *enum_ptr;
 	struct drm_property_blob *prop_blob;
-	uint32_t *blob_id_ptr;
+	uint32_t __user *blob_id_ptr;
 	uint64_t __user *values_ptr;
 	uint32_t __user *blob_length_ptr;
 
@@ -2620,7 +2750,7 @@
 	out_resp->flags = property->flags;
 
 	if ((out_resp->count_values >= value_count) && value_count) {
-		values_ptr = (uint64_t *)(unsigned long)out_resp->values_ptr;
+		values_ptr = (uint64_t __user *)(unsigned long)out_resp->values_ptr;
 		for (i = 0; i < value_count; i++) {
 			if (copy_to_user(values_ptr + i, &property->values[i], sizeof(uint64_t))) {
 				ret = -EFAULT;
@@ -2633,7 +2763,7 @@
 	if (property->flags & DRM_MODE_PROP_ENUM) {
 		if ((out_resp->count_enum_blobs >= enum_count) && enum_count) {
 			copied = 0;
-			enum_ptr = (struct drm_mode_property_enum *)(unsigned long)out_resp->enum_blob_ptr;
+			enum_ptr = (struct drm_mode_property_enum __user *)(unsigned long)out_resp->enum_blob_ptr;
 			list_for_each_entry(prop_enum, &property->enum_blob_list, head) {
 
 				if (copy_to_user(&enum_ptr[copied].value, &prop_enum->value, sizeof(uint64_t))) {
@@ -2655,8 +2785,8 @@
 	if (property->flags & DRM_MODE_PROP_BLOB) {
 		if ((out_resp->count_enum_blobs >= blob_count) && blob_count) {
 			copied = 0;
-			blob_id_ptr = (uint32_t *)(unsigned long)out_resp->enum_blob_ptr;
-			blob_length_ptr = (uint32_t *)(unsigned long)out_resp->values_ptr;
+			blob_id_ptr = (uint32_t __user *)(unsigned long)out_resp->enum_blob_ptr;
+			blob_length_ptr = (uint32_t __user *)(unsigned long)out_resp->values_ptr;
 
 			list_for_each_entry(prop_blob, &property->enum_blob_list, head) {
 				if (put_user(prop_blob->base.id, blob_id_ptr + copied)) {
@@ -2717,7 +2847,7 @@
 	struct drm_mode_get_blob *out_resp = data;
 	struct drm_property_blob *blob;
 	int ret = 0;
-	void *blob_ptr;
+	void __user *blob_ptr;
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
@@ -2731,7 +2861,7 @@
 	blob = obj_to_blob(obj);
 
 	if (out_resp->length == blob->length) {
-		blob_ptr = (void *)(unsigned long)out_resp->data;
+		blob_ptr = (void __user *)(unsigned long)out_resp->data;
 		if (copy_to_user(blob_ptr, blob->data, blob->length)){
 			ret = -EFAULT;
 			goto done;
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 2ce61d7..84a4a80 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -457,6 +457,30 @@
 EXPORT_SYMBOL(drm_crtc_helper_set_mode);
 
 
+static int
+drm_crtc_helper_disable(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_connector *connector;
+	struct drm_encoder *encoder;
+
+	/* Decouple all encoders and their attached connectors from this crtc */
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
+		if (encoder->crtc != crtc)
+			continue;
+
+		list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+			if (connector->encoder != encoder)
+				continue;
+
+			connector->encoder = NULL;
+		}
+	}
+
+	drm_helper_disable_unused_functions(dev);
+	return 0;
+}
+
 /**
  * drm_crtc_helper_set_config - set a new config from userspace
  * @crtc: CRTC to setup
@@ -485,6 +509,7 @@
 	struct drm_connector *save_connectors, *connector;
 	int count = 0, ro, fail = 0;
 	struct drm_crtc_helper_funcs *crtc_funcs;
+	struct drm_mode_set save_set;
 	int ret = 0;
 	int i;
 
@@ -510,8 +535,7 @@
 				(int)set->num_connectors, set->x, set->y);
 	} else {
 		DRM_DEBUG_KMS("[CRTC:%d] [NOFB]\n", set->crtc->base.id);
-		set->mode = NULL;
-		set->num_connectors = 0;
+		return drm_crtc_helper_disable(set->crtc);
 	}
 
 	dev = set->crtc->dev;
@@ -557,6 +581,12 @@
 		save_connectors[count++] = *connector;
 	}
 
+	save_set.crtc = set->crtc;
+	save_set.mode = &set->crtc->mode;
+	save_set.x = set->crtc->x;
+	save_set.y = set->crtc->y;
+	save_set.fb = set->crtc->fb;
+
 	/* We should be able to check here if the fb has the same properties
 	 * and then just flip_or_move it */
 	if (set->crtc->fb != set->fb) {
@@ -681,7 +711,7 @@
 			for (i = 0; i < set->num_connectors; i++) {
 				DRM_DEBUG_KMS("\t[CONNECTOR:%d:%s] set DPMS on\n", set->connectors[i]->base.id,
 					      drm_get_connector_name(set->connectors[i]));
-				set->connectors[i]->dpms = DRM_MODE_DPMS_ON;
+				set->connectors[i]->funcs->dpms(set->connectors[i], DRM_MODE_DPMS_ON);
 			}
 		}
 		drm_helper_disable_unused_functions(dev);
@@ -722,6 +752,12 @@
 		*connector = save_connectors[count++];
 	}
 
+	/* Try to restore the config */
+	if (mode_changed &&
+	    !drm_crtc_helper_set_mode(save_set.crtc, save_set.mode, save_set.x,
+				      save_set.y, save_set.fb))
+		DRM_ERROR("failed to restore config after modeset failure\n");
+
 	kfree(save_connectors);
 	kfree(save_encoders);
 	kfree(save_crtcs);
@@ -814,9 +850,14 @@
 int drm_helper_mode_fill_fb_struct(struct drm_framebuffer *fb,
 				   struct drm_mode_fb_cmd2 *mode_cmd)
 {
+	int i;
+
 	fb->width = mode_cmd->width;
 	fb->height = mode_cmd->height;
-	fb->pitch = mode_cmd->pitches[0];
+	for (i = 0; i < 4; i++) {
+		fb->pitches[i] = mode_cmd->pitches[i];
+		fb->offsets[i] = mode_cmd->offsets[i];
+	}
 	drm_fb_get_bpp_depth(mode_cmd->pixel_format, &fb->depth,
 				    &fb->bits_per_pixel);
 	fb->pixel_format = mode_cmd->pixel_format;
@@ -974,3 +1015,36 @@
 		queue_delayed_work(system_nrt_wq, &dev->mode_config.output_poll_work, 0);
 }
 EXPORT_SYMBOL(drm_helper_hpd_irq_event);
+
+
+/**
+ * drm_format_num_planes - get the number of planes for format
+ * @format: pixel format (DRM_FORMAT_*)
+ *
+ * RETURNS:
+ * The number of planes used by the specified pixel format.
+ */
+int drm_format_num_planes(uint32_t format)
+{
+	switch (format) {
+	case DRM_FORMAT_YUV410:
+	case DRM_FORMAT_YVU410:
+	case DRM_FORMAT_YUV411:
+	case DRM_FORMAT_YVU411:
+	case DRM_FORMAT_YUV420:
+	case DRM_FORMAT_YVU420:
+	case DRM_FORMAT_YUV422:
+	case DRM_FORMAT_YVU422:
+	case DRM_FORMAT_YUV444:
+	case DRM_FORMAT_YVU444:
+		return 3;
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV21:
+	case DRM_FORMAT_NV16:
+	case DRM_FORMAT_NV61:
+		return 2;
+	default:
+		return 1;
+	}
+}
+EXPORT_SYMBOL(drm_format_num_planes);
diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c
index d067c12..1c7a1c0 100644
--- a/drivers/gpu/drm/drm_debugfs.c
+++ b/drivers/gpu/drm/drm_debugfs.c
@@ -118,7 +118,10 @@
 		tmp->minor = minor;
 		tmp->dent = ent;
 		tmp->info_ent = &files[i];
-		list_add(&(tmp->list), &(minor->debugfs_nodes.list));
+
+		mutex_lock(&minor->debugfs_lock);
+		list_add(&tmp->list, &minor->debugfs_list);
+		mutex_unlock(&minor->debugfs_lock);
 	}
 	return 0;
 
@@ -146,7 +149,8 @@
 	char name[64];
 	int ret;
 
-	INIT_LIST_HEAD(&minor->debugfs_nodes.list);
+	INIT_LIST_HEAD(&minor->debugfs_list);
+	mutex_init(&minor->debugfs_lock);
 	sprintf(name, "%d", minor_id);
 	minor->debugfs_root = debugfs_create_dir(name, root);
 	if (!minor->debugfs_root) {
@@ -192,8 +196,9 @@
 	struct drm_info_node *tmp;
 	int i;
 
+	mutex_lock(&minor->debugfs_lock);
 	for (i = 0; i < count; i++) {
-		list_for_each_safe(pos, q, &minor->debugfs_nodes.list) {
+		list_for_each_safe(pos, q, &minor->debugfs_list) {
 			tmp = list_entry(pos, struct drm_info_node, list);
 			if (tmp->info_ent == &files[i]) {
 				debugfs_remove(tmp->dent);
@@ -202,6 +207,7 @@
 			}
 		}
 	}
+	mutex_unlock(&minor->debugfs_lock);
 	return 0;
 }
 EXPORT_SYMBOL(drm_debugfs_remove_files);
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index eaf25ff..bc5febe 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -125,7 +125,7 @@
 	DRM_IOCTL_DEF(DRM_IOCTL_SG_ALLOC, drm_sg_alloc_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF(DRM_IOCTL_SG_FREE, drm_sg_free, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_WAIT_VBLANK, drm_wait_vblank, 0),
+	DRM_IOCTL_DEF(DRM_IOCTL_WAIT_VBLANK, drm_wait_vblank, DRM_UNLOCKED),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_MODESET_CTL, drm_modeset_ctl, 0),
 
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 2283631..ece03fc 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -508,25 +508,10 @@
 cea_for_each_detailed_block(u8 *ext, detailed_cb *cb, void *closure)
 {
 	int i, n = 0;
-	u8 rev = ext[0x01], d = ext[0x02];
+	u8 d = ext[0x02];
 	u8 *det_base = ext + d;
 
-	switch (rev) {
-	case 0:
-		/* can't happen */
-		return;
-	case 1:
-		/* have to infer how many blocks we have, check pixel clock */
-		for (i = 0; i < 6; i++)
-			if (det_base[18*i] || det_base[18*i+1])
-				n++;
-		break;
-	default:
-		/* explicit count */
-		n = min(ext[0x03] & 0x0f, 6);
-		break;
-	}
-
+	n = (127 - d) / 18;
 	for (i = 0; i < n; i++)
 		cb((struct detailed_timing *)(det_base + 18 * i), closure);
 }
@@ -1474,26 +1459,29 @@
 	eld[18] = edid->prod_code[0];
 	eld[19] = edid->prod_code[1];
 
-	for (db = cea + 4; db < cea + cea[2]; db += dbl + 1) {
-		dbl = db[0] & 0x1f;
-
-		switch ((db[0] & 0xe0) >> 5) {
-		case AUDIO_BLOCK:	/* Audio Data Block, contains SADs */
-			sad_count = dbl / 3;
-			memcpy(eld + 20 + mnl, &db[1], dbl);
-			break;
-		case SPEAKER_BLOCK:	/* Speaker Allocation Data Block */
-			eld[7] = db[1];
-			break;
-		case VENDOR_BLOCK:
-			/* HDMI Vendor-Specific Data Block */
-			if (db[1] == 0x03 && db[2] == 0x0c && db[3] == 0)
-				parse_hdmi_vsdb(connector, db);
-			break;
-		default:
-			break;
+	if (cea[1] >= 3)
+		for (db = cea + 4; db < cea + cea[2]; db += dbl + 1) {
+			dbl = db[0] & 0x1f;
+			
+			switch ((db[0] & 0xe0) >> 5) {
+			case AUDIO_BLOCK:
+				/* Audio Data Block, contains SADs */
+				sad_count = dbl / 3;
+				memcpy(eld + 20 + mnl, &db[1], dbl);
+				break;
+			case SPEAKER_BLOCK:
+                                /* Speaker Allocation Data Block */
+				eld[7] = db[1];
+				break;
+			case VENDOR_BLOCK:
+				/* HDMI Vendor-Specific Data Block */
+				if (db[1] == 0x03 && db[2] == 0x0c && db[3] == 0)
+					parse_hdmi_vsdb(connector, db);
+				break;
+			default:
+				break;
+			}
 		}
-	}
 	eld[5] |= sad_count << 4;
 	eld[2] = (20 + mnl + sad_count * 3 + 3) / 4;
 
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index cb3794a..44a5d0a 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -110,10 +110,7 @@
 	/* Prevent vblank irq processing while disabling vblank irqs,
 	 * so no updates of timestamps or count can happen after we've
 	 * disabled. Needed to prevent races in case of delayed irq's.
-	 * Disable preemption, so vblank_time_lock is held as short as
-	 * possible, even under a kernel with PREEMPT_RT patches.
 	 */
-	preempt_disable();
 	spin_lock_irqsave(&dev->vblank_time_lock, irqflags);
 
 	dev->driver->disable_vblank(dev, crtc);
@@ -164,7 +161,6 @@
 	clear_vblank_timestamps(dev, crtc);
 
 	spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
-	preempt_enable();
 }
 
 static void vblank_disable_fn(unsigned long arg)
@@ -407,13 +403,16 @@
 	/*
 	 * Wake up any waiters so they don't hang.
 	 */
-	spin_lock_irqsave(&dev->vbl_lock, irqflags);
-	for (i = 0; i < dev->num_crtcs; i++) {
-		DRM_WAKEUP(&dev->vbl_queue[i]);
-		dev->vblank_enabled[i] = 0;
-		dev->last_vblank[i] = dev->driver->get_vblank_counter(dev, i);
+	if (dev->num_crtcs) {
+		spin_lock_irqsave(&dev->vbl_lock, irqflags);
+		for (i = 0; i < dev->num_crtcs; i++) {
+			DRM_WAKEUP(&dev->vbl_queue[i]);
+			dev->vblank_enabled[i] = 0;
+			dev->last_vblank[i] =
+				dev->driver->get_vblank_counter(dev, i);
+		}
+		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 	}
-	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 
 	if (!irq_enabled)
 		return -EINVAL;
@@ -886,10 +885,6 @@
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
 	/* Going from 0->1 means we have to enable interrupts again */
 	if (atomic_add_return(1, &dev->vblank_refcount[crtc]) == 1) {
-		/* Disable preemption while holding vblank_time_lock. Do
-		 * it explicitely to guard against PREEMPT_RT kernel.
-		 */
-		preempt_disable();
 		spin_lock_irqsave(&dev->vblank_time_lock, irqflags2);
 		if (!dev->vblank_enabled[crtc]) {
 			/* Enable vblank irqs under vblank_time_lock protection.
@@ -909,7 +904,6 @@
 			}
 		}
 		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags2);
-		preempt_enable();
 	} else {
 		if (!dev->vblank_enabled[crtc]) {
 			atomic_dec(&dev->vblank_refcount[crtc]);
@@ -1125,6 +1119,7 @@
 		trace_drm_vblank_event_delivered(current->pid, pipe,
 						 vblwait->request.sequence);
 	} else {
+		/* drm_handle_vblank_events will call drm_vblank_put */
 		list_add_tail(&e->base.link, &dev->vblank_event_list);
 		vblwait->reply.sequence = vblwait->request.sequence;
 	}
@@ -1205,8 +1200,12 @@
 		goto done;
 	}
 
-	if (flags & _DRM_VBLANK_EVENT)
+	if (flags & _DRM_VBLANK_EVENT) {
+		/* must hold on to the vblank ref until the event fires
+		 * drm_vblank_put will be called asynchronously
+		 */
 		return drm_queue_vblank_event(dev, crtc, vblwait, file_priv);
+	}
 
 	if ((flags & _DRM_VBLANK_NEXTONMISS) &&
 	    (seq - vblwait->request.sequence) <= (1<<23)) {
diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c
index 632ae24..c79c713 100644
--- a/drivers/gpu/drm/drm_lock.c
+++ b/drivers/gpu/drm/drm_lock.c
@@ -33,6 +33,7 @@
  * OTHER DEALINGS IN THE SOFTWARE.
  */
 
+#include <linux/export.h>
 #include "drmP.h"
 
 static int drm_notifier(void *priv);
@@ -345,6 +346,7 @@
 	}
 	spin_unlock_bh(&lock_data->spinlock);
 }
+EXPORT_SYMBOL(drm_idlelock_take);
 
 void drm_idlelock_release(struct drm_lock_data *lock_data)
 {
@@ -364,6 +366,7 @@
 	}
 	spin_unlock_bh(&lock_data->spinlock);
 }
+EXPORT_SYMBOL(drm_idlelock_release);
 
 int drm_i_have_hw_lock(struct drm_device *dev, struct drm_file *file_priv)
 {
diff --git a/drivers/gpu/drm/drm_sman.c b/drivers/gpu/drm/drm_sman.c
deleted file mode 100644
index cebce45..0000000
--- a/drivers/gpu/drm/drm_sman.c
+++ /dev/null
@@ -1,351 +0,0 @@
-/**************************************************************************
- *
- * Copyright 2006 Tungsten Graphics, Inc., Bismarck., ND., USA.
- * All Rights Reserved.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- *
- **************************************************************************/
-/*
- * Simple memory manager interface that keeps track on allocate regions on a
- * per "owner" basis. All regions associated with an "owner" can be released
- * with a simple call. Typically if the "owner" exists. The owner is any
- * "unsigned long" identifier. Can typically be a pointer to a file private
- * struct or a context identifier.
- *
- * Authors:
- * Thomas Hellström <thomas-at-tungstengraphics-dot-com>
- */
-
-#include <linux/export.h>
-#include "drm_sman.h"
-
-struct drm_owner_item {
-	struct drm_hash_item owner_hash;
-	struct list_head sman_list;
-	struct list_head mem_blocks;
-};
-
-void drm_sman_takedown(struct drm_sman * sman)
-{
-	drm_ht_remove(&sman->user_hash_tab);
-	drm_ht_remove(&sman->owner_hash_tab);
-	kfree(sman->mm);
-}
-
-EXPORT_SYMBOL(drm_sman_takedown);
-
-int
-drm_sman_init(struct drm_sman * sman, unsigned int num_managers,
-	      unsigned int user_order, unsigned int owner_order)
-{
-	int ret = 0;
-
-	sman->mm = kcalloc(num_managers, sizeof(*sman->mm), GFP_KERNEL);
-	if (!sman->mm) {
-		ret = -ENOMEM;
-		goto out;
-	}
-	sman->num_managers = num_managers;
-	INIT_LIST_HEAD(&sman->owner_items);
-	ret = drm_ht_create(&sman->owner_hash_tab, owner_order);
-	if (ret)
-		goto out1;
-	ret = drm_ht_create(&sman->user_hash_tab, user_order);
-	if (!ret)
-		goto out;
-
-	drm_ht_remove(&sman->owner_hash_tab);
-out1:
-	kfree(sman->mm);
-out:
-	return ret;
-}
-
-EXPORT_SYMBOL(drm_sman_init);
-
-static void *drm_sman_mm_allocate(void *private, unsigned long size,
-				  unsigned alignment)
-{
-	struct drm_mm *mm = (struct drm_mm *) private;
-	struct drm_mm_node *tmp;
-
-	tmp = drm_mm_search_free(mm, size, alignment, 1);
-	if (!tmp) {
-		return NULL;
-	}
-	tmp = drm_mm_get_block(tmp, size, alignment);
-	return tmp;
-}
-
-static void drm_sman_mm_free(void *private, void *ref)
-{
-	struct drm_mm_node *node = (struct drm_mm_node *) ref;
-
-	drm_mm_put_block(node);
-}
-
-static void drm_sman_mm_destroy(void *private)
-{
-	struct drm_mm *mm = (struct drm_mm *) private;
-	drm_mm_takedown(mm);
-	kfree(mm);
-}
-
-static unsigned long drm_sman_mm_offset(void *private, void *ref)
-{
-	struct drm_mm_node *node = (struct drm_mm_node *) ref;
-	return node->start;
-}
-
-int
-drm_sman_set_range(struct drm_sman * sman, unsigned int manager,
-		   unsigned long start, unsigned long size)
-{
-	struct drm_sman_mm *sman_mm;
-	struct drm_mm *mm;
-	int ret;
-
-	BUG_ON(manager >= sman->num_managers);
-
-	sman_mm = &sman->mm[manager];
-	mm = kzalloc(sizeof(*mm), GFP_KERNEL);
-	if (!mm) {
-		return -ENOMEM;
-	}
-	sman_mm->private = mm;
-	ret = drm_mm_init(mm, start, size);
-
-	if (ret) {
-		kfree(mm);
-		return ret;
-	}
-
-	sman_mm->allocate = drm_sman_mm_allocate;
-	sman_mm->free = drm_sman_mm_free;
-	sman_mm->destroy = drm_sman_mm_destroy;
-	sman_mm->offset = drm_sman_mm_offset;
-
-	return 0;
-}
-
-EXPORT_SYMBOL(drm_sman_set_range);
-
-int
-drm_sman_set_manager(struct drm_sman * sman, unsigned int manager,
-		     struct drm_sman_mm * allocator)
-{
-	BUG_ON(manager >= sman->num_managers);
-	sman->mm[manager] = *allocator;
-
-	return 0;
-}
-EXPORT_SYMBOL(drm_sman_set_manager);
-
-static struct drm_owner_item *drm_sman_get_owner_item(struct drm_sman * sman,
-						 unsigned long owner)
-{
-	int ret;
-	struct drm_hash_item *owner_hash_item;
-	struct drm_owner_item *owner_item;
-
-	ret = drm_ht_find_item(&sman->owner_hash_tab, owner, &owner_hash_item);
-	if (!ret) {
-		return drm_hash_entry(owner_hash_item, struct drm_owner_item,
-				      owner_hash);
-	}
-
-	owner_item = kzalloc(sizeof(*owner_item), GFP_KERNEL);
-	if (!owner_item)
-		goto out;
-
-	INIT_LIST_HEAD(&owner_item->mem_blocks);
-	owner_item->owner_hash.key = owner;
-	if (drm_ht_insert_item(&sman->owner_hash_tab, &owner_item->owner_hash))
-		goto out1;
-
-	list_add_tail(&owner_item->sman_list, &sman->owner_items);
-	return owner_item;
-
-out1:
-	kfree(owner_item);
-out:
-	return NULL;
-}
-
-struct drm_memblock_item *drm_sman_alloc(struct drm_sman *sman, unsigned int manager,
-				    unsigned long size, unsigned alignment,
-				    unsigned long owner)
-{
-	void *tmp;
-	struct drm_sman_mm *sman_mm;
-	struct drm_owner_item *owner_item;
-	struct drm_memblock_item *memblock;
-
-	BUG_ON(manager >= sman->num_managers);
-
-	sman_mm = &sman->mm[manager];
-	tmp = sman_mm->allocate(sman_mm->private, size, alignment);
-
-	if (!tmp) {
-		return NULL;
-	}
-
-	memblock = kzalloc(sizeof(*memblock), GFP_KERNEL);
-
-	if (!memblock)
-		goto out;
-
-	memblock->mm_info = tmp;
-	memblock->mm = sman_mm;
-	memblock->sman = sman;
-
-	if (drm_ht_just_insert_please
-	    (&sman->user_hash_tab, &memblock->user_hash,
-	     (unsigned long)memblock, 32, 0, 0))
-		goto out1;
-
-	owner_item = drm_sman_get_owner_item(sman, owner);
-	if (!owner_item)
-		goto out2;
-
-	list_add_tail(&memblock->owner_list, &owner_item->mem_blocks);
-
-	return memblock;
-
-out2:
-	drm_ht_remove_item(&sman->user_hash_tab, &memblock->user_hash);
-out1:
-	kfree(memblock);
-out:
-	sman_mm->free(sman_mm->private, tmp);
-
-	return NULL;
-}
-
-EXPORT_SYMBOL(drm_sman_alloc);
-
-static void drm_sman_free(struct drm_memblock_item *item)
-{
-	struct drm_sman *sman = item->sman;
-
-	list_del(&item->owner_list);
-	drm_ht_remove_item(&sman->user_hash_tab, &item->user_hash);
-	item->mm->free(item->mm->private, item->mm_info);
-	kfree(item);
-}
-
-int drm_sman_free_key(struct drm_sman *sman, unsigned int key)
-{
-	struct drm_hash_item *hash_item;
-	struct drm_memblock_item *memblock_item;
-
-	if (drm_ht_find_item(&sman->user_hash_tab, key, &hash_item))
-		return -EINVAL;
-
-	memblock_item = drm_hash_entry(hash_item, struct drm_memblock_item,
-				       user_hash);
-	drm_sman_free(memblock_item);
-	return 0;
-}
-
-EXPORT_SYMBOL(drm_sman_free_key);
-
-static void drm_sman_remove_owner(struct drm_sman *sman,
-				  struct drm_owner_item *owner_item)
-{
-	list_del(&owner_item->sman_list);
-	drm_ht_remove_item(&sman->owner_hash_tab, &owner_item->owner_hash);
-	kfree(owner_item);
-}
-
-int drm_sman_owner_clean(struct drm_sman *sman, unsigned long owner)
-{
-
-	struct drm_hash_item *hash_item;
-	struct drm_owner_item *owner_item;
-
-	if (drm_ht_find_item(&sman->owner_hash_tab, owner, &hash_item)) {
-		return -1;
-	}
-
-	owner_item = drm_hash_entry(hash_item, struct drm_owner_item, owner_hash);
-	if (owner_item->mem_blocks.next == &owner_item->mem_blocks) {
-		drm_sman_remove_owner(sman, owner_item);
-		return -1;
-	}
-
-	return 0;
-}
-
-EXPORT_SYMBOL(drm_sman_owner_clean);
-
-static void drm_sman_do_owner_cleanup(struct drm_sman *sman,
-				      struct drm_owner_item *owner_item)
-{
-	struct drm_memblock_item *entry, *next;
-
-	list_for_each_entry_safe(entry, next, &owner_item->mem_blocks,
-				 owner_list) {
-		drm_sman_free(entry);
-	}
-	drm_sman_remove_owner(sman, owner_item);
-}
-
-void drm_sman_owner_cleanup(struct drm_sman *sman, unsigned long owner)
-{
-
-	struct drm_hash_item *hash_item;
-	struct drm_owner_item *owner_item;
-
-	if (drm_ht_find_item(&sman->owner_hash_tab, owner, &hash_item)) {
-
-		return;
-	}
-
-	owner_item = drm_hash_entry(hash_item, struct drm_owner_item, owner_hash);
-	drm_sman_do_owner_cleanup(sman, owner_item);
-}
-
-EXPORT_SYMBOL(drm_sman_owner_cleanup);
-
-void drm_sman_cleanup(struct drm_sman *sman)
-{
-	struct drm_owner_item *entry, *next;
-	unsigned int i;
-	struct drm_sman_mm *sman_mm;
-
-	list_for_each_entry_safe(entry, next, &sman->owner_items, sman_list) {
-		drm_sman_do_owner_cleanup(sman, entry);
-	}
-	if (sman->mm) {
-		for (i = 0; i < sman->num_managers; ++i) {
-			sman_mm = &sman->mm[i];
-			if (sman_mm->private) {
-				sman_mm->destroy(sman_mm->private);
-				sman_mm->private = NULL;
-			}
-		}
-	}
-}
-
-EXPORT_SYMBOL(drm_sman_cleanup);
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 847466a..f9aaa56 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -18,3 +18,10 @@
 	help
 	  Choose this option if you want to use Exynos FIMD for DRM.
 	  If M is selected, the module will be called exynos_drm_fimd
+
+config DRM_EXYNOS_HDMI
+	tristate "Exynos DRM HDMI"
+	depends on DRM_EXYNOS
+	help
+	  Choose this option if you want to use Exynos HDMI for DRM.
+	  If M is selected, the module will be called exynos_drm_hdmi
diff --git a/drivers/gpu/drm/exynos/Makefile b/drivers/gpu/drm/exynos/Makefile
index 0496d3f..395e69c 100644
--- a/drivers/gpu/drm/exynos/Makefile
+++ b/drivers/gpu/drm/exynos/Makefile
@@ -5,7 +5,10 @@
 ccflags-y := -Iinclude/drm -Idrivers/gpu/drm/exynos
 exynosdrm-y := exynos_drm_drv.o exynos_drm_encoder.o exynos_drm_connector.o \
 		exynos_drm_crtc.o exynos_drm_fbdev.o exynos_drm_fb.o \
-		exynos_drm_buf.o exynos_drm_gem.o exynos_drm_core.o
+		exynos_drm_buf.o exynos_drm_gem.o exynos_drm_core.o \
+		exynos_drm_plane.o
 
 obj-$(CONFIG_DRM_EXYNOS) += exynosdrm.o
 obj-$(CONFIG_DRM_EXYNOS_FIMD) += exynos_drm_fimd.o
+obj-$(CONFIG_DRM_EXYNOS_HDMI) += exynos_hdmi.o exynos_mixer.o exynos_ddc.o \
+				 exynos_hdmiphy.o exynos_drm_hdmi.o
diff --git a/drivers/gpu/drm/exynos/exynos_ddc.c b/drivers/gpu/drm/exynos/exynos_ddc.c
new file mode 100644
index 0000000..84b614f
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_ddc.c
@@ -0,0 +1,58 @@
+/*
+ * Copyright (C) 2011 Samsung Electronics Co.Ltd
+ * Authors:
+ *	Seung-Woo Kim <sw0312.kim@samsung.com>
+ *	Inki Dae <inki.dae@samsung.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include "drmP.h"
+
+#include <linux/kernel.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+
+
+#include "exynos_drm_drv.h"
+#include "exynos_hdmi.h"
+
+static int s5p_ddc_probe(struct i2c_client *client,
+			const struct i2c_device_id *dev_id)
+{
+	hdmi_attach_ddc_client(client);
+
+	dev_info(&client->adapter->dev, "attached s5p_ddc "
+		"into i2c adapter successfully\n");
+
+	return 0;
+}
+
+static int s5p_ddc_remove(struct i2c_client *client)
+{
+	dev_info(&client->adapter->dev, "detached s5p_ddc "
+		"from i2c adapter successfully\n");
+
+	return 0;
+}
+
+static struct i2c_device_id ddc_idtable[] = {
+	{"s5p_ddc", 0},
+	{ },
+};
+
+struct i2c_driver ddc_driver = {
+	.driver = {
+		.name = "s5p_ddc",
+		.owner = THIS_MODULE,
+	},
+	.id_table	= ddc_idtable,
+	.probe		= s5p_ddc_probe,
+	.remove		= __devexit_p(s5p_ddc_remove),
+	.command		= NULL,
+};
+EXPORT_SYMBOL(ddc_driver);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.c b/drivers/gpu/drm/exynos/exynos_drm_buf.c
index 6f8afea..3cf785c 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_buf.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_buf.c
@@ -27,82 +27,83 @@
 #include "drm.h"
 
 #include "exynos_drm_drv.h"
+#include "exynos_drm_gem.h"
 #include "exynos_drm_buf.h"
 
-static DEFINE_MUTEX(exynos_drm_buf_lock);
-
 static int lowlevel_buffer_allocate(struct drm_device *dev,
-		struct exynos_drm_buf_entry *entry)
+		struct exynos_drm_gem_buf *buffer)
 {
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	entry->vaddr = dma_alloc_writecombine(dev->dev, entry->size,
-			(dma_addr_t *)&entry->paddr, GFP_KERNEL);
-	if (!entry->paddr) {
+	buffer->kvaddr = dma_alloc_writecombine(dev->dev, buffer->size,
+			&buffer->dma_addr, GFP_KERNEL);
+	if (!buffer->kvaddr) {
 		DRM_ERROR("failed to allocate buffer.\n");
 		return -ENOMEM;
 	}
 
-	DRM_DEBUG_KMS("allocated : vaddr(0x%x), paddr(0x%x), size(0x%x)\n",
-			(unsigned int)entry->vaddr, entry->paddr, entry->size);
+	DRM_DEBUG_KMS("vaddr(0x%lx), dma_addr(0x%lx), size(0x%lx)\n",
+			(unsigned long)buffer->kvaddr,
+			(unsigned long)buffer->dma_addr,
+			buffer->size);
 
 	return 0;
 }
 
 static void lowlevel_buffer_deallocate(struct drm_device *dev,
-		struct exynos_drm_buf_entry *entry)
+		struct exynos_drm_gem_buf *buffer)
 {
 	DRM_DEBUG_KMS("%s.\n", __FILE__);
 
-	if (entry->paddr && entry->vaddr && entry->size)
-		dma_free_writecombine(dev->dev, entry->size, entry->vaddr,
-				entry->paddr);
+	if (buffer->dma_addr && buffer->size)
+		dma_free_writecombine(dev->dev, buffer->size, buffer->kvaddr,
+				(dma_addr_t)buffer->dma_addr);
 	else
-		DRM_DEBUG_KMS("entry data is null.\n");
+		DRM_DEBUG_KMS("buffer data are invalid.\n");
 }
 
-struct exynos_drm_buf_entry *exynos_drm_buf_create(struct drm_device *dev,
+struct exynos_drm_gem_buf *exynos_drm_buf_create(struct drm_device *dev,
 		unsigned int size)
 {
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 
 	DRM_DEBUG_KMS("%s.\n", __FILE__);
+	DRM_DEBUG_KMS("desired size = 0x%x\n", size);
 
-	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
-	if (!entry) {
-		DRM_ERROR("failed to allocate exynos_drm_buf_entry.\n");
-		return ERR_PTR(-ENOMEM);
+	buffer = kzalloc(sizeof(*buffer), GFP_KERNEL);
+	if (!buffer) {
+		DRM_ERROR("failed to allocate exynos_drm_gem_buf.\n");
+		return NULL;
 	}
 
-	entry->size = size;
+	buffer->size = size;
 
 	/*
 	 * allocate memory region with size and set the memory information
-	 * to vaddr and paddr of a entry object.
+	 * to vaddr and dma_addr of a buffer object.
 	 */
-	if (lowlevel_buffer_allocate(dev, entry) < 0) {
-		kfree(entry);
-		entry = NULL;
-		return ERR_PTR(-ENOMEM);
+	if (lowlevel_buffer_allocate(dev, buffer) < 0) {
+		kfree(buffer);
+		return NULL;
 	}
 
-	return entry;
+	return buffer;
 }
 
 void exynos_drm_buf_destroy(struct drm_device *dev,
-		struct exynos_drm_buf_entry *entry)
+		struct exynos_drm_gem_buf *buffer)
 {
 	DRM_DEBUG_KMS("%s.\n", __FILE__);
 
-	if (!entry) {
-		DRM_DEBUG_KMS("entry is null.\n");
+	if (!buffer) {
+		DRM_DEBUG_KMS("buffer is null.\n");
 		return;
 	}
 
-	lowlevel_buffer_deallocate(dev, entry);
+	lowlevel_buffer_deallocate(dev, buffer);
 
-	kfree(entry);
-	entry = NULL;
+	kfree(buffer);
+	buffer = NULL;
 }
 
 MODULE_AUTHOR("Inki Dae <inki.dae@samsung.com>");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.h b/drivers/gpu/drm/exynos/exynos_drm_buf.h
index 045d59e..c913f2b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_buf.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_buf.h
@@ -26,28 +26,12 @@
 #ifndef _EXYNOS_DRM_BUF_H_
 #define _EXYNOS_DRM_BUF_H_
 
-/*
- * exynos drm buffer entry structure.
- *
- * @paddr: physical address of allocated memory.
- * @vaddr: kernel virtual address of allocated memory.
- * @size: size of allocated memory.
- */
-struct exynos_drm_buf_entry {
-	dma_addr_t paddr;
-	void __iomem *vaddr;
-	unsigned int size;
-};
-
 /* allocate physical memory. */
-struct exynos_drm_buf_entry *exynos_drm_buf_create(struct drm_device *dev,
+struct exynos_drm_gem_buf *exynos_drm_buf_create(struct drm_device *dev,
 		unsigned int size);
 
-/* get physical memory information of a drm framebuffer. */
-struct exynos_drm_buf_entry *exynos_drm_fb_get_buf(struct drm_framebuffer *fb);
-
 /* remove allocated physical memory. */
 void exynos_drm_buf_destroy(struct drm_device *dev,
-		struct exynos_drm_buf_entry *entry);
+		struct exynos_drm_gem_buf *buffer);
 
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c
index 985d9e7..d620b07 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_connector.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_connector.c
@@ -37,6 +37,8 @@
 
 struct exynos_drm_connector {
 	struct drm_connector	drm_connector;
+	uint32_t		encoder_id;
+	struct exynos_drm_manager *manager;
 };
 
 /* convert exynos_video_timings to drm_display_mode */
@@ -47,6 +49,7 @@
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
 	mode->clock = timing->pixclock / 1000;
+	mode->vrefresh = timing->refresh;
 
 	mode->hdisplay = timing->xres;
 	mode->hsync_start = mode->hdisplay + timing->left_margin;
@@ -57,6 +60,12 @@
 	mode->vsync_start = mode->vdisplay + timing->upper_margin;
 	mode->vsync_end = mode->vsync_start + timing->vsync_len;
 	mode->vtotal = mode->vsync_end + timing->lower_margin;
+
+	if (timing->vmode & FB_VMODE_INTERLACED)
+		mode->flags |= DRM_MODE_FLAG_INTERLACE;
+
+	if (timing->vmode & FB_VMODE_DOUBLE)
+		mode->flags |= DRM_MODE_FLAG_DBLSCAN;
 }
 
 /* convert drm_display_mode to exynos_video_timings */
@@ -69,7 +78,7 @@
 	memset(timing, 0, sizeof(*timing));
 
 	timing->pixclock = mode->clock * 1000;
-	timing->refresh = mode->vrefresh;
+	timing->refresh = drm_mode_vrefresh(mode);
 
 	timing->xres = mode->hdisplay;
 	timing->left_margin = mode->hsync_start - mode->hdisplay;
@@ -92,15 +101,16 @@
 
 static int exynos_drm_connector_get_modes(struct drm_connector *connector)
 {
-	struct exynos_drm_manager *manager =
-				exynos_drm_get_manager(connector->encoder);
-	struct exynos_drm_display *display = manager->display;
+	struct exynos_drm_connector *exynos_connector =
+					to_exynos_connector(connector);
+	struct exynos_drm_manager *manager = exynos_connector->manager;
+	struct exynos_drm_display_ops *display_ops = manager->display_ops;
 	unsigned int count;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	if (!display) {
-		DRM_DEBUG_KMS("display is null.\n");
+	if (!display_ops) {
+		DRM_DEBUG_KMS("display_ops is null.\n");
 		return 0;
 	}
 
@@ -112,7 +122,7 @@
 	 * P.S. in case of lcd panel, count is always 1 if success
 	 * because lcd panel has only one mode.
 	 */
-	if (display->get_edid) {
+	if (display_ops->get_edid) {
 		int ret;
 		void *edid;
 
@@ -122,7 +132,7 @@
 			return 0;
 		}
 
-		ret = display->get_edid(manager->dev, connector,
+		ret = display_ops->get_edid(manager->dev, connector,
 						edid, MAX_EDID);
 		if (ret < 0) {
 			DRM_ERROR("failed to get edid data.\n");
@@ -140,8 +150,8 @@
 		struct drm_display_mode *mode = drm_mode_create(connector->dev);
 		struct fb_videomode *timing;
 
-		if (display->get_timing)
-			timing = display->get_timing(manager->dev);
+		if (display_ops->get_timing)
+			timing = display_ops->get_timing(manager->dev);
 		else {
 			drm_mode_destroy(connector->dev, mode);
 			return 0;
@@ -162,9 +172,10 @@
 static int exynos_drm_connector_mode_valid(struct drm_connector *connector,
 					    struct drm_display_mode *mode)
 {
-	struct exynos_drm_manager *manager =
-				exynos_drm_get_manager(connector->encoder);
-	struct exynos_drm_display *display = manager->display;
+	struct exynos_drm_connector *exynos_connector =
+					to_exynos_connector(connector);
+	struct exynos_drm_manager *manager = exynos_connector->manager;
+	struct exynos_drm_display_ops *display_ops = manager->display_ops;
 	struct fb_videomode timing;
 	int ret = MODE_BAD;
 
@@ -172,8 +183,8 @@
 
 	convert_to_video_timing(&timing, mode);
 
-	if (display && display->check_timing)
-		if (!display->check_timing(manager->dev, (void *)&timing))
+	if (display_ops && display_ops->check_timing)
+		if (!display_ops->check_timing(manager->dev, (void *)&timing))
 			ret = MODE_OK;
 
 	return ret;
@@ -181,9 +192,25 @@
 
 struct drm_encoder *exynos_drm_best_encoder(struct drm_connector *connector)
 {
+	struct drm_device *dev = connector->dev;
+	struct exynos_drm_connector *exynos_connector =
+					to_exynos_connector(connector);
+	struct drm_mode_object *obj;
+	struct drm_encoder *encoder;
+
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	return connector->encoder;
+	obj = drm_mode_object_find(dev, exynos_connector->encoder_id,
+				   DRM_MODE_OBJECT_ENCODER);
+	if (!obj) {
+		DRM_DEBUG_KMS("Unknown ENCODER ID %d\n",
+				exynos_connector->encoder_id);
+		return NULL;
+	}
+
+	encoder = obj_to_encoder(obj);
+
+	return encoder;
 }
 
 static struct drm_connector_helper_funcs exynos_connector_helper_funcs = {
@@ -196,15 +223,17 @@
 static enum drm_connector_status
 exynos_drm_connector_detect(struct drm_connector *connector, bool force)
 {
-	struct exynos_drm_manager *manager =
-				exynos_drm_get_manager(connector->encoder);
-	struct exynos_drm_display *display = manager->display;
+	struct exynos_drm_connector *exynos_connector =
+					to_exynos_connector(connector);
+	struct exynos_drm_manager *manager = exynos_connector->manager;
+	struct exynos_drm_display_ops *display_ops =
+					manager->display_ops;
 	enum drm_connector_status status = connector_status_disconnected;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	if (display && display->is_connected) {
-		if (display->is_connected(manager->dev))
+	if (display_ops && display_ops->is_connected) {
+		if (display_ops->is_connected(manager->dev))
 			status = connector_status_connected;
 		else
 			status = connector_status_disconnected;
@@ -251,9 +280,11 @@
 
 	connector = &exynos_connector->drm_connector;
 
-	switch (manager->display->type) {
+	switch (manager->display_ops->type) {
 	case EXYNOS_DISPLAY_TYPE_HDMI:
 		type = DRM_MODE_CONNECTOR_HDMIA;
+		connector->interlace_allowed = true;
+		connector->polled = DRM_CONNECTOR_POLL_HPD;
 		break;
 	default:
 		type = DRM_MODE_CONNECTOR_Unknown;
@@ -267,7 +298,10 @@
 	if (err)
 		goto err_connector;
 
+	exynos_connector->encoder_id = encoder->base.id;
+	exynos_connector->manager = manager;
 	connector->encoder = encoder;
+
 	err = drm_mode_connector_attach_encoder(connector, encoder);
 	if (err) {
 		DRM_ERROR("failed to attach a connector to a encoder\n");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 9337e5e..e3861ac 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -29,36 +29,16 @@
 #include "drmP.h"
 #include "drm_crtc_helper.h"
 
+#include "exynos_drm_crtc.h"
 #include "exynos_drm_drv.h"
 #include "exynos_drm_fb.h"
 #include "exynos_drm_encoder.h"
-#include "exynos_drm_buf.h"
+#include "exynos_drm_gem.h"
 
 #define to_exynos_crtc(x)	container_of(x, struct exynos_drm_crtc,\
 				drm_crtc)
 
 /*
- * Exynos specific crtc postion structure.
- *
- * @fb_x: offset x on a framebuffer to be displyed
- *	- the unit is screen coordinates.
- * @fb_y: offset y on a framebuffer to be displayed
- *	- the unit is screen coordinates.
- * @crtc_x: offset x on hardware screen.
- * @crtc_y: offset y on hardware screen.
- * @crtc_w: width of hardware screen.
- * @crtc_h: height of hardware screen.
- */
-struct exynos_drm_crtc_pos {
-	unsigned int fb_x;
-	unsigned int fb_y;
-	unsigned int crtc_x;
-	unsigned int crtc_y;
-	unsigned int crtc_w;
-	unsigned int crtc_h;
-};
-
-/*
  * Exynos specific crtc structure.
  *
  * @drm_crtc: crtc object.
@@ -71,11 +51,13 @@
  *	drm framework doesn't support multiple irq yet.
  *	we can refer to the crtc to current hardware interrupt occured through
  *	this pipe value.
+ * @dpms: store the crtc dpms value
  */
 struct exynos_drm_crtc {
 	struct drm_crtc			drm_crtc;
 	struct exynos_drm_overlay	overlay;
 	unsigned int			pipe;
+	unsigned int			dpms;
 };
 
 static void exynos_drm_crtc_apply(struct drm_crtc *crtc)
@@ -85,31 +67,36 @@
 
 	exynos_drm_fn_encoder(crtc, overlay,
 			exynos_drm_encoder_crtc_mode_set);
-	exynos_drm_fn_encoder(crtc, NULL, exynos_drm_encoder_crtc_commit);
+	exynos_drm_fn_encoder(crtc, &exynos_crtc->pipe,
+			exynos_drm_encoder_crtc_commit);
 }
 
-static int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay,
-				       struct drm_framebuffer *fb,
-				       struct drm_display_mode *mode,
-				       struct exynos_drm_crtc_pos *pos)
+int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay,
+			      struct drm_framebuffer *fb,
+			      struct drm_display_mode *mode,
+			      struct exynos_drm_crtc_pos *pos)
 {
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 	unsigned int actual_w;
 	unsigned int actual_h;
+	int nr = exynos_drm_format_num_buffers(fb->pixel_format);
+	int i;
 
-	entry = exynos_drm_fb_get_buf(fb);
-	if (!entry) {
-		DRM_LOG_KMS("entry is null.\n");
-		return -EFAULT;
+	for (i = 0; i < nr; i++) {
+		buffer = exynos_drm_fb_buffer(fb, i);
+		if (!buffer) {
+			DRM_LOG_KMS("buffer is null\n");
+			return -EFAULT;
+		}
+
+		overlay->dma_addr[i] = buffer->dma_addr;
+		overlay->vaddr[i] = buffer->kvaddr;
+
+		DRM_DEBUG_KMS("buffer: %d, vaddr = 0x%lx, dma_addr = 0x%lx\n",
+				i, (unsigned long)overlay->vaddr[i],
+				(unsigned long)overlay->dma_addr[i]);
 	}
 
-	overlay->paddr = entry->paddr;
-	overlay->vaddr = entry->vaddr;
-
-	DRM_DEBUG_KMS("vaddr = 0x%lx, paddr = 0x%lx\n",
-			(unsigned long)overlay->vaddr,
-			(unsigned long)overlay->paddr);
-
 	actual_w = min((mode->hdisplay - pos->crtc_x), pos->crtc_w);
 	actual_h = min((mode->vdisplay - pos->crtc_y), pos->crtc_h);
 
@@ -119,7 +106,8 @@
 	overlay->fb_width = fb->width;
 	overlay->fb_height = fb->height;
 	overlay->bpp = fb->bits_per_pixel;
-	overlay->pitch = fb->pitch;
+	overlay->pitch = fb->pitches[0];
+	overlay->pixel_format = fb->pixel_format;
 
 	/* set overlay range to be displayed. */
 	overlay->crtc_x = pos->crtc_x;
@@ -171,9 +159,37 @@
 
 static void exynos_drm_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
-	DRM_DEBUG_KMS("%s\n", __FILE__);
+	struct drm_device *dev = crtc->dev;
+	struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc);
 
-	/* TODO */
+	DRM_DEBUG_KMS("crtc[%d] mode[%d]\n", crtc->base.id, mode);
+
+	if (exynos_crtc->dpms == mode) {
+		DRM_DEBUG_KMS("desired dpms mode is same as previous one.\n");
+		return;
+	}
+
+	mutex_lock(&dev->struct_mutex);
+
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		exynos_drm_fn_encoder(crtc, &mode,
+				exynos_drm_encoder_crtc_dpms);
+		exynos_crtc->dpms = mode;
+		break;
+	case DRM_MODE_DPMS_STANDBY:
+	case DRM_MODE_DPMS_SUSPEND:
+	case DRM_MODE_DPMS_OFF:
+		exynos_drm_fn_encoder(crtc, &mode,
+				exynos_drm_encoder_crtc_dpms);
+		exynos_crtc->dpms = mode;
+		break;
+	default:
+		DRM_ERROR("unspecified mode %d\n", mode);
+		break;
+	}
+
+	mutex_unlock(&dev->struct_mutex);
 }
 
 static void exynos_drm_crtc_prepare(struct drm_crtc *crtc)
@@ -185,9 +201,34 @@
 
 static void exynos_drm_crtc_commit(struct drm_crtc *crtc)
 {
+	struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc);
+
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	/* drm framework doesn't check NULL. */
+	/*
+	 * when set_crtc is requested from user or at booting time,
+	 * crtc->commit would be called without dpms call so if dpms is
+	 * no power on then crtc->dpms should be called
+	 * with DRM_MODE_DPMS_ON for the hardware power to be on.
+	 */
+	if (exynos_crtc->dpms != DRM_MODE_DPMS_ON) {
+		int mode = DRM_MODE_DPMS_ON;
+
+		/*
+		 * enable hardware(power on) to all encoders hdmi connected
+		 * to current crtc.
+		 */
+		exynos_drm_crtc_dpms(crtc, mode);
+		/*
+		 * enable dma to all encoders connected to current crtc and
+		 * lcd panel.
+		 */
+		exynos_drm_fn_encoder(crtc, &mode,
+					exynos_drm_encoder_dpms_from_crtc);
+	}
+
+	exynos_drm_fn_encoder(crtc, &exynos_crtc->pipe,
+			exynos_drm_encoder_crtc_commit);
 }
 
 static bool
@@ -342,6 +383,8 @@
 	}
 
 	exynos_crtc->pipe = nr;
+	exynos_crtc->dpms = DRM_MODE_DPMS_OFF;
+	exynos_crtc->overlay.zpos = DEFAULT_ZPOS;
 	crtc = &exynos_crtc->drm_crtc;
 
 	private->crtc[nr] = crtc;
@@ -355,9 +398,14 @@
 int exynos_drm_crtc_enable_vblank(struct drm_device *dev, int crtc)
 {
 	struct exynos_drm_private *private = dev->dev_private;
+	struct exynos_drm_crtc *exynos_crtc =
+		to_exynos_crtc(private->crtc[crtc]);
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
+	if (exynos_crtc->dpms != DRM_MODE_DPMS_ON)
+		return -EPERM;
+
 	exynos_drm_fn_encoder(private->crtc[crtc], &crtc,
 			exynos_drm_enable_vblank);
 
@@ -367,9 +415,14 @@
 void exynos_drm_crtc_disable_vblank(struct drm_device *dev, int crtc)
 {
 	struct exynos_drm_private *private = dev->dev_private;
+	struct exynos_drm_crtc *exynos_crtc =
+		to_exynos_crtc(private->crtc[crtc]);
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
+	if (exynos_crtc->dpms != DRM_MODE_DPMS_ON)
+		return;
+
 	exynos_drm_fn_encoder(private->crtc[crtc], &crtc,
 			exynos_drm_disable_vblank);
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.h b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
index c584042..25f72a6 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
@@ -35,4 +35,29 @@
 int exynos_drm_crtc_enable_vblank(struct drm_device *dev, int crtc);
 void exynos_drm_crtc_disable_vblank(struct drm_device *dev, int crtc);
 
+/*
+ * Exynos specific crtc postion structure.
+ *
+ * @fb_x: offset x on a framebuffer to be displyed
+ *	- the unit is screen coordinates.
+ * @fb_y: offset y on a framebuffer to be displayed
+ *	- the unit is screen coordinates.
+ * @crtc_x: offset x on hardware screen.
+ * @crtc_y: offset y on hardware screen.
+ * @crtc_w: width of hardware screen.
+ * @crtc_h: height of hardware screen.
+ */
+struct exynos_drm_crtc_pos {
+	unsigned int fb_x;
+	unsigned int fb_y;
+	unsigned int crtc_x;
+	unsigned int crtc_y;
+	unsigned int crtc_w;
+	unsigned int crtc_h;
+};
+
+int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay,
+			      struct drm_framebuffer *fb,
+			      struct drm_display_mode *mode,
+			      struct exynos_drm_crtc_pos *pos);
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 83810cb..35889ca 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -27,6 +27,7 @@
 
 #include "drmP.h"
 #include "drm.h"
+#include "drm_crtc_helper.h"
 
 #include <drm/exynos_drm.h>
 
@@ -35,13 +36,16 @@
 #include "exynos_drm_fbdev.h"
 #include "exynos_drm_fb.h"
 #include "exynos_drm_gem.h"
+#include "exynos_drm_plane.h"
 
-#define DRIVER_NAME	"exynos-drm"
+#define DRIVER_NAME	"exynos"
 #define DRIVER_DESC	"Samsung SoC DRM"
 #define DRIVER_DATE	"20110530"
 #define DRIVER_MAJOR	1
 #define DRIVER_MINOR	0
 
+#define VBLANK_OFF_DELAY	50000
+
 static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 {
 	struct exynos_drm_private *private;
@@ -61,6 +65,9 @@
 
 	drm_mode_config_init(dev);
 
+	/* init kms poll for handling hpd */
+	drm_kms_helper_poll_init(dev);
+
 	exynos_drm_mode_config_init(dev);
 
 	/*
@@ -73,6 +80,12 @@
 			goto err_crtc;
 	}
 
+	for (nr = 0; nr < MAX_PLANE; nr++) {
+		ret = exynos_plane_init(dev, nr);
+		if (ret)
+			goto err_crtc;
+	}
+
 	ret = drm_vblank_init(dev, MAX_CRTC);
 	if (ret)
 		goto err_crtc;
@@ -96,6 +109,8 @@
 		goto err_drm_device;
 	}
 
+	drm_vblank_offdelay = VBLANK_OFF_DELAY;
+
 	return 0;
 
 err_drm_device:
@@ -116,6 +131,7 @@
 	exynos_drm_fbdev_fini(dev);
 	exynos_drm_device_unregister(dev);
 	drm_vblank_cleanup(dev);
+	drm_kms_helper_poll_fini(dev);
 	drm_mode_config_cleanup(dev);
 	kfree(dev->dev_private);
 
@@ -158,6 +174,18 @@
 			DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(EXYNOS_GEM_MMAP,
 			exynos_drm_gem_mmap_ioctl, DRM_UNLOCKED | DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(EXYNOS_PLANE_SET_ZPOS, exynos_plane_set_zpos_ioctl,
+			DRM_UNLOCKED | DRM_AUTH),
+};
+
+static const struct file_operations exynos_drm_driver_fops = {
+	.owner		= THIS_MODULE,
+	.open		= drm_open,
+	.mmap		= exynos_drm_gem_mmap,
+	.poll		= drm_poll,
+	.read		= drm_read,
+	.unlocked_ioctl	= drm_ioctl,
+	.release	= drm_release,
 };
 
 static struct drm_driver exynos_drm_driver = {
@@ -177,15 +205,7 @@
 	.dumb_map_offset	= exynos_drm_gem_dumb_map_offset,
 	.dumb_destroy		= exynos_drm_gem_dumb_destroy,
 	.ioctls			= exynos_ioctls,
-	.fops = {
-		.owner		= THIS_MODULE,
-		.open		= drm_open,
-		.mmap		= exynos_drm_gem_mmap,
-		.poll		= drm_poll,
-		.read		= drm_read,
-		.unlocked_ioctl	= drm_ioctl,
-		.release	= drm_release,
-	},
+	.fops			= &exynos_drm_driver_fops,
 	.name	= DRIVER_NAME,
 	.desc	= DRIVER_DESC,
 	.date	= DRIVER_DATE,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index c03683f..e685e1e 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -29,14 +29,20 @@
 #ifndef _EXYNOS_DRM_DRV_H_
 #define _EXYNOS_DRM_DRV_H_
 
+#include <linux/module.h>
 #include "drm.h"
 
 #define MAX_CRTC	2
+#define MAX_PLANE	5
+#define MAX_FB_BUFFER	3
+#define DEFAULT_ZPOS	-1
 
 struct drm_device;
 struct exynos_drm_overlay;
 struct drm_connector;
 
+extern unsigned int drm_vblank_offdelay;
+
 /* this enumerates display type. */
 enum exynos_drm_output_type {
 	EXYNOS_DISPLAY_TYPE_NONE,
@@ -56,8 +62,8 @@
 struct exynos_drm_overlay_ops {
 	void (*mode_set)(struct device *subdrv_dev,
 			 struct exynos_drm_overlay *overlay);
-	void (*commit)(struct device *subdrv_dev);
-	void (*disable)(struct device *subdrv_dev);
+	void (*commit)(struct device *subdrv_dev, int zpos);
+	void (*disable)(struct device *subdrv_dev, int zpos);
 };
 
 /*
@@ -79,9 +85,11 @@
  * @scan_flag: interlace or progressive way.
  *	(it could be DRM_MODE_FLAG_*)
  * @bpp: pixel size.(in bit)
- * @paddr: bus(accessed by dma) physical memory address to this overlay
- *		and this is physically continuous.
- * @vaddr: virtual memory addresss to this overlay.
+ * @pixel_format: fourcc pixel format of this overlay
+ * @dma_addr: array of bus(accessed by dma) address to the memory region
+ *	      allocated for a overlay.
+ * @vaddr: array of virtual memory addresss to this overlay.
+ * @zpos: order of overlay layer(z position).
  * @default_win: a window to be enabled.
  * @color_key: color key on or off.
  * @index_color: if using color key feature then this value would be used
@@ -108,8 +116,10 @@
 	unsigned int scan_flag;
 	unsigned int bpp;
 	unsigned int pitch;
-	dma_addr_t paddr;
-	void __iomem *vaddr;
+	uint32_t pixel_format;
+	dma_addr_t dma_addr[MAX_FB_BUFFER];
+	void __iomem *vaddr[MAX_FB_BUFFER];
+	int zpos;
 
 	bool default_win;
 	bool color_key;
@@ -130,7 +140,7 @@
  * @check_timing: check if timing is valid or not.
  * @power_on: display device on or off.
  */
-struct exynos_drm_display {
+struct exynos_drm_display_ops {
 	enum exynos_drm_output_type type;
 	bool (*is_connected)(struct device *dev);
 	int (*get_edid)(struct device *dev, struct drm_connector *connector,
@@ -143,6 +153,8 @@
 /*
  * Exynos drm manager ops
  *
+ * @dpms: control device power.
+ * @apply: set timing, vblank and overlay data to registers.
  * @mode_set: convert drm_display_mode to hw specific display mode and
  *	      would be called by encoder->mode_set().
  * @commit: set current hw specific display mode to hw.
@@ -150,6 +162,8 @@
  * @disable_vblank: specific driver callback for disabling vblank interrupt.
  */
 struct exynos_drm_manager_ops {
+	void (*dpms)(struct device *subdrv_dev, int mode);
+	void (*apply)(struct device *subdrv_dev);
 	void (*mode_set)(struct device *subdrv_dev, void *mode);
 	void (*commit)(struct device *subdrv_dev);
 	int (*enable_vblank)(struct device *subdrv_dev);
@@ -178,7 +192,7 @@
 	int pipe;
 	struct exynos_drm_manager_ops *ops;
 	struct exynos_drm_overlay_ops *overlay_ops;
-	struct exynos_drm_display *display;
+	struct exynos_drm_display_ops *display_ops;
 };
 
 /*
diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.c b/drivers/gpu/drm/exynos/exynos_drm_encoder.c
index 7cf6fa8..86b93dd 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_encoder.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.c
@@ -42,30 +42,70 @@
  * @drm_encoder: encoder object.
  * @manager: specific encoder has its own manager to control a hardware
  *	appropriately and we can access a hardware drawing on this manager.
+ * @dpms: store the encoder dpms value.
  */
 struct exynos_drm_encoder {
 	struct drm_encoder		drm_encoder;
 	struct exynos_drm_manager	*manager;
+	int dpms;
 };
 
-static void exynos_drm_encoder_dpms(struct drm_encoder *encoder, int mode)
+static void exynos_drm_display_power(struct drm_encoder *encoder, int mode)
 {
 	struct drm_device *dev = encoder->dev;
 	struct drm_connector *connector;
 	struct exynos_drm_manager *manager = exynos_drm_get_manager(encoder);
 
-	DRM_DEBUG_KMS("%s, encoder dpms: %d\n", __FILE__, mode);
-
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		if (connector->encoder == encoder) {
-			struct exynos_drm_display *display = manager->display;
+			struct exynos_drm_display_ops *display_ops =
+							manager->display_ops;
 
-			if (display && display->power_on)
-				display->power_on(manager->dev, mode);
+			DRM_DEBUG_KMS("connector[%d] dpms[%d]\n",
+					connector->base.id, mode);
+			if (display_ops && display_ops->power_on)
+				display_ops->power_on(manager->dev, mode);
 		}
 	}
 }
 
+static void exynos_drm_encoder_dpms(struct drm_encoder *encoder, int mode)
+{
+	struct drm_device *dev = encoder->dev;
+	struct exynos_drm_manager *manager = exynos_drm_get_manager(encoder);
+	struct exynos_drm_manager_ops *manager_ops = manager->ops;
+	struct exynos_drm_encoder *exynos_encoder = to_exynos_encoder(encoder);
+
+	DRM_DEBUG_KMS("%s, encoder dpms: %d\n", __FILE__, mode);
+
+	if (exynos_encoder->dpms == mode) {
+		DRM_DEBUG_KMS("desired dpms mode is same as previous one.\n");
+		return;
+	}
+
+	mutex_lock(&dev->struct_mutex);
+
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		if (manager_ops && manager_ops->apply)
+			manager_ops->apply(manager->dev);
+		exynos_drm_display_power(encoder, mode);
+		exynos_encoder->dpms = mode;
+		break;
+	case DRM_MODE_DPMS_STANDBY:
+	case DRM_MODE_DPMS_SUSPEND:
+	case DRM_MODE_DPMS_OFF:
+		exynos_drm_display_power(encoder, mode);
+		exynos_encoder->dpms = mode;
+		break;
+	default:
+		DRM_ERROR("unspecified mode %d\n", mode);
+		break;
+	}
+
+	mutex_unlock(&dev->struct_mutex);
+}
+
 static bool
 exynos_drm_encoder_mode_fixup(struct drm_encoder *encoder,
 			       struct drm_display_mode *mode,
@@ -116,15 +156,11 @@
 {
 	struct exynos_drm_manager *manager = exynos_drm_get_manager(encoder);
 	struct exynos_drm_manager_ops *manager_ops = manager->ops;
-	struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
 	if (manager_ops && manager_ops->commit)
 		manager_ops->commit(manager->dev);
-
-	if (overlay_ops && overlay_ops->commit)
-		overlay_ops->commit(manager->dev);
 }
 
 static struct drm_crtc *
@@ -152,7 +188,6 @@
 	exynos_encoder->manager->pipe = -1;
 
 	drm_encoder_cleanup(encoder);
-	encoder->dev->mode_config.num_encoder--;
 	kfree(exynos_encoder);
 }
 
@@ -182,6 +217,7 @@
 		return NULL;
 	}
 
+	exynos_encoder->dpms = DRM_MODE_DPMS_OFF;
 	exynos_encoder->manager = manager;
 	encoder = &exynos_encoder->drm_encoder;
 	encoder->possible_crtcs = possible_crtcs;
@@ -208,10 +244,23 @@
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_encoder *encoder;
+	struct exynos_drm_private *private = dev->dev_private;
+	struct exynos_drm_manager *manager;
 
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
-		if (encoder->crtc != crtc)
-			continue;
+		/*
+		 * if crtc is detached from encoder, check pipe,
+		 * otherwise check crtc attached to encoder
+		 */
+		if (!encoder->crtc) {
+			manager = to_exynos_encoder(encoder)->manager;
+			if (manager->pipe < 0 ||
+					private->crtc[manager->pipe] != crtc)
+				continue;
+		} else {
+			if (encoder->crtc != crtc)
+				continue;
+		}
 
 		fn(encoder, data);
 	}
@@ -245,13 +294,83 @@
 		manager_ops->disable_vblank(manager->dev);
 }
 
-void exynos_drm_encoder_crtc_commit(struct drm_encoder *encoder, void *data)
+void exynos_drm_encoder_crtc_plane_commit(struct drm_encoder *encoder,
+					  void *data)
 {
 	struct exynos_drm_manager *manager =
 		to_exynos_encoder(encoder)->manager;
 	struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops;
+	int zpos = DEFAULT_ZPOS;
 
-	overlay_ops->commit(manager->dev);
+	if (data)
+		zpos = *(int *)data;
+
+	if (overlay_ops && overlay_ops->commit)
+		overlay_ops->commit(manager->dev, zpos);
+}
+
+void exynos_drm_encoder_crtc_commit(struct drm_encoder *encoder, void *data)
+{
+	struct exynos_drm_manager *manager =
+		to_exynos_encoder(encoder)->manager;
+	int crtc = *(int *)data;
+	int zpos = DEFAULT_ZPOS;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	/*
+	 * when crtc is detached from encoder, this pipe is used
+	 * to select manager operation
+	 */
+	manager->pipe = crtc;
+
+	exynos_drm_encoder_crtc_plane_commit(encoder, &zpos);
+}
+
+void exynos_drm_encoder_dpms_from_crtc(struct drm_encoder *encoder, void *data)
+{
+	struct exynos_drm_encoder *exynos_encoder = to_exynos_encoder(encoder);
+	int mode = *(int *)data;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	exynos_drm_encoder_dpms(encoder, mode);
+
+	exynos_encoder->dpms = mode;
+}
+
+void exynos_drm_encoder_crtc_dpms(struct drm_encoder *encoder, void *data)
+{
+	struct drm_device *dev = encoder->dev;
+	struct exynos_drm_encoder *exynos_encoder = to_exynos_encoder(encoder);
+	struct exynos_drm_manager *manager = exynos_encoder->manager;
+	struct exynos_drm_manager_ops *manager_ops = manager->ops;
+	struct drm_connector *connector;
+	int mode = *(int *)data;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (manager_ops && manager_ops->dpms)
+		manager_ops->dpms(manager->dev, mode);
+
+	/*
+	 * set current dpms mode to the connector connected to
+	 * current encoder. connector->dpms would be checked
+	 * at drm_helper_connector_dpms()
+	 */
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
+		if (connector->encoder == encoder)
+			connector->dpms = mode;
+
+	/*
+	 * if this condition is ok then it means that the crtc is already
+	 * detached from encoder and last function for detaching is properly
+	 * done, so clear pipe from manager to prevent repeated call.
+	 */
+	if (mode > DRM_MODE_DPMS_ON) {
+		if (!encoder->crtc)
+			manager->pipe = -1;
+	}
 }
 
 void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data)
@@ -261,7 +380,24 @@
 	struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops;
 	struct exynos_drm_overlay *overlay = data;
 
-	overlay_ops->mode_set(manager->dev, overlay);
+	if (overlay_ops && overlay_ops->mode_set)
+		overlay_ops->mode_set(manager->dev, overlay);
+}
+
+void exynos_drm_encoder_crtc_disable(struct drm_encoder *encoder, void *data)
+{
+	struct exynos_drm_manager *manager =
+		to_exynos_encoder(encoder)->manager;
+	struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops;
+	int zpos = DEFAULT_ZPOS;
+
+	DRM_DEBUG_KMS("\n");
+
+	if (data)
+		zpos = *(int *)data;
+
+	if (overlay_ops && overlay_ops->disable)
+		overlay_ops->disable(manager->dev, zpos);
 }
 
 MODULE_AUTHOR("Inki Dae <inki.dae@samsung.com>");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.h b/drivers/gpu/drm/exynos/exynos_drm_encoder.h
index 5ecd645..97b087a 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_encoder.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.h
@@ -39,7 +39,13 @@
 			    void (*fn)(struct drm_encoder *, void *));
 void exynos_drm_enable_vblank(struct drm_encoder *encoder, void *data);
 void exynos_drm_disable_vblank(struct drm_encoder *encoder, void *data);
+void exynos_drm_encoder_crtc_plane_commit(struct drm_encoder *encoder,
+					  void *data);
 void exynos_drm_encoder_crtc_commit(struct drm_encoder *encoder, void *data);
+void exynos_drm_encoder_dpms_from_crtc(struct drm_encoder *encoder,
+					void *data);
+void exynos_drm_encoder_crtc_dpms(struct drm_encoder *encoder, void *data);
 void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data);
+void exynos_drm_encoder_crtc_disable(struct drm_encoder *encoder, void *data);
 
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c
index 48d29cf..3733fe6 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c
@@ -29,9 +29,10 @@
 #include "drmP.h"
 #include "drm_crtc.h"
 #include "drm_crtc_helper.h"
+#include "drm_fb_helper.h"
 
+#include "exynos_drm_drv.h"
 #include "exynos_drm_fb.h"
-#include "exynos_drm_buf.h"
 #include "exynos_drm_gem.h"
 
 #define to_exynos_fb(x)	container_of(x, struct exynos_drm_fb, fb)
@@ -40,15 +41,11 @@
  * exynos specific framebuffer structure.
  *
  * @fb: drm framebuffer obejct.
- * @exynos_gem_obj: exynos specific gem object containing a gem object.
- * @entry: pointer to exynos drm buffer entry object.
- *	- containing only the information to physically continuous memory
- *	region allocated at default framebuffer creation.
+ * @exynos_gem_obj: array of exynos specific gem object containing a gem object.
  */
 struct exynos_drm_fb {
 	struct drm_framebuffer		fb;
-	struct exynos_drm_gem_obj	*exynos_gem_obj;
-	struct exynos_drm_buf_entry	*entry;
+	struct exynos_drm_gem_obj	*exynos_gem_obj[MAX_FB_BUFFER];
 };
 
 static void exynos_drm_fb_destroy(struct drm_framebuffer *fb)
@@ -59,13 +56,6 @@
 
 	drm_framebuffer_cleanup(fb);
 
-	/*
-	 * default framebuffer has no gem object so
-	 * a buffer of the default framebuffer should be released at here.
-	 */
-	if (!exynos_fb->exynos_gem_obj && exynos_fb->entry)
-		exynos_drm_buf_destroy(fb->dev, exynos_fb->entry);
-
 	kfree(exynos_fb);
 	exynos_fb = NULL;
 }
@@ -79,7 +69,7 @@
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
 	return drm_gem_handle_create(file_priv,
-			&exynos_fb->exynos_gem_obj->base, handle);
+			&exynos_fb->exynos_gem_obj[0]->base, handle);
 }
 
 static int exynos_drm_fb_dirty(struct drm_framebuffer *fb,
@@ -100,146 +90,110 @@
 	.dirty		= exynos_drm_fb_dirty,
 };
 
-static struct drm_framebuffer *
-exynos_drm_fb_init(struct drm_file *file_priv, struct drm_device *dev,
-		    struct drm_mode_fb_cmd *mode_cmd)
+struct drm_framebuffer *
+exynos_drm_framebuffer_init(struct drm_device *dev,
+			    struct drm_mode_fb_cmd2 *mode_cmd,
+			    struct drm_gem_object *obj)
 {
 	struct exynos_drm_fb *exynos_fb;
-	struct drm_framebuffer *fb;
-	struct exynos_drm_gem_obj *exynos_gem_obj = NULL;
-	struct drm_gem_object *obj;
-	unsigned int size;
 	int ret;
 
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
-	mode_cmd->pitch = max(mode_cmd->pitch,
-			mode_cmd->width * (mode_cmd->bpp >> 3));
-
-	DRM_LOG_KMS("drm fb create(%dx%d)\n",
-			mode_cmd->width, mode_cmd->height);
-
 	exynos_fb = kzalloc(sizeof(*exynos_fb), GFP_KERNEL);
 	if (!exynos_fb) {
-		DRM_ERROR("failed to allocate exynos drm framebuffer.\n");
+		DRM_ERROR("failed to allocate exynos drm framebuffer\n");
 		return ERR_PTR(-ENOMEM);
 	}
 
-	fb = &exynos_fb->fb;
-	ret = drm_framebuffer_init(dev, fb, &exynos_drm_fb_funcs);
+	ret = drm_framebuffer_init(dev, &exynos_fb->fb, &exynos_drm_fb_funcs);
 	if (ret) {
-		DRM_ERROR("failed to initialize framebuffer.\n");
-		goto err_init;
+		DRM_ERROR("failed to initialize framebuffer\n");
+		return ERR_PTR(ret);
 	}
 
-	DRM_LOG_KMS("create: fb id: %d\n", fb->base.id);
+	drm_helper_mode_fill_fb_struct(&exynos_fb->fb, mode_cmd);
+	exynos_fb->exynos_gem_obj[0] = to_exynos_gem_obj(obj);
 
-	size = mode_cmd->pitch * mode_cmd->height;
+	return &exynos_fb->fb;
+}
 
-	/*
-	 * mode_cmd->handle could be NULL at booting time or
-	 * with user request. if NULL, a new buffer or a gem object
-	 * would be allocated.
-	 */
-	if (!mode_cmd->handle) {
-		if (!file_priv) {
-			struct exynos_drm_buf_entry *entry;
+static struct drm_framebuffer *
+exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
+		      struct drm_mode_fb_cmd2 *mode_cmd)
+{
+	struct drm_gem_object *obj;
+	struct drm_framebuffer *fb;
+	struct exynos_drm_fb *exynos_fb;
+	int nr;
+	int i;
 
-			/*
-			 * in case that file_priv is NULL, it allocates
-			 * only buffer and this buffer would be used
-			 * for default framebuffer.
-			 */
-			entry = exynos_drm_buf_create(dev, size);
-			if (IS_ERR(entry)) {
-				ret = PTR_ERR(entry);
-				goto err_buffer;
-			}
+	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-			exynos_fb->entry = entry;
+	obj = drm_gem_object_lookup(dev, file_priv, mode_cmd->handles[0]);
+	if (!obj) {
+		DRM_ERROR("failed to lookup gem object\n");
+		return ERR_PTR(-ENOENT);
+	}
 
-			DRM_LOG_KMS("default fb: paddr = 0x%lx, size = 0x%x\n",
-					(unsigned long)entry->paddr, size);
+	drm_gem_object_unreference_unlocked(obj);
 
-			goto out;
-		} else {
-			exynos_gem_obj = exynos_drm_gem_create(file_priv, dev,
-							size,
-							&mode_cmd->handle);
-			if (IS_ERR(exynos_gem_obj)) {
-				ret = PTR_ERR(exynos_gem_obj);
-				goto err_buffer;
-			}
-		}
-	} else {
-		obj = drm_gem_object_lookup(dev, file_priv, mode_cmd->handle);
+	fb = exynos_drm_framebuffer_init(dev, mode_cmd, obj);
+	if (IS_ERR(fb))
+		return fb;
+
+	exynos_fb = to_exynos_fb(fb);
+	nr = exynos_drm_format_num_buffers(fb->pixel_format);
+
+	for (i = 1; i < nr; i++) {
+		obj = drm_gem_object_lookup(dev, file_priv,
+				mode_cmd->handles[i]);
 		if (!obj) {
-			DRM_ERROR("failed to lookup gem object.\n");
-			goto err_buffer;
+			DRM_ERROR("failed to lookup gem object\n");
+			exynos_drm_fb_destroy(fb);
+			return ERR_PTR(-ENOENT);
 		}
 
-		exynos_gem_obj = to_exynos_gem_obj(obj);
-
 		drm_gem_object_unreference_unlocked(obj);
+
+		exynos_fb->exynos_gem_obj[i] = to_exynos_gem_obj(obj);
 	}
 
-	/*
-	 * if got a exynos_gem_obj from either a handle or
-	 * a new creation then exynos_fb->exynos_gem_obj is NULL
-	 * so that default framebuffer has no its own gem object,
-	 * only its own buffer object.
-	 */
-	exynos_fb->entry = exynos_gem_obj->entry;
-
-	DRM_LOG_KMS("paddr = 0x%lx, size = 0x%x, gem object = 0x%x\n",
-			(unsigned long)exynos_fb->entry->paddr, size,
-			(unsigned int)&exynos_gem_obj->base);
-
-out:
-	exynos_fb->exynos_gem_obj = exynos_gem_obj;
-
-	drm_helper_mode_fill_fb_struct(fb, mode_cmd);
-
 	return fb;
-
-err_buffer:
-	drm_framebuffer_cleanup(fb);
-
-err_init:
-	kfree(exynos_fb);
-
-	return ERR_PTR(ret);
 }
 
-struct drm_framebuffer *exynos_drm_fb_create(struct drm_device *dev,
-					      struct drm_file *file_priv,
-					      struct drm_mode_fb_cmd *mode_cmd)
-{
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
-	return exynos_drm_fb_init(file_priv, dev, mode_cmd);
-}
-
-struct exynos_drm_buf_entry *exynos_drm_fb_get_buf(struct drm_framebuffer *fb)
+struct exynos_drm_gem_buf *exynos_drm_fb_buffer(struct drm_framebuffer *fb,
+						int index)
 {
 	struct exynos_drm_fb *exynos_fb = to_exynos_fb(fb);
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	entry = exynos_fb->entry;
-	if (!entry)
+	if (index >= MAX_FB_BUFFER)
 		return NULL;
 
-	DRM_DEBUG_KMS("vaddr = 0x%lx, paddr = 0x%lx\n",
-			(unsigned long)entry->vaddr,
-			(unsigned long)entry->paddr);
+	buffer = exynos_fb->exynos_gem_obj[index]->buffer;
+	if (!buffer)
+		return NULL;
 
-	return entry;
+	DRM_DEBUG_KMS("vaddr = 0x%lx, dma_addr = 0x%lx\n",
+			(unsigned long)buffer->kvaddr,
+			(unsigned long)buffer->dma_addr);
+
+	return buffer;
+}
+
+static void exynos_drm_output_poll_changed(struct drm_device *dev)
+{
+	struct exynos_drm_private *private = dev->dev_private;
+	struct drm_fb_helper *fb_helper = private->fb_helper;
+
+	if (fb_helper)
+		drm_fb_helper_hotplug_event(fb_helper);
 }
 
 static struct drm_mode_config_funcs exynos_drm_mode_config_funcs = {
-	.fb_create = exynos_drm_fb_create,
+	.fb_create = exynos_user_fb_create,
+	.output_poll_changed = exynos_drm_output_poll_changed,
 };
 
 void exynos_drm_mode_config_init(struct drm_device *dev)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.h b/drivers/gpu/drm/exynos/exynos_drm_fb.h
index eb35931..3ecb30d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.h
@@ -28,9 +28,27 @@
 #ifndef _EXYNOS_DRM_FB_H_
 #define _EXYNOS_DRM_FB_H
 
-struct drm_framebuffer *exynos_drm_fb_create(struct drm_device *dev,
-					      struct drm_file *filp,
-					      struct drm_mode_fb_cmd *mode_cmd);
+static inline int exynos_drm_format_num_buffers(uint32_t format)
+{
+	switch (format) {
+	case DRM_FORMAT_NV12M:
+	case DRM_FORMAT_NV12MT:
+		return 2;
+	case DRM_FORMAT_YUV420M:
+		return 3;
+	default:
+		return 1;
+	}
+}
+
+struct drm_framebuffer *
+exynos_drm_framebuffer_init(struct drm_device *dev,
+			    struct drm_mode_fb_cmd2 *mode_cmd,
+			    struct drm_gem_object *obj);
+
+/* get memory information of a drm framebuffer */
+struct exynos_drm_gem_buf *exynos_drm_fb_buffer(struct drm_framebuffer *fb,
+						 int index);
 
 void exynos_drm_mode_config_init(struct drm_device *dev);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index 1f4b3d1..d7ae29d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -33,7 +33,7 @@
 
 #include "exynos_drm_drv.h"
 #include "exynos_drm_fb.h"
-#include "exynos_drm_buf.h"
+#include "exynos_drm_gem.h"
 
 #define MAX_CONNECTOR		4
 #define PREFERRED_BPP		32
@@ -42,8 +42,8 @@
 				drm_fb_helper)
 
 struct exynos_drm_fbdev {
-	struct drm_fb_helper	drm_fb_helper;
-	struct drm_framebuffer	*fb;
+	struct drm_fb_helper		drm_fb_helper;
+	struct exynos_drm_gem_obj	*exynos_gem_obj;
 };
 
 static int exynos_drm_fbdev_set_par(struct fb_info *info)
@@ -85,36 +85,32 @@
 };
 
 static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
-				     struct drm_framebuffer *fb,
-				     unsigned int fb_width,
-				     unsigned int fb_height)
+				     struct drm_framebuffer *fb)
 {
 	struct fb_info *fbi = helper->fbdev;
 	struct drm_device *dev = helper->dev;
-	struct exynos_drm_fbdev *exynos_fb = to_exynos_fbdev(helper);
-	struct exynos_drm_buf_entry *entry;
-	unsigned int size = fb_width * fb_height * (fb->bits_per_pixel >> 3);
+	struct exynos_drm_gem_buf *buffer;
+	unsigned int size = fb->width * fb->height * (fb->bits_per_pixel >> 3);
 	unsigned long offset;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	exynos_fb->fb = fb;
+	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->depth);
+	drm_fb_helper_fill_var(fbi, helper, fb->width, fb->height);
 
-	drm_fb_helper_fill_fix(fbi, fb->pitch, fb->depth);
-	drm_fb_helper_fill_var(fbi, helper, fb_width, fb_height);
-
-	entry = exynos_drm_fb_get_buf(fb);
-	if (!entry) {
-		DRM_LOG_KMS("entry is null.\n");
+	/* RGB formats use only one buffer */
+	buffer = exynos_drm_fb_buffer(fb, 0);
+	if (!buffer) {
+		DRM_LOG_KMS("buffer is null.\n");
 		return -EFAULT;
 	}
 
 	offset = fbi->var.xoffset * (fb->bits_per_pixel >> 3);
-	offset += fbi->var.yoffset * fb->pitch;
+	offset += fbi->var.yoffset * fb->pitches[0];
 
-	dev->mode_config.fb_base = entry->paddr;
-	fbi->screen_base = entry->vaddr + offset;
-	fbi->fix.smem_start = entry->paddr + offset;
+	dev->mode_config.fb_base = (resource_size_t)buffer->dma_addr;
+	fbi->screen_base = buffer->kvaddr + offset;
+	fbi->fix.smem_start = (unsigned long)(buffer->dma_addr + offset);
 	fbi->screen_size = size;
 	fbi->fix.smem_len = size;
 
@@ -125,10 +121,12 @@
 				    struct drm_fb_helper_surface_size *sizes)
 {
 	struct exynos_drm_fbdev *exynos_fbdev = to_exynos_fbdev(helper);
+	struct exynos_drm_gem_obj *exynos_gem_obj;
 	struct drm_device *dev = helper->dev;
 	struct fb_info *fbi;
-	struct drm_mode_fb_cmd mode_cmd = { 0 };
+	struct drm_mode_fb_cmd2 mode_cmd = { 0 };
 	struct platform_device *pdev = dev->platformdev;
+	unsigned long size;
 	int ret;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
@@ -139,8 +137,9 @@
 
 	mode_cmd.width = sizes->surface_width;
 	mode_cmd.height = sizes->surface_height;
-	mode_cmd.bpp = sizes->surface_bpp;
-	mode_cmd.depth = sizes->surface_depth;
+	mode_cmd.pitches[0] = sizes->surface_width * (sizes->surface_bpp >> 3);
+	mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp,
+							  sizes->surface_depth);
 
 	mutex_lock(&dev->struct_mutex);
 
@@ -151,14 +150,23 @@
 		goto out;
 	}
 
-	exynos_fbdev->fb = exynos_drm_fb_create(dev, NULL, &mode_cmd);
-	if (IS_ERR_OR_NULL(exynos_fbdev->fb)) {
-		DRM_ERROR("failed to create drm framebuffer.\n");
-		ret = PTR_ERR(exynos_fbdev->fb);
+	size = mode_cmd.pitches[0] * mode_cmd.height;
+	exynos_gem_obj = exynos_drm_gem_create(dev, size);
+	if (IS_ERR(exynos_gem_obj)) {
+		ret = PTR_ERR(exynos_gem_obj);
 		goto out;
 	}
 
-	helper->fb = exynos_fbdev->fb;
+	exynos_fbdev->exynos_gem_obj = exynos_gem_obj;
+
+	helper->fb = exynos_drm_framebuffer_init(dev, &mode_cmd,
+			&exynos_gem_obj->base);
+	if (IS_ERR_OR_NULL(helper->fb)) {
+		DRM_ERROR("failed to create drm framebuffer.\n");
+		ret = PTR_ERR(helper->fb);
+		goto out;
+	}
+
 	helper->fbdev = fbi;
 
 	fbi->par = helper;
@@ -171,10 +179,11 @@
 		goto out;
 	}
 
-	ret = exynos_drm_fbdev_update(helper, helper->fb, sizes->fb_width,
-			sizes->fb_height);
-	if (ret < 0)
+	ret = exynos_drm_fbdev_update(helper, helper->fb);
+	if (ret < 0) {
 		fb_dealloc_cmap(&fbi->cmap);
+		goto out;
+	}
 
 /*
  * if failed, all resources allocated above would be released by
@@ -207,36 +216,43 @@
 {
 	struct drm_device *dev = helper->dev;
 	struct exynos_drm_fbdev *exynos_fbdev = to_exynos_fbdev(helper);
-	struct drm_framebuffer *fb = exynos_fbdev->fb;
-	struct drm_mode_fb_cmd mode_cmd = { 0 };
+	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct drm_framebuffer *fb = helper->fb;
+	struct drm_mode_fb_cmd2 mode_cmd = { 0 };
+	unsigned long size;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	if (helper->fb != fb) {
-		DRM_ERROR("drm framebuffer is different\n");
-		return -EINVAL;
-	}
-
 	if (exynos_drm_fbdev_is_samefb(fb, sizes))
 		return 0;
 
 	mode_cmd.width = sizes->surface_width;
 	mode_cmd.height = sizes->surface_height;
-	mode_cmd.bpp = sizes->surface_bpp;
-	mode_cmd.depth = sizes->surface_depth;
+	mode_cmd.pitches[0] = sizes->surface_width * (sizes->surface_bpp >> 3);
+	mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp,
+							  sizes->surface_depth);
+
+	if (exynos_fbdev->exynos_gem_obj)
+		exynos_drm_gem_destroy(exynos_fbdev->exynos_gem_obj);
 
 	if (fb->funcs->destroy)
 		fb->funcs->destroy(fb);
 
-	exynos_fbdev->fb = exynos_drm_fb_create(dev, NULL, &mode_cmd);
-	if (IS_ERR(exynos_fbdev->fb)) {
-		DRM_ERROR("failed to allocate fb.\n");
-		return PTR_ERR(exynos_fbdev->fb);
+	size = mode_cmd.pitches[0] * mode_cmd.height;
+	exynos_gem_obj = exynos_drm_gem_create(dev, size);
+	if (IS_ERR(exynos_gem_obj))
+		return PTR_ERR(exynos_gem_obj);
+
+	exynos_fbdev->exynos_gem_obj = exynos_gem_obj;
+
+	helper->fb = exynos_drm_framebuffer_init(dev, &mode_cmd,
+			&exynos_gem_obj->base);
+	if (IS_ERR_OR_NULL(helper->fb)) {
+		DRM_ERROR("failed to create drm framebuffer.\n");
+		return PTR_ERR(helper->fb);
 	}
 
-	helper->fb = exynos_fbdev->fb;
-	return exynos_drm_fbdev_update(helper, helper->fb, sizes->fb_width,
-			sizes->fb_height);
+	return exynos_drm_fbdev_update(helper, helper->fb);
 }
 
 static int exynos_drm_fbdev_probe(struct drm_fb_helper *helper,
@@ -369,6 +385,9 @@
 
 	fbdev = to_exynos_fbdev(private->fb_helper);
 
+	if (fbdev->exynos_gem_obj)
+		exynos_drm_gem_destroy(fbdev->exynos_gem_obj);
+
 	exynos_drm_fbdev_destroy(dev, private->fb_helper);
 	kfree(fbdev);
 	private->fb_helper = NULL;
@@ -405,6 +424,18 @@
 	fb_helper = private->fb_helper;
 
 	if (fb_helper) {
+		struct list_head temp_list;
+
+		INIT_LIST_HEAD(&temp_list);
+
+		/*
+		 * fb_helper is reintialized but kernel fb is reused
+		 * so kernel_fb_list need to be backuped and restored
+		 */
+		if (!list_empty(&fb_helper->kernel_fb_list))
+			list_replace_init(&fb_helper->kernel_fb_list,
+					&temp_list);
+
 		drm_fb_helper_fini(fb_helper);
 
 		ret = drm_fb_helper_init(dev, fb_helper,
@@ -414,6 +445,9 @@
 			return ret;
 		}
 
+		if (!list_empty(&temp_list))
+			list_replace(&temp_list, &fb_helper->kernel_fb_list);
+
 		ret = drm_fb_helper_single_add_all_connectors(fb_helper);
 		if (ret < 0) {
 			DRM_ERROR("failed to add fb helper to connectors\n");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 4659c88..ca83139 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -17,6 +17,7 @@
 #include <linux/module.h>
 #include <linux/platform_device.h>
 #include <linux/clk.h>
+#include <linux/pm_runtime.h>
 
 #include <drm/exynos_drm.h>
 #include <plat/regs-fb-v4.h>
@@ -64,10 +65,11 @@
 	unsigned int		fb_width;
 	unsigned int		fb_height;
 	unsigned int		bpp;
-	dma_addr_t		paddr;
+	dma_addr_t		dma_addr;
 	void __iomem		*vaddr;
 	unsigned int		buf_offsize;
 	unsigned int		line_size;	/* bytes */
+	bool			enabled;
 };
 
 struct fimd_context {
@@ -84,6 +86,8 @@
 	unsigned long			irq_flags;
 	u32				vidcon0;
 	u32				vidcon1;
+	bool				suspended;
+	struct mutex			lock;
 
 	struct fb_videomode		*timing;
 };
@@ -119,12 +123,12 @@
 {
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	/* TODO. */
+	/* TODO */
 
 	return 0;
 }
 
-static struct exynos_drm_display fimd_display = {
+static struct exynos_drm_display_ops fimd_display_ops = {
 	.type = EXYNOS_DISPLAY_TYPE_LCD,
 	.is_connected = fimd_display_is_connected,
 	.get_timing = fimd_get_timing,
@@ -132,12 +136,68 @@
 	.power_on = fimd_display_power_on,
 };
 
+static void fimd_dpms(struct device *subdrv_dev, int mode)
+{
+	struct fimd_context *ctx = get_fimd_context(subdrv_dev);
+
+	DRM_DEBUG_KMS("%s, %d\n", __FILE__, mode);
+
+	mutex_lock(&ctx->lock);
+
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		/*
+		 * enable fimd hardware only if suspended status.
+		 *
+		 * P.S. fimd_dpms function would be called at booting time so
+		 * clk_enable could be called double time.
+		 */
+		if (ctx->suspended)
+			pm_runtime_get_sync(subdrv_dev);
+		break;
+	case DRM_MODE_DPMS_STANDBY:
+	case DRM_MODE_DPMS_SUSPEND:
+	case DRM_MODE_DPMS_OFF:
+		pm_runtime_put_sync(subdrv_dev);
+		break;
+	default:
+		DRM_DEBUG_KMS("unspecified mode %d\n", mode);
+		break;
+	}
+
+	mutex_unlock(&ctx->lock);
+}
+
+static void fimd_apply(struct device *subdrv_dev)
+{
+	struct fimd_context *ctx = get_fimd_context(subdrv_dev);
+	struct exynos_drm_manager *mgr = &ctx->subdrv.manager;
+	struct exynos_drm_manager_ops *mgr_ops = mgr->ops;
+	struct exynos_drm_overlay_ops *ovl_ops = mgr->overlay_ops;
+	struct fimd_win_data *win_data;
+	int i;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	for (i = 0; i < WINDOWS_NR; i++) {
+		win_data = &ctx->win_data[i];
+		if (win_data->enabled && (ovl_ops && ovl_ops->commit))
+			ovl_ops->commit(subdrv_dev, i);
+	}
+
+	if (mgr_ops && mgr_ops->commit)
+		mgr_ops->commit(subdrv_dev);
+}
+
 static void fimd_commit(struct device *dev)
 {
 	struct fimd_context *ctx = get_fimd_context(dev);
 	struct fb_videomode *timing = ctx->timing;
 	u32 val;
 
+	if (ctx->suspended)
+		return;
+
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
 	/* setup polarity values from machine code. */
@@ -184,6 +244,9 @@
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
+	if (ctx->suspended)
+		return -EPERM;
+
 	if (!test_and_set_bit(0, &ctx->irq_flags)) {
 		val = readl(ctx->regs + VIDINTCON0);
 
@@ -208,6 +271,9 @@
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
+	if (ctx->suspended)
+		return;
+
 	if (test_and_clear_bit(0, &ctx->irq_flags)) {
 		val = readl(ctx->regs + VIDINTCON0);
 
@@ -219,6 +285,8 @@
 }
 
 static struct exynos_drm_manager_ops fimd_manager_ops = {
+	.dpms = fimd_dpms,
+	.apply = fimd_apply,
 	.commit = fimd_commit,
 	.enable_vblank = fimd_enable_vblank,
 	.disable_vblank = fimd_disable_vblank,
@@ -229,6 +297,7 @@
 {
 	struct fimd_context *ctx = get_fimd_context(dev);
 	struct fimd_win_data *win_data;
+	int win;
 	unsigned long offset;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
@@ -238,12 +307,19 @@
 		return;
 	}
 
+	win = overlay->zpos;
+	if (win == DEFAULT_ZPOS)
+		win = ctx->default_win;
+
+	if (win < 0 || win > WINDOWS_NR)
+		return;
+
 	offset = overlay->fb_x * (overlay->bpp >> 3);
 	offset += overlay->fb_y * overlay->pitch;
 
 	DRM_DEBUG_KMS("offset = 0x%lx, pitch = %x\n", offset, overlay->pitch);
 
-	win_data = &ctx->win_data[ctx->default_win];
+	win_data = &ctx->win_data[win];
 
 	win_data->offset_x = overlay->crtc_x;
 	win_data->offset_y = overlay->crtc_y;
@@ -251,8 +327,8 @@
 	win_data->ovl_height = overlay->crtc_height;
 	win_data->fb_width = overlay->fb_width;
 	win_data->fb_height = overlay->fb_height;
-	win_data->paddr = overlay->paddr + offset;
-	win_data->vaddr = overlay->vaddr + offset;
+	win_data->dma_addr = overlay->dma_addr[0] + offset;
+	win_data->vaddr = overlay->vaddr[0] + offset;
 	win_data->bpp = overlay->bpp;
 	win_data->buf_offsize = (overlay->fb_width - overlay->crtc_width) *
 				(overlay->bpp >> 3);
@@ -263,7 +339,7 @@
 	DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n",
 			win_data->ovl_width, win_data->ovl_height);
 	DRM_DEBUG_KMS("paddr = 0x%lx, vaddr = 0x%lx\n",
-			(unsigned long)win_data->paddr,
+			(unsigned long)win_data->dma_addr,
 			(unsigned long)win_data->vaddr);
 	DRM_DEBUG_KMS("fb_width = %d, crtc_width = %d\n",
 			overlay->fb_width, overlay->crtc_width);
@@ -346,15 +422,21 @@
 	writel(keycon1, ctx->regs + WKEYCON1_BASE(win));
 }
 
-static void fimd_win_commit(struct device *dev)
+static void fimd_win_commit(struct device *dev, int zpos)
 {
 	struct fimd_context *ctx = get_fimd_context(dev);
 	struct fimd_win_data *win_data;
-	int win = ctx->default_win;
+	int win = zpos;
 	unsigned long val, alpha, size;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
+	if (ctx->suspended)
+		return;
+
+	if (win == DEFAULT_ZPOS)
+		win = ctx->default_win;
+
 	if (win < 0 || win > WINDOWS_NR)
 		return;
 
@@ -376,16 +458,16 @@
 	writel(val, ctx->regs + SHADOWCON);
 
 	/* buffer start address */
-	val = win_data->paddr;
+	val = (unsigned long)win_data->dma_addr;
 	writel(val, ctx->regs + VIDWx_BUF_START(win, 0));
 
 	/* buffer end address */
 	size = win_data->fb_width * win_data->ovl_height * (win_data->bpp >> 3);
-	val = win_data->paddr + size;
+	val = (unsigned long)(win_data->dma_addr + size);
 	writel(val, ctx->regs + VIDWx_BUF_END(win, 0));
 
 	DRM_DEBUG_KMS("start addr = 0x%lx, end addr = 0x%lx, size = 0x%lx\n",
-			(unsigned long)win_data->paddr, val, size);
+			(unsigned long)win_data->dma_addr, val, size);
 	DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n",
 			win_data->ovl_width, win_data->ovl_height);
 
@@ -437,22 +519,32 @@
 	if (win != 0)
 		fimd_win_set_colkey(dev, win);
 
+	/* wincon */
+	val = readl(ctx->regs + WINCON(win));
+	val |= WINCONx_ENWIN;
+	writel(val, ctx->regs + WINCON(win));
+
 	/* Enable DMA channel and unprotect windows */
 	val = readl(ctx->regs + SHADOWCON);
 	val |= SHADOWCON_CHx_ENABLE(win);
 	val &= ~SHADOWCON_WINx_PROTECT(win);
 	writel(val, ctx->regs + SHADOWCON);
+
+	win_data->enabled = true;
 }
 
-static void fimd_win_disable(struct device *dev)
+static void fimd_win_disable(struct device *dev, int zpos)
 {
 	struct fimd_context *ctx = get_fimd_context(dev);
 	struct fimd_win_data *win_data;
-	int win = ctx->default_win;
+	int win = zpos;
 	u32 val;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
+	if (win == DEFAULT_ZPOS)
+		win = ctx->default_win;
+
 	if (win < 0 || win > WINDOWS_NR)
 		return;
 
@@ -473,6 +565,8 @@
 	val &= ~SHADOWCON_CHx_ENABLE(win);
 	val &= ~SHADOWCON_WINx_PROTECT(win);
 	writel(val, ctx->regs + SHADOWCON);
+
+	win_data->enabled = false;
 }
 
 static struct exynos_drm_overlay_ops fimd_overlay_ops = {
@@ -508,9 +602,17 @@
 		wake_up_interruptible(&e->base.file_priv->event_wait);
 	}
 
-	if (is_checked)
+	if (is_checked) {
 		drm_vblank_put(drm_dev, crtc);
 
+		/*
+		 * don't off vblank if vblank_disable_allowed is 1,
+		 * because vblank would be off by timer handler.
+		 */
+		if (!drm_dev->vblank_disable_allowed)
+			drm_vblank_off(drm_dev, crtc);
+	}
+
 	spin_unlock_irqrestore(&drm_dev->event_lock, flags);
 }
 
@@ -528,9 +630,14 @@
 		/* VSYNC interrupt */
 		writel(VIDINTCON1_INT_FRAME, ctx->regs + VIDINTCON1);
 
+	/* check the crtc is detached already from encoder */
+	if (manager->pipe < 0)
+		goto out;
+
 	drm_handle_vblank(drm_dev, manager->pipe);
 	fimd_finish_pageflip(drm_dev, manager->pipe);
 
+out:
 	return IRQ_HANDLED;
 }
 
@@ -551,7 +658,7 @@
 	/*
 	 * with vblank_disable_allowed = 1, vblank interrupt will be disabled
 	 * by drm timer once a current process gives up ownership of
-	 * vblank event.(drm_vblank_put function was called)
+	 * vblank event.(after drm_vblank_put function is called)
 	 */
 	drm_dev->vblank_disable_allowed = 1;
 
@@ -704,9 +811,6 @@
 
 	ctx->irq = res->start;
 
-	for (win = 0; win < WINDOWS_NR; win++)
-		fimd_clear_win(ctx, win);
-
 	ret = request_irq(ctx->irq, fimd_irq_handler, 0, "drm_fimd", ctx);
 	if (ret < 0) {
 		dev_err(dev, "irq request failed.\n");
@@ -731,10 +835,20 @@
 	subdrv->manager.pipe = -1;
 	subdrv->manager.ops = &fimd_manager_ops;
 	subdrv->manager.overlay_ops = &fimd_overlay_ops;
-	subdrv->manager.display = &fimd_display;
+	subdrv->manager.display_ops = &fimd_display_ops;
 	subdrv->manager.dev = dev;
 
+	mutex_init(&ctx->lock);
+
 	platform_set_drvdata(pdev, ctx);
+
+	pm_runtime_set_active(dev);
+	pm_runtime_enable(dev);
+	pm_runtime_get_sync(dev);
+
+	for (win = 0; win < WINDOWS_NR; win++)
+		fimd_clear_win(ctx, win);
+
 	exynos_drm_subdrv_register(subdrv);
 
 	return 0;
@@ -762,14 +876,25 @@
 
 static int __devexit fimd_remove(struct platform_device *pdev)
 {
+	struct device *dev = &pdev->dev;
 	struct fimd_context *ctx = platform_get_drvdata(pdev);
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
 	exynos_drm_subdrv_unregister(&ctx->subdrv);
 
+	if (ctx->suspended)
+		goto out;
+
 	clk_disable(ctx->lcd_clk);
 	clk_disable(ctx->bus_clk);
+
+	pm_runtime_set_suspended(dev);
+	pm_runtime_put_sync(dev);
+
+out:
+	pm_runtime_disable(dev);
+
 	clk_put(ctx->lcd_clk);
 	clk_put(ctx->bus_clk);
 
@@ -783,12 +908,102 @@
 	return 0;
 }
 
+#ifdef CONFIG_PM_SLEEP
+static int fimd_suspend(struct device *dev)
+{
+	int ret;
+
+	if (pm_runtime_suspended(dev))
+		return 0;
+
+	ret = pm_runtime_suspend(dev);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static int fimd_resume(struct device *dev)
+{
+	int ret;
+
+	ret = pm_runtime_resume(dev);
+	if (ret < 0) {
+		DRM_ERROR("failed to resume runtime pm.\n");
+		return ret;
+	}
+
+	pm_runtime_disable(dev);
+
+	ret = pm_runtime_set_active(dev);
+	if (ret < 0) {
+		DRM_ERROR("failed to active runtime pm.\n");
+		pm_runtime_enable(dev);
+		pm_runtime_suspend(dev);
+		return ret;
+	}
+
+	pm_runtime_enable(dev);
+
+	return 0;
+}
+#endif
+
+#ifdef CONFIG_PM_RUNTIME
+static int fimd_runtime_suspend(struct device *dev)
+{
+	struct fimd_context *ctx = get_fimd_context(dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	clk_disable(ctx->lcd_clk);
+	clk_disable(ctx->bus_clk);
+
+	ctx->suspended = true;
+	return 0;
+}
+
+static int fimd_runtime_resume(struct device *dev)
+{
+	struct fimd_context *ctx = get_fimd_context(dev);
+	int ret;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	ret = clk_enable(ctx->bus_clk);
+	if (ret < 0)
+		return ret;
+
+	ret = clk_enable(ctx->lcd_clk);
+	if  (ret < 0) {
+		clk_disable(ctx->bus_clk);
+		return ret;
+	}
+
+	ctx->suspended = false;
+
+	/* if vblank was enabled status, enable it again. */
+	if (test_and_clear_bit(0, &ctx->irq_flags))
+		fimd_enable_vblank(dev);
+
+	fimd_apply(dev);
+
+	return 0;
+}
+#endif
+
+static const struct dev_pm_ops fimd_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(fimd_suspend, fimd_resume)
+	SET_RUNTIME_PM_OPS(fimd_runtime_suspend, fimd_runtime_resume, NULL)
+};
+
 static struct platform_driver fimd_driver = {
 	.probe		= fimd_probe,
 	.remove		= __devexit_p(fimd_remove),
 	.driver		= {
 		.name	= "exynos4-fb",
 		.owner	= THIS_MODULE,
+		.pm	= &fimd_pm_ops,
 	},
 };
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index a8e7a88..025abb3 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -55,104 +55,128 @@
 	return out_msg;
 }
 
-static unsigned int get_gem_mmap_offset(struct drm_gem_object *obj)
+static int exynos_drm_gem_handle_create(struct drm_gem_object *obj,
+					struct drm_file *file_priv,
+					unsigned int *handle)
 {
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
-	return (unsigned int)obj->map_list.hash.key << PAGE_SHIFT;
-}
-
-struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_file *file_priv,
-		struct drm_device *dev, unsigned int size,
-		unsigned int *handle)
-{
-	struct exynos_drm_gem_obj *exynos_gem_obj;
-	struct exynos_drm_buf_entry *entry;
-	struct drm_gem_object *obj;
 	int ret;
 
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
-	size = roundup(size, PAGE_SIZE);
-
-	exynos_gem_obj = kzalloc(sizeof(*exynos_gem_obj), GFP_KERNEL);
-	if (!exynos_gem_obj) {
-		DRM_ERROR("failed to allocate exynos gem object.\n");
-		return ERR_PTR(-ENOMEM);
-	}
-
-	/* allocate the new buffer object and memory region. */
-	entry = exynos_drm_buf_create(dev, size);
-	if (!entry) {
-		kfree(exynos_gem_obj);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	exynos_gem_obj->entry = entry;
-
-	obj = &exynos_gem_obj->base;
-
-	ret = drm_gem_object_init(dev, obj, size);
-	if (ret < 0) {
-		DRM_ERROR("failed to initailize gem object.\n");
-		goto err_obj_init;
-	}
-
-	DRM_DEBUG_KMS("created file object = 0x%x\n", (unsigned int)obj->filp);
-
-	ret = drm_gem_create_mmap_offset(obj);
-	if (ret < 0) {
-		DRM_ERROR("failed to allocate mmap offset.\n");
-		goto err_create_mmap_offset;
-	}
-
 	/*
 	 * allocate a id of idr table where the obj is registered
 	 * and handle has the id what user can see.
 	 */
 	ret = drm_gem_handle_create(file_priv, obj, handle);
 	if (ret)
-		goto err_handle_create;
+		return ret;
 
 	DRM_DEBUG_KMS("gem handle = 0x%x\n", *handle);
 
 	/* drop reference from allocate - handle holds it now. */
 	drm_gem_object_unreference_unlocked(obj);
 
-	return exynos_gem_obj;
+	return 0;
+}
 
-err_handle_create:
-	drm_gem_free_mmap_offset(obj);
+void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj)
+{
+	struct drm_gem_object *obj;
 
-err_create_mmap_offset:
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (!exynos_gem_obj)
+		return;
+
+	obj = &exynos_gem_obj->base;
+
+	DRM_DEBUG_KMS("handle count = %d\n", atomic_read(&obj->handle_count));
+
+	exynos_drm_buf_destroy(obj->dev, exynos_gem_obj->buffer);
+
+	if (obj->map_list.map)
+		drm_gem_free_mmap_offset(obj);
+
+	/* release file pointer to gem object. */
 	drm_gem_object_release(obj);
 
-err_obj_init:
-	exynos_drm_buf_destroy(dev, exynos_gem_obj->entry);
-
 	kfree(exynos_gem_obj);
+}
 
-	return ERR_PTR(ret);
+static struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev,
+						      unsigned long size)
+{
+	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct drm_gem_object *obj;
+	int ret;
+
+	exynos_gem_obj = kzalloc(sizeof(*exynos_gem_obj), GFP_KERNEL);
+	if (!exynos_gem_obj) {
+		DRM_ERROR("failed to allocate exynos gem object\n");
+		return NULL;
+	}
+
+	obj = &exynos_gem_obj->base;
+
+	ret = drm_gem_object_init(dev, obj, size);
+	if (ret < 0) {
+		DRM_ERROR("failed to initialize gem object\n");
+		kfree(exynos_gem_obj);
+		return NULL;
+	}
+
+	DRM_DEBUG_KMS("created file object = 0x%x\n", (unsigned int)obj->filp);
+
+	return exynos_gem_obj;
+}
+
+struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev,
+						 unsigned long size)
+{
+	struct exynos_drm_gem_buf *buffer;
+	struct exynos_drm_gem_obj *exynos_gem_obj;
+
+	size = roundup(size, PAGE_SIZE);
+	DRM_DEBUG_KMS("%s: size = 0x%lx\n", __FILE__, size);
+
+	buffer = exynos_drm_buf_create(dev, size);
+	if (!buffer)
+		return ERR_PTR(-ENOMEM);
+
+	exynos_gem_obj = exynos_drm_gem_init(dev, size);
+	if (!exynos_gem_obj) {
+		exynos_drm_buf_destroy(dev, buffer);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	exynos_gem_obj->buffer = buffer;
+
+	return exynos_gem_obj;
 }
 
 int exynos_drm_gem_create_ioctl(struct drm_device *dev, void *data,
-		struct drm_file *file_priv)
+				struct drm_file *file_priv)
 {
 	struct drm_exynos_gem_create *args = data;
 	struct exynos_drm_gem_obj *exynos_gem_obj;
+	int ret;
 
-	DRM_DEBUG_KMS("%s : size = 0x%x\n", __FILE__, args->size);
+	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	exynos_gem_obj = exynos_drm_gem_create(file_priv, dev, args->size,
-			&args->handle);
+	exynos_gem_obj = exynos_drm_gem_create(dev, args->size);
 	if (IS_ERR(exynos_gem_obj))
 		return PTR_ERR(exynos_gem_obj);
 
+	ret = exynos_drm_gem_handle_create(&exynos_gem_obj->base, file_priv,
+			&args->handle);
+	if (ret) {
+		exynos_drm_gem_destroy(exynos_gem_obj);
+		return ret;
+	}
+
 	return 0;
 }
 
 int exynos_drm_gem_map_offset_ioctl(struct drm_device *dev, void *data,
-		struct drm_file *file_priv)
+				    struct drm_file *file_priv)
 {
 	struct drm_exynos_gem_map_off *args = data;
 
@@ -171,36 +195,37 @@
 }
 
 static int exynos_drm_gem_mmap_buffer(struct file *filp,
-		struct vm_area_struct *vma)
+				      struct vm_area_struct *vma)
 {
 	struct drm_gem_object *obj = filp->private_data;
 	struct exynos_drm_gem_obj *exynos_gem_obj = to_exynos_gem_obj(obj);
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 	unsigned long pfn, vm_size;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
 	vma->vm_flags |= (VM_IO | VM_RESERVED);
 
+	/* in case of direct mapping, always having non-cachable attribute */
 	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
 	vma->vm_file = filp;
 
 	vm_size = vma->vm_end - vma->vm_start;
 	/*
-	 * a entry contains information to physically continuous memory
+	 * a buffer contains information to physically continuous memory
 	 * allocated by user request or at framebuffer creation.
 	 */
-	entry = exynos_gem_obj->entry;
+	buffer = exynos_gem_obj->buffer;
 
 	/* check if user-requested size is valid. */
-	if (vm_size > entry->size)
+	if (vm_size > buffer->size)
 		return -EINVAL;
 
 	/*
 	 * get page frame number to physical memory to be mapped
 	 * to user space.
 	 */
-	pfn = exynos_gem_obj->entry->paddr >> PAGE_SHIFT;
+	pfn = ((unsigned long)exynos_gem_obj->buffer->dma_addr) >> PAGE_SHIFT;
 
 	DRM_DEBUG_KMS("pfn = 0x%lx\n", pfn);
 
@@ -218,7 +243,7 @@
 };
 
 int exynos_drm_gem_mmap_ioctl(struct drm_device *dev, void *data,
-		struct drm_file *file_priv)
+			      struct drm_file *file_priv)
 {
 	struct drm_exynos_gem_mmap *args = data;
 	struct drm_gem_object *obj;
@@ -264,32 +289,19 @@
 	return 0;
 }
 
-void exynos_drm_gem_free_object(struct drm_gem_object *gem_obj)
+void exynos_drm_gem_free_object(struct drm_gem_object *obj)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
-
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	DRM_DEBUG_KMS("handle count = %d\n",
-			atomic_read(&gem_obj->handle_count));
-
-	if (gem_obj->map_list.map)
-		drm_gem_free_mmap_offset(gem_obj);
-
-	/* release file pointer to gem object. */
-	drm_gem_object_release(gem_obj);
-
-	exynos_gem_obj = to_exynos_gem_obj(gem_obj);
-
-	exynos_drm_buf_destroy(gem_obj->dev, exynos_gem_obj->entry);
-
-	kfree(exynos_gem_obj);
+	exynos_drm_gem_destroy(to_exynos_gem_obj(obj));
 }
 
 int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
-		struct drm_device *dev, struct drm_mode_create_dumb *args)
+			       struct drm_device *dev,
+			       struct drm_mode_create_dumb *args)
 {
 	struct exynos_drm_gem_obj *exynos_gem_obj;
+	int ret;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
@@ -302,19 +314,27 @@
 	args->pitch = args->width * args->bpp >> 3;
 	args->size = args->pitch * args->height;
 
-	exynos_gem_obj = exynos_drm_gem_create(file_priv, dev, args->size,
-							&args->handle);
+	exynos_gem_obj = exynos_drm_gem_create(dev, args->size);
 	if (IS_ERR(exynos_gem_obj))
 		return PTR_ERR(exynos_gem_obj);
 
+	ret = exynos_drm_gem_handle_create(&exynos_gem_obj->base, file_priv,
+			&args->handle);
+	if (ret) {
+		exynos_drm_gem_destroy(exynos_gem_obj);
+		return ret;
+	}
+
 	return 0;
 }
 
 int exynos_drm_gem_dumb_map_offset(struct drm_file *file_priv,
-		struct drm_device *dev, uint32_t handle, uint64_t *offset)
+				   struct drm_device *dev, uint32_t handle,
+				   uint64_t *offset)
 {
 	struct exynos_drm_gem_obj *exynos_gem_obj;
 	struct drm_gem_object *obj;
+	int ret = 0;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
@@ -329,19 +349,46 @@
 	obj = drm_gem_object_lookup(dev, file_priv, handle);
 	if (!obj) {
 		DRM_ERROR("failed to lookup gem object.\n");
-		mutex_unlock(&dev->struct_mutex);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto unlock;
 	}
 
 	exynos_gem_obj = to_exynos_gem_obj(obj);
 
-	*offset = get_gem_mmap_offset(&exynos_gem_obj->base);
+	if (!exynos_gem_obj->base.map_list.map) {
+		ret = drm_gem_create_mmap_offset(&exynos_gem_obj->base);
+		if (ret)
+			goto out;
+	}
 
-	drm_gem_object_unreference(obj);
-
+	*offset = (u64)exynos_gem_obj->base.map_list.hash.key << PAGE_SHIFT;
 	DRM_DEBUG_KMS("offset = 0x%lx\n", (unsigned long)*offset);
 
+out:
+	drm_gem_object_unreference(obj);
+unlock:
 	mutex_unlock(&dev->struct_mutex);
+	return ret;
+}
+
+int exynos_drm_gem_dumb_destroy(struct drm_file *file_priv,
+				struct drm_device *dev,
+				unsigned int handle)
+{
+	int ret;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	/*
+	 * obj->refcount and obj->handle_count are decreased and
+	 * if both them are 0 then exynos_drm_gem_free_object()
+	 * would be called by callback to release resources.
+	 */
+	ret = drm_gem_handle_delete(file_priv, handle);
+	if (ret < 0) {
+		DRM_ERROR("failed to delete drm_gem_handle.\n");
+		return ret;
+	}
 
 	return 0;
 }
@@ -360,7 +407,8 @@
 
 	mutex_lock(&dev->struct_mutex);
 
-	pfn = (exynos_gem_obj->entry->paddr >> PAGE_SHIFT) + page_offset;
+	pfn = (((unsigned long)exynos_gem_obj->buffer->dma_addr) >>
+			PAGE_SHIFT) + page_offset;
 
 	ret = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address, pfn);
 
@@ -388,28 +436,6 @@
 	return ret;
 }
 
-
-int exynos_drm_gem_dumb_destroy(struct drm_file *file_priv,
-		struct drm_device *dev, unsigned int handle)
-{
-	int ret;
-
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
-	/*
-	 * obj->refcount and obj->handle_count are decreased and
-	 * if both them are 0 then exynos_drm_gem_free_object()
-	 * would be called by callback to release resources.
-	 */
-	ret = drm_gem_handle_delete(file_priv, handle);
-	if (ret < 0) {
-		DRM_ERROR("failed to delete drm_gem_handle.\n");
-		return ret;
-	}
-
-	return 0;
-}
-
 MODULE_AUTHOR("Inki Dae <inki.dae@samsung.com>");
 MODULE_DESCRIPTION("Samsung SoC DRM GEM Module");
 MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h
index e5fc014..67cdc91 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h
@@ -30,13 +30,29 @@
 			struct exynos_drm_gem_obj, base)
 
 /*
+ * exynos drm gem buffer structure.
+ *
+ * @kvaddr: kernel virtual address to allocated memory region.
+ * @dma_addr: bus address(accessed by dma) to allocated memory region.
+ *	- this address could be physical address without IOMMU and
+ *	device address with IOMMU.
+ * @size: size of allocated memory region.
+ */
+struct exynos_drm_gem_buf {
+	void __iomem		*kvaddr;
+	dma_addr_t		dma_addr;
+	unsigned long		size;
+};
+
+/*
  * exynos drm buffer structure.
  *
  * @base: a gem object.
  *	- a new handle to this gem object would be created
  *	by drm_gem_handle_create().
- * @entry: pointer to exynos drm buffer entry object.
- *	- containing the information to physically
+ * @buffer: a pointer to exynos_drm_gem_buffer object.
+ *	- contain the information to memory region allocated
+ *	by user request or at framebuffer creation.
  *	continuous memory region allocated by user request
  *	or at framebuffer creation.
  *
@@ -44,14 +60,16 @@
  *	user can access the buffer through kms_bo.handle.
  */
 struct exynos_drm_gem_obj {
-	struct drm_gem_object base;
-	struct exynos_drm_buf_entry *entry;
+	struct drm_gem_object		base;
+	struct exynos_drm_gem_buf	*buffer;
 };
 
-/* create a new buffer and get a new gem handle. */
-struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_file *file_priv,
-		struct drm_device *dev, unsigned int size,
-		unsigned int *handle);
+/* destroy a buffer with gem object */
+void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj);
+
+/* create a new buffer with gem object */
+struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev,
+						 unsigned long size);
 
 /*
  * request gem object creation and buffer allocation as the size
@@ -59,15 +77,18 @@
  * height and bpp.
  */
 int exynos_drm_gem_create_ioctl(struct drm_device *dev, void *data,
-		struct drm_file *file_priv);
+				struct drm_file *file_priv);
 
 /* get buffer offset to map to user space. */
 int exynos_drm_gem_map_offset_ioctl(struct drm_device *dev, void *data,
-		struct drm_file *file_priv);
+				    struct drm_file *file_priv);
 
-/* unmap a buffer from user space. */
-int exynos_drm_gem_munmap_ioctl(struct drm_device *dev, void *data,
-		struct drm_file *file_priv);
+/*
+ * mmap the physically continuous memory that a gem object contains
+ * to user space.
+ */
+int exynos_drm_gem_mmap_ioctl(struct drm_device *dev, void *data,
+			      struct drm_file *file_priv);
 
 /* initialize gem object. */
 int exynos_drm_gem_init_object(struct drm_gem_object *obj);
@@ -77,24 +98,13 @@
 
 /* create memory region for drm framebuffer. */
 int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
-		struct drm_device *dev, struct drm_mode_create_dumb *args);
+			       struct drm_device *dev,
+			       struct drm_mode_create_dumb *args);
 
 /* map memory region for drm framebuffer to user space. */
 int exynos_drm_gem_dumb_map_offset(struct drm_file *file_priv,
-		struct drm_device *dev, uint32_t handle, uint64_t *offset);
-
-/* page fault handler and mmap fault address(virtual) to physical memory. */
-int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
-
-/*
- * mmap the physically continuous memory that a gem object contains
- * to user space.
- */
-int exynos_drm_gem_mmap_ioctl(struct drm_device *dev, void *data,
-		struct drm_file *file_priv);
-
-/* set vm_flags and we can change the vm attribute to other one at here. */
-int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma);
+				   struct drm_device *dev, uint32_t handle,
+				   uint64_t *offset);
 
 /*
  * destroy memory region allocated.
@@ -102,6 +112,13 @@
  *	would be released by drm_gem_handle_delete().
  */
 int exynos_drm_gem_dumb_destroy(struct drm_file *file_priv,
-		struct drm_device *dev, unsigned int handle);
+				struct drm_device *dev,
+				unsigned int handle);
+
+/* page fault handler and mmap fault address(virtual) to physical memory. */
+int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
+
+/* set vm_flags and we can change the vm attribute to other one at here. */
+int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma);
 
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c
new file mode 100644
index 0000000..ed8a319e
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c
@@ -0,0 +1,439 @@
+/*
+ * Copyright (C) 2011 Samsung Electronics Co.Ltd
+ * Authors:
+ *	Inki Dae <inki.dae@samsung.com>
+ *	Seung-Woo Kim <sw0312.kim@samsung.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include "drmP.h"
+
+#include <linux/kernel.h>
+#include <linux/wait.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+
+#include <drm/exynos_drm.h>
+
+#include "exynos_drm_drv.h"
+#include "exynos_drm_hdmi.h"
+
+#define to_context(dev)		platform_get_drvdata(to_platform_device(dev))
+#define to_subdrv(dev)		to_context(dev)
+#define get_ctx_from_subdrv(subdrv)	container_of(subdrv,\
+					struct drm_hdmi_context, subdrv);
+
+/* these callback points shoud be set by specific drivers. */
+static struct exynos_hdmi_display_ops *hdmi_display_ops;
+static struct exynos_hdmi_manager_ops *hdmi_manager_ops;
+static struct exynos_hdmi_overlay_ops *hdmi_overlay_ops;
+
+struct drm_hdmi_context {
+	struct exynos_drm_subdrv	subdrv;
+	struct exynos_drm_hdmi_context	*hdmi_ctx;
+	struct exynos_drm_hdmi_context	*mixer_ctx;
+	struct work_struct		work;
+};
+
+void exynos_drm_display_ops_register(struct exynos_hdmi_display_ops
+					*display_ops)
+{
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (display_ops)
+		hdmi_display_ops = display_ops;
+}
+EXPORT_SYMBOL(exynos_drm_display_ops_register);
+
+void exynos_drm_manager_ops_register(struct exynos_hdmi_manager_ops
+					*manager_ops)
+{
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (manager_ops)
+		hdmi_manager_ops = manager_ops;
+}
+EXPORT_SYMBOL(exynos_drm_manager_ops_register);
+
+void exynos_drm_overlay_ops_register(struct exynos_hdmi_overlay_ops
+					*overlay_ops)
+{
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (overlay_ops)
+		hdmi_overlay_ops = overlay_ops;
+}
+EXPORT_SYMBOL(exynos_drm_overlay_ops_register);
+
+static bool drm_hdmi_is_connected(struct device *dev)
+{
+	struct drm_hdmi_context *ctx = to_context(dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_display_ops && hdmi_display_ops->is_connected)
+		return hdmi_display_ops->is_connected(ctx->hdmi_ctx->ctx);
+
+	return false;
+}
+
+static int drm_hdmi_get_edid(struct device *dev,
+		struct drm_connector *connector, u8 *edid, int len)
+{
+	struct drm_hdmi_context *ctx = to_context(dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_display_ops && hdmi_display_ops->get_edid)
+		return hdmi_display_ops->get_edid(ctx->hdmi_ctx->ctx,
+				connector, edid, len);
+
+	return 0;
+}
+
+static int drm_hdmi_check_timing(struct device *dev, void *timing)
+{
+	struct drm_hdmi_context *ctx = to_context(dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_display_ops && hdmi_display_ops->check_timing)
+		return hdmi_display_ops->check_timing(ctx->hdmi_ctx->ctx,
+				timing);
+
+	return 0;
+}
+
+static int drm_hdmi_power_on(struct device *dev, int mode)
+{
+	struct drm_hdmi_context *ctx = to_context(dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_display_ops && hdmi_display_ops->power_on)
+		return hdmi_display_ops->power_on(ctx->hdmi_ctx->ctx, mode);
+
+	return 0;
+}
+
+static struct exynos_drm_display_ops drm_hdmi_display_ops = {
+	.type = EXYNOS_DISPLAY_TYPE_HDMI,
+	.is_connected = drm_hdmi_is_connected,
+	.get_edid = drm_hdmi_get_edid,
+	.check_timing = drm_hdmi_check_timing,
+	.power_on = drm_hdmi_power_on,
+};
+
+static int drm_hdmi_enable_vblank(struct device *subdrv_dev)
+{
+	struct drm_hdmi_context *ctx = to_context(subdrv_dev);
+	struct exynos_drm_subdrv *subdrv = &ctx->subdrv;
+	struct exynos_drm_manager *manager = &subdrv->manager;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_overlay_ops && hdmi_overlay_ops->enable_vblank)
+		return hdmi_overlay_ops->enable_vblank(ctx->mixer_ctx->ctx,
+							manager->pipe);
+
+	return 0;
+}
+
+static void drm_hdmi_disable_vblank(struct device *subdrv_dev)
+{
+	struct drm_hdmi_context *ctx = to_context(subdrv_dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_overlay_ops && hdmi_overlay_ops->disable_vblank)
+		return hdmi_overlay_ops->disable_vblank(ctx->mixer_ctx->ctx);
+}
+
+static void drm_hdmi_mode_set(struct device *subdrv_dev, void *mode)
+{
+	struct drm_hdmi_context *ctx = to_context(subdrv_dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_manager_ops && hdmi_manager_ops->mode_set)
+		hdmi_manager_ops->mode_set(ctx->hdmi_ctx->ctx, mode);
+}
+
+static void drm_hdmi_commit(struct device *subdrv_dev)
+{
+	struct drm_hdmi_context *ctx = to_context(subdrv_dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_manager_ops && hdmi_manager_ops->commit)
+		hdmi_manager_ops->commit(ctx->hdmi_ctx->ctx);
+}
+
+static void drm_hdmi_dpms(struct device *subdrv_dev, int mode)
+{
+	struct drm_hdmi_context *ctx = to_context(subdrv_dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		break;
+	case DRM_MODE_DPMS_STANDBY:
+	case DRM_MODE_DPMS_SUSPEND:
+	case DRM_MODE_DPMS_OFF:
+		if (hdmi_manager_ops && hdmi_manager_ops->disable)
+			hdmi_manager_ops->disable(ctx->hdmi_ctx->ctx);
+		break;
+	default:
+		DRM_DEBUG_KMS("unkown dps mode: %d\n", mode);
+		break;
+	}
+}
+
+static struct exynos_drm_manager_ops drm_hdmi_manager_ops = {
+	.dpms = drm_hdmi_dpms,
+	.enable_vblank = drm_hdmi_enable_vblank,
+	.disable_vblank = drm_hdmi_disable_vblank,
+	.mode_set = drm_hdmi_mode_set,
+	.commit = drm_hdmi_commit,
+};
+
+static void drm_mixer_mode_set(struct device *subdrv_dev,
+		struct exynos_drm_overlay *overlay)
+{
+	struct drm_hdmi_context *ctx = to_context(subdrv_dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_overlay_ops && hdmi_overlay_ops->win_mode_set)
+		hdmi_overlay_ops->win_mode_set(ctx->mixer_ctx->ctx, overlay);
+}
+
+static void drm_mixer_commit(struct device *subdrv_dev, int zpos)
+{
+	struct drm_hdmi_context *ctx = to_context(subdrv_dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_overlay_ops && hdmi_overlay_ops->win_commit)
+		hdmi_overlay_ops->win_commit(ctx->mixer_ctx->ctx, zpos);
+}
+
+static void drm_mixer_disable(struct device *subdrv_dev, int zpos)
+{
+	struct drm_hdmi_context *ctx = to_context(subdrv_dev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	if (hdmi_overlay_ops && hdmi_overlay_ops->win_disable)
+		hdmi_overlay_ops->win_disable(ctx->mixer_ctx->ctx, zpos);
+}
+
+static struct exynos_drm_overlay_ops drm_hdmi_overlay_ops = {
+	.mode_set = drm_mixer_mode_set,
+	.commit = drm_mixer_commit,
+	.disable = drm_mixer_disable,
+};
+
+
+static int hdmi_subdrv_probe(struct drm_device *drm_dev,
+		struct device *dev)
+{
+	struct exynos_drm_subdrv *subdrv = to_subdrv(dev);
+	struct drm_hdmi_context *ctx;
+	struct platform_device *pdev = to_platform_device(dev);
+	struct exynos_drm_common_hdmi_pd *pd;
+	int ret;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	pd = pdev->dev.platform_data;
+
+	if (!pd) {
+		DRM_DEBUG_KMS("platform data is null.\n");
+		return -EFAULT;
+	}
+
+	if (!pd->hdmi_dev) {
+		DRM_DEBUG_KMS("hdmi device is null.\n");
+		return -EFAULT;
+	}
+
+	if (!pd->mixer_dev) {
+		DRM_DEBUG_KMS("mixer device is null.\n");
+		return -EFAULT;
+	}
+
+	ret = platform_driver_register(&hdmi_driver);
+	if (ret) {
+		DRM_DEBUG_KMS("failed to register hdmi driver.\n");
+		return ret;
+	}
+
+	ret = platform_driver_register(&mixer_driver);
+	if (ret) {
+		DRM_DEBUG_KMS("failed to register mixer driver.\n");
+		goto err_hdmidrv;
+	}
+
+	ctx = get_ctx_from_subdrv(subdrv);
+
+	ctx->hdmi_ctx = (struct exynos_drm_hdmi_context *)
+				to_context(pd->hdmi_dev);
+	if (!ctx->hdmi_ctx) {
+		DRM_DEBUG_KMS("hdmi context is null.\n");
+		ret = -EFAULT;
+		goto err_mixerdrv;
+	}
+
+	ctx->hdmi_ctx->drm_dev = drm_dev;
+
+	ctx->mixer_ctx = (struct exynos_drm_hdmi_context *)
+				to_context(pd->mixer_dev);
+	if (!ctx->mixer_ctx) {
+		DRM_DEBUG_KMS("mixer context is null.\n");
+		ret = -EFAULT;
+		goto err_mixerdrv;
+	}
+
+	ctx->mixer_ctx->drm_dev = drm_dev;
+
+	return 0;
+
+err_mixerdrv:
+	platform_driver_unregister(&mixer_driver);
+err_hdmidrv:
+	platform_driver_unregister(&hdmi_driver);
+	return ret;
+}
+
+static void hdmi_subdrv_remove(struct drm_device *drm_dev)
+{
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	platform_driver_unregister(&hdmi_driver);
+	platform_driver_unregister(&mixer_driver);
+}
+
+static void exynos_drm_hdmi_late_probe(struct work_struct *work)
+{
+	struct drm_hdmi_context *ctx = container_of(work,
+				struct drm_hdmi_context, work);
+
+	/*
+	 * this function calls subdrv->probe() so this must be called
+	 * after probe context.
+	 *
+	 * PS. subdrv->probe() will call platform_driver_register() to probe
+	 * hdmi and mixer driver.
+	 */
+	exynos_drm_subdrv_register(&ctx->subdrv);
+}
+
+static int __devinit exynos_drm_hdmi_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct exynos_drm_subdrv *subdrv;
+	struct drm_hdmi_context *ctx;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx) {
+		DRM_LOG_KMS("failed to alloc common hdmi context.\n");
+		return -ENOMEM;
+	}
+
+	subdrv = &ctx->subdrv;
+
+	subdrv->probe = hdmi_subdrv_probe;
+	subdrv->remove = hdmi_subdrv_remove;
+	subdrv->manager.pipe = -1;
+	subdrv->manager.ops = &drm_hdmi_manager_ops;
+	subdrv->manager.overlay_ops = &drm_hdmi_overlay_ops;
+	subdrv->manager.display_ops = &drm_hdmi_display_ops;
+	subdrv->manager.dev = dev;
+
+	platform_set_drvdata(pdev, subdrv);
+
+	INIT_WORK(&ctx->work, exynos_drm_hdmi_late_probe);
+
+	schedule_work(&ctx->work);
+
+	return 0;
+}
+
+static int hdmi_runtime_suspend(struct device *dev)
+{
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	return 0;
+}
+
+static int hdmi_runtime_resume(struct device *dev)
+{
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	return 0;
+}
+
+static const struct dev_pm_ops hdmi_pm_ops = {
+	.runtime_suspend = hdmi_runtime_suspend,
+	.runtime_resume	 = hdmi_runtime_resume,
+};
+
+static int __devexit exynos_drm_hdmi_remove(struct platform_device *pdev)
+{
+	struct drm_hdmi_context *ctx = platform_get_drvdata(pdev);
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	exynos_drm_subdrv_unregister(&ctx->subdrv);
+	kfree(ctx);
+
+	return 0;
+}
+
+static struct platform_driver exynos_drm_common_hdmi_driver = {
+	.probe		= exynos_drm_hdmi_probe,
+	.remove		= __devexit_p(exynos_drm_hdmi_remove),
+	.driver		= {
+		.name	= "exynos-drm-hdmi",
+		.owner	= THIS_MODULE,
+		.pm = &hdmi_pm_ops,
+	},
+};
+
+static int __init exynos_drm_hdmi_init(void)
+{
+	int ret;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	ret = platform_driver_register(&exynos_drm_common_hdmi_driver);
+	if (ret) {
+		DRM_DEBUG_KMS("failed to register hdmi common driver.\n");
+		return ret;
+	}
+
+	return ret;
+}
+
+static void __exit exynos_drm_hdmi_exit(void)
+{
+	platform_driver_unregister(&exynos_drm_common_hdmi_driver);
+}
+
+module_init(exynos_drm_hdmi_init);
+module_exit(exynos_drm_hdmi_exit);
+
+MODULE_AUTHOR("Inki Dae <inki.dae@samsung.com>");
+MODULE_AUTHOR("Seung-Woo Kim, <sw0312.kim@samsung.com>");
+MODULE_DESCRIPTION("Samsung SoC DRM HDMI Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.h b/drivers/gpu/drm/exynos/exynos_drm_hdmi.h
new file mode 100644
index 0000000..3c29f79
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.h
@@ -0,0 +1,73 @@
+/* exynos_drm_hdmi.h
+ *
+ * Copyright (c) 2011 Samsung Electronics Co., Ltd.
+ * Authoer: Inki Dae <inki.dae@samsung.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * VA LINUX SYSTEMS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _EXYNOS_DRM_HDMI_H_
+#define _EXYNOS_DRM_HDMI_H_
+
+/*
+ * exynos hdmi common context structure.
+ *
+ * @drm_dev: pointer to drm_device.
+ * @ctx: pointer to the context of specific device driver.
+ *	this context should be hdmi_context or mixer_context.
+ */
+struct exynos_drm_hdmi_context {
+	struct drm_device	*drm_dev;
+	void			*ctx;
+};
+
+struct exynos_hdmi_display_ops {
+	bool (*is_connected)(void *ctx);
+	int (*get_edid)(void *ctx, struct drm_connector *connector,
+			u8 *edid, int len);
+	int (*check_timing)(void *ctx, void *timing);
+	int (*power_on)(void *ctx, int mode);
+};
+
+struct exynos_hdmi_manager_ops {
+	void (*mode_set)(void *ctx, void *mode);
+	void (*commit)(void *ctx);
+	void (*disable)(void *ctx);
+};
+
+struct exynos_hdmi_overlay_ops {
+	int (*enable_vblank)(void *ctx, int pipe);
+	void (*disable_vblank)(void *ctx);
+	void (*win_mode_set)(void *ctx, struct exynos_drm_overlay *overlay);
+	void (*win_commit)(void *ctx, int zpos);
+	void (*win_disable)(void *ctx, int zpos);
+};
+
+extern struct platform_driver hdmi_driver;
+extern struct platform_driver mixer_driver;
+
+void exynos_drm_display_ops_register(struct exynos_hdmi_display_ops
+					*display_ops);
+void exynos_drm_manager_ops_register(struct exynos_hdmi_manager_ops
+					*manager_ops);
+void exynos_drm_overlay_ops_register(struct exynos_hdmi_overlay_ops
+					*overlay_ops);
+
+#endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c
new file mode 100644
index 0000000..c785e34
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c
@@ -0,0 +1,163 @@
+/*
+ * Copyright (C) 2011 Samsung Electronics Co.Ltd
+ * Authors: Joonyoung Shim <jy0922.shim@samsung.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include "drmP.h"
+
+#include "exynos_drm.h"
+#include "exynos_drm_crtc.h"
+#include "exynos_drm_drv.h"
+#include "exynos_drm_encoder.h"
+
+struct exynos_plane {
+	struct drm_plane		base;
+	struct exynos_drm_overlay	overlay;
+	bool				enabled;
+};
+
+static int
+exynos_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
+		     struct drm_framebuffer *fb, int crtc_x, int crtc_y,
+		     unsigned int crtc_w, unsigned int crtc_h,
+		     uint32_t src_x, uint32_t src_y,
+		     uint32_t src_w, uint32_t src_h)
+{
+	struct exynos_plane *exynos_plane =
+		container_of(plane, struct exynos_plane, base);
+	struct exynos_drm_overlay *overlay = &exynos_plane->overlay;
+	struct exynos_drm_crtc_pos pos;
+	unsigned int x = src_x >> 16;
+	unsigned int y = src_y >> 16;
+	int ret;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	memset(&pos, 0, sizeof(struct exynos_drm_crtc_pos));
+	pos.crtc_x = crtc_x;
+	pos.crtc_y = crtc_y;
+	pos.crtc_w = crtc_w;
+	pos.crtc_h = crtc_h;
+
+	pos.fb_x = x;
+	pos.fb_y = y;
+
+	/* TODO: scale feature */
+	ret = exynos_drm_overlay_update(overlay, fb, &crtc->mode, &pos);
+	if (ret < 0)
+		return ret;
+
+	exynos_drm_fn_encoder(crtc, overlay,
+			exynos_drm_encoder_crtc_mode_set);
+	exynos_drm_fn_encoder(crtc, &overlay->zpos,
+			exynos_drm_encoder_crtc_plane_commit);
+
+	exynos_plane->enabled = true;
+
+	return 0;
+}
+
+static int exynos_disable_plane(struct drm_plane *plane)
+{
+	struct exynos_plane *exynos_plane =
+		container_of(plane, struct exynos_plane, base);
+	struct exynos_drm_overlay *overlay = &exynos_plane->overlay;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	if (!exynos_plane->enabled)
+		return 0;
+
+	exynos_drm_fn_encoder(plane->crtc, &overlay->zpos,
+			exynos_drm_encoder_crtc_disable);
+
+	exynos_plane->enabled = false;
+	exynos_plane->overlay.zpos = DEFAULT_ZPOS;
+
+	return 0;
+}
+
+static void exynos_plane_destroy(struct drm_plane *plane)
+{
+	struct exynos_plane *exynos_plane =
+		container_of(plane, struct exynos_plane, base);
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	exynos_disable_plane(plane);
+	drm_plane_cleanup(plane);
+	kfree(exynos_plane);
+}
+
+static struct drm_plane_funcs exynos_plane_funcs = {
+	.update_plane	= exynos_update_plane,
+	.disable_plane	= exynos_disable_plane,
+	.destroy	= exynos_plane_destroy,
+};
+
+int exynos_plane_init(struct drm_device *dev, unsigned int nr)
+{
+	struct exynos_plane *exynos_plane;
+	uint32_t possible_crtcs;
+
+	exynos_plane = kzalloc(sizeof(struct exynos_plane), GFP_KERNEL);
+	if (!exynos_plane)
+		return -ENOMEM;
+
+	/* all CRTCs are available */
+	possible_crtcs = (1 << MAX_CRTC) - 1;
+
+	exynos_plane->overlay.zpos = DEFAULT_ZPOS;
+
+	/* TODO: format */
+	return drm_plane_init(dev, &exynos_plane->base, possible_crtcs,
+			      &exynos_plane_funcs, NULL, 0);
+}
+
+int exynos_plane_set_zpos_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file_priv)
+{
+	struct drm_exynos_plane_set_zpos *zpos_req = data;
+	struct drm_mode_object *obj;
+	struct drm_plane *plane;
+	struct exynos_plane *exynos_plane;
+	int ret = 0;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	if (!drm_core_check_feature(dev, DRIVER_MODESET))
+		return -EINVAL;
+
+	if (zpos_req->zpos < 0 || zpos_req->zpos >= MAX_PLANE) {
+		if (zpos_req->zpos != DEFAULT_ZPOS) {
+			DRM_ERROR("zpos not within limits\n");
+			return -EINVAL;
+		}
+	}
+
+	mutex_lock(&dev->mode_config.mutex);
+
+	obj = drm_mode_object_find(dev, zpos_req->plane_id,
+			DRM_MODE_OBJECT_PLANE);
+	if (!obj) {
+		DRM_DEBUG_KMS("Unknown plane ID %d\n",
+			      zpos_req->plane_id);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	plane = obj_to_plane(obj);
+	exynos_plane = container_of(plane, struct exynos_plane, base);
+
+	exynos_plane->overlay.zpos = zpos_req->zpos;
+
+out:
+	mutex_unlock(&dev->mode_config.mutex);
+	return ret;
+}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.h b/drivers/gpu/drm/exynos/exynos_drm_plane.h
new file mode 100644
index 0000000..16b71f8
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_drm_plane.h
@@ -0,0 +1,14 @@
+/*
+ * Copyright (C) 2011 Samsung Electronics Co.Ltd
+ * Authors: Joonyoung Shim <jy0922.shim@samsung.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+int exynos_plane_init(struct drm_device *dev, unsigned int nr);
+int exynos_plane_set_zpos_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file_priv);
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
new file mode 100644
index 0000000..f48f7ce
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -0,0 +1,1176 @@
+/*
+ * Copyright (C) 2011 Samsung Electronics Co.Ltd
+ * Authors:
+ * Seung-Woo Kim <sw0312.kim@samsung.com>
+ *	Inki Dae <inki.dae@samsung.com>
+ *	Joonyoung Shim <jy0922.shim@samsung.com>
+ *
+ * Based on drivers/media/video/s5p-tv/hdmi_drv.c
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include "drmP.h"
+#include "drm_edid.h"
+#include "drm_crtc_helper.h"
+
+#include "regs-hdmi.h"
+
+#include <linux/kernel.h>
+#include <linux/spinlock.h>
+#include <linux/wait.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/delay.h>
+#include <linux/pm_runtime.h>
+#include <linux/clk.h>
+#include <linux/regulator/consumer.h>
+
+#include <drm/exynos_drm.h>
+
+#include "exynos_drm_drv.h"
+#include "exynos_drm_hdmi.h"
+
+#include "exynos_hdmi.h"
+
+#define HDMI_OVERLAY_NUMBER	3
+#define get_hdmi_context(dev)	platform_get_drvdata(to_platform_device(dev))
+
+static const u8 hdmiphy_conf27[32] = {
+	0x01, 0x05, 0x00, 0xD8, 0x10, 0x1C, 0x30, 0x40,
+	0x6B, 0x10, 0x02, 0x51, 0xDF, 0xF2, 0x54, 0x87,
+	0x84, 0x00, 0x30, 0x38, 0x00, 0x08, 0x10, 0xE0,
+	0x22, 0x40, 0xE3, 0x26, 0x00, 0x00, 0x00, 0x00,
+};
+
+static const u8 hdmiphy_conf27_027[32] = {
+	0x01, 0x05, 0x00, 0xD4, 0x10, 0x9C, 0x09, 0x64,
+	0x6B, 0x10, 0x02, 0x51, 0xDF, 0xF2, 0x54, 0x87,
+	0x84, 0x00, 0x30, 0x38, 0x00, 0x08, 0x10, 0xE0,
+	0x22, 0x40, 0xE3, 0x26, 0x00, 0x00, 0x00, 0x00,
+};
+
+static const u8 hdmiphy_conf74_175[32] = {
+	0x01, 0x05, 0x00, 0xD8, 0x10, 0x9C, 0xef, 0x5B,
+	0x6D, 0x10, 0x01, 0x51, 0xef, 0xF3, 0x54, 0xb9,
+	0x84, 0x00, 0x30, 0x38, 0x00, 0x08, 0x10, 0xE0,
+	0x22, 0x40, 0xa5, 0x26, 0x01, 0x00, 0x00, 0x00,
+};
+
+static const u8 hdmiphy_conf74_25[32] = {
+	0x01, 0x05, 0x00, 0xd8, 0x10, 0x9c, 0xf8, 0x40,
+	0x6a, 0x10, 0x01, 0x51, 0xff, 0xf1, 0x54, 0xba,
+	0x84, 0x00, 0x10, 0x38, 0x00, 0x08, 0x10, 0xe0,
+	0x22, 0x40, 0xa4, 0x26, 0x01, 0x00, 0x00, 0x00,
+};
+
+static const u8 hdmiphy_conf148_5[32] = {
+	0x01, 0x05, 0x00, 0xD8, 0x10, 0x9C, 0xf8, 0x40,
+	0x6A, 0x18, 0x00, 0x51, 0xff, 0xF1, 0x54, 0xba,
+	0x84, 0x00, 0x10, 0x38, 0x00, 0x08, 0x10, 0xE0,
+	0x22, 0x40, 0xa4, 0x26, 0x02, 0x00, 0x00, 0x00,
+};
+
+struct hdmi_tg_regs {
+	u8 cmd;
+	u8 h_fsz_l;
+	u8 h_fsz_h;
+	u8 hact_st_l;
+	u8 hact_st_h;
+	u8 hact_sz_l;
+	u8 hact_sz_h;
+	u8 v_fsz_l;
+	u8 v_fsz_h;
+	u8 vsync_l;
+	u8 vsync_h;
+	u8 vsync2_l;
+	u8 vsync2_h;
+	u8 vact_st_l;
+	u8 vact_st_h;
+	u8 vact_sz_l;
+	u8 vact_sz_h;
+	u8 field_chg_l;
+	u8 field_chg_h;
+	u8 vact_st2_l;
+	u8 vact_st2_h;
+	u8 vsync_top_hdmi_l;
+	u8 vsync_top_hdmi_h;
+	u8 vsync_bot_hdmi_l;
+	u8 vsync_bot_hdmi_h;
+	u8 field_top_hdmi_l;
+	u8 field_top_hdmi_h;
+	u8 field_bot_hdmi_l;
+	u8 field_bot_hdmi_h;
+};
+
+struct hdmi_core_regs {
+	u8 h_blank[2];
+	u8 v_blank[3];
+	u8 h_v_line[3];
+	u8 vsync_pol[1];
+	u8 int_pro_mode[1];
+	u8 v_blank_f[3];
+	u8 h_sync_gen[3];
+	u8 v_sync_gen1[3];
+	u8 v_sync_gen2[3];
+	u8 v_sync_gen3[3];
+};
+
+struct hdmi_preset_conf {
+	struct hdmi_core_regs core;
+	struct hdmi_tg_regs tg;
+};
+
+static const struct hdmi_preset_conf hdmi_conf_480p = {
+	.core = {
+		.h_blank = {0x8a, 0x00},
+		.v_blank = {0x0d, 0x6a, 0x01},
+		.h_v_line = {0x0d, 0xa2, 0x35},
+		.vsync_pol = {0x01},
+		.int_pro_mode = {0x00},
+		.v_blank_f = {0x00, 0x00, 0x00},
+		.h_sync_gen = {0x0e, 0x30, 0x11},
+		.v_sync_gen1 = {0x0f, 0x90, 0x00},
+		/* other don't care */
+	},
+	.tg = {
+		0x00, /* cmd */
+		0x5a, 0x03, /* h_fsz */
+		0x8a, 0x00, 0xd0, 0x02, /* hact */
+		0x0d, 0x02, /* v_fsz */
+		0x01, 0x00, 0x33, 0x02, /* vsync */
+		0x2d, 0x00, 0xe0, 0x01, /* vact */
+		0x33, 0x02, /* field_chg */
+		0x49, 0x02, /* vact_st2 */
+		0x01, 0x00, 0x33, 0x02, /* vsync top/bot */
+		0x01, 0x00, 0x33, 0x02, /* field top/bot */
+	},
+};
+
+static const struct hdmi_preset_conf hdmi_conf_720p60 = {
+	.core = {
+		.h_blank = {0x72, 0x01},
+		.v_blank = {0xee, 0xf2, 0x00},
+		.h_v_line = {0xee, 0x22, 0x67},
+		.vsync_pol = {0x00},
+		.int_pro_mode = {0x00},
+		.v_blank_f = {0x00, 0x00, 0x00}, /* don't care */
+		.h_sync_gen = {0x6c, 0x50, 0x02},
+		.v_sync_gen1 = {0x0a, 0x50, 0x00},
+		.v_sync_gen2 = {0x01, 0x10, 0x00},
+		.v_sync_gen3 = {0x01, 0x10, 0x00},
+		/* other don't care */
+	},
+	.tg = {
+		0x00, /* cmd */
+		0x72, 0x06, /* h_fsz */
+		0x71, 0x01, 0x01, 0x05, /* hact */
+		0xee, 0x02, /* v_fsz */
+		0x01, 0x00, 0x33, 0x02, /* vsync */
+		0x1e, 0x00, 0xd0, 0x02, /* vact */
+		0x33, 0x02, /* field_chg */
+		0x49, 0x02, /* vact_st2 */
+		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
+		0x01, 0x00, 0x33, 0x02, /* field top/bot */
+	},
+};
+
+static const struct hdmi_preset_conf hdmi_conf_1080i50 = {
+	.core = {
+		.h_blank = {0xd0, 0x02},
+		.v_blank = {0x32, 0xB2, 0x00},
+		.h_v_line = {0x65, 0x04, 0xa5},
+		.vsync_pol = {0x00},
+		.int_pro_mode = {0x01},
+		.v_blank_f = {0x49, 0x2A, 0x23},
+		.h_sync_gen = {0x0E, 0xEA, 0x08},
+		.v_sync_gen1 = {0x07, 0x20, 0x00},
+		.v_sync_gen2 = {0x39, 0x42, 0x23},
+		.v_sync_gen3 = {0x38, 0x87, 0x73},
+		/* other don't care */
+	},
+	.tg = {
+		0x00, /* cmd */
+		0x50, 0x0A, /* h_fsz */
+		0xCF, 0x02, 0x81, 0x07, /* hact */
+		0x65, 0x04, /* v_fsz */
+		0x01, 0x00, 0x33, 0x02, /* vsync */
+		0x16, 0x00, 0x1c, 0x02, /* vact */
+		0x33, 0x02, /* field_chg */
+		0x49, 0x02, /* vact_st2 */
+		0x01, 0x00, 0x33, 0x02, /* vsync top/bot */
+		0x01, 0x00, 0x33, 0x02, /* field top/bot */
+	},
+};
+
+static const struct hdmi_preset_conf hdmi_conf_1080p50 = {
+	.core = {
+		.h_blank = {0xd0, 0x02},
+		.v_blank = {0x65, 0x6c, 0x01},
+		.h_v_line = {0x65, 0x04, 0xa5},
+		.vsync_pol = {0x00},
+		.int_pro_mode = {0x00},
+		.v_blank_f = {0x00, 0x00, 0x00}, /* don't care */
+		.h_sync_gen = {0x0e, 0xea, 0x08},
+		.v_sync_gen1 = {0x09, 0x40, 0x00},
+		.v_sync_gen2 = {0x01, 0x10, 0x00},
+		.v_sync_gen3 = {0x01, 0x10, 0x00},
+		/* other don't care */
+	},
+	.tg = {
+		0x00, /* cmd */
+		0x50, 0x0A, /* h_fsz */
+		0xCF, 0x02, 0x81, 0x07, /* hact */
+		0x65, 0x04, /* v_fsz */
+		0x01, 0x00, 0x33, 0x02, /* vsync */
+		0x2d, 0x00, 0x38, 0x04, /* vact */
+		0x33, 0x02, /* field_chg */
+		0x48, 0x02, /* vact_st2 */
+		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
+		0x01, 0x00, 0x33, 0x02, /* field top/bot */
+	},
+};
+
+static const struct hdmi_preset_conf hdmi_conf_1080i60 = {
+	.core = {
+		.h_blank = {0x18, 0x01},
+		.v_blank = {0x32, 0xB2, 0x00},
+		.h_v_line = {0x65, 0x84, 0x89},
+		.vsync_pol = {0x00},
+		.int_pro_mode = {0x01},
+		.v_blank_f = {0x49, 0x2A, 0x23},
+		.h_sync_gen = {0x56, 0x08, 0x02},
+		.v_sync_gen1 = {0x07, 0x20, 0x00},
+		.v_sync_gen2 = {0x39, 0x42, 0x23},
+		.v_sync_gen3 = {0xa4, 0x44, 0x4a},
+		/* other don't care */
+	},
+	.tg = {
+		0x00, /* cmd */
+		0x98, 0x08, /* h_fsz */
+		0x17, 0x01, 0x81, 0x07, /* hact */
+		0x65, 0x04, /* v_fsz */
+		0x01, 0x00, 0x33, 0x02, /* vsync */
+		0x16, 0x00, 0x1c, 0x02, /* vact */
+		0x33, 0x02, /* field_chg */
+		0x49, 0x02, /* vact_st2 */
+		0x01, 0x00, 0x33, 0x02, /* vsync top/bot */
+		0x01, 0x00, 0x33, 0x02, /* field top/bot */
+	},
+};
+
+static const struct hdmi_preset_conf hdmi_conf_1080p60 = {
+	.core = {
+		.h_blank = {0x18, 0x01},
+		.v_blank = {0x65, 0x6c, 0x01},
+		.h_v_line = {0x65, 0x84, 0x89},
+		.vsync_pol = {0x00},
+		.int_pro_mode = {0x00},
+		.v_blank_f = {0x00, 0x00, 0x00}, /* don't care */
+		.h_sync_gen = {0x56, 0x08, 0x02},
+		.v_sync_gen1 = {0x09, 0x40, 0x00},
+		.v_sync_gen2 = {0x01, 0x10, 0x00},
+		.v_sync_gen3 = {0x01, 0x10, 0x00},
+		/* other don't care */
+	},
+	.tg = {
+		0x00, /* cmd */
+		0x98, 0x08, /* h_fsz */
+		0x17, 0x01, 0x81, 0x07, /* hact */
+		0x65, 0x04, /* v_fsz */
+		0x01, 0x00, 0x33, 0x02, /* vsync */
+		0x2d, 0x00, 0x38, 0x04, /* vact */
+		0x33, 0x02, /* field_chg */
+		0x48, 0x02, /* vact_st2 */
+		0x01, 0x00, 0x01, 0x00, /* vsync top/bot */
+		0x01, 0x00, 0x33, 0x02, /* field top/bot */
+	},
+};
+
+static const struct hdmi_conf hdmi_confs[] = {
+	{ 1280, 720, 60, false, hdmiphy_conf74_25, &hdmi_conf_720p60 },
+	{ 1280, 720, 50, false, hdmiphy_conf74_25, &hdmi_conf_720p60 },
+	{ 720, 480, 60, false, hdmiphy_conf27_027, &hdmi_conf_480p },
+	{ 1920, 1080, 50, true, hdmiphy_conf74_25, &hdmi_conf_1080i50 },
+	{ 1920, 1080, 50, false, hdmiphy_conf148_5, &hdmi_conf_1080p50 },
+	{ 1920, 1080, 60, true, hdmiphy_conf74_25, &hdmi_conf_1080i60 },
+	{ 1920, 1080, 60, false, hdmiphy_conf148_5, &hdmi_conf_1080p60 },
+};
+
+
+static inline u32 hdmi_reg_read(struct hdmi_context *hdata, u32 reg_id)
+{
+	return readl(hdata->regs + reg_id);
+}
+
+static inline void hdmi_reg_writeb(struct hdmi_context *hdata,
+				 u32 reg_id, u8 value)
+{
+	writeb(value, hdata->regs + reg_id);
+}
+
+static inline void hdmi_reg_writemask(struct hdmi_context *hdata,
+				 u32 reg_id, u32 value, u32 mask)
+{
+	u32 old = readl(hdata->regs + reg_id);
+	value = (value & mask) | (old & ~mask);
+	writel(value, hdata->regs + reg_id);
+}
+
+static void hdmi_regs_dump(struct hdmi_context *hdata, char *prefix)
+{
+#define DUMPREG(reg_id) \
+	DRM_DEBUG_KMS("%s:" #reg_id " = %08x\n", prefix, \
+	readl(hdata->regs + reg_id))
+	DRM_DEBUG_KMS("%s: ---- CONTROL REGISTERS ----\n", prefix);
+	DUMPREG(HDMI_INTC_FLAG);
+	DUMPREG(HDMI_INTC_CON);
+	DUMPREG(HDMI_HPD_STATUS);
+	DUMPREG(HDMI_PHY_RSTOUT);
+	DUMPREG(HDMI_PHY_VPLL);
+	DUMPREG(HDMI_PHY_CMU);
+	DUMPREG(HDMI_CORE_RSTOUT);
+
+	DRM_DEBUG_KMS("%s: ---- CORE REGISTERS ----\n", prefix);
+	DUMPREG(HDMI_CON_0);
+	DUMPREG(HDMI_CON_1);
+	DUMPREG(HDMI_CON_2);
+	DUMPREG(HDMI_SYS_STATUS);
+	DUMPREG(HDMI_PHY_STATUS);
+	DUMPREG(HDMI_STATUS_EN);
+	DUMPREG(HDMI_HPD);
+	DUMPREG(HDMI_MODE_SEL);
+	DUMPREG(HDMI_HPD_GEN);
+	DUMPREG(HDMI_DC_CONTROL);
+	DUMPREG(HDMI_VIDEO_PATTERN_GEN);
+
+	DRM_DEBUG_KMS("%s: ---- CORE SYNC REGISTERS ----\n", prefix);
+	DUMPREG(HDMI_H_BLANK_0);
+	DUMPREG(HDMI_H_BLANK_1);
+	DUMPREG(HDMI_V_BLANK_0);
+	DUMPREG(HDMI_V_BLANK_1);
+	DUMPREG(HDMI_V_BLANK_2);
+	DUMPREG(HDMI_H_V_LINE_0);
+	DUMPREG(HDMI_H_V_LINE_1);
+	DUMPREG(HDMI_H_V_LINE_2);
+	DUMPREG(HDMI_VSYNC_POL);
+	DUMPREG(HDMI_INT_PRO_MODE);
+	DUMPREG(HDMI_V_BLANK_F_0);
+	DUMPREG(HDMI_V_BLANK_F_1);
+	DUMPREG(HDMI_V_BLANK_F_2);
+	DUMPREG(HDMI_H_SYNC_GEN_0);
+	DUMPREG(HDMI_H_SYNC_GEN_1);
+	DUMPREG(HDMI_H_SYNC_GEN_2);
+	DUMPREG(HDMI_V_SYNC_GEN_1_0);
+	DUMPREG(HDMI_V_SYNC_GEN_1_1);
+	DUMPREG(HDMI_V_SYNC_GEN_1_2);
+	DUMPREG(HDMI_V_SYNC_GEN_2_0);
+	DUMPREG(HDMI_V_SYNC_GEN_2_1);
+	DUMPREG(HDMI_V_SYNC_GEN_2_2);
+	DUMPREG(HDMI_V_SYNC_GEN_3_0);
+	DUMPREG(HDMI_V_SYNC_GEN_3_1);
+	DUMPREG(HDMI_V_SYNC_GEN_3_2);
+
+	DRM_DEBUG_KMS("%s: ---- TG REGISTERS ----\n", prefix);
+	DUMPREG(HDMI_TG_CMD);
+	DUMPREG(HDMI_TG_H_FSZ_L);
+	DUMPREG(HDMI_TG_H_FSZ_H);
+	DUMPREG(HDMI_TG_HACT_ST_L);
+	DUMPREG(HDMI_TG_HACT_ST_H);
+	DUMPREG(HDMI_TG_HACT_SZ_L);
+	DUMPREG(HDMI_TG_HACT_SZ_H);
+	DUMPREG(HDMI_TG_V_FSZ_L);
+	DUMPREG(HDMI_TG_V_FSZ_H);
+	DUMPREG(HDMI_TG_VSYNC_L);
+	DUMPREG(HDMI_TG_VSYNC_H);
+	DUMPREG(HDMI_TG_VSYNC2_L);
+	DUMPREG(HDMI_TG_VSYNC2_H);
+	DUMPREG(HDMI_TG_VACT_ST_L);
+	DUMPREG(HDMI_TG_VACT_ST_H);
+	DUMPREG(HDMI_TG_VACT_SZ_L);
+	DUMPREG(HDMI_TG_VACT_SZ_H);
+	DUMPREG(HDMI_TG_FIELD_CHG_L);
+	DUMPREG(HDMI_TG_FIELD_CHG_H);
+	DUMPREG(HDMI_TG_VACT_ST2_L);
+	DUMPREG(HDMI_TG_VACT_ST2_H);
+	DUMPREG(HDMI_TG_VSYNC_TOP_HDMI_L);
+	DUMPREG(HDMI_TG_VSYNC_TOP_HDMI_H);
+	DUMPREG(HDMI_TG_VSYNC_BOT_HDMI_L);
+	DUMPREG(HDMI_TG_VSYNC_BOT_HDMI_H);
+	DUMPREG(HDMI_TG_FIELD_TOP_HDMI_L);
+	DUMPREG(HDMI_TG_FIELD_TOP_HDMI_H);
+	DUMPREG(HDMI_TG_FIELD_BOT_HDMI_L);
+	DUMPREG(HDMI_TG_FIELD_BOT_HDMI_H);
+#undef DUMPREG
+}
+
+static int hdmi_conf_index(struct drm_display_mode *mode)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdmi_confs); ++i)
+		if (hdmi_confs[i].width == mode->hdisplay &&
+				hdmi_confs[i].height == mode->vdisplay &&
+				hdmi_confs[i].vrefresh == mode->vrefresh &&
+				hdmi_confs[i].interlace ==
+				((mode->flags & DRM_MODE_FLAG_INTERLACE) ?
+				 true : false))
+			return i;
+
+	return -1;
+}
+
+static bool hdmi_is_connected(void *ctx)
+{
+	struct hdmi_context *hdata = (struct hdmi_context *)ctx;
+	u32 val = hdmi_reg_read(hdata, HDMI_HPD_STATUS);
+
+	if (val)
+		return true;
+
+	return false;
+}
+
+static int hdmi_get_edid(void *ctx, struct drm_connector *connector,
+				u8 *edid, int len)
+{
+	struct edid *raw_edid;
+	struct hdmi_context *hdata = (struct hdmi_context *)ctx;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	if (!hdata->ddc_port)
+		return -ENODEV;
+
+	raw_edid = drm_get_edid(connector, hdata->ddc_port->adapter);
+	if (raw_edid) {
+		memcpy(edid, raw_edid, min((1 + raw_edid->extensions)
+					* EDID_LENGTH, len));
+		DRM_DEBUG_KMS("width[%d] x height[%d]\n",
+				raw_edid->width_cm, raw_edid->height_cm);
+	} else {
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static int hdmi_check_timing(void *ctx, void *timing)
+{
+	struct fb_videomode *check_timing = timing;
+	int i;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	DRM_DEBUG_KMS("[%d]x[%d] [%d]Hz [%x]\n", check_timing->xres,
+			check_timing->yres, check_timing->refresh,
+			check_timing->vmode);
+
+	for (i = 0; i < ARRAY_SIZE(hdmi_confs); ++i)
+		if (hdmi_confs[i].width == check_timing->xres &&
+			hdmi_confs[i].height == check_timing->yres &&
+			hdmi_confs[i].vrefresh == check_timing->refresh &&
+			hdmi_confs[i].interlace ==
+			((check_timing->vmode & FB_VMODE_INTERLACED) ?
+			 true : false))
+			return 0;
+
+	return -EINVAL;
+}
+
+static int hdmi_display_power_on(void *ctx, int mode)
+{
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		DRM_DEBUG_KMS("hdmi [on]\n");
+		break;
+	case DRM_MODE_DPMS_STANDBY:
+		break;
+	case DRM_MODE_DPMS_SUSPEND:
+		break;
+	case DRM_MODE_DPMS_OFF:
+		DRM_DEBUG_KMS("hdmi [off]\n");
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static struct exynos_hdmi_display_ops display_ops = {
+	.is_connected	= hdmi_is_connected,
+	.get_edid	= hdmi_get_edid,
+	.check_timing	= hdmi_check_timing,
+	.power_on	= hdmi_display_power_on,
+};
+
+static void hdmi_conf_reset(struct hdmi_context *hdata)
+{
+	/* disable hpd handle for drm */
+	hdata->hpd_handle = false;
+
+	/* resetting HDMI core */
+	hdmi_reg_writemask(hdata, HDMI_CORE_RSTOUT,  0, HDMI_CORE_SW_RSTOUT);
+	mdelay(10);
+	hdmi_reg_writemask(hdata, HDMI_CORE_RSTOUT, ~0, HDMI_CORE_SW_RSTOUT);
+	mdelay(10);
+
+	/* enable hpd handle for drm */
+	hdata->hpd_handle = true;
+}
+
+static void hdmi_conf_init(struct hdmi_context *hdata)
+{
+	/* disable hpd handle for drm */
+	hdata->hpd_handle = false;
+
+	/* enable HPD interrupts */
+	hdmi_reg_writemask(hdata, HDMI_INTC_CON, 0, HDMI_INTC_EN_GLOBAL |
+		HDMI_INTC_EN_HPD_PLUG | HDMI_INTC_EN_HPD_UNPLUG);
+	mdelay(10);
+	hdmi_reg_writemask(hdata, HDMI_INTC_CON, ~0, HDMI_INTC_EN_GLOBAL |
+		HDMI_INTC_EN_HPD_PLUG | HDMI_INTC_EN_HPD_UNPLUG);
+
+	/* choose HDMI mode */
+	hdmi_reg_writemask(hdata, HDMI_MODE_SEL,
+		HDMI_MODE_HDMI_EN, HDMI_MODE_MASK);
+	/* disable bluescreen */
+	hdmi_reg_writemask(hdata, HDMI_CON_0, 0, HDMI_BLUE_SCR_EN);
+	/* choose bluescreen (fecal) color */
+	hdmi_reg_writeb(hdata, HDMI_BLUE_SCREEN_0, 0x12);
+	hdmi_reg_writeb(hdata, HDMI_BLUE_SCREEN_1, 0x34);
+	hdmi_reg_writeb(hdata, HDMI_BLUE_SCREEN_2, 0x56);
+	/* enable AVI packet every vsync, fixes purple line problem */
+	hdmi_reg_writeb(hdata, HDMI_AVI_CON, 0x02);
+	/* force RGB, look to CEA-861-D, table 7 for more detail */
+	hdmi_reg_writeb(hdata, HDMI_AVI_BYTE(0), 0 << 5);
+	hdmi_reg_writemask(hdata, HDMI_CON_1, 0x10 << 5, 0x11 << 5);
+
+	hdmi_reg_writeb(hdata, HDMI_SPD_CON, 0x02);
+	hdmi_reg_writeb(hdata, HDMI_AUI_CON, 0x02);
+	hdmi_reg_writeb(hdata, HDMI_ACR_CON, 0x04);
+
+	/* enable hpd handle for drm */
+	hdata->hpd_handle = true;
+}
+
+static void hdmi_timing_apply(struct hdmi_context *hdata,
+				 const struct hdmi_preset_conf *conf)
+{
+	const struct hdmi_core_regs *core = &conf->core;
+	const struct hdmi_tg_regs *tg = &conf->tg;
+	int tries;
+
+	/* setting core registers */
+	hdmi_reg_writeb(hdata, HDMI_H_BLANK_0, core->h_blank[0]);
+	hdmi_reg_writeb(hdata, HDMI_H_BLANK_1, core->h_blank[1]);
+	hdmi_reg_writeb(hdata, HDMI_V_BLANK_0, core->v_blank[0]);
+	hdmi_reg_writeb(hdata, HDMI_V_BLANK_1, core->v_blank[1]);
+	hdmi_reg_writeb(hdata, HDMI_V_BLANK_2, core->v_blank[2]);
+	hdmi_reg_writeb(hdata, HDMI_H_V_LINE_0, core->h_v_line[0]);
+	hdmi_reg_writeb(hdata, HDMI_H_V_LINE_1, core->h_v_line[1]);
+	hdmi_reg_writeb(hdata, HDMI_H_V_LINE_2, core->h_v_line[2]);
+	hdmi_reg_writeb(hdata, HDMI_VSYNC_POL, core->vsync_pol[0]);
+	hdmi_reg_writeb(hdata, HDMI_INT_PRO_MODE, core->int_pro_mode[0]);
+	hdmi_reg_writeb(hdata, HDMI_V_BLANK_F_0, core->v_blank_f[0]);
+	hdmi_reg_writeb(hdata, HDMI_V_BLANK_F_1, core->v_blank_f[1]);
+	hdmi_reg_writeb(hdata, HDMI_V_BLANK_F_2, core->v_blank_f[2]);
+	hdmi_reg_writeb(hdata, HDMI_H_SYNC_GEN_0, core->h_sync_gen[0]);
+	hdmi_reg_writeb(hdata, HDMI_H_SYNC_GEN_1, core->h_sync_gen[1]);
+	hdmi_reg_writeb(hdata, HDMI_H_SYNC_GEN_2, core->h_sync_gen[2]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_1_0, core->v_sync_gen1[0]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_1_1, core->v_sync_gen1[1]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_1_2, core->v_sync_gen1[2]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_2_0, core->v_sync_gen2[0]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_2_1, core->v_sync_gen2[1]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_2_2, core->v_sync_gen2[2]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_3_0, core->v_sync_gen3[0]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_3_1, core->v_sync_gen3[1]);
+	hdmi_reg_writeb(hdata, HDMI_V_SYNC_GEN_3_2, core->v_sync_gen3[2]);
+	/* Timing generator registers */
+	hdmi_reg_writeb(hdata, HDMI_TG_H_FSZ_L, tg->h_fsz_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_H_FSZ_H, tg->h_fsz_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_HACT_ST_L, tg->hact_st_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_HACT_ST_H, tg->hact_st_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_HACT_SZ_L, tg->hact_sz_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_HACT_SZ_H, tg->hact_sz_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_V_FSZ_L, tg->v_fsz_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_V_FSZ_H, tg->v_fsz_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_L, tg->vsync_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_H, tg->vsync_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC2_L, tg->vsync2_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC2_H, tg->vsync2_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST_L, tg->vact_st_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST_H, tg->vact_st_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_SZ_L, tg->vact_sz_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_SZ_H, tg->vact_sz_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_CHG_L, tg->field_chg_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_CHG_H, tg->field_chg_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST2_L, tg->vact_st2_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_VACT_ST2_H, tg->vact_st2_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_TOP_HDMI_L, tg->vsync_top_hdmi_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_TOP_HDMI_H, tg->vsync_top_hdmi_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_BOT_HDMI_L, tg->vsync_bot_hdmi_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_VSYNC_BOT_HDMI_H, tg->vsync_bot_hdmi_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_TOP_HDMI_L, tg->field_top_hdmi_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_TOP_HDMI_H, tg->field_top_hdmi_h);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_BOT_HDMI_L, tg->field_bot_hdmi_l);
+	hdmi_reg_writeb(hdata, HDMI_TG_FIELD_BOT_HDMI_H, tg->field_bot_hdmi_h);
+
+	/* waiting for HDMIPHY's PLL to get to steady state */
+	for (tries = 100; tries; --tries) {
+		u32 val = hdmi_reg_read(hdata, HDMI_PHY_STATUS);
+		if (val & HDMI_PHY_STATUS_READY)
+			break;
+		mdelay(1);
+	}
+	/* steady state not achieved */
+	if (tries == 0) {
+		DRM_ERROR("hdmiphy's pll could not reach steady state.\n");
+		hdmi_regs_dump(hdata, "timing apply");
+	}
+
+	clk_disable(hdata->res.sclk_hdmi);
+	clk_set_parent(hdata->res.sclk_hdmi, hdata->res.sclk_hdmiphy);
+	clk_enable(hdata->res.sclk_hdmi);
+
+	/* enable HDMI and timing generator */
+	hdmi_reg_writemask(hdata, HDMI_CON_0, ~0, HDMI_EN);
+	if (core->int_pro_mode[0])
+		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN |
+				HDMI_FIELD_EN);
+	else
+		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN);
+}
+
+static void hdmiphy_conf_reset(struct hdmi_context *hdata)
+{
+	u8 buffer[2];
+
+	clk_disable(hdata->res.sclk_hdmi);
+	clk_set_parent(hdata->res.sclk_hdmi, hdata->res.sclk_pixel);
+	clk_enable(hdata->res.sclk_hdmi);
+
+	/* operation mode */
+	buffer[0] = 0x1f;
+	buffer[1] = 0x00;
+
+	if (hdata->hdmiphy_port)
+		i2c_master_send(hdata->hdmiphy_port, buffer, 2);
+
+	/* reset hdmiphy */
+	hdmi_reg_writemask(hdata, HDMI_PHY_RSTOUT, ~0, HDMI_PHY_SW_RSTOUT);
+	mdelay(10);
+	hdmi_reg_writemask(hdata, HDMI_PHY_RSTOUT,  0, HDMI_PHY_SW_RSTOUT);
+	mdelay(10);
+}
+
+static void hdmiphy_conf_apply(struct hdmi_context *hdata)
+{
+	u8 buffer[32];
+	u8 operation[2];
+	u8 read_buffer[32] = {0, };
+	int ret;
+	int i;
+
+	if (!hdata->hdmiphy_port) {
+		DRM_ERROR("hdmiphy is not attached\n");
+		return;
+	}
+
+	/* pixel clock */
+	memcpy(buffer, hdmi_confs[hdata->cur_conf].hdmiphy_data, 32);
+	ret = i2c_master_send(hdata->hdmiphy_port, buffer, 32);
+	if (ret != 32) {
+		DRM_ERROR("failed to configure HDMIPHY via I2C\n");
+		return;
+	}
+
+	mdelay(10);
+
+	/* operation mode */
+	operation[0] = 0x1f;
+	operation[1] = 0x80;
+
+	ret = i2c_master_send(hdata->hdmiphy_port, operation, 2);
+	if (ret != 2) {
+		DRM_ERROR("failed to enable hdmiphy\n");
+		return;
+	}
+
+	ret = i2c_master_recv(hdata->hdmiphy_port, read_buffer, 32);
+	if (ret < 0) {
+		DRM_ERROR("failed to read hdmiphy config\n");
+		return;
+	}
+
+	for (i = 0; i < ret; i++)
+		DRM_DEBUG_KMS("hdmiphy[0x%02x] write[0x%02x] - "
+			"recv [0x%02x]\n", i, buffer[i], read_buffer[i]);
+}
+
+static void hdmi_conf_apply(struct hdmi_context *hdata)
+{
+	const struct hdmi_preset_conf *conf =
+		  hdmi_confs[hdata->cur_conf].conf;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	hdmiphy_conf_reset(hdata);
+	hdmiphy_conf_apply(hdata);
+
+	hdmi_conf_reset(hdata);
+	hdmi_conf_init(hdata);
+
+	/* setting core registers */
+	hdmi_timing_apply(hdata, conf);
+
+	hdmi_regs_dump(hdata, "start");
+}
+
+static void hdmi_mode_set(void *ctx, void *mode)
+{
+	struct hdmi_context *hdata = (struct hdmi_context *)ctx;
+	int conf_idx;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	conf_idx = hdmi_conf_index(mode);
+	if (conf_idx >= 0 && conf_idx < ARRAY_SIZE(hdmi_confs))
+		hdata->cur_conf = conf_idx;
+	else
+		DRM_DEBUG_KMS("not supported mode\n");
+}
+
+static void hdmi_commit(void *ctx)
+{
+	struct hdmi_context *hdata = (struct hdmi_context *)ctx;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	hdmi_conf_apply(hdata);
+
+	hdata->enabled = true;
+}
+
+static void hdmi_disable(void *ctx)
+{
+	struct hdmi_context *hdata = (struct hdmi_context *)ctx;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	if (hdata->enabled) {
+		hdmiphy_conf_reset(hdata);
+		hdmi_conf_reset(hdata);
+	}
+}
+
+static struct exynos_hdmi_manager_ops manager_ops = {
+	.mode_set	= hdmi_mode_set,
+	.commit		= hdmi_commit,
+	.disable	= hdmi_disable,
+};
+
+/*
+ * Handle hotplug events outside the interrupt handler proper.
+ */
+static void hdmi_hotplug_func(struct work_struct *work)
+{
+	struct hdmi_context *hdata =
+		container_of(work, struct hdmi_context, hotplug_work);
+	struct exynos_drm_hdmi_context *ctx =
+		(struct exynos_drm_hdmi_context *)hdata->parent_ctx;
+
+	drm_helper_hpd_irq_event(ctx->drm_dev);
+}
+
+static irqreturn_t hdmi_irq_handler(int irq, void *arg)
+{
+	struct exynos_drm_hdmi_context *ctx = arg;
+	struct hdmi_context *hdata = (struct hdmi_context *)ctx->ctx;
+	u32 intc_flag;
+
+	intc_flag = hdmi_reg_read(hdata, HDMI_INTC_FLAG);
+	/* clearing flags for HPD plug/unplug */
+	if (intc_flag & HDMI_INTC_FLAG_HPD_UNPLUG) {
+		DRM_DEBUG_KMS("unplugged, handling:%d\n", hdata->hpd_handle);
+		hdmi_reg_writemask(hdata, HDMI_INTC_FLAG, ~0,
+			HDMI_INTC_FLAG_HPD_UNPLUG);
+	}
+	if (intc_flag & HDMI_INTC_FLAG_HPD_PLUG) {
+		DRM_DEBUG_KMS("plugged, handling:%d\n", hdata->hpd_handle);
+		hdmi_reg_writemask(hdata, HDMI_INTC_FLAG, ~0,
+			HDMI_INTC_FLAG_HPD_PLUG);
+	}
+
+	if (ctx->drm_dev && hdata->hpd_handle)
+		queue_work(hdata->wq, &hdata->hotplug_work);
+
+	return IRQ_HANDLED;
+}
+
+static int __devinit hdmi_resources_init(struct hdmi_context *hdata)
+{
+	struct device *dev = hdata->dev;
+	struct hdmi_resources *res = &hdata->res;
+	static char *supply[] = {
+		"hdmi-en",
+		"vdd",
+		"vdd_osc",
+		"vdd_pll",
+	};
+	int i, ret;
+
+	DRM_DEBUG_KMS("HDMI resource init\n");
+
+	memset(res, 0, sizeof *res);
+
+	/* get clocks, power */
+	res->hdmi = clk_get(dev, "hdmi");
+	if (IS_ERR_OR_NULL(res->hdmi)) {
+		DRM_ERROR("failed to get clock 'hdmi'\n");
+		goto fail;
+	}
+	res->sclk_hdmi = clk_get(dev, "sclk_hdmi");
+	if (IS_ERR_OR_NULL(res->sclk_hdmi)) {
+		DRM_ERROR("failed to get clock 'sclk_hdmi'\n");
+		goto fail;
+	}
+	res->sclk_pixel = clk_get(dev, "sclk_pixel");
+	if (IS_ERR_OR_NULL(res->sclk_pixel)) {
+		DRM_ERROR("failed to get clock 'sclk_pixel'\n");
+		goto fail;
+	}
+	res->sclk_hdmiphy = clk_get(dev, "sclk_hdmiphy");
+	if (IS_ERR_OR_NULL(res->sclk_hdmiphy)) {
+		DRM_ERROR("failed to get clock 'sclk_hdmiphy'\n");
+		goto fail;
+	}
+	res->hdmiphy = clk_get(dev, "hdmiphy");
+	if (IS_ERR_OR_NULL(res->hdmiphy)) {
+		DRM_ERROR("failed to get clock 'hdmiphy'\n");
+		goto fail;
+	}
+
+	clk_set_parent(res->sclk_hdmi, res->sclk_pixel);
+
+	res->regul_bulk = kzalloc(ARRAY_SIZE(supply) *
+		sizeof res->regul_bulk[0], GFP_KERNEL);
+	if (!res->regul_bulk) {
+		DRM_ERROR("failed to get memory for regulators\n");
+		goto fail;
+	}
+	for (i = 0; i < ARRAY_SIZE(supply); ++i) {
+		res->regul_bulk[i].supply = supply[i];
+		res->regul_bulk[i].consumer = NULL;
+	}
+	ret = regulator_bulk_get(dev, ARRAY_SIZE(supply), res->regul_bulk);
+	if (ret) {
+		DRM_ERROR("failed to get regulators\n");
+		goto fail;
+	}
+	res->regul_count = ARRAY_SIZE(supply);
+
+	return 0;
+fail:
+	DRM_ERROR("HDMI resource init - failed\n");
+	return -ENODEV;
+}
+
+static int hdmi_resources_cleanup(struct hdmi_context *hdata)
+{
+	struct hdmi_resources *res = &hdata->res;
+
+	regulator_bulk_free(res->regul_count, res->regul_bulk);
+	/* kfree is NULL-safe */
+	kfree(res->regul_bulk);
+	if (!IS_ERR_OR_NULL(res->hdmiphy))
+		clk_put(res->hdmiphy);
+	if (!IS_ERR_OR_NULL(res->sclk_hdmiphy))
+		clk_put(res->sclk_hdmiphy);
+	if (!IS_ERR_OR_NULL(res->sclk_pixel))
+		clk_put(res->sclk_pixel);
+	if (!IS_ERR_OR_NULL(res->sclk_hdmi))
+		clk_put(res->sclk_hdmi);
+	if (!IS_ERR_OR_NULL(res->hdmi))
+		clk_put(res->hdmi);
+	memset(res, 0, sizeof *res);
+
+	return 0;
+}
+
+static void hdmi_resource_poweron(struct hdmi_context *hdata)
+{
+	struct hdmi_resources *res = &hdata->res;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	/* turn HDMI power on */
+	regulator_bulk_enable(res->regul_count, res->regul_bulk);
+	/* power-on hdmi physical interface */
+	clk_enable(res->hdmiphy);
+	/* turn clocks on */
+	clk_enable(res->hdmi);
+	clk_enable(res->sclk_hdmi);
+
+	hdmiphy_conf_reset(hdata);
+	hdmi_conf_reset(hdata);
+	hdmi_conf_init(hdata);
+
+}
+
+static void hdmi_resource_poweroff(struct hdmi_context *hdata)
+{
+	struct hdmi_resources *res = &hdata->res;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	/* turn clocks off */
+	clk_disable(res->sclk_hdmi);
+	clk_disable(res->hdmi);
+	/* power-off hdmiphy */
+	clk_disable(res->hdmiphy);
+	/* turn HDMI power off */
+	regulator_bulk_disable(res->regul_count, res->regul_bulk);
+}
+
+static int hdmi_runtime_suspend(struct device *dev)
+{
+	struct exynos_drm_hdmi_context *ctx = get_hdmi_context(dev);
+
+	DRM_DEBUG_KMS("%s\n", __func__);
+
+	hdmi_resource_poweroff((struct hdmi_context *)ctx->ctx);
+
+	return 0;
+}
+
+static int hdmi_runtime_resume(struct device *dev)
+{
+	struct exynos_drm_hdmi_context *ctx = get_hdmi_context(dev);
+
+	DRM_DEBUG_KMS("%s\n", __func__);
+
+	hdmi_resource_poweron((struct hdmi_context *)ctx->ctx);
+
+	return 0;
+}
+
+static const struct dev_pm_ops hdmi_pm_ops = {
+	.runtime_suspend = hdmi_runtime_suspend,
+	.runtime_resume	 = hdmi_runtime_resume,
+};
+
+static struct i2c_client *hdmi_ddc, *hdmi_hdmiphy;
+
+void hdmi_attach_ddc_client(struct i2c_client *ddc)
+{
+	if (ddc)
+		hdmi_ddc = ddc;
+}
+EXPORT_SYMBOL(hdmi_attach_ddc_client);
+
+void hdmi_attach_hdmiphy_client(struct i2c_client *hdmiphy)
+{
+	if (hdmiphy)
+		hdmi_hdmiphy = hdmiphy;
+}
+EXPORT_SYMBOL(hdmi_attach_hdmiphy_client);
+
+static int __devinit hdmi_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct exynos_drm_hdmi_context *drm_hdmi_ctx;
+	struct hdmi_context *hdata;
+	struct exynos_drm_hdmi_pdata *pdata;
+	struct resource *res;
+	int ret;
+
+	DRM_DEBUG_KMS("[%d]\n", __LINE__);
+
+	pdata = pdev->dev.platform_data;
+	if (!pdata) {
+		DRM_ERROR("no platform data specified\n");
+		return -EINVAL;
+	}
+
+	drm_hdmi_ctx = kzalloc(sizeof(*drm_hdmi_ctx), GFP_KERNEL);
+	if (!drm_hdmi_ctx) {
+		DRM_ERROR("failed to allocate common hdmi context.\n");
+		return -ENOMEM;
+	}
+
+	hdata = kzalloc(sizeof(struct hdmi_context), GFP_KERNEL);
+	if (!hdata) {
+		DRM_ERROR("out of memory\n");
+		kfree(drm_hdmi_ctx);
+		return -ENOMEM;
+	}
+
+	drm_hdmi_ctx->ctx = (void *)hdata;
+	hdata->parent_ctx = (void *)drm_hdmi_ctx;
+
+	platform_set_drvdata(pdev, drm_hdmi_ctx);
+
+	hdata->default_win = pdata->default_win;
+	hdata->default_timing = &pdata->timing;
+	hdata->default_bpp = pdata->bpp;
+	hdata->dev = dev;
+
+	ret = hdmi_resources_init(hdata);
+	if (ret) {
+		ret = -EINVAL;
+		goto err_data;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		DRM_ERROR("failed to find registers\n");
+		ret = -ENOENT;
+		goto err_resource;
+	}
+
+	hdata->regs_res = request_mem_region(res->start, resource_size(res),
+					   dev_name(dev));
+	if (!hdata->regs_res) {
+		DRM_ERROR("failed to claim register region\n");
+		ret = -ENOENT;
+		goto err_resource;
+	}
+
+	hdata->regs = ioremap(res->start, resource_size(res));
+	if (!hdata->regs) {
+		DRM_ERROR("failed to map registers\n");
+		ret = -ENXIO;
+		goto err_req_region;
+	}
+
+	/* DDC i2c driver */
+	if (i2c_add_driver(&ddc_driver)) {
+		DRM_ERROR("failed to register ddc i2c driver\n");
+		ret = -ENOENT;
+		goto err_iomap;
+	}
+
+	hdata->ddc_port = hdmi_ddc;
+
+	/* hdmiphy i2c driver */
+	if (i2c_add_driver(&hdmiphy_driver)) {
+		DRM_ERROR("failed to register hdmiphy i2c driver\n");
+		ret = -ENOENT;
+		goto err_ddc;
+	}
+
+	hdata->hdmiphy_port = hdmi_hdmiphy;
+
+	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (res == NULL) {
+		DRM_ERROR("get interrupt resource failed.\n");
+		ret = -ENXIO;
+		goto err_hdmiphy;
+	}
+
+	/* create workqueue and hotplug work */
+	hdata->wq = alloc_workqueue("exynos-drm-hdmi",
+			WQ_UNBOUND | WQ_NON_REENTRANT, 1);
+	if (hdata->wq == NULL) {
+		DRM_ERROR("Failed to create workqueue.\n");
+		ret = -ENOMEM;
+		goto err_hdmiphy;
+	}
+	INIT_WORK(&hdata->hotplug_work, hdmi_hotplug_func);
+
+	/* register hpd interrupt */
+	ret = request_irq(res->start, hdmi_irq_handler, 0, "drm_hdmi",
+				drm_hdmi_ctx);
+	if (ret) {
+		DRM_ERROR("request interrupt failed.\n");
+		goto err_workqueue;
+	}
+	hdata->irq = res->start;
+
+	/* register specific callbacks to common hdmi. */
+	exynos_drm_display_ops_register(&display_ops);
+	exynos_drm_manager_ops_register(&manager_ops);
+
+	hdmi_resource_poweron(hdata);
+
+	return 0;
+
+err_workqueue:
+	destroy_workqueue(hdata->wq);
+err_hdmiphy:
+	i2c_del_driver(&hdmiphy_driver);
+err_ddc:
+	i2c_del_driver(&ddc_driver);
+err_iomap:
+	iounmap(hdata->regs);
+err_req_region:
+	release_resource(hdata->regs_res);
+	kfree(hdata->regs_res);
+err_resource:
+	hdmi_resources_cleanup(hdata);
+err_data:
+	kfree(hdata);
+	kfree(drm_hdmi_ctx);
+	return ret;
+}
+
+static int __devexit hdmi_remove(struct platform_device *pdev)
+{
+	struct exynos_drm_hdmi_context *ctx = platform_get_drvdata(pdev);
+	struct hdmi_context *hdata = (struct hdmi_context *)ctx->ctx;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	hdmi_resource_poweroff(hdata);
+
+	disable_irq(hdata->irq);
+	free_irq(hdata->irq, hdata);
+
+	cancel_work_sync(&hdata->hotplug_work);
+	destroy_workqueue(hdata->wq);
+
+	hdmi_resources_cleanup(hdata);
+
+	iounmap(hdata->regs);
+
+	release_resource(hdata->regs_res);
+	kfree(hdata->regs_res);
+
+	/* hdmiphy i2c driver */
+	i2c_del_driver(&hdmiphy_driver);
+	/* DDC i2c driver */
+	i2c_del_driver(&ddc_driver);
+
+	kfree(hdata);
+
+	return 0;
+}
+
+struct platform_driver hdmi_driver = {
+	.probe		= hdmi_probe,
+	.remove		= __devexit_p(hdmi_remove),
+	.driver		= {
+		.name	= "exynos4-hdmi",
+		.owner	= THIS_MODULE,
+		.pm = &hdmi_pm_ops,
+	},
+};
+EXPORT_SYMBOL(hdmi_driver);
+
+MODULE_AUTHOR("Seung-Woo Kim, <sw0312.kim@samsung.com>");
+MODULE_AUTHOR("Inki Dae <inki.dae@samsung.com>");
+MODULE_AUTHOR("Joonyoung Shim <jy0922.shim@samsung.com>");
+MODULE_DESCRIPTION("Samsung DRM HDMI core Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.h b/drivers/gpu/drm/exynos/exynos_hdmi.h
new file mode 100644
index 0000000..31d6cf8
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.h
@@ -0,0 +1,87 @@
+/*
+ *
+ * Copyright (c) 2011 Samsung Electronics Co., Ltd.
+ * Authors:
+ *	Inki Dae <inki.dae@samsung.com>
+ *	Seung-Woo Kim <sw0312.kim@samsung.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * VA LINUX SYSTEMS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _EXYNOS_HDMI_H_
+#define _EXYNOS_HDMI_H_
+
+struct hdmi_conf {
+	int width;
+	int height;
+	int vrefresh;
+	bool interlace;
+	const u8 *hdmiphy_data;
+	const struct hdmi_preset_conf *conf;
+};
+
+struct hdmi_resources {
+	struct clk *hdmi;
+	struct clk *sclk_hdmi;
+	struct clk *sclk_pixel;
+	struct clk *sclk_hdmiphy;
+	struct clk *hdmiphy;
+	struct regulator_bulk_data *regul_bulk;
+	int regul_count;
+};
+
+struct hdmi_context {
+	struct device			*dev;
+	struct drm_device		*drm_dev;
+	struct fb_videomode		*default_timing;
+	unsigned int			default_win;
+	unsigned int			default_bpp;
+	bool				hpd_handle;
+	bool				enabled;
+
+	struct resource			*regs_res;
+	/** base address of HDMI registers */
+	void __iomem *regs;
+	/** HDMI hotplug interrupt */
+	unsigned int irq;
+	/** workqueue for delayed work */
+	struct workqueue_struct *wq;
+	/** hotplug handling work */
+	struct work_struct hotplug_work;
+
+	struct i2c_client *ddc_port;
+	struct i2c_client *hdmiphy_port;
+
+	/** current hdmiphy conf index */
+	int cur_conf;
+	/** other resources */
+	struct hdmi_resources res;
+
+	void *parent_ctx;
+};
+
+
+void hdmi_attach_ddc_client(struct i2c_client *ddc);
+void hdmi_attach_hdmiphy_client(struct i2c_client *hdmiphy);
+
+extern struct i2c_driver hdmiphy_driver;
+extern struct i2c_driver ddc_driver;
+
+#endif
diff --git a/drivers/gpu/drm/exynos/exynos_hdmiphy.c b/drivers/gpu/drm/exynos/exynos_hdmiphy.c
new file mode 100644
index 0000000..9fe2995
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_hdmiphy.c
@@ -0,0 +1,58 @@
+/*
+ * Copyright (C) 2011 Samsung Electronics Co.Ltd
+ * Authors:
+ *	Seung-Woo Kim <sw0312.kim@samsung.com>
+ *	Inki Dae <inki.dae@samsung.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include "drmP.h"
+
+#include <linux/kernel.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+
+#include "exynos_drm_drv.h"
+#include "exynos_hdmi.h"
+
+
+static int hdmiphy_probe(struct i2c_client *client,
+	const struct i2c_device_id *id)
+{
+	hdmi_attach_hdmiphy_client(client);
+
+	dev_info(&client->adapter->dev, "attached s5p_hdmiphy "
+		"into i2c adapter successfully\n");
+
+	return 0;
+}
+
+static int hdmiphy_remove(struct i2c_client *client)
+{
+	dev_info(&client->adapter->dev, "detached s5p_hdmiphy "
+		"from i2c adapter successfully\n");
+
+	return 0;
+}
+
+static const struct i2c_device_id hdmiphy_id[] = {
+	{ "s5p_hdmiphy", 0 },
+	{ },
+};
+
+struct i2c_driver hdmiphy_driver = {
+	.driver = {
+		.name	= "s5p-hdmiphy",
+		.owner	= THIS_MODULE,
+	},
+	.id_table = hdmiphy_id,
+	.probe		= hdmiphy_probe,
+	.remove		= __devexit_p(hdmiphy_remove),
+	.command		= NULL,
+};
+EXPORT_SYMBOL(hdmiphy_driver);
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
new file mode 100644
index 0000000..ac24cff3
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -0,0 +1,1070 @@
+/*
+ * Copyright (C) 2011 Samsung Electronics Co.Ltd
+ * Authors:
+ * Seung-Woo Kim <sw0312.kim@samsung.com>
+ *	Inki Dae <inki.dae@samsung.com>
+ *	Joonyoung Shim <jy0922.shim@samsung.com>
+ *
+ * Based on drivers/media/video/s5p-tv/mixer_reg.c
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+
+#include "drmP.h"
+
+#include "regs-mixer.h"
+#include "regs-vp.h"
+
+#include <linux/kernel.h>
+#include <linux/spinlock.h>
+#include <linux/wait.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/delay.h>
+#include <linux/pm_runtime.h>
+#include <linux/clk.h>
+#include <linux/regulator/consumer.h>
+
+#include <drm/exynos_drm.h>
+
+#include "exynos_drm_drv.h"
+#include "exynos_drm_hdmi.h"
+#include "exynos_hdmi.h"
+#include "exynos_mixer.h"
+
+#define get_mixer_context(dev)	platform_get_drvdata(to_platform_device(dev))
+
+static const u8 filter_y_horiz_tap8[] = {
+	0,	-1,	-1,	-1,	-1,	-1,	-1,	-1,
+	-1,	-1,	-1,	-1,	-1,	0,	0,	0,
+	0,	2,	4,	5,	6,	6,	6,	6,
+	6,	5,	5,	4,	3,	2,	1,	1,
+	0,	-6,	-12,	-16,	-18,	-20,	-21,	-20,
+	-20,	-18,	-16,	-13,	-10,	-8,	-5,	-2,
+	127,	126,	125,	121,	114,	107,	99,	89,
+	79,	68,	57,	46,	35,	25,	16,	8,
+};
+
+static const u8 filter_y_vert_tap4[] = {
+	0,	-3,	-6,	-8,	-8,	-8,	-8,	-7,
+	-6,	-5,	-4,	-3,	-2,	-1,	-1,	0,
+	127,	126,	124,	118,	111,	102,	92,	81,
+	70,	59,	48,	37,	27,	19,	11,	5,
+	0,	5,	11,	19,	27,	37,	48,	59,
+	70,	81,	92,	102,	111,	118,	124,	126,
+	0,	0,	-1,	-1,	-2,	-3,	-4,	-5,
+	-6,	-7,	-8,	-8,	-8,	-8,	-6,	-3,
+};
+
+static const u8 filter_cr_horiz_tap4[] = {
+	0,	-3,	-6,	-8,	-8,	-8,	-8,	-7,
+	-6,	-5,	-4,	-3,	-2,	-1,	-1,	0,
+	127,	126,	124,	118,	111,	102,	92,	81,
+	70,	59,	48,	37,	27,	19,	11,	5,
+};
+
+static inline u32 vp_reg_read(struct mixer_resources *res, u32 reg_id)
+{
+	return readl(res->vp_regs + reg_id);
+}
+
+static inline void vp_reg_write(struct mixer_resources *res, u32 reg_id,
+				 u32 val)
+{
+	writel(val, res->vp_regs + reg_id);
+}
+
+static inline void vp_reg_writemask(struct mixer_resources *res, u32 reg_id,
+				 u32 val, u32 mask)
+{
+	u32 old = vp_reg_read(res, reg_id);
+
+	val = (val & mask) | (old & ~mask);
+	writel(val, res->vp_regs + reg_id);
+}
+
+static inline u32 mixer_reg_read(struct mixer_resources *res, u32 reg_id)
+{
+	return readl(res->mixer_regs + reg_id);
+}
+
+static inline void mixer_reg_write(struct mixer_resources *res, u32 reg_id,
+				 u32 val)
+{
+	writel(val, res->mixer_regs + reg_id);
+}
+
+static inline void mixer_reg_writemask(struct mixer_resources *res,
+				 u32 reg_id, u32 val, u32 mask)
+{
+	u32 old = mixer_reg_read(res, reg_id);
+
+	val = (val & mask) | (old & ~mask);
+	writel(val, res->mixer_regs + reg_id);
+}
+
+static void mixer_regs_dump(struct mixer_context *ctx)
+{
+#define DUMPREG(reg_id) \
+do { \
+	DRM_DEBUG_KMS(#reg_id " = %08x\n", \
+		(u32)readl(ctx->mixer_res.mixer_regs + reg_id)); \
+} while (0)
+
+	DUMPREG(MXR_STATUS);
+	DUMPREG(MXR_CFG);
+	DUMPREG(MXR_INT_EN);
+	DUMPREG(MXR_INT_STATUS);
+
+	DUMPREG(MXR_LAYER_CFG);
+	DUMPREG(MXR_VIDEO_CFG);
+
+	DUMPREG(MXR_GRAPHIC0_CFG);
+	DUMPREG(MXR_GRAPHIC0_BASE);
+	DUMPREG(MXR_GRAPHIC0_SPAN);
+	DUMPREG(MXR_GRAPHIC0_WH);
+	DUMPREG(MXR_GRAPHIC0_SXY);
+	DUMPREG(MXR_GRAPHIC0_DXY);
+
+	DUMPREG(MXR_GRAPHIC1_CFG);
+	DUMPREG(MXR_GRAPHIC1_BASE);
+	DUMPREG(MXR_GRAPHIC1_SPAN);
+	DUMPREG(MXR_GRAPHIC1_WH);
+	DUMPREG(MXR_GRAPHIC1_SXY);
+	DUMPREG(MXR_GRAPHIC1_DXY);
+#undef DUMPREG
+}
+
+static void vp_regs_dump(struct mixer_context *ctx)
+{
+#define DUMPREG(reg_id) \
+do { \
+	DRM_DEBUG_KMS(#reg_id " = %08x\n", \
+		(u32) readl(ctx->mixer_res.vp_regs + reg_id)); \
+} while (0)
+
+	DUMPREG(VP_ENABLE);
+	DUMPREG(VP_SRESET);
+	DUMPREG(VP_SHADOW_UPDATE);
+	DUMPREG(VP_FIELD_ID);
+	DUMPREG(VP_MODE);
+	DUMPREG(VP_IMG_SIZE_Y);
+	DUMPREG(VP_IMG_SIZE_C);
+	DUMPREG(VP_PER_RATE_CTRL);
+	DUMPREG(VP_TOP_Y_PTR);
+	DUMPREG(VP_BOT_Y_PTR);
+	DUMPREG(VP_TOP_C_PTR);
+	DUMPREG(VP_BOT_C_PTR);
+	DUMPREG(VP_ENDIAN_MODE);
+	DUMPREG(VP_SRC_H_POSITION);
+	DUMPREG(VP_SRC_V_POSITION);
+	DUMPREG(VP_SRC_WIDTH);
+	DUMPREG(VP_SRC_HEIGHT);
+	DUMPREG(VP_DST_H_POSITION);
+	DUMPREG(VP_DST_V_POSITION);
+	DUMPREG(VP_DST_WIDTH);
+	DUMPREG(VP_DST_HEIGHT);
+	DUMPREG(VP_H_RATIO);
+	DUMPREG(VP_V_RATIO);
+
+#undef DUMPREG
+}
+
+static inline void vp_filter_set(struct mixer_resources *res,
+		int reg_id, const u8 *data, unsigned int size)
+{
+	/* assure 4-byte align */
+	BUG_ON(size & 3);
+	for (; size; size -= 4, reg_id += 4, data += 4) {
+		u32 val = (data[0] << 24) |  (data[1] << 16) |
+			(data[2] << 8) | data[3];
+		vp_reg_write(res, reg_id, val);
+	}
+}
+
+static void vp_default_filter(struct mixer_resources *res)
+{
+	vp_filter_set(res, VP_POLY8_Y0_LL,
+		filter_y_horiz_tap8, sizeof filter_y_horiz_tap8);
+	vp_filter_set(res, VP_POLY4_Y0_LL,
+		filter_y_vert_tap4, sizeof filter_y_vert_tap4);
+	vp_filter_set(res, VP_POLY4_C0_LL,
+		filter_cr_horiz_tap4, sizeof filter_cr_horiz_tap4);
+}
+
+static void mixer_vsync_set_update(struct mixer_context *ctx, bool enable)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+
+	/* block update on vsync */
+	mixer_reg_writemask(res, MXR_STATUS, enable ?
+			MXR_STATUS_SYNC_ENABLE : 0, MXR_STATUS_SYNC_ENABLE);
+
+	vp_reg_write(res, VP_SHADOW_UPDATE, enable ?
+			VP_SHADOW_UPDATE_ENABLE : 0);
+}
+
+static void mixer_cfg_scan(struct mixer_context *ctx, unsigned int height)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+	u32 val;
+
+	/* choosing between interlace and progressive mode */
+	val = (ctx->interlace ? MXR_CFG_SCAN_INTERLACE :
+				MXR_CFG_SCAN_PROGRASSIVE);
+
+	/* choosing between porper HD and SD mode */
+	if (height == 480)
+		val |= MXR_CFG_SCAN_NTSC | MXR_CFG_SCAN_SD;
+	else if (height == 576)
+		val |= MXR_CFG_SCAN_PAL | MXR_CFG_SCAN_SD;
+	else if (height == 720)
+		val |= MXR_CFG_SCAN_HD_720 | MXR_CFG_SCAN_HD;
+	else if (height == 1080)
+		val |= MXR_CFG_SCAN_HD_1080 | MXR_CFG_SCAN_HD;
+	else
+		val |= MXR_CFG_SCAN_HD_720 | MXR_CFG_SCAN_HD;
+
+	mixer_reg_writemask(res, MXR_CFG, val, MXR_CFG_SCAN_MASK);
+}
+
+static void mixer_cfg_rgb_fmt(struct mixer_context *ctx, unsigned int height)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+	u32 val;
+
+	if (height == 480) {
+		val = MXR_CFG_RGB601_0_255;
+	} else if (height == 576) {
+		val = MXR_CFG_RGB601_0_255;
+	} else if (height == 720) {
+		val = MXR_CFG_RGB709_16_235;
+		mixer_reg_write(res, MXR_CM_COEFF_Y,
+				(1 << 30) | (94 << 20) | (314 << 10) |
+				(32 << 0));
+		mixer_reg_write(res, MXR_CM_COEFF_CB,
+				(972 << 20) | (851 << 10) | (225 << 0));
+		mixer_reg_write(res, MXR_CM_COEFF_CR,
+				(225 << 20) | (820 << 10) | (1004 << 0));
+	} else if (height == 1080) {
+		val = MXR_CFG_RGB709_16_235;
+		mixer_reg_write(res, MXR_CM_COEFF_Y,
+				(1 << 30) | (94 << 20) | (314 << 10) |
+				(32 << 0));
+		mixer_reg_write(res, MXR_CM_COEFF_CB,
+				(972 << 20) | (851 << 10) | (225 << 0));
+		mixer_reg_write(res, MXR_CM_COEFF_CR,
+				(225 << 20) | (820 << 10) | (1004 << 0));
+	} else {
+		val = MXR_CFG_RGB709_16_235;
+		mixer_reg_write(res, MXR_CM_COEFF_Y,
+				(1 << 30) | (94 << 20) | (314 << 10) |
+				(32 << 0));
+		mixer_reg_write(res, MXR_CM_COEFF_CB,
+				(972 << 20) | (851 << 10) | (225 << 0));
+		mixer_reg_write(res, MXR_CM_COEFF_CR,
+				(225 << 20) | (820 << 10) | (1004 << 0));
+	}
+
+	mixer_reg_writemask(res, MXR_CFG, val, MXR_CFG_RGB_FMT_MASK);
+}
+
+static void mixer_cfg_layer(struct mixer_context *ctx, int win, bool enable)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+	u32 val = enable ? ~0 : 0;
+
+	switch (win) {
+	case 0:
+		mixer_reg_writemask(res, MXR_CFG, val, MXR_CFG_GRP0_ENABLE);
+		break;
+	case 1:
+		mixer_reg_writemask(res, MXR_CFG, val, MXR_CFG_GRP1_ENABLE);
+		break;
+	case 2:
+		vp_reg_writemask(res, VP_ENABLE, val, VP_ENABLE_ON);
+		mixer_reg_writemask(res, MXR_CFG, val, MXR_CFG_VP_ENABLE);
+		break;
+	}
+}
+
+static void mixer_run(struct mixer_context *ctx)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+
+	mixer_reg_writemask(res, MXR_STATUS, ~0, MXR_STATUS_REG_RUN);
+
+	mixer_regs_dump(ctx);
+}
+
+static void vp_video_buffer(struct mixer_context *ctx, int win)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+	unsigned long flags;
+	struct hdmi_win_data *win_data;
+	unsigned int full_width, full_height, width, height;
+	unsigned int x_ratio, y_ratio;
+	unsigned int src_x_offset, src_y_offset, dst_x_offset, dst_y_offset;
+	unsigned int mode_width, mode_height;
+	unsigned int buf_num;
+	dma_addr_t luma_addr[2], chroma_addr[2];
+	bool tiled_mode = false;
+	bool crcb_mode = false;
+	u32 val;
+
+	win_data = &ctx->win_data[win];
+
+	switch (win_data->pixel_format) {
+	case DRM_FORMAT_NV12MT:
+		tiled_mode = true;
+	case DRM_FORMAT_NV12M:
+		crcb_mode = false;
+		buf_num = 2;
+		break;
+	/* TODO: single buffer format NV12, NV21 */
+	default:
+		/* ignore pixel format at disable time */
+		if (!win_data->dma_addr)
+			break;
+
+		DRM_ERROR("pixel format for vp is wrong [%d].\n",
+				win_data->pixel_format);
+		return;
+	}
+
+	full_width = win_data->fb_width;
+	full_height = win_data->fb_height;
+	width = win_data->crtc_width;
+	height = win_data->crtc_height;
+	mode_width = win_data->mode_width;
+	mode_height = win_data->mode_height;
+
+	/* scaling feature: (src << 16) / dst */
+	x_ratio = (width << 16) / width;
+	y_ratio = (height << 16) / height;
+
+	src_x_offset = win_data->fb_x;
+	src_y_offset = win_data->fb_y;
+	dst_x_offset = win_data->crtc_x;
+	dst_y_offset = win_data->crtc_y;
+
+	if (buf_num == 2) {
+		luma_addr[0] = win_data->dma_addr;
+		chroma_addr[0] = win_data->chroma_dma_addr;
+	} else {
+		luma_addr[0] = win_data->dma_addr;
+		chroma_addr[0] = win_data->dma_addr
+			+ (full_width * full_height);
+	}
+
+	if (win_data->scan_flags & DRM_MODE_FLAG_INTERLACE) {
+		ctx->interlace = true;
+		if (tiled_mode) {
+			luma_addr[1] = luma_addr[0] + 0x40;
+			chroma_addr[1] = chroma_addr[0] + 0x40;
+		} else {
+			luma_addr[1] = luma_addr[0] + full_width;
+			chroma_addr[1] = chroma_addr[0] + full_width;
+		}
+	} else {
+		ctx->interlace = false;
+		luma_addr[1] = 0;
+		chroma_addr[1] = 0;
+	}
+
+	spin_lock_irqsave(&res->reg_slock, flags);
+	mixer_vsync_set_update(ctx, false);
+
+	/* interlace or progressive scan mode */
+	val = (ctx->interlace ? ~0 : 0);
+	vp_reg_writemask(res, VP_MODE, val, VP_MODE_LINE_SKIP);
+
+	/* setup format */
+	val = (crcb_mode ? VP_MODE_NV21 : VP_MODE_NV12);
+	val |= (tiled_mode ? VP_MODE_MEM_TILED : VP_MODE_MEM_LINEAR);
+	vp_reg_writemask(res, VP_MODE, val, VP_MODE_FMT_MASK);
+
+	/* setting size of input image */
+	vp_reg_write(res, VP_IMG_SIZE_Y, VP_IMG_HSIZE(full_width) |
+		VP_IMG_VSIZE(full_height));
+	/* chroma height has to reduced by 2 to avoid chroma distorions */
+	vp_reg_write(res, VP_IMG_SIZE_C, VP_IMG_HSIZE(full_width) |
+		VP_IMG_VSIZE(full_height / 2));
+
+	vp_reg_write(res, VP_SRC_WIDTH, width);
+	vp_reg_write(res, VP_SRC_HEIGHT, height);
+	vp_reg_write(res, VP_SRC_H_POSITION,
+			VP_SRC_H_POSITION_VAL(src_x_offset));
+	vp_reg_write(res, VP_SRC_V_POSITION, src_y_offset);
+
+	vp_reg_write(res, VP_DST_WIDTH, width);
+	vp_reg_write(res, VP_DST_H_POSITION, dst_x_offset);
+	if (ctx->interlace) {
+		vp_reg_write(res, VP_DST_HEIGHT, height / 2);
+		vp_reg_write(res, VP_DST_V_POSITION, dst_y_offset / 2);
+	} else {
+		vp_reg_write(res, VP_DST_HEIGHT, height);
+		vp_reg_write(res, VP_DST_V_POSITION, dst_y_offset);
+	}
+
+	vp_reg_write(res, VP_H_RATIO, x_ratio);
+	vp_reg_write(res, VP_V_RATIO, y_ratio);
+
+	vp_reg_write(res, VP_ENDIAN_MODE, VP_ENDIAN_MODE_LITTLE);
+
+	/* set buffer address to vp */
+	vp_reg_write(res, VP_TOP_Y_PTR, luma_addr[0]);
+	vp_reg_write(res, VP_BOT_Y_PTR, luma_addr[1]);
+	vp_reg_write(res, VP_TOP_C_PTR, chroma_addr[0]);
+	vp_reg_write(res, VP_BOT_C_PTR, chroma_addr[1]);
+
+	mixer_cfg_scan(ctx, mode_height);
+	mixer_cfg_rgb_fmt(ctx, mode_height);
+	mixer_cfg_layer(ctx, win, true);
+	mixer_run(ctx);
+
+	mixer_vsync_set_update(ctx, true);
+	spin_unlock_irqrestore(&res->reg_slock, flags);
+
+	vp_regs_dump(ctx);
+}
+
+static void mixer_graph_buffer(struct mixer_context *ctx, int win)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+	unsigned long flags;
+	struct hdmi_win_data *win_data;
+	unsigned int full_width, width, height;
+	unsigned int x_ratio, y_ratio;
+	unsigned int src_x_offset, src_y_offset, dst_x_offset, dst_y_offset;
+	unsigned int mode_width, mode_height;
+	dma_addr_t dma_addr;
+	unsigned int fmt;
+	u32 val;
+
+	win_data = &ctx->win_data[win];
+
+	#define RGB565 4
+	#define ARGB1555 5
+	#define ARGB4444 6
+	#define ARGB8888 7
+
+	switch (win_data->bpp) {
+	case 16:
+		fmt = ARGB4444;
+		break;
+	case 32:
+		fmt = ARGB8888;
+		break;
+	default:
+		fmt = ARGB8888;
+	}
+
+	dma_addr = win_data->dma_addr;
+	full_width = win_data->fb_width;
+	width = win_data->crtc_width;
+	height = win_data->crtc_height;
+	mode_width = win_data->mode_width;
+	mode_height = win_data->mode_height;
+
+	/* 2x scaling feature */
+	x_ratio = 0;
+	y_ratio = 0;
+
+	src_x_offset = win_data->fb_x;
+	src_y_offset = win_data->fb_y;
+	dst_x_offset = win_data->crtc_x;
+	dst_y_offset = win_data->crtc_y;
+
+	/* converting dma address base and source offset */
+	dma_addr = dma_addr
+		+ (src_x_offset * win_data->bpp >> 3)
+		+ (src_y_offset * full_width * win_data->bpp >> 3);
+	src_x_offset = 0;
+	src_y_offset = 0;
+
+	if (win_data->scan_flags & DRM_MODE_FLAG_INTERLACE)
+		ctx->interlace = true;
+	else
+		ctx->interlace = false;
+
+	spin_lock_irqsave(&res->reg_slock, flags);
+	mixer_vsync_set_update(ctx, false);
+
+	/* setup format */
+	mixer_reg_writemask(res, MXR_GRAPHIC_CFG(win),
+		MXR_GRP_CFG_FORMAT_VAL(fmt), MXR_GRP_CFG_FORMAT_MASK);
+
+	/* setup geometry */
+	mixer_reg_write(res, MXR_GRAPHIC_SPAN(win), full_width);
+
+	val  = MXR_GRP_WH_WIDTH(width);
+	val |= MXR_GRP_WH_HEIGHT(height);
+	val |= MXR_GRP_WH_H_SCALE(x_ratio);
+	val |= MXR_GRP_WH_V_SCALE(y_ratio);
+	mixer_reg_write(res, MXR_GRAPHIC_WH(win), val);
+
+	/* setup offsets in source image */
+	val  = MXR_GRP_SXY_SX(src_x_offset);
+	val |= MXR_GRP_SXY_SY(src_y_offset);
+	mixer_reg_write(res, MXR_GRAPHIC_SXY(win), val);
+
+	/* setup offsets in display image */
+	val  = MXR_GRP_DXY_DX(dst_x_offset);
+	val |= MXR_GRP_DXY_DY(dst_y_offset);
+	mixer_reg_write(res, MXR_GRAPHIC_DXY(win), val);
+
+	/* set buffer address to mixer */
+	mixer_reg_write(res, MXR_GRAPHIC_BASE(win), dma_addr);
+
+	mixer_cfg_scan(ctx, mode_height);
+	mixer_cfg_rgb_fmt(ctx, mode_height);
+	mixer_cfg_layer(ctx, win, true);
+	mixer_run(ctx);
+
+	mixer_vsync_set_update(ctx, true);
+	spin_unlock_irqrestore(&res->reg_slock, flags);
+}
+
+static void vp_win_reset(struct mixer_context *ctx)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+	int tries = 100;
+
+	vp_reg_write(res, VP_SRESET, VP_SRESET_PROCESSING);
+	for (tries = 100; tries; --tries) {
+		/* waiting until VP_SRESET_PROCESSING is 0 */
+		if (~vp_reg_read(res, VP_SRESET) & VP_SRESET_PROCESSING)
+			break;
+		mdelay(10);
+	}
+	WARN(tries == 0, "failed to reset Video Processor\n");
+}
+
+static int mixer_enable_vblank(void *ctx, int pipe)
+{
+	struct mixer_context *mixer_ctx = ctx;
+	struct mixer_resources *res = &mixer_ctx->mixer_res;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	mixer_ctx->pipe = pipe;
+
+	/* enable vsync interrupt */
+	mixer_reg_writemask(res, MXR_INT_EN, MXR_INT_EN_VSYNC,
+			MXR_INT_EN_VSYNC);
+
+	return 0;
+}
+
+static void mixer_disable_vblank(void *ctx)
+{
+	struct mixer_context *mixer_ctx = ctx;
+	struct mixer_resources *res = &mixer_ctx->mixer_res;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	/* disable vsync interrupt */
+	mixer_reg_writemask(res, MXR_INT_EN, 0, MXR_INT_EN_VSYNC);
+}
+
+static void mixer_win_mode_set(void *ctx,
+			      struct exynos_drm_overlay *overlay)
+{
+	struct mixer_context *mixer_ctx = ctx;
+	struct hdmi_win_data *win_data;
+	int win;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	if (!overlay) {
+		DRM_ERROR("overlay is NULL\n");
+		return;
+	}
+
+	DRM_DEBUG_KMS("set [%d]x[%d] at (%d,%d) to [%d]x[%d] at (%d,%d)\n",
+				 overlay->fb_width, overlay->fb_height,
+				 overlay->fb_x, overlay->fb_y,
+				 overlay->crtc_width, overlay->crtc_height,
+				 overlay->crtc_x, overlay->crtc_y);
+
+	win = overlay->zpos;
+	if (win == DEFAULT_ZPOS)
+		win = mixer_ctx->default_win;
+
+	if (win < 0 || win > HDMI_OVERLAY_NUMBER) {
+		DRM_ERROR("overlay plane[%d] is wrong\n", win);
+		return;
+	}
+
+	win_data = &mixer_ctx->win_data[win];
+
+	win_data->dma_addr = overlay->dma_addr[0];
+	win_data->vaddr = overlay->vaddr[0];
+	win_data->chroma_dma_addr = overlay->dma_addr[1];
+	win_data->chroma_vaddr = overlay->vaddr[1];
+	win_data->pixel_format = overlay->pixel_format;
+	win_data->bpp = overlay->bpp;
+
+	win_data->crtc_x = overlay->crtc_x;
+	win_data->crtc_y = overlay->crtc_y;
+	win_data->crtc_width = overlay->crtc_width;
+	win_data->crtc_height = overlay->crtc_height;
+
+	win_data->fb_x = overlay->fb_x;
+	win_data->fb_y = overlay->fb_y;
+	win_data->fb_width = overlay->fb_width;
+	win_data->fb_height = overlay->fb_height;
+
+	win_data->mode_width = overlay->mode_width;
+	win_data->mode_height = overlay->mode_height;
+
+	win_data->scan_flags = overlay->scan_flag;
+}
+
+static void mixer_win_commit(void *ctx, int zpos)
+{
+	struct mixer_context *mixer_ctx = ctx;
+	int win = zpos;
+
+	DRM_DEBUG_KMS("[%d] %s, win: %d\n", __LINE__, __func__, win);
+
+	if (win == DEFAULT_ZPOS)
+		win = mixer_ctx->default_win;
+
+	if (win < 0 || win > HDMI_OVERLAY_NUMBER) {
+		DRM_ERROR("overlay plane[%d] is wrong\n", win);
+		return;
+	}
+
+	if (win > 1)
+		vp_video_buffer(mixer_ctx, win);
+	else
+		mixer_graph_buffer(mixer_ctx, win);
+}
+
+static void mixer_win_disable(void *ctx, int zpos)
+{
+	struct mixer_context *mixer_ctx = ctx;
+	struct mixer_resources *res = &mixer_ctx->mixer_res;
+	unsigned long flags;
+	int win = zpos;
+
+	DRM_DEBUG_KMS("[%d] %s, win: %d\n", __LINE__, __func__, win);
+
+	if (win == DEFAULT_ZPOS)
+		win = mixer_ctx->default_win;
+
+	if (win < 0 || win > HDMI_OVERLAY_NUMBER) {
+		DRM_ERROR("overlay plane[%d] is wrong\n", win);
+		return;
+	}
+
+	spin_lock_irqsave(&res->reg_slock, flags);
+	mixer_vsync_set_update(mixer_ctx, false);
+
+	mixer_cfg_layer(mixer_ctx, win, false);
+
+	mixer_vsync_set_update(mixer_ctx, true);
+	spin_unlock_irqrestore(&res->reg_slock, flags);
+}
+
+static struct exynos_hdmi_overlay_ops overlay_ops = {
+	.enable_vblank		= mixer_enable_vblank,
+	.disable_vblank		= mixer_disable_vblank,
+	.win_mode_set		= mixer_win_mode_set,
+	.win_commit		= mixer_win_commit,
+	.win_disable		= mixer_win_disable,
+};
+
+/* for pageflip event */
+static void mixer_finish_pageflip(struct drm_device *drm_dev, int crtc)
+{
+	struct exynos_drm_private *dev_priv = drm_dev->dev_private;
+	struct drm_pending_vblank_event *e, *t;
+	struct timeval now;
+	unsigned long flags;
+	bool is_checked = false;
+
+	spin_lock_irqsave(&drm_dev->event_lock, flags);
+
+	list_for_each_entry_safe(e, t, &dev_priv->pageflip_event_list,
+			base.link) {
+		/* if event's pipe isn't same as crtc then ignore it. */
+		if (crtc != e->pipe)
+			continue;
+
+		is_checked = true;
+		do_gettimeofday(&now);
+		e->event.sequence = 0;
+		e->event.tv_sec = now.tv_sec;
+		e->event.tv_usec = now.tv_usec;
+
+		list_move_tail(&e->base.link, &e->base.file_priv->event_list);
+		wake_up_interruptible(&e->base.file_priv->event_wait);
+	}
+
+	if (is_checked)
+		drm_vblank_put(drm_dev, crtc);
+
+	spin_unlock_irqrestore(&drm_dev->event_lock, flags);
+}
+
+static irqreturn_t mixer_irq_handler(int irq, void *arg)
+{
+	struct exynos_drm_hdmi_context *drm_hdmi_ctx = arg;
+	struct mixer_context *ctx =
+			(struct mixer_context *)drm_hdmi_ctx->ctx;
+	struct mixer_resources *res = &ctx->mixer_res;
+	u32 val, val_base;
+
+	spin_lock(&res->reg_slock);
+
+	/* read interrupt status for handling and clearing flags for VSYNC */
+	val = mixer_reg_read(res, MXR_INT_STATUS);
+
+	/* handling VSYNC */
+	if (val & MXR_INT_STATUS_VSYNC) {
+		/* interlace scan need to check shadow register */
+		if (ctx->interlace) {
+			val_base = mixer_reg_read(res, MXR_GRAPHIC_BASE_S(0));
+			if (ctx->win_data[0].dma_addr != val_base)
+				goto out;
+
+			val_base = mixer_reg_read(res, MXR_GRAPHIC_BASE_S(1));
+			if (ctx->win_data[1].dma_addr != val_base)
+				goto out;
+		}
+
+		drm_handle_vblank(drm_hdmi_ctx->drm_dev, ctx->pipe);
+		mixer_finish_pageflip(drm_hdmi_ctx->drm_dev, ctx->pipe);
+	}
+
+out:
+	/* clear interrupts */
+	if (~val & MXR_INT_EN_VSYNC) {
+		/* vsync interrupt use different bit for read and clear */
+		val &= ~MXR_INT_EN_VSYNC;
+		val |= MXR_INT_CLEAR_VSYNC;
+	}
+	mixer_reg_write(res, MXR_INT_STATUS, val);
+
+	spin_unlock(&res->reg_slock);
+
+	return IRQ_HANDLED;
+}
+
+static void mixer_win_reset(struct mixer_context *ctx)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+	unsigned long flags;
+	u32 val; /* value stored to register */
+
+	spin_lock_irqsave(&res->reg_slock, flags);
+	mixer_vsync_set_update(ctx, false);
+
+	mixer_reg_writemask(res, MXR_CFG, MXR_CFG_DST_HDMI, MXR_CFG_DST_MASK);
+
+	/* set output in RGB888 mode */
+	mixer_reg_writemask(res, MXR_CFG, MXR_CFG_OUT_RGB888, MXR_CFG_OUT_MASK);
+
+	/* 16 beat burst in DMA */
+	mixer_reg_writemask(res, MXR_STATUS, MXR_STATUS_16_BURST,
+		MXR_STATUS_BURST_MASK);
+
+	/* setting default layer priority: layer1 > video > layer0
+	 * because typical usage scenario would be
+	 * layer0 - framebuffer
+	 * video - video overlay
+	 * layer1 - OSD
+	 */
+	val  = MXR_LAYER_CFG_GRP0_VAL(1);
+	val |= MXR_LAYER_CFG_VP_VAL(2);
+	val |= MXR_LAYER_CFG_GRP1_VAL(3);
+	mixer_reg_write(res, MXR_LAYER_CFG, val);
+
+	/* setting background color */
+	mixer_reg_write(res, MXR_BG_COLOR0, 0x008080);
+	mixer_reg_write(res, MXR_BG_COLOR1, 0x008080);
+	mixer_reg_write(res, MXR_BG_COLOR2, 0x008080);
+
+	/* setting graphical layers */
+
+	val  = MXR_GRP_CFG_COLOR_KEY_DISABLE; /* no blank key */
+	val |= MXR_GRP_CFG_WIN_BLEND_EN;
+	val |= MXR_GRP_CFG_ALPHA_VAL(0xff); /* non-transparent alpha */
+
+	/* the same configuration for both layers */
+	mixer_reg_write(res, MXR_GRAPHIC_CFG(0), val);
+
+	val |= MXR_GRP_CFG_BLEND_PRE_MUL;
+	val |= MXR_GRP_CFG_PIXEL_BLEND_EN;
+	mixer_reg_write(res, MXR_GRAPHIC_CFG(1), val);
+
+	/* configuration of Video Processor Registers */
+	vp_win_reset(ctx);
+	vp_default_filter(res);
+
+	/* disable all layers */
+	mixer_reg_writemask(res, MXR_CFG, 0, MXR_CFG_GRP0_ENABLE);
+	mixer_reg_writemask(res, MXR_CFG, 0, MXR_CFG_GRP1_ENABLE);
+	mixer_reg_writemask(res, MXR_CFG, 0, MXR_CFG_VP_ENABLE);
+
+	mixer_vsync_set_update(ctx, true);
+	spin_unlock_irqrestore(&res->reg_slock, flags);
+}
+
+static void mixer_resource_poweron(struct mixer_context *ctx)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	clk_enable(res->mixer);
+	clk_enable(res->vp);
+	clk_enable(res->sclk_mixer);
+
+	mixer_win_reset(ctx);
+}
+
+static void mixer_resource_poweroff(struct mixer_context *ctx)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+
+	DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__);
+
+	clk_disable(res->mixer);
+	clk_disable(res->vp);
+	clk_disable(res->sclk_mixer);
+}
+
+static int mixer_runtime_resume(struct device *dev)
+{
+	struct exynos_drm_hdmi_context *ctx = get_mixer_context(dev);
+
+	DRM_DEBUG_KMS("resume - start\n");
+
+	mixer_resource_poweron((struct mixer_context *)ctx->ctx);
+
+	return 0;
+}
+
+static int mixer_runtime_suspend(struct device *dev)
+{
+	struct exynos_drm_hdmi_context *ctx = get_mixer_context(dev);
+
+	DRM_DEBUG_KMS("suspend - start\n");
+
+	mixer_resource_poweroff((struct mixer_context *)ctx->ctx);
+
+	return 0;
+}
+
+static const struct dev_pm_ops mixer_pm_ops = {
+	.runtime_suspend = mixer_runtime_suspend,
+	.runtime_resume	 = mixer_runtime_resume,
+};
+
+static int __devinit mixer_resources_init(struct exynos_drm_hdmi_context *ctx,
+				 struct platform_device *pdev)
+{
+	struct mixer_context *mixer_ctx =
+			(struct mixer_context *)ctx->ctx;
+	struct device *dev = &pdev->dev;
+	struct mixer_resources *mixer_res = &mixer_ctx->mixer_res;
+	struct resource *res;
+	int ret;
+
+	mixer_res->dev = dev;
+	spin_lock_init(&mixer_res->reg_slock);
+
+	mixer_res->mixer = clk_get(dev, "mixer");
+	if (IS_ERR_OR_NULL(mixer_res->mixer)) {
+		dev_err(dev, "failed to get clock 'mixer'\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+	mixer_res->vp = clk_get(dev, "vp");
+	if (IS_ERR_OR_NULL(mixer_res->vp)) {
+		dev_err(dev, "failed to get clock 'vp'\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+	mixer_res->sclk_mixer = clk_get(dev, "sclk_mixer");
+	if (IS_ERR_OR_NULL(mixer_res->sclk_mixer)) {
+		dev_err(dev, "failed to get clock 'sclk_mixer'\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+	mixer_res->sclk_hdmi = clk_get(dev, "sclk_hdmi");
+	if (IS_ERR_OR_NULL(mixer_res->sclk_hdmi)) {
+		dev_err(dev, "failed to get clock 'sclk_hdmi'\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+	mixer_res->sclk_dac = clk_get(dev, "sclk_dac");
+	if (IS_ERR_OR_NULL(mixer_res->sclk_dac)) {
+		dev_err(dev, "failed to get clock 'sclk_dac'\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "mxr");
+	if (res == NULL) {
+		dev_err(dev, "get memory resource failed.\n");
+		ret = -ENXIO;
+		goto fail;
+	}
+
+	clk_set_parent(mixer_res->sclk_mixer, mixer_res->sclk_hdmi);
+
+	mixer_res->mixer_regs = ioremap(res->start, resource_size(res));
+	if (mixer_res->mixer_regs == NULL) {
+		dev_err(dev, "register mapping failed.\n");
+		ret = -ENXIO;
+		goto fail;
+	}
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "vp");
+	if (res == NULL) {
+		dev_err(dev, "get memory resource failed.\n");
+		ret = -ENXIO;
+		goto fail_mixer_regs;
+	}
+
+	mixer_res->vp_regs = ioremap(res->start, resource_size(res));
+	if (mixer_res->vp_regs == NULL) {
+		dev_err(dev, "register mapping failed.\n");
+		ret = -ENXIO;
+		goto fail_mixer_regs;
+	}
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_IRQ, "irq");
+	if (res == NULL) {
+		dev_err(dev, "get interrupt resource failed.\n");
+		ret = -ENXIO;
+		goto fail_vp_regs;
+	}
+
+	ret = request_irq(res->start, mixer_irq_handler, 0, "drm_mixer", ctx);
+	if (ret) {
+		dev_err(dev, "request interrupt failed.\n");
+		goto fail_vp_regs;
+	}
+	mixer_res->irq = res->start;
+
+	return 0;
+
+fail_vp_regs:
+	iounmap(mixer_res->vp_regs);
+
+fail_mixer_regs:
+	iounmap(mixer_res->mixer_regs);
+
+fail:
+	if (!IS_ERR_OR_NULL(mixer_res->sclk_dac))
+		clk_put(mixer_res->sclk_dac);
+	if (!IS_ERR_OR_NULL(mixer_res->sclk_hdmi))
+		clk_put(mixer_res->sclk_hdmi);
+	if (!IS_ERR_OR_NULL(mixer_res->sclk_mixer))
+		clk_put(mixer_res->sclk_mixer);
+	if (!IS_ERR_OR_NULL(mixer_res->vp))
+		clk_put(mixer_res->vp);
+	if (!IS_ERR_OR_NULL(mixer_res->mixer))
+		clk_put(mixer_res->mixer);
+	mixer_res->dev = NULL;
+	return ret;
+}
+
+static void mixer_resources_cleanup(struct mixer_context *ctx)
+{
+	struct mixer_resources *res = &ctx->mixer_res;
+
+	disable_irq(res->irq);
+	free_irq(res->irq, ctx);
+
+	iounmap(res->vp_regs);
+	iounmap(res->mixer_regs);
+}
+
+static int __devinit mixer_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct exynos_drm_hdmi_context *drm_hdmi_ctx;
+	struct mixer_context *ctx;
+	int ret;
+
+	dev_info(dev, "probe start\n");
+
+	drm_hdmi_ctx = kzalloc(sizeof(*drm_hdmi_ctx), GFP_KERNEL);
+	if (!drm_hdmi_ctx) {
+		DRM_ERROR("failed to allocate common hdmi context.\n");
+		return -ENOMEM;
+	}
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx) {
+		DRM_ERROR("failed to alloc mixer context.\n");
+		kfree(drm_hdmi_ctx);
+		return -ENOMEM;
+	}
+
+	drm_hdmi_ctx->ctx = (void *)ctx;
+
+	platform_set_drvdata(pdev, drm_hdmi_ctx);
+
+	/* acquire resources: regs, irqs, clocks */
+	ret = mixer_resources_init(drm_hdmi_ctx, pdev);
+	if (ret)
+		goto fail;
+
+	/* register specific callback point to common hdmi. */
+	exynos_drm_overlay_ops_register(&overlay_ops);
+
+	mixer_resource_poweron(ctx);
+
+	return 0;
+
+
+fail:
+	dev_info(dev, "probe failed\n");
+	return ret;
+}
+
+static int mixer_remove(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct exynos_drm_hdmi_context *drm_hdmi_ctx =
+					platform_get_drvdata(pdev);
+	struct mixer_context *ctx = (struct mixer_context *)drm_hdmi_ctx->ctx;
+
+	dev_info(dev, "remove sucessful\n");
+
+	mixer_resource_poweroff(ctx);
+	mixer_resources_cleanup(ctx);
+
+	return 0;
+}
+
+struct platform_driver mixer_driver = {
+	.driver = {
+		.name = "s5p-mixer",
+		.owner = THIS_MODULE,
+		.pm = &mixer_pm_ops,
+	},
+	.probe = mixer_probe,
+	.remove = __devexit_p(mixer_remove),
+};
+EXPORT_SYMBOL(mixer_driver);
+
+MODULE_AUTHOR("Seung-Woo Kim, <sw0312.kim@samsung.com>");
+MODULE_AUTHOR("Inki Dae <inki.dae@samsung.com>");
+MODULE_AUTHOR("Joonyoung Shim <jy0922.shim@samsung.com>");
+MODULE_DESCRIPTION("Samsung DRM HDMI mixer Driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.h b/drivers/gpu/drm/exynos/exynos_mixer.h
new file mode 100644
index 0000000..cebacfe
--- /dev/null
+++ b/drivers/gpu/drm/exynos/exynos_mixer.h
@@ -0,0 +1,92 @@
+/*
+ *
+ * Copyright (c) 2011 Samsung Electronics Co., Ltd.
+ * Authors:
+ *	Seung-Woo Kim <sw0312.kim@samsung.com>
+ *	Inki Dae <inki.dae@samsung.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * VA LINUX SYSTEMS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _EXYNOS_MIXER_H_
+#define _EXYNOS_MIXER_H_
+
+#define HDMI_OVERLAY_NUMBER	3
+
+struct hdmi_win_data {
+	dma_addr_t		dma_addr;
+	void __iomem		*vaddr;
+	dma_addr_t		chroma_dma_addr;
+	void __iomem		*chroma_vaddr;
+	uint32_t		pixel_format;
+	unsigned int		bpp;
+	unsigned int		crtc_x;
+	unsigned int		crtc_y;
+	unsigned int		crtc_width;
+	unsigned int		crtc_height;
+	unsigned int		fb_x;
+	unsigned int		fb_y;
+	unsigned int		fb_width;
+	unsigned int		fb_height;
+	unsigned int		mode_width;
+	unsigned int		mode_height;
+	unsigned int		scan_flags;
+};
+
+struct mixer_resources {
+	struct device *dev;
+	/** interrupt index */
+	int irq;
+	/** pointer to Mixer registers */
+	void __iomem *mixer_regs;
+	/** pointer to Video Processor registers */
+	void __iomem *vp_regs;
+	/** spinlock for protection of registers */
+	spinlock_t reg_slock;
+	/** other resources */
+	struct clk *mixer;
+	struct clk *vp;
+	struct clk *sclk_mixer;
+	struct clk *sclk_hdmi;
+	struct clk *sclk_dac;
+};
+
+struct mixer_context {
+	unsigned int			default_win;
+	struct fb_videomode		*default_timing;
+	unsigned int			default_bpp;
+
+	/** mixer interrupt */
+	unsigned int irq;
+	/** current crtc pipe for vblank */
+	int pipe;
+	/** interlace scan mode */
+	bool interlace;
+	/** vp enabled status */
+	bool vp_enabled;
+
+	/** mixer and vp resources */
+	struct mixer_resources mixer_res;
+
+	/** overlay window data */
+	struct hdmi_win_data		win_data[HDMI_OVERLAY_NUMBER];
+};
+
+#endif
diff --git a/drivers/gpu/drm/exynos/regs-hdmi.h b/drivers/gpu/drm/exynos/regs-hdmi.h
new file mode 100644
index 0000000..72e6b52
--- /dev/null
+++ b/drivers/gpu/drm/exynos/regs-hdmi.h
@@ -0,0 +1,147 @@
+/*
+ *
+ *  Cloned from drivers/media/video/s5p-tv/regs-hdmi.h
+ *
+ * Copyright (c) 2010-2011 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com/
+ *
+ * HDMI register header file for Samsung TVOUT driver
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#ifndef SAMSUNG_REGS_HDMI_H
+#define SAMSUNG_REGS_HDMI_H
+
+/*
+ * Register part
+*/
+
+#define HDMI_CTRL_BASE(x)		((x) + 0x00000000)
+#define HDMI_CORE_BASE(x)		((x) + 0x00010000)
+#define HDMI_TG_BASE(x)			((x) + 0x00050000)
+
+/* Control registers */
+#define HDMI_INTC_CON			HDMI_CTRL_BASE(0x0000)
+#define HDMI_INTC_FLAG			HDMI_CTRL_BASE(0x0004)
+#define HDMI_HPD_STATUS			HDMI_CTRL_BASE(0x000C)
+#define HDMI_PHY_RSTOUT			HDMI_CTRL_BASE(0x0014)
+#define HDMI_PHY_VPLL			HDMI_CTRL_BASE(0x0018)
+#define HDMI_PHY_CMU			HDMI_CTRL_BASE(0x001C)
+#define HDMI_CORE_RSTOUT		HDMI_CTRL_BASE(0x0020)
+
+/* Core registers */
+#define HDMI_CON_0			HDMI_CORE_BASE(0x0000)
+#define HDMI_CON_1			HDMI_CORE_BASE(0x0004)
+#define HDMI_CON_2			HDMI_CORE_BASE(0x0008)
+#define HDMI_SYS_STATUS			HDMI_CORE_BASE(0x0010)
+#define HDMI_PHY_STATUS			HDMI_CORE_BASE(0x0014)
+#define HDMI_STATUS_EN			HDMI_CORE_BASE(0x0020)
+#define HDMI_HPD			HDMI_CORE_BASE(0x0030)
+#define HDMI_MODE_SEL			HDMI_CORE_BASE(0x0040)
+#define HDMI_BLUE_SCREEN_0		HDMI_CORE_BASE(0x0050)
+#define HDMI_BLUE_SCREEN_1		HDMI_CORE_BASE(0x0054)
+#define HDMI_BLUE_SCREEN_2		HDMI_CORE_BASE(0x0058)
+#define HDMI_H_BLANK_0			HDMI_CORE_BASE(0x00A0)
+#define HDMI_H_BLANK_1			HDMI_CORE_BASE(0x00A4)
+#define HDMI_V_BLANK_0			HDMI_CORE_BASE(0x00B0)
+#define HDMI_V_BLANK_1			HDMI_CORE_BASE(0x00B4)
+#define HDMI_V_BLANK_2			HDMI_CORE_BASE(0x00B8)
+#define HDMI_H_V_LINE_0			HDMI_CORE_BASE(0x00C0)
+#define HDMI_H_V_LINE_1			HDMI_CORE_BASE(0x00C4)
+#define HDMI_H_V_LINE_2			HDMI_CORE_BASE(0x00C8)
+#define HDMI_VSYNC_POL			HDMI_CORE_BASE(0x00E4)
+#define HDMI_INT_PRO_MODE		HDMI_CORE_BASE(0x00E8)
+#define HDMI_V_BLANK_F_0		HDMI_CORE_BASE(0x0110)
+#define HDMI_V_BLANK_F_1		HDMI_CORE_BASE(0x0114)
+#define HDMI_V_BLANK_F_2		HDMI_CORE_BASE(0x0118)
+#define HDMI_H_SYNC_GEN_0		HDMI_CORE_BASE(0x0120)
+#define HDMI_H_SYNC_GEN_1		HDMI_CORE_BASE(0x0124)
+#define HDMI_H_SYNC_GEN_2		HDMI_CORE_BASE(0x0128)
+#define HDMI_V_SYNC_GEN_1_0		HDMI_CORE_BASE(0x0130)
+#define HDMI_V_SYNC_GEN_1_1		HDMI_CORE_BASE(0x0134)
+#define HDMI_V_SYNC_GEN_1_2		HDMI_CORE_BASE(0x0138)
+#define HDMI_V_SYNC_GEN_2_0		HDMI_CORE_BASE(0x0140)
+#define HDMI_V_SYNC_GEN_2_1		HDMI_CORE_BASE(0x0144)
+#define HDMI_V_SYNC_GEN_2_2		HDMI_CORE_BASE(0x0148)
+#define HDMI_V_SYNC_GEN_3_0		HDMI_CORE_BASE(0x0150)
+#define HDMI_V_SYNC_GEN_3_1		HDMI_CORE_BASE(0x0154)
+#define HDMI_V_SYNC_GEN_3_2		HDMI_CORE_BASE(0x0158)
+#define HDMI_ACR_CON			HDMI_CORE_BASE(0x0180)
+#define HDMI_AVI_CON			HDMI_CORE_BASE(0x0300)
+#define HDMI_AVI_BYTE(n)		HDMI_CORE_BASE(0x0320 + 4 * (n))
+#define HDMI_DC_CONTROL			HDMI_CORE_BASE(0x05C0)
+#define HDMI_VIDEO_PATTERN_GEN		HDMI_CORE_BASE(0x05C4)
+#define HDMI_HPD_GEN			HDMI_CORE_BASE(0x05C8)
+#define HDMI_AUI_CON			HDMI_CORE_BASE(0x0360)
+#define HDMI_SPD_CON			HDMI_CORE_BASE(0x0400)
+
+/* Timing generator registers */
+#define HDMI_TG_CMD			HDMI_TG_BASE(0x0000)
+#define HDMI_TG_H_FSZ_L			HDMI_TG_BASE(0x0018)
+#define HDMI_TG_H_FSZ_H			HDMI_TG_BASE(0x001C)
+#define HDMI_TG_HACT_ST_L		HDMI_TG_BASE(0x0020)
+#define HDMI_TG_HACT_ST_H		HDMI_TG_BASE(0x0024)
+#define HDMI_TG_HACT_SZ_L		HDMI_TG_BASE(0x0028)
+#define HDMI_TG_HACT_SZ_H		HDMI_TG_BASE(0x002C)
+#define HDMI_TG_V_FSZ_L			HDMI_TG_BASE(0x0030)
+#define HDMI_TG_V_FSZ_H			HDMI_TG_BASE(0x0034)
+#define HDMI_TG_VSYNC_L			HDMI_TG_BASE(0x0038)
+#define HDMI_TG_VSYNC_H			HDMI_TG_BASE(0x003C)
+#define HDMI_TG_VSYNC2_L		HDMI_TG_BASE(0x0040)
+#define HDMI_TG_VSYNC2_H		HDMI_TG_BASE(0x0044)
+#define HDMI_TG_VACT_ST_L		HDMI_TG_BASE(0x0048)
+#define HDMI_TG_VACT_ST_H		HDMI_TG_BASE(0x004C)
+#define HDMI_TG_VACT_SZ_L		HDMI_TG_BASE(0x0050)
+#define HDMI_TG_VACT_SZ_H		HDMI_TG_BASE(0x0054)
+#define HDMI_TG_FIELD_CHG_L		HDMI_TG_BASE(0x0058)
+#define HDMI_TG_FIELD_CHG_H		HDMI_TG_BASE(0x005C)
+#define HDMI_TG_VACT_ST2_L		HDMI_TG_BASE(0x0060)
+#define HDMI_TG_VACT_ST2_H		HDMI_TG_BASE(0x0064)
+#define HDMI_TG_VSYNC_TOP_HDMI_L	HDMI_TG_BASE(0x0078)
+#define HDMI_TG_VSYNC_TOP_HDMI_H	HDMI_TG_BASE(0x007C)
+#define HDMI_TG_VSYNC_BOT_HDMI_L	HDMI_TG_BASE(0x0080)
+#define HDMI_TG_VSYNC_BOT_HDMI_H	HDMI_TG_BASE(0x0084)
+#define HDMI_TG_FIELD_TOP_HDMI_L	HDMI_TG_BASE(0x0088)
+#define HDMI_TG_FIELD_TOP_HDMI_H	HDMI_TG_BASE(0x008C)
+#define HDMI_TG_FIELD_BOT_HDMI_L	HDMI_TG_BASE(0x0090)
+#define HDMI_TG_FIELD_BOT_HDMI_H	HDMI_TG_BASE(0x0094)
+
+/*
+ * Bit definition part
+ */
+
+/* HDMI_INTC_CON */
+#define HDMI_INTC_EN_GLOBAL		(1 << 6)
+#define HDMI_INTC_EN_HPD_PLUG		(1 << 3)
+#define HDMI_INTC_EN_HPD_UNPLUG		(1 << 2)
+
+/* HDMI_INTC_FLAG */
+#define HDMI_INTC_FLAG_HPD_PLUG		(1 << 3)
+#define HDMI_INTC_FLAG_HPD_UNPLUG	(1 << 2)
+
+/* HDMI_PHY_RSTOUT */
+#define HDMI_PHY_SW_RSTOUT		(1 << 0)
+
+/* HDMI_CORE_RSTOUT */
+#define HDMI_CORE_SW_RSTOUT		(1 << 0)
+
+/* HDMI_CON_0 */
+#define HDMI_BLUE_SCR_EN		(1 << 5)
+#define HDMI_EN				(1 << 0)
+
+/* HDMI_PHY_STATUS */
+#define HDMI_PHY_STATUS_READY		(1 << 0)
+
+/* HDMI_MODE_SEL */
+#define HDMI_MODE_HDMI_EN		(1 << 1)
+#define HDMI_MODE_DVI_EN		(1 << 0)
+#define HDMI_MODE_MASK			(3 << 0)
+
+/* HDMI_TG_CMD */
+#define HDMI_TG_EN			(1 << 0)
+#define HDMI_FIELD_EN			(1 << 1)
+
+#endif /* SAMSUNG_REGS_HDMI_H */
diff --git a/drivers/gpu/drm/exynos/regs-mixer.h b/drivers/gpu/drm/exynos/regs-mixer.h
new file mode 100644
index 0000000..fd2f4d1
--- /dev/null
+++ b/drivers/gpu/drm/exynos/regs-mixer.h
@@ -0,0 +1,141 @@
+/*
+ *
+ *  Cloned from drivers/media/video/s5p-tv/regs-mixer.h
+ *
+ * Copyright (c) 2010-2011 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com/
+ *
+ * Mixer register header file for Samsung Mixer driver
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+#ifndef SAMSUNG_REGS_MIXER_H
+#define SAMSUNG_REGS_MIXER_H
+
+/*
+ * Register part
+ */
+#define MXR_STATUS			0x0000
+#define MXR_CFG				0x0004
+#define MXR_INT_EN			0x0008
+#define MXR_INT_STATUS			0x000C
+#define MXR_LAYER_CFG			0x0010
+#define MXR_VIDEO_CFG			0x0014
+#define MXR_GRAPHIC0_CFG		0x0020
+#define MXR_GRAPHIC0_BASE		0x0024
+#define MXR_GRAPHIC0_SPAN		0x0028
+#define MXR_GRAPHIC0_SXY		0x002C
+#define MXR_GRAPHIC0_WH			0x0030
+#define MXR_GRAPHIC0_DXY		0x0034
+#define MXR_GRAPHIC0_BLANK		0x0038
+#define MXR_GRAPHIC1_CFG		0x0040
+#define MXR_GRAPHIC1_BASE		0x0044
+#define MXR_GRAPHIC1_SPAN		0x0048
+#define MXR_GRAPHIC1_SXY		0x004C
+#define MXR_GRAPHIC1_WH			0x0050
+#define MXR_GRAPHIC1_DXY		0x0054
+#define MXR_GRAPHIC1_BLANK		0x0058
+#define MXR_BG_CFG			0x0060
+#define MXR_BG_COLOR0			0x0064
+#define MXR_BG_COLOR1			0x0068
+#define MXR_BG_COLOR2			0x006C
+#define MXR_CM_COEFF_Y			0x0080
+#define MXR_CM_COEFF_CB			0x0084
+#define MXR_CM_COEFF_CR			0x0088
+#define MXR_GRAPHIC0_BASE_S		0x2024
+#define MXR_GRAPHIC1_BASE_S		0x2044
+
+/* for parametrized access to layer registers */
+#define MXR_GRAPHIC_CFG(i)		(0x0020 + (i) * 0x20)
+#define MXR_GRAPHIC_BASE(i)		(0x0024 + (i) * 0x20)
+#define MXR_GRAPHIC_SPAN(i)		(0x0028 + (i) * 0x20)
+#define MXR_GRAPHIC_SXY(i)		(0x002C + (i) * 0x20)
+#define MXR_GRAPHIC_WH(i)		(0x0030 + (i) * 0x20)
+#define MXR_GRAPHIC_DXY(i)		(0x0034 + (i) * 0x20)
+#define MXR_GRAPHIC_BLANK(i)		(0x0038 + (i) * 0x20)
+#define MXR_GRAPHIC_BASE_S(i)		(0x2024 + (i) * 0x20)
+
+/*
+ * Bit definition part
+ */
+
+/* generates mask for range of bits */
+#define MXR_MASK(high_bit, low_bit) \
+	(((2 << ((high_bit) - (low_bit))) - 1) << (low_bit))
+
+#define MXR_MASK_VAL(val, high_bit, low_bit) \
+	(((val) << (low_bit)) & MXR_MASK(high_bit, low_bit))
+
+/* bits for MXR_STATUS */
+#define MXR_STATUS_16_BURST		(1 << 7)
+#define MXR_STATUS_BURST_MASK		(1 << 7)
+#define MXR_STATUS_BIG_ENDIAN		(1 << 3)
+#define MXR_STATUS_ENDIAN_MASK		(1 << 3)
+#define MXR_STATUS_SYNC_ENABLE		(1 << 2)
+#define MXR_STATUS_REG_RUN		(1 << 0)
+
+/* bits for MXR_CFG */
+#define MXR_CFG_RGB601_0_255		(0 << 9)
+#define MXR_CFG_RGB601_16_235		(1 << 9)
+#define MXR_CFG_RGB709_0_255		(2 << 9)
+#define MXR_CFG_RGB709_16_235		(3 << 9)
+#define MXR_CFG_RGB_FMT_MASK		0x600
+#define MXR_CFG_OUT_YUV444		(0 << 8)
+#define MXR_CFG_OUT_RGB888		(1 << 8)
+#define MXR_CFG_OUT_MASK		(1 << 8)
+#define MXR_CFG_DST_SDO			(0 << 7)
+#define MXR_CFG_DST_HDMI		(1 << 7)
+#define MXR_CFG_DST_MASK		(1 << 7)
+#define MXR_CFG_SCAN_HD_720		(0 << 6)
+#define MXR_CFG_SCAN_HD_1080		(1 << 6)
+#define MXR_CFG_GRP1_ENABLE		(1 << 5)
+#define MXR_CFG_GRP0_ENABLE		(1 << 4)
+#define MXR_CFG_VP_ENABLE		(1 << 3)
+#define MXR_CFG_SCAN_INTERLACE		(0 << 2)
+#define MXR_CFG_SCAN_PROGRASSIVE	(1 << 2)
+#define MXR_CFG_SCAN_NTSC		(0 << 1)
+#define MXR_CFG_SCAN_PAL		(1 << 1)
+#define MXR_CFG_SCAN_SD			(0 << 0)
+#define MXR_CFG_SCAN_HD			(1 << 0)
+#define MXR_CFG_SCAN_MASK		0x47
+
+/* bits for MXR_GRAPHICn_CFG */
+#define MXR_GRP_CFG_COLOR_KEY_DISABLE	(1 << 21)
+#define MXR_GRP_CFG_BLEND_PRE_MUL	(1 << 20)
+#define MXR_GRP_CFG_WIN_BLEND_EN	(1 << 17)
+#define MXR_GRP_CFG_PIXEL_BLEND_EN	(1 << 16)
+#define MXR_GRP_CFG_FORMAT_VAL(x)	MXR_MASK_VAL(x, 11, 8)
+#define MXR_GRP_CFG_FORMAT_MASK		MXR_GRP_CFG_FORMAT_VAL(~0)
+#define MXR_GRP_CFG_ALPHA_VAL(x)	MXR_MASK_VAL(x, 7, 0)
+
+/* bits for MXR_GRAPHICn_WH */
+#define MXR_GRP_WH_H_SCALE(x)		MXR_MASK_VAL(x, 28, 28)
+#define MXR_GRP_WH_V_SCALE(x)		MXR_MASK_VAL(x, 12, 12)
+#define MXR_GRP_WH_WIDTH(x)		MXR_MASK_VAL(x, 26, 16)
+#define MXR_GRP_WH_HEIGHT(x)		MXR_MASK_VAL(x, 10, 0)
+
+/* bits for MXR_GRAPHICn_SXY */
+#define MXR_GRP_SXY_SX(x)		MXR_MASK_VAL(x, 26, 16)
+#define MXR_GRP_SXY_SY(x)		MXR_MASK_VAL(x, 10, 0)
+
+/* bits for MXR_GRAPHICn_DXY */
+#define MXR_GRP_DXY_DX(x)		MXR_MASK_VAL(x, 26, 16)
+#define MXR_GRP_DXY_DY(x)		MXR_MASK_VAL(x, 10, 0)
+
+/* bits for MXR_INT_EN */
+#define MXR_INT_EN_VSYNC		(1 << 11)
+#define MXR_INT_EN_ALL			(0x0f << 8)
+
+/* bit for MXR_INT_STATUS */
+#define MXR_INT_CLEAR_VSYNC		(1 << 11)
+#define MXR_INT_STATUS_VSYNC		(1 << 0)
+
+/* bit for MXR_LAYER_CFG */
+#define MXR_LAYER_CFG_GRP1_VAL(x)	MXR_MASK_VAL(x, 11, 8)
+#define MXR_LAYER_CFG_GRP0_VAL(x)	MXR_MASK_VAL(x, 7, 4)
+#define MXR_LAYER_CFG_VP_VAL(x)		MXR_MASK_VAL(x, 3, 0)
+
+#endif /* SAMSUNG_REGS_MIXER_H */
+
diff --git a/drivers/gpu/drm/exynos/regs-vp.h b/drivers/gpu/drm/exynos/regs-vp.h
new file mode 100644
index 0000000..10b737a
--- /dev/null
+++ b/drivers/gpu/drm/exynos/regs-vp.h
@@ -0,0 +1,91 @@
+/*
+ *
+ *  Cloned from drivers/media/video/s5p-tv/regs-vp.h
+ *
+ * Copyright (c) 2010-2011 Samsung Electronics Co., Ltd.
+ *		http://www.samsung.com/
+ *
+ * Video processor register header file for Samsung Mixer driver
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef SAMSUNG_REGS_VP_H
+#define SAMSUNG_REGS_VP_H
+
+/*
+ * Register part
+ */
+
+#define VP_ENABLE			0x0000
+#define VP_SRESET			0x0004
+#define VP_SHADOW_UPDATE		0x0008
+#define VP_FIELD_ID			0x000C
+#define VP_MODE				0x0010
+#define VP_IMG_SIZE_Y			0x0014
+#define VP_IMG_SIZE_C			0x0018
+#define VP_PER_RATE_CTRL		0x001C
+#define VP_TOP_Y_PTR			0x0028
+#define VP_BOT_Y_PTR			0x002C
+#define VP_TOP_C_PTR			0x0030
+#define VP_BOT_C_PTR			0x0034
+#define VP_ENDIAN_MODE			0x03CC
+#define VP_SRC_H_POSITION		0x0044
+#define VP_SRC_V_POSITION		0x0048
+#define VP_SRC_WIDTH			0x004C
+#define VP_SRC_HEIGHT			0x0050
+#define VP_DST_H_POSITION		0x0054
+#define VP_DST_V_POSITION		0x0058
+#define VP_DST_WIDTH			0x005C
+#define VP_DST_HEIGHT			0x0060
+#define VP_H_RATIO			0x0064
+#define VP_V_RATIO			0x0068
+#define VP_POLY8_Y0_LL			0x006C
+#define VP_POLY4_Y0_LL			0x00EC
+#define VP_POLY4_C0_LL			0x012C
+
+/*
+ * Bit definition part
+ */
+
+/* generates mask for range of bits */
+
+#define VP_MASK(high_bit, low_bit) \
+	(((2 << ((high_bit) - (low_bit))) - 1) << (low_bit))
+
+#define VP_MASK_VAL(val, high_bit, low_bit) \
+	(((val) << (low_bit)) & VP_MASK(high_bit, low_bit))
+
+ /* VP_ENABLE */
+#define VP_ENABLE_ON			(1 << 0)
+
+/* VP_SRESET */
+#define VP_SRESET_PROCESSING		(1 << 0)
+
+/* VP_SHADOW_UPDATE */
+#define VP_SHADOW_UPDATE_ENABLE		(1 << 0)
+
+/* VP_MODE */
+#define VP_MODE_NV12			(0 << 6)
+#define VP_MODE_NV21			(1 << 6)
+#define VP_MODE_LINE_SKIP		(1 << 5)
+#define VP_MODE_MEM_LINEAR		(0 << 4)
+#define VP_MODE_MEM_TILED		(1 << 4)
+#define VP_MODE_FMT_MASK		(5 << 4)
+#define VP_MODE_FIELD_ID_AUTO_TOGGLING	(1 << 2)
+#define VP_MODE_2D_IPC			(1 << 1)
+
+/* VP_IMG_SIZE_Y */
+/* VP_IMG_SIZE_C */
+#define VP_IMG_HSIZE(x)			VP_MASK_VAL(x, 29, 16)
+#define VP_IMG_VSIZE(x)			VP_MASK_VAL(x, 13, 0)
+
+/* VP_SRC_H_POSITION */
+#define VP_SRC_H_POSITION_VAL(x)	VP_MASK_VAL(x, 14, 4)
+
+/* VP_ENDIAN_MODE */
+#define VP_ENDIAN_MODE_LITTLE		(1 << 0)
+
+#endif /* SAMSUNG_REGS_VP_H */
diff --git a/drivers/gpu/drm/gma500/Kconfig b/drivers/gpu/drm/gma500/Kconfig
index 78e46e8..754e14b 100644
--- a/drivers/gpu/drm/gma500/Kconfig
+++ b/drivers/gpu/drm/gma500/Kconfig
@@ -19,8 +19,9 @@
 	  platforms with LVDS ports. HDMI and MIPI are not currently
 	  supported.
 
-config DRM_CEDARVIEW
-	bool "Intel Cedarview support (Experimental)"
+config DRM_GMA3600
+	bool "Intel GMA3600/3650 support (Experimental)"
 	depends on DRM_GMA500
 	help
-	  Say yes to include support for Intel Cedarview platforms
+	  Say yes to include basic support for Intel GMA3600/3650 (Intel
+	  Cedar Trail) platforms.
diff --git a/drivers/gpu/drm/gma500/Makefile b/drivers/gpu/drm/gma500/Makefile
index 613c74f..81c103be 100644
--- a/drivers/gpu/drm/gma500/Makefile
+++ b/drivers/gpu/drm/gma500/Makefile
@@ -11,6 +11,7 @@
 	  gtt.o \
 	  intel_bios.o \
 	  intel_i2c.o \
+	  intel_gmbus.o \
 	  intel_opregion.o \
 	  mmu.o \
 	  power.o \
@@ -24,7 +25,7 @@
 	  psb_device.o \
 	  mid_bios.o
 
-gma500_gfx-$(CONFIG_DRM_CEDARVIEW) +=  cdv_device.o \
+gma500_gfx-$(CONFIG_DRM_GMA3600) +=  cdv_device.o \
 	  cdv_intel_crt.o \
 	  cdv_intel_display.o \
 	  cdv_intel_hdmi.o \
diff --git a/drivers/gpu/drm/gma500/accel_2d.c b/drivers/gpu/drm/gma500/accel_2d.c
index f0ce82a..d5ef1a5 100644
--- a/drivers/gpu/drm/gma500/accel_2d.c
+++ b/drivers/gpu/drm/gma500/accel_2d.c
@@ -253,7 +253,7 @@
 		return;
 
 	offset = psbfb->gtt->offset;
-	stride = fb->pitch;
+	stride = fb->pitches[0];
 
 	switch (fb->depth) {
 	case 8:
diff --git a/drivers/gpu/drm/gma500/cdv_device.c b/drivers/gpu/drm/gma500/cdv_device.c
index 7e8028a..4a5b099 100644
--- a/drivers/gpu/drm/gma500/cdv_device.c
+++ b/drivers/gpu/drm/gma500/cdv_device.c
@@ -327,9 +327,10 @@
 /* CDV is much like Poulsbo but has MID like SGX offsets and PM */
 
 const struct psb_ops cdv_chip_ops = {
-	.name = "Cedartrail",
+	.name = "GMA3600/3650",
 	.accel_2d = 0,
 	.pipes = 2,
+	.crtcs = 2,
 	.sgx_offset = MRST_SGX_OFFSET,
 	.chip_setup = cdv_chip_setup,
 
diff --git a/drivers/gpu/drm/gma500/cdv_intel_crt.c b/drivers/gpu/drm/gma500/cdv_intel_crt.c
index efda63b..6d0f10b 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_crt.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_crt.c
@@ -204,9 +204,10 @@
 
 static void cdv_intel_crt_destroy(struct drm_connector *connector)
 {
-	struct psb_intel_output *intel_output = to_psb_intel_output(connector);
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 
-	psb_intel_i2c_destroy(intel_output->ddc_bus);
+	psb_intel_i2c_destroy(psb_intel_encoder->ddc_bus);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
@@ -214,9 +215,9 @@
 
 static int cdv_intel_crt_get_modes(struct drm_connector *connector)
 {
-	struct psb_intel_output *intel_output =
-				to_psb_intel_output(connector);
-	return psb_intel_ddc_get_modes(intel_output);
+	struct psb_intel_encoder *psb_intel_encoder =
+				psb_intel_attached_encoder(connector);
+	return psb_intel_ddc_get_modes(connector, &psb_intel_encoder->ddc_bus->adapter);
 }
 
 static int cdv_intel_crt_set_property(struct drm_connector *connector,
@@ -266,27 +267,31 @@
 			struct psb_intel_mode_device *mode_dev)
 {
 
-	struct psb_intel_output *psb_intel_output;
+	struct psb_intel_connector *psb_intel_connector;
+	struct psb_intel_encoder *psb_intel_encoder;
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
 
 	u32 i2c_reg;
 
-	psb_intel_output = kzalloc(sizeof(struct psb_intel_output), GFP_KERNEL);
-	if (!psb_intel_output)
+	psb_intel_encoder = kzalloc(sizeof(struct psb_intel_encoder), GFP_KERNEL);
+	if (!psb_intel_encoder)
 		return;
 
-	psb_intel_output->mode_dev = mode_dev;
-	connector = &psb_intel_output->base;
+	psb_intel_connector = kzalloc(sizeof(struct psb_intel_connector), GFP_KERNEL);
+	if (!psb_intel_connector)
+		goto failed_connector;
+
+	connector = &psb_intel_connector->base;
 	drm_connector_init(dev, connector,
 		&cdv_intel_crt_connector_funcs, DRM_MODE_CONNECTOR_VGA);
 
-	encoder = &psb_intel_output->enc;
+	encoder = &psb_intel_encoder->base;
 	drm_encoder_init(dev, encoder,
 		&cdv_intel_crt_enc_funcs, DRM_MODE_ENCODER_DAC);
 
-	drm_mode_connector_attach_encoder(&psb_intel_output->base,
-					  &psb_intel_output->enc);
+	psb_intel_connector_attach_encoder(psb_intel_connector,
+					   psb_intel_encoder);
 
 	/* Set up the DDC bus. */
 	i2c_reg = GPIOA;
@@ -295,15 +300,15 @@
 	if (dev_priv->crt_ddc_bus != 0)
 		i2c_reg = dev_priv->crt_ddc_bus;
 	}*/
-	psb_intel_output->ddc_bus = psb_intel_i2c_create(dev,
-						i2c_reg, "CRTDDC_A");
-	if (!psb_intel_output->ddc_bus) {
+	psb_intel_encoder->ddc_bus = psb_intel_i2c_create(dev,
+							  i2c_reg, "CRTDDC_A");
+	if (!psb_intel_encoder->ddc_bus) {
 		dev_printk(KERN_ERR, &dev->pdev->dev, "DDC bus registration "
 			   "failed.\n");
 		goto failed_ddc;
 	}
 
-	psb_intel_output->type = INTEL_OUTPUT_ANALOG;
+	psb_intel_encoder->type = INTEL_OUTPUT_ANALOG;
 	/*
 	psb_intel_output->clone_mask = (1 << INTEL_ANALOG_CLONE_BIT);
 	psb_intel_output->crtc_mask = (1 << 0) | (1 << 1);
@@ -319,8 +324,10 @@
 
 	return;
 failed_ddc:
-	drm_encoder_cleanup(&psb_intel_output->enc);
-	drm_connector_cleanup(&psb_intel_output->base);
-	kfree(psb_intel_output);
+	drm_encoder_cleanup(&psb_intel_encoder->base);
+	drm_connector_cleanup(&psb_intel_connector->base);
+	kfree(psb_intel_connector);
+failed_connector:
+	kfree(psb_intel_encoder);
 	return;
 }
diff --git a/drivers/gpu/drm/gma500/cdv_intel_display.c b/drivers/gpu/drm/gma500/cdv_intel_display.c
index 7b97c60..18d1152 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_display.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_display.c
@@ -342,7 +342,7 @@
 }
 
 /*
- * Returns whether any output on the specified pipe is of the specified type
+ * Returns whether any encoder on the specified pipe is of the specified type
  */
 bool cdv_intel_pipe_has_type(struct drm_crtc *crtc, int type)
 {
@@ -352,9 +352,9 @@
 
 	list_for_each_entry(l_entry, &mode_config->connector_list, head) {
 		if (l_entry->encoder && l_entry->encoder->crtc == crtc) {
-			struct psb_intel_output *psb_intel_output =
-			    to_psb_intel_output(l_entry);
-			if (psb_intel_output->type == type)
+			struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(l_entry);
+			if (psb_intel_encoder->type == type)
 				return true;
 		}
 	}
@@ -507,9 +507,9 @@
 	if (ret < 0)
 		goto psb_intel_pipe_set_base_exit;
 	start = psbfb->gtt->offset;
-	offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8);
+	offset = y * crtc->fb->pitches[0] + x * (crtc->fb->bits_per_pixel / 8);
 
-	REG_WRITE(dspstride, crtc->fb->pitch);
+	REG_WRITE(dspstride, crtc->fb->pitches[0]);
 
 	dspcntr = REG_READ(dspcntr_reg);
 	dspcntr &= ~DISPPLANE_PIXFORMAT_MASK;
@@ -752,14 +752,14 @@
 	struct drm_connector *connector;
 
 	list_for_each_entry(connector, &mode_config->connector_list, head) {
-		struct psb_intel_output *psb_intel_output =
-		    to_psb_intel_output(connector);
+		struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 
 		if (!connector->encoder
 		    || connector->encoder->crtc != crtc)
 			continue;
 
-		switch (psb_intel_output->type) {
+		switch (psb_intel_encoder->type) {
 		case INTEL_OUTPUT_LVDS:
 			is_lvds = true;
 			break;
diff --git a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c
index cbca2b0..50d7cfb 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c
@@ -63,8 +63,8 @@
 			struct drm_display_mode *adjusted_mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
-	struct mid_intel_hdmi_priv *hdmi_priv = output->dev_priv;
+	struct psb_intel_encoder *psb_intel_encoder = to_psb_intel_encoder(encoder);
+	struct mid_intel_hdmi_priv *hdmi_priv = psb_intel_encoder->dev_priv;
 	u32 hdmib;
 	struct drm_crtc *crtc = encoder->crtc;
 	struct psb_intel_crtc *intel_crtc = to_psb_intel_crtc(crtc);
@@ -98,8 +98,9 @@
 static void cdv_hdmi_dpms(struct drm_encoder *encoder, int mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
-	struct mid_intel_hdmi_priv *hdmi_priv = output->dev_priv;
+	struct psb_intel_encoder *psb_intel_encoder =
+						to_psb_intel_encoder(encoder);
+	struct mid_intel_hdmi_priv *hdmi_priv = psb_intel_encoder->dev_priv;
 	u32 hdmib;
 
 	hdmib = REG_READ(hdmi_priv->hdmi_reg);
@@ -114,8 +115,9 @@
 static void cdv_hdmi_save(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct psb_intel_output *output = to_psb_intel_output(connector);
-	struct mid_intel_hdmi_priv *hdmi_priv = output->dev_priv;
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
+	struct mid_intel_hdmi_priv *hdmi_priv = psb_intel_encoder->dev_priv;
 
 	hdmi_priv->save_HDMIB = REG_READ(hdmi_priv->hdmi_reg);
 }
@@ -123,8 +125,9 @@
 static void cdv_hdmi_restore(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct psb_intel_output *output = to_psb_intel_output(connector);
-	struct mid_intel_hdmi_priv *hdmi_priv = output->dev_priv;
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
+	struct mid_intel_hdmi_priv *hdmi_priv = psb_intel_encoder->dev_priv;
 
 	REG_WRITE(hdmi_priv->hdmi_reg, hdmi_priv->save_HDMIB);
 	REG_READ(hdmi_priv->hdmi_reg);
@@ -133,14 +136,15 @@
 static enum drm_connector_status cdv_hdmi_detect(
 				struct drm_connector *connector, bool force)
 {
-	struct psb_intel_output *psb_intel_output =
-						to_psb_intel_output(connector);
-	struct mid_intel_hdmi_priv *hdmi_priv = psb_intel_output->dev_priv;
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
+	struct psb_intel_connector *psb_intel_connector =
+					to_psb_intel_connector(connector);
+	struct mid_intel_hdmi_priv *hdmi_priv = psb_intel_encoder->dev_priv;
 	struct edid *edid = NULL;
 	enum drm_connector_status status = connector_status_disconnected;
 
-	edid = drm_get_edid(&psb_intel_output->base,
-			 psb_intel_output->hdmi_i2c_adapter);
+	edid = drm_get_edid(connector, &psb_intel_encoder->i2c_bus->adapter);
 
 	hdmi_priv->has_hdmi_sink = false;
 	hdmi_priv->has_hdmi_audio = false;
@@ -153,7 +157,7 @@
 						drm_detect_monitor_audio(edid);
 		}
 
-		psb_intel_output->base.display_info.raw_edid = NULL;
+		psb_intel_connector->base.display_info.raw_edid = NULL;
 		kfree(edid);
 	}
 	return status;
@@ -220,17 +224,15 @@
  */
 static int cdv_hdmi_get_modes(struct drm_connector *connector)
 {
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 	struct edid *edid = NULL;
 	int ret = 0;
 
-	edid = drm_get_edid(&psb_intel_output->base,
-			 psb_intel_output->hdmi_i2c_adapter);
+	edid = drm_get_edid(connector, &psb_intel_encoder->i2c_bus->adapter);
 	if (edid) {
-		drm_mode_connector_update_edid_property(&psb_intel_output->
-							base, edid);
-		ret = drm_add_edid_modes(&psb_intel_output->base, edid);
+		drm_mode_connector_update_edid_property(connector, edid);
+		ret = drm_add_edid_modes(connector, edid);
 		kfree(edid);
 	}
 	return ret;
@@ -266,11 +268,11 @@
 
 static void cdv_hdmi_destroy(struct drm_connector *connector)
 {
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 
-	if (psb_intel_output->ddc_bus)
-		psb_intel_i2c_destroy(psb_intel_output->ddc_bus);
+	if (psb_intel_encoder->i2c_bus)
+		psb_intel_i2c_destroy(psb_intel_encoder->i2c_bus);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
@@ -304,34 +306,45 @@
 void cdv_hdmi_init(struct drm_device *dev,
 			struct psb_intel_mode_device *mode_dev, int reg)
 {
-	struct psb_intel_output *psb_intel_output;
+	struct psb_intel_encoder *psb_intel_encoder;
+	struct psb_intel_connector *psb_intel_connector;
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
 	struct mid_intel_hdmi_priv *hdmi_priv;
 	int ddc_bus;
 
-	psb_intel_output = kzalloc(sizeof(struct psb_intel_output) +
-			       sizeof(struct mid_intel_hdmi_priv), GFP_KERNEL);
-	if (!psb_intel_output)
+	psb_intel_encoder = kzalloc(sizeof(struct psb_intel_encoder),
+				    GFP_KERNEL);
+
+	if (!psb_intel_encoder)
 		return;
 
-	hdmi_priv = (struct mid_intel_hdmi_priv *)(psb_intel_output + 1);
-	psb_intel_output->mode_dev = mode_dev;
-	connector = &psb_intel_output->base;
-	encoder = &psb_intel_output->enc;
-	drm_connector_init(dev, &psb_intel_output->base,
+	psb_intel_connector = kzalloc(sizeof(struct psb_intel_connector),
+				      GFP_KERNEL);
+
+	if (!psb_intel_connector)
+		goto err_connector;
+
+	hdmi_priv = kzalloc(sizeof(struct mid_intel_hdmi_priv), GFP_KERNEL);
+
+	if (!hdmi_priv)
+		goto err_priv;
+
+	connector = &psb_intel_connector->base;
+	encoder = &psb_intel_encoder->base;
+	drm_connector_init(dev, connector,
 			   &cdv_hdmi_connector_funcs,
 			   DRM_MODE_CONNECTOR_DVID);
 
-	drm_encoder_init(dev, &psb_intel_output->enc, &psb_intel_lvds_enc_funcs,
+	drm_encoder_init(dev, encoder, &psb_intel_lvds_enc_funcs,
 			 DRM_MODE_ENCODER_TMDS);
 
-	drm_mode_connector_attach_encoder(&psb_intel_output->base,
-					  &psb_intel_output->enc);
-	psb_intel_output->type = INTEL_OUTPUT_HDMI;
+	psb_intel_connector_attach_encoder(psb_intel_connector,
+					   psb_intel_encoder);
+	psb_intel_encoder->type = INTEL_OUTPUT_HDMI;
 	hdmi_priv->hdmi_reg = reg;
 	hdmi_priv->has_hdmi_sink = false;
-	psb_intel_output->dev_priv = hdmi_priv;
+	psb_intel_encoder->dev_priv = hdmi_priv;
 
 	drm_encoder_helper_add(encoder, &cdv_hdmi_helper_funcs);
 	drm_connector_helper_add(connector,
@@ -341,7 +354,8 @@
 	connector->doublescan_allowed = false;
 
 	drm_connector_attach_property(connector,
-	    dev->mode_config.scaling_mode_property, DRM_MODE_SCALE_FULLSCREEN);
+				      dev->mode_config.scaling_mode_property,
+				      DRM_MODE_SCALE_FULLSCREEN);
 
 	switch (reg) {
 	case SDVOB:
@@ -356,21 +370,25 @@
 		break;
 	}
 
-	psb_intel_output->ddc_bus = psb_intel_i2c_create(dev,
+	psb_intel_encoder->i2c_bus = psb_intel_i2c_create(dev,
 				ddc_bus, (reg == SDVOB) ? "HDMIB" : "HDMIC");
 
-	if (!psb_intel_output->ddc_bus) {
+	if (!psb_intel_encoder->i2c_bus) {
 		dev_err(dev->dev, "No ddc adapter available!\n");
 		goto failed_ddc;
 	}
-	psb_intel_output->hdmi_i2c_adapter =
-				&(psb_intel_output->ddc_bus->adapter);
+
+	hdmi_priv->hdmi_i2c_adapter =
+				&(psb_intel_encoder->i2c_bus->adapter);
 	hdmi_priv->dev = dev;
 	drm_sysfs_connector_add(connector);
 	return;
 
 failed_ddc:
-	drm_encoder_cleanup(&psb_intel_output->enc);
-	drm_connector_cleanup(&psb_intel_output->base);
-	kfree(psb_intel_output);
+	drm_encoder_cleanup(encoder);
+	drm_connector_cleanup(connector);
+err_priv:
+	kfree(psb_intel_connector);
+err_connector:
+	kfree(psb_intel_encoder);
 }
diff --git a/drivers/gpu/drm/gma500/cdv_intel_lvds.c b/drivers/gpu/drm/gma500/cdv_intel_lvds.c
index 988b2d0..50e744b 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_lvds.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_lvds.c
@@ -195,8 +195,9 @@
  * Sets the power state for the panel.
  */
 static void cdv_intel_lvds_set_power(struct drm_device *dev,
-				 struct psb_intel_output *output, bool on)
+				     struct drm_encoder *encoder, bool on)
 {
+	struct drm_psb_private *dev_priv = dev->dev_private;
 	u32 pp_status;
 
 	if (!gma_power_begin(dev, true))
@@ -210,8 +211,7 @@
 		} while ((pp_status & PP_ON) == 0);
 
 		cdv_intel_lvds_set_backlight(dev,
-					 output->
-					 mode_dev->backlight_duty_cycle);
+				dev_priv->mode_dev.backlight_duty_cycle);
 	} else {
 		cdv_intel_lvds_set_backlight(dev, 0);
 
@@ -227,11 +227,10 @@
 static void cdv_intel_lvds_encoder_dpms(struct drm_encoder *encoder, int mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
 	if (mode == DRM_MODE_DPMS_ON)
-		cdv_intel_lvds_set_power(dev, output, true);
+		cdv_intel_lvds_set_power(dev, encoder, true);
 	else
-		cdv_intel_lvds_set_power(dev, output, false);
+		cdv_intel_lvds_set_power(dev, encoder, false);
 	/* XXX: We never power down the LVDS pairs. */
 }
 
@@ -244,12 +243,12 @@
 }
 
 int cdv_intel_lvds_mode_valid(struct drm_connector *connector,
-				 struct drm_display_mode *mode)
+			      struct drm_display_mode *mode)
 {
-	struct psb_intel_output *psb_intel_output =
-				to_psb_intel_output(connector);
+	struct drm_device *dev = connector->dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
 	struct drm_display_mode *fixed_mode =
-	    psb_intel_output->mode_dev->panel_fixed_mode;
+					dev_priv->mode_dev.panel_fixed_mode;
 
 	/* just in case */
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
@@ -272,9 +271,9 @@
 				  struct drm_display_mode *mode,
 				  struct drm_display_mode *adjusted_mode)
 {
-	struct psb_intel_mode_device *mode_dev =
-	    enc_to_psb_intel_output(encoder)->mode_dev;
 	struct drm_device *dev = encoder->dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 	struct drm_encoder *tmp_encoder;
 	struct drm_display_mode *panel_fixed_mode = mode_dev->panel_fixed_mode;
 
@@ -321,8 +320,8 @@
 static void cdv_intel_lvds_prepare(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
-	struct psb_intel_mode_device *mode_dev = output->mode_dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 
 	if (!gma_power_begin(dev, true))
 		return;
@@ -331,7 +330,7 @@
 	mode_dev->backlight_duty_cycle = (mode_dev->saveBLC_PWM_CTL &
 					  BACKLIGHT_DUTY_CYCLE_MASK);
 
-	cdv_intel_lvds_set_power(dev, output, false);
+	cdv_intel_lvds_set_power(dev, encoder, false);
 
 	gma_power_end(dev);
 }
@@ -339,14 +338,14 @@
 static void cdv_intel_lvds_commit(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
-	struct psb_intel_mode_device *mode_dev = output->mode_dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 
 	if (mode_dev->backlight_duty_cycle == 0)
 		mode_dev->backlight_duty_cycle =
 		    cdv_intel_lvds_get_max_backlight(dev);
 
-	cdv_intel_lvds_set_power(dev, output, true);
+	cdv_intel_lvds_set_power(dev, encoder, true);
 }
 
 static void cdv_intel_lvds_mode_set(struct drm_encoder *encoder,
@@ -401,13 +400,13 @@
 static int cdv_intel_lvds_get_modes(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
-	struct psb_intel_mode_device *mode_dev =
-					psb_intel_output->mode_dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 	int ret;
 
-	ret = psb_intel_ddc_get_modes(psb_intel_output);
+	ret = psb_intel_ddc_get_modes(connector, &psb_intel_encoder->i2c_bus->adapter);
 
 	if (ret)
 		return ret;
@@ -439,11 +438,11 @@
  */
 void cdv_intel_lvds_destroy(struct drm_connector *connector)
 {
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 
-	if (psb_intel_output->ddc_bus)
-		psb_intel_i2c_destroy(psb_intel_output->ddc_bus);
+	if (psb_intel_encoder->i2c_bus)
+		psb_intel_i2c_destroy(psb_intel_encoder->i2c_bus);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
@@ -565,7 +564,8 @@
 void cdv_intel_lvds_init(struct drm_device *dev,
 		     struct psb_intel_mode_device *mode_dev)
 {
-	struct psb_intel_output *psb_intel_output;
+	struct psb_intel_encoder *psb_intel_encoder;
+	struct psb_intel_connector *psb_intel_connector;
 	struct cdv_intel_lvds_priv *lvds_priv;
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
@@ -575,32 +575,38 @@
 	u32 lvds;
 	int pipe;
 
-	psb_intel_output = kzalloc(sizeof(struct psb_intel_output) +
-			sizeof(struct cdv_intel_lvds_priv), GFP_KERNEL);
-	if (!psb_intel_output)
+	psb_intel_encoder = kzalloc(sizeof(struct psb_intel_encoder),
+				    GFP_KERNEL);
+	if (!psb_intel_encoder)
 		return;
 
-	lvds_priv = (struct cdv_intel_lvds_priv *)(psb_intel_output + 1);
+	psb_intel_connector = kzalloc(sizeof(struct psb_intel_connector),
+				      GFP_KERNEL);
+	if (!psb_intel_connector)
+		goto failed_connector;
 
-	psb_intel_output->dev_priv = lvds_priv;
+	lvds_priv = kzalloc(sizeof(struct cdv_intel_lvds_priv), GFP_KERNEL);
+	if (!lvds_priv)
+		goto failed_lvds_priv;
 
-	psb_intel_output->mode_dev = mode_dev;
-	connector = &psb_intel_output->base;
-	encoder = &psb_intel_output->enc;
+	psb_intel_encoder->dev_priv = lvds_priv;
+
+	connector = &psb_intel_connector->base;
+	encoder = &psb_intel_encoder->base;
 
 
-	drm_connector_init(dev, &psb_intel_output->base,
+	drm_connector_init(dev, connector,
 			   &cdv_intel_lvds_connector_funcs,
 			   DRM_MODE_CONNECTOR_LVDS);
 
-	drm_encoder_init(dev, &psb_intel_output->enc,
+	drm_encoder_init(dev, encoder,
 			 &cdv_intel_lvds_enc_funcs,
 			 DRM_MODE_ENCODER_LVDS);
 
 
-	drm_mode_connector_attach_encoder(&psb_intel_output->base,
-					  &psb_intel_output->enc);
-	psb_intel_output->type = INTEL_OUTPUT_LVDS;
+	psb_intel_connector_attach_encoder(psb_intel_connector,
+					   psb_intel_encoder);
+	psb_intel_encoder->type = INTEL_OUTPUT_LVDS;
 
 	drm_encoder_helper_add(encoder, &cdv_intel_lvds_helper_funcs);
 	drm_connector_helper_add(connector,
@@ -621,16 +627,16 @@
 	 * Set up I2C bus
 	 * FIXME: distroy i2c_bus when exit
 	 */
-	psb_intel_output->i2c_bus = psb_intel_i2c_create(dev,
+	psb_intel_encoder->i2c_bus = psb_intel_i2c_create(dev,
 							 GPIOB,
 							 "LVDSBLC_B");
-	if (!psb_intel_output->i2c_bus) {
+	if (!psb_intel_encoder->i2c_bus) {
 		dev_printk(KERN_ERR,
 			&dev->pdev->dev, "I2C bus registration failed.\n");
 		goto failed_blc_i2c;
 	}
-	psb_intel_output->i2c_bus->slave_addr = 0x2C;
-	dev_priv->lvds_i2c_bus =  psb_intel_output->i2c_bus;
+	psb_intel_encoder->i2c_bus->slave_addr = 0x2C;
+	dev_priv->lvds_i2c_bus = psb_intel_encoder->i2c_bus;
 
 	/*
 	 * LVDS discovery:
@@ -643,10 +649,10 @@
 	 */
 
 	/* Set up the DDC bus. */
-	psb_intel_output->ddc_bus = psb_intel_i2c_create(dev,
+	psb_intel_encoder->ddc_bus = psb_intel_i2c_create(dev,
 							 GPIOC,
 							 "LVDSDDC_C");
-	if (!psb_intel_output->ddc_bus) {
+	if (!psb_intel_encoder->ddc_bus) {
 		dev_printk(KERN_ERR, &dev->pdev->dev,
 			   "DDC bus registration " "failed.\n");
 		goto failed_ddc;
@@ -656,7 +662,8 @@
 	 * Attempt to get the fixed panel mode from DDC.  Assume that the
 	 * preferred mode is the right one.
 	 */
-	psb_intel_ddc_get_modes(psb_intel_output);
+	psb_intel_ddc_get_modes(connector,
+				&psb_intel_encoder->ddc_bus->adapter);
 	list_for_each_entry(scan, &connector->probed_modes, head) {
 		if (scan->type & DRM_MODE_TYPE_PREFERRED) {
 			mode_dev->panel_fixed_mode =
@@ -707,15 +714,19 @@
 
 failed_find:
 	printk(KERN_ERR "Failed find\n");
-	if (psb_intel_output->ddc_bus)
-		psb_intel_i2c_destroy(psb_intel_output->ddc_bus);
+	if (psb_intel_encoder->ddc_bus)
+		psb_intel_i2c_destroy(psb_intel_encoder->ddc_bus);
 failed_ddc:
 	printk(KERN_ERR "Failed DDC\n");
-	if (psb_intel_output->i2c_bus)
-		psb_intel_i2c_destroy(psb_intel_output->i2c_bus);
+	if (psb_intel_encoder->i2c_bus)
+		psb_intel_i2c_destroy(psb_intel_encoder->i2c_bus);
 failed_blc_i2c:
 	printk(KERN_ERR "Failed BLC\n");
 	drm_encoder_cleanup(encoder);
 	drm_connector_cleanup(connector);
-	kfree(connector);
+	kfree(lvds_priv);
+failed_lvds_priv:
+	kfree(psb_intel_connector);
+failed_connector:
+	kfree(psb_intel_encoder);
 }
diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c
index 9ec1676..791c0ef 100644
--- a/drivers/gpu/drm/gma500/framebuffer.c
+++ b/drivers/gpu/drm/gma500/framebuffer.c
@@ -500,7 +500,7 @@
 		info->apertures->ranges[0].size = dev_priv->gtt.stolen_size;
 	}
 
-	drm_fb_helper_fill_fix(info, fb->pitch, fb->depth);
+	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->depth);
 	drm_fb_helper_fill_var(info, &fbdev->psb_fb_helper,
 				sizes->fb_width, sizes->fb_height);
 
@@ -747,13 +747,13 @@
 
 	list_for_each_entry(connector, &dev->mode_config.connector_list,
 			    head) {
-		struct psb_intel_output *psb_intel_output =
-		    to_psb_intel_output(connector);
-		struct drm_encoder *encoder = &psb_intel_output->enc;
+		struct psb_intel_encoder *psb_intel_encoder =
+			psb_intel_attached_encoder(connector);
+		struct drm_encoder *encoder = &psb_intel_encoder->base;
 		int crtc_mask = 0, clone_mask = 0;
 
 		/* valid crtcs */
-		switch (psb_intel_output->type) {
+		switch (psb_intel_encoder->type) {
 		case INTEL_OUTPUT_ANALOG:
 			crtc_mask = (1 << 0);
 			clone_mask = (1 << INTEL_OUTPUT_ANALOG);
diff --git a/drivers/gpu/drm/gma500/intel_gmbus.c b/drivers/gpu/drm/gma500/intel_gmbus.c
new file mode 100644
index 0000000..147584a
--- /dev/null
+++ b/drivers/gpu/drm/gma500/intel_gmbus.c
@@ -0,0 +1,493 @@
+/*
+ * Copyright (c) 2006 Dave Airlie <airlied@linux.ie>
+ * Copyright © 2006-2008,2010 Intel Corporation
+ *   Jesse Barnes <jesse.barnes@intel.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Authors:
+ *	Eric Anholt <eric@anholt.net>
+ *	Chris Wilson <chris@chris-wilson.co.uk>
+ */
+#include <linux/module.h>
+#include <linux/i2c.h>
+#include <linux/i2c-algo-bit.h>
+#include "drmP.h"
+#include "drm.h"
+#include "psb_intel_drv.h"
+#include "gma_drm.h"
+#include "psb_drv.h"
+#include "psb_intel_reg.h"
+
+#define _wait_for(COND, MS, W) ({ \
+	unsigned long timeout__ = jiffies + msecs_to_jiffies(MS);	\
+	int ret__ = 0;							\
+	while (! (COND)) {						\
+		if (time_after(jiffies, timeout__)) {			\
+			ret__ = -ETIMEDOUT;				\
+			break;						\
+		}							\
+		if (W && !(in_atomic() || in_dbg_master())) msleep(W);	\
+	}								\
+	ret__;								\
+})
+
+#define wait_for(COND, MS) _wait_for(COND, MS, 1)
+#define wait_for_atomic(COND, MS) _wait_for(COND, MS, 0)
+
+/* Intel GPIO access functions */
+
+#define I2C_RISEFALL_TIME 20
+
+static inline struct intel_gmbus *
+to_intel_gmbus(struct i2c_adapter *i2c)
+{
+	return container_of(i2c, struct intel_gmbus, adapter);
+}
+
+struct intel_gpio {
+	struct i2c_adapter adapter;
+	struct i2c_algo_bit_data algo;
+	struct drm_psb_private *dev_priv;
+	u32 reg;
+};
+
+void
+gma_intel_i2c_reset(struct drm_device *dev)
+{
+	REG_WRITE(GMBUS0, 0);
+}
+
+static void intel_i2c_quirk_set(struct drm_psb_private *dev_priv, bool enable)
+{
+	/* When using bit bashing for I2C, this bit needs to be set to 1 */
+	/* FIXME: We are never Pineview, right?
+
+	u32 val;
+
+	if (!IS_PINEVIEW(dev_priv->dev))
+		return;
+
+	val = REG_READ(DSPCLK_GATE_D);
+	if (enable)
+		val |= DPCUNIT_CLOCK_GATE_DISABLE;
+	else
+		val &= ~DPCUNIT_CLOCK_GATE_DISABLE;
+	REG_WRITE(DSPCLK_GATE_D, val);
+
+	return;
+	*/
+}
+
+static u32 get_reserved(struct intel_gpio *gpio)
+{
+	struct drm_psb_private *dev_priv = gpio->dev_priv;
+	struct drm_device *dev = dev_priv->dev;
+	u32 reserved = 0;
+
+	/* On most chips, these bits must be preserved in software. */
+	reserved = REG_READ(gpio->reg) &
+				     (GPIO_DATA_PULLUP_DISABLE |
+				      GPIO_CLOCK_PULLUP_DISABLE);
+
+	return reserved;
+}
+
+static int get_clock(void *data)
+{
+	struct intel_gpio *gpio = data;
+	struct drm_psb_private *dev_priv = gpio->dev_priv;
+	struct drm_device *dev = dev_priv->dev;
+	u32 reserved = get_reserved(gpio);
+	REG_WRITE(gpio->reg, reserved | GPIO_CLOCK_DIR_MASK);
+	REG_WRITE(gpio->reg, reserved);
+	return (REG_READ(gpio->reg) & GPIO_CLOCK_VAL_IN) != 0;
+}
+
+static int get_data(void *data)
+{
+	struct intel_gpio *gpio = data;
+	struct drm_psb_private *dev_priv = gpio->dev_priv;
+	struct drm_device *dev = dev_priv->dev;
+	u32 reserved = get_reserved(gpio);
+	REG_WRITE(gpio->reg, reserved | GPIO_DATA_DIR_MASK);
+	REG_WRITE(gpio->reg, reserved);
+	return (REG_READ(gpio->reg) & GPIO_DATA_VAL_IN) != 0;
+}
+
+static void set_clock(void *data, int state_high)
+{
+	struct intel_gpio *gpio = data;
+	struct drm_psb_private *dev_priv = gpio->dev_priv;
+	struct drm_device *dev = dev_priv->dev;
+	u32 reserved = get_reserved(gpio);
+	u32 clock_bits;
+
+	if (state_high)
+		clock_bits = GPIO_CLOCK_DIR_IN | GPIO_CLOCK_DIR_MASK;
+	else
+		clock_bits = GPIO_CLOCK_DIR_OUT | GPIO_CLOCK_DIR_MASK |
+			GPIO_CLOCK_VAL_MASK;
+
+	REG_WRITE(gpio->reg, reserved | clock_bits);
+	REG_READ(gpio->reg); /* Posting */
+}
+
+static void set_data(void *data, int state_high)
+{
+	struct intel_gpio *gpio = data;
+	struct drm_psb_private *dev_priv = gpio->dev_priv;
+	struct drm_device *dev = dev_priv->dev;
+	u32 reserved = get_reserved(gpio);
+	u32 data_bits;
+
+	if (state_high)
+		data_bits = GPIO_DATA_DIR_IN | GPIO_DATA_DIR_MASK;
+	else
+		data_bits = GPIO_DATA_DIR_OUT | GPIO_DATA_DIR_MASK |
+			GPIO_DATA_VAL_MASK;
+
+	REG_WRITE(gpio->reg, reserved | data_bits);
+	REG_READ(gpio->reg);
+}
+
+static struct i2c_adapter *
+intel_gpio_create(struct drm_psb_private *dev_priv, u32 pin)
+{
+	static const int map_pin_to_reg[] = {
+		0,
+		GPIOB,
+		GPIOA,
+		GPIOC,
+		GPIOD,
+		GPIOE,
+		0,
+		GPIOF,
+	};
+	struct intel_gpio *gpio;
+
+	if (pin >= ARRAY_SIZE(map_pin_to_reg) || !map_pin_to_reg[pin])
+		return NULL;
+
+	gpio = kzalloc(sizeof(struct intel_gpio), GFP_KERNEL);
+	if (gpio == NULL)
+		return NULL;
+
+	gpio->reg = map_pin_to_reg[pin];
+	gpio->dev_priv = dev_priv;
+
+	snprintf(gpio->adapter.name, sizeof(gpio->adapter.name),
+		 "gma500 GPIO%c", "?BACDE?F"[pin]);
+	gpio->adapter.owner = THIS_MODULE;
+	gpio->adapter.algo_data	= &gpio->algo;
+	gpio->adapter.dev.parent = &dev_priv->dev->pdev->dev;
+	gpio->algo.setsda = set_data;
+	gpio->algo.setscl = set_clock;
+	gpio->algo.getsda = get_data;
+	gpio->algo.getscl = get_clock;
+	gpio->algo.udelay = I2C_RISEFALL_TIME;
+	gpio->algo.timeout = usecs_to_jiffies(2200);
+	gpio->algo.data = gpio;
+
+	if (i2c_bit_add_bus(&gpio->adapter))
+		goto out_free;
+
+	return &gpio->adapter;
+
+out_free:
+	kfree(gpio);
+	return NULL;
+}
+
+static int
+intel_i2c_quirk_xfer(struct drm_psb_private *dev_priv,
+		     struct i2c_adapter *adapter,
+		     struct i2c_msg *msgs,
+		     int num)
+{
+	struct intel_gpio *gpio = container_of(adapter,
+					       struct intel_gpio,
+					       adapter);
+	int ret;
+
+	gma_intel_i2c_reset(dev_priv->dev);
+
+	intel_i2c_quirk_set(dev_priv, true);
+	set_data(gpio, 1);
+	set_clock(gpio, 1);
+	udelay(I2C_RISEFALL_TIME);
+
+	ret = adapter->algo->master_xfer(adapter, msgs, num);
+
+	set_data(gpio, 1);
+	set_clock(gpio, 1);
+	intel_i2c_quirk_set(dev_priv, false);
+
+	return ret;
+}
+
+static int
+gmbus_xfer(struct i2c_adapter *adapter,
+	   struct i2c_msg *msgs,
+	   int num)
+{
+	struct intel_gmbus *bus = container_of(adapter,
+					       struct intel_gmbus,
+					       adapter);
+	struct drm_psb_private *dev_priv = adapter->algo_data;
+	struct drm_device *dev = dev_priv->dev;
+	int i, reg_offset;
+
+	if (bus->force_bit)
+		return intel_i2c_quirk_xfer(dev_priv,
+					    bus->force_bit, msgs, num);
+
+	reg_offset = 0;
+
+	REG_WRITE(GMBUS0 + reg_offset, bus->reg0);
+
+	for (i = 0; i < num; i++) {
+		u16 len = msgs[i].len;
+		u8 *buf = msgs[i].buf;
+
+		if (msgs[i].flags & I2C_M_RD) {
+			REG_WRITE(GMBUS1 + reg_offset,
+				   GMBUS_CYCLE_WAIT | (i + 1 == num ? GMBUS_CYCLE_STOP : 0) |
+				   (len << GMBUS_BYTE_COUNT_SHIFT) |
+				   (msgs[i].addr << GMBUS_SLAVE_ADDR_SHIFT) |
+				   GMBUS_SLAVE_READ | GMBUS_SW_RDY);
+			REG_READ(GMBUS2+reg_offset);
+			do {
+				u32 val, loop = 0;
+
+				if (wait_for(REG_READ(GMBUS2 + reg_offset) & (GMBUS_SATOER | GMBUS_HW_RDY), 50))
+					goto timeout;
+				if (REG_READ(GMBUS2 + reg_offset) & GMBUS_SATOER)
+					goto clear_err;
+
+				val = REG_READ(GMBUS3 + reg_offset);
+				do {
+					*buf++ = val & 0xff;
+					val >>= 8;
+				} while (--len && ++loop < 4);
+			} while (len);
+		} else {
+			u32 val, loop;
+
+			val = loop = 0;
+			do {
+				val |= *buf++ << (8 * loop);
+			} while (--len && ++loop < 4);
+
+			REG_WRITE(GMBUS3 + reg_offset, val);
+			REG_WRITE(GMBUS1 + reg_offset,
+				   (i + 1 == num ? GMBUS_CYCLE_STOP : GMBUS_CYCLE_WAIT) |
+				   (msgs[i].len << GMBUS_BYTE_COUNT_SHIFT) |
+				   (msgs[i].addr << GMBUS_SLAVE_ADDR_SHIFT) |
+				   GMBUS_SLAVE_WRITE | GMBUS_SW_RDY);
+			REG_READ(GMBUS2+reg_offset);
+
+			while (len) {
+				if (wait_for(REG_READ(GMBUS2 + reg_offset) & (GMBUS_SATOER | GMBUS_HW_RDY), 50))
+					goto timeout;
+				if (REG_READ(GMBUS2 + reg_offset) & GMBUS_SATOER)
+					goto clear_err;
+
+				val = loop = 0;
+				do {
+					val |= *buf++ << (8 * loop);
+				} while (--len && ++loop < 4);
+
+				REG_WRITE(GMBUS3 + reg_offset, val);
+				REG_READ(GMBUS2+reg_offset);
+			}
+		}
+
+		if (i + 1 < num && wait_for(REG_READ(GMBUS2 + reg_offset) & (GMBUS_SATOER | GMBUS_HW_WAIT_PHASE), 50))
+			goto timeout;
+		if (REG_READ(GMBUS2 + reg_offset) & GMBUS_SATOER)
+			goto clear_err;
+	}
+
+	goto done;
+
+clear_err:
+	/* Toggle the Software Clear Interrupt bit. This has the effect
+	 * of resetting the GMBUS controller and so clearing the
+	 * BUS_ERROR raised by the slave's NAK.
+	 */
+	REG_WRITE(GMBUS1 + reg_offset, GMBUS_SW_CLR_INT);
+	REG_WRITE(GMBUS1 + reg_offset, 0);
+
+done:
+	/* Mark the GMBUS interface as disabled. We will re-enable it at the
+	 * start of the next xfer, till then let it sleep.
+	 */
+	REG_WRITE(GMBUS0 + reg_offset, 0);
+	return i;
+
+timeout:
+	DRM_INFO("GMBUS timed out, falling back to bit banging on pin %d [%s]\n",
+		 bus->reg0 & 0xff, bus->adapter.name);
+	REG_WRITE(GMBUS0 + reg_offset, 0);
+
+	/* Hardware may not support GMBUS over these pins? Try GPIO bitbanging instead. */
+	bus->force_bit = intel_gpio_create(dev_priv, bus->reg0 & 0xff);
+	if (!bus->force_bit)
+		return -ENOMEM;
+
+	return intel_i2c_quirk_xfer(dev_priv, bus->force_bit, msgs, num);
+}
+
+static u32 gmbus_func(struct i2c_adapter *adapter)
+{
+	struct intel_gmbus *bus = container_of(adapter,
+					       struct intel_gmbus,
+					       adapter);
+
+	if (bus->force_bit)
+		bus->force_bit->algo->functionality(bus->force_bit);
+
+	return (I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL |
+		/* I2C_FUNC_10BIT_ADDR | */
+		I2C_FUNC_SMBUS_READ_BLOCK_DATA |
+		I2C_FUNC_SMBUS_BLOCK_PROC_CALL);
+}
+
+static const struct i2c_algorithm gmbus_algorithm = {
+	.master_xfer	= gmbus_xfer,
+	.functionality	= gmbus_func
+};
+
+/**
+ * intel_gmbus_setup - instantiate all Intel i2c GMBuses
+ * @dev: DRM device
+ */
+int gma_intel_setup_gmbus(struct drm_device *dev)
+{
+	static const char *names[GMBUS_NUM_PORTS] = {
+		"disabled",
+		"ssc",
+		"vga",
+		"panel",
+		"dpc",
+		"dpb",
+		"reserved",
+		"dpd",
+	};
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	int ret, i;
+
+	dev_priv->gmbus = kcalloc(sizeof(struct intel_gmbus), GMBUS_NUM_PORTS,
+				  GFP_KERNEL);
+	if (dev_priv->gmbus == NULL)
+		return -ENOMEM;
+
+	for (i = 0; i < GMBUS_NUM_PORTS; i++) {
+		struct intel_gmbus *bus = &dev_priv->gmbus[i];
+
+		bus->adapter.owner = THIS_MODULE;
+		bus->adapter.class = I2C_CLASS_DDC;
+		snprintf(bus->adapter.name,
+			 sizeof(bus->adapter.name),
+			 "gma500 gmbus %s",
+			 names[i]);
+
+		bus->adapter.dev.parent = &dev->pdev->dev;
+		bus->adapter.algo_data	= dev_priv;
+
+		bus->adapter.algo = &gmbus_algorithm;
+		ret = i2c_add_adapter(&bus->adapter);
+		if (ret)
+			goto err;
+
+		/* By default use a conservative clock rate */
+		bus->reg0 = i | GMBUS_RATE_100KHZ;
+
+		/* XXX force bit banging until GMBUS is fully debugged */
+		bus->force_bit = intel_gpio_create(dev_priv, i);
+	}
+
+	gma_intel_i2c_reset(dev_priv->dev);
+
+	return 0;
+
+err:
+	while (--i) {
+		struct intel_gmbus *bus = &dev_priv->gmbus[i];
+		i2c_del_adapter(&bus->adapter);
+	}
+	kfree(dev_priv->gmbus);
+	dev_priv->gmbus = NULL;
+	return ret;
+}
+
+void gma_intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed)
+{
+	struct intel_gmbus *bus = to_intel_gmbus(adapter);
+
+	/* speed:
+	 * 0x0 = 100 KHz
+	 * 0x1 = 50 KHz
+	 * 0x2 = 400 KHz
+	 * 0x3 = 1000 Khz
+	 */
+	bus->reg0 = (bus->reg0 & ~(0x3 << 8)) | (speed << 8);
+}
+
+void gma_intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit)
+{
+	struct intel_gmbus *bus = to_intel_gmbus(adapter);
+
+	if (force_bit) {
+		if (bus->force_bit == NULL) {
+			struct drm_psb_private *dev_priv = adapter->algo_data;
+			bus->force_bit = intel_gpio_create(dev_priv,
+							   bus->reg0 & 0xff);
+		}
+	} else {
+		if (bus->force_bit) {
+			i2c_del_adapter(bus->force_bit);
+			kfree(bus->force_bit);
+			bus->force_bit = NULL;
+		}
+	}
+}
+
+void gma_intel_teardown_gmbus(struct drm_device *dev)
+{
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	int i;
+
+	if (dev_priv->gmbus == NULL)
+		return;
+
+	for (i = 0; i < GMBUS_NUM_PORTS; i++) {
+		struct intel_gmbus *bus = &dev_priv->gmbus[i];
+		if (bus->force_bit) {
+			i2c_del_adapter(bus->force_bit);
+			kfree(bus->force_bit);
+		}
+		i2c_del_adapter(&bus->adapter);
+	}
+
+	kfree(dev_priv->gmbus);
+	dev_priv->gmbus = NULL;
+}
diff --git a/drivers/gpu/drm/gma500/mid_bios.c b/drivers/gpu/drm/gma500/mid_bios.c
index 128e6bb..55fee10 100644
--- a/drivers/gpu/drm/gma500/mid_bios.c
+++ b/drivers/gpu/drm/gma500/mid_bios.c
@@ -45,6 +45,12 @@
 #define FB_SKU_100 0
 #define FB_SKU_100L 1
 #define FB_SKU_83 2
+	if (pci_root == NULL) {
+		WARN_ON(1);
+		return;
+	}
+
+
 	pci_write_config_dword(pci_root, 0xD0, FB_REG06);
 	pci_read_config_dword(pci_root, 0xD4, &fuse_value);
 
@@ -101,6 +107,10 @@
 	uint32_t platform_rev_id = 0;
 	struct pci_dev *pci_gfx_root = pci_get_bus_and_slot(0, PCI_DEVFN(2, 0));
 
+	if (pci_gfx_root == NULL) {
+		WARN_ON(1);
+		return;
+	}
 	pci_read_config_dword(pci_gfx_root, 0x08, &platform_rev_id);
 	dev_priv->platform_rev_id = (uint8_t) platform_rev_id;
 	pci_dev_put(pci_gfx_root);
@@ -138,6 +148,10 @@
 
 	/* get the virtual address of the vbt */
 	vbt_virtual = ioremap(addr, sizeof(*vbt));
+	if (vbt_virtual == NULL) {
+		vbt->size = 0;
+		return;
+	}
 
 	memcpy(vbt, vbt_virtual, sizeof(*vbt));
 	iounmap(vbt_virtual); /* Free virtual address space */
diff --git a/drivers/gpu/drm/gma500/oaktrail_crtc.c b/drivers/gpu/drm/gma500/oaktrail_crtc.c
index 8e15b5a..9d12a3e 100644
--- a/drivers/gpu/drm/gma500/oaktrail_crtc.c
+++ b/drivers/gpu/drm/gma500/oaktrail_crtc.c
@@ -313,9 +313,9 @@
 	bool is_crt = false, is_lvds = false, is_tv = false;
 	bool is_mipi = false;
 	struct drm_mode_config *mode_config = &dev->mode_config;
-	struct psb_intel_output *psb_intel_output = NULL;
+	struct psb_intel_encoder *psb_intel_encoder = NULL;
 	uint64_t scalingType = DRM_MODE_SCALE_FULLSCREEN;
-	struct drm_encoder *encoder;
+	struct drm_connector *connector;
 
 	if (!gma_power_begin(dev, true))
 		return 0;
@@ -327,13 +327,13 @@
 		adjusted_mode,
 		sizeof(struct drm_display_mode));
 
-	list_for_each_entry(encoder, &mode_config->encoder_list, head) {
-
-		if (encoder->crtc != crtc)
+	list_for_each_entry(connector, &mode_config->connector_list, head) {
+		if (!connector->encoder || connector->encoder->crtc != crtc)
 			continue;
 
-		psb_intel_output = enc_to_psb_intel_output(encoder);
-		switch (psb_intel_output->type) {
+		psb_intel_encoder = psb_intel_attached_encoder(connector);
+
+		switch (psb_intel_encoder->type) {
 		case INTEL_OUTPUT_LVDS:
 			is_lvds = true;
 			break;
@@ -363,8 +363,8 @@
 		  ((mode->crtc_hdisplay - 1) << 16) |
 		  (mode->crtc_vdisplay - 1));
 
-	if (psb_intel_output)
-		drm_connector_property_get_value(&psb_intel_output->base,
+	if (psb_intel_encoder)
+		drm_connector_property_get_value(connector,
 			dev->mode_config.scaling_mode_property, &scalingType);
 
 	if (scalingType == DRM_MODE_SCALE_NO_SCALE) {
@@ -543,9 +543,9 @@
 		return 0;
 
 	start = psbfb->gtt->offset;
-	offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8);
+	offset = y * crtc->fb->pitches[0] + x * (crtc->fb->bits_per_pixel / 8);
 
-	REG_WRITE(dspstride, crtc->fb->pitch);
+	REG_WRITE(dspstride, crtc->fb->pitches[0]);
 
 	dspcntr = REG_READ(dspcntr_reg);
 	dspcntr &= ~DISPPLANE_PIXFORMAT_MASK;
diff --git a/drivers/gpu/drm/gma500/oaktrail_hdmi.c b/drivers/gpu/drm/gma500/oaktrail_hdmi.c
index 6f423c0..36878a6 100644
--- a/drivers/gpu/drm/gma500/oaktrail_hdmi.c
+++ b/drivers/gpu/drm/gma500/oaktrail_hdmi.c
@@ -643,29 +643,33 @@
 void oaktrail_hdmi_init(struct drm_device *dev,
 					struct psb_intel_mode_device *mode_dev)
 {
-	struct psb_intel_output *psb_intel_output;
+	struct psb_intel_encoder *psb_intel_encoder;
+	struct psb_intel_connector *psb_intel_connector;
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
 
-	psb_intel_output = kzalloc(sizeof(struct psb_intel_output), GFP_KERNEL);
-	if (!psb_intel_output)
+	psb_intel_encoder = kzalloc(sizeof(struct psb_intel_encoder), GFP_KERNEL);
+	if (!psb_intel_encoder)
 		return;
 
-	psb_intel_output->mode_dev = mode_dev;
-	connector = &psb_intel_output->base;
-	encoder = &psb_intel_output->enc;
-	drm_connector_init(dev, &psb_intel_output->base,
+	psb_intel_connector = kzalloc(sizeof(struct psb_intel_connector), GFP_KERNEL);
+	if (!psb_intel_connector)
+		goto failed_connector;
+
+	connector = &psb_intel_connector->base;
+	encoder = &psb_intel_encoder->base;
+	drm_connector_init(dev, connector,
 			   &oaktrail_hdmi_connector_funcs,
 			   DRM_MODE_CONNECTOR_DVID);
 
-	drm_encoder_init(dev, &psb_intel_output->enc,
+	drm_encoder_init(dev, encoder,
 			 &oaktrail_hdmi_enc_funcs,
 			 DRM_MODE_ENCODER_TMDS);
 
-	drm_mode_connector_attach_encoder(&psb_intel_output->base,
-					  &psb_intel_output->enc);
+	psb_intel_connector_attach_encoder(psb_intel_connector,
+					   psb_intel_encoder);
 
-	psb_intel_output->type = INTEL_OUTPUT_HDMI;
+	psb_intel_encoder->type = INTEL_OUTPUT_HDMI;
 	drm_encoder_helper_add(encoder, &oaktrail_hdmi_helper_funcs);
 	drm_connector_helper_add(connector, &oaktrail_hdmi_connector_helper_funcs);
 
@@ -675,6 +679,9 @@
 	drm_sysfs_connector_add(connector);
 
 	return;
+
+failed_connector:
+	kfree(psb_intel_encoder);
 }
 
 static DEFINE_PCI_DEVICE_TABLE(hdmi_ids) = {
diff --git a/drivers/gpu/drm/gma500/oaktrail_lvds.c b/drivers/gpu/drm/gma500/oaktrail_lvds.c
index 69659ca..238bbe1 100644
--- a/drivers/gpu/drm/gma500/oaktrail_lvds.c
+++ b/drivers/gpu/drm/gma500/oaktrail_lvds.c
@@ -43,7 +43,8 @@
  * Sets the power state for the panel.
  */
 static void oaktrail_lvds_set_power(struct drm_device *dev,
-				struct psb_intel_output *output, bool on)
+				struct psb_intel_encoder *psb_intel_encoder,
+				bool on)
 {
 	u32 pp_status;
 	struct drm_psb_private *dev_priv = dev->dev_private;
@@ -77,12 +78,13 @@
 static void oaktrail_lvds_dpms(struct drm_encoder *encoder, int mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
+	struct psb_intel_encoder *psb_intel_encoder =
+						to_psb_intel_encoder(encoder);
 
 	if (mode == DRM_MODE_DPMS_ON)
-		oaktrail_lvds_set_power(dev, output, true);
+		oaktrail_lvds_set_power(dev, psb_intel_encoder, true);
 	else
-		oaktrail_lvds_set_power(dev, output, false);
+		oaktrail_lvds_set_power(dev, psb_intel_encoder, false);
 
 	/* XXX: We never power down the LVDS pairs. */
 }
@@ -91,10 +93,12 @@
 			       struct drm_display_mode *mode,
 			       struct drm_display_mode *adjusted_mode)
 {
-	struct psb_intel_mode_device *mode_dev =
-				enc_to_psb_intel_output(encoder)->mode_dev;
 	struct drm_device *dev = encoder->dev;
 	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
+	struct drm_mode_config *mode_config = &dev->mode_config;
+	struct drm_connector *connector = NULL;
+	struct drm_crtc *crtc = encoder->crtc;
 	u32 lvds_port;
 	uint64_t v = DRM_MODE_SCALE_FULLSCREEN;
 
@@ -118,8 +122,19 @@
 
 	REG_WRITE(LVDS, lvds_port);
 
+	/* Find the connector we're trying to set up */
+	list_for_each_entry(connector, &mode_config->connector_list, head) {
+		if (!connector->encoder || connector->encoder->crtc != crtc)
+			continue;
+	}
+
+	if (!connector) {
+		DRM_ERROR("Couldn't find connector when setting mode");
+		return;
+	}
+
 	drm_connector_property_get_value(
-		&enc_to_psb_intel_output(encoder)->base,
+		connector,
 		dev->mode_config.scaling_mode_property,
 		&v);
 
@@ -150,8 +165,10 @@
 static void oaktrail_lvds_prepare(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
-	struct psb_intel_mode_device *mode_dev = output->mode_dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_encoder *psb_intel_encoder =
+						to_psb_intel_encoder(encoder);
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 
 	if (!gma_power_begin(dev, true))
 		return;
@@ -159,7 +176,7 @@
 	mode_dev->saveBLC_PWM_CTL = REG_READ(BLC_PWM_CTL);
 	mode_dev->backlight_duty_cycle = (mode_dev->saveBLC_PWM_CTL &
 					  BACKLIGHT_DUTY_CYCLE_MASK);
-	oaktrail_lvds_set_power(dev, output, false);
+	oaktrail_lvds_set_power(dev, psb_intel_encoder, false);
 	gma_power_end(dev);
 }
 
@@ -185,13 +202,15 @@
 static void oaktrail_lvds_commit(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
-	struct psb_intel_mode_device *mode_dev = output->mode_dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_encoder *psb_intel_encoder =
+						to_psb_intel_encoder(encoder);
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 
 	if (mode_dev->backlight_duty_cycle == 0)
 		mode_dev->backlight_duty_cycle =
 					oaktrail_lvds_get_max_backlight(dev);
-	oaktrail_lvds_set_power(dev, output, true);
+	oaktrail_lvds_set_power(dev, psb_intel_encoder, true);
 }
 
 static const struct drm_encoder_helper_funcs oaktrail_lvds_helper_funcs = {
@@ -306,34 +325,37 @@
 void oaktrail_lvds_init(struct drm_device *dev,
 		    struct psb_intel_mode_device *mode_dev)
 {
-	struct psb_intel_output *psb_intel_output;
+	struct psb_intel_encoder *psb_intel_encoder;
+	struct psb_intel_connector *psb_intel_connector;
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
-	struct drm_psb_private *dev_priv =
-				(struct drm_psb_private *) dev->dev_private;
+	struct drm_psb_private *dev_priv = dev->dev_private;
 	struct edid *edid;
 	int ret = 0;
 	struct i2c_adapter *i2c_adap;
 	struct drm_display_mode *scan;	/* *modes, *bios_mode; */
 
-	psb_intel_output = kzalloc(sizeof(struct psb_intel_output), GFP_KERNEL);
-	if (!psb_intel_output)
+	psb_intel_encoder = kzalloc(sizeof(struct psb_intel_encoder), GFP_KERNEL);
+	if (!psb_intel_encoder)
 		return;
 
-	psb_intel_output->mode_dev = mode_dev;
-	connector = &psb_intel_output->base;
-	encoder = &psb_intel_output->enc;
+	psb_intel_connector = kzalloc(sizeof(struct psb_intel_connector), GFP_KERNEL);
+	if (!psb_intel_connector)
+		goto failed_connector;
+
+	connector = &psb_intel_connector->base;
+	encoder = &psb_intel_encoder->base;
 	dev_priv->is_lvds_on = true;
-	drm_connector_init(dev, &psb_intel_output->base,
+	drm_connector_init(dev, connector,
 			   &psb_intel_lvds_connector_funcs,
 			   DRM_MODE_CONNECTOR_LVDS);
 
-	drm_encoder_init(dev, &psb_intel_output->enc, &psb_intel_lvds_enc_funcs,
+	drm_encoder_init(dev, encoder, &psb_intel_lvds_enc_funcs,
 			 DRM_MODE_ENCODER_LVDS);
 
-	drm_mode_connector_attach_encoder(&psb_intel_output->base,
-					  &psb_intel_output->enc);
-	psb_intel_output->type = INTEL_OUTPUT_LVDS;
+	psb_intel_connector_attach_encoder(psb_intel_connector,
+					   psb_intel_encoder);
+	psb_intel_encoder->type = INTEL_OUTPUT_LVDS;
 
 	drm_encoder_helper_add(encoder, &oaktrail_lvds_helper_funcs);
 	drm_connector_helper_add(connector,
@@ -353,6 +375,8 @@
 	if (dev_priv->vbt_data.size != 0x00)
 		mode_dev->panel_wants_dither = (dev_priv->gct_data.
 			Panel_Port_Control & MRST_PANEL_8TO6_DITHER_ENABLE);
+        if (dev_priv->lvds_dither)
+                mode_dev->panel_wants_dither = 1;
 
 	/*
 	 * LVDS discovery:
@@ -411,13 +435,15 @@
 
 failed_find:
 	dev_dbg(dev->dev, "No LVDS modes found, disabling.\n");
-	if (psb_intel_output->ddc_bus)
-		psb_intel_i2c_destroy(psb_intel_output->ddc_bus);
+	if (psb_intel_encoder->ddc_bus)
+		psb_intel_i2c_destroy(psb_intel_encoder->ddc_bus);
 
 /* failed_ddc: */
 
 	drm_encoder_cleanup(encoder);
 	drm_connector_cleanup(connector);
-	kfree(connector);
+	kfree(psb_intel_connector);
+failed_connector:
+	kfree(psb_intel_encoder);
 }
 
diff --git a/drivers/gpu/drm/gma500/psb_device.c b/drivers/gpu/drm/gma500/psb_device.c
index 35eddef..e5f5906 100644
--- a/drivers/gpu/drm/gma500/psb_device.c
+++ b/drivers/gpu/drm/gma500/psb_device.c
@@ -290,11 +290,17 @@
 static int psb_chip_setup(struct drm_device *dev)
 {
 	psb_get_core_freq(dev);
+	gma_intel_setup_gmbus(dev);
 	gma_intel_opregion_init(dev);
 	psb_intel_init_bios(dev);
 	return 0;
 }
 
+static void psb_chip_teardown(struct drm_device *dev)
+{
+	gma_intel_teardown_gmbus(dev);
+}
+
 const struct psb_ops psb_chip_ops = {
 	.name = "Poulsbo",
 	.accel_2d = 1,
@@ -302,6 +308,7 @@
 	.crtcs = 2,
 	.sgx_offset = PSB_SGX_OFFSET,
 	.chip_setup = psb_chip_setup,
+	.chip_teardown = psb_chip_teardown,
 
 	.crtc_helper = &psb_intel_helper_funcs,
 	.crtc_funcs = &psb_intel_crtc_funcs,
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index add3156..f14768f 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -39,20 +39,16 @@
 
 static int drm_psb_trap_pagefaults;
 
-int drm_psb_no_fb;
-
 static int psb_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
 
-MODULE_PARM_DESC(no_fb, "Disable FBdev");
 MODULE_PARM_DESC(trap_pagefaults, "Error and reset on MMU pagefaults");
-module_param_named(no_fb, drm_psb_no_fb, int, 0600);
 module_param_named(trap_pagefaults, drm_psb_trap_pagefaults, int, 0600);
 
 
 static DEFINE_PCI_DEVICE_TABLE(pciidlist) = {
 	{ 0x8086, 0x8108, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &psb_chip_ops },
 	{ 0x8086, 0x8109, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &psb_chip_ops },
-#if defined(CONFIG_DRM_OAKTRAIL)
+#if defined(CONFIG_DRM_GMA600)
 	{ 0x8086, 0x4100, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &oaktrail_chip_ops},
 	{ 0x8086, 0x4101, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &oaktrail_chip_ops},
 	{ 0x8086, 0x4102, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &oaktrail_chip_ops},
@@ -61,8 +57,10 @@
 	{ 0x8086, 0x4105, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &oaktrail_chip_ops},
 	{ 0x8086, 0x4106, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &oaktrail_chip_ops},
 	{ 0x8086, 0x4107, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &oaktrail_chip_ops},
+	/* Atom E620 */
+	{ 0x8086, 0x4108, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &oaktrail_chip_ops},
 #endif
-#if defined(CONFIG_DRM_CDV)
+#if defined(CONFIG_DRM_GMA3600)
 	{ 0x8086, 0x0be0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &cdv_chip_ops},
 	{ 0x8086, 0x0be1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &cdv_chip_ops},
 	{ 0x8086, 0x0be2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &cdv_chip_ops},
@@ -211,8 +209,7 @@
 
 	gma_backlight_exit(dev);
 
-	if (drm_psb_no_fb == 0)
-		psb_modeset_cleanup(dev);
+	psb_modeset_cleanup(dev);
 
 	if (dev_priv) {
 		psb_lid_timer_takedown(dev_priv);
@@ -276,7 +273,7 @@
 	int ret = -ENOMEM;
 	uint32_t tt_pages;
 	struct drm_connector *connector;
-	struct psb_intel_output *psb_intel_output;
+	struct psb_intel_encoder *psb_intel_encoder;
 
 	dev_priv = kzalloc(sizeof(*dev_priv), GFP_KERNEL);
 	if (dev_priv == NULL)
@@ -381,18 +378,16 @@
 
 	dev->driver->get_vblank_counter = psb_get_vblank_counter;
 
-	if (drm_psb_no_fb == 0) {
-		psb_modeset_init(dev);
-		psb_fbdev_init(dev);
-		drm_kms_helper_poll_init(dev);
-	}
+	psb_modeset_init(dev);
+	psb_fbdev_init(dev);
+	drm_kms_helper_poll_init(dev);
 
 	/* Only add backlight support if we have LVDS output */
 	list_for_each_entry(connector, &dev->mode_config.connector_list,
 			    head) {
-		psb_intel_output = to_psb_intel_output(connector);
+		psb_intel_encoder = psb_intel_attached_encoder(connector);
 
-		switch (psb_intel_output->type) {
+		switch (psb_intel_encoder->type) {
 		case INTEL_OUTPUT_LVDS:
 		case INTEL_OUTPUT_MIPI:
 			ret = gma_backlight_init(dev);
diff --git a/drivers/gpu/drm/gma500/psb_drv.h b/drivers/gpu/drm/gma500/psb_drv.h
index 5ec8edf..eb1568a 100644
--- a/drivers/gpu/drm/gma500/psb_drv.h
+++ b/drivers/gpu/drm/gma500/psb_drv.h
@@ -260,6 +260,22 @@
 	int enabled;
 };
 
+struct sdvo_device_mapping {
+	u8 initialized;
+	u8 dvo_port;
+	u8 slave_addr;
+	u8 dvo_wiring;
+	u8 i2c_pin;
+	u8 i2c_speed;
+	u8 ddc_pin;
+};
+
+struct intel_gmbus {
+	struct i2c_adapter adapter;
+	struct i2c_adapter *force_bit;
+	u32 reg0;
+};
+
 struct psb_ops;
 
 #define PSB_NUM_PIPE		3
@@ -336,6 +352,18 @@
 	/* PCI revision ID for B0:D2:F0 */
 	uint8_t platform_rev_id;
 
+	/* gmbus */
+	struct intel_gmbus *gmbus;
+
+	/* Used by SDVO */
+	int crt_ddc_pin;
+	/* FIXME: The mappings should be parsed from bios but for now we can
+		  pretend there are no mappings available */
+	struct sdvo_device_mapping sdvo_mappings[2];
+	u32 hotplug_supported_mask;
+	struct drm_property *broadcast_rgb_property;
+	struct drm_property *force_audio_property;
+
 	/*
 	 * LVDS info
 	 */
@@ -346,7 +374,7 @@
 	struct drm_display_mode *sdvo_lvds_vbt_mode;
 
 	struct bdb_lvds_backlight *lvds_bl; /* LVDS backlight info from VBT */
-	struct psb_intel_i2c_chan *lvds_i2c_bus;
+	struct psb_intel_i2c_chan *lvds_i2c_bus; /* FIXME: Remove this? */
 
 	/* Feature bits from the VBIOS */
 	unsigned int int_tv_support:1;
diff --git a/drivers/gpu/drm/gma500/psb_intel_display.c b/drivers/gpu/drm/gma500/psb_intel_display.c
index ab65076..49e9835 100644
--- a/drivers/gpu/drm/gma500/psb_intel_display.c
+++ b/drivers/gpu/drm/gma500/psb_intel_display.c
@@ -214,9 +214,9 @@
 
 	list_for_each_entry(l_entry, &mode_config->connector_list, head) {
 		if (l_entry->encoder && l_entry->encoder->crtc == crtc) {
-			struct psb_intel_output *psb_intel_output =
-			    to_psb_intel_output(l_entry);
-			if (psb_intel_output->type == type)
+			struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(l_entry);
+			if (psb_intel_encoder->type == type)
 				return true;
 		}
 	}
@@ -365,9 +365,9 @@
 		goto psb_intel_pipe_set_base_exit;
 	start = psbfb->gtt->offset;
 
-	offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8);
+	offset = y * crtc->fb->pitches[0] + x * (crtc->fb->bits_per_pixel / 8);
 
-	REG_WRITE(dspstride, crtc->fb->pitch);
+	REG_WRITE(dspstride, crtc->fb->pitches[0]);
 
 	dspcntr = REG_READ(dspcntr_reg);
 	dspcntr &= ~DISPPLANE_PIXFORMAT_MASK;
@@ -552,6 +552,14 @@
 	encoder_funcs->dpms(encoder, DRM_MODE_DPMS_ON);
 }
 
+void psb_intel_encoder_destroy(struct drm_encoder *encoder)
+{
+	struct psb_intel_encoder *intel_encoder = to_psb_intel_encoder(encoder);
+
+	drm_encoder_cleanup(encoder);
+	kfree(intel_encoder);
+}
+
 static bool psb_intel_crtc_mode_fixup(struct drm_crtc *crtc,
 				  struct drm_display_mode *mode,
 				  struct drm_display_mode *adjusted_mode)
@@ -615,14 +623,14 @@
 	}
 
 	list_for_each_entry(connector, &mode_config->connector_list, head) {
-		struct psb_intel_output *psb_intel_output =
-		    to_psb_intel_output(connector);
+		struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 
 		if (!connector->encoder
 		    || connector->encoder->crtc != crtc)
 			continue;
 
-		switch (psb_intel_output->type) {
+		switch (psb_intel_encoder->type) {
 		case INTEL_OUTPUT_LVDS:
 			is_lvds = true;
 			break;
@@ -1402,9 +1410,9 @@
 
 	list_for_each_entry(connector, &dev->mode_config.connector_list,
 			    head) {
-		struct psb_intel_output *psb_intel_output =
-		    to_psb_intel_output(connector);
-		if (type_mask & (1 << psb_intel_output->type))
+		struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
+		if (type_mask & (1 << psb_intel_encoder->type))
 			index_mask |= (1 << entry);
 		entry++;
 	}
@@ -1423,9 +1431,16 @@
 */
 struct drm_encoder *psb_intel_best_encoder(struct drm_connector *connector)
 {
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 
-	return &psb_intel_output->enc;
+	return &psb_intel_encoder->base;
 }
 
+void psb_intel_connector_attach_encoder(struct psb_intel_connector *connector,
+					struct psb_intel_encoder *encoder)
+{
+	connector->encoder = encoder;
+	drm_mode_connector_attach_encoder(&connector->base,
+					  &encoder->base);
+}
diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h
index ac953ca..f40535e 100644
--- a/drivers/gpu/drm/gma500/psb_intel_drv.h
+++ b/drivers/gpu/drm/gma500/psb_intel_drv.h
@@ -39,6 +39,25 @@
 #define INTEL_I2C_BUS_DVO 1
 #define INTEL_I2C_BUS_SDVO 2
 
+/* Intel Pipe Clone Bit */
+#define INTEL_HDMIB_CLONE_BIT 1
+#define INTEL_HDMIC_CLONE_BIT 2
+#define INTEL_HDMID_CLONE_BIT 3
+#define INTEL_HDMIE_CLONE_BIT 4
+#define INTEL_HDMIF_CLONE_BIT 5
+#define INTEL_SDVO_NON_TV_CLONE_BIT 6
+#define INTEL_SDVO_TV_CLONE_BIT 7
+#define INTEL_SDVO_LVDS_CLONE_BIT 8
+#define INTEL_ANALOG_CLONE_BIT 9
+#define INTEL_TV_CLONE_BIT 10
+#define INTEL_DP_B_CLONE_BIT 11
+#define INTEL_DP_C_CLONE_BIT 12
+#define INTEL_DP_D_CLONE_BIT 13
+#define INTEL_LVDS_CLONE_BIT 14
+#define INTEL_DVO_TMDS_CLONE_BIT 15
+#define INTEL_DVO_LVDS_CLONE_BIT 16
+#define INTEL_EDP_CLONE_BIT 17
+
 /* these are outputs from the chip - integrated only
  * external chips are via DVO or SDVO output */
 #define INTEL_OUTPUT_UNUSED 0
@@ -56,6 +75,25 @@
 #define INTEL_DVO_CHIP_TMDS 2
 #define INTEL_DVO_CHIP_TVOUT 4
 
+#define INTEL_MODE_PIXEL_MULTIPLIER_SHIFT (0x0)
+#define INTEL_MODE_PIXEL_MULTIPLIER_MASK (0xf << INTEL_MODE_PIXEL_MULTIPLIER_SHIFT)
+
+static inline void
+psb_intel_mode_set_pixel_multiplier(struct drm_display_mode *mode,
+				int multiplier)
+{
+	mode->clock *= multiplier;
+	mode->private_flags |= multiplier;
+}
+
+static inline int
+psb_intel_mode_get_pixel_multiplier(const struct drm_display_mode *mode)
+{
+	return (mode->private_flags & INTEL_MODE_PIXEL_MULTIPLIER_MASK)
+	       >> INTEL_MODE_PIXEL_MULTIPLIER_SHIFT;
+}
+
+
 /*
  * Hold information useally put on the device driver privates here,
  * since it needs to be shared across multiple of devices drivers privates.
@@ -93,19 +131,24 @@
 	u8 slave_addr;
 };
 
-struct psb_intel_output {
-	struct drm_connector base;
-
-	struct drm_encoder enc;
+struct psb_intel_encoder {
+	struct drm_encoder base;
 	int type;
+	bool needs_tv_clock;
+	void (*hot_plug)(struct psb_intel_encoder *);
+	int crtc_mask;
+	int clone_mask;
+	void *dev_priv; /* For sdvo_priv, lvds_priv, etc... */
 
-	struct psb_intel_i2c_chan *i2c_bus;	/* for control functions */
-	struct psb_intel_i2c_chan *ddc_bus;	/* for DDC only stuff */
-	bool load_detect_temp;
-	void *dev_priv;
+	/* FIXME: Either make SDVO and LVDS store it's i2c here or give CDV it's
+	   own set of output privates */
+	struct psb_intel_i2c_chan *i2c_bus;
+	struct psb_intel_i2c_chan *ddc_bus;
+};
 
-	struct psb_intel_mode_device *mode_dev;
-	struct i2c_adapter *hdmi_i2c_adapter;	/* for control functions */
+struct psb_intel_connector {
+	struct drm_connector base;
+	struct psb_intel_encoder *encoder;
 };
 
 struct psb_intel_crtc_state {
@@ -156,23 +199,24 @@
 
 #define to_psb_intel_crtc(x)	\
 		container_of(x, struct psb_intel_crtc, base)
-#define to_psb_intel_output(x)	\
-		container_of(x, struct psb_intel_output, base)
-#define enc_to_psb_intel_output(x)	\
-		container_of(x, struct psb_intel_output, enc)
+#define to_psb_intel_connector(x) \
+		container_of(x, struct psb_intel_connector, base)
+#define to_psb_intel_encoder(x)	\
+		container_of(x, struct psb_intel_encoder, base)
 #define to_psb_intel_framebuffer(x)	\
 		container_of(x, struct psb_intel_framebuffer, base)
 
 struct psb_intel_i2c_chan *psb_intel_i2c_create(struct drm_device *dev,
 					const u32 reg, const char *name);
 void psb_intel_i2c_destroy(struct psb_intel_i2c_chan *chan);
-int psb_intel_ddc_get_modes(struct psb_intel_output *psb_intel_output);
-extern bool psb_intel_ddc_probe(struct psb_intel_output *psb_intel_output);
+int psb_intel_ddc_get_modes(struct drm_connector *connector,
+			    struct i2c_adapter *adapter);
+extern bool psb_intel_ddc_probe(struct i2c_adapter *adapter);
 
 extern void psb_intel_crtc_init(struct drm_device *dev, int pipe,
 			    struct psb_intel_mode_device *mode_dev);
 extern void psb_intel_crt_init(struct drm_device *dev);
-extern void psb_intel_sdvo_init(struct drm_device *dev, int output_device);
+extern bool psb_intel_sdvo_init(struct drm_device *dev, int output_device);
 extern void psb_intel_dvo_init(struct drm_device *dev);
 extern void psb_intel_tv_init(struct drm_device *dev);
 extern void psb_intel_lvds_init(struct drm_device *dev,
@@ -189,6 +233,17 @@
 extern void psb_intel_crtc_load_lut(struct drm_crtc *crtc);
 extern void psb_intel_encoder_prepare(struct drm_encoder *encoder);
 extern void psb_intel_encoder_commit(struct drm_encoder *encoder);
+extern void psb_intel_encoder_destroy(struct drm_encoder *encoder);
+
+static inline struct psb_intel_encoder *psb_intel_attached_encoder(
+						struct drm_connector *connector)
+{
+	return to_psb_intel_connector(connector)->encoder;
+}
+
+extern void psb_intel_connector_attach_encoder(
+					struct psb_intel_connector *connector,
+					struct psb_intel_encoder *encoder);
 
 extern struct drm_encoder *psb_intel_best_encoder(struct drm_connector
 					      *connector);
@@ -224,5 +279,11 @@
 extern void psb_intel_lvds_destroy(struct drm_connector *connector);
 extern const struct drm_encoder_funcs psb_intel_lvds_enc_funcs;
 
+/* intel_gmbus.c */
+extern void gma_intel_i2c_reset(struct drm_device *dev);
+extern int gma_intel_setup_gmbus(struct drm_device *dev);
+extern void gma_intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed);
+extern void gma_intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit);
+extern void gma_intel_teardown_gmbus(struct drm_device *dev);
 
 #endif				/* __INTEL_DRV_H__ */
diff --git a/drivers/gpu/drm/gma500/psb_intel_lvds.c b/drivers/gpu/drm/gma500/psb_intel_lvds.c
index 21022e1..a25e4ca 100644
--- a/drivers/gpu/drm/gma500/psb_intel_lvds.c
+++ b/drivers/gpu/drm/gma500/psb_intel_lvds.c
@@ -59,6 +59,9 @@
 	uint32_t savePFIT_CONTROL;
 	uint32_t savePFIT_PGM_RATIOS;
 	uint32_t saveBLC_PWM_CTL;
+
+	struct psb_intel_i2c_chan *i2c_bus;
+	struct psb_intel_i2c_chan *ddc_bus;
 };
 
 
@@ -214,9 +217,10 @@
 /*
  * Sets the power state for the panel.
  */
-static void psb_intel_lvds_set_power(struct drm_device *dev,
-				 struct psb_intel_output *output, bool on)
+static void psb_intel_lvds_set_power(struct drm_device *dev, bool on)
 {
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 	u32 pp_status;
 
 	if (!gma_power_begin(dev, true)) {
@@ -232,8 +236,7 @@
 		} while ((pp_status & PP_ON) == 0);
 
 		psb_intel_lvds_set_backlight(dev,
-					 output->
-					 mode_dev->backlight_duty_cycle);
+					     mode_dev->backlight_duty_cycle);
 	} else {
 		psb_intel_lvds_set_backlight(dev, 0);
 
@@ -250,12 +253,11 @@
 static void psb_intel_lvds_encoder_dpms(struct drm_encoder *encoder, int mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
 
 	if (mode == DRM_MODE_DPMS_ON)
-		psb_intel_lvds_set_power(dev, output, true);
+		psb_intel_lvds_set_power(dev, true);
 	else
-		psb_intel_lvds_set_power(dev, output, false);
+		psb_intel_lvds_set_power(dev, false);
 
 	/* XXX: We never power down the LVDS pairs. */
 }
@@ -265,10 +267,10 @@
 	struct drm_device *dev = connector->dev;
 	struct drm_psb_private *dev_priv =
 		(struct drm_psb_private *)dev->dev_private;
-	struct psb_intel_output *psb_intel_output =
-		to_psb_intel_output(connector);
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 	struct psb_intel_lvds_priv *lvds_priv =
-		(struct psb_intel_lvds_priv *)psb_intel_output->dev_priv;
+		(struct psb_intel_lvds_priv *)psb_intel_encoder->dev_priv;
 
 	lvds_priv->savePP_ON = REG_READ(LVDSPP_ON);
 	lvds_priv->savePP_OFF = REG_READ(LVDSPP_OFF);
@@ -305,10 +307,10 @@
 {
 	struct drm_device *dev = connector->dev;
 	u32 pp_status;
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 	struct psb_intel_lvds_priv *lvds_priv =
-		(struct psb_intel_lvds_priv *)psb_intel_output->dev_priv;
+		(struct psb_intel_lvds_priv *)psb_intel_encoder->dev_priv;
 
 	dev_dbg(dev->dev, "(0x%x, 0x%x, 0x%x, 0x%x, 0x%x, 0x%x)\n",
 			lvds_priv->savePP_ON,
@@ -346,13 +348,14 @@
 int psb_intel_lvds_mode_valid(struct drm_connector *connector,
 				 struct drm_display_mode *mode)
 {
-	struct psb_intel_output *psb_intel_output =
-				to_psb_intel_output(connector);
+	struct drm_psb_private *dev_priv = connector->dev->dev_private;
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
 	struct drm_display_mode *fixed_mode =
-	    psb_intel_output->mode_dev->panel_fixed_mode;
+					dev_priv->mode_dev.panel_fixed_mode;
 
-	if (psb_intel_output->type == INTEL_OUTPUT_MIPI2)
-		fixed_mode = psb_intel_output->mode_dev->panel_fixed_mode2;
+	if (psb_intel_encoder->type == INTEL_OUTPUT_MIPI2)
+		fixed_mode = dev_priv->mode_dev.panel_fixed_mode2;
 
 	/* just in case */
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
@@ -375,17 +378,17 @@
 				  struct drm_display_mode *mode,
 				  struct drm_display_mode *adjusted_mode)
 {
-	struct psb_intel_mode_device *mode_dev =
-	    enc_to_psb_intel_output(encoder)->mode_dev;
 	struct drm_device *dev = encoder->dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 	struct psb_intel_crtc *psb_intel_crtc =
 				to_psb_intel_crtc(encoder->crtc);
 	struct drm_encoder *tmp_encoder;
 	struct drm_display_mode *panel_fixed_mode = mode_dev->panel_fixed_mode;
-	struct psb_intel_output *psb_intel_output =
-					enc_to_psb_intel_output(encoder);
+	struct psb_intel_encoder *psb_intel_encoder =
+						to_psb_intel_encoder(encoder);
 
-	if (psb_intel_output->type == INTEL_OUTPUT_MIPI2)
+	if (psb_intel_encoder->type == INTEL_OUTPUT_MIPI2)
 		panel_fixed_mode = mode_dev->panel_fixed_mode2;
 
 	/* PSB requires the LVDS is on pipe B, MRST has only one pipe anyway */
@@ -440,8 +443,8 @@
 static void psb_intel_lvds_prepare(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
-	struct psb_intel_mode_device *mode_dev = output->mode_dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 
 	if (!gma_power_begin(dev, true))
 		return;
@@ -450,7 +453,7 @@
 	mode_dev->backlight_duty_cycle = (mode_dev->saveBLC_PWM_CTL &
 					  BACKLIGHT_DUTY_CYCLE_MASK);
 
-	psb_intel_lvds_set_power(dev, output, false);
+	psb_intel_lvds_set_power(dev, false);
 
 	gma_power_end(dev);
 }
@@ -458,14 +461,14 @@
 static void psb_intel_lvds_commit(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *output = enc_to_psb_intel_output(encoder);
-	struct psb_intel_mode_device *mode_dev = output->mode_dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
 
 	if (mode_dev->backlight_duty_cycle == 0)
 		mode_dev->backlight_duty_cycle =
 		    psb_intel_lvds_get_max_backlight(dev);
 
-	psb_intel_lvds_set_power(dev, output, true);
+	psb_intel_lvds_set_power(dev, true);
 }
 
 static void psb_intel_lvds_mode_set(struct drm_encoder *encoder,
@@ -520,14 +523,15 @@
 static int psb_intel_lvds_get_modes(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
-	struct psb_intel_mode_device *mode_dev =
-					psb_intel_output->mode_dev;
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_mode_device *mode_dev = &dev_priv->mode_dev;
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
+	struct psb_intel_lvds_priv *lvds_priv = psb_intel_encoder->dev_priv;
 	int ret = 0;
 
 	if (!IS_MRST(dev))
-		ret = psb_intel_ddc_get_modes(psb_intel_output);
+		ret = psb_intel_ddc_get_modes(connector, &lvds_priv->i2c_bus->adapter);
 
 	if (ret)
 		return ret;
@@ -560,11 +564,12 @@
  */
 void psb_intel_lvds_destroy(struct drm_connector *connector)
 {
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_encoder *psb_intel_encoder =
+					psb_intel_attached_encoder(connector);
+	struct psb_intel_lvds_priv *lvds_priv = psb_intel_encoder->dev_priv;
 
-	if (psb_intel_output->ddc_bus)
-		psb_intel_i2c_destroy(psb_intel_output->ddc_bus);
+	if (lvds_priv->ddc_bus)
+		psb_intel_i2c_destroy(lvds_priv->ddc_bus);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(connector);
@@ -693,9 +698,10 @@
  * modes we can display on the LVDS panel (if present).
  */
 void psb_intel_lvds_init(struct drm_device *dev,
-		     struct psb_intel_mode_device *mode_dev)
+			 struct psb_intel_mode_device *mode_dev)
 {
-	struct psb_intel_output *psb_intel_output;
+	struct psb_intel_encoder *psb_intel_encoder;
+	struct psb_intel_connector *psb_intel_connector;
 	struct psb_intel_lvds_priv *lvds_priv;
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
@@ -705,33 +711,43 @@
 	u32 lvds;
 	int pipe;
 
-	psb_intel_output = kzalloc(sizeof(struct psb_intel_output), GFP_KERNEL);
-	if (!psb_intel_output)
-		return;
+	psb_intel_encoder =
+			kzalloc(sizeof(struct psb_intel_encoder), GFP_KERNEL);
 
-	lvds_priv = kzalloc(sizeof(struct psb_intel_lvds_priv), GFP_KERNEL);
-	if (!lvds_priv) {
-		kfree(psb_intel_output);
-		dev_err(dev->dev, "LVDS private allocation error\n");
+	if (!psb_intel_encoder) {
+		dev_err(dev->dev, "psb_intel_encoder allocation error\n");
 		return;
 	}
 
-	psb_intel_output->dev_priv = lvds_priv;
-	psb_intel_output->mode_dev = mode_dev;
+	psb_intel_connector =
+		kzalloc(sizeof(struct psb_intel_connector), GFP_KERNEL);
 
-	connector = &psb_intel_output->base;
-	encoder = &psb_intel_output->enc;
-	drm_connector_init(dev, &psb_intel_output->base,
+	if (!psb_intel_connector) {
+		kfree(psb_intel_encoder);
+		dev_err(dev->dev, "psb_intel_connector allocation error\n");
+	}
+
+	lvds_priv = kzalloc(sizeof(struct psb_intel_lvds_priv), GFP_KERNEL);
+	if (!lvds_priv) {
+		dev_err(dev->dev, "LVDS private allocation error\n");
+		goto failed_connector;
+	}
+
+	psb_intel_encoder->dev_priv = lvds_priv;
+
+	connector = &psb_intel_connector->base;
+	encoder = &psb_intel_encoder->base;
+	drm_connector_init(dev, connector,
 			   &psb_intel_lvds_connector_funcs,
 			   DRM_MODE_CONNECTOR_LVDS);
 
-	drm_encoder_init(dev, &psb_intel_output->enc,
+	drm_encoder_init(dev, encoder,
 			 &psb_intel_lvds_enc_funcs,
 			 DRM_MODE_ENCODER_LVDS);
 
-	drm_mode_connector_attach_encoder(&psb_intel_output->base,
-					  &psb_intel_output->enc);
-	psb_intel_output->type = INTEL_OUTPUT_LVDS;
+	psb_intel_connector_attach_encoder(psb_intel_connector,
+					   psb_intel_encoder);
+	psb_intel_encoder->type = INTEL_OUTPUT_LVDS;
 
 	drm_encoder_helper_add(encoder, &psb_intel_lvds_helper_funcs);
 	drm_connector_helper_add(connector,
@@ -752,16 +768,14 @@
 	 * Set up I2C bus
 	 * FIXME: distroy i2c_bus when exit
 	 */
-	psb_intel_output->i2c_bus = psb_intel_i2c_create(dev,
-							 GPIOB,
-							 "LVDSBLC_B");
-	if (!psb_intel_output->i2c_bus) {
+	lvds_priv->i2c_bus = psb_intel_i2c_create(dev, GPIOB, "LVDSBLC_B");
+	if (!lvds_priv->i2c_bus) {
 		dev_printk(KERN_ERR,
 			&dev->pdev->dev, "I2C bus registration failed.\n");
 		goto failed_blc_i2c;
 	}
-	psb_intel_output->i2c_bus->slave_addr = 0x2C;
-	dev_priv->lvds_i2c_bus =  psb_intel_output->i2c_bus;
+	lvds_priv->i2c_bus->slave_addr = 0x2C;
+	dev_priv->lvds_i2c_bus =  lvds_priv->i2c_bus;
 
 	/*
 	 * LVDS discovery:
@@ -774,10 +788,8 @@
 	 */
 
 	/* Set up the DDC bus. */
-	psb_intel_output->ddc_bus = psb_intel_i2c_create(dev,
-							 GPIOC,
-							 "LVDSDDC_C");
-	if (!psb_intel_output->ddc_bus) {
+	lvds_priv->ddc_bus = psb_intel_i2c_create(dev, GPIOC, "LVDSDDC_C");
+	if (!lvds_priv->ddc_bus) {
 		dev_printk(KERN_ERR, &dev->pdev->dev,
 			   "DDC bus registration " "failed.\n");
 		goto failed_ddc;
@@ -787,7 +799,7 @@
 	 * Attempt to get the fixed panel mode from DDC.  Assume that the
 	 * preferred mode is the right one.
 	 */
-	psb_intel_ddc_get_modes(psb_intel_output);
+	psb_intel_ddc_get_modes(connector, &lvds_priv->ddc_bus->adapter);
 	list_for_each_entry(scan, &connector->probed_modes, head) {
 		if (scan->type & DRM_MODE_TYPE_PREFERRED) {
 			mode_dev->panel_fixed_mode =
@@ -841,14 +853,16 @@
 	return;
 
 failed_find:
-	if (psb_intel_output->ddc_bus)
-		psb_intel_i2c_destroy(psb_intel_output->ddc_bus);
+	if (lvds_priv->ddc_bus)
+		psb_intel_i2c_destroy(lvds_priv->ddc_bus);
 failed_ddc:
-	if (psb_intel_output->i2c_bus)
-		psb_intel_i2c_destroy(psb_intel_output->i2c_bus);
+	if (lvds_priv->i2c_bus)
+		psb_intel_i2c_destroy(lvds_priv->i2c_bus);
 failed_blc_i2c:
 	drm_encoder_cleanup(encoder);
 	drm_connector_cleanup(connector);
-	kfree(connector);
+failed_connector:
+	if (psb_intel_connector)
+		kfree(psb_intel_connector);
 }
 
diff --git a/drivers/gpu/drm/gma500/psb_intel_modes.c b/drivers/gpu/drm/gma500/psb_intel_modes.c
index bde1aff..4fca0d6 100644
--- a/drivers/gpu/drm/gma500/psb_intel_modes.c
+++ b/drivers/gpu/drm/gma500/psb_intel_modes.c
@@ -26,7 +26,7 @@
  * psb_intel_ddc_probe
  *
  */
-bool psb_intel_ddc_probe(struct psb_intel_output *psb_intel_output)
+bool psb_intel_ddc_probe(struct i2c_adapter *adapter)
 {
 	u8 out_buf[] = { 0x0, 0x0 };
 	u8 buf[2];
@@ -46,7 +46,7 @@
 		 }
 	};
 
-	ret = i2c_transfer(&psb_intel_output->ddc_bus->adapter, msgs, 2);
+	ret = i2c_transfer(adapter, msgs, 2);
 	if (ret == 2)
 		return true;
 
@@ -59,18 +59,16 @@
  *
  * Fetch the EDID information from @connector using the DDC bus.
  */
-int psb_intel_ddc_get_modes(struct psb_intel_output *psb_intel_output)
+int psb_intel_ddc_get_modes(struct drm_connector *connector,
+			    struct i2c_adapter *adapter)
 {
 	struct edid *edid;
 	int ret = 0;
 
-	edid =
-	    drm_get_edid(&psb_intel_output->base,
-			 &psb_intel_output->ddc_bus->adapter);
+	edid = drm_get_edid(connector, adapter);
 	if (edid) {
-		drm_mode_connector_update_edid_property(&psb_intel_output->
-							base, edid);
-		ret = drm_add_edid_modes(&psb_intel_output->base, edid);
+		drm_mode_connector_update_edid_property(connector, edid);
+		ret = drm_add_edid_modes(connector, edid);
 		kfree(edid);
 	}
 	return ret;
diff --git a/drivers/gpu/drm/gma500/psb_intel_reg.h b/drivers/gpu/drm/gma500/psb_intel_reg.h
index 1ac16aa..fcc0af0 100644
--- a/drivers/gpu/drm/gma500/psb_intel_reg.h
+++ b/drivers/gpu/drm/gma500/psb_intel_reg.h
@@ -17,6 +17,78 @@
 #ifndef __PSB_INTEL_REG_H__
 #define __PSB_INTEL_REG_H__
 
+/*
+ * GPIO regs
+ */
+#define GPIOA			0x5010
+#define GPIOB			0x5014
+#define GPIOC			0x5018
+#define GPIOD			0x501c
+#define GPIOE			0x5020
+#define GPIOF			0x5024
+#define GPIOG			0x5028
+#define GPIOH			0x502c
+# define GPIO_CLOCK_DIR_MASK		(1 << 0)
+# define GPIO_CLOCK_DIR_IN		(0 << 1)
+# define GPIO_CLOCK_DIR_OUT		(1 << 1)
+# define GPIO_CLOCK_VAL_MASK		(1 << 2)
+# define GPIO_CLOCK_VAL_OUT		(1 << 3)
+# define GPIO_CLOCK_VAL_IN		(1 << 4)
+# define GPIO_CLOCK_PULLUP_DISABLE	(1 << 5)
+# define GPIO_DATA_DIR_MASK		(1 << 8)
+# define GPIO_DATA_DIR_IN		(0 << 9)
+# define GPIO_DATA_DIR_OUT		(1 << 9)
+# define GPIO_DATA_VAL_MASK		(1 << 10)
+# define GPIO_DATA_VAL_OUT		(1 << 11)
+# define GPIO_DATA_VAL_IN		(1 << 12)
+# define GPIO_DATA_PULLUP_DISABLE	(1 << 13)
+
+#define GMBUS0			0x5100 /* clock/port select */
+#define   GMBUS_RATE_100KHZ	(0<<8)
+#define   GMBUS_RATE_50KHZ	(1<<8)
+#define   GMBUS_RATE_400KHZ	(2<<8) /* reserved on Pineview */
+#define   GMBUS_RATE_1MHZ	(3<<8) /* reserved on Pineview */
+#define   GMBUS_HOLD_EXT	(1<<7) /* 300ns hold time, rsvd on Pineview */
+#define   GMBUS_PORT_DISABLED	0
+#define   GMBUS_PORT_SSC	1
+#define   GMBUS_PORT_VGADDC	2
+#define   GMBUS_PORT_PANEL	3
+#define   GMBUS_PORT_DPC	4 /* HDMIC */
+#define   GMBUS_PORT_DPB	5 /* SDVO, HDMIB */
+				  /* 6 reserved */
+#define   GMBUS_PORT_DPD	7 /* HDMID */
+#define   GMBUS_NUM_PORTS       8
+#define GMBUS1			0x5104 /* command/status */
+#define   GMBUS_SW_CLR_INT	(1<<31)
+#define   GMBUS_SW_RDY		(1<<30)
+#define   GMBUS_ENT		(1<<29) /* enable timeout */
+#define   GMBUS_CYCLE_NONE	(0<<25)
+#define   GMBUS_CYCLE_WAIT	(1<<25)
+#define   GMBUS_CYCLE_INDEX	(2<<25)
+#define   GMBUS_CYCLE_STOP	(4<<25)
+#define   GMBUS_BYTE_COUNT_SHIFT 16
+#define   GMBUS_SLAVE_INDEX_SHIFT 8
+#define   GMBUS_SLAVE_ADDR_SHIFT 1
+#define   GMBUS_SLAVE_READ	(1<<0)
+#define   GMBUS_SLAVE_WRITE	(0<<0)
+#define GMBUS2			0x5108 /* status */
+#define   GMBUS_INUSE		(1<<15)
+#define   GMBUS_HW_WAIT_PHASE	(1<<14)
+#define   GMBUS_STALL_TIMEOUT	(1<<13)
+#define   GMBUS_INT		(1<<12)
+#define   GMBUS_HW_RDY		(1<<11)
+#define   GMBUS_SATOER		(1<<10)
+#define   GMBUS_ACTIVE		(1<<9)
+#define GMBUS3			0x510c /* data buffer bytes 3-0 */
+#define GMBUS4			0x5110 /* interrupt mask (Pineview+) */
+#define   GMBUS_SLAVE_TIMEOUT_EN (1<<4)
+#define   GMBUS_NAK_EN		(1<<3)
+#define   GMBUS_IDLE_EN		(1<<2)
+#define   GMBUS_HW_WAIT_EN	(1<<1)
+#define   GMBUS_HW_RDY_EN	(1<<0)
+#define GMBUS5			0x5120 /* byte index */
+#define   GMBUS_2BYTE_INDEX_EN	(1<<31)
+
 #define BLC_PWM_CTL		0x61254
 #define BLC_PWM_CTL2		0x61250
 #define BLC_PWM_CTL_C		0x62254
@@ -304,6 +376,8 @@
 #define SDVO_PIPE_B_SELECT		(1 << 30)
 #define SDVO_STALL_SELECT		(1 << 29)
 #define SDVO_INTERRUPT_ENABLE		(1 << 26)
+#define SDVO_COLOR_RANGE_16_235		(1 << 8)
+#define SDVO_AUDIO_ENABLE		(1 << 6)
 
 /**
  * 915G/GM SDVO pixel multiplier.
diff --git a/drivers/gpu/drm/gma500/psb_intel_sdvo.c b/drivers/gpu/drm/gma500/psb_intel_sdvo.c
index a4bad1a..4882b29 100644
--- a/drivers/gpu/drm/gma500/psb_intel_sdvo.c
+++ b/drivers/gpu/drm/gma500/psb_intel_sdvo.c
@@ -1,76 +1,245 @@
 /*
- * Copyright (c) 2006-2007 Intel Corporation
+ * Copyright 2006 Dave Airlie <airlied@linux.ie>
+ * Copyright © 2006-2007 Intel Corporation
+ *   Jesse Barnes <jesse.barnes@intel.com>
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
  *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
  *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
  *
  * Authors:
  *	Eric Anholt <eric@anholt.net>
  */
-
+#include <linux/module.h>
 #include <linux/i2c.h>
+#include <linux/slab.h>
 #include <linux/delay.h>
-/* #include <drm/drm_crtc.h> */
-#include <drm/drmP.h>
-#include "psb_drv.h"
+#include "drmP.h"
+#include "drm.h"
+#include "drm_crtc.h"
+#include "drm_edid.h"
 #include "psb_intel_drv.h"
-#include "psb_intel_reg.h"
+#include "gma_drm.h"
+#include "psb_drv.h"
 #include "psb_intel_sdvo_regs.h"
+#include "psb_intel_reg.h"
 
-struct psb_intel_sdvo_priv {
-	struct psb_intel_i2c_chan *i2c_bus;
-	int slaveaddr;
-	int output_device;
+#define SDVO_TMDS_MASK (SDVO_OUTPUT_TMDS0 | SDVO_OUTPUT_TMDS1)
+#define SDVO_RGB_MASK  (SDVO_OUTPUT_RGB0 | SDVO_OUTPUT_RGB1)
+#define SDVO_LVDS_MASK (SDVO_OUTPUT_LVDS0 | SDVO_OUTPUT_LVDS1)
+#define SDVO_TV_MASK   (SDVO_OUTPUT_CVBS0 | SDVO_OUTPUT_SVID0)
 
-	u16 active_outputs;
+#define SDVO_OUTPUT_MASK (SDVO_TMDS_MASK | SDVO_RGB_MASK | SDVO_LVDS_MASK |\
+                         SDVO_TV_MASK)
 
+#define IS_TV(c)	(c->output_flag & SDVO_TV_MASK)
+#define IS_TMDS(c)	(c->output_flag & SDVO_TMDS_MASK)
+#define IS_LVDS(c)	(c->output_flag & SDVO_LVDS_MASK)
+#define IS_TV_OR_LVDS(c) (c->output_flag & (SDVO_TV_MASK | SDVO_LVDS_MASK))
+
+
+static const char *tv_format_names[] = {
+	"NTSC_M"   , "NTSC_J"  , "NTSC_443",
+	"PAL_B"    , "PAL_D"   , "PAL_G"   ,
+	"PAL_H"    , "PAL_I"   , "PAL_M"   ,
+	"PAL_N"    , "PAL_NC"  , "PAL_60"  ,
+	"SECAM_B"  , "SECAM_D" , "SECAM_G" ,
+	"SECAM_K"  , "SECAM_K1", "SECAM_L" ,
+	"SECAM_60"
+};
+
+#define TV_FORMAT_NUM  (sizeof(tv_format_names) / sizeof(*tv_format_names))
+
+struct psb_intel_sdvo {
+	struct psb_intel_encoder base;
+
+	struct i2c_adapter *i2c;
+	u8 slave_addr;
+
+	struct i2c_adapter ddc;
+
+	/* Register for the SDVO device: SDVOB or SDVOC */
+	int sdvo_reg;
+
+	/* Active outputs controlled by this SDVO output */
+	uint16_t controlled_output;
+
+	/*
+	 * Capabilities of the SDVO device returned by
+	 * i830_sdvo_get_capabilities()
+	 */
 	struct psb_intel_sdvo_caps caps;
+
+	/* Pixel clock limitations reported by the SDVO device, in kHz */
 	int pixel_clock_min, pixel_clock_max;
 
-	int save_sdvo_mult;
-	u16 save_active_outputs;
-	struct psb_intel_sdvo_dtd save_input_dtd_1, save_input_dtd_2;
-	struct psb_intel_sdvo_dtd save_output_dtd[16];
-	u32 save_SDVOX;
-	u8 in_out_map[4];
+	/*
+	* For multiple function SDVO device,
+	* this is for current attached outputs.
+	*/
+	uint16_t attached_output;
 
-	u8 by_input_wiring;
-	u32 active_device;
+	/**
+	 * This is used to select the color range of RBG outputs in HDMI mode.
+	 * It is only valid when using TMDS encoding and 8 bit per color mode.
+	 */
+	uint32_t color_range;
+
+	/**
+	 * This is set if we're going to treat the device as TV-out.
+	 *
+	 * While we have these nice friendly flags for output types that ought
+	 * to decide this for us, the S-Video output on our HDMI+S-Video card
+	 * shows up as RGB1 (VGA).
+	 */
+	bool is_tv;
+
+	/* This is for current tv format name */
+	int tv_format_index;
+
+	/**
+	 * This is set if we treat the device as HDMI, instead of DVI.
+	 */
+	bool is_hdmi;
+	bool has_hdmi_monitor;
+	bool has_hdmi_audio;
+
+	/**
+	 * This is set if we detect output of sdvo device as LVDS and
+	 * have a valid fixed mode to use with the panel.
+	 */
+	bool is_lvds;
+
+	/**
+	 * This is sdvo fixed pannel mode pointer
+	 */
+	struct drm_display_mode *sdvo_lvds_fixed_mode;
+
+	/* DDC bus used by this SDVO encoder */
+	uint8_t ddc_bus;
+
+	/* Input timings for adjusted_mode */
+	struct psb_intel_sdvo_dtd input_dtd;
 };
 
+struct psb_intel_sdvo_connector {
+	struct psb_intel_connector base;
+
+	/* Mark the type of connector */
+	uint16_t output_flag;
+
+	int force_audio;
+
+	/* This contains all current supported TV format */
+	u8 tv_format_supported[TV_FORMAT_NUM];
+	int   format_supported_num;
+	struct drm_property *tv_format;
+
+	/* add the property for the SDVO-TV */
+	struct drm_property *left;
+	struct drm_property *right;
+	struct drm_property *top;
+	struct drm_property *bottom;
+	struct drm_property *hpos;
+	struct drm_property *vpos;
+	struct drm_property *contrast;
+	struct drm_property *saturation;
+	struct drm_property *hue;
+	struct drm_property *sharpness;
+	struct drm_property *flicker_filter;
+	struct drm_property *flicker_filter_adaptive;
+	struct drm_property *flicker_filter_2d;
+	struct drm_property *tv_chroma_filter;
+	struct drm_property *tv_luma_filter;
+	struct drm_property *dot_crawl;
+
+	/* add the property for the SDVO-TV/LVDS */
+	struct drm_property *brightness;
+
+	/* Add variable to record current setting for the above property */
+	u32	left_margin, right_margin, top_margin, bottom_margin;
+
+	/* this is to get the range of margin.*/
+	u32	max_hscan,  max_vscan;
+	u32	max_hpos, cur_hpos;
+	u32	max_vpos, cur_vpos;
+	u32	cur_brightness, max_brightness;
+	u32	cur_contrast,	max_contrast;
+	u32	cur_saturation, max_saturation;
+	u32	cur_hue,	max_hue;
+	u32	cur_sharpness,	max_sharpness;
+	u32	cur_flicker_filter,		max_flicker_filter;
+	u32	cur_flicker_filter_adaptive,	max_flicker_filter_adaptive;
+	u32	cur_flicker_filter_2d,		max_flicker_filter_2d;
+	u32	cur_tv_chroma_filter,	max_tv_chroma_filter;
+	u32	cur_tv_luma_filter,	max_tv_luma_filter;
+	u32	cur_dot_crawl,	max_dot_crawl;
+};
+
+static struct psb_intel_sdvo *to_psb_intel_sdvo(struct drm_encoder *encoder)
+{
+	return container_of(encoder, struct psb_intel_sdvo, base.base);
+}
+
+static struct psb_intel_sdvo *intel_attached_sdvo(struct drm_connector *connector)
+{
+	return container_of(psb_intel_attached_encoder(connector),
+			    struct psb_intel_sdvo, base);
+}
+
+static struct psb_intel_sdvo_connector *to_psb_intel_sdvo_connector(struct drm_connector *connector)
+{
+	return container_of(to_psb_intel_connector(connector), struct psb_intel_sdvo_connector, base);
+}
+
+static bool
+psb_intel_sdvo_output_setup(struct psb_intel_sdvo *psb_intel_sdvo, uint16_t flags);
+static bool
+psb_intel_sdvo_tv_create_property(struct psb_intel_sdvo *psb_intel_sdvo,
+			      struct psb_intel_sdvo_connector *psb_intel_sdvo_connector,
+			      int type);
+static bool
+psb_intel_sdvo_create_enhance_property(struct psb_intel_sdvo *psb_intel_sdvo,
+				   struct psb_intel_sdvo_connector *psb_intel_sdvo_connector);
+
 /**
  * Writes the SDVOB or SDVOC with the given value, but always writes both
  * SDVOB and SDVOC to work around apparent hardware issues (according to
  * comments in the BIOS).
  */
-void psb_intel_sdvo_write_sdvox(struct psb_intel_output *psb_intel_output,
-				u32 val)
+static void psb_intel_sdvo_write_sdvox(struct psb_intel_sdvo *psb_intel_sdvo, u32 val)
 {
-	struct drm_device *dev = psb_intel_output->base.dev;
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
+	struct drm_device *dev = psb_intel_sdvo->base.base.dev;
 	u32 bval = val, cval = val;
 	int i;
 
-	if (sdvo_priv->output_device == SDVOB)
+	if (psb_intel_sdvo->sdvo_reg == SDVOB) {
 		cval = REG_READ(SDVOC);
-	else
+	} else {
 		bval = REG_READ(SDVOB);
+	}
 	/*
 	 * Write the registers twice for luck. Sometimes,
 	 * writing them only once doesn't appear to 'stick'.
 	 * The BIOS does this too. Yay, magic
 	 */
-	for (i = 0; i < 2; i++) {
+	for (i = 0; i < 2; i++)
+	{
 		REG_WRITE(SDVOB, bval);
 		REG_READ(SDVOB);
 		REG_WRITE(SDVOC, cval);
@@ -78,61 +247,28 @@
 	}
 }
 
-static bool psb_intel_sdvo_read_byte(
-				struct psb_intel_output *psb_intel_output,
-				u8 addr, u8 *ch)
+static bool psb_intel_sdvo_read_byte(struct psb_intel_sdvo *psb_intel_sdvo, u8 addr, u8 *ch)
 {
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
-	u8 out_buf[2];
-	u8 buf[2];
+	struct i2c_msg msgs[] = {
+		{
+			.addr = psb_intel_sdvo->slave_addr,
+			.flags = 0,
+			.len = 1,
+			.buf = &addr,
+		},
+		{
+			.addr = psb_intel_sdvo->slave_addr,
+			.flags = I2C_M_RD,
+			.len = 1,
+			.buf = ch,
+		}
+	};
 	int ret;
 
-	struct i2c_msg msgs[] = {
-		{
-		 .addr = sdvo_priv->i2c_bus->slave_addr,
-		 .flags = 0,
-		 .len = 1,
-		 .buf = out_buf,
-		 },
-		{
-		 .addr = sdvo_priv->i2c_bus->slave_addr,
-		 .flags = I2C_M_RD,
-		 .len = 1,
-		 .buf = buf,
-		 }
-	};
-
-	out_buf[0] = addr;
-	out_buf[1] = 0;
-
-	ret = i2c_transfer(&sdvo_priv->i2c_bus->adapter, msgs, 2);
-	if (ret == 2) {
-		*ch = buf[0];
+	if ((ret = i2c_transfer(psb_intel_sdvo->i2c, msgs, 2)) == 2)
 		return true;
-	}
 
-	return false;
-}
-
-static bool psb_intel_sdvo_write_byte(
-			struct psb_intel_output *psb_intel_output,
-			int addr, u8 ch)
-{
-	u8 out_buf[2];
-	struct i2c_msg msgs[] = {
-		{
-		 .addr = psb_intel_output->i2c_bus->slave_addr,
-		 .flags = 0,
-		 .len = 2,
-		 .buf = out_buf,
-		 }
-	};
-
-	out_buf[0] = addr;
-	out_buf[1] = ch;
-
-	if (i2c_transfer(&psb_intel_output->i2c_bus->adapter, msgs, 1) == 1)
-		return true;
+	DRM_DEBUG_KMS("i2c transfer returned %d\n", ret);
 	return false;
 }
 
@@ -140,98 +276,147 @@
 /** Mapping of command numbers to names, for debug output */
 static const struct _sdvo_cmd_name {
 	u8 cmd;
-	char *name;
+	const char *name;
 } sdvo_cmd_names[] = {
-SDVO_CMD_NAME_ENTRY(SDVO_CMD_RESET),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_DEVICE_CAPS),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_FIRMWARE_REV),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_TRAINED_INPUTS),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_ACTIVE_OUTPUTS),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_ACTIVE_OUTPUTS),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_IN_OUT_MAP),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_IN_OUT_MAP),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_ATTACHED_DISPLAYS),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HOT_PLUG_SUPPORT),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_ACTIVE_HOT_PLUG),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_ACTIVE_HOT_PLUG),
-	    SDVO_CMD_NAME_ENTRY
-	    (SDVO_CMD_GET_INTERRUPT_EVENT_SOURCE),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_TARGET_INPUT),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_TARGET_OUTPUT),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_INPUT_TIMINGS_PART1),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_INPUT_TIMINGS_PART2),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_INPUT_TIMINGS_PART1),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_INPUT_TIMINGS_PART2),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_INPUT_TIMINGS_PART1),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_OUTPUT_TIMINGS_PART1),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_OUTPUT_TIMINGS_PART2),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_OUTPUT_TIMINGS_PART1),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_OUTPUT_TIMINGS_PART2),
-	    SDVO_CMD_NAME_ENTRY
-	    (SDVO_CMD_CREATE_PREFERRED_INPUT_TIMING),
-	    SDVO_CMD_NAME_ENTRY
-	    (SDVO_CMD_GET_PREFERRED_INPUT_TIMING_PART1),
-	    SDVO_CMD_NAME_ENTRY
-	    (SDVO_CMD_GET_PREFERRED_INPUT_TIMING_PART2),
-	    SDVO_CMD_NAME_ENTRY
-	    (SDVO_CMD_GET_INPUT_PIXEL_CLOCK_RANGE),
-	    SDVO_CMD_NAME_ENTRY
-	    (SDVO_CMD_GET_OUTPUT_PIXEL_CLOCK_RANGE),
-	    SDVO_CMD_NAME_ENTRY
-	    (SDVO_CMD_GET_SUPPORTED_CLOCK_RATE_MULTS),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_CLOCK_RATE_MULT),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_CLOCK_RATE_MULT),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SUPPORTED_TV_FORMATS),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_TV_FORMAT),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_TV_FORMAT),
-	    SDVO_CMD_NAME_ENTRY
-	    (SDVO_CMD_SET_TV_RESOLUTION_SUPPORT),
-	    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_CONTROL_BUS_SWITCH),};
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_RESET),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_DEVICE_CAPS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_FIRMWARE_REV),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_TRAINED_INPUTS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_ACTIVE_OUTPUTS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_ACTIVE_OUTPUTS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_IN_OUT_MAP),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_IN_OUT_MAP),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_ATTACHED_DISPLAYS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HOT_PLUG_SUPPORT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_ACTIVE_HOT_PLUG),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_ACTIVE_HOT_PLUG),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_INTERRUPT_EVENT_SOURCE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_TARGET_INPUT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_TARGET_OUTPUT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_INPUT_TIMINGS_PART1),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_INPUT_TIMINGS_PART2),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_INPUT_TIMINGS_PART1),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_INPUT_TIMINGS_PART2),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_INPUT_TIMINGS_PART1),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_OUTPUT_TIMINGS_PART1),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_OUTPUT_TIMINGS_PART2),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_OUTPUT_TIMINGS_PART1),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_OUTPUT_TIMINGS_PART2),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_CREATE_PREFERRED_INPUT_TIMING),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_PREFERRED_INPUT_TIMING_PART1),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_PREFERRED_INPUT_TIMING_PART2),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_INPUT_PIXEL_CLOCK_RANGE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_OUTPUT_PIXEL_CLOCK_RANGE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SUPPORTED_CLOCK_RATE_MULTS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_CLOCK_RATE_MULT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_CLOCK_RATE_MULT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SUPPORTED_TV_FORMATS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_TV_FORMAT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_TV_FORMAT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SUPPORTED_POWER_STATES),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_POWER_STATE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_ENCODER_POWER_STATE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_DISPLAY_POWER_STATE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_CONTROL_BUS_SWITCH),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SDTV_RESOLUTION_SUPPORT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SCALED_HDTV_RESOLUTION_SUPPORT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SUPPORTED_ENHANCEMENTS),
 
-#define SDVO_NAME(dev_priv) \
-		 ((dev_priv)->output_device == SDVOB ? "SDVOB" : "SDVOC")
-#define SDVO_PRIV(output)   ((struct psb_intel_sdvo_priv *) (output)->dev_priv)
+    /* Add the op code for SDVO enhancements */
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_HPOS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HPOS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_HPOS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_VPOS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_VPOS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_VPOS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_SATURATION),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SATURATION),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_SATURATION),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_HUE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HUE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_HUE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_CONTRAST),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_CONTRAST),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_CONTRAST),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_BRIGHTNESS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_BRIGHTNESS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_BRIGHTNESS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_OVERSCAN_H),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_OVERSCAN_H),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_OVERSCAN_H),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_OVERSCAN_V),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_OVERSCAN_V),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_OVERSCAN_V),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_FLICKER_FILTER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_FLICKER_FILTER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_FLICKER_FILTER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_FLICKER_FILTER_ADAPTIVE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_FLICKER_FILTER_ADAPTIVE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_FLICKER_FILTER_ADAPTIVE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_FLICKER_FILTER_2D),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_FLICKER_FILTER_2D),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_FLICKER_FILTER_2D),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_SHARPNESS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SHARPNESS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_SHARPNESS),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_DOT_CRAWL),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_DOT_CRAWL),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_TV_CHROMA_FILTER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_TV_CHROMA_FILTER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_TV_CHROMA_FILTER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_MAX_TV_LUMA_FILTER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_TV_LUMA_FILTER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_TV_LUMA_FILTER),
 
-static void psb_intel_sdvo_write_cmd(struct psb_intel_output *psb_intel_output,
-				     u8 cmd,
-				     void *args,
-				     int args_len)
+    /* HDMI op code */
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_SUPP_ENCODE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_ENCODE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_ENCODE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_PIXEL_REPLI),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_PIXEL_REPLI),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_COLORIMETRY_CAP),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_COLORIMETRY),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_COLORIMETRY),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_AUDIO_ENCRYPT_PREFER),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_AUDIO_STAT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_AUDIO_STAT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HBUF_INDEX),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_HBUF_INDEX),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HBUF_INFO),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HBUF_AV_SPLIT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_HBUF_AV_SPLIT),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HBUF_TXRATE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_HBUF_TXRATE),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_SET_HBUF_DATA),
+    SDVO_CMD_NAME_ENTRY(SDVO_CMD_GET_HBUF_DATA),
+};
+
+#define IS_SDVOB(reg)	(reg == SDVOB)
+#define SDVO_NAME(svdo) (IS_SDVOB((svdo)->sdvo_reg) ? "SDVOB" : "SDVOC")
+
+static void psb_intel_sdvo_debug_write(struct psb_intel_sdvo *psb_intel_sdvo, u8 cmd,
+				   const void *args, int args_len)
 {
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
 	int i;
 
-	if (0) {
-		printk(KERN_DEBUG "%s: W: %02X ", SDVO_NAME(sdvo_priv), cmd);
-		for (i = 0; i < args_len; i++)
-			printk(KERN_CONT "%02X ", ((u8 *) args)[i]);
-		for (; i < 8; i++)
-			printk(KERN_CONT "   ");
-		for (i = 0;
-		     i <
-		     sizeof(sdvo_cmd_names) / sizeof(sdvo_cmd_names[0]);
-		     i++) {
-			if (cmd == sdvo_cmd_names[i].cmd) {
-				printk(KERN_CONT
-					"(%s)", sdvo_cmd_names[i].name);
-				break;
-			}
+	DRM_DEBUG_KMS("%s: W: %02X ",
+				SDVO_NAME(psb_intel_sdvo), cmd);
+	for (i = 0; i < args_len; i++)
+		DRM_LOG_KMS("%02X ", ((u8 *)args)[i]);
+	for (; i < 8; i++)
+		DRM_LOG_KMS("   ");
+	for (i = 0; i < ARRAY_SIZE(sdvo_cmd_names); i++) {
+		if (cmd == sdvo_cmd_names[i].cmd) {
+			DRM_LOG_KMS("(%s)", sdvo_cmd_names[i].name);
+			break;
 		}
-		if (i ==
-		    sizeof(sdvo_cmd_names) / sizeof(sdvo_cmd_names[0]))
-			printk(KERN_CONT "(%02X)", cmd);
-		printk(KERN_CONT "\n");
 	}
-
-	for (i = 0; i < args_len; i++) {
-		psb_intel_sdvo_write_byte(psb_intel_output,
-					SDVO_I2C_ARG_0 - i,
-					((u8 *) args)[i]);
-	}
-
-	psb_intel_sdvo_write_byte(psb_intel_output, SDVO_I2C_OPCODE, cmd);
+	if (i == ARRAY_SIZE(sdvo_cmd_names))
+		DRM_LOG_KMS("(%02X)", cmd);
+	DRM_LOG_KMS("\n");
 }
 
-static const char *const cmd_status_names[] = {
+static const char *cmd_status_names[] = {
 	"Power on",
 	"Success",
 	"Not supported",
@@ -241,52 +426,111 @@
 	"Scaling not supported"
 };
 
-static u8 psb_intel_sdvo_read_response(
-				struct psb_intel_output *psb_intel_output,
-				void *response, int response_len)
+static bool psb_intel_sdvo_write_cmd(struct psb_intel_sdvo *psb_intel_sdvo, u8 cmd,
+				 const void *args, int args_len)
 {
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
-	int i;
-	u8 status;
-	u8 retry = 50;
+	u8 buf[args_len*2 + 2], status;
+	struct i2c_msg msgs[args_len + 3];
+	int i, ret;
 
-	while (retry--) {
-		/* Read the command response */
-		for (i = 0; i < response_len; i++) {
-			psb_intel_sdvo_read_byte(psb_intel_output,
-					     SDVO_I2C_RETURN_0 + i,
-					     &((u8 *) response)[i]);
-		}
+	psb_intel_sdvo_debug_write(psb_intel_sdvo, cmd, args, args_len);
 
-		/* read the return status */
-		psb_intel_sdvo_read_byte(psb_intel_output,
-					 SDVO_I2C_CMD_STATUS,
-					 &status);
+	for (i = 0; i < args_len; i++) {
+		msgs[i].addr = psb_intel_sdvo->slave_addr;
+		msgs[i].flags = 0;
+		msgs[i].len = 2;
+		msgs[i].buf = buf + 2 *i;
+		buf[2*i + 0] = SDVO_I2C_ARG_0 - i;
+		buf[2*i + 1] = ((u8*)args)[i];
+	}
+	msgs[i].addr = psb_intel_sdvo->slave_addr;
+	msgs[i].flags = 0;
+	msgs[i].len = 2;
+	msgs[i].buf = buf + 2*i;
+	buf[2*i + 0] = SDVO_I2C_OPCODE;
+	buf[2*i + 1] = cmd;
 
-		if (0) {
-			pr_debug("%s: R: ", SDVO_NAME(sdvo_priv));
-			for (i = 0; i < response_len; i++)
-				printk(KERN_CONT "%02X ", ((u8 *) response)[i]);
-			for (; i < 8; i++)
-				printk("   ");
-			if (status <= SDVO_CMD_STATUS_SCALING_NOT_SUPP)
-				printk(KERN_CONT "(%s)",
-					 cmd_status_names[status]);
-			else
-				printk(KERN_CONT "(??? %d)", status);
-			printk(KERN_CONT "\n");
-		}
+	/* the following two are to read the response */
+	status = SDVO_I2C_CMD_STATUS;
+	msgs[i+1].addr = psb_intel_sdvo->slave_addr;
+	msgs[i+1].flags = 0;
+	msgs[i+1].len = 1;
+	msgs[i+1].buf = &status;
 
-		if (status != SDVO_CMD_STATUS_PENDING)
-			return status;
+	msgs[i+2].addr = psb_intel_sdvo->slave_addr;
+	msgs[i+2].flags = I2C_M_RD;
+	msgs[i+2].len = 1;
+	msgs[i+2].buf = &status;
 
-		mdelay(50);
+	ret = i2c_transfer(psb_intel_sdvo->i2c, msgs, i+3);
+	if (ret < 0) {
+		DRM_DEBUG_KMS("I2c transfer returned %d\n", ret);
+		return false;
+	}
+	if (ret != i+3) {
+		/* failure in I2C transfer */
+		DRM_DEBUG_KMS("I2c transfer returned %d/%d\n", ret, i+3);
+		return false;
 	}
 
-	return status;
+	return true;
 }
 
-int psb_intel_sdvo_get_pixel_multiplier(struct drm_display_mode *mode)
+static bool psb_intel_sdvo_read_response(struct psb_intel_sdvo *psb_intel_sdvo,
+				     void *response, int response_len)
+{
+	u8 retry = 5;
+	u8 status;
+	int i;
+
+	DRM_DEBUG_KMS("%s: R: ", SDVO_NAME(psb_intel_sdvo));
+
+	/*
+	 * The documentation states that all commands will be
+	 * processed within 15µs, and that we need only poll
+	 * the status byte a maximum of 3 times in order for the
+	 * command to be complete.
+	 *
+	 * Check 5 times in case the hardware failed to read the docs.
+	 */
+	if (!psb_intel_sdvo_read_byte(psb_intel_sdvo,
+				  SDVO_I2C_CMD_STATUS,
+				  &status))
+		goto log_fail;
+
+	while (status == SDVO_CMD_STATUS_PENDING && retry--) {
+		udelay(15);
+		if (!psb_intel_sdvo_read_byte(psb_intel_sdvo,
+					  SDVO_I2C_CMD_STATUS,
+					  &status))
+			goto log_fail;
+	}
+
+	if (status <= SDVO_CMD_STATUS_SCALING_NOT_SUPP)
+		DRM_LOG_KMS("(%s)", cmd_status_names[status]);
+	else
+		DRM_LOG_KMS("(??? %d)", status);
+
+	if (status != SDVO_CMD_STATUS_SUCCESS)
+		goto log_fail;
+
+	/* Read the command response */
+	for (i = 0; i < response_len; i++) {
+		if (!psb_intel_sdvo_read_byte(psb_intel_sdvo,
+					  SDVO_I2C_RETURN_0 + i,
+					  &((u8 *)response)[i]))
+			goto log_fail;
+		DRM_LOG_KMS(" %02X", ((u8 *)response)[i]);
+	}
+	DRM_LOG_KMS("\n");
+	return true;
+
+log_fail:
+	DRM_LOG_KMS("... failed\n");
+	return false;
+}
+
+static int psb_intel_sdvo_get_pixel_multiplier(struct drm_display_mode *mode)
 {
 	if (mode->clock >= 100000)
 		return 1;
@@ -296,40 +540,38 @@
 		return 4;
 }
 
-/**
- * Don't check status code from this as it switches the bus back to the
- * SDVO chips which defeats the purpose of doing a bus switch in the first
- * place.
- */
-void psb_intel_sdvo_set_control_bus_switch(
-				struct psb_intel_output *psb_intel_output,
-				u8 target)
+static bool psb_intel_sdvo_set_control_bus_switch(struct psb_intel_sdvo *psb_intel_sdvo,
+					      u8 ddc_bus)
 {
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				 SDVO_CMD_SET_CONTROL_BUS_SWITCH,
-				 &target,
-				 1);
+	/* This must be the immediately preceding write before the i2c xfer */
+	return psb_intel_sdvo_write_cmd(psb_intel_sdvo,
+				    SDVO_CMD_SET_CONTROL_BUS_SWITCH,
+				    &ddc_bus, 1);
 }
 
-static bool psb_intel_sdvo_set_target_input(
-				struct psb_intel_output *psb_intel_output,
-				bool target_0, bool target_1)
+static bool psb_intel_sdvo_set_value(struct psb_intel_sdvo *psb_intel_sdvo, u8 cmd, const void *data, int len)
 {
-	struct psb_intel_sdvo_set_target_input_args targets = { 0 };
-	u8 status;
+	if (!psb_intel_sdvo_write_cmd(psb_intel_sdvo, cmd, data, len))
+		return false;
 
-	if (target_0 && target_1)
-		return SDVO_CMD_STATUS_NOTSUPP;
+	return psb_intel_sdvo_read_response(psb_intel_sdvo, NULL, 0);
+}
 
-	if (target_1)
-		targets.target_1 = 1;
+static bool
+psb_intel_sdvo_get_value(struct psb_intel_sdvo *psb_intel_sdvo, u8 cmd, void *value, int len)
+{
+	if (!psb_intel_sdvo_write_cmd(psb_intel_sdvo, cmd, NULL, 0))
+		return false;
 
-	psb_intel_sdvo_write_cmd(psb_intel_output, SDVO_CMD_SET_TARGET_INPUT,
-			     &targets, sizeof(targets));
+	return psb_intel_sdvo_read_response(psb_intel_sdvo, value, len);
+}
 
-	status = psb_intel_sdvo_read_response(psb_intel_output, NULL, 0);
-
-	return status == SDVO_CMD_STATUS_SUCCESS;
+static bool psb_intel_sdvo_set_target_input(struct psb_intel_sdvo *psb_intel_sdvo)
+{
+	struct psb_intel_sdvo_set_target_input_args targets = {0};
+	return psb_intel_sdvo_set_value(psb_intel_sdvo,
+				    SDVO_CMD_SET_TARGET_INPUT,
+				    &targets, sizeof(targets));
 }
 
 /**
@@ -338,19 +580,13 @@
  * This function is making an assumption about the layout of the response,
  * which should be checked against the docs.
  */
-static bool psb_intel_sdvo_get_trained_inputs(struct psb_intel_output
-					  *psb_intel_output, bool *input_1,
-					  bool *input_2)
+static bool psb_intel_sdvo_get_trained_inputs(struct psb_intel_sdvo *psb_intel_sdvo, bool *input_1, bool *input_2)
 {
 	struct psb_intel_sdvo_get_trained_inputs_response response;
-	u8 status;
 
-	psb_intel_sdvo_write_cmd(psb_intel_output, SDVO_CMD_GET_TRAINED_INPUTS,
-			     NULL, 0);
-	status =
-	    psb_intel_sdvo_read_response(psb_intel_output, &response,
-				     sizeof(response));
-	if (status != SDVO_CMD_STATUS_SUCCESS)
+	BUILD_BUG_ON(sizeof(response) != 1);
+	if (!psb_intel_sdvo_get_value(psb_intel_sdvo, SDVO_CMD_GET_TRAINED_INPUTS,
+				  &response, sizeof(response)))
 		return false;
 
 	*input_1 = response.input0_trained;
@@ -358,35 +594,18 @@
 	return true;
 }
 
-static bool psb_intel_sdvo_get_active_outputs(struct psb_intel_output
-					  *psb_intel_output, u16 *outputs)
+static bool psb_intel_sdvo_set_active_outputs(struct psb_intel_sdvo *psb_intel_sdvo,
+					  u16 outputs)
 {
-	u8 status;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output, SDVO_CMD_GET_ACTIVE_OUTPUTS,
-			     NULL, 0);
-	status =
-	    psb_intel_sdvo_read_response(psb_intel_output, outputs,
-				     sizeof(*outputs));
-
-	return status == SDVO_CMD_STATUS_SUCCESS;
+	return psb_intel_sdvo_set_value(psb_intel_sdvo,
+				    SDVO_CMD_SET_ACTIVE_OUTPUTS,
+				    &outputs, sizeof(outputs));
 }
 
-static bool psb_intel_sdvo_set_active_outputs(struct psb_intel_output
-					  *psb_intel_output, u16 outputs)
+static bool psb_intel_sdvo_set_encoder_power_state(struct psb_intel_sdvo *psb_intel_sdvo,
+					       int mode)
 {
-	u8 status;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output, SDVO_CMD_SET_ACTIVE_OUTPUTS,
-			     &outputs, sizeof(outputs));
-	status = psb_intel_sdvo_read_response(psb_intel_output, NULL, 0);
-	return status == SDVO_CMD_STATUS_SUCCESS;
-}
-
-static bool psb_intel_sdvo_set_encoder_power_state(struct psb_intel_output
-					       *psb_intel_output, int mode)
-{
-	u8 status, state = SDVO_ENCODER_STATE_ON;
+	u8 state = SDVO_ENCODER_STATE_ON;
 
 	switch (mode) {
 	case DRM_MODE_DPMS_ON:
@@ -403,284 +622,103 @@
 		break;
 	}
 
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-			     SDVO_CMD_SET_ENCODER_POWER_STATE, &state,
-			     sizeof(state));
-	status = psb_intel_sdvo_read_response(psb_intel_output, NULL, 0);
-
-	return status == SDVO_CMD_STATUS_SUCCESS;
+	return psb_intel_sdvo_set_value(psb_intel_sdvo,
+				    SDVO_CMD_SET_ENCODER_POWER_STATE, &state, sizeof(state));
 }
 
-static bool psb_intel_sdvo_get_input_pixel_clock_range(struct psb_intel_output
-						   *psb_intel_output,
+static bool psb_intel_sdvo_get_input_pixel_clock_range(struct psb_intel_sdvo *psb_intel_sdvo,
 						   int *clock_min,
 						   int *clock_max)
 {
 	struct psb_intel_sdvo_pixel_clock_range clocks;
-	u8 status;
 
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-			     SDVO_CMD_GET_INPUT_PIXEL_CLOCK_RANGE, NULL,
-			     0);
-
-	status =
-	    psb_intel_sdvo_read_response(psb_intel_output, &clocks,
-				     sizeof(clocks));
-
-	if (status != SDVO_CMD_STATUS_SUCCESS)
+	BUILD_BUG_ON(sizeof(clocks) != 4);
+	if (!psb_intel_sdvo_get_value(psb_intel_sdvo,
+				  SDVO_CMD_GET_INPUT_PIXEL_CLOCK_RANGE,
+				  &clocks, sizeof(clocks)))
 		return false;
 
 	/* Convert the values from units of 10 kHz to kHz. */
 	*clock_min = clocks.min * 10;
 	*clock_max = clocks.max * 10;
-
 	return true;
 }
 
-static bool psb_intel_sdvo_set_target_output(
-				struct psb_intel_output *psb_intel_output,
-				u16 outputs)
+static bool psb_intel_sdvo_set_target_output(struct psb_intel_sdvo *psb_intel_sdvo,
+					 u16 outputs)
 {
-	u8 status;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output, SDVO_CMD_SET_TARGET_OUTPUT,
-			     &outputs, sizeof(outputs));
-
-	status = psb_intel_sdvo_read_response(psb_intel_output, NULL, 0);
-	return status == SDVO_CMD_STATUS_SUCCESS;
+	return psb_intel_sdvo_set_value(psb_intel_sdvo,
+				    SDVO_CMD_SET_TARGET_OUTPUT,
+				    &outputs, sizeof(outputs));
 }
 
-static bool psb_intel_sdvo_get_timing(struct psb_intel_output *psb_intel_output,
-				  u8 cmd, struct psb_intel_sdvo_dtd *dtd)
+static bool psb_intel_sdvo_set_timing(struct psb_intel_sdvo *psb_intel_sdvo, u8 cmd,
+				  struct psb_intel_sdvo_dtd *dtd)
 {
-	u8 status;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output, cmd, NULL, 0);
-	status = psb_intel_sdvo_read_response(psb_intel_output, &dtd->part1,
-					  sizeof(dtd->part1));
-	if (status != SDVO_CMD_STATUS_SUCCESS)
-		return false;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output, cmd + 1, NULL, 0);
-	status = psb_intel_sdvo_read_response(psb_intel_output, &dtd->part2,
-					  sizeof(dtd->part2));
-	if (status != SDVO_CMD_STATUS_SUCCESS)
-		return false;
-
-	return true;
+	return psb_intel_sdvo_set_value(psb_intel_sdvo, cmd, &dtd->part1, sizeof(dtd->part1)) &&
+		psb_intel_sdvo_set_value(psb_intel_sdvo, cmd + 1, &dtd->part2, sizeof(dtd->part2));
 }
 
-static bool psb_intel_sdvo_get_input_timing(
-				struct psb_intel_output *psb_intel_output,
-				struct psb_intel_sdvo_dtd *dtd)
+static bool psb_intel_sdvo_set_input_timing(struct psb_intel_sdvo *psb_intel_sdvo,
+					 struct psb_intel_sdvo_dtd *dtd)
 {
-	return psb_intel_sdvo_get_timing(psb_intel_output,
-				     SDVO_CMD_GET_INPUT_TIMINGS_PART1,
-				     dtd);
+	return psb_intel_sdvo_set_timing(psb_intel_sdvo,
+				     SDVO_CMD_SET_INPUT_TIMINGS_PART1, dtd);
 }
 
-static bool psb_intel_sdvo_set_timing(
-				struct psb_intel_output *psb_intel_output,
-				u8 cmd,
-				struct psb_intel_sdvo_dtd *dtd)
+static bool psb_intel_sdvo_set_output_timing(struct psb_intel_sdvo *psb_intel_sdvo,
+					 struct psb_intel_sdvo_dtd *dtd)
 {
-	u8 status;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output, cmd, &dtd->part1,
-			     sizeof(dtd->part1));
-	status = psb_intel_sdvo_read_response(psb_intel_output, NULL, 0);
-	if (status != SDVO_CMD_STATUS_SUCCESS)
-		return false;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output, cmd + 1, &dtd->part2,
-			     sizeof(dtd->part2));
-	status = psb_intel_sdvo_read_response(psb_intel_output, NULL, 0);
-	if (status != SDVO_CMD_STATUS_SUCCESS)
-		return false;
-
-	return true;
+	return psb_intel_sdvo_set_timing(psb_intel_sdvo,
+				     SDVO_CMD_SET_OUTPUT_TIMINGS_PART1, dtd);
 }
 
-static bool psb_intel_sdvo_set_input_timing(
-				struct psb_intel_output *psb_intel_output,
-				struct psb_intel_sdvo_dtd *dtd)
+static bool
+psb_intel_sdvo_create_preferred_input_timing(struct psb_intel_sdvo *psb_intel_sdvo,
+					 uint16_t clock,
+					 uint16_t width,
+					 uint16_t height)
 {
-	return psb_intel_sdvo_set_timing(psb_intel_output,
-				     SDVO_CMD_SET_INPUT_TIMINGS_PART1,
-				     dtd);
+	struct psb_intel_sdvo_preferred_input_timing_args args;
+
+	memset(&args, 0, sizeof(args));
+	args.clock = clock;
+	args.width = width;
+	args.height = height;
+	args.interlace = 0;
+
+	if (psb_intel_sdvo->is_lvds &&
+	   (psb_intel_sdvo->sdvo_lvds_fixed_mode->hdisplay != width ||
+	    psb_intel_sdvo->sdvo_lvds_fixed_mode->vdisplay != height))
+		args.scaled = 1;
+
+	return psb_intel_sdvo_set_value(psb_intel_sdvo,
+				    SDVO_CMD_CREATE_PREFERRED_INPUT_TIMING,
+				    &args, sizeof(args));
 }
 
-static bool psb_intel_sdvo_set_output_timing(
-				struct psb_intel_output *psb_intel_output,
-				struct psb_intel_sdvo_dtd *dtd)
+static bool psb_intel_sdvo_get_preferred_input_timing(struct psb_intel_sdvo *psb_intel_sdvo,
+						  struct psb_intel_sdvo_dtd *dtd)
 {
-	return psb_intel_sdvo_set_timing(psb_intel_output,
-				     SDVO_CMD_SET_OUTPUT_TIMINGS_PART1,
-				     dtd);
+	BUILD_BUG_ON(sizeof(dtd->part1) != 8);
+	BUILD_BUG_ON(sizeof(dtd->part2) != 8);
+	return psb_intel_sdvo_get_value(psb_intel_sdvo, SDVO_CMD_GET_PREFERRED_INPUT_TIMING_PART1,
+				    &dtd->part1, sizeof(dtd->part1)) &&
+		psb_intel_sdvo_get_value(psb_intel_sdvo, SDVO_CMD_GET_PREFERRED_INPUT_TIMING_PART2,
+				     &dtd->part2, sizeof(dtd->part2));
 }
 
-static int psb_intel_sdvo_get_clock_rate_mult(struct psb_intel_output
-						*psb_intel_output)
+static bool psb_intel_sdvo_set_clock_rate_mult(struct psb_intel_sdvo *psb_intel_sdvo, u8 val)
 {
-	u8 response, status;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				 SDVO_CMD_GET_CLOCK_RATE_MULT,
-				 NULL,
-				 0);
-
-	status = psb_intel_sdvo_read_response(psb_intel_output, &response, 1);
-
-	if (status != SDVO_CMD_STATUS_SUCCESS) {
-		DRM_DEBUG("Couldn't get SDVO clock rate multiplier\n");
-		return SDVO_CLOCK_RATE_MULT_1X;
-	} else {
-		DRM_DEBUG("Current clock rate multiplier: %d\n", response);
-	}
-
-	return response;
+	return psb_intel_sdvo_set_value(psb_intel_sdvo, SDVO_CMD_SET_CLOCK_RATE_MULT, &val, 1);
 }
 
-static bool psb_intel_sdvo_set_clock_rate_mult(struct psb_intel_output
-						*psb_intel_output, u8 val)
+static void psb_intel_sdvo_get_dtd_from_mode(struct psb_intel_sdvo_dtd *dtd,
+					 const struct drm_display_mode *mode)
 {
-	u8 status;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				SDVO_CMD_SET_CLOCK_RATE_MULT,
-				&val,
-				1);
-
-	status = psb_intel_sdvo_read_response(psb_intel_output, NULL, 0);
-	if (status != SDVO_CMD_STATUS_SUCCESS)
-		return false;
-
-	return true;
-}
-
-static bool psb_sdvo_set_current_inoutmap(struct psb_intel_output *output,
-					  u32 in0outputmask,
-					  u32 in1outputmask)
-{
-	u8 byArgs[4];
-	u8 status;
-	int i;
-	struct psb_intel_sdvo_priv *sdvo_priv = output->dev_priv;
-
-	/* Make all fields of the  args/ret to zero */
-	memset(byArgs, 0, sizeof(byArgs));
-
-	/* Fill up the argument values; */
-	byArgs[0] = (u8) (in0outputmask & 0xFF);
-	byArgs[1] = (u8) ((in0outputmask >> 8) & 0xFF);
-	byArgs[2] = (u8) (in1outputmask & 0xFF);
-	byArgs[3] = (u8) ((in1outputmask >> 8) & 0xFF);
-
-
-	/*save inoutmap arg here*/
-	for (i = 0; i < 4; i++)
-		sdvo_priv->in_out_map[i] = byArgs[0];
-
-	psb_intel_sdvo_write_cmd(output, SDVO_CMD_SET_IN_OUT_MAP, byArgs, 4);
-	status = psb_intel_sdvo_read_response(output, NULL, 0);
-
-	if (status != SDVO_CMD_STATUS_SUCCESS)
-		return false;
-	return true;
-}
-
-
-static void psb_intel_sdvo_set_iomap(struct psb_intel_output *output)
-{
-	u32 dwCurrentSDVOIn0 = 0;
-	u32 dwCurrentSDVOIn1 = 0;
-	u32 dwDevMask = 0;
-
-
-	struct psb_intel_sdvo_priv *sdvo_priv = output->dev_priv;
-
-	/* Please DO NOT change the following code. */
-	/* SDVOB_IN0 or SDVOB_IN1 ==> sdvo_in0 */
-	/* SDVOC_IN0 or SDVOC_IN1 ==> sdvo_in1 */
-	if (sdvo_priv->by_input_wiring & (SDVOB_IN0 | SDVOC_IN0)) {
-		switch (sdvo_priv->active_device) {
-		case SDVO_DEVICE_LVDS:
-			dwDevMask = SDVO_OUTPUT_LVDS0 | SDVO_OUTPUT_LVDS1;
-			break;
-		case SDVO_DEVICE_TMDS:
-			dwDevMask = SDVO_OUTPUT_TMDS0 | SDVO_OUTPUT_TMDS1;
-			break;
-		case SDVO_DEVICE_TV:
-			dwDevMask =
-			SDVO_OUTPUT_YPRPB0 | SDVO_OUTPUT_SVID0 |
-			SDVO_OUTPUT_CVBS0 | SDVO_OUTPUT_YPRPB1 |
-			SDVO_OUTPUT_SVID1 | SDVO_OUTPUT_CVBS1 |
-			SDVO_OUTPUT_SCART0 | SDVO_OUTPUT_SCART1;
-			break;
-		case SDVO_DEVICE_CRT:
-			dwDevMask = SDVO_OUTPUT_RGB0 | SDVO_OUTPUT_RGB1;
-			break;
-		}
-		dwCurrentSDVOIn0 = (sdvo_priv->active_outputs & dwDevMask);
-	} else if (sdvo_priv->by_input_wiring & (SDVOB_IN1 | SDVOC_IN1)) {
-		switch (sdvo_priv->active_device) {
-		case SDVO_DEVICE_LVDS:
-			dwDevMask = SDVO_OUTPUT_LVDS0 | SDVO_OUTPUT_LVDS1;
-			break;
-		case SDVO_DEVICE_TMDS:
-			dwDevMask = SDVO_OUTPUT_TMDS0 | SDVO_OUTPUT_TMDS1;
-			break;
-		case SDVO_DEVICE_TV:
-			dwDevMask =
-			SDVO_OUTPUT_YPRPB0 | SDVO_OUTPUT_SVID0 |
-			SDVO_OUTPUT_CVBS0 | SDVO_OUTPUT_YPRPB1 |
-			SDVO_OUTPUT_SVID1 | SDVO_OUTPUT_CVBS1 |
-			SDVO_OUTPUT_SCART0 | SDVO_OUTPUT_SCART1;
-			break;
-		case SDVO_DEVICE_CRT:
-			dwDevMask = SDVO_OUTPUT_RGB0 | SDVO_OUTPUT_RGB1;
-			break;
-		}
-		dwCurrentSDVOIn1 = (sdvo_priv->active_outputs & dwDevMask);
-	}
-
-	psb_sdvo_set_current_inoutmap(output, dwCurrentSDVOIn0,
-					  dwCurrentSDVOIn1);
-}
-
-
-static bool psb_intel_sdvo_mode_fixup(struct drm_encoder *encoder,
-				  struct drm_display_mode *mode,
-				  struct drm_display_mode *adjusted_mode)
-{
-	/* Make the CRTC code factor in the SDVO pixel multiplier.  The SDVO
-	 * device will be told of the multiplier during mode_set.
-	 */
-	adjusted_mode->clock *= psb_intel_sdvo_get_pixel_multiplier(mode);
-	return true;
-}
-
-static void psb_intel_sdvo_mode_set(struct drm_encoder *encoder,
-				struct drm_display_mode *mode,
-				struct drm_display_mode *adjusted_mode)
-{
-	struct drm_device *dev = encoder->dev;
-	struct drm_crtc *crtc = encoder->crtc;
-	struct psb_intel_crtc *psb_intel_crtc = to_psb_intel_crtc(crtc);
-	struct psb_intel_output *psb_intel_output =
-					enc_to_psb_intel_output(encoder);
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
-	u16 width, height;
-	u16 h_blank_len, h_sync_len, v_blank_len, v_sync_len;
-	u16 h_sync_offset, v_sync_offset;
-	u32 sdvox;
-	struct psb_intel_sdvo_dtd output_dtd;
-	int sdvo_pixel_multiply;
-
-	if (!mode)
-		return;
-
-	psb_intel_sdvo_set_target_output(psb_intel_output, 0);
+	uint16_t width, height;
+	uint16_t h_blank_len, h_sync_len, v_blank_len, v_sync_len;
+	uint16_t h_sync_offset, v_sync_offset;
 
 	width = mode->crtc_hdisplay;
 	height = mode->crtc_vdisplay;
@@ -695,68 +733,332 @@
 	h_sync_offset = mode->crtc_hsync_start - mode->crtc_hblank_start;
 	v_sync_offset = mode->crtc_vsync_start - mode->crtc_vblank_start;
 
-	output_dtd.part1.clock = mode->clock / 10;
-	output_dtd.part1.h_active = width & 0xff;
-	output_dtd.part1.h_blank = h_blank_len & 0xff;
-	output_dtd.part1.h_high = (((width >> 8) & 0xf) << 4) |
-	    ((h_blank_len >> 8) & 0xf);
-	output_dtd.part1.v_active = height & 0xff;
-	output_dtd.part1.v_blank = v_blank_len & 0xff;
-	output_dtd.part1.v_high = (((height >> 8) & 0xf) << 4) |
-	    ((v_blank_len >> 8) & 0xf);
+	dtd->part1.clock = mode->clock / 10;
+	dtd->part1.h_active = width & 0xff;
+	dtd->part1.h_blank = h_blank_len & 0xff;
+	dtd->part1.h_high = (((width >> 8) & 0xf) << 4) |
+		((h_blank_len >> 8) & 0xf);
+	dtd->part1.v_active = height & 0xff;
+	dtd->part1.v_blank = v_blank_len & 0xff;
+	dtd->part1.v_high = (((height >> 8) & 0xf) << 4) |
+		((v_blank_len >> 8) & 0xf);
 
-	output_dtd.part2.h_sync_off = h_sync_offset;
-	output_dtd.part2.h_sync_width = h_sync_len & 0xff;
-	output_dtd.part2.v_sync_off_width = (v_sync_offset & 0xf) << 4 |
-	    (v_sync_len & 0xf);
-	output_dtd.part2.sync_off_width_high =
-	    ((h_sync_offset & 0x300) >> 2) | ((h_sync_len & 0x300) >> 4) |
-	    ((v_sync_offset & 0x30) >> 2) | ((v_sync_len & 0x30) >> 4);
+	dtd->part2.h_sync_off = h_sync_offset & 0xff;
+	dtd->part2.h_sync_width = h_sync_len & 0xff;
+	dtd->part2.v_sync_off_width = (v_sync_offset & 0xf) << 4 |
+		(v_sync_len & 0xf);
+	dtd->part2.sync_off_width_high = ((h_sync_offset & 0x300) >> 2) |
+		((h_sync_len & 0x300) >> 4) | ((v_sync_offset & 0x30) >> 2) |
+		((v_sync_len & 0x30) >> 4);
 
-	output_dtd.part2.dtd_flags = 0x18;
+	dtd->part2.dtd_flags = 0x18;
 	if (mode->flags & DRM_MODE_FLAG_PHSYNC)
-		output_dtd.part2.dtd_flags |= 0x2;
+		dtd->part2.dtd_flags |= 0x2;
 	if (mode->flags & DRM_MODE_FLAG_PVSYNC)
-		output_dtd.part2.dtd_flags |= 0x4;
+		dtd->part2.dtd_flags |= 0x4;
 
-	output_dtd.part2.sdvo_flags = 0;
-	output_dtd.part2.v_sync_off_high = v_sync_offset & 0xc0;
-	output_dtd.part2.reserved = 0;
+	dtd->part2.sdvo_flags = 0;
+	dtd->part2.v_sync_off_high = v_sync_offset & 0xc0;
+	dtd->part2.reserved = 0;
+}
 
-	/* Set the output timing to the screen */
-	psb_intel_sdvo_set_target_output(psb_intel_output,
-				     sdvo_priv->active_outputs);
+static void psb_intel_sdvo_get_mode_from_dtd(struct drm_display_mode * mode,
+					 const struct psb_intel_sdvo_dtd *dtd)
+{
+	mode->hdisplay = dtd->part1.h_active;
+	mode->hdisplay += ((dtd->part1.h_high >> 4) & 0x0f) << 8;
+	mode->hsync_start = mode->hdisplay + dtd->part2.h_sync_off;
+	mode->hsync_start += (dtd->part2.sync_off_width_high & 0xc0) << 2;
+	mode->hsync_end = mode->hsync_start + dtd->part2.h_sync_width;
+	mode->hsync_end += (dtd->part2.sync_off_width_high & 0x30) << 4;
+	mode->htotal = mode->hdisplay + dtd->part1.h_blank;
+	mode->htotal += (dtd->part1.h_high & 0xf) << 8;
 
-	/* Set the input timing to the screen. Assume always input 0. */
-	psb_intel_sdvo_set_target_input(psb_intel_output, true, false);
+	mode->vdisplay = dtd->part1.v_active;
+	mode->vdisplay += ((dtd->part1.v_high >> 4) & 0x0f) << 8;
+	mode->vsync_start = mode->vdisplay;
+	mode->vsync_start += (dtd->part2.v_sync_off_width >> 4) & 0xf;
+	mode->vsync_start += (dtd->part2.sync_off_width_high & 0x0c) << 2;
+	mode->vsync_start += dtd->part2.v_sync_off_high & 0xc0;
+	mode->vsync_end = mode->vsync_start +
+		(dtd->part2.v_sync_off_width & 0xf);
+	mode->vsync_end += (dtd->part2.sync_off_width_high & 0x3) << 4;
+	mode->vtotal = mode->vdisplay + dtd->part1.v_blank;
+	mode->vtotal += (dtd->part1.v_high & 0xf) << 8;
 
-	psb_intel_sdvo_set_output_timing(psb_intel_output, &output_dtd);
+	mode->clock = dtd->part1.clock * 10;
 
-	/* We would like to use i830_sdvo_create_preferred_input_timing() to
-	 * provide the device with a timing it can support, if it supports that
-	 * feature.  However, presumably we would need to adjust the CRTC to
-	 * output the preferred timing, and we don't support that currently.
-	 */
-	psb_intel_sdvo_set_input_timing(psb_intel_output, &output_dtd);
+	mode->flags &= ~(DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
+	if (dtd->part2.dtd_flags & 0x2)
+		mode->flags |= DRM_MODE_FLAG_PHSYNC;
+	if (dtd->part2.dtd_flags & 0x4)
+		mode->flags |= DRM_MODE_FLAG_PVSYNC;
+}
 
-	switch (psb_intel_sdvo_get_pixel_multiplier(mode)) {
-	case 1:
-		psb_intel_sdvo_set_clock_rate_mult(psb_intel_output,
-					       SDVO_CLOCK_RATE_MULT_1X);
-		break;
-	case 2:
-		psb_intel_sdvo_set_clock_rate_mult(psb_intel_output,
-					       SDVO_CLOCK_RATE_MULT_2X);
-		break;
-	case 4:
-		psb_intel_sdvo_set_clock_rate_mult(psb_intel_output,
-					       SDVO_CLOCK_RATE_MULT_4X);
-		break;
+static bool psb_intel_sdvo_check_supp_encode(struct psb_intel_sdvo *psb_intel_sdvo)
+{
+	struct psb_intel_sdvo_encode encode;
+
+	BUILD_BUG_ON(sizeof(encode) != 2);
+	return psb_intel_sdvo_get_value(psb_intel_sdvo,
+				  SDVO_CMD_GET_SUPP_ENCODE,
+				  &encode, sizeof(encode));
+}
+
+static bool psb_intel_sdvo_set_encode(struct psb_intel_sdvo *psb_intel_sdvo,
+				  uint8_t mode)
+{
+	return psb_intel_sdvo_set_value(psb_intel_sdvo, SDVO_CMD_SET_ENCODE, &mode, 1);
+}
+
+static bool psb_intel_sdvo_set_colorimetry(struct psb_intel_sdvo *psb_intel_sdvo,
+				       uint8_t mode)
+{
+	return psb_intel_sdvo_set_value(psb_intel_sdvo, SDVO_CMD_SET_COLORIMETRY, &mode, 1);
+}
+
+#if 0
+static void psb_intel_sdvo_dump_hdmi_buf(struct psb_intel_sdvo *psb_intel_sdvo)
+{
+	int i, j;
+	uint8_t set_buf_index[2];
+	uint8_t av_split;
+	uint8_t buf_size;
+	uint8_t buf[48];
+	uint8_t *pos;
+
+	psb_intel_sdvo_get_value(encoder, SDVO_CMD_GET_HBUF_AV_SPLIT, &av_split, 1);
+
+	for (i = 0; i <= av_split; i++) {
+		set_buf_index[0] = i; set_buf_index[1] = 0;
+		psb_intel_sdvo_write_cmd(encoder, SDVO_CMD_SET_HBUF_INDEX,
+				     set_buf_index, 2);
+		psb_intel_sdvo_write_cmd(encoder, SDVO_CMD_GET_HBUF_INFO, NULL, 0);
+		psb_intel_sdvo_read_response(encoder, &buf_size, 1);
+
+		pos = buf;
+		for (j = 0; j <= buf_size; j += 8) {
+			psb_intel_sdvo_write_cmd(encoder, SDVO_CMD_GET_HBUF_DATA,
+					     NULL, 0);
+			psb_intel_sdvo_read_response(encoder, pos, 8);
+			pos += 8;
+		}
+	}
+}
+#endif
+
+static bool psb_intel_sdvo_set_avi_infoframe(struct psb_intel_sdvo *psb_intel_sdvo)
+{
+	DRM_INFO("HDMI is not supported yet");
+
+	return false;
+#if 0
+	struct dip_infoframe avi_if = {
+		.type = DIP_TYPE_AVI,
+		.ver = DIP_VERSION_AVI,
+		.len = DIP_LEN_AVI,
+	};
+	uint8_t tx_rate = SDVO_HBUF_TX_VSYNC;
+	uint8_t set_buf_index[2] = { 1, 0 };
+	uint64_t *data = (uint64_t *)&avi_if;
+	unsigned i;
+
+	intel_dip_infoframe_csum(&avi_if);
+
+	if (!psb_intel_sdvo_set_value(psb_intel_sdvo,
+				  SDVO_CMD_SET_HBUF_INDEX,
+				  set_buf_index, 2))
+		return false;
+
+	for (i = 0; i < sizeof(avi_if); i += 8) {
+		if (!psb_intel_sdvo_set_value(psb_intel_sdvo,
+					  SDVO_CMD_SET_HBUF_DATA,
+					  data, 8))
+			return false;
+		data++;
 	}
 
+	return psb_intel_sdvo_set_value(psb_intel_sdvo,
+				    SDVO_CMD_SET_HBUF_TXRATE,
+				    &tx_rate, 1);
+#endif
+}
+
+static bool psb_intel_sdvo_set_tv_format(struct psb_intel_sdvo *psb_intel_sdvo)
+{
+	struct psb_intel_sdvo_tv_format format;
+	uint32_t format_map;
+
+	format_map = 1 << psb_intel_sdvo->tv_format_index;
+	memset(&format, 0, sizeof(format));
+	memcpy(&format, &format_map, min(sizeof(format), sizeof(format_map)));
+
+	BUILD_BUG_ON(sizeof(format) != 6);
+	return psb_intel_sdvo_set_value(psb_intel_sdvo,
+				    SDVO_CMD_SET_TV_FORMAT,
+				    &format, sizeof(format));
+}
+
+static bool
+psb_intel_sdvo_set_output_timings_from_mode(struct psb_intel_sdvo *psb_intel_sdvo,
+					struct drm_display_mode *mode)
+{
+	struct psb_intel_sdvo_dtd output_dtd;
+
+	if (!psb_intel_sdvo_set_target_output(psb_intel_sdvo,
+					  psb_intel_sdvo->attached_output))
+		return false;
+
+	psb_intel_sdvo_get_dtd_from_mode(&output_dtd, mode);
+	if (!psb_intel_sdvo_set_output_timing(psb_intel_sdvo, &output_dtd))
+		return false;
+
+	return true;
+}
+
+static bool
+psb_intel_sdvo_set_input_timings_for_mode(struct psb_intel_sdvo *psb_intel_sdvo,
+					struct drm_display_mode *mode,
+					struct drm_display_mode *adjusted_mode)
+{
+	/* Reset the input timing to the screen. Assume always input 0. */
+	if (!psb_intel_sdvo_set_target_input(psb_intel_sdvo))
+		return false;
+
+	if (!psb_intel_sdvo_create_preferred_input_timing(psb_intel_sdvo,
+						      mode->clock / 10,
+						      mode->hdisplay,
+						      mode->vdisplay))
+		return false;
+
+	if (!psb_intel_sdvo_get_preferred_input_timing(psb_intel_sdvo,
+						   &psb_intel_sdvo->input_dtd))
+		return false;
+
+	psb_intel_sdvo_get_mode_from_dtd(adjusted_mode, &psb_intel_sdvo->input_dtd);
+
+	drm_mode_set_crtcinfo(adjusted_mode, 0);
+	return true;
+}
+
+static bool psb_intel_sdvo_mode_fixup(struct drm_encoder *encoder,
+				  struct drm_display_mode *mode,
+				  struct drm_display_mode *adjusted_mode)
+{
+	struct psb_intel_sdvo *psb_intel_sdvo = to_psb_intel_sdvo(encoder);
+	int multiplier;
+
+	/* We need to construct preferred input timings based on our
+	 * output timings.  To do that, we have to set the output
+	 * timings, even though this isn't really the right place in
+	 * the sequence to do it. Oh well.
+	 */
+	if (psb_intel_sdvo->is_tv) {
+		if (!psb_intel_sdvo_set_output_timings_from_mode(psb_intel_sdvo, mode))
+			return false;
+
+		(void) psb_intel_sdvo_set_input_timings_for_mode(psb_intel_sdvo,
+							     mode,
+							     adjusted_mode);
+	} else if (psb_intel_sdvo->is_lvds) {
+		if (!psb_intel_sdvo_set_output_timings_from_mode(psb_intel_sdvo,
+							     psb_intel_sdvo->sdvo_lvds_fixed_mode))
+			return false;
+
+		(void) psb_intel_sdvo_set_input_timings_for_mode(psb_intel_sdvo,
+							     mode,
+							     adjusted_mode);
+	}
+
+	/* Make the CRTC code factor in the SDVO pixel multiplier.  The
+	 * SDVO device will factor out the multiplier during mode_set.
+	 */
+	multiplier = psb_intel_sdvo_get_pixel_multiplier(adjusted_mode);
+	psb_intel_mode_set_pixel_multiplier(adjusted_mode, multiplier);
+
+	return true;
+}
+
+static void psb_intel_sdvo_mode_set(struct drm_encoder *encoder,
+				struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)
+{
+	struct drm_device *dev = encoder->dev;
+	struct drm_crtc *crtc = encoder->crtc;
+	struct psb_intel_crtc *psb_intel_crtc = to_psb_intel_crtc(crtc);
+	struct psb_intel_sdvo *psb_intel_sdvo = to_psb_intel_sdvo(encoder);
+	u32 sdvox;
+	struct psb_intel_sdvo_in_out_map in_out;
+	struct psb_intel_sdvo_dtd input_dtd;
+	int pixel_multiplier = psb_intel_mode_get_pixel_multiplier(adjusted_mode);
+	int rate;
+
+	if (!mode)
+		return;
+
+	/* First, set the input mapping for the first input to our controlled
+	 * output. This is only correct if we're a single-input device, in
+	 * which case the first input is the output from the appropriate SDVO
+	 * channel on the motherboard.  In a two-input device, the first input
+	 * will be SDVOB and the second SDVOC.
+	 */
+	in_out.in0 = psb_intel_sdvo->attached_output;
+	in_out.in1 = 0;
+
+	psb_intel_sdvo_set_value(psb_intel_sdvo,
+			     SDVO_CMD_SET_IN_OUT_MAP,
+			     &in_out, sizeof(in_out));
+
+	/* Set the output timings to the screen */
+	if (!psb_intel_sdvo_set_target_output(psb_intel_sdvo,
+					  psb_intel_sdvo->attached_output))
+		return;
+
+	/* We have tried to get input timing in mode_fixup, and filled into
+	 * adjusted_mode.
+	 */
+	if (psb_intel_sdvo->is_tv || psb_intel_sdvo->is_lvds) {
+		input_dtd = psb_intel_sdvo->input_dtd;
+	} else {
+		/* Set the output timing to the screen */
+		if (!psb_intel_sdvo_set_target_output(psb_intel_sdvo,
+						  psb_intel_sdvo->attached_output))
+			return;
+
+		psb_intel_sdvo_get_dtd_from_mode(&input_dtd, adjusted_mode);
+		(void) psb_intel_sdvo_set_output_timing(psb_intel_sdvo, &input_dtd);
+	}
+
+	/* Set the input timing to the screen. Assume always input 0. */
+	if (!psb_intel_sdvo_set_target_input(psb_intel_sdvo))
+		return;
+
+	if (psb_intel_sdvo->has_hdmi_monitor) {
+		psb_intel_sdvo_set_encode(psb_intel_sdvo, SDVO_ENCODE_HDMI);
+		psb_intel_sdvo_set_colorimetry(psb_intel_sdvo,
+					   SDVO_COLORIMETRY_RGB256);
+		psb_intel_sdvo_set_avi_infoframe(psb_intel_sdvo);
+	} else
+		psb_intel_sdvo_set_encode(psb_intel_sdvo, SDVO_ENCODE_DVI);
+
+	if (psb_intel_sdvo->is_tv &&
+	    !psb_intel_sdvo_set_tv_format(psb_intel_sdvo))
+		return;
+
+	(void) psb_intel_sdvo_set_input_timing(psb_intel_sdvo, &input_dtd);
+
+	switch (pixel_multiplier) {
+	default:
+	case 1: rate = SDVO_CLOCK_RATE_MULT_1X; break;
+	case 2: rate = SDVO_CLOCK_RATE_MULT_2X; break;
+	case 4: rate = SDVO_CLOCK_RATE_MULT_4X; break;
+	}
+	if (!psb_intel_sdvo_set_clock_rate_mult(psb_intel_sdvo, rate))
+		return;
+
 	/* Set the SDVO control regs. */
-	sdvox = REG_READ(sdvo_priv->output_device);
-	switch (sdvo_priv->output_device) {
+	sdvox = REG_READ(psb_intel_sdvo->sdvo_reg);
+	switch (psb_intel_sdvo->sdvo_reg) {
 	case SDVOB:
 		sdvox &= SDVOB_PRESERVE_MASK;
 		break;
@@ -765,37 +1067,47 @@
 		break;
 	}
 	sdvox |= (9 << 19) | SDVO_BORDER_ENABLE;
+
 	if (psb_intel_crtc->pipe == 1)
 		sdvox |= SDVO_PIPE_B_SELECT;
+	if (psb_intel_sdvo->has_hdmi_audio)
+		sdvox |= SDVO_AUDIO_ENABLE;
 
-	sdvo_pixel_multiply = psb_intel_sdvo_get_pixel_multiplier(mode);
+	/* FIXME: Check if this is needed for PSB
+	sdvox |= (pixel_multiplier - 1) << SDVO_PORT_MULTIPLY_SHIFT;
+	*/
 
-	psb_intel_sdvo_write_sdvox(psb_intel_output, sdvox);
-
-	 psb_intel_sdvo_set_iomap(psb_intel_output);
+	if (input_dtd.part2.sdvo_flags & SDVO_NEED_TO_STALL)
+		sdvox |= SDVO_STALL_SELECT;
+	psb_intel_sdvo_write_sdvox(psb_intel_sdvo, sdvox);
 }
 
 static void psb_intel_sdvo_dpms(struct drm_encoder *encoder, int mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct psb_intel_output *psb_intel_output =
-					enc_to_psb_intel_output(encoder);
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
+	struct psb_intel_sdvo *psb_intel_sdvo = to_psb_intel_sdvo(encoder);
 	u32 temp;
 
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		DRM_DEBUG("DPMS_ON");
+		break;
+	case DRM_MODE_DPMS_OFF:
+		DRM_DEBUG("DPMS_OFF");
+		break;
+	default:
+		DRM_DEBUG("DPMS: %d", mode);
+	}
+
 	if (mode != DRM_MODE_DPMS_ON) {
-		psb_intel_sdvo_set_active_outputs(psb_intel_output, 0);
+		psb_intel_sdvo_set_active_outputs(psb_intel_sdvo, 0);
 		if (0)
-			psb_intel_sdvo_set_encoder_power_state(
-							psb_intel_output,
-							mode);
+			psb_intel_sdvo_set_encoder_power_state(psb_intel_sdvo, mode);
 
 		if (mode == DRM_MODE_DPMS_OFF) {
-			temp = REG_READ(sdvo_priv->output_device);
+			temp = REG_READ(psb_intel_sdvo->sdvo_reg);
 			if ((temp & SDVO_ENABLE) != 0) {
-				psb_intel_sdvo_write_sdvox(psb_intel_output,
-						       temp &
-						       ~SDVO_ENABLE);
+				psb_intel_sdvo_write_sdvox(psb_intel_sdvo, temp & ~SDVO_ENABLE);
 			}
 		}
 	} else {
@@ -803,185 +1115,112 @@
 		int i;
 		u8 status;
 
-		temp = REG_READ(sdvo_priv->output_device);
+		temp = REG_READ(psb_intel_sdvo->sdvo_reg);
 		if ((temp & SDVO_ENABLE) == 0)
-			psb_intel_sdvo_write_sdvox(psb_intel_output,
-					       temp | SDVO_ENABLE);
+			psb_intel_sdvo_write_sdvox(psb_intel_sdvo, temp | SDVO_ENABLE);
 		for (i = 0; i < 2; i++)
 			psb_intel_wait_for_vblank(dev);
 
-		status =
-		    psb_intel_sdvo_get_trained_inputs(psb_intel_output,
-							&input1,
-							&input2);
-
-
+		status = psb_intel_sdvo_get_trained_inputs(psb_intel_sdvo, &input1, &input2);
 		/* Warn if the device reported failure to sync.
 		 * A lot of SDVO devices fail to notify of sync, but it's
 		 * a given it the status is a success, we succeeded.
 		 */
 		if (status == SDVO_CMD_STATUS_SUCCESS && !input1) {
-			DRM_DEBUG
-			    ("First %s output reported failure to sync\n",
-			     SDVO_NAME(sdvo_priv));
+			DRM_DEBUG_KMS("First %s output reported failure to "
+					"sync\n", SDVO_NAME(psb_intel_sdvo));
 		}
 
 		if (0)
-			psb_intel_sdvo_set_encoder_power_state(
-							psb_intel_output,
-							mode);
-		psb_intel_sdvo_set_active_outputs(psb_intel_output,
-					      sdvo_priv->active_outputs);
+			psb_intel_sdvo_set_encoder_power_state(psb_intel_sdvo, mode);
+		psb_intel_sdvo_set_active_outputs(psb_intel_sdvo, psb_intel_sdvo->attached_output);
 	}
 	return;
 }
 
-static void psb_intel_sdvo_save(struct drm_connector *connector)
-{
-	struct drm_device *dev = connector->dev;
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
-	/*int o;*/
-
-	sdvo_priv->save_sdvo_mult =
-	    psb_intel_sdvo_get_clock_rate_mult(psb_intel_output);
-	psb_intel_sdvo_get_active_outputs(psb_intel_output,
-				      &sdvo_priv->save_active_outputs);
-
-	if (sdvo_priv->caps.sdvo_inputs_mask & 0x1) {
-		psb_intel_sdvo_set_target_input(psb_intel_output,
-						true,
-						false);
-		psb_intel_sdvo_get_input_timing(psb_intel_output,
-					    &sdvo_priv->save_input_dtd_1);
-	}
-
-	if (sdvo_priv->caps.sdvo_inputs_mask & 0x2) {
-		psb_intel_sdvo_set_target_input(psb_intel_output,
-						false,
-						true);
-		psb_intel_sdvo_get_input_timing(psb_intel_output,
-					    &sdvo_priv->save_input_dtd_2);
-	}
-	sdvo_priv->save_SDVOX = REG_READ(sdvo_priv->output_device);
-
-	/*TODO: save the in_out_map state*/
-}
-
-static void psb_intel_sdvo_restore(struct drm_connector *connector)
-{
-	struct drm_device *dev = connector->dev;
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
-	/*int o;*/
-	int i;
-	bool input1, input2;
-	u8 status;
-
-	psb_intel_sdvo_set_active_outputs(psb_intel_output, 0);
-
-	if (sdvo_priv->caps.sdvo_inputs_mask & 0x1) {
-		psb_intel_sdvo_set_target_input(psb_intel_output, true, false);
-		psb_intel_sdvo_set_input_timing(psb_intel_output,
-					    &sdvo_priv->save_input_dtd_1);
-	}
-
-	if (sdvo_priv->caps.sdvo_inputs_mask & 0x2) {
-		psb_intel_sdvo_set_target_input(psb_intel_output, false, true);
-		psb_intel_sdvo_set_input_timing(psb_intel_output,
-					    &sdvo_priv->save_input_dtd_2);
-	}
-
-	psb_intel_sdvo_set_clock_rate_mult(psb_intel_output,
-				       sdvo_priv->save_sdvo_mult);
-
-	REG_WRITE(sdvo_priv->output_device, sdvo_priv->save_SDVOX);
-
-	if (sdvo_priv->save_SDVOX & SDVO_ENABLE) {
-		for (i = 0; i < 2; i++)
-			psb_intel_wait_for_vblank(dev);
-		status =
-		    psb_intel_sdvo_get_trained_inputs(psb_intel_output,
-							&input1,
-							&input2);
-		if (status == SDVO_CMD_STATUS_SUCCESS && !input1)
-			DRM_DEBUG
-			    ("First %s output reported failure to sync\n",
-			     SDVO_NAME(sdvo_priv));
-	}
-
-	psb_intel_sdvo_set_active_outputs(psb_intel_output,
-				      sdvo_priv->save_active_outputs);
-
-	/*TODO: restore in_out_map*/
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				 SDVO_CMD_SET_IN_OUT_MAP,
-				 sdvo_priv->in_out_map,
-				 4);
-
-	psb_intel_sdvo_read_response(psb_intel_output, NULL, 0);
-}
-
 static int psb_intel_sdvo_mode_valid(struct drm_connector *connector,
 				 struct drm_display_mode *mode)
 {
-	struct psb_intel_output *psb_intel_output =
-				to_psb_intel_output(connector);
-	struct psb_intel_sdvo_priv *sdvo_priv = psb_intel_output->dev_priv;
+	struct psb_intel_sdvo *psb_intel_sdvo = intel_attached_sdvo(connector);
 
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return MODE_NO_DBLESCAN;
 
-	if (sdvo_priv->pixel_clock_min > mode->clock)
+	if (psb_intel_sdvo->pixel_clock_min > mode->clock)
 		return MODE_CLOCK_LOW;
 
-	if (sdvo_priv->pixel_clock_max < mode->clock)
+	if (psb_intel_sdvo->pixel_clock_max < mode->clock)
 		return MODE_CLOCK_HIGH;
 
+	if (psb_intel_sdvo->is_lvds) {
+		if (mode->hdisplay > psb_intel_sdvo->sdvo_lvds_fixed_mode->hdisplay)
+			return MODE_PANEL;
+
+		if (mode->vdisplay > psb_intel_sdvo->sdvo_lvds_fixed_mode->vdisplay)
+			return MODE_PANEL;
+	}
+
 	return MODE_OK;
 }
 
-static bool psb_intel_sdvo_get_capabilities(
-				struct psb_intel_output *psb_intel_output,
-				struct psb_intel_sdvo_caps *caps)
+static bool psb_intel_sdvo_get_capabilities(struct psb_intel_sdvo *psb_intel_sdvo, struct psb_intel_sdvo_caps *caps)
 {
-	u8 status;
-
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				 SDVO_CMD_GET_DEVICE_CAPS,
-				 NULL,
-				 0);
-	status = psb_intel_sdvo_read_response(psb_intel_output,
-						caps,
-						sizeof(*caps));
-	if (status != SDVO_CMD_STATUS_SUCCESS)
+	BUILD_BUG_ON(sizeof(*caps) != 8);
+	if (!psb_intel_sdvo_get_value(psb_intel_sdvo,
+				  SDVO_CMD_GET_DEVICE_CAPS,
+				  caps, sizeof(*caps)))
 		return false;
 
+	DRM_DEBUG_KMS("SDVO capabilities:\n"
+		      "  vendor_id: %d\n"
+		      "  device_id: %d\n"
+		      "  device_rev_id: %d\n"
+		      "  sdvo_version_major: %d\n"
+		      "  sdvo_version_minor: %d\n"
+		      "  sdvo_inputs_mask: %d\n"
+		      "  smooth_scaling: %d\n"
+		      "  sharp_scaling: %d\n"
+		      "  up_scaling: %d\n"
+		      "  down_scaling: %d\n"
+		      "  stall_support: %d\n"
+		      "  output_flags: %d\n",
+		      caps->vendor_id,
+		      caps->device_id,
+		      caps->device_rev_id,
+		      caps->sdvo_version_major,
+		      caps->sdvo_version_minor,
+		      caps->sdvo_inputs_mask,
+		      caps->smooth_scaling,
+		      caps->sharp_scaling,
+		      caps->up_scaling,
+		      caps->down_scaling,
+		      caps->stall_support,
+		      caps->output_flags);
+
 	return true;
 }
 
-struct drm_connector *psb_intel_sdvo_find(struct drm_device *dev, int sdvoB)
+/* No use! */
+#if 0
+struct drm_connector* psb_intel_sdvo_find(struct drm_device *dev, int sdvoB)
 {
 	struct drm_connector *connector = NULL;
-	struct psb_intel_output *iout = NULL;
-	struct psb_intel_sdvo_priv *sdvo;
+	struct psb_intel_sdvo *iout = NULL;
+	struct psb_intel_sdvo *sdvo;
 
 	/* find the sdvo connector */
-	list_for_each_entry(connector, &dev->mode_config.connector_list,
-			    head) {
-		iout = to_psb_intel_output(connector);
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+		iout = to_psb_intel_sdvo(connector);
 
 		if (iout->type != INTEL_OUTPUT_SDVO)
 			continue;
 
 		sdvo = iout->dev_priv;
 
-		if (sdvo->output_device == SDVOB && sdvoB)
+		if (sdvo->sdvo_reg == SDVOB && sdvoB)
 			return connector;
 
-		if (sdvo->output_device == SDVOC && !sdvoB)
+		if (sdvo->sdvo_reg == SDVOC && !sdvoB)
 			return connector;
 
 	}
@@ -993,112 +1232,607 @@
 {
 	u8 response[2];
 	u8 status;
-	struct psb_intel_output *psb_intel_output;
+	struct psb_intel_sdvo *psb_intel_sdvo;
+	DRM_DEBUG_KMS("\n");
 
 	if (!connector)
 		return 0;
 
-	psb_intel_output = to_psb_intel_output(connector);
+	psb_intel_sdvo = to_psb_intel_sdvo(connector);
 
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				 SDVO_CMD_GET_HOT_PLUG_SUPPORT,
-				 NULL,
-				 0);
-	status = psb_intel_sdvo_read_response(psb_intel_output,
-						&response,
-						2);
-
-	if (response[0] != 0)
-		return 1;
-
-	return 0;
+	return psb_intel_sdvo_get_value(psb_intel_sdvo, SDVO_CMD_GET_HOT_PLUG_SUPPORT,
+				    &response, 2) && response[0];
 }
 
 void psb_intel_sdvo_set_hotplug(struct drm_connector *connector, int on)
 {
 	u8 response[2];
 	u8 status;
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_sdvo *psb_intel_sdvo = to_psb_intel_sdvo(connector);
 
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				 SDVO_CMD_GET_ACTIVE_HOT_PLUG,
-				 NULL,
-				 0);
-	psb_intel_sdvo_read_response(psb_intel_output, &response, 2);
+	psb_intel_sdvo_write_cmd(psb_intel_sdvo, SDVO_CMD_GET_ACTIVE_HOT_PLUG, NULL, 0);
+	psb_intel_sdvo_read_response(psb_intel_sdvo, &response, 2);
 
 	if (on) {
-		psb_intel_sdvo_write_cmd(psb_intel_output,
-				     SDVO_CMD_GET_HOT_PLUG_SUPPORT, NULL,
-				     0);
-		status = psb_intel_sdvo_read_response(psb_intel_output,
-						      &response,
-						      2);
+		psb_intel_sdvo_write_cmd(psb_intel_sdvo, SDVO_CMD_GET_HOT_PLUG_SUPPORT, NULL, 0);
+		status = psb_intel_sdvo_read_response(psb_intel_sdvo, &response, 2);
 
-		psb_intel_sdvo_write_cmd(psb_intel_output,
-				     SDVO_CMD_SET_ACTIVE_HOT_PLUG,
-				     &response, 2);
+		psb_intel_sdvo_write_cmd(psb_intel_sdvo, SDVO_CMD_SET_ACTIVE_HOT_PLUG, &response, 2);
 	} else {
 		response[0] = 0;
 		response[1] = 0;
-		psb_intel_sdvo_write_cmd(psb_intel_output,
-				     SDVO_CMD_SET_ACTIVE_HOT_PLUG,
-				     &response, 2);
+		psb_intel_sdvo_write_cmd(psb_intel_sdvo, SDVO_CMD_SET_ACTIVE_HOT_PLUG, &response, 2);
 	}
 
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				 SDVO_CMD_GET_ACTIVE_HOT_PLUG,
-				 NULL,
-				 0);
-	psb_intel_sdvo_read_response(psb_intel_output, &response, 2);
+	psb_intel_sdvo_write_cmd(psb_intel_sdvo, SDVO_CMD_GET_ACTIVE_HOT_PLUG, NULL, 0);
+	psb_intel_sdvo_read_response(psb_intel_sdvo, &response, 2);
+}
+#endif
+
+static bool
+psb_intel_sdvo_multifunc_encoder(struct psb_intel_sdvo *psb_intel_sdvo)
+{
+	/* Is there more than one type of output? */
+	int caps = psb_intel_sdvo->caps.output_flags & 0xf;
+	return caps & -caps;
 }
 
-static enum drm_connector_status psb_intel_sdvo_detect(struct drm_connector
-						   *connector, bool force)
+static struct edid *
+psb_intel_sdvo_get_edid(struct drm_connector *connector)
 {
-	u8 response[2];
-	u8 status;
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_sdvo *sdvo = intel_attached_sdvo(connector);
+	return drm_get_edid(connector, &sdvo->ddc);
+}
 
-	psb_intel_sdvo_write_cmd(psb_intel_output,
-				 SDVO_CMD_GET_ATTACHED_DISPLAYS,
-				 NULL,
-				 0);
-	status = psb_intel_sdvo_read_response(psb_intel_output, &response, 2);
+/* Mac mini hack -- use the same DDC as the analog connector */
+static struct edid *
+psb_intel_sdvo_get_analog_edid(struct drm_connector *connector)
+{
+	struct drm_psb_private *dev_priv = connector->dev->dev_private;
 
-	DRM_DEBUG("SDVO response %d %d\n", response[0], response[1]);
-	if ((response[0] != 0) || (response[1] != 0))
-		return connector_status_connected;
-	else
+	return drm_get_edid(connector,
+			    &dev_priv->gmbus[dev_priv->crt_ddc_pin].adapter);
+	return NULL;
+}
+
+enum drm_connector_status
+psb_intel_sdvo_hdmi_sink_detect(struct drm_connector *connector)
+{
+	struct psb_intel_sdvo *psb_intel_sdvo = intel_attached_sdvo(connector);
+	enum drm_connector_status status;
+	struct edid *edid;
+
+	edid = psb_intel_sdvo_get_edid(connector);
+
+	if (edid == NULL && psb_intel_sdvo_multifunc_encoder(psb_intel_sdvo)) {
+		u8 ddc, saved_ddc = psb_intel_sdvo->ddc_bus;
+
+		/*
+		 * Don't use the 1 as the argument of DDC bus switch to get
+		 * the EDID. It is used for SDVO SPD ROM.
+		 */
+		for (ddc = psb_intel_sdvo->ddc_bus >> 1; ddc > 1; ddc >>= 1) {
+			psb_intel_sdvo->ddc_bus = ddc;
+			edid = psb_intel_sdvo_get_edid(connector);
+			if (edid)
+				break;
+		}
+		/*
+		 * If we found the EDID on the other bus,
+		 * assume that is the correct DDC bus.
+		 */
+		if (edid == NULL)
+			psb_intel_sdvo->ddc_bus = saved_ddc;
+	}
+
+	/*
+	 * When there is no edid and no monitor is connected with VGA
+	 * port, try to use the CRT ddc to read the EDID for DVI-connector.
+	 */
+	if (edid == NULL)
+		edid = psb_intel_sdvo_get_analog_edid(connector);
+
+	status = connector_status_unknown;
+	if (edid != NULL) {
+		/* DDC bus is shared, match EDID to connector type */
+		if (edid->input & DRM_EDID_INPUT_DIGITAL) {
+			status = connector_status_connected;
+			if (psb_intel_sdvo->is_hdmi) {
+				psb_intel_sdvo->has_hdmi_monitor = drm_detect_hdmi_monitor(edid);
+				psb_intel_sdvo->has_hdmi_audio = drm_detect_monitor_audio(edid);
+			}
+		} else
+			status = connector_status_disconnected;
+		connector->display_info.raw_edid = NULL;
+		kfree(edid);
+	}
+
+	if (status == connector_status_connected) {
+		struct psb_intel_sdvo_connector *psb_intel_sdvo_connector = to_psb_intel_sdvo_connector(connector);
+		if (psb_intel_sdvo_connector->force_audio)
+			psb_intel_sdvo->has_hdmi_audio = psb_intel_sdvo_connector->force_audio > 0;
+	}
+
+	return status;
+}
+
+static enum drm_connector_status
+psb_intel_sdvo_detect(struct drm_connector *connector, bool force)
+{
+	uint16_t response;
+	struct psb_intel_sdvo *psb_intel_sdvo = intel_attached_sdvo(connector);
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector = to_psb_intel_sdvo_connector(connector);
+	enum drm_connector_status ret;
+
+	if (!psb_intel_sdvo_write_cmd(psb_intel_sdvo,
+				  SDVO_CMD_GET_ATTACHED_DISPLAYS, NULL, 0))
+		return connector_status_unknown;
+
+	/* add 30ms delay when the output type might be TV */
+	if (psb_intel_sdvo->caps.output_flags &
+	    (SDVO_OUTPUT_SVID0 | SDVO_OUTPUT_CVBS0))
+		mdelay(30);
+
+	if (!psb_intel_sdvo_read_response(psb_intel_sdvo, &response, 2))
+		return connector_status_unknown;
+
+	DRM_DEBUG_KMS("SDVO response %d %d [%x]\n",
+		      response & 0xff, response >> 8,
+		      psb_intel_sdvo_connector->output_flag);
+
+	if (response == 0)
 		return connector_status_disconnected;
+
+	psb_intel_sdvo->attached_output = response;
+
+	psb_intel_sdvo->has_hdmi_monitor = false;
+	psb_intel_sdvo->has_hdmi_audio = false;
+
+	if ((psb_intel_sdvo_connector->output_flag & response) == 0)
+		ret = connector_status_disconnected;
+	else if (IS_TMDS(psb_intel_sdvo_connector))
+		ret = psb_intel_sdvo_hdmi_sink_detect(connector);
+	else {
+		struct edid *edid;
+
+		/* if we have an edid check it matches the connection */
+		edid = psb_intel_sdvo_get_edid(connector);
+		if (edid == NULL)
+			edid = psb_intel_sdvo_get_analog_edid(connector);
+		if (edid != NULL) {
+			if (edid->input & DRM_EDID_INPUT_DIGITAL)
+				ret = connector_status_disconnected;
+			else
+				ret = connector_status_connected;
+			connector->display_info.raw_edid = NULL;
+			kfree(edid);
+		} else
+			ret = connector_status_connected;
+	}
+
+	/* May update encoder flag for like clock for SDVO TV, etc.*/
+	if (ret == connector_status_connected) {
+		psb_intel_sdvo->is_tv = false;
+		psb_intel_sdvo->is_lvds = false;
+		psb_intel_sdvo->base.needs_tv_clock = false;
+
+		if (response & SDVO_TV_MASK) {
+			psb_intel_sdvo->is_tv = true;
+			psb_intel_sdvo->base.needs_tv_clock = true;
+		}
+		if (response & SDVO_LVDS_MASK)
+			psb_intel_sdvo->is_lvds = psb_intel_sdvo->sdvo_lvds_fixed_mode != NULL;
+	}
+
+	return ret;
+}
+
+static void psb_intel_sdvo_get_ddc_modes(struct drm_connector *connector)
+{
+	struct edid *edid;
+
+	/* set the bus switch and get the modes */
+	edid = psb_intel_sdvo_get_edid(connector);
+
+	/*
+	 * Mac mini hack.  On this device, the DVI-I connector shares one DDC
+	 * link between analog and digital outputs. So, if the regular SDVO
+	 * DDC fails, check to see if the analog output is disconnected, in
+	 * which case we'll look there for the digital DDC data.
+	 */
+	if (edid == NULL)
+		edid = psb_intel_sdvo_get_analog_edid(connector);
+
+	if (edid != NULL) {
+		struct psb_intel_sdvo_connector *psb_intel_sdvo_connector = to_psb_intel_sdvo_connector(connector);
+		bool monitor_is_digital = !!(edid->input & DRM_EDID_INPUT_DIGITAL);
+		bool connector_is_digital = !!IS_TMDS(psb_intel_sdvo_connector);
+
+		if (connector_is_digital == monitor_is_digital) {
+			drm_mode_connector_update_edid_property(connector, edid);
+			drm_add_edid_modes(connector, edid);
+		}
+
+		connector->display_info.raw_edid = NULL;
+		kfree(edid);
+	}
+}
+
+/*
+ * Set of SDVO TV modes.
+ * Note!  This is in reply order (see loop in get_tv_modes).
+ * XXX: all 60Hz refresh?
+ */
+static const struct drm_display_mode sdvo_tv_modes[] = {
+	{ DRM_MODE("320x200", DRM_MODE_TYPE_DRIVER, 5815, 320, 321, 384,
+		   416, 0, 200, 201, 232, 233, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("320x240", DRM_MODE_TYPE_DRIVER, 6814, 320, 321, 384,
+		   416, 0, 240, 241, 272, 273, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("400x300", DRM_MODE_TYPE_DRIVER, 9910, 400, 401, 464,
+		   496, 0, 300, 301, 332, 333, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("640x350", DRM_MODE_TYPE_DRIVER, 16913, 640, 641, 704,
+		   736, 0, 350, 351, 382, 383, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("640x400", DRM_MODE_TYPE_DRIVER, 19121, 640, 641, 704,
+		   736, 0, 400, 401, 432, 433, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 22654, 640, 641, 704,
+		   736, 0, 480, 481, 512, 513, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("704x480", DRM_MODE_TYPE_DRIVER, 24624, 704, 705, 768,
+		   800, 0, 480, 481, 512, 513, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("704x576", DRM_MODE_TYPE_DRIVER, 29232, 704, 705, 768,
+		   800, 0, 576, 577, 608, 609, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("720x350", DRM_MODE_TYPE_DRIVER, 18751, 720, 721, 784,
+		   816, 0, 350, 351, 382, 383, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("720x400", DRM_MODE_TYPE_DRIVER, 21199, 720, 721, 784,
+		   816, 0, 400, 401, 432, 433, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("720x480", DRM_MODE_TYPE_DRIVER, 25116, 720, 721, 784,
+		   816, 0, 480, 481, 512, 513, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("720x540", DRM_MODE_TYPE_DRIVER, 28054, 720, 721, 784,
+		   816, 0, 540, 541, 572, 573, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("720x576", DRM_MODE_TYPE_DRIVER, 29816, 720, 721, 784,
+		   816, 0, 576, 577, 608, 609, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("768x576", DRM_MODE_TYPE_DRIVER, 31570, 768, 769, 832,
+		   864, 0, 576, 577, 608, 609, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("800x600", DRM_MODE_TYPE_DRIVER, 34030, 800, 801, 864,
+		   896, 0, 600, 601, 632, 633, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("832x624", DRM_MODE_TYPE_DRIVER, 36581, 832, 833, 896,
+		   928, 0, 624, 625, 656, 657, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("920x766", DRM_MODE_TYPE_DRIVER, 48707, 920, 921, 984,
+		   1016, 0, 766, 767, 798, 799, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 53827, 1024, 1025, 1088,
+		   1120, 0, 768, 769, 800, 801, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+	{ DRM_MODE("1280x1024", DRM_MODE_TYPE_DRIVER, 87265, 1280, 1281, 1344,
+		   1376, 0, 1024, 1025, 1056, 1057, 0,
+		   DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) },
+};
+
+static void psb_intel_sdvo_get_tv_modes(struct drm_connector *connector)
+{
+	struct psb_intel_sdvo *psb_intel_sdvo = intel_attached_sdvo(connector);
+	struct psb_intel_sdvo_sdtv_resolution_request tv_res;
+	uint32_t reply = 0, format_map = 0;
+	int i;
+
+	/* Read the list of supported input resolutions for the selected TV
+	 * format.
+	 */
+	format_map = 1 << psb_intel_sdvo->tv_format_index;
+	memcpy(&tv_res, &format_map,
+	       min(sizeof(format_map), sizeof(struct psb_intel_sdvo_sdtv_resolution_request)));
+
+	if (!psb_intel_sdvo_set_target_output(psb_intel_sdvo, psb_intel_sdvo->attached_output))
+		return;
+
+	BUILD_BUG_ON(sizeof(tv_res) != 3);
+	if (!psb_intel_sdvo_write_cmd(psb_intel_sdvo,
+				  SDVO_CMD_GET_SDTV_RESOLUTION_SUPPORT,
+				  &tv_res, sizeof(tv_res)))
+		return;
+	if (!psb_intel_sdvo_read_response(psb_intel_sdvo, &reply, 3))
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(sdvo_tv_modes); i++)
+		if (reply & (1 << i)) {
+			struct drm_display_mode *nmode;
+			nmode = drm_mode_duplicate(connector->dev,
+						   &sdvo_tv_modes[i]);
+			if (nmode)
+				drm_mode_probed_add(connector, nmode);
+		}
+}
+
+static void psb_intel_sdvo_get_lvds_modes(struct drm_connector *connector)
+{
+	struct psb_intel_sdvo *psb_intel_sdvo = intel_attached_sdvo(connector);
+	struct drm_psb_private *dev_priv = connector->dev->dev_private;
+	struct drm_display_mode *newmode;
+
+	/*
+	 * Attempt to get the mode list from DDC.
+	 * Assume that the preferred modes are
+	 * arranged in priority order.
+	 */
+	psb_intel_ddc_get_modes(connector, psb_intel_sdvo->i2c);
+	if (list_empty(&connector->probed_modes) == false)
+		goto end;
+
+	/* Fetch modes from VBT */
+	if (dev_priv->sdvo_lvds_vbt_mode != NULL) {
+		newmode = drm_mode_duplicate(connector->dev,
+					     dev_priv->sdvo_lvds_vbt_mode);
+		if (newmode != NULL) {
+			/* Guarantee the mode is preferred */
+			newmode->type = (DRM_MODE_TYPE_PREFERRED |
+					 DRM_MODE_TYPE_DRIVER);
+			drm_mode_probed_add(connector, newmode);
+		}
+	}
+
+end:
+	list_for_each_entry(newmode, &connector->probed_modes, head) {
+		if (newmode->type & DRM_MODE_TYPE_PREFERRED) {
+			psb_intel_sdvo->sdvo_lvds_fixed_mode =
+				drm_mode_duplicate(connector->dev, newmode);
+
+			drm_mode_set_crtcinfo(psb_intel_sdvo->sdvo_lvds_fixed_mode,
+					      0);
+
+			psb_intel_sdvo->is_lvds = true;
+			break;
+		}
+	}
+
 }
 
 static int psb_intel_sdvo_get_modes(struct drm_connector *connector)
 {
-	struct psb_intel_output *psb_intel_output =
-					to_psb_intel_output(connector);
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector = to_psb_intel_sdvo_connector(connector);
 
-	/* set the bus switch and get the modes */
-	psb_intel_sdvo_set_control_bus_switch(psb_intel_output,
-					  SDVO_CONTROL_BUS_DDC2);
-	psb_intel_ddc_get_modes(psb_intel_output);
+	if (IS_TV(psb_intel_sdvo_connector))
+		psb_intel_sdvo_get_tv_modes(connector);
+	else if (IS_LVDS(psb_intel_sdvo_connector))
+		psb_intel_sdvo_get_lvds_modes(connector);
+	else
+		psb_intel_sdvo_get_ddc_modes(connector);
 
-	if (list_empty(&connector->probed_modes))
-		return 0;
-	return 1;
+	return !list_empty(&connector->probed_modes);
+}
+
+static void
+psb_intel_sdvo_destroy_enhance_property(struct drm_connector *connector)
+{
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector = to_psb_intel_sdvo_connector(connector);
+	struct drm_device *dev = connector->dev;
+
+	if (psb_intel_sdvo_connector->left)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->left);
+	if (psb_intel_sdvo_connector->right)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->right);
+	if (psb_intel_sdvo_connector->top)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->top);
+	if (psb_intel_sdvo_connector->bottom)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->bottom);
+	if (psb_intel_sdvo_connector->hpos)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->hpos);
+	if (psb_intel_sdvo_connector->vpos)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->vpos);
+	if (psb_intel_sdvo_connector->saturation)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->saturation);
+	if (psb_intel_sdvo_connector->contrast)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->contrast);
+	if (psb_intel_sdvo_connector->hue)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->hue);
+	if (psb_intel_sdvo_connector->sharpness)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->sharpness);
+	if (psb_intel_sdvo_connector->flicker_filter)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->flicker_filter);
+	if (psb_intel_sdvo_connector->flicker_filter_2d)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->flicker_filter_2d);
+	if (psb_intel_sdvo_connector->flicker_filter_adaptive)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->flicker_filter_adaptive);
+	if (psb_intel_sdvo_connector->tv_luma_filter)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->tv_luma_filter);
+	if (psb_intel_sdvo_connector->tv_chroma_filter)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->tv_chroma_filter);
+	if (psb_intel_sdvo_connector->dot_crawl)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->dot_crawl);
+	if (psb_intel_sdvo_connector->brightness)
+		drm_property_destroy(dev, psb_intel_sdvo_connector->brightness);
 }
 
 static void psb_intel_sdvo_destroy(struct drm_connector *connector)
 {
-	struct psb_intel_output *psb_intel_output =
-				to_psb_intel_output(connector);
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector = to_psb_intel_sdvo_connector(connector);
 
-	if (psb_intel_output->i2c_bus)
-		psb_intel_i2c_destroy(psb_intel_output->i2c_bus);
+	if (psb_intel_sdvo_connector->tv_format)
+		drm_property_destroy(connector->dev,
+				     psb_intel_sdvo_connector->tv_format);
+
+	psb_intel_sdvo_destroy_enhance_property(connector);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
-	kfree(psb_intel_output);
+	kfree(connector);
+}
+
+static bool psb_intel_sdvo_detect_hdmi_audio(struct drm_connector *connector)
+{
+	struct psb_intel_sdvo *psb_intel_sdvo = intel_attached_sdvo(connector);
+	struct edid *edid;
+	bool has_audio = false;
+
+	if (!psb_intel_sdvo->is_hdmi)
+		return false;
+
+	edid = psb_intel_sdvo_get_edid(connector);
+	if (edid != NULL && edid->input & DRM_EDID_INPUT_DIGITAL)
+		has_audio = drm_detect_monitor_audio(edid);
+
+	return has_audio;
+}
+
+static int
+psb_intel_sdvo_set_property(struct drm_connector *connector,
+			struct drm_property *property,
+			uint64_t val)
+{
+	struct psb_intel_sdvo *psb_intel_sdvo = intel_attached_sdvo(connector);
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector = to_psb_intel_sdvo_connector(connector);
+	struct drm_psb_private *dev_priv = connector->dev->dev_private;
+	uint16_t temp_value;
+	uint8_t cmd;
+	int ret;
+
+	ret = drm_connector_property_set_value(connector, property, val);
+	if (ret)
+		return ret;
+
+	if (property == dev_priv->force_audio_property) {
+		int i = val;
+		bool has_audio;
+
+		if (i == psb_intel_sdvo_connector->force_audio)
+			return 0;
+
+		psb_intel_sdvo_connector->force_audio = i;
+
+		if (i == 0)
+			has_audio = psb_intel_sdvo_detect_hdmi_audio(connector);
+		else
+			has_audio = i > 0;
+
+		if (has_audio == psb_intel_sdvo->has_hdmi_audio)
+			return 0;
+
+		psb_intel_sdvo->has_hdmi_audio = has_audio;
+		goto done;
+	}
+
+	if (property == dev_priv->broadcast_rgb_property) {
+		if (val == !!psb_intel_sdvo->color_range)
+			return 0;
+
+		psb_intel_sdvo->color_range = val ? SDVO_COLOR_RANGE_16_235 : 0;
+		goto done;
+	}
+
+#define CHECK_PROPERTY(name, NAME) \
+	if (psb_intel_sdvo_connector->name == property) { \
+		if (psb_intel_sdvo_connector->cur_##name == temp_value) return 0; \
+		if (psb_intel_sdvo_connector->max_##name < temp_value) return -EINVAL; \
+		cmd = SDVO_CMD_SET_##NAME; \
+		psb_intel_sdvo_connector->cur_##name = temp_value; \
+		goto set_value; \
+	}
+
+	if (property == psb_intel_sdvo_connector->tv_format) {
+		if (val >= TV_FORMAT_NUM)
+			return -EINVAL;
+
+		if (psb_intel_sdvo->tv_format_index ==
+		    psb_intel_sdvo_connector->tv_format_supported[val])
+			return 0;
+
+		psb_intel_sdvo->tv_format_index = psb_intel_sdvo_connector->tv_format_supported[val];
+		goto done;
+	} else if (IS_TV_OR_LVDS(psb_intel_sdvo_connector)) {
+		temp_value = val;
+		if (psb_intel_sdvo_connector->left == property) {
+			drm_connector_property_set_value(connector,
+							 psb_intel_sdvo_connector->right, val);
+			if (psb_intel_sdvo_connector->left_margin == temp_value)
+				return 0;
+
+			psb_intel_sdvo_connector->left_margin = temp_value;
+			psb_intel_sdvo_connector->right_margin = temp_value;
+			temp_value = psb_intel_sdvo_connector->max_hscan -
+				psb_intel_sdvo_connector->left_margin;
+			cmd = SDVO_CMD_SET_OVERSCAN_H;
+			goto set_value;
+		} else if (psb_intel_sdvo_connector->right == property) {
+			drm_connector_property_set_value(connector,
+							 psb_intel_sdvo_connector->left, val);
+			if (psb_intel_sdvo_connector->right_margin == temp_value)
+				return 0;
+
+			psb_intel_sdvo_connector->left_margin = temp_value;
+			psb_intel_sdvo_connector->right_margin = temp_value;
+			temp_value = psb_intel_sdvo_connector->max_hscan -
+				psb_intel_sdvo_connector->left_margin;
+			cmd = SDVO_CMD_SET_OVERSCAN_H;
+			goto set_value;
+		} else if (psb_intel_sdvo_connector->top == property) {
+			drm_connector_property_set_value(connector,
+							 psb_intel_sdvo_connector->bottom, val);
+			if (psb_intel_sdvo_connector->top_margin == temp_value)
+				return 0;
+
+			psb_intel_sdvo_connector->top_margin = temp_value;
+			psb_intel_sdvo_connector->bottom_margin = temp_value;
+			temp_value = psb_intel_sdvo_connector->max_vscan -
+				psb_intel_sdvo_connector->top_margin;
+			cmd = SDVO_CMD_SET_OVERSCAN_V;
+			goto set_value;
+		} else if (psb_intel_sdvo_connector->bottom == property) {
+			drm_connector_property_set_value(connector,
+							 psb_intel_sdvo_connector->top, val);
+			if (psb_intel_sdvo_connector->bottom_margin == temp_value)
+				return 0;
+
+			psb_intel_sdvo_connector->top_margin = temp_value;
+			psb_intel_sdvo_connector->bottom_margin = temp_value;
+			temp_value = psb_intel_sdvo_connector->max_vscan -
+				psb_intel_sdvo_connector->top_margin;
+			cmd = SDVO_CMD_SET_OVERSCAN_V;
+			goto set_value;
+		}
+		CHECK_PROPERTY(hpos, HPOS)
+		CHECK_PROPERTY(vpos, VPOS)
+		CHECK_PROPERTY(saturation, SATURATION)
+		CHECK_PROPERTY(contrast, CONTRAST)
+		CHECK_PROPERTY(hue, HUE)
+		CHECK_PROPERTY(brightness, BRIGHTNESS)
+		CHECK_PROPERTY(sharpness, SHARPNESS)
+		CHECK_PROPERTY(flicker_filter, FLICKER_FILTER)
+		CHECK_PROPERTY(flicker_filter_2d, FLICKER_FILTER_2D)
+		CHECK_PROPERTY(flicker_filter_adaptive, FLICKER_FILTER_ADAPTIVE)
+		CHECK_PROPERTY(tv_chroma_filter, TV_CHROMA_FILTER)
+		CHECK_PROPERTY(tv_luma_filter, TV_LUMA_FILTER)
+		CHECK_PROPERTY(dot_crawl, DOT_CRAWL)
+	}
+
+	return -EINVAL; /* unknown property */
+
+set_value:
+	if (!psb_intel_sdvo_set_value(psb_intel_sdvo, cmd, &temp_value, 2))
+		return -EIO;
+
+
+done:
+	if (psb_intel_sdvo->base.base.crtc) {
+		struct drm_crtc *crtc = psb_intel_sdvo->base.base.crtc;
+		drm_crtc_helper_set_mode(crtc, &crtc->mode, crtc->x,
+					 crtc->y, crtc->fb);
+	}
+
+	return 0;
+#undef CHECK_PROPERTY
 }
 
 static const struct drm_encoder_helper_funcs psb_intel_sdvo_helper_funcs = {
@@ -1111,183 +1845,773 @@
 
 static const struct drm_connector_funcs psb_intel_sdvo_connector_funcs = {
 	.dpms = drm_helper_connector_dpms,
-	.save = psb_intel_sdvo_save,
-	.restore = psb_intel_sdvo_restore,
 	.detect = psb_intel_sdvo_detect,
 	.fill_modes = drm_helper_probe_single_connector_modes,
+	.set_property = psb_intel_sdvo_set_property,
 	.destroy = psb_intel_sdvo_destroy,
 };
 
-static const struct drm_connector_helper_funcs
-				psb_intel_sdvo_connector_helper_funcs = {
+static const struct drm_connector_helper_funcs psb_intel_sdvo_connector_helper_funcs = {
 	.get_modes = psb_intel_sdvo_get_modes,
 	.mode_valid = psb_intel_sdvo_mode_valid,
 	.best_encoder = psb_intel_best_encoder,
 };
 
-void psb_intel_sdvo_enc_destroy(struct drm_encoder *encoder)
+static void psb_intel_sdvo_enc_destroy(struct drm_encoder *encoder)
 {
-	drm_encoder_cleanup(encoder);
+	struct psb_intel_sdvo *psb_intel_sdvo = to_psb_intel_sdvo(encoder);
+
+	if (psb_intel_sdvo->sdvo_lvds_fixed_mode != NULL)
+		drm_mode_destroy(encoder->dev,
+				 psb_intel_sdvo->sdvo_lvds_fixed_mode);
+
+	i2c_del_adapter(&psb_intel_sdvo->ddc);
+	psb_intel_encoder_destroy(encoder);
 }
 
 static const struct drm_encoder_funcs psb_intel_sdvo_enc_funcs = {
 	.destroy = psb_intel_sdvo_enc_destroy,
 };
 
-
-void psb_intel_sdvo_init(struct drm_device *dev, int output_device)
+static void
+psb_intel_sdvo_guess_ddc_bus(struct psb_intel_sdvo *sdvo)
 {
-	struct drm_connector *connector;
-	struct psb_intel_output *psb_intel_output;
-	struct psb_intel_sdvo_priv *sdvo_priv;
-	struct psb_intel_i2c_chan *i2cbus = NULL;
-	int connector_type;
-	u8 ch[0x40];
-	int i;
-	int encoder_type, output_id;
+	/* FIXME: At the moment, ddc_bus = 2 is the only thing that works.
+	 * We need to figure out if this is true for all available poulsbo
+	 * hardware, or if we need to fiddle with the guessing code above.
+	 * The problem might go away if we can parse sdvo mappings from bios */
+	sdvo->ddc_bus = 2;
 
-	psb_intel_output =
-	    kcalloc(sizeof(struct psb_intel_output) +
-		    sizeof(struct psb_intel_sdvo_priv), 1, GFP_KERNEL);
-	if (!psb_intel_output)
-		return;
+#if 0
+	uint16_t mask = 0;
+	unsigned int num_bits;
 
-	connector = &psb_intel_output->base;
-
-	drm_connector_init(dev, connector, &psb_intel_sdvo_connector_funcs,
-			   DRM_MODE_CONNECTOR_Unknown);
-	drm_connector_helper_add(connector,
-				 &psb_intel_sdvo_connector_helper_funcs);
-	sdvo_priv = (struct psb_intel_sdvo_priv *) (psb_intel_output + 1);
-	psb_intel_output->type = INTEL_OUTPUT_SDVO;
-
-	connector->interlace_allowed = 0;
-	connector->doublescan_allowed = 0;
-
-	/* setup the DDC bus. */
-	if (output_device == SDVOB)
-		i2cbus =
-		    psb_intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOB");
-	else
-		i2cbus =
-		    psb_intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOC");
-
-	if (!i2cbus)
-		goto err_connector;
-
-	sdvo_priv->i2c_bus = i2cbus;
-
-	if (output_device == SDVOB) {
-		output_id = 1;
-		sdvo_priv->by_input_wiring = SDVOB_IN0;
-		sdvo_priv->i2c_bus->slave_addr = 0x38;
-	} else {
-		output_id = 2;
-		sdvo_priv->i2c_bus->slave_addr = 0x39;
+	/* Make a mask of outputs less than or equal to our own priority in the
+	 * list.
+	 */
+	switch (sdvo->controlled_output) {
+	case SDVO_OUTPUT_LVDS1:
+		mask |= SDVO_OUTPUT_LVDS1;
+	case SDVO_OUTPUT_LVDS0:
+		mask |= SDVO_OUTPUT_LVDS0;
+	case SDVO_OUTPUT_TMDS1:
+		mask |= SDVO_OUTPUT_TMDS1;
+	case SDVO_OUTPUT_TMDS0:
+		mask |= SDVO_OUTPUT_TMDS0;
+	case SDVO_OUTPUT_RGB1:
+		mask |= SDVO_OUTPUT_RGB1;
+	case SDVO_OUTPUT_RGB0:
+		mask |= SDVO_OUTPUT_RGB0;
+		break;
 	}
 
-	sdvo_priv->output_device = output_device;
-	psb_intel_output->i2c_bus = i2cbus;
-	psb_intel_output->dev_priv = sdvo_priv;
+	/* Count bits to find what number we are in the priority list. */
+	mask &= sdvo->caps.output_flags;
+	num_bits = hweight16(mask);
+	/* If more than 3 outputs, default to DDC bus 3 for now. */
+	if (num_bits > 3)
+		num_bits = 3;
 
+	/* Corresponds to SDVO_CONTROL_BUS_DDCx */
+	sdvo->ddc_bus = 1 << num_bits;
+#endif
+}
+
+/**
+ * Choose the appropriate DDC bus for control bus switch command for this
+ * SDVO output based on the controlled output.
+ *
+ * DDC bus number assignment is in a priority order of RGB outputs, then TMDS
+ * outputs, then LVDS outputs.
+ */
+static void
+psb_intel_sdvo_select_ddc_bus(struct drm_psb_private *dev_priv,
+			  struct psb_intel_sdvo *sdvo, u32 reg)
+{
+	struct sdvo_device_mapping *mapping;
+
+	if (IS_SDVOB(reg))
+		mapping = &(dev_priv->sdvo_mappings[0]);
+	else
+		mapping = &(dev_priv->sdvo_mappings[1]);
+
+	if (mapping->initialized)
+		sdvo->ddc_bus = 1 << ((mapping->ddc_pin & 0xf0) >> 4);
+	else
+		psb_intel_sdvo_guess_ddc_bus(sdvo);
+}
+
+static void
+psb_intel_sdvo_select_i2c_bus(struct drm_psb_private *dev_priv,
+			  struct psb_intel_sdvo *sdvo, u32 reg)
+{
+	struct sdvo_device_mapping *mapping;
+	u8 pin, speed;
+
+	if (IS_SDVOB(reg))
+		mapping = &dev_priv->sdvo_mappings[0];
+	else
+		mapping = &dev_priv->sdvo_mappings[1];
+
+	pin = GMBUS_PORT_DPB;
+	speed = GMBUS_RATE_1MHZ >> 8;
+	if (mapping->initialized) {
+		pin = mapping->i2c_pin;
+		speed = mapping->i2c_speed;
+	}
+
+	if (pin < GMBUS_NUM_PORTS) {
+		sdvo->i2c = &dev_priv->gmbus[pin].adapter;
+		gma_intel_gmbus_set_speed(sdvo->i2c, speed);
+		gma_intel_gmbus_force_bit(sdvo->i2c, true);
+	} else
+		sdvo->i2c = &dev_priv->gmbus[GMBUS_PORT_DPB].adapter;
+}
+
+static bool
+psb_intel_sdvo_is_hdmi_connector(struct psb_intel_sdvo *psb_intel_sdvo, int device)
+{
+	return psb_intel_sdvo_check_supp_encode(psb_intel_sdvo);
+}
+
+static u8
+psb_intel_sdvo_get_slave_addr(struct drm_device *dev, int sdvo_reg)
+{
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct sdvo_device_mapping *my_mapping, *other_mapping;
+
+	if (IS_SDVOB(sdvo_reg)) {
+		my_mapping = &dev_priv->sdvo_mappings[0];
+		other_mapping = &dev_priv->sdvo_mappings[1];
+	} else {
+		my_mapping = &dev_priv->sdvo_mappings[1];
+		other_mapping = &dev_priv->sdvo_mappings[0];
+	}
+
+	/* If the BIOS described our SDVO device, take advantage of it. */
+	if (my_mapping->slave_addr)
+		return my_mapping->slave_addr;
+
+	/* If the BIOS only described a different SDVO device, use the
+	 * address that it isn't using.
+	 */
+	if (other_mapping->slave_addr) {
+		if (other_mapping->slave_addr == 0x70)
+			return 0x72;
+		else
+			return 0x70;
+	}
+
+	/* No SDVO device info is found for another DVO port,
+	 * so use mapping assumption we had before BIOS parsing.
+	 */
+	if (IS_SDVOB(sdvo_reg))
+		return 0x70;
+	else
+		return 0x72;
+}
+
+static void
+psb_intel_sdvo_connector_init(struct psb_intel_sdvo_connector *connector,
+			  struct psb_intel_sdvo *encoder)
+{
+	drm_connector_init(encoder->base.base.dev,
+			   &connector->base.base,
+			   &psb_intel_sdvo_connector_funcs,
+			   connector->base.base.connector_type);
+
+	drm_connector_helper_add(&connector->base.base,
+				 &psb_intel_sdvo_connector_helper_funcs);
+
+	connector->base.base.interlace_allowed = 0;
+	connector->base.base.doublescan_allowed = 0;
+	connector->base.base.display_info.subpixel_order = SubPixelHorizontalRGB;
+
+	psb_intel_connector_attach_encoder(&connector->base, &encoder->base);
+	drm_sysfs_connector_add(&connector->base.base);
+}
+
+static void
+psb_intel_sdvo_add_hdmi_properties(struct psb_intel_sdvo_connector *connector)
+{
+	/* FIXME: We don't support HDMI at the moment
+	struct drm_device *dev = connector->base.base.dev;
+
+	intel_attach_force_audio_property(&connector->base.base);
+	if (INTEL_INFO(dev)->gen >= 4 && IS_MOBILE(dev))
+		intel_attach_broadcast_rgb_property(&connector->base.base);
+	*/
+}
+
+static bool
+psb_intel_sdvo_dvi_init(struct psb_intel_sdvo *psb_intel_sdvo, int device)
+{
+	struct drm_encoder *encoder = &psb_intel_sdvo->base.base;
+	struct drm_connector *connector;
+	struct psb_intel_connector *intel_connector;
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector;
+
+	psb_intel_sdvo_connector = kzalloc(sizeof(struct psb_intel_sdvo_connector), GFP_KERNEL);
+	if (!psb_intel_sdvo_connector)
+		return false;
+
+	if (device == 0) {
+		psb_intel_sdvo->controlled_output |= SDVO_OUTPUT_TMDS0;
+		psb_intel_sdvo_connector->output_flag = SDVO_OUTPUT_TMDS0;
+	} else if (device == 1) {
+		psb_intel_sdvo->controlled_output |= SDVO_OUTPUT_TMDS1;
+		psb_intel_sdvo_connector->output_flag = SDVO_OUTPUT_TMDS1;
+	}
+
+	intel_connector = &psb_intel_sdvo_connector->base;
+	connector = &intel_connector->base;
+	// connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT;
+	encoder->encoder_type = DRM_MODE_ENCODER_TMDS;
+	connector->connector_type = DRM_MODE_CONNECTOR_DVID;
+
+	if (psb_intel_sdvo_is_hdmi_connector(psb_intel_sdvo, device)) {
+		connector->connector_type = DRM_MODE_CONNECTOR_HDMIA;
+		psb_intel_sdvo->is_hdmi = true;
+	}
+	psb_intel_sdvo->base.clone_mask = ((1 << INTEL_SDVO_NON_TV_CLONE_BIT) |
+				       (1 << INTEL_ANALOG_CLONE_BIT));
+
+	psb_intel_sdvo_connector_init(psb_intel_sdvo_connector, psb_intel_sdvo);
+	if (psb_intel_sdvo->is_hdmi)
+		psb_intel_sdvo_add_hdmi_properties(psb_intel_sdvo_connector);
+
+	return true;
+}
+
+static bool
+psb_intel_sdvo_tv_init(struct psb_intel_sdvo *psb_intel_sdvo, int type)
+{
+	struct drm_encoder *encoder = &psb_intel_sdvo->base.base;
+	struct drm_connector *connector;
+	struct psb_intel_connector *intel_connector;
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector;
+
+	psb_intel_sdvo_connector = kzalloc(sizeof(struct psb_intel_sdvo_connector), GFP_KERNEL);
+	if (!psb_intel_sdvo_connector)
+		return false;
+
+	intel_connector = &psb_intel_sdvo_connector->base;
+	connector = &intel_connector->base;
+	encoder->encoder_type = DRM_MODE_ENCODER_TVDAC;
+	connector->connector_type = DRM_MODE_CONNECTOR_SVIDEO;
+
+	psb_intel_sdvo->controlled_output |= type;
+	psb_intel_sdvo_connector->output_flag = type;
+
+	psb_intel_sdvo->is_tv = true;
+	psb_intel_sdvo->base.needs_tv_clock = true;
+	psb_intel_sdvo->base.clone_mask = 1 << INTEL_SDVO_TV_CLONE_BIT;
+
+	psb_intel_sdvo_connector_init(psb_intel_sdvo_connector, psb_intel_sdvo);
+
+	if (!psb_intel_sdvo_tv_create_property(psb_intel_sdvo, psb_intel_sdvo_connector, type))
+		goto err;
+
+	if (!psb_intel_sdvo_create_enhance_property(psb_intel_sdvo, psb_intel_sdvo_connector))
+		goto err;
+
+	return true;
+
+err:
+	psb_intel_sdvo_destroy(connector);
+	return false;
+}
+
+static bool
+psb_intel_sdvo_analog_init(struct psb_intel_sdvo *psb_intel_sdvo, int device)
+{
+	struct drm_encoder *encoder = &psb_intel_sdvo->base.base;
+	struct drm_connector *connector;
+	struct psb_intel_connector *intel_connector;
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector;
+
+	psb_intel_sdvo_connector = kzalloc(sizeof(struct psb_intel_sdvo_connector), GFP_KERNEL);
+	if (!psb_intel_sdvo_connector)
+		return false;
+
+	intel_connector = &psb_intel_sdvo_connector->base;
+	connector = &intel_connector->base;
+	connector->polled = DRM_CONNECTOR_POLL_CONNECT;
+	encoder->encoder_type = DRM_MODE_ENCODER_DAC;
+	connector->connector_type = DRM_MODE_CONNECTOR_VGA;
+
+	if (device == 0) {
+		psb_intel_sdvo->controlled_output |= SDVO_OUTPUT_RGB0;
+		psb_intel_sdvo_connector->output_flag = SDVO_OUTPUT_RGB0;
+	} else if (device == 1) {
+		psb_intel_sdvo->controlled_output |= SDVO_OUTPUT_RGB1;
+		psb_intel_sdvo_connector->output_flag = SDVO_OUTPUT_RGB1;
+	}
+
+	psb_intel_sdvo->base.clone_mask = ((1 << INTEL_SDVO_NON_TV_CLONE_BIT) |
+				       (1 << INTEL_ANALOG_CLONE_BIT));
+
+	psb_intel_sdvo_connector_init(psb_intel_sdvo_connector,
+				  psb_intel_sdvo);
+	return true;
+}
+
+static bool
+psb_intel_sdvo_lvds_init(struct psb_intel_sdvo *psb_intel_sdvo, int device)
+{
+	struct drm_encoder *encoder = &psb_intel_sdvo->base.base;
+	struct drm_connector *connector;
+	struct psb_intel_connector *intel_connector;
+	struct psb_intel_sdvo_connector *psb_intel_sdvo_connector;
+
+	psb_intel_sdvo_connector = kzalloc(sizeof(struct psb_intel_sdvo_connector), GFP_KERNEL);
+	if (!psb_intel_sdvo_connector)
+		return false;
+
+	intel_connector = &psb_intel_sdvo_connector->base;
+	connector = &intel_connector->base;
+	encoder->encoder_type = DRM_MODE_ENCODER_LVDS;
+	connector->connector_type = DRM_MODE_CONNECTOR_LVDS;
+
+	if (device == 0) {
+		psb_intel_sdvo->controlled_output |= SDVO_OUTPUT_LVDS0;
+		psb_intel_sdvo_connector->output_flag = SDVO_OUTPUT_LVDS0;
+	} else if (device == 1) {
+		psb_intel_sdvo->controlled_output |= SDVO_OUTPUT_LVDS1;
+		psb_intel_sdvo_connector->output_flag = SDVO_OUTPUT_LVDS1;
+	}
+
+	psb_intel_sdvo->base.clone_mask = ((1 << INTEL_ANALOG_CLONE_BIT) |
+				       (1 << INTEL_SDVO_LVDS_CLONE_BIT));
+
+	psb_intel_sdvo_connector_init(psb_intel_sdvo_connector, psb_intel_sdvo);
+	if (!psb_intel_sdvo_create_enhance_property(psb_intel_sdvo, psb_intel_sdvo_connector))
+		goto err;
+
+	return true;
+
+err:
+	psb_intel_sdvo_destroy(connector);
+	return false;
+}
+
+static bool
+psb_intel_sdvo_output_setup(struct psb_intel_sdvo *psb_intel_sdvo, uint16_t flags)
+{
+	psb_intel_sdvo->is_tv = false;
+	psb_intel_sdvo->base.needs_tv_clock = false;
+	psb_intel_sdvo->is_lvds = false;
+
+	/* SDVO requires XXX1 function may not exist unless it has XXX0 function.*/
+
+	if (flags & SDVO_OUTPUT_TMDS0)
+		if (!psb_intel_sdvo_dvi_init(psb_intel_sdvo, 0))
+			return false;
+
+	if ((flags & SDVO_TMDS_MASK) == SDVO_TMDS_MASK)
+		if (!psb_intel_sdvo_dvi_init(psb_intel_sdvo, 1))
+			return false;
+
+	/* TV has no XXX1 function block */
+	if (flags & SDVO_OUTPUT_SVID0)
+		if (!psb_intel_sdvo_tv_init(psb_intel_sdvo, SDVO_OUTPUT_SVID0))
+			return false;
+
+	if (flags & SDVO_OUTPUT_CVBS0)
+		if (!psb_intel_sdvo_tv_init(psb_intel_sdvo, SDVO_OUTPUT_CVBS0))
+			return false;
+
+	if (flags & SDVO_OUTPUT_RGB0)
+		if (!psb_intel_sdvo_analog_init(psb_intel_sdvo, 0))
+			return false;
+
+	if ((flags & SDVO_RGB_MASK) == SDVO_RGB_MASK)
+		if (!psb_intel_sdvo_analog_init(psb_intel_sdvo, 1))
+			return false;
+
+	if (flags & SDVO_OUTPUT_LVDS0)
+		if (!psb_intel_sdvo_lvds_init(psb_intel_sdvo, 0))
+			return false;
+
+	if ((flags & SDVO_LVDS_MASK) == SDVO_LVDS_MASK)
+		if (!psb_intel_sdvo_lvds_init(psb_intel_sdvo, 1))
+			return false;
+
+	if ((flags & SDVO_OUTPUT_MASK) == 0) {
+		unsigned char bytes[2];
+
+		psb_intel_sdvo->controlled_output = 0;
+		memcpy(bytes, &psb_intel_sdvo->caps.output_flags, 2);
+		DRM_DEBUG_KMS("%s: Unknown SDVO output type (0x%02x%02x)\n",
+			      SDVO_NAME(psb_intel_sdvo),
+			      bytes[0], bytes[1]);
+		return false;
+	}
+	psb_intel_sdvo->base.crtc_mask = (1 << 0) | (1 << 1);
+
+	return true;
+}
+
+static bool psb_intel_sdvo_tv_create_property(struct psb_intel_sdvo *psb_intel_sdvo,
+					  struct psb_intel_sdvo_connector *psb_intel_sdvo_connector,
+					  int type)
+{
+	struct drm_device *dev = psb_intel_sdvo->base.base.dev;
+	struct psb_intel_sdvo_tv_format format;
+	uint32_t format_map, i;
+
+	if (!psb_intel_sdvo_set_target_output(psb_intel_sdvo, type))
+		return false;
+
+	BUILD_BUG_ON(sizeof(format) != 6);
+	if (!psb_intel_sdvo_get_value(psb_intel_sdvo,
+				  SDVO_CMD_GET_SUPPORTED_TV_FORMATS,
+				  &format, sizeof(format)))
+		return false;
+
+	memcpy(&format_map, &format, min(sizeof(format_map), sizeof(format)));
+
+	if (format_map == 0)
+		return false;
+
+	psb_intel_sdvo_connector->format_supported_num = 0;
+	for (i = 0 ; i < TV_FORMAT_NUM; i++)
+		if (format_map & (1 << i))
+			psb_intel_sdvo_connector->tv_format_supported[psb_intel_sdvo_connector->format_supported_num++] = i;
+
+
+	psb_intel_sdvo_connector->tv_format =
+			drm_property_create(dev, DRM_MODE_PROP_ENUM,
+					    "mode", psb_intel_sdvo_connector->format_supported_num);
+	if (!psb_intel_sdvo_connector->tv_format)
+		return false;
+
+	for (i = 0; i < psb_intel_sdvo_connector->format_supported_num; i++)
+		drm_property_add_enum(
+				psb_intel_sdvo_connector->tv_format, i,
+				i, tv_format_names[psb_intel_sdvo_connector->tv_format_supported[i]]);
+
+	psb_intel_sdvo->tv_format_index = psb_intel_sdvo_connector->tv_format_supported[0];
+	drm_connector_attach_property(&psb_intel_sdvo_connector->base.base,
+				      psb_intel_sdvo_connector->tv_format, 0);
+	return true;
+
+}
+
+#define ENHANCEMENT(name, NAME) do { \
+	if (enhancements.name) { \
+		if (!psb_intel_sdvo_get_value(psb_intel_sdvo, SDVO_CMD_GET_MAX_##NAME, &data_value, 4) || \
+		    !psb_intel_sdvo_get_value(psb_intel_sdvo, SDVO_CMD_GET_##NAME, &response, 2)) \
+			return false; \
+		psb_intel_sdvo_connector->max_##name = data_value[0]; \
+		psb_intel_sdvo_connector->cur_##name = response; \
+		psb_intel_sdvo_connector->name = \
+			drm_property_create(dev, DRM_MODE_PROP_RANGE, #name, 2); \
+		if (!psb_intel_sdvo_connector->name) return false; \
+		psb_intel_sdvo_connector->name->values[0] = 0; \
+		psb_intel_sdvo_connector->name->values[1] = data_value[0]; \
+		drm_connector_attach_property(connector, \
+					      psb_intel_sdvo_connector->name, \
+					      psb_intel_sdvo_connector->cur_##name); \
+		DRM_DEBUG_KMS(#name ": max %d, default %d, current %d\n", \
+			      data_value[0], data_value[1], response); \
+	} \
+} while(0)
+
+static bool
+psb_intel_sdvo_create_enhance_property_tv(struct psb_intel_sdvo *psb_intel_sdvo,
+				      struct psb_intel_sdvo_connector *psb_intel_sdvo_connector,
+				      struct psb_intel_sdvo_enhancements_reply enhancements)
+{
+	struct drm_device *dev = psb_intel_sdvo->base.base.dev;
+	struct drm_connector *connector = &psb_intel_sdvo_connector->base.base;
+	uint16_t response, data_value[2];
+
+	/* when horizontal overscan is supported, Add the left/right  property */
+	if (enhancements.overscan_h) {
+		if (!psb_intel_sdvo_get_value(psb_intel_sdvo,
+					  SDVO_CMD_GET_MAX_OVERSCAN_H,
+					  &data_value, 4))
+			return false;
+
+		if (!psb_intel_sdvo_get_value(psb_intel_sdvo,
+					  SDVO_CMD_GET_OVERSCAN_H,
+					  &response, 2))
+			return false;
+
+		psb_intel_sdvo_connector->max_hscan = data_value[0];
+		psb_intel_sdvo_connector->left_margin = data_value[0] - response;
+		psb_intel_sdvo_connector->right_margin = psb_intel_sdvo_connector->left_margin;
+		psb_intel_sdvo_connector->left =
+			drm_property_create(dev, DRM_MODE_PROP_RANGE,
+					    "left_margin", 2);
+		if (!psb_intel_sdvo_connector->left)
+			return false;
+
+		psb_intel_sdvo_connector->left->values[0] = 0;
+		psb_intel_sdvo_connector->left->values[1] = data_value[0];
+		drm_connector_attach_property(connector,
+					      psb_intel_sdvo_connector->left,
+					      psb_intel_sdvo_connector->left_margin);
+
+		psb_intel_sdvo_connector->right =
+			drm_property_create(dev, DRM_MODE_PROP_RANGE,
+					    "right_margin", 2);
+		if (!psb_intel_sdvo_connector->right)
+			return false;
+
+		psb_intel_sdvo_connector->right->values[0] = 0;
+		psb_intel_sdvo_connector->right->values[1] = data_value[0];
+		drm_connector_attach_property(connector,
+					      psb_intel_sdvo_connector->right,
+					      psb_intel_sdvo_connector->right_margin);
+		DRM_DEBUG_KMS("h_overscan: max %d, "
+			      "default %d, current %d\n",
+			      data_value[0], data_value[1], response);
+	}
+
+	if (enhancements.overscan_v) {
+		if (!psb_intel_sdvo_get_value(psb_intel_sdvo,
+					  SDVO_CMD_GET_MAX_OVERSCAN_V,
+					  &data_value, 4))
+			return false;
+
+		if (!psb_intel_sdvo_get_value(psb_intel_sdvo,
+					  SDVO_CMD_GET_OVERSCAN_V,
+					  &response, 2))
+			return false;
+
+		psb_intel_sdvo_connector->max_vscan = data_value[0];
+		psb_intel_sdvo_connector->top_margin = data_value[0] - response;
+		psb_intel_sdvo_connector->bottom_margin = psb_intel_sdvo_connector->top_margin;
+		psb_intel_sdvo_connector->top =
+			drm_property_create(dev, DRM_MODE_PROP_RANGE,
+					    "top_margin", 2);
+		if (!psb_intel_sdvo_connector->top)
+			return false;
+
+		psb_intel_sdvo_connector->top->values[0] = 0;
+		psb_intel_sdvo_connector->top->values[1] = data_value[0];
+		drm_connector_attach_property(connector,
+					      psb_intel_sdvo_connector->top,
+					      psb_intel_sdvo_connector->top_margin);
+
+		psb_intel_sdvo_connector->bottom =
+			drm_property_create(dev, DRM_MODE_PROP_RANGE,
+					    "bottom_margin", 2);
+		if (!psb_intel_sdvo_connector->bottom)
+			return false;
+
+		psb_intel_sdvo_connector->bottom->values[0] = 0;
+		psb_intel_sdvo_connector->bottom->values[1] = data_value[0];
+		drm_connector_attach_property(connector,
+					      psb_intel_sdvo_connector->bottom,
+					      psb_intel_sdvo_connector->bottom_margin);
+		DRM_DEBUG_KMS("v_overscan: max %d, "
+			      "default %d, current %d\n",
+			      data_value[0], data_value[1], response);
+	}
+
+	ENHANCEMENT(hpos, HPOS);
+	ENHANCEMENT(vpos, VPOS);
+	ENHANCEMENT(saturation, SATURATION);
+	ENHANCEMENT(contrast, CONTRAST);
+	ENHANCEMENT(hue, HUE);
+	ENHANCEMENT(sharpness, SHARPNESS);
+	ENHANCEMENT(brightness, BRIGHTNESS);
+	ENHANCEMENT(flicker_filter, FLICKER_FILTER);
+	ENHANCEMENT(flicker_filter_adaptive, FLICKER_FILTER_ADAPTIVE);
+	ENHANCEMENT(flicker_filter_2d, FLICKER_FILTER_2D);
+	ENHANCEMENT(tv_chroma_filter, TV_CHROMA_FILTER);
+	ENHANCEMENT(tv_luma_filter, TV_LUMA_FILTER);
+
+	if (enhancements.dot_crawl) {
+		if (!psb_intel_sdvo_get_value(psb_intel_sdvo, SDVO_CMD_GET_DOT_CRAWL, &response, 2))
+			return false;
+
+		psb_intel_sdvo_connector->max_dot_crawl = 1;
+		psb_intel_sdvo_connector->cur_dot_crawl = response & 0x1;
+		psb_intel_sdvo_connector->dot_crawl =
+			drm_property_create(dev, DRM_MODE_PROP_RANGE, "dot_crawl", 2);
+		if (!psb_intel_sdvo_connector->dot_crawl)
+			return false;
+
+		psb_intel_sdvo_connector->dot_crawl->values[0] = 0;
+		psb_intel_sdvo_connector->dot_crawl->values[1] = 1;
+		drm_connector_attach_property(connector,
+					      psb_intel_sdvo_connector->dot_crawl,
+					      psb_intel_sdvo_connector->cur_dot_crawl);
+		DRM_DEBUG_KMS("dot crawl: current %d\n", response);
+	}
+
+	return true;
+}
+
+static bool
+psb_intel_sdvo_create_enhance_property_lvds(struct psb_intel_sdvo *psb_intel_sdvo,
+					struct psb_intel_sdvo_connector *psb_intel_sdvo_connector,
+					struct psb_intel_sdvo_enhancements_reply enhancements)
+{
+	struct drm_device *dev = psb_intel_sdvo->base.base.dev;
+	struct drm_connector *connector = &psb_intel_sdvo_connector->base.base;
+	uint16_t response, data_value[2];
+
+	ENHANCEMENT(brightness, BRIGHTNESS);
+
+	return true;
+}
+#undef ENHANCEMENT
+
+static bool psb_intel_sdvo_create_enhance_property(struct psb_intel_sdvo *psb_intel_sdvo,
+					       struct psb_intel_sdvo_connector *psb_intel_sdvo_connector)
+{
+	union {
+		struct psb_intel_sdvo_enhancements_reply reply;
+		uint16_t response;
+	} enhancements;
+
+	BUILD_BUG_ON(sizeof(enhancements) != 2);
+
+	enhancements.response = 0;
+	psb_intel_sdvo_get_value(psb_intel_sdvo,
+			     SDVO_CMD_GET_SUPPORTED_ENHANCEMENTS,
+			     &enhancements, sizeof(enhancements));
+	if (enhancements.response == 0) {
+		DRM_DEBUG_KMS("No enhancement is supported\n");
+		return true;
+	}
+
+	if (IS_TV(psb_intel_sdvo_connector))
+		return psb_intel_sdvo_create_enhance_property_tv(psb_intel_sdvo, psb_intel_sdvo_connector, enhancements.reply);
+	else if(IS_LVDS(psb_intel_sdvo_connector))
+		return psb_intel_sdvo_create_enhance_property_lvds(psb_intel_sdvo, psb_intel_sdvo_connector, enhancements.reply);
+	else
+		return true;
+}
+
+static int psb_intel_sdvo_ddc_proxy_xfer(struct i2c_adapter *adapter,
+				     struct i2c_msg *msgs,
+				     int num)
+{
+	struct psb_intel_sdvo *sdvo = adapter->algo_data;
+
+	if (!psb_intel_sdvo_set_control_bus_switch(sdvo, sdvo->ddc_bus))
+		return -EIO;
+
+	return sdvo->i2c->algo->master_xfer(sdvo->i2c, msgs, num);
+}
+
+static u32 psb_intel_sdvo_ddc_proxy_func(struct i2c_adapter *adapter)
+{
+	struct psb_intel_sdvo *sdvo = adapter->algo_data;
+	return sdvo->i2c->algo->functionality(sdvo->i2c);
+}
+
+static const struct i2c_algorithm psb_intel_sdvo_ddc_proxy = {
+	.master_xfer	= psb_intel_sdvo_ddc_proxy_xfer,
+	.functionality	= psb_intel_sdvo_ddc_proxy_func
+};
+
+static bool
+psb_intel_sdvo_init_ddc_proxy(struct psb_intel_sdvo *sdvo,
+			  struct drm_device *dev)
+{
+	sdvo->ddc.owner = THIS_MODULE;
+	sdvo->ddc.class = I2C_CLASS_DDC;
+	snprintf(sdvo->ddc.name, I2C_NAME_SIZE, "SDVO DDC proxy");
+	sdvo->ddc.dev.parent = &dev->pdev->dev;
+	sdvo->ddc.algo_data = sdvo;
+	sdvo->ddc.algo = &psb_intel_sdvo_ddc_proxy;
+
+	return i2c_add_adapter(&sdvo->ddc) == 0;
+}
+
+bool psb_intel_sdvo_init(struct drm_device *dev, int sdvo_reg)
+{
+	struct drm_psb_private *dev_priv = dev->dev_private;
+	struct psb_intel_encoder *psb_intel_encoder;
+	struct psb_intel_sdvo *psb_intel_sdvo;
+	int i;
+
+	psb_intel_sdvo = kzalloc(sizeof(struct psb_intel_sdvo), GFP_KERNEL);
+	if (!psb_intel_sdvo)
+		return false;
+
+	psb_intel_sdvo->sdvo_reg = sdvo_reg;
+	psb_intel_sdvo->slave_addr = psb_intel_sdvo_get_slave_addr(dev, sdvo_reg) >> 1;
+	psb_intel_sdvo_select_i2c_bus(dev_priv, psb_intel_sdvo, sdvo_reg);
+	if (!psb_intel_sdvo_init_ddc_proxy(psb_intel_sdvo, dev)) {
+		kfree(psb_intel_sdvo);
+		return false;
+	}
+
+	/* encoder type will be decided later */
+	psb_intel_encoder = &psb_intel_sdvo->base;
+	psb_intel_encoder->type = INTEL_OUTPUT_SDVO;
+	drm_encoder_init(dev, &psb_intel_encoder->base, &psb_intel_sdvo_enc_funcs, 0);
 
 	/* Read the regs to test if we can talk to the device */
 	for (i = 0; i < 0x40; i++) {
-		if (!psb_intel_sdvo_read_byte(psb_intel_output, i, &ch[i])) {
-			dev_dbg(dev->dev, "No SDVO device found on SDVO%c\n",
-				  output_device == SDVOB ? 'B' : 'C');
-			goto err_i2c;
+		u8 byte;
+
+		if (!psb_intel_sdvo_read_byte(psb_intel_sdvo, i, &byte)) {
+			DRM_DEBUG_KMS("No SDVO device found on SDVO%c\n",
+				      IS_SDVOB(sdvo_reg) ? 'B' : 'C');
+			goto err;
 		}
 	}
 
-	psb_intel_sdvo_get_capabilities(psb_intel_output, &sdvo_priv->caps);
+	if (IS_SDVOB(sdvo_reg))
+		dev_priv->hotplug_supported_mask |= SDVOB_HOTPLUG_INT_STATUS;
+	else
+		dev_priv->hotplug_supported_mask |= SDVOC_HOTPLUG_INT_STATUS;
 
-	memset(&sdvo_priv->active_outputs, 0,
-	       sizeof(sdvo_priv->active_outputs));
+	drm_encoder_helper_add(&psb_intel_encoder->base, &psb_intel_sdvo_helper_funcs);
 
-	/* TODO, CVBS, SVID, YPRPB & SCART outputs. */
-	if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_RGB0) {
-		sdvo_priv->active_outputs = SDVO_OUTPUT_RGB0;
-		sdvo_priv->active_device = SDVO_DEVICE_CRT;
-		connector->display_info.subpixel_order =
-		    SubPixelHorizontalRGB;
-		encoder_type = DRM_MODE_ENCODER_DAC;
-		connector_type = DRM_MODE_CONNECTOR_VGA;
-	} else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_RGB1) {
-		sdvo_priv->active_outputs = SDVO_OUTPUT_RGB1;
-		sdvo_priv->active_outputs = SDVO_DEVICE_CRT;
-		connector->display_info.subpixel_order =
-		    SubPixelHorizontalRGB;
-		encoder_type = DRM_MODE_ENCODER_DAC;
-		connector_type = DRM_MODE_CONNECTOR_VGA;
-	} else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_TMDS0) {
-		sdvo_priv->active_outputs = SDVO_OUTPUT_TMDS0;
-		sdvo_priv->active_device = SDVO_DEVICE_TMDS;
-		connector->display_info.subpixel_order =
-		    SubPixelHorizontalRGB;
-		encoder_type = DRM_MODE_ENCODER_TMDS;
-		connector_type = DRM_MODE_CONNECTOR_DVID;
-	} else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_TMDS1) {
-		sdvo_priv->active_outputs = SDVO_OUTPUT_TMDS1;
-		sdvo_priv->active_device = SDVO_DEVICE_TMDS;
-		connector->display_info.subpixel_order =
-		    SubPixelHorizontalRGB;
-		encoder_type = DRM_MODE_ENCODER_TMDS;
-		connector_type = DRM_MODE_CONNECTOR_DVID;
-	} else {
-		unsigned char bytes[2];
+	/* In default case sdvo lvds is false */
+	if (!psb_intel_sdvo_get_capabilities(psb_intel_sdvo, &psb_intel_sdvo->caps))
+		goto err;
 
-		memcpy(bytes, &sdvo_priv->caps.output_flags, 2);
-		dev_dbg(dev->dev, "%s: No active RGB or TMDS outputs (0x%02x%02x)\n",
-		     SDVO_NAME(sdvo_priv), bytes[0], bytes[1]);
-		goto err_i2c;
+	if (psb_intel_sdvo_output_setup(psb_intel_sdvo,
+				    psb_intel_sdvo->caps.output_flags) != true) {
+		DRM_DEBUG_KMS("SDVO output failed to setup on SDVO%c\n",
+			      IS_SDVOB(sdvo_reg) ? 'B' : 'C');
+		goto err;
 	}
 
-	drm_encoder_init(dev, &psb_intel_output->enc, &psb_intel_sdvo_enc_funcs,
-			 encoder_type);
-	drm_encoder_helper_add(&psb_intel_output->enc,
-			       &psb_intel_sdvo_helper_funcs);
-	connector->connector_type = connector_type;
-
-	drm_mode_connector_attach_encoder(&psb_intel_output->base,
-					  &psb_intel_output->enc);
-	drm_sysfs_connector_add(connector);
+	psb_intel_sdvo_select_ddc_bus(dev_priv, psb_intel_sdvo, sdvo_reg);
 
 	/* Set the input timing to the screen. Assume always input 0. */
-	psb_intel_sdvo_set_target_input(psb_intel_output, true, false);
+	if (!psb_intel_sdvo_set_target_input(psb_intel_sdvo))
+		goto err;
 
-	psb_intel_sdvo_get_input_pixel_clock_range(psb_intel_output,
-					       &sdvo_priv->pixel_clock_min,
-					       &sdvo_priv->
-					       pixel_clock_max);
+	if (!psb_intel_sdvo_get_input_pixel_clock_range(psb_intel_sdvo,
+						    &psb_intel_sdvo->pixel_clock_min,
+						    &psb_intel_sdvo->pixel_clock_max))
+		goto err;
 
+	DRM_DEBUG_KMS("%s device VID/DID: %02X:%02X.%02X, "
+			"clock range %dMHz - %dMHz, "
+			"input 1: %c, input 2: %c, "
+			"output 1: %c, output 2: %c\n",
+			SDVO_NAME(psb_intel_sdvo),
+			psb_intel_sdvo->caps.vendor_id, psb_intel_sdvo->caps.device_id,
+			psb_intel_sdvo->caps.device_rev_id,
+			psb_intel_sdvo->pixel_clock_min / 1000,
+			psb_intel_sdvo->pixel_clock_max / 1000,
+			(psb_intel_sdvo->caps.sdvo_inputs_mask & 0x1) ? 'Y' : 'N',
+			(psb_intel_sdvo->caps.sdvo_inputs_mask & 0x2) ? 'Y' : 'N',
+			/* check currently supported outputs */
+			psb_intel_sdvo->caps.output_flags &
+			(SDVO_OUTPUT_TMDS0 | SDVO_OUTPUT_RGB0) ? 'Y' : 'N',
+			psb_intel_sdvo->caps.output_flags &
+			(SDVO_OUTPUT_TMDS1 | SDVO_OUTPUT_RGB1) ? 'Y' : 'N');
+	return true;
 
-	dev_dbg(dev->dev, "%s device VID/DID: %02X:%02X.%02X, "
-		  "clock range %dMHz - %dMHz, "
-		  "input 1: %c, input 2: %c, "
-		  "output 1: %c, output 2: %c\n",
-		  SDVO_NAME(sdvo_priv),
-		  sdvo_priv->caps.vendor_id, sdvo_priv->caps.device_id,
-		  sdvo_priv->caps.device_rev_id,
-		  sdvo_priv->pixel_clock_min / 1000,
-		  sdvo_priv->pixel_clock_max / 1000,
-		  (sdvo_priv->caps.sdvo_inputs_mask & 0x1) ? 'Y' : 'N',
-		  (sdvo_priv->caps.sdvo_inputs_mask & 0x2) ? 'Y' : 'N',
-		  /* check currently supported outputs */
-		  sdvo_priv->caps.output_flags &
-		  (SDVO_OUTPUT_TMDS0 | SDVO_OUTPUT_RGB0) ? 'Y' : 'N',
-		  sdvo_priv->caps.output_flags &
-		  (SDVO_OUTPUT_TMDS1 | SDVO_OUTPUT_RGB1) ? 'Y' : 'N');
+err:
+	drm_encoder_cleanup(&psb_intel_encoder->base);
+	i2c_del_adapter(&psb_intel_sdvo->ddc);
+	kfree(psb_intel_sdvo);
 
-	psb_intel_output->ddc_bus = i2cbus;
-
-	return;
-
-err_i2c:
-	psb_intel_i2c_destroy(psb_intel_output->i2c_bus);
-err_connector:
-	drm_connector_cleanup(connector);
-	kfree(psb_intel_output);
-
-	return;
+	return false;
 }
diff --git a/drivers/gpu/drm/gma500/psb_intel_sdvo_regs.h b/drivers/gpu/drm/gma500/psb_intel_sdvo_regs.h
index 96862ea..600e797 100644
--- a/drivers/gpu/drm/gma500/psb_intel_sdvo_regs.h
+++ b/drivers/gpu/drm/gma500/psb_intel_sdvo_regs.h
@@ -1,25 +1,33 @@
 /*
- * SDVO command definitions and structures.
+ * Copyright ? 2006-2007 Intel Corporation
  *
- * Copyright (c) 2008, Intel Corporation
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
  *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
  *
  * Authors:
  *	Eric Anholt <eric@anholt.net>
  */
 
+/**
+ * @file SDVO command definitions and structures.
+ */
+
 #define SDVO_OUTPUT_FIRST   (0)
 #define SDVO_OUTPUT_TMDS0   (1 << 0)
 #define SDVO_OUTPUT_RGB0    (1 << 1)
@@ -38,62 +46,64 @@
 #define SDVO_OUTPUT_LAST    (14)
 
 struct psb_intel_sdvo_caps {
-	u8 vendor_id;
-	u8 device_id;
-	u8 device_rev_id;
-	u8 sdvo_version_major;
-	u8 sdvo_version_minor;
-	unsigned int sdvo_inputs_mask:2;
-	unsigned int smooth_scaling:1;
-	unsigned int sharp_scaling:1;
-	unsigned int up_scaling:1;
-	unsigned int down_scaling:1;
-	unsigned int stall_support:1;
-	unsigned int pad:1;
-	u16 output_flags;
-} __packed;
+    u8 vendor_id;
+    u8 device_id;
+    u8 device_rev_id;
+    u8 sdvo_version_major;
+    u8 sdvo_version_minor;
+    unsigned int sdvo_inputs_mask:2;
+    unsigned int smooth_scaling:1;
+    unsigned int sharp_scaling:1;
+    unsigned int up_scaling:1;
+    unsigned int down_scaling:1;
+    unsigned int stall_support:1;
+    unsigned int pad:1;
+    u16 output_flags;
+} __attribute__((packed));
 
 /** This matches the EDID DTD structure, more or less */
 struct psb_intel_sdvo_dtd {
-	struct {
-		u16 clock;	/**< pixel clock, in 10kHz units */
-		u8 h_active;	/**< lower 8 bits (pixels) */
-		u8 h_blank;	/**< lower 8 bits (pixels) */
-		u8 h_high;	/**< upper 4 bits each h_active, h_blank */
-		u8 v_active;	/**< lower 8 bits (lines) */
-		u8 v_blank;	/**< lower 8 bits (lines) */
-		u8 v_high;	/**< upper 4 bits each v_active, v_blank */
-	} part1;
+    struct {
+	u16 clock;		/**< pixel clock, in 10kHz units */
+	u8 h_active;		/**< lower 8 bits (pixels) */
+	u8 h_blank;		/**< lower 8 bits (pixels) */
+	u8 h_high;		/**< upper 4 bits each h_active, h_blank */
+	u8 v_active;		/**< lower 8 bits (lines) */
+	u8 v_blank;		/**< lower 8 bits (lines) */
+	u8 v_high;		/**< upper 4 bits each v_active, v_blank */
+    } part1;
 
-	struct {
-		u8 h_sync_off;
-			/**< lower 8 bits, from hblank start */
-		u8 h_sync_width;/**< lower 8 bits (pixels) */
+    struct {
+	u8 h_sync_off;	/**< lower 8 bits, from hblank start */
+	u8 h_sync_width;	/**< lower 8 bits (pixels) */
 	/** lower 4 bits each vsync offset, vsync width */
-		u8 v_sync_off_width;
+	u8 v_sync_off_width;
 	/**
 	 * 2 high bits of hsync offset, 2 high bits of hsync width,
 	 * bits 4-5 of vsync offset, and 2 high bits of vsync width.
 	 */
-		u8 sync_off_width_high;
-		u8 dtd_flags;
-		u8 sdvo_flags;
+	u8 sync_off_width_high;
+	u8 dtd_flags;
+	u8 sdvo_flags;
 	/** bits 6-7 of vsync offset at bits 6-7 */
-		u8 v_sync_off_high;
-		u8 reserved;
-	} part2;
-} __packed;
+	u8 v_sync_off_high;
+	u8 reserved;
+    } part2;
+} __attribute__((packed));
 
 struct psb_intel_sdvo_pixel_clock_range {
-	u16 min;		/**< pixel clock, in 10kHz units */
-	u16 max;		/**< pixel clock, in 10kHz units */
-} __packed;
+    u16 min;			/**< pixel clock, in 10kHz units */
+    u16 max;			/**< pixel clock, in 10kHz units */
+} __attribute__((packed));
 
 struct psb_intel_sdvo_preferred_input_timing_args {
-	u16 clock;
-	u16 width;
-	u16 height;
-} __packed;
+    u16 clock;
+    u16 width;
+    u16 height;
+    u8	interlace:1;
+    u8	scaled:1;
+    u8	pad:6;
+} __attribute__((packed));
 
 /* I2C registers for SDVO */
 #define SDVO_I2C_ARG_0				0x07
@@ -129,7 +139,7 @@
 
 #define SDVO_CMD_RESET					0x01
 
-/** Returns a struct psb_intel_sdvo_caps */
+/** Returns a struct intel_sdvo_caps */
 #define SDVO_CMD_GET_DEVICE_CAPS			0x02
 
 #define SDVO_CMD_GET_FIRMWARE_REV			0x86
@@ -144,19 +154,18 @@
  */
 #define SDVO_CMD_GET_TRAINED_INPUTS			0x03
 struct psb_intel_sdvo_get_trained_inputs_response {
-	unsigned int input0_trained:1;
-	unsigned int input1_trained:1;
-	unsigned int pad:6;
-} __packed;
+    unsigned int input0_trained:1;
+    unsigned int input1_trained:1;
+    unsigned int pad:6;
+} __attribute__((packed));
 
-/** Returns a struct psb_intel_sdvo_output_flags of active outputs. */
+/** Returns a struct intel_sdvo_output_flags of active outputs. */
 #define SDVO_CMD_GET_ACTIVE_OUTPUTS			0x04
 
 /**
  * Sets the current set of active outputs.
  *
- * Takes a struct psb_intel_sdvo_output_flags.
- * Must be preceded by a SET_IN_OUT_MAP
+ * Takes a struct intel_sdvo_output_flags.  Must be preceded by a SET_IN_OUT_MAP
  * on multi-output devices.
  */
 #define SDVO_CMD_SET_ACTIVE_OUTPUTS			0x05
@@ -164,9 +173,12 @@
 /**
  * Returns the current mapping of SDVO inputs to outputs on the device.
  *
- * Returns two struct psb_intel_sdvo_output_flags structures.
+ * Returns two struct intel_sdvo_output_flags structures.
  */
 #define SDVO_CMD_GET_IN_OUT_MAP				0x06
+struct psb_intel_sdvo_in_out_map {
+    u16 in0, in1;
+};
 
 /**
  * Sets the current mapping of SDVO inputs to outputs on the device.
@@ -176,32 +188,33 @@
 #define SDVO_CMD_SET_IN_OUT_MAP				0x07
 
 /**
- * Returns a struct psb_intel_sdvo_output_flags of attached displays.
+ * Returns a struct intel_sdvo_output_flags of attached displays.
  */
 #define SDVO_CMD_GET_ATTACHED_DISPLAYS			0x0b
 
 /**
- * Returns a struct psb_intel_sdvo_ouptut_flags of displays supporting hot plugging.
+ * Returns a struct intel_sdvo_ouptut_flags of displays supporting hot plugging.
  */
 #define SDVO_CMD_GET_HOT_PLUG_SUPPORT			0x0c
 
 /**
- * Takes a struct psb_intel_sdvo_output_flags.
+ * Takes a struct intel_sdvo_output_flags.
  */
 #define SDVO_CMD_SET_ACTIVE_HOT_PLUG			0x0d
 
 /**
- * Returns a struct psb_intel_sdvo_output_flags of displays with hot plug
+ * Returns a struct intel_sdvo_output_flags of displays with hot plug
  * interrupts enabled.
  */
 #define SDVO_CMD_GET_ACTIVE_HOT_PLUG			0x0e
 
 #define SDVO_CMD_GET_INTERRUPT_EVENT_SOURCE		0x0f
-struct psb_intel_sdvo_get_interrupt_event_source_response {
-	u16 interrupt_status;
-	unsigned int ambient_light_interrupt:1;
-	unsigned int pad:7;
-} __packed;
+struct intel_sdvo_get_interrupt_event_source_response {
+    u16 interrupt_status;
+    unsigned int ambient_light_interrupt:1;
+    unsigned int hdmi_audio_encrypt_change:1;
+    unsigned int pad:6;
+} __attribute__((packed));
 
 /**
  * Selects which input is affected by future input commands.
@@ -212,12 +225,12 @@
  */
 #define SDVO_CMD_SET_TARGET_INPUT			0x10
 struct psb_intel_sdvo_set_target_input_args {
-	unsigned int target_1:1;
-	unsigned int pad:7;
-} __packed;
+    unsigned int target_1:1;
+    unsigned int pad:7;
+} __attribute__((packed));
 
 /**
- * Takes a struct psb_intel_sdvo_output_flags of which outputs are targeted by
+ * Takes a struct intel_sdvo_output_flags of which outputs are targeted by
  * future output commands.
  *
  * Affected commands inclue SET_OUTPUT_TIMINGS_PART[12],
@@ -282,9 +295,9 @@
 #define SDVO_CMD_GET_PREFERRED_INPUT_TIMING_PART1	0x1b
 #define SDVO_CMD_GET_PREFERRED_INPUT_TIMING_PART2	0x1c
 
-/** Returns a struct psb_intel_sdvo_pixel_clock_range */
+/** Returns a struct intel_sdvo_pixel_clock_range */
 #define SDVO_CMD_GET_INPUT_PIXEL_CLOCK_RANGE		0x1d
-/** Returns a struct psb_intel_sdvo_pixel_clock_range */
+/** Returns a struct intel_sdvo_pixel_clock_range */
 #define SDVO_CMD_GET_OUTPUT_PIXEL_CLOCK_RANGE		0x1e
 
 /** Returns a byte bitfield containing SDVO_CLOCK_RATE_MULT_* flags */
@@ -299,40 +312,412 @@
 # define SDVO_CLOCK_RATE_MULT_4X				(1 << 3)
 
 #define SDVO_CMD_GET_SUPPORTED_TV_FORMATS		0x27
+/** 6 bytes of bit flags for TV formats shared by all TV format functions */
+struct psb_intel_sdvo_tv_format {
+    unsigned int ntsc_m:1;
+    unsigned int ntsc_j:1;
+    unsigned int ntsc_443:1;
+    unsigned int pal_b:1;
+    unsigned int pal_d:1;
+    unsigned int pal_g:1;
+    unsigned int pal_h:1;
+    unsigned int pal_i:1;
+
+    unsigned int pal_m:1;
+    unsigned int pal_n:1;
+    unsigned int pal_nc:1;
+    unsigned int pal_60:1;
+    unsigned int secam_b:1;
+    unsigned int secam_d:1;
+    unsigned int secam_g:1;
+    unsigned int secam_k:1;
+
+    unsigned int secam_k1:1;
+    unsigned int secam_l:1;
+    unsigned int secam_60:1;
+    unsigned int hdtv_std_smpte_240m_1080i_59:1;
+    unsigned int hdtv_std_smpte_240m_1080i_60:1;
+    unsigned int hdtv_std_smpte_260m_1080i_59:1;
+    unsigned int hdtv_std_smpte_260m_1080i_60:1;
+    unsigned int hdtv_std_smpte_274m_1080i_50:1;
+
+    unsigned int hdtv_std_smpte_274m_1080i_59:1;
+    unsigned int hdtv_std_smpte_274m_1080i_60:1;
+    unsigned int hdtv_std_smpte_274m_1080p_23:1;
+    unsigned int hdtv_std_smpte_274m_1080p_24:1;
+    unsigned int hdtv_std_smpte_274m_1080p_25:1;
+    unsigned int hdtv_std_smpte_274m_1080p_29:1;
+    unsigned int hdtv_std_smpte_274m_1080p_30:1;
+    unsigned int hdtv_std_smpte_274m_1080p_50:1;
+
+    unsigned int hdtv_std_smpte_274m_1080p_59:1;
+    unsigned int hdtv_std_smpte_274m_1080p_60:1;
+    unsigned int hdtv_std_smpte_295m_1080i_50:1;
+    unsigned int hdtv_std_smpte_295m_1080p_50:1;
+    unsigned int hdtv_std_smpte_296m_720p_59:1;
+    unsigned int hdtv_std_smpte_296m_720p_60:1;
+    unsigned int hdtv_std_smpte_296m_720p_50:1;
+    unsigned int hdtv_std_smpte_293m_480p_59:1;
+
+    unsigned int hdtv_std_smpte_170m_480i_59:1;
+    unsigned int hdtv_std_iturbt601_576i_50:1;
+    unsigned int hdtv_std_iturbt601_576p_50:1;
+    unsigned int hdtv_std_eia_7702a_480i_60:1;
+    unsigned int hdtv_std_eia_7702a_480p_60:1;
+    unsigned int pad:3;
+} __attribute__((packed));
 
 #define SDVO_CMD_GET_TV_FORMAT				0x28
 
 #define SDVO_CMD_SET_TV_FORMAT				0x29
 
+/** Returns the resolutiosn that can be used with the given TV format */
+#define SDVO_CMD_GET_SDTV_RESOLUTION_SUPPORT		0x83
+struct psb_intel_sdvo_sdtv_resolution_request {
+    unsigned int ntsc_m:1;
+    unsigned int ntsc_j:1;
+    unsigned int ntsc_443:1;
+    unsigned int pal_b:1;
+    unsigned int pal_d:1;
+    unsigned int pal_g:1;
+    unsigned int pal_h:1;
+    unsigned int pal_i:1;
+
+    unsigned int pal_m:1;
+    unsigned int pal_n:1;
+    unsigned int pal_nc:1;
+    unsigned int pal_60:1;
+    unsigned int secam_b:1;
+    unsigned int secam_d:1;
+    unsigned int secam_g:1;
+    unsigned int secam_k:1;
+
+    unsigned int secam_k1:1;
+    unsigned int secam_l:1;
+    unsigned int secam_60:1;
+    unsigned int pad:5;
+} __attribute__((packed));
+
+struct psb_intel_sdvo_sdtv_resolution_reply {
+    unsigned int res_320x200:1;
+    unsigned int res_320x240:1;
+    unsigned int res_400x300:1;
+    unsigned int res_640x350:1;
+    unsigned int res_640x400:1;
+    unsigned int res_640x480:1;
+    unsigned int res_704x480:1;
+    unsigned int res_704x576:1;
+
+    unsigned int res_720x350:1;
+    unsigned int res_720x400:1;
+    unsigned int res_720x480:1;
+    unsigned int res_720x540:1;
+    unsigned int res_720x576:1;
+    unsigned int res_768x576:1;
+    unsigned int res_800x600:1;
+    unsigned int res_832x624:1;
+
+    unsigned int res_920x766:1;
+    unsigned int res_1024x768:1;
+    unsigned int res_1280x1024:1;
+    unsigned int pad:5;
+} __attribute__((packed));
+
+/* Get supported resolution with squire pixel aspect ratio that can be
+   scaled for the requested HDTV format */
+#define SDVO_CMD_GET_SCALED_HDTV_RESOLUTION_SUPPORT		0x85
+
+struct psb_intel_sdvo_hdtv_resolution_request {
+    unsigned int hdtv_std_smpte_240m_1080i_59:1;
+    unsigned int hdtv_std_smpte_240m_1080i_60:1;
+    unsigned int hdtv_std_smpte_260m_1080i_59:1;
+    unsigned int hdtv_std_smpte_260m_1080i_60:1;
+    unsigned int hdtv_std_smpte_274m_1080i_50:1;
+    unsigned int hdtv_std_smpte_274m_1080i_59:1;
+    unsigned int hdtv_std_smpte_274m_1080i_60:1;
+    unsigned int hdtv_std_smpte_274m_1080p_23:1;
+
+    unsigned int hdtv_std_smpte_274m_1080p_24:1;
+    unsigned int hdtv_std_smpte_274m_1080p_25:1;
+    unsigned int hdtv_std_smpte_274m_1080p_29:1;
+    unsigned int hdtv_std_smpte_274m_1080p_30:1;
+    unsigned int hdtv_std_smpte_274m_1080p_50:1;
+    unsigned int hdtv_std_smpte_274m_1080p_59:1;
+    unsigned int hdtv_std_smpte_274m_1080p_60:1;
+    unsigned int hdtv_std_smpte_295m_1080i_50:1;
+
+    unsigned int hdtv_std_smpte_295m_1080p_50:1;
+    unsigned int hdtv_std_smpte_296m_720p_59:1;
+    unsigned int hdtv_std_smpte_296m_720p_60:1;
+    unsigned int hdtv_std_smpte_296m_720p_50:1;
+    unsigned int hdtv_std_smpte_293m_480p_59:1;
+    unsigned int hdtv_std_smpte_170m_480i_59:1;
+    unsigned int hdtv_std_iturbt601_576i_50:1;
+    unsigned int hdtv_std_iturbt601_576p_50:1;
+
+    unsigned int hdtv_std_eia_7702a_480i_60:1;
+    unsigned int hdtv_std_eia_7702a_480p_60:1;
+    unsigned int pad:6;
+} __attribute__((packed));
+
+struct psb_intel_sdvo_hdtv_resolution_reply {
+    unsigned int res_640x480:1;
+    unsigned int res_800x600:1;
+    unsigned int res_1024x768:1;
+    unsigned int res_1280x960:1;
+    unsigned int res_1400x1050:1;
+    unsigned int res_1600x1200:1;
+    unsigned int res_1920x1440:1;
+    unsigned int res_2048x1536:1;
+
+    unsigned int res_2560x1920:1;
+    unsigned int res_3200x2400:1;
+    unsigned int res_3840x2880:1;
+    unsigned int pad1:5;
+
+    unsigned int res_848x480:1;
+    unsigned int res_1064x600:1;
+    unsigned int res_1280x720:1;
+    unsigned int res_1360x768:1;
+    unsigned int res_1704x960:1;
+    unsigned int res_1864x1050:1;
+    unsigned int res_1920x1080:1;
+    unsigned int res_2128x1200:1;
+
+    unsigned int res_2560x1400:1;
+    unsigned int res_2728x1536:1;
+    unsigned int res_3408x1920:1;
+    unsigned int res_4264x2400:1;
+    unsigned int res_5120x2880:1;
+    unsigned int pad2:3;
+
+    unsigned int res_768x480:1;
+    unsigned int res_960x600:1;
+    unsigned int res_1152x720:1;
+    unsigned int res_1124x768:1;
+    unsigned int res_1536x960:1;
+    unsigned int res_1680x1050:1;
+    unsigned int res_1728x1080:1;
+    unsigned int res_1920x1200:1;
+
+    unsigned int res_2304x1440:1;
+    unsigned int res_2456x1536:1;
+    unsigned int res_3072x1920:1;
+    unsigned int res_3840x2400:1;
+    unsigned int res_4608x2880:1;
+    unsigned int pad3:3;
+
+    unsigned int res_1280x1024:1;
+    unsigned int pad4:7;
+
+    unsigned int res_1280x768:1;
+    unsigned int pad5:7;
+} __attribute__((packed));
+
+/* Get supported power state returns info for encoder and monitor, rely on
+   last SetTargetInput and SetTargetOutput calls */
 #define SDVO_CMD_GET_SUPPORTED_POWER_STATES		0x2a
+/* Get power state returns info for encoder and monitor, rely on last
+   SetTargetInput and SetTargetOutput calls */
+#define SDVO_CMD_GET_POWER_STATE			0x2b
 #define SDVO_CMD_GET_ENCODER_POWER_STATE		0x2b
 #define SDVO_CMD_SET_ENCODER_POWER_STATE		0x2c
 # define SDVO_ENCODER_STATE_ON					(1 << 0)
 # define SDVO_ENCODER_STATE_STANDBY				(1 << 1)
 # define SDVO_ENCODER_STATE_SUSPEND				(1 << 2)
 # define SDVO_ENCODER_STATE_OFF					(1 << 3)
+# define SDVO_MONITOR_STATE_ON					(1 << 4)
+# define SDVO_MONITOR_STATE_STANDBY				(1 << 5)
+# define SDVO_MONITOR_STATE_SUSPEND				(1 << 6)
+# define SDVO_MONITOR_STATE_OFF					(1 << 7)
 
-#define SDVO_CMD_SET_TV_RESOLUTION_SUPPORT		0x93
+#define SDVO_CMD_GET_MAX_PANEL_POWER_SEQUENCING		0x2d
+#define SDVO_CMD_GET_PANEL_POWER_SEQUENCING		0x2e
+#define SDVO_CMD_SET_PANEL_POWER_SEQUENCING		0x2f
+/**
+ * The panel power sequencing parameters are in units of milliseconds.
+ * The high fields are bits 8:9 of the 10-bit values.
+ */
+struct psb_sdvo_panel_power_sequencing {
+    u8 t0;
+    u8 t1;
+    u8 t2;
+    u8 t3;
+    u8 t4;
+
+    unsigned int t0_high:2;
+    unsigned int t1_high:2;
+    unsigned int t2_high:2;
+    unsigned int t3_high:2;
+
+    unsigned int t4_high:2;
+    unsigned int pad:6;
+} __attribute__((packed));
+
+#define SDVO_CMD_GET_MAX_BACKLIGHT_LEVEL		0x30
+struct sdvo_max_backlight_reply {
+    u8 max_value;
+    u8 default_value;
+} __attribute__((packed));
+
+#define SDVO_CMD_GET_BACKLIGHT_LEVEL			0x31
+#define SDVO_CMD_SET_BACKLIGHT_LEVEL			0x32
+
+#define SDVO_CMD_GET_AMBIENT_LIGHT			0x33
+struct sdvo_get_ambient_light_reply {
+    u16 trip_low;
+    u16 trip_high;
+    u16 value;
+} __attribute__((packed));
+#define SDVO_CMD_SET_AMBIENT_LIGHT			0x34
+struct sdvo_set_ambient_light_reply {
+    u16 trip_low;
+    u16 trip_high;
+    unsigned int enable:1;
+    unsigned int pad:7;
+} __attribute__((packed));
+
+/* Set display power state */
+#define SDVO_CMD_SET_DISPLAY_POWER_STATE		0x7d
+# define SDVO_DISPLAY_STATE_ON				(1 << 0)
+# define SDVO_DISPLAY_STATE_STANDBY			(1 << 1)
+# define SDVO_DISPLAY_STATE_SUSPEND			(1 << 2)
+# define SDVO_DISPLAY_STATE_OFF				(1 << 3)
+
+#define SDVO_CMD_GET_SUPPORTED_ENHANCEMENTS		0x84
+struct psb_intel_sdvo_enhancements_reply {
+    unsigned int flicker_filter:1;
+    unsigned int flicker_filter_adaptive:1;
+    unsigned int flicker_filter_2d:1;
+    unsigned int saturation:1;
+    unsigned int hue:1;
+    unsigned int brightness:1;
+    unsigned int contrast:1;
+    unsigned int overscan_h:1;
+
+    unsigned int overscan_v:1;
+    unsigned int hpos:1;
+    unsigned int vpos:1;
+    unsigned int sharpness:1;
+    unsigned int dot_crawl:1;
+    unsigned int dither:1;
+    unsigned int tv_chroma_filter:1;
+    unsigned int tv_luma_filter:1;
+} __attribute__((packed));
+
+/* Picture enhancement limits below are dependent on the current TV format,
+ * and thus need to be queried and set after it.
+ */
+#define SDVO_CMD_GET_MAX_FLICKER_FILTER			0x4d
+#define SDVO_CMD_GET_MAX_FLICKER_FILTER_ADAPTIVE	0x7b
+#define SDVO_CMD_GET_MAX_FLICKER_FILTER_2D		0x52
+#define SDVO_CMD_GET_MAX_SATURATION			0x55
+#define SDVO_CMD_GET_MAX_HUE				0x58
+#define SDVO_CMD_GET_MAX_BRIGHTNESS			0x5b
+#define SDVO_CMD_GET_MAX_CONTRAST			0x5e
+#define SDVO_CMD_GET_MAX_OVERSCAN_H			0x61
+#define SDVO_CMD_GET_MAX_OVERSCAN_V			0x64
+#define SDVO_CMD_GET_MAX_HPOS				0x67
+#define SDVO_CMD_GET_MAX_VPOS				0x6a
+#define SDVO_CMD_GET_MAX_SHARPNESS			0x6d
+#define SDVO_CMD_GET_MAX_TV_CHROMA_FILTER		0x74
+#define SDVO_CMD_GET_MAX_TV_LUMA_FILTER			0x77
+struct psb_intel_sdvo_enhancement_limits_reply {
+    u16 max_value;
+    u16 default_value;
+} __attribute__((packed));
+
+#define SDVO_CMD_GET_LVDS_PANEL_INFORMATION		0x7f
+#define SDVO_CMD_SET_LVDS_PANEL_INFORMATION		0x80
+# define SDVO_LVDS_COLOR_DEPTH_18			(0 << 0)
+# define SDVO_LVDS_COLOR_DEPTH_24			(1 << 0)
+# define SDVO_LVDS_CONNECTOR_SPWG			(0 << 2)
+# define SDVO_LVDS_CONNECTOR_OPENLDI			(1 << 2)
+# define SDVO_LVDS_SINGLE_CHANNEL			(0 << 4)
+# define SDVO_LVDS_DUAL_CHANNEL				(1 << 4)
+
+#define SDVO_CMD_GET_FLICKER_FILTER			0x4e
+#define SDVO_CMD_SET_FLICKER_FILTER			0x4f
+#define SDVO_CMD_GET_FLICKER_FILTER_ADAPTIVE		0x50
+#define SDVO_CMD_SET_FLICKER_FILTER_ADAPTIVE		0x51
+#define SDVO_CMD_GET_FLICKER_FILTER_2D			0x53
+#define SDVO_CMD_SET_FLICKER_FILTER_2D			0x54
+#define SDVO_CMD_GET_SATURATION				0x56
+#define SDVO_CMD_SET_SATURATION				0x57
+#define SDVO_CMD_GET_HUE				0x59
+#define SDVO_CMD_SET_HUE				0x5a
+#define SDVO_CMD_GET_BRIGHTNESS				0x5c
+#define SDVO_CMD_SET_BRIGHTNESS				0x5d
+#define SDVO_CMD_GET_CONTRAST				0x5f
+#define SDVO_CMD_SET_CONTRAST				0x60
+#define SDVO_CMD_GET_OVERSCAN_H				0x62
+#define SDVO_CMD_SET_OVERSCAN_H				0x63
+#define SDVO_CMD_GET_OVERSCAN_V				0x65
+#define SDVO_CMD_SET_OVERSCAN_V				0x66
+#define SDVO_CMD_GET_HPOS				0x68
+#define SDVO_CMD_SET_HPOS				0x69
+#define SDVO_CMD_GET_VPOS				0x6b
+#define SDVO_CMD_SET_VPOS				0x6c
+#define SDVO_CMD_GET_SHARPNESS				0x6e
+#define SDVO_CMD_SET_SHARPNESS				0x6f
+#define SDVO_CMD_GET_TV_CHROMA_FILTER			0x75
+#define SDVO_CMD_SET_TV_CHROMA_FILTER			0x76
+#define SDVO_CMD_GET_TV_LUMA_FILTER			0x78
+#define SDVO_CMD_SET_TV_LUMA_FILTER			0x79
+struct psb_intel_sdvo_enhancements_arg {
+    u16 value;
+}__attribute__((packed));
+
+#define SDVO_CMD_GET_DOT_CRAWL				0x70
+#define SDVO_CMD_SET_DOT_CRAWL				0x71
+# define SDVO_DOT_CRAWL_ON					(1 << 0)
+# define SDVO_DOT_CRAWL_DEFAULT_ON				(1 << 1)
+
+#define SDVO_CMD_GET_DITHER				0x72
+#define SDVO_CMD_SET_DITHER				0x73
+# define SDVO_DITHER_ON						(1 << 0)
+# define SDVO_DITHER_DEFAULT_ON					(1 << 1)
 
 #define SDVO_CMD_SET_CONTROL_BUS_SWITCH			0x7a
-# define SDVO_CONTROL_BUS_PROM				0x0
-# define SDVO_CONTROL_BUS_DDC1				0x1
-# define SDVO_CONTROL_BUS_DDC2				0x2
-# define SDVO_CONTROL_BUS_DDC3				0x3
+# define SDVO_CONTROL_BUS_PROM				(1 << 0)
+# define SDVO_CONTROL_BUS_DDC1				(1 << 1)
+# define SDVO_CONTROL_BUS_DDC2				(1 << 2)
+# define SDVO_CONTROL_BUS_DDC3				(1 << 3)
 
-/* SDVO Bus & SDVO Inputs wiring details*/
-/* Bit 0: Is SDVOB connected to In0 (1 = yes, 0 = no*/
-/* Bit 1: Is SDVOB connected to In1 (1 = yes, 0 = no*/
-/* Bit 2: Is SDVOC connected to In0 (1 = yes, 0 = no*/
-/* Bit 3: Is SDVOC connected to In1 (1 = yes, 0 = no*/
-#define SDVOB_IN0 0x01
-#define SDVOB_IN1 0x02
-#define SDVOC_IN0 0x04
-#define SDVOC_IN1 0x08
+/* HDMI op codes */
+#define SDVO_CMD_GET_SUPP_ENCODE	0x9d
+#define SDVO_CMD_GET_ENCODE		0x9e
+#define SDVO_CMD_SET_ENCODE		0x9f
+  #define SDVO_ENCODE_DVI	0x0
+  #define SDVO_ENCODE_HDMI	0x1
+#define SDVO_CMD_SET_PIXEL_REPLI	0x8b
+#define SDVO_CMD_GET_PIXEL_REPLI	0x8c
+#define SDVO_CMD_GET_COLORIMETRY_CAP	0x8d
+#define SDVO_CMD_SET_COLORIMETRY	0x8e
+  #define SDVO_COLORIMETRY_RGB256   0x0
+  #define SDVO_COLORIMETRY_RGB220   0x1
+  #define SDVO_COLORIMETRY_YCrCb422 0x3
+  #define SDVO_COLORIMETRY_YCrCb444 0x4
+#define SDVO_CMD_GET_COLORIMETRY	0x8f
+#define SDVO_CMD_GET_AUDIO_ENCRYPT_PREFER 0x90
+#define SDVO_CMD_SET_AUDIO_STAT		0x91
+#define SDVO_CMD_GET_AUDIO_STAT		0x92
+#define SDVO_CMD_SET_HBUF_INDEX		0x93
+#define SDVO_CMD_GET_HBUF_INDEX		0x94
+#define SDVO_CMD_GET_HBUF_INFO		0x95
+#define SDVO_CMD_SET_HBUF_AV_SPLIT	0x96
+#define SDVO_CMD_GET_HBUF_AV_SPLIT	0x97
+#define SDVO_CMD_SET_HBUF_DATA		0x98
+#define SDVO_CMD_GET_HBUF_DATA		0x99
+#define SDVO_CMD_SET_HBUF_TXRATE	0x9a
+#define SDVO_CMD_GET_HBUF_TXRATE	0x9b
+  #define SDVO_HBUF_TX_DISABLED	(0 << 6)
+  #define SDVO_HBUF_TX_ONCE	(2 << 6)
+  #define SDVO_HBUF_TX_VSYNC	(3 << 6)
+#define SDVO_CMD_GET_AUDIO_TX_INFO	0x9c
+#define SDVO_NEED_TO_STALL  (1 << 7)
 
-#define SDVO_DEVICE_NONE 0x00
-#define        SDVO_DEVICE_CRT 0x01
-#define        SDVO_DEVICE_TV 0x02
-#define        SDVO_DEVICE_LVDS 0x04
-#define        SDVO_DEVICE_TMDS 0x08
-
+struct psb_intel_sdvo_encode {
+    u8 dvi_rev;
+    u8 hdmi_rev;
+} __attribute__ ((packed));
diff --git a/drivers/gpu/drm/i810/i810_dma.c b/drivers/gpu/drm/i810/i810_dma.c
index 8f371e8..f7c17b2 100644
--- a/drivers/gpu/drm/i810/i810_dma.c
+++ b/drivers/gpu/drm/i810/i810_dma.c
@@ -222,8 +222,6 @@
 			pci_free_consistent(dev->pdev, PAGE_SIZE,
 					    dev_priv->hw_status_page,
 					    dev_priv->dma_status_page);
-			/* Need to rewrite hardware status page */
-			I810_WRITE(0x02080, 0x1ffff000);
 		}
 		kfree(dev->dev_private);
 		dev->dev_private = NULL;
@@ -888,7 +886,7 @@
 }
 
 /* Must be called with the lock held */
-static void i810_reclaim_buffers(struct drm_device *dev,
+void i810_driver_reclaim_buffers(struct drm_device *dev,
 				 struct drm_file *file_priv)
 {
 	struct drm_device_dma *dma = dev->dma;
@@ -1225,12 +1223,17 @@
 		if (dev_priv->page_flipping)
 			i810_do_cleanup_pageflip(dev);
 	}
-}
 
-void i810_driver_reclaim_buffers_locked(struct drm_device *dev,
-					struct drm_file *file_priv)
-{
-	i810_reclaim_buffers(dev, file_priv);
+	if (file_priv->master && file_priv->master->lock.hw_lock) {
+		drm_idlelock_take(&file_priv->master->lock);
+		i810_driver_reclaim_buffers(dev, file_priv);
+		drm_idlelock_release(&file_priv->master->lock);
+	} else {
+		/* master disappeared, clean up stuff anyway and hope nothing
+		 * goes wrong */
+		i810_driver_reclaim_buffers(dev, file_priv);
+	}
+
 }
 
 int i810_driver_dma_quiescent(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i810/i810_drv.c b/drivers/gpu/drm/i810/i810_drv.c
index ec12f7d..053f1ee 100644
--- a/drivers/gpu/drm/i810/i810_drv.c
+++ b/drivers/gpu/drm/i810/i810_drv.c
@@ -63,7 +63,6 @@
 	.lastclose = i810_driver_lastclose,
 	.preclose = i810_driver_preclose,
 	.device_is_agp = i810_driver_device_is_agp,
-	.reclaim_buffers_locked = i810_driver_reclaim_buffers_locked,
 	.dma_quiescent = i810_driver_dma_quiescent,
 	.ioctls = i810_ioctls,
 	.fops = &i810_driver_fops,
diff --git a/drivers/gpu/drm/i810/i810_drv.h b/drivers/gpu/drm/i810/i810_drv.h
index c9339f4..6e0acad 100644
--- a/drivers/gpu/drm/i810/i810_drv.h
+++ b/drivers/gpu/drm/i810/i810_drv.h
@@ -116,14 +116,12 @@
 
 				/* i810_dma.c */
 extern int i810_driver_dma_quiescent(struct drm_device *dev);
-extern void i810_driver_reclaim_buffers_locked(struct drm_device *dev,
-					       struct drm_file *file_priv);
+void i810_driver_reclaim_buffers(struct drm_device *dev,
+			         struct drm_file *file_priv);
 extern int i810_driver_load(struct drm_device *, unsigned long flags);
 extern void i810_driver_lastclose(struct drm_device *dev);
 extern void i810_driver_preclose(struct drm_device *dev,
 				 struct drm_file *file_priv);
-extern void i810_driver_reclaim_buffers_locked(struct drm_device *dev,
-					       struct drm_file *file_priv);
 extern int i810_driver_device_is_agp(struct drm_device *dev);
 
 extern long i810_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index d14b44e..004b048 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -62,6 +62,7 @@
 	const struct intel_device_info *info = INTEL_INFO(dev);
 
 	seq_printf(m, "gen: %d\n", info->gen);
+	seq_printf(m, "pch: %d\n", INTEL_PCH_TYPE(dev));
 #define B(x) seq_printf(m, #x ": %s\n", yesno(info->x))
 	B(is_mobile);
 	B(is_i85x);
@@ -636,11 +637,16 @@
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring;
+	int ret;
 
 	ring = &dev_priv->ring[(uintptr_t)node->info_ent->data];
 	if (ring->size == 0)
 		return 0;
 
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
 	seq_printf(m, "Ring %s:\n", ring->name);
 	seq_printf(m, "  Head :    %08x\n", I915_READ_HEAD(ring) & HEAD_ADDR);
 	seq_printf(m, "  Tail :    %08x\n", I915_READ_TAIL(ring) & TAIL_ADDR);
@@ -654,6 +660,8 @@
 	seq_printf(m, "  Control : %08x\n", I915_READ_CTL(ring));
 	seq_printf(m, "  Start :   %08x\n", I915_READ_START(ring));
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -842,7 +850,16 @@
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	u16 crstanddelay = I915_READ16(CRSTANDVID);
+	u16 crstanddelay;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	crstanddelay = I915_READ16(CRSTANDVID);
+
+	mutex_unlock(&dev->struct_mutex);
 
 	seq_printf(m, "w/ctx: %d, w/o ctx: %d\n", (crstanddelay >> 8) & 0x3f, (crstanddelay & 0x3f));
 
@@ -940,7 +957,11 @@
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	u32 delayfreq;
-	int i;
+	int ret, i;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
 
 	for (i = 0; i < 16; i++) {
 		delayfreq = I915_READ(PXVFREQ_BASE + i * 4);
@@ -948,6 +969,8 @@
 			   (delayfreq & PXVFREQ_PX_MASK) >> PXVFREQ_PX_SHIFT);
 	}
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -962,13 +985,19 @@
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	u32 inttoext;
-	int i;
+	int ret, i;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
 
 	for (i = 1; i <= 32; i++) {
 		inttoext = I915_READ(INTTOEXT_BASE_ILK + i * 4);
 		seq_printf(m, "INTTOEXT%02d: 0x%08x\n", i, inttoext);
 	}
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -977,9 +1006,19 @@
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	u32 rgvmodectl = I915_READ(MEMMODECTL);
-	u32 rstdbyctl = I915_READ(RSTDBYCTL);
-	u16 crstandvid = I915_READ16(CRSTANDVID);
+	u32 rgvmodectl, rstdbyctl;
+	u16 crstandvid;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	rgvmodectl = I915_READ(MEMMODECTL);
+	rstdbyctl = I915_READ(RSTDBYCTL);
+	crstandvid = I915_READ16(CRSTANDVID);
+
+	mutex_unlock(&dev->struct_mutex);
 
 	seq_printf(m, "HD boost: %s\n", (rgvmodectl & MEMMODE_BOOST_EN) ?
 		   "yes" : "no");
@@ -1167,9 +1206,16 @@
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
 
 	seq_printf(m, "GFXEC: %ld\n", (unsigned long)I915_READ(0x112f4));
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -1506,7 +1552,10 @@
 	node->minor = minor;
 	node->dent = ent;
 	node->info_ent = (void *) key;
-	list_add(&node->list, &minor->debugfs_nodes.list);
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index a9533c5..a9ae374 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1454,6 +1454,14 @@
 
 	diff1 = now - dev_priv->last_time1;
 
+	/* Prevent division-by-zero if we are asking too fast.
+	 * Also, we don't get interesting results if we are polling
+	 * faster than once in 10ms, so just return the saved value
+	 * in such cases.
+	 */
+	if (diff1 <= 10)
+		return dev_priv->chipset_power;
+
 	count1 = I915_READ(DMIEC);
 	count2 = I915_READ(DDREC);
 	count3 = I915_READ(CSIEC);
@@ -1484,6 +1492,8 @@
 	dev_priv->last_count1 = total_count;
 	dev_priv->last_time1 = now;
 
+	dev_priv->chipset_power = ret;
+
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 9f59270..22c8ab7 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -58,17 +58,17 @@
 MODULE_PARM_DESC(powersave,
 		"Enable powersavings, fbc, downclocking, etc. (default: true)");
 
-unsigned int i915_semaphores __read_mostly = 0;
+int i915_semaphores __read_mostly = -1;
 module_param_named(semaphores, i915_semaphores, int, 0600);
 MODULE_PARM_DESC(semaphores,
-		"Use semaphores for inter-ring sync (default: false)");
+		"Use semaphores for inter-ring sync (default: -1 (use per-chip defaults))");
 
-unsigned int i915_enable_rc6 __read_mostly = 0;
+int i915_enable_rc6 __read_mostly = -1;
 module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0600);
 MODULE_PARM_DESC(i915_enable_rc6,
-		"Enable power-saving render C-state 6 (default: true)");
+		"Enable power-saving render C-state 6 (default: -1 (use per-chip default)");
 
-unsigned int i915_enable_fbc __read_mostly = -1;
+int i915_enable_fbc __read_mostly = -1;
 module_param_named(i915_enable_fbc, i915_enable_fbc, int, 0600);
 MODULE_PARM_DESC(i915_enable_fbc,
 		"Enable frame buffer compression for power savings "
@@ -80,7 +80,7 @@
 		"Use panel (LVDS/eDP) downclocking for power savings "
 		"(default: false)");
 
-unsigned int i915_panel_use_ssc __read_mostly = -1;
+int i915_panel_use_ssc __read_mostly = -1;
 module_param_named(lvds_use_ssc, i915_panel_use_ssc, int, 0600);
 MODULE_PARM_DESC(lvds_use_ssc,
 		"Use Spread Spectrum Clock with panels [LVDS/eDP] "
@@ -107,7 +107,7 @@
 extern int intel_agp_enabled;
 
 #define INTEL_VGA_DEVICE(id, info) {		\
-	.class = PCI_CLASS_DISPLAY_VGA << 8,	\
+	.class = PCI_BASE_CLASS_DISPLAY << 16,	\
 	.class_mask = 0xff0000,			\
 	.vendor = 0x8086,			\
 	.device = id,				\
@@ -328,7 +328,7 @@
 	}
 }
 
-static void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
+void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
 {
 	int count;
 
@@ -344,6 +344,22 @@
 		udelay(10);
 }
 
+void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv)
+{
+	int count;
+
+	count = 0;
+	while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1))
+		udelay(10);
+
+	I915_WRITE_NOTRACE(FORCEWAKE_MT, (1<<16) | 1);
+	POSTING_READ(FORCEWAKE_MT);
+
+	count = 0;
+	while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1) == 0)
+		udelay(10);
+}
+
 /*
  * Generally this is called implicitly by the register read function. However,
  * if some sequence requires the GT to not power down then this function should
@@ -356,15 +372,21 @@
 
 	/* Forcewake is atomic in case we get in here without the lock */
 	if (atomic_add_return(1, &dev_priv->forcewake_count) == 1)
-		__gen6_gt_force_wake_get(dev_priv);
+		dev_priv->display.force_wake_get(dev_priv);
 }
 
-static void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
+void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
 {
 	I915_WRITE_NOTRACE(FORCEWAKE, 0);
 	POSTING_READ(FORCEWAKE);
 }
 
+void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv)
+{
+	I915_WRITE_NOTRACE(FORCEWAKE_MT, (1<<16) | 0);
+	POSTING_READ(FORCEWAKE_MT);
+}
+
 /*
  * see gen6_gt_force_wake_get()
  */
@@ -373,7 +395,7 @@
 	WARN_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
 
 	if (atomic_dec_and_test(&dev_priv->forcewake_count))
-		__gen6_gt_force_wake_put(dev_priv);
+		dev_priv->display.force_wake_put(dev_priv);
 }
 
 void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
@@ -804,8 +826,8 @@
 };
 
 static struct drm_driver driver = {
-	/* don't use mtrr's here, the Xserver or user space app should
-	 * deal with them for intel hardware.
+	/* Don't use MTRRs here; the Xserver or userspace app should
+	 * deal with them for Intel hardware.
 	 */
 	.driver_features =
 	    DRIVER_USE_AGP | DRIVER_REQUIRE_AGP | /* DRIVER_USE_MTRR |*/
@@ -904,8 +926,9 @@
 /* We give fast paths for the really cool registers */
 #define NEEDS_FORCE_WAKE(dev_priv, reg) \
 	(((dev_priv)->info->gen >= 6) && \
-	((reg) < 0x40000) && \
-	((reg) != FORCEWAKE))
+	 ((reg) < 0x40000) &&		 \
+	 ((reg) != FORCEWAKE) &&	 \
+	 ((reg) != ECOBUS))
 
 #define __i915_read(x, y) \
 u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg) { \
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 06a37f4..505af3f 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -107,6 +107,7 @@
 struct opregion_acpi;
 struct opregion_swsci;
 struct opregion_asle;
+struct drm_i915_private;
 
 struct intel_opregion {
 	struct opregion_header *header;
@@ -126,6 +127,9 @@
 	struct _drm_i915_sarea *sarea_priv;
 };
 #define I915_FENCE_REG_NONE -1
+#define I915_MAX_NUM_FENCES 16
+/* 16 fences + sign bit for FENCE_REG_NONE */
+#define I915_MAX_NUM_FENCE_BITS 5
 
 struct drm_i915_fence_reg {
 	struct list_head lru_list;
@@ -168,7 +172,7 @@
 	u32 instdone1;
 	u32 seqno;
 	u64 bbaddr;
-	u64 fence[16];
+	u64 fence[I915_MAX_NUM_FENCES];
 	struct timeval time;
 	struct drm_i915_error_object {
 		int page_count;
@@ -182,7 +186,7 @@
 		u32 gtt_offset;
 		u32 read_domains;
 		u32 write_domain;
-		s32 fence_reg:5;
+		s32 fence_reg:I915_MAX_NUM_FENCE_BITS;
 		s32 pinned:2;
 		u32 tiling:2;
 		u32 dirty:1;
@@ -218,6 +222,8 @@
 			  struct drm_i915_gem_object *obj);
 	int (*update_plane)(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			    int x, int y);
+	void (*force_wake_get)(struct drm_i915_private *dev_priv);
+	void (*force_wake_put)(struct drm_i915_private *dev_priv);
 	/* clock updates for mode set */
 	/* cursor updates */
 	/* render clock increase/decrease */
@@ -331,6 +337,8 @@
 	struct timer_list hangcheck_timer;
 	int hangcheck_count;
 	uint32_t last_acthd;
+	uint32_t last_acthd_bsd;
+	uint32_t last_acthd_blt;
 	uint32_t last_instdone;
 	uint32_t last_instdone1;
 
@@ -375,7 +383,7 @@
 	struct notifier_block lid_notifier;
 
 	int crt_ddc_pin;
-	struct drm_i915_fence_reg fence_regs[16]; /* assume 965 */
+	struct drm_i915_fence_reg fence_regs[I915_MAX_NUM_FENCES]; /* assume 965 */
 	int fence_reg_start; /* 4 if userland hasn't ioctl'd us yet */
 	int num_fence_regs; /* 8 on pre-965, 16 otherwise */
 
@@ -506,7 +514,7 @@
 	u8 saveAR[21];
 	u8 saveDACMASK;
 	u8 saveCR[37];
-	uint64_t saveFENCE[16];
+	uint64_t saveFENCE[I915_MAX_NUM_FENCES];
 	u32 saveCURACNTR;
 	u32 saveCURAPOS;
 	u32 saveCURABASE;
@@ -707,6 +715,7 @@
 
 	u64 last_count1;
 	unsigned long last_time1;
+	unsigned long chipset_power;
 	u64 last_count2;
 	struct timespec last_time2;
 	unsigned long gfx_power;
@@ -777,10 +786,8 @@
 	 * Fence register bits (if any) for this object.  Will be set
 	 * as needed when mapped into the GTT.
 	 * Protected by dev->struct_mutex.
-	 *
-	 * Size: 4 bits for 16 fences + sign (for FENCE_REG_NONE)
 	 */
-	signed int fence_reg:5;
+	signed int fence_reg:I915_MAX_NUM_FENCE_BITS;
 
 	/**
 	 * Advice: are the backing pages purgeable?
@@ -997,12 +1004,12 @@
 extern unsigned int i915_fbpercrtc __always_unused;
 extern int i915_panel_ignore_lid __read_mostly;
 extern unsigned int i915_powersave __read_mostly;
-extern unsigned int i915_semaphores __read_mostly;
+extern int i915_semaphores __read_mostly;
 extern unsigned int i915_lvds_downclock __read_mostly;
-extern unsigned int i915_panel_use_ssc __read_mostly;
+extern int i915_panel_use_ssc __read_mostly;
 extern int i915_vbt_sdvo_panel_type __read_mostly;
-extern unsigned int i915_enable_rc6 __read_mostly;
-extern unsigned int i915_enable_fbc __read_mostly;
+extern int i915_enable_rc6 __read_mostly;
+extern int i915_enable_fbc __read_mostly;
 extern bool i915_enable_hangcheck __read_mostly;
 
 extern int i915_suspend(struct drm_device *dev, pm_message_t state);
@@ -1307,6 +1314,11 @@
 extern void intel_detect_pch(struct drm_device *dev);
 extern int intel_trans_dp_port_sel(struct drm_crtc *crtc);
 
+extern void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv);
+extern void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv);
+extern void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv);
+extern void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv);
+
 /* overlay */
 #ifdef CONFIG_DEBUG_FS
 extern struct intel_overlay_error_state *intel_overlay_capture_error_state(struct drm_device *dev);
@@ -1351,8 +1363,9 @@
 /* We give fast paths for the really cool registers */
 #define NEEDS_FORCE_WAKE(dev_priv, reg) \
 	(((dev_priv)->info->gen >= 6) && \
-	((reg) < 0x40000) && \
-	((reg) != FORCEWAKE))
+	 ((reg) < 0x40000) &&		 \
+	 ((reg) != FORCEWAKE) &&	 \
+	 ((reg) != ECOBUS))
 
 #define __i915_read(x, y) \
 	u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 6651c36..8359dc7 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1396,7 +1396,7 @@
 
 	if (obj->base.size > dev_priv->mm.gtt_mappable_end) {
 		ret = -E2BIG;
-		goto unlock;
+		goto out;
 	}
 
 	if (obj->madv != I915_MADV_WILLNEED) {
@@ -1745,7 +1745,7 @@
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i;
 
-	for (i = 0; i < 16; i++) {
+	for (i = 0; i < dev_priv->num_fence_regs; i++) {
 		struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i];
 		struct drm_i915_gem_object *obj = reg->obj;
 
@@ -3512,9 +3512,11 @@
 			 * so emit a request to do so.
 			 */
 			request = kzalloc(sizeof(*request), GFP_KERNEL);
-			if (request)
+			if (request) {
 				ret = i915_add_request(obj->ring, NULL, request);
-			else
+				if (ret)
+					kfree(request);
+			} else
 				ret = -ENOMEM;
 		}
 
@@ -3613,7 +3615,7 @@
 	obj->base.write_domain = I915_GEM_DOMAIN_CPU;
 	obj->base.read_domains = I915_GEM_DOMAIN_CPU;
 
-	if (IS_GEN6(dev)) {
+	if (IS_GEN6(dev) || IS_GEN7(dev)) {
 		/* On Gen6, we can have the GPU use the LLC (the CPU
 		 * cache) for about a 10% performance improvement
 		 * compared to uncached.  Graphics requests other than
@@ -3877,7 +3879,7 @@
 	INIT_LIST_HEAD(&dev_priv->mm.gtt_list);
 	for (i = 0; i < I915_NUM_RINGS; i++)
 		init_ring_lists(&dev_priv->ring[i]);
-	for (i = 0; i < 16; i++)
+	for (i = 0; i < I915_MAX_NUM_FENCES; i++)
 		INIT_LIST_HEAD(&dev_priv->fence_regs[i].lru_list);
 	INIT_DELAYED_WORK(&dev_priv->mm.retire_work,
 			  i915_gem_retire_work_handler);
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 3693e83..c681dc1 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -32,6 +32,7 @@
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
+#include <linux/dma_remapping.h>
 
 struct change_domains {
 	uint32_t invalidate_domains;
@@ -746,6 +747,22 @@
 	return 0;
 }
 
+static bool
+intel_enable_semaphores(struct drm_device *dev)
+{
+	if (INTEL_INFO(dev)->gen < 6)
+		return 0;
+
+	if (i915_semaphores >= 0)
+		return i915_semaphores;
+
+	/* Enable semaphores on SNB when IO remapping is off */
+	if (INTEL_INFO(dev)->gen == 6)
+		return !intel_iommu_enabled;
+
+	return 1;
+}
+
 static int
 i915_gem_execbuffer_sync_rings(struct drm_i915_gem_object *obj,
 			       struct intel_ring_buffer *to)
@@ -758,7 +775,7 @@
 		return 0;
 
 	/* XXX gpu semaphores are implicated in various hard hangs on SNB */
-	if (INTEL_INFO(obj->base.dev)->gen < 6 || !i915_semaphores)
+	if (!intel_enable_semaphores(obj->base.dev))
 		return i915_gem_object_wait_rendering(obj);
 
 	idx = intel_ring_sync_index(from, to);
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 9ee2729..5849311 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -824,6 +824,7 @@
 
 	/* Fences */
 	switch (INTEL_INFO(dev)->gen) {
+	case 7:
 	case 6:
 		for (i = 0; i < 16; i++)
 			error->fence[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 + (i * 8));
@@ -1204,7 +1205,7 @@
 	} else {
 		int dspaddr = DSPADDR(intel_crtc->plane);
 		stall_detected = I915_READ(dspaddr) == (obj->gtt_offset +
-							crtc->y * crtc->fb->pitch +
+							crtc->y * crtc->fb->pitches[0] +
 							crtc->x * crtc->fb->bits_per_pixel/8);
 	}
 
@@ -1668,7 +1669,7 @@
 {
 	struct drm_device *dev = (struct drm_device *)data;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	uint32_t acthd, instdone, instdone1;
+	uint32_t acthd, instdone, instdone1, acthd_bsd, acthd_blt;
 	bool err = false;
 
 	if (!i915_enable_hangcheck)
@@ -1685,16 +1686,21 @@
 	}
 
 	if (INTEL_INFO(dev)->gen < 4) {
-		acthd = I915_READ(ACTHD);
 		instdone = I915_READ(INSTDONE);
 		instdone1 = 0;
 	} else {
-		acthd = I915_READ(ACTHD_I965);
 		instdone = I915_READ(INSTDONE_I965);
 		instdone1 = I915_READ(INSTDONE1);
 	}
+	acthd = intel_ring_get_active_head(&dev_priv->ring[RCS]);
+	acthd_bsd = HAS_BSD(dev) ?
+		intel_ring_get_active_head(&dev_priv->ring[VCS]) : 0;
+	acthd_blt = HAS_BLT(dev) ?
+		intel_ring_get_active_head(&dev_priv->ring[BCS]) : 0;
 
 	if (dev_priv->last_acthd == acthd &&
+	    dev_priv->last_acthd_bsd == acthd_bsd &&
+	    dev_priv->last_acthd_blt == acthd_blt &&
 	    dev_priv->last_instdone == instdone &&
 	    dev_priv->last_instdone1 == instdone1) {
 		if (dev_priv->hangcheck_count++ > 1) {
@@ -1726,6 +1732,8 @@
 		dev_priv->hangcheck_count = 0;
 
 		dev_priv->last_acthd = acthd;
+		dev_priv->last_acthd_bsd = acthd_bsd;
+		dev_priv->last_acthd_blt = acthd_blt;
 		dev_priv->last_instdone = instdone;
 		dev_priv->last_instdone1 = instdone1;
 	}
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 5a09416..194d987 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1553,12 +1553,21 @@
  */
 #define   PP_READY		(1 << 30)
 #define   PP_SEQUENCE_NONE	(0 << 28)
-#define   PP_SEQUENCE_ON	(1 << 28)
-#define   PP_SEQUENCE_OFF	(2 << 28)
-#define   PP_SEQUENCE_MASK	0x30000000
+#define   PP_SEQUENCE_POWER_UP	(1 << 28)
+#define   PP_SEQUENCE_POWER_DOWN (2 << 28)
+#define   PP_SEQUENCE_MASK	(3 << 28)
+#define   PP_SEQUENCE_SHIFT	28
 #define   PP_CYCLE_DELAY_ACTIVE	(1 << 27)
-#define   PP_SEQUENCE_STATE_ON_IDLE (1 << 3)
 #define   PP_SEQUENCE_STATE_MASK 0x0000000f
+#define   PP_SEQUENCE_STATE_OFF_IDLE	(0x0 << 0)
+#define   PP_SEQUENCE_STATE_OFF_S0_1	(0x1 << 0)
+#define   PP_SEQUENCE_STATE_OFF_S0_2	(0x2 << 0)
+#define   PP_SEQUENCE_STATE_OFF_S0_3	(0x3 << 0)
+#define   PP_SEQUENCE_STATE_ON_IDLE	(0x8 << 0)
+#define   PP_SEQUENCE_STATE_ON_S1_0	(0x9 << 0)
+#define   PP_SEQUENCE_STATE_ON_S1_2	(0xa << 0)
+#define   PP_SEQUENCE_STATE_ON_S1_3	(0xb << 0)
+#define   PP_SEQUENCE_STATE_RESET	(0xf << 0)
 #define PP_CONTROL	0x61204
 #define   POWER_TARGET_ON	(1 << 0)
 #define PP_ON_DELAYS	0x61208
@@ -2312,6 +2321,7 @@
 #define   PIPECONF_PROGRESSIVE	(0 << 21)
 #define   PIPECONF_INTERLACE_W_FIELD_INDICATION	(6 << 21)
 #define   PIPECONF_INTERLACE_FIELD_0_ONLY		(7 << 21)
+#define   PIPECONF_INTERLACE_MASK	(7 << 21)
 #define   PIPECONF_CXSR_DOWNCLOCK	(1<<16)
 #define   PIPECONF_BPP_MASK	(0x000000e0)
 #define   PIPECONF_BPP_8	(0<<5)
@@ -3294,10 +3304,10 @@
 /* or SDVOB */
 #define HDMIB   0xe1140
 #define  PORT_ENABLE    (1 << 31)
-#define  TRANSCODER_A   (0)
-#define  TRANSCODER_B   (1 << 30)
-#define  TRANSCODER(pipe)	((pipe) << 30)
-#define  TRANSCODER_MASK   (1 << 30)
+#define  TRANSCODER(pipe)       ((pipe) << 30)
+#define  TRANSCODER_CPT(pipe)   ((pipe) << 29)
+#define  TRANSCODER_MASK        (1 << 30)
+#define  TRANSCODER_MASK_CPT    (3 << 29)
 #define  COLOR_FORMAT_8bpc      (0)
 #define  COLOR_FORMAT_12bpc     (3 << 26)
 #define  SDVOB_HOTPLUG_ENABLE   (1 << 23)
@@ -3438,12 +3448,38 @@
 #define  EDP_LINK_TRAIN_800_1200MV_0DB_SNB_B	(0x38<<22)
 #define  EDP_LINK_TRAIN_VOL_EMP_MASK_SNB	(0x3f<<22)
 
+/* IVB */
+#define EDP_LINK_TRAIN_400MV_0DB_IVB		(0x24 <<22)
+#define EDP_LINK_TRAIN_400MV_3_5DB_IVB		(0x2a <<22)
+#define EDP_LINK_TRAIN_400MV_6DB_IVB		(0x2f <<22)
+#define EDP_LINK_TRAIN_600MV_0DB_IVB		(0x30 <<22)
+#define EDP_LINK_TRAIN_600MV_3_5DB_IVB		(0x36 <<22)
+#define EDP_LINK_TRAIN_800MV_0DB_IVB		(0x38 <<22)
+#define EDP_LINK_TRAIN_800MV_3_5DB_IVB		(0x33 <<22)
+
+/* legacy values */
+#define EDP_LINK_TRAIN_500MV_0DB_IVB		(0x00 <<22)
+#define EDP_LINK_TRAIN_1000MV_0DB_IVB		(0x20 <<22)
+#define EDP_LINK_TRAIN_500MV_3_5DB_IVB		(0x02 <<22)
+#define EDP_LINK_TRAIN_1000MV_3_5DB_IVB		(0x22 <<22)
+#define EDP_LINK_TRAIN_1000MV_6DB_IVB		(0x23 <<22)
+
+#define  EDP_LINK_TRAIN_VOL_EMP_MASK_IVB	(0x3f<<22)
+
 #define  FORCEWAKE				0xA18C
 #define  FORCEWAKE_ACK				0x130090
+#define  FORCEWAKE_MT				0xa188 /* multi-threaded */
+#define  FORCEWAKE_MT_ACK			0x130040
+#define  ECOBUS					0xa180
+#define    FORCEWAKE_MT_ENABLE			(1<<5)
 
 #define  GT_FIFO_FREE_ENTRIES			0x120008
 #define    GT_FIFO_NUM_RESERVED_ENTRIES		20
 
+#define GEN6_UCGCTL2				0x9404
+# define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE		(1 << 12)
+# define GEN6_RCCUNIT_CLOCK_GATE_DISABLE		(1 << 11)
+
 #define GEN6_RPNSWREQ				0xA008
 #define   GEN6_TURBO_DISABLE			(1<<31)
 #define   GEN6_FREQUENCY(x)			((x)<<25)
@@ -3534,17 +3570,17 @@
 #define G4X_ELD_ACK			(1 << 4)
 #define G4X_HDMIW_HDMIEDID		0x6210C
 
-#define GEN5_HDMIW_HDMIEDID_A		0xE2050
-#define GEN5_AUD_CNTL_ST_A		0xE20B4
-#define GEN5_ELD_BUFFER_SIZE		(0x1f << 10)
-#define GEN5_ELD_ADDRESS		(0x1f << 5)
-#define GEN5_ELD_ACK			(1 << 4)
-#define GEN5_AUD_CNTL_ST2		0xE20C0
-#define GEN5_ELD_VALIDB			(1 << 0)
-#define GEN5_CP_READYB			(1 << 1)
+#define IBX_HDMIW_HDMIEDID_A		0xE2050
+#define IBX_AUD_CNTL_ST_A		0xE20B4
+#define IBX_ELD_BUFFER_SIZE		(0x1f << 10)
+#define IBX_ELD_ADDRESS			(0x1f << 5)
+#define IBX_ELD_ACK			(1 << 4)
+#define IBX_AUD_CNTL_ST2		0xE20C0
+#define IBX_ELD_VALIDB			(1 << 0)
+#define IBX_CP_READYB			(1 << 1)
 
-#define GEN7_HDMIW_HDMIEDID_A		0xE5050
-#define GEN7_AUD_CNTRL_ST_A		0xE50B4
-#define GEN7_AUD_CNTRL_ST2		0xE50C0
+#define CPT_HDMIW_HDMIEDID_A		0xE5050
+#define CPT_AUD_CNTL_ST_A		0xE50B4
+#define CPT_AUD_CNTRL_ST2		0xE50C0
 
 #endif /* _I915_REG_H_ */
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index f8f602d..7886e4f 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -370,6 +370,7 @@
 
 	/* Fences */
 	switch (INTEL_INFO(dev)->gen) {
+	case 7:
 	case 6:
 		for (i = 0; i < 16; i++)
 			dev_priv->saveFENCE[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 + (i * 8));
@@ -404,6 +405,7 @@
 
 	/* Fences */
 	switch (INTEL_INFO(dev)->gen) {
+	case 7:
 	case 6:
 		for (i = 0; i < 16; i++)
 			I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + (i * 8), dev_priv->saveFENCE[i]);
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 0bc93c3..55a5b4c 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -38,8 +38,8 @@
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "drm_dp_helper.h"
-
 #include "drm_crtc_helper.h"
+#include <linux/dma_remapping.h>
 
 #define HAS_eDP (intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))
 
@@ -1511,8 +1511,8 @@
 	u32 fbc_ctl, fbc_ctl2;
 
 	cfb_pitch = dev_priv->cfb_size / FBC_LL_SIZE;
-	if (fb->pitch < cfb_pitch)
-		cfb_pitch = fb->pitch;
+	if (fb->pitches[0] < cfb_pitch)
+		cfb_pitch = fb->pitches[0];
 
 	/* FBC_CTL wants 64B units */
 	cfb_pitch = (cfb_pitch / 64) - 1;
@@ -2073,11 +2073,11 @@
 	I915_WRITE(reg, dspcntr);
 
 	Start = obj->gtt_offset;
-	Offset = y * fb->pitch + x * (fb->bits_per_pixel / 8);
+	Offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8);
 
 	DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n",
-		      Start, Offset, x, y, fb->pitch);
-	I915_WRITE(DSPSTRIDE(plane), fb->pitch);
+		      Start, Offset, x, y, fb->pitches[0]);
+	I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
 	if (INTEL_INFO(dev)->gen >= 4) {
 		I915_WRITE(DSPSURF(plane), Start);
 		I915_WRITE(DSPTILEOFF(plane), (y << 16) | x);
@@ -2154,11 +2154,11 @@
 	I915_WRITE(reg, dspcntr);
 
 	Start = obj->gtt_offset;
-	Offset = y * fb->pitch + x * (fb->bits_per_pixel / 8);
+	Offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8);
 
 	DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n",
-		      Start, Offset, x, y, fb->pitch);
-	I915_WRITE(DSPSTRIDE(plane), fb->pitch);
+		      Start, Offset, x, y, fb->pitches[0]);
+	I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
 	I915_WRITE(DSPSURF(plane), Start);
 	I915_WRITE(DSPTILEOFF(plane), (y << 16) | x);
 	I915_WRITE(DSPADDR(plane), Offset);
@@ -2933,7 +2933,8 @@
 
 	/* For PCH DP, enable TRANS_DP_CTL */
 	if (HAS_PCH_CPT(dev) &&
-	    intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) {
+	    (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT) ||
+	     intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))) {
 		u32 bpc = (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) >> 5;
 		reg = TRANS_DP_CTL(pipe);
 		temp = I915_READ(reg);
@@ -4669,6 +4670,7 @@
 /**
  * intel_choose_pipe_bpp_dither - figure out what color depth the pipe should send
  * @crtc: CRTC structure
+ * @mode: requested mode
  *
  * A pipe may be connected to one or more outputs.  Based on the depth of the
  * attached framebuffer, choose a good color depth to use on the pipe.
@@ -4680,13 +4682,15 @@
  *    HDMI supports only 8bpc or 12bpc, so clamp to 8bpc with dither for 10bpc
  *    Displays may support a restricted set as well, check EDID and clamp as
  *      appropriate.
+ *    DP may want to dither down to 6bpc to fit larger modes
  *
  * RETURNS:
  * Dithering requirement (i.e. false if display bpc and pipe bpc match,
  * true if they don't match).
  */
 static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
-					 unsigned int *pipe_bpp)
+					 unsigned int *pipe_bpp,
+					 struct drm_display_mode *mode)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4711,7 +4715,7 @@
 				lvds_bpc = 6;
 
 			if (lvds_bpc < display_bpc) {
-				DRM_DEBUG_DRIVER("clamping display bpc (was %d) to LVDS (%d)\n", display_bpc, lvds_bpc);
+				DRM_DEBUG_KMS("clamping display bpc (was %d) to LVDS (%d)\n", display_bpc, lvds_bpc);
 				display_bpc = lvds_bpc;
 			}
 			continue;
@@ -4722,7 +4726,7 @@
 			unsigned int edp_bpc = dev_priv->edp.bpp / 3;
 
 			if (edp_bpc < display_bpc) {
-				DRM_DEBUG_DRIVER("clamping display bpc (was %d) to eDP (%d)\n", display_bpc, edp_bpc);
+				DRM_DEBUG_KMS("clamping display bpc (was %d) to eDP (%d)\n", display_bpc, edp_bpc);
 				display_bpc = edp_bpc;
 			}
 			continue;
@@ -4737,7 +4741,7 @@
 			/* Don't use an invalid EDID bpc value */
 			if (connector->display_info.bpc &&
 			    connector->display_info.bpc < display_bpc) {
-				DRM_DEBUG_DRIVER("clamping display bpc (was %d) to EDID reported max of %d\n", display_bpc, connector->display_info.bpc);
+				DRM_DEBUG_KMS("clamping display bpc (was %d) to EDID reported max of %d\n", display_bpc, connector->display_info.bpc);
 				display_bpc = connector->display_info.bpc;
 			}
 		}
@@ -4748,15 +4752,20 @@
 		 */
 		if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
 			if (display_bpc > 8 && display_bpc < 12) {
-				DRM_DEBUG_DRIVER("forcing bpc to 12 for HDMI\n");
+				DRM_DEBUG_KMS("forcing bpc to 12 for HDMI\n");
 				display_bpc = 12;
 			} else {
-				DRM_DEBUG_DRIVER("forcing bpc to 8 for HDMI\n");
+				DRM_DEBUG_KMS("forcing bpc to 8 for HDMI\n");
 				display_bpc = 8;
 			}
 		}
 	}
 
+	if (mode->private_flags & INTEL_MODE_DP_FORCE_6BPC) {
+		DRM_DEBUG_KMS("Dithering DP to 6bpc\n");
+		display_bpc = 6;
+	}
+
 	/*
 	 * We could just drive the pipe at the highest bpc all the time and
 	 * enable dithering as needed, but that costs bandwidth.  So choose
@@ -4789,8 +4798,8 @@
 
 	display_bpc = min(display_bpc, bpc);
 
-	DRM_DEBUG_DRIVER("setting pipe bpc to %d (max display bpc %d)\n",
-			 bpc, display_bpc);
+	DRM_DEBUG_KMS("setting pipe bpc to %d (max display bpc %d)\n",
+		      bpc, display_bpc);
 
 	*pipe_bpp = display_bpc * 3;
 
@@ -5018,6 +5027,16 @@
 			pipeconf &= ~PIPECONF_DOUBLE_WIDE;
 	}
 
+	/* default to 8bpc */
+	pipeconf &= ~(PIPECONF_BPP_MASK | PIPECONF_DITHER_EN);
+	if (is_dp) {
+		if (mode->private_flags & INTEL_MODE_DP_FORCE_6BPC) {
+			pipeconf |= PIPECONF_BPP_6 |
+				    PIPECONF_DITHER_EN |
+				    PIPECONF_DITHER_TYPE_SP;
+		}
+	}
+
 	dpll |= DPLL_VCO_ENABLE;
 
 	DRM_DEBUG_KMS("Mode for pipe %c:\n", pipe == 0 ? 'A' : 'B');
@@ -5136,7 +5155,7 @@
 		adjusted_mode->crtc_vsync_end -= 1;
 		adjusted_mode->crtc_vsync_start -= 1;
 	} else
-		pipeconf &= ~PIPECONF_INTERLACE_W_FIELD_INDICATION; /* progressive */
+		pipeconf &= ~PIPECONF_INTERLACE_MASK; /* progressive */
 
 	I915_WRITE(HTOTAL(pipe),
 		   (adjusted_mode->crtc_hdisplay - 1) |
@@ -5479,7 +5498,7 @@
 	/* determine panel color depth */
 	temp = I915_READ(PIPECONF(pipe));
 	temp &= ~PIPE_BPC_MASK;
-	dither = intel_choose_pipe_bpp_dither(crtc, &pipe_bpp);
+	dither = intel_choose_pipe_bpp_dither(crtc, &pipe_bpp, mode);
 	switch (pipe_bpp) {
 	case 18:
 		temp |= PIPE_6BPC;
@@ -5671,7 +5690,7 @@
 	pipeconf &= ~PIPECONF_DITHER_TYPE_MASK;
 	if ((is_lvds && dev_priv->lvds_dither) || dither) {
 		pipeconf |= PIPECONF_DITHER_EN;
-		pipeconf |= PIPECONF_DITHER_TYPE_ST1;
+		pipeconf |= PIPECONF_DITHER_TYPE_SP;
 	}
 	if (is_dp || intel_encoder_is_pch_edp(&has_edp_encoder->base)) {
 		intel_dp_set_m_n(crtc, mode, adjusted_mode);
@@ -5811,6 +5830,35 @@
 	return ret;
 }
 
+static bool intel_eld_uptodate(struct drm_connector *connector,
+			       int reg_eldv, uint32_t bits_eldv,
+			       int reg_elda, uint32_t bits_elda,
+			       int reg_edid)
+{
+	struct drm_i915_private *dev_priv = connector->dev->dev_private;
+	uint8_t *eld = connector->eld;
+	uint32_t i;
+
+	i = I915_READ(reg_eldv);
+	i &= bits_eldv;
+
+	if (!eld[0])
+		return !i;
+
+	if (!i)
+		return false;
+
+	i = I915_READ(reg_elda);
+	i &= ~bits_elda;
+	I915_WRITE(reg_elda, i);
+
+	for (i = 0; i < eld[2]; i++)
+		if (I915_READ(reg_edid) != *((uint32_t *)eld + i))
+			return false;
+
+	return true;
+}
+
 static void g4x_write_eld(struct drm_connector *connector,
 			  struct drm_crtc *crtc)
 {
@@ -5827,6 +5875,12 @@
 	else
 		eldv = G4X_ELDV_DEVCTG;
 
+	if (intel_eld_uptodate(connector,
+			       G4X_AUD_CNTL_ST, eldv,
+			       G4X_AUD_CNTL_ST, G4X_ELD_ADDR,
+			       G4X_HDMIW_HDMIEDID))
+		return;
+
 	i = I915_READ(G4X_AUD_CNTL_ST);
 	i &= ~(eldv | G4X_ELD_ADDR);
 	len = (i >> 9) & 0x1f;		/* ELD buffer size */
@@ -5857,14 +5911,14 @@
 	int aud_cntl_st;
 	int aud_cntrl_st2;
 
-	if (IS_IVYBRIDGE(connector->dev)) {
-		hdmiw_hdmiedid = GEN7_HDMIW_HDMIEDID_A;
-		aud_cntl_st = GEN7_AUD_CNTRL_ST_A;
-		aud_cntrl_st2 = GEN7_AUD_CNTRL_ST2;
+	if (HAS_PCH_IBX(connector->dev)) {
+		hdmiw_hdmiedid = IBX_HDMIW_HDMIEDID_A;
+		aud_cntl_st = IBX_AUD_CNTL_ST_A;
+		aud_cntrl_st2 = IBX_AUD_CNTL_ST2;
 	} else {
-		hdmiw_hdmiedid = GEN5_HDMIW_HDMIEDID_A;
-		aud_cntl_st = GEN5_AUD_CNTL_ST_A;
-		aud_cntrl_st2 = GEN5_AUD_CNTL_ST2;
+		hdmiw_hdmiedid = CPT_HDMIW_HDMIEDID_A;
+		aud_cntl_st = CPT_AUD_CNTL_ST_A;
+		aud_cntrl_st2 = CPT_AUD_CNTRL_ST2;
 	}
 
 	i = to_intel_crtc(crtc)->pipe;
@@ -5878,14 +5932,25 @@
 	if (!i) {
 		DRM_DEBUG_DRIVER("Audio directed to unknown port\n");
 		/* operate blindly on all ports */
-		eldv = GEN5_ELD_VALIDB;
-		eldv |= GEN5_ELD_VALIDB << 4;
-		eldv |= GEN5_ELD_VALIDB << 8;
+		eldv = IBX_ELD_VALIDB;
+		eldv |= IBX_ELD_VALIDB << 4;
+		eldv |= IBX_ELD_VALIDB << 8;
 	} else {
 		DRM_DEBUG_DRIVER("ELD on port %c\n", 'A' + i);
-		eldv = GEN5_ELD_VALIDB << ((i - 1) * 4);
+		eldv = IBX_ELD_VALIDB << ((i - 1) * 4);
 	}
 
+	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) {
+		DRM_DEBUG_DRIVER("ELD: DisplayPort detected\n");
+		eld[5] |= (1 << 2);	/* Conn_Type, 0x1 = DisplayPort */
+	}
+
+	if (intel_eld_uptodate(connector,
+			       aud_cntrl_st2, eldv,
+			       aud_cntl_st, IBX_ELD_ADDRESS,
+			       hdmiw_hdmiedid))
+		return;
+
 	i = I915_READ(aud_cntrl_st2);
 	i &= ~eldv;
 	I915_WRITE(aud_cntrl_st2, i);
@@ -5893,13 +5958,8 @@
 	if (!eld[0])
 		return;
 
-	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) {
-		DRM_DEBUG_DRIVER("ELD: DisplayPort detected\n");
-		eld[5] |= (1 << 2);	/* Conn_Type, 0x1 = DisplayPort */
-	}
-
 	i = I915_READ(aud_cntl_st);
-	i &= ~GEN5_ELD_ADDRESS;
+	i &= ~IBX_ELD_ADDRESS;
 	I915_WRITE(aud_cntl_st, i);
 
 	len = min_t(uint8_t, eld[2], 21);	/* 84 bytes of hw ELD buffer */
@@ -6353,11 +6413,11 @@
 		return NULL;
 
 	fb = &dev_priv->fbdev->ifb.base;
-	if (fb->pitch < intel_framebuffer_pitch_for_width(mode->hdisplay,
-							  fb->bits_per_pixel))
+	if (fb->pitches[0] < intel_framebuffer_pitch_for_width(mode->hdisplay,
+							       fb->bits_per_pixel))
 		return NULL;
 
-	if (obj->base.size < mode->vdisplay * fb->pitch)
+	if (obj->base.size < mode->vdisplay * fb->pitches[0])
 		return NULL;
 
 	return fb;
@@ -6990,7 +7050,7 @@
 		goto out;
 
 	/* Offset into the new buffer for cases of shared fbs between CRTCs */
-	offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8;
+	offset = crtc->y * fb->pitches[0] + crtc->x * fb->bits_per_pixel/8;
 
 	ret = BEGIN_LP_RING(6);
 	if (ret)
@@ -7007,7 +7067,7 @@
 	OUT_RING(MI_NOOP);
 	OUT_RING(MI_DISPLAY_FLIP |
 		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	OUT_RING(fb->pitch);
+	OUT_RING(fb->pitches[0]);
 	OUT_RING(obj->gtt_offset + offset);
 	OUT_RING(MI_NOOP);
 	ADVANCE_LP_RING();
@@ -7031,7 +7091,7 @@
 		goto out;
 
 	/* Offset into the new buffer for cases of shared fbs between CRTCs */
-	offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8;
+	offset = crtc->y * fb->pitches[0] + crtc->x * fb->bits_per_pixel/8;
 
 	ret = BEGIN_LP_RING(6);
 	if (ret)
@@ -7045,7 +7105,7 @@
 	OUT_RING(MI_NOOP);
 	OUT_RING(MI_DISPLAY_FLIP_I915 |
 		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	OUT_RING(fb->pitch);
+	OUT_RING(fb->pitches[0]);
 	OUT_RING(obj->gtt_offset + offset);
 	OUT_RING(MI_NOOP);
 
@@ -7078,7 +7138,7 @@
 	 */
 	OUT_RING(MI_DISPLAY_FLIP |
 		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	OUT_RING(fb->pitch);
+	OUT_RING(fb->pitches[0]);
 	OUT_RING(obj->gtt_offset | obj->tiling_mode);
 
 	/* XXX Enabling the panel-fitter across page-flip is so far
@@ -7113,7 +7173,7 @@
 
 	OUT_RING(MI_DISPLAY_FLIP |
 		 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
-	OUT_RING(fb->pitch | obj->tiling_mode);
+	OUT_RING(fb->pitches[0] | obj->tiling_mode);
 	OUT_RING(obj->gtt_offset);
 
 	pf = I915_READ(PF_CTL(intel_crtc->pipe)) & PF_ENABLE;
@@ -7149,7 +7209,7 @@
 		goto out;
 
 	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | (intel_crtc->plane << 19));
-	intel_ring_emit(ring, (fb->pitch | obj->tiling_mode));
+	intel_ring_emit(ring, (fb->pitches[0] | obj->tiling_mode));
 	intel_ring_emit(ring, (obj->gtt_offset));
 	intel_ring_emit(ring, (MI_NOOP));
 	intel_ring_advance(ring);
@@ -7188,11 +7248,16 @@
 	work->old_fb_obj = intel_fb->obj;
 	INIT_WORK(&work->work, intel_unpin_work_fn);
 
+	ret = drm_vblank_get(dev, intel_crtc->pipe);
+	if (ret)
+		goto free_work;
+
 	/* We borrow the event spin lock for protecting unpin_work */
 	spin_lock_irqsave(&dev->event_lock, flags);
 	if (intel_crtc->unpin_work) {
 		spin_unlock_irqrestore(&dev->event_lock, flags);
 		kfree(work);
+		drm_vblank_put(dev, intel_crtc->pipe);
 
 		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
 		return -EBUSY;
@@ -7211,10 +7276,6 @@
 
 	crtc->fb = fb;
 
-	ret = drm_vblank_get(dev, intel_crtc->pipe);
-	if (ret)
-		goto cleanup_objs;
-
 	work->pending_flip_obj = obj;
 
 	work->enable_stall_check = true;
@@ -7237,7 +7298,6 @@
 
 cleanup_pending:
 	atomic_sub(1 << intel_crtc->plane, &work->old_fb_obj->pending_flip);
-cleanup_objs:
 	drm_gem_object_unreference(&work->old_fb_obj->base);
 	drm_gem_object_unreference(&obj->base);
 	mutex_unlock(&dev->struct_mutex);
@@ -7246,6 +7306,8 @@
 	intel_crtc->unpin_work = NULL;
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 
+	drm_vblank_put(dev, intel_crtc->pipe);
+free_work:
 	kfree(work);
 
 	return ret;
@@ -7891,6 +7953,33 @@
 	dev_priv->corr = (lcfuse & LCFUSE_HIV_MASK);
 }
 
+static bool intel_enable_rc6(struct drm_device *dev)
+{
+	/*
+	 * Respect the kernel parameter if it is set
+	 */
+	if (i915_enable_rc6 >= 0)
+		return i915_enable_rc6;
+
+	/*
+	 * Disable RC6 on Ironlake
+	 */
+	if (INTEL_INFO(dev)->gen == 5)
+		return 0;
+
+	/*
+	 * Enable rc6 on Sandybridge if DMA remapping is disabled
+	 */
+	if (INTEL_INFO(dev)->gen == 6) {
+		DRM_DEBUG_DRIVER("Sandybridge: intel_iommu_enabled %s -- RC6 %sabled\n",
+				 intel_iommu_enabled ? "true" : "false",
+				 !intel_iommu_enabled ? "en" : "dis");
+		return !intel_iommu_enabled;
+	}
+	DRM_DEBUG_DRIVER("RC6 enabled\n");
+	return 1;
+}
+
 void gen6_enable_rps(struct drm_i915_private *dev_priv)
 {
 	u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
@@ -7927,7 +8016,7 @@
 	I915_WRITE(GEN6_RC6p_THRESHOLD, 100000);
 	I915_WRITE(GEN6_RC6pp_THRESHOLD, 64000); /* unused */
 
-	if (i915_enable_rc6)
+	if (intel_enable_rc6(dev_priv->dev))
 		rc6_mask = GEN6_RC_CTL_RC6p_ENABLE |
 			GEN6_RC_CTL_RC6_ENABLE;
 
@@ -8153,6 +8242,20 @@
 	I915_WRITE(WM2_LP_ILK, 0);
 	I915_WRITE(WM1_LP_ILK, 0);
 
+	/* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock
+	 * gating disable must be set.  Failure to set it results in
+	 * flickering pixels due to Z write ordering failures after
+	 * some amount of runtime in the Mesa "fire" demo, and Unigine
+	 * Sanctuary and Tropics, and apparently anything else with
+	 * alpha test or pixel discard.
+	 *
+	 * According to the spec, bit 11 (RCCUNIT) must also be set,
+	 * but we didn't debug actual testcases to find it out.
+	 */
+	I915_WRITE(GEN6_UCGCTL2,
+		   GEN6_RCPBUNIT_CLOCK_GATE_DISABLE |
+		   GEN6_RCCUNIT_CLOCK_GATE_DISABLE);
+
 	/*
 	 * According to the spec the following bits should be
 	 * set in order to enable memory self-refresh and fbc:
@@ -8362,7 +8465,7 @@
 	/* rc6 disabled by default due to repeated reports of hanging during
 	 * boot and resume.
 	 */
-	if (!i915_enable_rc6)
+	if (!intel_enable_rc6(dev))
 		return;
 
 	mutex_lock(&dev->struct_mutex);
@@ -8481,6 +8584,28 @@
 
 	/* For FIFO watermark updates */
 	if (HAS_PCH_SPLIT(dev)) {
+		dev_priv->display.force_wake_get = __gen6_gt_force_wake_get;
+		dev_priv->display.force_wake_put = __gen6_gt_force_wake_put;
+
+		/* IVB configs may use multi-threaded forcewake */
+		if (IS_IVYBRIDGE(dev)) {
+			u32	ecobus;
+
+			mutex_lock(&dev->struct_mutex);
+			__gen6_gt_force_wake_mt_get(dev_priv);
+			ecobus = I915_READ(ECOBUS);
+			__gen6_gt_force_wake_mt_put(dev_priv);
+			mutex_unlock(&dev->struct_mutex);
+
+			if (ecobus & FORCEWAKE_MT_ENABLE) {
+				DRM_DEBUG_KMS("Using MT version of forcewake\n");
+				dev_priv->display.force_wake_get =
+					__gen6_gt_force_wake_mt_get;
+				dev_priv->display.force_wake_put =
+					__gen6_gt_force_wake_mt_put;
+			}
+		}
+
 		if (HAS_PCH_IBX(dev))
 			dev_priv->display.init_pch_clock_gating = ibx_init_clock_gating;
 		else if (HAS_PCH_CPT(dev))
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 09b318b..db3b461 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -59,7 +59,6 @@
 	struct i2c_algo_dp_aux_data algo;
 	bool is_pch_edp;
 	uint8_t	train_set[4];
-	uint8_t link_status[DP_LINK_STATUS_SIZE];
 	int panel_power_up_delay;
 	int panel_power_down_delay;
 	int panel_power_cycle_delay;
@@ -68,7 +67,6 @@
 	struct drm_display_mode *panel_fixed_mode;  /* for eDP */
 	struct delayed_work panel_vdd_work;
 	bool want_panel_vdd;
-	unsigned long panel_off_jiffies;
 };
 
 /**
@@ -157,16 +155,12 @@
 static int
 intel_dp_max_lane_count(struct intel_dp *intel_dp)
 {
-	int max_lane_count = 4;
-
-	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11) {
-		max_lane_count = intel_dp->dpcd[DP_MAX_LANE_COUNT] & 0x1f;
-		switch (max_lane_count) {
-		case 1: case 2: case 4:
-			break;
-		default:
-			max_lane_count = 4;
-		}
+	int max_lane_count = intel_dp->dpcd[DP_MAX_LANE_COUNT] & 0x1f;
+	switch (max_lane_count) {
+	case 1: case 2: case 4:
+		break;
+	default:
+		max_lane_count = 4;
 	}
 	return max_lane_count;
 }
@@ -214,13 +208,15 @@
  */
 
 static int
-intel_dp_link_required(struct intel_dp *intel_dp, int pixel_clock)
+intel_dp_link_required(struct intel_dp *intel_dp, int pixel_clock, int check_bpp)
 {
 	struct drm_crtc *crtc = intel_dp->base.base.crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int bpp = 24;
 
-	if (intel_crtc)
+	if (check_bpp)
+		bpp = check_bpp;
+	else if (intel_crtc)
 		bpp = intel_crtc->bpp;
 
 	return (pixel_clock * bpp + 9) / 10;
@@ -239,6 +235,7 @@
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
 	int max_link_clock = intel_dp_link_clock(intel_dp_max_link_bw(intel_dp));
 	int max_lanes = intel_dp_max_lane_count(intel_dp);
+	int max_rate, mode_rate;
 
 	if (is_edp(intel_dp) && intel_dp->panel_fixed_mode) {
 		if (mode->hdisplay > intel_dp->panel_fixed_mode->hdisplay)
@@ -248,9 +245,17 @@
 			return MODE_PANEL;
 	}
 
-	if (intel_dp_link_required(intel_dp, mode->clock)
-	    > intel_dp_max_data_rate(max_link_clock, max_lanes))
-		return MODE_CLOCK_HIGH;
+	mode_rate = intel_dp_link_required(intel_dp, mode->clock, 0);
+	max_rate = intel_dp_max_data_rate(max_link_clock, max_lanes);
+
+	if (mode_rate > max_rate) {
+			mode_rate = intel_dp_link_required(intel_dp,
+							   mode->clock, 18);
+			if (mode_rate > max_rate)
+				return MODE_CLOCK_HIGH;
+			else
+				mode->private_flags |= INTEL_MODE_DP_FORCE_6BPC;
+	}
 
 	if (mode->clock < 10000)
 		return MODE_CLOCK_LOW;
@@ -368,8 +373,8 @@
 	 * clock divider.
 	 */
 	if (is_cpu_edp(intel_dp)) {
-		if (IS_GEN6(dev))
-			aux_clock_divider = 200; /* SNB eDP input clock at 400Mhz */
+		if (IS_GEN6(dev) || IS_GEN7(dev))
+			aux_clock_divider = 200; /* SNB & IVB eDP input clock at 400Mhz */
 		else
 			aux_clock_divider = 225; /* eDP input clock at 450Mhz */
 	} else if (HAS_PCH_SPLIT(dev))
@@ -678,6 +683,7 @@
 	int lane_count, clock;
 	int max_lane_count = intel_dp_max_lane_count(intel_dp);
 	int max_clock = intel_dp_max_link_bw(intel_dp) == DP_LINK_BW_2_7 ? 1 : 0;
+	int bpp = mode->private_flags & INTEL_MODE_DP_FORCE_6BPC ? 18 : 0;
 	static int bws[2] = { DP_LINK_BW_1_62, DP_LINK_BW_2_7 };
 
 	if (is_edp(intel_dp) && intel_dp->panel_fixed_mode) {
@@ -695,7 +701,7 @@
 		for (clock = 0; clock <= max_clock; clock++) {
 			int link_avail = intel_dp_max_data_rate(intel_dp_link_clock(bws[clock]), lane_count);
 
-			if (intel_dp_link_required(intel_dp, mode->clock)
+			if (intel_dp_link_required(intel_dp, mode->clock, bpp)
 					<= link_avail) {
 				intel_dp->link_bw = bws[clock];
 				intel_dp->lane_count = lane_count;
@@ -768,12 +774,11 @@
 			continue;
 
 		intel_dp = enc_to_intel_dp(encoder);
-		if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT) {
+		if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT ||
+		    intel_dp->base.type == INTEL_OUTPUT_EDP)
+		{
 			lane_count = intel_dp->lane_count;
 			break;
-		} else if (is_edp(intel_dp)) {
-			lane_count = dev_priv->edp.lanes;
-			break;
 		}
 	}
 
@@ -810,6 +815,7 @@
 		  struct drm_display_mode *adjusted_mode)
 {
 	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct drm_crtc *crtc = intel_dp->base.base.crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -822,18 +828,32 @@
 			ironlake_edp_pll_off(encoder);
 	}
 
-	intel_dp->DP = DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
-	intel_dp->DP |= intel_dp->color_range;
+	/*
+	 * There are four kinds of DP registers:
+	 *
+	 * 	IBX PCH
+	 * 	SNB CPU
+	 *	IVB CPU
+	 * 	CPT PCH
+	 *
+	 * IBX PCH and CPU are the same for almost everything,
+	 * except that the CPU DP PLL is configured in this
+	 * register
+	 *
+	 * CPT PCH is quite different, having many bits moved
+	 * to the TRANS_DP_CTL register instead. That
+	 * configuration happens (oddly) in ironlake_pch_enable
+	 */
 
-	if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
-		intel_dp->DP |= DP_SYNC_HS_HIGH;
-	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
-		intel_dp->DP |= DP_SYNC_VS_HIGH;
+	/* Preserve the BIOS-computed detected bit. This is
+	 * supposed to be read-only.
+	 */
+	intel_dp->DP = I915_READ(intel_dp->output_reg) & DP_DETECTED;
+	intel_dp->DP |=  DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
 
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
-		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
-	else
-		intel_dp->DP |= DP_LINK_TRAIN_OFF;
+	/* Handle DP bits in common between all three register formats */
+
+	intel_dp->DP |= DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
 
 	switch (intel_dp->lane_count) {
 	case 1:
@@ -852,59 +872,124 @@
 		intel_dp->DP |= DP_AUDIO_OUTPUT_ENABLE;
 		intel_write_eld(encoder, adjusted_mode);
 	}
-
 	memset(intel_dp->link_configuration, 0, DP_LINK_CONFIGURATION_SIZE);
 	intel_dp->link_configuration[0] = intel_dp->link_bw;
 	intel_dp->link_configuration[1] = intel_dp->lane_count;
 	intel_dp->link_configuration[8] = DP_SET_ANSI_8B10B;
-
 	/*
 	 * Check for DPCD version > 1.1 and enhanced framing support
 	 */
 	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
 	    (intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP)) {
 		intel_dp->link_configuration[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
-		intel_dp->DP |= DP_ENHANCED_FRAMING;
 	}
 
-	/* CPT DP's pipe select is decided in TRANS_DP_CTL */
-	if (intel_crtc->pipe == 1 && !HAS_PCH_CPT(dev))
-		intel_dp->DP |= DP_PIPEB_SELECT;
+	/* Split out the IBX/CPU vs CPT settings */
 
-	if (is_cpu_edp(intel_dp)) {
+	if (is_cpu_edp(intel_dp) && IS_GEN7(dev)) {
+		if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
+			intel_dp->DP |= DP_SYNC_HS_HIGH;
+		if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
+			intel_dp->DP |= DP_SYNC_VS_HIGH;
+		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
+
+		if (intel_dp->link_configuration[1] & DP_LANE_COUNT_ENHANCED_FRAME_EN)
+			intel_dp->DP |= DP_ENHANCED_FRAMING;
+
+		intel_dp->DP |= intel_crtc->pipe << 29;
+
 		/* don't miss out required setting for eDP */
 		intel_dp->DP |= DP_PLL_ENABLE;
 		if (adjusted_mode->clock < 200000)
 			intel_dp->DP |= DP_PLL_FREQ_160MHZ;
 		else
 			intel_dp->DP |= DP_PLL_FREQ_270MHZ;
+	} else if (!HAS_PCH_CPT(dev) || is_cpu_edp(intel_dp)) {
+		intel_dp->DP |= intel_dp->color_range;
+
+		if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
+			intel_dp->DP |= DP_SYNC_HS_HIGH;
+		if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
+			intel_dp->DP |= DP_SYNC_VS_HIGH;
+		intel_dp->DP |= DP_LINK_TRAIN_OFF;
+
+		if (intel_dp->link_configuration[1] & DP_LANE_COUNT_ENHANCED_FRAME_EN)
+			intel_dp->DP |= DP_ENHANCED_FRAMING;
+
+		if (intel_crtc->pipe == 1)
+			intel_dp->DP |= DP_PIPEB_SELECT;
+
+		if (is_cpu_edp(intel_dp)) {
+			/* don't miss out required setting for eDP */
+			intel_dp->DP |= DP_PLL_ENABLE;
+			if (adjusted_mode->clock < 200000)
+				intel_dp->DP |= DP_PLL_FREQ_160MHZ;
+			else
+				intel_dp->DP |= DP_PLL_FREQ_270MHZ;
+		}
+	} else {
+		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
 	}
 }
 
+#define IDLE_ON_MASK		(PP_ON | 0 	  | PP_SEQUENCE_MASK | 0                     | PP_SEQUENCE_STATE_MASK)
+#define IDLE_ON_VALUE   	(PP_ON | 0 	  | PP_SEQUENCE_NONE | 0                     | PP_SEQUENCE_STATE_ON_IDLE)
+
+#define IDLE_OFF_MASK		(PP_ON | 0        | PP_SEQUENCE_MASK | 0                     | PP_SEQUENCE_STATE_MASK)
+#define IDLE_OFF_VALUE		(0     | 0        | PP_SEQUENCE_NONE | 0                     | PP_SEQUENCE_STATE_OFF_IDLE)
+
+#define IDLE_CYCLE_MASK		(PP_ON | 0        | PP_SEQUENCE_MASK | PP_CYCLE_DELAY_ACTIVE | PP_SEQUENCE_STATE_MASK)
+#define IDLE_CYCLE_VALUE	(0     | 0        | PP_SEQUENCE_NONE | 0                     | PP_SEQUENCE_STATE_OFF_IDLE)
+
+static void ironlake_wait_panel_status(struct intel_dp *intel_dp,
+				       u32 mask,
+				       u32 value)
+{
+	struct drm_device *dev = intel_dp->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	DRM_DEBUG_KMS("mask %08x value %08x status %08x control %08x\n",
+		      mask, value,
+		      I915_READ(PCH_PP_STATUS),
+		      I915_READ(PCH_PP_CONTROL));
+
+	if (_wait_for((I915_READ(PCH_PP_STATUS) & mask) == value, 5000, 10)) {
+		DRM_ERROR("Panel status timeout: status %08x control %08x\n",
+			  I915_READ(PCH_PP_STATUS),
+			  I915_READ(PCH_PP_CONTROL));
+	}
+}
+
+static void ironlake_wait_panel_on(struct intel_dp *intel_dp)
+{
+	DRM_DEBUG_KMS("Wait for panel power on\n");
+	ironlake_wait_panel_status(intel_dp, IDLE_ON_MASK, IDLE_ON_VALUE);
+}
+
 static void ironlake_wait_panel_off(struct intel_dp *intel_dp)
 {
-	unsigned long	off_time;
-	unsigned long	delay;
-
 	DRM_DEBUG_KMS("Wait for panel power off time\n");
+	ironlake_wait_panel_status(intel_dp, IDLE_OFF_MASK, IDLE_OFF_VALUE);
+}
 
-	if (ironlake_edp_have_panel_power(intel_dp) ||
-	    ironlake_edp_have_panel_vdd(intel_dp))
-	{
-		DRM_DEBUG_KMS("Panel still on, no delay needed\n");
-		return;
-	}
+static void ironlake_wait_panel_power_cycle(struct intel_dp *intel_dp)
+{
+	DRM_DEBUG_KMS("Wait for panel power cycle\n");
+	ironlake_wait_panel_status(intel_dp, IDLE_CYCLE_MASK, IDLE_CYCLE_VALUE);
+}
 
-	off_time = intel_dp->panel_off_jiffies + msecs_to_jiffies(intel_dp->panel_power_down_delay);
-	if (time_after(jiffies, off_time)) {
-		DRM_DEBUG_KMS("Time already passed");
-		return;
-	}
-	delay = jiffies_to_msecs(off_time - jiffies);
-	if (delay > intel_dp->panel_power_down_delay)
-		delay = intel_dp->panel_power_down_delay;
-	DRM_DEBUG_KMS("Waiting an additional %ld ms\n", delay);
-	msleep(delay);
+
+/* Read the current pp_control value, unlocking the register if it
+ * is locked
+ */
+
+static  u32 ironlake_get_pp_control(struct drm_i915_private *dev_priv)
+{
+	u32	control = I915_READ(PCH_PP_CONTROL);
+
+	control &= ~PANEL_UNLOCK_MASK;
+	control |= PANEL_UNLOCK_REGS;
+	return control;
 }
 
 static void ironlake_edp_panel_vdd_on(struct intel_dp *intel_dp)
@@ -921,15 +1006,16 @@
 	     "eDP VDD already requested on\n");
 
 	intel_dp->want_panel_vdd = true;
+
 	if (ironlake_edp_have_panel_vdd(intel_dp)) {
 		DRM_DEBUG_KMS("eDP VDD already on\n");
 		return;
 	}
 
-	ironlake_wait_panel_off(intel_dp);
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
+	if (!ironlake_edp_have_panel_power(intel_dp))
+		ironlake_wait_panel_power_cycle(intel_dp);
+
+	pp = ironlake_get_pp_control(dev_priv);
 	pp |= EDP_FORCE_VDD;
 	I915_WRITE(PCH_PP_CONTROL, pp);
 	POSTING_READ(PCH_PP_CONTROL);
@@ -952,9 +1038,7 @@
 	u32 pp;
 
 	if (!intel_dp->want_panel_vdd && ironlake_edp_have_panel_vdd(intel_dp)) {
-		pp = I915_READ(PCH_PP_CONTROL);
-		pp &= ~PANEL_UNLOCK_MASK;
-		pp |= PANEL_UNLOCK_REGS;
+		pp = ironlake_get_pp_control(dev_priv);
 		pp &= ~EDP_FORCE_VDD;
 		I915_WRITE(PCH_PP_CONTROL, pp);
 		POSTING_READ(PCH_PP_CONTROL);
@@ -962,7 +1046,8 @@
 		/* Make sure sequencer is idle before allowing subsequent activity */
 		DRM_DEBUG_KMS("PCH_PP_STATUS: 0x%08x PCH_PP_CONTROL: 0x%08x\n",
 			      I915_READ(PCH_PP_STATUS), I915_READ(PCH_PP_CONTROL));
-		intel_dp->panel_off_jiffies = jiffies;
+
+		msleep(intel_dp->panel_power_down_delay);
 	}
 }
 
@@ -972,9 +1057,9 @@
 						 struct intel_dp, panel_vdd_work);
 	struct drm_device *dev = intel_dp->base.base.dev;
 
-	mutex_lock(&dev->struct_mutex);
+	mutex_lock(&dev->mode_config.mutex);
 	ironlake_panel_vdd_off_sync(intel_dp);
-	mutex_unlock(&dev->struct_mutex);
+	mutex_unlock(&dev->mode_config.mutex);
 }
 
 static void ironlake_edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync)
@@ -984,7 +1069,7 @@
 
 	DRM_DEBUG_KMS("Turn eDP VDD off %d\n", intel_dp->want_panel_vdd);
 	WARN(!intel_dp->want_panel_vdd, "eDP VDD not forced on");
-	
+
 	intel_dp->want_panel_vdd = false;
 
 	if (sync) {
@@ -1000,23 +1085,25 @@
 	}
 }
 
-/* Returns true if the panel was already on when called */
 static void ironlake_edp_panel_on(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pp, idle_on_mask = PP_ON | PP_SEQUENCE_STATE_ON_IDLE;
+	u32 pp;
 
 	if (!is_edp(intel_dp))
 		return;
-	if (ironlake_edp_have_panel_power(intel_dp))
+
+	DRM_DEBUG_KMS("Turn eDP power on\n");
+
+	if (ironlake_edp_have_panel_power(intel_dp)) {
+		DRM_DEBUG_KMS("eDP power already on\n");
 		return;
+	}
 
-	ironlake_wait_panel_off(intel_dp);
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
+	ironlake_wait_panel_power_cycle(intel_dp);
 
+	pp = ironlake_get_pp_control(dev_priv);
 	if (IS_GEN5(dev)) {
 		/* ILK workaround: disable reset around power sequence */
 		pp &= ~PANEL_POWER_RESET;
@@ -1025,13 +1112,13 @@
 	}
 
 	pp |= POWER_TARGET_ON;
+	if (!IS_GEN5(dev))
+		pp |= PANEL_POWER_RESET;
+
 	I915_WRITE(PCH_PP_CONTROL, pp);
 	POSTING_READ(PCH_PP_CONTROL);
 
-	if (wait_for((I915_READ(PCH_PP_STATUS) & idle_on_mask) == idle_on_mask,
-		     5000))
-		DRM_ERROR("panel on wait timed out: 0x%08x\n",
-			  I915_READ(PCH_PP_STATUS));
+	ironlake_wait_panel_on(intel_dp);
 
 	if (IS_GEN5(dev)) {
 		pp |= PANEL_POWER_RESET; /* restore panel reset bit */
@@ -1040,46 +1127,25 @@
 	}
 }
 
-static void ironlake_edp_panel_off(struct drm_encoder *encoder)
+static void ironlake_edp_panel_off(struct intel_dp *intel_dp)
 {
-	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-	struct drm_device *dev = encoder->dev;
+	struct drm_device *dev = intel_dp->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pp, idle_off_mask = PP_ON | PP_SEQUENCE_MASK |
-		PP_CYCLE_DELAY_ACTIVE | PP_SEQUENCE_STATE_MASK;
+	u32 pp;
 
 	if (!is_edp(intel_dp))
 		return;
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
 
-	if (IS_GEN5(dev)) {
-		/* ILK workaround: disable reset around power sequence */
-		pp &= ~PANEL_POWER_RESET;
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
-	}
+	DRM_DEBUG_KMS("Turn eDP power off\n");
 
-	intel_dp->panel_off_jiffies = jiffies;
+	WARN(intel_dp->want_panel_vdd, "Cannot turn power off while VDD is on\n");
 
-	if (IS_GEN5(dev)) {
-		pp &= ~POWER_TARGET_ON;
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
-		pp &= ~POWER_TARGET_ON;
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
-		msleep(intel_dp->panel_power_cycle_delay);
+	pp = ironlake_get_pp_control(dev_priv);
+	pp &= ~(POWER_TARGET_ON | EDP_FORCE_VDD | PANEL_POWER_RESET | EDP_BLC_ENABLE);
+	I915_WRITE(PCH_PP_CONTROL, pp);
+	POSTING_READ(PCH_PP_CONTROL);
 
-		if (wait_for((I915_READ(PCH_PP_STATUS) & idle_off_mask) == 0, 5000))
-			DRM_ERROR("panel off wait timed out: 0x%08x\n",
-				  I915_READ(PCH_PP_STATUS));
-
-		pp |= PANEL_POWER_RESET; /* restore panel reset bit */
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
-	}
+	ironlake_wait_panel_off(intel_dp);
 }
 
 static void ironlake_edp_backlight_on(struct intel_dp *intel_dp)
@@ -1099,9 +1165,7 @@
 	 * allowing it to appear.
 	 */
 	msleep(intel_dp->backlight_on_delay);
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
+	pp = ironlake_get_pp_control(dev_priv);
 	pp |= EDP_BLC_ENABLE;
 	I915_WRITE(PCH_PP_CONTROL, pp);
 	POSTING_READ(PCH_PP_CONTROL);
@@ -1117,9 +1181,7 @@
 		return;
 
 	DRM_DEBUG_KMS("\n");
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
+	pp = ironlake_get_pp_control(dev_priv);
 	pp &= ~EDP_BLC_ENABLE;
 	I915_WRITE(PCH_PP_CONTROL, pp);
 	POSTING_READ(PCH_PP_CONTROL);
@@ -1187,17 +1249,18 @@
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
+	ironlake_edp_backlight_off(intel_dp);
+	ironlake_edp_panel_off(intel_dp);
+
 	/* Wake up the sink first */
 	ironlake_edp_panel_vdd_on(intel_dp);
 	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
+	intel_dp_link_down(intel_dp);
 	ironlake_edp_panel_vdd_off(intel_dp, false);
 
 	/* Make sure the panel is off before trying to
 	 * change the mode
 	 */
-	ironlake_edp_backlight_off(intel_dp);
-	intel_dp_link_down(intel_dp);
-	ironlake_edp_panel_off(encoder);
 }
 
 static void intel_dp_commit(struct drm_encoder *encoder)
@@ -1211,7 +1274,6 @@
 	intel_dp_start_link_train(intel_dp);
 	ironlake_edp_panel_on(intel_dp);
 	ironlake_edp_panel_vdd_off(intel_dp, true);
-
 	intel_dp_complete_link_train(intel_dp);
 	ironlake_edp_backlight_on(intel_dp);
 
@@ -1230,16 +1292,20 @@
 	uint32_t dp_reg = I915_READ(intel_dp->output_reg);
 
 	if (mode != DRM_MODE_DPMS_ON) {
+		ironlake_edp_backlight_off(intel_dp);
+		ironlake_edp_panel_off(intel_dp);
+
 		ironlake_edp_panel_vdd_on(intel_dp);
-		if (is_edp(intel_dp))
-			ironlake_edp_backlight_off(intel_dp);
 		intel_dp_sink_dpms(intel_dp, mode);
 		intel_dp_link_down(intel_dp);
-		ironlake_edp_panel_off(encoder);
-		if (is_edp(intel_dp) && !is_pch_edp(intel_dp))
-			ironlake_edp_pll_off(encoder);
 		ironlake_edp_panel_vdd_off(intel_dp, false);
+
+		if (is_cpu_edp(intel_dp))
+			ironlake_edp_pll_off(encoder);
 	} else {
+		if (is_cpu_edp(intel_dp))
+			ironlake_edp_pll_on(encoder);
+
 		ironlake_edp_panel_vdd_on(intel_dp);
 		intel_dp_sink_dpms(intel_dp, mode);
 		if (!(dp_reg & DP_PORT_EN)) {
@@ -1247,7 +1313,6 @@
 			ironlake_edp_panel_on(intel_dp);
 			ironlake_edp_panel_vdd_off(intel_dp, true);
 			intel_dp_complete_link_train(intel_dp);
-			ironlake_edp_backlight_on(intel_dp);
 		} else
 			ironlake_edp_panel_vdd_off(intel_dp, false);
 		ironlake_edp_backlight_on(intel_dp);
@@ -1285,11 +1350,11 @@
  * link status information
  */
 static bool
-intel_dp_get_link_status(struct intel_dp *intel_dp)
+intel_dp_get_link_status(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
 	return intel_dp_aux_native_read_retry(intel_dp,
 					      DP_LANE0_1_STATUS,
-					      intel_dp->link_status,
+					      link_status,
 					      DP_LINK_STATUS_SIZE);
 }
 
@@ -1301,27 +1366,25 @@
 }
 
 static uint8_t
-intel_get_adjust_request_voltage(uint8_t link_status[DP_LINK_STATUS_SIZE],
+intel_get_adjust_request_voltage(uint8_t adjust_request[2],
 				 int lane)
 {
-	int	    i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1);
 	int	    s = ((lane & 1) ?
 			 DP_ADJUST_VOLTAGE_SWING_LANE1_SHIFT :
 			 DP_ADJUST_VOLTAGE_SWING_LANE0_SHIFT);
-	uint8_t l = intel_dp_link_status(link_status, i);
+	uint8_t l = adjust_request[lane>>1];
 
 	return ((l >> s) & 3) << DP_TRAIN_VOLTAGE_SWING_SHIFT;
 }
 
 static uint8_t
-intel_get_adjust_request_pre_emphasis(uint8_t link_status[DP_LINK_STATUS_SIZE],
+intel_get_adjust_request_pre_emphasis(uint8_t adjust_request[2],
 				      int lane)
 {
-	int	    i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1);
 	int	    s = ((lane & 1) ?
 			 DP_ADJUST_PRE_EMPHASIS_LANE1_SHIFT :
 			 DP_ADJUST_PRE_EMPHASIS_LANE0_SHIFT);
-	uint8_t l = intel_dp_link_status(link_status, i);
+	uint8_t l = adjust_request[lane>>1];
 
 	return ((l >> s) & 3) << DP_TRAIN_PRE_EMPHASIS_SHIFT;
 }
@@ -1343,34 +1406,63 @@
  * These are source-specific values; current Intel hardware supports
  * a maximum voltage of 800mV and a maximum pre-emphasis of 6dB
  */
-#define I830_DP_VOLTAGE_MAX	    DP_TRAIN_VOLTAGE_SWING_800
 
 static uint8_t
-intel_dp_pre_emphasis_max(uint8_t voltage_swing)
+intel_dp_voltage_max(struct intel_dp *intel_dp)
 {
-	switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
-	case DP_TRAIN_VOLTAGE_SWING_400:
-		return DP_TRAIN_PRE_EMPHASIS_6;
-	case DP_TRAIN_VOLTAGE_SWING_600:
-		return DP_TRAIN_PRE_EMPHASIS_6;
-	case DP_TRAIN_VOLTAGE_SWING_800:
-		return DP_TRAIN_PRE_EMPHASIS_3_5;
-	case DP_TRAIN_VOLTAGE_SWING_1200:
-	default:
-		return DP_TRAIN_PRE_EMPHASIS_0;
+	struct drm_device *dev = intel_dp->base.base.dev;
+
+	if (IS_GEN7(dev) && is_cpu_edp(intel_dp))
+		return DP_TRAIN_VOLTAGE_SWING_800;
+	else if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+		return DP_TRAIN_VOLTAGE_SWING_1200;
+	else
+		return DP_TRAIN_VOLTAGE_SWING_800;
+}
+
+static uint8_t
+intel_dp_pre_emphasis_max(struct intel_dp *intel_dp, uint8_t voltage_swing)
+{
+	struct drm_device *dev = intel_dp->base.base.dev;
+
+	if (IS_GEN7(dev) && is_cpu_edp(intel_dp)) {
+		switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			return DP_TRAIN_PRE_EMPHASIS_6;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			return DP_TRAIN_PRE_EMPHASIS_3_5;
+		default:
+			return DP_TRAIN_PRE_EMPHASIS_0;
+		}
+	} else {
+		switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			return DP_TRAIN_PRE_EMPHASIS_6;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			return DP_TRAIN_PRE_EMPHASIS_6;
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			return DP_TRAIN_PRE_EMPHASIS_3_5;
+		case DP_TRAIN_VOLTAGE_SWING_1200:
+		default:
+			return DP_TRAIN_PRE_EMPHASIS_0;
+		}
 	}
 }
 
 static void
-intel_get_adjust_train(struct intel_dp *intel_dp)
+intel_get_adjust_train(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
 	uint8_t v = 0;
 	uint8_t p = 0;
 	int lane;
+	uint8_t	*adjust_request = link_status + (DP_ADJUST_REQUEST_LANE0_1 - DP_LANE0_1_STATUS);
+	uint8_t voltage_max;
+	uint8_t preemph_max;
 
 	for (lane = 0; lane < intel_dp->lane_count; lane++) {
-		uint8_t this_v = intel_get_adjust_request_voltage(intel_dp->link_status, lane);
-		uint8_t this_p = intel_get_adjust_request_pre_emphasis(intel_dp->link_status, lane);
+		uint8_t this_v = intel_get_adjust_request_voltage(adjust_request, lane);
+		uint8_t this_p = intel_get_adjust_request_pre_emphasis(adjust_request, lane);
 
 		if (this_v > v)
 			v = this_v;
@@ -1378,18 +1470,20 @@
 			p = this_p;
 	}
 
-	if (v >= I830_DP_VOLTAGE_MAX)
-		v = I830_DP_VOLTAGE_MAX | DP_TRAIN_MAX_SWING_REACHED;
+	voltage_max = intel_dp_voltage_max(intel_dp);
+	if (v >= voltage_max)
+		v = voltage_max | DP_TRAIN_MAX_SWING_REACHED;
 
-	if (p >= intel_dp_pre_emphasis_max(v))
-		p = intel_dp_pre_emphasis_max(v) | DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
+	preemph_max = intel_dp_pre_emphasis_max(intel_dp, v);
+	if (p >= preemph_max)
+		p = preemph_max | DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
 
 	for (lane = 0; lane < 4; lane++)
 		intel_dp->train_set[lane] = v | p;
 }
 
 static uint32_t
-intel_dp_signal_levels(uint8_t train_set, int lane_count)
+intel_dp_signal_levels(uint8_t train_set)
 {
 	uint32_t	signal_levels = 0;
 
@@ -1454,13 +1548,43 @@
 	}
 }
 
+/* Gen7's DP voltage swing and pre-emphasis control */
+static uint32_t
+intel_gen7_edp_signal_levels(uint8_t train_set)
+{
+	int signal_levels = train_set & (DP_TRAIN_VOLTAGE_SWING_MASK |
+					 DP_TRAIN_PRE_EMPHASIS_MASK);
+	switch (signal_levels) {
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_0:
+		return EDP_LINK_TRAIN_400MV_0DB_IVB;
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return EDP_LINK_TRAIN_400MV_3_5DB_IVB;
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_6:
+		return EDP_LINK_TRAIN_400MV_6DB_IVB;
+
+	case DP_TRAIN_VOLTAGE_SWING_600 | DP_TRAIN_PRE_EMPHASIS_0:
+		return EDP_LINK_TRAIN_600MV_0DB_IVB;
+	case DP_TRAIN_VOLTAGE_SWING_600 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return EDP_LINK_TRAIN_600MV_3_5DB_IVB;
+
+	case DP_TRAIN_VOLTAGE_SWING_800 | DP_TRAIN_PRE_EMPHASIS_0:
+		return EDP_LINK_TRAIN_800MV_0DB_IVB;
+	case DP_TRAIN_VOLTAGE_SWING_800 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return EDP_LINK_TRAIN_800MV_3_5DB_IVB;
+
+	default:
+		DRM_DEBUG_KMS("Unsupported voltage swing/pre-emphasis level:"
+			      "0x%x\n", signal_levels);
+		return EDP_LINK_TRAIN_500MV_0DB_IVB;
+	}
+}
+
 static uint8_t
 intel_get_lane_status(uint8_t link_status[DP_LINK_STATUS_SIZE],
 		      int lane)
 {
-	int i = DP_LANE0_1_STATUS + (lane >> 1);
 	int s = (lane & 1) * 4;
-	uint8_t l = intel_dp_link_status(link_status, i);
+	uint8_t l = link_status[lane>>1];
 
 	return (l >> s) & 0xf;
 }
@@ -1485,18 +1609,18 @@
 			 DP_LANE_CHANNEL_EQ_DONE|\
 			 DP_LANE_SYMBOL_LOCKED)
 static bool
-intel_channel_eq_ok(struct intel_dp *intel_dp)
+intel_channel_eq_ok(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
 	uint8_t lane_align;
 	uint8_t lane_status;
 	int lane;
 
-	lane_align = intel_dp_link_status(intel_dp->link_status,
+	lane_align = intel_dp_link_status(link_status,
 					  DP_LANE_ALIGN_STATUS_UPDATED);
 	if ((lane_align & DP_INTERLANE_ALIGN_DONE) == 0)
 		return false;
 	for (lane = 0; lane < intel_dp->lane_count; lane++) {
-		lane_status = intel_get_lane_status(intel_dp->link_status, lane);
+		lane_status = intel_get_lane_status(link_status, lane);
 		if ((lane_status & CHANNEL_EQ_BITS) != CHANNEL_EQ_BITS)
 			return false;
 	}
@@ -1521,8 +1645,9 @@
 
 	ret = intel_dp_aux_native_write(intel_dp,
 					DP_TRAINING_LANE0_SET,
-					intel_dp->train_set, 4);
-	if (ret != 4)
+					intel_dp->train_set,
+					intel_dp->lane_count);
+	if (ret != intel_dp->lane_count)
 		return false;
 
 	return true;
@@ -1538,7 +1663,7 @@
 	int i;
 	uint8_t voltage;
 	bool clock_recovery = false;
-	int tries;
+	int voltage_tries, loop_tries;
 	u32 reg;
 	uint32_t DP = intel_dp->DP;
 
@@ -1559,26 +1684,35 @@
 				  DP_LINK_CONFIGURATION_SIZE);
 
 	DP |= DP_PORT_EN;
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+
+	if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
 		DP &= ~DP_LINK_TRAIN_MASK_CPT;
 	else
 		DP &= ~DP_LINK_TRAIN_MASK;
 	memset(intel_dp->train_set, 0, 4);
 	voltage = 0xff;
-	tries = 0;
+	voltage_tries = 0;
+	loop_tries = 0;
 	clock_recovery = false;
 	for (;;) {
 		/* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */
+		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
 		uint32_t    signal_levels;
-		if (IS_GEN6(dev) && is_edp(intel_dp)) {
+
+
+		if (IS_GEN7(dev) && is_cpu_edp(intel_dp)) {
+			signal_levels = intel_gen7_edp_signal_levels(intel_dp->train_set[0]);
+			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_IVB) | signal_levels;
+		} else if (IS_GEN6(dev) && is_cpu_edp(intel_dp)) {
 			signal_levels = intel_gen6_edp_signal_levels(intel_dp->train_set[0]);
 			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_SNB) | signal_levels;
 		} else {
-			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0], intel_dp->lane_count);
+			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0]);
+			DRM_DEBUG_KMS("training pattern 1 signal levels %08x\n", signal_levels);
 			DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
 		}
 
-		if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+		if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
 			reg = DP | DP_LINK_TRAIN_PAT_1_CPT;
 		else
 			reg = DP | DP_LINK_TRAIN_PAT_1;
@@ -1590,10 +1724,13 @@
 		/* Set training pattern 1 */
 
 		udelay(100);
-		if (!intel_dp_get_link_status(intel_dp))
+		if (!intel_dp_get_link_status(intel_dp, link_status)) {
+			DRM_ERROR("failed to get link status\n");
 			break;
+		}
 
-		if (intel_clock_recovery_ok(intel_dp->link_status, intel_dp->lane_count)) {
+		if (intel_clock_recovery_ok(link_status, intel_dp->lane_count)) {
+			DRM_DEBUG_KMS("clock recovery OK\n");
 			clock_recovery = true;
 			break;
 		}
@@ -1602,20 +1739,30 @@
 		for (i = 0; i < intel_dp->lane_count; i++)
 			if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0)
 				break;
-		if (i == intel_dp->lane_count)
-			break;
+		if (i == intel_dp->lane_count) {
+			++loop_tries;
+			if (loop_tries == 5) {
+				DRM_DEBUG_KMS("too many full retries, give up\n");
+				break;
+			}
+			memset(intel_dp->train_set, 0, 4);
+			voltage_tries = 0;
+			continue;
+		}
 
 		/* Check to see if we've tried the same voltage 5 times */
 		if ((intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK) == voltage) {
-			++tries;
-			if (tries == 5)
+			++voltage_tries;
+			if (voltage_tries == 5) {
+				DRM_DEBUG_KMS("too many voltage retries, give up\n");
 				break;
+			}
 		} else
-			tries = 0;
+			voltage_tries = 0;
 		voltage = intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
 
 		/* Compute new intel_dp->train_set as requested by target */
-		intel_get_adjust_train(intel_dp);
+		intel_get_adjust_train(intel_dp, link_status);
 	}
 
 	intel_dp->DP = DP;
@@ -1638,6 +1785,7 @@
 	for (;;) {
 		/* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */
 		uint32_t    signal_levels;
+		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
 
 		if (cr_tries > 5) {
 			DRM_ERROR("failed to train DP, aborting\n");
@@ -1645,15 +1793,18 @@
 			break;
 		}
 
-		if (IS_GEN6(dev) && is_edp(intel_dp)) {
+		if (IS_GEN7(dev) && is_cpu_edp(intel_dp)) {
+			signal_levels = intel_gen7_edp_signal_levels(intel_dp->train_set[0]);
+			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_IVB) | signal_levels;
+		} else if (IS_GEN6(dev) && is_cpu_edp(intel_dp)) {
 			signal_levels = intel_gen6_edp_signal_levels(intel_dp->train_set[0]);
 			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_SNB) | signal_levels;
 		} else {
-			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0], intel_dp->lane_count);
+			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0]);
 			DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
 		}
 
-		if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+		if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
 			reg = DP | DP_LINK_TRAIN_PAT_2_CPT;
 		else
 			reg = DP | DP_LINK_TRAIN_PAT_2;
@@ -1665,17 +1816,17 @@
 			break;
 
 		udelay(400);
-		if (!intel_dp_get_link_status(intel_dp))
+		if (!intel_dp_get_link_status(intel_dp, link_status))
 			break;
 
 		/* Make sure clock is still ok */
-		if (!intel_clock_recovery_ok(intel_dp->link_status, intel_dp->lane_count)) {
+		if (!intel_clock_recovery_ok(link_status, intel_dp->lane_count)) {
 			intel_dp_start_link_train(intel_dp);
 			cr_tries++;
 			continue;
 		}
 
-		if (intel_channel_eq_ok(intel_dp)) {
+		if (intel_channel_eq_ok(intel_dp, link_status)) {
 			channel_eq = true;
 			break;
 		}
@@ -1690,11 +1841,11 @@
 		}
 
 		/* Compute new intel_dp->train_set as requested by target */
-		intel_get_adjust_train(intel_dp);
+		intel_get_adjust_train(intel_dp, link_status);
 		++tries;
 	}
 
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+	if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
 		reg = DP | DP_LINK_TRAIN_OFF_CPT;
 	else
 		reg = DP | DP_LINK_TRAIN_OFF;
@@ -1724,7 +1875,7 @@
 		udelay(100);
 	}
 
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp)) {
+	if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) {
 		DP &= ~DP_LINK_TRAIN_MASK_CPT;
 		I915_WRITE(intel_dp->output_reg, DP | DP_LINK_TRAIN_PAT_IDLE_CPT);
 	} else {
@@ -1735,8 +1886,12 @@
 
 	msleep(17);
 
-	if (is_edp(intel_dp))
-		DP |= DP_LINK_TRAIN_OFF;
+	if (is_edp(intel_dp)) {
+		if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
+			DP |= DP_LINK_TRAIN_OFF_CPT;
+		else
+			DP |= DP_LINK_TRAIN_OFF;
+	}
 
 	if (!HAS_PCH_CPT(dev) &&
 	    I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) {
@@ -1771,6 +1926,7 @@
 			intel_wait_for_vblank(dev, to_intel_crtc(crtc)->pipe);
 	}
 
+	DP &= ~DP_AUDIO_OUTPUT_ENABLE;
 	I915_WRITE(intel_dp->output_reg, DP & ~DP_PORT_EN);
 	POSTING_READ(intel_dp->output_reg);
 	msleep(intel_dp->panel_power_down_delay);
@@ -1822,6 +1978,7 @@
 intel_dp_check_link_status(struct intel_dp *intel_dp)
 {
 	u8 sink_irq_vector;
+	u8 link_status[DP_LINK_STATUS_SIZE];
 
 	if (intel_dp->dpms_mode != DRM_MODE_DPMS_ON)
 		return;
@@ -1830,7 +1987,7 @@
 		return;
 
 	/* Try to read receiver status if the link appears to be up */
-	if (!intel_dp_get_link_status(intel_dp)) {
+	if (!intel_dp_get_link_status(intel_dp, link_status)) {
 		intel_dp_link_down(intel_dp);
 		return;
 	}
@@ -1855,7 +2012,7 @@
 			DRM_DEBUG_DRIVER("CP or sink specific irq unhandled\n");
 	}
 
-	if (!intel_channel_eq_ok(intel_dp)) {
+	if (!intel_channel_eq_ok(intel_dp, link_status)) {
 		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
 			      drm_get_encoder_name(&intel_dp->base.base));
 		intel_dp_start_link_train(intel_dp);
@@ -2179,7 +2336,8 @@
 			continue;
 
 		intel_dp = enc_to_intel_dp(encoder);
-		if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT)
+		if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT ||
+		    intel_dp->base.type == INTEL_OUTPUT_EDP)
 			return intel_dp->output_reg;
 	}
 
@@ -2321,7 +2479,7 @@
 
 		cur.t8 = (pp_on & PANEL_LIGHT_ON_DELAY_MASK) >>
 			PANEL_LIGHT_ON_DELAY_SHIFT;
-		
+
 		cur.t9 = (pp_off & PANEL_LIGHT_OFF_DELAY_MASK) >>
 			PANEL_LIGHT_OFF_DELAY_SHIFT;
 
@@ -2354,11 +2512,10 @@
 		DRM_DEBUG_KMS("backlight on delay %d, off delay %d\n",
 			      intel_dp->backlight_on_delay, intel_dp->backlight_off_delay);
 
-		intel_dp->panel_off_jiffies = jiffies - intel_dp->panel_power_down_delay;
-
 		ironlake_edp_panel_vdd_on(intel_dp);
 		ret = intel_dp_get_dpcd(intel_dp);
 		ironlake_edp_panel_vdd_off(intel_dp, false);
+
 		if (ret) {
 			if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11)
 				dev_priv->no_aux_handshake =
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 23c5622..82a459b 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -110,6 +110,7 @@
 /* drm_display_mode->private_flags */
 #define INTEL_MODE_PIXEL_MULTIPLIER_SHIFT (0x0)
 #define INTEL_MODE_PIXEL_MULTIPLIER_MASK (0xf << INTEL_MODE_PIXEL_MULTIPLIER_SHIFT)
+#define INTEL_MODE_DP_FORCE_6BPC (0x10)
 
 static inline void
 intel_mode_set_pixel_multiplier(struct drm_display_mode *mode,
diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c
index dc1db4f..f02fc71 100644
--- a/drivers/gpu/drm/i915/intel_fb.c
+++ b/drivers/gpu/drm/i915/intel_fb.c
@@ -149,7 +149,7 @@
 
 //	memset(info->screen_base, 0, size);
 
-	drm_fb_helper_fill_fix(info, fb->pitch, fb->depth);
+	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->depth);
 	drm_fb_helper_fill_var(info, &ifbdev->helper, sizes->fb_width, sizes->fb_height);
 
 	info->pixmap.size = 64*1024;
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index d4f5a0b..64541f7 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -269,6 +269,10 @@
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	u32 temp;
+	u32 enable_bits = SDVO_ENABLE;
+
+	if (intel_hdmi->has_audio)
+		enable_bits |= SDVO_AUDIO_ENABLE;
 
 	temp = I915_READ(intel_hdmi->sdvox_reg);
 
@@ -281,9 +285,9 @@
 	}
 
 	if (mode != DRM_MODE_DPMS_ON) {
-		temp &= ~SDVO_ENABLE;
+		temp &= ~enable_bits;
 	} else {
-		temp |= SDVO_ENABLE;
+		temp |= enable_bits;
 	}
 
 	I915_WRITE(intel_hdmi->sdvox_reg, temp);
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 42f165a..e441911 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -715,6 +715,14 @@
 			DMI_MATCH(DMI_PRODUCT_NAME, "EB1007"),
 		},
 	},
+	{
+		.callback = intel_no_lvds_dmi_callback,
+		.ident = "Asus AT5NM10T-I",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "AT5NM10T-I"),
+		},
+	},
 
 	{ }	/* terminating entry */
 };
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 499d4c0..04d79fd 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -178,13 +178,10 @@
 	if (HAS_PCH_SPLIT(dev)) {
 		max >>= 16;
 	} else {
-		if (IS_PINEVIEW(dev)) {
+		if (INTEL_INFO(dev)->gen < 4)
 			max >>= 17;
-		} else {
+		else
 			max >>= 16;
-			if (INTEL_INFO(dev)->gen < 4)
-				max &= ~1;
-		}
 
 		if (is_backlight_combination_mode(dev))
 			max *= 0xff;
@@ -203,13 +200,12 @@
 		val = I915_READ(BLC_PWM_CPU_CTL) & BACKLIGHT_DUTY_CYCLE_MASK;
 	} else {
 		val = I915_READ(BLC_PWM_CTL) & BACKLIGHT_DUTY_CYCLE_MASK;
-		if (IS_PINEVIEW(dev))
+		if (INTEL_INFO(dev)->gen < 4)
 			val >>= 1;
 
 		if (is_backlight_combination_mode(dev)) {
 			u8 lbpc;
 
-			val &= ~1;
 			pci_read_config_byte(dev->pdev, PCI_LBPC, &lbpc);
 			val *= lbpc;
 		}
@@ -246,11 +242,9 @@
 	}
 
 	tmp = I915_READ(BLC_PWM_CTL);
-	if (IS_PINEVIEW(dev)) {
-		tmp &= ~(BACKLIGHT_DUTY_CYCLE_MASK - 1);
+	if (INTEL_INFO(dev)->gen < 4) 
 		level <<= 1;
-	} else
-		tmp &= ~BACKLIGHT_DUTY_CYCLE_MASK;
+	tmp &= ~BACKLIGHT_DUTY_CYCLE_MASK;
 	I915_WRITE(BLC_PWM_CTL, tmp | level);
 }
 
@@ -326,7 +320,8 @@
 static int intel_panel_get_brightness(struct backlight_device *bd)
 {
 	struct drm_device *dev = bl_get_data(bd);
-	return intel_panel_get_backlight(dev);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	return dev_priv->backlight_level;
 }
 
 static const struct backlight_ops intel_panel_bl_ops = {
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 3003fb2..f7b9268 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -50,6 +50,7 @@
 #define IS_TMDS(c)	(c->output_flag & SDVO_TMDS_MASK)
 #define IS_LVDS(c)	(c->output_flag & SDVO_LVDS_MASK)
 #define IS_TV_OR_LVDS(c) (c->output_flag & (SDVO_TV_MASK | SDVO_LVDS_MASK))
+#define IS_DIGITAL(c) (c->output_flag & (SDVO_TMDS_MASK | SDVO_LVDS_MASK))
 
 
 static const char *tv_format_names[] = {
@@ -1086,8 +1087,12 @@
 		}
 		sdvox |= (9 << 19) | SDVO_BORDER_ENABLE;
 	}
-	if (intel_crtc->pipe == 1)
-		sdvox |= SDVO_PIPE_B_SELECT;
+
+	if (INTEL_PCH_TYPE(dev) >= PCH_CPT)
+		sdvox |= TRANSCODER_CPT(intel_crtc->pipe);
+	else
+		sdvox |= TRANSCODER(intel_crtc->pipe);
+
 	if (intel_sdvo->has_hdmi_audio)
 		sdvox |= SDVO_AUDIO_ENABLE;
 
@@ -1314,6 +1319,18 @@
 	return status;
 }
 
+static bool
+intel_sdvo_connector_matches_edid(struct intel_sdvo_connector *sdvo,
+				  struct edid *edid)
+{
+	bool monitor_is_digital = !!(edid->input & DRM_EDID_INPUT_DIGITAL);
+	bool connector_is_digital = !!IS_DIGITAL(sdvo);
+
+	DRM_DEBUG_KMS("connector_is_digital? %d, monitor_is_digital? %d\n",
+		      connector_is_digital, monitor_is_digital);
+	return connector_is_digital == monitor_is_digital;
+}
+
 static enum drm_connector_status
 intel_sdvo_detect(struct drm_connector *connector, bool force)
 {
@@ -1358,10 +1375,12 @@
 		if (edid == NULL)
 			edid = intel_sdvo_get_analog_edid(connector);
 		if (edid != NULL) {
-			if (edid->input & DRM_EDID_INPUT_DIGITAL)
-				ret = connector_status_disconnected;
-			else
+			if (intel_sdvo_connector_matches_edid(intel_sdvo_connector,
+							      edid))
 				ret = connector_status_connected;
+			else
+				ret = connector_status_disconnected;
+
 			connector->display_info.raw_edid = NULL;
 			kfree(edid);
 		} else
@@ -1402,11 +1421,8 @@
 		edid = intel_sdvo_get_analog_edid(connector);
 
 	if (edid != NULL) {
-		struct intel_sdvo_connector *intel_sdvo_connector = to_intel_sdvo_connector(connector);
-		bool monitor_is_digital = !!(edid->input & DRM_EDID_INPUT_DIGITAL);
-		bool connector_is_digital = !!IS_TMDS(intel_sdvo_connector);
-
-		if (connector_is_digital == monitor_is_digital) {
+		if (intel_sdvo_connector_matches_edid(to_intel_sdvo_connector(connector),
+						      edid)) {
 			drm_mode_connector_update_edid_property(connector, edid);
 			drm_add_edid_modes(connector, edid);
 		}
diff --git a/drivers/gpu/drm/nouveau/Makefile b/drivers/gpu/drm/nouveau/Makefile
index 35ef5b1..9f27e3d 100644
--- a/drivers/gpu/drm/nouveau/Makefile
+++ b/drivers/gpu/drm/nouveau/Makefile
@@ -9,9 +9,9 @@
              nouveau_bo.o nouveau_fence.o nouveau_gem.o nouveau_ttm.o \
              nouveau_hw.o nouveau_calc.o nouveau_bios.o nouveau_i2c.o \
              nouveau_display.o nouveau_connector.o nouveau_fbcon.o \
-             nouveau_dp.o nouveau_ramht.o \
+             nouveau_hdmi.o nouveau_dp.o nouveau_ramht.o \
 	     nouveau_pm.o nouveau_volt.o nouveau_perf.o nouveau_temp.o \
-	     nouveau_mm.o nouveau_vm.o \
+	     nouveau_mm.o nouveau_vm.o nouveau_mxm.o nouveau_gpio.o \
              nv04_timer.o \
              nv04_mc.o nv40_mc.o nv50_mc.o \
              nv04_fb.o nv10_fb.o nv30_fb.o nv40_fb.o nv50_fb.o nvc0_fb.o \
@@ -19,9 +19,12 @@
              nv04_graph.o nv10_graph.o nv20_graph.o \
              nv40_graph.o nv50_graph.o nvc0_graph.o \
              nv40_grctx.o nv50_grctx.o nvc0_grctx.o \
-             nv84_crypt.o \
+             nv84_crypt.o nv98_crypt.o \
              nva3_copy.o nvc0_copy.o \
              nv31_mpeg.o nv50_mpeg.o \
+             nv84_bsp.o \
+             nv84_vp.o \
+             nv98_ppp.o \
              nv04_instmem.o nv50_instmem.o nvc0_instmem.o \
              nv04_dac.o nv04_dfp.o nv04_tv.o nv17_tv.o nv17_tv_modes.o \
              nv04_crtc.o nv04_display.o nv04_cursor.o \
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c
index 032a820..e5cbead 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.c
@@ -27,6 +27,7 @@
 #include "nouveau_drv.h"
 #include "nouveau_hw.h"
 #include "nouveau_encoder.h"
+#include "nouveau_gpio.h"
 
 #include <linux/io-mapping.h>
 
@@ -34,9 +35,6 @@
 #define NV_CIO_CRE_44_HEADA 0x0
 #define NV_CIO_CRE_44_HEADB 0x3
 #define FEATURE_MOBILE 0x10	/* also FEATURE_QUADRO for BMP */
-#define LEGACY_I2C_CRT 0x80
-#define LEGACY_I2C_PANEL 0x81
-#define LEGACY_I2C_TV 0x82
 
 #define EDID1_LEN 128
 
@@ -640,10 +638,9 @@
 nv50_pll_set(struct drm_device *dev, uint32_t reg, uint32_t clk)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	uint32_t reg0 = nv_rd32(dev, reg + 0);
-	uint32_t reg1 = nv_rd32(dev, reg + 4);
 	struct nouveau_pll_vals pll;
 	struct pll_lims pll_limits;
+	u32 ctrl, mask, coef;
 	int ret;
 
 	ret = get_pll_limits(dev, reg, &pll_limits);
@@ -654,15 +651,20 @@
 	if (!clk)
 		return -ERANGE;
 
-	reg0 = (reg0 & 0xfff8ffff) | (pll.log2P << 16);
-	reg1 = (reg1 & 0xffff0000) | (pll.N1 << 8) | pll.M1;
-
-	if (dev_priv->vbios.execute) {
-		still_alive();
-		nv_wr32(dev, reg + 4, reg1);
-		nv_wr32(dev, reg + 0, reg0);
+	coef = pll.N1 << 8 | pll.M1;
+	ctrl = pll.log2P << 16;
+	mask = 0x00070000;
+	if (reg == 0x004008) {
+		mask |= 0x01f80000;
+		ctrl |= (pll_limits.log2p_bias << 19);
+		ctrl |= (pll.log2P << 22);
 	}
 
+	if (!dev_priv->vbios.execute)
+		return 0;
+
+	nv_mask(dev, reg + 0, mask, ctrl);
+	nv_wr32(dev, reg + 4, coef);
 	return 0;
 }
 
@@ -719,115 +721,19 @@
 	return dcb_entry;
 }
 
-static int
-read_dcb_i2c_entry(struct drm_device *dev, int dcb_version, uint8_t *i2ctable, int index, struct dcb_i2c_entry *i2c)
-{
-	uint8_t dcb_i2c_ver = dcb_version, headerlen = 0, entry_len = 4;
-	int i2c_entries = DCB_MAX_NUM_I2C_ENTRIES;
-	int recordoffset = 0, rdofs = 1, wrofs = 0;
-	uint8_t port_type = 0;
-
-	if (!i2ctable)
-		return -EINVAL;
-
-	if (dcb_version >= 0x30) {
-		if (i2ctable[0] != dcb_version) /* necessary? */
-			NV_WARN(dev,
-				"DCB I2C table version mismatch (%02X vs %02X)\n",
-				i2ctable[0], dcb_version);
-		dcb_i2c_ver = i2ctable[0];
-		headerlen = i2ctable[1];
-		if (i2ctable[2] <= DCB_MAX_NUM_I2C_ENTRIES)
-			i2c_entries = i2ctable[2];
-		else
-			NV_WARN(dev,
-				"DCB I2C table has more entries than indexable "
-				"(%d entries, max %d)\n", i2ctable[2],
-				DCB_MAX_NUM_I2C_ENTRIES);
-		entry_len = i2ctable[3];
-		/* [4] is i2c_default_indices, read in parse_dcb_table() */
-	}
-	/*
-	 * It's your own fault if you call this function on a DCB 1.1 BIOS --
-	 * the test below is for DCB 1.2
-	 */
-	if (dcb_version < 0x14) {
-		recordoffset = 2;
-		rdofs = 0;
-		wrofs = 1;
-	}
-
-	if (index == 0xf)
-		return 0;
-	if (index >= i2c_entries) {
-		NV_ERROR(dev, "DCB I2C index too big (%d >= %d)\n",
-			 index, i2ctable[2]);
-		return -ENOENT;
-	}
-	if (i2ctable[headerlen + entry_len * index + 3] == 0xff) {
-		NV_ERROR(dev, "DCB I2C entry invalid\n");
-		return -EINVAL;
-	}
-
-	if (dcb_i2c_ver >= 0x30) {
-		port_type = i2ctable[headerlen + recordoffset + 3 + entry_len * index];
-
-		/*
-		 * Fixup for chips using same address offset for read and
-		 * write.
-		 */
-		if (port_type == 4)	/* seen on C51 */
-			rdofs = wrofs = 1;
-		if (port_type >= 5)	/* G80+ */
-			rdofs = wrofs = 0;
-	}
-
-	if (dcb_i2c_ver >= 0x40) {
-		if (port_type != 5 && port_type != 6)
-			NV_WARN(dev, "DCB I2C table has port type %d\n", port_type);
-
-		i2c->entry = ROM32(i2ctable[headerlen + recordoffset + entry_len * index]);
-	}
-
-	i2c->port_type = port_type;
-	i2c->read = i2ctable[headerlen + recordoffset + rdofs + entry_len * index];
-	i2c->write = i2ctable[headerlen + recordoffset + wrofs + entry_len * index];
-
-	return 0;
-}
-
 static struct nouveau_i2c_chan *
 init_i2c_device_find(struct drm_device *dev, int i2c_index)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct dcb_table *dcb = &dev_priv->vbios.dcb;
-
 	if (i2c_index == 0xff) {
+		struct drm_nouveau_private *dev_priv = dev->dev_private;
+		struct dcb_table *dcb = &dev_priv->vbios.dcb;
 		/* note: dcb_entry_idx_from_crtchead needs pre-script set-up */
-		int idx = dcb_entry_idx_from_crtchead(dev), shift = 0;
-		int default_indices = dcb->i2c_default_indices;
+		int idx = dcb_entry_idx_from_crtchead(dev);
 
+		i2c_index = NV_I2C_DEFAULT(0);
 		if (idx != 0x7f && dcb->entry[idx].i2c_upper_default)
-			shift = 4;
-
-		i2c_index = (default_indices >> shift) & 0xf;
+			i2c_index = NV_I2C_DEFAULT(1);
 	}
-	if (i2c_index == 0x80)	/* g80+ */
-		i2c_index = dcb->i2c_default_indices & 0xf;
-	else
-	if (i2c_index == 0x81)
-		i2c_index = (dcb->i2c_default_indices & 0xf0) >> 4;
-
-	if (i2c_index >= DCB_MAX_NUM_I2C_ENTRIES) {
-		NV_ERROR(dev, "invalid i2c_index 0x%x\n", i2c_index);
-		return NULL;
-	}
-
-	/* Make sure i2c table entry has been parsed, it may not
-	 * have been if this is a bus not referenced by a DCB encoder
-	 */
-	read_dcb_i2c_entry(dev, dcb->version, dcb->i2c_table,
-			   i2c_index, &dcb->i2c[i2c_index]);
 
 	return nouveau_i2c_find(dev, i2c_index);
 }
@@ -1195,13 +1101,9 @@
 
 	switch (cond) {
 	case 0:
-	{
-		struct dcb_connector_table_entry *ent =
-			&bios->dcb.connector.entry[dcb->connector];
-
-		if (ent->type != DCB_CONNECTOR_eDP)
+		entry = dcb_conn(dev, dcb->connector);
+		if (!entry || entry[0] != DCB_CONNECTOR_eDP)
 			iexec->execute = false;
-	}
 		break;
 	case 1:
 	case 2:
@@ -3223,49 +3125,6 @@
 	return 1;
 }
 
-static void
-init_gpio_unknv50(struct nvbios *bios, struct dcb_gpio_entry *gpio)
-{
-	const uint32_t nv50_gpio_ctl[2] = { 0xe100, 0xe28c };
-	u32 r, s, v;
-
-	/* Not a clue, needs de-magicing */
-	r = nv50_gpio_ctl[gpio->line >> 4];
-	s = (gpio->line & 0x0f);
-	v = bios_rd32(bios, r) & ~(0x00010001 << s);
-	switch ((gpio->entry & 0x06000000) >> 25) {
-	case 1:
-		v |= (0x00000001 << s);
-		break;
-	case 2:
-		v |= (0x00010000 << s);
-		break;
-	default:
-		break;
-	}
-
-	bios_wr32(bios, r, v);
-}
-
-static void
-init_gpio_unknvd0(struct nvbios *bios, struct dcb_gpio_entry *gpio)
-{
-	u32 v, i;
-
-	v  = bios_rd32(bios, 0x00d610 + (gpio->line * 4));
-	v &= 0xffffff00;
-	v |= (gpio->entry & 0x00ff0000) >> 16;
-	bios_wr32(bios, 0x00d610 + (gpio->line * 4), v);
-
-	i = (gpio->entry & 0x1f000000) >> 24;
-	if (i) {
-		v  = bios_rd32(bios, 0x00d640 + ((i - 1) * 4));
-		v &= 0xffffff00;
-		v |= gpio->line;
-		bios_wr32(bios, 0x00d640 + ((i - 1) * 4), v);
-	}
-}
-
 static int
 init_gpio(struct nvbios *bios, uint16_t offset, struct init_exec *iexec)
 {
@@ -3278,35 +3137,8 @@
 	 * each GPIO according to various values listed in each entry
 	 */
 
-	struct drm_nouveau_private *dev_priv = bios->dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
-	int i;
-
-	if (dev_priv->card_type < NV_50) {
-		NV_ERROR(bios->dev, "INIT_GPIO on unsupported chipset\n");
-		return 1;
-	}
-
-	if (!iexec->execute)
-		return 1;
-
-	for (i = 0; i < bios->dcb.gpio.entries; i++) {
-		struct dcb_gpio_entry *gpio = &bios->dcb.gpio.entry[i];
-
-		BIOSLOG(bios, "0x%04X: Entry: 0x%08X\n", offset, gpio->entry);
-
-		BIOSLOG(bios, "0x%04X: set gpio 0x%02x, state %d\n",
-			offset, gpio->tag, gpio->state_default);
-
-		if (!bios->execute)
-			continue;
-
-		pgpio->set(bios->dev, gpio->tag, gpio->state_default);
-		if (dev_priv->card_type < NV_D0)
-			init_gpio_unknv50(bios, gpio);
-		else
-			init_gpio_unknvd0(bios, gpio);
-	}
+	if (iexec->execute && bios->execute)
+		nouveau_gpio_reset(bios->dev);
 
 	return 1;
 }
@@ -4403,18 +4235,6 @@
 		break;
 	}
 
-	/* Dell Latitude D620 reports a too-high value for the dual-link
-	 * transition freq, causing us to program the panel incorrectly.
-	 *
-	 * It doesn't appear the VBIOS actually uses its transition freq
-	 * (90000kHz), instead it uses the "Number of LVDS channels" field
-	 * out of the panel ID structure (http://www.spwg.org/).
-	 *
-	 * For the moment, a quirk will do :)
-	 */
-	if (nv_match_device(dev, 0x01d7, 0x1028, 0x01c2))
-		bios->fp.duallink_transition_clk = 80000;
-
 	/* set dual_link flag for EDID case */
 	if (pxclk && (chip_version < 0x25 || chip_version > 0x28))
 		bios->fp.dual_link = (pxclk >= bios->fp.duallink_transition_clk);
@@ -4537,7 +4357,7 @@
 	NV_DEBUG_KMS(dev, "Searching for output entry for %d %d %d\n",
 			dcbent->type, dcbent->location, dcbent->or);
 	for (i = 0; i < table[3]; i++) {
-		otable = ROMPTR(bios, table[table[1] + (i * table[2])]);
+		otable = ROMPTR(dev, table[table[1] + (i * table[2])]);
 		if (otable && bios_encoder_match(dcbent, ROM32(otable[0])))
 			break;
 	}
@@ -4715,7 +4535,7 @@
 	{ PLL_CORE  , 0x004028 },
 	{ PLL_SHADER, 0x004020 },
 	{ PLL_MEMORY, 0x004008 },
-	{ PLL_UNK05 , 0x004030 },
+	{ PLL_VDEC  , 0x004030 },
 	{ PLL_UNK41 , 0x00e818 },
 	{ PLL_VPLL0 , 0x614100 },
 	{ PLL_VPLL1 , 0x614900 },
@@ -5481,6 +5301,9 @@
 	struct nvbios *bios = &dev_priv->vbios;
 	u8 entries, *entry;
 
+	if (bios->type != NVBIOS_BIT)
+		return -ENODEV;
+
 	entries = bios->data[bios->offset + 10];
 	entry   = &bios->data[bios->offset + 12];
 	while (entries--) {
@@ -5489,7 +5312,7 @@
 			bit->version = entry[1];
 			bit->length = ROM16(entry[2]);
 			bit->offset = ROM16(entry[4]);
-			bit->data = ROMPTR(bios, entry[4]);
+			bit->data = ROMPTR(dev, entry[4]);
 			return 0;
 		}
 
@@ -5594,10 +5417,6 @@
 	uint16_t legacy_scripts_offset, legacy_i2c_offset;
 
 	/* load needed defaults in case we can't parse this info */
-	bios->dcb.i2c[0].write = NV_CIO_CRE_DDC_WR__INDEX;
-	bios->dcb.i2c[0].read = NV_CIO_CRE_DDC_STATUS__INDEX;
-	bios->dcb.i2c[1].write = NV_CIO_CRE_DDC0_WR__INDEX;
-	bios->dcb.i2c[1].read = NV_CIO_CRE_DDC0_STATUS__INDEX;
 	bios->digital_min_front_porch = 0x4b;
 	bios->fmaxvco = 256000;
 	bios->fminvco = 128000;
@@ -5705,14 +5524,6 @@
 	bios->legacy.i2c_indices.crt = bios->data[legacy_i2c_offset];
 	bios->legacy.i2c_indices.tv = bios->data[legacy_i2c_offset + 1];
 	bios->legacy.i2c_indices.panel = bios->data[legacy_i2c_offset + 2];
-	if (bios->data[legacy_i2c_offset + 4])
-		bios->dcb.i2c[0].write = bios->data[legacy_i2c_offset + 4];
-	if (bios->data[legacy_i2c_offset + 5])
-		bios->dcb.i2c[0].read = bios->data[legacy_i2c_offset + 5];
-	if (bios->data[legacy_i2c_offset + 6])
-		bios->dcb.i2c[1].write = bios->data[legacy_i2c_offset + 6];
-	if (bios->data[legacy_i2c_offset + 7])
-		bios->dcb.i2c[1].read = bios->data[legacy_i2c_offset + 7];
 
 	if (bmplength > 74) {
 		bios->fmaxvco = ROM32(bmp[67]);
@@ -5763,286 +5574,128 @@
 	return 0;
 }
 
-static struct dcb_gpio_entry *
-new_gpio_entry(struct nvbios *bios)
+void *
+dcb_table(struct drm_device *dev)
 {
-	struct drm_device *dev = bios->dev;
-	struct dcb_gpio_table *gpio = &bios->dcb.gpio;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	u8 *dcb = NULL;
 
-	if (gpio->entries >= DCB_MAX_NUM_GPIO_ENTRIES) {
-		NV_ERROR(dev, "exceeded maximum number of gpio entries!!\n");
+	if (dev_priv->card_type > NV_04)
+		dcb = ROMPTR(dev, dev_priv->vbios.data[0x36]);
+	if (!dcb) {
+		NV_WARNONCE(dev, "No DCB data found in VBIOS\n");
 		return NULL;
 	}
 
-	return &gpio->entry[gpio->entries++];
+	if (dcb[0] >= 0x41) {
+		NV_WARNONCE(dev, "DCB version 0x%02x unknown\n", dcb[0]);
+		return NULL;
+	} else
+	if (dcb[0] >= 0x30) {
+		if (ROM32(dcb[6]) == 0x4edcbdcb)
+			return dcb;
+	} else
+	if (dcb[0] >= 0x20) {
+		if (ROM32(dcb[4]) == 0x4edcbdcb)
+			return dcb;
+	} else
+	if (dcb[0] >= 0x15) {
+		if (!memcmp(&dcb[-7], "DEV_REC", 7))
+			return dcb;
+	} else {
+		/*
+		 * v1.4 (some NV15/16, NV11+) seems the same as v1.5, but
+		 * always has the same single (crt) entry, even when tv-out
+		 * present, so the conclusion is this version cannot really
+		 * be used.
+		 *
+		 * v1.2 tables (some NV6/10, and NV15+) normally have the
+		 * same 5 entries, which are not specific to the card and so
+		 * no use.
+		 *
+		 * v1.2 does have an I2C table that read_dcb_i2c_table can
+		 * handle, but cards exist (nv11 in #14821) with a bad i2c
+		 * table pointer, so use the indices parsed in
+		 * parse_bmp_structure.
+		 *
+		 * v1.1 (NV5+, maybe some NV4) is entirely unhelpful
+		 */
+		NV_WARNONCE(dev, "No useful DCB data in VBIOS\n");
+		return NULL;
+	}
+
+	NV_WARNONCE(dev, "DCB header validation failed\n");
+	return NULL;
 }
 
-struct dcb_gpio_entry *
-nouveau_bios_gpio_entry(struct drm_device *dev, enum dcb_gpio_tag tag)
+void *
+dcb_outp(struct drm_device *dev, u8 idx)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nvbios *bios = &dev_priv->vbios;
-	int i;
-
-	for (i = 0; i < bios->dcb.gpio.entries; i++) {
-		if (bios->dcb.gpio.entry[i].tag != tag)
-			continue;
-
-		return &bios->dcb.gpio.entry[i];
+	u8 *dcb = dcb_table(dev);
+	if (dcb && dcb[0] >= 0x30) {
+		if (idx < dcb[2])
+			return dcb + dcb[1] + (idx * dcb[3]);
+	} else
+	if (dcb && dcb[0] >= 0x20) {
+		u8 *i2c = ROMPTR(dev, dcb[2]);
+		u8 *ent = dcb + 8 + (idx * 8);
+		if (i2c && ent < i2c)
+			return ent;
+	} else
+	if (dcb && dcb[0] >= 0x15) {
+		u8 *i2c = ROMPTR(dev, dcb[2]);
+		u8 *ent = dcb + 4 + (idx * 10);
+		if (i2c && ent < i2c)
+			return ent;
 	}
 
 	return NULL;
 }
 
-static void
-parse_dcb_gpio_table(struct nvbios *bios)
+int
+dcb_outp_foreach(struct drm_device *dev, void *data,
+		 int (*exec)(struct drm_device *, void *, int idx, u8 *outp))
 {
-	struct drm_device *dev = bios->dev;
-	struct dcb_gpio_entry *e;
-	u8 headerlen, entries, recordlen;
-	u8 *dcb, *gpio = NULL, *entry;
-	int i;
+	int ret, idx = -1;
+	u8 *outp = NULL;
+	while ((outp = dcb_outp(dev, ++idx))) {
+		if (ROM32(outp[0]) == 0x00000000)
+			break; /* seen on an NV11 with DCB v1.5 */
+		if (ROM32(outp[0]) == 0xffffffff)
+			break; /* seen on an NV17 with DCB v2.0 */
 
-	dcb = ROMPTR(bios, bios->data[0x36]);
-	if (dcb[0] >= 0x30) {
-		gpio = ROMPTR(bios, dcb[10]);
-		if (!gpio)
-			goto no_table;
-
-		headerlen = gpio[1];
-		entries   = gpio[2];
-		recordlen = gpio[3];
-	} else
-	if (dcb[0] >= 0x22 && dcb[-1] >= 0x13) {
-		gpio = ROMPTR(bios, dcb[-15]);
-		if (!gpio)
-			goto no_table;
-
-		headerlen = 3;
-		entries   = gpio[2];
-		recordlen = gpio[1];
-	} else
-	if (dcb[0] >= 0x22) {
-		/* No GPIO table present, parse the TVDAC GPIO data. */
-		uint8_t *tvdac_gpio = &dcb[-5];
-
-		if (tvdac_gpio[0] & 1) {
-			e = new_gpio_entry(bios);
-			e->tag = DCB_GPIO_TVDAC0;
-			e->line = tvdac_gpio[1] >> 4;
-			e->invert = tvdac_gpio[0] & 2;
-		}
-
-		goto no_table;
-	} else {
-		NV_DEBUG(dev, "no/unknown gpio table on DCB 0x%02x\n", dcb[0]);
-		goto no_table;
-	}
-
-	entry = gpio + headerlen;
-	for (i = 0; i < entries; i++, entry += recordlen) {
-		e = new_gpio_entry(bios);
-		if (!e)
-			break;
-
-		if (gpio[0] < 0x40) {
-			e->entry = ROM16(entry[0]);
-			e->tag = (e->entry & 0x07e0) >> 5;
-			if (e->tag == 0x3f) {
-				bios->dcb.gpio.entries--;
-				continue;
-			}
-
-			e->line = (e->entry & 0x001f);
-			e->invert = ((e->entry & 0xf800) >> 11) != 4;
-		} else {
-			e->entry = ROM32(entry[0]);
-			e->tag = (e->entry & 0x0000ff00) >> 8;
-			if (e->tag == 0xff) {
-				bios->dcb.gpio.entries--;
-				continue;
-			}
-
-			e->line = (e->entry & 0x0000001f) >> 0;
-			if (gpio[0] == 0x40) {
-				e->state_default = (e->entry & 0x01000000) >> 24;
-				e->state[0] = (e->entry & 0x18000000) >> 27;
-				e->state[1] = (e->entry & 0x60000000) >> 29;
-			} else {
-				e->state_default = (e->entry & 0x00000080) >> 7;
-				e->state[0] = (entry[4] >> 4) & 3;
-				e->state[1] = (entry[4] >> 6) & 3;
-			}
-		}
-	}
-
-no_table:
-	/* Apple iMac G4 NV18 */
-	if (nv_match_device(dev, 0x0189, 0x10de, 0x0010)) {
-		e = new_gpio_entry(bios);
-		if (e) {
-			e->tag = DCB_GPIO_TVDAC0;
-			e->line = 4;
-		}
-	}
-}
-
-struct dcb_connector_table_entry *
-nouveau_bios_connector_entry(struct drm_device *dev, int index)
-{
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nvbios *bios = &dev_priv->vbios;
-	struct dcb_connector_table_entry *cte;
-
-	if (index >= bios->dcb.connector.entries)
-		return NULL;
-
-	cte = &bios->dcb.connector.entry[index];
-	if (cte->type == 0xff)
-		return NULL;
-
-	return cte;
-}
-
-static enum dcb_connector_type
-divine_connector_type(struct nvbios *bios, int index)
-{
-	struct dcb_table *dcb = &bios->dcb;
-	unsigned encoders = 0, type = DCB_CONNECTOR_NONE;
-	int i;
-
-	for (i = 0; i < dcb->entries; i++) {
-		if (dcb->entry[i].connector == index)
-			encoders |= (1 << dcb->entry[i].type);
-	}
-
-	if (encoders & (1 << OUTPUT_DP)) {
-		if (encoders & (1 << OUTPUT_TMDS))
-			type = DCB_CONNECTOR_DP;
-		else
-			type = DCB_CONNECTOR_eDP;
-	} else
-	if (encoders & (1 << OUTPUT_TMDS)) {
-		if (encoders & (1 << OUTPUT_ANALOG))
-			type = DCB_CONNECTOR_DVI_I;
-		else
-			type = DCB_CONNECTOR_DVI_D;
-	} else
-	if (encoders & (1 << OUTPUT_ANALOG)) {
-		type = DCB_CONNECTOR_VGA;
-	} else
-	if (encoders & (1 << OUTPUT_LVDS)) {
-		type = DCB_CONNECTOR_LVDS;
-	} else
-	if (encoders & (1 << OUTPUT_TV)) {
-		type = DCB_CONNECTOR_TV_0;
-	}
-
-	return type;
-}
-
-static void
-apply_dcb_connector_quirks(struct nvbios *bios, int idx)
-{
-	struct dcb_connector_table_entry *cte = &bios->dcb.connector.entry[idx];
-	struct drm_device *dev = bios->dev;
-
-	/* Gigabyte NX85T */
-	if (nv_match_device(dev, 0x0421, 0x1458, 0x344c)) {
-		if (cte->type == DCB_CONNECTOR_HDMI_1)
-			cte->type = DCB_CONNECTOR_DVI_I;
-	}
-
-	/* Gigabyte GV-NX86T512H */
-	if (nv_match_device(dev, 0x0402, 0x1458, 0x3455)) {
-		if (cte->type == DCB_CONNECTOR_HDMI_1)
-			cte->type = DCB_CONNECTOR_DVI_I;
-	}
-}
-
-static const u8 hpd_gpio[16] = {
-	0xff, 0x07, 0x08, 0xff, 0xff, 0x51, 0x52, 0xff,
-	0xff, 0xff, 0xff, 0xff, 0xff, 0x5e, 0x5f, 0x60,
-};
-
-static void
-parse_dcb_connector_table(struct nvbios *bios)
-{
-	struct drm_device *dev = bios->dev;
-	struct dcb_connector_table *ct = &bios->dcb.connector;
-	struct dcb_connector_table_entry *cte;
-	uint8_t *conntab = &bios->data[bios->dcb.connector_table_ptr];
-	uint8_t *entry;
-	int i;
-
-	if (!bios->dcb.connector_table_ptr) {
-		NV_DEBUG_KMS(dev, "No DCB connector table present\n");
-		return;
-	}
-
-	NV_INFO(dev, "DCB connector table: VHER 0x%02x %d %d %d\n",
-		conntab[0], conntab[1], conntab[2], conntab[3]);
-	if ((conntab[0] != 0x30 && conntab[0] != 0x40) ||
-	    (conntab[3] != 2 && conntab[3] != 4)) {
-		NV_ERROR(dev, "  Unknown!  Please report.\n");
-		return;
-	}
-
-	ct->entries = conntab[2];
-
-	entry = conntab + conntab[1];
-	cte = &ct->entry[0];
-	for (i = 0; i < conntab[2]; i++, entry += conntab[3], cte++) {
-		cte->index = i;
-		if (conntab[3] == 2)
-			cte->entry = ROM16(entry[0]);
-		else
-			cte->entry = ROM32(entry[0]);
-
-		cte->type  = (cte->entry & 0x000000ff) >> 0;
-		cte->index2 = (cte->entry & 0x00000f00) >> 8;
-
-		cte->gpio_tag = ffs((cte->entry & 0x07033000) >> 12);
-		cte->gpio_tag = hpd_gpio[cte->gpio_tag];
-
-		if (cte->type == 0xff)
+		if ((outp[0] & 0x0f) == OUTPUT_UNUSED)
 			continue;
-
-		apply_dcb_connector_quirks(bios, i);
-
-		NV_INFO(dev, "  %d: 0x%08x: type 0x%02x idx %d tag 0x%02x\n",
-			i, cte->entry, cte->type, cte->index, cte->gpio_tag);
-
-		/* check for known types, fallback to guessing the type
-		 * from attached encoders if we hit an unknown.
-		 */
-		switch (cte->type) {
-		case DCB_CONNECTOR_VGA:
-		case DCB_CONNECTOR_TV_0:
-		case DCB_CONNECTOR_TV_1:
-		case DCB_CONNECTOR_TV_3:
-		case DCB_CONNECTOR_DVI_I:
-		case DCB_CONNECTOR_DVI_D:
-		case DCB_CONNECTOR_LVDS:
-		case DCB_CONNECTOR_LVDS_SPWG:
-		case DCB_CONNECTOR_DP:
-		case DCB_CONNECTOR_eDP:
-		case DCB_CONNECTOR_HDMI_0:
-		case DCB_CONNECTOR_HDMI_1:
+		if ((outp[0] & 0x0f) == OUTPUT_EOL)
 			break;
-		default:
-			cte->type = divine_connector_type(bios, cte->index);
-			NV_WARN(dev, "unknown type, using 0x%02x\n", cte->type);
-			break;
-		}
 
-		if (nouveau_override_conntype) {
-			int type = divine_connector_type(bios, cte->index);
-			if (type != cte->type)
-				NV_WARN(dev, " -> type 0x%02x\n", cte->type);
-		}
-
+		ret = exec(dev, data, idx, outp);
+		if (ret)
+			return ret;
 	}
+
+	return 0;
+}
+
+u8 *
+dcb_conntab(struct drm_device *dev)
+{
+	u8 *dcb = dcb_table(dev);
+	if (dcb && dcb[0] >= 0x30 && dcb[1] >= 0x16) {
+		u8 *conntab = ROMPTR(dev, dcb[0x14]);
+		if (conntab && conntab[0] >= 0x30 && conntab[0] <= 0x40)
+			return conntab;
+	}
+	return NULL;
+}
+
+u8 *
+dcb_conn(struct drm_device *dev, u8 idx)
+{
+	u8 *conntab = dcb_conntab(dev);
+	if (conntab && idx < conntab[2])
+		return conntab + conntab[1] + (idx * conntab[3]);
+	return NULL;
 }
 
 static struct dcb_entry *new_dcb_entry(struct dcb_table *dcb)
@@ -6075,8 +5728,7 @@
 	entry->type = conn & 0xf;
 	entry->i2c_index = (conn >> 4) & 0xf;
 	entry->heads = (conn >> 8) & 0xf;
-	if (dcb->version >= 0x40)
-		entry->connector = (conn >> 12) & 0xf;
+	entry->connector = (conn >> 12) & 0xf;
 	entry->bus = (conn >> 16) & 0xf;
 	entry->location = (conn >> 20) & 0x3;
 	entry->or = (conn >> 24) & 0xf;
@@ -6248,25 +5900,6 @@
 	return true;
 }
 
-static bool parse_dcb_entry(struct drm_device *dev, struct dcb_table *dcb,
-			    uint32_t conn, uint32_t conf)
-{
-	struct dcb_entry *entry = new_dcb_entry(dcb);
-	bool ret;
-
-	if (dcb->version >= 0x20)
-		ret = parse_dcb20_entry(dev, dcb, conn, conf, entry);
-	else
-		ret = parse_dcb15_entry(dev, dcb, conn, conf, entry);
-	if (!ret)
-		return ret;
-
-	read_dcb_i2c_entry(dev, dcb->version, dcb->i2c_table,
-			   entry->i2c_index, &dcb->i2c[entry->i2c_index]);
-
-	return true;
-}
-
 static
 void merge_like_dcb_entries(struct drm_device *dev, struct dcb_table *dcb)
 {
@@ -6427,154 +6060,118 @@
 #endif
 
 	/* Make up some sane defaults */
-	fabricate_dcb_output(dcb, OUTPUT_ANALOG, LEGACY_I2C_CRT, 1, 1);
+	fabricate_dcb_output(dcb, OUTPUT_ANALOG,
+			     bios->legacy.i2c_indices.crt, 1, 1);
 
 	if (nv04_tv_identify(dev, bios->legacy.i2c_indices.tv) >= 0)
-		fabricate_dcb_output(dcb, OUTPUT_TV, LEGACY_I2C_TV,
+		fabricate_dcb_output(dcb, OUTPUT_TV,
+				     bios->legacy.i2c_indices.tv,
 				     all_heads, 0);
 
 	else if (bios->tmds.output0_script_ptr ||
 		 bios->tmds.output1_script_ptr)
-		fabricate_dcb_output(dcb, OUTPUT_TMDS, LEGACY_I2C_PANEL,
+		fabricate_dcb_output(dcb, OUTPUT_TMDS,
+				     bios->legacy.i2c_indices.panel,
 				     all_heads, 1);
 }
 
 static int
+parse_dcb_entry(struct drm_device *dev, void *data, int idx, u8 *outp)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct dcb_table *dcb = &dev_priv->vbios.dcb;
+	u32 conf = (dcb->version >= 0x20) ? ROM32(outp[4]) : ROM32(outp[6]);
+	u32 conn = ROM32(outp[0]);
+	bool ret;
+
+	if (apply_dcb_encoder_quirks(dev, idx, &conn, &conf)) {
+		struct dcb_entry *entry = new_dcb_entry(dcb);
+
+		NV_TRACEWARN(dev, "DCB outp %02d: %08x %08x\n", idx, conn, conf);
+
+		if (dcb->version >= 0x20)
+			ret = parse_dcb20_entry(dev, dcb, conn, conf, entry);
+		else
+			ret = parse_dcb15_entry(dev, dcb, conn, conf, entry);
+		if (!ret)
+			return 1; /* stop parsing */
+
+		/* Ignore the I2C index for on-chip TV-out, as there
+		 * are cards with bogus values (nv31m in bug 23212),
+		 * and it's otherwise useless.
+		 */
+		if (entry->type == OUTPUT_TV &&
+		    entry->location == DCB_LOC_ON_CHIP)
+			entry->i2c_index = 0x0f;
+	}
+
+	return 0;
+}
+
+static void
+dcb_fake_connectors(struct nvbios *bios)
+{
+	struct dcb_table *dcbt = &bios->dcb;
+	u8 map[16] = { };
+	int i, idx = 0;
+
+	/* heuristic: if we ever get a non-zero connector field, assume
+	 * that all the indices are valid and we don't need fake them.
+	 */
+	for (i = 0; i < dcbt->entries; i++) {
+		if (dcbt->entry[i].connector)
+			return;
+	}
+
+	/* no useful connector info available, we need to make it up
+	 * ourselves.  the rule here is: anything on the same i2c bus
+	 * is considered to be on the same connector.  any output
+	 * without an associated i2c bus is assigned its own unique
+	 * connector index.
+	 */
+	for (i = 0; i < dcbt->entries; i++) {
+		u8 i2c = dcbt->entry[i].i2c_index;
+		if (i2c == 0x0f) {
+			dcbt->entry[i].connector = idx++;
+		} else {
+			if (!map[i2c])
+				map[i2c] = ++idx;
+			dcbt->entry[i].connector = map[i2c] - 1;
+		}
+	}
+
+	/* if we created more than one connector, destroy the connector
+	 * table - just in case it has random, rather than stub, entries.
+	 */
+	if (i > 1) {
+		u8 *conntab = dcb_conntab(bios->dev);
+		if (conntab)
+			conntab[0] = 0x00;
+	}
+}
+
+static int
 parse_dcb_table(struct drm_device *dev, struct nvbios *bios)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct dcb_table *dcb = &bios->dcb;
-	uint16_t dcbptr = 0, i2ctabptr = 0;
-	uint8_t *dcbtable;
-	uint8_t headerlen = 0x4, entries = DCB_MAX_NUM_ENTRIES;
-	bool configblock = true;
-	int recordlength = 8, confofs = 4;
-	int i;
+	u8 *dcbt, *conn;
+	int idx;
 
-	/* get the offset from 0x36 */
-	if (dev_priv->card_type > NV_04) {
-		dcbptr = ROM16(bios->data[0x36]);
-		if (dcbptr == 0x0000)
-			NV_WARN(dev, "No output data (DCB) found in BIOS\n");
-	}
-
-	/* this situation likely means a really old card, pre DCB */
-	if (dcbptr == 0x0) {
-		fabricate_dcb_encoder_table(dev, bios);
-		return 0;
-	}
-
-	dcbtable = &bios->data[dcbptr];
-
-	/* get DCB version */
-	dcb->version = dcbtable[0];
-	NV_TRACE(dev, "Found Display Configuration Block version %d.%d\n",
-		 dcb->version >> 4, dcb->version & 0xf);
-
-	if (dcb->version >= 0x20) { /* NV17+ */
-		uint32_t sig;
-
-		if (dcb->version >= 0x30) { /* NV40+ */
-			headerlen = dcbtable[1];
-			entries = dcbtable[2];
-			recordlength = dcbtable[3];
-			i2ctabptr = ROM16(dcbtable[4]);
-			sig = ROM32(dcbtable[6]);
-			dcb->gpio_table_ptr = ROM16(dcbtable[10]);
-			dcb->connector_table_ptr = ROM16(dcbtable[20]);
-		} else {
-			i2ctabptr = ROM16(dcbtable[2]);
-			sig = ROM32(dcbtable[4]);
-			headerlen = 8;
+	dcbt = dcb_table(dev);
+	if (!dcbt) {
+		/* handle pre-DCB boards */
+		if (bios->type == NVBIOS_BMP) {
+			fabricate_dcb_encoder_table(dev, bios);
+			return 0;
 		}
 
-		if (sig != 0x4edcbdcb) {
-			NV_ERROR(dev, "Bad Display Configuration Block "
-					"signature (%08X)\n", sig);
-			return -EINVAL;
-		}
-	} else if (dcb->version >= 0x15) { /* some NV11 and NV20 */
-		char sig[8] = { 0 };
-
-		strncpy(sig, (char *)&dcbtable[-7], 7);
-		i2ctabptr = ROM16(dcbtable[2]);
-		recordlength = 10;
-		confofs = 6;
-
-		if (strcmp(sig, "DEV_REC")) {
-			NV_ERROR(dev, "Bad Display Configuration Block "
-					"signature (%s)\n", sig);
-			return -EINVAL;
-		}
-	} else {
-		/*
-		 * v1.4 (some NV15/16, NV11+) seems the same as v1.5, but always
-		 * has the same single (crt) entry, even when tv-out present, so
-		 * the conclusion is this version cannot really be used.
-		 * v1.2 tables (some NV6/10, and NV15+) normally have the same
-		 * 5 entries, which are not specific to the card and so no use.
-		 * v1.2 does have an I2C table that read_dcb_i2c_table can
-		 * handle, but cards exist (nv11 in #14821) with a bad i2c table
-		 * pointer, so use the indices parsed in parse_bmp_structure.
-		 * v1.1 (NV5+, maybe some NV4) is entirely unhelpful
-		 */
-		NV_TRACEWARN(dev, "No useful information in BIOS output table; "
-				  "adding all possible outputs\n");
-		fabricate_dcb_encoder_table(dev, bios);
-		return 0;
+		return -EINVAL;
 	}
 
-	if (!i2ctabptr)
-		NV_WARN(dev, "No pointer to DCB I2C port table\n");
-	else {
-		dcb->i2c_table = &bios->data[i2ctabptr];
-		if (dcb->version >= 0x30)
-			dcb->i2c_default_indices = dcb->i2c_table[4];
+	NV_TRACE(dev, "DCB version %d.%d\n", dcbt[0] >> 4, dcbt[0] & 0xf);
 
-		/*
-		 * Parse the "management" I2C bus, used for hardware
-		 * monitoring and some external TMDS transmitters.
-		 */
-		if (dcb->version >= 0x22) {
-			int idx = (dcb->version >= 0x40 ?
-				   dcb->i2c_default_indices & 0xf :
-				   2);
-
-			read_dcb_i2c_entry(dev, dcb->version, dcb->i2c_table,
-					   idx, &dcb->i2c[idx]);
-		}
-	}
-
-	if (entries > DCB_MAX_NUM_ENTRIES)
-		entries = DCB_MAX_NUM_ENTRIES;
-
-	for (i = 0; i < entries; i++) {
-		uint32_t connection, config = 0;
-
-		connection = ROM32(dcbtable[headerlen + recordlength * i]);
-		if (configblock)
-			config = ROM32(dcbtable[headerlen + confofs + recordlength * i]);
-
-		/* seen on an NV11 with DCB v1.5 */
-		if (connection == 0x00000000)
-			break;
-
-		/* seen on an NV17 with DCB v2.0 */
-		if (connection == 0xffffffff)
-			break;
-
-		if ((connection & 0x0000000f) == 0x0000000f)
-			continue;
-
-		if (!apply_dcb_encoder_quirks(dev, i, &connection, &config))
-			continue;
-
-		NV_TRACEWARN(dev, "Raw DCB entry %d: %08x %08x\n",
-			     dcb->entries, connection, config);
-
-		if (!parse_dcb_entry(dev, dcb, connection, config))
-			break;
-	}
+	dcb->version = dcbt[0];
+	dcb_outp_foreach(dev, NULL, parse_dcb_entry);
 
 	/*
 	 * apart for v2.1+ not being known for requiring merging, this
@@ -6586,77 +6183,19 @@
 	if (!dcb->entries)
 		return -ENXIO;
 
-	parse_dcb_gpio_table(bios);
-	parse_dcb_connector_table(bios);
-	return 0;
-}
-
-static void
-fixup_legacy_connector(struct nvbios *bios)
-{
-	struct dcb_table *dcb = &bios->dcb;
-	int i, i2c, i2c_conn[DCB_MAX_NUM_I2C_ENTRIES] = { };
-
-	/*
-	 * DCB 3.0 also has the table in most cases, but there are some cards
-	 * where the table is filled with stub entries, and the DCB entriy
-	 * indices are all 0.  We don't need the connector indices on pre-G80
-	 * chips (yet?) so limit the use to DCB 4.0 and above.
-	 */
-	if (dcb->version >= 0x40)
-		return;
-
-	dcb->connector.entries = 0;
-
-	/*
-	 * No known connector info before v3.0, so make it up.  the rule here
-	 * is: anything on the same i2c bus is considered to be on the same
-	 * connector.  any output without an associated i2c bus is assigned
-	 * its own unique connector index.
-	 */
-	for (i = 0; i < dcb->entries; i++) {
-		/*
-		 * Ignore the I2C index for on-chip TV-out, as there
-		 * are cards with bogus values (nv31m in bug 23212),
-		 * and it's otherwise useless.
-		 */
-		if (dcb->entry[i].type == OUTPUT_TV &&
-		    dcb->entry[i].location == DCB_LOC_ON_CHIP)
-			dcb->entry[i].i2c_index = 0xf;
-		i2c = dcb->entry[i].i2c_index;
-
-		if (i2c_conn[i2c]) {
-			dcb->entry[i].connector = i2c_conn[i2c] - 1;
-			continue;
+	/* dump connector table entries to log, if any exist */
+	idx = -1;
+	while ((conn = dcb_conn(dev, ++idx))) {
+		if (conn[0] != 0xff) {
+			NV_TRACE(dev, "DCB conn %02d: ", idx);
+			if (dcb_conntab(dev)[3] < 4)
+				printk("%04x\n", ROM16(conn[0]));
+			else
+				printk("%08x\n", ROM32(conn[0]));
 		}
-
-		dcb->entry[i].connector = dcb->connector.entries++;
-		if (i2c != 0xf)
-			i2c_conn[i2c] = dcb->connector.entries;
 	}
-
-	/* Fake the connector table as well as just connector indices */
-	for (i = 0; i < dcb->connector.entries; i++) {
-		dcb->connector.entry[i].index = i;
-		dcb->connector.entry[i].type = divine_connector_type(bios, i);
-		dcb->connector.entry[i].gpio_tag = 0xff;
-	}
-}
-
-static void
-fixup_legacy_i2c(struct nvbios *bios)
-{
-	struct dcb_table *dcb = &bios->dcb;
-	int i;
-
-	for (i = 0; i < dcb->entries; i++) {
-		if (dcb->entry[i].i2c_index == LEGACY_I2C_CRT)
-			dcb->entry[i].i2c_index = bios->legacy.i2c_indices.crt;
-		if (dcb->entry[i].i2c_index == LEGACY_I2C_PANEL)
-			dcb->entry[i].i2c_index = bios->legacy.i2c_indices.panel;
-		if (dcb->entry[i].i2c_index == LEGACY_I2C_TV)
-			dcb->entry[i].i2c_index = bios->legacy.i2c_indices.tv;
-	}
+	dcb_fake_connectors(bios);
+	return 0;
 }
 
 static int load_nv17_hwsq_ucode_entry(struct drm_device *dev, struct nvbios *bios, uint16_t hwsq_offset, int entry)
@@ -6875,19 +6414,6 @@
 	return ret;
 }
 
-static void
-nouveau_bios_i2c_devices_takedown(struct drm_device *dev)
-{
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nvbios *bios = &dev_priv->vbios;
-	struct dcb_i2c_entry *entry;
-	int i;
-
-	entry = &bios->dcb.i2c[0];
-	for (i = 0; i < DCB_MAX_NUM_I2C_ENTRIES; i++, entry++)
-		nouveau_i2c_fini(dev, entry);
-}
-
 static bool
 nouveau_bios_posted(struct drm_device *dev)
 {
@@ -6924,12 +6450,17 @@
 	if (ret)
 		return ret;
 
-	ret = parse_dcb_table(dev, bios);
+	ret = nouveau_i2c_init(dev);
 	if (ret)
 		return ret;
 
-	fixup_legacy_i2c(bios);
-	fixup_legacy_connector(bios);
+	ret = nouveau_mxm_init(dev);
+	if (ret)
+		return ret;
+
+	ret = parse_dcb_table(dev, bios);
+	if (ret)
+		return ret;
 
 	if (!bios->major_version)	/* we don't run version 0 bios */
 		return 0;
@@ -6967,5 +6498,6 @@
 void
 nouveau_bios_takedown(struct drm_device *dev)
 {
-	nouveau_bios_i2c_devices_takedown(dev);
+	nouveau_mxm_fini(dev);
+	nouveau_i2c_fini(dev);
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.h b/drivers/gpu/drm/nouveau/nouveau_bios.h
index 8adb69e..1e382ad 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.h
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.h
@@ -34,9 +34,14 @@
 
 #define DCB_LOC_ON_CHIP 0
 
-#define ROM16(x) le16_to_cpu(*(uint16_t *)&(x))
-#define ROM32(x) le32_to_cpu(*(uint32_t *)&(x))
-#define ROMPTR(bios, x) (ROM16(x) ? &(bios)->data[ROM16(x)] : NULL)
+#define ROM16(x) le16_to_cpu(*(u16 *)&(x))
+#define ROM32(x) le32_to_cpu(*(u32 *)&(x))
+#define ROM48(x) ({ u8 *p = &(x); (u64)ROM16(p[4]) << 32 | ROM32(p[0]); })
+#define ROM64(x) le64_to_cpu(*(u64 *)&(x))
+#define ROMPTR(d,x) ({            \
+	struct drm_nouveau_private *dev_priv = (d)->dev_private; \
+	ROM16(x) ? &dev_priv->vbios.data[ROM16(x)] : NULL; \
+})
 
 struct bit_entry {
 	uint8_t  id;
@@ -48,30 +53,12 @@
 
 int bit_table(struct drm_device *, u8 id, struct bit_entry *);
 
-struct dcb_i2c_entry {
-	uint32_t entry;
-	uint8_t port_type;
-	uint8_t read, write;
-	struct nouveau_i2c_chan *chan;
-};
-
 enum dcb_gpio_tag {
 	DCB_GPIO_TVDAC0 = 0xc,
 	DCB_GPIO_TVDAC1 = 0x2d,
-};
-
-struct dcb_gpio_entry {
-	enum dcb_gpio_tag tag;
-	int line;
-	bool invert;
-	uint32_t entry;
-	uint8_t state_default;
-	uint8_t state[2];
-};
-
-struct dcb_gpio_table {
-	int entries;
-	struct dcb_gpio_entry entry[DCB_MAX_NUM_GPIO_ENTRIES];
+	DCB_GPIO_PWM_FAN = 0x9,
+	DCB_GPIO_FAN_SENSE = 0x3d,
+	DCB_GPIO_UNUSED = 0xff
 };
 
 enum dcb_connector_type {
@@ -90,20 +77,6 @@
 	DCB_CONNECTOR_NONE = 0xff
 };
 
-struct dcb_connector_table_entry {
-	uint8_t index;
-	uint32_t entry;
-	enum dcb_connector_type type;
-	uint8_t index2;
-	uint8_t gpio_tag;
-	void *drm;
-};
-
-struct dcb_connector_table {
-	int entries;
-	struct dcb_connector_table_entry entry[DCB_MAX_NUM_CONNECTOR_ENTRIES];
-};
-
 enum dcb_type {
 	OUTPUT_ANALOG = 0,
 	OUTPUT_TV = 1,
@@ -111,6 +84,7 @@
 	OUTPUT_LVDS = 3,
 	OUTPUT_DP = 6,
 	OUTPUT_EOL = 14, /* DCB 4.0+, appears to be end-of-list */
+	OUTPUT_UNUSED = 15,
 	OUTPUT_ANY = -1
 };
 
@@ -155,18 +129,8 @@
 
 struct dcb_table {
 	uint8_t version;
-
 	int entries;
 	struct dcb_entry entry[DCB_MAX_NUM_ENTRIES];
-
-	uint8_t *i2c_table;
-	uint8_t i2c_default_indices;
-	struct dcb_i2c_entry i2c[DCB_MAX_NUM_I2C_ENTRIES];
-
-	uint16_t gpio_table_ptr;
-	struct dcb_gpio_table gpio;
-	uint16_t connector_table_ptr;
-	struct dcb_connector_table connector;
 };
 
 enum nouveau_or {
@@ -195,7 +159,7 @@
 	PLL_SHADER = 0x02,
 	PLL_UNK03  = 0x03,
 	PLL_MEMORY = 0x04,
-	PLL_UNK05  = 0x05,
+	PLL_VDEC   = 0x05,
 	PLL_UNK40  = 0x40,
 	PLL_UNK41  = 0x41,
 	PLL_UNK42  = 0x42,
@@ -333,4 +297,11 @@
 	} legacy;
 };
 
+void *dcb_table(struct drm_device *);
+void *dcb_outp(struct drm_device *, u8 idx);
+int dcb_outp_foreach(struct drm_device *, void *data,
+		     int (*)(struct drm_device *, void *, int idx, u8 *outp));
+u8 *dcb_conntab(struct drm_device *);
+u8 *dcb_conn(struct drm_device *, u8 idx);
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index b1b33a1..8cf4a48 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -153,7 +153,7 @@
 
 	if (dev_priv->card_type == NV_10 &&
 	    nvbo->tile_mode && (type & TTM_PL_FLAG_VRAM) &&
-	    nvbo->bo.mem.num_pages < vram_pages / 2) {
+	    nvbo->bo.mem.num_pages < vram_pages / 4) {
 		/*
 		 * Make sure that the color and depth buffers are handled
 		 * by independent memory controller units. Up to a 9x
@@ -682,8 +682,7 @@
 	if (mem->mem_type == TTM_PL_VRAM)
 		nouveau_vm_map(vma, node);
 	else
-		nouveau_vm_map_sg(vma, 0, mem->num_pages << PAGE_SHIFT,
-				  node, node->pages);
+		nouveau_vm_map_sg(vma, 0, mem->num_pages << PAGE_SHIFT, node);
 
 	return 0;
 }
@@ -810,7 +809,6 @@
 static void
 nouveau_bo_move_ntfy(struct ttm_buffer_object *bo, struct ttm_mem_reg *new_mem)
 {
-	struct nouveau_mem *node = new_mem->mm_node;
 	struct nouveau_bo *nvbo = nouveau_bo(bo);
 	struct nouveau_vma *vma;
 
@@ -822,7 +820,7 @@
 		    nvbo->page_shift == vma->vm->spg_shift) {
 			nouveau_vm_map_sg(vma, 0, new_mem->
 					  num_pages << PAGE_SHIFT,
-					  node, node->pages);
+					  new_mem->mm_node);
 		} else {
 			nouveau_vm_unmap(vma);
 		}
@@ -1173,7 +1171,7 @@
 		nouveau_vm_map(vma, nvbo->bo.mem.mm_node);
 	else
 	if (nvbo->bo.mem.mem_type == TTM_PL_TT)
-		nouveau_vm_map_sg(vma, 0, size, node, node->pages);
+		nouveau_vm_map_sg(vma, 0, size, node);
 
 	list_add_tail(&vma->head, &nvbo->vma_list);
 	vma->refcount = 1;
diff --git a/drivers/gpu/drm/nouveau/nouveau_channel.c b/drivers/gpu/drm/nouveau/nouveau_channel.c
index a319d564..a018def 100644
--- a/drivers/gpu/drm/nouveau/nouveau_channel.c
+++ b/drivers/gpu/drm/nouveau/nouveau_channel.c
@@ -158,6 +158,7 @@
 	INIT_LIST_HEAD(&chan->nvsw.vbl_wait);
 	INIT_LIST_HEAD(&chan->nvsw.flip);
 	INIT_LIST_HEAD(&chan->fence.pending);
+	spin_lock_init(&chan->fence.lock);
 
 	/* setup channel's memory and vm */
 	ret = nouveau_gpuobj_channel_init(chan, vram_handle, gart_handle);
@@ -186,6 +187,8 @@
 	nouveau_dma_pre_init(chan);
 	chan->user_put = 0x40;
 	chan->user_get = 0x44;
+	if (dev_priv->card_type >= NV_50)
+                chan->user_get_hi = 0x60;
 
 	/* disable the fifo caches */
 	pfifo->reassign(dev, false);
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index e0d275e..f3ce34b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -35,6 +35,7 @@
 #include "nouveau_encoder.h"
 #include "nouveau_crtc.h"
 #include "nouveau_connector.h"
+#include "nouveau_gpio.h"
 #include "nouveau_hw.h"
 
 static void nouveau_connector_hotplug(void *, int);
@@ -78,29 +79,11 @@
 	return NULL;
 }
 
-/*TODO: This could use improvement, and learn to handle the fixed
- *      BIOS tables etc.  It's fine currently, for its only user.
- */
-int
-nouveau_connector_bpp(struct drm_connector *connector)
-{
-	struct nouveau_connector *nv_connector = nouveau_connector(connector);
-
-	if (nv_connector->edid && nv_connector->edid->revision >= 4) {
-		u8 bpc = ((nv_connector->edid->input & 0x70) >> 3) + 4;
-		if (bpc > 4)
-			return bpc;
-	}
-
-	return 18;
-}
-
 static void
 nouveau_connector_destroy(struct drm_connector *connector)
 {
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
 	struct drm_nouveau_private *dev_priv;
-	struct nouveau_gpio_engine *pgpio;
 	struct drm_device *dev;
 
 	if (!nv_connector)
@@ -110,10 +93,9 @@
 	dev_priv = dev->dev_private;
 	NV_DEBUG_KMS(dev, "\n");
 
-	pgpio = &dev_priv->engine.gpio;
-	if (pgpio->irq_unregister) {
-		pgpio->irq_unregister(dev, nv_connector->dcb->gpio_tag,
-				      nouveau_connector_hotplug, connector);
+	if (nv_connector->hpd != DCB_GPIO_UNUSED) {
+		nouveau_gpio_isr_del(dev, 0, nv_connector->hpd, 0xff,
+				     nouveau_connector_hotplug, connector);
 	}
 
 	kfree(nv_connector->edid);
@@ -198,6 +180,10 @@
 		return;
 	nv_connector->detected_encoder = nv_encoder;
 
+	if (dev_priv->card_type >= NV_50) {
+		connector->interlace_allowed = true;
+		connector->doublescan_allowed = true;
+	} else
 	if (nv_encoder->dcb->type == OUTPUT_LVDS ||
 	    nv_encoder->dcb->type == OUTPUT_TMDS) {
 		connector->doublescan_allowed = false;
@@ -214,7 +200,7 @@
 			connector->interlace_allowed = true;
 	}
 
-	if (nv_connector->dcb->type == DCB_CONNECTOR_DVI_I) {
+	if (nv_connector->type == DCB_CONNECTOR_DVI_I) {
 		drm_connector_property_set_value(connector,
 			dev->mode_config.dvi_i_subconnector_property,
 			nv_encoder->dcb->type == OUTPUT_TMDS ?
@@ -397,7 +383,7 @@
 	struct nouveau_encoder *nv_encoder;
 	int type;
 
-	if (nv_connector->dcb->type == DCB_CONNECTOR_DVI_I) {
+	if (nv_connector->type == DCB_CONNECTOR_DVI_I) {
 		if (connector->force == DRM_FORCE_ON_DIGITAL)
 			type = OUTPUT_TMDS;
 		else
@@ -420,15 +406,21 @@
 nouveau_connector_set_property(struct drm_connector *connector,
 			       struct drm_property *property, uint64_t value)
 {
+	struct drm_nouveau_private *dev_priv = connector->dev->dev_private;
+	struct nouveau_display_engine *disp = &dev_priv->engine.display;
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
 	struct nouveau_encoder *nv_encoder = nv_connector->detected_encoder;
 	struct drm_encoder *encoder = to_drm_encoder(nv_encoder);
 	struct drm_device *dev = connector->dev;
+	struct nouveau_crtc *nv_crtc;
 	int ret;
 
+	nv_crtc = NULL;
+	if (connector->encoder && connector->encoder->crtc)
+		nv_crtc = nouveau_crtc(connector->encoder->crtc);
+
 	/* Scaling mode */
 	if (property == dev->mode_config.scaling_mode_property) {
-		struct nouveau_crtc *nv_crtc = NULL;
 		bool modeset = false;
 
 		switch (value) {
@@ -454,8 +446,6 @@
 			modeset = true;
 		nv_connector->scaling_mode = value;
 
-		if (connector->encoder && connector->encoder->crtc)
-			nv_crtc = nouveau_crtc(connector->encoder->crtc);
 		if (!nv_crtc)
 			return 0;
 
@@ -467,7 +457,7 @@
 			if (!ret)
 				return -EINVAL;
 		} else {
-			ret = nv_crtc->set_scale(nv_crtc, value, true);
+			ret = nv_crtc->set_scale(nv_crtc, true);
 			if (ret)
 				return ret;
 		}
@@ -475,23 +465,58 @@
 		return 0;
 	}
 
+	/* Underscan */
+	if (property == disp->underscan_property) {
+		if (nv_connector->underscan != value) {
+			nv_connector->underscan = value;
+			if (!nv_crtc || !nv_crtc->set_scale)
+				return 0;
+
+			return nv_crtc->set_scale(nv_crtc, true);
+		}
+
+		return 0;
+	}
+
+	if (property == disp->underscan_hborder_property) {
+		if (nv_connector->underscan_hborder != value) {
+			nv_connector->underscan_hborder = value;
+			if (!nv_crtc || !nv_crtc->set_scale)
+				return 0;
+
+			return nv_crtc->set_scale(nv_crtc, true);
+		}
+
+		return 0;
+	}
+
+	if (property == disp->underscan_vborder_property) {
+		if (nv_connector->underscan_vborder != value) {
+			nv_connector->underscan_vborder = value;
+			if (!nv_crtc || !nv_crtc->set_scale)
+				return 0;
+
+			return nv_crtc->set_scale(nv_crtc, true);
+		}
+
+		return 0;
+	}
+
 	/* Dithering */
-	if (property == dev->mode_config.dithering_mode_property) {
-		struct nouveau_crtc *nv_crtc = NULL;
-
-		if (value == DRM_MODE_DITHERING_ON)
-			nv_connector->use_dithering = true;
-		else
-			nv_connector->use_dithering = false;
-
-		if (connector->encoder && connector->encoder->crtc)
-			nv_crtc = nouveau_crtc(connector->encoder->crtc);
-
+	if (property == disp->dithering_mode) {
+		nv_connector->dithering_mode = value;
 		if (!nv_crtc || !nv_crtc->set_dither)
 			return 0;
 
-		return nv_crtc->set_dither(nv_crtc, nv_connector->use_dithering,
-					   true);
+		return nv_crtc->set_dither(nv_crtc, true);
+	}
+
+	if (property == disp->dithering_depth) {
+		nv_connector->dithering_depth = value;
+		if (!nv_crtc || !nv_crtc->set_dither)
+			return 0;
+
+		return nv_crtc->set_dither(nv_crtc, true);
 	}
 
 	if (nv_encoder && nv_encoder->dcb->type == OUTPUT_TV)
@@ -602,6 +627,46 @@
 	return modes;
 }
 
+static void
+nouveau_connector_detect_depth(struct drm_connector *connector)
+{
+	struct drm_nouveau_private *dev_priv = connector->dev->dev_private;
+	struct nouveau_connector *nv_connector = nouveau_connector(connector);
+	struct nouveau_encoder *nv_encoder = nv_connector->detected_encoder;
+	struct nvbios *bios = &dev_priv->vbios;
+	struct drm_display_mode *mode = nv_connector->native_mode;
+	bool duallink;
+
+	/* if the edid is feeling nice enough to provide this info, use it */
+	if (nv_connector->edid && connector->display_info.bpc)
+		return;
+
+	/* if not, we're out of options unless we're LVDS, default to 6bpc */
+	connector->display_info.bpc = 6;
+	if (nv_encoder->dcb->type != OUTPUT_LVDS)
+		return;
+
+	/* LVDS: panel straps */
+	if (bios->fp_no_ddc) {
+		if (bios->fp.if_is_24bit)
+			connector->display_info.bpc = 8;
+		return;
+	}
+
+	/* LVDS: DDC panel, need to first determine the number of links to
+	 * know which if_is_24bit flag to check...
+	 */
+	if (nv_connector->edid &&
+	    nv_connector->type == DCB_CONNECTOR_LVDS_SPWG)
+		duallink = ((u8 *)nv_connector->edid)[121] == 2;
+	else
+		duallink = mode->clock >= bios->fp.duallink_transition_clk;
+
+	if ((!duallink && (bios->fp.strapless_is_24bit & 1)) ||
+	    ( duallink && (bios->fp.strapless_is_24bit & 2)))
+		connector->display_info.bpc = 8;
+}
+
 static int
 nouveau_connector_get_modes(struct drm_connector *connector)
 {
@@ -631,6 +696,12 @@
 		nv_connector->native_mode = drm_mode_duplicate(dev, &mode);
 	}
 
+	/* Determine display colour depth for everything except LVDS now,
+	 * DP requires this before mode_valid() is called.
+	 */
+	if (connector->connector_type != DRM_MODE_CONNECTOR_LVDS)
+		nouveau_connector_detect_depth(connector);
+
 	/* Find the native mode if this is a digital panel, if we didn't
 	 * find any modes through DDC previously add the native mode to
 	 * the list of modes.
@@ -646,12 +717,19 @@
 		ret = 1;
 	}
 
+	/* Determine LVDS colour depth, must happen after determining
+	 * "native" mode as some VBIOS tables require us to use the
+	 * pixel clock as part of the lookup...
+	 */
+	if (connector->connector_type == DRM_MODE_CONNECTOR_LVDS)
+		nouveau_connector_detect_depth(connector);
+
 	if (nv_encoder->dcb->type == OUTPUT_TV)
 		ret = get_slave_funcs(encoder)->get_modes(encoder, connector);
 
-	if (nv_connector->dcb->type == DCB_CONNECTOR_LVDS ||
-	    nv_connector->dcb->type == DCB_CONNECTOR_LVDS_SPWG ||
-	    nv_connector->dcb->type == DCB_CONNECTOR_eDP)
+	if (nv_connector->type == DCB_CONNECTOR_LVDS ||
+	    nv_connector->type == DCB_CONNECTOR_LVDS_SPWG ||
+	    nv_connector->type == DCB_CONNECTOR_eDP)
 		ret += nouveau_connector_scaler_modes_add(connector);
 
 	return ret;
@@ -710,7 +788,7 @@
 	case OUTPUT_DP:
 		max_clock  = nv_encoder->dp.link_nr;
 		max_clock *= nv_encoder->dp.link_bw;
-		clock = clock * nouveau_connector_bpp(connector) / 8;
+		clock = clock * (connector->display_info.bpc * 3) / 10;
 		break;
 	default:
 		BUG_ON(1);
@@ -768,66 +846,144 @@
 	.force = nouveau_connector_force
 };
 
+static int
+drm_conntype_from_dcb(enum dcb_connector_type dcb)
+{
+	switch (dcb) {
+	case DCB_CONNECTOR_VGA      : return DRM_MODE_CONNECTOR_VGA;
+	case DCB_CONNECTOR_TV_0     :
+	case DCB_CONNECTOR_TV_1     :
+	case DCB_CONNECTOR_TV_3     : return DRM_MODE_CONNECTOR_TV;
+	case DCB_CONNECTOR_DVI_I    : return DRM_MODE_CONNECTOR_DVII;
+	case DCB_CONNECTOR_DVI_D    : return DRM_MODE_CONNECTOR_DVID;
+	case DCB_CONNECTOR_LVDS     :
+	case DCB_CONNECTOR_LVDS_SPWG: return DRM_MODE_CONNECTOR_LVDS;
+	case DCB_CONNECTOR_DP       : return DRM_MODE_CONNECTOR_DisplayPort;
+	case DCB_CONNECTOR_eDP      : return DRM_MODE_CONNECTOR_eDP;
+	case DCB_CONNECTOR_HDMI_0   :
+	case DCB_CONNECTOR_HDMI_1   : return DRM_MODE_CONNECTOR_HDMIA;
+	default:
+		break;
+	}
+
+	return DRM_MODE_CONNECTOR_Unknown;
+}
+
 struct drm_connector *
 nouveau_connector_create(struct drm_device *dev, int index)
 {
 	const struct drm_connector_funcs *funcs = &nouveau_connector_funcs;
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+	struct nouveau_display_engine *disp = &dev_priv->engine.display;
 	struct nouveau_connector *nv_connector = NULL;
-	struct dcb_connector_table_entry *dcb = NULL;
 	struct drm_connector *connector;
 	int type, ret = 0;
+	bool dummy;
 
 	NV_DEBUG_KMS(dev, "\n");
 
-	if (index >= dev_priv->vbios.dcb.connector.entries)
-		return ERR_PTR(-EINVAL);
-
-	dcb = &dev_priv->vbios.dcb.connector.entry[index];
-	if (dcb->drm)
-		return dcb->drm;
-
-	switch (dcb->type) {
-	case DCB_CONNECTOR_VGA:
-		type = DRM_MODE_CONNECTOR_VGA;
-		break;
-	case DCB_CONNECTOR_TV_0:
-	case DCB_CONNECTOR_TV_1:
-	case DCB_CONNECTOR_TV_3:
-		type = DRM_MODE_CONNECTOR_TV;
-		break;
-	case DCB_CONNECTOR_DVI_I:
-		type = DRM_MODE_CONNECTOR_DVII;
-		break;
-	case DCB_CONNECTOR_DVI_D:
-		type = DRM_MODE_CONNECTOR_DVID;
-		break;
-	case DCB_CONNECTOR_HDMI_0:
-	case DCB_CONNECTOR_HDMI_1:
-		type = DRM_MODE_CONNECTOR_HDMIA;
-		break;
-	case DCB_CONNECTOR_LVDS:
-	case DCB_CONNECTOR_LVDS_SPWG:
-		type = DRM_MODE_CONNECTOR_LVDS;
-		funcs = &nouveau_connector_funcs_lvds;
-		break;
-	case DCB_CONNECTOR_DP:
-		type = DRM_MODE_CONNECTOR_DisplayPort;
-		break;
-	case DCB_CONNECTOR_eDP:
-		type = DRM_MODE_CONNECTOR_eDP;
-		break;
-	default:
-		NV_ERROR(dev, "unknown connector type: 0x%02x!!\n", dcb->type);
-		return ERR_PTR(-EINVAL);
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+		nv_connector = nouveau_connector(connector);
+		if (nv_connector->index == index)
+			return connector;
 	}
 
 	nv_connector = kzalloc(sizeof(*nv_connector), GFP_KERNEL);
 	if (!nv_connector)
 		return ERR_PTR(-ENOMEM);
-	nv_connector->dcb = dcb;
+
 	connector = &nv_connector->base;
+	nv_connector->index = index;
+
+	/* attempt to parse vbios connector type and hotplug gpio */
+	nv_connector->dcb = dcb_conn(dev, index);
+	if (nv_connector->dcb) {
+		static const u8 hpd[16] = {
+			0xff, 0x07, 0x08, 0xff, 0xff, 0x51, 0x52, 0xff,
+			0xff, 0xff, 0xff, 0xff, 0xff, 0x5e, 0x5f, 0x60,
+		};
+
+		u32 entry = ROM16(nv_connector->dcb[0]);
+		if (dcb_conntab(dev)[3] >= 4)
+			entry |= (u32)ROM16(nv_connector->dcb[2]) << 16;
+
+		nv_connector->hpd = ffs((entry & 0x07033000) >> 12);
+		nv_connector->hpd = hpd[nv_connector->hpd];
+
+		nv_connector->type = nv_connector->dcb[0];
+		if (drm_conntype_from_dcb(nv_connector->type) ==
+					  DRM_MODE_CONNECTOR_Unknown) {
+			NV_WARN(dev, "unknown connector type %02x\n",
+				nv_connector->type);
+			nv_connector->type = DCB_CONNECTOR_NONE;
+		}
+
+		/* Gigabyte NX85T */
+		if (nv_match_device(dev, 0x0421, 0x1458, 0x344c)) {
+			if (nv_connector->type == DCB_CONNECTOR_HDMI_1)
+				nv_connector->type = DCB_CONNECTOR_DVI_I;
+		}
+
+		/* Gigabyte GV-NX86T512H */
+		if (nv_match_device(dev, 0x0402, 0x1458, 0x3455)) {
+			if (nv_connector->type == DCB_CONNECTOR_HDMI_1)
+				nv_connector->type = DCB_CONNECTOR_DVI_I;
+		}
+	} else {
+		nv_connector->type = DCB_CONNECTOR_NONE;
+		nv_connector->hpd = DCB_GPIO_UNUSED;
+	}
+
+	/* no vbios data, or an unknown dcb connector type - attempt to
+	 * figure out something suitable ourselves
+	 */
+	if (nv_connector->type == DCB_CONNECTOR_NONE) {
+		struct drm_nouveau_private *dev_priv = dev->dev_private;
+		struct dcb_table *dcbt = &dev_priv->vbios.dcb;
+		u32 encoders = 0;
+		int i;
+
+		for (i = 0; i < dcbt->entries; i++) {
+			if (dcbt->entry[i].connector == nv_connector->index)
+				encoders |= (1 << dcbt->entry[i].type);
+		}
+
+		if (encoders & (1 << OUTPUT_DP)) {
+			if (encoders & (1 << OUTPUT_TMDS))
+				nv_connector->type = DCB_CONNECTOR_DP;
+			else
+				nv_connector->type = DCB_CONNECTOR_eDP;
+		} else
+		if (encoders & (1 << OUTPUT_TMDS)) {
+			if (encoders & (1 << OUTPUT_ANALOG))
+				nv_connector->type = DCB_CONNECTOR_DVI_I;
+			else
+				nv_connector->type = DCB_CONNECTOR_DVI_D;
+		} else
+		if (encoders & (1 << OUTPUT_ANALOG)) {
+			nv_connector->type = DCB_CONNECTOR_VGA;
+		} else
+		if (encoders & (1 << OUTPUT_LVDS)) {
+			nv_connector->type = DCB_CONNECTOR_LVDS;
+		} else
+		if (encoders & (1 << OUTPUT_TV)) {
+			nv_connector->type = DCB_CONNECTOR_TV_0;
+		}
+	}
+
+	type = drm_conntype_from_dcb(nv_connector->type);
+	if (type == DRM_MODE_CONNECTOR_LVDS) {
+		ret = nouveau_bios_parse_lvds_table(dev, 0, &dummy, &dummy);
+		if (ret) {
+			NV_ERROR(dev, "Error parsing LVDS table, disabling\n");
+			kfree(nv_connector);
+			return ERR_PTR(ret);
+		}
+
+		funcs = &nouveau_connector_funcs_lvds;
+	} else {
+		funcs = &nouveau_connector_funcs;
+	}
 
 	/* defaults, will get overridden in detect() */
 	connector->interlace_allowed = false;
@@ -836,28 +992,29 @@
 	drm_connector_init(dev, connector, funcs, type);
 	drm_connector_helper_add(connector, &nouveau_connector_helper_funcs);
 
-	/* Check if we need dithering enabled */
-	if (connector->connector_type == DRM_MODE_CONNECTOR_LVDS) {
-		bool dummy, is_24bit = false;
-
-		ret = nouveau_bios_parse_lvds_table(dev, 0, &dummy, &is_24bit);
-		if (ret) {
-			NV_ERROR(dev, "Error parsing LVDS table, disabling "
-				 "LVDS\n");
-			goto fail;
-		}
-
-		nv_connector->use_dithering = !is_24bit;
-	}
-
 	/* Init DVI-I specific properties */
-	if (dcb->type == DCB_CONNECTOR_DVI_I) {
-		drm_mode_create_dvi_i_properties(dev);
+	if (nv_connector->type == DCB_CONNECTOR_DVI_I)
 		drm_connector_attach_property(connector, dev->mode_config.dvi_i_subconnector_property, 0);
-		drm_connector_attach_property(connector, dev->mode_config.dvi_i_select_subconnector_property, 0);
+
+	/* Add overscan compensation options to digital outputs */
+	if (disp->underscan_property &&
+	    (nv_connector->type == DCB_CONNECTOR_DVI_D ||
+	     nv_connector->type == DCB_CONNECTOR_DVI_I ||
+	     nv_connector->type == DCB_CONNECTOR_HDMI_0 ||
+	     nv_connector->type == DCB_CONNECTOR_HDMI_1 ||
+	     nv_connector->type == DCB_CONNECTOR_DP)) {
+		drm_connector_attach_property(connector,
+					      disp->underscan_property,
+					      UNDERSCAN_OFF);
+		drm_connector_attach_property(connector,
+					      disp->underscan_hborder_property,
+					      0);
+		drm_connector_attach_property(connector,
+					      disp->underscan_vborder_property,
+					      0);
 	}
 
-	switch (dcb->type) {
+	switch (nv_connector->type) {
 	case DCB_CONNECTOR_VGA:
 		if (dev_priv->card_type >= NV_50) {
 			drm_connector_attach_property(connector,
@@ -876,32 +1033,32 @@
 		drm_connector_attach_property(connector,
 				dev->mode_config.scaling_mode_property,
 				nv_connector->scaling_mode);
-		drm_connector_attach_property(connector,
-				dev->mode_config.dithering_mode_property,
-				nv_connector->use_dithering ?
-				DRM_MODE_DITHERING_ON : DRM_MODE_DITHERING_OFF);
+		if (disp->dithering_mode) {
+			nv_connector->dithering_mode = DITHERING_MODE_AUTO;
+			drm_connector_attach_property(connector,
+						disp->dithering_mode,
+						nv_connector->dithering_mode);
+		}
+		if (disp->dithering_depth) {
+			nv_connector->dithering_depth = DITHERING_DEPTH_AUTO;
+			drm_connector_attach_property(connector,
+						disp->dithering_depth,
+						nv_connector->dithering_depth);
+		}
 		break;
 	}
 
-	if (nv_connector->dcb->gpio_tag != 0xff && pgpio->irq_register) {
-		pgpio->irq_register(dev, nv_connector->dcb->gpio_tag,
-				    nouveau_connector_hotplug, connector);
-
-		connector->polled = DRM_CONNECTOR_POLL_HPD;
-	} else {
-		connector->polled = DRM_CONNECTOR_POLL_CONNECT;
+	connector->polled = DRM_CONNECTOR_POLL_CONNECT;
+	if (nv_connector->hpd != DCB_GPIO_UNUSED) {
+		ret = nouveau_gpio_isr_add(dev, 0, nv_connector->hpd, 0xff,
+					   nouveau_connector_hotplug,
+					   connector);
+		if (ret == 0)
+			connector->polled = DRM_CONNECTOR_POLL_HPD;
 	}
 
 	drm_sysfs_connector_add(connector);
-
-	dcb->drm = connector;
-	return dcb->drm;
-
-fail:
-	drm_connector_cleanup(connector);
-	kfree(connector);
-	return ERR_PTR(ret);
-
+	return connector;
 }
 
 static void
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h
index 711b1e9..e485702 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.h
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.h
@@ -30,13 +30,43 @@
 #include "drm_edid.h"
 #include "nouveau_i2c.h"
 
+enum nouveau_underscan_type {
+	UNDERSCAN_OFF,
+	UNDERSCAN_ON,
+	UNDERSCAN_AUTO,
+};
+
+/* the enum values specifically defined here match nv50/nvd0 hw values, and
+ * the code relies on this
+ */
+enum nouveau_dithering_mode {
+	DITHERING_MODE_OFF = 0x00,
+	DITHERING_MODE_ON = 0x01,
+	DITHERING_MODE_DYNAMIC2X2 = 0x10 | DITHERING_MODE_ON,
+	DITHERING_MODE_STATIC2X2 = 0x18 | DITHERING_MODE_ON,
+	DITHERING_MODE_TEMPORAL = 0x20 | DITHERING_MODE_ON,
+	DITHERING_MODE_AUTO
+};
+
+enum nouveau_dithering_depth {
+	DITHERING_DEPTH_6BPC = 0x00,
+	DITHERING_DEPTH_8BPC = 0x02,
+	DITHERING_DEPTH_AUTO
+};
+
 struct nouveau_connector {
 	struct drm_connector base;
+	enum dcb_connector_type type;
+	u8 index;
+	u8 *dcb;
+	u8 hpd;
 
-	struct dcb_connector_table_entry *dcb;
-
+	int dithering_mode;
+	int dithering_depth;
 	int scaling_mode;
-	bool use_dithering;
+	enum nouveau_underscan_type underscan;
+	u32 underscan_hborder;
+	u32 underscan_vborder;
 
 	struct nouveau_encoder *detected_encoder;
 	struct edid *edid;
diff --git a/drivers/gpu/drm/nouveau/nouveau_crtc.h b/drivers/gpu/drm/nouveau/nouveau_crtc.h
index bf8e128..686f6b4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_crtc.h
+++ b/drivers/gpu/drm/nouveau/nouveau_crtc.h
@@ -32,8 +32,6 @@
 
 	int index;
 
-	struct drm_display_mode *mode;
-
 	uint32_t dpms_saved_fp_control;
 	uint32_t fp_users;
 	int saturation;
@@ -67,8 +65,8 @@
 		int depth;
 	} lut;
 
-	int (*set_dither)(struct nouveau_crtc *crtc, bool on, bool update);
-	int (*set_scale)(struct nouveau_crtc *crtc, int mode, bool update);
+	int (*set_dither)(struct nouveau_crtc *crtc, bool update);
+	int (*set_scale)(struct nouveau_crtc *crtc, bool update);
 };
 
 static inline struct nouveau_crtc *nouveau_crtc(struct drm_crtc *crtc)
diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
index f52c2db..fa2ec49 100644
--- a/drivers/gpu/drm/nouveau/nouveau_debugfs.c
+++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
@@ -44,7 +44,7 @@
 	seq_printf(m, "channel id    : %d\n", chan->id);
 
 	seq_printf(m, "cpu fifo state:\n");
-	seq_printf(m, "          base: 0x%08x\n", chan->pushbuf_base);
+	seq_printf(m, "          base: 0x%10llx\n", chan->pushbuf_base);
 	seq_printf(m, "           max: 0x%08x\n", chan->dma.max << 2);
 	seq_printf(m, "           cur: 0x%08x\n", chan->dma.cur << 2);
 	seq_printf(m, "           put: 0x%08x\n", chan->dma.put << 2);
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 7687a77..3cb52bc5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -32,6 +32,8 @@
 #include "nouveau_hw.h"
 #include "nouveau_crtc.h"
 #include "nouveau_dma.h"
+#include "nouveau_connector.h"
+#include "nouveau_gpio.h"
 #include "nv50_display.h"
 
 static void
@@ -107,14 +109,14 @@
 
 		if (!tile_flags) {
 			if (dev_priv->card_type < NV_D0)
-				nv_fb->r_pitch = 0x00100000 | fb->pitch;
+				nv_fb->r_pitch = 0x00100000 | fb->pitches[0];
 			else
-				nv_fb->r_pitch = 0x01000000 | fb->pitch;
+				nv_fb->r_pitch = 0x01000000 | fb->pitches[0];
 		} else {
 			u32 mode = nvbo->tile_mode;
 			if (dev_priv->card_type >= NV_C0)
 				mode >>= 4;
-			nv_fb->r_pitch = ((fb->pitch / 4) << 4) | mode;
+			nv_fb->r_pitch = ((fb->pitches[0] / 4) << 4) | mode;
 		}
 	}
 
@@ -147,11 +149,186 @@
 	return &nouveau_fb->base;
 }
 
-const struct drm_mode_config_funcs nouveau_mode_config_funcs = {
+static const struct drm_mode_config_funcs nouveau_mode_config_funcs = {
 	.fb_create = nouveau_user_framebuffer_create,
 	.output_poll_changed = nouveau_fbcon_output_poll_changed,
 };
 
+
+struct drm_prop_enum_list {
+	u8 gen_mask;
+	int type;
+	char *name;
+};
+
+static struct drm_prop_enum_list underscan[] = {
+	{ 6, UNDERSCAN_AUTO, "auto" },
+	{ 6, UNDERSCAN_OFF, "off" },
+	{ 6, UNDERSCAN_ON, "on" },
+	{}
+};
+
+static struct drm_prop_enum_list dither_mode[] = {
+	{ 7, DITHERING_MODE_AUTO, "auto" },
+	{ 7, DITHERING_MODE_OFF, "off" },
+	{ 1, DITHERING_MODE_ON, "on" },
+	{ 6, DITHERING_MODE_STATIC2X2, "static 2x2" },
+	{ 6, DITHERING_MODE_DYNAMIC2X2, "dynamic 2x2" },
+	{ 4, DITHERING_MODE_TEMPORAL, "temporal" },
+	{}
+};
+
+static struct drm_prop_enum_list dither_depth[] = {
+	{ 6, DITHERING_DEPTH_AUTO, "auto" },
+	{ 6, DITHERING_DEPTH_6BPC, "6 bpc" },
+	{ 6, DITHERING_DEPTH_8BPC, "8 bpc" },
+	{}
+};
+
+#define PROP_ENUM(p,gen,n,list) do {                                           \
+	struct drm_prop_enum_list *l = (list);                                 \
+	int c = 0;                                                             \
+	while (l->gen_mask) {                                                  \
+		if (l->gen_mask & (1 << (gen)))                                \
+			c++;                                                   \
+		l++;                                                           \
+	}                                                                      \
+	if (c) {                                                               \
+		p = drm_property_create(dev, DRM_MODE_PROP_ENUM, n, c);        \
+		l = (list);                                                    \
+		c = 0;                                                         \
+		while (p && l->gen_mask) {                                     \
+			if (l->gen_mask & (1 << (gen))) {                      \
+				drm_property_add_enum(p, c, l->type, l->name); \
+				c++;                                           \
+			}                                                      \
+			l++;                                                   \
+		}                                                              \
+	}                                                                      \
+} while(0)
+
+int
+nouveau_display_init(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_display_engine *disp = &dev_priv->engine.display;
+	struct drm_connector *connector;
+	int ret;
+
+	ret = disp->init(dev);
+	if (ret)
+		return ret;
+
+	drm_kms_helper_poll_enable(dev);
+
+	/* enable hotplug interrupts */
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+		struct nouveau_connector *conn = nouveau_connector(connector);
+		nouveau_gpio_irq(dev, 0, conn->hpd, 0xff, true);
+	}
+
+	return ret;
+}
+
+void
+nouveau_display_fini(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_display_engine *disp = &dev_priv->engine.display;
+	struct drm_connector *connector;
+
+	/* disable hotplug interrupts */
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+		struct nouveau_connector *conn = nouveau_connector(connector);
+		nouveau_gpio_irq(dev, 0, conn->hpd, 0xff, false);
+	}
+
+	drm_kms_helper_poll_disable(dev);
+	disp->fini(dev);
+}
+
+int
+nouveau_display_create(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_display_engine *disp = &dev_priv->engine.display;
+	int ret, gen;
+
+	drm_mode_config_init(dev);
+	drm_mode_create_scaling_mode_property(dev);
+	drm_mode_create_dvi_i_properties(dev);
+
+	if (dev_priv->card_type < NV_50)
+		gen = 0;
+	else
+	if (dev_priv->card_type < NV_D0)
+		gen = 1;
+	else
+		gen = 2;
+
+	PROP_ENUM(disp->dithering_mode, gen, "dithering mode", dither_mode);
+	PROP_ENUM(disp->dithering_depth, gen, "dithering depth", dither_depth);
+	PROP_ENUM(disp->underscan_property, gen, "underscan", underscan);
+
+	disp->underscan_hborder_property =
+		drm_property_create(dev, DRM_MODE_PROP_RANGE,
+				    "underscan hborder", 2);
+	disp->underscan_hborder_property->values[0] = 0;
+	disp->underscan_hborder_property->values[1] = 128;
+
+	disp->underscan_vborder_property =
+		drm_property_create(dev, DRM_MODE_PROP_RANGE,
+				    "underscan vborder", 2);
+	disp->underscan_vborder_property->values[0] = 0;
+	disp->underscan_vborder_property->values[1] = 128;
+
+	dev->mode_config.funcs = (void *)&nouveau_mode_config_funcs;
+	dev->mode_config.fb_base = pci_resource_start(dev->pdev, 1);
+
+	dev->mode_config.min_width = 0;
+	dev->mode_config.min_height = 0;
+	if (dev_priv->card_type < NV_10) {
+		dev->mode_config.max_width = 2048;
+		dev->mode_config.max_height = 2048;
+	} else
+	if (dev_priv->card_type < NV_50) {
+		dev->mode_config.max_width = 4096;
+		dev->mode_config.max_height = 4096;
+	} else {
+		dev->mode_config.max_width = 8192;
+		dev->mode_config.max_height = 8192;
+	}
+
+	drm_kms_helper_poll_init(dev);
+	drm_kms_helper_poll_disable(dev);
+
+	ret = disp->create(dev);
+	if (ret)
+		return ret;
+
+	if (dev->mode_config.num_crtc) {
+		ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+void
+nouveau_display_destroy(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_display_engine *disp = &dev_priv->engine.display;
+
+	drm_vblank_cleanup(dev);
+
+	disp->destroy(dev);
+
+	drm_kms_helper_poll_fini(dev);
+	drm_mode_config_cleanup(dev);
+}
+
 int
 nouveau_vblank_enable(struct drm_device *dev, int crtc)
 {
@@ -294,7 +471,7 @@
 	/* Initialize a page flip struct */
 	*s = (struct nouveau_page_flip_state)
 		{ { }, event, nouveau_crtc(crtc)->index,
-		  fb->bits_per_pixel, fb->pitch, crtc->x, crtc->y,
+		  fb->bits_per_pixel, fb->pitches[0], crtc->x, crtc->y,
 		  new_bo->bo.offset };
 
 	/* Choose the channel the flip will be handled in */
@@ -305,7 +482,10 @@
 
 	/* Emit a page flip */
 	if (dev_priv->card_type >= NV_50) {
-		ret = nv50_display_flip_next(crtc, fb, chan);
+		if (dev_priv->card_type >= NV_D0)
+			ret = nvd0_display_flip_next(crtc, fb, chan, 0);
+		else
+			ret = nv50_display_flip_next(crtc, fb, chan);
 		if (ret) {
 			nouveau_channel_put(&chan);
 			goto fail_unreserve;
@@ -369,3 +549,48 @@
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 	return 0;
 }
+
+int
+nouveau_display_dumb_create(struct drm_file *file_priv, struct drm_device *dev,
+			    struct drm_mode_create_dumb *args)
+{
+	struct nouveau_bo *bo;
+	int ret;
+
+	args->pitch = roundup(args->width * (args->bpp / 8), 256);
+	args->size = args->pitch * args->height;
+	args->size = roundup(args->size, PAGE_SIZE);
+
+	ret = nouveau_gem_new(dev, args->size, 0, TTM_PL_FLAG_VRAM, 0, 0, &bo);
+	if (ret)
+		return ret;
+
+	ret = drm_gem_handle_create(file_priv, bo->gem, &args->handle);
+	drm_gem_object_unreference_unlocked(bo->gem);
+	return ret;
+}
+
+int
+nouveau_display_dumb_destroy(struct drm_file *file_priv, struct drm_device *dev,
+			     uint32_t handle)
+{
+	return drm_gem_handle_delete(file_priv, handle);
+}
+
+int
+nouveau_display_dumb_map_offset(struct drm_file *file_priv,
+				struct drm_device *dev,
+				uint32_t handle, uint64_t *poffset)
+{
+	struct drm_gem_object *gem;
+
+	gem = drm_gem_object_lookup(dev, file_priv, handle);
+	if (gem) {
+		struct nouveau_bo *bo = gem->driver_private;
+		*poffset = bo->bo.addr_space_offset;
+		drm_gem_object_unreference_unlocked(gem);
+		return 0;
+	}
+
+	return -ENOENT;
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.c b/drivers/gpu/drm/nouveau/nouveau_dma.c
index 00bc6ea..4c2e4e5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dma.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dma.c
@@ -134,11 +134,13 @@
  *  -EBUSY if timeout exceeded
  */
 static inline int
-READ_GET(struct nouveau_channel *chan, uint32_t *prev_get, uint32_t *timeout)
+READ_GET(struct nouveau_channel *chan, uint64_t *prev_get, int *timeout)
 {
-	uint32_t val;
+	uint64_t val;
 
 	val = nvchan_rd32(chan, chan->user_get);
+        if (chan->user_get_hi)
+                val |= (uint64_t)nvchan_rd32(chan, chan->user_get_hi) << 32;
 
 	/* reset counter as long as GET is still advancing, this is
 	 * to avoid misdetecting a GPU lockup if the GPU happens to
@@ -218,8 +220,8 @@
 static int
 nv50_dma_wait(struct nouveau_channel *chan, int slots, int count)
 {
-	uint32_t cnt = 0, prev_get = 0;
-	int ret;
+	uint64_t prev_get = 0;
+	int ret, cnt = 0;
 
 	ret = nv50_dma_push_wait(chan, slots + 1);
 	if (unlikely(ret))
@@ -261,8 +263,8 @@
 int
 nouveau_dma_wait(struct nouveau_channel *chan, int slots, int size)
 {
-	uint32_t prev_get = 0, cnt = 0;
-	int get;
+	uint64_t prev_get = 0;
+	int cnt = 0, get;
 
 	if (chan->dma.ib_max)
 		return nv50_dma_wait(chan, slots, size);
diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c
index de5efe7..9b93b70 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dp.c
@@ -29,6 +29,7 @@
 #include "nouveau_connector.h"
 #include "nouveau_encoder.h"
 #include "nouveau_crtc.h"
+#include "nouveau_gpio.h"
 
 /******************************************************************************
  * aux channel util functions
@@ -273,8 +274,6 @@
 u8 *
 nouveau_dp_bios_data(struct drm_device *dev, struct dcb_entry *dcb, u8 **entry)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nvbios *bios = &dev_priv->vbios;
 	struct bit_entry d;
 	u8 *table;
 	int i;
@@ -289,7 +288,7 @@
 		return NULL;
 	}
 
-	table = ROMPTR(bios, d.data[0]);
+	table = ROMPTR(dev, d.data[0]);
 	if (!table) {
 		NV_ERROR(dev, "displayport table pointer invalid\n");
 		return NULL;
@@ -306,7 +305,7 @@
 	}
 
 	for (i = 0; i < table[3]; i++) {
-		*entry = ROMPTR(bios, table[table[1] + (i * table[2])]);
+		*entry = ROMPTR(dev, table[table[1] + (i * table[2])]);
 		if (*entry && bios_encoder_match(dcb, ROM32((*entry)[0])))
 			return table;
 	}
@@ -336,7 +335,6 @@
 static void
 dp_set_link_config(struct drm_device *dev, struct dp_state *dp)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	int or = dp->or, link = dp->link;
 	u8 *entry, sink[2];
 	u32 dp_ctrl;
@@ -360,7 +358,7 @@
 	 * table, that has (among other things) pointers to more scripts that
 	 * need to be executed, this time depending on link speed.
 	 */
-	entry = ROMPTR(&dev_priv->vbios, dp->entry[10]);
+	entry = ROMPTR(dev, dp->entry[10]);
 	if (entry) {
 		if (dp->table[0] < 0x30) {
 			while (dp->link_bw < (ROM16(entry[0]) * 10))
@@ -559,8 +557,6 @@
 bool
 nouveau_dp_link_train(struct drm_encoder *encoder, u32 datarate)
 {
-	struct drm_nouveau_private *dev_priv = encoder->dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc);
 	struct nouveau_connector *nv_connector =
@@ -581,7 +577,7 @@
 
 	dp.dcb = nv_encoder->dcb;
 	dp.crtc = nv_crtc->index;
-	dp.auxch = auxch->rd;
+	dp.auxch = auxch->drive;
 	dp.or = nv_encoder->or;
 	dp.link = !(nv_encoder->dcb->sorconf.link & 1);
 	dp.dpcd = nv_encoder->dp.dpcd;
@@ -590,7 +586,7 @@
 	 * we take during link training (DP_SET_POWER is one), we need
 	 * to ignore them for the moment to avoid races.
 	 */
-	pgpio->irq_enable(dev, nv_connector->dcb->gpio_tag, false);
+	nouveau_gpio_irq(dev, 0, nv_connector->hpd, 0xff, false);
 
 	/* enable down-spreading, if possible */
 	if (dp.table[1] >= 16) {
@@ -639,7 +635,7 @@
 	nouveau_bios_run_init_table(dev, ROM16(dp.entry[8]), dp.dcb, dp.crtc);
 
 	/* re-enable hotplug detect */
-	pgpio->irq_enable(dev, nv_connector->dcb->gpio_tag, true);
+	nouveau_gpio_irq(dev, 0, nv_connector->hpd, 0xff, true);
 	return true;
 }
 
@@ -656,7 +652,7 @@
 	if (!auxch)
 		return false;
 
-	ret = auxch_tx(dev, auxch->rd, 9, DP_DPCD_REV, dpcd, 8);
+	ret = auxch_tx(dev, auxch->drive, 9, DP_DPCD_REV, dpcd, 8);
 	if (ret)
 		return false;
 
@@ -684,7 +680,7 @@
 nouveau_dp_auxch(struct nouveau_i2c_chan *auxch, int cmd, int addr,
 		 uint8_t *data, int data_nr)
 {
-	return auxch_tx(auxch->dev, auxch->rd, cmd, addr, data, data_nr);
+	return auxch_tx(auxch->dev, auxch->drive, cmd, addr, data, data_nr);
 }
 
 static int
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.c b/drivers/gpu/drm/nouveau/nouveau_drv.c
index d661bc5..e4a7cfe 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.c
@@ -124,6 +124,10 @@
 int nouveau_ctxfw;
 module_param_named(ctxfw, nouveau_ctxfw, int, 0400);
 
+MODULE_PARM_DESC(ctxfw, "Santise DCB table according to MXM-SIS\n");
+int nouveau_mxmdcb = 1;
+module_param_named(mxmdcb, nouveau_mxmdcb, int, 0400);
+
 int nouveau_fbpercrtc;
 #if 0
 module_param_named(fbpercrtc, nouveau_fbpercrtc, int, 0400);
@@ -178,8 +182,11 @@
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
-	NV_INFO(dev, "Disabling fbcon acceleration...\n");
-	nouveau_fbcon_save_disable_accel(dev);
+	NV_INFO(dev, "Disabling display...\n");
+	nouveau_display_fini(dev);
+
+	NV_INFO(dev, "Disabling fbcon...\n");
+	nouveau_fbcon_set_suspend(dev, 1);
 
 	NV_INFO(dev, "Unpinning framebuffer(s)...\n");
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
@@ -220,7 +227,7 @@
 
 		ret = dev_priv->eng[e]->fini(dev, e, true);
 		if (ret) {
-			NV_ERROR(dev, "... engine %d failed: %d\n", i, ret);
+			NV_ERROR(dev, "... engine %d failed: %d\n", e, ret);
 			goto out_abort;
 		}
 	}
@@ -246,10 +253,6 @@
 		pci_set_power_state(pdev, PCI_D3hot);
 	}
 
-	console_lock();
-	nouveau_fbcon_set_suspend(dev, 1);
-	console_unlock();
-	nouveau_fbcon_restore_accel(dev);
 	return 0;
 
 out_abort:
@@ -275,8 +278,6 @@
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
 		return 0;
 
-	nouveau_fbcon_save_disable_accel(dev);
-
 	NV_INFO(dev, "We're back, enabling device...\n");
 	pci_set_power_state(pdev, PCI_D0);
 	pci_restore_state(pdev);
@@ -296,8 +297,6 @@
 	if (ret)
 		return ret;
 
-	nouveau_pm_resume(dev);
-
 	if (dev_priv->gart_info.type == NOUVEAU_GART_AGP) {
 		ret = nouveau_mem_init_agp(dev);
 		if (ret) {
@@ -337,6 +336,8 @@
 		}
 	}
 
+	nouveau_pm_resume(dev);
+
 	NV_INFO(dev, "Restoring mode...\n");
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		struct nouveau_framebuffer *nouveau_fb;
@@ -358,7 +359,19 @@
 			NV_ERROR(dev, "Could not pin/map cursor.\n");
 	}
 
-	engine->display.init(dev);
+	nouveau_fbcon_set_suspend(dev, 0);
+	nouveau_fbcon_zfill_all(dev);
+
+	nouveau_display_init(dev);
+
+	/* Force CLUT to get re-loaded during modeset */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+
+		nv_crtc->lut.depth = 0;
+	}
+
+	drm_helper_resume_force_mode(dev);
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
@@ -369,22 +382,6 @@
 						 nv_crtc->cursor_saved_y);
 	}
 
-	/* Force CLUT to get re-loaded during modeset */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-
-		nv_crtc->lut.depth = 0;
-	}
-
-	console_lock();
-	nouveau_fbcon_set_suspend(dev, 0);
-	console_unlock();
-
-	nouveau_fbcon_zfill_all(dev);
-
-	drm_helper_resume_force_mode(dev);
-
-	nouveau_fbcon_restore_accel(dev);
 	return 0;
 }
 
@@ -434,6 +431,10 @@
 	.gem_open_object = nouveau_gem_object_open,
 	.gem_close_object = nouveau_gem_object_close,
 
+	.dumb_create = nouveau_display_dumb_create,
+	.dumb_map_offset = nouveau_display_dumb_map_offset,
+	.dumb_destroy = nouveau_display_dumb_destroy,
+
 	.name = DRIVER_NAME,
 	.desc = DRIVER_DESC,
 #ifdef GIT_REVISION
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index 0c53e39..38134a9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -163,6 +163,9 @@
 #define NVOBJ_ENGINE_COPY0	3
 #define NVOBJ_ENGINE_COPY1	4
 #define NVOBJ_ENGINE_MPEG	5
+#define NVOBJ_ENGINE_PPP	NVOBJ_ENGINE_MPEG
+#define NVOBJ_ENGINE_BSP	6
+#define NVOBJ_ENGINE_VP		7
 #define NVOBJ_ENGINE_DISPLAY	15
 #define NVOBJ_ENGINE_NR		16
 
@@ -229,6 +232,7 @@
 	/* mapping of the regs controlling the fifo */
 	void __iomem *user;
 	uint32_t user_get;
+	uint32_t user_get_hi;
 	uint32_t user_put;
 
 	/* Fencing */
@@ -246,7 +250,7 @@
 	struct nouveau_gpuobj *pushbuf;
 	struct nouveau_bo     *pushbuf_bo;
 	struct nouveau_vma     pushbuf_vma;
-	uint32_t               pushbuf_base;
+	uint64_t               pushbuf_base;
 
 	/* Notifier memory */
 	struct nouveau_bo *notifier_bo;
@@ -393,24 +397,25 @@
 	int (*early_init)(struct drm_device *);
 	void (*late_takedown)(struct drm_device *);
 	int (*create)(struct drm_device *);
-	int (*init)(struct drm_device *);
 	void (*destroy)(struct drm_device *);
+	int (*init)(struct drm_device *);
+	void (*fini)(struct drm_device *);
+
+	struct drm_property *dithering_mode;
+	struct drm_property *dithering_depth;
+	struct drm_property *underscan_property;
+	struct drm_property *underscan_hborder_property;
+	struct drm_property *underscan_vborder_property;
 };
 
 struct nouveau_gpio_engine {
-	void *priv;
-
-	int  (*init)(struct drm_device *);
-	void (*takedown)(struct drm_device *);
-
-	int  (*get)(struct drm_device *, enum dcb_gpio_tag);
-	int  (*set)(struct drm_device *, enum dcb_gpio_tag, int state);
-
-	int  (*irq_register)(struct drm_device *, enum dcb_gpio_tag,
-			     void (*)(void *, int), void *);
-	void (*irq_unregister)(struct drm_device *, enum dcb_gpio_tag,
-			       void (*)(void *, int), void *);
-	bool (*irq_enable)(struct drm_device *, enum dcb_gpio_tag, bool on);
+	spinlock_t lock;
+	struct list_head isr;
+	int (*init)(struct drm_device *);
+	void (*fini)(struct drm_device *);
+	int (*drive)(struct drm_device *, int line, int dir, int out);
+	int (*sense)(struct drm_device *, int line);
+	void (*irq_enable)(struct drm_device *, int line, bool);
 };
 
 struct nouveau_pm_voltage_level {
@@ -484,7 +489,7 @@
 	u32 copy;
 	u32 daemon;
 	u32 vdec;
-	u32 unk05;	/* nv50:nva3, roughly.. */
+	u32 dom6;
 	u32 unka0;	/* nva3:nvc0 */
 	u32 hub01;	/* nvc0- */
 	u32 hub06;	/* nvc0- */
@@ -518,6 +523,12 @@
 	int nr_timing;
 };
 
+struct nouveau_pm_fan {
+	u32 min_duty;
+	u32 max_duty;
+	u32 pwm_freq;
+};
+
 struct nouveau_pm_engine {
 	struct nouveau_pm_voltage voltage;
 	struct nouveau_pm_level perflvl[NOUVEAU_PM_MAX_LEVEL];
@@ -525,6 +536,8 @@
 	struct nouveau_pm_memtimings memtimings;
 	struct nouveau_pm_temp_sensor_constants sensor_constants;
 	struct nouveau_pm_threshold_temp threshold_temp;
+	struct nouveau_pm_fan fan;
+	u32 pwm_divisor;
 
 	struct nouveau_pm_level boot;
 	struct nouveau_pm_level *cur;
@@ -532,19 +545,14 @@
 	struct device *hwmon;
 	struct notifier_block acpi_nb;
 
-	int (*clock_get)(struct drm_device *, u32 id);
-	void *(*clock_pre)(struct drm_device *, struct nouveau_pm_level *,
-			   u32 id, int khz);
-	void (*clock_set)(struct drm_device *, void *);
-
 	int  (*clocks_get)(struct drm_device *, struct nouveau_pm_level *);
 	void *(*clocks_pre)(struct drm_device *, struct nouveau_pm_level *);
-	void (*clocks_set)(struct drm_device *, void *);
+	int (*clocks_set)(struct drm_device *, void *);
 
 	int (*voltage_get)(struct drm_device *);
 	int (*voltage_set)(struct drm_device *, int voltage);
-	int (*fanspeed_get)(struct drm_device *);
-	int (*fanspeed_set)(struct drm_device *, int fanspeed);
+	int (*pwm_get)(struct drm_device *, int line, u32*, u32*);
+	int (*pwm_set)(struct drm_device *, int line, u32, u32);
 	int (*temp_get)(struct drm_device *);
 };
 
@@ -780,6 +788,8 @@
 	struct nouveau_vm *chan_vm;
 
 	struct nvbios vbios;
+	u8 *mxms;
+	struct list_head i2c_ports;
 
 	struct nv04_mode_state mode_reg;
 	struct nv04_mode_state saved_reg;
@@ -850,6 +860,7 @@
 extern int nouveau_perflvl_wr;
 extern int nouveau_msi;
 extern int nouveau_ctxfw;
+extern int nouveau_mxmdcb;
 
 extern int nouveau_pci_suspend(struct pci_dev *pdev, pm_message_t pm_state);
 extern int nouveau_pci_resume(struct pci_dev *pdev);
@@ -1075,8 +1086,6 @@
 extern void nouveau_bios_run_init_table(struct drm_device *, uint16_t table,
 					struct dcb_entry *, int crtc);
 extern void nouveau_bios_init_exec(struct drm_device *, uint16_t table);
-extern struct dcb_gpio_entry *nouveau_bios_gpio_entry(struct drm_device *,
-						      enum dcb_gpio_tag);
 extern struct dcb_connector_table_entry *
 nouveau_bios_connector_entry(struct drm_device *, int index);
 extern u32 get_pll_register(struct drm_device *, enum pll_types);
@@ -1094,11 +1103,18 @@
 			    enum LVDS_script, int pxclk);
 bool bios_encoder_match(struct dcb_entry *, u32 hash);
 
+/* nouveau_mxm.c */
+int  nouveau_mxm_init(struct drm_device *dev);
+void nouveau_mxm_fini(struct drm_device *dev);
+
 /* nouveau_ttm.c */
 int nouveau_ttm_global_init(struct drm_nouveau_private *);
 void nouveau_ttm_global_release(struct drm_nouveau_private *);
 int nouveau_ttm_mmap(struct file *, struct vm_area_struct *);
 
+/* nouveau_hdmi.c */
+void nouveau_hdmi_mode_set(struct drm_encoder *, struct drm_display_mode *);
+
 /* nouveau_dp.c */
 int nouveau_dp_auxch(struct nouveau_i2c_chan *auxch, int cmd, int addr,
 		     uint8_t *data, int data_nr);
@@ -1225,6 +1241,9 @@
 /* nv84_crypt.c */
 extern int  nv84_crypt_create(struct drm_device *);
 
+/* nv98_crypt.c */
+extern int  nv98_crypt_create(struct drm_device *dev);
+
 /* nva3_copy.c */
 extern int  nva3_copy_create(struct drm_device *dev);
 
@@ -1237,6 +1256,17 @@
 /* nv50_mpeg.c */
 extern int  nv50_mpeg_create(struct drm_device *dev);
 
+/* nv84_bsp.c */
+/* nv98_bsp.c */
+extern int  nv84_bsp_create(struct drm_device *dev);
+
+/* nv84_vp.c */
+/* nv98_vp.c */
+extern int  nv84_vp_create(struct drm_device *dev);
+
+/* nv98_ppp.c */
+extern int  nv98_ppp_create(struct drm_device *dev);
+
 /* nv04_instmem.c */
 extern int  nv04_instmem_init(struct drm_device *);
 extern void nv04_instmem_takedown(struct drm_device *);
@@ -1314,13 +1344,19 @@
 extern int nv04_display_early_init(struct drm_device *);
 extern void nv04_display_late_takedown(struct drm_device *);
 extern int nv04_display_create(struct drm_device *);
-extern int nv04_display_init(struct drm_device *);
 extern void nv04_display_destroy(struct drm_device *);
+extern int nv04_display_init(struct drm_device *);
+extern void nv04_display_fini(struct drm_device *);
 
 /* nvd0_display.c */
 extern int nvd0_display_create(struct drm_device *);
-extern int nvd0_display_init(struct drm_device *);
 extern void nvd0_display_destroy(struct drm_device *);
+extern int nvd0_display_init(struct drm_device *);
+extern void nvd0_display_fini(struct drm_device *);
+struct nouveau_bo *nvd0_display_crtc_sema(struct drm_device *, int crtc);
+void nvd0_display_flip_stop(struct drm_crtc *);
+int nvd0_display_flip_next(struct drm_crtc *, struct drm_framebuffer *,
+			   struct nouveau_channel *, u32 swap_interval);
 
 /* nv04_crtc.c */
 extern int nv04_crtc_create(struct drm_device *, int index);
@@ -1415,31 +1451,40 @@
 				  struct drm_file *);
 
 /* nouveau_display.c */
+int nouveau_display_create(struct drm_device *dev);
+void nouveau_display_destroy(struct drm_device *dev);
+int nouveau_display_init(struct drm_device *dev);
+void nouveau_display_fini(struct drm_device *dev);
 int nouveau_vblank_enable(struct drm_device *dev, int crtc);
 void nouveau_vblank_disable(struct drm_device *dev, int crtc);
 int nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			   struct drm_pending_vblank_event *event);
 int nouveau_finish_page_flip(struct nouveau_channel *,
 			     struct nouveau_page_flip_state *);
+int nouveau_display_dumb_create(struct drm_file *, struct drm_device *,
+				struct drm_mode_create_dumb *args);
+int nouveau_display_dumb_map_offset(struct drm_file *, struct drm_device *,
+				    uint32_t handle, uint64_t *offset);
+int nouveau_display_dumb_destroy(struct drm_file *, struct drm_device *,
+				 uint32_t handle);
 
 /* nv10_gpio.c */
-int nv10_gpio_get(struct drm_device *dev, enum dcb_gpio_tag tag);
-int nv10_gpio_set(struct drm_device *dev, enum dcb_gpio_tag tag, int state);
+int nv10_gpio_init(struct drm_device *dev);
+void nv10_gpio_fini(struct drm_device *dev);
+int nv10_gpio_drive(struct drm_device *dev, int line, int dir, int out);
+int nv10_gpio_sense(struct drm_device *dev, int line);
+void nv10_gpio_irq_enable(struct drm_device *, int line, bool on);
 
 /* nv50_gpio.c */
 int nv50_gpio_init(struct drm_device *dev);
 void nv50_gpio_fini(struct drm_device *dev);
-int nv50_gpio_get(struct drm_device *dev, enum dcb_gpio_tag tag);
-int nv50_gpio_set(struct drm_device *dev, enum dcb_gpio_tag tag, int state);
-int nvd0_gpio_get(struct drm_device *dev, enum dcb_gpio_tag tag);
-int nvd0_gpio_set(struct drm_device *dev, enum dcb_gpio_tag tag, int state);
-int  nv50_gpio_irq_register(struct drm_device *, enum dcb_gpio_tag,
-			    void (*)(void *, int), void *);
-void nv50_gpio_irq_unregister(struct drm_device *, enum dcb_gpio_tag,
-			      void (*)(void *, int), void *);
-bool nv50_gpio_irq_enable(struct drm_device *, enum dcb_gpio_tag, bool on);
+int nv50_gpio_drive(struct drm_device *dev, int line, int dir, int out);
+int nv50_gpio_sense(struct drm_device *dev, int line);
+void nv50_gpio_irq_enable(struct drm_device *, int line, bool on);
+int nvd0_gpio_drive(struct drm_device *dev, int line, int dir, int out);
+int nvd0_gpio_sense(struct drm_device *dev, int line);
 
-/* nv50_calc. */
+/* nv50_calc.c */
 int nv50_calc_pll(struct drm_device *, struct pll_lims *, int clk,
 		  int *N1, int *M1, int *N2, int *M2, int *P);
 int nva3_calc_pll(struct drm_device *, struct pll_lims *,
@@ -1562,6 +1607,13 @@
 #define NV_TRACEWARN(d, fmt, arg...) NV_PRINTK(KERN_NOTICE, d, fmt, ##arg)
 #define NV_TRACE(d, fmt, arg...) NV_PRINTK(KERN_INFO, d, fmt, ##arg)
 #define NV_WARN(d, fmt, arg...) NV_PRINTK(KERN_WARNING, d, fmt, ##arg)
+#define NV_WARNONCE(d, fmt, arg...) do {                                       \
+	static int _warned = 0;                                                \
+	if (!_warned) {                                                        \
+		NV_WARN(d, fmt, ##arg);                                        \
+		_warned = 1;                                                   \
+	}                                                                      \
+} while(0)
 
 /* nouveau_reg_debug bitmask */
 enum {
diff --git a/drivers/gpu/drm/nouveau/nouveau_fb.h b/drivers/gpu/drm/nouveau/nouveau_fb.h
index f4dd301..f3fb649 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fb.h
+++ b/drivers/gpu/drm/nouveau/nouveau_fb.h
@@ -42,8 +42,6 @@
 	return container_of(fb, struct nouveau_framebuffer, base);
 }
 
-extern const struct drm_mode_config_funcs nouveau_mode_config_funcs;
-
 int nouveau_framebuffer_init(struct drm_device *dev, struct nouveau_framebuffer *nouveau_fb,
 			     struct drm_mode_fb_cmd2 *mode_cmd, struct nouveau_bo *nvbo);
 #endif /* __NOUVEAU_FB_H__ */
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index d663065..9892218 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -36,6 +36,7 @@
 #include <linux/init.h>
 #include <linux/screen_info.h>
 #include <linux/vga_switcheroo.h>
+#include <linux/console.h>
 
 #include "drmP.h"
 #include "drm.h"
@@ -370,7 +371,7 @@
 	info->screen_base = nvbo_kmap_obj_iovirtual(nouveau_fb->nvbo);
 	info->screen_size = size;
 
-	drm_fb_helper_fill_fix(info, fb->pitch, fb->depth);
+	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->depth);
 	drm_fb_helper_fill_var(info, &nfbdev->helper, sizes->fb_width, sizes->fb_height);
 
 	/* Set aperture base/size for vesafb takeover */
@@ -488,6 +489,7 @@
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_fbdev *nfbdev;
+	int preferred_bpp;
 	int ret;
 
 	nfbdev = kzalloc(sizeof(struct nouveau_fbdev), GFP_KERNEL);
@@ -506,7 +508,15 @@
 	}
 
 	drm_fb_helper_single_add_all_connectors(&nfbdev->helper);
-	drm_fb_helper_initial_config(&nfbdev->helper, 32);
+
+	if (dev_priv->vram_size <= 32 * 1024 * 1024)
+		preferred_bpp = 8;
+	else if (dev_priv->vram_size <= 64 * 1024 * 1024)
+		preferred_bpp = 16;
+	else
+		preferred_bpp = 32;
+
+	drm_fb_helper_initial_config(&nfbdev->helper, preferred_bpp);
 	return 0;
 }
 
@@ -539,7 +549,13 @@
 void nouveau_fbcon_set_suspend(struct drm_device *dev, int state)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	console_lock();
+	if (state == 0)
+		nouveau_fbcon_save_disable_accel(dev);
 	fb_set_suspend(dev_priv->nfbdev->helper.fbdev, state);
+	if (state == 1)
+		nouveau_fbcon_restore_accel(dev);
+	console_unlock();
 }
 
 void nouveau_fbcon_zfill_all(struct drm_device *dev)
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 81116cf..2f6daae 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -539,8 +539,6 @@
 			return ret;
 	}
 
-	INIT_LIST_HEAD(&chan->fence.pending);
-	spin_lock_init(&chan->fence.lock);
 	atomic_set(&chan->fence.last_sequence_irq, 0);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_gpio.c b/drivers/gpu/drm/nouveau/nouveau_gpio.c
new file mode 100644
index 0000000..a580cc6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_gpio.c
@@ -0,0 +1,400 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "drmP.h"
+#include "nouveau_drv.h"
+#include "nouveau_i2c.h"
+#include "nouveau_gpio.h"
+
+static u8 *
+dcb_gpio_table(struct drm_device *dev)
+{
+	u8 *dcb = dcb_table(dev);
+	if (dcb) {
+		if (dcb[0] >= 0x30 && dcb[1] >= 0x0c)
+			return ROMPTR(dev, dcb[0x0a]);
+		if (dcb[0] >= 0x22 && dcb[-1] >= 0x13)
+			return ROMPTR(dev, dcb[-15]);
+	}
+	return NULL;
+}
+
+static u8 *
+dcb_gpio_entry(struct drm_device *dev, int idx, int ent, u8 *version)
+{
+	u8 *table = dcb_gpio_table(dev);
+	if (table) {
+		*version = table[0];
+		if (*version < 0x30 && ent < table[2])
+			return table + 3 + (ent * table[1]);
+		else if (ent < table[2])
+			return table + table[1] + (ent * table[3]);
+	}
+	return NULL;
+}
+
+int
+nouveau_gpio_drive(struct drm_device *dev, int idx, int line, int dir, int out)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+
+	return pgpio->drive ? pgpio->drive(dev, line, dir, out) : -ENODEV;
+}
+
+int
+nouveau_gpio_sense(struct drm_device *dev, int idx, int line)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+
+	return pgpio->sense ? pgpio->sense(dev, line) : -ENODEV;
+}
+
+int
+nouveau_gpio_find(struct drm_device *dev, int idx, u8 func, u8 line,
+		  struct gpio_func *gpio)
+{
+	u8 *table, *entry, version;
+	int i = -1;
+
+	if (line == 0xff && func == 0xff)
+		return -EINVAL;
+
+	while ((entry = dcb_gpio_entry(dev, idx, ++i, &version))) {
+		if (version < 0x40) {
+			u16 data = ROM16(entry[0]);
+			*gpio = (struct gpio_func) {
+				.line = (data & 0x001f) >> 0,
+				.func = (data & 0x07e0) >> 5,
+				.log[0] = (data & 0x1800) >> 11,
+				.log[1] = (data & 0x6000) >> 13,
+			};
+		} else
+		if (version < 0x41) {
+			*gpio = (struct gpio_func) {
+				.line = entry[0] & 0x1f,
+				.func = entry[1],
+				.log[0] = (entry[3] & 0x18) >> 3,
+				.log[1] = (entry[3] & 0x60) >> 5,
+			};
+		} else {
+			*gpio = (struct gpio_func) {
+				.line = entry[0] & 0x3f,
+				.func = entry[1],
+				.log[0] = (entry[4] & 0x30) >> 4,
+				.log[1] = (entry[4] & 0xc0) >> 6,
+			};
+		}
+
+		if ((line == 0xff || line == gpio->line) &&
+		    (func == 0xff || func == gpio->func))
+			return 0;
+	}
+
+	/* DCB 2.2, fixed TVDAC GPIO data */
+	if ((table = dcb_table(dev)) && table[0] >= 0x22) {
+		if (func == DCB_GPIO_TVDAC0) {
+			*gpio = (struct gpio_func) {
+				.func = DCB_GPIO_TVDAC0,
+				.line = table[-4] >> 4,
+				.log[0] = !!(table[-5] & 2),
+				.log[1] =  !(table[-5] & 2),
+			};
+			return 0;
+		}
+	}
+
+	/* Apple iMac G4 NV18 */
+	if (nv_match_device(dev, 0x0189, 0x10de, 0x0010)) {
+		if (func == DCB_GPIO_TVDAC0) {
+			*gpio = (struct gpio_func) {
+				.func = DCB_GPIO_TVDAC0,
+				.line = 4,
+				.log[0] = 0,
+				.log[1] = 1,
+			};
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+int
+nouveau_gpio_set(struct drm_device *dev, int idx, u8 tag, u8 line, int state)
+{
+	struct gpio_func gpio;
+	int ret;
+
+	ret = nouveau_gpio_find(dev, idx, tag, line, &gpio);
+	if (ret == 0) {
+		int dir = !!(gpio.log[state] & 0x02);
+		int out = !!(gpio.log[state] & 0x01);
+		ret = nouveau_gpio_drive(dev, idx, gpio.line, dir, out);
+	}
+
+	return ret;
+}
+
+int
+nouveau_gpio_get(struct drm_device *dev, int idx, u8 tag, u8 line)
+{
+	struct gpio_func gpio;
+	int ret;
+
+	ret = nouveau_gpio_find(dev, idx, tag, line, &gpio);
+	if (ret == 0) {
+		ret = nouveau_gpio_sense(dev, idx, gpio.line);
+		if (ret >= 0)
+			ret = (ret == (gpio.log[1] & 1));
+	}
+
+	return ret;
+}
+
+int
+nouveau_gpio_irq(struct drm_device *dev, int idx, u8 tag, u8 line, bool on)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+	struct gpio_func gpio;
+	int ret;
+
+	ret = nouveau_gpio_find(dev, idx, tag, line, &gpio);
+	if (ret == 0) {
+		if (idx == 0 && pgpio->irq_enable)
+			pgpio->irq_enable(dev, gpio.line, on);
+		else
+			ret = -ENODEV;
+	}
+
+	return ret;
+}
+
+struct gpio_isr {
+	struct drm_device *dev;
+	struct list_head head;
+	struct work_struct work;
+	int idx;
+	struct gpio_func func;
+	void (*handler)(void *, int);
+	void *data;
+	bool inhibit;
+};
+
+static void
+nouveau_gpio_isr_bh(struct work_struct *work)
+{
+	struct gpio_isr *isr = container_of(work, struct gpio_isr, work);
+	struct drm_device *dev = isr->dev;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+	unsigned long flags;
+	int state;
+
+	state = nouveau_gpio_get(dev, isr->idx, isr->func.func, isr->func.line);
+	if (state >= 0)
+		isr->handler(isr->data, state);
+
+	spin_lock_irqsave(&pgpio->lock, flags);
+	isr->inhibit = false;
+	spin_unlock_irqrestore(&pgpio->lock, flags);
+}
+
+void
+nouveau_gpio_isr(struct drm_device *dev, int idx, u32 line_mask)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+	struct gpio_isr *isr;
+
+	if (idx != 0)
+		return;
+
+	spin_lock(&pgpio->lock);
+	list_for_each_entry(isr, &pgpio->isr, head) {
+		if (line_mask & (1 << isr->func.line)) {
+			if (isr->inhibit)
+				continue;
+			isr->inhibit = true;
+			schedule_work(&isr->work);
+		}
+	}
+	spin_unlock(&pgpio->lock);
+}
+
+int
+nouveau_gpio_isr_add(struct drm_device *dev, int idx, u8 tag, u8 line,
+		     void (*handler)(void *, int), void *data)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+	struct gpio_isr *isr;
+	unsigned long flags;
+	int ret;
+
+	isr = kzalloc(sizeof(*isr), GFP_KERNEL);
+	if (!isr)
+		return -ENOMEM;
+
+	ret = nouveau_gpio_find(dev, idx, tag, line, &isr->func);
+	if (ret) {
+		kfree(isr);
+		return ret;
+	}
+
+	INIT_WORK(&isr->work, nouveau_gpio_isr_bh);
+	isr->dev = dev;
+	isr->handler = handler;
+	isr->data = data;
+	isr->idx = idx;
+
+	spin_lock_irqsave(&pgpio->lock, flags);
+	list_add(&isr->head, &pgpio->isr);
+	spin_unlock_irqrestore(&pgpio->lock, flags);
+	return 0;
+}
+
+void
+nouveau_gpio_isr_del(struct drm_device *dev, int idx, u8 tag, u8 line,
+		     void (*handler)(void *, int), void *data)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+	struct gpio_isr *isr, *tmp;
+	struct gpio_func func;
+	unsigned long flags;
+	LIST_HEAD(tofree);
+	int ret;
+
+	ret = nouveau_gpio_find(dev, idx, tag, line, &func);
+	if (ret == 0) {
+		spin_lock_irqsave(&pgpio->lock, flags);
+		list_for_each_entry_safe(isr, tmp, &pgpio->isr, head) {
+			if (memcmp(&isr->func, &func, sizeof(func)) ||
+			    isr->idx != idx ||
+			    isr->handler != handler || isr->data != data)
+				continue;
+			list_move(&isr->head, &tofree);
+		}
+		spin_unlock_irqrestore(&pgpio->lock, flags);
+
+		list_for_each_entry_safe(isr, tmp, &tofree, head) {
+			flush_work_sync(&isr->work);
+			kfree(isr);
+		}
+	}
+}
+
+int
+nouveau_gpio_create(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+
+	INIT_LIST_HEAD(&pgpio->isr);
+	spin_lock_init(&pgpio->lock);
+
+	return nouveau_gpio_init(dev);
+}
+
+void
+nouveau_gpio_destroy(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+
+	nouveau_gpio_fini(dev);
+	BUG_ON(!list_empty(&pgpio->isr));
+}
+
+int
+nouveau_gpio_init(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+	int ret = 0;
+
+	if (pgpio->init)
+		ret = pgpio->init(dev);
+
+	return ret;
+}
+
+void
+nouveau_gpio_fini(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+
+	if (pgpio->fini)
+		pgpio->fini(dev);
+}
+
+void
+nouveau_gpio_reset(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	u8 *entry, version;
+	int ent = -1;
+
+	while ((entry = dcb_gpio_entry(dev, 0, ++ent, &version))) {
+		u8 func = 0xff, line, defs, unk0, unk1;
+		if (version >= 0x41) {
+			defs = !!(entry[0] & 0x80);
+			line = entry[0] & 0x3f;
+			func = entry[1];
+			unk0 = entry[2];
+			unk1 = entry[3] & 0x1f;
+		} else
+		if (version >= 0x40) {
+			line = entry[0] & 0x1f;
+			func = entry[1];
+			defs = !!(entry[3] & 0x01);
+			unk0 = !!(entry[3] & 0x02);
+			unk1 = !!(entry[3] & 0x04);
+		} else {
+			break;
+		}
+
+		if (func == 0xff)
+			continue;
+
+		nouveau_gpio_func_set(dev, func, defs);
+
+		if (dev_priv->card_type >= NV_D0) {
+			nv_mask(dev, 0x00d610 + (line * 4), 0xff, unk0);
+			if (unk1--)
+				nv_mask(dev, 0x00d640 + (unk1 * 4), 0xff, line);
+		} else
+		if (dev_priv->card_type >= NV_50) {
+			static const u32 regs[] = { 0xe100, 0xe28c };
+			u32 val = (unk1 << 16) | unk0;
+			u32 reg = regs[line >> 4]; line &= 0x0f;
+
+			nv_mask(dev, reg, 0x00010001 << line, val << line);
+		}
+	}
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_gpio.h b/drivers/gpu/drm/nouveau/nouveau_gpio.h
new file mode 100644
index 0000000..64c5cb0
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_gpio.h
@@ -0,0 +1,71 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef __NOUVEAU_GPIO_H__
+#define __NOUVEAU_GPIO_H__
+
+struct gpio_func {
+	u8 func;
+	u8 line;
+	u8 log[2];
+};
+
+/* nouveau_gpio.c */
+int  nouveau_gpio_create(struct drm_device *);
+void nouveau_gpio_destroy(struct drm_device *);
+int  nouveau_gpio_init(struct drm_device *);
+void nouveau_gpio_fini(struct drm_device *);
+void nouveau_gpio_reset(struct drm_device *);
+int  nouveau_gpio_drive(struct drm_device *, int idx, int line,
+			int dir, int out);
+int  nouveau_gpio_sense(struct drm_device *, int idx, int line);
+int  nouveau_gpio_find(struct drm_device *, int idx, u8 tag, u8 line,
+		       struct gpio_func *);
+int  nouveau_gpio_set(struct drm_device *, int idx, u8 tag, u8 line, int state);
+int  nouveau_gpio_get(struct drm_device *, int idx, u8 tag, u8 line);
+int  nouveau_gpio_irq(struct drm_device *, int idx, u8 tag, u8 line, bool on);
+void nouveau_gpio_isr(struct drm_device *, int idx, u32 mask);
+int  nouveau_gpio_isr_add(struct drm_device *, int idx, u8 tag, u8 line,
+			  void (*)(void *, int state), void *data);
+void nouveau_gpio_isr_del(struct drm_device *, int idx, u8 tag, u8 line,
+			  void (*)(void *, int state), void *data);
+
+static inline bool
+nouveau_gpio_func_valid(struct drm_device *dev, u8 tag)
+{
+	struct gpio_func func;
+	return (nouveau_gpio_find(dev, 0, tag, 0xff, &func)) == 0;
+}
+
+static inline int
+nouveau_gpio_func_set(struct drm_device *dev, u8 tag, int state)
+{
+	return nouveau_gpio_set(dev, 0, tag, 0xff, state);
+}
+
+static inline int
+nouveau_gpio_func_get(struct drm_device *dev, u8 tag)
+{
+	return nouveau_gpio_get(dev, 0, tag, 0xff);
+}
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_hdmi.c b/drivers/gpu/drm/nouveau/nouveau_hdmi.c
new file mode 100644
index 0000000..59ea1c1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_hdmi.c
@@ -0,0 +1,258 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "drmP.h"
+#include "nouveau_drv.h"
+#include "nouveau_connector.h"
+#include "nouveau_encoder.h"
+#include "nouveau_crtc.h"
+
+static bool
+hdmi_sor(struct drm_encoder *encoder)
+{
+	struct drm_nouveau_private *dev_priv = encoder->dev->dev_private;
+	if (dev_priv->chipset < 0xa3)
+		return false;
+	return true;
+}
+
+static inline u32
+hdmi_base(struct drm_encoder *encoder)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc);
+	if (!hdmi_sor(encoder))
+		return 0x616500 + (nv_crtc->index * 0x800);
+	return 0x61c500 + (nv_encoder->or * 0x800);
+}
+
+static void
+hdmi_wr32(struct drm_encoder *encoder, u32 reg, u32 val)
+{
+	nv_wr32(encoder->dev, hdmi_base(encoder) + reg, val);
+}
+
+static u32
+hdmi_rd32(struct drm_encoder *encoder, u32 reg)
+{
+	return nv_rd32(encoder->dev, hdmi_base(encoder) + reg);
+}
+
+static u32
+hdmi_mask(struct drm_encoder *encoder, u32 reg, u32 mask, u32 val)
+{
+	u32 tmp = hdmi_rd32(encoder, reg);
+	hdmi_wr32(encoder, reg, (tmp & ~mask) | val);
+	return tmp;
+}
+
+static void
+nouveau_audio_disconnect(struct drm_encoder *encoder)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct drm_device *dev = encoder->dev;
+	u32 or = nv_encoder->or * 0x800;
+
+	if (hdmi_sor(encoder)) {
+		nv_mask(dev, 0x61c448 + or, 0x00000003, 0x00000000);
+	}
+}
+
+static void
+nouveau_audio_mode_set(struct drm_encoder *encoder,
+		       struct drm_display_mode *mode)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_connector *nv_connector;
+	struct drm_device *dev = encoder->dev;
+	u32 or = nv_encoder->or * 0x800;
+	int i;
+
+	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	if (!drm_detect_monitor_audio(nv_connector->edid)) {
+		nouveau_audio_disconnect(encoder);
+		return;
+	}
+
+	if (hdmi_sor(encoder)) {
+		nv_mask(dev, 0x61c448 + or, 0x00000001, 0x00000001);
+
+		drm_edid_to_eld(&nv_connector->base, nv_connector->edid);
+		if (nv_connector->base.eld[0]) {
+			u8 *eld = nv_connector->base.eld;
+			for (i = 0; i < eld[2] * 4; i++)
+				nv_wr32(dev, 0x61c440 + or, (i << 8) | eld[i]);
+			for (i = eld[2] * 4; i < 0x60; i++)
+				nv_wr32(dev, 0x61c440 + or, (i << 8) | 0x00);
+			nv_mask(dev, 0x61c448 + or, 0x00000002, 0x00000002);
+		}
+	}
+}
+
+static void
+nouveau_hdmi_infoframe(struct drm_encoder *encoder, u32 ctrl, u8 *frame)
+{
+	/* calculate checksum for the infoframe */
+	u8 sum = 0, i;
+	for (i = 0; i < frame[2]; i++)
+		sum += frame[i];
+	frame[3] = 256 - sum;
+
+	/* disable infoframe, and write header */
+	hdmi_mask(encoder, ctrl + 0x00, 0x00000001, 0x00000000);
+	hdmi_wr32(encoder, ctrl + 0x08, *(u32 *)frame & 0xffffff);
+
+	/* register scans tell me the audio infoframe has only one set of
+	 * subpack regs, according to tegra (gee nvidia, it'd be nice if we
+	 * could get those docs too!), the hdmi block pads out the rest of
+	 * the packet on its own.
+	 */
+	if (ctrl == 0x020)
+		frame[2] = 6;
+
+	/* write out checksum and data, weird weird 7 byte register pairs */
+	for (i = 0; i < frame[2] + 1; i += 7) {
+		u32 rsubpack = ctrl + 0x0c + ((i / 7) * 8);
+		u32 *subpack = (u32 *)&frame[3 + i];
+		hdmi_wr32(encoder, rsubpack + 0, subpack[0]);
+		hdmi_wr32(encoder, rsubpack + 4, subpack[1] & 0xffffff);
+	}
+
+	/* enable the infoframe */
+	hdmi_mask(encoder, ctrl, 0x00000001, 0x00000001);
+}
+
+static void
+nouveau_hdmi_video_infoframe(struct drm_encoder *encoder,
+			     struct drm_display_mode *mode)
+{
+	const u8 Y = 0, A = 0, B = 0, S = 0, C = 0, M = 0, R = 0;
+	const u8 ITC = 0, EC = 0, Q = 0, SC = 0, VIC = 0, PR = 0;
+	const u8 bar_top = 0, bar_bottom = 0, bar_left = 0, bar_right = 0;
+	u8 frame[20];
+
+	frame[0x00] = 0x82; /* AVI infoframe */
+	frame[0x01] = 0x02; /* version */
+	frame[0x02] = 0x0d; /* length */
+	frame[0x03] = 0x00;
+	frame[0x04] = (Y << 5) | (A << 4) | (B << 2) | S;
+	frame[0x05] = (C << 6) | (M << 4) | R;
+	frame[0x06] = (ITC << 7) | (EC << 4) | (Q << 2) | SC;
+	frame[0x07] = VIC;
+	frame[0x08] = PR;
+	frame[0x09] = bar_top & 0xff;
+	frame[0x0a] = bar_top >> 8;
+	frame[0x0b] = bar_bottom & 0xff;
+	frame[0x0c] = bar_bottom >> 8;
+	frame[0x0d] = bar_left & 0xff;
+	frame[0x0e] = bar_left >> 8;
+	frame[0x0f] = bar_right & 0xff;
+	frame[0x10] = bar_right >> 8;
+	frame[0x11] = 0x00;
+	frame[0x12] = 0x00;
+	frame[0x13] = 0x00;
+
+	nouveau_hdmi_infoframe(encoder, 0x020, frame);
+}
+
+static void
+nouveau_hdmi_audio_infoframe(struct drm_encoder *encoder,
+			     struct drm_display_mode *mode)
+{
+	const u8 CT = 0x00, CC = 0x01, ceaSS = 0x00, SF = 0x00, FMT = 0x00;
+	const u8 CA = 0x00, DM_INH = 0, LSV = 0x00;
+	u8 frame[12];
+
+	frame[0x00] = 0x84;	/* Audio infoframe */
+	frame[0x01] = 0x01;	/* version */
+	frame[0x02] = 0x0a;	/* length */
+	frame[0x03] = 0x00;
+	frame[0x04] = (CT << 4) | CC;
+	frame[0x05] = (SF << 2) | ceaSS;
+	frame[0x06] = FMT;
+	frame[0x07] = CA;
+	frame[0x08] = (DM_INH << 7) | (LSV << 3);
+	frame[0x09] = 0x00;
+	frame[0x0a] = 0x00;
+	frame[0x0b] = 0x00;
+
+	nouveau_hdmi_infoframe(encoder, 0x000, frame);
+}
+
+static void
+nouveau_hdmi_disconnect(struct drm_encoder *encoder)
+{
+	nouveau_audio_disconnect(encoder);
+
+	/* disable audio and avi infoframes */
+	hdmi_mask(encoder, 0x000, 0x00000001, 0x00000000);
+	hdmi_mask(encoder, 0x020, 0x00000001, 0x00000000);
+
+	/* disable hdmi */
+	hdmi_mask(encoder, 0x0a4, 0x40000000, 0x00000000);
+}
+
+void
+nouveau_hdmi_mode_set(struct drm_encoder *encoder,
+		      struct drm_display_mode *mode)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_connector *nv_connector;
+	struct drm_device *dev = encoder->dev;
+	u32 max_ac_packet, rekey;
+
+	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	if (!mode || !nv_connector || !nv_connector->edid ||
+	    !drm_detect_hdmi_monitor(nv_connector->edid)) {
+		nouveau_hdmi_disconnect(encoder);
+		return;
+	}
+
+	nouveau_hdmi_video_infoframe(encoder, mode);
+	nouveau_hdmi_audio_infoframe(encoder, mode);
+
+	hdmi_mask(encoder, 0x0d0, 0x00070001, 0x00010001); /* SPARE, HW_CTS */
+	hdmi_mask(encoder, 0x068, 0x00010101, 0x00000000); /* ACR_CTRL, ?? */
+	hdmi_mask(encoder, 0x078, 0x80000000, 0x80000000); /* ACR_0441_ENABLE */
+
+	nv_mask(dev, 0x61733c, 0x00100000, 0x00100000); /* RESETF */
+	nv_mask(dev, 0x61733c, 0x10000000, 0x10000000); /* LOOKUP_EN */
+	nv_mask(dev, 0x61733c, 0x00100000, 0x00000000); /* !RESETF */
+
+	/* value matches nvidia binary driver, and tegra constant */
+	rekey = 56;
+
+	max_ac_packet  = mode->htotal - mode->hdisplay;
+	max_ac_packet -= rekey;
+	max_ac_packet -= 18; /* constant from tegra */
+	max_ac_packet /= 32;
+
+	/* enable hdmi */
+	hdmi_mask(encoder, 0x0a4, 0x5f1f003f, 0x40000000 | /* enable */
+					      0x1f000000 | /* unknown */
+					      max_ac_packet << 16 |
+					      rekey);
+
+	nouveau_audio_mode_set(encoder, mode);
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_hwsq.h b/drivers/gpu/drm/nouveau/nouveau_hwsq.h
new file mode 100644
index 0000000..6976875
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_hwsq.h
@@ -0,0 +1,115 @@
+/*
+ * Copyright 2010 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#ifndef __NOUVEAU_HWSQ_H__
+#define __NOUVEAU_HWSQ_H__
+
+struct hwsq_ucode {
+	u8 data[0x200];
+	union {
+		u8  *u08;
+		u16 *u16;
+		u32 *u32;
+	} ptr;
+	u16 len;
+
+	u32 reg;
+	u32 val;
+};
+
+static inline void
+hwsq_init(struct hwsq_ucode *hwsq)
+{
+	hwsq->ptr.u08 = hwsq->data;
+	hwsq->reg = 0xffffffff;
+	hwsq->val = 0xffffffff;
+}
+
+static inline void
+hwsq_fini(struct hwsq_ucode *hwsq)
+{
+	do {
+		*hwsq->ptr.u08++ = 0x7f;
+		hwsq->len = hwsq->ptr.u08 - hwsq->data;
+	} while (hwsq->len & 3);
+	hwsq->ptr.u08 = hwsq->data;
+}
+
+static inline void
+hwsq_usec(struct hwsq_ucode *hwsq, u8 usec)
+{
+	u32 shift = 0;
+	while (usec & ~3) {
+		usec >>= 2;
+		shift++;
+	}
+
+	*hwsq->ptr.u08++ = (shift << 2) | usec;
+}
+
+static inline void
+hwsq_setf(struct hwsq_ucode *hwsq, u8 flag, int val)
+{
+	flag += 0x80;
+	if (val >= 0)
+		flag += 0x20;
+	if (val >= 1)
+		flag += 0x20;
+	*hwsq->ptr.u08++ = flag;
+}
+
+static inline void
+hwsq_op5f(struct hwsq_ucode *hwsq, u8 v0, u8 v1)
+{
+	*hwsq->ptr.u08++ = 0x5f;
+	*hwsq->ptr.u08++ = v0;
+	*hwsq->ptr.u08++ = v1;
+}
+
+static inline void
+hwsq_wr32(struct hwsq_ucode *hwsq, u32 reg, u32 val)
+{
+	if (val != hwsq->val) {
+		if ((val & 0xffff0000) == (hwsq->val & 0xffff0000)) {
+			*hwsq->ptr.u08++ = 0x42;
+			*hwsq->ptr.u16++ = (val & 0x0000ffff);
+		} else {
+			*hwsq->ptr.u08++ = 0xe2;
+			*hwsq->ptr.u32++ = val;
+		}
+
+		hwsq->val = val;
+	}
+
+	if ((reg & 0xffff0000) == (hwsq->reg & 0xffff0000)) {
+		*hwsq->ptr.u08++ = 0x40;
+		*hwsq->ptr.u16++ = (reg & 0x0000ffff);
+	} else {
+		*hwsq->ptr.u08++ = 0xe0;
+		*hwsq->ptr.u32++ = reg;
+	}
+	hwsq->reg = reg;
+}
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_i2c.c b/drivers/gpu/drm/nouveau/nouveau_i2c.c
index c6143df..820ae7f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_i2c.c
+++ b/drivers/gpu/drm/nouveau/nouveau_i2c.c
@@ -29,262 +29,465 @@
 #include "nouveau_i2c.h"
 #include "nouveau_hw.h"
 
-static void
-nv04_i2c_setscl(void *data, int state)
-{
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
-	uint8_t val;
+#define T_TIMEOUT  2200000
+#define T_RISEFALL 1000
+#define T_HOLD     5000
 
-	val = (NVReadVgaCrtc(dev, 0, i2c->wr) & 0xd0) | (state ? 0x20 : 0);
-	NVWriteVgaCrtc(dev, 0, i2c->wr, val | 0x01);
+static void
+i2c_drive_scl(void *data, int state)
+{
+	struct nouveau_i2c_chan *port = data;
+	if (port->type == 0) {
+		u8 val = NVReadVgaCrtc(port->dev, 0, port->drive);
+		if (state) val |= 0x20;
+		else	   val &= 0xdf;
+		NVWriteVgaCrtc(port->dev, 0, port->drive, val | 0x01);
+	} else
+	if (port->type == 4) {
+		nv_mask(port->dev, port->drive, 0x2f, state ? 0x21 : 0x01);
+	} else
+	if (port->type == 5) {
+		if (state) port->state |= 0x01;
+		else	   port->state &= 0xfe;
+		nv_wr32(port->dev, port->drive, 4 | port->state);
+	}
 }
 
 static void
-nv04_i2c_setsda(void *data, int state)
+i2c_drive_sda(void *data, int state)
 {
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
-	uint8_t val;
-
-	val = (NVReadVgaCrtc(dev, 0, i2c->wr) & 0xe0) | (state ? 0x10 : 0);
-	NVWriteVgaCrtc(dev, 0, i2c->wr, val | 0x01);
+	struct nouveau_i2c_chan *port = data;
+	if (port->type == 0) {
+		u8 val = NVReadVgaCrtc(port->dev, 0, port->drive);
+		if (state) val |= 0x10;
+		else	   val &= 0xef;
+		NVWriteVgaCrtc(port->dev, 0, port->drive, val | 0x01);
+	} else
+	if (port->type == 4) {
+		nv_mask(port->dev, port->drive, 0x1f, state ? 0x11 : 0x01);
+	} else
+	if (port->type == 5) {
+		if (state) port->state |= 0x02;
+		else	   port->state &= 0xfd;
+		nv_wr32(port->dev, port->drive, 4 | port->state);
+	}
 }
 
 static int
-nv04_i2c_getscl(void *data)
+i2c_sense_scl(void *data)
 {
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
-
-	return !!(NVReadVgaCrtc(dev, 0, i2c->rd) & 4);
+	struct nouveau_i2c_chan *port = data;
+	struct drm_nouveau_private *dev_priv = port->dev->dev_private;
+	if (port->type == 0) {
+		return !!(NVReadVgaCrtc(port->dev, 0, port->sense) & 0x04);
+	} else
+	if (port->type == 4) {
+		return !!(nv_rd32(port->dev, port->sense) & 0x00040000);
+	} else
+	if (port->type == 5) {
+		if (dev_priv->card_type < NV_D0)
+			return !!(nv_rd32(port->dev, port->sense) & 0x01);
+		else
+			return !!(nv_rd32(port->dev, port->sense) & 0x10);
+	}
+	return 0;
 }
 
 static int
-nv04_i2c_getsda(void *data)
+i2c_sense_sda(void *data)
 {
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
-
-	return !!(NVReadVgaCrtc(dev, 0, i2c->rd) & 8);
+	struct nouveau_i2c_chan *port = data;
+	struct drm_nouveau_private *dev_priv = port->dev->dev_private;
+	if (port->type == 0) {
+		return !!(NVReadVgaCrtc(port->dev, 0, port->sense) & 0x08);
+	} else
+	if (port->type == 4) {
+		return !!(nv_rd32(port->dev, port->sense) & 0x00080000);
+	} else
+	if (port->type == 5) {
+		if (dev_priv->card_type < NV_D0)
+			return !!(nv_rd32(port->dev, port->sense) & 0x02);
+		else
+			return !!(nv_rd32(port->dev, port->sense) & 0x20);
+	}
+	return 0;
 }
 
 static void
-nv4e_i2c_setscl(void *data, int state)
+i2c_delay(struct nouveau_i2c_chan *port, u32 nsec)
 {
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
-	uint8_t val;
+	udelay((nsec + 500) / 1000);
+}
 
-	val = (nv_rd32(dev, i2c->wr) & 0xd0) | (state ? 0x20 : 0);
-	nv_wr32(dev, i2c->wr, val | 0x01);
+static bool
+i2c_raise_scl(struct nouveau_i2c_chan *port)
+{
+	u32 timeout = T_TIMEOUT / T_RISEFALL;
+
+	i2c_drive_scl(port, 1);
+	do {
+		i2c_delay(port, T_RISEFALL);
+	} while (!i2c_sense_scl(port) && --timeout);
+
+	return timeout != 0;
+}
+
+static int
+i2c_start(struct nouveau_i2c_chan *port)
+{
+	int ret = 0;
+
+	port->state  = i2c_sense_scl(port);
+	port->state |= i2c_sense_sda(port) << 1;
+	if (port->state != 3) {
+		i2c_drive_scl(port, 0);
+		i2c_drive_sda(port, 1);
+		if (!i2c_raise_scl(port))
+			ret = -EBUSY;
+	}
+
+	i2c_drive_sda(port, 0);
+	i2c_delay(port, T_HOLD);
+	i2c_drive_scl(port, 0);
+	i2c_delay(port, T_HOLD);
+	return ret;
 }
 
 static void
-nv4e_i2c_setsda(void *data, int state)
+i2c_stop(struct nouveau_i2c_chan *port)
 {
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
-	uint8_t val;
+	i2c_drive_scl(port, 0);
+	i2c_drive_sda(port, 0);
+	i2c_delay(port, T_RISEFALL);
 
-	val = (nv_rd32(dev, i2c->wr) & 0xe0) | (state ? 0x10 : 0);
-	nv_wr32(dev, i2c->wr, val | 0x01);
+	i2c_drive_scl(port, 1);
+	i2c_delay(port, T_HOLD);
+	i2c_drive_sda(port, 1);
+	i2c_delay(port, T_HOLD);
 }
 
 static int
-nv4e_i2c_getscl(void *data)
+i2c_bitw(struct nouveau_i2c_chan *port, int sda)
 {
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
+	i2c_drive_sda(port, sda);
+	i2c_delay(port, T_RISEFALL);
 
-	return !!((nv_rd32(dev, i2c->rd) >> 16) & 4);
+	if (!i2c_raise_scl(port))
+		return -ETIMEDOUT;
+	i2c_delay(port, T_HOLD);
+
+	i2c_drive_scl(port, 0);
+	i2c_delay(port, T_HOLD);
+	return 0;
 }
 
 static int
-nv4e_i2c_getsda(void *data)
+i2c_bitr(struct nouveau_i2c_chan *port)
 {
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
+	int sda;
 
-	return !!((nv_rd32(dev, i2c->rd) >> 16) & 8);
+	i2c_drive_sda(port, 1);
+	i2c_delay(port, T_RISEFALL);
+
+	if (!i2c_raise_scl(port))
+		return -ETIMEDOUT;
+	i2c_delay(port, T_HOLD);
+
+	sda = i2c_sense_sda(port);
+
+	i2c_drive_scl(port, 0);
+	i2c_delay(port, T_HOLD);
+	return sda;
 }
 
+static int
+i2c_get_byte(struct nouveau_i2c_chan *port, u8 *byte, bool last)
+{
+	int i, bit;
+
+	*byte = 0;
+	for (i = 7; i >= 0; i--) {
+		bit = i2c_bitr(port);
+		if (bit < 0)
+			return bit;
+		*byte |= bit << i;
+	}
+
+	return i2c_bitw(port, last ? 1 : 0);
+}
+
+static int
+i2c_put_byte(struct nouveau_i2c_chan *port, u8 byte)
+{
+	int i, ret;
+	for (i = 7; i >= 0; i--) {
+		ret = i2c_bitw(port, !!(byte & (1 << i)));
+		if (ret < 0)
+			return ret;
+	}
+
+	ret = i2c_bitr(port);
+	if (ret == 1) /* nack */
+		ret = -EIO;
+	return ret;
+}
+
+static int
+i2c_addr(struct nouveau_i2c_chan *port, struct i2c_msg *msg)
+{
+	u32 addr = msg->addr << 1;
+	if (msg->flags & I2C_M_RD)
+		addr |= 1;
+	return i2c_put_byte(port, addr);
+}
+
+static int
+i2c_bit_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
+{
+	struct nouveau_i2c_chan *port = (struct nouveau_i2c_chan *)adap;
+	struct i2c_msg *msg = msgs;
+	int ret = 0, mcnt = num;
+
+	while (!ret && mcnt--) {
+		u8 remaining = msg->len;
+		u8 *ptr = msg->buf;
+
+		ret = i2c_start(port);
+		if (ret == 0)
+			ret = i2c_addr(port, msg);
+
+		if (msg->flags & I2C_M_RD) {
+			while (!ret && remaining--)
+				ret = i2c_get_byte(port, ptr++, !remaining);
+		} else {
+			while (!ret && remaining--)
+				ret = i2c_put_byte(port, *ptr++);
+		}
+
+		msg++;
+	}
+
+	i2c_stop(port);
+	return (ret < 0) ? ret : num;
+}
+
+static u32
+i2c_bit_func(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
+}
+
+const struct i2c_algorithm i2c_bit_algo = {
+	.master_xfer = i2c_bit_xfer,
+	.functionality = i2c_bit_func
+};
+
 static const uint32_t nv50_i2c_port[] = {
 	0x00e138, 0x00e150, 0x00e168, 0x00e180,
 	0x00e254, 0x00e274, 0x00e764, 0x00e780,
 	0x00e79c, 0x00e7b8
 };
-#define NV50_I2C_PORTS ARRAY_SIZE(nv50_i2c_port)
 
-static int
-nv50_i2c_getscl(void *data)
+static u8 *
+i2c_table(struct drm_device *dev, u8 *version)
 {
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
+	u8 *dcb = dcb_table(dev), *i2c = NULL;
+	if (dcb) {
+		if (dcb[0] >= 0x15)
+			i2c = ROMPTR(dev, dcb[2]);
+		if (dcb[0] >= 0x30)
+			i2c = ROMPTR(dev, dcb[4]);
+	}
 
-	return !!(nv_rd32(dev, i2c->rd) & 1);
-}
+	/* early revisions had no version number, use dcb version */
+	if (i2c) {
+		*version = dcb[0];
+		if (*version >= 0x30)
+			*version = i2c[0];
+	}
 
-
-static int
-nv50_i2c_getsda(void *data)
-{
-	struct nouveau_i2c_chan *i2c = data;
-	struct drm_device *dev = i2c->dev;
-
-	return !!(nv_rd32(dev, i2c->rd) & 2);
-}
-
-static void
-nv50_i2c_setscl(void *data, int state)
-{
-	struct nouveau_i2c_chan *i2c = data;
-
-	nv_wr32(i2c->dev, i2c->wr, 4 | (i2c->data ? 2 : 0) | (state ? 1 : 0));
-}
-
-static void
-nv50_i2c_setsda(void *data, int state)
-{
-	struct nouveau_i2c_chan *i2c = data;
-
-	nv_mask(i2c->dev, i2c->wr, 0x00000006, 4 | (state ? 2 : 0));
-	i2c->data = state;
-}
-
-static int
-nvd0_i2c_getscl(void *data)
-{
-	struct nouveau_i2c_chan *i2c = data;
-	return !!(nv_rd32(i2c->dev, i2c->rd) & 0x10);
-}
-
-static int
-nvd0_i2c_getsda(void *data)
-{
-	struct nouveau_i2c_chan *i2c = data;
-	return !!(nv_rd32(i2c->dev, i2c->rd) & 0x20);
+	return i2c;
 }
 
 int
-nouveau_i2c_init(struct drm_device *dev, struct dcb_i2c_entry *entry, int index)
+nouveau_i2c_init(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_i2c_chan *i2c;
-	int ret;
+	struct nvbios *bios = &dev_priv->vbios;
+	struct nouveau_i2c_chan *port;
+	u8 *i2c, *entry, legacy[2][4] = {};
+	u8 version, entries, recordlen;
+	int ret, i;
 
-	if (entry->chan)
-		return -EEXIST;
+	INIT_LIST_HEAD(&dev_priv->i2c_ports);
 
-	if (dev_priv->card_type >= NV_50 &&
-	    dev_priv->card_type <= NV_C0 && entry->read >= NV50_I2C_PORTS) {
-		NV_ERROR(dev, "unknown i2c port %d\n", entry->read);
-		return -EINVAL;
+	i2c = i2c_table(dev, &version);
+	if (!i2c) {
+		u8 *bmp = &bios->data[bios->offset];
+		if (bios->type != NVBIOS_BMP)
+			return -ENODEV;
+
+		legacy[0][0] = NV_CIO_CRE_DDC_WR__INDEX;
+		legacy[0][1] = NV_CIO_CRE_DDC_STATUS__INDEX;
+		legacy[1][0] = NV_CIO_CRE_DDC0_WR__INDEX;
+		legacy[1][1] = NV_CIO_CRE_DDC0_STATUS__INDEX;
+
+		/* BMP (from v4.0) has i2c info in the structure, it's in a
+		 * fixed location on earlier VBIOS
+		 */
+		if (bmp[5] < 4)
+			i2c = &bios->data[0x48];
+		else
+			i2c = &bmp[0x36];
+
+		if (i2c[4]) legacy[0][0] = i2c[4];
+		if (i2c[5]) legacy[0][1] = i2c[5];
+		if (i2c[6]) legacy[1][0] = i2c[6];
+		if (i2c[7]) legacy[1][1] = i2c[7];
 	}
 
-	i2c = kzalloc(sizeof(*i2c), GFP_KERNEL);
-	if (i2c == NULL)
-		return -ENOMEM;
-
-	switch (entry->port_type) {
-	case 0:
-		i2c->bit.setsda = nv04_i2c_setsda;
-		i2c->bit.setscl = nv04_i2c_setscl;
-		i2c->bit.getsda = nv04_i2c_getsda;
-		i2c->bit.getscl = nv04_i2c_getscl;
-		i2c->rd = entry->read;
-		i2c->wr = entry->write;
-		break;
-	case 4:
-		i2c->bit.setsda = nv4e_i2c_setsda;
-		i2c->bit.setscl = nv4e_i2c_setscl;
-		i2c->bit.getsda = nv4e_i2c_getsda;
-		i2c->bit.getscl = nv4e_i2c_getscl;
-		i2c->rd = 0x600800 + entry->read;
-		i2c->wr = 0x600800 + entry->write;
-		break;
-	case 5:
-		i2c->bit.setsda = nv50_i2c_setsda;
-		i2c->bit.setscl = nv50_i2c_setscl;
-		if (dev_priv->card_type < NV_D0) {
-			i2c->bit.getsda = nv50_i2c_getsda;
-			i2c->bit.getscl = nv50_i2c_getscl;
-			i2c->rd = nv50_i2c_port[entry->read];
-			i2c->wr = i2c->rd;
-		} else {
-			i2c->bit.getsda = nvd0_i2c_getsda;
-			i2c->bit.getscl = nvd0_i2c_getscl;
-			i2c->rd = 0x00d014 + (entry->read * 0x20);
-			i2c->wr = i2c->rd;
-		}
-		break;
-	case 6:
-		i2c->rd = entry->read;
-		i2c->wr = entry->write;
-		break;
-	default:
-		NV_ERROR(dev, "DCB I2C port type %d unknown\n",
-			 entry->port_type);
-		kfree(i2c);
-		return -EINVAL;
-	}
-
-	snprintf(i2c->adapter.name, sizeof(i2c->adapter.name),
-		 "nouveau-%s-%d", pci_name(dev->pdev), index);
-	i2c->adapter.owner = THIS_MODULE;
-	i2c->adapter.dev.parent = &dev->pdev->dev;
-	i2c->dev = dev;
-	i2c_set_adapdata(&i2c->adapter, i2c);
-
-	if (entry->port_type < 6) {
-		i2c->adapter.algo_data = &i2c->bit;
-		i2c->bit.udelay = 40;
-		i2c->bit.timeout = usecs_to_jiffies(5000);
-		i2c->bit.data = i2c;
-		ret = i2c_bit_add_bus(&i2c->adapter);
+	if (i2c && version >= 0x30) {
+		entry     = i2c[1] + i2c;
+		entries   = i2c[2];
+		recordlen = i2c[3];
+	} else
+	if (i2c) {
+		entry     = i2c;
+		entries   = 16;
+		recordlen = 4;
 	} else {
-		i2c->adapter.algo = &nouveau_dp_i2c_algo;
-		ret = i2c_add_adapter(&i2c->adapter);
+		entry     = legacy[0];
+		entries   = 2;
+		recordlen = 4;
 	}
 
-	if (ret) {
-		NV_ERROR(dev, "Failed to register i2c %d\n", index);
-		kfree(i2c);
-		return ret;
+	for (i = 0; i < entries; i++, entry += recordlen) {
+		port = kzalloc(sizeof(*port), GFP_KERNEL);
+		if (port == NULL) {
+			nouveau_i2c_fini(dev);
+			return -ENOMEM;
+		}
+
+		port->type = entry[3];
+		if (version < 0x30) {
+			port->type &= 0x07;
+			if (port->type == 0x07)
+				port->type = 0xff;
+		}
+
+		if (port->type == 0xff) {
+			kfree(port);
+			continue;
+		}
+
+		switch (port->type) {
+		case 0: /* NV04:NV50 */
+			port->drive = entry[0];
+			port->sense = entry[1];
+			port->adapter.algo = &i2c_bit_algo;
+			break;
+		case 4: /* NV4E */
+			port->drive = 0x600800 + entry[1];
+			port->sense = port->drive;
+			port->adapter.algo = &i2c_bit_algo;
+			break;
+		case 5: /* NV50- */
+			port->drive = entry[0] & 0x0f;
+			if (dev_priv->card_type < NV_D0) {
+				if (port->drive >= ARRAY_SIZE(nv50_i2c_port))
+					break;
+				port->drive = nv50_i2c_port[port->drive];
+				port->sense = port->drive;
+			} else {
+				port->drive = 0x00d014 + (port->drive * 0x20);
+				port->sense = port->drive;
+			}
+			port->adapter.algo = &i2c_bit_algo;
+			break;
+		case 6: /* NV50- DP AUX */
+			port->drive = entry[0];
+			port->sense = port->drive;
+			port->adapter.algo = &nouveau_dp_i2c_algo;
+			break;
+		default:
+			break;
+		}
+
+		if (!port->adapter.algo) {
+			NV_ERROR(dev, "I2C%d: type %d index %x/%x unknown\n",
+				 i, port->type, port->drive, port->sense);
+			kfree(port);
+			continue;
+		}
+
+		snprintf(port->adapter.name, sizeof(port->adapter.name),
+			 "nouveau-%s-%d", pci_name(dev->pdev), i);
+		port->adapter.owner = THIS_MODULE;
+		port->adapter.dev.parent = &dev->pdev->dev;
+		port->dev = dev;
+		port->index = i;
+		port->dcb = ROM32(entry[0]);
+		i2c_set_adapdata(&port->adapter, i2c);
+
+		ret = i2c_add_adapter(&port->adapter);
+		if (ret) {
+			NV_ERROR(dev, "I2C%d: failed register: %d\n", i, ret);
+			kfree(port);
+			continue;
+		}
+
+		list_add_tail(&port->head, &dev_priv->i2c_ports);
 	}
 
-	entry->chan = i2c;
 	return 0;
 }
 
 void
-nouveau_i2c_fini(struct drm_device *dev, struct dcb_i2c_entry *entry)
+nouveau_i2c_fini(struct drm_device *dev)
 {
-	if (!entry->chan)
-		return;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_i2c_chan *port, *tmp;
 
-	i2c_del_adapter(&entry->chan->adapter);
-	kfree(entry->chan);
-	entry->chan = NULL;
+	list_for_each_entry_safe(port, tmp, &dev_priv->i2c_ports, head) {
+		i2c_del_adapter(&port->adapter);
+		kfree(port);
+	}
 }
 
 struct nouveau_i2c_chan *
-nouveau_i2c_find(struct drm_device *dev, int index)
+nouveau_i2c_find(struct drm_device *dev, u8 index)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct dcb_i2c_entry *i2c = &dev_priv->vbios.dcb.i2c[index];
+	struct nouveau_i2c_chan *port;
 
-	if (index >= DCB_MAX_NUM_I2C_ENTRIES)
+	if (index == NV_I2C_DEFAULT(0) ||
+	    index == NV_I2C_DEFAULT(1)) {
+		u8 version, *i2c = i2c_table(dev, &version);
+		if (i2c && version >= 0x30) {
+			if (index == NV_I2C_DEFAULT(0))
+				index = (i2c[4] & 0x0f);
+			else
+				index = (i2c[4] & 0xf0) >> 4;
+		} else {
+			index = 2;
+		}
+	}
+
+	list_for_each_entry(port, &dev_priv->i2c_ports, head) {
+		if (port->index == index)
+			break;
+	}
+
+	if (&port->head == &dev_priv->i2c_ports)
 		return NULL;
 
-	if (dev_priv->card_type >= NV_50 && (i2c->entry & 0x00000100)) {
-		uint32_t reg = 0xe500, val;
-
-		if (i2c->port_type == 6) {
-			reg += i2c->read * 0x50;
+	if (dev_priv->card_type >= NV_50 && (port->dcb & 0x00000100)) {
+		u32 reg = 0x00e500, val;
+		if (port->type == 6) {
+			reg += port->drive * 0x50;
 			val  = 0x2002;
 		} else {
-			reg += ((i2c->entry & 0x1e00) >> 9) * 0x50;
+			reg += ((port->dcb & 0x1e00) >> 9) * 0x50;
 			val  = 0xe001;
 		}
 
@@ -294,9 +497,7 @@
 		nv_mask(dev, reg + 0x00, 0x0000f003, val);
 	}
 
-	if (!i2c->chan && nouveau_i2c_init(dev, i2c, index))
-		return NULL;
-	return i2c->chan;
+	return port;
 }
 
 bool
@@ -331,8 +532,12 @@
 	struct nouveau_i2c_chan *i2c = nouveau_i2c_find(dev, index);
 	int i;
 
-	NV_DEBUG(dev, "Probing %ss on I2C bus: %d\n", what, index);
+	if (!i2c) {
+		NV_DEBUG(dev, "No bus when probing %s on %d\n", what, index);
+		return -ENODEV;
+	}
 
+	NV_DEBUG(dev, "Probing %ss on I2C bus: %d\n", what, i2c->index);
 	for (i = 0; info[i].addr; i++) {
 		if (nouveau_probe_i2c_addr(i2c, info[i].addr) &&
 		    (!match || match(i2c, &info[i]))) {
@@ -342,6 +547,5 @@
 	}
 
 	NV_DEBUG(dev, "No devices found.\n");
-
 	return -ENODEV;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_i2c.h b/drivers/gpu/drm/nouveau/nouveau_i2c.h
index 422b62f..4d2e4e9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_i2c.h
+++ b/drivers/gpu/drm/nouveau/nouveau_i2c.h
@@ -27,20 +27,25 @@
 #include <linux/i2c-algo-bit.h>
 #include "drm_dp_helper.h"
 
-struct dcb_i2c_entry;
+#define NV_I2C_PORT(n)    (0x00 + (n))
+#define NV_I2C_PORT_NUM    0x10
+#define NV_I2C_DEFAULT(n) (0x80 + (n))
 
 struct nouveau_i2c_chan {
 	struct i2c_adapter adapter;
 	struct drm_device *dev;
-	struct i2c_algo_bit_data bit;
-	unsigned rd;
-	unsigned wr;
-	unsigned data;
+	struct list_head head;
+	u8  index;
+	u8  type;
+	u32 dcb;
+	u32 drive;
+	u32 sense;
+	u32 state;
 };
 
-int nouveau_i2c_init(struct drm_device *, struct dcb_i2c_entry *, int index);
-void nouveau_i2c_fini(struct drm_device *, struct dcb_i2c_entry *);
-struct nouveau_i2c_chan *nouveau_i2c_find(struct drm_device *, int index);
+int  nouveau_i2c_init(struct drm_device *);
+void nouveau_i2c_fini(struct drm_device *);
+struct nouveau_i2c_chan *nouveau_i2c_find(struct drm_device *, u8 index);
 bool nouveau_probe_i2c_addr(struct nouveau_i2c_chan *i2c, int addr);
 int nouveau_i2c_identify(struct drm_device *dev, const char *what,
 			 struct i2c_board_info *info,
diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.c b/drivers/gpu/drm/nouveau/nouveau_mem.c
index 37fcaa2..c3a5745 100644
--- a/drivers/gpu/drm/nouveau/nouveau_mem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_mem.c
@@ -644,10 +644,10 @@
 			return;
 
 		if (P.version == 1)
-			hdr = (struct nouveau_pm_tbl_header *) ROMPTR(bios, P.data[4]);
+			hdr = (struct nouveau_pm_tbl_header *) ROMPTR(dev, P.data[4]);
 		else
 		if (P.version == 2)
-			hdr = (struct nouveau_pm_tbl_header *) ROMPTR(bios, P.data[8]);
+			hdr = (struct nouveau_pm_tbl_header *) ROMPTR(dev, P.data[8]);
 		else {
 			NV_WARN(dev, "unknown mem for BIT P %d\n", P.version);
 		}
diff --git a/drivers/gpu/drm/nouveau/nouveau_mxm.c b/drivers/gpu/drm/nouveau/nouveau_mxm.c
new file mode 100644
index 0000000..8bccddf
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nouveau_mxm.c
@@ -0,0 +1,677 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <linux/acpi.h>
+
+#include "drmP.h"
+#include "nouveau_drv.h"
+
+#define MXM_DBG(dev, fmt, args...) NV_DEBUG((dev), "MXM: " fmt, ##args)
+#define MXM_MSG(dev, fmt, args...) NV_INFO((dev), "MXM: " fmt, ##args)
+
+static u8 *
+mxms_data(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	return dev_priv->mxms;
+
+}
+
+static u16
+mxms_version(struct drm_device *dev)
+{
+	u8 *mxms = mxms_data(dev);
+	u16 version = (mxms[4] << 8) | mxms[5];
+	switch (version ) {
+	case 0x0200:
+	case 0x0201:
+	case 0x0300:
+		return version;
+	default:
+		break;
+	}
+
+	MXM_DBG(dev, "unknown version %d.%d\n", mxms[4], mxms[5]);
+	return 0x0000;
+}
+
+static u16
+mxms_headerlen(struct drm_device *dev)
+{
+	return 8;
+}
+
+static u16
+mxms_structlen(struct drm_device *dev)
+{
+	return *(u16 *)&mxms_data(dev)[6];
+}
+
+static bool
+mxms_checksum(struct drm_device *dev)
+{
+	u16 size = mxms_headerlen(dev) + mxms_structlen(dev);
+	u8 *mxms = mxms_data(dev), sum = 0;
+	while (size--)
+		sum += *mxms++;
+	if (sum) {
+		MXM_DBG(dev, "checksum invalid\n");
+		return false;
+	}
+	return true;
+}
+
+static bool
+mxms_valid(struct drm_device *dev)
+{
+	u8 *mxms = mxms_data(dev);
+	if (*(u32 *)mxms != 0x5f4d584d) {
+		MXM_DBG(dev, "signature invalid\n");
+		return false;
+	}
+
+	if (!mxms_version(dev) || !mxms_checksum(dev))
+		return false;
+
+	return true;
+}
+
+static bool
+mxms_foreach(struct drm_device *dev, u8 types,
+	     bool (*exec)(struct drm_device *, u8 *, void *), void *info)
+{
+	u8 *mxms = mxms_data(dev);
+	u8 *desc = mxms + mxms_headerlen(dev);
+	u8 *fini = desc + mxms_structlen(dev) - 1;
+	while (desc < fini) {
+		u8 type = desc[0] & 0x0f;
+		u8 headerlen = 0;
+		u8 recordlen = 0;
+		u8 entries = 0;
+
+		switch (type) {
+		case 0: /* Output Device Structure */
+			if (mxms_version(dev) >= 0x0300)
+				headerlen = 8;
+			else
+				headerlen = 6;
+			break;
+		case 1: /* System Cooling Capability Structure */
+		case 2: /* Thermal Structure */
+		case 3: /* Input Power Structure */
+			headerlen = 4;
+			break;
+		case 4: /* GPIO Device Structure */
+			headerlen = 4;
+			recordlen = 2;
+			entries   = (ROM32(desc[0]) & 0x01f00000) >> 20;
+			break;
+		case 5: /* Vendor Specific Structure */
+			headerlen = 8;
+			break;
+		case 6: /* Backlight Control Structure */
+			if (mxms_version(dev) >= 0x0300) {
+				headerlen = 4;
+				recordlen = 8;
+				entries   = (desc[1] & 0xf0) >> 4;
+			} else {
+				headerlen = 8;
+			}
+			break;
+		case 7: /* Fan Control Structure */
+			headerlen = 8;
+			recordlen = 4;
+			entries   = desc[1] & 0x07;
+			break;
+		default:
+			MXM_DBG(dev, "unknown descriptor type %d\n", type);
+			return false;
+		}
+
+		if ((drm_debug & DRM_UT_DRIVER) && (exec == NULL)) {
+			static const char * mxms_desc_name[] = {
+				"ODS", "SCCS", "TS", "IPS",
+				"GSD", "VSS", "BCS", "FCS",
+			};
+			u8 *dump = desc;
+			int i, j;
+
+			MXM_DBG(dev, "%4s: ", mxms_desc_name[type]);
+			for (j = headerlen - 1; j >= 0; j--)
+				printk("%02x", dump[j]);
+			printk("\n");
+			dump += headerlen;
+
+			for (i = 0; i < entries; i++, dump += recordlen) {
+				MXM_DBG(dev, "      ");
+				for (j = recordlen - 1; j >= 0; j--)
+					printk("%02x", dump[j]);
+				printk("\n");
+			}
+		}
+
+		if (types & (1 << type)) {
+			if (!exec(dev, desc, info))
+				return false;
+		}
+
+		desc += headerlen + (entries * recordlen);
+	}
+
+	return true;
+}
+
+static u8 *
+mxm_table(struct drm_device *dev, u8 *size)
+{
+	struct bit_entry x;
+
+	if (bit_table(dev, 'x', &x)) {
+		MXM_DBG(dev, "BIT 'x' table not present\n");
+		return NULL;
+	}
+
+	if (x.version != 1 || x.length < 3) {
+		MXM_MSG(dev, "BIT x table %d/%d unknown\n",
+			x.version, x.length);
+		return NULL;
+	}
+
+	*size = x.length;
+	return x.data;
+}
+
+/* These map MXM v2.x digital connection values to the appropriate SOR/link,
+ * hopefully they're correct for all boards within the same chipset...
+ *
+ * MXM v3.x VBIOS are nicer and provide pointers to these tables.
+ */
+static u8 nv84_sor_map[16] = {
+	0x00, 0x12, 0x22, 0x11, 0x32, 0x31, 0x11, 0x31,
+	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+
+static u8 nv92_sor_map[16] = {
+	0x00, 0x12, 0x22, 0x11, 0x32, 0x31, 0x11, 0x31,
+	0x11, 0x31, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+
+static u8 nv94_sor_map[16] = {
+	0x00, 0x14, 0x24, 0x11, 0x34, 0x31, 0x11, 0x31,
+	0x11, 0x31, 0x12, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+
+static u8 nv96_sor_map[16] = {
+	0x00, 0x14, 0x24, 0x00, 0x34, 0x00, 0x11, 0x31,
+	0x11, 0x31, 0x12, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+
+static u8 nv98_sor_map[16] = {
+	0x00, 0x14, 0x12, 0x11, 0x00, 0x31, 0x11, 0x31,
+	0x11, 0x31, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00
+};
+
+static u8
+mxm_sor_map(struct drm_device *dev, u8 conn)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	u8 len, *mxm = mxm_table(dev, &len);
+	if (mxm && len >= 6) {
+		u8 *map = ROMPTR(dev, mxm[4]);
+		if (map) {
+			if (map[0] == 0x10) {
+				if (conn < map[3])
+					return map[map[1] + conn];
+				return 0x00;
+			}
+
+			MXM_MSG(dev, "unknown sor map 0x%02x\n", map[0]);
+		}
+	}
+
+	if (dev_priv->chipset == 0x84 || dev_priv->chipset == 0x86)
+		return nv84_sor_map[conn];
+	if (dev_priv->chipset == 0x92)
+		return nv92_sor_map[conn];
+	if (dev_priv->chipset == 0x94)
+		return nv94_sor_map[conn];
+	if (dev_priv->chipset == 0x96)
+		return nv96_sor_map[conn];
+	if (dev_priv->chipset == 0x98)
+		return nv98_sor_map[conn];
+
+	MXM_MSG(dev, "missing sor map\n");
+	return 0x00;
+}
+
+static u8
+mxm_ddc_map(struct drm_device *dev, u8 port)
+{
+	u8 len, *mxm = mxm_table(dev, &len);
+	if (mxm && len >= 8) {
+		u8 *map = ROMPTR(dev, mxm[6]);
+		if (map) {
+			if (map[0] == 0x10) {
+				if (port < map[3])
+					return map[map[1] + port];
+				return 0x00;
+			}
+
+			MXM_MSG(dev, "unknown ddc map 0x%02x\n", map[0]);
+		}
+	}
+
+	/* v2.x: directly write port as dcb i2cidx */
+	return (port << 4) | port;
+}
+
+struct mxms_odev {
+	u8 outp_type;
+	u8 conn_type;
+	u8 ddc_port;
+	u8 dig_conn;
+};
+
+static void
+mxms_output_device(struct drm_device *dev, u8 *pdata, struct mxms_odev *desc)
+{
+	u64 data = ROM32(pdata[0]);
+	if (mxms_version(dev) >= 0x0300)
+		data |= (u64)ROM16(pdata[4]) << 32;
+
+	desc->outp_type = (data & 0x00000000000000f0ULL) >> 4;
+	desc->ddc_port  = (data & 0x0000000000000f00ULL) >> 8;
+	desc->conn_type = (data & 0x000000000001f000ULL) >> 12;
+	desc->dig_conn  = (data & 0x0000000000780000ULL) >> 19;
+}
+
+struct context {
+	u32 *outp;
+	struct mxms_odev desc;
+};
+
+static bool
+mxm_match_tmds_partner(struct drm_device *dev, u8 *data, void *info)
+{
+	struct context *ctx = info;
+	struct mxms_odev desc;
+
+	mxms_output_device(dev, data, &desc);
+	if (desc.outp_type == 2 &&
+	    desc.dig_conn == ctx->desc.dig_conn)
+		return false;
+	return true;
+}
+
+static bool
+mxm_match_dcb(struct drm_device *dev, u8 *data, void *info)
+{
+	struct context *ctx = info;
+	u64 desc = *(u64 *)data;
+
+	mxms_output_device(dev, data, &ctx->desc);
+
+	/* match dcb encoder type to mxm-ods device type */
+	if ((ctx->outp[0] & 0x0000000f) != ctx->desc.outp_type)
+		return true;
+
+	/* digital output, have some extra stuff to match here, there's a
+	 * table in the vbios that provides a mapping from the mxm digital
+	 * connection enum values to SOR/link
+	 */
+	if ((desc & 0x00000000000000f0) >= 0x20) {
+		/* check against sor index */
+		u8 link = mxm_sor_map(dev, ctx->desc.dig_conn);
+		if ((ctx->outp[0] & 0x0f000000) != (link & 0x0f) << 24)
+			return true;
+
+		/* check dcb entry has a compatible link field */
+		link = (link & 0x30) >> 4;
+		if ((link & ((ctx->outp[1] & 0x00000030) >> 4)) != link)
+			return true;
+	}
+
+	/* mark this descriptor accounted for by setting invalid device type,
+	 * except of course some manufactures don't follow specs properly and
+	 * we need to avoid killing off the TMDS function on DP connectors
+	 * if MXM-SIS is missing an entry for it.
+	 */
+	data[0] &= ~0xf0;
+	if (ctx->desc.outp_type == 6 && ctx->desc.conn_type == 6 &&
+	    mxms_foreach(dev, 0x01, mxm_match_tmds_partner, ctx)) {
+		data[0] |= 0x20; /* modify descriptor to match TMDS now */
+	} else {
+		data[0] |= 0xf0;
+	}
+
+	return false;
+}
+
+static int
+mxm_dcb_sanitise_entry(struct drm_device *dev, void *data, int idx, u8 *dcbe)
+{
+	struct context ctx = { .outp = (u32 *)dcbe };
+	u8 type, i2cidx, link;
+	u8 *conn;
+
+	/* look for an output device structure that matches this dcb entry.
+	 * if one isn't found, disable it.
+	 */
+	if (mxms_foreach(dev, 0x01, mxm_match_dcb, &ctx)) {
+		MXM_DBG(dev, "disable %d: 0x%08x 0x%08x\n",
+			idx, ctx.outp[0], ctx.outp[1]);
+		ctx.outp[0] |= 0x0000000f;
+		return 0;
+	}
+
+	/* modify the output's ddc/aux port, there's a pointer to a table
+	 * with the mapping from mxm ddc/aux port to dcb i2c_index in the
+	 * vbios mxm table
+	 */
+	i2cidx = mxm_ddc_map(dev, ctx.desc.ddc_port);
+	if ((ctx.outp[0] & 0x0000000f) != OUTPUT_DP)
+		i2cidx = (i2cidx & 0x0f) << 4;
+	else
+		i2cidx = (i2cidx & 0xf0);
+
+	if (i2cidx != 0xf0) {
+		ctx.outp[0] &= ~0x000000f0;
+		ctx.outp[0] |= i2cidx;
+	}
+
+	/* override dcb sorconf.link, based on what mxm data says */
+	switch (ctx.desc.outp_type) {
+	case 0x00: /* Analog CRT */
+	case 0x01: /* Analog TV/HDTV */
+		break;
+	default:
+		link = mxm_sor_map(dev, ctx.desc.dig_conn) & 0x30;
+		ctx.outp[1] &= ~0x00000030;
+		ctx.outp[1] |= link;
+		break;
+	}
+
+	/* we may need to fixup various other vbios tables based on what
+	 * the descriptor says the connector type should be.
+	 *
+	 * in a lot of cases, the vbios tables will claim DVI-I is possible,
+	 * and the mxm data says the connector is really HDMI.  another
+	 * common example is DP->eDP.
+	 */
+	conn = dcb_conn(dev, (ctx.outp[0] & 0x0000f000) >> 12);
+	type = conn[0];
+	switch (ctx.desc.conn_type) {
+	case 0x01: /* LVDS */
+		ctx.outp[1] |= 0x00000004; /* use_power_scripts */
+		/* XXX: modify default link width in LVDS table */
+		break;
+	case 0x02: /* HDMI */
+		type = DCB_CONNECTOR_HDMI_1;
+		break;
+	case 0x03: /* DVI-D */
+		type = DCB_CONNECTOR_DVI_D;
+		break;
+	case 0x0e: /* eDP, falls through to DPint */
+		ctx.outp[1] |= 0x00010000;
+	case 0x07: /* DP internal, wtf is this?? HP8670w */
+		ctx.outp[1] |= 0x00000004; /* use_power_scripts? */
+		type = DCB_CONNECTOR_eDP;
+		break;
+	default:
+		break;
+	}
+
+	if (mxms_version(dev) >= 0x0300)
+		conn[0] = type;
+
+	return 0;
+}
+
+static bool
+mxm_show_unmatched(struct drm_device *dev, u8 *data, void *info)
+{
+	u64 desc = *(u64 *)data;
+	if ((desc & 0xf0) != 0xf0)
+		MXM_MSG(dev, "unmatched output device 0x%016llx\n", desc);
+	return true;
+}
+
+static void
+mxm_dcb_sanitise(struct drm_device *dev)
+{
+	u8 *dcb = dcb_table(dev);
+	if (!dcb || dcb[0] != 0x40) {
+		MXM_DBG(dev, "unsupported DCB version\n");
+		return;
+	}
+
+	dcb_outp_foreach(dev, NULL, mxm_dcb_sanitise_entry);
+	mxms_foreach(dev, 0x01, mxm_show_unmatched, NULL);
+}
+
+static bool
+mxm_shadow_rom_fetch(struct nouveau_i2c_chan *i2c, u8 addr,
+		     u8 offset, u8 size, u8 *data)
+{
+	struct i2c_msg msgs[] = {
+		{ .addr = addr, .flags = 0, .len = 1, .buf = &offset },
+		{ .addr = addr, .flags = I2C_M_RD, .len = size, .buf = data, },
+	};
+
+	return i2c_transfer(&i2c->adapter, msgs, 2) == 2;
+}
+
+static bool
+mxm_shadow_rom(struct drm_device *dev, u8 version)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_i2c_chan *i2c = NULL;
+	u8 i2cidx, mxms[6], addr, size;
+
+	i2cidx = mxm_ddc_map(dev, 1 /* LVDS_DDC */) & 0x0f;
+	if (i2cidx < 0x0f)
+		i2c = nouveau_i2c_find(dev, i2cidx);
+	if (!i2c)
+		return false;
+
+	addr = 0x54;
+	if (!mxm_shadow_rom_fetch(i2c, addr, 0, 6, mxms)) {
+		addr = 0x56;
+		if (!mxm_shadow_rom_fetch(i2c, addr, 0, 6, mxms))
+			return false;
+	}
+
+	dev_priv->mxms = mxms;
+	size = mxms_headerlen(dev) + mxms_structlen(dev);
+	dev_priv->mxms = kmalloc(size, GFP_KERNEL);
+
+	if (dev_priv->mxms &&
+	    mxm_shadow_rom_fetch(i2c, addr, 0, size, dev_priv->mxms))
+		return true;
+
+	kfree(dev_priv->mxms);
+	dev_priv->mxms = NULL;
+	return false;
+}
+
+#if defined(CONFIG_ACPI)
+static bool
+mxm_shadow_dsm(struct drm_device *dev, u8 version)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	static char muid[] = {
+		0x00, 0xA4, 0x04, 0x40, 0x7D, 0x91, 0xF2, 0x4C,
+		0xB8, 0x9C, 0x79, 0xB6, 0x2F, 0xD5, 0x56, 0x65
+	};
+	u32 mxms_args[] = { 0x00000000 };
+	union acpi_object args[4] = {
+		/* _DSM MUID */
+		{ .buffer.type = 3,
+		  .buffer.length = sizeof(muid),
+		  .buffer.pointer = muid,
+		},
+		/* spec says this can be zero to mean "highest revision", but
+		 * of course there's at least one bios out there which fails
+		 * unless you pass in exactly the version it supports..
+		 */
+		{ .integer.type = ACPI_TYPE_INTEGER,
+		  .integer.value = (version & 0xf0) << 4 | (version & 0x0f),
+		},
+		/* MXMS function */
+		{ .integer.type = ACPI_TYPE_INTEGER,
+		  .integer.value = 0x00000010,
+		},
+		/* Pointer to MXMS arguments */
+		{ .buffer.type = ACPI_TYPE_BUFFER,
+		  .buffer.length = sizeof(mxms_args),
+		  .buffer.pointer = (char *)mxms_args,
+		},
+	};
+	struct acpi_object_list list = { ARRAY_SIZE(args), args };
+	struct acpi_buffer retn = { ACPI_ALLOCATE_BUFFER, NULL };
+	union acpi_object *obj;
+	acpi_handle handle;
+	int ret;
+
+	handle = DEVICE_ACPI_HANDLE(&dev->pdev->dev);
+	if (!handle)
+		return false;
+
+	ret = acpi_evaluate_object(handle, "_DSM", &list, &retn);
+	if (ret) {
+		MXM_DBG(dev, "DSM MXMS failed: %d\n", ret);
+		return false;
+	}
+
+	obj = retn.pointer;
+	if (obj->type == ACPI_TYPE_BUFFER) {
+		dev_priv->mxms = kmemdup(obj->buffer.pointer,
+					 obj->buffer.length, GFP_KERNEL);
+	} else
+	if (obj->type == ACPI_TYPE_INTEGER) {
+		MXM_DBG(dev, "DSM MXMS returned 0x%llx\n", obj->integer.value);
+	}
+
+	kfree(obj);
+	return dev_priv->mxms != NULL;
+}
+#endif
+
+#if defined(CONFIG_ACPI_WMI) || defined(CONFIG_ACPI_WMI_MODULE)
+
+#define WMI_WMMX_GUID "F6CB5C3C-9CAE-4EBD-B577-931EA32A2CC0"
+
+static bool
+mxm_shadow_wmi(struct drm_device *dev, u8 version)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	u32 mxms_args[] = { 0x534D584D /* MXMS */, version, 0 };
+	struct acpi_buffer args = { sizeof(mxms_args), mxms_args };
+	struct acpi_buffer retn = { ACPI_ALLOCATE_BUFFER, NULL };
+	union acpi_object *obj;
+	acpi_status status;
+
+	if (!wmi_has_guid(WMI_WMMX_GUID))
+		return false;
+
+	status = wmi_evaluate_method(WMI_WMMX_GUID, 0, 0, &args, &retn);
+	if (ACPI_FAILURE(status)) {
+		MXM_DBG(dev, "WMMX MXMS returned %d\n", status);
+		return false;
+	}
+
+	obj = retn.pointer;
+	if (obj->type == ACPI_TYPE_BUFFER) {
+		dev_priv->mxms = kmemdup(obj->buffer.pointer,
+					 obj->buffer.length, GFP_KERNEL);
+	}
+
+	kfree(obj);
+	return dev_priv->mxms != NULL;
+}
+#endif
+
+struct mxm_shadow_h {
+	const char *name;
+	bool (*exec)(struct drm_device *, u8 version);
+} _mxm_shadow[] = {
+	{ "ROM", mxm_shadow_rom },
+#if defined(CONFIG_ACPI)
+	{ "DSM", mxm_shadow_dsm },
+#endif
+#if defined(CONFIG_ACPI_WMI) || defined(CONFIG_ACPI_WMI_MODULE)
+	{ "WMI", mxm_shadow_wmi },
+#endif
+	{}
+};
+
+static int
+mxm_shadow(struct drm_device *dev, u8 version)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct mxm_shadow_h *shadow = _mxm_shadow;
+	do {
+		MXM_DBG(dev, "checking %s\n", shadow->name);
+		if (shadow->exec(dev, version)) {
+			if (mxms_valid(dev))
+				return 0;
+			kfree(dev_priv->mxms);
+			dev_priv->mxms = NULL;
+		}
+	} while ((++shadow)->name);
+	return -ENOENT;
+}
+
+int
+nouveau_mxm_init(struct drm_device *dev)
+{
+	u8 mxm_size, *mxm = mxm_table(dev, &mxm_size);
+	if (!mxm || !mxm[0]) {
+		MXM_MSG(dev, "no VBIOS data, nothing to do\n");
+		return 0;
+	}
+
+	MXM_MSG(dev, "BIOS version %d.%d\n", mxm[0] >> 4, mxm[0] & 0x0f);
+
+	if (mxm_shadow(dev, mxm[0])) {
+		MXM_MSG(dev, "failed to locate valid SIS\n");
+		return -EINVAL;
+	}
+
+	MXM_MSG(dev, "MXMS Version %d.%d\n",
+		mxms_version(dev) >> 8, mxms_version(dev) & 0xff);
+	mxms_foreach(dev, 0, NULL, NULL);
+
+	if (nouveau_mxmdcb)
+		mxm_dcb_sanitise(dev);
+	return 0;
+}
+
+void
+nouveau_mxm_fini(struct drm_device *dev)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	kfree(dev_priv->mxms);
+	dev_priv->mxms = NULL;
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_notifier.c b/drivers/gpu/drm/nouveau/nouveau_notifier.c
index 6abdbe6..2ef883c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_notifier.c
+++ b/drivers/gpu/drm/nouveau/nouveau_notifier.c
@@ -115,7 +115,7 @@
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_gpuobj *nobj = NULL;
 	struct drm_mm_node *mem;
-	uint32_t offset;
+	uint64_t offset;
 	int target, ret;
 
 	mem = drm_mm_search_free_in_range(&chan->notifier_heap, size, 0,
diff --git a/drivers/gpu/drm/nouveau/nouveau_object.c b/drivers/gpu/drm/nouveau/nouveau_object.c
index 02222c5..cc419fae 100644
--- a/drivers/gpu/drm/nouveau/nouveau_object.c
+++ b/drivers/gpu/drm/nouveau/nouveau_object.c
@@ -680,7 +680,7 @@
 		return ret;
 	}
 
-	ret = drm_mm_init(&chan->ramin_heap, base, size);
+	ret = drm_mm_init(&chan->ramin_heap, base, size - base);
 	if (ret) {
 		NV_ERROR(dev, "Error creating PRAMIN heap: %d\n", ret);
 		nouveau_gpuobj_ref(NULL, &chan->ramin);
@@ -723,14 +723,14 @@
 	nv_wo32(chan->ramin, 0x020c, 0x000000ff);
 
 	/* map display semaphore buffers into channel's vm */
-	if (dev_priv->card_type >= NV_D0)
-		return 0;
+	for (i = 0; i < dev->mode_config.num_crtc; i++) {
+		struct nouveau_bo *bo;
+		if (dev_priv->card_type >= NV_D0)
+			bo = nvd0_display_crtc_sema(dev, i);
+		else
+			bo = nv50_display(dev)->crtc[i].sem.bo;
 
-	for (i = 0; i < 2; i++) {
-		struct nv50_display_crtc *dispc = &nv50_display(dev)->crtc[i];
-
-		ret = nouveau_bo_vma_add(dispc->sem.bo, chan->vm,
-					 &chan->dispc_vma[i]);
+		ret = nouveau_bo_vma_add(bo, chan->vm, &chan->dispc_vma[i]);
 		if (ret)
 			return ret;
 	}
@@ -879,9 +879,14 @@
 
 	NV_DEBUG(dev, "ch%d\n", chan->id);
 
-	if (dev_priv->card_type >= NV_50 && dev_priv->card_type <= NV_C0) {
+	if (dev_priv->card_type >= NV_D0) {
+		for (i = 0; i < dev->mode_config.num_crtc; i++) {
+			struct nouveau_bo *bo = nvd0_display_crtc_sema(dev, i);
+			nouveau_bo_vma_del(bo, &chan->dispc_vma[i]);
+		}
+	} else
+	if (dev_priv->card_type >= NV_50) {
 		struct nv50_display *disp = nv50_display(dev);
-
 		for (i = 0; i < dev->mode_config.num_crtc; i++) {
 			struct nv50_display_crtc *dispc = &disp->crtc[i];
 			nouveau_bo_vma_del(dispc->sem.bo, &chan->dispc_vma[i]);
diff --git a/drivers/gpu/drm/nouveau/nouveau_perf.c b/drivers/gpu/drm/nouveau/nouveau_perf.c
index 9f178aa..58f4973 100644
--- a/drivers/gpu/drm/nouveau/nouveau_perf.c
+++ b/drivers/gpu/drm/nouveau/nouveau_perf.c
@@ -41,7 +41,7 @@
 		return;
 	}
 
-	perf = ROMPTR(bios, bmp[0x73]);
+	perf = ROMPTR(dev, bmp[0x73]);
 	if (!perf) {
 		NV_DEBUG(dev, "No memclock table pointer found.\n");
 		return;
@@ -87,7 +87,7 @@
 	 * ramcfg to select the correct subentry
 	 */
 	if (P->version == 2) {
-		u8 *tmap = ROMPTR(bios, P->data[4]);
+		u8 *tmap = ROMPTR(dev, P->data[4]);
 		if (!tmap) {
 			NV_DEBUG(dev, "no timing map pointer\n");
 			return NULL;
@@ -140,7 +140,6 @@
 		     struct nouveau_pm_level *perflvl)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nvbios *bios = &dev_priv->vbios;
 	u8 *vmap;
 	int id;
 
@@ -165,7 +164,7 @@
 		return;
 	}
 
-	vmap = ROMPTR(bios, P->data[32]);
+	vmap = ROMPTR(dev, P->data[32]);
 	if (!vmap) {
 		NV_DEBUG(dev, "volt map table pointer invalid\n");
 		return;
@@ -200,12 +199,14 @@
 			return;
 		}
 
-		perf = ROMPTR(bios, P.data[0]);
+		perf = ROMPTR(dev, P.data[0]);
 		version   = perf[0];
 		headerlen = perf[1];
 		if (version < 0x40) {
 			recordlen = perf[3] + (perf[4] * perf[5]);
 			entries   = perf[2];
+
+			pm->pwm_divisor = ROM16(perf[6]);
 		} else {
 			recordlen = perf[2] + (perf[3] * perf[4]);
 			entries   = perf[5];
@@ -216,7 +217,7 @@
 			return;
 		}
 
-		perf = ROMPTR(bios, bios->data[bios->offset + 0x94]);
+		perf = ROMPTR(dev, bios->data[bios->offset + 0x94]);
 		if (!perf) {
 			NV_DEBUG(dev, "perf table pointer invalid\n");
 			return;
@@ -239,7 +240,7 @@
 	if(version == 0x15) {
 		memtimings->timing =
 				kcalloc(entries, sizeof(*memtimings->timing), GFP_KERNEL);
-		if(!memtimings) {
+		if (!memtimings->timing) {
 			NV_WARN(dev,"Could not allocate memtiming table\n");
 			return;
 		}
@@ -283,7 +284,6 @@
 				perflvl->memory = ROM16(entry[11]) * 1000;
 			else
 				perflvl->memory = ROM16(entry[11]) * 2000;
-
 			break;
 		case 0x25:
 			perflvl->fanspeed = entry[4];
@@ -300,8 +300,8 @@
 			perflvl->core = ROM16(entry[8]) * 1000;
 			perflvl->shader = ROM16(entry[10]) * 1000;
 			perflvl->memory = ROM16(entry[12]) * 1000;
-			/*XXX: confirm on 0x35 */
-			perflvl->unk05 = ROM16(entry[16]) * 1000;
+			perflvl->vdec = ROM16(entry[16]) * 1000;
+			perflvl->dom6 = ROM16(entry[20]) * 1000;
 			break;
 		case 0x40:
 #define subent(n) (ROM16(entry[perf[2] + ((n) * perf[3])]) & 0xfff) * 1000
diff --git a/drivers/gpu/drm/nouveau/nouveau_pm.c b/drivers/gpu/drm/nouveau/nouveau_pm.c
index a539fd2..aba3362 100644
--- a/drivers/gpu/drm/nouveau/nouveau_pm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_pm.c
@@ -26,6 +26,7 @@
 
 #include "nouveau_drv.h"
 #include "nouveau_pm.h"
+#include "nouveau_gpio.h"
 
 #ifdef CONFIG_ACPI
 #include <linux/acpi.h>
@@ -35,22 +36,95 @@
 #include <linux/hwmon-sysfs.h>
 
 static int
-nouveau_pm_clock_set(struct drm_device *dev, struct nouveau_pm_level *perflvl,
-		     u8 id, u32 khz)
+nouveau_pwmfan_get(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
-	void *pre_state;
+	struct gpio_func gpio;
+	u32 divs, duty;
+	int ret;
 
-	if (khz == 0)
-		return 0;
+	if (!pm->pwm_get)
+		return -ENODEV;
 
-	pre_state = pm->clock_pre(dev, perflvl, id, khz);
-	if (IS_ERR(pre_state))
-		return PTR_ERR(pre_state);
+	ret = nouveau_gpio_find(dev, 0, DCB_GPIO_PWM_FAN, 0xff, &gpio);
+	if (ret == 0) {
+		ret = pm->pwm_get(dev, gpio.line, &divs, &duty);
+		if (ret == 0) {
+			divs = max(divs, duty);
+			if (dev_priv->card_type <= NV_40 || (gpio.log[0] & 1))
+				duty = divs - duty;
+			return (duty * 100) / divs;
+		}
 
-	if (pre_state)
-		pm->clock_set(dev, pre_state);
+		return nouveau_gpio_func_get(dev, gpio.func) * 100;
+	}
+
+	return -ENODEV;
+}
+
+static int
+nouveau_pwmfan_set(struct drm_device *dev, int percent)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
+	struct gpio_func gpio;
+	u32 divs, duty;
+	int ret;
+
+	if (!pm->pwm_set)
+		return -ENODEV;
+
+	ret = nouveau_gpio_find(dev, 0, DCB_GPIO_PWM_FAN, 0xff, &gpio);
+	if (ret == 0) {
+		divs = pm->pwm_divisor;
+		if (pm->fan.pwm_freq) {
+			/*XXX: PNVIO clock more than likely... */
+			divs = 135000 / pm->fan.pwm_freq;
+			if (dev_priv->chipset < 0xa3)
+				divs /= 4;
+		}
+
+		duty = ((divs * percent) + 99) / 100;
+		if (dev_priv->card_type <= NV_40 || (gpio.log[0] & 1))
+			duty = divs - duty;
+
+		return pm->pwm_set(dev, gpio.line, divs, duty);
+	}
+
+	return -ENODEV;
+}
+
+static int
+nouveau_pm_perflvl_aux(struct drm_device *dev, struct nouveau_pm_level *perflvl,
+		       struct nouveau_pm_level *a, struct nouveau_pm_level *b)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
+	int ret;
+
+	/*XXX: not on all boards, we should control based on temperature
+	 *     on recent boards..  or maybe on some other factor we don't
+	 *     know about?
+	 */
+	if (a->fanspeed && b->fanspeed && b->fanspeed > a->fanspeed) {
+		ret = nouveau_pwmfan_set(dev, perflvl->fanspeed);
+		if (ret && ret != -ENODEV) {
+			NV_ERROR(dev, "fanspeed set failed: %d\n", ret);
+			return ret;
+		}
+	}
+
+	if (pm->voltage.supported && pm->voltage_set) {
+		if (perflvl->volt_min && b->volt_min > a->volt_min) {
+			ret = pm->voltage_set(dev, perflvl->volt_min);
+			if (ret) {
+				NV_ERROR(dev, "voltage set failed: %d\n", ret);
+				return ret;
+			}
+		}
+	}
+
 	return 0;
 }
 
@@ -59,31 +133,24 @@
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
+	void *state;
 	int ret;
 
 	if (perflvl == pm->cur)
 		return 0;
 
-	if (pm->voltage.supported && pm->voltage_set && perflvl->volt_min) {
-		ret = pm->voltage_set(dev, perflvl->volt_min);
-		if (ret) {
-			NV_ERROR(dev, "voltage_set %d failed: %d\n",
-				 perflvl->volt_min, ret);
-		}
-	}
+	ret = nouveau_pm_perflvl_aux(dev, perflvl, pm->cur, perflvl);
+	if (ret)
+		return ret;
 
-	if (pm->clocks_pre) {
-		void *state = pm->clocks_pre(dev, perflvl);
-		if (IS_ERR(state))
-			return PTR_ERR(state);
-		pm->clocks_set(dev, state);
-	} else
-	if (pm->clock_set) {
-		nouveau_pm_clock_set(dev, perflvl, PLL_CORE, perflvl->core);
-		nouveau_pm_clock_set(dev, perflvl, PLL_SHADER, perflvl->shader);
-		nouveau_pm_clock_set(dev, perflvl, PLL_MEMORY, perflvl->memory);
-		nouveau_pm_clock_set(dev, perflvl, PLL_UNK05, perflvl->unk05);
-	}
+	state = pm->clocks_pre(dev, perflvl);
+	if (IS_ERR(state))
+		return PTR_ERR(state);
+	pm->clocks_set(dev, state);
+
+	ret = nouveau_pm_perflvl_aux(dev, perflvl, perflvl, pm->cur);
+	if (ret)
+		return ret;
 
 	pm->cur = perflvl;
 	return 0;
@@ -130,28 +197,9 @@
 
 	memset(perflvl, 0, sizeof(*perflvl));
 
-	if (pm->clocks_get) {
-		ret = pm->clocks_get(dev, perflvl);
-		if (ret)
-			return ret;
-	} else
-	if (pm->clock_get) {
-		ret = pm->clock_get(dev, PLL_CORE);
-		if (ret > 0)
-			perflvl->core = ret;
-
-		ret = pm->clock_get(dev, PLL_MEMORY);
-		if (ret > 0)
-			perflvl->memory = ret;
-
-		ret = pm->clock_get(dev, PLL_SHADER);
-		if (ret > 0)
-			perflvl->shader = ret;
-
-		ret = pm->clock_get(dev, PLL_UNK05);
-		if (ret > 0)
-			perflvl->unk05 = ret;
-	}
+	ret = pm->clocks_get(dev, perflvl);
+	if (ret)
+		return ret;
 
 	if (pm->voltage.supported && pm->voltage_get) {
 		ret = pm->voltage_get(dev);
@@ -161,6 +209,10 @@
 		}
 	}
 
+	ret = nouveau_pwmfan_get(dev);
+	if (ret > 0)
+		perflvl->fanspeed = ret;
+
 	return 0;
 }
 
@@ -412,6 +464,172 @@
 						nouveau_hwmon_show_update_rate,
 						NULL, 0);
 
+static ssize_t
+nouveau_hwmon_show_fan0_input(struct device *d, struct device_attribute *attr,
+			      char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_timer_engine *ptimer = &dev_priv->engine.timer;
+	struct gpio_func gpio;
+	u32 cycles, cur, prev;
+	u64 start;
+	int ret;
+
+	ret = nouveau_gpio_find(dev, 0, DCB_GPIO_FAN_SENSE, 0xff, &gpio);
+	if (ret)
+		return ret;
+
+	/* Monitor the GPIO input 0x3b for 250ms.
+	 * When the fan spins, it changes the value of GPIO FAN_SENSE.
+	 * We get 4 changes (0 -> 1 -> 0 -> 1 -> [...]) per complete rotation.
+	 */
+	start = ptimer->read(dev);
+	prev = nouveau_gpio_sense(dev, 0, gpio.line);
+	cycles = 0;
+	do {
+		cur = nouveau_gpio_sense(dev, 0, gpio.line);
+		if (prev != cur) {
+			cycles++;
+			prev = cur;
+		}
+
+		usleep_range(500, 1000); /* supports 0 < rpm < 7500 */
+	} while (ptimer->read(dev) - start < 250000000);
+
+	/* interpolate to get rpm */
+	return sprintf(buf, "%i\n", cycles / 4 * 4 * 60);
+}
+static SENSOR_DEVICE_ATTR(fan0_input, S_IRUGO, nouveau_hwmon_show_fan0_input,
+			  NULL, 0);
+
+static ssize_t
+nouveau_hwmon_get_pwm0(struct device *d, struct device_attribute *a, char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	int ret;
+
+	ret = nouveau_pwmfan_get(dev);
+	if (ret < 0)
+		return ret;
+
+	return sprintf(buf, "%i\n", ret);
+}
+
+static ssize_t
+nouveau_hwmon_set_pwm0(struct device *d, struct device_attribute *a,
+		       const char *buf, size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
+	int ret = -ENODEV;
+	long value;
+
+	if (nouveau_perflvl_wr != 7777)
+		return -EPERM;
+
+	if (strict_strtol(buf, 10, &value) == -EINVAL)
+		return -EINVAL;
+
+	if (value < pm->fan.min_duty)
+		value = pm->fan.min_duty;
+	if (value > pm->fan.max_duty)
+		value = pm->fan.max_duty;
+
+	ret = nouveau_pwmfan_set(dev, value);
+	if (ret)
+		return ret;
+
+	return count;
+}
+
+static SENSOR_DEVICE_ATTR(pwm0, S_IRUGO | S_IWUSR,
+			  nouveau_hwmon_get_pwm0,
+			  nouveau_hwmon_set_pwm0, 0);
+
+static ssize_t
+nouveau_hwmon_get_pwm0_min(struct device *d,
+			   struct device_attribute *a, char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
+
+	return sprintf(buf, "%i\n", pm->fan.min_duty);
+}
+
+static ssize_t
+nouveau_hwmon_set_pwm0_min(struct device *d, struct device_attribute *a,
+			   const char *buf, size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
+	long value;
+
+	if (strict_strtol(buf, 10, &value) == -EINVAL)
+		return -EINVAL;
+
+	if (value < 0)
+		value = 0;
+
+	if (pm->fan.max_duty - value < 10)
+		value = pm->fan.max_duty - 10;
+
+	if (value < 10)
+		pm->fan.min_duty = 10;
+	else
+		pm->fan.min_duty = value;
+
+	return count;
+}
+
+static SENSOR_DEVICE_ATTR(pwm0_min, S_IRUGO | S_IWUSR,
+			  nouveau_hwmon_get_pwm0_min,
+			  nouveau_hwmon_set_pwm0_min, 0);
+
+static ssize_t
+nouveau_hwmon_get_pwm0_max(struct device *d,
+			   struct device_attribute *a, char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
+
+	return sprintf(buf, "%i\n", pm->fan.max_duty);
+}
+
+static ssize_t
+nouveau_hwmon_set_pwm0_max(struct device *d, struct device_attribute *a,
+			   const char *buf, size_t count)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
+	long value;
+
+	if (strict_strtol(buf, 10, &value) == -EINVAL)
+		return -EINVAL;
+
+	if (value < 0)
+		value = 0;
+
+	if (value - pm->fan.min_duty < 10)
+		value = pm->fan.min_duty + 10;
+
+	if (value > 100)
+		pm->fan.max_duty = 100;
+	else
+		pm->fan.max_duty = value;
+
+	return count;
+}
+
+static SENSOR_DEVICE_ATTR(pwm0_max, S_IRUGO | S_IWUSR,
+			  nouveau_hwmon_get_pwm0_max,
+			  nouveau_hwmon_set_pwm0_max, 0);
+
 static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_temp1_input.dev_attr.attr,
 	&sensor_dev_attr_temp1_max.dev_attr.attr,
@@ -420,10 +638,26 @@
 	&sensor_dev_attr_update_rate.dev_attr.attr,
 	NULL
 };
+static struct attribute *hwmon_fan_rpm_attributes[] = {
+	&sensor_dev_attr_fan0_input.dev_attr.attr,
+	NULL
+};
+static struct attribute *hwmon_pwm_fan_attributes[] = {
+	&sensor_dev_attr_pwm0.dev_attr.attr,
+	&sensor_dev_attr_pwm0_min.dev_attr.attr,
+	&sensor_dev_attr_pwm0_max.dev_attr.attr,
+	NULL
+};
 
 static const struct attribute_group hwmon_attrgroup = {
 	.attrs = hwmon_attributes,
 };
+static const struct attribute_group hwmon_fan_rpm_attrgroup = {
+	.attrs = hwmon_fan_rpm_attributes,
+};
+static const struct attribute_group hwmon_pwm_fan_attrgroup = {
+	.attrs = hwmon_pwm_fan_attributes,
+};
 #endif
 
 static int
@@ -433,7 +667,7 @@
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_pm_engine *pm = &dev_priv->engine.pm;
 	struct device *hwmon_dev;
-	int ret;
+	int ret = 0;
 
 	if (!pm->temp_get)
 		return -ENODEV;
@@ -446,17 +680,46 @@
 		return ret;
 	}
 	dev_set_drvdata(hwmon_dev, dev);
+
+	/* default sysfs entries */
 	ret = sysfs_create_group(&dev->pdev->dev.kobj, &hwmon_attrgroup);
 	if (ret) {
-		NV_ERROR(dev,
-			"Unable to create hwmon sysfs file: %d\n", ret);
-		hwmon_device_unregister(hwmon_dev);
-		return ret;
+		if (ret)
+			goto error;
+	}
+
+	/* if the card has a pwm fan */
+	/*XXX: incorrect, need better detection for this, some boards have
+	 *     the gpio entries for pwm fan control even when there's no
+	 *     actual fan connected to it... therm table? */
+	if (nouveau_pwmfan_get(dev) >= 0) {
+		ret = sysfs_create_group(&dev->pdev->dev.kobj,
+					 &hwmon_pwm_fan_attrgroup);
+		if (ret)
+			goto error;
+	}
+
+	/* if the card can read the fan rpm */
+	if (nouveau_gpio_func_valid(dev, DCB_GPIO_FAN_SENSE)) {
+		ret = sysfs_create_group(&dev->pdev->dev.kobj,
+					 &hwmon_fan_rpm_attrgroup);
+		if (ret)
+			goto error;
 	}
 
 	pm->hwmon = hwmon_dev;
-#endif
+
 	return 0;
+
+error:
+	NV_ERROR(dev, "Unable to create some hwmon sysfs files: %d\n", ret);
+	hwmon_device_unregister(hwmon_dev);
+	pm->hwmon = NULL;
+	return ret;
+#else
+	pm->hwmon = NULL;
+	return 0;
+#endif
 }
 
 static void
@@ -468,6 +731,9 @@
 
 	if (pm->hwmon) {
 		sysfs_remove_group(&dev->pdev->dev.kobj, &hwmon_attrgroup);
+		sysfs_remove_group(&dev->pdev->dev.kobj, &hwmon_pwm_fan_attrgroup);
+		sysfs_remove_group(&dev->pdev->dev.kobj, &hwmon_fan_rpm_attrgroup);
+
 		hwmon_device_unregister(pm->hwmon);
 	}
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_pm.h b/drivers/gpu/drm/nouveau/nouveau_pm.h
index 8ac02cd..2f8e14f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_pm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_pm.h
@@ -47,29 +47,33 @@
 void nouveau_mem_timing_fini(struct drm_device *);
 
 /* nv04_pm.c */
-int nv04_pm_clock_get(struct drm_device *, u32 id);
-void *nv04_pm_clock_pre(struct drm_device *, struct nouveau_pm_level *,
-			u32 id, int khz);
-void nv04_pm_clock_set(struct drm_device *, void *);
+int nv04_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
+void *nv04_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
+int nv04_pm_clocks_set(struct drm_device *, void *);
 
 /* nv40_pm.c */
 int nv40_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
 void *nv40_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
-void nv40_pm_clocks_set(struct drm_device *, void *);
+int nv40_pm_clocks_set(struct drm_device *, void *);
+int nv40_pm_pwm_get(struct drm_device *, int, u32 *, u32 *);
+int nv40_pm_pwm_set(struct drm_device *, int, u32, u32);
 
 /* nv50_pm.c */
-int nv50_pm_clock_get(struct drm_device *, u32 id);
-void *nv50_pm_clock_pre(struct drm_device *, struct nouveau_pm_level *,
-			u32 id, int khz);
-void nv50_pm_clock_set(struct drm_device *, void *);
+int nv50_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
+void *nv50_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
+int nv50_pm_clocks_set(struct drm_device *, void *);
+int nv50_pm_pwm_get(struct drm_device *, int, u32 *, u32 *);
+int nv50_pm_pwm_set(struct drm_device *, int, u32, u32);
 
 /* nva3_pm.c */
 int nva3_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
 void *nva3_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
-void nva3_pm_clocks_set(struct drm_device *, void *);
+int nva3_pm_clocks_set(struct drm_device *, void *);
 
 /* nvc0_pm.c */
 int nvc0_pm_clocks_get(struct drm_device *, struct nouveau_pm_level *);
+void *nvc0_pm_clocks_pre(struct drm_device *, struct nouveau_pm_level *);
+int nvc0_pm_clocks_set(struct drm_device *, void *);
 
 /* nouveau_temp.c */
 void nouveau_temp_init(struct drm_device *dev);
diff --git a/drivers/gpu/drm/nouveau/nouveau_state.c b/drivers/gpu/drm/nouveau/nouveau_state.c
index 82478e0..f5e9891 100644
--- a/drivers/gpu/drm/nouveau/nouveau_state.c
+++ b/drivers/gpu/drm/nouveau/nouveau_state.c
@@ -36,6 +36,7 @@
 #include "nouveau_drm.h"
 #include "nouveau_fbcon.h"
 #include "nouveau_ramht.h"
+#include "nouveau_gpio.h"
 #include "nouveau_pm.h"
 #include "nv50_display.h"
 
@@ -80,16 +81,12 @@
 		engine->display.early_init	= nv04_display_early_init;
 		engine->display.late_takedown	= nv04_display_late_takedown;
 		engine->display.create		= nv04_display_create;
-		engine->display.init		= nv04_display_init;
 		engine->display.destroy		= nv04_display_destroy;
-		engine->gpio.init		= nouveau_stub_init;
-		engine->gpio.takedown		= nouveau_stub_takedown;
-		engine->gpio.get		= NULL;
-		engine->gpio.set		= NULL;
-		engine->gpio.irq_enable		= NULL;
-		engine->pm.clock_get		= nv04_pm_clock_get;
-		engine->pm.clock_pre		= nv04_pm_clock_pre;
-		engine->pm.clock_set		= nv04_pm_clock_set;
+		engine->display.init		= nv04_display_init;
+		engine->display.fini		= nv04_display_fini;
+		engine->pm.clocks_get		= nv04_pm_clocks_get;
+		engine->pm.clocks_pre		= nv04_pm_clocks_pre;
+		engine->pm.clocks_set		= nv04_pm_clocks_set;
 		engine->vram.init		= nouveau_mem_detect;
 		engine->vram.takedown		= nouveau_stub_takedown;
 		engine->vram.flags_valid	= nouveau_mem_flags_valid;
@@ -129,16 +126,14 @@
 		engine->display.early_init	= nv04_display_early_init;
 		engine->display.late_takedown	= nv04_display_late_takedown;
 		engine->display.create		= nv04_display_create;
-		engine->display.init		= nv04_display_init;
 		engine->display.destroy		= nv04_display_destroy;
-		engine->gpio.init		= nouveau_stub_init;
-		engine->gpio.takedown		= nouveau_stub_takedown;
-		engine->gpio.get		= nv10_gpio_get;
-		engine->gpio.set		= nv10_gpio_set;
-		engine->gpio.irq_enable		= NULL;
-		engine->pm.clock_get		= nv04_pm_clock_get;
-		engine->pm.clock_pre		= nv04_pm_clock_pre;
-		engine->pm.clock_set		= nv04_pm_clock_set;
+		engine->display.init		= nv04_display_init;
+		engine->display.fini		= nv04_display_fini;
+		engine->gpio.drive		= nv10_gpio_drive;
+		engine->gpio.sense		= nv10_gpio_sense;
+		engine->pm.clocks_get		= nv04_pm_clocks_get;
+		engine->pm.clocks_pre		= nv04_pm_clocks_pre;
+		engine->pm.clocks_set		= nv04_pm_clocks_set;
 		engine->vram.init		= nouveau_mem_detect;
 		engine->vram.takedown		= nouveau_stub_takedown;
 		engine->vram.flags_valid	= nouveau_mem_flags_valid;
@@ -178,16 +173,14 @@
 		engine->display.early_init	= nv04_display_early_init;
 		engine->display.late_takedown	= nv04_display_late_takedown;
 		engine->display.create		= nv04_display_create;
-		engine->display.init		= nv04_display_init;
 		engine->display.destroy		= nv04_display_destroy;
-		engine->gpio.init		= nouveau_stub_init;
-		engine->gpio.takedown		= nouveau_stub_takedown;
-		engine->gpio.get		= nv10_gpio_get;
-		engine->gpio.set		= nv10_gpio_set;
-		engine->gpio.irq_enable		= NULL;
-		engine->pm.clock_get		= nv04_pm_clock_get;
-		engine->pm.clock_pre		= nv04_pm_clock_pre;
-		engine->pm.clock_set		= nv04_pm_clock_set;
+		engine->display.init		= nv04_display_init;
+		engine->display.fini		= nv04_display_fini;
+		engine->gpio.drive		= nv10_gpio_drive;
+		engine->gpio.sense		= nv10_gpio_sense;
+		engine->pm.clocks_get		= nv04_pm_clocks_get;
+		engine->pm.clocks_pre		= nv04_pm_clocks_pre;
+		engine->pm.clocks_set		= nv04_pm_clocks_set;
 		engine->vram.init		= nouveau_mem_detect;
 		engine->vram.takedown		= nouveau_stub_takedown;
 		engine->vram.flags_valid	= nouveau_mem_flags_valid;
@@ -227,16 +220,14 @@
 		engine->display.early_init	= nv04_display_early_init;
 		engine->display.late_takedown	= nv04_display_late_takedown;
 		engine->display.create		= nv04_display_create;
-		engine->display.init		= nv04_display_init;
 		engine->display.destroy		= nv04_display_destroy;
-		engine->gpio.init		= nouveau_stub_init;
-		engine->gpio.takedown		= nouveau_stub_takedown;
-		engine->gpio.get		= nv10_gpio_get;
-		engine->gpio.set		= nv10_gpio_set;
-		engine->gpio.irq_enable		= NULL;
-		engine->pm.clock_get		= nv04_pm_clock_get;
-		engine->pm.clock_pre		= nv04_pm_clock_pre;
-		engine->pm.clock_set		= nv04_pm_clock_set;
+		engine->display.init		= nv04_display_init;
+		engine->display.fini		= nv04_display_fini;
+		engine->gpio.drive		= nv10_gpio_drive;
+		engine->gpio.sense		= nv10_gpio_sense;
+		engine->pm.clocks_get		= nv04_pm_clocks_get;
+		engine->pm.clocks_pre		= nv04_pm_clocks_pre;
+		engine->pm.clocks_set		= nv04_pm_clocks_set;
 		engine->pm.voltage_get		= nouveau_voltage_gpio_get;
 		engine->pm.voltage_set		= nouveau_voltage_gpio_set;
 		engine->vram.init		= nouveau_mem_detect;
@@ -279,19 +270,22 @@
 		engine->display.early_init	= nv04_display_early_init;
 		engine->display.late_takedown	= nv04_display_late_takedown;
 		engine->display.create		= nv04_display_create;
-		engine->display.init		= nv04_display_init;
 		engine->display.destroy		= nv04_display_destroy;
-		engine->gpio.init		= nouveau_stub_init;
-		engine->gpio.takedown		= nouveau_stub_takedown;
-		engine->gpio.get		= nv10_gpio_get;
-		engine->gpio.set		= nv10_gpio_set;
-		engine->gpio.irq_enable		= NULL;
+		engine->display.init		= nv04_display_init;
+		engine->display.fini		= nv04_display_fini;
+		engine->gpio.init		= nv10_gpio_init;
+		engine->gpio.fini		= nv10_gpio_fini;
+		engine->gpio.drive		= nv10_gpio_drive;
+		engine->gpio.sense		= nv10_gpio_sense;
+		engine->gpio.irq_enable		= nv10_gpio_irq_enable;
 		engine->pm.clocks_get		= nv40_pm_clocks_get;
 		engine->pm.clocks_pre		= nv40_pm_clocks_pre;
 		engine->pm.clocks_set		= nv40_pm_clocks_set;
 		engine->pm.voltage_get		= nouveau_voltage_gpio_get;
 		engine->pm.voltage_set		= nouveau_voltage_gpio_set;
 		engine->pm.temp_get		= nv40_temp_get;
+		engine->pm.pwm_get		= nv40_pm_pwm_get;
+		engine->pm.pwm_set		= nv40_pm_pwm_set;
 		engine->vram.init		= nouveau_mem_detect;
 		engine->vram.takedown		= nouveau_stub_takedown;
 		engine->vram.flags_valid	= nouveau_mem_flags_valid;
@@ -334,14 +328,13 @@
 		engine->display.early_init	= nv50_display_early_init;
 		engine->display.late_takedown	= nv50_display_late_takedown;
 		engine->display.create		= nv50_display_create;
-		engine->display.init		= nv50_display_init;
 		engine->display.destroy		= nv50_display_destroy;
+		engine->display.init		= nv50_display_init;
+		engine->display.fini		= nv50_display_fini;
 		engine->gpio.init		= nv50_gpio_init;
-		engine->gpio.takedown		= nv50_gpio_fini;
-		engine->gpio.get		= nv50_gpio_get;
-		engine->gpio.set		= nv50_gpio_set;
-		engine->gpio.irq_register	= nv50_gpio_irq_register;
-		engine->gpio.irq_unregister	= nv50_gpio_irq_unregister;
+		engine->gpio.fini		= nv50_gpio_fini;
+		engine->gpio.drive		= nv50_gpio_drive;
+		engine->gpio.sense		= nv50_gpio_sense;
 		engine->gpio.irq_enable		= nv50_gpio_irq_enable;
 		switch (dev_priv->chipset) {
 		case 0x84:
@@ -354,9 +347,9 @@
 		case 0xaa:
 		case 0xac:
 		case 0x50:
-			engine->pm.clock_get	= nv50_pm_clock_get;
-			engine->pm.clock_pre	= nv50_pm_clock_pre;
-			engine->pm.clock_set	= nv50_pm_clock_set;
+			engine->pm.clocks_get	= nv50_pm_clocks_get;
+			engine->pm.clocks_pre	= nv50_pm_clocks_pre;
+			engine->pm.clocks_set	= nv50_pm_clocks_set;
 			break;
 		default:
 			engine->pm.clocks_get	= nva3_pm_clocks_get;
@@ -370,6 +363,8 @@
 			engine->pm.temp_get	= nv84_temp_get;
 		else
 			engine->pm.temp_get	= nv40_temp_get;
+		engine->pm.pwm_get		= nv50_pm_pwm_get;
+		engine->pm.pwm_set		= nv50_pm_pwm_set;
 		engine->vram.init		= nv50_vram_init;
 		engine->vram.takedown		= nv50_vram_fini;
 		engine->vram.get		= nv50_vram_new;
@@ -407,14 +402,13 @@
 		engine->display.early_init	= nv50_display_early_init;
 		engine->display.late_takedown	= nv50_display_late_takedown;
 		engine->display.create		= nv50_display_create;
-		engine->display.init		= nv50_display_init;
 		engine->display.destroy		= nv50_display_destroy;
+		engine->display.init		= nv50_display_init;
+		engine->display.fini		= nv50_display_fini;
 		engine->gpio.init		= nv50_gpio_init;
-		engine->gpio.takedown		= nouveau_stub_takedown;
-		engine->gpio.get		= nv50_gpio_get;
-		engine->gpio.set		= nv50_gpio_set;
-		engine->gpio.irq_register	= nv50_gpio_irq_register;
-		engine->gpio.irq_unregister	= nv50_gpio_irq_unregister;
+		engine->gpio.fini		= nv50_gpio_fini;
+		engine->gpio.drive		= nv50_gpio_drive;
+		engine->gpio.sense		= nv50_gpio_sense;
 		engine->gpio.irq_enable		= nv50_gpio_irq_enable;
 		engine->vram.init		= nvc0_vram_init;
 		engine->vram.takedown		= nv50_vram_fini;
@@ -423,8 +417,12 @@
 		engine->vram.flags_valid	= nvc0_vram_flags_valid;
 		engine->pm.temp_get		= nv84_temp_get;
 		engine->pm.clocks_get		= nvc0_pm_clocks_get;
+		engine->pm.clocks_pre		= nvc0_pm_clocks_pre;
+		engine->pm.clocks_set		= nvc0_pm_clocks_set;
 		engine->pm.voltage_get		= nouveau_voltage_gpio_get;
 		engine->pm.voltage_set		= nouveau_voltage_gpio_set;
+		engine->pm.pwm_get		= nv50_pm_pwm_get;
+		engine->pm.pwm_set		= nv50_pm_pwm_set;
 		break;
 	case 0xd0:
 		engine->instmem.init		= nvc0_instmem_init;
@@ -457,21 +455,23 @@
 		engine->display.early_init	= nouveau_stub_init;
 		engine->display.late_takedown	= nouveau_stub_takedown;
 		engine->display.create		= nvd0_display_create;
-		engine->display.init		= nvd0_display_init;
 		engine->display.destroy		= nvd0_display_destroy;
+		engine->display.init		= nvd0_display_init;
+		engine->display.fini		= nvd0_display_fini;
 		engine->gpio.init		= nv50_gpio_init;
-		engine->gpio.takedown		= nouveau_stub_takedown;
-		engine->gpio.get		= nvd0_gpio_get;
-		engine->gpio.set		= nvd0_gpio_set;
-		engine->gpio.irq_register	= nv50_gpio_irq_register;
-		engine->gpio.irq_unregister	= nv50_gpio_irq_unregister;
+		engine->gpio.fini		= nv50_gpio_fini;
+		engine->gpio.drive		= nvd0_gpio_drive;
+		engine->gpio.sense		= nvd0_gpio_sense;
 		engine->gpio.irq_enable		= nv50_gpio_irq_enable;
 		engine->vram.init		= nvc0_vram_init;
 		engine->vram.takedown		= nv50_vram_fini;
 		engine->vram.get		= nvc0_vram_new;
 		engine->vram.put		= nv50_vram_del;
 		engine->vram.flags_valid	= nvc0_vram_flags_valid;
+		engine->pm.temp_get		= nv84_temp_get;
 		engine->pm.clocks_get		= nvc0_pm_clocks_get;
+		engine->pm.clocks_pre		= nvc0_pm_clocks_pre;
+		engine->pm.clocks_set		= nvc0_pm_clocks_set;
 		engine->pm.voltage_get		= nouveau_voltage_gpio_get;
 		engine->pm.voltage_set		= nouveau_voltage_gpio_set;
 		break;
@@ -579,6 +579,14 @@
 	if (ret)
 		goto out_display_early;
 
+	/* workaround an odd issue on nvc1 by disabling the device's
+	 * nosnoop capability.  hopefully won't cause issues until a
+	 * better fix is found - assuming there is one...
+	 */
+	if (dev_priv->chipset == 0xc1) {
+		nv_mask(dev, 0x00088080, 0x00000800, 0x00000000);
+	}
+
 	nouveau_pm_init(dev);
 
 	ret = engine->vram.init(dev);
@@ -607,7 +615,7 @@
 		goto out_gart;
 
 	/* PGPIO */
-	ret = engine->gpio.init(dev);
+	ret = nouveau_gpio_create(dev);
 	if (ret)
 		goto out_mc;
 
@@ -640,6 +648,7 @@
 			nv50_graph_create(dev);
 			break;
 		case NV_C0:
+		case NV_D0:
 			nvc0_graph_create(dev);
 			break;
 		default:
@@ -655,6 +664,11 @@
 		case 0xa0:
 			nv84_crypt_create(dev);
 			break;
+		case 0x98:
+		case 0xaa:
+		case 0xac:
+			nv98_crypt_create(dev);
+			break;
 		}
 
 		switch (dev_priv->card_type) {
@@ -676,15 +690,25 @@
 			break;
 		}
 
+		if (dev_priv->chipset >= 0xa3 || dev_priv->chipset == 0x98) {
+			nv84_bsp_create(dev);
+			nv84_vp_create(dev);
+			nv98_ppp_create(dev);
+		} else
+		if (dev_priv->chipset >= 0x84) {
+			nv50_mpeg_create(dev);
+			nv84_bsp_create(dev);
+			nv84_vp_create(dev);
+		} else
+		if (dev_priv->chipset >= 0x50) {
+			nv50_mpeg_create(dev);
+		} else
 		if (dev_priv->card_type == NV_40 ||
 		    dev_priv->chipset == 0x31 ||
 		    dev_priv->chipset == 0x34 ||
-		    dev_priv->chipset == 0x36)
+		    dev_priv->chipset == 0x36) {
 			nv31_mpeg_create(dev);
-		else
-		if (dev_priv->card_type == NV_50 &&
-		    (dev_priv->chipset < 0x98 || dev_priv->chipset == 0xa0))
-			nv50_mpeg_create(dev);
+		}
 
 		for (e = 0; e < NVOBJ_ENGINE_NR; e++) {
 			if (dev_priv->eng[e]) {
@@ -704,27 +728,7 @@
 	if (ret)
 		goto out_fifo;
 
-	/* initialise general modesetting */
-	drm_mode_config_init(dev);
-	drm_mode_create_scaling_mode_property(dev);
-	drm_mode_create_dithering_property(dev);
-	dev->mode_config.funcs = (void *)&nouveau_mode_config_funcs;
-	dev->mode_config.fb_base = pci_resource_start(dev->pdev, 1);
-	dev->mode_config.min_width = 0;
-	dev->mode_config.min_height = 0;
-	if (dev_priv->card_type < NV_10) {
-		dev->mode_config.max_width = 2048;
-		dev->mode_config.max_height = 2048;
-	} else
-	if (dev_priv->card_type < NV_50) {
-		dev->mode_config.max_width = 4096;
-		dev->mode_config.max_height = 4096;
-	} else {
-		dev->mode_config.max_width = 8192;
-		dev->mode_config.max_height = 8192;
-	}
-
-	ret = engine->display.create(dev);
+	ret = nouveau_display_create(dev);
 	if (ret)
 		goto out_irq;
 
@@ -744,12 +748,11 @@
 	}
 
 	if (dev->mode_config.num_crtc) {
-		ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
+		ret = nouveau_display_init(dev);
 		if (ret)
 			goto out_chan;
 
 		nouveau_fbcon_init(dev);
-		drm_kms_helper_poll_init(dev);
 	}
 
 	return 0;
@@ -760,7 +763,7 @@
 	nouveau_fence_fini(dev);
 out_disp:
 	nouveau_backlight_exit(dev);
-	engine->display.destroy(dev);
+	nouveau_display_destroy(dev);
 out_irq:
 	nouveau_irq_fini(dev);
 out_fifo:
@@ -780,7 +783,7 @@
 out_timer:
 	engine->timer.takedown(dev);
 out_gpio:
-	engine->gpio.takedown(dev);
+	nouveau_gpio_destroy(dev);
 out_mc:
 	engine->mc.takedown(dev);
 out_gart:
@@ -810,9 +813,8 @@
 	int e;
 
 	if (dev->mode_config.num_crtc) {
-		drm_kms_helper_poll_fini(dev);
 		nouveau_fbcon_fini(dev);
-		drm_vblank_cleanup(dev);
+		nouveau_display_fini(dev);
 	}
 
 	if (dev_priv->channel) {
@@ -821,8 +823,7 @@
 	}
 
 	nouveau_backlight_exit(dev);
-	engine->display.destroy(dev);
-	drm_mode_config_cleanup(dev);
+	nouveau_display_destroy(dev);
 
 	if (!dev_priv->noaccel) {
 		engine->fifo.takedown(dev);
@@ -835,7 +836,7 @@
 	}
 	engine->fb.takedown(dev);
 	engine->timer.takedown(dev);
-	engine->gpio.takedown(dev);
+	nouveau_gpio_destroy(dev);
 	engine->mc.takedown(dev);
 	engine->display.late_takedown(dev);
 
@@ -1102,8 +1103,7 @@
 	dev_priv->noaccel = !!nouveau_noaccel;
 	if (nouveau_noaccel == -1) {
 		switch (dev_priv->chipset) {
-		case 0xc1: /* known broken */
-		case 0xc8: /* never tested */
+		case 0xd9: /* known broken */
 			NV_INFO(dev, "acceleration disabled by default, pass "
 				     "noaccel=0 to force enable\n");
 			dev_priv->noaccel = true;
@@ -1229,7 +1229,7 @@
 		getparam->value = 1;
 		break;
 	case NOUVEAU_GETPARAM_HAS_PAGEFLIP:
-		getparam->value = dev_priv->card_type < NV_D0;
+		getparam->value = 1;
 		break;
 	case NOUVEAU_GETPARAM_GRAPH_UNITS:
 		/* NV40 and NV50 versions are quite different, but register
diff --git a/drivers/gpu/drm/nouveau/nouveau_temp.c b/drivers/gpu/drm/nouveau/nouveau_temp.c
index 5a46446..0f5a301 100644
--- a/drivers/gpu/drm/nouveau/nouveau_temp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_temp.c
@@ -55,6 +55,10 @@
 	temps->down_clock = 100;
 	temps->fan_boost = 90;
 
+	/* Set the default range for the pwm fan */
+	pm->fan.min_duty = 30;
+	pm->fan.max_duty = 100;
+
 	/* Set the known default values to setup the temperature sensor */
 	if (dev_priv->card_type >= NV_40) {
 		switch (dev_priv->chipset) {
@@ -156,11 +160,26 @@
 		case 0x13:
 			sensor->slope_div = value;
 			break;
+		case 0x22:
+			pm->fan.min_duty = value & 0xff;
+			pm->fan.max_duty = (value & 0xff00) >> 8;
+			break;
+		case 0x26:
+			pm->fan.pwm_freq = value;
+			break;
 		}
 		temp += recordlen;
 	}
 
 	nouveau_temp_safety_checks(dev);
+
+	/* check the fan min/max settings */
+	if (pm->fan.min_duty < 10)
+		pm->fan.min_duty = 10;
+	if (pm->fan.max_duty > 100)
+		pm->fan.max_duty = 100;
+	if (pm->fan.max_duty < pm->fan.min_duty)
+		pm->fan.max_duty = pm->fan.min_duty;
 }
 
 static int
@@ -267,8 +286,6 @@
 static void
 nouveau_temp_probe_i2c(struct drm_device *dev)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct dcb_table *dcb = &dev_priv->vbios.dcb;
 	struct i2c_board_info info[] = {
 		{ I2C_BOARD_INFO("w83l785ts", 0x2d) },
 		{ I2C_BOARD_INFO("w83781d", 0x2d) },
@@ -277,11 +294,9 @@
 		{ I2C_BOARD_INFO("lm99", 0x4c) },
 		{ }
 	};
-	int idx = (dcb->version >= 0x40 ?
-		   dcb->i2c_default_indices & 0xf : 2);
 
 	nouveau_i2c_identify(dev, "monitoring device", info,
-			     probe_monitoring_device, idx);
+			     probe_monitoring_device, NV_I2C_DEFAULT(0));
 }
 
 void
@@ -297,9 +312,9 @@
 			return;
 
 		if (P.version == 1)
-			temp = ROMPTR(bios, P.data[12]);
+			temp = ROMPTR(dev, P.data[12]);
 		else if (P.version == 2)
-			temp = ROMPTR(bios, P.data[16]);
+			temp = ROMPTR(dev, P.data[16]);
 		else
 			NV_WARN(dev, "unknown temp for BIT P %d\n", P.version);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_vm.c b/drivers/gpu/drm/nouveau/nouveau_vm.c
index ef0832b..2bf6c03 100644
--- a/drivers/gpu/drm/nouveau/nouveau_vm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_vm.c
@@ -78,9 +78,10 @@
 
 void
 nouveau_vm_map_sg(struct nouveau_vma *vma, u64 delta, u64 length,
-		  struct nouveau_mem *mem, dma_addr_t *list)
+		  struct nouveau_mem *mem)
 {
 	struct nouveau_vm *vm = vma->vm;
+	dma_addr_t *list = mem->pages;
 	int big = vma->node->type != vm->spg_shift;
 	u32 offset = vma->node->offset + (delta >> 12);
 	u32 bits = vma->node->type - 12;
diff --git a/drivers/gpu/drm/nouveau/nouveau_vm.h b/drivers/gpu/drm/nouveau/nouveau_vm.h
index 6ce995f..4fb6e72 100644
--- a/drivers/gpu/drm/nouveau/nouveau_vm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_vm.h
@@ -89,7 +89,7 @@
 void nouveau_vm_unmap(struct nouveau_vma *);
 void nouveau_vm_unmap_at(struct nouveau_vma *, u64 offset, u64 length);
 void nouveau_vm_map_sg(struct nouveau_vma *, u64 offset, u64 length,
-		       struct nouveau_mem *, dma_addr_t *);
+		       struct nouveau_mem *);
 
 /* nv50_vm.c */
 void nv50_vm_map_pgt(struct nouveau_gpuobj *pgd, u32 pde,
diff --git a/drivers/gpu/drm/nouveau/nouveau_volt.c b/drivers/gpu/drm/nouveau/nouveau_volt.c
index 86d03e1..b010cb9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_volt.c
+++ b/drivers/gpu/drm/nouveau/nouveau_volt.c
@@ -26,6 +26,7 @@
 
 #include "nouveau_drv.h"
 #include "nouveau_pm.h"
+#include "nouveau_gpio.h"
 
 static const enum dcb_gpio_tag vidtag[] = { 0x04, 0x05, 0x06, 0x1a, 0x73 };
 static int nr_vidtag = sizeof(vidtag) / sizeof(vidtag[0]);
@@ -34,7 +35,6 @@
 nouveau_voltage_gpio_get(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *gpio = &dev_priv->engine.gpio;
 	struct nouveau_pm_voltage *volt = &dev_priv->engine.pm.voltage;
 	u8 vid = 0;
 	int i;
@@ -43,7 +43,7 @@
 		if (!(volt->vid_mask & (1 << i)))
 			continue;
 
-		vid |= gpio->get(dev, vidtag[i]) << i;
+		vid |= nouveau_gpio_func_get(dev, vidtag[i]) << i;
 	}
 
 	return nouveau_volt_lvl_lookup(dev, vid);
@@ -53,7 +53,6 @@
 nouveau_voltage_gpio_set(struct drm_device *dev, int voltage)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *gpio = &dev_priv->engine.gpio;
 	struct nouveau_pm_voltage *volt = &dev_priv->engine.pm.voltage;
 	int vid, i;
 
@@ -65,7 +64,7 @@
 		if (!(volt->vid_mask & (1 << i)))
 			continue;
 
-		gpio->set(dev, vidtag[i], !!(vid & (1 << i)));
+		nouveau_gpio_func_set(dev, vidtag[i], !!(vid & (1 << i)));
 	}
 
 	return 0;
@@ -117,10 +116,10 @@
 			return;
 
 		if (P.version == 1)
-			volt = ROMPTR(bios, P.data[16]);
+			volt = ROMPTR(dev, P.data[16]);
 		else
 		if (P.version == 2)
-			volt = ROMPTR(bios, P.data[12]);
+			volt = ROMPTR(dev, P.data[12]);
 		else {
 			NV_WARN(dev, "unknown volt for BIT P %d\n", P.version);
 		}
@@ -130,7 +129,7 @@
 			return;
 		}
 
-		volt = ROMPTR(bios, bios->data[bios->offset + 0x98]);
+		volt = ROMPTR(dev, bios->data[bios->offset + 0x98]);
 	}
 
 	if (!volt) {
@@ -194,7 +193,7 @@
 			return;
 		}
 
-		if (!nouveau_bios_gpio_entry(dev, vidtag[i])) {
+		if (!nouveau_gpio_func_valid(dev, vidtag[i])) {
 			NV_DEBUG(dev, "vid bit %d has no gpio tag\n", i);
 			return;
 		}
diff --git a/drivers/gpu/drm/nouveau/nv04_crtc.c b/drivers/gpu/drm/nouveau/nv04_crtc.c
index 5e45398..728d075 100644
--- a/drivers/gpu/drm/nouveau/nv04_crtc.c
+++ b/drivers/gpu/drm/nouveau/nv04_crtc.c
@@ -364,7 +364,7 @@
 	regp->CRTC[NV_CIO_CR_VRE_INDEX] = 1 << 5 | XLATE(vertEnd, 0, NV_CIO_CR_VRE_3_0);
 	regp->CRTC[NV_CIO_CR_VDE_INDEX] = vertDisplay;
 	/* framebuffer can be larger than crtc scanout area. */
-	regp->CRTC[NV_CIO_CR_OFFSET_INDEX] = fb->pitch / 8;
+	regp->CRTC[NV_CIO_CR_OFFSET_INDEX] = fb->pitches[0] / 8;
 	regp->CRTC[NV_CIO_CR_ULINE_INDEX] = 0x00;
 	regp->CRTC[NV_CIO_CR_VBS_INDEX] = vertBlankStart;
 	regp->CRTC[NV_CIO_CR_VBE_INDEX] = vertBlankEnd;
@@ -377,9 +377,9 @@
 
 	/* framebuffer can be larger than crtc scanout area. */
 	regp->CRTC[NV_CIO_CRE_RPC0_INDEX] =
-		XLATE(fb->pitch / 8, 8, NV_CIO_CRE_RPC0_OFFSET_10_8);
+		XLATE(fb->pitches[0] / 8, 8, NV_CIO_CRE_RPC0_OFFSET_10_8);
 	regp->CRTC[NV_CIO_CRE_42] =
-		XLATE(fb->pitch / 8, 11, NV_CIO_CRE_42_OFFSET_11);
+		XLATE(fb->pitches[0] / 8, 11, NV_CIO_CRE_42_OFFSET_11);
 	regp->CRTC[NV_CIO_CRE_RPC1_INDEX] = mode->crtc_hdisplay < 1280 ?
 					    MASK(NV_CIO_CRE_RPC1_LARGE) : 0x00;
 	regp->CRTC[NV_CIO_CRE_LSR_INDEX] = XLATE(horizBlankEnd, 6, NV_CIO_CRE_LSR_HBE_6) |
@@ -835,18 +835,18 @@
 	NVWriteRAMDAC(dev, nv_crtc->index, NV_PRAMDAC_GENERAL_CONTROL,
 		      regp->ramdac_gen_ctrl);
 
-	regp->CRTC[NV_CIO_CR_OFFSET_INDEX] = drm_fb->pitch >> 3;
+	regp->CRTC[NV_CIO_CR_OFFSET_INDEX] = drm_fb->pitches[0] >> 3;
 	regp->CRTC[NV_CIO_CRE_RPC0_INDEX] =
-		XLATE(drm_fb->pitch >> 3, 8, NV_CIO_CRE_RPC0_OFFSET_10_8);
+		XLATE(drm_fb->pitches[0] >> 3, 8, NV_CIO_CRE_RPC0_OFFSET_10_8);
 	regp->CRTC[NV_CIO_CRE_42] =
-		XLATE(drm_fb->pitch / 8, 11, NV_CIO_CRE_42_OFFSET_11);
+		XLATE(drm_fb->pitches[0] / 8, 11, NV_CIO_CRE_42_OFFSET_11);
 	crtc_wr_cio_state(crtc, regp, NV_CIO_CRE_RPC0_INDEX);
 	crtc_wr_cio_state(crtc, regp, NV_CIO_CR_OFFSET_INDEX);
 	crtc_wr_cio_state(crtc, regp, NV_CIO_CRE_42);
 
 	/* Update the framebuffer location. */
 	regp->fb_start = nv_crtc->fb.offset & ~3;
-	regp->fb_start += (y * drm_fb->pitch) + (x * drm_fb->bits_per_pixel / 8);
+	regp->fb_start += (y * drm_fb->pitches[0]) + (x * drm_fb->bits_per_pixel / 8);
 	nv_set_crtc_base(dev, nv_crtc->index, regp->fb_start);
 
 	/* Update the arbitration parameters. */
diff --git a/drivers/gpu/drm/nouveau/nv04_dac.c b/drivers/gpu/drm/nouveau/nv04_dac.c
index e000455..8300266 100644
--- a/drivers/gpu/drm/nouveau/nv04_dac.c
+++ b/drivers/gpu/drm/nouveau/nv04_dac.c
@@ -32,6 +32,7 @@
 #include "nouveau_connector.h"
 #include "nouveau_crtc.h"
 #include "nouveau_hw.h"
+#include "nouveau_gpio.h"
 #include "nvreg.h"
 
 int nv04_dac_output_offset(struct drm_encoder *encoder)
@@ -220,7 +221,6 @@
 {
 	struct drm_device *dev = encoder->dev;
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *gpio = &dev_priv->engine.gpio;
 	struct dcb_entry *dcb = nouveau_encoder(encoder)->dcb;
 	uint32_t sample, testval, regoffset = nv04_dac_output_offset(encoder);
 	uint32_t saved_powerctrl_2 = 0, saved_powerctrl_4 = 0, saved_routput,
@@ -252,11 +252,11 @@
 		nvWriteMC(dev, NV_PBUS_POWERCTRL_4, saved_powerctrl_4 & 0xffffffcf);
 	}
 
-	saved_gpio1 = gpio->get(dev, DCB_GPIO_TVDAC1);
-	saved_gpio0 = gpio->get(dev, DCB_GPIO_TVDAC0);
+	saved_gpio1 = nouveau_gpio_func_get(dev, DCB_GPIO_TVDAC1);
+	saved_gpio0 = nouveau_gpio_func_get(dev, DCB_GPIO_TVDAC0);
 
-	gpio->set(dev, DCB_GPIO_TVDAC1, dcb->type == OUTPUT_TV);
-	gpio->set(dev, DCB_GPIO_TVDAC0, dcb->type == OUTPUT_TV);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC1, dcb->type == OUTPUT_TV);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC0, dcb->type == OUTPUT_TV);
 
 	msleep(4);
 
@@ -306,8 +306,8 @@
 		nvWriteMC(dev, NV_PBUS_POWERCTRL_4, saved_powerctrl_4);
 	nvWriteMC(dev, NV_PBUS_POWERCTRL_2, saved_powerctrl_2);
 
-	gpio->set(dev, DCB_GPIO_TVDAC1, saved_gpio1);
-	gpio->set(dev, DCB_GPIO_TVDAC0, saved_gpio0);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC1, saved_gpio1);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC0, saved_gpio0);
 
 	return sample;
 }
diff --git a/drivers/gpu/drm/nouveau/nv04_dfp.c b/drivers/gpu/drm/nouveau/nv04_dfp.c
index 12098bf..2258746 100644
--- a/drivers/gpu/drm/nouveau/nv04_dfp.c
+++ b/drivers/gpu/drm/nouveau/nv04_dfp.c
@@ -289,6 +289,7 @@
 	struct nouveau_connector *nv_connector = nouveau_crtc_connector_get(nv_crtc);
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
 	struct drm_display_mode *output_mode = &nv_encoder->mode;
+	struct drm_connector *connector = &nv_connector->base;
 	uint32_t mode_ratio, panel_ratio;
 
 	NV_DEBUG_KMS(dev, "Output mode on CRTC %d:\n", nv_crtc->index);
@@ -340,10 +341,15 @@
 	    output_mode->clock > 165000)
 		regp->fp_control |= (2 << 24);
 	if (nv_encoder->dcb->type == OUTPUT_LVDS) {
-		bool duallink, dummy;
+		bool duallink = false, dummy;
+		if (nv_connector->edid &&
+		    nv_connector->type == DCB_CONNECTOR_LVDS_SPWG) {
+			duallink = (((u8 *)nv_connector->edid)[121] == 2);
+		} else {
+			nouveau_bios_parse_lvds_table(dev, output_mode->clock,
+						      &duallink, &dummy);
+		}
 
-		nouveau_bios_parse_lvds_table(dev, output_mode->clock,
-					      &duallink, &dummy);
 		if (duallink)
 			regp->fp_control |= (8 << 28);
 	} else
@@ -407,7 +413,9 @@
 	}
 
 	/* Output property. */
-	if (nv_connector->use_dithering) {
+	if ((nv_connector->dithering_mode == DITHERING_MODE_ON) ||
+	    (nv_connector->dithering_mode == DITHERING_MODE_AUTO &&
+	     encoder->crtc->fb->depth > connector->display_info.bpc * 3)) {
 		if (dev_priv->chipset == 0x11)
 			regp->dither = savep->dither | 0x00010000;
 		else {
diff --git a/drivers/gpu/drm/nouveau/nv04_display.c b/drivers/gpu/drm/nouveau/nv04_display.c
index 6bd8518..7047d37 100644
--- a/drivers/gpu/drm/nouveau/nv04_display.c
+++ b/drivers/gpu/drm/nouveau/nv04_display.c
@@ -243,6 +243,11 @@
 	return 0;
 }
 
+void
+nv04_display_fini(struct drm_device *dev)
+{
+}
+
 static void
 nv04_vblank_crtc0_isr(struct drm_device *dev)
 {
diff --git a/drivers/gpu/drm/nouveau/nv04_pm.c b/drivers/gpu/drm/nouveau/nv04_pm.c
index 9ae92a8..6e75899 100644
--- a/drivers/gpu/drm/nouveau/nv04_pm.c
+++ b/drivers/gpu/drm/nouveau/nv04_pm.c
@@ -27,68 +27,111 @@
 #include "nouveau_hw.h"
 #include "nouveau_pm.h"
 
-struct nv04_pm_state {
+int
+nv04_pm_clocks_get(struct drm_device *dev, struct nouveau_pm_level *perflvl)
+{
+	int ret;
+
+	ret = nouveau_hw_get_clock(dev, PLL_CORE);
+	if (ret < 0)
+		return ret;
+	perflvl->core = ret;
+
+	ret = nouveau_hw_get_clock(dev, PLL_MEMORY);
+	if (ret < 0)
+		return ret;
+	perflvl->memory = ret;
+
+	return 0;
+}
+
+struct nv04_pm_clock {
 	struct pll_lims pll;
 	struct nouveau_pll_vals calc;
 };
 
-int
-nv04_pm_clock_get(struct drm_device *dev, u32 id)
+struct nv04_pm_state {
+	struct nv04_pm_clock core;
+	struct nv04_pm_clock memory;
+};
+
+static int
+calc_pll(struct drm_device *dev, u32 id, int khz, struct nv04_pm_clock *clk)
 {
-	return nouveau_hw_get_clock(dev, id);
+	int ret;
+
+	ret = get_pll_limits(dev, id, &clk->pll);
+	if (ret)
+		return ret;
+
+	ret = nouveau_calc_pll_mnp(dev, &clk->pll, khz, &clk->calc);
+	if (!ret)
+		return -EINVAL;
+
+	return 0;
 }
 
 void *
-nv04_pm_clock_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl,
-		  u32 id, int khz)
+nv04_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
 {
-	struct nv04_pm_state *state;
+	struct nv04_pm_state *info;
 	int ret;
 
-	state = kzalloc(sizeof(*state), GFP_KERNEL);
-	if (!state)
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (!info)
 		return ERR_PTR(-ENOMEM);
 
-	ret = get_pll_limits(dev, id, &state->pll);
-	if (ret) {
-		kfree(state);
-		return (ret == -ENOENT) ? NULL : ERR_PTR(ret);
+	ret = calc_pll(dev, PLL_CORE, perflvl->core, &info->core);
+	if (ret)
+		goto error;
+
+	if (perflvl->memory) {
+		ret = calc_pll(dev, PLL_MEMORY, perflvl->memory, &info->memory);
+		if (ret)
+			goto error;
 	}
 
-	ret = nouveau_calc_pll_mnp(dev, &state->pll, khz, &state->calc);
-	if (!ret) {
-		kfree(state);
-		return ERR_PTR(-EINVAL);
-	}
-
-	return state;
+	return info;
+error:
+	kfree(info);
+	return ERR_PTR(ret);
 }
 
-void
-nv04_pm_clock_set(struct drm_device *dev, void *pre_state)
+static void
+prog_pll(struct drm_device *dev, struct nv04_pm_clock *clk)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_timer_engine *ptimer = &dev_priv->engine.timer;
-	struct nv04_pm_state *state = pre_state;
-	u32 reg = state->pll.reg;
+	u32 reg = clk->pll.reg;
 
 	/* thank the insane nouveau_hw_setpll() interface for this */
 	if (dev_priv->card_type >= NV_40)
 		reg += 4;
 
-	nouveau_hw_setpll(dev, reg, &state->calc);
-
-	if (dev_priv->card_type < NV_30 && reg == NV_PRAMDAC_MPLL_COEFF) {
-		if (dev_priv->card_type == NV_20)
-			nv_mask(dev, 0x1002c4, 0, 1 << 20);
-
-		/* Reset the DLLs */
-		nv_mask(dev, 0x1002c0, 0, 1 << 8);
-	}
-
-	if (reg == NV_PRAMDAC_NVPLL_COEFF)
-		ptimer->init(dev);
-
-	kfree(state);
+	nouveau_hw_setpll(dev, reg, &clk->calc);
 }
 
+int
+nv04_pm_clocks_set(struct drm_device *dev, void *pre_state)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nouveau_timer_engine *ptimer = &dev_priv->engine.timer;
+	struct nv04_pm_state *state = pre_state;
+
+	prog_pll(dev, &state->core);
+
+	if (state->memory.pll.reg) {
+		prog_pll(dev, &state->memory);
+		if (dev_priv->card_type < NV_30) {
+			if (dev_priv->card_type == NV_20)
+				nv_mask(dev, 0x1002c4, 0, 1 << 20);
+
+			/* Reset the DLLs */
+			nv_mask(dev, 0x1002c0, 0, 1 << 8);
+		}
+	}
+
+	ptimer->init(dev);
+
+	kfree(state);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nv04_timer.c b/drivers/gpu/drm/nouveau/nv04_timer.c
index 263301b..55c9452 100644
--- a/drivers/gpu/drm/nouveau/nv04_timer.c
+++ b/drivers/gpu/drm/nouveau/nv04_timer.c
@@ -2,6 +2,7 @@
 #include "drm.h"
 #include "nouveau_drv.h"
 #include "nouveau_drm.h"
+#include "nouveau_hw.h"
 
 int
 nv04_timer_init(struct drm_device *dev)
@@ -17,7 +18,7 @@
 
 	/* determine base clock for timer source */
 	if (dev_priv->chipset < 0x40) {
-		n = dev_priv->engine.pm.clock_get(dev, PLL_CORE);
+		n = nouveau_hw_get_clock(dev, PLL_CORE);
 	} else
 	if (dev_priv->chipset == 0x40) {
 		/*XXX: figure this out */
diff --git a/drivers/gpu/drm/nouveau/nv10_gpio.c b/drivers/gpu/drm/nouveau/nv10_gpio.c
index 007fc29..550ad3f 100644
--- a/drivers/gpu/drm/nouveau/nv10_gpio.c
+++ b/drivers/gpu/drm/nouveau/nv10_gpio.c
@@ -27,66 +27,97 @@
 #include "drmP.h"
 #include "nouveau_drv.h"
 #include "nouveau_hw.h"
+#include "nouveau_gpio.h"
 
-static bool
-get_gpio_location(struct dcb_gpio_entry *ent, uint32_t *reg, uint32_t *shift,
-		  uint32_t *mask)
+int
+nv10_gpio_sense(struct drm_device *dev, int line)
 {
-	if (ent->line < 2) {
-		*reg = NV_PCRTC_GPIO;
-		*shift = ent->line * 16;
-		*mask = 0x11;
-
-	} else if (ent->line < 10) {
-		*reg = NV_PCRTC_GPIO_EXT;
-		*shift = (ent->line - 2) * 4;
-		*mask = 0x3;
-
-	} else if (ent->line < 14) {
-		*reg = NV_PCRTC_850;
-		*shift = (ent->line - 10) * 4;
-		*mask = 0x3;
-
-	} else {
-		return false;
+	if (line < 2) {
+		line = line * 16;
+		line = NVReadCRTC(dev, 0, NV_PCRTC_GPIO) >> line;
+		return !!(line & 0x0100);
+	} else
+	if (line < 10) {
+		line = (line - 2) * 4;
+		line = NVReadCRTC(dev, 0, NV_PCRTC_GPIO_EXT) >> line;
+		return !!(line & 0x04);
+	} else
+	if (line < 14) {
+		line = (line - 10) * 4;
+		line = NVReadCRTC(dev, 0, NV_PCRTC_850) >> line;
+		return !!(line & 0x04);
 	}
 
-	return true;
+	return -EINVAL;
 }
 
 int
-nv10_gpio_get(struct drm_device *dev, enum dcb_gpio_tag tag)
+nv10_gpio_drive(struct drm_device *dev, int line, int dir, int out)
 {
-	struct dcb_gpio_entry *ent = nouveau_bios_gpio_entry(dev, tag);
-	uint32_t reg, shift, mask, value;
+	u32 reg, mask, data;
 
-	if (!ent)
-		return -ENODEV;
+	if (line < 2) {
+		line = line * 16;
+		reg  = NV_PCRTC_GPIO;
+		mask = 0x00000011;
+		data = (dir << 4) | out;
+	} else
+	if (line < 10) {
+		line = (line - 2) * 4;
+		reg  = NV_PCRTC_GPIO_EXT;
+		mask = 0x00000003 << ((line - 2) * 4);
+		data = (dir << 1) | out;
+	} else
+	if (line < 14) {
+		line = (line - 10) * 4;
+		reg  = NV_PCRTC_850;
+		mask = 0x00000003;
+		data = (dir << 1) | out;
+	} else {
+		return -EINVAL;
+	}
 
-	if (!get_gpio_location(ent, &reg, &shift, &mask))
-		return -ENODEV;
-
-	value = NVReadCRTC(dev, 0, reg) >> shift;
-
-	return (ent->invert ? 1 : 0) ^ (value & 1);
-}
-
-int
-nv10_gpio_set(struct drm_device *dev, enum dcb_gpio_tag tag, int state)
-{
-	struct dcb_gpio_entry *ent = nouveau_bios_gpio_entry(dev, tag);
-	uint32_t reg, shift, mask, value;
-
-	if (!ent)
-		return -ENODEV;
-
-	if (!get_gpio_location(ent, &reg, &shift, &mask))
-		return -ENODEV;
-
-	value = ((ent->invert ? 1 : 0) ^ (state ? 1 : 0)) << shift;
-	mask = ~(mask << shift);
-
-	NVWriteCRTC(dev, 0, reg, value | (NVReadCRTC(dev, 0, reg) & mask));
-
+	mask = NVReadCRTC(dev, 0, reg) & ~(mask << line);
+	NVWriteCRTC(dev, 0, reg, mask | (data << line));
 	return 0;
 }
+
+void
+nv10_gpio_irq_enable(struct drm_device *dev, int line, bool on)
+{
+	u32 mask = 0x00010001 << line;
+
+	nv_wr32(dev, 0x001104, mask);
+	nv_mask(dev, 0x001144, mask, on ? mask : 0);
+}
+
+static void
+nv10_gpio_isr(struct drm_device *dev)
+{
+	u32 intr = nv_rd32(dev, 0x1104);
+	u32 hi = (intr & 0x0000ffff) >> 0;
+	u32 lo = (intr & 0xffff0000) >> 16;
+
+	nouveau_gpio_isr(dev, 0, hi | lo);
+
+	nv_wr32(dev, 0x001104, intr);
+}
+
+int
+nv10_gpio_init(struct drm_device *dev)
+{
+	nv_wr32(dev, 0x001140, 0x00000000);
+	nv_wr32(dev, 0x001100, 0xffffffff);
+	nv_wr32(dev, 0x001144, 0x00000000);
+	nv_wr32(dev, 0x001104, 0xffffffff);
+	nouveau_irq_register(dev, 28, nv10_gpio_isr); /* PBUS */
+	return 0;
+}
+
+void
+nv10_gpio_fini(struct drm_device *dev)
+{
+	nv_wr32(dev, 0x001140, 0x00000000);
+	nv_wr32(dev, 0x001144, 0x00000000);
+	nouveau_irq_unregister(dev, 28);
+}
diff --git a/drivers/gpu/drm/nouveau/nv17_tv.c b/drivers/gpu/drm/nouveau/nv17_tv.c
index 3900ceb..696d7e7 100644
--- a/drivers/gpu/drm/nouveau/nv17_tv.c
+++ b/drivers/gpu/drm/nouveau/nv17_tv.c
@@ -30,6 +30,7 @@
 #include "nouveau_encoder.h"
 #include "nouveau_connector.h"
 #include "nouveau_crtc.h"
+#include "nouveau_gpio.h"
 #include "nouveau_hw.h"
 #include "nv17_tv.h"
 
@@ -37,7 +38,6 @@
 {
 	struct drm_device *dev = encoder->dev;
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *gpio = &dev_priv->engine.gpio;
 	uint32_t testval, regoffset = nv04_dac_output_offset(encoder);
 	uint32_t gpio0, gpio1, fp_htotal, fp_hsync_start, fp_hsync_end,
 		fp_control, test_ctrl, dacclk, ctv_14, ctv_1c, ctv_6c;
@@ -53,8 +53,8 @@
 	head = (dacclk & 0x100) >> 8;
 
 	/* Save the previous state. */
-	gpio1 = gpio->get(dev, DCB_GPIO_TVDAC1);
-	gpio0 = gpio->get(dev, DCB_GPIO_TVDAC0);
+	gpio1 = nouveau_gpio_func_get(dev, DCB_GPIO_TVDAC1);
+	gpio0 = nouveau_gpio_func_get(dev, DCB_GPIO_TVDAC0);
 	fp_htotal = NVReadRAMDAC(dev, head, NV_PRAMDAC_FP_HTOTAL);
 	fp_hsync_start = NVReadRAMDAC(dev, head, NV_PRAMDAC_FP_HSYNC_START);
 	fp_hsync_end = NVReadRAMDAC(dev, head, NV_PRAMDAC_FP_HSYNC_END);
@@ -65,8 +65,8 @@
 	ctv_6c = NVReadRAMDAC(dev, head, 0x680c6c);
 
 	/* Prepare the DAC for load detection.  */
-	gpio->set(dev, DCB_GPIO_TVDAC1, true);
-	gpio->set(dev, DCB_GPIO_TVDAC0, true);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC1, true);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC0, true);
 
 	NVWriteRAMDAC(dev, head, NV_PRAMDAC_FP_HTOTAL, 1343);
 	NVWriteRAMDAC(dev, head, NV_PRAMDAC_FP_HSYNC_START, 1047);
@@ -111,8 +111,8 @@
 	NVWriteRAMDAC(dev, head, NV_PRAMDAC_FP_HSYNC_END, fp_hsync_end);
 	NVWriteRAMDAC(dev, head, NV_PRAMDAC_FP_HSYNC_START, fp_hsync_start);
 	NVWriteRAMDAC(dev, head, NV_PRAMDAC_FP_HTOTAL, fp_htotal);
-	gpio->set(dev, DCB_GPIO_TVDAC1, gpio1);
-	gpio->set(dev, DCB_GPIO_TVDAC0, gpio0);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC1, gpio1);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC0, gpio0);
 
 	return sample;
 }
@@ -357,8 +357,6 @@
 static void  nv17_tv_dpms(struct drm_encoder *encoder, int mode)
 {
 	struct drm_device *dev = encoder->dev;
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *gpio = &dev_priv->engine.gpio;
 	struct nv17_tv_state *regs = &to_tv_enc(encoder)->state;
 	struct nv17_tv_norm_params *tv_norm = get_tv_norm(encoder);
 
@@ -383,8 +381,8 @@
 
 	nv_load_ptv(dev, regs, 200);
 
-	gpio->set(dev, DCB_GPIO_TVDAC1, mode == DRM_MODE_DPMS_ON);
-	gpio->set(dev, DCB_GPIO_TVDAC0, mode == DRM_MODE_DPMS_ON);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC1, mode == DRM_MODE_DPMS_ON);
+	nouveau_gpio_func_set(dev, DCB_GPIO_TVDAC0, mode == DRM_MODE_DPMS_ON);
 
 	nv04_dac_update_dacclk(encoder, mode == DRM_MODE_DPMS_ON);
 }
diff --git a/drivers/gpu/drm/nouveau/nv40_pm.c b/drivers/gpu/drm/nouveau/nv40_pm.c
index bbc0b9c..c761538 100644
--- a/drivers/gpu/drm/nouveau/nv40_pm.c
+++ b/drivers/gpu/drm/nouveau/nv40_pm.c
@@ -57,12 +57,14 @@
 	int P = (ctrl & 0x00070000) >> 16;
 	u32 ref = 27000, clk = 0;
 
-	if (ctrl & 0x80000000)
+	if ((ctrl & 0x80000000) && M1) {
 		clk = ref * N1 / M1;
-
-	if (!(ctrl & 0x00000100)) {
-		if (ctrl & 0x40000000)
-			clk = clk * N2 / M2;
+		if ((ctrl & 0x40000100) == 0x40000000) {
+			if (M2)
+				clk = clk * N2 / M2;
+			else
+				clk = 0;
+		}
 	}
 
 	return clk >> P;
@@ -177,6 +179,11 @@
 	}
 
 	/* memory clock */
+	if (!perflvl->memory) {
+		info->mpll_ctrl = 0x00000000;
+		goto out;
+	}
+
 	ret = nv40_calc_pll(dev, 0x004020, &pll, perflvl->memory,
 			    &N1, &M1, &N2, &M2, &log2P);
 	if (ret < 0)
@@ -215,7 +222,7 @@
 	return true;
 }
 
-void
+int
 nv40_pm_clocks_set(struct drm_device *dev, void *pre_state)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
@@ -224,7 +231,7 @@
 	struct bit_entry M;
 	u32 crtc_mask = 0;
 	u8 sr1[2];
-	int i;
+	int i, ret = -EAGAIN;
 
 	/* determine which CRTCs are active, fetch VGA_SR1 for each */
 	for (i = 0; i < 2; i++) {
@@ -256,6 +263,8 @@
 	if (!nv_wait_cb(dev, nv40_pm_gr_idle, dev))
 		goto resume;
 
+	ret = 0;
+
 	/* set engine clocks */
 	nv_mask(dev, 0x00c040, 0x00000333, 0x00000000);
 	nv_wr32(dev, 0x004004, info->npll_coef);
@@ -264,6 +273,9 @@
 	mdelay(5);
 	nv_mask(dev, 0x00c040, 0x00000333, info->ctrl);
 
+	if (!info->mpll_ctrl)
+		goto resume;
+
 	/* wait for vblank start on active crtcs, disable memory access */
 	for (i = 0; i < 2; i++) {
 		if (!(crtc_mask & (1 << i)))
@@ -335,4 +347,48 @@
 	spin_unlock_irqrestore(&dev_priv->context_switch_lock, flags);
 
 	kfree(info);
+	return ret;
+}
+
+int
+nv40_pm_pwm_get(struct drm_device *dev, int line, u32 *divs, u32 *duty)
+{
+	if (line == 2) {
+		u32 reg = nv_rd32(dev, 0x0010f0);
+		if (reg & 0x80000000) {
+			*duty = (reg & 0x7fff0000) >> 16;
+			*divs = (reg & 0x00007fff);
+			return 0;
+		}
+	} else
+	if (line == 9) {
+		u32 reg = nv_rd32(dev, 0x0015f4);
+		if (reg & 0x80000000) {
+			*divs = nv_rd32(dev, 0x0015f8);
+			*duty = (reg & 0x7fffffff);
+			return 0;
+		}
+	} else {
+		NV_ERROR(dev, "unknown pwm ctrl for gpio %d\n", line);
+		return -ENODEV;
+	}
+
+	return -EINVAL;
+}
+
+int
+nv40_pm_pwm_set(struct drm_device *dev, int line, u32 divs, u32 duty)
+{
+	if (line == 2) {
+		nv_wr32(dev, 0x0010f0, 0x80000000 | (duty << 16) | divs);
+	} else
+	if (line == 9) {
+		nv_wr32(dev, 0x0015f8, divs);
+		nv_wr32(dev, 0x0015f4, duty | 0x80000000);
+	} else {
+		NV_ERROR(dev, "unknown pwm ctrl for gpio %d\n", line);
+		return -ENODEV;
+	}
+
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nv50_crtc.c b/drivers/gpu/drm/nouveau/nv50_crtc.c
index 882080e..8f6c2ac 100644
--- a/drivers/gpu/drm/nouveau/nv50_crtc.c
+++ b/drivers/gpu/drm/nouveau/nv50_crtc.c
@@ -132,33 +132,42 @@
 }
 
 static int
-nv50_crtc_set_dither(struct nouveau_crtc *nv_crtc, bool on, bool update)
+nv50_crtc_set_dither(struct nouveau_crtc *nv_crtc, bool update)
 {
-	struct drm_device *dev = nv_crtc->base.dev;
-	struct nouveau_channel *evo = nv50_display(dev)->master;
-	int ret;
+	struct nouveau_channel *evo = nv50_display(nv_crtc->base.dev)->master;
+	struct nouveau_connector *nv_connector;
+	struct drm_connector *connector;
+	int head = nv_crtc->index, ret;
+	u32 mode = 0x00;
 
-	NV_DEBUG_KMS(dev, "\n");
+	nv_connector = nouveau_crtc_connector_get(nv_crtc);
+	connector = &nv_connector->base;
+	if (nv_connector->dithering_mode == DITHERING_MODE_AUTO) {
+		if (nv_crtc->base.fb->depth > connector->display_info.bpc * 3)
+			mode = DITHERING_MODE_DYNAMIC2X2;
+	} else {
+		mode = nv_connector->dithering_mode;
+	}
+
+	if (nv_connector->dithering_depth == DITHERING_DEPTH_AUTO) {
+		if (connector->display_info.bpc >= 8)
+			mode |= DITHERING_DEPTH_8BPC;
+	} else {
+		mode |= nv_connector->dithering_depth;
+	}
 
 	ret = RING_SPACE(evo, 2 + (update ? 2 : 0));
-	if (ret) {
-		NV_ERROR(dev, "no space while setting dither\n");
-		return ret;
+	if (ret == 0) {
+		BEGIN_RING(evo, 0, NV50_EVO_CRTC(head, DITHER_CTRL), 1);
+		OUT_RING  (evo, mode);
+		if (update) {
+			BEGIN_RING(evo, 0, NV50_EVO_UPDATE, 1);
+			OUT_RING  (evo, 0);
+			FIRE_RING (evo);
+		}
 	}
 
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, DITHER_CTRL), 1);
-	if (on)
-		OUT_RING(evo, NV50_EVO_CRTC_DITHER_CTRL_ON);
-	else
-		OUT_RING(evo, NV50_EVO_CRTC_DITHER_CTRL_OFF);
-
-	if (update) {
-		BEGIN_RING(evo, 0, NV50_EVO_UPDATE, 1);
-		OUT_RING(evo, 0);
-		FIRE_RING(evo);
-	}
-
-	return 0;
+	return ret;
 }
 
 struct nouveau_connector *
@@ -180,80 +189,103 @@
 }
 
 static int
-nv50_crtc_set_scale(struct nouveau_crtc *nv_crtc, int scaling_mode, bool update)
+nv50_crtc_set_scale(struct nouveau_crtc *nv_crtc, bool update)
 {
-	struct nouveau_connector *nv_connector =
-		nouveau_crtc_connector_get(nv_crtc);
-	struct drm_device *dev = nv_crtc->base.dev;
+	struct nouveau_connector *nv_connector;
+	struct drm_crtc *crtc = &nv_crtc->base;
+	struct drm_device *dev = crtc->dev;
 	struct nouveau_channel *evo = nv50_display(dev)->master;
-	struct drm_display_mode *native_mode = NULL;
-	struct drm_display_mode *mode = &nv_crtc->base.mode;
-	uint32_t outX, outY, horiz, vert;
-	int ret;
+	struct drm_display_mode *umode = &crtc->mode;
+	struct drm_display_mode *omode;
+	int scaling_mode, ret;
+	u32 ctrl = 0, oX, oY;
 
 	NV_DEBUG_KMS(dev, "\n");
 
-	switch (scaling_mode) {
-	case DRM_MODE_SCALE_NONE:
-		break;
-	default:
-		if (!nv_connector || !nv_connector->native_mode) {
-			NV_ERROR(dev, "No native mode, forcing panel scaling\n");
-			scaling_mode = DRM_MODE_SCALE_NONE;
-		} else {
-			native_mode = nv_connector->native_mode;
-		}
-		break;
+	nv_connector = nouveau_crtc_connector_get(nv_crtc);
+	if (!nv_connector || !nv_connector->native_mode) {
+		NV_ERROR(dev, "no native mode, forcing panel scaling\n");
+		scaling_mode = DRM_MODE_SCALE_NONE;
+	} else {
+		scaling_mode = nv_connector->scaling_mode;
 	}
 
-	switch (scaling_mode) {
-	case DRM_MODE_SCALE_ASPECT:
-		horiz = (native_mode->hdisplay << 19) / mode->hdisplay;
-		vert = (native_mode->vdisplay << 19) / mode->vdisplay;
+	/* start off at the resolution we programmed the crtc for, this
+	 * effectively handles NONE/FULL scaling
+	 */
+	if (scaling_mode != DRM_MODE_SCALE_NONE)
+		omode = nv_connector->native_mode;
+	else
+		omode = umode;
 
-		if (vert > horiz) {
-			outX = (mode->hdisplay * horiz) >> 19;
-			outY = (mode->vdisplay * horiz) >> 19;
+	oX = omode->hdisplay;
+	oY = omode->vdisplay;
+	if (omode->flags & DRM_MODE_FLAG_DBLSCAN)
+		oY *= 2;
+
+	/* add overscan compensation if necessary, will keep the aspect
+	 * ratio the same as the backend mode unless overridden by the
+	 * user setting both hborder and vborder properties.
+	 */
+	if (nv_connector && ( nv_connector->underscan == UNDERSCAN_ON ||
+			     (nv_connector->underscan == UNDERSCAN_AUTO &&
+			      nv_connector->edid &&
+			      drm_detect_hdmi_monitor(nv_connector->edid)))) {
+		u32 bX = nv_connector->underscan_hborder;
+		u32 bY = nv_connector->underscan_vborder;
+		u32 aspect = (oY << 19) / oX;
+
+		if (bX) {
+			oX -= (bX * 2);
+			if (bY) oY -= (bY * 2);
+			else    oY  = ((oX * aspect) + (aspect / 2)) >> 19;
 		} else {
-			outX = (mode->hdisplay * vert) >> 19;
-			outY = (mode->vdisplay * vert) >> 19;
+			oX -= (oX >> 4) + 32;
+			if (bY) oY -= (bY * 2);
+			else    oY  = ((oX * aspect) + (aspect / 2)) >> 19;
 		}
-		break;
-	case DRM_MODE_SCALE_FULLSCREEN:
-		outX = native_mode->hdisplay;
-		outY = native_mode->vdisplay;
-		break;
+	}
+
+	/* handle CENTER/ASPECT scaling, taking into account the areas
+	 * removed already for overscan compensation
+	 */
+	switch (scaling_mode) {
 	case DRM_MODE_SCALE_CENTER:
-	case DRM_MODE_SCALE_NONE:
+		oX = min((u32)umode->hdisplay, oX);
+		oY = min((u32)umode->vdisplay, oY);
+		/* fall-through */
+	case DRM_MODE_SCALE_ASPECT:
+		if (oY < oX) {
+			u32 aspect = (umode->hdisplay << 19) / umode->vdisplay;
+			oX = ((oY * aspect) + (aspect / 2)) >> 19;
+		} else {
+			u32 aspect = (umode->vdisplay << 19) / umode->hdisplay;
+			oY = ((oX * aspect) + (aspect / 2)) >> 19;
+		}
+		break;
 	default:
-		outX = mode->hdisplay;
-		outY = mode->vdisplay;
 		break;
 	}
 
-	ret = RING_SPACE(evo, update ? 7 : 5);
+	if (umode->hdisplay != oX || umode->vdisplay != oY ||
+	    umode->flags & DRM_MODE_FLAG_INTERLACE ||
+	    umode->flags & DRM_MODE_FLAG_DBLSCAN)
+		ctrl |= NV50_EVO_CRTC_SCALE_CTRL_ACTIVE;
+
+	ret = RING_SPACE(evo, 5);
 	if (ret)
 		return ret;
 
-	/* Got a better name for SCALER_ACTIVE? */
-	/* One day i've got to really figure out why this is needed. */
 	BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, SCALE_CTRL), 1);
-	if ((mode->flags & DRM_MODE_FLAG_DBLSCAN) ||
-	    (mode->flags & DRM_MODE_FLAG_INTERLACE) ||
-	    mode->hdisplay != outX || mode->vdisplay != outY) {
-		OUT_RING(evo, NV50_EVO_CRTC_SCALE_CTRL_ACTIVE);
-	} else {
-		OUT_RING(evo, NV50_EVO_CRTC_SCALE_CTRL_INACTIVE);
-	}
-
+	OUT_RING  (evo, ctrl);
 	BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, SCALE_RES1), 2);
-	OUT_RING(evo, outY << 16 | outX);
-	OUT_RING(evo, outY << 16 | outX);
+	OUT_RING  (evo, oY << 16 | oX);
+	OUT_RING  (evo, oY << 16 | oX);
 
 	if (update) {
-		BEGIN_RING(evo, 0, NV50_EVO_UPDATE, 1);
-		OUT_RING(evo, 0);
-		FIRE_RING(evo);
+		nv50_display_flip_stop(crtc);
+		nv50_display_sync(dev);
+		nv50_display_flip_next(crtc, crtc->fb, NULL);
 	}
 
 	return 0;
@@ -333,7 +365,6 @@
 	nouveau_bo_ref(NULL, &nv_crtc->lut.nvbo);
 	nouveau_bo_unmap(nv_crtc->cursor.nvbo);
 	nouveau_bo_ref(NULL, &nv_crtc->cursor.nvbo);
-	kfree(nv_crtc->mode);
 	kfree(nv_crtc);
 }
 
@@ -441,39 +472,6 @@
 {
 }
 
-static int
-nv50_crtc_wait_complete(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_timer_engine *ptimer = &dev_priv->engine.timer;
-	struct nv50_display *disp = nv50_display(dev);
-	struct nouveau_channel *evo = disp->master;
-	u64 start;
-	int ret;
-
-	ret = RING_SPACE(evo, 6);
-	if (ret)
-		return ret;
-	BEGIN_RING(evo, 0, 0x0084, 1);
-	OUT_RING  (evo, 0x80000000);
-	BEGIN_RING(evo, 0, 0x0080, 1);
-	OUT_RING  (evo, 0);
-	BEGIN_RING(evo, 0, 0x0084, 1);
-	OUT_RING  (evo, 0x00000000);
-
-	nv_wo32(disp->ntfy, 0x000, 0x00000000);
-	FIRE_RING (evo);
-
-	start = ptimer->read(dev);
-	do {
-		if (nv_ro32(disp->ntfy, 0x000))
-			return 0;
-	} while (ptimer->read(dev) - start < 2000000000ULL);
-
-	return -EBUSY;
-}
-
 static void
 nv50_crtc_prepare(struct drm_crtc *crtc)
 {
@@ -497,7 +495,7 @@
 
 	nv50_crtc_blank(nv_crtc, false);
 	drm_vblank_post_modeset(dev, nv_crtc->index);
-	nv50_crtc_wait_complete(crtc);
+	nv50_display_sync(dev);
 	nv50_display_flip_next(crtc, crtc->fb, NULL);
 }
 
@@ -593,90 +591,76 @@
 }
 
 static int
-nv50_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *mode,
-		   struct drm_display_mode *adjusted_mode, int x, int y,
+nv50_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *umode,
+		   struct drm_display_mode *mode, int x, int y,
 		   struct drm_framebuffer *old_fb)
 {
 	struct drm_device *dev = crtc->dev;
 	struct nouveau_channel *evo = nv50_display(dev)->master;
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-	struct nouveau_connector *nv_connector = NULL;
-	uint32_t hsync_dur,  vsync_dur, hsync_start_to_end, vsync_start_to_end;
-	uint32_t hunk1, vunk1, vunk2a, vunk2b;
+	u32 head = nv_crtc->index * 0x400;
+	u32 ilace = (mode->flags & DRM_MODE_FLAG_INTERLACE) ? 2 : 1;
+	u32 vscan = (mode->flags & DRM_MODE_FLAG_DBLSCAN) ? 2 : 1;
+	u32 hactive, hsynce, hbackp, hfrontp, hblanke, hblanks;
+	u32 vactive, vsynce, vbackp, vfrontp, vblanke, vblanks;
+	u32 vblan2e = 0, vblan2s = 1;
 	int ret;
 
-	/* Find the connector attached to this CRTC */
-	nv_connector = nouveau_crtc_connector_get(nv_crtc);
+	/* hw timing description looks like this:
+	 *
+	 * <sync> <back porch> <---------display---------> <front porch>
+	 * ______
+	 *       |____________|---------------------------|____________|
+	 *
+	 *       ^ synce      ^ blanke                    ^ blanks     ^ active
+	 *
+	 * interlaced modes also have 2 additional values pointing at the end
+	 * and start of the next field's blanking period.
+	 */
 
-	*nv_crtc->mode = *adjusted_mode;
+	hactive = mode->htotal;
+	hsynce  = mode->hsync_end - mode->hsync_start - 1;
+	hbackp  = mode->htotal - mode->hsync_end;
+	hblanke = hsynce + hbackp;
+	hfrontp = mode->hsync_start - mode->hdisplay;
+	hblanks = mode->htotal - hfrontp - 1;
 
-	NV_DEBUG_KMS(dev, "index %d\n", nv_crtc->index);
-
-	hsync_dur = adjusted_mode->hsync_end - adjusted_mode->hsync_start;
-	vsync_dur = adjusted_mode->vsync_end - adjusted_mode->vsync_start;
-	hsync_start_to_end = adjusted_mode->htotal - adjusted_mode->hsync_start;
-	vsync_start_to_end = adjusted_mode->vtotal - adjusted_mode->vsync_start;
-	/* I can't give this a proper name, anyone else can? */
-	hunk1 = adjusted_mode->htotal -
-		adjusted_mode->hsync_start + adjusted_mode->hdisplay;
-	vunk1 = adjusted_mode->vtotal -
-		adjusted_mode->vsync_start + adjusted_mode->vdisplay;
-	/* Another strange value, this time only for interlaced adjusted_modes. */
-	vunk2a = 2 * adjusted_mode->vtotal -
-		 adjusted_mode->vsync_start + adjusted_mode->vdisplay;
-	vunk2b = adjusted_mode->vtotal -
-		 adjusted_mode->vsync_start + adjusted_mode->vtotal;
-
-	if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) {
-		vsync_dur /= 2;
-		vsync_start_to_end  /= 2;
-		vunk1 /= 2;
-		vunk2a /= 2;
-		vunk2b /= 2;
-		/* magic */
-		if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN) {
-			vsync_start_to_end -= 1;
-			vunk1 -= 1;
-			vunk2a -= 1;
-			vunk2b -= 1;
-		}
+	vactive = mode->vtotal * vscan / ilace;
+	vsynce  = ((mode->vsync_end - mode->vsync_start) * vscan / ilace) - 1;
+	vbackp  = (mode->vtotal - mode->vsync_end) * vscan / ilace;
+	vblanke = vsynce + vbackp;
+	vfrontp = (mode->vsync_start - mode->vdisplay) * vscan / ilace;
+	vblanks = vactive - vfrontp - 1;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE) {
+		vblan2e = vactive + vsynce + vbackp;
+		vblan2s = vblan2e + (mode->vdisplay * vscan / ilace);
+		vactive = (vactive * 2) + 1;
 	}
 
-	ret = RING_SPACE(evo, 17);
-	if (ret)
-		return ret;
-
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, CLOCK), 2);
-	OUT_RING(evo, adjusted_mode->clock | 0x800000);
-	OUT_RING(evo, (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) ? 2 : 0);
-
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, DISPLAY_START), 5);
-	OUT_RING(evo, 0);
-	OUT_RING(evo, (adjusted_mode->vtotal << 16) | adjusted_mode->htotal);
-	OUT_RING(evo, (vsync_dur - 1) << 16 | (hsync_dur - 1));
-	OUT_RING(evo, (vsync_start_to_end - 1) << 16 |
-			(hsync_start_to_end - 1));
-	OUT_RING(evo, (vunk1 - 1) << 16 | (hunk1 - 1));
-
-	if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) {
-		BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, UNK0824), 1);
-		OUT_RING(evo, (vunk2b - 1) << 16 | (vunk2a - 1));
-	} else {
-		OUT_RING(evo, 0);
-		OUT_RING(evo, 0);
+	ret = RING_SPACE(evo, 18);
+	if (ret == 0) {
+		BEGIN_RING(evo, 0, 0x0804 + head, 2);
+		OUT_RING  (evo, 0x00800000 | mode->clock);
+		OUT_RING  (evo, (ilace == 2) ? 2 : 0);
+		BEGIN_RING(evo, 0, 0x0810 + head, 6);
+		OUT_RING  (evo, 0x00000000); /* border colour */
+		OUT_RING  (evo, (vactive << 16) | hactive);
+		OUT_RING  (evo, ( vsynce << 16) | hsynce);
+		OUT_RING  (evo, (vblanke << 16) | hblanke);
+		OUT_RING  (evo, (vblanks << 16) | hblanks);
+		OUT_RING  (evo, (vblan2e << 16) | vblan2s);
+		BEGIN_RING(evo, 0, 0x082c + head, 1);
+		OUT_RING  (evo, 0x00000000);
+		BEGIN_RING(evo, 0, 0x0900 + head, 1);
+		OUT_RING  (evo, 0x00000311); /* makes sync channel work */
+		BEGIN_RING(evo, 0, 0x08c8 + head, 1);
+		OUT_RING  (evo, (umode->vdisplay << 16) | umode->hdisplay);
+		BEGIN_RING(evo, 0, 0x08d4 + head, 1);
+		OUT_RING  (evo, 0x00000000); /* screen position */
 	}
 
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, UNK082C), 1);
-	OUT_RING(evo, 0);
-
-	/* This is the actual resolution of the mode. */
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, REAL_RES), 1);
-	OUT_RING(evo, (mode->vdisplay << 16) | mode->hdisplay);
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(nv_crtc->index, SCALE_CENTER_OFFSET), 1);
-	OUT_RING(evo, NV50_EVO_CRTC_SCALE_CENTER_OFFSET_VAL(0, 0));
-
-	nv_crtc->set_dither(nv_crtc, nv_connector->use_dithering, false);
-	nv_crtc->set_scale(nv_crtc, nv_connector->scaling_mode, false);
+	nv_crtc->set_dither(nv_crtc, false);
+	nv_crtc->set_scale(nv_crtc, false);
 
 	return nv50_crtc_do_mode_set_base(crtc, old_fb, x, y, false);
 }
@@ -692,7 +676,7 @@
 	if (ret)
 		return ret;
 
-	ret = nv50_crtc_wait_complete(crtc);
+	ret = nv50_display_sync(crtc->dev);
 	if (ret)
 		return ret;
 
@@ -711,7 +695,7 @@
 	if (ret)
 		return ret;
 
-	return nv50_crtc_wait_complete(crtc);
+	return nv50_display_sync(crtc->dev);
 }
 
 static const struct drm_crtc_helper_funcs nv50_crtc_helper_funcs = {
@@ -737,12 +721,6 @@
 	if (!nv_crtc)
 		return -ENOMEM;
 
-	nv_crtc->mode = kzalloc(sizeof(*nv_crtc->mode), GFP_KERNEL);
-	if (!nv_crtc->mode) {
-		kfree(nv_crtc);
-		return -ENOMEM;
-	}
-
 	/* Default CLUT parameters, will be activated on the hw upon
 	 * first mode set.
 	 */
@@ -764,7 +742,6 @@
 	}
 
 	if (ret) {
-		kfree(nv_crtc->mode);
 		kfree(nv_crtc);
 		return ret;
 	}
diff --git a/drivers/gpu/drm/nouveau/nv50_dac.c b/drivers/gpu/drm/nouveau/nv50_dac.c
index 808f3ec..a0f2beb 100644
--- a/drivers/gpu/drm/nouveau/nv50_dac.c
+++ b/drivers/gpu/drm/nouveau/nv50_dac.c
@@ -200,11 +200,6 @@
 }
 
 static void
-nv50_dac_prepare(struct drm_encoder *encoder)
-{
-}
-
-static void
 nv50_dac_commit(struct drm_encoder *encoder)
 {
 }
@@ -266,7 +261,7 @@
 	.save = nv50_dac_save,
 	.restore = nv50_dac_restore,
 	.mode_fixup = nv50_dac_mode_fixup,
-	.prepare = nv50_dac_prepare,
+	.prepare = nv50_dac_disconnect,
 	.commit = nv50_dac_commit,
 	.mode_set = nv50_dac_mode_set,
 	.get_crtc = nv50_dac_crtc_get,
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index d23ca00..7ba28e0 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -50,9 +50,53 @@
 	return 4;
 }
 
+static int
+evo_icmd(struct drm_device *dev, int ch, u32 mthd, u32 data)
+{
+	int ret = 0;
+	nv_mask(dev, 0x610300 + (ch * 0x08), 0x00000001, 0x00000001);
+	nv_wr32(dev, 0x610304 + (ch * 0x08), data);
+	nv_wr32(dev, 0x610300 + (ch * 0x08), 0x80000001 | mthd);
+	if (!nv_wait(dev, 0x610300 + (ch * 0x08), 0x80000000, 0x00000000))
+		ret = -EBUSY;
+	if (ret || (nouveau_reg_debug & NOUVEAU_REG_DEBUG_EVO))
+		NV_INFO(dev, "EvoPIO: %d 0x%04x 0x%08x\n", ch, mthd, data);
+	nv_mask(dev, 0x610300 + (ch * 0x08), 0x00000001, 0x00000000);
+	return ret;
+}
+
 int
 nv50_display_early_init(struct drm_device *dev)
 {
+	u32 ctrl = nv_rd32(dev, 0x610200);
+	int i;
+
+	/* check if master evo channel is already active, a good a sign as any
+	 * that the display engine is in a weird state (hibernate/kexec), if
+	 * it is, do our best to reset the display engine...
+	 */
+	if ((ctrl & 0x00000003) == 0x00000003) {
+		NV_INFO(dev, "PDISP: EVO(0) 0x%08x, resetting...\n", ctrl);
+
+		/* deactivate both heads first, PDISP will disappear forever
+		 * (well, until you power cycle) on some boards as soon as
+		 * PMC_ENABLE is hit unless they are..
+		 */
+		for (i = 0; i < 2; i++) {
+			evo_icmd(dev, 0, 0x0880 + (i * 0x400), 0x05000000);
+			evo_icmd(dev, 0, 0x089c + (i * 0x400), 0);
+			evo_icmd(dev, 0, 0x0840 + (i * 0x400), 0);
+			evo_icmd(dev, 0, 0x0844 + (i * 0x400), 0);
+			evo_icmd(dev, 0, 0x085c + (i * 0x400), 0);
+			evo_icmd(dev, 0, 0x0874 + (i * 0x400), 0);
+		}
+		evo_icmd(dev, 0, 0x0080, 0);
+
+		/* reset PDISP */
+		nv_mask(dev, 0x000200, 0x40000000, 0x00000000);
+		nv_mask(dev, 0x000200, 0x40000000, 0x40000000);
+	}
+
 	return 0;
 }
 
@@ -62,11 +106,40 @@
 }
 
 int
-nv50_display_init(struct drm_device *dev)
+nv50_display_sync(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
-	struct drm_connector *connector;
+	struct nouveau_timer_engine *ptimer = &dev_priv->engine.timer;
+	struct nv50_display *disp = nv50_display(dev);
+	struct nouveau_channel *evo = disp->master;
+	u64 start;
+	int ret;
+
+	ret = RING_SPACE(evo, 6);
+	if (ret == 0) {
+		BEGIN_RING(evo, 0, 0x0084, 1);
+		OUT_RING  (evo, 0x80000000);
+		BEGIN_RING(evo, 0, 0x0080, 1);
+		OUT_RING  (evo, 0);
+		BEGIN_RING(evo, 0, 0x0084, 1);
+		OUT_RING  (evo, 0x00000000);
+
+		nv_wo32(disp->ntfy, 0x000, 0x00000000);
+		FIRE_RING (evo);
+
+		start = ptimer->read(dev);
+		do {
+			if (nv_ro32(disp->ntfy, 0x000))
+				return 0;
+		} while (ptimer->read(dev) - start < 2000000000ULL);
+	}
+
+	return -EBUSY;
+}
+
+int
+nv50_display_init(struct drm_device *dev)
+{
 	struct nouveau_channel *evo;
 	int ret, i;
 	u32 val;
@@ -161,16 +234,6 @@
 		     NV50_PDISPLAY_INTR_EN_1_CLK_UNK20 |
 		     NV50_PDISPLAY_INTR_EN_1_CLK_UNK40);
 
-	/* enable hotplug interrupts */
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-		struct nouveau_connector *conn = nouveau_connector(connector);
-
-		if (conn->dcb->gpio_tag == 0xff)
-			continue;
-
-		pgpio->irq_enable(dev, conn->dcb->gpio_tag, true);
-	}
-
 	ret = nv50_evo_init(dev);
 	if (ret)
 		return ret;
@@ -178,36 +241,19 @@
 
 	nv_wr32(dev, NV50_PDISPLAY_OBJECTS, (evo->ramin->vinst >> 8) | 9);
 
-	ret = RING_SPACE(evo, 15);
+	ret = RING_SPACE(evo, 3);
 	if (ret)
 		return ret;
 	BEGIN_RING(evo, 0, NV50_EVO_UNK84, 2);
-	OUT_RING(evo, NV50_EVO_UNK84_NOTIFY_DISABLED);
-	OUT_RING(evo, NvEvoSync);
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(0, FB_DMA), 1);
-	OUT_RING(evo, NV50_EVO_CRTC_FB_DMA_HANDLE_NONE);
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(0, UNK0800), 1);
-	OUT_RING(evo, 0);
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(0, DISPLAY_START), 1);
-	OUT_RING(evo, 0);
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(0, UNK082C), 1);
-	OUT_RING(evo, 0);
-	/* required to make display sync channels not hate life */
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(0, UNK900), 1);
-	OUT_RING  (evo, 0x00000311);
-	BEGIN_RING(evo, 0, NV50_EVO_CRTC(1, UNK900), 1);
-	OUT_RING  (evo, 0x00000311);
-	FIRE_RING(evo);
-	if (!nv_wait(dev, 0x640004, 0xffffffff, evo->dma.put << 2))
-		NV_ERROR(dev, "evo pushbuf stalled\n");
+	OUT_RING  (evo, NV50_EVO_UNK84_NOTIFY_DISABLED);
+	OUT_RING  (evo, NvEvoSync);
 
-
-	return 0;
+	return nv50_display_sync(dev);
 }
 
-static int nv50_display_disable(struct drm_device *dev)
+void
+nv50_display_fini(struct drm_device *dev)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nv50_display *disp = nv50_display(dev);
 	struct nouveau_channel *evo = disp->master;
 	struct drm_crtc *drm_crtc;
@@ -270,18 +316,10 @@
 
 	/* disable interrupts. */
 	nv_wr32(dev, NV50_PDISPLAY_INTR_EN_1, 0x00000000);
-
-	/* disable hotplug interrupts */
-	nv_wr32(dev, 0xe054, 0xffffffff);
-	nv_wr32(dev, 0xe050, 0x00000000);
-	if (dev_priv->chipset >= 0x90) {
-		nv_wr32(dev, 0xe074, 0xffffffff);
-		nv_wr32(dev, 0xe070, 0x00000000);
-	}
-	return 0;
 }
 
-int nv50_display_create(struct drm_device *dev)
+int
+nv50_display_create(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct dcb_table *dcb = &dev_priv->vbios.dcb;
@@ -341,7 +379,7 @@
 	tasklet_init(&priv->tasklet, nv50_display_bh, (unsigned long)dev);
 	nouveau_irq_register(dev, 26, nv50_display_isr);
 
-	ret = nv50_display_init(dev);
+	ret = nv50_evo_create(dev);
 	if (ret) {
 		nv50_display_destroy(dev);
 		return ret;
@@ -357,7 +395,7 @@
 
 	NV_DEBUG_KMS(dev, "\n");
 
-	nv50_display_disable(dev);
+	nv50_evo_destroy(dev);
 	nouveau_irq_unregister(dev, 26);
 	kfree(disp);
 }
@@ -521,7 +559,7 @@
 		} else {
 			/* determine number of lvds links */
 			if (nv_connector && nv_connector->edid &&
-			    nv_connector->dcb->type == DCB_CONNECTOR_LVDS_SPWG) {
+			    nv_connector->type == DCB_CONNECTOR_LVDS_SPWG) {
 				/* http://www.spwg.org */
 				if (((u8 *)nv_connector->edid)[121] == 2)
 					script |= 0x0100;
@@ -616,7 +654,7 @@
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nv50_display *disp = nv50_display(dev);
 	u32 unk30 = nv_rd32(dev, 0x610030), mc;
-	int i, crtc, or, type = OUTPUT_ANY;
+	int i, crtc, or = 0, type = OUTPUT_ANY;
 
 	NV_DEBUG_KMS(dev, "0x610030: 0x%08x\n", unk30);
 	disp->irq.dcb = NULL;
@@ -708,7 +746,7 @@
 	struct nv50_display *disp = nv50_display(dev);
 	u32 unk30 = nv_rd32(dev, 0x610030), tmp, pclk, script, mc = 0;
 	struct dcb_entry *dcb;
-	int i, crtc, or, type = OUTPUT_ANY;
+	int i, crtc, or = 0, type = OUTPUT_ANY;
 
 	NV_DEBUG_KMS(dev, "0x610030: 0x%08x\n", unk30);
 	dcb = disp->irq.dcb;
@@ -722,8 +760,8 @@
 	if (crtc >= 0) {
 		pclk  = nv_rd32(dev, NV50_PDISPLAY_CRTC_P(crtc, CLOCK));
 		pclk &= 0x003fffff;
-
-		nv50_crtc_set_clock(dev, crtc, pclk);
+		if (pclk)
+			nv50_crtc_set_clock(dev, crtc, pclk);
 
 		tmp = nv_rd32(dev, NV50_PDISPLAY_CRTC_CLK_CTRL2(crtc));
 		tmp &= ~0x000000f;
diff --git a/drivers/gpu/drm/nouveau/nv50_display.h b/drivers/gpu/drm/nouveau/nv50_display.h
index c2da503..95874f7 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.h
+++ b/drivers/gpu/drm/nouveau/nv50_display.h
@@ -69,14 +69,18 @@
 void nv50_display_late_takedown(struct drm_device *dev);
 int nv50_display_create(struct drm_device *dev);
 int nv50_display_init(struct drm_device *dev);
+void nv50_display_fini(struct drm_device *dev);
 void nv50_display_destroy(struct drm_device *dev);
 int nv50_crtc_blank(struct nouveau_crtc *, bool blank);
 int nv50_crtc_set_clock(struct drm_device *, int head, int pclk);
 
+int  nv50_display_sync(struct drm_device *);
 int  nv50_display_flip_next(struct drm_crtc *, struct drm_framebuffer *,
 			    struct nouveau_channel *chan);
 void nv50_display_flip_stop(struct drm_crtc *);
 
+int  nv50_evo_create(struct drm_device *dev);
+void nv50_evo_destroy(struct drm_device *dev);
 int  nv50_evo_init(struct drm_device *dev);
 void nv50_evo_fini(struct drm_device *dev);
 void nv50_evo_dmaobj_init(struct nouveau_gpuobj *, u32 memtype, u64 base,
diff --git a/drivers/gpu/drm/nouveau/nv50_evo.c b/drivers/gpu/drm/nouveau/nv50_evo.c
index c99d975..9b962e9 100644
--- a/drivers/gpu/drm/nouveau/nv50_evo.c
+++ b/drivers/gpu/drm/nouveau/nv50_evo.c
@@ -218,7 +218,7 @@
 	}
 }
 
-static void
+void
 nv50_evo_destroy(struct drm_device *dev)
 {
 	struct nv50_display *disp = nv50_display(dev);
@@ -235,7 +235,7 @@
 	nv50_evo_channel_del(&disp->master);
 }
 
-static int
+int
 nv50_evo_create(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
@@ -388,12 +388,6 @@
 	struct nv50_display *disp = nv50_display(dev);
 	int ret, i;
 
-	if (!disp->master) {
-		ret = nv50_evo_create(dev);
-		if (ret)
-			return ret;
-	}
-
 	ret = nv50_evo_channel_init(disp->master);
 	if (ret)
 		return ret;
@@ -420,6 +414,4 @@
 
 	if (disp->master)
 		nv50_evo_channel_fini(disp->master);
-
-	nv50_evo_destroy(dev);
 }
diff --git a/drivers/gpu/drm/nouveau/nv50_fifo.c b/drivers/gpu/drm/nouveau/nv50_fifo.c
index c34a074..3bc2a56 100644
--- a/drivers/gpu/drm/nouveau/nv50_fifo.c
+++ b/drivers/gpu/drm/nouveau/nv50_fifo.c
@@ -230,6 +230,7 @@
 	struct drm_device *dev = chan->dev;
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_gpuobj *ramfc = NULL;
+        uint64_t ib_offset = chan->pushbuf_base + chan->dma.ib_base * 4;
 	unsigned long flags;
 	int ret;
 
@@ -280,8 +281,9 @@
 	nv_wo32(ramfc, 0x7c, 0x30000001);
 	nv_wo32(ramfc, 0x78, 0x00000000);
 	nv_wo32(ramfc, 0x3c, 0x403f6078);
-	nv_wo32(ramfc, 0x50, chan->pushbuf_base + chan->dma.ib_base * 4);
-	nv_wo32(ramfc, 0x54, drm_order(chan->dma.ib_max + 1) << 16);
+	nv_wo32(ramfc, 0x50, lower_32_bits(ib_offset));
+	nv_wo32(ramfc, 0x54, upper_32_bits(ib_offset) |
+                drm_order(chan->dma.ib_max + 1) << 16);
 
 	if (dev_priv->chipset != 0x50) {
 		nv_wo32(chan->ramin, 0, chan->id);
diff --git a/drivers/gpu/drm/nouveau/nv50_gpio.c b/drivers/gpu/drm/nouveau/nv50_gpio.c
index 793a5cc..f429e6a 100644
--- a/drivers/gpu/drm/nouveau/nv50_gpio.c
+++ b/drivers/gpu/drm/nouveau/nv50_gpio.c
@@ -25,229 +25,95 @@
 #include "drmP.h"
 #include "nouveau_drv.h"
 #include "nouveau_hw.h"
+#include "nouveau_gpio.h"
 
 #include "nv50_display.h"
 
-static void nv50_gpio_isr(struct drm_device *dev);
-static void nv50_gpio_isr_bh(struct work_struct *work);
-
-struct nv50_gpio_priv {
-	struct list_head handlers;
-	spinlock_t lock;
-};
-
-struct nv50_gpio_handler {
-	struct drm_device *dev;
-	struct list_head head;
-	struct work_struct work;
-	bool inhibit;
-
-	struct dcb_gpio_entry *gpio;
-
-	void (*handler)(void *data, int state);
-	void *data;
-};
-
 static int
-nv50_gpio_location(struct dcb_gpio_entry *gpio, uint32_t *reg, uint32_t *shift)
+nv50_gpio_location(int line, u32 *reg, u32 *shift)
 {
 	const uint32_t nv50_gpio_reg[4] = { 0xe104, 0xe108, 0xe280, 0xe284 };
 
-	if (gpio->line >= 32)
+	if (line >= 32)
 		return -EINVAL;
 
-	*reg = nv50_gpio_reg[gpio->line >> 3];
-	*shift = (gpio->line & 7) << 2;
+	*reg = nv50_gpio_reg[line >> 3];
+	*shift = (line & 7) << 2;
 	return 0;
 }
 
 int
-nv50_gpio_get(struct drm_device *dev, enum dcb_gpio_tag tag)
+nv50_gpio_drive(struct drm_device *dev, int line, int dir, int out)
 {
-	struct dcb_gpio_entry *gpio;
-	uint32_t r, s, v;
+	u32 reg, shift;
 
-	gpio = nouveau_bios_gpio_entry(dev, tag);
-	if (!gpio)
-		return -ENOENT;
-
-	if (nv50_gpio_location(gpio, &r, &s))
+	if (nv50_gpio_location(line, &reg, &shift))
 		return -EINVAL;
 
-	v = nv_rd32(dev, r) >> (s + 2);
-	return ((v & 1) == (gpio->state[1] & 1));
+	nv_mask(dev, reg, 7 << shift, (((dir ^ 1) << 1) | out) << shift);
+	return 0;
 }
 
 int
-nv50_gpio_set(struct drm_device *dev, enum dcb_gpio_tag tag, int state)
+nv50_gpio_sense(struct drm_device *dev, int line)
 {
-	struct dcb_gpio_entry *gpio;
-	uint32_t r, s, v;
+	u32 reg, shift;
 
-	gpio = nouveau_bios_gpio_entry(dev, tag);
-	if (!gpio)
-		return -ENOENT;
-
-	if (nv50_gpio_location(gpio, &r, &s))
+	if (nv50_gpio_location(line, &reg, &shift))
 		return -EINVAL;
 
-	v  = nv_rd32(dev, r) & ~(0x3 << s);
-	v |= (gpio->state[state] ^ 2) << s;
-	nv_wr32(dev, r, v);
-	return 0;
-}
-
-int
-nvd0_gpio_get(struct drm_device *dev, enum dcb_gpio_tag tag)
-{
-	struct dcb_gpio_entry *gpio;
-	u32 v;
-
-	gpio = nouveau_bios_gpio_entry(dev, tag);
-	if (!gpio)
-		return -ENOENT;
-
-	v  = nv_rd32(dev, 0x00d610 + (gpio->line * 4));
-	v &= 0x00004000;
-	return (!!v == (gpio->state[1] & 1));
-}
-
-int
-nvd0_gpio_set(struct drm_device *dev, enum dcb_gpio_tag tag, int state)
-{
-	struct dcb_gpio_entry *gpio;
-	u32 v;
-
-	gpio = nouveau_bios_gpio_entry(dev, tag);
-	if (!gpio)
-		return -ENOENT;
-
-	v = gpio->state[state] ^ 2;
-
-	nv_mask(dev, 0x00d610 + (gpio->line * 4), 0x00003000, v << 12);
-	return 0;
-}
-
-int
-nv50_gpio_irq_register(struct drm_device *dev, enum dcb_gpio_tag tag,
-		       void (*handler)(void *, int), void *data)
-{
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
-	struct nv50_gpio_priv *priv = pgpio->priv;
-	struct nv50_gpio_handler *gpioh;
-	struct dcb_gpio_entry *gpio;
-	unsigned long flags;
-
-	gpio = nouveau_bios_gpio_entry(dev, tag);
-	if (!gpio)
-		return -ENOENT;
-
-	gpioh = kzalloc(sizeof(*gpioh), GFP_KERNEL);
-	if (!gpioh)
-		return -ENOMEM;
-
-	INIT_WORK(&gpioh->work, nv50_gpio_isr_bh);
-	gpioh->dev  = dev;
-	gpioh->gpio = gpio;
-	gpioh->handler = handler;
-	gpioh->data = data;
-
-	spin_lock_irqsave(&priv->lock, flags);
-	list_add(&gpioh->head, &priv->handlers);
-	spin_unlock_irqrestore(&priv->lock, flags);
-	return 0;
+	return !!(nv_rd32(dev, reg) & (4 << shift));
 }
 
 void
-nv50_gpio_irq_unregister(struct drm_device *dev, enum dcb_gpio_tag tag,
-			 void (*handler)(void *, int), void *data)
+nv50_gpio_irq_enable(struct drm_device *dev, int line, bool on)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
-	struct nv50_gpio_priv *priv = pgpio->priv;
-	struct nv50_gpio_handler *gpioh, *tmp;
-	struct dcb_gpio_entry *gpio;
-	LIST_HEAD(tofree);
-	unsigned long flags;
-
-	gpio = nouveau_bios_gpio_entry(dev, tag);
-	if (!gpio)
-		return;
-
-	spin_lock_irqsave(&priv->lock, flags);
-	list_for_each_entry_safe(gpioh, tmp, &priv->handlers, head) {
-		if (gpioh->gpio != gpio ||
-		    gpioh->handler != handler ||
-		    gpioh->data != data)
-			continue;
-		list_move(&gpioh->head, &tofree);
-	}
-	spin_unlock_irqrestore(&priv->lock, flags);
-
-	list_for_each_entry_safe(gpioh, tmp, &tofree, head) {
-		flush_work_sync(&gpioh->work);
-		kfree(gpioh);
-	}
-}
-
-bool
-nv50_gpio_irq_enable(struct drm_device *dev, enum dcb_gpio_tag tag, bool on)
-{
-	struct dcb_gpio_entry *gpio;
-	u32 reg, mask;
-
-	gpio = nouveau_bios_gpio_entry(dev, tag);
-	if (!gpio)
-		return false;
-
-	reg  = gpio->line < 16 ? 0xe050 : 0xe070;
-	mask = 0x00010001 << (gpio->line & 0xf);
+	u32 reg  = line < 16 ? 0xe050 : 0xe070;
+	u32 mask = 0x00010001 << (line & 0xf);
 
 	nv_wr32(dev, reg + 4, mask);
-	reg = nv_mask(dev, reg + 0, mask, on ? mask : 0);
-	return (reg & mask) == mask;
+	nv_mask(dev, reg + 0, mask, on ? mask : 0);
 }
 
-static int
-nv50_gpio_create(struct drm_device *dev)
+int
+nvd0_gpio_drive(struct drm_device *dev, int line, int dir, int out)
 {
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
-	struct nv50_gpio_priv *priv;
-
-	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
-
-	INIT_LIST_HEAD(&priv->handlers);
-	spin_lock_init(&priv->lock);
-	pgpio->priv = priv;
+	u32 data = ((dir ^ 1) << 13) | (out << 12);
+	nv_mask(dev, 0x00d610 + (line * 4), 0x00003000, data);
+	nv_mask(dev, 0x00d604, 0x00000001, 0x00000001); /* update? */
 	return 0;
 }
 
+int
+nvd0_gpio_sense(struct drm_device *dev, int line)
+{
+	return !!(nv_rd32(dev, 0x00d610 + (line * 4)) & 0x00004000);
+}
+
 static void
-nv50_gpio_destroy(struct drm_device *dev)
+nv50_gpio_isr(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
+	u32 intr0, intr1 = 0;
+	u32 hi, lo;
 
-	kfree(pgpio->priv);
-	pgpio->priv = NULL;
+	intr0 = nv_rd32(dev, 0xe054) & nv_rd32(dev, 0xe050);
+	if (dev_priv->chipset >= 0x90)
+		intr1 = nv_rd32(dev, 0xe074) & nv_rd32(dev, 0xe070);
+
+	hi = (intr0 & 0x0000ffff) | (intr1 << 16);
+	lo = (intr0 >> 16) | (intr1 & 0xffff0000);
+	nouveau_gpio_isr(dev, 0, hi | lo);
+
+	nv_wr32(dev, 0xe054, intr0);
+	if (dev_priv->chipset >= 0x90)
+		nv_wr32(dev, 0xe074, intr1);
 }
 
 int
 nv50_gpio_init(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
-	int ret;
-
-	if (!pgpio->priv) {
-		ret = nv50_gpio_create(dev);
-		if (ret)
-			return ret;
-	}
 
 	/* disable, and ack any pending gpio interrupts */
 	nv_wr32(dev, 0xe050, 0x00000000);
@@ -270,64 +136,4 @@
 	if (dev_priv->chipset >= 0x90)
 		nv_wr32(dev, 0xe070, 0x00000000);
 	nouveau_irq_unregister(dev, 21);
-
-	nv50_gpio_destroy(dev);
-}
-
-static void
-nv50_gpio_isr_bh(struct work_struct *work)
-{
-	struct nv50_gpio_handler *gpioh =
-		container_of(work, struct nv50_gpio_handler, work);
-	struct drm_nouveau_private *dev_priv = gpioh->dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
-	struct nv50_gpio_priv *priv = pgpio->priv;
-	unsigned long flags;
-	int state;
-
-	state = pgpio->get(gpioh->dev, gpioh->gpio->tag);
-	if (state < 0)
-		return;
-
-	gpioh->handler(gpioh->data, state);
-
-	spin_lock_irqsave(&priv->lock, flags);
-	gpioh->inhibit = false;
-	spin_unlock_irqrestore(&priv->lock, flags);
-}
-
-static void
-nv50_gpio_isr(struct drm_device *dev)
-{
-	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	struct nouveau_gpio_engine *pgpio = &dev_priv->engine.gpio;
-	struct nv50_gpio_priv *priv = pgpio->priv;
-	struct nv50_gpio_handler *gpioh;
-	u32 intr0, intr1 = 0;
-	u32 hi, lo, ch;
-
-	intr0 = nv_rd32(dev, 0xe054) & nv_rd32(dev, 0xe050);
-	if (dev_priv->chipset >= 0x90)
-		intr1 = nv_rd32(dev, 0xe074) & nv_rd32(dev, 0xe070);
-
-	hi = (intr0 & 0x0000ffff) | (intr1 << 16);
-	lo = (intr0 >> 16) | (intr1 & 0xffff0000);
-	ch = hi | lo;
-
-	nv_wr32(dev, 0xe054, intr0);
-	if (dev_priv->chipset >= 0x90)
-		nv_wr32(dev, 0xe074, intr1);
-
-	spin_lock(&priv->lock);
-	list_for_each_entry(gpioh, &priv->handlers, head) {
-		if (!(ch & (1 << gpioh->gpio->line)))
-			continue;
-
-		if (gpioh->inhibit)
-			continue;
-		gpioh->inhibit = true;
-
-		schedule_work(&gpioh->work);
-	}
-	spin_unlock(&priv->lock);
 }
diff --git a/drivers/gpu/drm/nouveau/nv50_graph.c b/drivers/gpu/drm/nouveau/nv50_graph.c
index 8c979b3..33d5711 100644
--- a/drivers/gpu/drm/nouveau/nv50_graph.c
+++ b/drivers/gpu/drm/nouveau/nv50_graph.c
@@ -131,8 +131,8 @@
 	NV_DEBUG(dev, "\n");
 
 	/* master reset */
-	nv_mask(dev, 0x000200, 0x00200100, 0x00000000);
-	nv_mask(dev, 0x000200, 0x00200100, 0x00200100);
+	nv_mask(dev, 0x000200, 0x00201000, 0x00000000);
+	nv_mask(dev, 0x000200, 0x00201000, 0x00201000);
 	nv_wr32(dev, 0x40008c, 0x00000004); /* HW_CTX_SWITCH_ENABLED */
 
 	/* reset/enable traps and interrupts */
@@ -616,9 +616,9 @@
 			}
 			break;
 		case 7: /* MP error */
-			if (ustatus & 0x00010000) {
+			if (ustatus & 0x04030000) {
 				nv50_pgraph_mp_trap(dev, i, display);
-				ustatus &= ~0x00010000;
+				ustatus &= ~0x04030000;
 			}
 			break;
 		case 8: /* TPDMA error */
diff --git a/drivers/gpu/drm/nouveau/nv50_grctx.c b/drivers/gpu/drm/nouveau/nv50_grctx.c
index d05c2c3..4b46d69 100644
--- a/drivers/gpu/drm/nouveau/nv50_grctx.c
+++ b/drivers/gpu/drm/nouveau/nv50_grctx.c
@@ -601,7 +601,7 @@
 					gr_def(ctx, offset + 0x1c, 0x00880000);
 					break;
 				case 0x86:
-					gr_def(ctx, offset + 0x1c, 0x008c0000);
+					gr_def(ctx, offset + 0x1c, 0x018c0000);
 					break;
 				case 0x92:
 				case 0x96:
diff --git a/drivers/gpu/drm/nouveau/nv50_pm.c b/drivers/gpu/drm/nouveau/nv50_pm.c
index 3d5a86b..0393721 100644
--- a/drivers/gpu/drm/nouveau/nv50_pm.c
+++ b/drivers/gpu/drm/nouveau/nv50_pm.c
@@ -25,122 +25,745 @@
 #include "drmP.h"
 #include "nouveau_drv.h"
 #include "nouveau_bios.h"
+#include "nouveau_hw.h"
 #include "nouveau_pm.h"
+#include "nouveau_hwsq.h"
 
-struct nv50_pm_state {
-	struct nouveau_pm_level *perflvl;
-	struct pll_lims pll;
-	enum pll_types type;
-	int N, M, P;
+enum clk_src {
+	clk_src_crystal,
+	clk_src_href,
+	clk_src_hclk,
+	clk_src_hclkm3,
+	clk_src_hclkm3d2,
+	clk_src_host,
+	clk_src_nvclk,
+	clk_src_sclk,
+	clk_src_mclk,
+	clk_src_vdec,
+	clk_src_dom6
 };
 
-int
-nv50_pm_clock_get(struct drm_device *dev, u32 id)
+static u32 read_clk(struct drm_device *, enum clk_src);
+
+static u32
+read_div(struct drm_device *dev)
 {
-	struct pll_lims pll;
-	int P, N, M, ret;
-	u32 reg0, reg1;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
 
-	ret = get_pll_limits(dev, id, &pll);
-	if (ret)
-		return ret;
+	switch (dev_priv->chipset) {
+	case 0x50: /* it exists, but only has bit 31, not the dividers.. */
+	case 0x84:
+	case 0x86:
+	case 0x98:
+	case 0xa0:
+		return nv_rd32(dev, 0x004700);
+	case 0x92:
+	case 0x94:
+	case 0x96:
+		return nv_rd32(dev, 0x004800);
+	default:
+		return 0x00000000;
+	}
+}
 
-	reg0 = nv_rd32(dev, pll.reg + 0);
-	reg1 = nv_rd32(dev, pll.reg + 4);
+static u32
+read_pll_src(struct drm_device *dev, u32 base)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	u32 coef, ref = read_clk(dev, clk_src_crystal);
+	u32 rsel = nv_rd32(dev, 0x00e18c);
+	int P, N, M, id;
 
-	if ((reg0 & 0x80000000) == 0) {
-		if (id == PLL_SHADER) {
-			NV_DEBUG(dev, "Shader PLL is disabled. "
-				"Shader clock is twice the core\n");
-			ret = nv50_pm_clock_get(dev, PLL_CORE);
-			if (ret > 0)
-				return ret << 1;
-		} else if (id == PLL_MEMORY) {
-			NV_DEBUG(dev, "Memory PLL is disabled. "
-				"Memory clock is equal to the ref_clk\n");
-			return pll.refclk;
+	switch (dev_priv->chipset) {
+	case 0x50:
+	case 0xa0:
+		switch (base) {
+		case 0x4020:
+		case 0x4028: id = !!(rsel & 0x00000004); break;
+		case 0x4008: id = !!(rsel & 0x00000008); break;
+		case 0x4030: id = 0; break;
+		default:
+			NV_ERROR(dev, "ref: bad pll 0x%06x\n", base);
+			return 0;
+		}
+
+		coef = nv_rd32(dev, 0x00e81c + (id * 0x0c));
+		ref *=  (coef & 0x01000000) ? 2 : 4;
+		P    =  (coef & 0x00070000) >> 16;
+		N    = ((coef & 0x0000ff00) >> 8) + 1;
+		M    = ((coef & 0x000000ff) >> 0) + 1;
+		break;
+	case 0x84:
+	case 0x86:
+	case 0x92:
+		coef = nv_rd32(dev, 0x00e81c);
+		P    = (coef & 0x00070000) >> 16;
+		N    = (coef & 0x0000ff00) >> 8;
+		M    = (coef & 0x000000ff) >> 0;
+		break;
+	case 0x94:
+	case 0x96:
+	case 0x98:
+		rsel = nv_rd32(dev, 0x00c050);
+		switch (base) {
+		case 0x4020: rsel = (rsel & 0x00000003) >> 0; break;
+		case 0x4008: rsel = (rsel & 0x0000000c) >> 2; break;
+		case 0x4028: rsel = (rsel & 0x00001800) >> 11; break;
+		case 0x4030: rsel = 3; break;
+		default:
+			NV_ERROR(dev, "ref: bad pll 0x%06x\n", base);
+			return 0;
+		}
+
+		switch (rsel) {
+		case 0: id = 1; break;
+		case 1: return read_clk(dev, clk_src_crystal);
+		case 2: return read_clk(dev, clk_src_href);
+		case 3: id = 0; break;
+		}
+
+		coef =  nv_rd32(dev, 0x00e81c + (id * 0x28));
+		P    = (nv_rd32(dev, 0x00e824 + (id * 0x28)) >> 16) & 7;
+		P   += (coef & 0x00070000) >> 16;
+		N    = (coef & 0x0000ff00) >> 8;
+		M    = (coef & 0x000000ff) >> 0;
+		break;
+	default:
+		BUG_ON(1);
+	}
+
+	if (M)
+		return (ref * N / M) >> P;
+	return 0;
+}
+
+static u32
+read_pll_ref(struct drm_device *dev, u32 base)
+{
+	u32 src, mast = nv_rd32(dev, 0x00c040);
+
+	switch (base) {
+	case 0x004028:
+		src = !!(mast & 0x00200000);
+		break;
+	case 0x004020:
+		src = !!(mast & 0x00400000);
+		break;
+	case 0x004008:
+		src = !!(mast & 0x00010000);
+		break;
+	case 0x004030:
+		src = !!(mast & 0x02000000);
+		break;
+	case 0x00e810:
+		return read_clk(dev, clk_src_crystal);
+	default:
+		NV_ERROR(dev, "bad pll 0x%06x\n", base);
+		return 0;
+	}
+
+	if (src)
+		return read_clk(dev, clk_src_href);
+	return read_pll_src(dev, base);
+}
+
+static u32
+read_pll(struct drm_device *dev, u32 base)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	u32 mast = nv_rd32(dev, 0x00c040);
+	u32 ctrl = nv_rd32(dev, base + 0);
+	u32 coef = nv_rd32(dev, base + 4);
+	u32 ref = read_pll_ref(dev, base);
+	u32 clk = 0;
+	int N1, N2, M1, M2;
+
+	if (base == 0x004028 && (mast & 0x00100000)) {
+		/* wtf, appears to only disable post-divider on nva0 */
+		if (dev_priv->chipset != 0xa0)
+			return read_clk(dev, clk_src_dom6);
+	}
+
+	N2 = (coef & 0xff000000) >> 24;
+	M2 = (coef & 0x00ff0000) >> 16;
+	N1 = (coef & 0x0000ff00) >> 8;
+	M1 = (coef & 0x000000ff);
+	if ((ctrl & 0x80000000) && M1) {
+		clk = ref * N1 / M1;
+		if ((ctrl & 0x40000100) == 0x40000000) {
+			if (M2)
+				clk = clk * N2 / M2;
+			else
+				clk = 0;
 		}
 	}
 
-	P = (reg0 & 0x00070000) >> 16;
-	N = (reg1 & 0x0000ff00) >> 8;
-	M = (reg1 & 0x000000ff);
+	return clk;
+}
 
-	return ((pll.refclk * N / M) >> P);
+static u32
+read_clk(struct drm_device *dev, enum clk_src src)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	u32 mast = nv_rd32(dev, 0x00c040);
+	u32 P = 0;
+
+	switch (src) {
+	case clk_src_crystal:
+		return dev_priv->crystal;
+	case clk_src_href:
+		return 100000; /* PCIE reference clock */
+	case clk_src_hclk:
+		return read_clk(dev, clk_src_href) * 27778 / 10000;
+	case clk_src_hclkm3:
+		return read_clk(dev, clk_src_hclk) * 3;
+	case clk_src_hclkm3d2:
+		return read_clk(dev, clk_src_hclk) * 3 / 2;
+	case clk_src_host:
+		switch (mast & 0x30000000) {
+		case 0x00000000: return read_clk(dev, clk_src_href);
+		case 0x10000000: break;
+		case 0x20000000: /* !0x50 */
+		case 0x30000000: return read_clk(dev, clk_src_hclk);
+		}
+		break;
+	case clk_src_nvclk:
+		if (!(mast & 0x00100000))
+			P = (nv_rd32(dev, 0x004028) & 0x00070000) >> 16;
+		switch (mast & 0x00000003) {
+		case 0x00000000: return read_clk(dev, clk_src_crystal) >> P;
+		case 0x00000001: return read_clk(dev, clk_src_dom6);
+		case 0x00000002: return read_pll(dev, 0x004020) >> P;
+		case 0x00000003: return read_pll(dev, 0x004028) >> P;
+		}
+		break;
+	case clk_src_sclk:
+		P = (nv_rd32(dev, 0x004020) & 0x00070000) >> 16;
+		switch (mast & 0x00000030) {
+		case 0x00000000:
+			if (mast & 0x00000080)
+				return read_clk(dev, clk_src_host) >> P;
+			return read_clk(dev, clk_src_crystal) >> P;
+		case 0x00000010: break;
+		case 0x00000020: return read_pll(dev, 0x004028) >> P;
+		case 0x00000030: return read_pll(dev, 0x004020) >> P;
+		}
+		break;
+	case clk_src_mclk:
+		P = (nv_rd32(dev, 0x004008) & 0x00070000) >> 16;
+		if (nv_rd32(dev, 0x004008) & 0x00000200) {
+			switch (mast & 0x0000c000) {
+			case 0x00000000:
+				return read_clk(dev, clk_src_crystal) >> P;
+			case 0x00008000:
+			case 0x0000c000:
+				return read_clk(dev, clk_src_href) >> P;
+			}
+		} else {
+			return read_pll(dev, 0x004008) >> P;
+		}
+		break;
+	case clk_src_vdec:
+		P = (read_div(dev) & 0x00000700) >> 8;
+		switch (dev_priv->chipset) {
+		case 0x84:
+		case 0x86:
+		case 0x92:
+		case 0x94:
+		case 0x96:
+		case 0xa0:
+			switch (mast & 0x00000c00) {
+			case 0x00000000:
+				if (dev_priv->chipset == 0xa0) /* wtf?? */
+					return read_clk(dev, clk_src_nvclk) >> P;
+				return read_clk(dev, clk_src_crystal) >> P;
+			case 0x00000400:
+				return 0;
+			case 0x00000800:
+				if (mast & 0x01000000)
+					return read_pll(dev, 0x004028) >> P;
+				return read_pll(dev, 0x004030) >> P;
+			case 0x00000c00:
+				return read_clk(dev, clk_src_nvclk) >> P;
+			}
+			break;
+		case 0x98:
+			switch (mast & 0x00000c00) {
+			case 0x00000000:
+				return read_clk(dev, clk_src_nvclk) >> P;
+			case 0x00000400:
+				return 0;
+			case 0x00000800:
+				return read_clk(dev, clk_src_hclkm3d2) >> P;
+			case 0x00000c00:
+				return read_clk(dev, clk_src_mclk) >> P;
+			}
+			break;
+		}
+		break;
+	case clk_src_dom6:
+		switch (dev_priv->chipset) {
+		case 0x50:
+		case 0xa0:
+			return read_pll(dev, 0x00e810) >> 2;
+		case 0x84:
+		case 0x86:
+		case 0x92:
+		case 0x94:
+		case 0x96:
+		case 0x98:
+			P = (read_div(dev) & 0x00000007) >> 0;
+			switch (mast & 0x0c000000) {
+			case 0x00000000: return read_clk(dev, clk_src_href);
+			case 0x04000000: break;
+			case 0x08000000: return read_clk(dev, clk_src_hclk);
+			case 0x0c000000:
+				return read_clk(dev, clk_src_hclkm3) >> P;
+			}
+			break;
+		default:
+			break;
+		}
+	default:
+		break;
+	}
+
+	NV_DEBUG(dev, "unknown clock source %d 0x%08x\n", src, mast);
+	return 0;
+}
+
+int
+nv50_pm_clocks_get(struct drm_device *dev, struct nouveau_pm_level *perflvl)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	if (dev_priv->chipset == 0xaa ||
+	    dev_priv->chipset == 0xac)
+		return 0;
+
+	perflvl->core   = read_clk(dev, clk_src_nvclk);
+	perflvl->shader = read_clk(dev, clk_src_sclk);
+	perflvl->memory = read_clk(dev, clk_src_mclk);
+	if (dev_priv->chipset != 0x50) {
+		perflvl->vdec = read_clk(dev, clk_src_vdec);
+		perflvl->dom6 = read_clk(dev, clk_src_dom6);
+	}
+
+	return 0;
+}
+
+struct nv50_pm_state {
+	struct hwsq_ucode mclk_hwsq;
+	u32 mscript;
+
+	u32 emast;
+	u32 nctrl;
+	u32 ncoef;
+	u32 sctrl;
+	u32 scoef;
+
+	u32 amast;
+	u32 pdivs;
+};
+
+static u32
+calc_pll(struct drm_device *dev, u32 reg, struct pll_lims *pll,
+	 u32 clk, int *N1, int *M1, int *log2P)
+{
+	struct nouveau_pll_vals coef;
+	int ret;
+
+	ret = get_pll_limits(dev, reg, pll);
+	if (ret)
+		return 0;
+
+	pll->vco2.maxfreq = 0;
+	pll->refclk = read_pll_ref(dev, reg);
+	if (!pll->refclk)
+		return 0;
+
+	ret = nouveau_calc_pll_mnp(dev, pll, clk, &coef);
+	if (ret == 0)
+		return 0;
+
+	*N1 = coef.N1;
+	*M1 = coef.M1;
+	*log2P = coef.log2P;
+	return ret;
+}
+
+static inline u32
+calc_div(u32 src, u32 target, int *div)
+{
+	u32 clk0 = src, clk1 = src;
+	for (*div = 0; *div <= 7; (*div)++) {
+		if (clk0 <= target) {
+			clk1 = clk0 << (*div ? 1 : 0);
+			break;
+		}
+		clk0 >>= 1;
+	}
+
+	if (target - clk0 <= clk1 - target)
+		return clk0;
+	(*div)--;
+	return clk1;
+}
+
+static inline u32
+clk_same(u32 a, u32 b)
+{
+	return ((a / 1000) == (b / 1000));
+}
+
+static int
+calc_mclk(struct drm_device *dev, u32 freq, struct hwsq_ucode *hwsq)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct pll_lims pll;
+	u32 mast = nv_rd32(dev, 0x00c040);
+	u32 ctrl = nv_rd32(dev, 0x004008);
+	u32 coef = nv_rd32(dev, 0x00400c);
+	u32 orig = ctrl;
+	u32 crtc_mask = 0;
+	int N, M, P;
+	int ret, i;
+
+	/* use pcie refclock if possible, otherwise use mpll */
+	ctrl &= ~0x81ff0200;
+	if (clk_same(freq, read_clk(dev, clk_src_href))) {
+		ctrl |= 0x00000200 | (pll.log2p_bias << 19);
+	} else {
+		ret = calc_pll(dev, 0x4008, &pll, freq, &N, &M, &P);
+		if (ret == 0)
+			return -EINVAL;
+
+		ctrl |= 0x80000000 | (P << 22) | (P << 16);
+		ctrl |= pll.log2p_bias << 19;
+		coef  = (N << 8) | M;
+	}
+
+	mast &= ~0xc0000000; /* get MCLK_2 from HREF */
+	mast |=  0x0000c000; /* use MCLK_2 as MPLL_BYPASS clock */
+
+	/* determine active crtcs */
+	for (i = 0; i < 2; i++) {
+		if (nv_rd32(dev, NV50_PDISPLAY_CRTC_C(i, CLOCK)))
+			crtc_mask |= (1 << i);
+	}
+
+	/* build the ucode which will reclock the memory for us */
+	hwsq_init(hwsq);
+	if (crtc_mask) {
+		hwsq_op5f(hwsq, crtc_mask, 0x00); /* wait for scanout */
+		hwsq_op5f(hwsq, crtc_mask, 0x01); /* wait for vblank */
+	}
+	if (dev_priv->chipset >= 0x92)
+		hwsq_wr32(hwsq, 0x611200, 0x00003300); /* disable scanout */
+	hwsq_setf(hwsq, 0x10, 0); /* disable bus access */
+	hwsq_op5f(hwsq, 0x00, 0x01); /* no idea :s */
+
+	/* prepare memory controller */
+	hwsq_wr32(hwsq, 0x1002d4, 0x00000001); /* precharge banks and idle */
+	hwsq_wr32(hwsq, 0x1002d0, 0x00000001); /* force refresh */
+	hwsq_wr32(hwsq, 0x100210, 0x00000000); /* stop the automatic refresh */
+	hwsq_wr32(hwsq, 0x1002dc, 0x00000001); /* start self refresh mode */
+
+	/* reclock memory */
+	hwsq_wr32(hwsq, 0xc040, mast);
+	hwsq_wr32(hwsq, 0x4008, orig | 0x00000200); /* bypass MPLL */
+	hwsq_wr32(hwsq, 0x400c, coef);
+	hwsq_wr32(hwsq, 0x4008, ctrl);
+
+	/* restart memory controller */
+	hwsq_wr32(hwsq, 0x1002d4, 0x00000001); /* precharge banks and idle */
+	hwsq_wr32(hwsq, 0x1002dc, 0x00000000); /* stop self refresh mode */
+	hwsq_wr32(hwsq, 0x100210, 0x80000000); /* restart automatic refresh */
+	hwsq_usec(hwsq, 12); /* wait for the PLL to stabilize */
+
+	hwsq_usec(hwsq, 48); /* may be unnecessary: causes flickering */
+	hwsq_setf(hwsq, 0x10, 1); /* enable bus access */
+	hwsq_op5f(hwsq, 0x00, 0x00); /* no idea, reverse of 0x00, 0x01? */
+	if (dev_priv->chipset >= 0x92)
+		hwsq_wr32(hwsq, 0x611200, 0x00003330); /* enable scanout */
+	hwsq_fini(hwsq);
+	return 0;
 }
 
 void *
-nv50_pm_clock_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl,
-		  u32 id, int khz)
+nv50_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
 {
-	struct nv50_pm_state *state;
-	int dummy, ret;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nv50_pm_state *info;
+	struct pll_lims pll;
+	int ret = -EINVAL;
+	int N, M, P1, P2;
+	u32 clk, out;
 
-	state = kzalloc(sizeof(*state), GFP_KERNEL);
-	if (!state)
+	if (dev_priv->chipset == 0xaa ||
+	    dev_priv->chipset == 0xac)
+		return ERR_PTR(-ENODEV);
+
+	info = kmalloc(sizeof(*info), GFP_KERNEL);
+	if (!info)
 		return ERR_PTR(-ENOMEM);
-	state->type = id;
-	state->perflvl = perflvl;
 
-	ret = get_pll_limits(dev, id, &state->pll);
-	if (ret < 0) {
-		kfree(state);
-		return (ret == -ENOENT) ? NULL : ERR_PTR(ret);
+	/* core: for the moment at least, always use nvpll */
+	clk = calc_pll(dev, 0x4028, &pll, perflvl->core, &N, &M, &P1);
+	if (clk == 0)
+		goto error;
+
+	info->emast = 0x00000003;
+	info->nctrl = 0x80000000 | (P1 << 19) | (P1 << 16);
+	info->ncoef = (N << 8) | M;
+
+	/* shader: tie to nvclk if possible, otherwise use spll.  have to be
+	 * very careful that the shader clock is at least twice the core, or
+	 * some chipsets will be very unhappy.  i expect most or all of these
+	 * cases will be handled by tying to nvclk, but it's possible there's
+	 * corners
+	 */
+	if (P1-- && perflvl->shader == (perflvl->core << 1)) {
+		info->emast |= 0x00000020;
+		info->sctrl  = 0x00000000 | (P1 << 19) | (P1 << 16);
+		info->scoef  = nv_rd32(dev, 0x004024);
+	} else {
+		clk = calc_pll(dev, 0x4020, &pll, perflvl->shader, &N, &M, &P1);
+		if (clk == 0)
+			goto error;
+
+		info->emast |= 0x00000030;
+		info->sctrl  = 0x80000000 | (P1 << 19) | (P1 << 16);
+		info->scoef  = (N << 8) | M;
 	}
 
-	ret = nv50_calc_pll(dev, &state->pll, khz, &state->N, &state->M,
-			    &dummy, &dummy, &state->P);
-	if (ret < 0) {
-		kfree(state);
-		return ERR_PTR(ret);
+	/* memory: build hwsq ucode which we'll use to reclock memory */
+	info->mclk_hwsq.len = 0;
+	if (perflvl->memory) {
+		clk = calc_mclk(dev, perflvl->memory, &info->mclk_hwsq);
+		if (clk < 0) {
+			ret = clk;
+			goto error;
+		}
+
+		info->mscript = perflvl->memscript;
 	}
 
-	return state;
+	/* vdec: avoid modifying xpll until we know exactly how the other
+	 * clock domains work, i suspect at least some of them can also be
+	 * tied to xpll...
+	 */
+	info->amast = nv_rd32(dev, 0x00c040);
+	info->pdivs = read_div(dev);
+	if (perflvl->vdec) {
+		/* see how close we can get using nvclk as a source */
+		clk = calc_div(perflvl->core, perflvl->vdec, &P1);
+
+		/* see how close we can get using xpll/hclk as a source */
+		if (dev_priv->chipset != 0x98)
+			out = read_pll(dev, 0x004030);
+		else
+			out = read_clk(dev, clk_src_hclkm3d2);
+		out = calc_div(out, perflvl->vdec, &P2);
+
+		/* select whichever gets us closest */
+		info->amast &= ~0x00000c00;
+		info->pdivs &= ~0x00000700;
+		if (abs((int)perflvl->vdec - clk) <=
+		    abs((int)perflvl->vdec - out)) {
+			if (dev_priv->chipset != 0x98)
+				info->amast |= 0x00000c00;
+			info->pdivs |= P1 << 8;
+		} else {
+			info->amast |= 0x00000800;
+			info->pdivs |= P2 << 8;
+		}
+	}
+
+	/* dom6: nfi what this is, but we're limited to various combinations
+	 * of the host clock frequency
+	 */
+	if (perflvl->dom6) {
+		info->amast &= ~0x0c000000;
+		if (clk_same(perflvl->dom6, read_clk(dev, clk_src_href))) {
+			info->amast |= 0x00000000;
+		} else
+		if (clk_same(perflvl->dom6, read_clk(dev, clk_src_hclk))) {
+			info->amast |= 0x08000000;
+		} else {
+			clk = read_clk(dev, clk_src_hclk) * 3;
+			clk = calc_div(clk, perflvl->dom6, &P1);
+
+			info->amast |= 0x0c000000;
+			info->pdivs  = (info->pdivs & ~0x00000007) | P1;
+		}
+	}
+
+	return info;
+error:
+	kfree(info);
+	return ERR_PTR(ret);
 }
 
-void
-nv50_pm_clock_set(struct drm_device *dev, void *pre_state)
+static int
+prog_mclk(struct drm_device *dev, struct hwsq_ucode *hwsq)
 {
-	struct nv50_pm_state *state = pre_state;
-	struct nouveau_pm_level *perflvl = state->perflvl;
-	u32 reg = state->pll.reg, tmp;
-	struct bit_entry BIT_M;
-	u16 script;
-	int N = state->N;
-	int M = state->M;
-	int P = state->P;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	u32 hwsq_data, hwsq_kick;
+	int i;
 
-	if (state->type == PLL_MEMORY && perflvl->memscript &&
-	    bit_table(dev, 'M', &BIT_M) == 0 &&
-	    BIT_M.version == 1 && BIT_M.length >= 0x0b) {
-		script = ROM16(BIT_M.data[0x05]);
-		if (script)
-			nouveau_bios_run_init_table(dev, script, NULL, -1);
-		script = ROM16(BIT_M.data[0x07]);
-		if (script)
-			nouveau_bios_run_init_table(dev, script, NULL, -1);
-		script = ROM16(BIT_M.data[0x09]);
-		if (script)
-			nouveau_bios_run_init_table(dev, script, NULL, -1);
-
-		nouveau_bios_run_init_table(dev, perflvl->memscript, NULL, -1);
+	if (dev_priv->chipset < 0x90) {
+		hwsq_data = 0x001400;
+		hwsq_kick = 0x00000003;
+	} else {
+		hwsq_data = 0x080000;
+		hwsq_kick = 0x00000001;
 	}
 
-	if (state->type == PLL_MEMORY) {
-		nv_wr32(dev, 0x100210, 0);
-		nv_wr32(dev, 0x1002dc, 1);
+	/* upload hwsq ucode */
+	nv_mask(dev, 0x001098, 0x00000008, 0x00000000);
+	nv_wr32(dev, 0x001304, 0x00000000);
+	for (i = 0; i < hwsq->len / 4; i++)
+		nv_wr32(dev, hwsq_data + (i * 4), hwsq->ptr.u32[i]);
+	nv_mask(dev, 0x001098, 0x00000018, 0x00000018);
+
+	/* launch, and wait for completion */
+	nv_wr32(dev, 0x00130c, hwsq_kick);
+	if (!nv_wait(dev, 0x001308, 0x00000100, 0x00000000)) {
+		NV_ERROR(dev, "hwsq ucode exec timed out\n");
+		NV_ERROR(dev, "0x001308: 0x%08x\n", nv_rd32(dev, 0x001308));
+		for (i = 0; i < hwsq->len / 4; i++) {
+			NV_ERROR(dev, "0x%06x: 0x%08x\n", 0x1400 + (i * 4),
+				 nv_rd32(dev, 0x001400 + (i * 4)));
+		}
+
+		return -EIO;
 	}
 
-	tmp  = nv_rd32(dev, reg + 0) & 0xfff8ffff;
-	tmp |= 0x80000000 | (P << 16);
-	nv_wr32(dev, reg + 0, tmp);
-	nv_wr32(dev, reg + 4, (N << 8) | M);
-
-	if (state->type == PLL_MEMORY) {
-		nv_wr32(dev, 0x1002dc, 0);
-		nv_wr32(dev, 0x100210, 0x80000000);
-	}
-
-	kfree(state);
+	return 0;
 }
 
+int
+nv50_pm_clocks_set(struct drm_device *dev, void *data)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nv50_pm_state *info = data;
+	struct bit_entry M;
+	int ret = 0;
+
+	/* halt and idle execution engines */
+	nv_mask(dev, 0x002504, 0x00000001, 0x00000001);
+	if (!nv_wait(dev, 0x002504, 0x00000010, 0x00000010))
+		goto error;
+
+	/* memory: it is *very* important we change this first, the ucode
+	 * we build in pre() now has hardcoded 0xc040 values, which can't
+	 * change before we execute it or the engine clocks may end up
+	 * messed up.
+	 */
+	if (info->mclk_hwsq.len) {
+		/* execute some scripts that do ??? from the vbios.. */
+		if (!bit_table(dev, 'M', &M) && M.version == 1) {
+			if (M.length >= 6)
+				nouveau_bios_init_exec(dev, ROM16(M.data[5]));
+			if (M.length >= 8)
+				nouveau_bios_init_exec(dev, ROM16(M.data[7]));
+			if (M.length >= 10)
+				nouveau_bios_init_exec(dev, ROM16(M.data[9]));
+			nouveau_bios_init_exec(dev, info->mscript);
+		}
+
+		ret = prog_mclk(dev, &info->mclk_hwsq);
+		if (ret)
+			goto resume;
+	}
+
+	/* reclock vdec/dom6 */
+	nv_mask(dev, 0x00c040, 0x00000c00, 0x00000000);
+	switch (dev_priv->chipset) {
+	case 0x92:
+	case 0x94:
+	case 0x96:
+		nv_mask(dev, 0x004800, 0x00000707, info->pdivs);
+		break;
+	default:
+		nv_mask(dev, 0x004700, 0x00000707, info->pdivs);
+		break;
+	}
+	nv_mask(dev, 0x00c040, 0x0c000c00, info->amast);
+
+	/* core/shader: make sure sclk/nvclk are disconnected from their
+	 * plls (nvclk to dom6, sclk to hclk), modify the plls, and
+	 * reconnect sclk/nvclk to their new clock source
+	 */
+	if (dev_priv->chipset < 0x92)
+		nv_mask(dev, 0x00c040, 0x001000b0, 0x00100080); /* grrr! */
+	else
+		nv_mask(dev, 0x00c040, 0x000000b3, 0x00000081);
+	nv_mask(dev, 0x004020, 0xc03f0100, info->sctrl);
+	nv_wr32(dev, 0x004024, info->scoef);
+	nv_mask(dev, 0x004028, 0xc03f0100, info->nctrl);
+	nv_wr32(dev, 0x00402c, info->ncoef);
+	nv_mask(dev, 0x00c040, 0x00100033, info->emast);
+
+	goto resume;
+error:
+	ret = -EBUSY;
+resume:
+	nv_mask(dev, 0x002504, 0x00000001, 0x00000000);
+	kfree(info);
+	return ret;
+}
+
+static int
+pwm_info(struct drm_device *dev, int *line, int *ctrl, int *indx)
+{
+	if (*line == 0x04) {
+		*ctrl = 0x00e100;
+		*line = 4;
+		*indx = 0;
+	} else
+	if (*line == 0x09) {
+		*ctrl = 0x00e100;
+		*line = 9;
+		*indx = 1;
+	} else
+	if (*line == 0x10) {
+		*ctrl = 0x00e28c;
+		*line = 0;
+		*indx = 0;
+	} else {
+		NV_ERROR(dev, "unknown pwm ctrl for gpio %d\n", *line);
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+int
+nv50_pm_pwm_get(struct drm_device *dev, int line, u32 *divs, u32 *duty)
+{
+	int ctrl, id, ret = pwm_info(dev, &line, &ctrl, &id);
+	if (ret)
+		return ret;
+
+	if (nv_rd32(dev, ctrl) & (1 << line)) {
+		*divs = nv_rd32(dev, 0x00e114 + (id * 8));
+		*duty = nv_rd32(dev, 0x00e118 + (id * 8));
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+int
+nv50_pm_pwm_set(struct drm_device *dev, int line, u32 divs, u32 duty)
+{
+	int ctrl, id, ret = pwm_info(dev, &line, &ctrl, &id);
+	if (ret)
+		return ret;
+
+	nv_mask(dev, ctrl, 0x00010001 << line, 0x00000001 << line);
+	nv_wr32(dev, 0x00e114 + (id * 8), divs);
+	nv_wr32(dev, 0x00e118 + (id * 8), duty | 0x80000000);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nv50_sor.c b/drivers/gpu/drm/nouveau/nv50_sor.c
index 2633aa8..c4423ba 100644
--- a/drivers/gpu/drm/nouveau/nv50_sor.c
+++ b/drivers/gpu/drm/nouveau/nv50_sor.c
@@ -60,6 +60,8 @@
 	BEGIN_RING(evo, 0, NV50_EVO_UPDATE, 1);
 	OUT_RING  (evo, 0);
 
+	nouveau_hdmi_mode_set(encoder, NULL);
+
 	nv_encoder->crtc = NULL;
 	nv_encoder->last_dpms = DRM_MODE_DPMS_OFF;
 }
@@ -172,6 +174,12 @@
 static void
 nv50_sor_prepare(struct drm_encoder *encoder)
 {
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	nv50_sor_disconnect(encoder);
+	if (nv_encoder->dcb->type == OUTPUT_DP) {
+		/* avoid race between link training and supervisor intr */
+		nv50_display_sync(encoder->dev);
+	}
 }
 
 static void
@@ -180,8 +188,8 @@
 }
 
 static void
-nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
-		  struct drm_display_mode *adjusted_mode)
+nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
+		  struct drm_display_mode *mode)
 {
 	struct nouveau_channel *evo = nv50_display(encoder->dev)->master;
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
@@ -193,24 +201,27 @@
 
 	NV_DEBUG_KMS(dev, "or %d type %d -> crtc %d\n",
 		     nv_encoder->or, nv_encoder->dcb->type, crtc->index);
+	nv_encoder->crtc = encoder->crtc;
 
 	switch (nv_encoder->dcb->type) {
 	case OUTPUT_TMDS:
 		if (nv_encoder->dcb->sorconf.link & 1) {
-			if (adjusted_mode->clock < 165000)
+			if (mode->clock < 165000)
 				mode_ctl = 0x0100;
 			else
 				mode_ctl = 0x0500;
 		} else
 			mode_ctl = 0x0200;
+
+		nouveau_hdmi_mode_set(encoder, mode);
 		break;
 	case OUTPUT_DP:
 		nv_connector = nouveau_encoder_connector_get(nv_encoder);
 		if (nv_connector && nv_connector->base.display_info.bpc == 6) {
-			nv_encoder->dp.datarate = crtc->mode->clock * 18 / 8;
+			nv_encoder->dp.datarate = mode->clock * 18 / 8;
 			mode_ctl |= 0x00020000;
 		} else {
-			nv_encoder->dp.datarate = crtc->mode->clock * 24 / 8;
+			nv_encoder->dp.datarate = mode->clock * 24 / 8;
 			mode_ctl |= 0x00050000;
 		}
 
@@ -228,10 +239,10 @@
 	else
 		mode_ctl |= NV50_EVO_SOR_MODE_CTRL_CRTC0;
 
-	if (adjusted_mode->flags & DRM_MODE_FLAG_NHSYNC)
+	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
 		mode_ctl |= NV50_EVO_SOR_MODE_CTRL_NHSYNC;
 
-	if (adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC)
+	if (mode->flags & DRM_MODE_FLAG_NVSYNC)
 		mode_ctl |= NV50_EVO_SOR_MODE_CTRL_NVSYNC;
 
 	nv50_sor_dpms(encoder, DRM_MODE_DPMS_ON);
@@ -239,12 +250,11 @@
 	ret = RING_SPACE(evo, 2);
 	if (ret) {
 		NV_ERROR(dev, "no space while connecting SOR\n");
+		nv_encoder->crtc = NULL;
 		return;
 	}
 	BEGIN_RING(evo, 0, NV50_EVO_SOR(nv_encoder->or, MODE_CTRL), 1);
 	OUT_RING(evo, mode_ctl);
-
-	nv_encoder->crtc = encoder->crtc;
 }
 
 static struct drm_crtc *
diff --git a/drivers/gpu/drm/nouveau/nv50_vm.c b/drivers/gpu/drm/nouveau/nv50_vm.c
index 40b84f2..6f38cea 100644
--- a/drivers/gpu/drm/nouveau/nv50_vm.c
+++ b/drivers/gpu/drm/nouveau/nv50_vm.c
@@ -48,7 +48,7 @@
 			phys |= 0x60;
 		else if (coverage <= 64 * 1024 * 1024)
 			phys |= 0x40;
-		else if (coverage < 128 * 1024 * 1024)
+		else if (coverage <= 128 * 1024 * 1024)
 			phys |= 0x20;
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nv50_vram.c b/drivers/gpu/drm/nouveau/nv50_vram.c
index 9da2383..2e45e57 100644
--- a/drivers/gpu/drm/nouveau/nv50_vram.c
+++ b/drivers/gpu/drm/nouveau/nv50_vram.c
@@ -160,7 +160,7 @@
 	colbits  =  (r4 & 0x0000f000) >> 12;
 	rowbitsa = ((r4 & 0x000f0000) >> 16) + 8;
 	rowbitsb = ((r4 & 0x00f00000) >> 20) + 8;
-	banks    = ((r4 & 0x01000000) ? 8 : 4);
+	banks    = 1 << (((r4 & 0x03000000) >> 24) + 2);
 
 	rowsize = parts * banks * (1 << colbits) * 8;
 	predicted = rowsize << rowbitsa;
diff --git a/drivers/gpu/drm/nouveau/nv84_bsp.c b/drivers/gpu/drm/nouveau/nv84_bsp.c
new file mode 100644
index 0000000..7487573
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nv84_bsp.c
@@ -0,0 +1,83 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "drmP.h"
+#include "nouveau_drv.h"
+#include "nouveau_util.h"
+#include "nouveau_vm.h"
+#include "nouveau_ramht.h"
+
+/*XXX: This stub is currently used on NV98+ also, as soon as this becomes
+ *     more than just an enable/disable stub this needs to be split out to
+ *     nv98_bsp.c...
+ */
+
+struct nv84_bsp_engine {
+	struct nouveau_exec_engine base;
+};
+
+static int
+nv84_bsp_fini(struct drm_device *dev, int engine, bool suspend)
+{
+	if (!(nv_rd32(dev, 0x000200) & 0x00008000))
+		return 0;
+
+	nv_mask(dev, 0x000200, 0x00008000, 0x00000000);
+	return 0;
+}
+
+static int
+nv84_bsp_init(struct drm_device *dev, int engine)
+{
+	nv_mask(dev, 0x000200, 0x00008000, 0x00000000);
+	nv_mask(dev, 0x000200, 0x00008000, 0x00008000);
+	return 0;
+}
+
+static void
+nv84_bsp_destroy(struct drm_device *dev, int engine)
+{
+	struct nv84_bsp_engine *pbsp = nv_engine(dev, engine);
+
+	NVOBJ_ENGINE_DEL(dev, BSP);
+
+	kfree(pbsp);
+}
+
+int
+nv84_bsp_create(struct drm_device *dev)
+{
+	struct nv84_bsp_engine *pbsp;
+
+	pbsp = kzalloc(sizeof(*pbsp), GFP_KERNEL);
+	if (!pbsp)
+		return -ENOMEM;
+
+	pbsp->base.destroy = nv84_bsp_destroy;
+	pbsp->base.init = nv84_bsp_init;
+	pbsp->base.fini = nv84_bsp_fini;
+
+	NVOBJ_ENGINE_ADD(dev, BSP, &pbsp->base);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nv84_vp.c b/drivers/gpu/drm/nouveau/nv84_vp.c
new file mode 100644
index 0000000..6570d30
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nv84_vp.c
@@ -0,0 +1,83 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "drmP.h"
+#include "nouveau_drv.h"
+#include "nouveau_util.h"
+#include "nouveau_vm.h"
+#include "nouveau_ramht.h"
+
+/*XXX: This stub is currently used on NV98+ also, as soon as this becomes
+ *     more than just an enable/disable stub this needs to be split out to
+ *     nv98_vp.c...
+ */
+
+struct nv84_vp_engine {
+	struct nouveau_exec_engine base;
+};
+
+static int
+nv84_vp_fini(struct drm_device *dev, int engine, bool suspend)
+{
+	if (!(nv_rd32(dev, 0x000200) & 0x00020000))
+		return 0;
+
+	nv_mask(dev, 0x000200, 0x00020000, 0x00000000);
+	return 0;
+}
+
+static int
+nv84_vp_init(struct drm_device *dev, int engine)
+{
+	nv_mask(dev, 0x000200, 0x00020000, 0x00000000);
+	nv_mask(dev, 0x000200, 0x00020000, 0x00020000);
+	return 0;
+}
+
+static void
+nv84_vp_destroy(struct drm_device *dev, int engine)
+{
+	struct nv84_vp_engine *pvp = nv_engine(dev, engine);
+
+	NVOBJ_ENGINE_DEL(dev, VP);
+
+	kfree(pvp);
+}
+
+int
+nv84_vp_create(struct drm_device *dev)
+{
+	struct nv84_vp_engine *pvp;
+
+	pvp = kzalloc(sizeof(*pvp), GFP_KERNEL);
+	if (!pvp)
+		return -ENOMEM;
+
+	pvp->base.destroy = nv84_vp_destroy;
+	pvp->base.init = nv84_vp_init;
+	pvp->base.fini = nv84_vp_fini;
+
+	NVOBJ_ENGINE_ADD(dev, VP, &pvp->base);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nv98_crypt.c b/drivers/gpu/drm/nouveau/nv98_crypt.c
new file mode 100644
index 0000000..db94ff0
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nv98_crypt.c
@@ -0,0 +1,78 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "drmP.h"
+#include "nouveau_drv.h"
+#include "nouveau_util.h"
+#include "nouveau_vm.h"
+#include "nouveau_ramht.h"
+
+struct nv98_crypt_engine {
+	struct nouveau_exec_engine base;
+};
+
+static int
+nv98_crypt_fini(struct drm_device *dev, int engine, bool suspend)
+{
+	if (!(nv_rd32(dev, 0x000200) & 0x00004000))
+		return 0;
+
+	nv_mask(dev, 0x000200, 0x00004000, 0x00000000);
+	return 0;
+}
+
+static int
+nv98_crypt_init(struct drm_device *dev, int engine)
+{
+	nv_mask(dev, 0x000200, 0x00004000, 0x00000000);
+	nv_mask(dev, 0x000200, 0x00004000, 0x00004000);
+	return 0;
+}
+
+static void
+nv98_crypt_destroy(struct drm_device *dev, int engine)
+{
+	struct nv98_crypt_engine *pcrypt = nv_engine(dev, engine);
+
+	NVOBJ_ENGINE_DEL(dev, CRYPT);
+
+	kfree(pcrypt);
+}
+
+int
+nv98_crypt_create(struct drm_device *dev)
+{
+	struct nv98_crypt_engine *pcrypt;
+
+	pcrypt = kzalloc(sizeof(*pcrypt), GFP_KERNEL);
+	if (!pcrypt)
+		return -ENOMEM;
+
+	pcrypt->base.destroy = nv98_crypt_destroy;
+	pcrypt->base.init = nv98_crypt_init;
+	pcrypt->base.fini = nv98_crypt_fini;
+
+	NVOBJ_ENGINE_ADD(dev, CRYPT, &pcrypt->base);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nv98_ppp.c b/drivers/gpu/drm/nouveau/nv98_ppp.c
new file mode 100644
index 0000000..a987dd6
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nv98_ppp.c
@@ -0,0 +1,78 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "drmP.h"
+#include "nouveau_drv.h"
+#include "nouveau_util.h"
+#include "nouveau_vm.h"
+#include "nouveau_ramht.h"
+
+struct nv98_ppp_engine {
+	struct nouveau_exec_engine base;
+};
+
+static int
+nv98_ppp_fini(struct drm_device *dev, int engine, bool suspend)
+{
+	if (!(nv_rd32(dev, 0x000200) & 0x00000002))
+		return 0;
+
+	nv_mask(dev, 0x000200, 0x00000002, 0x00000000);
+	return 0;
+}
+
+static int
+nv98_ppp_init(struct drm_device *dev, int engine)
+{
+	nv_mask(dev, 0x000200, 0x00000002, 0x00000000);
+	nv_mask(dev, 0x000200, 0x00000002, 0x00000002);
+	return 0;
+}
+
+static void
+nv98_ppp_destroy(struct drm_device *dev, int engine)
+{
+	struct nv98_ppp_engine *pppp = nv_engine(dev, engine);
+
+	NVOBJ_ENGINE_DEL(dev, PPP);
+
+	kfree(pppp);
+}
+
+int
+nv98_ppp_create(struct drm_device *dev)
+{
+	struct nv98_ppp_engine *pppp;
+
+	pppp = kzalloc(sizeof(*pppp), GFP_KERNEL);
+	if (!pppp)
+		return -ENOMEM;
+
+	pppp->base.destroy = nv98_ppp_destroy;
+	pppp->base.init = nv98_ppp_init;
+	pppp->base.fini = nv98_ppp_fini;
+
+	NVOBJ_ENGINE_ADD(dev, PPP, &pppp->base);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nva3_copy.fuc b/drivers/gpu/drm/nouveau/nva3_copy.fuc
index eaf35f8..abc3662 100644
--- a/drivers/gpu/drm/nouveau/nva3_copy.fuc
+++ b/drivers/gpu/drm/nouveau/nva3_copy.fuc
@@ -31,8 +31,9 @@
  */
 
 ifdef(`NVA3',
-.section nva3_pcopy_data,
-.section nvc0_pcopy_data
+.section #nva3_pcopy_data
+,
+.section #nvc0_pcopy_data
 )
 
 ctx_object:                   .b32 0
@@ -42,7 +43,7 @@
 ctx_dma_src:                  .b32 0
 ctx_dma_dst:                  .b32 0
 ,)
-.equ ctx_dma_count 3
+.equ #ctx_dma_count 3
 ctx_query_address_high:       .b32 0
 ctx_query_address_low:        .b32 0
 ctx_query_counter:            .b32 0
@@ -78,64 +79,65 @@
 dispatch_table:
 // mthd 0x0000, NAME
 .b16 0x000 1
-.b32 ctx_object                     ~0xffffffff
+.b32 #ctx_object                     ~0xffffffff
 // mthd 0x0100, NOP
 .b16 0x040 1
-.b32 0x00010000 + cmd_nop           ~0xffffffff
+.b32 0x00010000 + #cmd_nop           ~0xffffffff
 // mthd 0x0140, PM_TRIGGER
 .b16 0x050 1
-.b32 0x00010000 + cmd_pm_trigger    ~0xffffffff
+.b32 0x00010000 + #cmd_pm_trigger    ~0xffffffff
 ifdef(`NVA3', `
 // mthd 0x0180-0x018c, DMA_
-.b16 0x060 ctx_dma_count
+.b16 0x060 #ctx_dma_count
 dispatch_dma:
-.b32 0x00010000 + cmd_dma           ~0xffffffff
-.b32 0x00010000 + cmd_dma           ~0xffffffff
-.b32 0x00010000 + cmd_dma           ~0xffffffff
+.b32 0x00010000 + #cmd_dma           ~0xffffffff
+.b32 0x00010000 + #cmd_dma           ~0xffffffff
+.b32 0x00010000 + #cmd_dma           ~0xffffffff
 ',)
 // mthd 0x0200-0x0218, SRC_TILE
 .b16 0x80 7
-.b32 ctx_src_tile_mode              ~0x00000fff
-.b32 ctx_src_xsize                  ~0x0007ffff
-.b32 ctx_src_ysize                  ~0x00001fff
-.b32 ctx_src_zsize                  ~0x000007ff
-.b32 ctx_src_zoff                   ~0x00000fff
-.b32 ctx_src_xoff                   ~0x0007ffff
-.b32 ctx_src_yoff                   ~0x00001fff
+.b32 #ctx_src_tile_mode              ~0x00000fff
+.b32 #ctx_src_xsize                  ~0x0007ffff
+.b32 #ctx_src_ysize                  ~0x00001fff
+.b32 #ctx_src_zsize                  ~0x000007ff
+.b32 #ctx_src_zoff                   ~0x00000fff
+.b32 #ctx_src_xoff                   ~0x0007ffff
+.b32 #ctx_src_yoff                   ~0x00001fff
 // mthd 0x0220-0x0238, DST_TILE
 .b16 0x88 7
-.b32 ctx_dst_tile_mode              ~0x00000fff
-.b32 ctx_dst_xsize                  ~0x0007ffff
-.b32 ctx_dst_ysize                  ~0x00001fff
-.b32 ctx_dst_zsize                  ~0x000007ff
-.b32 ctx_dst_zoff                   ~0x00000fff
-.b32 ctx_dst_xoff                   ~0x0007ffff
-.b32 ctx_dst_yoff                   ~0x00001fff
+.b32 #ctx_dst_tile_mode              ~0x00000fff
+.b32 #ctx_dst_xsize                  ~0x0007ffff
+.b32 #ctx_dst_ysize                  ~0x00001fff
+.b32 #ctx_dst_zsize                  ~0x000007ff
+.b32 #ctx_dst_zoff                   ~0x00000fff
+.b32 #ctx_dst_xoff                   ~0x0007ffff
+.b32 #ctx_dst_yoff                   ~0x00001fff
 // mthd 0x0300-0x0304, EXEC, WRCACHE_FLUSH
 .b16 0xc0 2
-.b32 0x00010000 + cmd_exec          ~0xffffffff
-.b32 0x00010000 + cmd_wrcache_flush ~0xffffffff
+.b32 0x00010000 + #cmd_exec          ~0xffffffff
+.b32 0x00010000 + #cmd_wrcache_flush ~0xffffffff
 // mthd 0x030c-0x0340, various stuff
 .b16 0xc3 14
-.b32 ctx_src_address_high           ~0x000000ff
-.b32 ctx_src_address_low            ~0xfffffff0
-.b32 ctx_dst_address_high           ~0x000000ff
-.b32 ctx_dst_address_low            ~0xfffffff0
-.b32 ctx_src_pitch                  ~0x0007ffff
-.b32 ctx_dst_pitch                  ~0x0007ffff
-.b32 ctx_xcnt                       ~0x0000ffff
-.b32 ctx_ycnt                       ~0x00001fff
-.b32 ctx_format                     ~0x0333ffff
-.b32 ctx_swz_const0                 ~0xffffffff
-.b32 ctx_swz_const1                 ~0xffffffff
-.b32 ctx_query_address_high         ~0x000000ff
-.b32 ctx_query_address_low          ~0xffffffff
-.b32 ctx_query_counter              ~0xffffffff
+.b32 #ctx_src_address_high           ~0x000000ff
+.b32 #ctx_src_address_low            ~0xfffffff0
+.b32 #ctx_dst_address_high           ~0x000000ff
+.b32 #ctx_dst_address_low            ~0xfffffff0
+.b32 #ctx_src_pitch                  ~0x0007ffff
+.b32 #ctx_dst_pitch                  ~0x0007ffff
+.b32 #ctx_xcnt                       ~0x0000ffff
+.b32 #ctx_ycnt                       ~0x00001fff
+.b32 #ctx_format                     ~0x0333ffff
+.b32 #ctx_swz_const0                 ~0xffffffff
+.b32 #ctx_swz_const1                 ~0xffffffff
+.b32 #ctx_query_address_high         ~0x000000ff
+.b32 #ctx_query_address_low          ~0xffffffff
+.b32 #ctx_query_counter              ~0xffffffff
 .b16 0x800 0
 
 ifdef(`NVA3',
-.section nva3_pcopy_code,
-.section nvc0_pcopy_code
+.section #nva3_pcopy_code
+,
+.section #nvc0_pcopy_code
 )
 
 main:
@@ -143,12 +145,12 @@
    mov $sp $r0
 
    // setup i0 handler and route fifo and ctxswitch to it
-   mov $r1 ih
+   mov $r1 #ih
    mov $iv0 $r1
    mov $r1 0x400
    movw $r2 0xfff3
    sethi $r2 0
-   iowr I[$r2 + 0x300] $r2
+   iowr I[$r1 + 0x300] $r2
 
    // enable interrupts
    or $r2 0xc
@@ -164,19 +166,19 @@
    bset $flags $p0
    spin:
       sleep $p0
-      bra spin
+      bra #spin
 
 // i0 handler
 ih:
    iord $r1 I[$r0 + 0x200]
 
    and $r2 $r1 0x00000008
-   bra e ih_no_chsw
-      call chsw
+   bra e #ih_no_chsw
+      call #chsw
    ih_no_chsw:
    and $r2 $r1 0x00000004
-   bra e ih_no_cmd
-      call dispatch
+   bra e #ih_no_cmd
+      call #dispatch
 
    ih_no_cmd:
    and $r1 $r1 0x0000000c
@@ -235,9 +237,9 @@
    sethi $r4 0x60000
 
    // swap!
-   bra $p1 swctx_load
+   bra $p1 #swctx_load
       xdst $r0 $r4
-      bra swctx_done
+      bra #swctx_done
    swctx_load:
       xdld $r0 $r4
    swctx_done:
@@ -251,9 +253,9 @@
 
    // if it's active, unload it and return
    xbit $r15 $r3 0x1e
-   bra e chsw_no_unload
+   bra e #chsw_no_unload
       bclr $flags $p1
-      call swctx
+      call #swctx
       bclr $r3 0x1e
       iowr I[$r2] $r3
       mov $r4 1
@@ -266,20 +268,20 @@
 
    // is there a channel waiting to be loaded?
    xbit $r13 $r3 0x1e
-   bra e chsw_finish_load
+   bra e #chsw_finish_load
       bset $flags $p1
-      call swctx
+      call #swctx
 ifdef(`NVA3',
       // load dma objects back into TARGET regs
-      mov $r5 ctx_dma
-      mov $r6 ctx_dma_count
+      mov $r5 #ctx_dma
+      mov $r6 #ctx_dma_count
       chsw_load_ctx_dma:
          ld b32 $r7 D[$r5 + $r6 * 4]
          add b32 $r8 $r6 0x180
          shl b32 $r8 8
          iowr I[$r8] $r7
          sub b32 $r6 1
-         bra nc chsw_load_ctx_dma
+         bra nc #chsw_load_ctx_dma
 ,)
 
    chsw_finish_load:
@@ -297,7 +299,7 @@
    shl b32 $r2 0x10
 
    // lookup method in the dispatch table, ILLEGAL_MTHD if not found
-   mov $r5 dispatch_table
+   mov $r5 #dispatch_table
    clear b32 $r6
    clear b32 $r7
    dispatch_loop:
@@ -305,14 +307,14 @@
       ld b16 $r7 D[$r5 + 2]
       add b32 $r5 4
       cmpu b32 $r4 $r6
-      bra c dispatch_illegal_mthd
+      bra c #dispatch_illegal_mthd
       add b32 $r7 $r6
       cmpu b32 $r4 $r7
-      bra c dispatch_valid_mthd
+      bra c #dispatch_valid_mthd
       sub b32 $r7 $r6
       shl b32 $r7 3
       add b32 $r5 $r7
-      bra dispatch_loop
+      bra #dispatch_loop
 
    // ensure no bits set in reserved fields, INVALID_BITFIELD
    dispatch_valid_mthd:
@@ -322,20 +324,20 @@
    ld b32 $r5 D[$r4 + 4]
    and $r5 $r3
    cmpu b32 $r5 0
-   bra ne dispatch_invalid_bitfield
+   bra ne #dispatch_invalid_bitfield
 
    // depending on dispatch flags: execute method, or save data as state
    ld b16 $r5 D[$r4 + 0]
    ld b16 $r6 D[$r4 + 2]
    cmpu b32 $r6 0
-   bra ne dispatch_cmd
+   bra ne #dispatch_cmd
       st b32 D[$r5] $r3
-      bra dispatch_done
+      bra #dispatch_done
    dispatch_cmd:
       bclr $flags $p1
       call $r5
-      bra $p1 dispatch_error
-      bra dispatch_done
+      bra $p1 #dispatch_error
+      bra #dispatch_done
 
    dispatch_invalid_bitfield:
    or $r2 2
@@ -353,7 +355,7 @@
       iord $r2 I[$r0 + 0x200]
       and $r2 0x40
       cmpu b32 $r2 0
-      bra ne hostirq_wait
+      bra ne #hostirq_wait
 
    dispatch_done:
    mov $r2 0x1d00
@@ -409,10 +411,10 @@
 //       $r2: hostirq state
 //       $r3: data
 cmd_dma:
-   sub b32 $r4 dispatch_dma
+   sub b32 $r4 #dispatch_dma
    shr b32 $r4 1
    bset $r3 0x1e
-   st b32 D[$r4 + ctx_dma] $r3
+   st b32 D[$r4 + #ctx_dma] $r3
    add b32 $r4 0x600
    shl b32 $r4 6
    iowr I[$r4] $r3
@@ -430,7 +432,7 @@
    st b32 D[$sp + 0x0c] $r0
 
    // extract cpp, src_ncomp and dst_ncomp from FORMAT
-   ld b32 $r4 D[$r0 + ctx_format]
+   ld b32 $r4 D[$r0 + #ctx_format]
    extr $r5 $r4 16:17
    add b32 $r5 1
    extr $r6 $r4 20:21
@@ -448,22 +450,22 @@
       clear b32 $r11
       bpc_loop:
          cmpu b8 $r10 4
-         bra nc cmp_c0
+         bra nc #cmp_c0
             mulu $r12 $r10 $r5
             add b32 $r12 $r11
             bset $flags $p2
-            bra bpc_next
+            bra #bpc_next
          cmp_c0:
-         bra ne cmp_c1
+         bra ne #cmp_c1
             mov $r12 0x10
             add b32 $r12 $r11
-            bra bpc_next
+            bra #bpc_next
          cmp_c1:
          cmpu b8 $r10 6
-         bra nc cmp_zero
+         bra nc #cmp_zero
             mov $r12 0x14
             add b32 $r12 $r11
-            bra bpc_next
+            bra #bpc_next
          cmp_zero:
             mov $r12 0x80
          bpc_next:
@@ -471,22 +473,22 @@
          add b32 $r8 1
          add b32 $r11 1
          cmpu b32 $r11 $r5
-         bra c bpc_loop
+         bra c #bpc_loop
       add b32 $r9 1
       cmpu b32 $r9 $r7
-      bra c ncomp_loop
+      bra c #ncomp_loop
 
    // SRC_XCNT = (xcnt * src_cpp), or 0 if no src ref in swz (hw will hang)
    mulu $r6 $r5
-   st b32 D[$r0 + ctx_src_cpp] $r6
-   ld b32 $r8 D[$r0 + ctx_xcnt]
+   st b32 D[$r0 + #ctx_src_cpp] $r6
+   ld b32 $r8 D[$r0 + #ctx_xcnt]
    mulu $r6 $r8
-   bra $p2 dst_xcnt
+   bra $p2 #dst_xcnt
    clear b32 $r6
 
    dst_xcnt:
    mulu $r7 $r5
-   st b32 D[$r0 + ctx_dst_cpp] $r7
+   st b32 D[$r0 + #ctx_dst_cpp] $r7
    mulu $r7 $r8
 
    mov $r5 0x810
@@ -494,10 +496,10 @@
    iowr I[$r5 + 0x000] $r6
    iowr I[$r5 + 0x100] $r7
    add b32 $r5 0x800
-   ld b32 $r6 D[$r0 + ctx_dst_cpp]
+   ld b32 $r6 D[$r0 + #ctx_dst_cpp]
    sub b32 $r6 1
    shl b32 $r6 8
-   ld b32 $r7 D[$r0 + ctx_src_cpp]
+   ld b32 $r7 D[$r0 + #ctx_src_cpp]
    sub b32 $r7 1
    or $r6 $r7
    iowr I[$r5 + 0x000] $r6
@@ -511,9 +513,9 @@
    ld b32 $r6 D[$sp + 0x0c]
    iowr I[$r5 + 0x300] $r6
    add b32 $r5 0x400
-   ld b32 $r6 D[$r0 + ctx_swz_const0]
+   ld b32 $r6 D[$r0 + #ctx_swz_const0]
    iowr I[$r5 + 0x000] $r6
-   ld b32 $r6 D[$r0 + ctx_swz_const1]
+   ld b32 $r6 D[$r0 + #ctx_swz_const1]
    iowr I[$r5 + 0x100] $r6
    add $sp 0x10
    ret
@@ -543,7 +545,7 @@
 //
 cmd_exec_set_surface_tiled:
    // translate TILE_MODE into Tp, Th, Td shift values
-   ld b32 $r7 D[$r5 + ctx_src_tile_mode]
+   ld b32 $r7 D[$r5 + #ctx_src_tile_mode]
    extr $r9 $r7 8:11
    extr $r8 $r7 4:7
 ifdef(`NVA3',
@@ -553,9 +555,9 @@
 )
    extr $r7 $r7 0:3
    cmp b32 $r7 0xe
-   bra ne xtile64
+   bra ne #xtile64
    mov $r7 4
-   bra xtileok
+   bra #xtileok
    xtile64:
    xbit $r7 $flags $p2
    add b32 $r7 17
@@ -565,8 +567,8 @@
 
    // Op = (x * cpp) & ((1 << Tp) - 1)
    // Tx = (x * cpp) >> Tp
-   ld b32 $r10 D[$r5 + ctx_src_xoff]
-   ld b32 $r11 D[$r5 + ctx_src_cpp]
+   ld b32 $r10 D[$r5 + #ctx_src_xoff]
+   ld b32 $r11 D[$r5 + #ctx_src_cpp]
    mulu $r10 $r11
    mov $r11 1
    shl b32 $r11 $r7
@@ -576,7 +578,7 @@
 
    // Tyo = y & ((1 << Th) - 1)
    // Ty  = y >> Th
-   ld b32 $r13 D[$r5 + ctx_src_yoff]
+   ld b32 $r13 D[$r5 + #ctx_src_yoff]
    mov $r14 1
    shl b32 $r14 $r8
    sub b32 $r14 1
@@ -598,8 +600,8 @@
    add b32 $r12 $r11
 
    // nTx = ((w * cpp) + ((1 << Tp) - 1) >> Tp)
-   ld b32 $r15 D[$r5 + ctx_src_xsize]
-   ld b32 $r11 D[$r5 + ctx_src_cpp]
+   ld b32 $r15 D[$r5 + #ctx_src_xsize]
+   ld b32 $r11 D[$r5 + #ctx_src_cpp]
    mulu $r15 $r11
    mov $r11 1
    shl b32 $r11 $r7
@@ -609,7 +611,7 @@
    push $r15
 
    // nTy = (h + ((1 << Th) - 1)) >> Th
-   ld b32 $r15 D[$r5 + ctx_src_ysize]
+   ld b32 $r15 D[$r5 + #ctx_src_ysize]
    mov $r11 1
    shl b32 $r11 $r8
    sub b32 $r11 1
@@ -629,7 +631,7 @@
    // Tz  = z >> Td
    // Op += Tzo << Tys
    // Ts  = Tys + Td
-   ld b32 $r8 D[$r5 + ctx_src_zoff]
+   ld b32 $r8 D[$r5 + #ctx_src_zoff]
    mov $r14 1
    shl b32 $r14 $r9
    sub b32 $r14 1
@@ -656,8 +658,8 @@
 
    // SRC_ADDRESS_LOW   = (Ot + Op) & 0xffffffff
    // CFG_ADDRESS_HIGH |= ((Ot + Op) >> 32) << 16
-   ld b32 $r7 D[$r5 + ctx_src_address_low]
-   ld b32 $r8 D[$r5 + ctx_src_address_high]
+   ld b32 $r7 D[$r5 + #ctx_src_address_low]
+   ld b32 $r8 D[$r5 + #ctx_src_address_high]
    add b32 $r10 $r12
    add b32 $r7 $r10
    adc b32 $r8 0
@@ -677,14 +679,14 @@
    xbit $r6 $flags $p2
    add b32 $r6 0x202
    shl b32 $r6 8
-   ld b32 $r7 D[$r5 + ctx_src_address_low]
+   ld b32 $r7 D[$r5 + #ctx_src_address_low]
    iowr I[$r6 + 0x000] $r7
    add b32 $r6 0x400
-   ld b32 $r7 D[$r5 + ctx_src_address_high]
+   ld b32 $r7 D[$r5 + #ctx_src_address_high]
    shl b32 $r7 16
    iowr I[$r6 + 0x000] $r7
    add b32 $r6 0x400
-   ld b32 $r7 D[$r5 + ctx_src_pitch]
+   ld b32 $r7 D[$r5 + #ctx_src_pitch]
    iowr I[$r6 + 0x000] $r7
    ret
 
@@ -697,7 +699,7 @@
    loop:
       iord $r1 I[$r0]
       and $r1 1
-      bra ne loop
+      bra ne #loop
    pop $r1
    pop $r0
    ret
@@ -705,18 +707,18 @@
 cmd_exec_query:
    // if QUERY_SHORT not set, write out { -, 0, TIME_LO, TIME_HI }
    xbit $r4 $r3 13
-   bra ne query_counter
-      call cmd_exec_wait
+   bra ne #query_counter
+      call #cmd_exec_wait
       mov $r4 0x80c
       shl b32 $r4 6
-      ld b32 $r5 D[$r0 + ctx_query_address_low]
+      ld b32 $r5 D[$r0 + #ctx_query_address_low]
       add b32 $r5 4
       iowr I[$r4 + 0x000] $r5
       iowr I[$r4 + 0x100] $r0
       mov $r5 0xc
       iowr I[$r4 + 0x200] $r5
       add b32 $r4 0x400
-      ld b32 $r5 D[$r0 + ctx_query_address_high]
+      ld b32 $r5 D[$r0 + #ctx_query_address_high]
       shl b32 $r5 16
       iowr I[$r4 + 0x000] $r5
       add b32 $r4 0x500
@@ -741,16 +743,16 @@
 
    // write COUNTER
    query_counter:
-   call cmd_exec_wait
+   call #cmd_exec_wait
    mov $r4 0x80c
    shl b32 $r4 6
-   ld b32 $r5 D[$r0 + ctx_query_address_low]
+   ld b32 $r5 D[$r0 + #ctx_query_address_low]
    iowr I[$r4 + 0x000] $r5
    iowr I[$r4 + 0x100] $r0
    mov $r5 0x4
    iowr I[$r4 + 0x200] $r5
    add b32 $r4 0x400
-   ld b32 $r5 D[$r0 + ctx_query_address_high]
+   ld b32 $r5 D[$r0 + #ctx_query_address_high]
    shl b32 $r5 16
    iowr I[$r4 + 0x000] $r5
    add b32 $r4 0x500
@@ -759,7 +761,7 @@
    mov $r5 0x00001110
    sethi $r5 0x13120000
    iowr I[$r4 + 0x100] $r5
-   ld b32 $r5 D[$r0 + ctx_query_counter]
+   ld b32 $r5 D[$r0 + #ctx_query_counter]
    add b32 $r4 0x500
    iowr I[$r4 + 0x000] $r5
    mov $r5 0x00002601
@@ -787,22 +789,22 @@
 //       $r2: hostirq state
 //       $r3: data
 cmd_exec:
-   call cmd_exec_wait
+   call #cmd_exec_wait
 
    // if format requested, call function to calculate it, otherwise
    // fill in cpp/xcnt for both surfaces as if (cpp == 1)
    xbit $r15 $r3 0
-   bra e cmd_exec_no_format
-      call cmd_exec_set_format
+   bra e #cmd_exec_no_format
+      call #cmd_exec_set_format
       mov $r4 0x200
-      bra cmd_exec_init_src_surface
+      bra #cmd_exec_init_src_surface
    cmd_exec_no_format:
       mov $r6 0x810
       shl b32 $r6 6
       mov $r7 1
-      st b32 D[$r0 + ctx_src_cpp] $r7
-      st b32 D[$r0 + ctx_dst_cpp] $r7
-      ld b32 $r7 D[$r0 + ctx_xcnt]
+      st b32 D[$r0 + #ctx_src_cpp] $r7
+      st b32 D[$r0 + #ctx_dst_cpp] $r7
+      ld b32 $r7 D[$r0 + #ctx_xcnt]
       iowr I[$r6 + 0x000] $r7
       iowr I[$r6 + 0x100] $r7
       clear b32 $r4
@@ -811,28 +813,28 @@
    bclr $flags $p2
    clear b32 $r5
    xbit $r15 $r3 4
-   bra e src_tiled
-      call cmd_exec_set_surface_linear
-      bra cmd_exec_init_dst_surface
+   bra e #src_tiled
+      call #cmd_exec_set_surface_linear
+      bra #cmd_exec_init_dst_surface
    src_tiled:
-      call cmd_exec_set_surface_tiled
+      call #cmd_exec_set_surface_tiled
       bset $r4 7
 
    cmd_exec_init_dst_surface:
    bset $flags $p2
-   mov $r5 ctx_dst_address_high - ctx_src_address_high
+   mov $r5 #ctx_dst_address_high - #ctx_src_address_high
    xbit $r15 $r3 8
-   bra e dst_tiled
-      call cmd_exec_set_surface_linear
-      bra cmd_exec_kick
+   bra e #dst_tiled
+      call #cmd_exec_set_surface_linear
+      bra #cmd_exec_kick
    dst_tiled:
-      call cmd_exec_set_surface_tiled
+      call #cmd_exec_set_surface_tiled
       bset $r4 8
 
    cmd_exec_kick:
    mov $r5 0x800
    shl b32 $r5 6
-   ld b32 $r6 D[$r0 + ctx_ycnt]
+   ld b32 $r6 D[$r0 + #ctx_ycnt]
    iowr I[$r5 + 0x100] $r6
    mov $r6 0x0041
    // SRC_TARGET = 1, DST_TARGET = 2
@@ -842,8 +844,8 @@
 
    // if requested, queue up a QUERY write after the copy has completed
    xbit $r15 $r3 12
-   bra e cmd_exec_done
-      call cmd_exec_query
+   bra e #cmd_exec_done
+      call #cmd_exec_query
 
    cmd_exec_done:
    ret
diff --git a/drivers/gpu/drm/nouveau/nva3_copy.fuc.h b/drivers/gpu/drm/nouveau/nva3_copy.fuc.h
index 2731de2..1f33fbd 100644
--- a/drivers/gpu/drm/nouveau/nva3_copy.fuc.h
+++ b/drivers/gpu/drm/nouveau/nva3_copy.fuc.h
@@ -152,7 +152,7 @@
 	0xf10010fe,
 	0xf1040017,
 	0xf0fff327,
-	0x22d00023,
+	0x12d00023,
 	0x0c25f0c0,
 	0xf40012d0,
 	0x17f11031,
diff --git a/drivers/gpu/drm/nouveau/nva3_pm.c b/drivers/gpu/drm/nouveau/nva3_pm.c
index 618c144..9e636e6 100644
--- a/drivers/gpu/drm/nouveau/nva3_pm.c
+++ b/drivers/gpu/drm/nouveau/nva3_pm.c
@@ -287,12 +287,13 @@
 	return false;
 }
 
-void
+int
 nva3_pm_clocks_set(struct drm_device *dev, void *pre_state)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nva3_pm_state *info = pre_state;
 	unsigned long flags;
+	int ret = -EAGAIN;
 
 	/* prevent any new grctx switches from starting */
 	spin_lock_irqsave(&dev_priv->context_switch_lock, flags);
@@ -328,6 +329,8 @@
 		nv_wr32(dev, 0x100210, 0x80000000);
 	}
 
+	ret = 0;
+
 cleanup:
 	/* unfreeze PFIFO */
 	nv_mask(dev, 0x002504, 0x00000001, 0x00000000);
@@ -339,4 +342,5 @@
 		nv_mask(dev, 0x400824, 0x10000000, 0x10000000);
 	spin_unlock_irqrestore(&dev_priv->context_switch_lock, flags);
 	kfree(info);
+	return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nvc0_copy.fuc.h b/drivers/gpu/drm/nouveau/nvc0_copy.fuc.h
index 4199038..a8d1745 100644
--- a/drivers/gpu/drm/nouveau/nvc0_copy.fuc.h
+++ b/drivers/gpu/drm/nouveau/nvc0_copy.fuc.h
@@ -145,7 +145,7 @@
 	0xf10010fe,
 	0xf1040017,
 	0xf0fff327,
-	0x22d00023,
+	0x12d00023,
 	0x0c25f0c0,
 	0xf40012d0,
 	0x17f11031,
diff --git a/drivers/gpu/drm/nouveau/nvc0_graph.c b/drivers/gpu/drm/nouveau/nvc0_graph.c
index bbdbc51..8ee3963 100644
--- a/drivers/gpu/drm/nouveau/nvc0_graph.c
+++ b/drivers/gpu/drm/nouveau/nvc0_graph.c
@@ -157,8 +157,8 @@
 	struct nvc0_graph_priv *priv = nv_engine(chan->dev, NVOBJ_ENGINE_GR);
 	struct nvc0_graph_chan *grch = chan->engctx[NVOBJ_ENGINE_GR];
 	struct drm_device *dev = chan->dev;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	int i = 0, gpc, tp, ret;
-	u32 magic;
 
 	ret = nouveau_gpuobj_new(dev, chan, 0x2000, 256, NVOBJ_FLAG_VM,
 				 &grch->unk408004);
@@ -207,14 +207,37 @@
 	nv_wo32(grch->mmio, i++ * 4, 0x0041880c);
 	nv_wo32(grch->mmio, i++ * 4, 0x80000018);
 
-	magic = 0x02180000;
-	nv_wo32(grch->mmio, i++ * 4, 0x00405830);
-	nv_wo32(grch->mmio, i++ * 4, magic);
-	for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
-		for (tp = 0; tp < priv->tp_nr[gpc]; tp++, magic += 0x0324) {
-			u32 reg = 0x504520 + (gpc * 0x8000) + (tp * 0x0800);
-			nv_wo32(grch->mmio, i++ * 4, reg);
-			nv_wo32(grch->mmio, i++ * 4, magic);
+	if (dev_priv->chipset != 0xc1) {
+		u32 magic = 0x02180000;
+		nv_wo32(grch->mmio, i++ * 4, 0x00405830);
+		nv_wo32(grch->mmio, i++ * 4, magic);
+		for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x520);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, magic);
+				magic += 0x0324;
+			}
+		}
+	} else {
+		u32 magic = 0x02180000;
+		nv_wo32(grch->mmio, i++ * 4, 0x00405830);
+		nv_wo32(grch->mmio, i++ * 4, magic | 0x0000218);
+		nv_wo32(grch->mmio, i++ * 4, 0x004064c4);
+		nv_wo32(grch->mmio, i++ * 4, 0x0086ffff);
+		for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x520);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, (1 << 28) | magic);
+				magic += 0x0324;
+			}
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x544);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, magic);
+				magic += 0x0324;
+			}
 		}
 	}
 
@@ -358,6 +381,8 @@
 	u8  tpnr[GPC_MAX];
 	int i, gpc, tpc;
 
+	nv_wr32(dev, TP_UNIT(0, 0, 0x5c), 1); /* affects TFB offset queries */
+
 	/*
 	 *      TP      ROP UNKVAL(magic_not_rop_nr)
 	 * 450: 4/0/0/0 2        3
@@ -850,14 +875,16 @@
 	case 0xcf: /* 4/0/0/0, 3 */
 		priv->magic_not_rop_nr = 0x03;
 		break;
+	case 0xd9: /* 1/0/0/0, 1 */
+		priv->magic_not_rop_nr = 0x01;
+		break;
 	}
 
 	if (!priv->magic_not_rop_nr) {
 		NV_ERROR(dev, "PGRAPH: unknown config: %d/%d/%d/%d, %d\n",
 			 priv->tp_nr[0], priv->tp_nr[1], priv->tp_nr[2],
 			 priv->tp_nr[3], priv->rop_nr);
-		/* use 0xc3's values... */
-		priv->magic_not_rop_nr = 0x03;
+		priv->magic_not_rop_nr = 0x00;
 	}
 
 	NVOBJ_CLASS(dev, 0x902d, GR); /* 2D */
diff --git a/drivers/gpu/drm/nouveau/nvc0_graph.fuc b/drivers/gpu/drm/nouveau/nvc0_graph.fuc
index 2a4b6dc..e6b2288 100644
--- a/drivers/gpu/drm/nouveau/nvc0_graph.fuc
+++ b/drivers/gpu/drm/nouveau/nvc0_graph.fuc
@@ -71,9 +71,9 @@
 	ld b32 $r9 D[$r13 + 0x4]	// PUT
 	xor $r8 8
 	cmpu b32 $r8 $r9
-	bra ne queue_put_next
+	bra ne #queue_put_next
 		mov $r15 E_CMD_OVERFLOW
-		call error
+		call #error
 		ret
 
 	// store cmd/data on queue
@@ -104,7 +104,7 @@
 	ld b32 $r8 D[$r13 + 0x0]	// GET
 	ld b32 $r9 D[$r13 + 0x4]	// PUT
 	cmpu b32 $r8 $r9
-	bra e queue_get_done
+	bra e #queue_get_done
 		// fetch first cmd/data pair
 		and $r9 $r8 7
 		shl b32 $r9 3
@@ -135,9 +135,9 @@
 	nv_rd32_wait:
 		iord $r12 I[$r11 + 0x000]
 		xbit $r12 $r12 31
-		bra ne nv_rd32_wait
+		bra ne #nv_rd32_wait
 	mov $r10 6			// DONE_MMIO_RD
-	call wait_doneo
+	call #wait_doneo
 	iord $r15 I[$r11 + 0x100]	// MMIO_RDVAL
 	ret
 
@@ -157,7 +157,7 @@
 	nv_wr32_wait:
 		iord $r12 I[$r11 + 0x000]
 		xbit $r12 $r12 31
-		bra ne nv_wr32_wait
+		bra ne #nv_wr32_wait
 	ret
 
 // (re)set watchdog timer
@@ -193,7 +193,7 @@
 		shl b32 $r8 6
 		iord $r8 I[$r8 + 0x000]	// DONE
 		xbit $r8 $r8 $r10
-		bra $2 wait_done_$1
+		bra $2 #wait_done_$1
 	trace_clr(T_WAIT)
 	ret
 ')
@@ -216,7 +216,7 @@
 		add b32 $r9 $r8
 		add b32 $r14 4
 		cmpu b32 $r14 $r15
-		bra ne nv_mmctx_size_loop
+		bra ne #nv_mmctx_size_loop
 	mov b32 $r15 $r9
 	ret
 
@@ -238,12 +238,12 @@
 	shl b32 $r8 6
 	clear b32 $r9
 	or $r11 $r11
-	bra e mmctx_base_disabled
+	bra e #mmctx_base_disabled
 		iowr I[$r8 + 0x000] $r11	// MMCTX_BASE
 		bset $r9 0			// BASE_EN
 	mmctx_base_disabled:
 	or $r14 $r14
-	bra e mmctx_multi_disabled
+	bra e #mmctx_multi_disabled
 		iowr I[$r8 + 0x200] $r14 	// MMCTX_MULTI_STRIDE
 		iowr I[$r8 + 0x300] $r15 	// MMCTX_MULTI_MASK
 		bset $r9 1			// MULTI_EN
@@ -264,7 +264,7 @@
 		mmctx_wait_free:
 			iord $r14 I[$r8 + 0x000] // MMCTX_CTRL
 			and $r14 0x1f
-			bra e mmctx_wait_free
+			bra e #mmctx_wait_free
 
 		// queue up an entry
 		ld b32 $r14 D[$r12]
@@ -272,19 +272,19 @@
 		iowr I[$r8 + 0x300] $r14
 		add b32 $r12 4
 		cmpu b32 $r12 $r13
-		bra ne mmctx_exec_loop
+		bra ne #mmctx_exec_loop
 
 	xbit $r11 $r10 2
-	bra ne mmctx_stop
+	bra ne #mmctx_stop
 		// wait for queue to empty
 		mmctx_fini_wait:
 			iord $r11 I[$r8 + 0x000]	// MMCTX_CTRL
 			and $r11 0x1f
 			cmpu b32 $r11 0x10
-			bra ne mmctx_fini_wait
+			bra ne #mmctx_fini_wait
 		mov $r10 2				// DONE_MMCTX
-		call wait_donez
-		bra mmctx_done
+		call #wait_donez
+		bra #mmctx_done
 	mmctx_stop:
 		xbit $r11 $r10 0
 		shl b32 $r11 16			// DIR
@@ -295,7 +295,7 @@
 			// wait for STOP_TRIGGER to clear
 			iord $r11 I[$r8 + 0x000] // MMCTX_CTRL
 			xbit $r11 $r11 18
-			bra ne mmctx_stop_wait
+			bra ne #mmctx_stop_wait
 	mmctx_done:
 	trace_clr(T_MMCTX)
 	ret
@@ -305,7 +305,7 @@
 strand_wait:
 	push $r10
 	mov $r10 2
-	call wait_donez
+	call #wait_donez
 	pop $r10
 	ret
 
@@ -316,7 +316,7 @@
 	sethi $r8 0x20000
 	mov $r9 0xc
 	iowr I[$r8] $r9
-	call strand_wait
+	call #strand_wait
 	ret
 
 // unknown - call after issuing strand commands
@@ -326,7 +326,7 @@
 	sethi $r8 0x20000
 	mov $r9 0xd
 	iowr I[$r8] $r9
-	call strand_wait
+	call #strand_wait
 	ret
 
 // Selects strand set?!
@@ -341,11 +341,11 @@
 	iowr I[$r10 + 0x000] $r12		// 0x93c = 0xf
 	mov $r12 0xb
 	iowr I[$r11 + 0x000] $r12		// 0x928 = 0xb
-	call strand_wait
+	call #strand_wait
 	iowr I[$r10 + 0x000] $r14		// 0x93c = <id>
 	mov $r12 0xa
 	iowr I[$r11 + 0x000] $r12		// 0x928 = 0xa
-	call strand_wait
+	call #strand_wait
 	ret
 
 // Initialise strand context data
@@ -357,22 +357,22 @@
 //
 strand_ctx_init:
 	trace_set(T_STRINIT)
-	call strand_pre
+	call #strand_pre
 	mov $r14 3
-	call strand_set
+	call #strand_set
 	mov $r10 0x46fc
 	sethi $r10 0x20000
 	add b32 $r11 $r10 0x400
 	iowr I[$r10 + 0x100] $r0	// STRAND_FIRST_GENE = 0
 	mov $r12 1
 	iowr I[$r11 + 0x000] $r12	// STRAND_CMD = LATCH_FIRST_GENE
-	call strand_wait
+	call #strand_wait
 	sub b32 $r12 $r0 1
 	iowr I[$r10 + 0x000] $r12	// STRAND_GENE_CNT = 0xffffffff
 	mov $r12 2
 	iowr I[$r11 + 0x000] $r12	// STRAND_CMD = LATCH_GENE_CNT
-	call strand_wait
-	call strand_post
+	call #strand_wait
+	call #strand_post
 
 	// read the size of each strand, poke the context offset of
 	// each into STRAND_{SAVE,LOAD}_SWBASE now, no need to worry
@@ -391,7 +391,7 @@
 		add b32 $r14 $r10
 		add b32 $r8 4
 		sub b32 $r9 1
-		bra ne ctx_init_strand_loop
+		bra ne #ctx_init_strand_loop
 
 	shl b32 $r14 8
 	sub b32 $r15 $r14 $r15
diff --git a/drivers/gpu/drm/nouveau/nvc0_graph.h b/drivers/gpu/drm/nouveau/nvc0_graph.h
index 636fe98..91d44ea6 100644
--- a/drivers/gpu/drm/nouveau/nvc0_graph.h
+++ b/drivers/gpu/drm/nouveau/nvc0_graph.h
@@ -87,6 +87,7 @@
 	case 0xc1:
 		return 0x9197;
 	case 0xc8:
+	case 0xd9:
 		return 0x9297;
 	default:
 		return 0;
diff --git a/drivers/gpu/drm/nouveau/nvc0_grctx.c b/drivers/gpu/drm/nouveau/nvc0_grctx.c
index dd0e6a7..de77842 100644
--- a/drivers/gpu/drm/nouveau/nvc0_grctx.c
+++ b/drivers/gpu/drm/nouveau/nvc0_grctx.c
@@ -1268,6 +1268,17 @@
 static void
 nvc0_grctx_generate_90c0(struct drm_device *dev)
 {
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	int i;
+
+	for (i = 0; dev_priv->chipset == 0xd9 && i < 4; i++) {
+		nv_mthd(dev, 0x90c0, 0x2700 + (i * 0x40), 0x00000000);
+		nv_mthd(dev, 0x90c0, 0x2720 + (i * 0x40), 0x00000000);
+		nv_mthd(dev, 0x90c0, 0x2704 + (i * 0x40), 0x00000000);
+		nv_mthd(dev, 0x90c0, 0x2724 + (i * 0x40), 0x00000000);
+		nv_mthd(dev, 0x90c0, 0x2708 + (i * 0x40), 0x00000000);
+		nv_mthd(dev, 0x90c0, 0x2728 + (i * 0x40), 0x00000000);
+	}
 	nv_mthd(dev, 0x90c0, 0x270c, 0x00000000);
 	nv_mthd(dev, 0x90c0, 0x272c, 0x00000000);
 	nv_mthd(dev, 0x90c0, 0x274c, 0x00000000);
@@ -1276,6 +1287,12 @@
 	nv_mthd(dev, 0x90c0, 0x27ac, 0x00000000);
 	nv_mthd(dev, 0x90c0, 0x27cc, 0x00000000);
 	nv_mthd(dev, 0x90c0, 0x27ec, 0x00000000);
+	for (i = 0; dev_priv->chipset == 0xd9 && i < 4; i++) {
+		nv_mthd(dev, 0x90c0, 0x2710 + (i * 0x40), 0x00014000);
+		nv_mthd(dev, 0x90c0, 0x2730 + (i * 0x40), 0x00014000);
+		nv_mthd(dev, 0x90c0, 0x2714 + (i * 0x40), 0x00000040);
+		nv_mthd(dev, 0x90c0, 0x2734 + (i * 0x40), 0x00000040);
+	}
 	nv_mthd(dev, 0x90c0, 0x030c, 0x00000001);
 	nv_mthd(dev, 0x90c0, 0x1944, 0x00000000);
 	nv_mthd(dev, 0x90c0, 0x0758, 0x00000100);
@@ -1471,14 +1488,20 @@
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 
-	if (dev_priv->chipset != 0xc1) {
-		nv_wr32(dev, 0x405800, 0x078000bf);
-		nv_wr32(dev, 0x405830, 0x02180000);
-	} else {
+	if (dev_priv->chipset == 0xd9) {
 		nv_wr32(dev, 0x405800, 0x0f8000bf);
 		nv_wr32(dev, 0x405830, 0x02180218);
+		nv_wr32(dev, 0x405834, 0x08000000);
+	} else
+	if (dev_priv->chipset == 0xc1) {
+		nv_wr32(dev, 0x405800, 0x0f8000bf);
+		nv_wr32(dev, 0x405830, 0x02180218);
+		nv_wr32(dev, 0x405834, 0x00000000);
+	} else {
+		nv_wr32(dev, 0x405800, 0x078000bf);
+		nv_wr32(dev, 0x405830, 0x02180000);
+		nv_wr32(dev, 0x405834, 0x00000000);
 	}
-	nv_wr32(dev, 0x405834, 0x00000000);
 	nv_wr32(dev, 0x405838, 0x00000000);
 	nv_wr32(dev, 0x405854, 0x00000000);
 	nv_wr32(dev, 0x405870, 0x00000001);
@@ -1509,7 +1532,10 @@
 	nv_wr32(dev, 0x4064ac, 0x00003fff);
 	nv_wr32(dev, 0x4064b4, 0x00000000);
 	nv_wr32(dev, 0x4064b8, 0x00000000);
-	if (dev_priv->chipset == 0xc1) {
+	if (dev_priv->chipset == 0xd9)
+		nv_wr32(dev, 0x4064bc, 0x00000000);
+	if (dev_priv->chipset == 0xc1 ||
+	    dev_priv->chipset == 0xd9) {
 		nv_wr32(dev, 0x4064c0, 0x80140078);
 		nv_wr32(dev, 0x4064c4, 0x0086ffff);
 	}
@@ -1550,10 +1576,23 @@
 	/* ROPC_BROADCAST */
 	nv_wr32(dev, 0x408800, 0x02802a3c);
 	nv_wr32(dev, 0x408804, 0x00000040);
-	nv_wr32(dev, 0x408808, chipset != 0xc1 ? 0x0003e00d : 0x1003e005);
-	nv_wr32(dev, 0x408900, 0x3080b801);
-	nv_wr32(dev, 0x408904, chipset != 0xc1 ? 0x02000001 : 0x62000001);
-	nv_wr32(dev, 0x408908, 0x00c80929);
+	if (chipset == 0xd9) {
+		nv_wr32(dev, 0x408808, 0x1043e005);
+		nv_wr32(dev, 0x408900, 0x3080b801);
+		nv_wr32(dev, 0x408904, 0x1043e005);
+		nv_wr32(dev, 0x408908, 0x00c8102f);
+	} else
+	if (chipset == 0xc1) {
+		nv_wr32(dev, 0x408808, 0x1003e005);
+		nv_wr32(dev, 0x408900, 0x3080b801);
+		nv_wr32(dev, 0x408904, 0x62000001);
+		nv_wr32(dev, 0x408908, 0x00c80929);
+	} else {
+		nv_wr32(dev, 0x408808, 0x0003e00d);
+		nv_wr32(dev, 0x408900, 0x3080b801);
+		nv_wr32(dev, 0x408904, 0x02000001);
+		nv_wr32(dev, 0x408908, 0x00c80929);
+	}
 	nv_wr32(dev, 0x40890c, 0x00000000);
 	nv_wr32(dev, 0x408980, 0x0000011d);
 }
@@ -1572,7 +1611,7 @@
 	nv_wr32(dev, 0x418408, 0x00000000);
 	nv_wr32(dev, 0x41840c, 0x00001008);
 	nv_wr32(dev, 0x418410, 0x0fff0fff);
-	nv_wr32(dev, 0x418414, 0x00200fff);
+	nv_wr32(dev, 0x418414, chipset != 0xd9 ? 0x00200fff : 0x02200fff);
 	nv_wr32(dev, 0x418450, 0x00000000);
 	nv_wr32(dev, 0x418454, 0x00000000);
 	nv_wr32(dev, 0x418458, 0x00000000);
@@ -1587,14 +1626,17 @@
 	nv_wr32(dev, 0x418700, 0x00000002);
 	nv_wr32(dev, 0x418704, 0x00000080);
 	nv_wr32(dev, 0x418708, 0x00000000);
-	nv_wr32(dev, 0x41870c, 0x07c80000);
+	nv_wr32(dev, 0x41870c, chipset != 0xd9 ? 0x07c80000 : 0x00000000);
 	nv_wr32(dev, 0x418710, 0x00000000);
-	nv_wr32(dev, 0x418800, 0x0006860a);
+	nv_wr32(dev, 0x418800, chipset != 0xd9 ? 0x0006860a : 0x7006860a);
 	nv_wr32(dev, 0x418808, 0x00000000);
 	nv_wr32(dev, 0x41880c, 0x00000000);
 	nv_wr32(dev, 0x418810, 0x00000000);
 	nv_wr32(dev, 0x418828, 0x00008442);
-	nv_wr32(dev, 0x418830, chipset != 0xc1 ? 0x00000001 : 0x10000001);
+	if (chipset == 0xc1 || chipset == 0xd9)
+		nv_wr32(dev, 0x418830, 0x10000001);
+	else
+		nv_wr32(dev, 0x418830, 0x00000001);
 	nv_wr32(dev, 0x4188d8, 0x00000008);
 	nv_wr32(dev, 0x4188e0, 0x01000000);
 	nv_wr32(dev, 0x4188e8, 0x00000000);
@@ -1602,7 +1644,12 @@
 	nv_wr32(dev, 0x4188f0, 0x00000000);
 	nv_wr32(dev, 0x4188f4, 0x00000000);
 	nv_wr32(dev, 0x4188f8, 0x00000000);
-	nv_wr32(dev, 0x4188fc, chipset != 0xc1 ? 0x00100000 : 0x00100018);
+	if (chipset == 0xd9)
+		nv_wr32(dev, 0x4188fc, 0x20100008);
+	else if (chipset == 0xc1)
+		nv_wr32(dev, 0x4188fc, 0x00100018);
+	else
+		nv_wr32(dev, 0x4188fc, 0x00100000);
 	nv_wr32(dev, 0x41891c, 0x00ff00ff);
 	nv_wr32(dev, 0x418924, 0x00000000);
 	nv_wr32(dev, 0x418928, 0x00ffff00);
@@ -1616,7 +1663,7 @@
 		nv_wr32(dev, 0x418a14 + (i * 0x20), 0x00000000);
 		nv_wr32(dev, 0x418a18 + (i * 0x20), 0x00000000);
 	}
-	nv_wr32(dev, 0x418b00, 0x00000000);
+	nv_wr32(dev, 0x418b00, chipset != 0xd9 ? 0x00000000 : 0x00000006);
 	nv_wr32(dev, 0x418b08, 0x0a418820);
 	nv_wr32(dev, 0x418b0c, 0x062080e6);
 	nv_wr32(dev, 0x418b10, 0x020398a4);
@@ -1633,7 +1680,7 @@
 	nv_wr32(dev, 0x418c24, 0x00000000);
 	nv_wr32(dev, 0x418c28, 0x00000000);
 	nv_wr32(dev, 0x418c2c, 0x00000000);
-	if (chipset == 0xc1)
+	if (chipset == 0xc1 || chipset == 0xd9)
 		nv_wr32(dev, 0x418c6c, 0x00000001);
 	nv_wr32(dev, 0x418c80, 0x20200004);
 	nv_wr32(dev, 0x418c8c, 0x00000001);
@@ -1653,7 +1700,10 @@
 	nv_wr32(dev, 0x419818, 0x00000000);
 	nv_wr32(dev, 0x41983c, 0x00038bc7);
 	nv_wr32(dev, 0x419848, 0x00000000);
-	nv_wr32(dev, 0x419864, chipset != 0xc1 ? 0x0000012a : 0x00000129);
+	if (chipset == 0xc1 || chipset == 0xd9)
+		nv_wr32(dev, 0x419864, 0x00000129);
+	else
+		nv_wr32(dev, 0x419864, 0x0000012a);
 	nv_wr32(dev, 0x419888, 0x00000000);
 	nv_wr32(dev, 0x419a00, 0x000001f0);
 	nv_wr32(dev, 0x419a04, 0x00000001);
@@ -1663,7 +1713,9 @@
 	nv_wr32(dev, 0x419a14, 0x00000200);
 	nv_wr32(dev, 0x419a1c, 0x00000000);
 	nv_wr32(dev, 0x419a20, 0x00000800);
-	if (chipset != 0xc0 && chipset != 0xc8)
+	if (chipset == 0xd9)
+		nv_wr32(dev, 0x00419ac4, 0x0017f440);
+	else if (chipset != 0xc0 && chipset != 0xc8)
 		nv_wr32(dev, 0x00419ac4, 0x0007f440);
 	nv_wr32(dev, 0x419b00, 0x0a418820);
 	nv_wr32(dev, 0x419b04, 0x062080e6);
@@ -1672,21 +1724,33 @@
 	nv_wr32(dev, 0x419b10, 0x0a418820);
 	nv_wr32(dev, 0x419b14, 0x000000e6);
 	nv_wr32(dev, 0x419bd0, 0x00900103);
-	nv_wr32(dev, 0x419be0, chipset != 0xc1 ? 0x00000001 : 0x00400001);
+	if (chipset == 0xc1 || chipset == 0xd9)
+		nv_wr32(dev, 0x419be0, 0x00400001);
+	else
+		nv_wr32(dev, 0x419be0, 0x00000001);
 	nv_wr32(dev, 0x419be4, 0x00000000);
-	nv_wr32(dev, 0x419c00, 0x00000002);
+	nv_wr32(dev, 0x419c00, chipset != 0xd9 ? 0x00000002 : 0x0000000a);
 	nv_wr32(dev, 0x419c04, 0x00000006);
 	nv_wr32(dev, 0x419c08, 0x00000002);
 	nv_wr32(dev, 0x419c20, 0x00000000);
-	if (chipset == 0xce || chipset == 0xcf)
+	if (dev_priv->chipset == 0xd9) {
+		nv_wr32(dev, 0x419c24, 0x00084210);
+		nv_wr32(dev, 0x419c28, 0x3cf3cf3c);
 		nv_wr32(dev, 0x419cb0, 0x00020048);
-	else
+	} else
+	if (chipset == 0xce || chipset == 0xcf) {
+		nv_wr32(dev, 0x419cb0, 0x00020048);
+	} else {
 		nv_wr32(dev, 0x419cb0, 0x00060048);
+	}
 	nv_wr32(dev, 0x419ce8, 0x00000000);
 	nv_wr32(dev, 0x419cf4, 0x00000183);
-	nv_wr32(dev, 0x419d20, chipset != 0xc1 ? 0x02180000 : 0x12180000);
+	if (chipset == 0xc1 || chipset == 0xd9)
+		nv_wr32(dev, 0x419d20, 0x12180000);
+	else
+		nv_wr32(dev, 0x419d20, 0x02180000);
 	nv_wr32(dev, 0x419d24, 0x00001fff);
-	if (chipset == 0xc1)
+	if (chipset == 0xc1 || chipset == 0xd9)
 		nv_wr32(dev, 0x419d44, 0x02180218);
 	nv_wr32(dev, 0x419e04, 0x00000000);
 	nv_wr32(dev, 0x419e08, 0x00000000);
@@ -1812,6 +1876,7 @@
 		/* calculate first set of magics */
 		memcpy(tpnr, priv->tp_nr, sizeof(priv->tp_nr));
 
+		gpc = -1;
 		for (tp = 0; tp < priv->tp_total; tp++) {
 			do {
 				gpc = (gpc + 1) % priv->gpc_nr;
@@ -1861,30 +1926,26 @@
 
 	if (1) {
 		u32 tp_mask = 0, tp_set = 0;
-		u8  tpnr[GPC_MAX];
+		u8  tpnr[GPC_MAX], a, b;
 
 		memcpy(tpnr, priv->tp_nr, sizeof(priv->tp_nr));
 		for (gpc = 0; gpc < priv->gpc_nr; gpc++)
 			tp_mask |= ((1 << priv->tp_nr[gpc]) - 1) << (gpc * 8);
 
-		gpc = -1;
-		for (i = 0, gpc = -1; i < 32; i++) {
-			int ltp = i * (priv->tp_total - 1) / 32;
+		for (i = 0, gpc = -1, b = -1; i < 32; i++) {
+			a = (i * (priv->tp_total - 1)) / 32;
+			if (a != b) {
+				b = a;
+				do {
+					gpc = (gpc + 1) % priv->gpc_nr;
+				} while (!tpnr[gpc]);
+				tp = priv->tp_nr[gpc] - tpnr[gpc]--;
 
-			do {
-				gpc = (gpc + 1) % priv->gpc_nr;
-			} while (!tpnr[gpc]);
-			tp = priv->tp_nr[gpc] - tpnr[gpc]--;
+				tp_set |= 1 << ((gpc * 8) + tp);
+			}
 
-			tp_set |= 1 << ((gpc * 8) + tp);
-
-			do {
-				nv_wr32(dev, 0x406800 + (i * 0x20), tp_set);
-				tp_set ^= tp_mask;
-				nv_wr32(dev, 0x406c00 + (i * 0x20), tp_set);
-				tp_set ^= tp_mask;
-			} while (ltp == (++i * (priv->tp_total - 1) / 32));
-			i--;
+			nv_wr32(dev, 0x406800 + (i * 0x20), tp_set);
+			nv_wr32(dev, 0x406c00 + (i * 0x20), tp_set ^ tp_mask);
 		}
 	}
 
@@ -1989,6 +2050,10 @@
 	nv_icmd(dev, 0x00000215, 0x00000040);
 	nv_icmd(dev, 0x00000216, 0x00000040);
 	nv_icmd(dev, 0x00000217, 0x00000040);
+	if (dev_priv->chipset == 0xd9) {
+		for (i = 0x0400; i <= 0x0417; i++)
+			nv_icmd(dev, i, 0x00000040);
+	}
 	nv_icmd(dev, 0x00000218, 0x0000c080);
 	nv_icmd(dev, 0x00000219, 0x0000c080);
 	nv_icmd(dev, 0x0000021a, 0x0000c080);
@@ -1997,6 +2062,10 @@
 	nv_icmd(dev, 0x0000021d, 0x0000c080);
 	nv_icmd(dev, 0x0000021e, 0x0000c080);
 	nv_icmd(dev, 0x0000021f, 0x0000c080);
+	if (dev_priv->chipset == 0xd9) {
+		for (i = 0x0440; i <= 0x0457; i++)
+			nv_icmd(dev, i, 0x0000c080);
+	}
 	nv_icmd(dev, 0x000000ad, 0x0000013e);
 	nv_icmd(dev, 0x000000e1, 0x00000010);
 	nv_icmd(dev, 0x00000290, 0x00000000);
@@ -2559,7 +2628,8 @@
 	nv_icmd(dev, 0x0000053f, 0xffff0000);
 	nv_icmd(dev, 0x00000585, 0x0000003f);
 	nv_icmd(dev, 0x00000576, 0x00000003);
-	if (dev_priv->chipset == 0xc1)
+	if (dev_priv->chipset == 0xc1 ||
+	    dev_priv->chipset == 0xd9)
 		nv_icmd(dev, 0x0000057b, 0x00000059);
 	nv_icmd(dev, 0x00000586, 0x00000040);
 	nv_icmd(dev, 0x00000582, 0x00000080);
@@ -2661,6 +2731,8 @@
 	nv_icmd(dev, 0x00000957, 0x00000003);
 	nv_icmd(dev, 0x0000095e, 0x20164010);
 	nv_icmd(dev, 0x0000095f, 0x00000020);
+	if (dev_priv->chipset == 0xd9)
+		nv_icmd(dev, 0x0000097d, 0x00000020);
 	nv_icmd(dev, 0x00000683, 0x00000006);
 	nv_icmd(dev, 0x00000685, 0x003fffff);
 	nv_icmd(dev, 0x00000687, 0x00000c48);
diff --git a/drivers/gpu/drm/nouveau/nvc0_grgpc.fuc b/drivers/gpu/drm/nouveau/nvc0_grgpc.fuc
index 06f5e26..15272be 100644
--- a/drivers/gpu/drm/nouveau/nvc0_grgpc.fuc
+++ b/drivers/gpu/drm/nouveau/nvc0_grgpc.fuc
@@ -32,7 +32,7 @@
  * - watchdog timer around ctx operations
  */
 
-.section nvc0_grgpc_data
+.section #nvc0_grgpc_data
 include(`nvc0_graph.fuc')
 gpc_id:			.b32 0
 gpc_mmio_list_head:	.b32 0
@@ -48,40 +48,45 @@
 // chipset descriptions
 chipsets:
 .b8  0xc0 0 0 0
-.b16 nvc0_gpc_mmio_head
-.b16 nvc0_gpc_mmio_tail
-.b16 nvc0_tpc_mmio_head
-.b16 nvc0_tpc_mmio_tail
+.b16 #nvc0_gpc_mmio_head
+.b16 #nvc0_gpc_mmio_tail
+.b16 #nvc0_tpc_mmio_head
+.b16 #nvc0_tpc_mmio_tail
 .b8  0xc1 0 0 0
-.b16 nvc0_gpc_mmio_head
-.b16 nvc1_gpc_mmio_tail
-.b16 nvc0_tpc_mmio_head
-.b16 nvc1_tpc_mmio_tail
+.b16 #nvc0_gpc_mmio_head
+.b16 #nvc1_gpc_mmio_tail
+.b16 #nvc0_tpc_mmio_head
+.b16 #nvc1_tpc_mmio_tail
 .b8  0xc3 0 0 0
-.b16 nvc0_gpc_mmio_head
-.b16 nvc0_gpc_mmio_tail
-.b16 nvc0_tpc_mmio_head
-.b16 nvc3_tpc_mmio_tail
+.b16 #nvc0_gpc_mmio_head
+.b16 #nvc0_gpc_mmio_tail
+.b16 #nvc0_tpc_mmio_head
+.b16 #nvc3_tpc_mmio_tail
 .b8  0xc4 0 0 0
-.b16 nvc0_gpc_mmio_head
-.b16 nvc0_gpc_mmio_tail
-.b16 nvc0_tpc_mmio_head
-.b16 nvc3_tpc_mmio_tail
+.b16 #nvc0_gpc_mmio_head
+.b16 #nvc0_gpc_mmio_tail
+.b16 #nvc0_tpc_mmio_head
+.b16 #nvc3_tpc_mmio_tail
 .b8  0xc8 0 0 0
-.b16 nvc0_gpc_mmio_head
-.b16 nvc0_gpc_mmio_tail
-.b16 nvc0_tpc_mmio_head
-.b16 nvc0_tpc_mmio_tail
+.b16 #nvc0_gpc_mmio_head
+.b16 #nvc0_gpc_mmio_tail
+.b16 #nvc0_tpc_mmio_head
+.b16 #nvc0_tpc_mmio_tail
 .b8  0xce 0 0 0
-.b16 nvc0_gpc_mmio_head
-.b16 nvc0_gpc_mmio_tail
-.b16 nvc0_tpc_mmio_head
-.b16 nvc3_tpc_mmio_tail
+.b16 #nvc0_gpc_mmio_head
+.b16 #nvc0_gpc_mmio_tail
+.b16 #nvc0_tpc_mmio_head
+.b16 #nvc3_tpc_mmio_tail
 .b8  0xcf 0 0 0
-.b16 nvc0_gpc_mmio_head
-.b16 nvc0_gpc_mmio_tail
-.b16 nvc0_tpc_mmio_head
-.b16 nvcf_tpc_mmio_tail
+.b16 #nvc0_gpc_mmio_head
+.b16 #nvc0_gpc_mmio_tail
+.b16 #nvc0_tpc_mmio_head
+.b16 #nvcf_tpc_mmio_tail
+.b8  0xd9 0 0 0
+.b16 #nvd9_gpc_mmio_head
+.b16 #nvd9_gpc_mmio_tail
+.b16 #nvd9_tpc_mmio_head
+.b16 #nvd9_tpc_mmio_tail
 .b8  0 0 0 0
 
 // GPC mmio lists
@@ -114,6 +119,35 @@
 mmctx_data(0x000c6c, 1);
 nvc1_gpc_mmio_tail:
 
+nvd9_gpc_mmio_head:
+mmctx_data(0x000380, 1)
+mmctx_data(0x000400, 2)
+mmctx_data(0x00040c, 3)
+mmctx_data(0x000450, 9)
+mmctx_data(0x000600, 1)
+mmctx_data(0x000684, 1)
+mmctx_data(0x000700, 5)
+mmctx_data(0x000800, 1)
+mmctx_data(0x000808, 3)
+mmctx_data(0x000828, 1)
+mmctx_data(0x000830, 1)
+mmctx_data(0x0008d8, 1)
+mmctx_data(0x0008e0, 1)
+mmctx_data(0x0008e8, 6)
+mmctx_data(0x00091c, 1)
+mmctx_data(0x000924, 3)
+mmctx_data(0x000b00, 1)
+mmctx_data(0x000b08, 6)
+mmctx_data(0x000bb8, 1)
+mmctx_data(0x000c08, 1)
+mmctx_data(0x000c10, 8)
+mmctx_data(0x000c6c, 1)
+mmctx_data(0x000c80, 1)
+mmctx_data(0x000c8c, 1)
+mmctx_data(0x001000, 3)
+mmctx_data(0x001014, 1)
+nvd9_gpc_mmio_tail:
+
 // TPC mmio lists
 nvc0_tpc_mmio_head:
 mmctx_data(0x000018, 1)
@@ -146,9 +180,34 @@
 mmctx_data(0x000544, 1)
 nvc1_tpc_mmio_tail:
 
+nvd9_tpc_mmio_head:
+mmctx_data(0x000018, 1)
+mmctx_data(0x00003c, 1)
+mmctx_data(0x000048, 1)
+mmctx_data(0x000064, 1)
+mmctx_data(0x000088, 1)
+mmctx_data(0x000200, 6)
+mmctx_data(0x00021c, 2)
+mmctx_data(0x0002c4, 1)
+mmctx_data(0x000300, 6)
+mmctx_data(0x0003d0, 1)
+mmctx_data(0x0003e0, 2)
+mmctx_data(0x000400, 3)
+mmctx_data(0x000420, 3)
+mmctx_data(0x0004b0, 1)
+mmctx_data(0x0004e8, 1)
+mmctx_data(0x0004f4, 1)
+mmctx_data(0x000520, 2)
+mmctx_data(0x000544, 1)
+mmctx_data(0x000604, 4)
+mmctx_data(0x000644, 20)
+mmctx_data(0x000698, 1)
+mmctx_data(0x0006e0, 1)
+mmctx_data(0x000750, 3)
+nvd9_tpc_mmio_tail:
 
-.section nvc0_grgpc_code
-bra init
+.section #nvc0_grgpc_code
+bra #init
 define(`include_code')
 include(`nvc0_graph.fuc')
 
@@ -160,10 +219,10 @@
 	push $r14
 	mov $r14 -0x67ec 	// 0x9814
 	sethi $r14 0x400000
-	call nv_wr32		// HUB_CTXCTL_CC_SCRATCH[5] = error code
+	call #nv_wr32		// HUB_CTXCTL_CC_SCRATCH[5] = error code
 	add b32 $r14 0x41c
 	mov $r15 1
-	call nv_wr32		// HUB_CTXCTL_INTR_UP_SET
+	call #nv_wr32		// HUB_CTXCTL_INTR_UP_SET
 	pop $r14
 	ret
 
@@ -190,7 +249,7 @@
 	iowr I[$r1 + 0x000] $r2		// FIFO_ENABLE
 
 	// setup i0 handler, and route all interrupts to it
-	mov $r1 ih
+	mov $r1 #ih
 	mov $iv0 $r1
 	mov $r1 0x400
 	iowr I[$r1 + 0x300] $r0		// INTR_DISPATCH
@@ -210,24 +269,24 @@
 	and $r2 0x1f
 	shl b32 $r3 $r2
 	sub b32 $r3 1
-	st b32 D[$r0 + tpc_count] $r2
-	st b32 D[$r0 + tpc_mask] $r3
+	st b32 D[$r0 + #tpc_count] $r2
+	st b32 D[$r0 + #tpc_mask] $r3
 	add b32 $r1 0x400
 	iord $r2 I[$r1 + 0x000]		// MYINDEX
-	st b32 D[$r0 + gpc_id] $r2
+	st b32 D[$r0 + #gpc_id] $r2
 
 	// find context data for this chipset
 	mov $r2 0x800
 	shl b32 $r2 6
 	iord $r2 I[$r2 + 0x000]		// CC_SCRATCH[0]
-	mov $r1 chipsets - 12
+	mov $r1 #chipsets - 12
 	init_find_chipset:
 		add b32 $r1 12
 		ld b32 $r3 D[$r1 + 0x00]
 		cmpu b32 $r3 $r2
-		bra e init_context
+		bra e #init_context
 		cmpu b32 $r3 0
-		bra ne init_find_chipset
+		bra ne #init_find_chipset
 		// unknown chipset
 		ret
 
@@ -253,19 +312,19 @@
 	clear b32 $r15
 	ld b16 $r14 D[$r1 + 4]
 	ld b16 $r15 D[$r1 + 6]
-	st b16 D[$r0 + gpc_mmio_list_head] $r14
-	st b16 D[$r0 + gpc_mmio_list_tail] $r15
-	call mmctx_size
+	st b16 D[$r0 + #gpc_mmio_list_head] $r14
+	st b16 D[$r0 + #gpc_mmio_list_tail] $r15
+	call #mmctx_size
 	add b32 $r2 $r15
 	add b32 $r3 $r15
 
 	// calculate per-TPC mmio context size, store the list pointers
 	ld b16 $r14 D[$r1 + 8]
 	ld b16 $r15 D[$r1 + 10]
-	st b16 D[$r0 + tpc_mmio_list_head] $r14
-	st b16 D[$r0 + tpc_mmio_list_tail] $r15
-	call mmctx_size
-	ld b32 $r14 D[$r0 + tpc_count]
+	st b16 D[$r0 + #tpc_mmio_list_head] $r14
+	st b16 D[$r0 + #tpc_mmio_list_tail] $r15
+	call #mmctx_size
+	ld b32 $r14 D[$r0 + #tpc_count]
 	mulu $r14 $r15
 	add b32 $r2 $r14
 	add b32 $r3 $r14
@@ -283,7 +342,7 @@
 
 	// calculate size of strand context data
 	mov b32 $r15 $r2
-	call strand_ctx_init
+	call #strand_ctx_init
 	add b32 $r3 $r15
 
 	// save context size, and tell HUB we're done
@@ -301,13 +360,13 @@
 main:
 	bset $flags $p0
 	sleep $p0
-	mov $r13 cmd_queue
-	call queue_get
-	bra $p1 main
+	mov $r13 #cmd_queue
+	call #queue_get
+	bra $p1 #main
 
 	// 0x0000-0x0003 are all context transfers
 	cmpu b32 $r14 0x04
-	bra nc main_not_ctx_xfer
+	bra nc #main_not_ctx_xfer
 		// fetch $flags and mask off $p1/$p2
 		mov $r1 $flags
 		mov $r2 0x0006
@@ -318,14 +377,14 @@
 		or $r1 $r14
 		mov $flags $r1
 		// transfer context data
-		call ctx_xfer
-		bra main
+		call #ctx_xfer
+		bra #main
 
 	main_not_ctx_xfer:
 	shl b32 $r15 $r14 16
 	or $r15 E_BAD_COMMAND
-	call error
-	bra main
+	call #error
+	bra #main
 
 // interrupt handler
 ih:
@@ -342,13 +401,13 @@
 	// incoming fifo command?
 	iord $r10 I[$r0 + 0x200]	// INTR
 	and $r11 $r10 0x00000004
-	bra e ih_no_fifo
+	bra e #ih_no_fifo
 		// queue incoming fifo command for later processing
 		mov $r11 0x1900
-		mov $r13 cmd_queue
+		mov $r13 #cmd_queue
 		iord $r14 I[$r11 + 0x100]	// FIFO_CMD
 		iord $r15 I[$r11 + 0x000]	// FIFO_DATA
-		call queue_put
+		call #queue_put
 		add b32 $r11 0x400
 		mov $r14 1
 		iowr I[$r11 + 0x000] $r14	// FIFO_ACK
@@ -374,11 +433,11 @@
 //
 hub_barrier_done:
 	mov $r15 1
-	ld b32 $r14 D[$r0 + gpc_id]
+	ld b32 $r14 D[$r0 + #gpc_id]
 	shl b32 $r15 $r14
 	mov $r14 -0x6be8 	// 0x409418 - HUB_BAR_SET
 	sethi $r14 0x400000
-	call nv_wr32
+	call #nv_wr32
 	ret
 
 // Disables various things, waits a bit, and re-enables them..
@@ -395,7 +454,7 @@
 	mov $r15 8
 	ctx_redswitch_delay:
 		sub b32 $r15 1
-		bra ne ctx_redswitch_delay
+		bra ne #ctx_redswitch_delay
 	mov $r15 0xa20
 	iowr I[$r14] $r15	// GPC_RED_SWITCH = UNK11, ENABLE, POWER
 	ret
@@ -413,8 +472,8 @@
 	mov $r1 0xa04
 	shl b32 $r1 6
 	iowr I[$r1 + 0x000] $r15// MEM_BASE
-	bra not $p1 ctx_xfer_not_load
-		call ctx_redswitch
+	bra not $p1 #ctx_xfer_not_load
+		call #ctx_redswitch
 	ctx_xfer_not_load:
 
 	// strands
@@ -422,7 +481,7 @@
 	sethi $r1 0x20000
 	mov $r2 0xc
 	iowr I[$r1] $r2		// STRAND_CMD(0x3f) = 0x0c
-	call strand_wait
+	call #strand_wait
 	mov $r2 0x47fc
 	sethi $r2 0x20000
 	iowr I[$r2] $r0		// STRAND_FIRST_GENE(0x3f) = 0x00
@@ -435,46 +494,46 @@
 	or $r10 2		// first
 	mov $r11 0x0000
 	sethi $r11 0x500000
-	ld b32 $r12 D[$r0 + gpc_id]
+	ld b32 $r12 D[$r0 + #gpc_id]
 	shl b32 $r12 15
 	add b32 $r11 $r12	// base = NV_PGRAPH_GPCn
-	ld b32 $r12 D[$r0 + gpc_mmio_list_head]
-	ld b32 $r13 D[$r0 + gpc_mmio_list_tail]
+	ld b32 $r12 D[$r0 + #gpc_mmio_list_head]
+	ld b32 $r13 D[$r0 + #gpc_mmio_list_tail]
 	mov $r14 0		// not multi
-	call mmctx_xfer
+	call #mmctx_xfer
 
 	// per-TPC mmio context
 	xbit $r10 $flags $p1	// direction
 	or $r10 4		// last
 	mov $r11 0x4000
 	sethi $r11 0x500000	// base = NV_PGRAPH_GPC0_TPC0
-	ld b32 $r12 D[$r0 + gpc_id]
+	ld b32 $r12 D[$r0 + #gpc_id]
 	shl b32 $r12 15
 	add b32 $r11 $r12	// base = NV_PGRAPH_GPCn_TPC0
-	ld b32 $r12 D[$r0 + tpc_mmio_list_head]
-	ld b32 $r13 D[$r0 + tpc_mmio_list_tail]
-	ld b32 $r15 D[$r0 + tpc_mask]
+	ld b32 $r12 D[$r0 + #tpc_mmio_list_head]
+	ld b32 $r13 D[$r0 + #tpc_mmio_list_tail]
+	ld b32 $r15 D[$r0 + #tpc_mask]
 	mov $r14 0x800		// stride = 0x800
-	call mmctx_xfer
+	call #mmctx_xfer
 
 	// wait for strands to finish
-	call strand_wait
+	call #strand_wait
 
 	// if load, or a save without a load following, do some
 	// unknown stuff that's done after finishing a block of
 	// strand commands
-	bra $p1 ctx_xfer_post
-	bra not $p2 ctx_xfer_done
+	bra $p1 #ctx_xfer_post
+	bra not $p2 #ctx_xfer_done
 	ctx_xfer_post:
 		mov $r1 0x4afc
 		sethi $r1 0x20000
 		mov $r2 0xd
 		iowr I[$r1] $r2		// STRAND_CMD(0x3f) = 0x0d
-		call strand_wait
+		call #strand_wait
 
 	// mark completion in HUB's barrier
 	ctx_xfer_done:
-	call hub_barrier_done
+	call #hub_barrier_done
 	ret
 
 .align 256
diff --git a/drivers/gpu/drm/nouveau/nvc0_grgpc.fuc.h b/drivers/gpu/drm/nouveau/nvc0_grgpc.fuc.h
index 6f82032..a988b8a 100644
--- a/drivers/gpu/drm/nouveau/nvc0_grgpc.fuc.h
+++ b/drivers/gpu/drm/nouveau/nvc0_grgpc.fuc.h
@@ -25,26 +25,29 @@
 	0x00000000,
 	0x00000000,
 	0x000000c0,
-	0x011c00bc,
-	0x01700120,
+	0x012800c8,
+	0x01e40194,
 	0x000000c1,
-	0x012000bc,
-	0x01840120,
+	0x012c00c8,
+	0x01f80194,
 	0x000000c3,
-	0x011c00bc,
-	0x01800120,
+	0x012800c8,
+	0x01f40194,
 	0x000000c4,
-	0x011c00bc,
-	0x01800120,
+	0x012800c8,
+	0x01f40194,
 	0x000000c8,
-	0x011c00bc,
-	0x01700120,
+	0x012800c8,
+	0x01e40194,
 	0x000000ce,
-	0x011c00bc,
-	0x01800120,
+	0x012800c8,
+	0x01f40194,
 	0x000000cf,
-	0x011c00bc,
-	0x017c0120,
+	0x012800c8,
+	0x01f00194,
+	0x000000d9,
+	0x0194012c,
+	0x025401f8,
 	0x00000000,
 	0x00000380,
 	0x14000400,
@@ -71,6 +74,32 @@
 	0x08001000,
 	0x00001014,
 	0x00000c6c,
+	0x00000380,
+	0x04000400,
+	0x0800040c,
+	0x20000450,
+	0x00000600,
+	0x00000684,
+	0x10000700,
+	0x00000800,
+	0x08000808,
+	0x00000828,
+	0x00000830,
+	0x000008d8,
+	0x000008e0,
+	0x140008e8,
+	0x0000091c,
+	0x08000924,
+	0x00000b00,
+	0x14000b08,
+	0x00000bb8,
+	0x00000c08,
+	0x1c000c10,
+	0x00000c6c,
+	0x00000c80,
+	0x00000c8c,
+	0x08001000,
+	0x00001014,
 	0x00000018,
 	0x0000003c,
 	0x00000048,
@@ -96,6 +125,29 @@
 	0x000006e0,
 	0x000004bc,
 	0x00000544,
+	0x00000018,
+	0x0000003c,
+	0x00000048,
+	0x00000064,
+	0x00000088,
+	0x14000200,
+	0x0400021c,
+	0x000002c4,
+	0x14000300,
+	0x000003d0,
+	0x040003e0,
+	0x08000400,
+	0x08000420,
+	0x000004b0,
+	0x000004e8,
+	0x000004f4,
+	0x04000520,
+	0x00000544,
+	0x0c000604,
+	0x4c000644,
+	0x00000698,
+	0x000006e0,
+	0x08000750,
 };
 
 uint32_t nvc0_grgpc_code[] = {
diff --git a/drivers/gpu/drm/nouveau/nvc0_grhub.fuc b/drivers/gpu/drm/nouveau/nvc0_grhub.fuc
index e4f8c7e..98acddb 100644
--- a/drivers/gpu/drm/nouveau/nvc0_grhub.fuc
+++ b/drivers/gpu/drm/nouveau/nvc0_grhub.fuc
@@ -27,7 +27,7 @@
  *    m4 nvc0_grhub.fuc | envyas -a -w -m fuc -V nva3 -o nvc0_grhub.fuc.h
  */
 
-.section nvc0_grhub_data
+.section #nvc0_grhub_data
 include(`nvc0_graph.fuc')
 gpc_count:		.b32 0
 rop_count:		.b32 0
@@ -39,26 +39,29 @@
 
 chipsets:
 .b8  0xc0 0 0 0
-.b16 nvc0_hub_mmio_head
-.b16 nvc0_hub_mmio_tail
+.b16 #nvc0_hub_mmio_head
+.b16 #nvc0_hub_mmio_tail
 .b8  0xc1 0 0 0
-.b16 nvc0_hub_mmio_head
-.b16 nvc1_hub_mmio_tail
+.b16 #nvc0_hub_mmio_head
+.b16 #nvc1_hub_mmio_tail
 .b8  0xc3 0 0 0
-.b16 nvc0_hub_mmio_head
-.b16 nvc0_hub_mmio_tail
+.b16 #nvc0_hub_mmio_head
+.b16 #nvc0_hub_mmio_tail
 .b8  0xc4 0 0 0
-.b16 nvc0_hub_mmio_head
-.b16 nvc0_hub_mmio_tail
+.b16 #nvc0_hub_mmio_head
+.b16 #nvc0_hub_mmio_tail
 .b8  0xc8 0 0 0
-.b16 nvc0_hub_mmio_head
-.b16 nvc0_hub_mmio_tail
+.b16 #nvc0_hub_mmio_head
+.b16 #nvc0_hub_mmio_tail
 .b8  0xce 0 0 0
-.b16 nvc0_hub_mmio_head
-.b16 nvc0_hub_mmio_tail
+.b16 #nvc0_hub_mmio_head
+.b16 #nvc0_hub_mmio_tail
 .b8  0xcf 0 0 0
-.b16 nvc0_hub_mmio_head
-.b16 nvc0_hub_mmio_tail
+.b16 #nvc0_hub_mmio_head
+.b16 #nvc0_hub_mmio_tail
+.b8  0xd9 0 0 0
+.b16 #nvd9_hub_mmio_head
+.b16 #nvd9_hub_mmio_tail
 .b8  0 0 0 0
 
 nvc0_hub_mmio_head:
@@ -105,6 +108,48 @@
 mmctx_data(0x4064c0, 2)
 nvc1_hub_mmio_tail:
 
+nvd9_hub_mmio_head:
+mmctx_data(0x17e91c, 2)
+mmctx_data(0x400204, 2)
+mmctx_data(0x404004, 10)
+mmctx_data(0x404044, 1)
+mmctx_data(0x404094, 14)
+mmctx_data(0x4040d0, 7)
+mmctx_data(0x4040f8, 1)
+mmctx_data(0x404130, 3)
+mmctx_data(0x404150, 3)
+mmctx_data(0x404164, 2)
+mmctx_data(0x404178, 2)
+mmctx_data(0x404200, 8)
+mmctx_data(0x404404, 14)
+mmctx_data(0x404460, 4)
+mmctx_data(0x404480, 1)
+mmctx_data(0x404498, 1)
+mmctx_data(0x404604, 4)
+mmctx_data(0x404618, 32)
+mmctx_data(0x404698, 21)
+mmctx_data(0x4046f0, 2)
+mmctx_data(0x404700, 22)
+mmctx_data(0x405800, 1)
+mmctx_data(0x405830, 3)
+mmctx_data(0x405854, 1)
+mmctx_data(0x405870, 4)
+mmctx_data(0x405a00, 2)
+mmctx_data(0x405a18, 1)
+mmctx_data(0x406020, 1)
+mmctx_data(0x406028, 4)
+mmctx_data(0x4064a8, 2)
+mmctx_data(0x4064b4, 5)
+mmctx_data(0x407804, 1)
+mmctx_data(0x40780c, 6)
+mmctx_data(0x4078bc, 1)
+mmctx_data(0x408000, 7)
+mmctx_data(0x408064, 1)
+mmctx_data(0x408800, 3)
+mmctx_data(0x408900, 4)
+mmctx_data(0x408980, 1)
+nvd9_hub_mmio_tail:
+
 .align 256
 chan_data:
 chan_mmio_count:	.b32 0
@@ -113,8 +158,8 @@
 .align 256
 xfer_data: 		.b32 0
 
-.section nvc0_grhub_code
-bra init
+.section #nvc0_grhub_code
+bra #init
 define(`include_code')
 include(`nvc0_graph.fuc')
 
@@ -157,7 +202,7 @@
 	iowr I[$r1 + 0x000] $r2	// FIFO_ENABLE
 
 	// setup i0 handler, and route all interrupts to it
-	mov $r1 ih
+	mov $r1 #ih
 	mov $iv0 $r1
 	mov $r1 0x400
 	iowr I[$r1 + 0x300] $r0	// INTR_DISPATCH
@@ -201,11 +246,11 @@
 	// fetch enabled GPC/ROP counts
 	mov $r14 -0x69fc	// 0x409604
 	sethi $r14 0x400000
-	call nv_rd32
+	call #nv_rd32
 	extr $r1 $r15 16:20
-	st b32 D[$r0 + rop_count] $r1
+	st b32 D[$r0 + #rop_count] $r1
 	and $r15 0x1f
-	st b32 D[$r0 + gpc_count] $r15
+	st b32 D[$r0 + #gpc_count] $r15
 
 	// set BAR_REQMASK to GPC mask
 	mov $r1 1
@@ -220,14 +265,14 @@
 	mov $r2 0x800
 	shl b32 $r2 6
 	iord $r2 I[$r2 + 0x000]		// CC_SCRATCH[0]
-	mov $r15 chipsets - 8
+	mov $r15 #chipsets - 8
 	init_find_chipset:
 		add b32 $r15 8
 		ld b32 $r3 D[$r15 + 0x00]
 		cmpu b32 $r3 $r2
-		bra e init_context
+		bra e #init_context
 		cmpu b32 $r3 0
-		bra ne init_find_chipset
+		bra ne #init_find_chipset
 		// unknown chipset
 		ret
 
@@ -239,9 +284,9 @@
 	ld b16 $r14 D[$r15 + 4]
 	ld b16 $r15 D[$r15 + 6]
 	sethi $r14 0
-	st b32 D[$r0 + hub_mmio_list_head] $r14
-	st b32 D[$r0 + hub_mmio_list_tail] $r15
-	call mmctx_size
+	st b32 D[$r0 + #hub_mmio_list_head] $r14
+	st b32 D[$r0 + #hub_mmio_list_tail] $r15
+	call #mmctx_size
 
 	// set mmctx base addresses now so we don't have to do it later,
 	// they don't (currently) ever change
@@ -260,7 +305,7 @@
 	add b32 $r1 1
 	shl b32 $r1 8
 	mov b32 $r15 $r1
-	call strand_ctx_init
+	call #strand_ctx_init
 	add b32 $r1 $r15
 
 	// initialise each GPC in sequence by passing in the offset of its
@@ -271,40 +316,40 @@
 	// when it has completed, and return the size of its context data
 	// in GPCn_CC_SCRATCH[1]
 	//
-	ld b32 $r3 D[$r0 + gpc_count]
+	ld b32 $r3 D[$r0 + #gpc_count]
 	mov $r4 0x2000
 	sethi $r4 0x500000
 	init_gpc:
 		// setup, and start GPC ucode running
 		add b32 $r14 $r4 0x804
 		mov b32 $r15 $r1
-		call nv_wr32			// CC_SCRATCH[1] = ctx offset
+		call #nv_wr32			// CC_SCRATCH[1] = ctx offset
 		add b32 $r14 $r4 0x800
 		mov b32 $r15 $r2
-		call nv_wr32			// CC_SCRATCH[0] = chipset
+		call #nv_wr32			// CC_SCRATCH[0] = chipset
 		add b32 $r14 $r4 0x10c
 		clear b32 $r15
-		call nv_wr32
+		call #nv_wr32
 		add b32 $r14 $r4 0x104
-		call nv_wr32			// ENTRY
+		call #nv_wr32			// ENTRY
 		add b32 $r14 $r4 0x100
 		mov $r15 2			// CTRL_START_TRIGGER
-		call nv_wr32			// CTRL
+		call #nv_wr32			// CTRL
 
 		// wait for it to complete, and adjust context size
 		add b32 $r14 $r4 0x800
 		init_gpc_wait:
-			call nv_rd32
+			call #nv_rd32
 			xbit $r15 $r15 31
-			bra e init_gpc_wait
+			bra e #init_gpc_wait
 		add b32 $r14 $r4 0x804
-		call nv_rd32
+		call #nv_rd32
 		add b32 $r1 $r15
 
 		// next!
 		add b32 $r4 0x8000
 		sub b32 $r3 1
-		bra ne init_gpc
+		bra ne #init_gpc
 
 	// save context size, and tell host we're ready
 	mov $r2 0x800
@@ -322,13 +367,13 @@
 	// sleep until we have something to do
 	bset $flags $p0
 	sleep $p0
-	mov $r13 cmd_queue
-	call queue_get
-	bra $p1 main
+	mov $r13 #cmd_queue
+	call #queue_get
+	bra $p1 #main
 
 	// context switch, requested by GPU?
 	cmpu b32 $r14 0x4001
-	bra ne main_not_ctx_switch
+	bra ne #main_not_ctx_switch
 		trace_set(T_AUTO)
 		mov $r1 0xb00
 		shl b32 $r1 6
@@ -336,39 +381,39 @@
 		iord $r1 I[$r1 + 0x000]		// CHAN_CUR
 
 		xbit $r3 $r1 31
-		bra e chsw_no_prev
+		bra e #chsw_no_prev
 			xbit $r3 $r2 31
-			bra e chsw_prev_no_next
+			bra e #chsw_prev_no_next
 				push $r2
 				mov b32 $r2 $r1
 				trace_set(T_SAVE)
 				bclr $flags $p1
 				bset $flags $p2
-				call ctx_xfer
+				call #ctx_xfer
 				trace_clr(T_SAVE);
 				pop $r2
 				trace_set(T_LOAD);
 				bset $flags $p1
-				call ctx_xfer
+				call #ctx_xfer
 				trace_clr(T_LOAD);
-				bra chsw_done
+				bra #chsw_done
 			chsw_prev_no_next:
 				push $r2
 				mov b32 $r2 $r1
 				bclr $flags $p1
 				bclr $flags $p2
-				call ctx_xfer
+				call #ctx_xfer
 				pop $r2
 				mov $r1 0xb00
 				shl b32 $r1 6
 				iowr I[$r1] $r2
-				bra chsw_done
+				bra #chsw_done
 		chsw_no_prev:
 			xbit $r3 $r2 31
-			bra e chsw_done
+			bra e #chsw_done
 				bset $flags $p1
 				bclr $flags $p2
-				call ctx_xfer
+				call #ctx_xfer
 
 		// ack the context switch request
 		chsw_done:
@@ -377,32 +422,32 @@
 		mov $r2 1
 		iowr I[$r1 + 0x000] $r2		// 0x409b0c
 		trace_clr(T_AUTO)
-		bra main
+		bra #main
 
 	// request to set current channel? (*not* a context switch)
 	main_not_ctx_switch:
 	cmpu b32 $r14 0x0001
-	bra ne main_not_ctx_chan
+	bra ne #main_not_ctx_chan
 		mov b32 $r2 $r15
-		call ctx_chan
-		bra main_done
+		call #ctx_chan
+		bra #main_done
 
 	// request to store current channel context?
 	main_not_ctx_chan:
 	cmpu b32 $r14 0x0002
-	bra ne main_not_ctx_save
+	bra ne #main_not_ctx_save
 		trace_set(T_SAVE)
 		bclr $flags $p1
 		bclr $flags $p2
-		call ctx_xfer
+		call #ctx_xfer
 		trace_clr(T_SAVE)
-		bra main_done
+		bra #main_done
 
 	main_not_ctx_save:
 		shl b32 $r15 $r14 16
 		or $r15 E_BAD_COMMAND
-		call error
-		bra main
+		call #error
+		bra #main
 
 	main_done:
 	mov $r1 0x820
@@ -410,7 +455,7 @@
 	clear b32 $r2
 	bset $r2 31
 	iowr I[$r1 + 0x000] $r2		// CC_SCRATCH[0] |= 0x80000000
-	bra main
+	bra #main
 
 // interrupt handler
 ih:
@@ -427,13 +472,13 @@
 	// incoming fifo command?
 	iord $r10 I[$r0 + 0x200]	// INTR
 	and $r11 $r10 0x00000004
-	bra e ih_no_fifo
+	bra e #ih_no_fifo
 		// queue incoming fifo command for later processing
 		mov $r11 0x1900
-		mov $r13 cmd_queue
+		mov $r13 #cmd_queue
 		iord $r14 I[$r11 + 0x100]	// FIFO_CMD
 		iord $r15 I[$r11 + 0x000]	// FIFO_DATA
-		call queue_put
+		call #queue_put
 		add b32 $r11 0x400
 		mov $r14 1
 		iowr I[$r11 + 0x000] $r14	// FIFO_ACK
@@ -441,18 +486,18 @@
 	// context switch request?
 	ih_no_fifo:
 	and $r11 $r10 0x00000100
-	bra e ih_no_ctxsw
+	bra e #ih_no_ctxsw
 		// enqueue a context switch for later processing
-		mov $r13 cmd_queue
+		mov $r13 #cmd_queue
 		mov $r14 0x4001
-		call queue_put
+		call #queue_put
 
 	// anything we didn't handle, bring it to the host's attention
 	ih_no_ctxsw:
 	mov $r11 0x104
 	not b32 $r11
 	and $r11 $r10 $r11
-	bra e ih_no_other
+	bra e #ih_no_other
 		mov $r10 0xc1c
 		shl b32 $r10 6
 		iowr I[$r10] $r11	// INTR_UP_SET
@@ -478,11 +523,11 @@
 	mov $r14 0x4160
 	sethi $r14 0x400000
 	mov $r15 1
-	call nv_wr32
+	call #nv_wr32
 	ctx_4160s_wait:
-		call nv_rd32
+		call #nv_rd32
 		xbit $r15 $r15 4
-		bra e ctx_4160s_wait
+		bra e #ctx_4160s_wait
 	ret
 
 // Without clearing again at end of xfer, some things cause PGRAPH
@@ -492,7 +537,7 @@
 	mov $r14 0x4160
 	sethi $r14 0x400000
 	clear b32 $r15
-	call nv_wr32
+	call #nv_wr32
 	ret
 
 // Again, not real sure
@@ -503,7 +548,7 @@
 	mov $r14 0x4170
 	sethi $r14 0x400000
 	or $r15 0x10
-	call nv_wr32
+	call #nv_wr32
 	ret
 
 // Waits for a ctx_4170s() call to complete
@@ -511,9 +556,9 @@
 ctx_4170w:
 	mov $r14 0x4170
 	sethi $r14 0x400000
-	call nv_rd32
+	call #nv_rd32
 	and $r15 0x10
-	bra ne ctx_4170w
+	bra ne #ctx_4170w
 	ret
 
 // Disables various things, waits a bit, and re-enables them..
@@ -530,7 +575,7 @@
 	mov $r15 8
 	ctx_redswitch_delay:
 		sub b32 $r15 1
-		bra ne ctx_redswitch_delay
+		bra ne #ctx_redswitch_delay
 	mov $r15 0x770
 	iowr I[$r14] $r15	// HUB_RED_SWITCH = ENABLE_ALL, POWER_ALL
 	ret
@@ -546,10 +591,10 @@
 	iowr I[$r14] $r15	// HUB(0x86c) = val
 	mov $r14 -0x75ec
 	sethi $r14 0x400000
-	call nv_wr32		// ROP(0xa14) = val
+	call #nv_wr32		// ROP(0xa14) = val
 	mov $r14 -0x5794
 	sethi $r14 0x410000
-	call nv_wr32		// GPC(0x86c) = val
+	call #nv_wr32		// GPC(0x86c) = val
 	ret
 
 // ctx_load - load's a channel's ctxctl data, and selects its vm
@@ -561,7 +606,7 @@
 
 	// switch to channel, somewhat magic in parts..
 	mov $r10 12		// DONE_UNK12
-	call wait_donez
+	call #wait_donez
 	mov $r1 0xa24
 	shl b32 $r1 6
 	iowr I[$r1 + 0x000] $r0	// 0x409a24
@@ -576,7 +621,7 @@
 	ctx_chan_wait_0:
 		iord $r4 I[$r1 + 0x100]
 		and $r4 0x1f
-		bra ne ctx_chan_wait_0
+		bra ne #ctx_chan_wait_0
 	iowr I[$r3 + 0x000] $r2	// CHAN_CUR
 
 	// load channel header, fetch PGRAPH context pointer
@@ -595,19 +640,19 @@
 	sethi $r2 0x80000000
 	iowr I[$r1 + 0x000] $r2		// MEM_TARGET = vram
 	mov $r1 0x10			// chan + 0x0210
-	mov $r2 xfer_data
+	mov $r2 #xfer_data
 	sethi $r2 0x00020000		// 16 bytes
 	xdld $r1 $r2
 	xdwait
 	trace_clr(T_LCHAN)
 
 	// update current context
-	ld b32 $r1 D[$r0 + xfer_data + 4]
+	ld b32 $r1 D[$r0 + #xfer_data + 4]
 	shl b32 $r1 24
-	ld b32 $r2 D[$r0 + xfer_data + 0]
+	ld b32 $r2 D[$r0 + #xfer_data + 0]
 	shr b32 $r2 8
 	or $r1 $r2
-	st b32 D[$r0 + ctx_current] $r1
+	st b32 D[$r0 + #ctx_current] $r1
 
 	// set transfer base to start of context, and fetch context header
 	trace_set(T_LCTXH)
@@ -618,7 +663,7 @@
 	mov $r1 0xa20
 	shl b32 $r1 6
 	iowr I[$r1 + 0x000] $r2		// MEM_TARGET = vm
-	mov $r1 chan_data
+	mov $r1 #chan_data
 	sethi $r1 0x00060000		// 256 bytes
 	xdld $r0 $r1
 	xdwait
@@ -635,10 +680,10 @@
 // In: $r2 channel address
 //
 ctx_chan:
-	call ctx_4160s
-	call ctx_load
+	call #ctx_4160s
+	call #ctx_load
 	mov $r10 12			// DONE_UNK12
-	call wait_donez
+	call #wait_donez
 	mov $r1 0xa10
 	shl b32 $r1 6
 	mov $r2 5
@@ -646,8 +691,8 @@
 	ctx_chan_wait:
 		iord $r2 I[$r1 + 0x000]
 		or $r2 $r2
-		bra ne ctx_chan_wait
-	call ctx_4160c
+		bra ne #ctx_chan_wait
+	call #ctx_4160c
 	ret
 
 // Execute per-context state overrides list
@@ -661,7 +706,7 @@
 //
 ctx_mmio_exec:
 	// set transfer base to be the mmio list
-	ld b32 $r3 D[$r0 + chan_mmio_address]
+	ld b32 $r3 D[$r0 + #chan_mmio_address]
 	mov $r2 0xa04
 	shl b32 $r2 6
 	iowr I[$r2 + 0x000] $r3		// MEM_BASE
@@ -670,31 +715,31 @@
 	ctx_mmio_loop:
 		// fetch next 256 bytes of mmio list if necessary
 		and $r4 $r3 0xff
-		bra ne ctx_mmio_pull
-			mov $r5 xfer_data
+		bra ne #ctx_mmio_pull
+			mov $r5 #xfer_data
 			sethi $r5 0x00060000	// 256 bytes
 			xdld $r3 $r5
 			xdwait
 
 		// execute a single list entry
 		ctx_mmio_pull:
-		ld b32 $r14 D[$r4 + xfer_data + 0x00]
-		ld b32 $r15 D[$r4 + xfer_data + 0x04]
-		call nv_wr32
+		ld b32 $r14 D[$r4 + #xfer_data + 0x00]
+		ld b32 $r15 D[$r4 + #xfer_data + 0x04]
+		call #nv_wr32
 
 		// next!
 		add b32 $r3 8
 		sub b32 $r1 1
-		bra ne ctx_mmio_loop
+		bra ne #ctx_mmio_loop
 
 	// set transfer base back to the current context
 	ctx_mmio_done:
-	ld b32 $r3 D[$r0 + ctx_current]
+	ld b32 $r3 D[$r0 + #ctx_current]
 	iowr I[$r2 + 0x000] $r3		// MEM_BASE
 
 	// disable the mmio list now, we don't need/want to execute it again
-	st b32 D[$r0 + chan_mmio_count] $r0
-	mov $r1 chan_data
+	st b32 D[$r0 + #chan_mmio_count] $r0
+	mov $r1 #chan_data
 	sethi $r1 0x00060000		// 256 bytes
 	xdst $r0 $r1
 	xdwait
@@ -709,46 +754,46 @@
 //		on load it means: "a save preceeded this load"
 //
 ctx_xfer:
-	bra not $p1 ctx_xfer_pre
-	bra $p2 ctx_xfer_pre_load
+	bra not $p1 #ctx_xfer_pre
+	bra $p2 #ctx_xfer_pre_load
 	ctx_xfer_pre:
 		mov $r15 0x10
-		call ctx_86c
-		call ctx_4160s
-		bra not $p1 ctx_xfer_exec
+		call #ctx_86c
+		call #ctx_4160s
+		bra not $p1 #ctx_xfer_exec
 
 	ctx_xfer_pre_load:
 		mov $r15 2
-		call ctx_4170s
-		call ctx_4170w
-		call ctx_redswitch
+		call #ctx_4170s
+		call #ctx_4170w
+		call #ctx_redswitch
 		clear b32 $r15
-		call ctx_4170s
-		call ctx_load
+		call #ctx_4170s
+		call #ctx_load
 
 	// fetch context pointer, and initiate xfer on all GPCs
 	ctx_xfer_exec:
-	ld b32 $r1 D[$r0 + ctx_current]
+	ld b32 $r1 D[$r0 + #ctx_current]
 	mov $r2 0x414
 	shl b32 $r2 6
 	iowr I[$r2 + 0x000] $r0	// BAR_STATUS = reset
 	mov $r14 -0x5b00
 	sethi $r14 0x410000
 	mov b32 $r15 $r1
-	call nv_wr32		// GPC_BCAST_WRCMD_DATA = ctx pointer
+	call #nv_wr32		// GPC_BCAST_WRCMD_DATA = ctx pointer
 	add b32 $r14 4
 	xbit $r15 $flags $p1
 	xbit $r2 $flags $p2
 	shl b32 $r2 1
 	or $r15 $r2
-	call nv_wr32		// GPC_BCAST_WRCMD_CMD = GPC_XFER(type)
+	call #nv_wr32		// GPC_BCAST_WRCMD_CMD = GPC_XFER(type)
 
 	// strands
 	mov $r1 0x4afc
 	sethi $r1 0x20000
 	mov $r2 0xc
 	iowr I[$r1] $r2		// STRAND_CMD(0x3f) = 0x0c
-	call strand_wait
+	call #strand_wait
 	mov $r2 0x47fc
 	sethi $r2 0x20000
 	iowr I[$r2] $r0		// STRAND_FIRST_GENE(0x3f) = 0x00
@@ -760,22 +805,22 @@
 	xbit $r10 $flags $p1	// direction
 	or $r10 6		// first, last
 	mov $r11 0		// base = 0
-	ld b32 $r12 D[$r0 + hub_mmio_list_head]
-	ld b32 $r13 D[$r0 + hub_mmio_list_tail]
+	ld b32 $r12 D[$r0 + #hub_mmio_list_head]
+	ld b32 $r13 D[$r0 + #hub_mmio_list_tail]
 	mov $r14 0		// not multi
-	call mmctx_xfer
+	call #mmctx_xfer
 
 	// wait for GPCs to all complete
 	mov $r10 8		// DONE_BAR
-	call wait_doneo
+	call #wait_doneo
 
 	// wait for strand xfer to complete
-	call strand_wait
+	call #strand_wait
 
 	// post-op
-	bra $p1 ctx_xfer_post
+	bra $p1 #ctx_xfer_post
 		mov $r10 12		// DONE_UNK12
-		call wait_donez
+		call #wait_donez
 		mov $r1 0xa10
 		shl b32 $r1 6
 		mov $r2 5
@@ -783,27 +828,27 @@
 		ctx_xfer_post_save_wait:
 			iord $r2 I[$r1]
 			or $r2 $r2
-			bra ne ctx_xfer_post_save_wait
+			bra ne #ctx_xfer_post_save_wait
 
-	bra $p2 ctx_xfer_done
+	bra $p2 #ctx_xfer_done
 	ctx_xfer_post:
 		mov $r15 2
-		call ctx_4170s
+		call #ctx_4170s
 		clear b32 $r15
-		call ctx_86c
-		call strand_post
-		call ctx_4170w
+		call #ctx_86c
+		call #strand_post
+		call #ctx_4170w
 		clear b32 $r15
-		call ctx_4170s
+		call #ctx_4170s
 
-		bra not $p1 ctx_xfer_no_post_mmio
-		ld b32 $r1 D[$r0 + chan_mmio_count]
+		bra not $p1 #ctx_xfer_no_post_mmio
+		ld b32 $r1 D[$r0 + #chan_mmio_count]
 		or $r1 $r1
-		bra e ctx_xfer_no_post_mmio
-			call ctx_mmio_exec
+		bra e #ctx_xfer_no_post_mmio
+			call #ctx_mmio_exec
 
 		ctx_xfer_no_post_mmio:
-		call ctx_4160c
+		call #ctx_4160c
 
 	ctx_xfer_done:
 	ret
diff --git a/drivers/gpu/drm/nouveau/nvc0_grhub.fuc.h b/drivers/gpu/drm/nouveau/nvc0_grhub.fuc.h
index 241d326..c5ed307 100644
--- a/drivers/gpu/drm/nouveau/nvc0_grhub.fuc.h
+++ b/drivers/gpu/drm/nouveau/nvc0_grhub.fuc.h
@@ -23,19 +23,21 @@
 	0x00000000,
 	0x00000000,
 	0x000000c0,
-	0x01340098,
+	0x013c00a0,
 	0x000000c1,
-	0x01380098,
+	0x014000a0,
 	0x000000c3,
-	0x01340098,
+	0x013c00a0,
 	0x000000c4,
-	0x01340098,
+	0x013c00a0,
 	0x000000c8,
-	0x01340098,
+	0x013c00a0,
 	0x000000ce,
-	0x01340098,
+	0x013c00a0,
 	0x000000cf,
-	0x01340098,
+	0x013c00a0,
+	0x000000d9,
+	0x01dc0140,
 	0x00000000,
 	0x0417e91c,
 	0x04400204,
@@ -77,47 +79,45 @@
 	0x0c408900,
 	0x00408980,
 	0x044064c0,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
-	0x00000000,
+	0x0417e91c,
+	0x04400204,
+	0x24404004,
+	0x00404044,
+	0x34404094,
+	0x184040d0,
+	0x004040f8,
+	0x08404130,
+	0x08404150,
+	0x04404164,
+	0x04404178,
+	0x1c404200,
+	0x34404404,
+	0x0c404460,
+	0x00404480,
+	0x00404498,
+	0x0c404604,
+	0x7c404618,
+	0x50404698,
+	0x044046f0,
+	0x54404700,
+	0x00405800,
+	0x08405830,
+	0x00405854,
+	0x0c405870,
+	0x04405a00,
+	0x00405a18,
+	0x00406020,
+	0x0c406028,
+	0x044064a8,
+	0x104064b4,
+	0x00407804,
+	0x1440780c,
+	0x004078bc,
+	0x18408000,
+	0x00408064,
+	0x08408800,
+	0x0c408900,
+	0x00408980,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/nvc0_pm.c b/drivers/gpu/drm/nouveau/nvc0_pm.c
index 929aded..e9992f6 100644
--- a/drivers/gpu/drm/nouveau/nvc0_pm.c
+++ b/drivers/gpu/drm/nouveau/nvc0_pm.c
@@ -153,3 +153,240 @@
 	perflvl->vdec   = read_clk(dev, 0x0e);
 	return 0;
 }
+
+struct nvc0_pm_clock {
+	u32 freq;
+	u32 ssel;
+	u32 mdiv;
+	u32 dsrc;
+	u32 ddiv;
+	u32 coef;
+};
+
+struct nvc0_pm_state {
+	struct nvc0_pm_clock eng[16];
+};
+
+static u32
+calc_div(struct drm_device *dev, int clk, u32 ref, u32 freq, u32 *ddiv)
+{
+	u32 div = min((ref * 2) / freq, (u32)65);
+	if (div < 2)
+		div = 2;
+
+	*ddiv = div - 2;
+	return (ref * 2) / div;
+}
+
+static u32
+calc_src(struct drm_device *dev, int clk, u32 freq, u32 *dsrc, u32 *ddiv)
+{
+	u32 sclk;
+
+	/* use one of the fixed frequencies if possible */
+	*ddiv = 0x00000000;
+	switch (freq) {
+	case  27000:
+	case 108000:
+		*dsrc = 0x00000000;
+		if (freq == 108000)
+			*dsrc |= 0x00030000;
+		return freq;
+	case 100000:
+		*dsrc = 0x00000002;
+		return freq;
+	default:
+		*dsrc = 0x00000003;
+		break;
+	}
+
+	/* otherwise, calculate the closest divider */
+	sclk = read_vco(dev, clk);
+	if (clk < 7)
+		sclk = calc_div(dev, clk, sclk, freq, ddiv);
+	return sclk;
+}
+
+static u32
+calc_pll(struct drm_device *dev, int clk, u32 freq, u32 *coef)
+{
+	struct pll_lims limits;
+	int N, M, P, ret;
+
+	ret = get_pll_limits(dev, 0x137000 + (clk * 0x20), &limits);
+	if (ret)
+		return 0;
+
+	limits.refclk = read_div(dev, clk, 0x137120, 0x137140);
+	if (!limits.refclk)
+		return 0;
+
+	ret = nva3_calc_pll(dev, &limits, freq, &N, NULL, &M, &P);
+	if (ret <= 0)
+		return 0;
+
+	*coef = (P << 16) | (N << 8) | M;
+	return ret;
+}
+
+/* A (likely rather simplified and incomplete) view of the clock tree
+ *
+ * Key:
+ *
+ * S: source select
+ * D: divider
+ * P: pll
+ * F: switch
+ *
+ * Engine clocks:
+ *
+ * 137250(D) ---- 137100(F0) ---- 137160(S)/1371d0(D) ------------------- ref
+ *                      (F1) ---- 1370X0(P) ---- 137120(S)/137140(D) ---- ref
+ *
+ * Not all registers exist for all clocks.  For example: clocks >= 8 don't
+ * have their own PLL (all tied to clock 7's PLL when in PLL mode), nor do
+ * they have the divider at 1371d0, though the source selection at 137160
+ * still exists.  You must use the divider at 137250 for these instead.
+ *
+ * Memory clock:
+ *
+ * TBD, read_mem() above is likely very wrong...
+ *
+ */
+
+static int
+calc_clk(struct drm_device *dev, int clk, struct nvc0_pm_clock *info, u32 freq)
+{
+	u32 src0, div0, div1D, div1P = 0;
+	u32 clk0, clk1 = 0;
+
+	/* invalid clock domain */
+	if (!freq)
+		return 0;
+
+	/* first possible path, using only dividers */
+	clk0 = calc_src(dev, clk, freq, &src0, &div0);
+	clk0 = calc_div(dev, clk, clk0, freq, &div1D);
+
+	/* see if we can get any closer using PLLs */
+	if (clk0 != freq) {
+		if (clk < 7)
+			clk1 = calc_pll(dev, clk, freq, &info->coef);
+		else
+			clk1 = read_pll(dev, 0x1370e0);
+		clk1 = calc_div(dev, clk, clk1, freq, &div1P);
+	}
+
+	/* select the method which gets closest to target freq */
+	if (abs((int)freq - clk0) <= abs((int)freq - clk1)) {
+		info->dsrc = src0;
+		if (div0) {
+			info->ddiv |= 0x80000000;
+			info->ddiv |= div0 << 8;
+			info->ddiv |= div0;
+		}
+		if (div1D) {
+			info->mdiv |= 0x80000000;
+			info->mdiv |= div1D;
+		}
+		info->ssel = 0;
+		info->freq = clk0;
+	} else {
+		if (div1P) {
+			info->mdiv |= 0x80000000;
+			info->mdiv |= div1P << 8;
+		}
+		info->ssel = (1 << clk);
+		info->freq = clk1;
+	}
+
+	return 0;
+}
+
+void *
+nvc0_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
+{
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
+	struct nvc0_pm_state *info;
+	int ret;
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (!info)
+		return ERR_PTR(-ENOMEM);
+
+	/* NFI why this is still in the performance table, the ROPCs appear
+	 * to get their clock from clock 2 ("hub07", actually hub05 on this
+	 * chip, but, anyway...) as well.  nvatiming confirms hub05 and ROP
+	 * are always the same freq with the binary driver even when the
+	 * performance table says they should differ.
+	 */
+	if (dev_priv->chipset == 0xd9)
+		perflvl->rop = 0;
+
+	if ((ret = calc_clk(dev, 0x00, &info->eng[0x00], perflvl->shader)) ||
+	    (ret = calc_clk(dev, 0x01, &info->eng[0x01], perflvl->rop)) ||
+	    (ret = calc_clk(dev, 0x02, &info->eng[0x02], perflvl->hub07)) ||
+	    (ret = calc_clk(dev, 0x07, &info->eng[0x07], perflvl->hub06)) ||
+	    (ret = calc_clk(dev, 0x08, &info->eng[0x08], perflvl->hub01)) ||
+	    (ret = calc_clk(dev, 0x09, &info->eng[0x09], perflvl->copy)) ||
+	    (ret = calc_clk(dev, 0x0c, &info->eng[0x0c], perflvl->daemon)) ||
+	    (ret = calc_clk(dev, 0x0e, &info->eng[0x0e], perflvl->vdec))) {
+		kfree(info);
+		return ERR_PTR(ret);
+	}
+
+	return info;
+}
+
+static void
+prog_clk(struct drm_device *dev, int clk, struct nvc0_pm_clock *info)
+{
+	/* program dividers at 137160/1371d0 first */
+	if (clk < 7 && !info->ssel) {
+		nv_mask(dev, 0x1371d0 + (clk * 0x04), 0x80003f3f, info->ddiv);
+		nv_wr32(dev, 0x137160 + (clk * 0x04), info->dsrc);
+	}
+
+	/* switch clock to non-pll mode */
+	nv_mask(dev, 0x137100, (1 << clk), 0x00000000);
+	nv_wait(dev, 0x137100, (1 << clk), 0x00000000);
+
+	/* reprogram pll */
+	if (clk < 7) {
+		/* make sure it's disabled first... */
+		u32 base = 0x137000 + (clk * 0x20);
+		u32 ctrl = nv_rd32(dev, base + 0x00);
+		if (ctrl & 0x00000001) {
+			nv_mask(dev, base + 0x00, 0x00000004, 0x00000000);
+			nv_mask(dev, base + 0x00, 0x00000001, 0x00000000);
+		}
+		/* program it to new values, if necessary */
+		if (info->ssel) {
+			nv_wr32(dev, base + 0x04, info->coef);
+			nv_mask(dev, base + 0x00, 0x00000001, 0x00000001);
+			nv_wait(dev, base + 0x00, 0x00020000, 0x00020000);
+			nv_mask(dev, base + 0x00, 0x00020004, 0x00000004);
+		}
+	}
+
+	/* select pll/non-pll mode, and program final clock divider */
+	nv_mask(dev, 0x137100, (1 << clk), info->ssel);
+	nv_wait(dev, 0x137100, (1 << clk), info->ssel);
+	nv_mask(dev, 0x137250 + (clk * 0x04), 0x00003f3f, info->mdiv);
+}
+
+int
+nvc0_pm_clocks_set(struct drm_device *dev, void *data)
+{
+	struct nvc0_pm_state *info = data;
+	int i;
+
+	for (i = 0; i < 16; i++) {
+		if (!info->eng[i].freq)
+			continue;
+		prog_clk(dev, i, &info->eng[i]);
+	}
+
+	kfree(info);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvc0_vram.c b/drivers/gpu/drm/nouveau/nvc0_vram.c
index edbfe93..ce984d5 100644
--- a/drivers/gpu/drm/nouveau/nvc0_vram.c
+++ b/drivers/gpu/drm/nouveau/nvc0_vram.c
@@ -43,7 +43,7 @@
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 	0, 0, 0, 3, 3, 3, 3, 1, 1, 1, 1, 0, 0, 0, 0, 0,
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3,
-	3, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3,
+	3, 3, 3, 1, 0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3,
 	3, 3, 0, 0, 0, 0, 0, 0, 3, 0, 0, 3, 0, 3, 0, 3,
 	3, 0, 3, 3, 3, 3, 3, 0, 0, 3, 0, 3, 0, 3, 3, 0,
 	3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 0, 1, 1, 0
@@ -110,22 +110,26 @@
 	u32 bsize = nv_rd32(dev, 0x10f20c);
 	u32 offset, length;
 	bool uniform = true;
-	int ret, i;
+	int ret, part;
 
 	NV_DEBUG(dev, "0x100800: 0x%08x\n", nv_rd32(dev, 0x100800));
 	NV_DEBUG(dev, "parts 0x%08x bcast_mem_amount 0x%08x\n", parts, bsize);
 
 	/* read amount of vram attached to each memory controller */
-	for (i = 0; i < parts; i++) {
-		u32 psize = nv_rd32(dev, 0x11020c + (i * 0x1000));
+	part = 0;
+	while (parts) {
+		u32 psize = nv_rd32(dev, 0x11020c + (part++ * 0x1000));
+		if (psize == 0)
+			continue;
+		parts--;
+
 		if (psize != bsize) {
 			if (psize < bsize)
 				bsize = psize;
 			uniform = false;
 		}
 
-		NV_DEBUG(dev, "%d: mem_amount 0x%08x\n", i, psize);
-
+		NV_DEBUG(dev, "%d: mem_amount 0x%08x\n", part, psize);
 		dev_priv->vram_size += (u64)psize << 20;
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nvd0_display.c b/drivers/gpu/drm/nouveau/nvd0_display.c
index 23d63b4..d2ba2f0 100644
--- a/drivers/gpu/drm/nouveau/nvd0_display.c
+++ b/drivers/gpu/drm/nouveau/nvd0_display.c
@@ -35,12 +35,34 @@
 #include "nouveau_fb.h"
 #include "nv50_display.h"
 
+#define EVO_DMA_NR 9
+
+#define EVO_MASTER  (0x00)
+#define EVO_FLIP(c) (0x01 + (c))
+#define EVO_OVLY(c) (0x05 + (c))
+#define EVO_OIMM(c) (0x09 + (c))
+#define EVO_CURS(c) (0x0d + (c))
+
+/* offsets in shared sync bo of various structures */
+#define EVO_SYNC(c, o) ((c) * 0x0100 + (o))
+#define EVO_MAST_NTFY     EVO_SYNC(  0, 0x00)
+#define EVO_FLIP_SEM0(c)  EVO_SYNC((c), 0x00)
+#define EVO_FLIP_SEM1(c)  EVO_SYNC((c), 0x10)
+
+struct evo {
+	int idx;
+	dma_addr_t handle;
+	u32 *ptr;
+	struct {
+		u32 offset;
+		u16 value;
+	} sem;
+};
+
 struct nvd0_display {
 	struct nouveau_gpuobj *mem;
-	struct {
-		dma_addr_t handle;
-		u32 *ptr;
-	} evo[1];
+	struct nouveau_bo *sync;
+	struct evo evo[9];
 
 	struct tasklet_struct tasklet;
 	u32 modeset;
@@ -53,6 +75,15 @@
 	return dev_priv->engine.display.priv;
 }
 
+static struct drm_crtc *
+nvd0_display_crtc_get(struct drm_encoder *encoder)
+{
+	return nouveau_encoder(encoder)->crtc;
+}
+
+/******************************************************************************
+ * EVO channel helpers
+ *****************************************************************************/
 static inline int
 evo_icmd(struct drm_device *dev, int id, u32 mthd, u32 data)
 {
@@ -84,6 +115,9 @@
 		put = 0;
 	}
 
+	if (nouveau_reg_debug & NOUVEAU_REG_DEBUG_EVO)
+		NV_INFO(dev, "Evo%d: %p START\n", id, disp->evo[id].ptr + put);
+
 	return disp->evo[id].ptr + put;
 }
 
@@ -91,40 +125,264 @@
 evo_kick(u32 *push, struct drm_device *dev, int id)
 {
 	struct nvd0_display *disp = nvd0_display(dev);
+
+	if (nouveau_reg_debug & NOUVEAU_REG_DEBUG_EVO) {
+		u32 curp = nv_rd32(dev, 0x640000 + (id * 0x1000)) >> 2;
+		u32 *cur = disp->evo[id].ptr + curp;
+
+		while (cur < push)
+			NV_INFO(dev, "Evo%d: 0x%08x\n", id, *cur++);
+		NV_INFO(dev, "Evo%d: %p KICK!\n", id, push);
+	}
+
 	nv_wr32(dev, 0x640000 + (id * 0x1000), (push - disp->evo[id].ptr) << 2);
 }
 
 #define evo_mthd(p,m,s) *((p)++) = (((s) << 18) | (m))
 #define evo_data(p,d)   *((p)++) = (d)
 
-static struct drm_crtc *
-nvd0_display_crtc_get(struct drm_encoder *encoder)
+static int
+evo_init_dma(struct drm_device *dev, int ch)
 {
-	return nouveau_encoder(encoder)->crtc;
+	struct nvd0_display *disp = nvd0_display(dev);
+	u32 flags;
+
+	flags = 0x00000000;
+	if (ch == EVO_MASTER)
+		flags |= 0x01000000;
+
+	nv_wr32(dev, 0x610494 + (ch * 0x0010), (disp->evo[ch].handle >> 8) | 3);
+	nv_wr32(dev, 0x610498 + (ch * 0x0010), 0x00010000);
+	nv_wr32(dev, 0x61049c + (ch * 0x0010), 0x00000001);
+	nv_mask(dev, 0x610490 + (ch * 0x0010), 0x00000010, 0x00000010);
+	nv_wr32(dev, 0x640000 + (ch * 0x1000), 0x00000000);
+	nv_wr32(dev, 0x610490 + (ch * 0x0010), 0x00000013 | flags);
+	if (!nv_wait(dev, 0x610490 + (ch * 0x0010), 0x80000000, 0x00000000)) {
+		NV_ERROR(dev, "PDISP: ch%d 0x%08x\n", ch,
+			      nv_rd32(dev, 0x610490 + (ch * 0x0010)));
+		return -EBUSY;
+	}
+
+	nv_mask(dev, 0x610090, (1 << ch), (1 << ch));
+	nv_mask(dev, 0x6100a0, (1 << ch), (1 << ch));
+	return 0;
+}
+
+static void
+evo_fini_dma(struct drm_device *dev, int ch)
+{
+	if (!(nv_rd32(dev, 0x610490 + (ch * 0x0010)) & 0x00000010))
+		return;
+
+	nv_mask(dev, 0x610490 + (ch * 0x0010), 0x00000010, 0x00000000);
+	nv_mask(dev, 0x610490 + (ch * 0x0010), 0x00000003, 0x00000000);
+	nv_wait(dev, 0x610490 + (ch * 0x0010), 0x80000000, 0x00000000);
+	nv_mask(dev, 0x610090, (1 << ch), 0x00000000);
+	nv_mask(dev, 0x6100a0, (1 << ch), 0x00000000);
+}
+
+static inline void
+evo_piow(struct drm_device *dev, int ch, u16 mthd, u32 data)
+{
+	nv_wr32(dev, 0x640000 + (ch * 0x1000) + mthd, data);
+}
+
+static int
+evo_init_pio(struct drm_device *dev, int ch)
+{
+	nv_wr32(dev, 0x610490 + (ch * 0x0010), 0x00000001);
+	if (!nv_wait(dev, 0x610490 + (ch * 0x0010), 0x00010000, 0x00010000)) {
+		NV_ERROR(dev, "PDISP: ch%d 0x%08x\n", ch,
+			      nv_rd32(dev, 0x610490 + (ch * 0x0010)));
+		return -EBUSY;
+	}
+
+	nv_mask(dev, 0x610090, (1 << ch), (1 << ch));
+	nv_mask(dev, 0x6100a0, (1 << ch), (1 << ch));
+	return 0;
+}
+
+static void
+evo_fini_pio(struct drm_device *dev, int ch)
+{
+	if (!(nv_rd32(dev, 0x610490 + (ch * 0x0010)) & 0x00000001))
+		return;
+
+	nv_mask(dev, 0x610490 + (ch * 0x0010), 0x00000010, 0x00000010);
+	nv_mask(dev, 0x610490 + (ch * 0x0010), 0x00000001, 0x00000000);
+	nv_wait(dev, 0x610490 + (ch * 0x0010), 0x00010000, 0x00000000);
+	nv_mask(dev, 0x610090, (1 << ch), 0x00000000);
+	nv_mask(dev, 0x6100a0, (1 << ch), 0x00000000);
+}
+
+static bool
+evo_sync_wait(void *data)
+{
+	return nouveau_bo_rd32(data, EVO_MAST_NTFY) != 0x00000000;
+}
+
+static int
+evo_sync(struct drm_device *dev, int ch)
+{
+	struct nvd0_display *disp = nvd0_display(dev);
+	u32 *push = evo_wait(dev, ch, 8);
+	if (push) {
+		nouveau_bo_wr32(disp->sync, EVO_MAST_NTFY, 0x00000000);
+		evo_mthd(push, 0x0084, 1);
+		evo_data(push, 0x80000000 | EVO_MAST_NTFY);
+		evo_mthd(push, 0x0080, 2);
+		evo_data(push, 0x00000000);
+		evo_data(push, 0x00000000);
+		evo_kick(push, dev, ch);
+		if (nv_wait_cb(dev, evo_sync_wait, disp->sync))
+			return 0;
+	}
+
+	return -EBUSY;
+}
+
+/******************************************************************************
+ * Page flipping channel
+ *****************************************************************************/
+struct nouveau_bo *
+nvd0_display_crtc_sema(struct drm_device *dev, int crtc)
+{
+	return nvd0_display(dev)->sync;
+}
+
+void
+nvd0_display_flip_stop(struct drm_crtc *crtc)
+{
+	struct nvd0_display *disp = nvd0_display(crtc->dev);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+	struct evo *evo = &disp->evo[EVO_FLIP(nv_crtc->index)];
+	u32 *push;
+
+	push = evo_wait(crtc->dev, evo->idx, 8);
+	if (push) {
+		evo_mthd(push, 0x0084, 1);
+		evo_data(push, 0x00000000);
+		evo_mthd(push, 0x0094, 1);
+		evo_data(push, 0x00000000);
+		evo_mthd(push, 0x00c0, 1);
+		evo_data(push, 0x00000000);
+		evo_mthd(push, 0x0080, 1);
+		evo_data(push, 0x00000000);
+		evo_kick(push, crtc->dev, evo->idx);
+	}
+}
+
+int
+nvd0_display_flip_next(struct drm_crtc *crtc, struct drm_framebuffer *fb,
+		       struct nouveau_channel *chan, u32 swap_interval)
+{
+	struct nouveau_framebuffer *nv_fb = nouveau_framebuffer(fb);
+	struct nvd0_display *disp = nvd0_display(crtc->dev);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+	struct evo *evo = &disp->evo[EVO_FLIP(nv_crtc->index)];
+	u64 offset;
+	u32 *push;
+	int ret;
+
+	swap_interval <<= 4;
+	if (swap_interval == 0)
+		swap_interval |= 0x100;
+
+	push = evo_wait(crtc->dev, evo->idx, 128);
+	if (unlikely(push == NULL))
+		return -EBUSY;
+
+	/* synchronise with the rendering channel, if necessary */
+	if (likely(chan)) {
+		ret = RING_SPACE(chan, 10);
+		if (ret)
+			return ret;
+
+		offset  = chan->dispc_vma[nv_crtc->index].offset;
+		offset += evo->sem.offset;
+
+		BEGIN_NVC0(chan, 2, NvSubM2MF, 0x0010, 4);
+		OUT_RING  (chan, upper_32_bits(offset));
+		OUT_RING  (chan, lower_32_bits(offset));
+		OUT_RING  (chan, 0xf00d0000 | evo->sem.value);
+		OUT_RING  (chan, 0x1002);
+		BEGIN_NVC0(chan, 2, NvSubM2MF, 0x0010, 4);
+		OUT_RING  (chan, upper_32_bits(offset));
+		OUT_RING  (chan, lower_32_bits(offset ^ 0x10));
+		OUT_RING  (chan, 0x74b1e000);
+		OUT_RING  (chan, 0x1001);
+		FIRE_RING (chan);
+	} else {
+		nouveau_bo_wr32(disp->sync, evo->sem.offset / 4,
+				0xf00d0000 | evo->sem.value);
+		evo_sync(crtc->dev, EVO_MASTER);
+	}
+
+	/* queue the flip */
+	evo_mthd(push, 0x0100, 1);
+	evo_data(push, 0xfffe0000);
+	evo_mthd(push, 0x0084, 1);
+	evo_data(push, swap_interval);
+	if (!(swap_interval & 0x00000100)) {
+		evo_mthd(push, 0x00e0, 1);
+		evo_data(push, 0x40000000);
+	}
+	evo_mthd(push, 0x0088, 4);
+	evo_data(push, evo->sem.offset);
+	evo_data(push, 0xf00d0000 | evo->sem.value);
+	evo_data(push, 0x74b1e000);
+	evo_data(push, NvEvoSync);
+	evo_mthd(push, 0x00a0, 2);
+	evo_data(push, 0x00000000);
+	evo_data(push, 0x00000000);
+	evo_mthd(push, 0x00c0, 1);
+	evo_data(push, nv_fb->r_dma);
+	evo_mthd(push, 0x0110, 2);
+	evo_data(push, 0x00000000);
+	evo_data(push, 0x00000000);
+	evo_mthd(push, 0x0400, 5);
+	evo_data(push, nv_fb->nvbo->bo.offset >> 8);
+	evo_data(push, 0);
+	evo_data(push, (fb->height << 16) | fb->width);
+	evo_data(push, nv_fb->r_pitch);
+	evo_data(push, nv_fb->r_format);
+	evo_mthd(push, 0x0080, 1);
+	evo_data(push, 0x00000000);
+	evo_kick(push, crtc->dev, evo->idx);
+
+	evo->sem.offset ^= 0x10;
+	evo->sem.value++;
+	return 0;
 }
 
 /******************************************************************************
  * CRTC
  *****************************************************************************/
 static int
-nvd0_crtc_set_dither(struct nouveau_crtc *nv_crtc, bool on, bool update)
+nvd0_crtc_set_dither(struct nouveau_crtc *nv_crtc, bool update)
 {
 	struct drm_device *dev = nv_crtc->base.dev;
-	u32 *push, mode;
+	struct nouveau_connector *nv_connector;
+	struct drm_connector *connector;
+	u32 *push, mode = 0x00;
 
-	mode = 0x00000000;
-	if (on) {
-		/* 0x11: 6bpc dynamic 2x2
-		 * 0x13: 8bpc dynamic 2x2
-		 * 0x19: 6bpc static 2x2
-		 * 0x1b: 8bpc static 2x2
-		 * 0x21: 6bpc temporal
-		 * 0x23: 8bpc temporal
-		 */
-		mode = 0x00000011;
+	nv_connector = nouveau_crtc_connector_get(nv_crtc);
+	connector = &nv_connector->base;
+	if (nv_connector->dithering_mode == DITHERING_MODE_AUTO) {
+		if (nv_crtc->base.fb->depth > connector->display_info.bpc * 3)
+			mode = DITHERING_MODE_DYNAMIC2X2;
+	} else {
+		mode = nv_connector->dithering_mode;
 	}
 
-	push = evo_wait(dev, 0, 4);
+	if (nv_connector->dithering_depth == DITHERING_DEPTH_AUTO) {
+		if (connector->display_info.bpc >= 8)
+			mode |= DITHERING_DEPTH_8BPC;
+	} else {
+		mode |= nv_connector->dithering_depth;
+	}
+
+	push = evo_wait(dev, EVO_MASTER, 4);
 	if (push) {
 		evo_mthd(push, 0x0490 + (nv_crtc->index * 0x300), 1);
 		evo_data(push, mode);
@@ -132,63 +390,98 @@
 			evo_mthd(push, 0x0080, 1);
 			evo_data(push, 0x00000000);
 		}
-		evo_kick(push, dev, 0);
+		evo_kick(push, dev, EVO_MASTER);
 	}
 
 	return 0;
 }
 
 static int
-nvd0_crtc_set_scale(struct nouveau_crtc *nv_crtc, int type, bool update)
+nvd0_crtc_set_scale(struct nouveau_crtc *nv_crtc, bool update)
 {
-	struct drm_display_mode *mode = &nv_crtc->base.mode;
+	struct drm_display_mode *omode, *umode = &nv_crtc->base.mode;
 	struct drm_device *dev = nv_crtc->base.dev;
+	struct drm_crtc *crtc = &nv_crtc->base;
 	struct nouveau_connector *nv_connector;
-	u32 *push, outX, outY;
+	int mode = DRM_MODE_SCALE_NONE;
+	u32 oX, oY, *push;
 
-	outX = mode->hdisplay;
-	outY = mode->vdisplay;
-
+	/* start off at the resolution we programmed the crtc for, this
+	 * effectively handles NONE/FULL scaling
+	 */
 	nv_connector = nouveau_crtc_connector_get(nv_crtc);
-	if (nv_connector && nv_connector->native_mode) {
-		struct drm_display_mode *native = nv_connector->native_mode;
-		u32 xratio = (native->hdisplay << 19) / mode->hdisplay;
-		u32 yratio = (native->vdisplay << 19) / mode->vdisplay;
+	if (nv_connector && nv_connector->native_mode)
+		mode = nv_connector->scaling_mode;
 
-		switch (type) {
-		case DRM_MODE_SCALE_ASPECT:
-			if (xratio > yratio) {
-				outX = (mode->hdisplay * yratio) >> 19;
-				outY = (mode->vdisplay * yratio) >> 19;
-			} else {
-				outX = (mode->hdisplay * xratio) >> 19;
-				outY = (mode->vdisplay * xratio) >> 19;
-			}
-			break;
-		case DRM_MODE_SCALE_FULLSCREEN:
-			outX = native->hdisplay;
-			outY = native->vdisplay;
-			break;
-		default:
-			break;
+	if (mode != DRM_MODE_SCALE_NONE)
+		omode = nv_connector->native_mode;
+	else
+		omode = umode;
+
+	oX = omode->hdisplay;
+	oY = omode->vdisplay;
+	if (omode->flags & DRM_MODE_FLAG_DBLSCAN)
+		oY *= 2;
+
+	/* add overscan compensation if necessary, will keep the aspect
+	 * ratio the same as the backend mode unless overridden by the
+	 * user setting both hborder and vborder properties.
+	 */
+	if (nv_connector && ( nv_connector->underscan == UNDERSCAN_ON ||
+			     (nv_connector->underscan == UNDERSCAN_AUTO &&
+			      nv_connector->edid &&
+			      drm_detect_hdmi_monitor(nv_connector->edid)))) {
+		u32 bX = nv_connector->underscan_hborder;
+		u32 bY = nv_connector->underscan_vborder;
+		u32 aspect = (oY << 19) / oX;
+
+		if (bX) {
+			oX -= (bX * 2);
+			if (bY) oY -= (bY * 2);
+			else    oY  = ((oX * aspect) + (aspect / 2)) >> 19;
+		} else {
+			oX -= (oX >> 4) + 32;
+			if (bY) oY -= (bY * 2);
+			else    oY  = ((oX * aspect) + (aspect / 2)) >> 19;
 		}
 	}
 
-	push = evo_wait(dev, 0, 16);
+	/* handle CENTER/ASPECT scaling, taking into account the areas
+	 * removed already for overscan compensation
+	 */
+	switch (mode) {
+	case DRM_MODE_SCALE_CENTER:
+		oX = min((u32)umode->hdisplay, oX);
+		oY = min((u32)umode->vdisplay, oY);
+		/* fall-through */
+	case DRM_MODE_SCALE_ASPECT:
+		if (oY < oX) {
+			u32 aspect = (umode->hdisplay << 19) / umode->vdisplay;
+			oX = ((oY * aspect) + (aspect / 2)) >> 19;
+		} else {
+			u32 aspect = (umode->vdisplay << 19) / umode->hdisplay;
+			oY = ((oX * aspect) + (aspect / 2)) >> 19;
+		}
+		break;
+	default:
+		break;
+	}
+
+	push = evo_wait(dev, EVO_MASTER, 8);
 	if (push) {
 		evo_mthd(push, 0x04c0 + (nv_crtc->index * 0x300), 3);
-		evo_data(push, (outY << 16) | outX);
-		evo_data(push, (outY << 16) | outX);
-		evo_data(push, (outY << 16) | outX);
+		evo_data(push, (oY << 16) | oX);
+		evo_data(push, (oY << 16) | oX);
+		evo_data(push, (oY << 16) | oX);
 		evo_mthd(push, 0x0494 + (nv_crtc->index * 0x300), 1);
 		evo_data(push, 0x00000000);
 		evo_mthd(push, 0x04b8 + (nv_crtc->index * 0x300), 1);
-		evo_data(push, (mode->vdisplay << 16) | mode->hdisplay);
+		evo_data(push, (umode->vdisplay << 16) | umode->hdisplay);
+		evo_kick(push, dev, EVO_MASTER);
 		if (update) {
-			evo_mthd(push, 0x0080, 1);
-			evo_data(push, 0x00000000);
+			nvd0_display_flip_stop(crtc);
+			nvd0_display_flip_next(crtc, crtc->fb, NULL, 1);
 		}
-		evo_kick(push, dev, 0);
 	}
 
 	return 0;
@@ -201,7 +494,7 @@
 	struct nouveau_framebuffer *nvfb = nouveau_framebuffer(fb);
 	u32 *push;
 
-	push = evo_wait(fb->dev, 0, 16);
+	push = evo_wait(fb->dev, EVO_MASTER, 16);
 	if (push) {
 		evo_mthd(push, 0x0460 + (nv_crtc->index * 0x300), 1);
 		evo_data(push, nvfb->nvbo->bo.offset >> 8);
@@ -216,7 +509,7 @@
 			evo_mthd(push, 0x0080, 1);
 			evo_data(push, 0x00000000);
 		}
-		evo_kick(push, fb->dev, 0);
+		evo_kick(push, fb->dev, EVO_MASTER);
 	}
 
 	nv_crtc->fb.tile_flags = nvfb->r_dma;
@@ -227,7 +520,7 @@
 nvd0_crtc_cursor_show(struct nouveau_crtc *nv_crtc, bool show, bool update)
 {
 	struct drm_device *dev = nv_crtc->base.dev;
-	u32 *push = evo_wait(dev, 0, 16);
+	u32 *push = evo_wait(dev, EVO_MASTER, 16);
 	if (push) {
 		if (show) {
 			evo_mthd(push, 0x0480 + (nv_crtc->index * 0x300), 2);
@@ -247,7 +540,7 @@
 			evo_data(push, 0x00000000);
 		}
 
-		evo_kick(push, dev, 0);
+		evo_kick(push, dev, EVO_MASTER);
 	}
 }
 
@@ -262,7 +555,9 @@
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
 	u32 *push;
 
-	push = evo_wait(crtc->dev, 0, 2);
+	nvd0_display_flip_stop(crtc);
+
+	push = evo_wait(crtc->dev, EVO_MASTER, 2);
 	if (push) {
 		evo_mthd(push, 0x0474 + (nv_crtc->index * 0x300), 1);
 		evo_data(push, 0x00000000);
@@ -270,7 +565,7 @@
 		evo_data(push, 0x03000000);
 		evo_mthd(push, 0x045c + (nv_crtc->index * 0x300), 1);
 		evo_data(push, 0x00000000);
-		evo_kick(push, crtc->dev, 0);
+		evo_kick(push, crtc->dev, EVO_MASTER);
 	}
 
 	nvd0_crtc_cursor_show(nv_crtc, false, false);
@@ -282,7 +577,7 @@
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
 	u32 *push;
 
-	push = evo_wait(crtc->dev, 0, 32);
+	push = evo_wait(crtc->dev, EVO_MASTER, 32);
 	if (push) {
 		evo_mthd(push, 0x0474 + (nv_crtc->index * 0x300), 1);
 		evo_data(push, nv_crtc->fb.tile_flags);
@@ -295,10 +590,11 @@
 		evo_data(push, NvEvoVRAM);
 		evo_mthd(push, 0x0430 + (nv_crtc->index * 0x300), 1);
 		evo_data(push, 0xffffff00);
-		evo_kick(push, crtc->dev, 0);
+		evo_kick(push, crtc->dev, EVO_MASTER);
 	}
 
-	nvd0_crtc_cursor_show(nv_crtc, nv_crtc->cursor.visible, true);
+	nvd0_crtc_cursor_show(nv_crtc, nv_crtc->cursor.visible, false);
+	nvd0_display_flip_next(crtc, crtc->fb, NULL, 1);
 }
 
 static bool
@@ -333,21 +629,35 @@
 {
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
 	struct nouveau_connector *nv_connector;
-	u32 htotal = mode->htotal;
-	u32 vtotal = mode->vtotal;
-	u32 hsyncw = mode->hsync_end - mode->hsync_start - 1;
-	u32 vsyncw = mode->vsync_end - mode->vsync_start - 1;
-	u32 hfrntp = mode->hsync_start - mode->hdisplay;
-	u32 vfrntp = mode->vsync_start - mode->vdisplay;
-	u32 hbackp = mode->htotal - mode->hsync_end;
-	u32 vbackp = mode->vtotal - mode->vsync_end;
-	u32 hss2be = hsyncw + hbackp;
-	u32 vss2be = vsyncw + vbackp;
-	u32 hss2de = htotal - hfrntp;
-	u32 vss2de = vtotal - vfrntp;
+	u32 ilace = (mode->flags & DRM_MODE_FLAG_INTERLACE) ? 2 : 1;
+	u32 vscan = (mode->flags & DRM_MODE_FLAG_DBLSCAN) ? 2 : 1;
+	u32 hactive, hsynce, hbackp, hfrontp, hblanke, hblanks;
+	u32 vactive, vsynce, vbackp, vfrontp, vblanke, vblanks;
+	u32 vblan2e = 0, vblan2s = 1;
+	u32 magic = 0x31ec6000;
 	u32 syncs, *push;
 	int ret;
 
+	hactive = mode->htotal;
+	hsynce  = mode->hsync_end - mode->hsync_start - 1;
+	hbackp  = mode->htotal - mode->hsync_end;
+	hblanke = hsynce + hbackp;
+	hfrontp = mode->hsync_start - mode->hdisplay;
+	hblanks = mode->htotal - hfrontp - 1;
+
+	vactive = mode->vtotal * vscan / ilace;
+	vsynce  = ((mode->vsync_end - mode->vsync_start) * vscan / ilace) - 1;
+	vbackp  = (mode->vtotal - mode->vsync_end) * vscan / ilace;
+	vblanke = vsynce + vbackp;
+	vfrontp = (mode->vsync_start - mode->vdisplay) * vscan / ilace;
+	vblanks = vactive - vfrontp - 1;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE) {
+		vblan2e = vactive + vsynce + vbackp;
+		vblan2s = vblan2e + (mode->vdisplay * vscan / ilace);
+		vactive = (vactive * 2) + 1;
+		magic  |= 0x00000001;
+	}
+
 	syncs = 0x00000001;
 	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
 		syncs |= 0x00000008;
@@ -358,28 +668,33 @@
 	if (ret)
 		return ret;
 
-	push = evo_wait(crtc->dev, 0, 64);
+	push = evo_wait(crtc->dev, EVO_MASTER, 64);
 	if (push) {
-		evo_mthd(push, 0x0410 + (nv_crtc->index * 0x300), 5);
+		evo_mthd(push, 0x0410 + (nv_crtc->index * 0x300), 6);
 		evo_data(push, 0x00000000);
-		evo_data(push, (vtotal << 16) | htotal);
-		evo_data(push, (vsyncw << 16) | hsyncw);
-		evo_data(push, (vss2be << 16) | hss2be);
-		evo_data(push, (vss2de << 16) | hss2de);
+		evo_data(push, (vactive << 16) | hactive);
+		evo_data(push, ( vsynce << 16) | hsynce);
+		evo_data(push, (vblanke << 16) | hblanke);
+		evo_data(push, (vblanks << 16) | hblanks);
+		evo_data(push, (vblan2e << 16) | vblan2s);
 		evo_mthd(push, 0x042c + (nv_crtc->index * 0x300), 1);
 		evo_data(push, 0x00000000); /* ??? */
 		evo_mthd(push, 0x0450 + (nv_crtc->index * 0x300), 3);
 		evo_data(push, mode->clock * 1000);
 		evo_data(push, 0x00200000); /* ??? */
 		evo_data(push, mode->clock * 1000);
-		evo_mthd(push, 0x0404 + (nv_crtc->index * 0x300), 1);
+		evo_mthd(push, 0x0404 + (nv_crtc->index * 0x300), 2);
 		evo_data(push, syncs);
-		evo_kick(push, crtc->dev, 0);
+		evo_data(push, magic);
+		evo_mthd(push, 0x04d0 + (nv_crtc->index * 0x300), 2);
+		evo_data(push, 0x00000311);
+		evo_data(push, 0x00000100);
+		evo_kick(push, crtc->dev, EVO_MASTER);
 	}
 
 	nv_connector = nouveau_crtc_connector_get(nv_crtc);
-	nvd0_crtc_set_dither(nv_crtc, nv_connector->use_dithering, false);
-	nvd0_crtc_set_scale(nv_crtc, nv_connector->scaling_mode, false);
+	nvd0_crtc_set_dither(nv_crtc, false);
+	nvd0_crtc_set_scale(nv_crtc, false);
 	nvd0_crtc_set_image(nv_crtc, crtc->fb, x, y, false);
 	return 0;
 }
@@ -400,7 +715,9 @@
 	if (ret)
 		return ret;
 
+	nvd0_display_flip_stop(crtc);
 	nvd0_crtc_set_image(nv_crtc, crtc->fb, x, y, true);
+	nvd0_display_flip_next(crtc, crtc->fb, NULL, 1);
 	return 0;
 }
 
@@ -410,6 +727,7 @@
 			       enum mode_set_atomic state)
 {
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
+	nvd0_display_flip_stop(crtc);
 	nvd0_crtc_set_image(nv_crtc, fb, x, y, true);
 	return 0;
 }
@@ -472,10 +790,10 @@
 nvd0_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 {
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-	const u32 data = (y << 16) | x;
+	int ch = EVO_CURS(nv_crtc->index);
 
-	nv_wr32(crtc->dev, 0x64d084 + (nv_crtc->index * 0x1000), data);
-	nv_wr32(crtc->dev, 0x64d080 + (nv_crtc->index * 0x1000), 0x00000000);
+	evo_piow(crtc->dev, ch, 0x0084, (y << 16) | x);
+	evo_piow(crtc->dev, ch, 0x0080, 0x00000000);
 	return 0;
 }
 
@@ -525,6 +843,7 @@
 	.gamma_set = nvd0_crtc_gamma_set,
 	.set_config = drm_crtc_helper_set_config,
 	.destroy = nvd0_crtc_destroy,
+	.page_flip = nouveau_crtc_page_flip,
 };
 
 static void
@@ -659,12 +978,12 @@
 
 	nvd0_dac_dpms(encoder, DRM_MODE_DPMS_ON);
 
-	push = evo_wait(encoder->dev, 0, 4);
+	push = evo_wait(encoder->dev, EVO_MASTER, 4);
 	if (push) {
 		evo_mthd(push, 0x0180 + (nv_encoder->or * 0x20), 2);
 		evo_data(push, 1 << nv_crtc->index);
 		evo_data(push, 0x00ff);
-		evo_kick(push, encoder->dev, 0);
+		evo_kick(push, encoder->dev, EVO_MASTER);
 	}
 
 	nv_encoder->crtc = encoder->crtc;
@@ -680,13 +999,13 @@
 	if (nv_encoder->crtc) {
 		nvd0_crtc_prepare(nv_encoder->crtc);
 
-		push = evo_wait(dev, 0, 4);
+		push = evo_wait(dev, EVO_MASTER, 4);
 		if (push) {
 			evo_mthd(push, 0x0180 + (nv_encoder->or * 0x20), 1);
 			evo_data(push, 0x00000000);
 			evo_mthd(push, 0x0080, 1);
 			evo_data(push, 0x00000000);
-			evo_kick(push, dev, 0);
+			evo_kick(push, dev, EVO_MASTER);
 		}
 
 		nv_encoder->crtc = NULL;
@@ -760,6 +1079,108 @@
 }
 
 /******************************************************************************
+ * Audio
+ *****************************************************************************/
+static void
+nvd0_audio_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_connector *nv_connector;
+	struct drm_device *dev = encoder->dev;
+	int i, or = nv_encoder->or * 0x30;
+
+	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	if (!drm_detect_monitor_audio(nv_connector->edid))
+		return;
+
+	nv_mask(dev, 0x10ec10 + or, 0x80000003, 0x80000001);
+
+	drm_edid_to_eld(&nv_connector->base, nv_connector->edid);
+	if (nv_connector->base.eld[0]) {
+		u8 *eld = nv_connector->base.eld;
+
+		for (i = 0; i < eld[2] * 4; i++)
+			nv_wr32(dev, 0x10ec00 + or, (i << 8) | eld[i]);
+		for (i = eld[2] * 4; i < 0x60; i++)
+			nv_wr32(dev, 0x10ec00 + or, (i << 8) | 0x00);
+
+		nv_mask(dev, 0x10ec10 + or, 0x80000002, 0x80000002);
+	}
+}
+
+static void
+nvd0_audio_disconnect(struct drm_encoder *encoder)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct drm_device *dev = encoder->dev;
+	int or = nv_encoder->or * 0x30;
+
+	nv_mask(dev, 0x10ec10 + or, 0x80000003, 0x80000000);
+}
+
+/******************************************************************************
+ * HDMI
+ *****************************************************************************/
+static void
+nvd0_hdmi_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc);
+	struct nouveau_connector *nv_connector;
+	struct drm_device *dev = encoder->dev;
+	int head = nv_crtc->index * 0x800;
+	u32 rekey = 56; /* binary driver, and tegra constant */
+	u32 max_ac_packet;
+
+	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	if (!drm_detect_hdmi_monitor(nv_connector->edid))
+		return;
+
+	max_ac_packet  = mode->htotal - mode->hdisplay;
+	max_ac_packet -= rekey;
+	max_ac_packet -= 18; /* constant from tegra */
+	max_ac_packet /= 32;
+
+	/* AVI InfoFrame */
+	nv_mask(dev, 0x616714 + head, 0x00000001, 0x00000000);
+	nv_wr32(dev, 0x61671c + head, 0x000d0282);
+	nv_wr32(dev, 0x616720 + head, 0x0000006f);
+	nv_wr32(dev, 0x616724 + head, 0x00000000);
+	nv_wr32(dev, 0x616728 + head, 0x00000000);
+	nv_wr32(dev, 0x61672c + head, 0x00000000);
+	nv_mask(dev, 0x616714 + head, 0x00000001, 0x00000001);
+
+	/* ??? InfoFrame? */
+	nv_mask(dev, 0x6167a4 + head, 0x00000001, 0x00000000);
+	nv_wr32(dev, 0x6167ac + head, 0x00000010);
+	nv_mask(dev, 0x6167a4 + head, 0x00000001, 0x00000001);
+
+	/* HDMI_CTRL */
+	nv_mask(dev, 0x616798 + head, 0x401f007f, 0x40000000 | rekey |
+						  max_ac_packet << 16);
+
+	/* NFI, audio doesn't work without it though.. */
+	nv_mask(dev, 0x616548 + head, 0x00000070, 0x00000000);
+
+	nvd0_audio_mode_set(encoder, mode);
+}
+
+static void
+nvd0_hdmi_disconnect(struct drm_encoder *encoder)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc);
+	struct drm_device *dev = encoder->dev;
+	int head = nv_crtc->index * 0x800;
+
+	nvd0_audio_disconnect(encoder);
+
+	nv_mask(dev, 0x616798 + head, 0x40000000, 0x00000000);
+	nv_mask(dev, 0x6167a4 + head, 0x00000001, 0x00000000);
+	nv_mask(dev, 0x616714 + head, 0x00000001, 0x00000000);
+}
+
+/******************************************************************************
  * SOR
  *****************************************************************************/
 static void
@@ -780,7 +1201,7 @@
 			continue;
 
 		if (nv_partner != nv_encoder &&
-		    nv_partner->dcb->or == nv_encoder->or) {
+		    nv_partner->dcb->or == nv_encoder->dcb->or) {
 			if (nv_partner->last_dpms == DRM_MODE_DPMS_ON)
 				return;
 			break;
@@ -829,7 +1250,8 @@
 nvd0_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 		  struct drm_display_mode *mode)
 {
-	struct drm_nouveau_private *dev_priv = encoder->dev->dev_private;
+	struct drm_device *dev = encoder->dev;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc);
 	struct nouveau_connector *nv_connector;
@@ -852,6 +1274,8 @@
 		or_config = (mode_ctrl & 0x00000f00) >> 8;
 		if (mode->clock >= 165000)
 			or_config |= 0x0100;
+
+		nvd0_hdmi_mode_set(encoder, mode);
 		break;
 	case OUTPUT_LVDS:
 		or_config = (mode_ctrl & 0x00000f00) >> 8;
@@ -861,7 +1285,7 @@
 			if (bios->fp.if_is_24bit)
 				or_config |= 0x0200;
 		} else {
-			if (nv_connector->dcb->type == DCB_CONNECTOR_LVDS_SPWG) {
+			if (nv_connector->type == DCB_CONNECTOR_LVDS_SPWG) {
 				if (((u8 *)nv_connector->edid)[121] == 2)
 					or_config |= 0x0100;
 			} else
@@ -889,12 +1313,12 @@
 
 	nvd0_sor_dpms(encoder, DRM_MODE_DPMS_ON);
 
-	push = evo_wait(encoder->dev, 0, 4);
+	push = evo_wait(dev, EVO_MASTER, 4);
 	if (push) {
 		evo_mthd(push, 0x0200 + (nv_encoder->or * 0x20), 2);
 		evo_data(push, mode_ctrl);
 		evo_data(push, or_config);
-		evo_kick(push, encoder->dev, 0);
+		evo_kick(push, dev, EVO_MASTER);
 	}
 
 	nv_encoder->crtc = encoder->crtc;
@@ -910,15 +1334,17 @@
 	if (nv_encoder->crtc) {
 		nvd0_crtc_prepare(nv_encoder->crtc);
 
-		push = evo_wait(dev, 0, 4);
+		push = evo_wait(dev, EVO_MASTER, 4);
 		if (push) {
 			evo_mthd(push, 0x0200 + (nv_encoder->or * 0x20), 1);
 			evo_data(push, 0x00000000);
 			evo_mthd(push, 0x0080, 1);
 			evo_data(push, 0x00000000);
-			evo_kick(push, dev, 0);
+			evo_kick(push, dev, EVO_MASTER);
 		}
 
+		nvd0_hdmi_disconnect(encoder);
+
 		nv_encoder->crtc = NULL;
 		nv_encoder->last_dpms = DRM_MODE_DPMS_OFF;
 	}
@@ -1159,6 +1585,12 @@
 	struct nvd0_display *disp = nvd0_display(dev);
 	u32 intr = nv_rd32(dev, 0x610088);
 
+	if (intr & 0x00000001) {
+		u32 stat = nv_rd32(dev, 0x61008c);
+		nv_wr32(dev, 0x61008c, stat);
+		intr &= ~0x00000001;
+	}
+
 	if (intr & 0x00000002) {
 		u32 stat = nv_rd32(dev, 0x61009c);
 		int chid = ffs(stat) - 1;
@@ -1215,38 +1647,29 @@
 /******************************************************************************
  * Init
  *****************************************************************************/
-static void
+void
 nvd0_display_fini(struct drm_device *dev)
 {
 	int i;
 
-	/* fini cursors */
-	for (i = 14; i >= 13; i--) {
-		if (!(nv_rd32(dev, 0x610490 + (i * 0x10)) & 0x00000001))
-			continue;
-
-		nv_mask(dev, 0x610490 + (i * 0x10), 0x00000001, 0x00000000);
-		nv_wait(dev, 0x610490 + (i * 0x10), 0x00010000, 0x00000000);
-		nv_mask(dev, 0x610090, 1 << i, 0x00000000);
-		nv_mask(dev, 0x6100a0, 1 << i, 0x00000000);
+	/* fini cursors + overlays + flips */
+	for (i = 1; i >= 0; i--) {
+		evo_fini_pio(dev, EVO_CURS(i));
+		evo_fini_pio(dev, EVO_OIMM(i));
+		evo_fini_dma(dev, EVO_OVLY(i));
+		evo_fini_dma(dev, EVO_FLIP(i));
 	}
 
 	/* fini master */
-	if (nv_rd32(dev, 0x610490) & 0x00000010) {
-		nv_mask(dev, 0x610490, 0x00000010, 0x00000000);
-		nv_mask(dev, 0x610490, 0x00000003, 0x00000000);
-		nv_wait(dev, 0x610490, 0x80000000, 0x00000000);
-		nv_mask(dev, 0x610090, 0x00000001, 0x00000000);
-		nv_mask(dev, 0x6100a0, 0x00000001, 0x00000000);
-	}
+	evo_fini_dma(dev, EVO_MASTER);
 }
 
 int
 nvd0_display_init(struct drm_device *dev)
 {
 	struct nvd0_display *disp = nvd0_display(dev);
+	int ret, i;
 	u32 *push;
-	int i;
 
 	if (nv_rd32(dev, 0x6100ac) & 0x00000100) {
 		nv_wr32(dev, 0x6100ac, 0x00000100);
@@ -1271,7 +1694,7 @@
 		nv_wr32(dev, 0x6301c4 + (i * 0x800), sor);
 	}
 
-	for (i = 0; i < 2; i++) {
+	for (i = 0; i < dev->mode_config.num_crtc; i++) {
 		u32 crtc0 = nv_rd32(dev, 0x616104 + (i * 0x800));
 		u32 crtc1 = nv_rd32(dev, 0x616108 + (i * 0x800));
 		u32 crtc2 = nv_rd32(dev, 0x61610c + (i * 0x800));
@@ -1285,36 +1708,24 @@
 	nv_mask(dev, 0x6100b0, 0x00000307, 0x00000307);
 
 	/* init master */
-	nv_wr32(dev, 0x610494, (disp->evo[0].handle >> 8) | 3);
-	nv_wr32(dev, 0x610498, 0x00010000);
-	nv_wr32(dev, 0x61049c, 0x00000001);
-	nv_mask(dev, 0x610490, 0x00000010, 0x00000010);
-	nv_wr32(dev, 0x640000, 0x00000000);
-	nv_wr32(dev, 0x610490, 0x01000013);
-	if (!nv_wait(dev, 0x610490, 0x80000000, 0x00000000)) {
-		NV_ERROR(dev, "PDISP: master 0x%08x\n",
-			 nv_rd32(dev, 0x610490));
-		return -EBUSY;
-	}
-	nv_mask(dev, 0x610090, 0x00000001, 0x00000001);
-	nv_mask(dev, 0x6100a0, 0x00000001, 0x00000001);
+	ret = evo_init_dma(dev, EVO_MASTER);
+	if (ret)
+		goto error;
 
-	/* init cursors */
-	for (i = 13; i <= 14; i++) {
-		nv_wr32(dev, 0x610490 + (i * 0x10), 0x00000001);
-		if (!nv_wait(dev, 0x610490 + (i * 0x10), 0x00010000, 0x00010000)) {
-			NV_ERROR(dev, "PDISP: curs%d 0x%08x\n", i,
-				 nv_rd32(dev, 0x610490 + (i * 0x10)));
-			return -EBUSY;
-		}
-
-		nv_mask(dev, 0x610090, 1 << i, 1 << i);
-		nv_mask(dev, 0x6100a0, 1 << i, 1 << i);
+	/* init flips + overlays + cursors */
+	for (i = 0; i < dev->mode_config.num_crtc; i++) {
+		if ((ret = evo_init_dma(dev, EVO_FLIP(i))) ||
+		    (ret = evo_init_dma(dev, EVO_OVLY(i))) ||
+		    (ret = evo_init_pio(dev, EVO_OIMM(i))) ||
+		    (ret = evo_init_pio(dev, EVO_CURS(i))))
+			goto error;
 	}
 
-	push = evo_wait(dev, 0, 32);
-	if (!push)
-		return -EBUSY;
+	push = evo_wait(dev, EVO_MASTER, 32);
+	if (!push) {
+		ret = -EBUSY;
+		goto error;
+	}
 	evo_mthd(push, 0x0088, 1);
 	evo_data(push, NvEvoSync);
 	evo_mthd(push, 0x0084, 1);
@@ -1323,9 +1734,12 @@
 	evo_data(push, 0x80000000);
 	evo_mthd(push, 0x008c, 1);
 	evo_data(push, 0x00000000);
-	evo_kick(push, dev, 0);
+	evo_kick(push, dev, EVO_MASTER);
 
-	return 0;
+error:
+	if (ret)
+		nvd0_display_fini(dev);
+	return ret;
 }
 
 void
@@ -1334,11 +1748,16 @@
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nvd0_display *disp = nvd0_display(dev);
 	struct pci_dev *pdev = dev->pdev;
+	int i;
 
-	nvd0_display_fini(dev);
+	for (i = 0; i < EVO_DMA_NR; i++) {
+		struct evo *evo = &disp->evo[i];
+		pci_free_consistent(pdev, PAGE_SIZE, evo->ptr, evo->handle);
+	}
 
-	pci_free_consistent(pdev, PAGE_SIZE, disp->evo[0].ptr, disp->evo[0].handle);
 	nouveau_gpuobj_ref(NULL, &disp->mem);
+	nouveau_bo_unmap(disp->sync);
+	nouveau_bo_ref(NULL, &disp->sync);
 	nouveau_irq_unregister(dev, 26);
 
 	dev_priv->engine.display.priv = NULL;
@@ -1410,61 +1829,83 @@
 	tasklet_init(&disp->tasklet, nvd0_display_bh, (unsigned long)dev);
 	nouveau_irq_register(dev, 26, nvd0_display_intr);
 
+	/* small shared memory area we use for notifiers and semaphores */
+	ret = nouveau_bo_new(dev, 4096, 0x1000, TTM_PL_FLAG_VRAM,
+			     0, 0x0000, &disp->sync);
+	if (!ret) {
+		ret = nouveau_bo_pin(disp->sync, TTM_PL_FLAG_VRAM);
+		if (!ret)
+			ret = nouveau_bo_map(disp->sync);
+		if (ret)
+			nouveau_bo_ref(NULL, &disp->sync);
+	}
+
+	if (ret)
+		goto out;
+
 	/* hash table and dma objects for the memory areas we care about */
 	ret = nouveau_gpuobj_new(dev, NULL, 0x4000, 0x10000,
 				 NVOBJ_FLAG_ZERO_ALLOC, &disp->mem);
 	if (ret)
 		goto out;
 
-	nv_wo32(disp->mem, 0x1000, 0x00000049);
-	nv_wo32(disp->mem, 0x1004, (disp->mem->vinst + 0x2000) >> 8);
-	nv_wo32(disp->mem, 0x1008, (disp->mem->vinst + 0x2fff) >> 8);
-	nv_wo32(disp->mem, 0x100c, 0x00000000);
-	nv_wo32(disp->mem, 0x1010, 0x00000000);
-	nv_wo32(disp->mem, 0x1014, 0x00000000);
-	nv_wo32(disp->mem, 0x0000, NvEvoSync);
-	nv_wo32(disp->mem, 0x0004, (0x1000 << 9) | 0x00000001);
+	/* create evo dma channels */
+	for (i = 0; i < EVO_DMA_NR; i++) {
+		struct evo *evo = &disp->evo[i];
+		u64 offset = disp->sync->bo.offset;
+		u32 dmao = 0x1000 + (i * 0x100);
+		u32 hash = 0x0000 + (i * 0x040);
 
-	nv_wo32(disp->mem, 0x1020, 0x00000049);
-	nv_wo32(disp->mem, 0x1024, 0x00000000);
-	nv_wo32(disp->mem, 0x1028, (dev_priv->vram_size - 1) >> 8);
-	nv_wo32(disp->mem, 0x102c, 0x00000000);
-	nv_wo32(disp->mem, 0x1030, 0x00000000);
-	nv_wo32(disp->mem, 0x1034, 0x00000000);
-	nv_wo32(disp->mem, 0x0008, NvEvoVRAM);
-	nv_wo32(disp->mem, 0x000c, (0x1020 << 9) | 0x00000001);
+		evo->idx = i;
+		evo->sem.offset = EVO_SYNC(evo->idx, 0x00);
+		evo->ptr = pci_alloc_consistent(pdev, PAGE_SIZE, &evo->handle);
+		if (!evo->ptr) {
+			ret = -ENOMEM;
+			goto out;
+		}
 
-	nv_wo32(disp->mem, 0x1040, 0x00000009);
-	nv_wo32(disp->mem, 0x1044, 0x00000000);
-	nv_wo32(disp->mem, 0x1048, (dev_priv->vram_size - 1) >> 8);
-	nv_wo32(disp->mem, 0x104c, 0x00000000);
-	nv_wo32(disp->mem, 0x1050, 0x00000000);
-	nv_wo32(disp->mem, 0x1054, 0x00000000);
-	nv_wo32(disp->mem, 0x0010, NvEvoVRAM_LP);
-	nv_wo32(disp->mem, 0x0014, (0x1040 << 9) | 0x00000001);
+		nv_wo32(disp->mem, dmao + 0x00, 0x00000049);
+		nv_wo32(disp->mem, dmao + 0x04, (offset + 0x0000) >> 8);
+		nv_wo32(disp->mem, dmao + 0x08, (offset + 0x0fff) >> 8);
+		nv_wo32(disp->mem, dmao + 0x0c, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x10, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x14, 0x00000000);
+		nv_wo32(disp->mem, hash + 0x00, NvEvoSync);
+		nv_wo32(disp->mem, hash + 0x04, 0x00000001 | (i << 27) |
+						((dmao + 0x00) << 9));
 
-	nv_wo32(disp->mem, 0x1060, 0x0fe00009);
-	nv_wo32(disp->mem, 0x1064, 0x00000000);
-	nv_wo32(disp->mem, 0x1068, (dev_priv->vram_size - 1) >> 8);
-	nv_wo32(disp->mem, 0x106c, 0x00000000);
-	nv_wo32(disp->mem, 0x1070, 0x00000000);
-	nv_wo32(disp->mem, 0x1074, 0x00000000);
-	nv_wo32(disp->mem, 0x0018, NvEvoFB32);
-	nv_wo32(disp->mem, 0x001c, (0x1060 << 9) | 0x00000001);
+		nv_wo32(disp->mem, dmao + 0x20, 0x00000049);
+		nv_wo32(disp->mem, dmao + 0x24, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x28, (dev_priv->vram_size - 1) >> 8);
+		nv_wo32(disp->mem, dmao + 0x2c, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x30, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x34, 0x00000000);
+		nv_wo32(disp->mem, hash + 0x08, NvEvoVRAM);
+		nv_wo32(disp->mem, hash + 0x0c, 0x00000001 | (i << 27) |
+						((dmao + 0x20) << 9));
 
-	pinstmem->flush(dev);
+		nv_wo32(disp->mem, dmao + 0x40, 0x00000009);
+		nv_wo32(disp->mem, dmao + 0x44, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x48, (dev_priv->vram_size - 1) >> 8);
+		nv_wo32(disp->mem, dmao + 0x4c, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x50, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x54, 0x00000000);
+		nv_wo32(disp->mem, hash + 0x10, NvEvoVRAM_LP);
+		nv_wo32(disp->mem, hash + 0x14, 0x00000001 | (i << 27) |
+						((dmao + 0x40) << 9));
 
-	/* push buffers for evo channels */
-	disp->evo[0].ptr =
-		pci_alloc_consistent(pdev, PAGE_SIZE, &disp->evo[0].handle);
-	if (!disp->evo[0].ptr) {
-		ret = -ENOMEM;
-		goto out;
+		nv_wo32(disp->mem, dmao + 0x60, 0x0fe00009);
+		nv_wo32(disp->mem, dmao + 0x64, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x68, (dev_priv->vram_size - 1) >> 8);
+		nv_wo32(disp->mem, dmao + 0x6c, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x70, 0x00000000);
+		nv_wo32(disp->mem, dmao + 0x74, 0x00000000);
+		nv_wo32(disp->mem, hash + 0x18, NvEvoFB32);
+		nv_wo32(disp->mem, hash + 0x1c, 0x00000001 | (i << 27) |
+						((dmao + 0x60) << 9));
 	}
 
-	ret = nvd0_display_init(dev);
-	if (ret)
-		goto out;
+	pinstmem->flush(dev);
 
 out:
 	if (ret)
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index cf8b4bc..2139fe8 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -70,7 +70,8 @@
 	r200.o radeon_legacy_tv.o r600_cs.o r600_blit.o r600_blit_shaders.o \
 	r600_blit_kms.o radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o \
 	evergreen.o evergreen_cs.o evergreen_blit_shaders.o evergreen_blit_kms.o \
-	radeon_trace_points.o ni.o cayman_blit_shaders.o atombios_encoders.o
+	radeon_trace_points.o ni.o cayman_blit_shaders.o atombios_encoders.o \
+	radeon_semaphore.o radeon_sa.o
 
 radeon-$(CONFIG_COMPAT) += radeon_ioc32.o
 radeon-$(CONFIG_VGA_SWITCHEROO) += radeon_atpx_handler.o
@@ -78,4 +79,4 @@
 
 obj-$(CONFIG_DRM_RADEON)+= radeon.o
 
-CFLAGS_radeon_trace_points.o := -I$(src)
\ No newline at end of file
+CFLAGS_radeon_trace_points.o := -I$(src)
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 87921c8..0fda830 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -554,7 +554,7 @@
 		if (encoder->crtc == crtc) {
 			radeon_encoder = to_radeon_encoder(encoder);
 			connector = radeon_get_connector_for_encoder(encoder);
-			if (connector)
+			if (connector && connector->display_info.bpc)
 				bpc = connector->display_info.bpc;
 			encoder_mode = atombios_get_encoder_mode(encoder);
 			if ((radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT | ATOM_DEVICE_DFP_SUPPORT)) ||
@@ -1107,9 +1107,40 @@
 		return -EINVAL;
 	}
 
-	if (tiling_flags & RADEON_TILING_MACRO)
+	if (tiling_flags & RADEON_TILING_MACRO) {
+		if (rdev->family >= CHIP_CAYMAN)
+			tmp = rdev->config.cayman.tile_config;
+		else
+			tmp = rdev->config.evergreen.tile_config;
+
+		switch ((tmp & 0xf0) >> 4) {
+		case 0: /* 4 banks */
+			fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_4_BANK);
+			break;
+		case 1: /* 8 banks */
+		default:
+			fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_8_BANK);
+			break;
+		case 2: /* 16 banks */
+			fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_16_BANK);
+			break;
+		}
+
+		switch ((tmp & 0xf000) >> 12) {
+		case 0: /* 1KB rows */
+		default:
+			fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_1KB);
+			break;
+		case 1: /* 2KB rows */
+			fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_2KB);
+			break;
+		case 2: /* 4KB rows */
+			fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_4KB);
+			break;
+		}
+
 		fb_format |= EVERGREEN_GRPH_ARRAY_MODE(EVERGREEN_GRPH_ARRAY_2D_TILED_THIN1);
-	else if (tiling_flags & RADEON_TILING_MICRO)
+	} else if (tiling_flags & RADEON_TILING_MICRO)
 		fb_format |= EVERGREEN_GRPH_ARRAY_MODE(EVERGREEN_GRPH_ARRAY_1D_TILED_THIN1);
 
 	switch (radeon_crtc->crtc_id) {
@@ -1153,7 +1184,7 @@
 	WREG32(EVERGREEN_GRPH_X_END + radeon_crtc->crtc_offset, target_fb->width);
 	WREG32(EVERGREEN_GRPH_Y_END + radeon_crtc->crtc_offset, target_fb->height);
 
-	fb_pitch_pixels = target_fb->pitch / (target_fb->bits_per_pixel / 8);
+	fb_pitch_pixels = target_fb->pitches[0] / (target_fb->bits_per_pixel / 8);
 	WREG32(EVERGREEN_GRPH_PITCH + radeon_crtc->crtc_offset, fb_pitch_pixels);
 	WREG32(EVERGREEN_GRPH_ENABLE + radeon_crtc->crtc_offset, 1);
 
@@ -1322,7 +1353,7 @@
 	WREG32(AVIVO_D1GRPH_X_END + radeon_crtc->crtc_offset, target_fb->width);
 	WREG32(AVIVO_D1GRPH_Y_END + radeon_crtc->crtc_offset, target_fb->height);
 
-	fb_pitch_pixels = target_fb->pitch / (target_fb->bits_per_pixel / 8);
+	fb_pitch_pixels = target_fb->pitches[0] / (target_fb->bits_per_pixel / 8);
 	WREG32(AVIVO_D1GRPH_PITCH + radeon_crtc->crtc_offset, fb_pitch_pixels);
 	WREG32(AVIVO_D1GRPH_ENABLE + radeon_crtc->crtc_offset, 1);
 
@@ -1522,12 +1553,6 @@
 				     struct drm_display_mode *mode,
 				     struct drm_display_mode *adjusted_mode)
 {
-	struct drm_device *dev = crtc->dev;
-	struct radeon_device *rdev = dev->dev_private;
-
-	/* adjust pm to upcoming mode change */
-	radeon_pm_compute_clocks(rdev);
-
 	if (!radeon_crtc_scaling_mode_fixup(crtc, mode, adjusted_mode))
 		return false;
 	return true;
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index a0de485..6fb335a 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -283,7 +283,7 @@
 		}
 	}
 
-	DRM_ERROR("aux i2c too many retries, giving up\n");
+	DRM_DEBUG_KMS("aux i2c too many retries, giving up\n");
 	return -EREMOTEIO;
 }
 
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index a7da8d4..ccde2c9 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -40,6 +40,8 @@
 static void evergreen_gpu_init(struct radeon_device *rdev);
 void evergreen_fini(struct radeon_device *rdev);
 void evergreen_pcie_gen2_enable(struct radeon_device *rdev);
+extern void cayman_cp_int_cntl_setup(struct radeon_device *rdev,
+				     int ring, u32 cp_int_cntl);
 
 void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev)
 {
@@ -82,6 +84,7 @@
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset);
+	int i;
 
 	/* Lock the graphics update lock */
 	tmp |= EVERGREEN_GRPH_UPDATE_LOCK;
@@ -99,7 +102,11 @@
 	       (u32)crtc_base);
 
 	/* Wait for update_pending to go high. */
-	while (!(RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING));
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		if (RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING)
+			break;
+		udelay(1);
+	}
 	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
@@ -157,6 +164,57 @@
 	return actual_temp * 1000;
 }
 
+void sumo_pm_init_profile(struct radeon_device *rdev)
+{
+	int idx;
+
+	/* default */
+	rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_off_ps_idx = rdev->pm.default_power_state_index;
+	rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_on_ps_idx = rdev->pm.default_power_state_index;
+	rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_on_cm_idx = 0;
+
+	/* low,mid sh/mh */
+	if (rdev->flags & RADEON_IS_MOBILITY)
+		idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
+	else
+		idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+
+	rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0;
+
+	rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0;
+
+	rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 0;
+
+	rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 0;
+
+	/* high sh/mh */
+	idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+	rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_cm_idx =
+		rdev->pm.power_state[idx].num_clock_modes - 1;
+
+	rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_cm_idx =
+		rdev->pm.power_state[idx].num_clock_modes - 1;
+}
+
 void evergreen_pm_misc(struct radeon_device *rdev)
 {
 	int req_ps_idx = rdev->pm.requested_power_state_index;
@@ -1219,7 +1277,7 @@
 		WREG32(MC_VM_SYSTEM_APERTURE_HIGH_ADDR,
 			rdev->mc.vram_end >> 12);
 	}
-	WREG32(MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR, 0);
+	WREG32(MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR, rdev->vram_scratch.gpu_addr >> 12);
 	if (rdev->flags & RADEON_IS_IGP) {
 		tmp = RREG32(MC_FUS_VM_FB_OFFSET) & 0x000FFFFF;
 		tmp |= ((rdev->mc.vram_end >> 20) & 0xF) << 24;
@@ -1255,18 +1313,20 @@
  */
 void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
+	struct radeon_ring *ring = &rdev->ring[ib->fence->ring];
+
 	/* set to DX10/11 mode */
-	radeon_ring_write(rdev, PACKET3(PACKET3_MODE_CONTROL, 0));
-	radeon_ring_write(rdev, 1);
+	radeon_ring_write(ring, PACKET3(PACKET3_MODE_CONTROL, 0));
+	radeon_ring_write(ring, 1);
 	/* FIXME: implement */
-	radeon_ring_write(rdev, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
+	radeon_ring_write(ring,
 #ifdef __BIG_ENDIAN
 			  (2 << 0) |
 #endif
 			  (ib->gpu_addr & 0xFFFFFFFC));
-	radeon_ring_write(rdev, upper_32_bits(ib->gpu_addr) & 0xFF);
-	radeon_ring_write(rdev, ib->length_dw);
+	radeon_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFF);
+	radeon_ring_write(ring, ib->length_dw);
 }
 
 
@@ -1304,71 +1364,73 @@
 
 static int evergreen_cp_start(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r, i;
 	uint32_t cp_me;
 
-	r = radeon_ring_lock(rdev, 7);
+	r = radeon_ring_lock(rdev, ring, 7);
 	if (r) {
 		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
 		return r;
 	}
-	radeon_ring_write(rdev, PACKET3(PACKET3_ME_INITIALIZE, 5));
-	radeon_ring_write(rdev, 0x1);
-	radeon_ring_write(rdev, 0x0);
-	radeon_ring_write(rdev, rdev->config.evergreen.max_hw_contexts - 1);
-	radeon_ring_write(rdev, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_write(ring, PACKET3(PACKET3_ME_INITIALIZE, 5));
+	radeon_ring_write(ring, 0x1);
+	radeon_ring_write(ring, 0x0);
+	radeon_ring_write(ring, rdev->config.evergreen.max_hw_contexts - 1);
+	radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_unlock_commit(rdev, ring);
 
 	cp_me = 0xff;
 	WREG32(CP_ME_CNTL, cp_me);
 
-	r = radeon_ring_lock(rdev, evergreen_default_size + 19);
+	r = radeon_ring_lock(rdev, ring, evergreen_default_size + 19);
 	if (r) {
 		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
 		return r;
 	}
 
 	/* setup clear context state */
-	radeon_ring_write(rdev, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	radeon_ring_write(rdev, PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
+	radeon_ring_write(ring, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	radeon_ring_write(ring, PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
 
 	for (i = 0; i < evergreen_default_size; i++)
-		radeon_ring_write(rdev, evergreen_default_state[i]);
+		radeon_ring_write(ring, evergreen_default_state[i]);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	radeon_ring_write(rdev, PACKET3_PREAMBLE_END_CLEAR_STATE);
+	radeon_ring_write(ring, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	radeon_ring_write(ring, PACKET3_PREAMBLE_END_CLEAR_STATE);
 
 	/* set clear context state */
-	radeon_ring_write(rdev, PACKET3(PACKET3_CLEAR_STATE, 0));
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_CLEAR_STATE, 0));
+	radeon_ring_write(ring, 0);
 
 	/* SQ_VTX_BASE_VTX_LOC */
-	radeon_ring_write(rdev, 0xc0026f00);
-	radeon_ring_write(rdev, 0x00000000);
-	radeon_ring_write(rdev, 0x00000000);
-	radeon_ring_write(rdev, 0x00000000);
+	radeon_ring_write(ring, 0xc0026f00);
+	radeon_ring_write(ring, 0x00000000);
+	radeon_ring_write(ring, 0x00000000);
+	radeon_ring_write(ring, 0x00000000);
 
 	/* Clear consts */
-	radeon_ring_write(rdev, 0xc0036f00);
-	radeon_ring_write(rdev, 0x00000bc4);
-	radeon_ring_write(rdev, 0xffffffff);
-	radeon_ring_write(rdev, 0xffffffff);
-	radeon_ring_write(rdev, 0xffffffff);
+	radeon_ring_write(ring, 0xc0036f00);
+	radeon_ring_write(ring, 0x00000bc4);
+	radeon_ring_write(ring, 0xffffffff);
+	radeon_ring_write(ring, 0xffffffff);
+	radeon_ring_write(ring, 0xffffffff);
 
-	radeon_ring_write(rdev, 0xc0026900);
-	radeon_ring_write(rdev, 0x00000316);
-	radeon_ring_write(rdev, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
-	radeon_ring_write(rdev, 0x00000010); /*  */
+	radeon_ring_write(ring, 0xc0026900);
+	radeon_ring_write(ring, 0x00000316);
+	radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
+	radeon_ring_write(ring, 0x00000010); /*  */
 
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_unlock_commit(rdev, ring);
 
 	return 0;
 }
 
 int evergreen_cp_resume(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 tmp;
 	u32 rb_bufsz;
 	int r;
@@ -1386,13 +1448,13 @@
 	RREG32(GRBM_SOFT_RESET);
 
 	/* Set ring buffer size */
-	rb_bufsz = drm_order(rdev->cp.ring_size / 8);
+	rb_bufsz = drm_order(ring->ring_size / 8);
 	tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
 #ifdef __BIG_ENDIAN
 	tmp |= BUF_SWAP_32BIT;
 #endif
 	WREG32(CP_RB_CNTL, tmp);
-	WREG32(CP_SEM_WAIT_TIMER, 0x4);
+	WREG32(CP_SEM_WAIT_TIMER, 0x0);
 
 	/* Set the write pointer delay */
 	WREG32(CP_RB_WPTR_DELAY, 0);
@@ -1400,8 +1462,8 @@
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB_CNTL, tmp | RB_RPTR_WR_ENA);
 	WREG32(CP_RB_RPTR_WR, 0);
-	rdev->cp.wptr = 0;
-	WREG32(CP_RB_WPTR, rdev->cp.wptr);
+	ring->wptr = 0;
+	WREG32(CP_RB_WPTR, ring->wptr);
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB_RPTR_ADDR,
@@ -1419,16 +1481,16 @@
 	mdelay(1);
 	WREG32(CP_RB_CNTL, tmp);
 
-	WREG32(CP_RB_BASE, rdev->cp.gpu_addr >> 8);
+	WREG32(CP_RB_BASE, ring->gpu_addr >> 8);
 	WREG32(CP_DEBUG, (1 << 27) | (1 << 28));
 
-	rdev->cp.rptr = RREG32(CP_RB_RPTR);
+	ring->rptr = RREG32(CP_RB_RPTR);
 
 	evergreen_cp_start(rdev);
-	rdev->cp.ready = true;
-	r = radeon_ring_test(rdev);
+	ring->ready = true;
+	r = radeon_ring_test(rdev, ring);
 	if (r) {
-		rdev->cp.ready = false;
+		ring->ready = false;
 		return r;
 	}
 	return 0;
@@ -2297,7 +2359,7 @@
 	return 0;
 }
 
-bool evergreen_gpu_is_lockup(struct radeon_device *rdev)
+bool evergreen_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	u32 srbm_status;
 	u32 grbm_status;
@@ -2310,19 +2372,19 @@
 	grbm_status_se0 = RREG32(GRBM_STATUS_SE0);
 	grbm_status_se1 = RREG32(GRBM_STATUS_SE1);
 	if (!(grbm_status & GUI_ACTIVE)) {
-		r100_gpu_lockup_update(lockup, &rdev->cp);
+		r100_gpu_lockup_update(lockup, ring);
 		return false;
 	}
 	/* force CP activities */
-	r = radeon_ring_lock(rdev, 2);
+	r = radeon_ring_lock(rdev, ring, 2);
 	if (!r) {
 		/* PACKET2 NOP */
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_unlock_commit(rdev);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_unlock_commit(rdev, ring);
 	}
-	rdev->cp.rptr = RREG32(CP_RB_RPTR);
-	return r100_gpu_cp_is_lockup(rdev, lockup, &rdev->cp);
+	ring->rptr = RREG32(CP_RB_RPTR);
+	return r100_gpu_cp_is_lockup(rdev, lockup, ring);
 }
 
 static int evergreen_gpu_soft_reset(struct radeon_device *rdev)
@@ -2414,7 +2476,13 @@
 {
 	u32 tmp;
 
-	WREG32(CP_INT_CNTL, CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
+	if (rdev->family >= CHIP_CAYMAN) {
+		cayman_cp_int_cntl_setup(rdev, 0,
+					 CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
+		cayman_cp_int_cntl_setup(rdev, 1, 0);
+		cayman_cp_int_cntl_setup(rdev, 2, 0);
+	} else
+		WREG32(CP_INT_CNTL, CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE);
 	WREG32(GRBM_INT_CNTL, 0);
 	WREG32(INT_MASK + EVERGREEN_CRTC0_REGISTER_OFFSET, 0);
 	WREG32(INT_MASK + EVERGREEN_CRTC1_REGISTER_OFFSET, 0);
@@ -2459,6 +2527,7 @@
 int evergreen_irq_set(struct radeon_device *rdev)
 {
 	u32 cp_int_cntl = CNTX_BUSY_INT_ENABLE | CNTX_EMPTY_INT_ENABLE;
+	u32 cp_int_cntl1 = 0, cp_int_cntl2 = 0;
 	u32 crtc1 = 0, crtc2 = 0, crtc3 = 0, crtc4 = 0, crtc5 = 0, crtc6 = 0;
 	u32 hpd1, hpd2, hpd3, hpd4, hpd5, hpd6;
 	u32 grbm_int_cntl = 0;
@@ -2483,11 +2552,28 @@
 	hpd5 = RREG32(DC_HPD5_INT_CONTROL) & ~DC_HPDx_INT_EN;
 	hpd6 = RREG32(DC_HPD6_INT_CONTROL) & ~DC_HPDx_INT_EN;
 
-	if (rdev->irq.sw_int) {
-		DRM_DEBUG("evergreen_irq_set: sw int\n");
-		cp_int_cntl |= RB_INT_ENABLE;
-		cp_int_cntl |= TIME_STAMP_INT_ENABLE;
+	if (rdev->family >= CHIP_CAYMAN) {
+		/* enable CP interrupts on all rings */
+		if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
+			DRM_DEBUG("evergreen_irq_set: sw int gfx\n");
+			cp_int_cntl |= TIME_STAMP_INT_ENABLE;
+		}
+		if (rdev->irq.sw_int[CAYMAN_RING_TYPE_CP1_INDEX]) {
+			DRM_DEBUG("evergreen_irq_set: sw int cp1\n");
+			cp_int_cntl1 |= TIME_STAMP_INT_ENABLE;
+		}
+		if (rdev->irq.sw_int[CAYMAN_RING_TYPE_CP2_INDEX]) {
+			DRM_DEBUG("evergreen_irq_set: sw int cp2\n");
+			cp_int_cntl2 |= TIME_STAMP_INT_ENABLE;
+		}
+	} else {
+		if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
+			DRM_DEBUG("evergreen_irq_set: sw int gfx\n");
+			cp_int_cntl |= RB_INT_ENABLE;
+			cp_int_cntl |= TIME_STAMP_INT_ENABLE;
+		}
 	}
+
 	if (rdev->irq.crtc_vblank_int[0] ||
 	    rdev->irq.pflip[0]) {
 		DRM_DEBUG("evergreen_irq_set: vblank 0\n");
@@ -2547,7 +2633,12 @@
 		grbm_int_cntl |= GUI_IDLE_INT_ENABLE;
 	}
 
-	WREG32(CP_INT_CNTL, cp_int_cntl);
+	if (rdev->family >= CHIP_CAYMAN) {
+		cayman_cp_int_cntl_setup(rdev, 0, cp_int_cntl);
+		cayman_cp_int_cntl_setup(rdev, 1, cp_int_cntl1);
+		cayman_cp_int_cntl_setup(rdev, 2, cp_int_cntl2);
+	} else
+		WREG32(CP_INT_CNTL, cp_int_cntl);
 	WREG32(GRBM_INT_CNTL, grbm_int_cntl);
 
 	WREG32(INT_MASK + EVERGREEN_CRTC0_REGISTER_OFFSET, crtc1);
@@ -2962,11 +3053,24 @@
 		case 177: /* CP_INT in IB1 */
 		case 178: /* CP_INT in IB2 */
 			DRM_DEBUG("IH: CP int: 0x%08x\n", src_data);
-			radeon_fence_process(rdev);
+			radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX);
 			break;
 		case 181: /* CP EOP event */
 			DRM_DEBUG("IH: CP EOP\n");
-			radeon_fence_process(rdev);
+			if (rdev->family >= CHIP_CAYMAN) {
+				switch (src_data) {
+				case 0:
+					radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX);
+					break;
+				case 1:
+					radeon_fence_process(rdev, CAYMAN_RING_TYPE_CP1_INDEX);
+					break;
+				case 2:
+					radeon_fence_process(rdev, CAYMAN_RING_TYPE_CP2_INDEX);
+					break;
+				}
+			} else
+				radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX);
 			break;
 		case 233: /* GUI IDLE */
 			DRM_DEBUG("IH: GUI idle\n");
@@ -2996,6 +3100,7 @@
 
 static int evergreen_startup(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 
 	/* enable pcie gen2 link */
@@ -3050,6 +3155,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	r = r600_irq_init(rdev);
 	if (r) {
@@ -3059,7 +3170,9 @@
 	}
 	evergreen_irq_set(rdev);
 
-	r = radeon_ring_init(rdev, rdev->cp.ring_size);
+	r = radeon_ring_init(rdev, ring, ring->ring_size, RADEON_WB_CP_RPTR_OFFSET,
+			     R600_CP_RB_RPTR, R600_CP_RB_WPTR,
+			     0, 0xfffff, RADEON_CP_PACKET2);
 	if (r)
 		return r;
 	r = evergreen_cp_load_microcode(rdev);
@@ -3069,6 +3182,16 @@
 	if (r)
 		return r;
 
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r600_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		DRM_ERROR("radeon: failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = r600_audio_init(rdev);
 	if (r) {
 		DRM_ERROR("radeon: audio init failed\n");
@@ -3094,38 +3217,30 @@
 	/* post card */
 	atom_asic_init(rdev->mode_info.atom_context);
 
+	rdev->accel_working = true;
 	r = evergreen_startup(rdev);
 	if (r) {
 		DRM_ERROR("evergreen startup failed on resume\n");
 		return r;
 	}
 
-	r = r600_ib_test(rdev);
-	if (r) {
-		DRM_ERROR("radeon: failed testing IB (%d).\n", r);
-		return r;
-	}
-
-	r = r600_audio_init(rdev);
-	if (r) {
-		DRM_ERROR("radeon: audio resume failed\n");
-		return r;
-	}
-
 	return r;
 
 }
 
 int evergreen_suspend(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+
 	r600_audio_fini(rdev);
 	/* FIXME: we should wait for ring to be empty */
+	radeon_ib_pool_suspend(rdev);
+	r600_blit_suspend(rdev);
 	r700_cp_stop(rdev);
-	rdev->cp.ready = false;
+	ring->ready = false;
 	evergreen_irq_suspend(rdev);
 	radeon_wb_disable(rdev);
 	evergreen_pcie_gart_disable(rdev);
-	r600_blit_suspend(rdev);
 
 	return 0;
 }
@@ -3200,8 +3315,8 @@
 	if (r)
 		return r;
 
-	rdev->cp.ring_obj = NULL;
-	r600_ring_init(rdev, 1024 * 1024);
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ring_obj = NULL;
+	r600_ring_init(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX], 1024 * 1024);
 
 	rdev->ih.ring_obj = NULL;
 	r600_ih_ring_init(rdev, 64 * 1024);
@@ -3210,29 +3325,24 @@
 	if (r)
 		return r;
 
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = evergreen_startup(rdev);
 	if (r) {
 		dev_err(rdev->dev, "disabling GPU acceleration\n");
 		r700_cp_fini(rdev);
 		r600_irq_fini(rdev);
 		radeon_wb_fini(rdev);
+		r100_ib_fini(rdev);
 		radeon_irq_kms_fini(rdev);
 		evergreen_pcie_gart_fini(rdev);
 		rdev->accel_working = false;
 	}
-	if (rdev->accel_working) {
-		r = radeon_ib_pool_init(rdev);
-		if (r) {
-			DRM_ERROR("radeon: failed initializing IB pool (%d).\n", r);
-			rdev->accel_working = false;
-		}
-		r = r600_ib_test(rdev);
-		if (r) {
-			DRM_ERROR("radeon: failed testing IB (%d).\n", r);
-			rdev->accel_working = false;
-		}
-	}
 	return 0;
 }
 
@@ -3243,11 +3353,12 @@
 	r700_cp_fini(rdev);
 	r600_irq_fini(rdev);
 	radeon_wb_fini(rdev);
-	radeon_ib_pool_fini(rdev);
+	r100_ib_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	evergreen_pcie_gart_fini(rdev);
 	r600_vram_scratch_fini(rdev);
 	radeon_gem_fini(rdev);
+	radeon_semaphore_driver_fini(rdev);
 	radeon_fence_driver_fini(rdev);
 	radeon_agp_fini(rdev);
 	radeon_bo_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
index 914e5af..2379849 100644
--- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c
+++ b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
@@ -49,6 +49,7 @@
 set_render_target(struct radeon_device *rdev, int format,
 		  int w, int h, u64 gpu_addr)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 cb_color_info;
 	int pitch, slice;
 
@@ -62,23 +63,23 @@
 	pitch = (w / 8) - 1;
 	slice = ((w * h) / 64) - 1;
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 15));
-	radeon_ring_write(rdev, (CB_COLOR0_BASE - PACKET3_SET_CONTEXT_REG_START) >> 2);
-	radeon_ring_write(rdev, gpu_addr >> 8);
-	radeon_ring_write(rdev, pitch);
-	radeon_ring_write(rdev, slice);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, cb_color_info);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, (w - 1) | ((h - 1) << 16));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 15));
+	radeon_ring_write(ring, (CB_COLOR0_BASE - PACKET3_SET_CONTEXT_REG_START) >> 2);
+	radeon_ring_write(ring, gpu_addr >> 8);
+	radeon_ring_write(ring, pitch);
+	radeon_ring_write(ring, slice);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, cb_color_info);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, (w - 1) | ((h - 1) << 16));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
 }
 
 /* emits 5dw */
@@ -87,6 +88,7 @@
 		    u32 sync_type, u32 size,
 		    u64 mc_addr)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 cp_coher_size;
 
 	if (size == 0xffffffff)
@@ -99,39 +101,40 @@
 		 * to the RB directly. For IBs, the CP programs this as part of the
 		 * surface_sync packet.
 		 */
-		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-		radeon_ring_write(rdev, (0x85e8 - PACKET3_SET_CONFIG_REG_START) >> 2);
-		radeon_ring_write(rdev, 0); /* CP_COHER_CNTL2 */
+		radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		radeon_ring_write(ring, (0x85e8 - PACKET3_SET_CONFIG_REG_START) >> 2);
+		radeon_ring_write(ring, 0); /* CP_COHER_CNTL2 */
 	}
-	radeon_ring_write(rdev, PACKET3(PACKET3_SURFACE_SYNC, 3));
-	radeon_ring_write(rdev, sync_type);
-	radeon_ring_write(rdev, cp_coher_size);
-	radeon_ring_write(rdev, mc_addr >> 8);
-	radeon_ring_write(rdev, 10); /* poll interval */
+	radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
+	radeon_ring_write(ring, sync_type);
+	radeon_ring_write(ring, cp_coher_size);
+	radeon_ring_write(ring, mc_addr >> 8);
+	radeon_ring_write(ring, 10); /* poll interval */
 }
 
 /* emits 11dw + 1 surface sync = 16dw */
 static void
 set_shaders(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u64 gpu_addr;
 
 	/* VS */
 	gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.vs_offset;
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 3));
-	radeon_ring_write(rdev, (SQ_PGM_START_VS - PACKET3_SET_CONTEXT_REG_START) >> 2);
-	radeon_ring_write(rdev, gpu_addr >> 8);
-	radeon_ring_write(rdev, 2);
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 3));
+	radeon_ring_write(ring, (SQ_PGM_START_VS - PACKET3_SET_CONTEXT_REG_START) >> 2);
+	radeon_ring_write(ring, gpu_addr >> 8);
+	radeon_ring_write(ring, 2);
+	radeon_ring_write(ring, 0);
 
 	/* PS */
 	gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.ps_offset;
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 4));
-	radeon_ring_write(rdev, (SQ_PGM_START_PS - PACKET3_SET_CONTEXT_REG_START) >> 2);
-	radeon_ring_write(rdev, gpu_addr >> 8);
-	radeon_ring_write(rdev, 1);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 2);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 4));
+	radeon_ring_write(ring, (SQ_PGM_START_PS - PACKET3_SET_CONTEXT_REG_START) >> 2);
+	radeon_ring_write(ring, gpu_addr >> 8);
+	radeon_ring_write(ring, 1);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 2);
 
 	gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.vs_offset;
 	cp_set_surface_sync(rdev, PACKET3_SH_ACTION_ENA, 512, gpu_addr);
@@ -141,6 +144,7 @@
 static void
 set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 sq_vtx_constant_word2, sq_vtx_constant_word3;
 
 	/* high addr, stride */
@@ -155,16 +159,16 @@
 		SQ_VTCX_SEL_Z(SQ_SEL_Z) |
 		SQ_VTCX_SEL_W(SQ_SEL_W);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_RESOURCE, 8));
-	radeon_ring_write(rdev, 0x580);
-	radeon_ring_write(rdev, gpu_addr & 0xffffffff);
-	radeon_ring_write(rdev, 48 - 1); /* size */
-	radeon_ring_write(rdev, sq_vtx_constant_word2);
-	radeon_ring_write(rdev, sq_vtx_constant_word3);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, S__SQ_CONSTANT_TYPE(SQ_TEX_VTX_VALID_BUFFER));
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_RESOURCE, 8));
+	radeon_ring_write(ring, 0x580);
+	radeon_ring_write(ring, gpu_addr & 0xffffffff);
+	radeon_ring_write(ring, 48 - 1); /* size */
+	radeon_ring_write(ring, sq_vtx_constant_word2);
+	radeon_ring_write(ring, sq_vtx_constant_word3);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, S__SQ_CONSTANT_TYPE(SQ_TEX_VTX_VALID_BUFFER));
 
 	if ((rdev->family == CHIP_CEDAR) ||
 	    (rdev->family == CHIP_PALM) ||
@@ -185,6 +189,7 @@
 		 int format, int w, int h, int pitch,
 		 u64 gpu_addr, u32 size)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 sq_tex_resource_word0, sq_tex_resource_word1;
 	u32 sq_tex_resource_word4, sq_tex_resource_word7;
 
@@ -208,16 +213,16 @@
 	cp_set_surface_sync(rdev,
 			    PACKET3_TC_ACTION_ENA, size, gpu_addr);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_RESOURCE, 8));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, sq_tex_resource_word0);
-	radeon_ring_write(rdev, sq_tex_resource_word1);
-	radeon_ring_write(rdev, gpu_addr >> 8);
-	radeon_ring_write(rdev, gpu_addr >> 8);
-	radeon_ring_write(rdev, sq_tex_resource_word4);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, sq_tex_resource_word7);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_RESOURCE, 8));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, sq_tex_resource_word0);
+	radeon_ring_write(ring, sq_tex_resource_word1);
+	radeon_ring_write(ring, gpu_addr >> 8);
+	radeon_ring_write(ring, gpu_addr >> 8);
+	radeon_ring_write(ring, sq_tex_resource_word4);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, sq_tex_resource_word7);
 }
 
 /* emits 12 */
@@ -225,6 +230,7 @@
 set_scissors(struct radeon_device *rdev, int x1, int y1,
 	     int x2, int y2)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	/* workaround some hw bugs */
 	if (x2 == 0)
 		x1 = 1;
@@ -235,43 +241,44 @@
 			x2 = 2;
 	}
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
-	radeon_ring_write(rdev, (PA_SC_SCREEN_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
-	radeon_ring_write(rdev, (x1 << 0) | (y1 << 16));
-	radeon_ring_write(rdev, (x2 << 0) | (y2 << 16));
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
+	radeon_ring_write(ring, (PA_SC_SCREEN_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
+	radeon_ring_write(ring, (x1 << 0) | (y1 << 16));
+	radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
-	radeon_ring_write(rdev, (PA_SC_GENERIC_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
-	radeon_ring_write(rdev, (x1 << 0) | (y1 << 16) | (1 << 31));
-	radeon_ring_write(rdev, (x2 << 0) | (y2 << 16));
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
+	radeon_ring_write(ring, (PA_SC_GENERIC_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
+	radeon_ring_write(ring, (x1 << 0) | (y1 << 16) | (1 << 31));
+	radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
-	radeon_ring_write(rdev, (PA_SC_WINDOW_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
-	radeon_ring_write(rdev, (x1 << 0) | (y1 << 16) | (1 << 31));
-	radeon_ring_write(rdev, (x2 << 0) | (y2 << 16));
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
+	radeon_ring_write(ring, (PA_SC_WINDOW_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_START) >> 2);
+	radeon_ring_write(ring, (x1 << 0) | (y1 << 16) | (1 << 31));
+	radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
 }
 
 /* emits 10 */
 static void
 draw_auto(struct radeon_device *rdev)
 {
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-	radeon_ring_write(rdev, (VGT_PRIMITIVE_TYPE - PACKET3_SET_CONFIG_REG_START) >> 2);
-	radeon_ring_write(rdev, DI_PT_RECTLIST);
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+	radeon_ring_write(ring, (VGT_PRIMITIVE_TYPE - PACKET3_SET_CONFIG_REG_START) >> 2);
+	radeon_ring_write(ring, DI_PT_RECTLIST);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_INDEX_TYPE, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET3(PACKET3_INDEX_TYPE, 0));
+	radeon_ring_write(ring,
 #ifdef __BIG_ENDIAN
 			  (2 << 2) |
 #endif
 			  DI_INDEX_SIZE_16_BIT);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_NUM_INSTANCES, 0));
-	radeon_ring_write(rdev, 1);
+	radeon_ring_write(ring, PACKET3(PACKET3_NUM_INSTANCES, 0));
+	radeon_ring_write(ring, 1);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_DRAW_INDEX_AUTO, 1));
-	radeon_ring_write(rdev, 3);
-	radeon_ring_write(rdev, DI_SRC_SEL_AUTO_INDEX);
+	radeon_ring_write(ring, PACKET3(PACKET3_DRAW_INDEX_AUTO, 1));
+	radeon_ring_write(ring, 3);
+	radeon_ring_write(ring, DI_SRC_SEL_AUTO_INDEX);
 
 }
 
@@ -279,6 +286,7 @@
 static void
 set_default_state(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 sq_config, sq_gpr_resource_mgmt_1, sq_gpr_resource_mgmt_2, sq_gpr_resource_mgmt_3;
 	u32 sq_thread_resource_mgmt, sq_thread_resource_mgmt_2;
 	u32 sq_stack_resource_mgmt_1, sq_stack_resource_mgmt_2, sq_stack_resource_mgmt_3;
@@ -292,8 +300,8 @@
 	int dwords;
 
 	/* set clear context state */
-	radeon_ring_write(rdev, PACKET3(PACKET3_CLEAR_STATE, 0));
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_CLEAR_STATE, 0));
+	radeon_ring_write(ring, 0);
 
 	if (rdev->family < CHIP_CAYMAN) {
 		switch (rdev->family) {
@@ -550,60 +558,60 @@
 					    NUM_LS_STACK_ENTRIES(num_ls_stack_entries));
 
 		/* disable dyn gprs */
-		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-		radeon_ring_write(rdev, (SQ_DYN_GPR_CNTL_PS_FLUSH_REQ - PACKET3_SET_CONFIG_REG_START) >> 2);
-		radeon_ring_write(rdev, 0);
+		radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		radeon_ring_write(ring, (SQ_DYN_GPR_CNTL_PS_FLUSH_REQ - PACKET3_SET_CONFIG_REG_START) >> 2);
+		radeon_ring_write(ring, 0);
 
 		/* setup LDS */
-		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-		radeon_ring_write(rdev, (SQ_LDS_RESOURCE_MGMT - PACKET3_SET_CONFIG_REG_START) >> 2);
-		radeon_ring_write(rdev, 0x10001000);
+		radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		radeon_ring_write(ring, (SQ_LDS_RESOURCE_MGMT - PACKET3_SET_CONFIG_REG_START) >> 2);
+		radeon_ring_write(ring, 0x10001000);
 
 		/* SQ config */
-		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 11));
-		radeon_ring_write(rdev, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_START) >> 2);
-		radeon_ring_write(rdev, sq_config);
-		radeon_ring_write(rdev, sq_gpr_resource_mgmt_1);
-		radeon_ring_write(rdev, sq_gpr_resource_mgmt_2);
-		radeon_ring_write(rdev, sq_gpr_resource_mgmt_3);
-		radeon_ring_write(rdev, 0);
-		radeon_ring_write(rdev, 0);
-		radeon_ring_write(rdev, sq_thread_resource_mgmt);
-		radeon_ring_write(rdev, sq_thread_resource_mgmt_2);
-		radeon_ring_write(rdev, sq_stack_resource_mgmt_1);
-		radeon_ring_write(rdev, sq_stack_resource_mgmt_2);
-		radeon_ring_write(rdev, sq_stack_resource_mgmt_3);
+		radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 11));
+		radeon_ring_write(ring, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_START) >> 2);
+		radeon_ring_write(ring, sq_config);
+		radeon_ring_write(ring, sq_gpr_resource_mgmt_1);
+		radeon_ring_write(ring, sq_gpr_resource_mgmt_2);
+		radeon_ring_write(ring, sq_gpr_resource_mgmt_3);
+		radeon_ring_write(ring, 0);
+		radeon_ring_write(ring, 0);
+		radeon_ring_write(ring, sq_thread_resource_mgmt);
+		radeon_ring_write(ring, sq_thread_resource_mgmt_2);
+		radeon_ring_write(ring, sq_stack_resource_mgmt_1);
+		radeon_ring_write(ring, sq_stack_resource_mgmt_2);
+		radeon_ring_write(ring, sq_stack_resource_mgmt_3);
 	}
 
 	/* CONTEXT_CONTROL */
-	radeon_ring_write(rdev, 0xc0012800);
-	radeon_ring_write(rdev, 0x80000000);
-	radeon_ring_write(rdev, 0x80000000);
+	radeon_ring_write(ring, 0xc0012800);
+	radeon_ring_write(ring, 0x80000000);
+	radeon_ring_write(ring, 0x80000000);
 
 	/* SQ_VTX_BASE_VTX_LOC */
-	radeon_ring_write(rdev, 0xc0026f00);
-	radeon_ring_write(rdev, 0x00000000);
-	radeon_ring_write(rdev, 0x00000000);
-	radeon_ring_write(rdev, 0x00000000);
+	radeon_ring_write(ring, 0xc0026f00);
+	radeon_ring_write(ring, 0x00000000);
+	radeon_ring_write(ring, 0x00000000);
+	radeon_ring_write(ring, 0x00000000);
 
 	/* SET_SAMPLER */
-	radeon_ring_write(rdev, 0xc0036e00);
-	radeon_ring_write(rdev, 0x00000000);
-	radeon_ring_write(rdev, 0x00000012);
-	radeon_ring_write(rdev, 0x00000000);
-	radeon_ring_write(rdev, 0x00000000);
+	radeon_ring_write(ring, 0xc0036e00);
+	radeon_ring_write(ring, 0x00000000);
+	radeon_ring_write(ring, 0x00000012);
+	radeon_ring_write(ring, 0x00000000);
+	radeon_ring_write(ring, 0x00000000);
 
 	/* set to DX10/11 mode */
-	radeon_ring_write(rdev, PACKET3(PACKET3_MODE_CONTROL, 0));
-	radeon_ring_write(rdev, 1);
+	radeon_ring_write(ring, PACKET3(PACKET3_MODE_CONTROL, 0));
+	radeon_ring_write(ring, 1);
 
 	/* emit an IB pointing at default state */
 	dwords = ALIGN(rdev->r600_blit.state_len, 0x10);
 	gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.state_offset;
-	radeon_ring_write(rdev, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
-	radeon_ring_write(rdev, gpu_addr & 0xFFFFFFFC);
-	radeon_ring_write(rdev, upper_32_bits(gpu_addr) & 0xFF);
-	radeon_ring_write(rdev, dwords);
+	radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
+	radeon_ring_write(ring, gpu_addr & 0xFFFFFFFC);
+	radeon_ring_write(ring, upper_32_bits(gpu_addr) & 0xFF);
+	radeon_ring_write(ring, dwords);
 
 }
 
diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index 7fdfa8e..cd4590a 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -38,6 +38,7 @@
 	u32			group_size;
 	u32			nbanks;
 	u32			npipes;
+	u32			row_size;
 	/* value we track */
 	u32			nsamples;
 	u32			cb_color_base_last[12];
@@ -77,6 +78,44 @@
 	struct radeon_bo	*db_s_write_bo;
 };
 
+static u32 evergreen_cs_get_aray_mode(u32 tiling_flags)
+{
+	if (tiling_flags & RADEON_TILING_MACRO)
+		return ARRAY_2D_TILED_THIN1;
+	else if (tiling_flags & RADEON_TILING_MICRO)
+		return ARRAY_1D_TILED_THIN1;
+	else
+		return ARRAY_LINEAR_GENERAL;
+}
+
+static u32 evergreen_cs_get_num_banks(u32 nbanks)
+{
+	switch (nbanks) {
+	case 2:
+		return ADDR_SURF_2_BANK;
+	case 4:
+		return ADDR_SURF_4_BANK;
+	case 8:
+	default:
+		return ADDR_SURF_8_BANK;
+	case 16:
+		return ADDR_SURF_16_BANK;
+	}
+}
+
+static u32 evergreen_cs_get_tile_split(u32 row_size)
+{
+	switch (row_size) {
+	case 1:
+	default:
+		return ADDR_SURF_TILE_SPLIT_1KB;
+	case 2:
+		return ADDR_SURF_TILE_SPLIT_2KB;
+	case 4:
+		return ADDR_SURF_TILE_SPLIT_4KB;
+	}
+}
+
 static void evergreen_cs_track_init(struct evergreen_cs_track *track)
 {
 	int i;
@@ -480,21 +519,22 @@
 		}
 		break;
 	case DB_Z_INFO:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
-			return -EINVAL;
-		}
 		track->db_z_info = radeon_get_ib_value(p, idx);
-		ib[idx] &= ~Z_ARRAY_MODE(0xf);
-		track->db_z_info &= ~Z_ARRAY_MODE(0xf);
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
-			ib[idx] |= Z_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-			track->db_z_info |= Z_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-		} else {
-			ib[idx] |= Z_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
-			track->db_z_info |= Z_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
+		if (!p->keep_tiling_flags) {
+			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				dev_warn(p->dev, "bad SET_CONTEXT_REG "
+						"0x%04X\n", reg);
+				return -EINVAL;
+			}
+			ib[idx] &= ~Z_ARRAY_MODE(0xf);
+			track->db_z_info &= ~Z_ARRAY_MODE(0xf);
+			ib[idx] |= Z_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+			track->db_z_info |= Z_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
+				ib[idx] |= DB_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks));
+				ib[idx] |= DB_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size));
+			}
 		}
 		break;
 	case DB_STENCIL_INFO:
@@ -607,40 +647,34 @@
 	case CB_COLOR5_INFO:
 	case CB_COLOR6_INFO:
 	case CB_COLOR7_INFO:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
-			return -EINVAL;
-		}
 		tmp = (reg - CB_COLOR0_INFO) / 0x3c;
 		track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
-			ib[idx] |= CB_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-			track->cb_color_info[tmp] |= CB_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-		} else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) {
-			ib[idx] |= CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
-			track->cb_color_info[tmp] |= CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
+		if (!p->keep_tiling_flags) {
+			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				dev_warn(p->dev, "bad SET_CONTEXT_REG "
+						"0x%04X\n", reg);
+				return -EINVAL;
+			}
+			ib[idx] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+			track->cb_color_info[tmp] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
 		}
 		break;
 	case CB_COLOR8_INFO:
 	case CB_COLOR9_INFO:
 	case CB_COLOR10_INFO:
 	case CB_COLOR11_INFO:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
-			return -EINVAL;
-		}
 		tmp = ((reg - CB_COLOR8_INFO) / 0x1c) + 8;
 		track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
-			ib[idx] |= CB_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-			track->cb_color_info[tmp] |= CB_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-		} else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) {
-			ib[idx] |= CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
-			track->cb_color_info[tmp] |= CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
+		if (!p->keep_tiling_flags) {
+			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				dev_warn(p->dev, "bad SET_CONTEXT_REG "
+						"0x%04X\n", reg);
+				return -EINVAL;
+			}
+			ib[idx] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+			track->cb_color_info[tmp] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
 		}
 		break;
 	case CB_COLOR0_PITCH:
@@ -695,6 +729,16 @@
 	case CB_COLOR9_ATTRIB:
 	case CB_COLOR10_ATTRIB:
 	case CB_COLOR11_ATTRIB:
+		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		if (r) {
+			dev_warn(p->dev, "bad SET_CONTEXT_REG "
+					"0x%04X\n", reg);
+			return -EINVAL;
+		}
+		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
+			ib[idx] |= CB_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks));
+			ib[idx] |= CB_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size));
+		}
 		break;
 	case CB_COLOR0_DIM:
 	case CB_COLOR1_DIM:
@@ -1311,10 +1355,16 @@
 					return -EINVAL;
 				}
 				ib[idx+1+(i*8)+2] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
-				if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-					ib[idx+1+(i*8)+1] |= TEX_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-				else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-					ib[idx+1+(i*8)+1] |= TEX_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
+				if (!p->keep_tiling_flags) {
+					ib[idx+1+(i*8)+1] |=
+						TEX_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+					if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
+						ib[idx+1+(i*8)+6] |=
+							TEX_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size));
+						ib[idx+1+(i*8)+7] |=
+							TEX_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks));
+					}
+				}
 				texture = reloc->robj;
 				/* tex mip base */
 				r = evergreen_cs_packet_next_reloc(p, &reloc);
@@ -1414,6 +1464,7 @@
 {
 	struct radeon_cs_packet pkt;
 	struct evergreen_cs_track *track;
+	u32 tmp;
 	int r;
 
 	if (p->track == NULL) {
@@ -1422,9 +1473,63 @@
 		if (track == NULL)
 			return -ENOMEM;
 		evergreen_cs_track_init(track);
-		track->npipes = p->rdev->config.evergreen.tiling_npipes;
-		track->nbanks = p->rdev->config.evergreen.tiling_nbanks;
-		track->group_size = p->rdev->config.evergreen.tiling_group_size;
+		if (p->rdev->family >= CHIP_CAYMAN)
+			tmp = p->rdev->config.cayman.tile_config;
+		else
+			tmp = p->rdev->config.evergreen.tile_config;
+
+		switch (tmp & 0xf) {
+		case 0:
+			track->npipes = 1;
+			break;
+		case 1:
+		default:
+			track->npipes = 2;
+			break;
+		case 2:
+			track->npipes = 4;
+			break;
+		case 3:
+			track->npipes = 8;
+			break;
+		}
+
+		switch ((tmp & 0xf0) >> 4) {
+		case 0:
+			track->nbanks = 4;
+			break;
+		case 1:
+		default:
+			track->nbanks = 8;
+			break;
+		case 2:
+			track->nbanks = 16;
+			break;
+		}
+
+		switch ((tmp & 0xf00) >> 8) {
+		case 0:
+			track->group_size = 256;
+			break;
+		case 1:
+		default:
+			track->group_size = 512;
+			break;
+		}
+
+		switch ((tmp & 0xf000) >> 12) {
+		case 0:
+			track->row_size = 1;
+			break;
+		case 1:
+		default:
+			track->row_size = 2;
+			break;
+		case 2:
+			track->row_size = 4;
+			break;
+		}
+
 		p->track = track;
 	}
 	do {
diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h
index fd38ba4..4215de9 100644
--- a/drivers/gpu/drm/radeon/evergreen_reg.h
+++ b/drivers/gpu/drm/radeon/evergreen_reg.h
@@ -50,6 +50,17 @@
 #       define EVERGREEN_GRPH_DEPTH_8BPP                0
 #       define EVERGREEN_GRPH_DEPTH_16BPP               1
 #       define EVERGREEN_GRPH_DEPTH_32BPP               2
+#       define EVERGREEN_GRPH_NUM_BANKS(x)              (((x) & 0x3) << 2)
+#       define EVERGREEN_ADDR_SURF_2_BANK               0
+#       define EVERGREEN_ADDR_SURF_4_BANK               1
+#       define EVERGREEN_ADDR_SURF_8_BANK               2
+#       define EVERGREEN_ADDR_SURF_16_BANK              3
+#       define EVERGREEN_GRPH_Z(x)                      (((x) & 0x3) << 4)
+#       define EVERGREEN_GRPH_BANK_WIDTH(x)             (((x) & 0x3) << 6)
+#       define EVERGREEN_ADDR_SURF_BANK_WIDTH_1         0
+#       define EVERGREEN_ADDR_SURF_BANK_WIDTH_2         1
+#       define EVERGREEN_ADDR_SURF_BANK_WIDTH_4         2
+#       define EVERGREEN_ADDR_SURF_BANK_WIDTH_8         3
 #       define EVERGREEN_GRPH_FORMAT(x)                 (((x) & 0x7) << 8)
 /* 8 BPP */
 #       define EVERGREEN_GRPH_FORMAT_INDEXED            0
@@ -69,6 +80,24 @@
 #       define EVERGREEN_GRPH_FORMAT_8B_BGRA1010102     5
 #       define EVERGREEN_GRPH_FORMAT_RGB111110          6
 #       define EVERGREEN_GRPH_FORMAT_BGR101111          7
+#       define EVERGREEN_GRPH_BANK_HEIGHT(x)            (((x) & 0x3) << 11)
+#       define EVERGREEN_ADDR_SURF_BANK_HEIGHT_1        0
+#       define EVERGREEN_ADDR_SURF_BANK_HEIGHT_2        1
+#       define EVERGREEN_ADDR_SURF_BANK_HEIGHT_4        2
+#       define EVERGREEN_ADDR_SURF_BANK_HEIGHT_8        3
+#       define EVERGREEN_GRPH_TILE_SPLIT(x)             (((x) & 0x7) << 13)
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_64B       0
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_128B      1
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_256B      2
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_512B      3
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_1KB       4
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_2KB       5
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_4KB       6
+#       define EVERGREEN_GRPH_MACRO_TILE_ASPECT(x)      (((x) & 0x3) << 18)
+#       define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_1  0
+#       define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_2  1
+#       define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_4  2
+#       define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_8  3
 #       define EVERGREEN_GRPH_ARRAY_MODE(x)             (((x) & 0x7) << 20)
 #       define EVERGREEN_GRPH_ARRAY_LINEAR_GENERAL      0
 #       define EVERGREEN_GRPH_ARRAY_LINEAR_ALIGNED      1
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index b937c49..e00039e 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -899,6 +899,10 @@
 #define DB_HTILE_DATA_BASE				0x28014
 #define DB_Z_INFO					0x28040
 #       define Z_ARRAY_MODE(x)                          ((x) << 4)
+#       define DB_TILE_SPLIT(x)                         (((x) & 0x7) << 8)
+#       define DB_NUM_BANKS(x)                          (((x) & 0x3) << 12)
+#       define DB_BANK_WIDTH(x)                         (((x) & 0x3) << 16)
+#       define DB_BANK_HEIGHT(x)                        (((x) & 0x3) << 20)
 #define DB_STENCIL_INFO					0x28044
 #define DB_Z_READ_BASE					0x28048
 #define DB_STENCIL_READ_BASE				0x2804c
@@ -951,6 +955,29 @@
 #	define CB_SF_EXPORT_FULL			0
 #	define CB_SF_EXPORT_NORM			1
 #define	CB_COLOR0_ATTRIB				0x28c74
+#       define CB_TILE_SPLIT(x)                         (((x) & 0x7) << 5)
+#       define ADDR_SURF_TILE_SPLIT_64B                 0
+#       define ADDR_SURF_TILE_SPLIT_128B                1
+#       define ADDR_SURF_TILE_SPLIT_256B                2
+#       define ADDR_SURF_TILE_SPLIT_512B                3
+#       define ADDR_SURF_TILE_SPLIT_1KB                 4
+#       define ADDR_SURF_TILE_SPLIT_2KB                 5
+#       define ADDR_SURF_TILE_SPLIT_4KB                 6
+#       define CB_NUM_BANKS(x)                          (((x) & 0x3) << 10)
+#       define ADDR_SURF_2_BANK                         0
+#       define ADDR_SURF_4_BANK                         1
+#       define ADDR_SURF_8_BANK                         2
+#       define ADDR_SURF_16_BANK                        3
+#       define CB_BANK_WIDTH(x)                         (((x) & 0x3) << 13)
+#       define ADDR_SURF_BANK_WIDTH_1                   0
+#       define ADDR_SURF_BANK_WIDTH_2                   1
+#       define ADDR_SURF_BANK_WIDTH_4                   2
+#       define ADDR_SURF_BANK_WIDTH_8                   3
+#       define CB_BANK_HEIGHT(x)                        (((x) & 0x3) << 16)
+#       define ADDR_SURF_BANK_HEIGHT_1                  0
+#       define ADDR_SURF_BANK_HEIGHT_2                  1
+#       define ADDR_SURF_BANK_HEIGHT_4                  2
+#       define ADDR_SURF_BANK_HEIGHT_8                  3
 #define	CB_COLOR0_DIM					0x28c78
 /* only CB0-7 blocks have these regs */
 #define	CB_COLOR0_CMASK					0x28c7c
@@ -1137,7 +1164,11 @@
 #	define SQ_SEL_1					5
 #define SQ_TEX_RESOURCE_WORD5_0                         0x30014
 #define SQ_TEX_RESOURCE_WORD6_0                         0x30018
+#       define TEX_TILE_SPLIT(x)                        (((x) & 0x7) << 29)
 #define SQ_TEX_RESOURCE_WORD7_0                         0x3001c
+#       define TEX_BANK_WIDTH(x)                        (((x) & 0x3) << 8)
+#       define TEX_BANK_HEIGHT(x)                       (((x) & 0x3) << 10)
+#       define TEX_NUM_BANKS(x)                         (((x) & 0x3) << 16)
 
 #define SQ_VTX_CONSTANT_WORD0_0				0x30000
 #define SQ_VTX_CONSTANT_WORD1_0				0x30004
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 0e57998..d89b2eb 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1006,9 +1006,39 @@
 	radeon_gart_fini(rdev);
 }
 
+void cayman_cp_int_cntl_setup(struct radeon_device *rdev,
+			      int ring, u32 cp_int_cntl)
+{
+	u32 srbm_gfx_cntl = RREG32(SRBM_GFX_CNTL) & ~3;
+
+	WREG32(SRBM_GFX_CNTL, srbm_gfx_cntl | (ring & 3));
+	WREG32(CP_INT_CNTL, cp_int_cntl);
+}
+
 /*
  * CP.
  */
+void cayman_fence_ring_emit(struct radeon_device *rdev,
+			    struct radeon_fence *fence)
+{
+	struct radeon_ring *ring = &rdev->ring[fence->ring];
+	u64 addr = rdev->fence_drv[fence->ring].gpu_addr;
+
+	/* flush read cache over gart */
+	radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
+	radeon_ring_write(ring, PACKET3_TC_ACTION_ENA | PACKET3_SH_ACTION_ENA);
+	radeon_ring_write(ring, 0xFFFFFFFF);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 10); /* poll interval */
+	/* EVENT_WRITE_EOP - flush caches, send int */
+	radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
+	radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT_TS) | EVENT_INDEX(5));
+	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
+	radeon_ring_write(ring, fence->seq);
+	radeon_ring_write(ring, 0);
+}
+
 static void cayman_cp_enable(struct radeon_device *rdev, bool enable)
 {
 	if (enable)
@@ -1049,63 +1079,64 @@
 
 static int cayman_cp_start(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r, i;
 
-	r = radeon_ring_lock(rdev, 7);
+	r = radeon_ring_lock(rdev, ring, 7);
 	if (r) {
 		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
 		return r;
 	}
-	radeon_ring_write(rdev, PACKET3(PACKET3_ME_INITIALIZE, 5));
-	radeon_ring_write(rdev, 0x1);
-	radeon_ring_write(rdev, 0x0);
-	radeon_ring_write(rdev, rdev->config.cayman.max_hw_contexts - 1);
-	radeon_ring_write(rdev, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_write(ring, PACKET3(PACKET3_ME_INITIALIZE, 5));
+	radeon_ring_write(ring, 0x1);
+	radeon_ring_write(ring, 0x0);
+	radeon_ring_write(ring, rdev->config.cayman.max_hw_contexts - 1);
+	radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_unlock_commit(rdev, ring);
 
 	cayman_cp_enable(rdev, true);
 
-	r = radeon_ring_lock(rdev, cayman_default_size + 19);
+	r = radeon_ring_lock(rdev, ring, cayman_default_size + 19);
 	if (r) {
 		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
 		return r;
 	}
 
 	/* setup clear context state */
-	radeon_ring_write(rdev, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	radeon_ring_write(rdev, PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
+	radeon_ring_write(ring, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	radeon_ring_write(ring, PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
 
 	for (i = 0; i < cayman_default_size; i++)
-		radeon_ring_write(rdev, cayman_default_state[i]);
+		radeon_ring_write(ring, cayman_default_state[i]);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	radeon_ring_write(rdev, PACKET3_PREAMBLE_END_CLEAR_STATE);
+	radeon_ring_write(ring, PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	radeon_ring_write(ring, PACKET3_PREAMBLE_END_CLEAR_STATE);
 
 	/* set clear context state */
-	radeon_ring_write(rdev, PACKET3(PACKET3_CLEAR_STATE, 0));
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_CLEAR_STATE, 0));
+	radeon_ring_write(ring, 0);
 
 	/* SQ_VTX_BASE_VTX_LOC */
-	radeon_ring_write(rdev, 0xc0026f00);
-	radeon_ring_write(rdev, 0x00000000);
-	radeon_ring_write(rdev, 0x00000000);
-	radeon_ring_write(rdev, 0x00000000);
+	radeon_ring_write(ring, 0xc0026f00);
+	radeon_ring_write(ring, 0x00000000);
+	radeon_ring_write(ring, 0x00000000);
+	radeon_ring_write(ring, 0x00000000);
 
 	/* Clear consts */
-	radeon_ring_write(rdev, 0xc0036f00);
-	radeon_ring_write(rdev, 0x00000bc4);
-	radeon_ring_write(rdev, 0xffffffff);
-	radeon_ring_write(rdev, 0xffffffff);
-	radeon_ring_write(rdev, 0xffffffff);
+	radeon_ring_write(ring, 0xc0036f00);
+	radeon_ring_write(ring, 0x00000bc4);
+	radeon_ring_write(ring, 0xffffffff);
+	radeon_ring_write(ring, 0xffffffff);
+	radeon_ring_write(ring, 0xffffffff);
 
-	radeon_ring_write(rdev, 0xc0026900);
-	radeon_ring_write(rdev, 0x00000316);
-	radeon_ring_write(rdev, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
-	radeon_ring_write(rdev, 0x00000010); /*  */
+	radeon_ring_write(ring, 0xc0026900);
+	radeon_ring_write(ring, 0x00000316);
+	radeon_ring_write(ring, 0x0000000e); /* VGT_VERTEX_REUSE_BLOCK_CNTL */
+	radeon_ring_write(ring, 0x00000010); /*  */
 
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_unlock_commit(rdev, ring);
 
 	/* XXX init other rings */
 
@@ -1115,11 +1146,12 @@
 static void cayman_cp_fini(struct radeon_device *rdev)
 {
 	cayman_cp_enable(rdev, false);
-	radeon_ring_fini(rdev);
+	radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
 }
 
 int cayman_cp_resume(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring;
 	u32 tmp;
 	u32 rb_bufsz;
 	int r;
@@ -1136,7 +1168,7 @@
 	WREG32(GRBM_SOFT_RESET, 0);
 	RREG32(GRBM_SOFT_RESET);
 
-	WREG32(CP_SEM_WAIT_TIMER, 0x4);
+	WREG32(CP_SEM_WAIT_TIMER, 0x0);
 
 	/* Set the write pointer delay */
 	WREG32(CP_RB_WPTR_DELAY, 0);
@@ -1145,7 +1177,8 @@
 
 	/* ring 0 - compute and gfx */
 	/* Set ring buffer size */
-	rb_bufsz = drm_order(rdev->cp.ring_size / 8);
+	ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+	rb_bufsz = drm_order(ring->ring_size / 8);
 	tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
 #ifdef __BIG_ENDIAN
 	tmp |= BUF_SWAP_32BIT;
@@ -1154,8 +1187,8 @@
 
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB0_CNTL, tmp | RB_RPTR_WR_ENA);
-	rdev->cp.wptr = 0;
-	WREG32(CP_RB0_WPTR, rdev->cp.wptr);
+	ring->wptr = 0;
+	WREG32(CP_RB0_WPTR, ring->wptr);
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB0_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC);
@@ -1172,13 +1205,14 @@
 	mdelay(1);
 	WREG32(CP_RB0_CNTL, tmp);
 
-	WREG32(CP_RB0_BASE, rdev->cp.gpu_addr >> 8);
+	WREG32(CP_RB0_BASE, ring->gpu_addr >> 8);
 
-	rdev->cp.rptr = RREG32(CP_RB0_RPTR);
+	ring->rptr = RREG32(CP_RB0_RPTR);
 
 	/* ring1  - compute only */
 	/* Set ring buffer size */
-	rb_bufsz = drm_order(rdev->cp1.ring_size / 8);
+	ring = &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX];
+	rb_bufsz = drm_order(ring->ring_size / 8);
 	tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
 #ifdef __BIG_ENDIAN
 	tmp |= BUF_SWAP_32BIT;
@@ -1187,8 +1221,8 @@
 
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB1_CNTL, tmp | RB_RPTR_WR_ENA);
-	rdev->cp1.wptr = 0;
-	WREG32(CP_RB1_WPTR, rdev->cp1.wptr);
+	ring->wptr = 0;
+	WREG32(CP_RB1_WPTR, ring->wptr);
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB1_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP1_RPTR_OFFSET) & 0xFFFFFFFC);
@@ -1197,13 +1231,14 @@
 	mdelay(1);
 	WREG32(CP_RB1_CNTL, tmp);
 
-	WREG32(CP_RB1_BASE, rdev->cp1.gpu_addr >> 8);
+	WREG32(CP_RB1_BASE, ring->gpu_addr >> 8);
 
-	rdev->cp1.rptr = RREG32(CP_RB1_RPTR);
+	ring->rptr = RREG32(CP_RB1_RPTR);
 
 	/* ring2 - compute only */
 	/* Set ring buffer size */
-	rb_bufsz = drm_order(rdev->cp2.ring_size / 8);
+	ring = &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX];
+	rb_bufsz = drm_order(ring->ring_size / 8);
 	tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
 #ifdef __BIG_ENDIAN
 	tmp |= BUF_SWAP_32BIT;
@@ -1212,8 +1247,8 @@
 
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB2_CNTL, tmp | RB_RPTR_WR_ENA);
-	rdev->cp2.wptr = 0;
-	WREG32(CP_RB2_WPTR, rdev->cp2.wptr);
+	ring->wptr = 0;
+	WREG32(CP_RB2_WPTR, ring->wptr);
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB2_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP2_RPTR_OFFSET) & 0xFFFFFFFC);
@@ -1222,28 +1257,28 @@
 	mdelay(1);
 	WREG32(CP_RB2_CNTL, tmp);
 
-	WREG32(CP_RB2_BASE, rdev->cp2.gpu_addr >> 8);
+	WREG32(CP_RB2_BASE, ring->gpu_addr >> 8);
 
-	rdev->cp2.rptr = RREG32(CP_RB2_RPTR);
+	ring->rptr = RREG32(CP_RB2_RPTR);
 
 	/* start the rings */
 	cayman_cp_start(rdev);
-	rdev->cp.ready = true;
-	rdev->cp1.ready = true;
-	rdev->cp2.ready = true;
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = true;
+	rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX].ready = false;
+	rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX].ready = false;
 	/* this only test cp0 */
-	r = radeon_ring_test(rdev);
+	r = radeon_ring_test(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
 	if (r) {
-		rdev->cp.ready = false;
-		rdev->cp1.ready = false;
-		rdev->cp2.ready = false;
+		rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
+		rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX].ready = false;
+		rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX].ready = false;
 		return r;
 	}
 
 	return 0;
 }
 
-bool cayman_gpu_is_lockup(struct radeon_device *rdev)
+bool cayman_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	u32 srbm_status;
 	u32 grbm_status;
@@ -1256,20 +1291,20 @@
 	grbm_status_se0 = RREG32(GRBM_STATUS_SE0);
 	grbm_status_se1 = RREG32(GRBM_STATUS_SE1);
 	if (!(grbm_status & GUI_ACTIVE)) {
-		r100_gpu_lockup_update(lockup, &rdev->cp);
+		r100_gpu_lockup_update(lockup, ring);
 		return false;
 	}
 	/* force CP activities */
-	r = radeon_ring_lock(rdev, 2);
+	r = radeon_ring_lock(rdev, ring, 2);
 	if (!r) {
 		/* PACKET2 NOP */
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_unlock_commit(rdev);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_unlock_commit(rdev, ring);
 	}
 	/* XXX deal with CP0,1,2 */
-	rdev->cp.rptr = RREG32(CP_RB0_RPTR);
-	return r100_gpu_cp_is_lockup(rdev, lockup, &rdev->cp);
+	ring->rptr = RREG32(ring->rptr_reg);
+	return r100_gpu_cp_is_lockup(rdev, lockup, ring);
 }
 
 static int cayman_gpu_soft_reset(struct radeon_device *rdev)
@@ -1338,6 +1373,7 @@
 
 static int cayman_startup(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 
 	/* enable pcie gen2 link */
@@ -1378,6 +1414,24 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
+	r = radeon_fence_driver_start_ring(rdev, CAYMAN_RING_TYPE_CP1_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
+	r = radeon_fence_driver_start_ring(rdev, CAYMAN_RING_TYPE_CP2_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	r = r600_irq_init(rdev);
 	if (r) {
@@ -1387,7 +1441,9 @@
 	}
 	evergreen_irq_set(rdev);
 
-	r = radeon_ring_init(rdev, rdev->cp.ring_size);
+	r = radeon_ring_init(rdev, ring, ring->ring_size, RADEON_WB_CP_RPTR_OFFSET,
+			     CP_RB0_RPTR, CP_RB0_WPTR,
+			     0, 0xfffff, RADEON_CP_PACKET2);
 	if (r)
 		return r;
 	r = cayman_cp_load_microcode(rdev);
@@ -1397,6 +1453,17 @@
 	if (r)
 		return r;
 
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r600_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		DRM_ERROR("radeon: failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
+		return r;
+	}
+
 	return 0;
 }
 
@@ -1411,32 +1478,25 @@
 	/* post card */
 	atom_asic_init(rdev->mode_info.atom_context);
 
+	rdev->accel_working = true;
 	r = cayman_startup(rdev);
 	if (r) {
 		DRM_ERROR("cayman startup failed on resume\n");
 		return r;
 	}
-
-	r = r600_ib_test(rdev);
-	if (r) {
-		DRM_ERROR("radeon: failled testing IB (%d).\n", r);
-		return r;
-	}
-
 	return r;
-
 }
 
 int cayman_suspend(struct radeon_device *rdev)
 {
 	/* FIXME: we should wait for ring to be empty */
+	radeon_ib_pool_suspend(rdev);
+	r600_blit_suspend(rdev);
 	cayman_cp_enable(rdev, false);
-	rdev->cp.ready = false;
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
 	evergreen_irq_suspend(rdev);
 	radeon_wb_disable(rdev);
 	cayman_pcie_gart_disable(rdev);
-	r600_blit_suspend(rdev);
-
 	return 0;
 }
 
@@ -1448,6 +1508,7 @@
  */
 int cayman_init(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 
 	/* This don't do much */
@@ -1500,8 +1561,8 @@
 	if (r)
 		return r;
 
-	rdev->cp.ring_obj = NULL;
-	r600_ring_init(rdev, 1024 * 1024);
+	ring->ring_obj = NULL;
+	r600_ring_init(rdev, ring, 1024 * 1024);
 
 	rdev->ih.ring_obj = NULL;
 	r600_ih_ring_init(rdev, 64 * 1024);
@@ -1510,29 +1571,24 @@
 	if (r)
 		return r;
 
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = cayman_startup(rdev);
 	if (r) {
 		dev_err(rdev->dev, "disabling GPU acceleration\n");
 		cayman_cp_fini(rdev);
 		r600_irq_fini(rdev);
 		radeon_wb_fini(rdev);
+		r100_ib_fini(rdev);
 		radeon_irq_kms_fini(rdev);
 		cayman_pcie_gart_fini(rdev);
 		rdev->accel_working = false;
 	}
-	if (rdev->accel_working) {
-		r = radeon_ib_pool_init(rdev);
-		if (r) {
-			DRM_ERROR("radeon: failed initializing IB pool (%d).\n", r);
-			rdev->accel_working = false;
-		}
-		r = r600_ib_test(rdev);
-		if (r) {
-			DRM_ERROR("radeon: failed testing IB (%d).\n", r);
-			rdev->accel_working = false;
-		}
-	}
 
 	/* Don't start up if the MC ucode is missing.
 	 * The default clocks and voltages before the MC ucode
@@ -1552,11 +1608,12 @@
 	cayman_cp_fini(rdev);
 	r600_irq_fini(rdev);
 	radeon_wb_fini(rdev);
-	radeon_ib_pool_fini(rdev);
+	r100_ib_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	cayman_pcie_gart_fini(rdev);
 	r600_vram_scratch_fini(rdev);
 	radeon_gem_fini(rdev);
+	radeon_semaphore_driver_fini(rdev);
 	radeon_fence_driver_fini(rdev);
 	radeon_bo_fini(rdev);
 	radeon_atombios_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/nid.h b/drivers/gpu/drm/radeon/nid.h
index 4672869..0d3f52c 100644
--- a/drivers/gpu/drm/radeon/nid.h
+++ b/drivers/gpu/drm/radeon/nid.h
@@ -42,6 +42,9 @@
 #define CAYMAN_MAX_TCC_MASK          0xFF
 
 #define DMIF_ADDR_CONFIG  				0xBD4
+#define	SRBM_GFX_CNTL				        0x0E44
+#define		RINGID(x)					(((x) & 0x3) << 0)
+#define		VMID(x)						(((x) & 0x7) << 0)
 #define	SRBM_STATUS				        0x0E50
 
 #define VM_CONTEXT0_REQUEST_RESPONSE			0x1470
@@ -394,6 +397,12 @@
 #define	CP_RB0_RPTR_ADDR				0xC10C
 #define	CP_RB0_RPTR_ADDR_HI				0xC110
 #define	CP_RB0_WPTR					0xC114
+
+#define CP_INT_CNTL                                     0xC124
+#       define CNTX_BUSY_INT_ENABLE                     (1 << 19)
+#       define CNTX_EMPTY_INT_ENABLE                    (1 << 20)
+#       define TIME_STAMP_INT_ENABLE                    (1 << 26)
+
 #define	CP_RB1_BASE					0xC180
 #define	CP_RB1_CNTL					0xC184
 #define	CP_RB1_RPTR_ADDR				0xC188
@@ -411,6 +420,10 @@
 #define	CP_ME_RAM_DATA					0xC160
 #define	CP_DEBUG					0xC1FC
 
+#define VGT_EVENT_INITIATOR                             0x28a90
+#       define CACHE_FLUSH_AND_INV_EVENT_TS                     (0x14 << 0)
+#       define CACHE_FLUSH_AND_INV_EVENT                        (0x16 << 0)
+
 /*
  * PM4
  */
@@ -494,7 +507,27 @@
 #define		PACKET3_ME_INITIALIZE_DEVICE_ID(x) ((x) << 16)
 #define	PACKET3_COND_WRITE				0x45
 #define	PACKET3_EVENT_WRITE				0x46
+#define		EVENT_TYPE(x)                           ((x) << 0)
+#define		EVENT_INDEX(x)                          ((x) << 8)
+                /* 0 - any non-TS event
+		 * 1 - ZPASS_DONE
+		 * 2 - SAMPLE_PIPELINESTAT
+		 * 3 - SAMPLE_STREAMOUTSTAT*
+		 * 4 - *S_PARTIAL_FLUSH
+		 * 5 - TS events
+		 */
 #define	PACKET3_EVENT_WRITE_EOP				0x47
+#define		DATA_SEL(x)                             ((x) << 29)
+                /* 0 - discard
+		 * 1 - send low 32bit data
+		 * 2 - send 64bit data
+		 * 3 - send 64bit counter value
+		 */
+#define		INT_SEL(x)                              ((x) << 24)
+                /* 0 - none
+		 * 1 - interrupt only (DATA_SEL = 0)
+		 * 2 - interrupt when data write is confirmed
+		 */
 #define	PACKET3_EVENT_WRITE_EOS				0x48
 #define	PACKET3_PREAMBLE_CNTL				0x4A
 #              define PACKET3_PREAMBLE_BEGIN_CLEAR_STATE     (2 << 28)
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index ad158ea..657040b 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -187,13 +187,18 @@
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = ((u32)crtc_base) | RADEON_CRTC_OFFSET__OFFSET_LOCK;
+	int i;
 
 	/* Lock the graphics update lock */
 	/* update the scanout addresses */
 	WREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset, tmp);
 
 	/* Wait for update_pending to go high. */
-	while (!(RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET));
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		if (RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET)
+			break;
+		udelay(1);
+	}
 	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
@@ -662,7 +667,7 @@
 		WREG32(R_000040_GEN_INT_CNTL, 0);
 		return -EINVAL;
 	}
-	if (rdev->irq.sw_int) {
+	if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
 		tmp |= RADEON_SW_INT_ENABLE;
 	}
 	if (rdev->irq.gui_idle) {
@@ -734,7 +739,7 @@
 	while (status) {
 		/* SW interrupt */
 		if (status & RADEON_SW_INT_TEST) {
-			radeon_fence_process(rdev);
+			radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX);
 		}
 		/* gui idle interrupt */
 		if (status & RADEON_GUI_IDLE_STAT) {
@@ -806,25 +811,36 @@
 void r100_fence_ring_emit(struct radeon_device *rdev,
 			  struct radeon_fence *fence)
 {
+	struct radeon_ring *ring = &rdev->ring[fence->ring];
+
 	/* We have to make sure that caches are flushed before
 	 * CPU might read something from VRAM. */
-	radeon_ring_write(rdev, PACKET0(RADEON_RB3D_DSTCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, RADEON_RB3D_DC_FLUSH_ALL);
-	radeon_ring_write(rdev, PACKET0(RADEON_RB3D_ZCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, RADEON_RB3D_ZC_FLUSH_ALL);
+	radeon_ring_write(ring, PACKET0(RADEON_RB3D_DSTCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, RADEON_RB3D_DC_FLUSH_ALL);
+	radeon_ring_write(ring, PACKET0(RADEON_RB3D_ZCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, RADEON_RB3D_ZC_FLUSH_ALL);
 	/* Wait until IDLE & CLEAN */
-	radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
-	radeon_ring_write(rdev, RADEON_WAIT_2D_IDLECLEAN | RADEON_WAIT_3D_IDLECLEAN);
-	radeon_ring_write(rdev, PACKET0(RADEON_HOST_PATH_CNTL, 0));
-	radeon_ring_write(rdev, rdev->config.r100.hdp_cntl |
+	radeon_ring_write(ring, PACKET0(RADEON_WAIT_UNTIL, 0));
+	radeon_ring_write(ring, RADEON_WAIT_2D_IDLECLEAN | RADEON_WAIT_3D_IDLECLEAN);
+	radeon_ring_write(ring, PACKET0(RADEON_HOST_PATH_CNTL, 0));
+	radeon_ring_write(ring, rdev->config.r100.hdp_cntl |
 				RADEON_HDP_READ_BUFFER_INVALIDATE);
-	radeon_ring_write(rdev, PACKET0(RADEON_HOST_PATH_CNTL, 0));
-	radeon_ring_write(rdev, rdev->config.r100.hdp_cntl);
+	radeon_ring_write(ring, PACKET0(RADEON_HOST_PATH_CNTL, 0));
+	radeon_ring_write(ring, rdev->config.r100.hdp_cntl);
 	/* Emit fence sequence & fire IRQ */
-	radeon_ring_write(rdev, PACKET0(rdev->fence_drv.scratch_reg, 0));
-	radeon_ring_write(rdev, fence->seq);
-	radeon_ring_write(rdev, PACKET0(RADEON_GEN_INT_STATUS, 0));
-	radeon_ring_write(rdev, RADEON_SW_INT_FIRE);
+	radeon_ring_write(ring, PACKET0(rdev->fence_drv[fence->ring].scratch_reg, 0));
+	radeon_ring_write(ring, fence->seq);
+	radeon_ring_write(ring, PACKET0(RADEON_GEN_INT_STATUS, 0));
+	radeon_ring_write(ring, RADEON_SW_INT_FIRE);
+}
+
+void r100_semaphore_ring_emit(struct radeon_device *rdev,
+			      struct radeon_ring *ring,
+			      struct radeon_semaphore *semaphore,
+			      bool emit_wait)
+{
+	/* Unused on older asics, since we don't have semaphores or multiple rings */
+	BUG();
 }
 
 int r100_copy_blit(struct radeon_device *rdev,
@@ -833,6 +849,7 @@
 		   unsigned num_gpu_pages,
 		   struct radeon_fence *fence)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	uint32_t cur_pages;
 	uint32_t stride_bytes = RADEON_GPU_PAGE_SIZE;
 	uint32_t pitch;
@@ -850,7 +867,7 @@
 
 	/* Ask for enough room for blit + flush + fence */
 	ndw = 64 + (10 * num_loops);
-	r = radeon_ring_lock(rdev, ndw);
+	r = radeon_ring_lock(rdev, ring, ndw);
 	if (r) {
 		DRM_ERROR("radeon: moving bo (%d) asking for %u dw.\n", r, ndw);
 		return -EINVAL;
@@ -864,8 +881,8 @@
 
 		/* pages are in Y direction - height
 		   page width in X direction - width */
-		radeon_ring_write(rdev, PACKET3(PACKET3_BITBLT_MULTI, 8));
-		radeon_ring_write(rdev,
+		radeon_ring_write(ring, PACKET3(PACKET3_BITBLT_MULTI, 8));
+		radeon_ring_write(ring,
 				  RADEON_GMC_SRC_PITCH_OFFSET_CNTL |
 				  RADEON_GMC_DST_PITCH_OFFSET_CNTL |
 				  RADEON_GMC_SRC_CLIPPING |
@@ -877,26 +894,26 @@
 				  RADEON_DP_SRC_SOURCE_MEMORY |
 				  RADEON_GMC_CLR_CMP_CNTL_DIS |
 				  RADEON_GMC_WR_MSK_DIS);
-		radeon_ring_write(rdev, (pitch << 22) | (src_offset >> 10));
-		radeon_ring_write(rdev, (pitch << 22) | (dst_offset >> 10));
-		radeon_ring_write(rdev, (0x1fff) | (0x1fff << 16));
-		radeon_ring_write(rdev, 0);
-		radeon_ring_write(rdev, (0x1fff) | (0x1fff << 16));
-		radeon_ring_write(rdev, num_gpu_pages);
-		radeon_ring_write(rdev, num_gpu_pages);
-		radeon_ring_write(rdev, cur_pages | (stride_pixels << 16));
+		radeon_ring_write(ring, (pitch << 22) | (src_offset >> 10));
+		radeon_ring_write(ring, (pitch << 22) | (dst_offset >> 10));
+		radeon_ring_write(ring, (0x1fff) | (0x1fff << 16));
+		radeon_ring_write(ring, 0);
+		radeon_ring_write(ring, (0x1fff) | (0x1fff << 16));
+		radeon_ring_write(ring, num_gpu_pages);
+		radeon_ring_write(ring, num_gpu_pages);
+		radeon_ring_write(ring, cur_pages | (stride_pixels << 16));
 	}
-	radeon_ring_write(rdev, PACKET0(RADEON_DSTCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, RADEON_RB2D_DC_FLUSH_ALL);
-	radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(RADEON_DSTCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, RADEON_RB2D_DC_FLUSH_ALL);
+	radeon_ring_write(ring, PACKET0(RADEON_WAIT_UNTIL, 0));
+	radeon_ring_write(ring,
 			  RADEON_WAIT_2D_IDLECLEAN |
 			  RADEON_WAIT_HOST_IDLECLEAN |
 			  RADEON_WAIT_DMA_GUI_IDLE);
 	if (fence) {
 		r = radeon_fence_emit(rdev, fence);
 	}
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_unlock_commit(rdev, ring);
 	return r;
 }
 
@@ -917,19 +934,20 @@
 
 void r100_ring_start(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 
-	r = radeon_ring_lock(rdev, 2);
+	r = radeon_ring_lock(rdev, ring, 2);
 	if (r) {
 		return;
 	}
-	radeon_ring_write(rdev, PACKET0(RADEON_ISYNC_CNTL, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(RADEON_ISYNC_CNTL, 0));
+	radeon_ring_write(ring,
 			  RADEON_ISYNC_ANY2D_IDLE3D |
 			  RADEON_ISYNC_ANY3D_IDLE2D |
 			  RADEON_ISYNC_WAIT_IDLEGUI |
 			  RADEON_ISYNC_CPSCRATCH_IDLEGUI);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_unlock_commit(rdev, ring);
 }
 
 
@@ -1030,6 +1048,7 @@
 
 int r100_cp_init(struct radeon_device *rdev, unsigned ring_size)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	unsigned rb_bufsz;
 	unsigned rb_blksz;
 	unsigned max_fetch;
@@ -1055,7 +1074,9 @@
 	rb_bufsz = drm_order(ring_size / 8);
 	ring_size = (1 << (rb_bufsz + 1)) * 4;
 	r100_cp_load_microcode(rdev);
-	r = radeon_ring_init(rdev, ring_size);
+	r = radeon_ring_init(rdev, ring, ring_size, RADEON_WB_CP_RPTR_OFFSET,
+			     RADEON_CP_RB_RPTR, RADEON_CP_RB_WPTR,
+			     0, 0x7fffff, RADEON_CP_PACKET2);
 	if (r) {
 		return r;
 	}
@@ -1064,7 +1085,7 @@
 	rb_blksz = 9;
 	/* cp will read 128bytes at a time (4 dwords) */
 	max_fetch = 1;
-	rdev->cp.align_mask = 16 - 1;
+	ring->align_mask = 16 - 1;
 	/* Write to CP_RB_WPTR will be delayed for pre_write_timer clocks */
 	pre_write_timer = 64;
 	/* Force CP_RB_WPTR write if written more than one time before the
@@ -1094,13 +1115,13 @@
 	WREG32(RADEON_CP_RB_CNTL, tmp | RADEON_RB_NO_UPDATE);
 
 	/* Set ring address */
-	DRM_INFO("radeon: ring at 0x%016lX\n", (unsigned long)rdev->cp.gpu_addr);
-	WREG32(RADEON_CP_RB_BASE, rdev->cp.gpu_addr);
+	DRM_INFO("radeon: ring at 0x%016lX\n", (unsigned long)ring->gpu_addr);
+	WREG32(RADEON_CP_RB_BASE, ring->gpu_addr);
 	/* Force read & write ptr to 0 */
 	WREG32(RADEON_CP_RB_CNTL, tmp | RADEON_RB_RPTR_WR_ENA | RADEON_RB_NO_UPDATE);
 	WREG32(RADEON_CP_RB_RPTR_WR, 0);
-	rdev->cp.wptr = 0;
-	WREG32(RADEON_CP_RB_WPTR, rdev->cp.wptr);
+	ring->wptr = 0;
+	WREG32(RADEON_CP_RB_WPTR, ring->wptr);
 
 	/* set the wb address whether it's enabled or not */
 	WREG32(R_00070C_CP_RB_RPTR_ADDR,
@@ -1116,7 +1137,7 @@
 
 	WREG32(RADEON_CP_RB_CNTL, tmp);
 	udelay(10);
-	rdev->cp.rptr = RREG32(RADEON_CP_RB_RPTR);
+	ring->rptr = RREG32(RADEON_CP_RB_RPTR);
 	/* Set cp mode to bus mastering & enable cp*/
 	WREG32(RADEON_CP_CSQ_MODE,
 	       REG_SET(RADEON_INDIRECT2_START, indirect2_start) |
@@ -1125,12 +1146,12 @@
 	WREG32(RADEON_CP_CSQ_MODE, 0x00004D4D);
 	WREG32(RADEON_CP_CSQ_CNTL, RADEON_CSQ_PRIBM_INDBM);
 	radeon_ring_start(rdev);
-	r = radeon_ring_test(rdev);
+	r = radeon_ring_test(rdev, ring);
 	if (r) {
 		DRM_ERROR("radeon: cp isn't working (%d).\n", r);
 		return r;
 	}
-	rdev->cp.ready = true;
+	ring->ready = true;
 	radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
 	return 0;
 }
@@ -1142,7 +1163,7 @@
 	}
 	/* Disable ring */
 	r100_cp_disable(rdev);
-	radeon_ring_fini(rdev);
+	radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
 	DRM_INFO("radeon: cp finalized\n");
 }
 
@@ -1150,7 +1171,7 @@
 {
 	/* Disable ring */
 	radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size);
-	rdev->cp.ready = false;
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
 	WREG32(RADEON_CP_CSQ_MODE, 0);
 	WREG32(RADEON_CP_CSQ_CNTL, 0);
 	WREG32(R_000770_SCRATCH_UMSK, 0);
@@ -1160,13 +1181,6 @@
 	}
 }
 
-void r100_cp_commit(struct radeon_device *rdev)
-{
-	WREG32(RADEON_CP_RB_WPTR, rdev->cp.wptr);
-	(void)RREG32(RADEON_CP_RB_WPTR);
-}
-
-
 /*
  * CS functions
  */
@@ -2094,9 +2108,9 @@
 	return -1;
 }
 
-void r100_gpu_lockup_update(struct r100_gpu_lockup *lockup, struct radeon_cp *cp)
+void r100_gpu_lockup_update(struct r100_gpu_lockup *lockup, struct radeon_ring *ring)
 {
-	lockup->last_cp_rptr = cp->rptr;
+	lockup->last_cp_rptr = ring->rptr;
 	lockup->last_jiffies = jiffies;
 }
 
@@ -2121,20 +2135,20 @@
  * false positive when CP is just gived nothing to do.
  *
  **/
-bool r100_gpu_cp_is_lockup(struct radeon_device *rdev, struct r100_gpu_lockup *lockup, struct radeon_cp *cp)
+bool r100_gpu_cp_is_lockup(struct radeon_device *rdev, struct r100_gpu_lockup *lockup, struct radeon_ring *ring)
 {
 	unsigned long cjiffies, elapsed;
 
 	cjiffies = jiffies;
 	if (!time_after(cjiffies, lockup->last_jiffies)) {
 		/* likely a wrap around */
-		lockup->last_cp_rptr = cp->rptr;
+		lockup->last_cp_rptr = ring->rptr;
 		lockup->last_jiffies = jiffies;
 		return false;
 	}
-	if (cp->rptr != lockup->last_cp_rptr) {
+	if (ring->rptr != lockup->last_cp_rptr) {
 		/* CP is still working no lockup */
-		lockup->last_cp_rptr = cp->rptr;
+		lockup->last_cp_rptr = ring->rptr;
 		lockup->last_jiffies = jiffies;
 		return false;
 	}
@@ -2147,26 +2161,26 @@
 	return false;
 }
 
-bool r100_gpu_is_lockup(struct radeon_device *rdev)
+bool r100_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	u32 rbbm_status;
 	int r;
 
 	rbbm_status = RREG32(R_000E40_RBBM_STATUS);
 	if (!G_000E40_GUI_ACTIVE(rbbm_status)) {
-		r100_gpu_lockup_update(&rdev->config.r100.lockup, &rdev->cp);
+		r100_gpu_lockup_update(&rdev->config.r100.lockup, ring);
 		return false;
 	}
 	/* force CP activities */
-	r = radeon_ring_lock(rdev, 2);
+	r = radeon_ring_lock(rdev, ring, 2);
 	if (!r) {
 		/* PACKET2 NOP */
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_unlock_commit(rdev);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_unlock_commit(rdev, ring);
 	}
-	rdev->cp.rptr = RREG32(RADEON_CP_RB_RPTR);
-	return r100_gpu_cp_is_lockup(rdev, &rdev->config.r100.lockup, &rdev->cp);
+	ring->rptr = RREG32(ring->rptr_reg);
+	return r100_gpu_cp_is_lockup(rdev, &rdev->config.r100.lockup, ring);
 }
 
 void r100_bm_disable(struct radeon_device *rdev)
@@ -2574,21 +2588,22 @@
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	uint32_t rdp, wdp;
 	unsigned count, i, j;
 
-	radeon_ring_free_size(rdev);
+	radeon_ring_free_size(rdev, ring);
 	rdp = RREG32(RADEON_CP_RB_RPTR);
 	wdp = RREG32(RADEON_CP_RB_WPTR);
-	count = (rdp + rdev->cp.ring_size - wdp) & rdev->cp.ptr_mask;
+	count = (rdp + ring->ring_size - wdp) & ring->ptr_mask;
 	seq_printf(m, "CP_STAT 0x%08x\n", RREG32(RADEON_CP_STAT));
 	seq_printf(m, "CP_RB_WPTR 0x%08x\n", wdp);
 	seq_printf(m, "CP_RB_RPTR 0x%08x\n", rdp);
-	seq_printf(m, "%u free dwords in ring\n", rdev->cp.ring_free_dw);
+	seq_printf(m, "%u free dwords in ring\n", ring->ring_free_dw);
 	seq_printf(m, "%u dwords in ring\n", count);
 	for (j = 0; j <= count; j++) {
-		i = (rdp + j) & rdev->cp.ptr_mask;
-		seq_printf(m, "r[%04d]=0x%08x\n", i, rdev->cp.ring[i]);
+		i = (rdp + j) & ring->ptr_mask;
+		seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]);
 	}
 	return 0;
 }
@@ -3630,7 +3645,7 @@
 	}
 }
 
-int r100_ring_test(struct radeon_device *rdev)
+int r100_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	uint32_t scratch;
 	uint32_t tmp = 0;
@@ -3643,15 +3658,15 @@
 		return r;
 	}
 	WREG32(scratch, 0xCAFEDEAD);
-	r = radeon_ring_lock(rdev, 2);
+	r = radeon_ring_lock(rdev, ring, 2);
 	if (r) {
 		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
 		radeon_scratch_free(rdev, scratch);
 		return r;
 	}
-	radeon_ring_write(rdev, PACKET0(scratch, 0));
-	radeon_ring_write(rdev, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_write(ring, PACKET0(scratch, 0));
+	radeon_ring_write(ring, 0xDEADBEEF);
+	radeon_ring_unlock_commit(rdev, ring);
 	for (i = 0; i < rdev->usec_timeout; i++) {
 		tmp = RREG32(scratch);
 		if (tmp == 0xDEADBEEF) {
@@ -3672,9 +3687,11 @@
 
 void r100_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
-	radeon_ring_write(rdev, PACKET0(RADEON_CP_IB_BASE, 1));
-	radeon_ring_write(rdev, ib->gpu_addr);
-	radeon_ring_write(rdev, ib->length_dw);
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+
+	radeon_ring_write(ring, PACKET0(RADEON_CP_IB_BASE, 1));
+	radeon_ring_write(ring, ib->gpu_addr);
+	radeon_ring_write(ring, ib->length_dw);
 }
 
 int r100_ib_test(struct radeon_device *rdev)
@@ -3691,7 +3708,7 @@
 		return r;
 	}
 	WREG32(scratch, 0xCAFEDEAD);
-	r = radeon_ib_get(rdev, &ib);
+	r = radeon_ib_get(rdev, RADEON_RING_TYPE_GFX_INDEX, &ib);
 	if (r) {
 		return r;
 	}
@@ -3735,34 +3752,16 @@
 
 void r100_ib_fini(struct radeon_device *rdev)
 {
+	radeon_ib_pool_suspend(rdev);
 	radeon_ib_pool_fini(rdev);
 }
 
-int r100_ib_init(struct radeon_device *rdev)
-{
-	int r;
-
-	r = radeon_ib_pool_init(rdev);
-	if (r) {
-		dev_err(rdev->dev, "failed initializing IB pool (%d).\n", r);
-		r100_ib_fini(rdev);
-		return r;
-	}
-	r = r100_ib_test(rdev);
-	if (r) {
-		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
-		r100_ib_fini(rdev);
-		return r;
-	}
-	return 0;
-}
-
 void r100_mc_stop(struct radeon_device *rdev, struct r100_mc_save *save)
 {
 	/* Shutdown CP we shouldn't need to do that but better be safe than
 	 * sorry
 	 */
-	rdev->cp.ready = false;
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
 	WREG32(R_000740_CP_CSQ_CNTL, 0);
 
 	/* Save few CRTC registers */
@@ -3900,6 +3899,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	r100_irq_set(rdev);
 	rdev->config.r100.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
@@ -3909,11 +3914,18 @@
 		dev_err(rdev->dev, "failed initializing CP (%d).\n", r);
 		return r;
 	}
-	r = r100_ib_init(rdev);
+
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r100_ib_test(rdev);
 	if (r) {
-		dev_err(rdev->dev, "failed initializing IB (%d).\n", r);
+		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
 		return r;
 	}
+
 	return 0;
 }
 
@@ -3936,11 +3948,14 @@
 	r100_clock_startup(rdev);
 	/* Initialize surface registers */
 	radeon_surface_init(rdev);
+
+	rdev->accel_working = true;
 	return r100_startup(rdev);
 }
 
 int r100_suspend(struct radeon_device *rdev)
 {
+	radeon_ib_pool_suspend(rdev);
 	r100_cp_disable(rdev);
 	radeon_wb_disable(rdev);
 	r100_irq_disable(rdev);
@@ -4059,7 +4074,14 @@
 			return r;
 	}
 	r100_set_safe_registers(rdev);
+
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = r100_startup(rdev);
 	if (r) {
 		/* Somethings want wront with the accel init stop accel */
diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index a1f3ba0..eba4cbf 100644
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -87,6 +87,7 @@
 		  unsigned num_gpu_pages,
 		  struct radeon_fence *fence)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	uint32_t size;
 	uint32_t cur_size;
 	int i, num_loops;
@@ -95,33 +96,33 @@
 	/* radeon pitch is /64 */
 	size = num_gpu_pages << RADEON_GPU_PAGE_SHIFT;
 	num_loops = DIV_ROUND_UP(size, 0x1FFFFF);
-	r = radeon_ring_lock(rdev, num_loops * 4 + 64);
+	r = radeon_ring_lock(rdev, ring, num_loops * 4 + 64);
 	if (r) {
 		DRM_ERROR("radeon: moving bo (%d).\n", r);
 		return r;
 	}
 	/* Must wait for 2D idle & clean before DMA or hangs might happen */
-	radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
-	radeon_ring_write(rdev, (1 << 16));
+	radeon_ring_write(ring, PACKET0(RADEON_WAIT_UNTIL, 0));
+	radeon_ring_write(ring, (1 << 16));
 	for (i = 0; i < num_loops; i++) {
 		cur_size = size;
 		if (cur_size > 0x1FFFFF) {
 			cur_size = 0x1FFFFF;
 		}
 		size -= cur_size;
-		radeon_ring_write(rdev, PACKET0(0x720, 2));
-		radeon_ring_write(rdev, src_offset);
-		radeon_ring_write(rdev, dst_offset);
-		radeon_ring_write(rdev, cur_size | (1 << 31) | (1 << 30));
+		radeon_ring_write(ring, PACKET0(0x720, 2));
+		radeon_ring_write(ring, src_offset);
+		radeon_ring_write(ring, dst_offset);
+		radeon_ring_write(ring, cur_size | (1 << 31) | (1 << 30));
 		src_offset += cur_size;
 		dst_offset += cur_size;
 	}
-	radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
-	radeon_ring_write(rdev, RADEON_WAIT_DMA_GUI_IDLE);
+	radeon_ring_write(ring, PACKET0(RADEON_WAIT_UNTIL, 0));
+	radeon_ring_write(ring, RADEON_WAIT_DMA_GUI_IDLE);
 	if (fence) {
 		r = radeon_fence_emit(rdev, fence);
 	}
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_unlock_commit(rdev, ring);
 	return r;
 }
 
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 400b26d..8ad5c64 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -175,37 +175,40 @@
 void r300_fence_ring_emit(struct radeon_device *rdev,
 			  struct radeon_fence *fence)
 {
+	struct radeon_ring *ring = &rdev->ring[fence->ring];
+
 	/* Who ever call radeon_fence_emit should call ring_lock and ask
 	 * for enough space (today caller are ib schedule and buffer move) */
 	/* Write SC register so SC & US assert idle */
-	radeon_ring_write(rdev, PACKET0(R300_RE_SCISSORS_TL, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, PACKET0(R300_RE_SCISSORS_BR, 0));
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET0(R300_RE_SCISSORS_TL, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, PACKET0(R300_RE_SCISSORS_BR, 0));
+	radeon_ring_write(ring, 0);
 	/* Flush 3D cache */
-	radeon_ring_write(rdev, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, R300_RB3D_DC_FLUSH);
-	radeon_ring_write(rdev, PACKET0(R300_RB3D_ZCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, R300_ZC_FLUSH);
+	radeon_ring_write(ring, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, R300_RB3D_DC_FLUSH);
+	radeon_ring_write(ring, PACKET0(R300_RB3D_ZCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, R300_ZC_FLUSH);
 	/* Wait until IDLE & CLEAN */
-	radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
-	radeon_ring_write(rdev, (RADEON_WAIT_3D_IDLECLEAN |
+	radeon_ring_write(ring, PACKET0(RADEON_WAIT_UNTIL, 0));
+	radeon_ring_write(ring, (RADEON_WAIT_3D_IDLECLEAN |
 				 RADEON_WAIT_2D_IDLECLEAN |
 				 RADEON_WAIT_DMA_GUI_IDLE));
-	radeon_ring_write(rdev, PACKET0(RADEON_HOST_PATH_CNTL, 0));
-	radeon_ring_write(rdev, rdev->config.r300.hdp_cntl |
+	radeon_ring_write(ring, PACKET0(RADEON_HOST_PATH_CNTL, 0));
+	radeon_ring_write(ring, rdev->config.r300.hdp_cntl |
 				RADEON_HDP_READ_BUFFER_INVALIDATE);
-	radeon_ring_write(rdev, PACKET0(RADEON_HOST_PATH_CNTL, 0));
-	radeon_ring_write(rdev, rdev->config.r300.hdp_cntl);
+	radeon_ring_write(ring, PACKET0(RADEON_HOST_PATH_CNTL, 0));
+	radeon_ring_write(ring, rdev->config.r300.hdp_cntl);
 	/* Emit fence sequence & fire IRQ */
-	radeon_ring_write(rdev, PACKET0(rdev->fence_drv.scratch_reg, 0));
-	radeon_ring_write(rdev, fence->seq);
-	radeon_ring_write(rdev, PACKET0(RADEON_GEN_INT_STATUS, 0));
-	radeon_ring_write(rdev, RADEON_SW_INT_FIRE);
+	radeon_ring_write(ring, PACKET0(rdev->fence_drv[fence->ring].scratch_reg, 0));
+	radeon_ring_write(ring, fence->seq);
+	radeon_ring_write(ring, PACKET0(RADEON_GEN_INT_STATUS, 0));
+	radeon_ring_write(ring, RADEON_SW_INT_FIRE);
 }
 
 void r300_ring_start(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	unsigned gb_tile_config;
 	int r;
 
@@ -227,44 +230,44 @@
 		break;
 	}
 
-	r = radeon_ring_lock(rdev, 64);
+	r = radeon_ring_lock(rdev, ring, 64);
 	if (r) {
 		return;
 	}
-	radeon_ring_write(rdev, PACKET0(RADEON_ISYNC_CNTL, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(RADEON_ISYNC_CNTL, 0));
+	radeon_ring_write(ring,
 			  RADEON_ISYNC_ANY2D_IDLE3D |
 			  RADEON_ISYNC_ANY3D_IDLE2D |
 			  RADEON_ISYNC_WAIT_IDLEGUI |
 			  RADEON_ISYNC_CPSCRATCH_IDLEGUI);
-	radeon_ring_write(rdev, PACKET0(R300_GB_TILE_CONFIG, 0));
-	radeon_ring_write(rdev, gb_tile_config);
-	radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(R300_GB_TILE_CONFIG, 0));
+	radeon_ring_write(ring, gb_tile_config);
+	radeon_ring_write(ring, PACKET0(RADEON_WAIT_UNTIL, 0));
+	radeon_ring_write(ring,
 			  RADEON_WAIT_2D_IDLECLEAN |
 			  RADEON_WAIT_3D_IDLECLEAN);
-	radeon_ring_write(rdev, PACKET0(R300_DST_PIPE_CONFIG, 0));
-	radeon_ring_write(rdev, R300_PIPE_AUTO_CONFIG);
-	radeon_ring_write(rdev, PACKET0(R300_GB_SELECT, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, PACKET0(R300_GB_ENABLE, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, R300_RB3D_DC_FLUSH | R300_RB3D_DC_FREE);
-	radeon_ring_write(rdev, PACKET0(R300_RB3D_ZCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, R300_ZC_FLUSH | R300_ZC_FREE);
-	radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(R300_DST_PIPE_CONFIG, 0));
+	radeon_ring_write(ring, R300_PIPE_AUTO_CONFIG);
+	radeon_ring_write(ring, PACKET0(R300_GB_SELECT, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, PACKET0(R300_GB_ENABLE, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, R300_RB3D_DC_FLUSH | R300_RB3D_DC_FREE);
+	radeon_ring_write(ring, PACKET0(R300_RB3D_ZCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, R300_ZC_FLUSH | R300_ZC_FREE);
+	radeon_ring_write(ring, PACKET0(RADEON_WAIT_UNTIL, 0));
+	radeon_ring_write(ring,
 			  RADEON_WAIT_2D_IDLECLEAN |
 			  RADEON_WAIT_3D_IDLECLEAN);
-	radeon_ring_write(rdev, PACKET0(R300_GB_AA_CONFIG, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, R300_RB3D_DC_FLUSH | R300_RB3D_DC_FREE);
-	radeon_ring_write(rdev, PACKET0(R300_RB3D_ZCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, R300_ZC_FLUSH | R300_ZC_FREE);
-	radeon_ring_write(rdev, PACKET0(R300_GB_MSPOS0, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(R300_GB_AA_CONFIG, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, R300_RB3D_DC_FLUSH | R300_RB3D_DC_FREE);
+	radeon_ring_write(ring, PACKET0(R300_RB3D_ZCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, R300_ZC_FLUSH | R300_ZC_FREE);
+	radeon_ring_write(ring, PACKET0(R300_GB_MSPOS0, 0));
+	radeon_ring_write(ring,
 			  ((6 << R300_MS_X0_SHIFT) |
 			   (6 << R300_MS_Y0_SHIFT) |
 			   (6 << R300_MS_X1_SHIFT) |
@@ -273,8 +276,8 @@
 			   (6 << R300_MS_Y2_SHIFT) |
 			   (6 << R300_MSBD0_Y_SHIFT) |
 			   (6 << R300_MSBD0_X_SHIFT)));
-	radeon_ring_write(rdev, PACKET0(R300_GB_MSPOS1, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(R300_GB_MSPOS1, 0));
+	radeon_ring_write(ring,
 			  ((6 << R300_MS_X3_SHIFT) |
 			   (6 << R300_MS_Y3_SHIFT) |
 			   (6 << R300_MS_X4_SHIFT) |
@@ -282,16 +285,16 @@
 			   (6 << R300_MS_X5_SHIFT) |
 			   (6 << R300_MS_Y5_SHIFT) |
 			   (6 << R300_MSBD1_SHIFT)));
-	radeon_ring_write(rdev, PACKET0(R300_GA_ENHANCE, 0));
-	radeon_ring_write(rdev, R300_GA_DEADLOCK_CNTL | R300_GA_FASTSYNC_CNTL);
-	radeon_ring_write(rdev, PACKET0(R300_GA_POLY_MODE, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(R300_GA_ENHANCE, 0));
+	radeon_ring_write(ring, R300_GA_DEADLOCK_CNTL | R300_GA_FASTSYNC_CNTL);
+	radeon_ring_write(ring, PACKET0(R300_GA_POLY_MODE, 0));
+	radeon_ring_write(ring,
 			  R300_FRONT_PTYPE_TRIANGE | R300_BACK_PTYPE_TRIANGE);
-	radeon_ring_write(rdev, PACKET0(R300_GA_ROUND_MODE, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(R300_GA_ROUND_MODE, 0));
+	radeon_ring_write(ring,
 			  R300_GEOMETRY_ROUND_NEAREST |
 			  R300_COLOR_ROUND_NEAREST);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_unlock_commit(rdev, ring);
 }
 
 void r300_errata(struct radeon_device *rdev)
@@ -375,26 +378,26 @@
 		 rdev->num_gb_pipes, rdev->num_z_pipes);
 }
 
-bool r300_gpu_is_lockup(struct radeon_device *rdev)
+bool r300_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	u32 rbbm_status;
 	int r;
 
 	rbbm_status = RREG32(R_000E40_RBBM_STATUS);
 	if (!G_000E40_GUI_ACTIVE(rbbm_status)) {
-		r100_gpu_lockup_update(&rdev->config.r300.lockup, &rdev->cp);
+		r100_gpu_lockup_update(&rdev->config.r300.lockup, ring);
 		return false;
 	}
 	/* force CP activities */
-	r = radeon_ring_lock(rdev, 2);
+	r = radeon_ring_lock(rdev, ring, 2);
 	if (!r) {
 		/* PACKET2 NOP */
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_unlock_commit(rdev);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_unlock_commit(rdev, ring);
 	}
-	rdev->cp.rptr = RREG32(RADEON_CP_RB_RPTR);
-	return r100_gpu_cp_is_lockup(rdev, &rdev->config.r300.lockup, &rdev->cp);
+	ring->rptr = RREG32(RADEON_CP_RB_RPTR);
+	return r100_gpu_cp_is_lockup(rdev, &rdev->config.r300.lockup, ring);
 }
 
 int r300_asic_reset(struct radeon_device *rdev)
@@ -701,16 +704,21 @@
 			return r;
 		}
 
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-			tile_flags |= R300_TXO_MACRO_TILE;
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-			tile_flags |= R300_TXO_MICRO_TILE;
-		else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
-			tile_flags |= R300_TXO_MICRO_TILE_SQUARE;
+		if (p->keep_tiling_flags) {
+			ib[idx] = (idx_value & 31) | /* keep the 1st 5 bits */
+				  ((idx_value & ~31) + (u32)reloc->lobj.gpu_offset);
+		} else {
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+				tile_flags |= R300_TXO_MACRO_TILE;
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+				tile_flags |= R300_TXO_MICRO_TILE;
+			else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
+				tile_flags |= R300_TXO_MICRO_TILE_SQUARE;
 
-		tmp = idx_value + ((u32)reloc->lobj.gpu_offset);
-		tmp |= tile_flags;
-		ib[idx] = tmp;
+			tmp = idx_value + ((u32)reloc->lobj.gpu_offset);
+			tmp |= tile_flags;
+			ib[idx] = tmp;
+		}
 		track->textures[i].robj = reloc->robj;
 		track->tex_dirty = true;
 		break;
@@ -760,24 +768,26 @@
 		/* RB3D_COLORPITCH1 */
 		/* RB3D_COLORPITCH2 */
 		/* RB3D_COLORPITCH3 */
-		r = r100_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
-			return r;
+		if (!p->keep_tiling_flags) {
+			r = r100_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
+					  idx, reg);
+				r100_cs_dump_packet(p, pkt);
+				return r;
+			}
+
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+				tile_flags |= R300_COLOR_TILE_ENABLE;
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+				tile_flags |= R300_COLOR_MICROTILE_ENABLE;
+			else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
+				tile_flags |= R300_COLOR_MICROTILE_SQUARE_ENABLE;
+
+			tmp = idx_value & ~(0x7 << 16);
+			tmp |= tile_flags;
+			ib[idx] = tmp;
 		}
-
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-			tile_flags |= R300_COLOR_TILE_ENABLE;
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-			tile_flags |= R300_COLOR_MICROTILE_ENABLE;
-		else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
-			tile_flags |= R300_COLOR_MICROTILE_SQUARE_ENABLE;
-
-		tmp = idx_value & ~(0x7 << 16);
-		tmp |= tile_flags;
-		ib[idx] = tmp;
 		i = (reg - 0x4E38) >> 2;
 		track->cb[i].pitch = idx_value & 0x3FFE;
 		switch (((idx_value >> 21) & 0xF)) {
@@ -843,25 +853,26 @@
 		break;
 	case 0x4F24:
 		/* ZB_DEPTHPITCH */
-		r = r100_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
-			return r;
+		if (!p->keep_tiling_flags) {
+			r = r100_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
+					  idx, reg);
+				r100_cs_dump_packet(p, pkt);
+				return r;
+			}
+
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+				tile_flags |= R300_DEPTHMACROTILE_ENABLE;
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+				tile_flags |= R300_DEPTHMICROTILE_TILED;
+			else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
+				tile_flags |= R300_DEPTHMICROTILE_TILED_SQUARE;
+
+			tmp = idx_value & ~(0x7 << 16);
+			tmp |= tile_flags;
+			ib[idx] = tmp;
 		}
-
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-			tile_flags |= R300_DEPTHMACROTILE_ENABLE;
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-			tile_flags |= R300_DEPTHMICROTILE_TILED;
-		else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
-			tile_flags |= R300_DEPTHMICROTILE_TILED_SQUARE;
-
-		tmp = idx_value & ~(0x7 << 16);
-		tmp |= tile_flags;
-		ib[idx] = tmp;
-
 		track->zb.pitch = idx_value & 0x3FFC;
 		track->zb_dirty = true;
 		break;
@@ -1388,6 +1399,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	r100_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
@@ -1397,11 +1414,18 @@
 		dev_err(rdev->dev, "failed initializing CP (%d).\n", r);
 		return r;
 	}
-	r = r100_ib_init(rdev);
+
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r100_ib_test(rdev);
 	if (r) {
-		dev_err(rdev->dev, "failed initializing IB (%d).\n", r);
+		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
 		return r;
 	}
+
 	return 0;
 }
 
@@ -1426,11 +1450,14 @@
 	r300_clock_startup(rdev);
 	/* Initialize surface registers */
 	radeon_surface_init(rdev);
+
+	rdev->accel_working = true;
 	return r300_startup(rdev);
 }
 
 int r300_suspend(struct radeon_device *rdev)
 {
+	radeon_ib_pool_suspend(rdev);
 	r100_cp_disable(rdev);
 	radeon_wb_disable(rdev);
 	r100_irq_disable(rdev);
@@ -1531,7 +1558,14 @@
 			return r;
 	}
 	r300_set_reg_safe(rdev);
+
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = r300_startup(rdev);
 	if (r) {
 		/* Somethings want wront with the accel init stop accel */
diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index 417fab8..666e28f 100644
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -199,6 +199,8 @@
 
 static void r420_cp_errata_init(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+
 	/* RV410 and R420 can lock up if CP DMA to host memory happens
 	 * while the 2D engine is busy.
 	 *
@@ -206,22 +208,24 @@
 	 * of the CP init, apparently.
 	 */
 	radeon_scratch_get(rdev, &rdev->config.r300.resync_scratch);
-	radeon_ring_lock(rdev, 8);
-	radeon_ring_write(rdev, PACKET0(R300_CP_RESYNC_ADDR, 1));
-	radeon_ring_write(rdev, rdev->config.r300.resync_scratch);
-	radeon_ring_write(rdev, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_lock(rdev, ring, 8);
+	radeon_ring_write(ring, PACKET0(R300_CP_RESYNC_ADDR, 1));
+	radeon_ring_write(ring, rdev->config.r300.resync_scratch);
+	radeon_ring_write(ring, 0xDEADBEEF);
+	radeon_ring_unlock_commit(rdev, ring);
 }
 
 static void r420_cp_errata_fini(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+
 	/* Catch the RESYNC we dispatched all the way back,
 	 * at the very beginning of the CP init.
 	 */
-	radeon_ring_lock(rdev, 8);
-	radeon_ring_write(rdev, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, R300_RB3D_DC_FINISH);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_lock(rdev, ring, 8);
+	radeon_ring_write(ring, PACKET0(R300_RB3D_DSTCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, R300_RB3D_DC_FINISH);
+	radeon_ring_unlock_commit(rdev, ring);
 	radeon_scratch_free(rdev, rdev->config.r300.resync_scratch);
 }
 
@@ -254,6 +258,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	r100_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
@@ -264,11 +274,18 @@
 		return r;
 	}
 	r420_cp_errata_init(rdev);
-	r = r100_ib_init(rdev);
+
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r100_ib_test(rdev);
 	if (r) {
-		dev_err(rdev->dev, "failed initializing IB (%d).\n", r);
+		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
 		return r;
 	}
+
 	return 0;
 }
 
@@ -297,11 +314,14 @@
 	r420_clock_resume(rdev);
 	/* Initialize surface registers */
 	radeon_surface_init(rdev);
+
+	rdev->accel_working = true;
 	return r420_startup(rdev);
 }
 
 int r420_suspend(struct radeon_device *rdev)
 {
+	radeon_ib_pool_suspend(rdev);
 	r420_cp_errata_fini(rdev);
 	r100_cp_disable(rdev);
 	radeon_wb_disable(rdev);
@@ -414,7 +434,14 @@
 			return r;
 	}
 	r420_set_reg_safe(rdev);
+
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = r420_startup(rdev);
 	if (r) {
 		/* Somethings want wront with the accel init stop accel */
diff --git a/drivers/gpu/drm/radeon/r520.c b/drivers/gpu/drm/radeon/r520.c
index 3081d07..4ae1615 100644
--- a/drivers/gpu/drm/radeon/r520.c
+++ b/drivers/gpu/drm/radeon/r520.c
@@ -187,6 +187,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	rs600_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
@@ -196,9 +202,15 @@
 		dev_err(rdev->dev, "failed initializing CP (%d).\n", r);
 		return r;
 	}
-	r = r100_ib_init(rdev);
+
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r100_ib_test(rdev);
 	if (r) {
-		dev_err(rdev->dev, "failed initializing IB (%d).\n", r);
+		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
 		return r;
 	}
 	return 0;
@@ -223,6 +235,8 @@
 	rv515_clock_startup(rdev);
 	/* Initialize surface registers */
 	radeon_surface_init(rdev);
+
+	rdev->accel_working = true;
 	return r520_startup(rdev);
 }
 
@@ -292,7 +306,14 @@
 	if (r)
 		return r;
 	rv515_set_safe_registers(rdev);
+
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = r520_startup(rdev);
 	if (r) {
 		/* Somethings want wront with the accel init stop accel */
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 19afc43..951566f 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -288,24 +288,6 @@
 		  pcie_lanes);
 }
 
-static int r600_pm_get_type_index(struct radeon_device *rdev,
-				  enum radeon_pm_state_type ps_type,
-				  int instance)
-{
-	int i;
-	int found_instance = -1;
-
-	for (i = 0; i < rdev->pm.num_power_states; i++) {
-		if (rdev->pm.power_state[i].type == ps_type) {
-			found_instance++;
-			if (found_instance == instance)
-				return i;
-		}
-	}
-	/* return default if no match */
-	return rdev->pm.default_power_state_index;
-}
-
 void rs780_pm_init_profile(struct radeon_device *rdev)
 {
 	if (rdev->pm.num_power_states == 2) {
@@ -421,6 +403,8 @@
 
 void r600_pm_init_profile(struct radeon_device *rdev)
 {
+	int idx;
+
 	if (rdev->family == CHIP_R600) {
 		/* XXX */
 		/* default */
@@ -502,81 +486,43 @@
 			rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_off_cm_idx = 0;
 			rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_on_cm_idx = 2;
 			/* low sh */
-			if (rdev->flags & RADEON_IS_MOBILITY) {
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0;
-			} else {
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0;
-			}
+			if (rdev->flags & RADEON_IS_MOBILITY)
+				idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
+			else
+				idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+			rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0;
+			rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0;
 			/* mid sh */
-			if (rdev->flags & RADEON_IS_MOBILITY) {
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 1;
-			} else {
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 1;
-			}
+			rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0;
+			rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 1;
 			/* high sh */
-			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_ps_idx =
-				r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_ps_idx =
-				r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+			idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_ps_idx = idx;
 			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_cm_idx = 0;
 			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_cm_idx = 2;
 			/* low mh */
-			if (rdev->flags & RADEON_IS_MOBILITY) {
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0;
-			} else {
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0;
-			}
+			if (rdev->flags & RADEON_IS_MOBILITY)
+				idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
+			else
+				idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
+			rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0;
+			rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0;
 			/* mid mh */
-			if (rdev->flags & RADEON_IS_MOBILITY) {
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 1;
-			} else {
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 1;
-			}
+			rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0;
+			rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 1;
 			/* high mh */
-			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_ps_idx =
-				r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_ps_idx =
-				r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
+			idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
+			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_ps_idx = idx;
 			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_cm_idx = 0;
 			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_cm_idx = 2;
 		}
@@ -1398,7 +1344,7 @@
 	return 0;
 }
 
-bool r600_gpu_is_lockup(struct radeon_device *rdev)
+bool r600_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	u32 srbm_status;
 	u32 grbm_status;
@@ -1415,19 +1361,19 @@
 	grbm_status = RREG32(R_008010_GRBM_STATUS);
 	grbm_status2 = RREG32(R_008014_GRBM_STATUS2);
 	if (!G_008010_GUI_ACTIVE(grbm_status)) {
-		r100_gpu_lockup_update(lockup, &rdev->cp);
+		r100_gpu_lockup_update(lockup, ring);
 		return false;
 	}
 	/* force CP activities */
-	r = radeon_ring_lock(rdev, 2);
+	r = radeon_ring_lock(rdev, ring, 2);
 	if (!r) {
 		/* PACKET2 NOP */
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_write(rdev, 0x80000000);
-		radeon_ring_unlock_commit(rdev);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_write(ring, 0x80000000);
+		radeon_ring_unlock_commit(rdev, ring);
 	}
-	rdev->cp.rptr = RREG32(R600_CP_RB_RPTR);
-	return r100_gpu_cp_is_lockup(rdev, lockup, &rdev->cp);
+	ring->rptr = RREG32(ring->rptr_reg);
+	return r100_gpu_cp_is_lockup(rdev, lockup, ring);
 }
 
 int r600_asic_reset(struct radeon_device *rdev)
@@ -2198,27 +2144,28 @@
 
 int r600_cp_start(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 	uint32_t cp_me;
 
-	r = radeon_ring_lock(rdev, 7);
+	r = radeon_ring_lock(rdev, ring, 7);
 	if (r) {
 		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
 		return r;
 	}
-	radeon_ring_write(rdev, PACKET3(PACKET3_ME_INITIALIZE, 5));
-	radeon_ring_write(rdev, 0x1);
+	radeon_ring_write(ring, PACKET3(PACKET3_ME_INITIALIZE, 5));
+	radeon_ring_write(ring, 0x1);
 	if (rdev->family >= CHIP_RV770) {
-		radeon_ring_write(rdev, 0x0);
-		radeon_ring_write(rdev, rdev->config.rv770.max_hw_contexts - 1);
+		radeon_ring_write(ring, 0x0);
+		radeon_ring_write(ring, rdev->config.rv770.max_hw_contexts - 1);
 	} else {
-		radeon_ring_write(rdev, 0x3);
-		radeon_ring_write(rdev, rdev->config.r600.max_hw_contexts - 1);
+		radeon_ring_write(ring, 0x3);
+		radeon_ring_write(ring, rdev->config.r600.max_hw_contexts - 1);
 	}
-	radeon_ring_write(rdev, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_write(ring, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_unlock_commit(rdev, ring);
 
 	cp_me = 0xff;
 	WREG32(R_0086D8_CP_ME_CNTL, cp_me);
@@ -2227,6 +2174,7 @@
 
 int r600_cp_resume(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 tmp;
 	u32 rb_bufsz;
 	int r;
@@ -2238,13 +2186,13 @@
 	WREG32(GRBM_SOFT_RESET, 0);
 
 	/* Set ring buffer size */
-	rb_bufsz = drm_order(rdev->cp.ring_size / 8);
+	rb_bufsz = drm_order(ring->ring_size / 8);
 	tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
 #ifdef __BIG_ENDIAN
 	tmp |= BUF_SWAP_32BIT;
 #endif
 	WREG32(CP_RB_CNTL, tmp);
-	WREG32(CP_SEM_WAIT_TIMER, 0x4);
+	WREG32(CP_SEM_WAIT_TIMER, 0x0);
 
 	/* Set the write pointer delay */
 	WREG32(CP_RB_WPTR_DELAY, 0);
@@ -2252,8 +2200,8 @@
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB_CNTL, tmp | RB_RPTR_WR_ENA);
 	WREG32(CP_RB_RPTR_WR, 0);
-	rdev->cp.wptr = 0;
-	WREG32(CP_RB_WPTR, rdev->cp.wptr);
+	ring->wptr = 0;
+	WREG32(CP_RB_WPTR, ring->wptr);
 
 	/* set the wb address whether it's enabled or not */
 	WREG32(CP_RB_RPTR_ADDR,
@@ -2271,42 +2219,36 @@
 	mdelay(1);
 	WREG32(CP_RB_CNTL, tmp);
 
-	WREG32(CP_RB_BASE, rdev->cp.gpu_addr >> 8);
+	WREG32(CP_RB_BASE, ring->gpu_addr >> 8);
 	WREG32(CP_DEBUG, (1 << 27) | (1 << 28));
 
-	rdev->cp.rptr = RREG32(CP_RB_RPTR);
+	ring->rptr = RREG32(CP_RB_RPTR);
 
 	r600_cp_start(rdev);
-	rdev->cp.ready = true;
-	r = radeon_ring_test(rdev);
+	ring->ready = true;
+	r = radeon_ring_test(rdev, ring);
 	if (r) {
-		rdev->cp.ready = false;
+		ring->ready = false;
 		return r;
 	}
 	return 0;
 }
 
-void r600_cp_commit(struct radeon_device *rdev)
-{
-	WREG32(CP_RB_WPTR, rdev->cp.wptr);
-	(void)RREG32(CP_RB_WPTR);
-}
-
-void r600_ring_init(struct radeon_device *rdev, unsigned ring_size)
+void r600_ring_init(struct radeon_device *rdev, struct radeon_ring *ring, unsigned ring_size)
 {
 	u32 rb_bufsz;
 
 	/* Align ring size */
 	rb_bufsz = drm_order(ring_size / 8);
 	ring_size = (1 << (rb_bufsz + 1)) * 4;
-	rdev->cp.ring_size = ring_size;
-	rdev->cp.align_mask = 16 - 1;
+	ring->ring_size = ring_size;
+	ring->align_mask = 16 - 1;
 }
 
 void r600_cp_fini(struct radeon_device *rdev)
 {
 	r600_cp_stop(rdev);
-	radeon_ring_fini(rdev);
+	radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
 }
 
 
@@ -2325,11 +2267,11 @@
 	}
 }
 
-int r600_ring_test(struct radeon_device *rdev)
+int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	uint32_t scratch;
 	uint32_t tmp = 0;
-	unsigned i;
+	unsigned i, ridx = radeon_ring_index(rdev, ring);
 	int r;
 
 	r = radeon_scratch_get(rdev, &scratch);
@@ -2338,16 +2280,16 @@
 		return r;
 	}
 	WREG32(scratch, 0xCAFEDEAD);
-	r = radeon_ring_lock(rdev, 3);
+	r = radeon_ring_lock(rdev, ring, 3);
 	if (r) {
-		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
+		DRM_ERROR("radeon: cp failed to lock ring %d (%d).\n", ridx, r);
 		radeon_scratch_free(rdev, scratch);
 		return r;
 	}
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-	radeon_ring_write(rdev, ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2));
-	radeon_ring_write(rdev, 0xDEADBEEF);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+	radeon_ring_write(ring, ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2));
+	radeon_ring_write(ring, 0xDEADBEEF);
+	radeon_ring_unlock_commit(rdev, ring);
 	for (i = 0; i < rdev->usec_timeout; i++) {
 		tmp = RREG32(scratch);
 		if (tmp == 0xDEADBEEF)
@@ -2355,10 +2297,10 @@
 		DRM_UDELAY(1);
 	}
 	if (i < rdev->usec_timeout) {
-		DRM_INFO("ring test succeeded in %d usecs\n", i);
+		DRM_INFO("ring test on %d succeeded in %d usecs\n", ridx, i);
 	} else {
-		DRM_ERROR("radeon: ring test failed (scratch(0x%04X)=0x%08X)\n",
-			  scratch, tmp);
+		DRM_ERROR("radeon: ring %d test failed (scratch(0x%04X)=0x%08X)\n",
+			  ridx, scratch, tmp);
 		r = -EINVAL;
 	}
 	radeon_scratch_free(rdev, scratch);
@@ -2368,49 +2310,63 @@
 void r600_fence_ring_emit(struct radeon_device *rdev,
 			  struct radeon_fence *fence)
 {
+	struct radeon_ring *ring = &rdev->ring[fence->ring];
+
 	if (rdev->wb.use_event) {
-		u64 addr = rdev->wb.gpu_addr + R600_WB_EVENT_OFFSET +
-			(u64)(rdev->fence_drv.scratch_reg - rdev->scratch.reg_base);
+		u64 addr = rdev->fence_drv[fence->ring].gpu_addr;
 		/* flush read cache over gart */
-		radeon_ring_write(rdev, PACKET3(PACKET3_SURFACE_SYNC, 3));
-		radeon_ring_write(rdev, PACKET3_TC_ACTION_ENA |
+		radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
+		radeon_ring_write(ring, PACKET3_TC_ACTION_ENA |
 					PACKET3_VC_ACTION_ENA |
 					PACKET3_SH_ACTION_ENA);
-		radeon_ring_write(rdev, 0xFFFFFFFF);
-		radeon_ring_write(rdev, 0);
-		radeon_ring_write(rdev, 10); /* poll interval */
+		radeon_ring_write(ring, 0xFFFFFFFF);
+		radeon_ring_write(ring, 0);
+		radeon_ring_write(ring, 10); /* poll interval */
 		/* EVENT_WRITE_EOP - flush caches, send int */
-		radeon_ring_write(rdev, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
-		radeon_ring_write(rdev, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT_TS) | EVENT_INDEX(5));
-		radeon_ring_write(rdev, addr & 0xffffffff);
-		radeon_ring_write(rdev, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
-		radeon_ring_write(rdev, fence->seq);
-		radeon_ring_write(rdev, 0);
+		radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
+		radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT_TS) | EVENT_INDEX(5));
+		radeon_ring_write(ring, addr & 0xffffffff);
+		radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
+		radeon_ring_write(ring, fence->seq);
+		radeon_ring_write(ring, 0);
 	} else {
 		/* flush read cache over gart */
-		radeon_ring_write(rdev, PACKET3(PACKET3_SURFACE_SYNC, 3));
-		radeon_ring_write(rdev, PACKET3_TC_ACTION_ENA |
+		radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
+		radeon_ring_write(ring, PACKET3_TC_ACTION_ENA |
 					PACKET3_VC_ACTION_ENA |
 					PACKET3_SH_ACTION_ENA);
-		radeon_ring_write(rdev, 0xFFFFFFFF);
-		radeon_ring_write(rdev, 0);
-		radeon_ring_write(rdev, 10); /* poll interval */
-		radeon_ring_write(rdev, PACKET3(PACKET3_EVENT_WRITE, 0));
-		radeon_ring_write(rdev, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT) | EVENT_INDEX(0));
+		radeon_ring_write(ring, 0xFFFFFFFF);
+		radeon_ring_write(ring, 0);
+		radeon_ring_write(ring, 10); /* poll interval */
+		radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE, 0));
+		radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT) | EVENT_INDEX(0));
 		/* wait for 3D idle clean */
-		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-		radeon_ring_write(rdev, (WAIT_UNTIL - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
-		radeon_ring_write(rdev, WAIT_3D_IDLE_bit | WAIT_3D_IDLECLEAN_bit);
+		radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		radeon_ring_write(ring, (WAIT_UNTIL - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
+		radeon_ring_write(ring, WAIT_3D_IDLE_bit | WAIT_3D_IDLECLEAN_bit);
 		/* Emit fence sequence & fire IRQ */
-		radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-		radeon_ring_write(rdev, ((rdev->fence_drv.scratch_reg - PACKET3_SET_CONFIG_REG_OFFSET) >> 2));
-		radeon_ring_write(rdev, fence->seq);
+		radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		radeon_ring_write(ring, ((rdev->fence_drv[fence->ring].scratch_reg - PACKET3_SET_CONFIG_REG_OFFSET) >> 2));
+		radeon_ring_write(ring, fence->seq);
 		/* CP_INTERRUPT packet 3 no longer exists, use packet 0 */
-		radeon_ring_write(rdev, PACKET0(CP_INT_STATUS, 0));
-		radeon_ring_write(rdev, RB_INT_STAT);
+		radeon_ring_write(ring, PACKET0(CP_INT_STATUS, 0));
+		radeon_ring_write(ring, RB_INT_STAT);
 	}
 }
 
+void r600_semaphore_ring_emit(struct radeon_device *rdev,
+			      struct radeon_ring *ring,
+			      struct radeon_semaphore *semaphore,
+			      bool emit_wait)
+{
+	uint64_t addr = semaphore->gpu_addr;
+	unsigned sel = emit_wait ? PACKET3_SEM_SEL_WAIT : PACKET3_SEM_SEL_SIGNAL;
+
+	radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1));
+	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | sel);
+}
+
 int r600_copy_blit(struct radeon_device *rdev,
 		   uint64_t src_offset,
 		   uint64_t dst_offset,
@@ -2463,6 +2419,7 @@
 
 int r600_startup(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 
 	/* enable pcie gen2 link */
@@ -2501,6 +2458,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	r = r600_irq_init(rdev);
 	if (r) {
@@ -2510,7 +2473,10 @@
 	}
 	r600_irq_set(rdev);
 
-	r = radeon_ring_init(rdev, rdev->cp.ring_size);
+	r = radeon_ring_init(rdev, ring, ring->ring_size, RADEON_WB_CP_RPTR_OFFSET,
+			     R600_CP_RB_RPTR, R600_CP_RB_WPTR,
+			     0, 0xfffff, RADEON_CP_PACKET2);
+
 	if (r)
 		return r;
 	r = r600_cp_load_microcode(rdev);
@@ -2520,6 +2486,17 @@
 	if (r)
 		return r;
 
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r600_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		DRM_ERROR("radeon: failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
+		return r;
+	}
+
 	return 0;
 }
 
@@ -2548,18 +2525,13 @@
 	/* post card */
 	atom_asic_init(rdev->mode_info.atom_context);
 
+	rdev->accel_working = true;
 	r = r600_startup(rdev);
 	if (r) {
 		DRM_ERROR("r600 startup failed on resume\n");
 		return r;
 	}
 
-	r = r600_ib_test(rdev);
-	if (r) {
-		DRM_ERROR("radeon: failed testing IB (%d).\n", r);
-		return r;
-	}
-
 	r = r600_audio_init(rdev);
 	if (r) {
 		DRM_ERROR("radeon: audio resume failed\n");
@@ -2572,13 +2544,14 @@
 int r600_suspend(struct radeon_device *rdev)
 {
 	r600_audio_fini(rdev);
+	radeon_ib_pool_suspend(rdev);
+	r600_blit_suspend(rdev);
 	/* FIXME: we should wait for ring to be empty */
 	r600_cp_stop(rdev);
-	rdev->cp.ready = false;
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
 	r600_irq_suspend(rdev);
 	radeon_wb_disable(rdev);
 	r600_pcie_gart_disable(rdev);
-	r600_blit_suspend(rdev);
 
 	return 0;
 }
@@ -2649,8 +2622,8 @@
 	if (r)
 		return r;
 
-	rdev->cp.ring_obj = NULL;
-	r600_ring_init(rdev, 1024 * 1024);
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ring_obj = NULL;
+	r600_ring_init(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX], 1024 * 1024);
 
 	rdev->ih.ring_obj = NULL;
 	r600_ih_ring_init(rdev, 64 * 1024);
@@ -2659,30 +2632,24 @@
 	if (r)
 		return r;
 
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = r600_startup(rdev);
 	if (r) {
 		dev_err(rdev->dev, "disabling GPU acceleration\n");
 		r600_cp_fini(rdev);
 		r600_irq_fini(rdev);
 		radeon_wb_fini(rdev);
+		r100_ib_fini(rdev);
 		radeon_irq_kms_fini(rdev);
 		r600_pcie_gart_fini(rdev);
 		rdev->accel_working = false;
 	}
-	if (rdev->accel_working) {
-		r = radeon_ib_pool_init(rdev);
-		if (r) {
-			dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-			rdev->accel_working = false;
-		} else {
-			r = r600_ib_test(rdev);
-			if (r) {
-				dev_err(rdev->dev, "IB test failed (%d).\n", r);
-				rdev->accel_working = false;
-			}
-		}
-	}
 
 	r = r600_audio_init(rdev);
 	if (r)
@@ -2697,12 +2664,13 @@
 	r600_cp_fini(rdev);
 	r600_irq_fini(rdev);
 	radeon_wb_fini(rdev);
-	radeon_ib_pool_fini(rdev);
+	r100_ib_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	r600_pcie_gart_fini(rdev);
 	r600_vram_scratch_fini(rdev);
 	radeon_agp_fini(rdev);
 	radeon_gem_fini(rdev);
+	radeon_semaphore_driver_fini(rdev);
 	radeon_fence_driver_fini(rdev);
 	radeon_bo_fini(rdev);
 	radeon_atombios_fini(rdev);
@@ -2716,18 +2684,20 @@
  */
 void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
+	struct radeon_ring *ring = &rdev->ring[ib->fence->ring];
+
 	/* FIXME: implement */
-	radeon_ring_write(rdev, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
+	radeon_ring_write(ring,
 #ifdef __BIG_ENDIAN
 			  (2 << 0) |
 #endif
 			  (ib->gpu_addr & 0xFFFFFFFC));
-	radeon_ring_write(rdev, upper_32_bits(ib->gpu_addr) & 0xFF);
-	radeon_ring_write(rdev, ib->length_dw);
+	radeon_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFF);
+	radeon_ring_write(ring, ib->length_dw);
 }
 
-int r600_ib_test(struct radeon_device *rdev)
+int r600_ib_test(struct radeon_device *rdev, int ring)
 {
 	struct radeon_ib *ib;
 	uint32_t scratch;
@@ -2741,7 +2711,7 @@
 		return r;
 	}
 	WREG32(scratch, 0xCAFEDEAD);
-	r = radeon_ib_get(rdev, &ib);
+	r = radeon_ib_get(rdev, ring, &ib);
 	if (r) {
 		DRM_ERROR("radeon: failed to get ib (%d).\n", r);
 		return r;
@@ -2782,7 +2752,7 @@
 		DRM_UDELAY(1);
 	}
 	if (i < rdev->usec_timeout) {
-		DRM_INFO("ib test succeeded in %u usecs\n", i);
+		DRM_INFO("ib test on ring %d succeeded in %u usecs\n", ib->fence->ring, i);
 	} else {
 		DRM_ERROR("radeon: ib test failed (scratch(0x%04X)=0x%08X)\n",
 			  scratch, tmp);
@@ -3129,7 +3099,7 @@
 		hpd3 = RREG32(DC_HOT_PLUG_DETECT3_INT_CONTROL) & ~DC_HPDx_INT_EN;
 	}
 
-	if (rdev->irq.sw_int) {
+	if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
 		DRM_DEBUG("r600_irq_set: sw int\n");
 		cp_int_cntl |= RB_INT_ENABLE;
 		cp_int_cntl |= TIME_STAMP_INT_ENABLE;
@@ -3513,11 +3483,11 @@
 		case 177: /* CP_INT in IB1 */
 		case 178: /* CP_INT in IB2 */
 			DRM_DEBUG("IH: CP int: 0x%08x\n", src_data);
-			radeon_fence_process(rdev);
+			radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX);
 			break;
 		case 181: /* CP EOP event */
 			DRM_DEBUG("IH: CP EOP\n");
-			radeon_fence_process(rdev);
+			radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX);
 			break;
 		case 233: /* GUI IDLE */
 			DRM_DEBUG("IH: GUI idle\n");
@@ -3550,30 +3520,6 @@
  */
 #if defined(CONFIG_DEBUG_FS)
 
-static int r600_debugfs_cp_ring_info(struct seq_file *m, void *data)
-{
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct radeon_device *rdev = dev->dev_private;
-	unsigned count, i, j;
-
-	radeon_ring_free_size(rdev);
-	count = (rdev->cp.ring_size / 4) - rdev->cp.ring_free_dw;
-	seq_printf(m, "CP_STAT 0x%08x\n", RREG32(CP_STAT));
-	seq_printf(m, "CP_RB_WPTR 0x%08x\n", RREG32(CP_RB_WPTR));
-	seq_printf(m, "CP_RB_RPTR 0x%08x\n", RREG32(CP_RB_RPTR));
-	seq_printf(m, "driver's copy of the CP_RB_WPTR 0x%08x\n", rdev->cp.wptr);
-	seq_printf(m, "driver's copy of the CP_RB_RPTR 0x%08x\n", rdev->cp.rptr);
-	seq_printf(m, "%u free dwords in ring\n", rdev->cp.ring_free_dw);
-	seq_printf(m, "%u dwords in ring\n", count);
-	i = rdev->cp.rptr;
-	for (j = 0; j <= count; j++) {
-		seq_printf(m, "r[%04d]=0x%08x\n", i, rdev->cp.ring[i]);
-		i = (i + 1) & rdev->cp.ptr_mask;
-	}
-	return 0;
-}
-
 static int r600_debugfs_mc_info(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
@@ -3587,7 +3533,6 @@
 
 static struct drm_info_list r600_mc_info_list[] = {
 	{"r600_mc_info", r600_debugfs_mc_info, 0, NULL},
-	{"r600_ring_info", r600_debugfs_cp_ring_info, 0, NULL},
 };
 #endif
 
diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c
index e09d281..02a7574 100644
--- a/drivers/gpu/drm/radeon/r600_blit_kms.c
+++ b/drivers/gpu/drm/radeon/r600_blit_kms.c
@@ -50,6 +50,7 @@
 set_render_target(struct radeon_device *rdev, int format,
 		  int w, int h, u64 gpu_addr)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 cb_color_info;
 	int pitch, slice;
 
@@ -63,38 +64,38 @@
 	pitch = (w / 8) - 1;
 	slice = ((w * h) / 64) - 1;
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (CB_COLOR0_BASE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, gpu_addr >> 8);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (CB_COLOR0_BASE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, gpu_addr >> 8);
 
 	if (rdev->family > CHIP_R600 && rdev->family < CHIP_RV770) {
-		radeon_ring_write(rdev, PACKET3(PACKET3_SURFACE_BASE_UPDATE, 0));
-		radeon_ring_write(rdev, 2 << 0);
+		radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_BASE_UPDATE, 0));
+		radeon_ring_write(ring, 2 << 0);
 	}
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (CB_COLOR0_SIZE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, (pitch << 0) | (slice << 10));
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (CB_COLOR0_SIZE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, (pitch << 0) | (slice << 10));
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (CB_COLOR0_VIEW - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (CB_COLOR0_VIEW - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, 0);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (CB_COLOR0_INFO - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, cb_color_info);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (CB_COLOR0_INFO - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, cb_color_info);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (CB_COLOR0_TILE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (CB_COLOR0_TILE - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, 0);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (CB_COLOR0_FRAG - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (CB_COLOR0_FRAG - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, 0);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (CB_COLOR0_MASK - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (CB_COLOR0_MASK - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, 0);
 }
 
 /* emits 5dw */
@@ -103,6 +104,7 @@
 		    u32 sync_type, u32 size,
 		    u64 mc_addr)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 cp_coher_size;
 
 	if (size == 0xffffffff)
@@ -110,17 +112,18 @@
 	else
 		cp_coher_size = ((size + 255) >> 8);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SURFACE_SYNC, 3));
-	radeon_ring_write(rdev, sync_type);
-	radeon_ring_write(rdev, cp_coher_size);
-	radeon_ring_write(rdev, mc_addr >> 8);
-	radeon_ring_write(rdev, 10); /* poll interval */
+	radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
+	radeon_ring_write(ring, sync_type);
+	radeon_ring_write(ring, cp_coher_size);
+	radeon_ring_write(ring, mc_addr >> 8);
+	radeon_ring_write(ring, 10); /* poll interval */
 }
 
 /* emits 21dw + 1 surface sync = 26dw */
 static void
 set_shaders(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u64 gpu_addr;
 	u32 sq_pgm_resources;
 
@@ -129,35 +132,35 @@
 
 	/* VS */
 	gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.vs_offset;
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (SQ_PGM_START_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, gpu_addr >> 8);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (SQ_PGM_START_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, gpu_addr >> 8);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (SQ_PGM_RESOURCES_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, sq_pgm_resources);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (SQ_PGM_RESOURCES_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, sq_pgm_resources);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (SQ_PGM_CF_OFFSET_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (SQ_PGM_CF_OFFSET_VS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, 0);
 
 	/* PS */
 	gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.ps_offset;
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (SQ_PGM_START_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, gpu_addr >> 8);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (SQ_PGM_START_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, gpu_addr >> 8);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (SQ_PGM_RESOURCES_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, sq_pgm_resources | (1 << 28));
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (SQ_PGM_RESOURCES_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, sq_pgm_resources | (1 << 28));
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (SQ_PGM_EXPORTS_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, 2);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (SQ_PGM_EXPORTS_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, 2);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
-	radeon_ring_write(rdev, (SQ_PGM_CF_OFFSET_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, 0);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 1));
+	radeon_ring_write(ring, (SQ_PGM_CF_OFFSET_PS - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, 0);
 
 	gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.vs_offset;
 	cp_set_surface_sync(rdev, PACKET3_SH_ACTION_ENA, 512, gpu_addr);
@@ -167,6 +170,7 @@
 static void
 set_vtx_resource(struct radeon_device *rdev, u64 gpu_addr)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 sq_vtx_constant_word2;
 
 	sq_vtx_constant_word2 = SQ_VTXC_BASE_ADDR_HI(upper_32_bits(gpu_addr) & 0xff) |
@@ -175,15 +179,15 @@
 	sq_vtx_constant_word2 |=  SQ_VTXC_ENDIAN_SWAP(SQ_ENDIAN_8IN32);
 #endif
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_RESOURCE, 7));
-	radeon_ring_write(rdev, 0x460);
-	radeon_ring_write(rdev, gpu_addr & 0xffffffff);
-	radeon_ring_write(rdev, 48 - 1);
-	radeon_ring_write(rdev, sq_vtx_constant_word2);
-	radeon_ring_write(rdev, 1 << 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, SQ_TEX_VTX_VALID_BUFFER << 30);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_RESOURCE, 7));
+	radeon_ring_write(ring, 0x460);
+	radeon_ring_write(ring, gpu_addr & 0xffffffff);
+	radeon_ring_write(ring, 48 - 1);
+	radeon_ring_write(ring, sq_vtx_constant_word2);
+	radeon_ring_write(ring, 1 << 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, SQ_TEX_VTX_VALID_BUFFER << 30);
 
 	if ((rdev->family == CHIP_RV610) ||
 	    (rdev->family == CHIP_RV620) ||
@@ -203,6 +207,7 @@
 		 int format, int w, int h, int pitch,
 		 u64 gpu_addr, u32 size)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	uint32_t sq_tex_resource_word0, sq_tex_resource_word1, sq_tex_resource_word4;
 
 	if (h < 1)
@@ -225,15 +230,15 @@
 	cp_set_surface_sync(rdev,
 			    PACKET3_TC_ACTION_ENA, size, gpu_addr);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_RESOURCE, 7));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, sq_tex_resource_word0);
-	radeon_ring_write(rdev, sq_tex_resource_word1);
-	radeon_ring_write(rdev, gpu_addr >> 8);
-	radeon_ring_write(rdev, gpu_addr >> 8);
-	radeon_ring_write(rdev, sq_tex_resource_word4);
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, SQ_TEX_VTX_VALID_TEXTURE << 30);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_RESOURCE, 7));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, sq_tex_resource_word0);
+	radeon_ring_write(ring, sq_tex_resource_word1);
+	radeon_ring_write(ring, gpu_addr >> 8);
+	radeon_ring_write(ring, gpu_addr >> 8);
+	radeon_ring_write(ring, sq_tex_resource_word4);
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, SQ_TEX_VTX_VALID_TEXTURE << 30);
 }
 
 /* emits 12 */
@@ -241,43 +246,45 @@
 set_scissors(struct radeon_device *rdev, int x1, int y1,
 	     int x2, int y2)
 {
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
-	radeon_ring_write(rdev, (PA_SC_SCREEN_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, (x1 << 0) | (y1 << 16));
-	radeon_ring_write(rdev, (x2 << 0) | (y2 << 16));
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
+	radeon_ring_write(ring, (PA_SC_SCREEN_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, (x1 << 0) | (y1 << 16));
+	radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
-	radeon_ring_write(rdev, (PA_SC_GENERIC_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, (x1 << 0) | (y1 << 16) | (1 << 31));
-	radeon_ring_write(rdev, (x2 << 0) | (y2 << 16));
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
+	radeon_ring_write(ring, (PA_SC_GENERIC_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, (x1 << 0) | (y1 << 16) | (1 << 31));
+	radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
-	radeon_ring_write(rdev, (PA_SC_WINDOW_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, (x1 << 0) | (y1 << 16) | (1 << 31));
-	radeon_ring_write(rdev, (x2 << 0) | (y2 << 16));
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
+	radeon_ring_write(ring, (PA_SC_WINDOW_SCISSOR_TL - PACKET3_SET_CONTEXT_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, (x1 << 0) | (y1 << 16) | (1 << 31));
+	radeon_ring_write(ring, (x2 << 0) | (y2 << 16));
 }
 
 /* emits 10 */
 static void
 draw_auto(struct radeon_device *rdev)
 {
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-	radeon_ring_write(rdev, (VGT_PRIMITIVE_TYPE - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, DI_PT_RECTLIST);
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+	radeon_ring_write(ring, (VGT_PRIMITIVE_TYPE - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, DI_PT_RECTLIST);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_INDEX_TYPE, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET3(PACKET3_INDEX_TYPE, 0));
+	radeon_ring_write(ring,
 #ifdef __BIG_ENDIAN
 			  (2 << 2) |
 #endif
 			  DI_INDEX_SIZE_16_BIT);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_NUM_INSTANCES, 0));
-	radeon_ring_write(rdev, 1);
+	radeon_ring_write(ring, PACKET3(PACKET3_NUM_INSTANCES, 0));
+	radeon_ring_write(ring, 1);
 
-	radeon_ring_write(rdev, PACKET3(PACKET3_DRAW_INDEX_AUTO, 1));
-	radeon_ring_write(rdev, 3);
-	radeon_ring_write(rdev, DI_SRC_SEL_AUTO_INDEX);
+	radeon_ring_write(ring, PACKET3(PACKET3_DRAW_INDEX_AUTO, 1));
+	radeon_ring_write(ring, 3);
+	radeon_ring_write(ring, DI_SRC_SEL_AUTO_INDEX);
 
 }
 
@@ -285,6 +292,7 @@
 static void
 set_default_state(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	u32 sq_config, sq_gpr_resource_mgmt_1, sq_gpr_resource_mgmt_2;
 	u32 sq_thread_resource_mgmt, sq_stack_resource_mgmt_1, sq_stack_resource_mgmt_2;
 	int num_ps_gprs, num_vs_gprs, num_temp_gprs, num_gs_gprs, num_es_gprs;
@@ -440,24 +448,24 @@
 	/* emit an IB pointing at default state */
 	dwords = ALIGN(rdev->r600_blit.state_len, 0x10);
 	gpu_addr = rdev->r600_blit.shader_gpu_addr + rdev->r600_blit.state_offset;
-	radeon_ring_write(rdev, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
+	radeon_ring_write(ring,
 #ifdef __BIG_ENDIAN
 			  (2 << 0) |
 #endif
 			  (gpu_addr & 0xFFFFFFFC));
-	radeon_ring_write(rdev, upper_32_bits(gpu_addr) & 0xFF);
-	radeon_ring_write(rdev, dwords);
+	radeon_ring_write(ring, upper_32_bits(gpu_addr) & 0xFF);
+	radeon_ring_write(ring, dwords);
 
 	/* SQ config */
-	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 6));
-	radeon_ring_write(rdev, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
-	radeon_ring_write(rdev, sq_config);
-	radeon_ring_write(rdev, sq_gpr_resource_mgmt_1);
-	radeon_ring_write(rdev, sq_gpr_resource_mgmt_2);
-	radeon_ring_write(rdev, sq_thread_resource_mgmt);
-	radeon_ring_write(rdev, sq_stack_resource_mgmt_1);
-	radeon_ring_write(rdev, sq_stack_resource_mgmt_2);
+	radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 6));
+	radeon_ring_write(ring, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
+	radeon_ring_write(ring, sq_config);
+	radeon_ring_write(ring, sq_gpr_resource_mgmt_1);
+	radeon_ring_write(ring, sq_gpr_resource_mgmt_2);
+	radeon_ring_write(ring, sq_thread_resource_mgmt);
+	radeon_ring_write(ring, sq_stack_resource_mgmt_1);
+	radeon_ring_write(ring, sq_stack_resource_mgmt_2);
 }
 
 static uint32_t i2f(uint32_t input)
@@ -614,7 +622,7 @@
 static int r600_vb_ib_get(struct radeon_device *rdev)
 {
 	int r;
-	r = radeon_ib_get(rdev, &rdev->r600_blit.vb_ib);
+	r = radeon_ib_get(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->r600_blit.vb_ib);
 	if (r) {
 		DRM_ERROR("failed to get IB for vertex buffer\n");
 		return r;
@@ -679,6 +687,7 @@
 
 int r600_blit_prepare_copy(struct radeon_device *rdev, unsigned num_gpu_pages)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 	int ring_size;
 	int num_loops = 0;
@@ -699,7 +708,7 @@
 	/* calculate number of loops correctly */
 	ring_size = num_loops * dwords_per_loop;
 	ring_size += rdev->r600_blit.ring_size_common;
-	r = radeon_ring_lock(rdev, ring_size);
+	r = radeon_ring_lock(rdev, ring, ring_size);
 	if (r)
 		return r;
 
@@ -718,7 +727,7 @@
 	if (fence)
 		r = radeon_fence_emit(rdev, fence);
 
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_unlock_commit(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
 }
 
 void r600_kms_blit_copy(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/r600_cp.c b/drivers/gpu/drm/radeon/r600_cp.c
index c9db493..84c5462 100644
--- a/drivers/gpu/drm/radeon/r600_cp.c
+++ b/drivers/gpu/drm/radeon/r600_cp.c
@@ -1815,7 +1815,7 @@
 		     dev_priv->ring.size_l2qw);
 #endif
 
-	RADEON_WRITE(R600_CP_SEM_WAIT_TIMER, 0x4);
+	RADEON_WRITE(R600_CP_SEM_WAIT_TIMER, 0x0);
 
 	/* Set the write pointer delay */
 	RADEON_WRITE(R600_CP_RB_WPTR_DELAY, 0);
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 0a2e023..cb1acff 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -941,7 +941,8 @@
 		track->db_depth_control = radeon_get_ib_value(p, idx);
 		break;
 	case R_028010_DB_DEPTH_INFO:
-		if (r600_cs_packet_next_is_pkt3_nop(p)) {
+		if (!p->keep_tiling_flags &&
+		    r600_cs_packet_next_is_pkt3_nop(p)) {
 			r = r600_cs_packet_next_reloc(p, &reloc);
 			if (r) {
 				dev_warn(p->dev, "bad SET_CONTEXT_REG "
@@ -992,7 +993,8 @@
 	case R_0280B4_CB_COLOR5_INFO:
 	case R_0280B8_CB_COLOR6_INFO:
 	case R_0280BC_CB_COLOR7_INFO:
-		if (r600_cs_packet_next_is_pkt3_nop(p)) {
+		if (!p->keep_tiling_flags &&
+		     r600_cs_packet_next_is_pkt3_nop(p)) {
 			r = r600_cs_packet_next_reloc(p, &reloc);
 			if (r) {
 				dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
@@ -1291,10 +1293,12 @@
 	mip_offset <<= 8;
 
 	word0 = radeon_get_ib_value(p, idx + 0);
-	if (tiling_flags & RADEON_TILING_MACRO)
-		word0 |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
-	else if (tiling_flags & RADEON_TILING_MICRO)
-		word0 |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+	if (!p->keep_tiling_flags) {
+		if (tiling_flags & RADEON_TILING_MACRO)
+			word0 |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
+		else if (tiling_flags & RADEON_TILING_MICRO)
+			word0 |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+	}
 	word1 = radeon_get_ib_value(p, idx + 1);
 	w0 = G_038000_TEX_WIDTH(word0) + 1;
 	h0 = G_038004_TEX_HEIGHT(word1) + 1;
@@ -1621,10 +1625,12 @@
 					return -EINVAL;
 				}
 				base_offset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
-				if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-					ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
-				else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-					ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+				if (!p->keep_tiling_flags) {
+					if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+						ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
+					else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+						ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+				}
 				texture = reloc->robj;
 				/* tex mip base */
 				r = r600_cs_packet_next_reloc(p, &reloc);
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index bfe1b5d..3ee1fd7 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -831,6 +831,8 @@
 #define	PACKET3_STRMOUT_BUFFER_UPDATE			0x34
 #define	PACKET3_INDIRECT_BUFFER_MP			0x38
 #define	PACKET3_MEM_SEMAPHORE				0x39
+#              define PACKET3_SEM_SEL_SIGNAL	    (0x6 << 29)
+#              define PACKET3_SEM_SEL_WAIT	    (0x7 << 29)
 #define	PACKET3_MPEG_INDEX				0x3A
 #define	PACKET3_WAIT_REG_MEM				0x3C
 #define	PACKET3_MEM_WRITE				0x3D
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 83b2e01..f29edbf 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -107,6 +107,17 @@
 #define RADEONFB_CONN_LIMIT		4
 #define RADEON_BIOS_NUM_SCRATCH		8
 
+/* max number of rings */
+#define RADEON_NUM_RINGS 3
+
+/* internal ring indices */
+/* r1xx+ has gfx CP ring */
+#define RADEON_RING_TYPE_GFX_INDEX  0
+
+/* cayman has 2 compute CP rings */
+#define CAYMAN_RING_TYPE_CP1_INDEX 1
+#define CAYMAN_RING_TYPE_CP2_INDEX 2
+
 /*
  * Errata workarounds.
  */
@@ -192,14 +203,15 @@
  */
 struct radeon_fence_driver {
 	uint32_t			scratch_reg;
+	uint64_t			gpu_addr;
+	volatile uint32_t		*cpu_addr;
 	atomic_t			seq;
 	uint32_t			last_seq;
 	unsigned long			last_jiffies;
 	unsigned long			last_timeout;
 	wait_queue_head_t		queue;
-	rwlock_t			lock;
 	struct list_head		created;
-	struct list_head		emited;
+	struct list_head		emitted;
 	struct list_head		signaled;
 	bool				initialized;
 };
@@ -210,21 +222,51 @@
 	struct list_head		list;
 	/* protected by radeon_fence.lock */
 	uint32_t			seq;
-	bool				emited;
+	bool				emitted;
 	bool				signaled;
+	/* RB, DMA, etc. */
+	int				ring;
 };
 
+int radeon_fence_driver_start_ring(struct radeon_device *rdev, int ring);
 int radeon_fence_driver_init(struct radeon_device *rdev);
 void radeon_fence_driver_fini(struct radeon_device *rdev);
-int radeon_fence_create(struct radeon_device *rdev, struct radeon_fence **fence);
+int radeon_fence_create(struct radeon_device *rdev, struct radeon_fence **fence, int ring);
 int radeon_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence);
-void radeon_fence_process(struct radeon_device *rdev);
+void radeon_fence_process(struct radeon_device *rdev, int ring);
 bool radeon_fence_signaled(struct radeon_fence *fence);
 int radeon_fence_wait(struct radeon_fence *fence, bool interruptible);
-int radeon_fence_wait_next(struct radeon_device *rdev);
-int radeon_fence_wait_last(struct radeon_device *rdev);
+int radeon_fence_wait_next(struct radeon_device *rdev, int ring);
+int radeon_fence_wait_last(struct radeon_device *rdev, int ring);
 struct radeon_fence *radeon_fence_ref(struct radeon_fence *fence);
 void radeon_fence_unref(struct radeon_fence **fence);
+int radeon_fence_count_emitted(struct radeon_device *rdev, int ring);
+
+/*
+ * Semaphores.
+ */
+struct radeon_ring;
+
+struct radeon_semaphore_driver {
+	rwlock_t		lock;
+	struct list_head	free;
+};
+
+struct radeon_semaphore {
+	struct radeon_bo	*robj;
+	struct list_head	list;
+	uint64_t		gpu_addr;
+};
+
+void radeon_semaphore_driver_fini(struct radeon_device *rdev);
+int radeon_semaphore_create(struct radeon_device *rdev,
+			    struct radeon_semaphore **semaphore);
+void radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring,
+				  struct radeon_semaphore *semaphore);
+void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring,
+				struct radeon_semaphore *semaphore);
+void radeon_semaphore_free(struct radeon_device *rdev,
+			   struct radeon_semaphore *semaphore);
 
 /*
  * Tiling registers
@@ -274,6 +316,48 @@
 	u32			tiling_flags;
 };
 
+/* sub-allocation manager, it has to be protected by another lock.
+ * By conception this is an helper for other part of the driver
+ * like the indirect buffer or semaphore, which both have their
+ * locking.
+ *
+ * Principe is simple, we keep a list of sub allocation in offset
+ * order (first entry has offset == 0, last entry has the highest
+ * offset).
+ *
+ * When allocating new object we first check if there is room at
+ * the end total_size - (last_object_offset + last_object_size) >=
+ * alloc_size. If so we allocate new object there.
+ *
+ * When there is not enough room at the end, we start waiting for
+ * each sub object until we reach object_offset+object_size >=
+ * alloc_size, this object then become the sub object we return.
+ *
+ * Alignment can't be bigger than page size.
+ *
+ * Hole are not considered for allocation to keep things simple.
+ * Assumption is that there won't be hole (all object on same
+ * alignment).
+ */
+struct radeon_sa_manager {
+	struct radeon_bo	*bo;
+	struct list_head	sa_bo;
+	unsigned		size;
+	uint64_t		gpu_addr;
+	void			*cpu_ptr;
+	uint32_t		domain;
+};
+
+struct radeon_sa_bo;
+
+/* sub-allocation buffer */
+struct radeon_sa_bo {
+	struct list_head		list;
+	struct radeon_sa_manager	*manager;
+	unsigned			offset;
+	unsigned			size;
+};
+
 /*
  * GEM objects.
  */
@@ -433,7 +517,7 @@
 
 struct radeon_irq {
 	bool		installed;
-	bool		sw_int;
+	bool		sw_int[RADEON_NUM_RINGS];
 	bool		crtc_vblank_int[RADEON_MAX_CRTCS];
 	bool		pflip[RADEON_MAX_CRTCS];
 	wait_queue_head_t	vblank_queue;
@@ -443,7 +527,7 @@
 	wait_queue_head_t	idle_queue;
 	bool		hdmi[RADEON_MAX_HDMI_BLOCKS];
 	spinlock_t sw_lock;
-	int sw_refcount;
+	int sw_refcount[RADEON_NUM_RINGS];
 	union radeon_irq_stat_regs stat_regs;
 	spinlock_t pflip_lock[RADEON_MAX_CRTCS];
 	int pflip_refcount[RADEON_MAX_CRTCS];
@@ -451,22 +535,22 @@
 
 int radeon_irq_kms_init(struct radeon_device *rdev);
 void radeon_irq_kms_fini(struct radeon_device *rdev);
-void radeon_irq_kms_sw_irq_get(struct radeon_device *rdev);
-void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev);
+void radeon_irq_kms_sw_irq_get(struct radeon_device *rdev, int ring);
+void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev, int ring);
 void radeon_irq_kms_pflip_irq_get(struct radeon_device *rdev, int crtc);
 void radeon_irq_kms_pflip_irq_put(struct radeon_device *rdev, int crtc);
 
 /*
- * CP & ring.
+ * CP & rings.
  */
+
 struct radeon_ib {
-	struct list_head	list;
+	struct radeon_sa_bo	sa_bo;
 	unsigned		idx;
-	uint64_t		gpu_addr;
-	struct radeon_fence	*fence;
-	uint32_t		*ptr;
 	uint32_t		length_dw;
-	bool			free;
+	uint64_t		gpu_addr;
+	uint32_t		*ptr;
+	struct radeon_fence	*fence;
 };
 
 /*
@@ -474,20 +558,22 @@
  * mutex protects scheduled_ibs, ready, alloc_bm
  */
 struct radeon_ib_pool {
-	struct mutex		mutex;
-	struct radeon_bo	*robj;
-	struct list_head	bogus_ib;
-	struct radeon_ib	ibs[RADEON_IB_POOL_SIZE];
-	bool			ready;
-	unsigned		head_id;
+	struct mutex			mutex;
+	struct radeon_sa_manager	sa_manager;
+	struct radeon_ib		ibs[RADEON_IB_POOL_SIZE];
+	bool				ready;
+	unsigned			head_id;
 };
 
-struct radeon_cp {
+struct radeon_ring {
 	struct radeon_bo	*ring_obj;
 	volatile uint32_t	*ring;
 	unsigned		rptr;
+	unsigned		rptr_offs;
+	unsigned		rptr_reg;
 	unsigned		wptr;
 	unsigned		wptr_old;
+	unsigned		wptr_reg;
 	unsigned		ring_size;
 	unsigned		ring_free_dw;
 	int			count_dw;
@@ -496,6 +582,9 @@
 	uint32_t		ptr_mask;
 	struct mutex		mutex;
 	bool			ready;
+	u32			ptr_reg_shift;
+	u32			ptr_reg_mask;
+	u32			nop;
 };
 
 /*
@@ -505,6 +594,7 @@
 	struct radeon_bo	*ring_obj;
 	volatile uint32_t	*ring;
 	unsigned		rptr;
+	unsigned		rptr_offs;
 	unsigned		wptr;
 	unsigned		wptr_old;
 	unsigned		ring_size;
@@ -548,23 +638,27 @@
 
 void r600_blit_suspend(struct radeon_device *rdev);
 
-int radeon_ib_get(struct radeon_device *rdev, struct radeon_ib **ib);
+int radeon_ib_get(struct radeon_device *rdev, int ring, struct radeon_ib **ib);
 void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib **ib);
 int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib);
 int radeon_ib_pool_init(struct radeon_device *rdev);
 void radeon_ib_pool_fini(struct radeon_device *rdev);
+int radeon_ib_pool_start(struct radeon_device *rdev);
+int radeon_ib_pool_suspend(struct radeon_device *rdev);
 int radeon_ib_test(struct radeon_device *rdev);
-extern void radeon_ib_bogus_add(struct radeon_device *rdev, struct radeon_ib *ib);
 /* Ring access between begin & end cannot sleep */
-void radeon_ring_free_size(struct radeon_device *rdev);
-int radeon_ring_alloc(struct radeon_device *rdev, unsigned ndw);
-int radeon_ring_lock(struct radeon_device *rdev, unsigned ndw);
-void radeon_ring_commit(struct radeon_device *rdev);
-void radeon_ring_unlock_commit(struct radeon_device *rdev);
-void radeon_ring_unlock_undo(struct radeon_device *rdev);
-int radeon_ring_test(struct radeon_device *rdev);
-int radeon_ring_init(struct radeon_device *rdev, unsigned ring_size);
-void radeon_ring_fini(struct radeon_device *rdev);
+int radeon_ring_index(struct radeon_device *rdev, struct radeon_ring *cp);
+void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *cp);
+int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
+int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
+void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *cp);
+void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *cp);
+void radeon_ring_unlock_undo(struct radeon_device *rdev, struct radeon_ring *cp);
+int radeon_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
+int radeon_ring_init(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ring_size,
+		     unsigned rptr_offs, unsigned rptr_reg, unsigned wptr_reg,
+		     u32 ptr_reg_shift, u32 ptr_reg_mask, u32 nop);
+void radeon_ring_fini(struct radeon_device *rdev, struct radeon_ring *cp);
 
 
 /*
@@ -610,7 +704,8 @@
 	struct radeon_ib	*ib;
 	void			*track;
 	unsigned		family;
-	int parser_error;
+	int			parser_error;
+	bool			keep_tiling_flags;
 };
 
 extern int radeon_cs_update_pages(struct radeon_cs_parser *p, int pg_idx);
@@ -783,8 +878,7 @@
 
 struct radeon_power_state {
 	enum radeon_pm_state_type type;
-	/* XXX: use a define for num clock modes */
-	struct radeon_pm_clock_info clock_info[8];
+	struct radeon_pm_clock_info *clock_info;
 	/* number of valid clock modes in this power state */
 	int num_clock_modes;
 	struct radeon_pm_clock_info *default_clock_mode;
@@ -854,6 +948,9 @@
 	struct device	        *int_hwmon_dev;
 };
 
+int radeon_pm_get_type_index(struct radeon_device *rdev,
+			     enum radeon_pm_state_type ps_type,
+			     int instance);
 
 /*
  * Benchmarking
@@ -865,11 +962,20 @@
  * Testing
  */
 void radeon_test_moves(struct radeon_device *rdev);
+void radeon_test_ring_sync(struct radeon_device *rdev,
+			   struct radeon_ring *cpA,
+			   struct radeon_ring *cpB);
+void radeon_test_syncing(struct radeon_device *rdev);
 
 
 /*
  * Debugfs
  */
+struct radeon_debugfs {
+	struct drm_info_list	*files;
+	unsigned		num_files;
+};
+
 int radeon_debugfs_add_files(struct radeon_device *rdev,
 			     struct drm_info_list *files,
 			     unsigned nfiles);
@@ -885,21 +991,26 @@
 	int (*resume)(struct radeon_device *rdev);
 	int (*suspend)(struct radeon_device *rdev);
 	void (*vga_set_state)(struct radeon_device *rdev, bool state);
-	bool (*gpu_is_lockup)(struct radeon_device *rdev);
+	bool (*gpu_is_lockup)(struct radeon_device *rdev, struct radeon_ring *cp);
 	int (*asic_reset)(struct radeon_device *rdev);
 	void (*gart_tlb_flush)(struct radeon_device *rdev);
 	int (*gart_set_page)(struct radeon_device *rdev, int i, uint64_t addr);
 	int (*cp_init)(struct radeon_device *rdev, unsigned ring_size);
 	void (*cp_fini)(struct radeon_device *rdev);
 	void (*cp_disable)(struct radeon_device *rdev);
-	void (*cp_commit)(struct radeon_device *rdev);
 	void (*ring_start)(struct radeon_device *rdev);
-	int (*ring_test)(struct radeon_device *rdev);
-	void (*ring_ib_execute)(struct radeon_device *rdev, struct radeon_ib *ib);
+
+	struct {
+		void (*ib_execute)(struct radeon_device *rdev, struct radeon_ib *ib);
+		void (*emit_fence)(struct radeon_device *rdev, struct radeon_fence *fence);
+		void (*emit_semaphore)(struct radeon_device *rdev, struct radeon_ring *cp,
+				       struct radeon_semaphore *semaphore, bool emit_wait);
+	} ring[RADEON_NUM_RINGS];
+
+	int (*ring_test)(struct radeon_device *rdev, struct radeon_ring *cp);
 	int (*irq_set)(struct radeon_device *rdev);
 	int (*irq_process)(struct radeon_device *rdev);
 	u32 (*get_vblank_counter)(struct radeon_device *rdev, int crtc);
-	void (*fence_ring_emit)(struct radeon_device *rdev, struct radeon_fence *fence);
 	int (*cs_parse)(struct radeon_cs_parser *p);
 	int (*copy_blit)(struct radeon_device *rdev,
 			 uint64_t src_offset,
@@ -1141,6 +1252,48 @@
 	u64				gpu_addr;
 };
 
+
+/*
+ * Mutex which allows recursive locking from the same process.
+ */
+struct radeon_mutex {
+	struct mutex		mutex;
+	struct task_struct	*owner;
+	int			level;
+};
+
+static inline void radeon_mutex_init(struct radeon_mutex *mutex)
+{
+	mutex_init(&mutex->mutex);
+	mutex->owner = NULL;
+	mutex->level = 0;
+}
+
+static inline void radeon_mutex_lock(struct radeon_mutex *mutex)
+{
+	if (mutex_trylock(&mutex->mutex)) {
+		/* The mutex was unlocked before, so it's ours now */
+		mutex->owner = current;
+	} else if (mutex->owner != current) {
+		/* Another process locked the mutex, take it */
+		mutex_lock(&mutex->mutex);
+		mutex->owner = current;
+	}
+	/* Otherwise the mutex was already locked by this process */
+
+	mutex->level++;
+}
+
+static inline void radeon_mutex_unlock(struct radeon_mutex *mutex)
+{
+	if (--mutex->level > 0)
+		return;
+
+	mutex->owner = NULL;
+	mutex_unlock(&mutex->mutex);
+}
+
+
 /*
  * Core structure, functions and helpers.
  */
@@ -1185,18 +1338,17 @@
 	struct radeon_mode_info		mode_info;
 	struct radeon_scratch		scratch;
 	struct radeon_mman		mman;
-	struct radeon_fence_driver	fence_drv;
-	struct radeon_cp		cp;
-	/* cayman compute rings */
-	struct radeon_cp		cp1;
-	struct radeon_cp		cp2;
+	rwlock_t			fence_lock;
+	struct radeon_fence_driver	fence_drv[RADEON_NUM_RINGS];
+	struct radeon_semaphore_driver	semaphore_drv;
+	struct radeon_ring		ring[RADEON_NUM_RINGS];
 	struct radeon_ib_pool		ib_pool;
 	struct radeon_irq		irq;
 	struct radeon_asic		*asic;
 	struct radeon_gem		gem;
 	struct radeon_pm		pm;
 	uint32_t			bios_scratch[RADEON_BIOS_NUM_SCRATCH];
-	struct mutex			cs_mutex;
+	struct radeon_mutex		cs_mutex;
 	struct radeon_wb		wb;
 	struct radeon_dummy_page	dummy_page;
 	bool				gpu_lockup;
@@ -1233,6 +1385,9 @@
 	struct drm_file *cmask_filp;
 	/* i2c buses */
 	struct radeon_i2c_chan *i2c_bus[RADEON_MAX_I2C_BUS];
+	/* debugfs */
+	struct radeon_debugfs	debugfs[RADEON_DEBUGFS_MAX_COMPONENTS];
+	unsigned 		debugfs_count;
 };
 
 int radeon_device_init(struct radeon_device *rdev,
@@ -1368,18 +1523,17 @@
 /*
  * RING helpers.
  */
-
 #if DRM_DEBUG_CODE == 0
-static inline void radeon_ring_write(struct radeon_device *rdev, uint32_t v)
+static inline void radeon_ring_write(struct radeon_ring *ring, uint32_t v)
 {
-	rdev->cp.ring[rdev->cp.wptr++] = v;
-	rdev->cp.wptr &= rdev->cp.ptr_mask;
-	rdev->cp.count_dw--;
-	rdev->cp.ring_free_dw--;
+	ring->ring[ring->wptr++] = v;
+	ring->wptr &= ring->ptr_mask;
+	ring->count_dw--;
+	ring->ring_free_dw--;
 }
 #else
 /* With debugging this is just too big to inline */
-void radeon_ring_write(struct radeon_device *rdev, uint32_t v);
+void radeon_ring_write(struct radeon_ring *ring, uint32_t v);
 #endif
 
 /*
@@ -1391,18 +1545,18 @@
 #define radeon_suspend(rdev) (rdev)->asic->suspend((rdev))
 #define radeon_cs_parse(p) rdev->asic->cs_parse((p))
 #define radeon_vga_set_state(rdev, state) (rdev)->asic->vga_set_state((rdev), (state))
-#define radeon_gpu_is_lockup(rdev) (rdev)->asic->gpu_is_lockup((rdev))
+#define radeon_gpu_is_lockup(rdev, cp) (rdev)->asic->gpu_is_lockup((rdev), (cp))
 #define radeon_asic_reset(rdev) (rdev)->asic->asic_reset((rdev))
 #define radeon_gart_tlb_flush(rdev) (rdev)->asic->gart_tlb_flush((rdev))
 #define radeon_gart_set_page(rdev, i, p) (rdev)->asic->gart_set_page((rdev), (i), (p))
-#define radeon_cp_commit(rdev) (rdev)->asic->cp_commit((rdev))
 #define radeon_ring_start(rdev) (rdev)->asic->ring_start((rdev))
-#define radeon_ring_test(rdev) (rdev)->asic->ring_test((rdev))
-#define radeon_ring_ib_execute(rdev, ib) (rdev)->asic->ring_ib_execute((rdev), (ib))
+#define radeon_ring_test(rdev, cp) (rdev)->asic->ring_test((rdev), (cp))
+#define radeon_ring_ib_execute(rdev, r, ib) (rdev)->asic->ring[(r)].ib_execute((rdev), (ib))
 #define radeon_irq_set(rdev) (rdev)->asic->irq_set((rdev))
 #define radeon_irq_process(rdev) (rdev)->asic->irq_process((rdev))
 #define radeon_get_vblank_counter(rdev, crtc) (rdev)->asic->get_vblank_counter((rdev), (crtc))
-#define radeon_fence_ring_emit(rdev, fence) (rdev)->asic->fence_ring_emit((rdev), (fence))
+#define radeon_fence_ring_emit(rdev, r, fence) (rdev)->asic->ring[(r)].emit_fence((rdev), (fence))
+#define radeon_semaphore_ring_emit(rdev, r, cp, semaphore, emit_wait) (rdev)->asic->ring[(r)].emit_semaphore((rdev), (cp), (semaphore), (emit_wait))
 #define radeon_copy_blit(rdev, s, d, np, f) (rdev)->asic->copy_blit((rdev), (s), (d), (np), (f))
 #define radeon_copy_dma(rdev, s, d, np, f) (rdev)->asic->copy_dma((rdev), (s), (d), (np), (f))
 #define radeon_copy(rdev, s, d, np, f) (rdev)->asic->copy((rdev), (s), (d), (np), (f))
diff --git a/drivers/gpu/drm/radeon/radeon_acpi.c b/drivers/gpu/drm/radeon/radeon_acpi.c
index 3f6636b..3516a60 100644
--- a/drivers/gpu/drm/radeon/radeon_acpi.c
+++ b/drivers/gpu/drm/radeon/radeon_acpi.c
@@ -35,7 +35,8 @@
 
 	/* Fail only if calling the method fails and ATIF is supported */
 	if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) {
-		printk(KERN_DEBUG "failed to evaluate ATIF got %s\n", acpi_format_exception(status));
+		DRM_DEBUG_DRIVER("failed to evaluate ATIF got %s\n",
+				 acpi_format_exception(status));
 		kfree(buffer.pointer);
 		return 1;
 	}
@@ -50,13 +51,13 @@
 	acpi_handle handle;
 	int ret;
 
-	/* No need to proceed if we're sure that ATIF is not supported */
-	if (!ASIC_IS_AVIVO(rdev) || !rdev->bios)
-		return 0;
-
 	/* Get the device handle */
 	handle = DEVICE_ACPI_HANDLE(&rdev->pdev->dev);
 
+	/* No need to proceed if we're sure that ATIF is not supported */
+	if (!ASIC_IS_AVIVO(rdev) || !rdev->bios || !handle)
+		return 0;
+
 	/* Call the ATIF method */
 	ret = radeon_atif_call(handle);
 	if (ret)
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index e294456..8493d40 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -138,14 +138,18 @@
 	.asic_reset = &r100_asic_reset,
 	.gart_tlb_flush = &r100_pci_gart_tlb_flush,
 	.gart_set_page = &r100_pci_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &r100_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r100_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r100_irq_set,
 	.irq_process = &r100_irq_process,
 	.get_vblank_counter = &r100_get_vblank_counter,
-	.fence_ring_emit = &r100_fence_ring_emit,
 	.cs_parse = &r100_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = NULL,
@@ -186,14 +190,18 @@
 	.asic_reset = &r100_asic_reset,
 	.gart_tlb_flush = &r100_pci_gart_tlb_flush,
 	.gart_set_page = &r100_pci_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &r100_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r100_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r100_irq_set,
 	.irq_process = &r100_irq_process,
 	.get_vblank_counter = &r100_get_vblank_counter,
-	.fence_ring_emit = &r100_fence_ring_emit,
 	.cs_parse = &r100_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -233,14 +241,18 @@
 	.asic_reset = &r300_asic_reset,
 	.gart_tlb_flush = &r100_pci_gart_tlb_flush,
 	.gart_set_page = &r100_pci_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &r300_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r300_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r100_irq_set,
 	.irq_process = &r100_irq_process,
 	.get_vblank_counter = &r100_get_vblank_counter,
-	.fence_ring_emit = &r300_fence_ring_emit,
 	.cs_parse = &r300_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -281,14 +293,18 @@
 	.asic_reset = &r300_asic_reset,
 	.gart_tlb_flush = &rv370_pcie_gart_tlb_flush,
 	.gart_set_page = &rv370_pcie_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &r300_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r300_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r100_irq_set,
 	.irq_process = &r100_irq_process,
 	.get_vblank_counter = &r100_get_vblank_counter,
-	.fence_ring_emit = &r300_fence_ring_emit,
 	.cs_parse = &r300_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -328,14 +344,18 @@
 	.asic_reset = &r300_asic_reset,
 	.gart_tlb_flush = &rv370_pcie_gart_tlb_flush,
 	.gart_set_page = &rv370_pcie_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &r300_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r300_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r100_irq_set,
 	.irq_process = &r100_irq_process,
 	.get_vblank_counter = &r100_get_vblank_counter,
-	.fence_ring_emit = &r300_fence_ring_emit,
 	.cs_parse = &r300_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -376,14 +396,18 @@
 	.asic_reset = &r300_asic_reset,
 	.gart_tlb_flush = &rs400_gart_tlb_flush,
 	.gart_set_page = &rs400_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &r300_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r300_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r100_irq_set,
 	.irq_process = &r100_irq_process,
 	.get_vblank_counter = &r100_get_vblank_counter,
-	.fence_ring_emit = &r300_fence_ring_emit,
 	.cs_parse = &r300_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -424,14 +448,18 @@
 	.asic_reset = &rs600_asic_reset,
 	.gart_tlb_flush = &rs600_gart_tlb_flush,
 	.gart_set_page = &rs600_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &r300_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r300_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &rs600_irq_set,
 	.irq_process = &rs600_irq_process,
 	.get_vblank_counter = &rs600_get_vblank_counter,
-	.fence_ring_emit = &r300_fence_ring_emit,
 	.cs_parse = &r300_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -472,14 +500,18 @@
 	.asic_reset = &rs600_asic_reset,
 	.gart_tlb_flush = &rs400_gart_tlb_flush,
 	.gart_set_page = &rs400_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &r300_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r300_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &rs600_irq_set,
 	.irq_process = &rs600_irq_process,
 	.get_vblank_counter = &rs600_get_vblank_counter,
-	.fence_ring_emit = &r300_fence_ring_emit,
 	.cs_parse = &r300_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -520,14 +552,18 @@
 	.asic_reset = &rs600_asic_reset,
 	.gart_tlb_flush = &rv370_pcie_gart_tlb_flush,
 	.gart_set_page = &rv370_pcie_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &rv515_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r300_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &rs600_irq_set,
 	.irq_process = &rs600_irq_process,
 	.get_vblank_counter = &rs600_get_vblank_counter,
-	.fence_ring_emit = &r300_fence_ring_emit,
 	.cs_parse = &r300_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -568,14 +604,18 @@
 	.asic_reset = &rs600_asic_reset,
 	.gart_tlb_flush = &rv370_pcie_gart_tlb_flush,
 	.gart_set_page = &rv370_pcie_gart_set_page,
-	.cp_commit = &r100_cp_commit,
 	.ring_start = &rv515_ring_start,
 	.ring_test = &r100_ring_test,
-	.ring_ib_execute = &r100_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r100_ring_ib_execute,
+			.emit_fence = &r300_fence_ring_emit,
+			.emit_semaphore = &r100_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &rs600_irq_set,
 	.irq_process = &rs600_irq_process,
 	.get_vblank_counter = &rs600_get_vblank_counter,
-	.fence_ring_emit = &r300_fence_ring_emit,
 	.cs_parse = &r300_cs_parse,
 	.copy_blit = &r100_copy_blit,
 	.copy_dma = &r200_copy_dma,
@@ -611,18 +651,22 @@
 	.fini = &r600_fini,
 	.suspend = &r600_suspend,
 	.resume = &r600_resume,
-	.cp_commit = &r600_cp_commit,
 	.vga_set_state = &r600_vga_set_state,
 	.gpu_is_lockup = &r600_gpu_is_lockup,
 	.asic_reset = &r600_asic_reset,
 	.gart_tlb_flush = &r600_pcie_gart_tlb_flush,
 	.gart_set_page = &rs600_gart_set_page,
 	.ring_test = &r600_ring_test,
-	.ring_ib_execute = &r600_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r600_ring_ib_execute,
+			.emit_fence = &r600_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r600_irq_set,
 	.irq_process = &r600_irq_process,
 	.get_vblank_counter = &rs600_get_vblank_counter,
-	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &r600_cs_parse,
 	.copy_blit = &r600_copy_blit,
 	.copy_dma = NULL,
@@ -658,18 +702,22 @@
 	.fini = &r600_fini,
 	.suspend = &r600_suspend,
 	.resume = &r600_resume,
-	.cp_commit = &r600_cp_commit,
 	.gpu_is_lockup = &r600_gpu_is_lockup,
 	.vga_set_state = &r600_vga_set_state,
 	.asic_reset = &r600_asic_reset,
 	.gart_tlb_flush = &r600_pcie_gart_tlb_flush,
 	.gart_set_page = &rs600_gart_set_page,
 	.ring_test = &r600_ring_test,
-	.ring_ib_execute = &r600_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r600_ring_ib_execute,
+			.emit_fence = &r600_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r600_irq_set,
 	.irq_process = &r600_irq_process,
 	.get_vblank_counter = &rs600_get_vblank_counter,
-	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &r600_cs_parse,
 	.copy_blit = &r600_copy_blit,
 	.copy_dma = NULL,
@@ -705,18 +753,22 @@
 	.fini = &rv770_fini,
 	.suspend = &rv770_suspend,
 	.resume = &rv770_resume,
-	.cp_commit = &r600_cp_commit,
 	.asic_reset = &r600_asic_reset,
 	.gpu_is_lockup = &r600_gpu_is_lockup,
 	.vga_set_state = &r600_vga_set_state,
 	.gart_tlb_flush = &r600_pcie_gart_tlb_flush,
 	.gart_set_page = &rs600_gart_set_page,
 	.ring_test = &r600_ring_test,
-	.ring_ib_execute = &r600_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &r600_ring_ib_execute,
+			.emit_fence = &r600_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &r600_irq_set,
 	.irq_process = &r600_irq_process,
 	.get_vblank_counter = &rs600_get_vblank_counter,
-	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &r600_cs_parse,
 	.copy_blit = &r600_copy_blit,
 	.copy_dma = NULL,
@@ -752,18 +804,22 @@
 	.fini = &evergreen_fini,
 	.suspend = &evergreen_suspend,
 	.resume = &evergreen_resume,
-	.cp_commit = &r600_cp_commit,
 	.gpu_is_lockup = &evergreen_gpu_is_lockup,
 	.asic_reset = &evergreen_asic_reset,
 	.vga_set_state = &r600_vga_set_state,
 	.gart_tlb_flush = &evergreen_pcie_gart_tlb_flush,
 	.gart_set_page = &rs600_gart_set_page,
 	.ring_test = &r600_ring_test,
-	.ring_ib_execute = &evergreen_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &evergreen_ring_ib_execute,
+			.emit_fence = &r600_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &evergreen_irq_set,
 	.irq_process = &evergreen_irq_process,
 	.get_vblank_counter = &evergreen_get_vblank_counter,
-	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
 	.copy_blit = &r600_copy_blit,
 	.copy_dma = NULL,
@@ -799,18 +855,22 @@
 	.fini = &evergreen_fini,
 	.suspend = &evergreen_suspend,
 	.resume = &evergreen_resume,
-	.cp_commit = &r600_cp_commit,
 	.gpu_is_lockup = &evergreen_gpu_is_lockup,
 	.asic_reset = &evergreen_asic_reset,
 	.vga_set_state = &r600_vga_set_state,
 	.gart_tlb_flush = &evergreen_pcie_gart_tlb_flush,
 	.gart_set_page = &rs600_gart_set_page,
 	.ring_test = &r600_ring_test,
-	.ring_ib_execute = &evergreen_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &evergreen_ring_ib_execute,
+			.emit_fence = &r600_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &evergreen_irq_set,
 	.irq_process = &evergreen_irq_process,
 	.get_vblank_counter = &evergreen_get_vblank_counter,
-	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
 	.copy_blit = &r600_copy_blit,
 	.copy_dma = NULL,
@@ -834,7 +894,7 @@
 	.pm_misc = &evergreen_pm_misc,
 	.pm_prepare = &evergreen_pm_prepare,
 	.pm_finish = &evergreen_pm_finish,
-	.pm_init_profile = &rs780_pm_init_profile,
+	.pm_init_profile = &sumo_pm_init_profile,
 	.pm_get_dynpm_state = &r600_pm_get_dynpm_state,
 	.pre_page_flip = &evergreen_pre_page_flip,
 	.page_flip = &evergreen_page_flip,
@@ -846,18 +906,22 @@
 	.fini = &evergreen_fini,
 	.suspend = &evergreen_suspend,
 	.resume = &evergreen_resume,
-	.cp_commit = &r600_cp_commit,
 	.gpu_is_lockup = &evergreen_gpu_is_lockup,
 	.asic_reset = &evergreen_asic_reset,
 	.vga_set_state = &r600_vga_set_state,
 	.gart_tlb_flush = &evergreen_pcie_gart_tlb_flush,
 	.gart_set_page = &rs600_gart_set_page,
 	.ring_test = &r600_ring_test,
-	.ring_ib_execute = &evergreen_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &evergreen_ring_ib_execute,
+			.emit_fence = &r600_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &evergreen_irq_set,
 	.irq_process = &evergreen_irq_process,
 	.get_vblank_counter = &evergreen_get_vblank_counter,
-	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
 	.copy_blit = &r600_copy_blit,
 	.copy_dma = NULL,
@@ -893,18 +957,32 @@
 	.fini = &cayman_fini,
 	.suspend = &cayman_suspend,
 	.resume = &cayman_resume,
-	.cp_commit = &r600_cp_commit,
 	.gpu_is_lockup = &cayman_gpu_is_lockup,
 	.asic_reset = &cayman_asic_reset,
 	.vga_set_state = &r600_vga_set_state,
 	.gart_tlb_flush = &cayman_pcie_gart_tlb_flush,
 	.gart_set_page = &rs600_gart_set_page,
 	.ring_test = &r600_ring_test,
-	.ring_ib_execute = &evergreen_ring_ib_execute,
+	.ring = {
+		[RADEON_RING_TYPE_GFX_INDEX] = {
+			.ib_execute = &evergreen_ring_ib_execute,
+			.emit_fence = &cayman_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		},
+		[CAYMAN_RING_TYPE_CP1_INDEX] = {
+			.ib_execute = &r600_ring_ib_execute,
+			.emit_fence = &cayman_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		},
+		[CAYMAN_RING_TYPE_CP2_INDEX] = {
+			.ib_execute = &r600_ring_ib_execute,
+			.emit_fence = &cayman_fence_ring_emit,
+			.emit_semaphore = &r600_semaphore_ring_emit,
+		}
+	},
 	.irq_set = &evergreen_irq_set,
 	.irq_process = &evergreen_irq_process,
 	.get_vblank_counter = &evergreen_get_vblank_counter,
-	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
 	.copy_blit = &r600_copy_blit,
 	.copy_dma = NULL,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 85f14f0..c002ed1c 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -58,17 +58,20 @@
 int r100_suspend(struct radeon_device *rdev);
 int r100_resume(struct radeon_device *rdev);
 void r100_vga_set_state(struct radeon_device *rdev, bool state);
-bool r100_gpu_is_lockup(struct radeon_device *rdev);
+bool r100_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int r100_asic_reset(struct radeon_device *rdev);
 u32 r100_get_vblank_counter(struct radeon_device *rdev, int crtc);
 void r100_pci_gart_tlb_flush(struct radeon_device *rdev);
 int r100_pci_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
-void r100_cp_commit(struct radeon_device *rdev);
 void r100_ring_start(struct radeon_device *rdev);
 int r100_irq_set(struct radeon_device *rdev);
 int r100_irq_process(struct radeon_device *rdev);
 void r100_fence_ring_emit(struct radeon_device *rdev,
 			  struct radeon_fence *fence);
+void r100_semaphore_ring_emit(struct radeon_device *rdev,
+			      struct radeon_ring *cp,
+			      struct radeon_semaphore *semaphore,
+			      bool emit_wait);
 int r100_cs_parse(struct radeon_cs_parser *p);
 void r100_pll_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 uint32_t r100_pll_rreg(struct radeon_device *rdev, uint32_t reg);
@@ -83,7 +86,7 @@
 void r100_clear_surface_reg(struct radeon_device *rdev, int reg);
 void r100_bandwidth_update(struct radeon_device *rdev);
 void r100_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
-int r100_ring_test(struct radeon_device *rdev);
+int r100_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
 void r100_hpd_init(struct radeon_device *rdev);
 void r100_hpd_fini(struct radeon_device *rdev);
 bool r100_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd);
@@ -101,12 +104,12 @@
 int r100_debugfs_mc_info_init(struct radeon_device *rdev);
 int r100_gui_wait_for_idle(struct radeon_device *rdev);
 void r100_gpu_lockup_update(struct r100_gpu_lockup *lockup,
-			    struct radeon_cp *cp);
+			    struct radeon_ring *cp);
 bool r100_gpu_cp_is_lockup(struct radeon_device *rdev,
 			   struct r100_gpu_lockup *lockup,
-			   struct radeon_cp *cp);
+			   struct radeon_ring *cp);
 void r100_ib_fini(struct radeon_device *rdev);
-int r100_ib_init(struct radeon_device *rdev);
+int r100_ib_test(struct radeon_device *rdev);
 void r100_irq_disable(struct radeon_device *rdev);
 void r100_mc_stop(struct radeon_device *rdev, struct r100_mc_save *save);
 void r100_mc_resume(struct radeon_device *rdev, struct r100_mc_save *save);
@@ -154,7 +157,7 @@
 extern void r300_fini(struct radeon_device *rdev);
 extern int r300_suspend(struct radeon_device *rdev);
 extern int r300_resume(struct radeon_device *rdev);
-extern bool r300_gpu_is_lockup(struct radeon_device *rdev);
+extern bool r300_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 extern int r300_asic_reset(struct radeon_device *rdev);
 extern void r300_ring_start(struct radeon_device *rdev);
 extern void r300_fence_ring_emit(struct radeon_device *rdev,
@@ -293,22 +296,25 @@
 void r600_vga_set_state(struct radeon_device *rdev, bool state);
 int r600_wb_init(struct radeon_device *rdev);
 void r600_wb_fini(struct radeon_device *rdev);
-void r600_cp_commit(struct radeon_device *rdev);
 void r600_pcie_gart_tlb_flush(struct radeon_device *rdev);
 uint32_t r600_pciep_rreg(struct radeon_device *rdev, uint32_t reg);
 void r600_pciep_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 int r600_cs_parse(struct radeon_cs_parser *p);
 void r600_fence_ring_emit(struct radeon_device *rdev,
 			  struct radeon_fence *fence);
-bool r600_gpu_is_lockup(struct radeon_device *rdev);
+void r600_semaphore_ring_emit(struct radeon_device *rdev,
+			      struct radeon_ring *cp,
+			      struct radeon_semaphore *semaphore,
+			      bool emit_wait);
+bool r600_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int r600_asic_reset(struct radeon_device *rdev);
 int r600_set_surface_reg(struct radeon_device *rdev, int reg,
 			 uint32_t tiling_flags, uint32_t pitch,
 			 uint32_t offset, uint32_t obj_size);
 void r600_clear_surface_reg(struct radeon_device *rdev, int reg);
-int r600_ib_test(struct radeon_device *rdev);
+int r600_ib_test(struct radeon_device *rdev, int ring);
 void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
-int r600_ring_test(struct radeon_device *rdev);
+int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
 int r600_copy_blit(struct radeon_device *rdev,
 		   uint64_t src_offset, uint64_t dst_offset,
 		   unsigned num_gpu_pages, struct radeon_fence *fence);
@@ -328,7 +334,7 @@
 bool r600_card_posted(struct radeon_device *rdev);
 void r600_cp_stop(struct radeon_device *rdev);
 int r600_cp_start(struct radeon_device *rdev);
-void r600_ring_init(struct radeon_device *rdev, unsigned ring_size);
+void r600_ring_init(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ring_size);
 int r600_cp_resume(struct radeon_device *rdev);
 void r600_cp_fini(struct radeon_device *rdev);
 int r600_count_pipe_bits(uint32_t val);
@@ -397,7 +403,7 @@
 void evergreen_fini(struct radeon_device *rdev);
 int evergreen_suspend(struct radeon_device *rdev);
 int evergreen_resume(struct radeon_device *rdev);
-bool evergreen_gpu_is_lockup(struct radeon_device *rdev);
+bool evergreen_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int evergreen_asic_reset(struct radeon_device *rdev);
 void evergreen_bandwidth_update(struct radeon_device *rdev);
 void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
@@ -413,6 +419,7 @@
 extern void evergreen_pm_misc(struct radeon_device *rdev);
 extern void evergreen_pm_prepare(struct radeon_device *rdev);
 extern void evergreen_pm_finish(struct radeon_device *rdev);
+extern void sumo_pm_init_profile(struct radeon_device *rdev);
 extern void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc);
 extern u32 evergreen_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
 extern void evergreen_post_page_flip(struct radeon_device *rdev, int crtc);
@@ -422,12 +429,14 @@
 /*
  * cayman
  */
+void cayman_fence_ring_emit(struct radeon_device *rdev,
+			    struct radeon_fence *fence);
 void cayman_pcie_gart_tlb_flush(struct radeon_device *rdev);
 int cayman_init(struct radeon_device *rdev);
 void cayman_fini(struct radeon_device *rdev);
 int cayman_suspend(struct radeon_device *rdev);
 int cayman_resume(struct radeon_device *rdev);
-bool cayman_gpu_is_lockup(struct radeon_device *rdev);
+bool cayman_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int cayman_asic_reset(struct radeon_device *rdev);
 
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 08d0b94..d24baf3 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -62,6 +62,87 @@
 	struct _ATOM_SUPPORTED_DEVICES_INFO_2d1 info_2d1;
 };
 
+static void radeon_lookup_i2c_gpio_quirks(struct radeon_device *rdev,
+					  ATOM_GPIO_I2C_ASSIGMENT *gpio,
+					  u8 index)
+{
+	/* r4xx mask is technically not used by the hw, so patch in the legacy mask bits */
+	if ((rdev->family == CHIP_R420) ||
+	    (rdev->family == CHIP_R423) ||
+	    (rdev->family == CHIP_RV410)) {
+		if ((le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x0018) ||
+		    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x0019) ||
+		    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x001a)) {
+			gpio->ucClkMaskShift = 0x19;
+			gpio->ucDataMaskShift = 0x18;
+		}
+	}
+
+	/* some evergreen boards have bad data for this entry */
+	if (ASIC_IS_DCE4(rdev)) {
+		if ((index == 7) &&
+		    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1936) &&
+		    (gpio->sucI2cId.ucAccess == 0)) {
+			gpio->sucI2cId.ucAccess = 0x97;
+			gpio->ucDataMaskShift = 8;
+			gpio->ucDataEnShift = 8;
+			gpio->ucDataY_Shift = 8;
+			gpio->ucDataA_Shift = 8;
+		}
+	}
+
+	/* some DCE3 boards have bad data for this entry */
+	if (ASIC_IS_DCE3(rdev)) {
+		if ((index == 4) &&
+		    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1fda) &&
+		    (gpio->sucI2cId.ucAccess == 0x94))
+			gpio->sucI2cId.ucAccess = 0x14;
+	}
+}
+
+static struct radeon_i2c_bus_rec radeon_get_bus_rec_for_i2c_gpio(ATOM_GPIO_I2C_ASSIGMENT *gpio)
+{
+	struct radeon_i2c_bus_rec i2c;
+
+	memset(&i2c, 0, sizeof(struct radeon_i2c_bus_rec));
+
+	i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4;
+	i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4;
+	i2c.en_clk_reg = le16_to_cpu(gpio->usClkEnRegisterIndex) * 4;
+	i2c.en_data_reg = le16_to_cpu(gpio->usDataEnRegisterIndex) * 4;
+	i2c.y_clk_reg = le16_to_cpu(gpio->usClkY_RegisterIndex) * 4;
+	i2c.y_data_reg = le16_to_cpu(gpio->usDataY_RegisterIndex) * 4;
+	i2c.a_clk_reg = le16_to_cpu(gpio->usClkA_RegisterIndex) * 4;
+	i2c.a_data_reg = le16_to_cpu(gpio->usDataA_RegisterIndex) * 4;
+	i2c.mask_clk_mask = (1 << gpio->ucClkMaskShift);
+	i2c.mask_data_mask = (1 << gpio->ucDataMaskShift);
+	i2c.en_clk_mask = (1 << gpio->ucClkEnShift);
+	i2c.en_data_mask = (1 << gpio->ucDataEnShift);
+	i2c.y_clk_mask = (1 << gpio->ucClkY_Shift);
+	i2c.y_data_mask = (1 << gpio->ucDataY_Shift);
+	i2c.a_clk_mask = (1 << gpio->ucClkA_Shift);
+	i2c.a_data_mask = (1 << gpio->ucDataA_Shift);
+
+	if (gpio->sucI2cId.sbfAccess.bfHW_Capable)
+		i2c.hw_capable = true;
+	else
+		i2c.hw_capable = false;
+
+	if (gpio->sucI2cId.ucAccess == 0xa0)
+		i2c.mm_i2c = true;
+	else
+		i2c.mm_i2c = false;
+
+	i2c.i2c_id = gpio->sucI2cId.ucAccess;
+
+	if (i2c.mask_clk_reg)
+		i2c.valid = true;
+	else
+		i2c.valid = false;
+
+	return i2c;
+}
+
 static struct radeon_i2c_bus_rec radeon_lookup_i2c_gpio(struct radeon_device *rdev,
 							       uint8_t id)
 {
@@ -85,59 +166,10 @@
 		for (i = 0; i < num_indices; i++) {
 			gpio = &i2c_info->asGPIO_Info[i];
 
-			/* some evergreen boards have bad data for this entry */
-			if (ASIC_IS_DCE4(rdev)) {
-				if ((i == 7) &&
-				    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1936) &&
-				    (gpio->sucI2cId.ucAccess == 0)) {
-					gpio->sucI2cId.ucAccess = 0x97;
-					gpio->ucDataMaskShift = 8;
-					gpio->ucDataEnShift = 8;
-					gpio->ucDataY_Shift = 8;
-					gpio->ucDataA_Shift = 8;
-				}
-			}
-
-			/* some DCE3 boards have bad data for this entry */
-			if (ASIC_IS_DCE3(rdev)) {
-				if ((i == 4) &&
-				    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1fda) &&
-				    (gpio->sucI2cId.ucAccess == 0x94))
-					gpio->sucI2cId.ucAccess = 0x14;
-			}
+			radeon_lookup_i2c_gpio_quirks(rdev, gpio, i);
 
 			if (gpio->sucI2cId.ucAccess == id) {
-				i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4;
-				i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4;
-				i2c.en_clk_reg = le16_to_cpu(gpio->usClkEnRegisterIndex) * 4;
-				i2c.en_data_reg = le16_to_cpu(gpio->usDataEnRegisterIndex) * 4;
-				i2c.y_clk_reg = le16_to_cpu(gpio->usClkY_RegisterIndex) * 4;
-				i2c.y_data_reg = le16_to_cpu(gpio->usDataY_RegisterIndex) * 4;
-				i2c.a_clk_reg = le16_to_cpu(gpio->usClkA_RegisterIndex) * 4;
-				i2c.a_data_reg = le16_to_cpu(gpio->usDataA_RegisterIndex) * 4;
-				i2c.mask_clk_mask = (1 << gpio->ucClkMaskShift);
-				i2c.mask_data_mask = (1 << gpio->ucDataMaskShift);
-				i2c.en_clk_mask = (1 << gpio->ucClkEnShift);
-				i2c.en_data_mask = (1 << gpio->ucDataEnShift);
-				i2c.y_clk_mask = (1 << gpio->ucClkY_Shift);
-				i2c.y_data_mask = (1 << gpio->ucDataY_Shift);
-				i2c.a_clk_mask = (1 << gpio->ucClkA_Shift);
-				i2c.a_data_mask = (1 << gpio->ucDataA_Shift);
-
-				if (gpio->sucI2cId.sbfAccess.bfHW_Capable)
-					i2c.hw_capable = true;
-				else
-					i2c.hw_capable = false;
-
-				if (gpio->sucI2cId.ucAccess == 0xa0)
-					i2c.mm_i2c = true;
-				else
-					i2c.mm_i2c = false;
-
-				i2c.i2c_id = gpio->sucI2cId.ucAccess;
-
-				if (i2c.mask_clk_reg)
-					i2c.valid = true;
+				i2c = radeon_get_bus_rec_for_i2c_gpio(gpio);
 				break;
 			}
 		}
@@ -157,8 +189,6 @@
 	int i, num_indices;
 	char stmp[32];
 
-	memset(&i2c, 0, sizeof(struct radeon_i2c_bus_rec));
-
 	if (atom_parse_data_header(ctx, index, &size, NULL, NULL, &data_offset)) {
 		i2c_info = (struct _ATOM_GPIO_I2C_INFO *)(ctx->bios + data_offset);
 
@@ -167,60 +197,12 @@
 
 		for (i = 0; i < num_indices; i++) {
 			gpio = &i2c_info->asGPIO_Info[i];
-			i2c.valid = false;
 
-			/* some evergreen boards have bad data for this entry */
-			if (ASIC_IS_DCE4(rdev)) {
-				if ((i == 7) &&
-				    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1936) &&
-				    (gpio->sucI2cId.ucAccess == 0)) {
-					gpio->sucI2cId.ucAccess = 0x97;
-					gpio->ucDataMaskShift = 8;
-					gpio->ucDataEnShift = 8;
-					gpio->ucDataY_Shift = 8;
-					gpio->ucDataA_Shift = 8;
-				}
-			}
+			radeon_lookup_i2c_gpio_quirks(rdev, gpio, i);
 
-			/* some DCE3 boards have bad data for this entry */
-			if (ASIC_IS_DCE3(rdev)) {
-				if ((i == 4) &&
-				    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1fda) &&
-				    (gpio->sucI2cId.ucAccess == 0x94))
-					gpio->sucI2cId.ucAccess = 0x14;
-			}
+			i2c = radeon_get_bus_rec_for_i2c_gpio(gpio);
 
-			i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4;
-			i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4;
-			i2c.en_clk_reg = le16_to_cpu(gpio->usClkEnRegisterIndex) * 4;
-			i2c.en_data_reg = le16_to_cpu(gpio->usDataEnRegisterIndex) * 4;
-			i2c.y_clk_reg = le16_to_cpu(gpio->usClkY_RegisterIndex) * 4;
-			i2c.y_data_reg = le16_to_cpu(gpio->usDataY_RegisterIndex) * 4;
-			i2c.a_clk_reg = le16_to_cpu(gpio->usClkA_RegisterIndex) * 4;
-			i2c.a_data_reg = le16_to_cpu(gpio->usDataA_RegisterIndex) * 4;
-			i2c.mask_clk_mask = (1 << gpio->ucClkMaskShift);
-			i2c.mask_data_mask = (1 << gpio->ucDataMaskShift);
-			i2c.en_clk_mask = (1 << gpio->ucClkEnShift);
-			i2c.en_data_mask = (1 << gpio->ucDataEnShift);
-			i2c.y_clk_mask = (1 << gpio->ucClkY_Shift);
-			i2c.y_data_mask = (1 << gpio->ucDataY_Shift);
-			i2c.a_clk_mask = (1 << gpio->ucClkA_Shift);
-			i2c.a_data_mask = (1 << gpio->ucDataA_Shift);
-
-			if (gpio->sucI2cId.sbfAccess.bfHW_Capable)
-				i2c.hw_capable = true;
-			else
-				i2c.hw_capable = false;
-
-			if (gpio->sucI2cId.ucAccess == 0xa0)
-				i2c.mm_i2c = true;
-			else
-				i2c.mm_i2c = false;
-
-			i2c.i2c_id = gpio->sucI2cId.ucAccess;
-
-			if (i2c.mask_clk_reg) {
-				i2c.valid = true;
+			if (i2c.valid) {
 				sprintf(stmp, "0x%x", i2c.i2c_id);
 				rdev->i2c_bus[i] = radeon_i2c_create(rdev->ddev, &i2c, stmp);
 			}
@@ -1996,10 +1978,14 @@
 		return state_index;
 	/* last mode is usually default, array is low to high */
 	for (i = 0; i < num_modes; i++) {
+		rdev->pm.power_state[state_index].clock_info =
+			kzalloc(sizeof(struct radeon_pm_clock_info) * 1, GFP_KERNEL);
+		if (!rdev->pm.power_state[state_index].clock_info)
+			return state_index;
+		rdev->pm.power_state[state_index].num_clock_modes = 1;
 		rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
 		switch (frev) {
 		case 1:
-			rdev->pm.power_state[state_index].num_clock_modes = 1;
 			rdev->pm.power_state[state_index].clock_info[0].mclk =
 				le16_to_cpu(power_info->info.asPowerPlayInfo[i].usMemoryClock);
 			rdev->pm.power_state[state_index].clock_info[0].sclk =
@@ -2035,7 +2021,6 @@
 			state_index++;
 			break;
 		case 2:
-			rdev->pm.power_state[state_index].num_clock_modes = 1;
 			rdev->pm.power_state[state_index].clock_info[0].mclk =
 				le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulMemoryClock);
 			rdev->pm.power_state[state_index].clock_info[0].sclk =
@@ -2072,7 +2057,6 @@
 			state_index++;
 			break;
 		case 3:
-			rdev->pm.power_state[state_index].num_clock_modes = 1;
 			rdev->pm.power_state[state_index].clock_info[0].mclk =
 				le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulMemoryClock);
 			rdev->pm.power_state[state_index].clock_info[0].sclk =
@@ -2257,7 +2241,7 @@
 		rdev->pm.default_power_state_index = state_index;
 		rdev->pm.power_state[state_index].default_clock_mode =
 			&rdev->pm.power_state[state_index].clock_info[mode_index - 1];
-		if (ASIC_IS_DCE5(rdev)) {
+		if (ASIC_IS_DCE5(rdev) && !(rdev->flags & RADEON_IS_IGP)) {
 			/* NI chips post without MC ucode, so default clocks are strobe mode only */
 			rdev->pm.default_sclk = rdev->pm.power_state[state_index].clock_info[0].sclk;
 			rdev->pm.default_mclk = rdev->pm.power_state[state_index].clock_info[0].mclk;
@@ -2377,17 +2361,31 @@
 			 le16_to_cpu(power_info->pplib.usNonClockInfoArrayOffset) +
 			 (power_state->v1.ucNonClockStateIndex *
 			  power_info->pplib.ucNonClockSize));
-		for (j = 0; j < (power_info->pplib.ucStateEntrySize - 1); j++) {
-			clock_info = (union pplib_clock_info *)
-				(mode_info->atom_context->bios + data_offset +
-				 le16_to_cpu(power_info->pplib.usClockInfoArrayOffset) +
-				 (power_state->v1.ucClockStateIndices[j] *
-				  power_info->pplib.ucClockInfoSize));
-			valid = radeon_atombios_parse_pplib_clock_info(rdev,
-								       state_index, mode_index,
-								       clock_info);
-			if (valid)
-				mode_index++;
+		rdev->pm.power_state[i].clock_info = kzalloc(sizeof(struct radeon_pm_clock_info) *
+							     ((power_info->pplib.ucStateEntrySize - 1) ?
+							      (power_info->pplib.ucStateEntrySize - 1) : 1),
+							     GFP_KERNEL);
+		if (!rdev->pm.power_state[i].clock_info)
+			return state_index;
+		if (power_info->pplib.ucStateEntrySize - 1) {
+			for (j = 0; j < (power_info->pplib.ucStateEntrySize - 1); j++) {
+				clock_info = (union pplib_clock_info *)
+					(mode_info->atom_context->bios + data_offset +
+					 le16_to_cpu(power_info->pplib.usClockInfoArrayOffset) +
+					 (power_state->v1.ucClockStateIndices[j] *
+					  power_info->pplib.ucClockInfoSize));
+				valid = radeon_atombios_parse_pplib_clock_info(rdev,
+									       state_index, mode_index,
+									       clock_info);
+				if (valid)
+					mode_index++;
+			}
+		} else {
+			rdev->pm.power_state[state_index].clock_info[0].mclk =
+				rdev->clock.default_mclk;
+			rdev->pm.power_state[state_index].clock_info[0].sclk =
+				rdev->clock.default_sclk;
+			mode_index++;
 		}
 		rdev->pm.power_state[state_index].num_clock_modes = mode_index;
 		if (mode_index) {
@@ -2456,18 +2454,32 @@
 		non_clock_array_index = i; /* power_state->v2.nonClockInfoIndex */
 		non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
 			&non_clock_info_array->nonClockInfo[non_clock_array_index];
-		for (j = 0; j < power_state->v2.ucNumDPMLevels; j++) {
-			clock_array_index = power_state->v2.clockInfoIndex[j];
-			/* XXX this might be an inagua bug... */
-			if (clock_array_index >= clock_info_array->ucNumEntries)
-				continue;
-			clock_info = (union pplib_clock_info *)
-				&clock_info_array->clockInfo[clock_array_index];
-			valid = radeon_atombios_parse_pplib_clock_info(rdev,
-								       state_index, mode_index,
-								       clock_info);
-			if (valid)
-				mode_index++;
+		rdev->pm.power_state[i].clock_info = kzalloc(sizeof(struct radeon_pm_clock_info) *
+							     (power_state->v2.ucNumDPMLevels ?
+							      power_state->v2.ucNumDPMLevels : 1),
+							     GFP_KERNEL);
+		if (!rdev->pm.power_state[i].clock_info)
+			return state_index;
+		if (power_state->v2.ucNumDPMLevels) {
+			for (j = 0; j < power_state->v2.ucNumDPMLevels; j++) {
+				clock_array_index = power_state->v2.clockInfoIndex[j];
+				/* XXX this might be an inagua bug... */
+				if (clock_array_index >= clock_info_array->ucNumEntries)
+					continue;
+				clock_info = (union pplib_clock_info *)
+					&clock_info_array->clockInfo[clock_array_index];
+				valid = radeon_atombios_parse_pplib_clock_info(rdev,
+									       state_index, mode_index,
+									       clock_info);
+				if (valid)
+					mode_index++;
+			}
+		} else {
+			rdev->pm.power_state[state_index].clock_info[0].mclk =
+				rdev->clock.default_mclk;
+			rdev->pm.power_state[state_index].clock_info[0].sclk =
+				rdev->clock.default_sclk;
+			mode_index++;
 		}
 		rdev->pm.power_state[state_index].num_clock_modes = mode_index;
 		if (mode_index) {
@@ -2524,19 +2536,23 @@
 	} else {
 		rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state), GFP_KERNEL);
 		if (rdev->pm.power_state) {
-			/* add the default mode */
-			rdev->pm.power_state[state_index].type =
-				POWER_STATE_TYPE_DEFAULT;
-			rdev->pm.power_state[state_index].num_clock_modes = 1;
-			rdev->pm.power_state[state_index].clock_info[0].mclk = rdev->clock.default_mclk;
-			rdev->pm.power_state[state_index].clock_info[0].sclk = rdev->clock.default_sclk;
-			rdev->pm.power_state[state_index].default_clock_mode =
-				&rdev->pm.power_state[state_index].clock_info[0];
-			rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
-			rdev->pm.power_state[state_index].pcie_lanes = 16;
-			rdev->pm.default_power_state_index = state_index;
-			rdev->pm.power_state[state_index].flags = 0;
-			state_index++;
+			rdev->pm.power_state[0].clock_info =
+				kzalloc(sizeof(struct radeon_pm_clock_info) * 1, GFP_KERNEL);
+			if (rdev->pm.power_state[0].clock_info) {
+				/* add the default mode */
+				rdev->pm.power_state[state_index].type =
+					POWER_STATE_TYPE_DEFAULT;
+				rdev->pm.power_state[state_index].num_clock_modes = 1;
+				rdev->pm.power_state[state_index].clock_info[0].mclk = rdev->clock.default_mclk;
+				rdev->pm.power_state[state_index].clock_info[0].sclk = rdev->clock.default_sclk;
+				rdev->pm.power_state[state_index].default_clock_mode =
+					&rdev->pm.power_state[state_index].clock_info[0];
+				rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
+				rdev->pm.power_state[state_index].pcie_lanes = 16;
+				rdev->pm.default_power_state_index = state_index;
+				rdev->pm.power_state[state_index].flags = 0;
+				state_index++;
+			}
 		}
 	}
 
diff --git a/drivers/gpu/drm/radeon/radeon_benchmark.c b/drivers/gpu/drm/radeon/radeon_benchmark.c
index 5cafc90..4f749a69 100644
--- a/drivers/gpu/drm/radeon/radeon_benchmark.c
+++ b/drivers/gpu/drm/radeon/radeon_benchmark.c
@@ -43,7 +43,7 @@
 
 	start_jiffies = jiffies;
 	for (i = 0; i < n; i++) {
-		r = radeon_fence_create(rdev, &fence);
+		r = radeon_fence_create(rdev, &fence, RADEON_RING_TYPE_GFX_INDEX);
 		if (r)
 			return r;
 
@@ -98,7 +98,7 @@
 	struct radeon_bo *sobj = NULL;
 	uint64_t saddr, daddr;
 	int r, n;
-	unsigned int time;
+	int time;
 
 	n = RADEON_BENCHMARK_ITERATIONS;
 	r = radeon_bo_create(rdev, size, PAGE_SIZE, true, sdomain, &sobj);
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index 8bf83c4..81fc100 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -2563,14 +2563,17 @@
 
 	/* allocate 2 power states */
 	rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state) * 2, GFP_KERNEL);
-	if (!rdev->pm.power_state) {
-		rdev->pm.default_power_state_index = state_index;
-		rdev->pm.num_power_states = 0;
-
-		rdev->pm.current_power_state_index = rdev->pm.default_power_state_index;
-		rdev->pm.current_clock_mode_index = 0;
-		return;
-	}
+	if (rdev->pm.power_state) {
+		/* allocate 1 clock mode per state */
+		rdev->pm.power_state[0].clock_info =
+			kzalloc(sizeof(struct radeon_pm_clock_info) * 1, GFP_KERNEL);
+		rdev->pm.power_state[1].clock_info =
+			kzalloc(sizeof(struct radeon_pm_clock_info) * 1, GFP_KERNEL);
+		if (!rdev->pm.power_state[0].clock_info ||
+		    !rdev->pm.power_state[1].clock_info)
+			goto pm_failed;
+	} else
+		goto pm_failed;
 
 	/* check for a thermal chip */
 	offset = combios_get_table_offset(dev, COMBIOS_OVERDRIVE_INFO_TABLE);
@@ -2735,6 +2738,14 @@
 
 	rdev->pm.current_power_state_index = rdev->pm.default_power_state_index;
 	rdev->pm.current_clock_mode_index = 0;
+	return;
+
+pm_failed:
+	rdev->pm.default_power_state_index = state_index;
+	rdev->pm.num_power_states = 0;
+
+	rdev->pm.current_power_state_index = rdev->pm.default_power_state_index;
+	rdev->pm.current_clock_mode_index = 0;
 }
 
 void radeon_external_tmds_setup(struct drm_encoder *encoder)
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index fae00c0..09ef486 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -58,7 +58,7 @@
 
 		duplicate = false;
 		r = (struct drm_radeon_cs_reloc *)&chunk->kdata[i*4];
-		for (j = 0; j < p->nrelocs; j++) {
+		for (j = 0; j < i; j++) {
 			if (r->handle == p->relocs[j].handle) {
 				p->relocs_ptr[i] = &p->relocs[j];
 				duplicate = true;
@@ -84,7 +84,8 @@
 			p->relocs[i].flags = r->flags;
 			radeon_bo_list_add_object(&p->relocs[i].lobj,
 						  &p->validated);
-		}
+		} else
+			p->relocs[i].handle = 0;
 	}
 	return radeon_bo_list_validate(&p->validated);
 }
@@ -93,7 +94,7 @@
 {
 	struct drm_radeon_cs *cs = data;
 	uint64_t *chunk_array_ptr;
-	unsigned size, i;
+	unsigned size, i, flags = 0;
 
 	if (!cs->num_chunks) {
 		return 0;
@@ -140,6 +141,10 @@
 			if (p->chunks[i].length_dw == 0)
 				return -EINVAL;
 		}
+		if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS &&
+		    !p->chunks[i].length_dw) {
+			return -EINVAL;
+		}
 
 		p->chunks[i].length_dw = user_chunk.length_dw;
 		p->chunks[i].user_ptr = (void __user *)(unsigned long)user_chunk.chunk_data;
@@ -155,6 +160,9 @@
 					       p->chunks[i].user_ptr, size)) {
 				return -EFAULT;
 			}
+			if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS) {
+				flags = p->chunks[i].kdata[0];
+			}
 		} else {
 			p->chunks[i].kpage[0] = kmalloc(PAGE_SIZE, GFP_KERNEL);
 			p->chunks[i].kpage[1] = kmalloc(PAGE_SIZE, GFP_KERNEL);
@@ -174,6 +182,8 @@
 			  p->chunks[p->chunk_ib_idx].length_dw);
 		return -EINVAL;
 	}
+
+	p->keep_tiling_flags = (flags & RADEON_CS_KEEP_TILING_FLAGS) != 0;
 	return 0;
 }
 
@@ -222,7 +232,7 @@
 	struct radeon_cs_chunk *ib_chunk;
 	int r;
 
-	mutex_lock(&rdev->cs_mutex);
+	radeon_mutex_lock(&rdev->cs_mutex);
 	/* initialize parser */
 	memset(&parser, 0, sizeof(struct radeon_cs_parser));
 	parser.filp = filp;
@@ -233,14 +243,14 @@
 	if (r) {
 		DRM_ERROR("Failed to initialize parser !\n");
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
-	r =  radeon_ib_get(rdev, &parser.ib);
+	r =  radeon_ib_get(rdev, RADEON_RING_TYPE_GFX_INDEX, &parser.ib);
 	if (r) {
 		DRM_ERROR("Failed to get ib !\n");
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	r = radeon_cs_parser_relocs(&parser);
@@ -248,7 +258,7 @@
 		if (r != -ERESTARTSYS)
 			DRM_ERROR("Failed to parse relocation %d!\n", r);
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	/* Copy the packet into the IB, the parser will read from the
@@ -260,14 +270,14 @@
 	if (r || parser.parser_error) {
 		DRM_ERROR("Invalid command stream !\n");
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	r = radeon_cs_finish_pages(&parser);
 	if (r) {
 		DRM_ERROR("Invalid command stream !\n");
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	r = radeon_ib_schedule(rdev, parser.ib);
@@ -275,7 +285,7 @@
 		DRM_ERROR("Failed to schedule IB !\n");
 	}
 	radeon_cs_parser_fini(&parser, r);
-	mutex_unlock(&rdev->cs_mutex);
+	radeon_mutex_unlock(&rdev->cs_mutex);
 	return r;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 7c31321..e81c333 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -716,19 +716,22 @@
 
 	/* mutex initialization are all done here so we
 	 * can recall function without having locking issues */
-	mutex_init(&rdev->cs_mutex);
+	radeon_mutex_init(&rdev->cs_mutex);
 	mutex_init(&rdev->ib_pool.mutex);
-	mutex_init(&rdev->cp.mutex);
+	for (i = 0; i < RADEON_NUM_RINGS; ++i)
+		mutex_init(&rdev->ring[i].mutex);
 	mutex_init(&rdev->dc_hw_i2c_mutex);
 	if (rdev->family >= CHIP_R600)
 		spin_lock_init(&rdev->ih.lock);
 	mutex_init(&rdev->gem.mutex);
 	mutex_init(&rdev->pm.mutex);
 	mutex_init(&rdev->vram_mutex);
-	rwlock_init(&rdev->fence_drv.lock);
+	rwlock_init(&rdev->fence_lock);
+	rwlock_init(&rdev->semaphore_drv.lock);
 	INIT_LIST_HEAD(&rdev->gem.objects);
 	init_waitqueue_head(&rdev->irq.vblank_queue);
 	init_waitqueue_head(&rdev->irq.idle_queue);
+	INIT_LIST_HEAD(&rdev->semaphore_drv.free);
 
 	/* Set asic functions */
 	r = radeon_asic_init(rdev);
@@ -820,15 +823,20 @@
 		if (r)
 			return r;
 	}
-	if (radeon_testing) {
+	if ((radeon_testing & 1)) {
 		radeon_test_moves(rdev);
 	}
+	if ((radeon_testing & 2)) {
+		radeon_test_syncing(rdev);
+	}
 	if (radeon_benchmarking) {
 		radeon_benchmark(rdev, radeon_benchmarking);
 	}
 	return 0;
 }
 
+static void radeon_debugfs_remove_files(struct radeon_device *rdev);
+
 void radeon_device_fini(struct radeon_device *rdev)
 {
 	DRM_INFO("radeon: finishing device.\n");
@@ -843,6 +851,7 @@
 	rdev->rio_mem = NULL;
 	iounmap(rdev->rmmio);
 	rdev->rmmio = NULL;
+	radeon_debugfs_remove_files(rdev);
 }
 
 
@@ -854,7 +863,7 @@
 	struct radeon_device *rdev;
 	struct drm_crtc *crtc;
 	struct drm_connector *connector;
-	int r;
+	int i, r;
 
 	if (dev == NULL || dev->dev_private == NULL) {
 		return -ENODEV;
@@ -893,7 +902,8 @@
 	/* evict vram memory */
 	radeon_bo_evict_vram(rdev);
 	/* wait for gpu to finish processing current batch */
-	radeon_fence_wait_last(rdev);
+	for (i = 0; i < RADEON_NUM_RINGS; i++)
+		radeon_fence_wait_last(rdev, i);
 
 	radeon_save_bios_scratch_regs(rdev);
 
@@ -961,6 +971,9 @@
 	int r;
 	int resched;
 
+	/* Prevent CS ioctl from interfering */
+	radeon_mutex_lock(&rdev->cs_mutex);
+
 	radeon_save_bios_scratch_regs(rdev);
 	/* block TTM */
 	resched = ttm_bo_lock_delayed_workqueue(&rdev->mman.bdev);
@@ -973,10 +986,15 @@
 		radeon_restore_bios_scratch_regs(rdev);
 		drm_helper_resume_force_mode(rdev->ddev);
 		ttm_bo_unlock_delayed_workqueue(&rdev->mman.bdev, resched);
-		return 0;
 	}
-	/* bad news, how to tell it to userspace ? */
-	dev_info(rdev->dev, "GPU reset failed\n");
+
+	radeon_mutex_unlock(&rdev->cs_mutex);
+
+	if (r) {
+		/* bad news, how to tell it to userspace ? */
+		dev_info(rdev->dev, "GPU reset failed\n");
+	}
+
 	return r;
 }
 
@@ -984,36 +1002,29 @@
 /*
  * Debugfs
  */
-struct radeon_debugfs {
-	struct drm_info_list	*files;
-	unsigned		num_files;
-};
-static struct radeon_debugfs _radeon_debugfs[RADEON_DEBUGFS_MAX_COMPONENTS];
-static unsigned _radeon_debugfs_count = 0;
-
 int radeon_debugfs_add_files(struct radeon_device *rdev,
 			     struct drm_info_list *files,
 			     unsigned nfiles)
 {
 	unsigned i;
 
-	for (i = 0; i < _radeon_debugfs_count; i++) {
-		if (_radeon_debugfs[i].files == files) {
+	for (i = 0; i < rdev->debugfs_count; i++) {
+		if (rdev->debugfs[i].files == files) {
 			/* Already registered */
 			return 0;
 		}
 	}
 
-	i = _radeon_debugfs_count + 1;
+	i = rdev->debugfs_count + 1;
 	if (i > RADEON_DEBUGFS_MAX_COMPONENTS) {
 		DRM_ERROR("Reached maximum number of debugfs components.\n");
 		DRM_ERROR("Report so we increase "
 		          "RADEON_DEBUGFS_MAX_COMPONENTS.\n");
 		return -EINVAL;
 	}
-	_radeon_debugfs[_radeon_debugfs_count].files = files;
-	_radeon_debugfs[_radeon_debugfs_count].num_files = nfiles;
-	_radeon_debugfs_count = i;
+	rdev->debugfs[rdev->debugfs_count].files = files;
+	rdev->debugfs[rdev->debugfs_count].num_files = nfiles;
+	rdev->debugfs_count = i;
 #if defined(CONFIG_DEBUG_FS)
 	drm_debugfs_create_files(files, nfiles,
 				 rdev->ddev->control->debugfs_root,
@@ -1025,6 +1036,22 @@
 	return 0;
 }
 
+static void radeon_debugfs_remove_files(struct radeon_device *rdev)
+{
+#if defined(CONFIG_DEBUG_FS)
+	unsigned i;
+
+	for (i = 0; i < rdev->debugfs_count; i++) {
+		drm_debugfs_remove_files(rdev->debugfs[i].files,
+					 rdev->debugfs[i].num_files,
+					 rdev->ddev->control);
+		drm_debugfs_remove_files(rdev->debugfs[i].files,
+					 rdev->debugfs[i].num_files,
+					 rdev->ddev->primary);
+	}
+#endif
+}
+
 #if defined(CONFIG_DEBUG_FS)
 int radeon_debugfs_init(struct drm_minor *minor)
 {
@@ -1033,11 +1060,5 @@
 
 void radeon_debugfs_cleanup(struct drm_minor *minor)
 {
-	unsigned i;
-
-	for (i = 0; i < _radeon_debugfs_count; i++) {
-		drm_debugfs_remove_files(_radeon_debugfs[i].files,
-					 _radeon_debugfs[i].num_files, minor);
-	}
 }
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 96d9ba9..d3ffc18 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -406,7 +406,7 @@
 	if (!ASIC_IS_AVIVO(rdev)) {
 		/* crtc offset is from display base addr not FB location */
 		base -= radeon_crtc->legacy_display_base_addr;
-		pitch_pixels = fb->pitch / (fb->bits_per_pixel / 8);
+		pitch_pixels = fb->pitches[0] / (fb->bits_per_pixel / 8);
 
 		if (tiling_flags & RADEON_TILING_MACRO) {
 			if (ASIC_IS_R300(rdev)) {
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index e42c34b..c3ef1d2 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -53,9 +53,10 @@
  *   2.9.0 - r600 tiling (s3tc,rgtc) working, SET_PREDICATION packet 3 on r600 + eg, backend query
  *   2.10.0 - fusion 2D tiling
  *   2.11.0 - backend map, initial compute support for the CS checker
+ *   2.12.0 - RADEON_CS_KEEP_TILING_FLAGS
  */
 #define KMS_DRIVER_MAJOR	2
-#define KMS_DRIVER_MINOR	11
+#define KMS_DRIVER_MINOR	12
 #define KMS_DRIVER_PATCHLEVEL	0
 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags);
 int radeon_driver_unload_kms(struct drm_device *dev);
diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c
index 06e413e..4b27efa 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -233,13 +233,12 @@
 		switch (radeon_encoder->encoder_id) {
 		case ENCODER_OBJECT_ID_TRAVIS:
 		case ENCODER_OBJECT_ID_NUTMEG:
-			return true;
+			return radeon_encoder->encoder_id;
 		default:
-			return false;
+			return ENCODER_OBJECT_ID_NONE;
 		}
 	}
-
-	return false;
+	return ENCODER_OBJECT_ID_NONE;
 }
 
 void radeon_panel_mode_fixup(struct drm_encoder *encoder,
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index 0dc749e..cf2bf35 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -232,7 +232,7 @@
 
 	strcpy(info->fix.id, "radeondrmfb");
 
-	drm_fb_helper_fill_fix(info, fb->pitch, fb->depth);
+	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->depth);
 
 	info->flags = FBINFO_DEFAULT | FBINFO_CAN_FORCE_OUTPUT;
 	info->fbops = &radeonfb_ops;
@@ -275,7 +275,7 @@
 	DRM_INFO("vram apper at 0x%lX\n",  (unsigned long)rdev->mc.aper_base);
 	DRM_INFO("size %lu\n", (unsigned long)radeon_bo_size(rbo));
 	DRM_INFO("fb depth is %d\n", fb->depth);
-	DRM_INFO("   pitch is %d\n", fb->pitch);
+	DRM_INFO("   pitch is %d\n", fb->pitches[0]);
 
 	vga_switcheroo_client_fb_set(rdev->ddev->pdev, info);
 	return 0;
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index 76ec0e9..ae9e3da 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -40,32 +40,24 @@
 #include "radeon.h"
 #include "radeon_trace.h"
 
-static void radeon_fence_write(struct radeon_device *rdev, u32 seq)
+static void radeon_fence_write(struct radeon_device *rdev, u32 seq, int ring)
 {
 	if (rdev->wb.enabled) {
-		u32 scratch_index;
-		if (rdev->wb.use_event)
-			scratch_index = R600_WB_EVENT_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
-		else
-			scratch_index = RADEON_WB_SCRATCH_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
-		rdev->wb.wb[scratch_index/4] = cpu_to_le32(seq);
-	} else
-		WREG32(rdev->fence_drv.scratch_reg, seq);
+		*rdev->fence_drv[ring].cpu_addr = cpu_to_le32(seq);
+	} else {
+		WREG32(rdev->fence_drv[ring].scratch_reg, seq);
+	}
 }
 
-static u32 radeon_fence_read(struct radeon_device *rdev)
+static u32 radeon_fence_read(struct radeon_device *rdev, int ring)
 {
-	u32 seq;
+	u32 seq = 0;
 
 	if (rdev->wb.enabled) {
-		u32 scratch_index;
-		if (rdev->wb.use_event)
-			scratch_index = R600_WB_EVENT_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
-		else
-			scratch_index = RADEON_WB_SCRATCH_OFFSET + rdev->fence_drv.scratch_reg - rdev->scratch.reg_base;
-		seq = le32_to_cpu(rdev->wb.wb[scratch_index/4]);
-	} else
-		seq = RREG32(rdev->fence_drv.scratch_reg);
+		seq = le32_to_cpu(*rdev->fence_drv[ring].cpu_addr);
+	} else {
+		seq = RREG32(rdev->fence_drv[ring].scratch_reg);
+	}
 	return seq;
 }
 
@@ -73,28 +65,28 @@
 {
 	unsigned long irq_flags;
 
-	write_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-	if (fence->emited) {
-		write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	write_lock_irqsave(&rdev->fence_lock, irq_flags);
+	if (fence->emitted) {
+		write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 		return 0;
 	}
-	fence->seq = atomic_add_return(1, &rdev->fence_drv.seq);
-	if (!rdev->cp.ready)
+	fence->seq = atomic_add_return(1, &rdev->fence_drv[fence->ring].seq);
+	if (!rdev->ring[fence->ring].ready)
 		/* FIXME: cp is not running assume everythings is done right
 		 * away
 		 */
-		radeon_fence_write(rdev, fence->seq);
+		radeon_fence_write(rdev, fence->seq, fence->ring);
 	else
-		radeon_fence_ring_emit(rdev, fence);
+		radeon_fence_ring_emit(rdev, fence->ring, fence);
 
 	trace_radeon_fence_emit(rdev->ddev, fence->seq);
-	fence->emited = true;
-	list_move_tail(&fence->list, &rdev->fence_drv.emited);
-	write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	fence->emitted = true;
+	list_move_tail(&fence->list, &rdev->fence_drv[fence->ring].emitted);
+	write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 	return 0;
 }
 
-static bool radeon_fence_poll_locked(struct radeon_device *rdev)
+static bool radeon_fence_poll_locked(struct radeon_device *rdev, int ring)
 {
 	struct radeon_fence *fence;
 	struct list_head *i, *n;
@@ -102,34 +94,34 @@
 	bool wake = false;
 	unsigned long cjiffies;
 
-	seq = radeon_fence_read(rdev);
-	if (seq != rdev->fence_drv.last_seq) {
-		rdev->fence_drv.last_seq = seq;
-		rdev->fence_drv.last_jiffies = jiffies;
-		rdev->fence_drv.last_timeout = RADEON_FENCE_JIFFIES_TIMEOUT;
+	seq = radeon_fence_read(rdev, ring);
+	if (seq != rdev->fence_drv[ring].last_seq) {
+		rdev->fence_drv[ring].last_seq = seq;
+		rdev->fence_drv[ring].last_jiffies = jiffies;
+		rdev->fence_drv[ring].last_timeout = RADEON_FENCE_JIFFIES_TIMEOUT;
 	} else {
 		cjiffies = jiffies;
-		if (time_after(cjiffies, rdev->fence_drv.last_jiffies)) {
-			cjiffies -= rdev->fence_drv.last_jiffies;
-			if (time_after(rdev->fence_drv.last_timeout, cjiffies)) {
+		if (time_after(cjiffies, rdev->fence_drv[ring].last_jiffies)) {
+			cjiffies -= rdev->fence_drv[ring].last_jiffies;
+			if (time_after(rdev->fence_drv[ring].last_timeout, cjiffies)) {
 				/* update the timeout */
-				rdev->fence_drv.last_timeout -= cjiffies;
+				rdev->fence_drv[ring].last_timeout -= cjiffies;
 			} else {
 				/* the 500ms timeout is elapsed we should test
 				 * for GPU lockup
 				 */
-				rdev->fence_drv.last_timeout = 1;
+				rdev->fence_drv[ring].last_timeout = 1;
 			}
 		} else {
 			/* wrap around update last jiffies, we will just wait
 			 * a little longer
 			 */
-			rdev->fence_drv.last_jiffies = cjiffies;
+			rdev->fence_drv[ring].last_jiffies = cjiffies;
 		}
 		return false;
 	}
 	n = NULL;
-	list_for_each(i, &rdev->fence_drv.emited) {
+	list_for_each(i, &rdev->fence_drv[ring].emitted) {
 		fence = list_entry(i, struct radeon_fence, list);
 		if (fence->seq == seq) {
 			n = i;
@@ -141,11 +133,11 @@
 		i = n;
 		do {
 			n = i->prev;
-			list_move_tail(i, &rdev->fence_drv.signaled);
+			list_move_tail(i, &rdev->fence_drv[ring].signaled);
 			fence = list_entry(i, struct radeon_fence, list);
 			fence->signaled = true;
 			i = n;
-		} while (i != &rdev->fence_drv.emited);
+		} while (i != &rdev->fence_drv[ring].emitted);
 		wake = true;
 	}
 	return wake;
@@ -157,14 +149,16 @@
         struct radeon_fence *fence;
 
 	fence = container_of(kref, struct radeon_fence, kref);
-	write_lock_irqsave(&fence->rdev->fence_drv.lock, irq_flags);
+	write_lock_irqsave(&fence->rdev->fence_lock, irq_flags);
 	list_del(&fence->list);
-	fence->emited = false;
-	write_unlock_irqrestore(&fence->rdev->fence_drv.lock, irq_flags);
+	fence->emitted = false;
+	write_unlock_irqrestore(&fence->rdev->fence_lock, irq_flags);
 	kfree(fence);
 }
 
-int radeon_fence_create(struct radeon_device *rdev, struct radeon_fence **fence)
+int radeon_fence_create(struct radeon_device *rdev,
+			struct radeon_fence **fence,
+			int ring)
 {
 	unsigned long irq_flags;
 
@@ -174,18 +168,18 @@
 	}
 	kref_init(&((*fence)->kref));
 	(*fence)->rdev = rdev;
-	(*fence)->emited = false;
+	(*fence)->emitted = false;
 	(*fence)->signaled = false;
 	(*fence)->seq = 0;
+	(*fence)->ring = ring;
 	INIT_LIST_HEAD(&(*fence)->list);
 
-	write_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-	list_add_tail(&(*fence)->list, &rdev->fence_drv.created);
-	write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	write_lock_irqsave(&rdev->fence_lock, irq_flags);
+	list_add_tail(&(*fence)->list, &rdev->fence_drv[ring].created);
+	write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 	return 0;
 }
 
-
 bool radeon_fence_signaled(struct radeon_fence *fence)
 {
 	unsigned long irq_flags;
@@ -197,21 +191,21 @@
 	if (fence->rdev->gpu_lockup)
 		return true;
 
-	write_lock_irqsave(&fence->rdev->fence_drv.lock, irq_flags);
+	write_lock_irqsave(&fence->rdev->fence_lock, irq_flags);
 	signaled = fence->signaled;
 	/* if we are shuting down report all fence as signaled */
 	if (fence->rdev->shutdown) {
 		signaled = true;
 	}
-	if (!fence->emited) {
-		WARN(1, "Querying an unemited fence : %p !\n", fence);
+	if (!fence->emitted) {
+		WARN(1, "Querying an unemitted fence : %p !\n", fence);
 		signaled = true;
 	}
 	if (!signaled) {
-		radeon_fence_poll_locked(fence->rdev);
+		radeon_fence_poll_locked(fence->rdev, fence->ring);
 		signaled = fence->signaled;
 	}
-	write_unlock_irqrestore(&fence->rdev->fence_drv.lock, irq_flags);
+	write_unlock_irqrestore(&fence->rdev->fence_lock, irq_flags);
 	return signaled;
 }
 
@@ -230,24 +224,24 @@
 	if (radeon_fence_signaled(fence)) {
 		return 0;
 	}
-	timeout = rdev->fence_drv.last_timeout;
+	timeout = rdev->fence_drv[fence->ring].last_timeout;
 retry:
 	/* save current sequence used to check for GPU lockup */
-	seq = rdev->fence_drv.last_seq;
+	seq = rdev->fence_drv[fence->ring].last_seq;
 	trace_radeon_fence_wait_begin(rdev->ddev, seq);
 	if (intr) {
-		radeon_irq_kms_sw_irq_get(rdev);
-		r = wait_event_interruptible_timeout(rdev->fence_drv.queue,
+		radeon_irq_kms_sw_irq_get(rdev, fence->ring);
+		r = wait_event_interruptible_timeout(rdev->fence_drv[fence->ring].queue,
 				radeon_fence_signaled(fence), timeout);
-		radeon_irq_kms_sw_irq_put(rdev);
+		radeon_irq_kms_sw_irq_put(rdev, fence->ring);
 		if (unlikely(r < 0)) {
 			return r;
 		}
 	} else {
-		radeon_irq_kms_sw_irq_get(rdev);
-		r = wait_event_timeout(rdev->fence_drv.queue,
+		radeon_irq_kms_sw_irq_get(rdev, fence->ring);
+		r = wait_event_timeout(rdev->fence_drv[fence->ring].queue,
 			 radeon_fence_signaled(fence), timeout);
-		radeon_irq_kms_sw_irq_put(rdev);
+		radeon_irq_kms_sw_irq_put(rdev, fence->ring);
 	}
 	trace_radeon_fence_wait_end(rdev->ddev, seq);
 	if (unlikely(!radeon_fence_signaled(fence))) {
@@ -258,10 +252,11 @@
 			timeout = r;
 			goto retry;
 		}
-		/* don't protect read access to rdev->fence_drv.last_seq
+		/* don't protect read access to rdev->fence_drv[t].last_seq
 		 * if we experiencing a lockup the value doesn't change
 		 */
-		if (seq == rdev->fence_drv.last_seq && radeon_gpu_is_lockup(rdev)) {
+		if (seq == rdev->fence_drv[fence->ring].last_seq &&
+		    radeon_gpu_is_lockup(rdev, &rdev->ring[fence->ring])) {
 			/* good news we believe it's a lockup */
 			printk(KERN_WARNING "GPU lockup (waiting for 0x%08X last fence id 0x%08X)\n",
 			     fence->seq, seq);
@@ -272,20 +267,20 @@
 			r = radeon_gpu_reset(rdev);
 			if (r)
 				return r;
-			radeon_fence_write(rdev, fence->seq);
+			radeon_fence_write(rdev, fence->seq, fence->ring);
 			rdev->gpu_lockup = false;
 		}
 		timeout = RADEON_FENCE_JIFFIES_TIMEOUT;
-		write_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-		rdev->fence_drv.last_timeout = RADEON_FENCE_JIFFIES_TIMEOUT;
-		rdev->fence_drv.last_jiffies = jiffies;
-		write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+		write_lock_irqsave(&rdev->fence_lock, irq_flags);
+		rdev->fence_drv[fence->ring].last_timeout = RADEON_FENCE_JIFFIES_TIMEOUT;
+		rdev->fence_drv[fence->ring].last_jiffies = jiffies;
+		write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 		goto retry;
 	}
 	return 0;
 }
 
-int radeon_fence_wait_next(struct radeon_device *rdev)
+int radeon_fence_wait_next(struct radeon_device *rdev, int ring)
 {
 	unsigned long irq_flags;
 	struct radeon_fence *fence;
@@ -294,21 +289,21 @@
 	if (rdev->gpu_lockup) {
 		return 0;
 	}
-	write_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-	if (list_empty(&rdev->fence_drv.emited)) {
-		write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	write_lock_irqsave(&rdev->fence_lock, irq_flags);
+	if (list_empty(&rdev->fence_drv[ring].emitted)) {
+		write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 		return 0;
 	}
-	fence = list_entry(rdev->fence_drv.emited.next,
+	fence = list_entry(rdev->fence_drv[ring].emitted.next,
 			   struct radeon_fence, list);
 	radeon_fence_ref(fence);
-	write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 	r = radeon_fence_wait(fence, false);
 	radeon_fence_unref(&fence);
 	return r;
 }
 
-int radeon_fence_wait_last(struct radeon_device *rdev)
+int radeon_fence_wait_last(struct radeon_device *rdev, int ring)
 {
 	unsigned long irq_flags;
 	struct radeon_fence *fence;
@@ -317,15 +312,15 @@
 	if (rdev->gpu_lockup) {
 		return 0;
 	}
-	write_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-	if (list_empty(&rdev->fence_drv.emited)) {
-		write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	write_lock_irqsave(&rdev->fence_lock, irq_flags);
+	if (list_empty(&rdev->fence_drv[ring].emitted)) {
+		write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 		return 0;
 	}
-	fence = list_entry(rdev->fence_drv.emited.prev,
+	fence = list_entry(rdev->fence_drv[ring].emitted.prev,
 			   struct radeon_fence, list);
 	radeon_fence_ref(fence);
-	write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 	r = radeon_fence_wait(fence, false);
 	radeon_fence_unref(&fence);
 	return r;
@@ -347,39 +342,95 @@
 	}
 }
 
-void radeon_fence_process(struct radeon_device *rdev)
+void radeon_fence_process(struct radeon_device *rdev, int ring)
 {
 	unsigned long irq_flags;
 	bool wake;
 
-	write_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-	wake = radeon_fence_poll_locked(rdev);
-	write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	write_lock_irqsave(&rdev->fence_lock, irq_flags);
+	wake = radeon_fence_poll_locked(rdev, ring);
+	write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 	if (wake) {
-		wake_up_all(&rdev->fence_drv.queue);
+		wake_up_all(&rdev->fence_drv[ring].queue);
 	}
 }
 
+int radeon_fence_count_emitted(struct radeon_device *rdev, int ring)
+{
+	unsigned long irq_flags;
+	int not_processed = 0;
+
+	read_lock_irqsave(&rdev->fence_lock, irq_flags);
+	if (!rdev->fence_drv[ring].initialized)
+		return 0;
+
+	if (!list_empty(&rdev->fence_drv[ring].emitted)) {
+		struct list_head *ptr;
+		list_for_each(ptr, &rdev->fence_drv[ring].emitted) {
+			/* count up to 3, that's enought info */
+			if (++not_processed >= 3)
+				break;
+		}
+	}
+	read_unlock_irqrestore(&rdev->fence_lock, irq_flags);
+	return not_processed;
+}
+
+int radeon_fence_driver_start_ring(struct radeon_device *rdev, int ring)
+{
+	unsigned long irq_flags;
+	uint64_t index;
+	int r;
+
+	write_lock_irqsave(&rdev->fence_lock, irq_flags);
+	radeon_scratch_free(rdev, rdev->fence_drv[ring].scratch_reg);
+	if (rdev->wb.use_event) {
+		rdev->fence_drv[ring].scratch_reg = 0;
+		index = R600_WB_EVENT_OFFSET + ring * 4;
+	} else {
+		r = radeon_scratch_get(rdev, &rdev->fence_drv[ring].scratch_reg);
+		if (r) {
+			dev_err(rdev->dev, "fence failed to get scratch register\n");
+			write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
+			return r;
+		}
+		index = RADEON_WB_SCRATCH_OFFSET +
+			rdev->fence_drv[ring].scratch_reg -
+			rdev->scratch.reg_base;
+	}
+	rdev->fence_drv[ring].cpu_addr = &rdev->wb.wb[index/4];
+	rdev->fence_drv[ring].gpu_addr = rdev->wb.gpu_addr + index;
+	radeon_fence_write(rdev, atomic_read(&rdev->fence_drv[ring].seq), ring);
+	rdev->fence_drv[ring].initialized = true;
+	DRM_INFO("fence driver on ring %d use gpu addr 0x%08Lx and cpu addr 0x%p\n",
+		 ring, rdev->fence_drv[ring].gpu_addr, rdev->fence_drv[ring].cpu_addr);
+	write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
+	return 0;
+}
+
+static void radeon_fence_driver_init_ring(struct radeon_device *rdev, int ring)
+{
+	rdev->fence_drv[ring].scratch_reg = -1;
+	rdev->fence_drv[ring].cpu_addr = NULL;
+	rdev->fence_drv[ring].gpu_addr = 0;
+	atomic_set(&rdev->fence_drv[ring].seq, 0);
+	INIT_LIST_HEAD(&rdev->fence_drv[ring].created);
+	INIT_LIST_HEAD(&rdev->fence_drv[ring].emitted);
+	INIT_LIST_HEAD(&rdev->fence_drv[ring].signaled);
+	init_waitqueue_head(&rdev->fence_drv[ring].queue);
+	rdev->fence_drv[ring].initialized = false;
+}
+
 int radeon_fence_driver_init(struct radeon_device *rdev)
 {
 	unsigned long irq_flags;
-	int r;
+	int ring;
 
-	write_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-	r = radeon_scratch_get(rdev, &rdev->fence_drv.scratch_reg);
-	if (r) {
-		dev_err(rdev->dev, "fence failed to get scratch register\n");
-		write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
-		return r;
+	write_lock_irqsave(&rdev->fence_lock, irq_flags);
+	for (ring = 0; ring < RADEON_NUM_RINGS; ring++) {
+		radeon_fence_driver_init_ring(rdev, ring);
 	}
-	radeon_fence_write(rdev, 0);
-	atomic_set(&rdev->fence_drv.seq, 0);
-	INIT_LIST_HEAD(&rdev->fence_drv.created);
-	INIT_LIST_HEAD(&rdev->fence_drv.emited);
-	INIT_LIST_HEAD(&rdev->fence_drv.signaled);
-	init_waitqueue_head(&rdev->fence_drv.queue);
-	rdev->fence_drv.initialized = true;
-	write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
+	write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
 	if (radeon_debugfs_fence_init(rdev)) {
 		dev_err(rdev->dev, "fence debugfs file creation failed\n");
 	}
@@ -389,14 +440,18 @@
 void radeon_fence_driver_fini(struct radeon_device *rdev)
 {
 	unsigned long irq_flags;
+	int ring;
 
-	if (!rdev->fence_drv.initialized)
-		return;
-	wake_up_all(&rdev->fence_drv.queue);
-	write_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-	radeon_scratch_free(rdev, rdev->fence_drv.scratch_reg);
-	write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
-	rdev->fence_drv.initialized = false;
+	for (ring = 0; ring < RADEON_NUM_RINGS; ring++) {
+		if (!rdev->fence_drv[ring].initialized)
+			continue;
+		radeon_fence_wait_last(rdev, ring);
+		wake_up_all(&rdev->fence_drv[ring].queue);
+		write_lock_irqsave(&rdev->fence_lock, irq_flags);
+		radeon_scratch_free(rdev, rdev->fence_drv[ring].scratch_reg);
+		write_unlock_irqrestore(&rdev->fence_lock, irq_flags);
+		rdev->fence_drv[ring].initialized = false;
+	}
 }
 
 
@@ -410,14 +465,21 @@
 	struct drm_device *dev = node->minor->dev;
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_fence *fence;
+	int i;
 
-	seq_printf(m, "Last signaled fence 0x%08X\n",
-		   radeon_fence_read(rdev));
-	if (!list_empty(&rdev->fence_drv.emited)) {
-		   fence = list_entry(rdev->fence_drv.emited.prev,
-				      struct radeon_fence, list);
-		   seq_printf(m, "Last emited fence %p with 0x%08X\n",
-			      fence,  fence->seq);
+	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+		if (!rdev->fence_drv[i].initialized)
+			continue;
+
+		seq_printf(m, "--- ring %d ---\n", i);
+		seq_printf(m, "Last signaled fence 0x%08X\n",
+			   radeon_fence_read(rdev, i));
+		if (!list_empty(&rdev->fence_drv[i].emitted)) {
+			fence = list_entry(rdev->fence_drv[i].emitted.prev,
+					   struct radeon_fence, list);
+			seq_printf(m, "Last emitted fence %p with 0x%08X\n",
+				   fence,  fence->seq);
+		}
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index aa1ca2d..ae32197 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -152,6 +152,7 @@
 	struct radeon_device *rdev = dev->dev_private;
 	struct drm_radeon_gem_info *args = data;
 	struct ttm_mem_type_manager *man;
+	unsigned i;
 
 	man = &rdev->mman.bdev.man[TTM_PL_VRAM];
 
@@ -160,8 +161,9 @@
 	if (rdev->stollen_vga_memory)
 		args->vram_visible -= radeon_bo_size(rdev->stollen_vga_memory);
 	args->vram_visible -= radeon_fbdev_total_size(rdev);
-	args->gart_size = rdev->mc.gtt_size - rdev->cp.ring_size - 4096 -
-		RADEON_IB_POOL_SIZE*64*1024;
+	args->gart_size = rdev->mc.gtt_size - 4096 - RADEON_IB_POOL_SIZE*64*1024;
+	for(i = 0; i < RADEON_NUM_RINGS; ++i)
+		args->gart_size -= rdev->ring[i].ring_size;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index 8f86aeb..be38921 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -65,7 +65,8 @@
 	unsigned i;
 
 	/* Disable *all* interrupts */
-	rdev->irq.sw_int = false;
+	for (i = 0; i < RADEON_NUM_RINGS; i++)
+		rdev->irq.sw_int[i] = false;
 	rdev->irq.gui_idle = false;
 	for (i = 0; i < RADEON_MAX_HPD_PINS; i++)
 		rdev->irq.hpd[i] = false;
@@ -81,9 +82,11 @@
 int radeon_driver_irq_postinstall_kms(struct drm_device *dev)
 {
 	struct radeon_device *rdev = dev->dev_private;
+	unsigned i;
 
 	dev->max_vblank_count = 0x001fffff;
-	rdev->irq.sw_int = true;
+	for (i = 0; i < RADEON_NUM_RINGS; i++)
+		rdev->irq.sw_int[i] = true;
 	radeon_irq_set(rdev);
 	return 0;
 }
@@ -97,7 +100,8 @@
 		return;
 	}
 	/* Disable *all* interrupts */
-	rdev->irq.sw_int = false;
+	for (i = 0; i < RADEON_NUM_RINGS; i++)
+		rdev->irq.sw_int[i] = false;
 	rdev->irq.gui_idle = false;
 	for (i = 0; i < RADEON_MAX_HPD_PINS; i++)
 		rdev->irq.hpd[i] = false;
@@ -194,26 +198,26 @@
 	flush_work_sync(&rdev->hotplug_work);
 }
 
-void radeon_irq_kms_sw_irq_get(struct radeon_device *rdev)
+void radeon_irq_kms_sw_irq_get(struct radeon_device *rdev, int ring)
 {
 	unsigned long irqflags;
 
 	spin_lock_irqsave(&rdev->irq.sw_lock, irqflags);
-	if (rdev->ddev->irq_enabled && (++rdev->irq.sw_refcount == 1)) {
-		rdev->irq.sw_int = true;
+	if (rdev->ddev->irq_enabled && (++rdev->irq.sw_refcount[ring] == 1)) {
+		rdev->irq.sw_int[ring] = true;
 		radeon_irq_set(rdev);
 	}
 	spin_unlock_irqrestore(&rdev->irq.sw_lock, irqflags);
 }
 
-void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev)
+void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev, int ring)
 {
 	unsigned long irqflags;
 
 	spin_lock_irqsave(&rdev->irq.sw_lock, irqflags);
-	BUG_ON(rdev->ddev->irq_enabled && rdev->irq.sw_refcount <= 0);
-	if (rdev->ddev->irq_enabled && (--rdev->irq.sw_refcount == 0)) {
-		rdev->irq.sw_int = false;
+	BUG_ON(rdev->ddev->irq_enabled && rdev->irq.sw_refcount[ring] <= 0);
+	if (rdev->ddev->irq_enabled && (--rdev->irq.sw_refcount[ring] == 0)) {
+		rdev->irq.sw_int[ring] = false;
 		radeon_irq_set(rdev);
 	}
 	spin_unlock_irqrestore(&rdev->irq.sw_lock, irqflags);
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index 41a5d48..25a19c4 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -437,7 +437,7 @@
 
 	crtc_offset_cntl = 0;
 
-	pitch_pixels = target_fb->pitch / (target_fb->bits_per_pixel / 8);
+	pitch_pixels = target_fb->pitches[0] / (target_fb->bits_per_pixel / 8);
 	crtc_pitch  = (((pitch_pixels * target_fb->bits_per_pixel) +
 			((target_fb->bits_per_pixel * 8) - 1)) /
 		       (target_fb->bits_per_pixel * 8));
@@ -991,12 +991,6 @@
 				   struct drm_display_mode *mode,
 				   struct drm_display_mode *adjusted_mode)
 {
-	struct drm_device *dev = crtc->dev;
-	struct radeon_device *rdev = dev->dev_private;
-
-	/* adjust pm to upcoming mode change */
-	radeon_pm_compute_clocks(rdev);
-
 	if (!radeon_crtc_scaling_mode_fixup(crtc, mode, adjusted_mode))
 		return false;
 	return true;
diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h
index b07f0f9..cc236fb 100644
--- a/drivers/gpu/drm/radeon/radeon_object.h
+++ b/drivers/gpu/drm/radeon/radeon_object.h
@@ -128,4 +128,24 @@
 					struct ttm_mem_reg *mem);
 extern int radeon_bo_fault_reserve_notify(struct ttm_buffer_object *bo);
 extern int radeon_bo_get_surface_reg(struct radeon_bo *bo);
+
+/*
+ * sub allocation
+ */
+extern int radeon_sa_bo_manager_init(struct radeon_device *rdev,
+				     struct radeon_sa_manager *sa_manager,
+				     unsigned size, u32 domain);
+extern void radeon_sa_bo_manager_fini(struct radeon_device *rdev,
+				      struct radeon_sa_manager *sa_manager);
+extern int radeon_sa_bo_manager_start(struct radeon_device *rdev,
+				      struct radeon_sa_manager *sa_manager);
+extern int radeon_sa_bo_manager_suspend(struct radeon_device *rdev,
+					struct radeon_sa_manager *sa_manager);
+extern int radeon_sa_bo_new(struct radeon_device *rdev,
+			    struct radeon_sa_manager *sa_manager,
+			    struct radeon_sa_bo *sa_bo,
+			    unsigned size, unsigned align);
+extern void radeon_sa_bo_free(struct radeon_device *rdev,
+			      struct radeon_sa_bo *sa_bo);
+
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 6fabe89..095148e 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -53,6 +53,24 @@
 
 #define ACPI_AC_CLASS           "ac_adapter"
 
+int radeon_pm_get_type_index(struct radeon_device *rdev,
+			     enum radeon_pm_state_type ps_type,
+			     int instance)
+{
+	int i;
+	int found_instance = -1;
+
+	for (i = 0; i < rdev->pm.num_power_states; i++) {
+		if (rdev->pm.power_state[i].type == ps_type) {
+			found_instance++;
+			if (found_instance == instance)
+				return i;
+		}
+	}
+	/* return default if no match */
+	return rdev->pm.default_power_state_index;
+}
+
 #ifdef CONFIG_ACPI
 static int radeon_acpi_event(struct notifier_block *nb,
 			     unsigned long val,
@@ -234,7 +252,10 @@
 
 	mutex_lock(&rdev->ddev->struct_mutex);
 	mutex_lock(&rdev->vram_mutex);
-	mutex_lock(&rdev->cp.mutex);
+	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+		if (rdev->ring[i].ring_obj)
+			mutex_lock(&rdev->ring[i].mutex);
+	}
 
 	/* gui idle int has issues on older chips it seems */
 	if (rdev->family >= CHIP_R600) {
@@ -250,12 +271,13 @@
 			radeon_irq_set(rdev);
 		}
 	} else {
-		if (rdev->cp.ready) {
+		struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+		if (ring->ready) {
 			struct radeon_fence *fence;
-			radeon_ring_alloc(rdev, 64);
-			radeon_fence_create(rdev, &fence);
+			radeon_ring_alloc(rdev, ring, 64);
+			radeon_fence_create(rdev, &fence, radeon_ring_index(rdev, ring));
 			radeon_fence_emit(rdev, fence);
-			radeon_ring_commit(rdev);
+			radeon_ring_commit(rdev, ring);
 			radeon_fence_wait(fence, false);
 			radeon_fence_unref(&fence);
 		}
@@ -289,7 +311,10 @@
 
 	rdev->pm.dynpm_planned_action = DYNPM_ACTION_NONE;
 
-	mutex_unlock(&rdev->cp.mutex);
+	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+		if (rdev->ring[i].ring_obj)
+			mutex_unlock(&rdev->ring[i].mutex);
+	}
 	mutex_unlock(&rdev->vram_mutex);
 	mutex_unlock(&rdev->ddev->struct_mutex);
 }
@@ -777,19 +802,14 @@
 	resched = ttm_bo_lock_delayed_workqueue(&rdev->mman.bdev);
 	mutex_lock(&rdev->pm.mutex);
 	if (rdev->pm.dynpm_state == DYNPM_STATE_ACTIVE) {
-		unsigned long irq_flags;
 		int not_processed = 0;
+		int i;
 
-		read_lock_irqsave(&rdev->fence_drv.lock, irq_flags);
-		if (!list_empty(&rdev->fence_drv.emited)) {
-			struct list_head *ptr;
-			list_for_each(ptr, &rdev->fence_drv.emited) {
-				/* count up to 3, that's enought info */
-				if (++not_processed >= 3)
-					break;
-			}
+		for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+			not_processed += radeon_fence_count_emitted(rdev, i);
+			if (not_processed >= 3)
+				break;
 		}
-		read_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags);
 
 		if (not_processed >= 3) { /* should upclock */
 			if (rdev->pm.dynpm_planned_action == DYNPM_ACTION_DOWNCLOCK) {
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 49d5820..f6a4fbd 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -34,6 +34,7 @@
 #include "atom.h"
 
 int radeon_debugfs_ib_init(struct radeon_device *rdev);
+int radeon_debugfs_ring_init(struct radeon_device *rdev);
 
 u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx)
 {
@@ -60,105 +61,103 @@
 	return idx_value;
 }
 
-void radeon_ring_write(struct radeon_device *rdev, uint32_t v)
+void radeon_ring_write(struct radeon_ring *ring, uint32_t v)
 {
 #if DRM_DEBUG_CODE
-	if (rdev->cp.count_dw <= 0) {
+	if (ring->count_dw <= 0) {
 		DRM_ERROR("radeon: writting more dword to ring than expected !\n");
 	}
 #endif
-	rdev->cp.ring[rdev->cp.wptr++] = v;
-	rdev->cp.wptr &= rdev->cp.ptr_mask;
-	rdev->cp.count_dw--;
-	rdev->cp.ring_free_dw--;
-}
-
-void radeon_ib_bogus_cleanup(struct radeon_device *rdev)
-{
-	struct radeon_ib *ib, *n;
-
-	list_for_each_entry_safe(ib, n, &rdev->ib_pool.bogus_ib, list) {
-		list_del(&ib->list);
-		vfree(ib->ptr);
-		kfree(ib);
-	}
-}
-
-void radeon_ib_bogus_add(struct radeon_device *rdev, struct radeon_ib *ib)
-{
-	struct radeon_ib *bib;
-
-	bib = kmalloc(sizeof(*bib), GFP_KERNEL);
-	if (bib == NULL)
-		return;
-	bib->ptr = vmalloc(ib->length_dw * 4);
-	if (bib->ptr == NULL) {
-		kfree(bib);
-		return;
-	}
-	memcpy(bib->ptr, ib->ptr, ib->length_dw * 4);
-	bib->length_dw = ib->length_dw;
-	mutex_lock(&rdev->ib_pool.mutex);
-	list_add_tail(&bib->list, &rdev->ib_pool.bogus_ib);
-	mutex_unlock(&rdev->ib_pool.mutex);
+	ring->ring[ring->wptr++] = v;
+	ring->wptr &= ring->ptr_mask;
+	ring->count_dw--;
+	ring->ring_free_dw--;
 }
 
 /*
  * IB.
  */
-int radeon_ib_get(struct radeon_device *rdev, struct radeon_ib **ib)
+static bool radeon_ib_try_free(struct radeon_device *rdev,
+			       struct radeon_ib *ib)
+{
+	bool done = false;
+
+	/* only free ib which have been emited */
+	if (ib->fence && ib->fence->emitted) {
+		if (radeon_fence_signaled(ib->fence)) {
+			radeon_fence_unref(&ib->fence);
+			radeon_sa_bo_free(rdev, &ib->sa_bo);
+			done = true;
+		}
+	}
+	return done;
+}
+
+int radeon_ib_get(struct radeon_device *rdev, int ring, struct radeon_ib **ib)
 {
 	struct radeon_fence *fence;
-	struct radeon_ib *nib;
-	int r = 0, i, c;
+	unsigned cretry = 0;
+	int r = 0, i, idx;
 
 	*ib = NULL;
-	r = radeon_fence_create(rdev, &fence);
+
+	r = radeon_fence_create(rdev, &fence, ring);
 	if (r) {
 		dev_err(rdev->dev, "failed to create fence for new IB\n");
 		return r;
 	}
+
 	mutex_lock(&rdev->ib_pool.mutex);
-	for (i = rdev->ib_pool.head_id, c = 0, nib = NULL; c < RADEON_IB_POOL_SIZE; c++, i++) {
-		i &= (RADEON_IB_POOL_SIZE - 1);
-		if (rdev->ib_pool.ibs[i].free) {
-			nib = &rdev->ib_pool.ibs[i];
-			break;
-		}
-	}
-	if (nib == NULL) {
-		/* This should never happen, it means we allocated all
-		 * IB and haven't scheduled one yet, return EBUSY to
-		 * userspace hoping that on ioctl recall we get better
-		 * luck
-		 */
-		dev_err(rdev->dev, "no free indirect buffer !\n");
+	idx = rdev->ib_pool.head_id;
+retry:
+	if (cretry > 5) {
+		dev_err(rdev->dev, "failed to get an ib after 5 retry\n");
 		mutex_unlock(&rdev->ib_pool.mutex);
 		radeon_fence_unref(&fence);
-		return -EBUSY;
+		return -ENOMEM;
 	}
-	rdev->ib_pool.head_id = (nib->idx + 1) & (RADEON_IB_POOL_SIZE - 1);
-	nib->free = false;
-	if (nib->fence) {
-		mutex_unlock(&rdev->ib_pool.mutex);
-		r = radeon_fence_wait(nib->fence, false);
-		if (r) {
-			dev_err(rdev->dev, "error waiting fence of IB(%u:0x%016lX:%u)\n",
-				nib->idx, (unsigned long)nib->gpu_addr, nib->length_dw);
-			mutex_lock(&rdev->ib_pool.mutex);
-			nib->free = true;
-			mutex_unlock(&rdev->ib_pool.mutex);
-			radeon_fence_unref(&fence);
-			return r;
+	cretry++;
+	for (i = 0; i < RADEON_IB_POOL_SIZE; i++) {
+		radeon_ib_try_free(rdev, &rdev->ib_pool.ibs[idx]);
+		if (rdev->ib_pool.ibs[idx].fence == NULL) {
+			r = radeon_sa_bo_new(rdev, &rdev->ib_pool.sa_manager,
+					     &rdev->ib_pool.ibs[idx].sa_bo,
+					     64*1024, 64);
+			if (!r) {
+				*ib = &rdev->ib_pool.ibs[idx];
+				(*ib)->ptr = rdev->ib_pool.sa_manager.cpu_ptr;
+				(*ib)->ptr += ((*ib)->sa_bo.offset >> 2);
+				(*ib)->gpu_addr = rdev->ib_pool.sa_manager.gpu_addr;
+				(*ib)->gpu_addr += (*ib)->sa_bo.offset;
+				(*ib)->fence = fence;
+				/* ib are most likely to be allocated in a ring fashion
+				 * thus rdev->ib_pool.head_id should be the id of the
+				 * oldest ib
+				 */
+				rdev->ib_pool.head_id = (1 + idx);
+				rdev->ib_pool.head_id &= (RADEON_IB_POOL_SIZE - 1);
+				mutex_unlock(&rdev->ib_pool.mutex);
+				return 0;
+			}
 		}
-		mutex_lock(&rdev->ib_pool.mutex);
+		idx = (idx + 1) & (RADEON_IB_POOL_SIZE - 1);
 	}
-	radeon_fence_unref(&nib->fence);
-	nib->fence = fence;
-	nib->length_dw = 0;
+	/* this should be rare event, ie all ib scheduled none signaled yet.
+	 */
+	for (i = 0; i < RADEON_IB_POOL_SIZE; i++) {
+		if (rdev->ib_pool.ibs[idx].fence) {
+			r = radeon_fence_wait(rdev->ib_pool.ibs[idx].fence, false);
+			if (!r) {
+				goto retry;
+			}
+			/* an error happened */
+			break;
+		}
+		idx = (idx + 1) & (RADEON_IB_POOL_SIZE - 1);
+	}
 	mutex_unlock(&rdev->ib_pool.mutex);
-	*ib = nib;
-	return 0;
+	radeon_fence_unref(&fence);
+	return r;
 }
 
 void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib **ib)
@@ -169,247 +168,255 @@
 	if (tmp == NULL) {
 		return;
 	}
-	if (!tmp->fence->emited)
-		radeon_fence_unref(&tmp->fence);
 	mutex_lock(&rdev->ib_pool.mutex);
-	tmp->free = true;
+	if (tmp->fence && !tmp->fence->emitted) {
+		radeon_sa_bo_free(rdev, &tmp->sa_bo);
+		radeon_fence_unref(&tmp->fence);
+	}
 	mutex_unlock(&rdev->ib_pool.mutex);
 }
 
 int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib)
 {
+	struct radeon_ring *ring = &rdev->ring[ib->fence->ring];
 	int r = 0;
 
-	if (!ib->length_dw || !rdev->cp.ready) {
+	if (!ib->length_dw || !ring->ready) {
 		/* TODO: Nothings in the ib we should report. */
 		DRM_ERROR("radeon: couldn't schedule IB(%u).\n", ib->idx);
 		return -EINVAL;
 	}
 
 	/* 64 dwords should be enough for fence too */
-	r = radeon_ring_lock(rdev, 64);
+	r = radeon_ring_lock(rdev, ring, 64);
 	if (r) {
 		DRM_ERROR("radeon: scheduling IB failed (%d).\n", r);
 		return r;
 	}
-	radeon_ring_ib_execute(rdev, ib);
+	radeon_ring_ib_execute(rdev, ib->fence->ring, ib);
 	radeon_fence_emit(rdev, ib->fence);
-	mutex_lock(&rdev->ib_pool.mutex);
-	/* once scheduled IB is considered free and protected by the fence */
-	ib->free = true;
-	mutex_unlock(&rdev->ib_pool.mutex);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_unlock_commit(rdev, ring);
 	return 0;
 }
 
 int radeon_ib_pool_init(struct radeon_device *rdev)
 {
-	void *ptr;
-	uint64_t gpu_addr;
-	int i;
-	int r = 0;
+	int i, r;
 
-	if (rdev->ib_pool.robj)
+	mutex_lock(&rdev->ib_pool.mutex);
+	if (rdev->ib_pool.ready) {
+		mutex_unlock(&rdev->ib_pool.mutex);
 		return 0;
-	INIT_LIST_HEAD(&rdev->ib_pool.bogus_ib);
-	/* Allocate 1M object buffer */
-	r = radeon_bo_create(rdev, RADEON_IB_POOL_SIZE*64*1024,
-			     PAGE_SIZE, true, RADEON_GEM_DOMAIN_GTT,
-			     &rdev->ib_pool.robj);
-	if (r) {
-		DRM_ERROR("radeon: failed to ib pool (%d).\n", r);
-		return r;
 	}
-	r = radeon_bo_reserve(rdev->ib_pool.robj, false);
-	if (unlikely(r != 0))
-		return r;
-	r = radeon_bo_pin(rdev->ib_pool.robj, RADEON_GEM_DOMAIN_GTT, &gpu_addr);
-	if (r) {
-		radeon_bo_unreserve(rdev->ib_pool.robj);
-		DRM_ERROR("radeon: failed to pin ib pool (%d).\n", r);
-		return r;
-	}
-	r = radeon_bo_kmap(rdev->ib_pool.robj, &ptr);
-	radeon_bo_unreserve(rdev->ib_pool.robj);
-	if (r) {
-		DRM_ERROR("radeon: failed to map ib pool (%d).\n", r);
-		return r;
-	}
-	for (i = 0; i < RADEON_IB_POOL_SIZE; i++) {
-		unsigned offset;
 
-		offset = i * 64 * 1024;
-		rdev->ib_pool.ibs[i].gpu_addr = gpu_addr + offset;
-		rdev->ib_pool.ibs[i].ptr = ptr + offset;
+	r = radeon_sa_bo_manager_init(rdev, &rdev->ib_pool.sa_manager,
+				      RADEON_IB_POOL_SIZE*64*1024,
+				      RADEON_GEM_DOMAIN_GTT);
+	if (r) {
+		mutex_unlock(&rdev->ib_pool.mutex);
+		return r;
+	}
+
+	for (i = 0; i < RADEON_IB_POOL_SIZE; i++) {
+		rdev->ib_pool.ibs[i].fence = NULL;
 		rdev->ib_pool.ibs[i].idx = i;
 		rdev->ib_pool.ibs[i].length_dw = 0;
-		rdev->ib_pool.ibs[i].free = true;
+		INIT_LIST_HEAD(&rdev->ib_pool.ibs[i].sa_bo.list);
 	}
 	rdev->ib_pool.head_id = 0;
 	rdev->ib_pool.ready = true;
 	DRM_INFO("radeon: ib pool ready.\n");
+
 	if (radeon_debugfs_ib_init(rdev)) {
 		DRM_ERROR("Failed to register debugfs file for IB !\n");
 	}
-	return r;
+	if (radeon_debugfs_ring_init(rdev)) {
+		DRM_ERROR("Failed to register debugfs file for rings !\n");
+	}
+	mutex_unlock(&rdev->ib_pool.mutex);
+	return 0;
 }
 
 void radeon_ib_pool_fini(struct radeon_device *rdev)
 {
-	int r;
-	struct radeon_bo *robj;
+	unsigned i;
 
-	if (!rdev->ib_pool.ready) {
-		return;
-	}
 	mutex_lock(&rdev->ib_pool.mutex);
-	radeon_ib_bogus_cleanup(rdev);
-	robj = rdev->ib_pool.robj;
-	rdev->ib_pool.robj = NULL;
-	mutex_unlock(&rdev->ib_pool.mutex);
-
-	if (robj) {
-		r = radeon_bo_reserve(robj, false);
-		if (likely(r == 0)) {
-			radeon_bo_kunmap(robj);
-			radeon_bo_unpin(robj);
-			radeon_bo_unreserve(robj);
+	if (rdev->ib_pool.ready) {
+		for (i = 0; i < RADEON_IB_POOL_SIZE; i++) {
+			radeon_sa_bo_free(rdev, &rdev->ib_pool.ibs[i].sa_bo);
+			radeon_fence_unref(&rdev->ib_pool.ibs[i].fence);
 		}
-		radeon_bo_unref(&robj);
+		radeon_sa_bo_manager_fini(rdev, &rdev->ib_pool.sa_manager);
+		rdev->ib_pool.ready = false;
 	}
+	mutex_unlock(&rdev->ib_pool.mutex);
 }
 
+int radeon_ib_pool_start(struct radeon_device *rdev)
+{
+	return radeon_sa_bo_manager_start(rdev, &rdev->ib_pool.sa_manager);
+}
+
+int radeon_ib_pool_suspend(struct radeon_device *rdev)
+{
+	return radeon_sa_bo_manager_suspend(rdev, &rdev->ib_pool.sa_manager);
+}
 
 /*
  * Ring.
  */
-void radeon_ring_free_size(struct radeon_device *rdev)
+int radeon_ring_index(struct radeon_device *rdev, struct radeon_ring *ring)
 {
-	if (rdev->wb.enabled)
-		rdev->cp.rptr = le32_to_cpu(rdev->wb.wb[RADEON_WB_CP_RPTR_OFFSET/4]);
-	else {
-		if (rdev->family >= CHIP_R600)
-			rdev->cp.rptr = RREG32(R600_CP_RB_RPTR);
-		else
-			rdev->cp.rptr = RREG32(RADEON_CP_RB_RPTR);
+	/* r1xx-r5xx only has CP ring */
+	if (rdev->family < CHIP_R600)
+		return RADEON_RING_TYPE_GFX_INDEX;
+
+	if (rdev->family >= CHIP_CAYMAN) {
+		if (ring == &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX])
+			return CAYMAN_RING_TYPE_CP1_INDEX;
+		else if (ring == &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX])
+			return CAYMAN_RING_TYPE_CP2_INDEX;
 	}
+	return RADEON_RING_TYPE_GFX_INDEX;
+}
+
+void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *ring)
+{
+	u32 rptr;
+
+	if (rdev->wb.enabled)
+		rptr = le32_to_cpu(rdev->wb.wb[ring->rptr_offs/4]);
+	else
+		rptr = RREG32(ring->rptr_reg);
+	ring->rptr = (rptr & ring->ptr_reg_mask) >> ring->ptr_reg_shift;
 	/* This works because ring_size is a power of 2 */
-	rdev->cp.ring_free_dw = (rdev->cp.rptr + (rdev->cp.ring_size / 4));
-	rdev->cp.ring_free_dw -= rdev->cp.wptr;
-	rdev->cp.ring_free_dw &= rdev->cp.ptr_mask;
-	if (!rdev->cp.ring_free_dw) {
-		rdev->cp.ring_free_dw = rdev->cp.ring_size / 4;
+	ring->ring_free_dw = (ring->rptr + (ring->ring_size / 4));
+	ring->ring_free_dw -= ring->wptr;
+	ring->ring_free_dw &= ring->ptr_mask;
+	if (!ring->ring_free_dw) {
+		ring->ring_free_dw = ring->ring_size / 4;
 	}
 }
 
-int radeon_ring_alloc(struct radeon_device *rdev, unsigned ndw)
+
+int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *ring, unsigned ndw)
 {
 	int r;
 
 	/* Align requested size with padding so unlock_commit can
 	 * pad safely */
-	ndw = (ndw + rdev->cp.align_mask) & ~rdev->cp.align_mask;
-	while (ndw > (rdev->cp.ring_free_dw - 1)) {
-		radeon_ring_free_size(rdev);
-		if (ndw < rdev->cp.ring_free_dw) {
+	ndw = (ndw + ring->align_mask) & ~ring->align_mask;
+	while (ndw > (ring->ring_free_dw - 1)) {
+		radeon_ring_free_size(rdev, ring);
+		if (ndw < ring->ring_free_dw) {
 			break;
 		}
-		r = radeon_fence_wait_next(rdev);
+		r = radeon_fence_wait_next(rdev, radeon_ring_index(rdev, ring));
 		if (r)
 			return r;
 	}
-	rdev->cp.count_dw = ndw;
-	rdev->cp.wptr_old = rdev->cp.wptr;
+	ring->count_dw = ndw;
+	ring->wptr_old = ring->wptr;
 	return 0;
 }
 
-int radeon_ring_lock(struct radeon_device *rdev, unsigned ndw)
+int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *ring, unsigned ndw)
 {
 	int r;
 
-	mutex_lock(&rdev->cp.mutex);
-	r = radeon_ring_alloc(rdev, ndw);
+	mutex_lock(&ring->mutex);
+	r = radeon_ring_alloc(rdev, ring, ndw);
 	if (r) {
-		mutex_unlock(&rdev->cp.mutex);
+		mutex_unlock(&ring->mutex);
 		return r;
 	}
 	return 0;
 }
 
-void radeon_ring_commit(struct radeon_device *rdev)
+void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	unsigned count_dw_pad;
 	unsigned i;
 
 	/* We pad to match fetch size */
-	count_dw_pad = (rdev->cp.align_mask + 1) -
-		       (rdev->cp.wptr & rdev->cp.align_mask);
+	count_dw_pad = (ring->align_mask + 1) -
+		       (ring->wptr & ring->align_mask);
 	for (i = 0; i < count_dw_pad; i++) {
-		radeon_ring_write(rdev, 2 << 30);
+		radeon_ring_write(ring, ring->nop);
 	}
 	DRM_MEMORYBARRIER();
-	radeon_cp_commit(rdev);
+	WREG32(ring->wptr_reg, (ring->wptr << ring->ptr_reg_shift) & ring->ptr_reg_mask);
+	(void)RREG32(ring->wptr_reg);
 }
 
-void radeon_ring_unlock_commit(struct radeon_device *rdev)
+void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *ring)
 {
-	radeon_ring_commit(rdev);
-	mutex_unlock(&rdev->cp.mutex);
+	radeon_ring_commit(rdev, ring);
+	mutex_unlock(&ring->mutex);
 }
 
-void radeon_ring_unlock_undo(struct radeon_device *rdev)
+void radeon_ring_unlock_undo(struct radeon_device *rdev, struct radeon_ring *ring)
 {
-	rdev->cp.wptr = rdev->cp.wptr_old;
-	mutex_unlock(&rdev->cp.mutex);
+	ring->wptr = ring->wptr_old;
+	mutex_unlock(&ring->mutex);
 }
 
-int radeon_ring_init(struct radeon_device *rdev, unsigned ring_size)
+int radeon_ring_init(struct radeon_device *rdev, struct radeon_ring *ring, unsigned ring_size,
+		     unsigned rptr_offs, unsigned rptr_reg, unsigned wptr_reg,
+		     u32 ptr_reg_shift, u32 ptr_reg_mask, u32 nop)
 {
 	int r;
 
-	rdev->cp.ring_size = ring_size;
+	ring->ring_size = ring_size;
+	ring->rptr_offs = rptr_offs;
+	ring->rptr_reg = rptr_reg;
+	ring->wptr_reg = wptr_reg;
+	ring->ptr_reg_shift = ptr_reg_shift;
+	ring->ptr_reg_mask = ptr_reg_mask;
+	ring->nop = nop;
 	/* Allocate ring buffer */
-	if (rdev->cp.ring_obj == NULL) {
-		r = radeon_bo_create(rdev, rdev->cp.ring_size, PAGE_SIZE, true,
+	if (ring->ring_obj == NULL) {
+		r = radeon_bo_create(rdev, ring->ring_size, PAGE_SIZE, true,
 					RADEON_GEM_DOMAIN_GTT,
-					&rdev->cp.ring_obj);
+					&ring->ring_obj);
 		if (r) {
 			dev_err(rdev->dev, "(%d) ring create failed\n", r);
 			return r;
 		}
-		r = radeon_bo_reserve(rdev->cp.ring_obj, false);
+		r = radeon_bo_reserve(ring->ring_obj, false);
 		if (unlikely(r != 0))
 			return r;
-		r = radeon_bo_pin(rdev->cp.ring_obj, RADEON_GEM_DOMAIN_GTT,
-					&rdev->cp.gpu_addr);
+		r = radeon_bo_pin(ring->ring_obj, RADEON_GEM_DOMAIN_GTT,
+					&ring->gpu_addr);
 		if (r) {
-			radeon_bo_unreserve(rdev->cp.ring_obj);
+			radeon_bo_unreserve(ring->ring_obj);
 			dev_err(rdev->dev, "(%d) ring pin failed\n", r);
 			return r;
 		}
-		r = radeon_bo_kmap(rdev->cp.ring_obj,
-				       (void **)&rdev->cp.ring);
-		radeon_bo_unreserve(rdev->cp.ring_obj);
+		r = radeon_bo_kmap(ring->ring_obj,
+				       (void **)&ring->ring);
+		radeon_bo_unreserve(ring->ring_obj);
 		if (r) {
 			dev_err(rdev->dev, "(%d) ring map failed\n", r);
 			return r;
 		}
 	}
-	rdev->cp.ptr_mask = (rdev->cp.ring_size / 4) - 1;
-	rdev->cp.ring_free_dw = rdev->cp.ring_size / 4;
+	ring->ptr_mask = (ring->ring_size / 4) - 1;
+	ring->ring_free_dw = ring->ring_size / 4;
 	return 0;
 }
 
-void radeon_ring_fini(struct radeon_device *rdev)
+void radeon_ring_fini(struct radeon_device *rdev, struct radeon_ring *ring)
 {
 	int r;
 	struct radeon_bo *ring_obj;
 
-	mutex_lock(&rdev->cp.mutex);
-	ring_obj = rdev->cp.ring_obj;
-	rdev->cp.ring = NULL;
-	rdev->cp.ring_obj = NULL;
-	mutex_unlock(&rdev->cp.mutex);
+	mutex_lock(&ring->mutex);
+	ring_obj = ring->ring_obj;
+	ring->ring = NULL;
+	ring->ring_obj = NULL;
+	mutex_unlock(&ring->mutex);
 
 	if (ring_obj) {
 		r = radeon_bo_reserve(ring_obj, false);
@@ -422,11 +429,46 @@
 	}
 }
 
-
 /*
  * Debugfs info
  */
 #if defined(CONFIG_DEBUG_FS)
+
+static int radeon_debugfs_ring_info(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	int ridx = *(int*)node->info_ent->data;
+	struct radeon_ring *ring = &rdev->ring[ridx];
+	unsigned count, i, j;
+
+	radeon_ring_free_size(rdev, ring);
+	count = (ring->ring_size / 4) - ring->ring_free_dw;
+	seq_printf(m, "wptr(0x%04x): 0x%08x\n", ring->wptr_reg, RREG32(ring->wptr_reg));
+	seq_printf(m, "rptr(0x%04x): 0x%08x\n", ring->rptr_reg, RREG32(ring->rptr_reg));
+	seq_printf(m, "driver's copy of the wptr: 0x%08x\n", ring->wptr);
+	seq_printf(m, "driver's copy of the rptr: 0x%08x\n", ring->rptr);
+	seq_printf(m, "%u free dwords in ring\n", ring->ring_free_dw);
+	seq_printf(m, "%u dwords in ring\n", count);
+	i = ring->rptr;
+	for (j = 0; j <= count; j++) {
+		seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]);
+		i = (i + 1) & ring->ptr_mask;
+	}
+	return 0;
+}
+
+static int radeon_ring_type_gfx_index = RADEON_RING_TYPE_GFX_INDEX;
+static int cayman_ring_type_cp1_index = CAYMAN_RING_TYPE_CP1_INDEX;
+static int cayman_ring_type_cp2_index = CAYMAN_RING_TYPE_CP2_INDEX;
+
+static struct drm_info_list radeon_debugfs_ring_info_list[] = {
+	{"radeon_ring_gfx", radeon_debugfs_ring_info, 0, &radeon_ring_type_gfx_index},
+	{"radeon_ring_cp1", radeon_debugfs_ring_info, 0, &cayman_ring_type_cp1_index},
+	{"radeon_ring_cp2", radeon_debugfs_ring_info, 0, &cayman_ring_type_cp2_index},
+};
+
 static int radeon_debugfs_ib_info(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
@@ -445,49 +487,25 @@
 	return 0;
 }
 
-static int radeon_debugfs_ib_bogus_info(struct seq_file *m, void *data)
-{
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
-	struct radeon_device *rdev = node->info_ent->data;
-	struct radeon_ib *ib;
-	unsigned i;
-
-	mutex_lock(&rdev->ib_pool.mutex);
-	if (list_empty(&rdev->ib_pool.bogus_ib)) {
-		mutex_unlock(&rdev->ib_pool.mutex);
-		seq_printf(m, "no bogus IB recorded\n");
-		return 0;
-	}
-	ib = list_first_entry(&rdev->ib_pool.bogus_ib, struct radeon_ib, list);
-	list_del_init(&ib->list);
-	mutex_unlock(&rdev->ib_pool.mutex);
-	seq_printf(m, "IB size %05u dwords\n", ib->length_dw);
-	for (i = 0; i < ib->length_dw; i++) {
-		seq_printf(m, "[%05u]=0x%08X\n", i, ib->ptr[i]);
-	}
-	vfree(ib->ptr);
-	kfree(ib);
-	return 0;
-}
-
 static struct drm_info_list radeon_debugfs_ib_list[RADEON_IB_POOL_SIZE];
 static char radeon_debugfs_ib_names[RADEON_IB_POOL_SIZE][32];
-
-static struct drm_info_list radeon_debugfs_ib_bogus_info_list[] = {
-	{"radeon_ib_bogus", radeon_debugfs_ib_bogus_info, 0, NULL},
-};
 #endif
 
+int radeon_debugfs_ring_init(struct radeon_device *rdev)
+{
+#if defined(CONFIG_DEBUG_FS)
+	return radeon_debugfs_add_files(rdev, radeon_debugfs_ring_info_list,
+					ARRAY_SIZE(radeon_debugfs_ring_info_list));
+#else
+	return 0;
+#endif
+}
+
 int radeon_debugfs_ib_init(struct radeon_device *rdev)
 {
 #if defined(CONFIG_DEBUG_FS)
 	unsigned i;
-	int r;
 
-	radeon_debugfs_ib_bogus_info_list[0].data = rdev;
-	r = radeon_debugfs_add_files(rdev, radeon_debugfs_ib_bogus_info_list, 1);
-	if (r)
-		return r;
 	for (i = 0; i < RADEON_IB_POOL_SIZE; i++) {
 		sprintf(radeon_debugfs_ib_names[i], "radeon_ib_%04u", i);
 		radeon_debugfs_ib_list[i].name = radeon_debugfs_ib_names[i];
diff --git a/drivers/gpu/drm/radeon/radeon_sa.c b/drivers/gpu/drm/radeon/radeon_sa.c
new file mode 100644
index 0000000..4cce47e
--- /dev/null
+++ b/drivers/gpu/drm/radeon/radeon_sa.c
@@ -0,0 +1,189 @@
+/*
+ * Copyright 2011 Red Hat Inc.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the
+ * "Software"), to deal in the Software without restriction, including
+ * without limitation the rights to use, copy, modify, merge, publish,
+ * distribute, sub license, and/or sell copies of the Software, and to
+ * permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
+ * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+ * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+ * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * The above copyright notice and this permission notice (including the
+ * next paragraph) shall be included in all copies or substantial portions
+ * of the Software.
+ *
+ */
+/*
+ * Authors:
+ *    Jerome Glisse <glisse@freedesktop.org>
+ */
+#include "drmP.h"
+#include "drm.h"
+#include "radeon.h"
+
+int radeon_sa_bo_manager_init(struct radeon_device *rdev,
+			      struct radeon_sa_manager *sa_manager,
+			      unsigned size, u32 domain)
+{
+	int r;
+
+	sa_manager->bo = NULL;
+	sa_manager->size = size;
+	sa_manager->domain = domain;
+	INIT_LIST_HEAD(&sa_manager->sa_bo);
+
+	r = radeon_bo_create(rdev, size, RADEON_GPU_PAGE_SIZE, true,
+			     RADEON_GEM_DOMAIN_CPU, &sa_manager->bo);
+	if (r) {
+		dev_err(rdev->dev, "(%d) failed to allocate bo for manager\n", r);
+		return r;
+	}
+
+	return r;
+}
+
+void radeon_sa_bo_manager_fini(struct radeon_device *rdev,
+			       struct radeon_sa_manager *sa_manager)
+{
+	struct radeon_sa_bo *sa_bo, *tmp;
+
+	if (!list_empty(&sa_manager->sa_bo)) {
+		dev_err(rdev->dev, "sa_manager is not empty, clearing anyway\n");
+	}
+	list_for_each_entry_safe(sa_bo, tmp, &sa_manager->sa_bo, list) {
+		list_del_init(&sa_bo->list);
+	}
+	radeon_bo_unref(&sa_manager->bo);
+	sa_manager->size = 0;
+}
+
+int radeon_sa_bo_manager_start(struct radeon_device *rdev,
+			       struct radeon_sa_manager *sa_manager)
+{
+	int r;
+
+	if (sa_manager->bo == NULL) {
+		dev_err(rdev->dev, "no bo for sa manager\n");
+		return -EINVAL;
+	}
+
+	/* map the buffer */
+	r = radeon_bo_reserve(sa_manager->bo, false);
+	if (r) {
+		dev_err(rdev->dev, "(%d) failed to reserve manager bo\n", r);
+		return r;
+	}
+	r = radeon_bo_pin(sa_manager->bo, sa_manager->domain, &sa_manager->gpu_addr);
+	if (r) {
+		radeon_bo_unreserve(sa_manager->bo);
+		dev_err(rdev->dev, "(%d) failed to pin manager bo\n", r);
+		return r;
+	}
+	r = radeon_bo_kmap(sa_manager->bo, &sa_manager->cpu_ptr);
+	radeon_bo_unreserve(sa_manager->bo);
+	return r;
+}
+
+int radeon_sa_bo_manager_suspend(struct radeon_device *rdev,
+				 struct radeon_sa_manager *sa_manager)
+{
+	int r;
+
+	if (sa_manager->bo == NULL) {
+		dev_err(rdev->dev, "no bo for sa manager\n");
+		return -EINVAL;
+	}
+
+	r = radeon_bo_reserve(sa_manager->bo, false);
+	if (!r) {
+		radeon_bo_kunmap(sa_manager->bo);
+		radeon_bo_unpin(sa_manager->bo);
+		radeon_bo_unreserve(sa_manager->bo);
+	}
+	return r;
+}
+
+/*
+ * Principe is simple, we keep a list of sub allocation in offset
+ * order (first entry has offset == 0, last entry has the highest
+ * offset).
+ *
+ * When allocating new object we first check if there is room at
+ * the end total_size - (last_object_offset + last_object_size) >=
+ * alloc_size. If so we allocate new object there.
+ *
+ * When there is not enough room at the end, we start waiting for
+ * each sub object until we reach object_offset+object_size >=
+ * alloc_size, this object then become the sub object we return.
+ *
+ * Alignment can't be bigger than page size
+ */
+int radeon_sa_bo_new(struct radeon_device *rdev,
+		     struct radeon_sa_manager *sa_manager,
+		     struct radeon_sa_bo *sa_bo,
+		     unsigned size, unsigned align)
+{
+	struct radeon_sa_bo *tmp;
+	struct list_head *head;
+	unsigned offset = 0, wasted = 0;
+
+	BUG_ON(align > RADEON_GPU_PAGE_SIZE);
+	BUG_ON(size > sa_manager->size);
+
+	/* no one ? */
+	head = sa_manager->sa_bo.prev;
+	if (list_empty(&sa_manager->sa_bo)) {
+		goto out;
+	}
+
+	/* look for a hole big enough */
+	offset = 0;
+	list_for_each_entry(tmp, &sa_manager->sa_bo, list) {
+		/* room before this object ? */
+		if ((tmp->offset - offset) >= size) {
+			head = tmp->list.prev;
+			goto out;
+		}
+		offset = tmp->offset + tmp->size;
+		wasted = offset % align;
+		if (wasted) {
+			wasted = align - wasted;
+		}
+		offset += wasted;
+	}
+	/* room at the end ? */
+	head = sa_manager->sa_bo.prev;
+	tmp = list_entry(head, struct radeon_sa_bo, list);
+	offset = tmp->offset + tmp->size;
+	wasted = offset % align;
+	if (wasted) {
+		wasted = align - wasted;
+	}
+	offset += wasted;
+	if ((sa_manager->size - offset) < size) {
+		/* failed to find somethings big enough */
+		return -ENOMEM;
+	}
+
+out:
+	sa_bo->manager = sa_manager;
+	sa_bo->offset = offset;
+	sa_bo->size = size;
+	list_add(&sa_bo->list, head);
+	return 0;
+}
+
+void radeon_sa_bo_free(struct radeon_device *rdev, struct radeon_sa_bo *sa_bo)
+{
+	list_del_init(&sa_bo->list);
+}
diff --git a/drivers/gpu/drm/radeon/radeon_semaphore.c b/drivers/gpu/drm/radeon/radeon_semaphore.c
new file mode 100644
index 0000000..f8cf044
--- /dev/null
+++ b/drivers/gpu/drm/radeon/radeon_semaphore.c
@@ -0,0 +1,161 @@
+/*
+ * Copyright 2011 Christian König.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the
+ * "Software"), to deal in the Software without restriction, including
+ * without limitation the rights to use, copy, modify, merge, publish,
+ * distribute, sub license, and/or sell copies of the Software, and to
+ * permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
+ * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+ * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+ * USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * The above copyright notice and this permission notice (including the
+ * next paragraph) shall be included in all copies or substantial portions
+ * of the Software.
+ *
+ */
+/*
+ * Authors:
+ *    Christian König <deathsimple@vodafone.de>
+ */
+#include "drmP.h"
+#include "drm.h"
+#include "radeon.h"
+
+static int allocate_semaphores(struct radeon_device *rdev)
+{
+	const unsigned long bo_size = PAGE_SIZE * 4;
+
+	struct radeon_bo *bo;
+	struct list_head new_entrys;
+	unsigned long irq_flags;
+	uint64_t gpu_addr;
+	void *map;
+	int i, r;
+
+	r = radeon_bo_create(rdev, bo_size, RADEON_GPU_PAGE_SIZE, true,
+			     RADEON_GEM_DOMAIN_GTT, &bo);
+	if (r) {
+		dev_err(rdev->dev, "(%d) failed to allocate semaphore bo\n", r);
+		return r;
+	}
+
+	r = radeon_bo_reserve(bo, false);
+	if (r) {
+		radeon_bo_unref(&bo);
+		dev_err(rdev->dev, "(%d) failed to reserve semaphore bo\n", r);
+		return r;
+	}
+
+	r = radeon_bo_kmap(bo, &map);
+	if (r) {
+		radeon_bo_unreserve(bo);
+		radeon_bo_unref(&bo);
+		dev_err(rdev->dev, "(%d) semaphore map failed\n", r);
+		return r;
+	}
+	memset(map, 0, bo_size);
+	radeon_bo_kunmap(bo);
+
+	r = radeon_bo_pin(bo, RADEON_GEM_DOMAIN_VRAM, &gpu_addr);
+	if (r) {
+		radeon_bo_unreserve(bo);
+		radeon_bo_unref(&bo);
+		dev_err(rdev->dev, "(%d) semaphore pin failed\n", r);
+		return r;
+	}
+
+	INIT_LIST_HEAD(&new_entrys);
+	for (i = 0; i < bo_size/8; ++i) {
+		struct radeon_semaphore *sem = kmalloc(sizeof(struct radeon_semaphore), GFP_KERNEL);
+		ttm_bo_reference(&bo->tbo);
+		sem->robj = bo;
+		sem->gpu_addr = gpu_addr;
+		gpu_addr += 8;
+		list_add_tail(&sem->list, &new_entrys);
+	}
+
+	radeon_bo_unreserve(bo);
+	radeon_bo_unref(&bo);
+
+	write_lock_irqsave(&rdev->semaphore_drv.lock, irq_flags);
+	list_splice_tail(&new_entrys, &rdev->semaphore_drv.free);
+	write_unlock_irqrestore(&rdev->semaphore_drv.lock, irq_flags);
+
+	DRM_INFO("%d new semaphores allocated\n", (int)(bo_size/8));
+
+	return 0;
+}
+
+int radeon_semaphore_create(struct radeon_device *rdev,
+			    struct radeon_semaphore **semaphore)
+{
+	unsigned long irq_flags;
+
+	write_lock_irqsave(&rdev->semaphore_drv.lock, irq_flags);
+	if (list_empty(&rdev->semaphore_drv.free)) {
+		int r;
+		write_unlock_irqrestore(&rdev->semaphore_drv.lock, irq_flags);
+		r = allocate_semaphores(rdev);
+		if (r)
+			return r;
+		write_lock_irqsave(&rdev->semaphore_drv.lock, irq_flags);
+	}
+
+	*semaphore = list_first_entry(&rdev->semaphore_drv.free, struct radeon_semaphore, list);
+	list_del(&(*semaphore)->list);
+
+	write_unlock_irqrestore(&rdev->semaphore_drv.lock, irq_flags);
+	return 0;
+}
+
+void radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring,
+			          struct radeon_semaphore *semaphore)
+{
+	radeon_semaphore_ring_emit(rdev, ring, &rdev->ring[ring], semaphore, false);
+}
+
+void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring,
+			        struct radeon_semaphore *semaphore)
+{
+	radeon_semaphore_ring_emit(rdev, ring, &rdev->ring[ring], semaphore, true);
+}
+
+void radeon_semaphore_free(struct radeon_device *rdev,
+			  struct radeon_semaphore *semaphore)
+{
+	unsigned long irq_flags;
+
+	write_lock_irqsave(&rdev->semaphore_drv.lock, irq_flags);
+	list_add_tail(&semaphore->list, &rdev->semaphore_drv.free);
+	write_unlock_irqrestore(&rdev->semaphore_drv.lock, irq_flags);
+}
+
+void radeon_semaphore_driver_fini(struct radeon_device *rdev)
+{
+	struct radeon_semaphore *i, *n;
+	struct list_head entrys;
+	unsigned long irq_flags;
+
+	INIT_LIST_HEAD(&entrys);
+	write_lock_irqsave(&rdev->semaphore_drv.lock, irq_flags);
+	if (!list_empty(&rdev->semaphore_drv.free)) {
+		list_splice(&rdev->semaphore_drv.free, &entrys);
+	}
+	INIT_LIST_HEAD(&rdev->semaphore_drv.free);
+	write_unlock_irqrestore(&rdev->semaphore_drv.lock, irq_flags);
+
+	list_for_each_entry_safe(i, n, &entrys, list) {
+		radeon_bo_unref(&i->robj);
+		kfree(i);
+	}
+}
diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c
index 602fa35..210d99f 100644
--- a/drivers/gpu/drm/radeon/radeon_test.c
+++ b/drivers/gpu/drm/radeon/radeon_test.c
@@ -42,7 +42,9 @@
 	/* Number of tests =
 	 * (Total GTT - IB pool - writeback page - ring buffers) / test size
 	 */
-	n = rdev->mc.gtt_size - RADEON_IB_POOL_SIZE*64*1024 - rdev->cp.ring_size;
+	n = rdev->mc.gtt_size - RADEON_IB_POOL_SIZE*64*1024;
+	for (i = 0; i < RADEON_NUM_RINGS; ++i)
+		n -= rdev->ring[i].ring_size;
 	if (rdev->wb.wb_obj)
 		n -= RADEON_GPU_PAGE_SIZE;
 	if (rdev->ih.ring_obj)
@@ -104,7 +106,7 @@
 
 		radeon_bo_kunmap(gtt_obj[i]);
 
-		r = radeon_fence_create(rdev, &fence);
+		r = radeon_fence_create(rdev, &fence, RADEON_RING_TYPE_GFX_INDEX);
 		if (r) {
 			DRM_ERROR("Failed to create GTT->VRAM fence %d\n", i);
 			goto out_cleanup;
@@ -153,7 +155,7 @@
 
 		radeon_bo_kunmap(vram_obj);
 
-		r = radeon_fence_create(rdev, &fence);
+		r = radeon_fence_create(rdev, &fence, RADEON_RING_TYPE_GFX_INDEX);
 		if (r) {
 			DRM_ERROR("Failed to create VRAM->GTT fence %d\n", i);
 			goto out_cleanup;
@@ -232,3 +234,262 @@
 		printk(KERN_WARNING "Error while testing BO move.\n");
 	}
 }
+
+void radeon_test_ring_sync(struct radeon_device *rdev,
+			   struct radeon_ring *ringA,
+			   struct radeon_ring *ringB)
+{
+	struct radeon_fence *fence1 = NULL, *fence2 = NULL;
+	struct radeon_semaphore *semaphore = NULL;
+	int ridxA = radeon_ring_index(rdev, ringA);
+	int ridxB = radeon_ring_index(rdev, ringB);
+	int r;
+
+	r = radeon_fence_create(rdev, &fence1, ridxA);
+	if (r) {
+		DRM_ERROR("Failed to create sync fence 1\n");
+		goto out_cleanup;
+	}
+	r = radeon_fence_create(rdev, &fence2, ridxA);
+	if (r) {
+		DRM_ERROR("Failed to create sync fence 2\n");
+		goto out_cleanup;
+	}
+
+	r = radeon_semaphore_create(rdev, &semaphore);
+	if (r) {
+		DRM_ERROR("Failed to create semaphore\n");
+		goto out_cleanup;
+	}
+
+	r = radeon_ring_lock(rdev, ringA, 64);
+	if (r) {
+		DRM_ERROR("Failed to lock ring A %d\n", ridxA);
+		goto out_cleanup;
+	}
+	radeon_semaphore_emit_wait(rdev, ridxA, semaphore);
+	radeon_fence_emit(rdev, fence1);
+	radeon_semaphore_emit_wait(rdev, ridxA, semaphore);
+	radeon_fence_emit(rdev, fence2);
+	radeon_ring_unlock_commit(rdev, ringA);
+
+	mdelay(1000);
+
+	if (radeon_fence_signaled(fence1)) {
+		DRM_ERROR("Fence 1 signaled without waiting for semaphore.\n");
+		goto out_cleanup;
+	}
+
+	r = radeon_ring_lock(rdev, ringB, 64);
+	if (r) {
+		DRM_ERROR("Failed to lock ring B %p\n", ringB);
+		goto out_cleanup;
+	}
+	radeon_semaphore_emit_signal(rdev, ridxB, semaphore);
+	radeon_ring_unlock_commit(rdev, ringB);
+
+	r = radeon_fence_wait(fence1, false);
+	if (r) {
+		DRM_ERROR("Failed to wait for sync fence 1\n");
+		goto out_cleanup;
+	}
+
+	mdelay(1000);
+
+	if (radeon_fence_signaled(fence2)) {
+		DRM_ERROR("Fence 2 signaled without waiting for semaphore.\n");
+		goto out_cleanup;
+	}
+
+	r = radeon_ring_lock(rdev, ringB, 64);
+	if (r) {
+		DRM_ERROR("Failed to lock ring B %p\n", ringB);
+		goto out_cleanup;
+	}
+	radeon_semaphore_emit_signal(rdev, ridxB, semaphore);
+	radeon_ring_unlock_commit(rdev, ringB);
+
+	r = radeon_fence_wait(fence2, false);
+	if (r) {
+		DRM_ERROR("Failed to wait for sync fence 1\n");
+		goto out_cleanup;
+	}
+
+out_cleanup:
+	if (semaphore)
+		radeon_semaphore_free(rdev, semaphore);
+
+	if (fence1)
+		radeon_fence_unref(&fence1);
+
+	if (fence2)
+		radeon_fence_unref(&fence2);
+
+	if (r)
+		printk(KERN_WARNING "Error while testing ring sync (%d).\n", r);
+}
+
+void radeon_test_ring_sync2(struct radeon_device *rdev,
+			    struct radeon_ring *ringA,
+			    struct radeon_ring *ringB,
+			    struct radeon_ring *ringC)
+{
+	struct radeon_fence *fenceA = NULL, *fenceB = NULL;
+	struct radeon_semaphore *semaphore = NULL;
+	int ridxA = radeon_ring_index(rdev, ringA);
+	int ridxB = radeon_ring_index(rdev, ringB);
+	int ridxC = radeon_ring_index(rdev, ringC);
+	bool sigA, sigB;
+	int i, r;
+
+	r = radeon_fence_create(rdev, &fenceA, ridxA);
+	if (r) {
+		DRM_ERROR("Failed to create sync fence 1\n");
+		goto out_cleanup;
+	}
+	r = radeon_fence_create(rdev, &fenceB, ridxB);
+	if (r) {
+		DRM_ERROR("Failed to create sync fence 2\n");
+		goto out_cleanup;
+	}
+
+	r = radeon_semaphore_create(rdev, &semaphore);
+	if (r) {
+		DRM_ERROR("Failed to create semaphore\n");
+		goto out_cleanup;
+	}
+
+	r = radeon_ring_lock(rdev, ringA, 64);
+	if (r) {
+		DRM_ERROR("Failed to lock ring A %d\n", ridxA);
+		goto out_cleanup;
+	}
+	radeon_semaphore_emit_wait(rdev, ridxA, semaphore);
+	radeon_fence_emit(rdev, fenceA);
+	radeon_ring_unlock_commit(rdev, ringA);
+
+	r = radeon_ring_lock(rdev, ringB, 64);
+	if (r) {
+		DRM_ERROR("Failed to lock ring B %d\n", ridxB);
+		goto out_cleanup;
+	}
+	radeon_semaphore_emit_wait(rdev, ridxB, semaphore);
+	radeon_fence_emit(rdev, fenceB);
+	radeon_ring_unlock_commit(rdev, ringB);
+
+	mdelay(1000);
+
+	if (radeon_fence_signaled(fenceA)) {
+		DRM_ERROR("Fence A signaled without waiting for semaphore.\n");
+		goto out_cleanup;
+	}
+	if (radeon_fence_signaled(fenceB)) {
+		DRM_ERROR("Fence A signaled without waiting for semaphore.\n");
+		goto out_cleanup;
+	}
+
+	r = radeon_ring_lock(rdev, ringC, 64);
+	if (r) {
+		DRM_ERROR("Failed to lock ring B %p\n", ringC);
+		goto out_cleanup;
+	}
+	radeon_semaphore_emit_signal(rdev, ridxC, semaphore);
+	radeon_ring_unlock_commit(rdev, ringC);
+
+	for (i = 0; i < 30; ++i) {
+		mdelay(100);
+		sigA = radeon_fence_signaled(fenceA);
+		sigB = radeon_fence_signaled(fenceB);
+		if (sigA || sigB)
+			break;
+	}
+
+	if (!sigA && !sigB) {
+		DRM_ERROR("Neither fence A nor B has been signaled\n");
+		goto out_cleanup;
+	} else if (sigA && sigB) {
+		DRM_ERROR("Both fence A and B has been signaled\n");
+		goto out_cleanup;
+	}
+
+	DRM_INFO("Fence %c was first signaled\n", sigA ? 'A' : 'B');
+
+	r = radeon_ring_lock(rdev, ringC, 64);
+	if (r) {
+		DRM_ERROR("Failed to lock ring B %p\n", ringC);
+		goto out_cleanup;
+	}
+	radeon_semaphore_emit_signal(rdev, ridxC, semaphore);
+	radeon_ring_unlock_commit(rdev, ringC);
+
+	mdelay(1000);
+
+	r = radeon_fence_wait(fenceA, false);
+	if (r) {
+		DRM_ERROR("Failed to wait for sync fence A\n");
+		goto out_cleanup;
+	}
+	r = radeon_fence_wait(fenceB, false);
+	if (r) {
+		DRM_ERROR("Failed to wait for sync fence B\n");
+		goto out_cleanup;
+	}
+
+out_cleanup:
+	if (semaphore)
+		radeon_semaphore_free(rdev, semaphore);
+
+	if (fenceA)
+		radeon_fence_unref(&fenceA);
+
+	if (fenceB)
+		radeon_fence_unref(&fenceB);
+
+	if (r)
+		printk(KERN_WARNING "Error while testing ring sync (%d).\n", r);
+}
+
+void radeon_test_syncing(struct radeon_device *rdev)
+{
+	int i, j, k;
+
+	for (i = 1; i < RADEON_NUM_RINGS; ++i) {
+		struct radeon_ring *ringA = &rdev->ring[i];
+		if (!ringA->ready)
+			continue;
+
+		for (j = 0; j < i; ++j) {
+			struct radeon_ring *ringB = &rdev->ring[j];
+			if (!ringB->ready)
+				continue;
+
+			DRM_INFO("Testing syncing between rings %d and %d...\n", i, j);
+			radeon_test_ring_sync(rdev, ringA, ringB);
+
+			DRM_INFO("Testing syncing between rings %d and %d...\n", j, i);
+			radeon_test_ring_sync(rdev, ringB, ringA);
+
+			for (k = 0; k < j; ++k) {
+				struct radeon_ring *ringC = &rdev->ring[k];
+
+				DRM_INFO("Testing syncing between rings %d, %d and %d...\n", i, j, k);
+				radeon_test_ring_sync2(rdev, ringA, ringB, ringC);
+
+				DRM_INFO("Testing syncing between rings %d, %d and %d...\n", i, k, j);
+				radeon_test_ring_sync2(rdev, ringA, ringC, ringB);
+
+				DRM_INFO("Testing syncing between rings %d, %d and %d...\n", j, i, k);
+				radeon_test_ring_sync2(rdev, ringB, ringA, ringC);
+
+				DRM_INFO("Testing syncing between rings %d, %d and %d...\n", j, k, i);
+				radeon_test_ring_sync2(rdev, ringB, ringC, ringA);
+
+				DRM_INFO("Testing syncing between rings %d, %d and %d...\n", k, i, j);
+				radeon_test_ring_sync2(rdev, ringC, ringA, ringB);
+
+				DRM_INFO("Testing syncing between rings %d, %d and %d...\n", k, j, i);
+				radeon_test_ring_sync2(rdev, ringC, ringB, ringA);
+			}
+		}
+	}
+}
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index e111a38..b0ebaf8 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -188,7 +188,7 @@
 	rbo = container_of(bo, struct radeon_bo, tbo);
 	switch (bo->mem.mem_type) {
 	case TTM_PL_VRAM:
-		if (rbo->rdev->cp.ready == false)
+		if (rbo->rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready == false)
 			radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_CPU);
 		else
 			radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_GTT);
@@ -226,7 +226,7 @@
 	int r;
 
 	rdev = radeon_get_rdev(bo->bdev);
-	r = radeon_fence_create(rdev, &fence);
+	r = radeon_fence_create(rdev, &fence, RADEON_RING_TYPE_GFX_INDEX);
 	if (unlikely(r)) {
 		return r;
 	}
@@ -255,7 +255,7 @@
 		DRM_ERROR("Unknown placement %d\n", old_mem->mem_type);
 		return -EINVAL;
 	}
-	if (!rdev->cp.ready) {
+	if (!rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready) {
 		DRM_ERROR("Trying to move memory with CP turned off.\n");
 		return -EINVAL;
 	}
@@ -380,7 +380,7 @@
 		radeon_move_null(bo, new_mem);
 		return 0;
 	}
-	if (!rdev->cp.ready || rdev->asic->copy == NULL) {
+	if (!rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready || rdev->asic->copy == NULL) {
 		/* use memcpy */
 		goto memcpy;
 	}
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index 06b90c8..b0ce84a 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -410,6 +410,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	r100_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
@@ -419,11 +425,18 @@
 		dev_err(rdev->dev, "failed initializing CP (%d).\n", r);
 		return r;
 	}
-	r = r100_ib_init(rdev);
+
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r100_ib_test(rdev);
 	if (r) {
-		dev_err(rdev->dev, "failed initializing IB (%d).\n", r);
+		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
 		return r;
 	}
+
 	return 0;
 }
 
@@ -447,11 +460,14 @@
 	r300_clock_startup(rdev);
 	/* Initialize surface registers */
 	radeon_surface_init(rdev);
+
+	rdev->accel_working = true;
 	return rs400_startup(rdev);
 }
 
 int rs400_suspend(struct radeon_device *rdev)
 {
+	radeon_ib_pool_suspend(rdev);
 	r100_cp_disable(rdev);
 	radeon_wb_disable(rdev);
 	r100_irq_disable(rdev);
@@ -530,7 +546,14 @@
 	if (r)
 		return r;
 	r300_set_reg_safe(rdev);
+
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = rs400_startup(rdev);
 	if (r) {
 		/* Somethings want wront with the accel init stop accel */
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 481b99e..ca6d5b6 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -62,6 +62,7 @@
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
+	int i;
 
 	/* Lock the graphics update lock */
 	tmp |= AVIVO_D1GRPH_UPDATE_LOCK;
@@ -74,7 +75,11 @@
 	       (u32)crtc_base);
 
 	/* Wait for update_pending to go high. */
-	while (!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING));
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		if (RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING)
+			break;
+		udelay(1);
+	}
 	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
@@ -544,7 +549,7 @@
 		WREG32(R_000040_GEN_INT_CNTL, 0);
 		return -EINVAL;
 	}
-	if (rdev->irq.sw_int) {
+	if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
 		tmp |= S_000040_SW_INT_EN(1);
 	}
 	if (rdev->irq.gui_idle) {
@@ -637,7 +642,7 @@
 	while (status || rdev->irq.stat_regs.r500.disp_int) {
 		/* SW interrupt */
 		if (G_000044_SW_INT(status)) {
-			radeon_fence_process(rdev);
+			radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX);
 		}
 		/* GUI idle */
 		if (G_000040_GUI_IDLE(status)) {
@@ -844,6 +849,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	rs600_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
@@ -853,11 +864,6 @@
 		dev_err(rdev->dev, "failed initializing CP (%d).\n", r);
 		return r;
 	}
-	r = r100_ib_init(rdev);
-	if (r) {
-		dev_err(rdev->dev, "failed initializing IB (%d).\n", r);
-		return r;
-	}
 
 	r = r600_audio_init(rdev);
 	if (r) {
@@ -865,6 +871,17 @@
 		return r;
 	}
 
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r100_ib_test(rdev);
+	if (r) {
+		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
+		return r;
+	}
+
 	return 0;
 }
 
@@ -886,11 +903,14 @@
 	rv515_clock_startup(rdev);
 	/* Initialize surface registers */
 	radeon_surface_init(rdev);
+
+	rdev->accel_working = true;
 	return rs600_startup(rdev);
 }
 
 int rs600_suspend(struct radeon_device *rdev)
 {
+	radeon_ib_pool_suspend(rdev);
 	r600_audio_fini(rdev);
 	r100_cp_disable(rdev);
 	radeon_wb_disable(rdev);
@@ -971,7 +991,14 @@
 	if (r)
 		return r;
 	rs600_set_safe_registers(rdev);
+
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = rs600_startup(rdev);
 	if (r) {
 		/* Somethings want wront with the accel init stop accel */
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index a9049ed..4f24a0f 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -621,6 +621,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	rs600_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
@@ -630,11 +636,6 @@
 		dev_err(rdev->dev, "failed initializing CP (%d).\n", r);
 		return r;
 	}
-	r = r100_ib_init(rdev);
-	if (r) {
-		dev_err(rdev->dev, "failed initializing IB (%d).\n", r);
-		return r;
-	}
 
 	r = r600_audio_init(rdev);
 	if (r) {
@@ -642,6 +643,17 @@
 		return r;
 	}
 
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r100_ib_test(rdev);
+	if (r) {
+		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
+		return r;
+	}
+
 	return 0;
 }
 
@@ -663,11 +675,14 @@
 	rv515_clock_startup(rdev);
 	/* Initialize surface registers */
 	radeon_surface_init(rdev);
+
+	rdev->accel_working = true;
 	return rs690_startup(rdev);
 }
 
 int rs690_suspend(struct radeon_device *rdev)
 {
+	radeon_ib_pool_suspend(rdev);
 	r600_audio_fini(rdev);
 	r100_cp_disable(rdev);
 	radeon_wb_disable(rdev);
@@ -749,7 +764,14 @@
 	if (r)
 		return r;
 	rs600_set_safe_registers(rdev);
+
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = rs690_startup(rdev);
 	if (r) {
 		/* Somethings want wront with the accel init stop accel */
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 6613ee9..880637f 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -55,44 +55,45 @@
 
 void rv515_ring_start(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 
-	r = radeon_ring_lock(rdev, 64);
+	r = radeon_ring_lock(rdev, ring, 64);
 	if (r) {
 		return;
 	}
-	radeon_ring_write(rdev, PACKET0(ISYNC_CNTL, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(ISYNC_CNTL, 0));
+	radeon_ring_write(ring,
 			  ISYNC_ANY2D_IDLE3D |
 			  ISYNC_ANY3D_IDLE2D |
 			  ISYNC_WAIT_IDLEGUI |
 			  ISYNC_CPSCRATCH_IDLEGUI);
-	radeon_ring_write(rdev, PACKET0(WAIT_UNTIL, 0));
-	radeon_ring_write(rdev, WAIT_2D_IDLECLEAN | WAIT_3D_IDLECLEAN);
-	radeon_ring_write(rdev, PACKET0(R300_DST_PIPE_CONFIG, 0));
-	radeon_ring_write(rdev, R300_PIPE_AUTO_CONFIG);
-	radeon_ring_write(rdev, PACKET0(GB_SELECT, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, PACKET0(GB_ENABLE, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, PACKET0(R500_SU_REG_DEST, 0));
-	radeon_ring_write(rdev, (1 << rdev->num_gb_pipes) - 1);
-	radeon_ring_write(rdev, PACKET0(VAP_INDEX_OFFSET, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, PACKET0(RB3D_DSTCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, RB3D_DC_FLUSH | RB3D_DC_FREE);
-	radeon_ring_write(rdev, PACKET0(ZB_ZCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, ZC_FLUSH | ZC_FREE);
-	radeon_ring_write(rdev, PACKET0(WAIT_UNTIL, 0));
-	radeon_ring_write(rdev, WAIT_2D_IDLECLEAN | WAIT_3D_IDLECLEAN);
-	radeon_ring_write(rdev, PACKET0(GB_AA_CONFIG, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_write(rdev, PACKET0(RB3D_DSTCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, RB3D_DC_FLUSH | RB3D_DC_FREE);
-	radeon_ring_write(rdev, PACKET0(ZB_ZCACHE_CTLSTAT, 0));
-	radeon_ring_write(rdev, ZC_FLUSH | ZC_FREE);
-	radeon_ring_write(rdev, PACKET0(GB_MSPOS0, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(WAIT_UNTIL, 0));
+	radeon_ring_write(ring, WAIT_2D_IDLECLEAN | WAIT_3D_IDLECLEAN);
+	radeon_ring_write(ring, PACKET0(R300_DST_PIPE_CONFIG, 0));
+	radeon_ring_write(ring, R300_PIPE_AUTO_CONFIG);
+	radeon_ring_write(ring, PACKET0(GB_SELECT, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, PACKET0(GB_ENABLE, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, PACKET0(R500_SU_REG_DEST, 0));
+	radeon_ring_write(ring, (1 << rdev->num_gb_pipes) - 1);
+	radeon_ring_write(ring, PACKET0(VAP_INDEX_OFFSET, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, PACKET0(RB3D_DSTCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, RB3D_DC_FLUSH | RB3D_DC_FREE);
+	radeon_ring_write(ring, PACKET0(ZB_ZCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, ZC_FLUSH | ZC_FREE);
+	radeon_ring_write(ring, PACKET0(WAIT_UNTIL, 0));
+	radeon_ring_write(ring, WAIT_2D_IDLECLEAN | WAIT_3D_IDLECLEAN);
+	radeon_ring_write(ring, PACKET0(GB_AA_CONFIG, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_write(ring, PACKET0(RB3D_DSTCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, RB3D_DC_FLUSH | RB3D_DC_FREE);
+	radeon_ring_write(ring, PACKET0(ZB_ZCACHE_CTLSTAT, 0));
+	radeon_ring_write(ring, ZC_FLUSH | ZC_FREE);
+	radeon_ring_write(ring, PACKET0(GB_MSPOS0, 0));
+	radeon_ring_write(ring,
 			  ((6 << MS_X0_SHIFT) |
 			   (6 << MS_Y0_SHIFT) |
 			   (6 << MS_X1_SHIFT) |
@@ -101,8 +102,8 @@
 			   (6 << MS_Y2_SHIFT) |
 			   (6 << MSBD0_Y_SHIFT) |
 			   (6 << MSBD0_X_SHIFT)));
-	radeon_ring_write(rdev, PACKET0(GB_MSPOS1, 0));
-	radeon_ring_write(rdev,
+	radeon_ring_write(ring, PACKET0(GB_MSPOS1, 0));
+	radeon_ring_write(ring,
 			  ((6 << MS_X3_SHIFT) |
 			   (6 << MS_Y3_SHIFT) |
 			   (6 << MS_X4_SHIFT) |
@@ -110,15 +111,15 @@
 			   (6 << MS_X5_SHIFT) |
 			   (6 << MS_Y5_SHIFT) |
 			   (6 << MSBD1_SHIFT)));
-	radeon_ring_write(rdev, PACKET0(GA_ENHANCE, 0));
-	radeon_ring_write(rdev, GA_DEADLOCK_CNTL | GA_FASTSYNC_CNTL);
-	radeon_ring_write(rdev, PACKET0(GA_POLY_MODE, 0));
-	radeon_ring_write(rdev, FRONT_PTYPE_TRIANGE | BACK_PTYPE_TRIANGE);
-	radeon_ring_write(rdev, PACKET0(GA_ROUND_MODE, 0));
-	radeon_ring_write(rdev, GEOMETRY_ROUND_NEAREST | COLOR_ROUND_NEAREST);
-	radeon_ring_write(rdev, PACKET0(0x20C8, 0));
-	radeon_ring_write(rdev, 0);
-	radeon_ring_unlock_commit(rdev);
+	radeon_ring_write(ring, PACKET0(GA_ENHANCE, 0));
+	radeon_ring_write(ring, GA_DEADLOCK_CNTL | GA_FASTSYNC_CNTL);
+	radeon_ring_write(ring, PACKET0(GA_POLY_MODE, 0));
+	radeon_ring_write(ring, FRONT_PTYPE_TRIANGE | BACK_PTYPE_TRIANGE);
+	radeon_ring_write(ring, PACKET0(GA_ROUND_MODE, 0));
+	radeon_ring_write(ring, GEOMETRY_ROUND_NEAREST | COLOR_ROUND_NEAREST);
+	radeon_ring_write(ring, PACKET0(0x20C8, 0));
+	radeon_ring_write(ring, 0);
+	radeon_ring_unlock_commit(rdev, ring);
 }
 
 int rv515_mc_wait_for_idle(struct radeon_device *rdev)
@@ -392,6 +393,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	rs600_irq_set(rdev);
 	rdev->config.r300.hdp_cntl = RREG32(RADEON_HOST_PATH_CNTL);
@@ -401,9 +408,15 @@
 		dev_err(rdev->dev, "failed initializing CP (%d).\n", r);
 		return r;
 	}
-	r = r100_ib_init(rdev);
+
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r100_ib_test(rdev);
 	if (r) {
-		dev_err(rdev->dev, "failed initializing IB (%d).\n", r);
+		dev_err(rdev->dev, "failed testing IB (%d).\n", r);
+		rdev->accel_working = false;
 		return r;
 	}
 	return 0;
@@ -428,6 +441,8 @@
 	rv515_clock_startup(rdev);
 	/* Initialize surface registers */
 	radeon_surface_init(rdev);
+
+	rdev->accel_working = true;
 	return rv515_startup(rdev);
 }
 
@@ -524,7 +539,14 @@
 	if (r)
 		return r;
 	rv515_set_safe_registers(rdev);
+
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = rv515_startup(rdev);
 	if (r) {
 		/* Somethings want wront with the accel init stop accel */
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index a983f41..a1668b6 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -47,6 +47,7 @@
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
+	int i;
 
 	/* Lock the graphics update lock */
 	tmp |= AVIVO_D1GRPH_UPDATE_LOCK;
@@ -66,7 +67,11 @@
 	       (u32)crtc_base);
 
 	/* Wait for update_pending to go high. */
-	while (!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING));
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		if (RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING)
+			break;
+		udelay(1);
+	}
 	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
@@ -352,7 +357,7 @@
 void r700_cp_fini(struct radeon_device *rdev)
 {
 	r700_cp_stop(rdev);
-	radeon_ring_fini(rdev);
+	radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
 }
 
 /*
@@ -1038,6 +1043,7 @@
 
 static int rv770_startup(struct radeon_device *rdev)
 {
+	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	int r;
 
 	/* enable pcie gen2 link */
@@ -1077,6 +1083,12 @@
 	if (r)
 		return r;
 
+	r = radeon_fence_driver_start_ring(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "failed initializing CP fences (%d).\n", r);
+		return r;
+	}
+
 	/* Enable IRQ */
 	r = r600_irq_init(rdev);
 	if (r) {
@@ -1086,7 +1098,9 @@
 	}
 	r600_irq_set(rdev);
 
-	r = radeon_ring_init(rdev, rdev->cp.ring_size);
+	r = radeon_ring_init(rdev, ring, ring->ring_size, RADEON_WB_CP_RPTR_OFFSET,
+			     R600_CP_RB_RPTR, R600_CP_RB_WPTR,
+			     0, 0xfffff, RADEON_CP_PACKET2);
 	if (r)
 		return r;
 	r = rv770_cp_load_microcode(rdev);
@@ -1096,6 +1110,17 @@
 	if (r)
 		return r;
 
+	r = radeon_ib_pool_start(rdev);
+	if (r)
+		return r;
+
+	r = r600_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX);
+	if (r) {
+		dev_err(rdev->dev, "IB test failed (%d).\n", r);
+		rdev->accel_working = false;
+		return r;
+	}
+
 	return 0;
 }
 
@@ -1110,18 +1135,13 @@
 	/* post card */
 	atom_asic_init(rdev->mode_info.atom_context);
 
+	rdev->accel_working = true;
 	r = rv770_startup(rdev);
 	if (r) {
 		DRM_ERROR("r600 startup failed on resume\n");
 		return r;
 	}
 
-	r = r600_ib_test(rdev);
-	if (r) {
-		DRM_ERROR("radeon: failed testing IB (%d).\n", r);
-		return r;
-	}
-
 	r = r600_audio_init(rdev);
 	if (r) {
 		dev_err(rdev->dev, "radeon: audio init failed\n");
@@ -1135,13 +1155,14 @@
 int rv770_suspend(struct radeon_device *rdev)
 {
 	r600_audio_fini(rdev);
+	radeon_ib_pool_suspend(rdev);
+	r600_blit_suspend(rdev);
 	/* FIXME: we should wait for ring to be empty */
 	r700_cp_stop(rdev);
-	rdev->cp.ready = false;
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
 	r600_irq_suspend(rdev);
 	radeon_wb_disable(rdev);
 	rv770_pcie_gart_disable(rdev);
-	r600_blit_suspend(rdev);
 
 	return 0;
 }
@@ -1210,8 +1231,8 @@
 	if (r)
 		return r;
 
-	rdev->cp.ring_obj = NULL;
-	r600_ring_init(rdev, 1024 * 1024);
+	rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ring_obj = NULL;
+	r600_ring_init(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX], 1024 * 1024);
 
 	rdev->ih.ring_obj = NULL;
 	r600_ih_ring_init(rdev, 64 * 1024);
@@ -1220,30 +1241,24 @@
 	if (r)
 		return r;
 
+	r = radeon_ib_pool_init(rdev);
 	rdev->accel_working = true;
+	if (r) {
+		dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
+		rdev->accel_working = false;
+	}
+
 	r = rv770_startup(rdev);
 	if (r) {
 		dev_err(rdev->dev, "disabling GPU acceleration\n");
 		r700_cp_fini(rdev);
 		r600_irq_fini(rdev);
 		radeon_wb_fini(rdev);
+		r100_ib_fini(rdev);
 		radeon_irq_kms_fini(rdev);
 		rv770_pcie_gart_fini(rdev);
 		rdev->accel_working = false;
 	}
-	if (rdev->accel_working) {
-		r = radeon_ib_pool_init(rdev);
-		if (r) {
-			dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-			rdev->accel_working = false;
-		} else {
-			r = r600_ib_test(rdev);
-			if (r) {
-				dev_err(rdev->dev, "IB test failed (%d).\n", r);
-				rdev->accel_working = false;
-			}
-		}
-	}
 
 	r = r600_audio_init(rdev);
 	if (r) {
@@ -1260,11 +1275,12 @@
 	r700_cp_fini(rdev);
 	r600_irq_fini(rdev);
 	radeon_wb_fini(rdev);
-	radeon_ib_pool_fini(rdev);
+	r100_ib_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	rv770_pcie_gart_fini(rdev);
 	r600_vram_scratch_fini(rdev);
 	radeon_gem_fini(rdev);
+	radeon_semaphore_driver_fini(rdev);
 	radeon_fence_driver_fini(rdev);
 	radeon_agp_fini(rdev);
 	radeon_bo_fini(rdev);
diff --git a/drivers/gpu/drm/sis/sis_drv.c b/drivers/gpu/drm/sis/sis_drv.c
index bda96a8..06da063 100644
--- a/drivers/gpu/drm/sis/sis_drv.c
+++ b/drivers/gpu/drm/sis/sis_drv.c
@@ -48,9 +48,7 @@
 
 	dev->dev_private = (void *)dev_priv;
 	dev_priv->chipset = chipset;
-	ret = drm_sman_init(&dev_priv->sman, 2, 12, 8);
-	if (ret)
-		kfree(dev_priv);
+	idr_init(&dev->object_name_idr);
 
 	return ret;
 }
@@ -59,7 +57,9 @@
 {
 	drm_sis_private_t *dev_priv = dev->dev_private;
 
-	drm_sman_takedown(&dev_priv->sman);
+	idr_remove_all(&dev_priv->object_idr);
+	idr_destroy(&dev_priv->object_idr);
+
 	kfree(dev_priv);
 
 	return 0;
@@ -76,10 +76,35 @@
 	.llseek = noop_llseek,
 };
 
+static int sis_driver_open(struct drm_device *dev, struct drm_file *file)
+{
+	struct sis_file_private *file_priv;
+
+	DRM_DEBUG_DRIVER("\n");
+	file_priv = kmalloc(sizeof(*file_priv), GFP_KERNEL);
+	if (!file_priv)
+		return -ENOMEM;
+
+	file->driver_priv = file_priv;
+
+	INIT_LIST_HEAD(&file_priv->obj_list);
+
+	return 0;
+}
+
+void sis_driver_postclose(struct drm_device *dev, struct drm_file *file)
+{
+	struct sis_file_private *file_priv = file->driver_priv;
+
+	kfree(file_priv);
+}
+
 static struct drm_driver driver = {
 	.driver_features = DRIVER_USE_AGP | DRIVER_USE_MTRR,
 	.load = sis_driver_load,
 	.unload = sis_driver_unload,
+	.open = sis_driver_open,
+	.postclose = sis_driver_postclose,
 	.dma_quiescent = sis_idle,
 	.reclaim_buffers = NULL,
 	.reclaim_buffers_idlelocked = sis_reclaim_buffers_locked,
diff --git a/drivers/gpu/drm/sis/sis_drv.h b/drivers/gpu/drm/sis/sis_drv.h
index 194303c..573758b 100644
--- a/drivers/gpu/drm/sis/sis_drv.h
+++ b/drivers/gpu/drm/sis/sis_drv.h
@@ -44,7 +44,7 @@
 	SIS_CHIP_315 = 1,
 };
 
-#include "drm_sman.h"
+#include "drm_mm.h"
 
 
 #define SIS_BASE (dev_priv->mmio)
@@ -54,12 +54,15 @@
 typedef struct drm_sis_private {
 	drm_local_map_t *mmio;
 	unsigned int idle_fault;
-	struct drm_sman sman;
 	unsigned int chipset;
 	int vram_initialized;
 	int agp_initialized;
 	unsigned long vram_offset;
 	unsigned long agp_offset;
+	struct drm_mm vram_mm;
+	struct drm_mm agp_mm;
+	/** Mapping of userspace keys to mm objects */
+	struct idr object_idr;
 } drm_sis_private_t;
 
 extern int sis_idle(struct drm_device *dev);
diff --git a/drivers/gpu/drm/sis/sis_mm.c b/drivers/gpu/drm/sis/sis_mm.c
index 7fe2b63..46cb2bc 100644
--- a/drivers/gpu/drm/sis/sis_mm.c
+++ b/drivers/gpu/drm/sis/sis_mm.c
@@ -41,40 +41,18 @@
 #define AGP_TYPE 1
 
 
+struct sis_memblock {
+	struct drm_mm_node mm_node;
+	struct sis_memreq req;
+	struct list_head owner_list;
+};
+
 #if defined(CONFIG_FB_SIS) || defined(CONFIG_FB_SIS_MODULE)
 /* fb management via fb device */
 
 #define SIS_MM_ALIGN_SHIFT 0
 #define SIS_MM_ALIGN_MASK 0
 
-static void *sis_sman_mm_allocate(void *private, unsigned long size,
-				  unsigned alignment)
-{
-	struct sis_memreq req;
-
-	req.size = size;
-	sis_malloc(&req);
-	if (req.size == 0)
-		return NULL;
-	else
-		return (void *)(unsigned long)~req.offset;
-}
-
-static void sis_sman_mm_free(void *private, void *ref)
-{
-	sis_free(~((unsigned long)ref));
-}
-
-static void sis_sman_mm_destroy(void *private)
-{
-	;
-}
-
-static unsigned long sis_sman_mm_offset(void *private, void *ref)
-{
-	return ~((unsigned long)ref);
-}
-
 #else /* CONFIG_FB_SIS[_MODULE] */
 
 #define SIS_MM_ALIGN_SHIFT 4
@@ -86,30 +64,11 @@
 {
 	drm_sis_private_t *dev_priv = dev->dev_private;
 	drm_sis_fb_t *fb = data;
-	int ret;
 
 	mutex_lock(&dev->struct_mutex);
-#if defined(CONFIG_FB_SIS) || defined(CONFIG_FB_SIS_MODULE)
-	{
-		struct drm_sman_mm sman_mm;
-		sman_mm.private = (void *)0xFFFFFFFF;
-		sman_mm.allocate = sis_sman_mm_allocate;
-		sman_mm.free = sis_sman_mm_free;
-		sman_mm.destroy = sis_sman_mm_destroy;
-		sman_mm.offset = sis_sman_mm_offset;
-		ret =
-		    drm_sman_set_manager(&dev_priv->sman, VIDEO_TYPE, &sman_mm);
-	}
-#else
-	ret = drm_sman_set_range(&dev_priv->sman, VIDEO_TYPE, 0,
-				 fb->size >> SIS_MM_ALIGN_SHIFT);
-#endif
-
-	if (ret) {
-		DRM_ERROR("VRAM memory manager initialisation error\n");
-		mutex_unlock(&dev->struct_mutex);
-		return ret;
-	}
+	/* Unconditionally init the drm_mm, even though we don't use it when the
+	 * fb sis driver is available - make cleanup easier. */
+	drm_mm_init(&dev_priv->vram_mm, 0, fb->size >> SIS_MM_ALIGN_SHIFT);
 
 	dev_priv->vram_initialized = 1;
 	dev_priv->vram_offset = fb->offset;
@@ -120,13 +79,15 @@
 	return 0;
 }
 
-static int sis_drm_alloc(struct drm_device *dev, struct drm_file *file_priv,
+static int sis_drm_alloc(struct drm_device *dev, struct drm_file *file,
 			 void *data, int pool)
 {
 	drm_sis_private_t *dev_priv = dev->dev_private;
 	drm_sis_mem_t *mem = data;
-	int retval = 0;
-	struct drm_memblock_item *item;
+	int retval = 0, user_key;
+	struct sis_memblock *item;
+	struct sis_file_private *file_priv = file->driver_priv;
+	unsigned long offset;
 
 	mutex_lock(&dev->struct_mutex);
 
@@ -138,25 +99,68 @@
 		return -EINVAL;
 	}
 
-	mem->size = (mem->size + SIS_MM_ALIGN_MASK) >> SIS_MM_ALIGN_SHIFT;
-	item = drm_sman_alloc(&dev_priv->sman, pool, mem->size, 0,
-			      (unsigned long)file_priv);
-
-	mutex_unlock(&dev->struct_mutex);
-	if (item) {
-		mem->offset = ((pool == 0) ?
-			      dev_priv->vram_offset : dev_priv->agp_offset) +
-		    (item->mm->
-		     offset(item->mm, item->mm_info) << SIS_MM_ALIGN_SHIFT);
-		mem->free = item->user_hash.key;
-		mem->size = mem->size << SIS_MM_ALIGN_SHIFT;
-	} else {
-		mem->offset = 0;
-		mem->size = 0;
-		mem->free = 0;
+	item = kzalloc(sizeof(*item), GFP_KERNEL);
+	if (!item) {
 		retval = -ENOMEM;
+		goto fail_alloc;
 	}
 
+	mem->size = (mem->size + SIS_MM_ALIGN_MASK) >> SIS_MM_ALIGN_SHIFT;
+	if (pool == AGP_TYPE) {
+		retval = drm_mm_insert_node(&dev_priv->agp_mm,
+					    &item->mm_node,
+					    mem->size, 0);
+		offset = item->mm_node.start;
+	} else {
+#if defined(CONFIG_FB_SIS) || defined(CONFIG_FB_SIS_MODULE)
+		item->req.size = mem->size;
+		sis_malloc(&item->req);
+		if (item->req.size == 0)
+			retval = -ENOMEM;
+		offset = item->req.offset;
+#else
+		retval = drm_mm_insert_node(&dev_priv->vram_mm,
+					    &item->mm_node,
+					    mem->size, 0);
+		offset = item->mm_node.start;
+#endif
+	}
+	if (retval)
+		goto fail_alloc;
+
+again:
+	if (idr_pre_get(&dev_priv->object_idr, GFP_KERNEL) == 0) {
+		retval = -ENOMEM;
+		goto fail_idr;
+	}
+
+	retval = idr_get_new_above(&dev_priv->object_idr, item, 1, &user_key);
+	if (retval == -EAGAIN)
+		goto again;
+	if (retval)
+		goto fail_idr;
+
+	list_add(&item->owner_list, &file_priv->obj_list);
+	mutex_unlock(&dev->struct_mutex);
+
+	mem->offset = ((pool == 0) ?
+		      dev_priv->vram_offset : dev_priv->agp_offset) +
+	    (offset << SIS_MM_ALIGN_SHIFT);
+	mem->free = user_key;
+	mem->size = mem->size << SIS_MM_ALIGN_SHIFT;
+
+	return 0;
+
+fail_idr:
+	drm_mm_remove_node(&item->mm_node);
+fail_alloc:
+	kfree(item);
+	mutex_unlock(&dev->struct_mutex);
+
+	mem->offset = 0;
+	mem->size = 0;
+	mem->free = 0;
+
 	DRM_DEBUG("alloc %d, size = %d, offset = %d\n", pool, mem->size,
 		  mem->offset);
 
@@ -167,10 +171,25 @@
 {
 	drm_sis_private_t *dev_priv = dev->dev_private;
 	drm_sis_mem_t *mem = data;
+	struct sis_memblock *obj;
 	int ret;
 
 	mutex_lock(&dev->struct_mutex);
-	ret = drm_sman_free_key(&dev_priv->sman, mem->free);
+	obj = idr_find(&dev_priv->object_idr, mem->free);
+	if (obj == NULL) {
+		mutex_unlock(&dev->struct_mutex);
+		return -EINVAL;
+	}
+
+	idr_remove(&dev_priv->object_idr, mem->free);
+	list_del(&obj->owner_list);
+	if (drm_mm_node_allocated(&obj->mm_node))
+		drm_mm_remove_node(&obj->mm_node);
+#if defined(CONFIG_FB_SIS) || defined(CONFIG_FB_SIS_MODULE)
+	else
+		sis_free(obj->req.offset);
+#endif
+	kfree(obj);
 	mutex_unlock(&dev->struct_mutex);
 	DRM_DEBUG("free = 0x%lx\n", mem->free);
 
@@ -188,18 +207,10 @@
 {
 	drm_sis_private_t *dev_priv = dev->dev_private;
 	drm_sis_agp_t *agp = data;
-	int ret;
 	dev_priv = dev->dev_private;
 
 	mutex_lock(&dev->struct_mutex);
-	ret = drm_sman_set_range(&dev_priv->sman, AGP_TYPE, 0,
-				 agp->size >> SIS_MM_ALIGN_SHIFT);
-
-	if (ret) {
-		DRM_ERROR("AGP memory manager initialisation error\n");
-		mutex_unlock(&dev->struct_mutex);
-		return ret;
-	}
+	drm_mm_init(&dev_priv->agp_mm, 0, agp->size >> SIS_MM_ALIGN_SHIFT);
 
 	dev_priv->agp_initialized = 1;
 	dev_priv->agp_offset = agp->offset;
@@ -293,20 +304,26 @@
 		return;
 
 	mutex_lock(&dev->struct_mutex);
-	drm_sman_cleanup(&dev_priv->sman);
-	dev_priv->vram_initialized = 0;
-	dev_priv->agp_initialized = 0;
+	if (dev_priv->vram_initialized) {
+		drm_mm_takedown(&dev_priv->vram_mm);
+		dev_priv->vram_initialized = 0;
+	}
+	if (dev_priv->agp_initialized) {
+		drm_mm_takedown(&dev_priv->agp_mm);
+		dev_priv->agp_initialized = 0;
+	}
 	dev_priv->mmio = NULL;
 	mutex_unlock(&dev->struct_mutex);
 }
 
 void sis_reclaim_buffers_locked(struct drm_device *dev,
-				struct drm_file *file_priv)
+				struct drm_file *file)
 {
-	drm_sis_private_t *dev_priv = dev->dev_private;
+	struct sis_file_private *file_priv = file->driver_priv;
+	struct sis_memblock *entry, *next;
 
 	mutex_lock(&dev->struct_mutex);
-	if (drm_sman_owner_clean(&dev_priv->sman, (unsigned long)file_priv)) {
+	if (list_empty(&file_priv->obj_list)) {
 		mutex_unlock(&dev->struct_mutex);
 		return;
 	}
@@ -314,7 +331,18 @@
 	if (dev->driver->dma_quiescent)
 		dev->driver->dma_quiescent(dev);
 
-	drm_sman_owner_cleanup(&dev_priv->sman, (unsigned long)file_priv);
+
+	list_for_each_entry_safe(entry, next, &file_priv->obj_list,
+				 owner_list) {
+		list_del(&entry->owner_list);
+		if (drm_mm_node_allocated(&entry->mm_node))
+			drm_mm_remove_node(&entry->mm_node);
+#if defined(CONFIG_FB_SIS) || defined(CONFIG_FB_SIS_MODULE)
+		else
+			sis_free(entry->req.offset);
+#endif
+		kfree(entry);
+	}
 	mutex_unlock(&dev->struct_mutex);
 	return;
 }
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 119b6e3..2f0eab6 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -562,10 +562,16 @@
 		return ret;
 
 	spin_lock(&glob->lru_lock);
+
+	if (unlikely(list_empty(&bo->ddestroy))) {
+		spin_unlock(&glob->lru_lock);
+		return 0;
+	}
+
 	ret = ttm_bo_reserve_locked(bo, interruptible,
 				    no_wait_reserve, false, 0);
 
-	if (unlikely(ret != 0) || list_empty(&bo->ddestroy)) {
+	if (unlikely(ret != 0)) {
 		spin_unlock(&glob->lru_lock);
 		return ret;
 	}
diff --git a/drivers/gpu/drm/via/via_drv.c b/drivers/gpu/drm/via/via_drv.c
index fb43fd3..02661f3 100644
--- a/drivers/gpu/drm/via/via_drv.c
+++ b/drivers/gpu/drm/via/via_drv.c
@@ -30,6 +30,29 @@
 
 #include "drm_pciids.h"
 
+static int via_driver_open(struct drm_device *dev, struct drm_file *file)
+{
+	struct via_file_private *file_priv;
+
+	DRM_DEBUG_DRIVER("\n");
+	file_priv = kmalloc(sizeof(*file_priv), GFP_KERNEL);
+	if (!file_priv)
+		return -ENOMEM;
+
+	file->driver_priv = file_priv;
+
+	INIT_LIST_HEAD(&file_priv->obj_list);
+
+	return 0;
+}
+
+void via_driver_postclose(struct drm_device *dev, struct drm_file *file)
+{
+	struct via_file_private *file_priv = file->driver_priv;
+
+	kfree(file_priv);
+}
+
 static struct pci_device_id pciidlist[] = {
 	viadrv_PCI_IDS
 };
@@ -51,6 +74,8 @@
 	    DRIVER_IRQ_SHARED,
 	.load = via_driver_load,
 	.unload = via_driver_unload,
+	.open = via_driver_open,
+	.postclose = via_driver_postclose,
 	.context_dtor = via_final_context,
 	.get_vblank_counter = via_get_vblank_counter,
 	.enable_vblank = via_enable_vblank,
diff --git a/drivers/gpu/drm/via/via_drv.h b/drivers/gpu/drm/via/via_drv.h
index 9cf87d9..88edacc 100644
--- a/drivers/gpu/drm/via/via_drv.h
+++ b/drivers/gpu/drm/via/via_drv.h
@@ -24,7 +24,7 @@
 #ifndef _VIA_DRV_H_
 #define _VIA_DRV_H_
 
-#include "drm_sman.h"
+#include "drm_mm.h"
 #define DRIVER_AUTHOR	"Various"
 
 #define DRIVER_NAME		"via"
@@ -88,9 +88,12 @@
 	uint32_t irq_pending_mask;
 	int *irq_map;
 	unsigned int idle_fault;
-	struct drm_sman sman;
 	int vram_initialized;
+	struct drm_mm vram_mm;
 	int agp_initialized;
+	struct drm_mm agp_mm;
+	/** Mapping of userspace keys to mm objects */
+	struct idr object_idr;
 	unsigned long vram_offset;
 	unsigned long agp_offset;
 	drm_via_blitq_t blit_queues[VIA_NUM_BLIT_ENGINES];
diff --git a/drivers/gpu/drm/via/via_map.c b/drivers/gpu/drm/via/via_map.c
index 6cca9a7..a2ab343 100644
--- a/drivers/gpu/drm/via/via_map.c
+++ b/drivers/gpu/drm/via/via_map.c
@@ -104,15 +104,10 @@
 
 	dev_priv->chipset = chipset;
 
-	ret = drm_sman_init(&dev_priv->sman, 2, 12, 8);
-	if (ret) {
-		kfree(dev_priv);
-		return ret;
-	}
+	idr_init(&dev->object_name_idr);
 
 	ret = drm_vblank_init(dev, 1);
 	if (ret) {
-		drm_sman_takedown(&dev_priv->sman);
 		kfree(dev_priv);
 		return ret;
 	}
@@ -124,7 +119,8 @@
 {
 	drm_via_private_t *dev_priv = dev->dev_private;
 
-	drm_sman_takedown(&dev_priv->sman);
+	idr_remove_all(&dev_priv->object_idr);
+	idr_destroy(&dev_priv->object_idr);
 
 	kfree(dev_priv);
 
diff --git a/drivers/gpu/drm/via/via_mm.c b/drivers/gpu/drm/via/via_mm.c
index 6cc2dad..bedb23d 100644
--- a/drivers/gpu/drm/via/via_mm.c
+++ b/drivers/gpu/drm/via/via_mm.c
@@ -28,26 +28,22 @@
 #include "drmP.h"
 #include "via_drm.h"
 #include "via_drv.h"
-#include "drm_sman.h"
 
 #define VIA_MM_ALIGN_SHIFT 4
 #define VIA_MM_ALIGN_MASK ((1 << VIA_MM_ALIGN_SHIFT) - 1)
 
+struct via_memblock {
+	struct drm_mm_node mm_node;
+	struct list_head owner_list;
+};
+
 int via_agp_init(struct drm_device *dev, void *data, struct drm_file *file_priv)
 {
 	drm_via_agp_t *agp = data;
 	drm_via_private_t *dev_priv = (drm_via_private_t *) dev->dev_private;
-	int ret;
 
 	mutex_lock(&dev->struct_mutex);
-	ret = drm_sman_set_range(&dev_priv->sman, VIA_MEM_AGP, 0,
-				 agp->size >> VIA_MM_ALIGN_SHIFT);
-
-	if (ret) {
-		DRM_ERROR("AGP memory manager initialisation error\n");
-		mutex_unlock(&dev->struct_mutex);
-		return ret;
-	}
+	drm_mm_init(&dev_priv->agp_mm, 0, agp->size >> VIA_MM_ALIGN_SHIFT);
 
 	dev_priv->agp_initialized = 1;
 	dev_priv->agp_offset = agp->offset;
@@ -61,17 +57,9 @@
 {
 	drm_via_fb_t *fb = data;
 	drm_via_private_t *dev_priv = (drm_via_private_t *) dev->dev_private;
-	int ret;
 
 	mutex_lock(&dev->struct_mutex);
-	ret = drm_sman_set_range(&dev_priv->sman, VIA_MEM_VIDEO, 0,
-				 fb->size >> VIA_MM_ALIGN_SHIFT);
-
-	if (ret) {
-		DRM_ERROR("VRAM memory manager initialisation error\n");
-		mutex_unlock(&dev->struct_mutex);
-		return ret;
-	}
+	drm_mm_init(&dev_priv->vram_mm, 0, fb->size >> VIA_MM_ALIGN_SHIFT);
 
 	dev_priv->vram_initialized = 1;
 	dev_priv->vram_offset = fb->offset;
@@ -108,19 +96,25 @@
 		return;
 
 	mutex_lock(&dev->struct_mutex);
-	drm_sman_cleanup(&dev_priv->sman);
-	dev_priv->vram_initialized = 0;
-	dev_priv->agp_initialized = 0;
+	if (dev_priv->vram_initialized) {
+		drm_mm_takedown(&dev_priv->vram_mm);
+		dev_priv->vram_initialized = 0;
+	}
+	if (dev_priv->agp_initialized) {
+		drm_mm_takedown(&dev_priv->agp_mm);
+		dev_priv->agp_initialized = 0;
+	}
 	mutex_unlock(&dev->struct_mutex);
 }
 
 int via_mem_alloc(struct drm_device *dev, void *data,
-		  struct drm_file *file_priv)
+		  struct drm_file *file)
 {
 	drm_via_mem_t *mem = data;
-	int retval = 0;
-	struct drm_memblock_item *item;
+	int retval = 0, user_key;
+	struct via_memblock *item;
 	drm_via_private_t *dev_priv = (drm_via_private_t *) dev->dev_private;
+	struct via_file_private *file_priv = file->driver_priv;
 	unsigned long tmpSize;
 
 	if (mem->type > VIA_MEM_AGP) {
@@ -136,24 +130,57 @@
 		return -EINVAL;
 	}
 
-	tmpSize = (mem->size + VIA_MM_ALIGN_MASK) >> VIA_MM_ALIGN_SHIFT;
-	item = drm_sman_alloc(&dev_priv->sman, mem->type, tmpSize, 0,
-			      (unsigned long)file_priv);
-	mutex_unlock(&dev->struct_mutex);
-	if (item) {
-		mem->offset = ((mem->type == VIA_MEM_VIDEO) ?
-			      dev_priv->vram_offset : dev_priv->agp_offset) +
-		    (item->mm->
-		     offset(item->mm, item->mm_info) << VIA_MM_ALIGN_SHIFT);
-		mem->index = item->user_hash.key;
-	} else {
-		mem->offset = 0;
-		mem->size = 0;
-		mem->index = 0;
-		DRM_DEBUG("Video memory allocation failed\n");
+	item = kzalloc(sizeof(*item), GFP_KERNEL);
+	if (!item) {
 		retval = -ENOMEM;
+		goto fail_alloc;
 	}
 
+	tmpSize = (mem->size + VIA_MM_ALIGN_MASK) >> VIA_MM_ALIGN_SHIFT;
+	if (mem->type == VIA_MEM_AGP)
+		retval = drm_mm_insert_node(&dev_priv->agp_mm,
+					    &item->mm_node,
+					    tmpSize, 0);
+	else
+		retval = drm_mm_insert_node(&dev_priv->vram_mm,
+					    &item->mm_node,
+					    tmpSize, 0);
+	if (retval)
+		goto fail_alloc;
+
+again:
+	if (idr_pre_get(&dev_priv->object_idr, GFP_KERNEL) == 0) {
+		retval = -ENOMEM;
+		goto fail_idr;
+	}
+
+	retval = idr_get_new_above(&dev_priv->object_idr, item, 1, &user_key);
+	if (retval == -EAGAIN)
+		goto again;
+	if (retval)
+		goto fail_idr;
+
+	list_add(&item->owner_list, &file_priv->obj_list);
+	mutex_unlock(&dev->struct_mutex);
+
+	mem->offset = ((mem->type == VIA_MEM_VIDEO) ?
+		      dev_priv->vram_offset : dev_priv->agp_offset) +
+	    ((item->mm_node.start) << VIA_MM_ALIGN_SHIFT);
+	mem->index = user_key;
+
+	return 0;
+
+fail_idr:
+	drm_mm_remove_node(&item->mm_node);
+fail_alloc:
+	kfree(item);
+	mutex_unlock(&dev->struct_mutex);
+
+	mem->offset = 0;
+	mem->size = 0;
+	mem->index = 0;
+	DRM_DEBUG("Video memory allocation failed\n");
+
 	return retval;
 }
 
@@ -161,11 +188,22 @@
 {
 	drm_via_private_t *dev_priv = dev->dev_private;
 	drm_via_mem_t *mem = data;
+	struct via_memblock *obj;
 	int ret;
 
 	mutex_lock(&dev->struct_mutex);
-	ret = drm_sman_free_key(&dev_priv->sman, mem->index);
+	obj = idr_find(&dev_priv->object_idr, mem->index);
+	if (obj == NULL) {
+		mutex_unlock(&dev->struct_mutex);
+		return -EINVAL;
+	}
+
+	idr_remove(&dev_priv->object_idr, mem->index);
+	list_del(&obj->owner_list);
+	drm_mm_remove_node(&obj->mm_node);
+	kfree(obj);
 	mutex_unlock(&dev->struct_mutex);
+
 	DRM_DEBUG("free = 0x%lx\n", mem->index);
 
 	return ret;
@@ -173,12 +211,13 @@
 
 
 void via_reclaim_buffers_locked(struct drm_device *dev,
-				struct drm_file *file_priv)
+				struct drm_file *file)
 {
-	drm_via_private_t *dev_priv = dev->dev_private;
+	struct via_file_private *file_priv = file->driver_priv;
+	struct via_memblock *entry, *next;
 
 	mutex_lock(&dev->struct_mutex);
-	if (drm_sman_owner_clean(&dev_priv->sman, (unsigned long)file_priv)) {
+	if (list_empty(&file_priv->obj_list)) {
 		mutex_unlock(&dev->struct_mutex);
 		return;
 	}
@@ -186,7 +225,12 @@
 	if (dev->driver->dma_quiescent)
 		dev->driver->dma_quiescent(dev);
 
-	drm_sman_owner_cleanup(&dev_priv->sman, (unsigned long)file_priv);
+	list_for_each_entry_safe(entry, next, &file_priv->obj_list,
+				 owner_list) {
+		list_del(&entry->owner_list);
+		drm_mm_remove_node(&entry->mm_node);
+		kfree(entry);
+	}
 	mutex_unlock(&dev->struct_mutex);
 	return;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 8cca91a..dc27970 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -390,6 +390,11 @@
 			     struct ttm_object_file *tfile,
 			     int id,
 			     struct vmw_resource **p_res);
+extern int vmw_user_lookup_handle(struct vmw_private *dev_priv,
+				  struct ttm_object_file *tfile,
+				  uint32_t handle,
+				  struct vmw_surface **out_surf,
+				  struct vmw_dma_buffer **out_buf);
 extern void vmw_surface_res_free(struct vmw_resource *res);
 extern int vmw_surface_init(struct vmw_private *dev_priv,
 			    struct vmw_surface *srf,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
index 03bbc2a..a0c2f12 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
@@ -33,6 +33,7 @@
 {
 	__le32 __iomem *fifo_mem = dev_priv->mmio_virt;
 	uint32_t fifo_min, hwversion;
+	const struct vmw_fifo_state *fifo = &dev_priv->fifo;
 
 	if (!(dev_priv->capabilities & SVGA_CAP_EXTENDED_FIFO))
 		return false;
@@ -41,7 +42,12 @@
 	if (fifo_min <= SVGA_FIFO_3D_HWVERSION * sizeof(unsigned int))
 		return false;
 
-	hwversion = ioread32(fifo_mem + SVGA_FIFO_3D_HWVERSION);
+	hwversion = ioread32(fifo_mem +
+			     ((fifo->capabilities &
+			       SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ?
+			      SVGA_FIFO_3D_HWVERSION_REVISED :
+			      SVGA_FIFO_3D_HWVERSION));
+
 	if (hwversion == 0)
 		return false;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index 3f63435..66917c6 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -58,8 +58,14 @@
 	case DRM_VMW_PARAM_FIFO_HW_VERSION:
 	{
 		__le32 __iomem *fifo_mem = dev_priv->mmio_virt;
+		const struct vmw_fifo_state *fifo = &dev_priv->fifo;
 
-		param->value = ioread32(fifo_mem + SVGA_FIFO_3D_HWVERSION);
+		param->value =
+			ioread32(fifo_mem +
+				 ((fifo->capabilities &
+				   SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ?
+				  SVGA_FIFO_3D_HWVERSION_REVISED :
+				  SVGA_FIFO_3D_HWVERSION));
 		break;
 	}
 	default:
@@ -140,7 +146,7 @@
 		goto out_clips;
 	}
 
-	clips = kzalloc(num_clips * sizeof(*clips), GFP_KERNEL);
+	clips = kcalloc(num_clips, sizeof(*clips), GFP_KERNEL);
 	if (clips == NULL) {
 		DRM_ERROR("Failed to allocate clip rect list.\n");
 		ret = -ENOMEM;
@@ -166,13 +172,7 @@
 		ret = -EINVAL;
 		goto out_no_fb;
 	}
-
 	vfb = vmw_framebuffer_to_vfb(obj_to_fb(obj));
-	if (!vfb->dmabuf) {
-		DRM_ERROR("Framebuffer not dmabuf backed.\n");
-		ret = -EINVAL;
-		goto out_no_fb;
-	}
 
 	ret = ttm_read_lock(&vmaster->lock, true);
 	if (unlikely(ret != 0))
@@ -232,7 +232,7 @@
 		goto out_clips;
 	}
 
-	clips = kzalloc(num_clips * sizeof(*clips), GFP_KERNEL);
+	clips = kcalloc(num_clips, sizeof(*clips), GFP_KERNEL);
 	if (clips == NULL) {
 		DRM_ERROR("Failed to allocate clip rect list.\n");
 		ret = -ENOMEM;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index 760d04ae..c4bdef3 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -31,6 +31,44 @@
 /* Might need a hrtimer here? */
 #define VMWGFX_PRESENT_RATE ((HZ / 60 > 0) ? HZ / 60 : 1)
 
+
+struct vmw_clip_rect {
+	int x1, x2, y1, y2;
+};
+
+/**
+ * Clip @num_rects number of @rects against @clip storing the
+ * results in @out_rects and the number of passed rects in @out_num.
+ */
+void vmw_clip_cliprects(struct drm_clip_rect *rects,
+			int num_rects,
+			struct vmw_clip_rect clip,
+			SVGASignedRect *out_rects,
+			int *out_num)
+{
+	int i, k;
+
+	for (i = 0, k = 0; i < num_rects; i++) {
+		int x1 = max_t(int, clip.x1, rects[i].x1);
+		int y1 = max_t(int, clip.y1, rects[i].y1);
+		int x2 = min_t(int, clip.x2, rects[i].x2);
+		int y2 = min_t(int, clip.y2, rects[i].y2);
+
+		if (x1 >= x2)
+			continue;
+		if (y1 >= y2)
+			continue;
+
+		out_rects[k].left   = x1;
+		out_rects[k].top    = y1;
+		out_rects[k].right  = x2;
+		out_rects[k].bottom = y2;
+		k++;
+	}
+
+	*out_num = k;
+}
+
 void vmw_display_unit_cleanup(struct vmw_display_unit *du)
 {
 	if (du->cursor_surface)
@@ -82,6 +120,43 @@
 	return 0;
 }
 
+int vmw_cursor_update_dmabuf(struct vmw_private *dev_priv,
+			     struct vmw_dma_buffer *dmabuf,
+			     u32 width, u32 height,
+			     u32 hotspotX, u32 hotspotY)
+{
+	struct ttm_bo_kmap_obj map;
+	unsigned long kmap_offset;
+	unsigned long kmap_num;
+	void *virtual;
+	bool dummy;
+	int ret;
+
+	kmap_offset = 0;
+	kmap_num = (width*height*4 + PAGE_SIZE - 1) >> PAGE_SHIFT;
+
+	ret = ttm_bo_reserve(&dmabuf->base, true, false, false, 0);
+	if (unlikely(ret != 0)) {
+		DRM_ERROR("reserve failed\n");
+		return -EINVAL;
+	}
+
+	ret = ttm_bo_kmap(&dmabuf->base, kmap_offset, kmap_num, &map);
+	if (unlikely(ret != 0))
+		goto err_unreserve;
+
+	virtual = ttm_kmap_obj_virtual(&map, &dummy);
+	ret = vmw_cursor_update_image(dev_priv, virtual, width, height,
+				      hotspotX, hotspotY);
+
+	ttm_bo_kunmap(&map);
+err_unreserve:
+	ttm_bo_unreserve(&dmabuf->base);
+
+	return ret;
+}
+
+
 void vmw_cursor_update_position(struct vmw_private *dev_priv,
 				bool show, int x, int y)
 {
@@ -105,25 +180,26 @@
 	struct vmw_dma_buffer *dmabuf = NULL;
 	int ret;
 
+	/* A lot of the code assumes this */
+	if (handle && (width != 64 || height != 64))
+		return -EINVAL;
+
 	if (handle) {
-		ret = vmw_user_surface_lookup_handle(dev_priv, tfile,
-						     handle, &surface);
-		if (!ret) {
-			if (!surface->snooper.image) {
-				DRM_ERROR("surface not suitable for cursor\n");
-				vmw_surface_unreference(&surface);
-				return -EINVAL;
-			}
-		} else {
-			ret = vmw_user_dmabuf_lookup(tfile,
-						     handle, &dmabuf);
-			if (ret) {
-				DRM_ERROR("failed to find surface or dmabuf: %i\n", ret);
-				return -EINVAL;
-			}
+		ret = vmw_user_lookup_handle(dev_priv, tfile,
+					     handle, &surface, &dmabuf);
+		if (ret) {
+			DRM_ERROR("failed to find surface or dmabuf: %i\n", ret);
+			return -EINVAL;
 		}
 	}
 
+	/* need to do this before taking down old image */
+	if (surface && !surface->snooper.image) {
+		DRM_ERROR("surface not suitable for cursor\n");
+		vmw_surface_unreference(&surface);
+		return -EINVAL;
+	}
+
 	/* takedown old cursor */
 	if (du->cursor_surface) {
 		du->cursor_surface->snooper.crtc = NULL;
@@ -142,36 +218,11 @@
 		vmw_cursor_update_image(dev_priv, surface->snooper.image,
 					64, 64, du->hotspot_x, du->hotspot_y);
 	} else if (dmabuf) {
-		struct ttm_bo_kmap_obj map;
-		unsigned long kmap_offset;
-		unsigned long kmap_num;
-		void *virtual;
-		bool dummy;
-
 		/* vmw_user_surface_lookup takes one reference */
 		du->cursor_dmabuf = dmabuf;
 
-		kmap_offset = 0;
-		kmap_num = (64*64*4) >> PAGE_SHIFT;
-
-		ret = ttm_bo_reserve(&dmabuf->base, true, false, false, 0);
-		if (unlikely(ret != 0)) {
-			DRM_ERROR("reserve failed\n");
-			return -EINVAL;
-		}
-
-		ret = ttm_bo_kmap(&dmabuf->base, kmap_offset, kmap_num, &map);
-		if (unlikely(ret != 0))
-			goto err_unreserve;
-
-		virtual = ttm_kmap_obj_virtual(&map, &dummy);
-		vmw_cursor_update_image(dev_priv, virtual, 64, 64,
-					du->hotspot_x, du->hotspot_y);
-
-		ttm_bo_kunmap(&map);
-err_unreserve:
-		ttm_bo_unreserve(&dmabuf->base);
-
+		ret = vmw_cursor_update_dmabuf(dev_priv, dmabuf, width, height,
+					       du->hotspot_x, du->hotspot_y);
 	} else {
 		vmw_cursor_update_position(dev_priv, false, 0, 0);
 		return 0;
@@ -373,8 +424,9 @@
 				struct drm_clip_rect *clips,
 				unsigned num_clips, int inc)
 {
-	struct drm_clip_rect *clips_ptr;
 	struct vmw_display_unit *units[VMWGFX_NUM_DISPLAY_UNITS];
+	struct drm_clip_rect *clips_ptr;
+	struct drm_clip_rect *tmp;
 	struct drm_crtc *crtc;
 	size_t fifo_size;
 	int i, num_units;
@@ -387,7 +439,6 @@
 	} *cmd;
 	SVGASignedRect *blits;
 
-
 	num_units = 0;
 	list_for_each_entry(crtc, &dev_priv->dev->mode_config.crtc_list,
 			    head) {
@@ -398,20 +449,32 @@
 
 	BUG_ON(!clips || !num_clips);
 
+	tmp = kzalloc(sizeof(*tmp) * num_clips, GFP_KERNEL);
+	if (unlikely(tmp == NULL)) {
+		DRM_ERROR("Temporary cliprect memory alloc failed.\n");
+		return -ENOMEM;
+	}
+
 	fifo_size = sizeof(*cmd) + sizeof(SVGASignedRect) * num_clips;
 	cmd = kzalloc(fifo_size, GFP_KERNEL);
 	if (unlikely(cmd == NULL)) {
 		DRM_ERROR("Temporary fifo memory alloc failed.\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto out_free_tmp;
 	}
 
+	/* setup blits pointer */
+	blits = (SVGASignedRect *)&cmd[1];
+
+	/* initial clip region */
 	left = clips->x1;
 	right = clips->x2;
 	top = clips->y1;
 	bottom = clips->y2;
 
-	clips_ptr = clips;
-	for (i = 1; i < num_clips; i++, clips_ptr += inc) {
+	/* skip the first clip rect */
+	for (i = 1, clips_ptr = clips + inc;
+	     i < num_clips; i++, clips_ptr += inc) {
 		left = min_t(int, left, (int)clips_ptr->x1);
 		right = max_t(int, right, (int)clips_ptr->x2);
 		top = min_t(int, top, (int)clips_ptr->y1);
@@ -429,45 +492,60 @@
 	cmd->body.srcRect.bottom = bottom;
 
 	clips_ptr = clips;
-	blits = (SVGASignedRect *)&cmd[1];
 	for (i = 0; i < num_clips; i++, clips_ptr += inc) {
-		blits[i].left   = clips_ptr->x1 - left;
-		blits[i].right  = clips_ptr->x2 - left;
-		blits[i].top    = clips_ptr->y1 - top;
-		blits[i].bottom = clips_ptr->y2 - top;
+		tmp[i].x1 = clips_ptr->x1 - left;
+		tmp[i].x2 = clips_ptr->x2 - left;
+		tmp[i].y1 = clips_ptr->y1 - top;
+		tmp[i].y2 = clips_ptr->y2 - top;
 	}
 
 	/* do per unit writing, reuse fifo for each */
 	for (i = 0; i < num_units; i++) {
 		struct vmw_display_unit *unit = units[i];
-		int clip_x1 = left - unit->crtc.x;
-		int clip_y1 = top - unit->crtc.y;
-		int clip_x2 = right - unit->crtc.x;
-		int clip_y2 = bottom - unit->crtc.y;
+		struct vmw_clip_rect clip;
+		int num;
+
+		clip.x1 = left - unit->crtc.x;
+		clip.y1 = top - unit->crtc.y;
+		clip.x2 = right - unit->crtc.x;
+		clip.y2 = bottom - unit->crtc.y;
 
 		/* skip any crtcs that misses the clip region */
-		if (clip_x1 >= unit->crtc.mode.hdisplay ||
-		    clip_y1 >= unit->crtc.mode.vdisplay ||
-		    clip_x2 <= 0 || clip_y2 <= 0)
+		if (clip.x1 >= unit->crtc.mode.hdisplay ||
+		    clip.y1 >= unit->crtc.mode.vdisplay ||
+		    clip.x2 <= 0 || clip.y2 <= 0)
 			continue;
 
+		/*
+		 * In order for the clip rects to be correctly scaled
+		 * the src and dest rects needs to be the same size.
+		 */
+		cmd->body.destRect.left = clip.x1;
+		cmd->body.destRect.right = clip.x2;
+		cmd->body.destRect.top = clip.y1;
+		cmd->body.destRect.bottom = clip.y2;
+
+		/* create a clip rect of the crtc in dest coords */
+		clip.x2 = unit->crtc.mode.hdisplay - clip.x1;
+		clip.y2 = unit->crtc.mode.vdisplay - clip.y1;
+		clip.x1 = 0 - clip.x1;
+		clip.y1 = 0 - clip.y1;
+
 		/* need to reset sid as it is changed by execbuf */
 		cmd->body.srcImage.sid = cpu_to_le32(framebuffer->user_handle);
-
 		cmd->body.destScreenId = unit->unit;
 
-		/*
-		 * The blit command is a lot more resilient then the
-		 * readback command when it comes to clip rects. So its
-		 * okay to go out of bounds.
-		 */
+		/* clip and write blits to cmd stream */
+		vmw_clip_cliprects(tmp, num_clips, clip, blits, &num);
 
-		cmd->body.destRect.left = clip_x1;
-		cmd->body.destRect.right = clip_x2;
-		cmd->body.destRect.top = clip_y1;
-		cmd->body.destRect.bottom = clip_y2;
+		/* if no cliprects hit skip this */
+		if (num == 0)
+			continue;
 
 
+		/* recalculate package length */
+		fifo_size = sizeof(*cmd) + sizeof(SVGASignedRect) * num;
+		cmd->header.size = cpu_to_le32(fifo_size - sizeof(cmd->header));
 		ret = vmw_execbuf_process(file_priv, dev_priv, NULL, cmd,
 					  fifo_size, 0, NULL);
 
@@ -475,7 +553,10 @@
 			break;
 	}
 
+
 	kfree(cmd);
+out_free_tmp:
+	kfree(tmp);
 
 	return ret;
 }
@@ -551,6 +632,10 @@
 	 * Sanity checks.
 	 */
 
+	/* Surface must be marked as a scanout. */
+	if (unlikely(!surface->scanout))
+		return -EINVAL;
+
 	if (unlikely(surface->mip_levels[0] != 1 ||
 		     surface->num_sizes != 1 ||
 		     surface->sizes[0].width < mode_cmd->width ||
@@ -605,7 +690,7 @@
 
 	/* XXX get the first 3 from the surface info */
 	vfbs->base.base.bits_per_pixel = mode_cmd->bpp;
-	vfbs->base.base.pitch = mode_cmd->pitch;
+	vfbs->base.base.pitches[0] = mode_cmd->pitch;
 	vfbs->base.base.depth = mode_cmd->depth;
 	vfbs->base.base.width = mode_cmd->width;
 	vfbs->base.base.height = mode_cmd->height;
@@ -719,7 +804,7 @@
 	cmd->body.format.bitsPerPixel = framebuffer->base.bits_per_pixel;
 	cmd->body.format.colorDepth = depth;
 	cmd->body.format.reserved = 0;
-	cmd->body.bytesPerLine = framebuffer->base.pitch;
+	cmd->body.bytesPerLine = framebuffer->base.pitches[0];
 	cmd->body.ptr.gmrId = framebuffer->user_handle;
 	cmd->body.ptr.offset = 0;
 
@@ -777,6 +862,7 @@
 			int clip_y1 = clips_ptr->y1 - unit->crtc.y;
 			int clip_x2 = clips_ptr->x2 - unit->crtc.x;
 			int clip_y2 = clips_ptr->y2 - unit->crtc.y;
+			int move_x, move_y;
 
 			/* skip any crtcs that misses the clip region */
 			if (clip_x1 >= unit->crtc.mode.hdisplay ||
@@ -784,12 +870,21 @@
 			    clip_x2 <= 0 || clip_y2 <= 0)
 				continue;
 
+			/* clip size to crtc size */
+			clip_x2 = min_t(int, clip_x2, unit->crtc.mode.hdisplay);
+			clip_y2 = min_t(int, clip_y2, unit->crtc.mode.vdisplay);
+
+			/* translate both src and dest to bring clip into screen */
+			move_x = min_t(int, clip_x1, 0);
+			move_y = min_t(int, clip_y1, 0);
+
+			/* actual translate done here */
 			blits[hit_num].header = SVGA_CMD_BLIT_GMRFB_TO_SCREEN;
 			blits[hit_num].body.destScreenId = unit->unit;
-			blits[hit_num].body.srcOrigin.x = clips_ptr->x1;
-			blits[hit_num].body.srcOrigin.y = clips_ptr->y1;
-			blits[hit_num].body.destRect.left = clip_x1;
-			blits[hit_num].body.destRect.top = clip_y1;
+			blits[hit_num].body.srcOrigin.x = clips_ptr->x1 - move_x;
+			blits[hit_num].body.srcOrigin.y = clips_ptr->y1 - move_y;
+			blits[hit_num].body.destRect.left = clip_x1 - move_x;
+			blits[hit_num].body.destRect.top = clip_y1 - move_y;
 			blits[hit_num].body.destRect.right = clip_x2;
 			blits[hit_num].body.destRect.bottom = clip_y2;
 			hit_num++;
@@ -961,7 +1056,7 @@
 	}
 
 	vfbd->base.base.bits_per_pixel = mode_cmd->bpp;
-	vfbd->base.base.pitch = mode_cmd->pitch;
+	vfbd->base.base.pitches[0] = mode_cmd->pitch;
 	vfbd->base.base.depth = mode_cmd->depth;
 	vfbd->base.base.width = mode_cmd->width;
 	vfbd->base.base.height = mode_cmd->height;
@@ -1040,42 +1135,29 @@
 	 * End conditioned code.
 	 */
 
-	ret = vmw_user_surface_lookup_handle(dev_priv, tfile,
-					     mode_cmd.handle, &surface);
+	/* returns either a dmabuf or surface */
+	ret = vmw_user_lookup_handle(dev_priv, tfile,
+				     mode_cmd.handle,
+				     &surface, &bo);
 	if (ret)
-		goto try_dmabuf;
+		goto err_out;
 
-	if (!surface->scanout)
-		goto err_not_scanout;
+	/* Create the new framebuffer depending one what we got back */
+	if (bo)
+		ret = vmw_kms_new_framebuffer_dmabuf(dev_priv, bo, &vfb,
+						     &mode_cmd);
+	else if (surface)
+		ret = vmw_kms_new_framebuffer_surface(dev_priv, file_priv,
+						      surface, &vfb, &mode_cmd);
+	else
+		BUG();
 
-	ret = vmw_kms_new_framebuffer_surface(dev_priv, file_priv, surface,
-					      &vfb, &mode_cmd);
-
-	/* vmw_user_surface_lookup takes one ref so does new_fb */
-	vmw_surface_unreference(&surface);
-
-	if (ret) {
-		DRM_ERROR("failed to create vmw_framebuffer: %i\n", ret);
-		ttm_base_object_unref(&user_obj);
-		return ERR_PTR(ret);
-	} else
-		vfb->user_obj = user_obj;
-	return &vfb->base;
-
-try_dmabuf:
-	DRM_INFO("%s: trying buffer\n", __func__);
-
-	ret = vmw_user_dmabuf_lookup(tfile, mode_cmd.handle, &bo);
-	if (ret) {
-		DRM_ERROR("failed to find buffer: %i\n", ret);
-		return ERR_PTR(-ENOENT);
-	}
-
-	ret = vmw_kms_new_framebuffer_dmabuf(dev_priv, bo, &vfb,
-					     &mode_cmd);
-
-	/* vmw_user_dmabuf_lookup takes one ref so does new_fb */
-	vmw_dmabuf_unreference(&bo);
+err_out:
+	/* vmw_user_lookup_handle takes one ref so does new_fb */
+	if (bo)
+		vmw_dmabuf_unreference(&bo);
+	if (surface)
+		vmw_surface_unreference(&surface);
 
 	if (ret) {
 		DRM_ERROR("failed to create vmw_framebuffer: %i\n", ret);
@@ -1085,14 +1167,6 @@
 		vfb->user_obj = user_obj;
 
 	return &vfb->base;
-
-err_not_scanout:
-	DRM_ERROR("surface not marked as scanout\n");
-	/* vmw_user_surface_lookup takes one ref */
-	vmw_surface_unreference(&surface);
-	ttm_base_object_unref(&user_obj);
-
-	return ERR_PTR(-EINVAL);
 }
 
 static struct drm_mode_config_funcs vmw_kms_funcs = {
@@ -1109,10 +1183,12 @@
 		    uint32_t num_clips)
 {
 	struct vmw_display_unit *units[VMWGFX_NUM_DISPLAY_UNITS];
+	struct drm_clip_rect *tmp;
 	struct drm_crtc *crtc;
 	size_t fifo_size;
 	int i, k, num_units;
 	int ret = 0; /* silence warning */
+	int left, right, top, bottom;
 
 	struct {
 		SVGA3dCmdHeader header;
@@ -1130,60 +1206,95 @@
 	BUG_ON(surface == NULL);
 	BUG_ON(!clips || !num_clips);
 
+	tmp = kzalloc(sizeof(*tmp) * num_clips, GFP_KERNEL);
+	if (unlikely(tmp == NULL)) {
+		DRM_ERROR("Temporary cliprect memory alloc failed.\n");
+		return -ENOMEM;
+	}
+
 	fifo_size = sizeof(*cmd) + sizeof(SVGASignedRect) * num_clips;
 	cmd = kmalloc(fifo_size, GFP_KERNEL);
 	if (unlikely(cmd == NULL)) {
 		DRM_ERROR("Failed to allocate temporary fifo memory.\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto out_free_tmp;
+	}
+
+	left = clips->x;
+	right = clips->x + clips->w;
+	top = clips->y;
+	bottom = clips->y + clips->h;
+
+	for (i = 1; i < num_clips; i++) {
+		left = min_t(int, left, (int)clips[i].x);
+		right = max_t(int, right, (int)clips[i].x + clips[i].w);
+		top = min_t(int, top, (int)clips[i].y);
+		bottom = max_t(int, bottom, (int)clips[i].y + clips[i].h);
 	}
 
 	/* only need to do this once */
 	memset(cmd, 0, fifo_size);
 	cmd->header.id = cpu_to_le32(SVGA_3D_CMD_BLIT_SURFACE_TO_SCREEN);
-	cmd->header.size = cpu_to_le32(fifo_size - sizeof(cmd->header));
-
-	cmd->body.srcRect.left = 0;
-	cmd->body.srcRect.right = surface->sizes[0].width;
-	cmd->body.srcRect.top = 0;
-	cmd->body.srcRect.bottom = surface->sizes[0].height;
 
 	blits = (SVGASignedRect *)&cmd[1];
+
+	cmd->body.srcRect.left = left;
+	cmd->body.srcRect.right = right;
+	cmd->body.srcRect.top = top;
+	cmd->body.srcRect.bottom = bottom;
+
 	for (i = 0; i < num_clips; i++) {
-		blits[i].left   = clips[i].x;
-		blits[i].right  = clips[i].x + clips[i].w;
-		blits[i].top    = clips[i].y;
-		blits[i].bottom = clips[i].y + clips[i].h;
+		tmp[i].x1 = clips[i].x - left;
+		tmp[i].x2 = clips[i].x + clips[i].w - left;
+		tmp[i].y1 = clips[i].y - top;
+		tmp[i].y2 = clips[i].y + clips[i].h - top;
 	}
 
 	for (k = 0; k < num_units; k++) {
 		struct vmw_display_unit *unit = units[k];
-		int clip_x1 = destX - unit->crtc.x;
-		int clip_y1 = destY - unit->crtc.y;
-		int clip_x2 = clip_x1 + surface->sizes[0].width;
-		int clip_y2 = clip_y1 + surface->sizes[0].height;
+		struct vmw_clip_rect clip;
+		int num;
+
+		clip.x1 = left + destX - unit->crtc.x;
+		clip.y1 = top + destY - unit->crtc.y;
+		clip.x2 = right + destX - unit->crtc.x;
+		clip.y2 = bottom + destY - unit->crtc.y;
 
 		/* skip any crtcs that misses the clip region */
-		if (clip_x1 >= unit->crtc.mode.hdisplay ||
-		    clip_y1 >= unit->crtc.mode.vdisplay ||
-		    clip_x2 <= 0 || clip_y2 <= 0)
+		if (clip.x1 >= unit->crtc.mode.hdisplay ||
+		    clip.y1 >= unit->crtc.mode.vdisplay ||
+		    clip.x2 <= 0 || clip.y2 <= 0)
 			continue;
 
+		/*
+		 * In order for the clip rects to be correctly scaled
+		 * the src and dest rects needs to be the same size.
+		 */
+		cmd->body.destRect.left = clip.x1;
+		cmd->body.destRect.right = clip.x2;
+		cmd->body.destRect.top = clip.y1;
+		cmd->body.destRect.bottom = clip.y2;
+
+		/* create a clip rect of the crtc in dest coords */
+		clip.x2 = unit->crtc.mode.hdisplay - clip.x1;
+		clip.y2 = unit->crtc.mode.vdisplay - clip.y1;
+		clip.x1 = 0 - clip.x1;
+		clip.y1 = 0 - clip.y1;
+
 		/* need to reset sid as it is changed by execbuf */
 		cmd->body.srcImage.sid = sid;
-
 		cmd->body.destScreenId = unit->unit;
 
-		/*
-		 * The blit command is a lot more resilient then the
-		 * readback command when it comes to clip rects. So its
-		 * okay to go out of bounds.
-		 */
+		/* clip and write blits to cmd stream */
+		vmw_clip_cliprects(tmp, num_clips, clip, blits, &num);
 
-		cmd->body.destRect.left = clip_x1;
-		cmd->body.destRect.right = clip_x2;
-		cmd->body.destRect.top = clip_y1;
-		cmd->body.destRect.bottom = clip_y2;
+		/* if no cliprects hit skip this */
+		if (num == 0)
+			continue;
 
+		/* recalculate package length */
+		fifo_size = sizeof(*cmd) + sizeof(SVGASignedRect) * num;
+		cmd->header.size = cpu_to_le32(fifo_size - sizeof(cmd->header));
 		ret = vmw_execbuf_process(file_priv, dev_priv, NULL, cmd,
 					  fifo_size, 0, NULL);
 
@@ -1192,6 +1303,8 @@
 	}
 
 	kfree(cmd);
+out_free_tmp:
+	kfree(tmp);
 
 	return ret;
 }
@@ -1243,7 +1356,7 @@
 	cmd->body.format.bitsPerPixel = vfb->base.bits_per_pixel;
 	cmd->body.format.colorDepth = vfb->base.depth;
 	cmd->body.format.reserved = 0;
-	cmd->body.bytesPerLine = vfb->base.pitch;
+	cmd->body.bytesPerLine = vfb->base.pitches[0];
 	cmd->body.ptr.gmrId = vfb->user_handle;
 	cmd->body.ptr.offset = 0;
 
@@ -1331,7 +1444,10 @@
 	 * drm_encoder_cleanup which takes the lock we deadlock.
 	 */
 	drm_mode_config_cleanup(dev_priv->dev);
-	vmw_kms_close_legacy_display_system(dev_priv);
+	if (dev_priv->sou_priv)
+		vmw_kms_close_screen_object_display(dev_priv);
+	else
+		vmw_kms_close_legacy_display_system(dev_priv);
 	return 0;
 }
 
@@ -1809,7 +1925,8 @@
 	}
 
 	rects_size = arg->num_outputs * sizeof(struct drm_vmw_rect);
-	rects = kzalloc(rects_size, GFP_KERNEL);
+	rects = kcalloc(arg->num_outputs, sizeof(struct drm_vmw_rect),
+			GFP_KERNEL);
 	if (unlikely(!rects)) {
 		ret = -ENOMEM;
 		goto out_unlock;
@@ -1824,10 +1941,10 @@
 	}
 
 	for (i = 0; i < arg->num_outputs; ++i) {
-		if (rects->x < 0 ||
-		    rects->y < 0 ||
-		    rects->x + rects->w > mode_config->max_width ||
-		    rects->y + rects->h > mode_config->max_height) {
+		if (rects[i].x < 0 ||
+		    rects[i].y < 0 ||
+		    rects[i].x + rects[i].w > mode_config->max_width ||
+		    rects[i].y + rects[i].h > mode_config->max_height) {
 			DRM_ERROR("Invalid GUI layout.\n");
 			ret = -EINVAL;
 			goto out_free;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
index 055b844..a4f7f03 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
@@ -63,9 +63,14 @@
 int vmw_cursor_update_image(struct vmw_private *dev_priv,
 			    u32 *image, u32 width, u32 height,
 			    u32 hotspotX, u32 hotspotY);
+int vmw_cursor_update_dmabuf(struct vmw_private *dev_priv,
+			     struct vmw_dma_buffer *dmabuf,
+			     u32 width, u32 height,
+			     u32 hotspotX, u32 hotspotY);
 void vmw_cursor_update_position(struct vmw_private *dev_priv,
 				bool show, int x, int y);
 
+
 /**
  * Base class display unit.
  *
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
index 90c5e39..f77b184 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
@@ -74,9 +74,10 @@
 {
 	struct vmw_legacy_display *lds = dev_priv->ldu_priv;
 	struct vmw_legacy_display_unit *entry;
+	struct vmw_display_unit *du = NULL;
 	struct drm_framebuffer *fb = NULL;
 	struct drm_crtc *crtc = NULL;
-	int i = 0;
+	int i = 0, ret;
 
 	/* If there is no display topology the host just assumes
 	 * that the guest will set the same layout as the host.
@@ -94,7 +95,7 @@
 			return 0;
 		fb = entry->base.crtc.fb;
 
-		return vmw_kms_write_svga(dev_priv, w, h, fb->pitch,
+		return vmw_kms_write_svga(dev_priv, w, h, fb->pitches[0],
 					  fb->bits_per_pixel, fb->depth);
 	}
 
@@ -102,7 +103,7 @@
 		entry = list_entry(lds->active.next, typeof(*entry), active);
 		fb = entry->base.crtc.fb;
 
-		vmw_kms_write_svga(dev_priv, fb->width, fb->height, fb->pitch,
+		vmw_kms_write_svga(dev_priv, fb->width, fb->height, fb->pitches[0],
 				   fb->bits_per_pixel, fb->depth);
 	}
 
@@ -129,6 +130,25 @@
 
 	lds->last_num_active = lds->num_active;
 
+
+	/* Find the first du with a cursor. */
+	list_for_each_entry(entry, &lds->active, active) {
+		du = &entry->base;
+
+		if (!du->cursor_dmabuf)
+			continue;
+
+		ret = vmw_cursor_update_dmabuf(dev_priv,
+					       du->cursor_dmabuf,
+					       64, 64,
+					       du->hotspot_x,
+					       du->hotspot_y);
+		if (ret == 0)
+			break;
+
+		DRM_ERROR("Could not update cursor image\n");
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index 2eb84a5..a37abb5 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -1190,6 +1190,29 @@
 		write_unlock(lock);
 }
 
+/**
+ * Helper function that looks either a surface or dmabuf.
+ *
+ * The pointer this pointed at by out_surf and out_buf needs to be null.
+ */
+int vmw_user_lookup_handle(struct vmw_private *dev_priv,
+			   struct ttm_object_file *tfile,
+			   uint32_t handle,
+			   struct vmw_surface **out_surf,
+			   struct vmw_dma_buffer **out_buf)
+{
+	int ret;
+
+	BUG_ON(*out_surf || *out_buf);
+
+	ret = vmw_user_surface_lookup_handle(dev_priv, tfile, handle, out_surf);
+	if (!ret)
+		return 0;
+
+	ret = vmw_user_dmabuf_lookup(tfile, handle, out_buf);
+	return ret;
+}
+
 
 int vmw_user_surface_lookup_handle(struct vmw_private *dev_priv,
 				   struct ttm_object_file *tfile,
diff --git a/drivers/gpu/vga/vgaarb.c b/drivers/gpu/vga/vgaarb.c
index c72f1c0..111d956 100644
--- a/drivers/gpu/vga/vgaarb.c
+++ b/drivers/gpu/vga/vgaarb.c
@@ -465,31 +465,29 @@
 	while (new_bus) {
 		new_bridge = new_bus->self;
 
-		if (new_bridge) {
-			/* go through list of devices already registered */
-			list_for_each_entry(same_bridge_vgadev, &vga_list, list) {
-				bus = same_bridge_vgadev->pdev->bus;
+		/* go through list of devices already registered */
+		list_for_each_entry(same_bridge_vgadev, &vga_list, list) {
+			bus = same_bridge_vgadev->pdev->bus;
+			bridge = bus->self;
+
+			/* see if the share a bridge with this device */
+			if (new_bridge == bridge) {
+				/* if their direct parent bridge is the same
+				   as any bridge of this device then it can't be used
+				   for that device */
+				same_bridge_vgadev->bridge_has_one_vga = false;
+			}
+
+			/* now iterate the previous devices bridge hierarchy */
+			/* if the new devices parent bridge is in the other devices
+			   hierarchy then we can't use it to control this device */
+			while (bus) {
 				bridge = bus->self;
-
-				/* see if the share a bridge with this device */
-				if (new_bridge == bridge) {
-					/* if their direct parent bridge is the same
-					   as any bridge of this device then it can't be used
-					   for that device */
-					same_bridge_vgadev->bridge_has_one_vga = false;
+				if (bridge) {
+					if (bridge == vgadev->pdev->bus->self)
+						vgadev->bridge_has_one_vga = false;
 				}
-
-				/* now iterate the previous devices bridge hierarchy */
-				/* if the new devices parent bridge is in the other devices
-				   hierarchy then we can't use it to control this device */
-				while (bus) {
-					bridge = bus->self;
-					if (bridge) {
-						if (bridge == vgadev->pdev->bus->self)
-							vgadev->bridge_has_one_vga = false;
-					}
-					bus = bus->parent;
-				}
+				bus = bus->parent;
 			}
 		}
 		new_bus = new_bus->parent;
@@ -993,14 +991,20 @@
 				uc = &priv->cards[i];
 		}
 
-		if (!uc)
-			return -EINVAL;
+		if (!uc) {
+			ret_val = -EINVAL;
+			goto done;
+		}
 
-		if (io_state & VGA_RSRC_LEGACY_IO && uc->io_cnt == 0)
-			return -EINVAL;
+		if (io_state & VGA_RSRC_LEGACY_IO && uc->io_cnt == 0) {
+			ret_val = -EINVAL;
+			goto done;
+		}
 
-		if (io_state & VGA_RSRC_LEGACY_MEM && uc->mem_cnt == 0)
-			return -EINVAL;
+		if (io_state & VGA_RSRC_LEGACY_MEM && uc->mem_cnt == 0) {
+			ret_val = -EINVAL;
+			goto done;
+		}
 
 		vga_put(pdev, io_state);
 
diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c
index 848a56c..af35384 100644
--- a/drivers/hid/hid-core.c
+++ b/drivers/hid/hid-core.c
@@ -1771,8 +1771,8 @@
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ESSENTIAL_REALITY, USB_DEVICE_ID_ESSENTIAL_REALITY_P5) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ETT, USB_DEVICE_ID_TC5UH) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ETT, USB_DEVICE_ID_TC4UM) },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH, 0x0001) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH, 0x0002) },
-	{ HID_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH, 0x0003) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_GENERAL_TOUCH, 0x0004) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_GLAB, USB_DEVICE_ID_4_PHIDGETSERVO_30) },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_GLAB, USB_DEVICE_ID_1_PHIDGETSERVO_30) },
diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index 06ce996..4a441a6 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h
@@ -266,7 +266,7 @@
 #define USB_DEVICE_ID_GAMERON_DUAL_PCS_ADAPTOR	0x0002
 
 #define USB_VENDOR_ID_GENERAL_TOUCH	0x0dfc
-#define USB_DEVICE_ID_GENERAL_TOUCH_WIN7_TWOFINGERS 0x0001
+#define USB_DEVICE_ID_GENERAL_TOUCH_WIN7_TWOFINGERS 0x0003
 
 #define USB_VENDOR_ID_GLAB		0x06c2
 #define USB_DEVICE_ID_4_PHIDGETSERVO_30	0x0038
diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index 9ec854a..91be41f 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -315,7 +315,7 @@
 
 config SENSORS_EXYNOS4_TMU
 	tristate "Temperature sensor on Samsung EXYNOS4"
-	depends on EXYNOS4_DEV_TMU
+	depends on ARCH_EXYNOS4
 	help
 	  If you say yes here you get support for TMU (Thermal Managment
 	  Unit) on SAMSUNG EXYNOS4 series of SoC.
diff --git a/drivers/hwmon/ad7314.c b/drivers/hwmon/ad7314.c
index 318e38e..5d760f3 100644
--- a/drivers/hwmon/ad7314.c
+++ b/drivers/hwmon/ad7314.c
@@ -160,7 +160,6 @@
 static struct spi_driver ad7314_driver = {
 	.driver = {
 		.name = "ad7314",
-		.bus = &spi_bus_type,
 		.owner = THIS_MODULE,
 	},
 	.probe = ad7314_probe,
diff --git a/drivers/hwmon/ads7871.c b/drivers/hwmon/ads7871.c
index 5231934..04450f8 100644
--- a/drivers/hwmon/ads7871.c
+++ b/drivers/hwmon/ads7871.c
@@ -227,7 +227,6 @@
 static struct spi_driver ads7871_driver = {
 	.driver = {
 		.name = DEVICE_NAME,
-		.bus = &spi_bus_type,
 		.owner = THIS_MODULE,
 	},
 
diff --git a/drivers/hwmon/exynos4_tmu.c b/drivers/hwmon/exynos4_tmu.c
index faa0884..f2359a0 100644
--- a/drivers/hwmon/exynos4_tmu.c
+++ b/drivers/hwmon/exynos4_tmu.c
@@ -506,17 +506,7 @@
 	.resume = exynos4_tmu_resume,
 };
 
-static int __init exynos4_tmu_driver_init(void)
-{
-	return platform_driver_register(&exynos4_tmu_driver);
-}
-module_init(exynos4_tmu_driver_init);
-
-static void __exit exynos4_tmu_driver_exit(void)
-{
-	platform_driver_unregister(&exynos4_tmu_driver);
-}
-module_exit(exynos4_tmu_driver_exit);
+module_platform_driver(exynos4_tmu_driver);
 
 MODULE_DESCRIPTION("EXYNOS4 TMU Driver");
 MODULE_AUTHOR("Donggeun Kim <dg77.kim@samsung.com>");
diff --git a/drivers/hwmon/gpio-fan.c b/drivers/hwmon/gpio-fan.c
index 89aa9fb..9ba38f3 100644
--- a/drivers/hwmon/gpio-fan.c
+++ b/drivers/hwmon/gpio-fan.c
@@ -539,18 +539,7 @@
 	},
 };
 
-static int __init gpio_fan_init(void)
-{
-	return platform_driver_register(&gpio_fan_driver);
-}
-
-static void __exit gpio_fan_exit(void)
-{
-	platform_driver_unregister(&gpio_fan_driver);
-}
-
-module_init(gpio_fan_init);
-module_exit(gpio_fan_exit);
+module_platform_driver(gpio_fan_driver);
 
 MODULE_AUTHOR("Simon Guinot <sguinot@lacie.com>");
 MODULE_DESCRIPTION("GPIO FAN driver");
diff --git a/drivers/hwmon/jz4740-hwmon.c b/drivers/hwmon/jz4740-hwmon.c
index fea292d..5253d23 100644
--- a/drivers/hwmon/jz4740-hwmon.c
+++ b/drivers/hwmon/jz4740-hwmon.c
@@ -59,7 +59,7 @@
 {
 	struct jz4740_hwmon *hwmon = dev_get_drvdata(dev);
 	struct completion *completion = &hwmon->read_completion;
-	unsigned long t;
+	long t;
 	unsigned long val;
 	int ret;
 
@@ -203,7 +203,7 @@
 	return 0;
 }
 
-struct platform_driver jz4740_hwmon_driver = {
+static struct platform_driver jz4740_hwmon_driver = {
 	.probe	= jz4740_hwmon_probe,
 	.remove = __devexit_p(jz4740_hwmon_remove),
 	.driver = {
@@ -212,17 +212,7 @@
 	},
 };
 
-static int __init jz4740_hwmon_init(void)
-{
-	return platform_driver_register(&jz4740_hwmon_driver);
-}
-module_init(jz4740_hwmon_init);
-
-static void __exit jz4740_hwmon_exit(void)
-{
-	platform_driver_unregister(&jz4740_hwmon_driver);
-}
-module_exit(jz4740_hwmon_exit);
+module_platform_driver(jz4740_hwmon_driver);
 
 MODULE_DESCRIPTION("JZ4740 SoC HWMON driver");
 MODULE_AUTHOR("Lars-Peter Clausen <lars@metafoo.de>");
diff --git a/drivers/hwmon/ntc_thermistor.c b/drivers/hwmon/ntc_thermistor.c
index eab1161..9b382ec 100644
--- a/drivers/hwmon/ntc_thermistor.c
+++ b/drivers/hwmon/ntc_thermistor.c
@@ -432,19 +432,7 @@
 	.id_table = ntc_thermistor_id,
 };
 
-static int __init ntc_thermistor_init(void)
-{
-	return platform_driver_register(&ntc_thermistor_driver);
-}
-
-module_init(ntc_thermistor_init);
-
-static void __exit ntc_thermistor_cleanup(void)
-{
-	platform_driver_unregister(&ntc_thermistor_driver);
-}
-
-module_exit(ntc_thermistor_cleanup);
+module_platform_driver(ntc_thermistor_driver);
 
 MODULE_DESCRIPTION("NTC Thermistor Driver");
 MODULE_AUTHOR("MyungJoo Ham <myungjoo.ham@samsung.com>");
diff --git a/drivers/hwmon/s3c-hwmon.c b/drivers/hwmon/s3c-hwmon.c
index b39f52e..f6c26d1 100644
--- a/drivers/hwmon/s3c-hwmon.c
+++ b/drivers/hwmon/s3c-hwmon.c
@@ -393,18 +393,7 @@
 	.remove		= __devexit_p(s3c_hwmon_remove),
 };
 
-static int __init s3c_hwmon_init(void)
-{
-	return platform_driver_register(&s3c_hwmon_driver);
-}
-
-static void __exit s3c_hwmon_exit(void)
-{
-	platform_driver_unregister(&s3c_hwmon_driver);
-}
-
-module_init(s3c_hwmon_init);
-module_exit(s3c_hwmon_exit);
+module_platform_driver(s3c_hwmon_driver);
 
 MODULE_AUTHOR("Ben Dooks <ben@simtec.co.uk>");
 MODULE_DESCRIPTION("S3C ADC HWMon driver");
diff --git a/drivers/hwmon/sch5627.c b/drivers/hwmon/sch5627.c
index e3b5c60..79b6dab 100644
--- a/drivers/hwmon/sch5627.c
+++ b/drivers/hwmon/sch5627.c
@@ -590,19 +590,8 @@
 	.remove		= sch5627_remove,
 };
 
-static int __init sch5627_init(void)
-{
-	return platform_driver_register(&sch5627_driver);
-}
-
-static void __exit sch5627_exit(void)
-{
-	platform_driver_unregister(&sch5627_driver);
-}
+module_platform_driver(sch5627_driver);
 
 MODULE_DESCRIPTION("SMSC SCH5627 Hardware Monitoring Driver");
 MODULE_AUTHOR("Hans de Goede <hdegoede@redhat.com>");
 MODULE_LICENSE("GPL");
-
-module_init(sch5627_init);
-module_exit(sch5627_exit);
diff --git a/drivers/hwmon/sch5636.c b/drivers/hwmon/sch5636.c
index 244407a..9d5236f 100644
--- a/drivers/hwmon/sch5636.c
+++ b/drivers/hwmon/sch5636.c
@@ -521,19 +521,8 @@
 	.remove		= sch5636_remove,
 };
 
-static int __init sch5636_init(void)
-{
-	return platform_driver_register(&sch5636_driver);
-}
-
-static void __exit sch5636_exit(void)
-{
-	platform_driver_unregister(&sch5636_driver);
-}
+module_platform_driver(sch5636_driver);
 
 MODULE_DESCRIPTION("SMSC SCH5636 Hardware Monitoring Driver");
 MODULE_AUTHOR("Hans de Goede <hdegoede@redhat.com>");
 MODULE_LICENSE("GPL");
-
-module_init(sch5636_init);
-module_exit(sch5636_exit);
diff --git a/drivers/hwmon/twl4030-madc-hwmon.c b/drivers/hwmon/twl4030-madc-hwmon.c
index 5724074..0018c7d 100644
--- a/drivers/hwmon/twl4030-madc-hwmon.c
+++ b/drivers/hwmon/twl4030-madc-hwmon.c
@@ -136,19 +136,7 @@
 		   },
 };
 
-static int __init twl4030_madc_hwmon_init(void)
-{
-	return platform_driver_register(&twl4030_madc_hwmon_driver);
-}
-
-module_init(twl4030_madc_hwmon_init);
-
-static void __exit twl4030_madc_hwmon_exit(void)
-{
-	platform_driver_unregister(&twl4030_madc_hwmon_driver);
-}
-
-module_exit(twl4030_madc_hwmon_exit);
+module_platform_driver(twl4030_madc_hwmon_driver);
 
 MODULE_DESCRIPTION("TWL4030 ADC Hwmon driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/hwmon/ultra45_env.c b/drivers/hwmon/ultra45_env.c
index 3cd07bf..b9a87e8 100644
--- a/drivers/hwmon/ultra45_env.c
+++ b/drivers/hwmon/ultra45_env.c
@@ -309,15 +309,4 @@
 	.remove		= __devexit_p(env_remove),
 };
 
-static int __init env_init(void)
-{
-	return platform_driver_register(&env_driver);
-}
-
-static void __exit env_exit(void)
-{
-	platform_driver_unregister(&env_driver);
-}
-
-module_init(env_init);
-module_exit(env_exit);
+module_platform_driver(env_driver);
diff --git a/drivers/hwmon/wm831x-hwmon.c b/drivers/hwmon/wm831x-hwmon.c
index 97b1f83..9b598ed 100644
--- a/drivers/hwmon/wm831x-hwmon.c
+++ b/drivers/hwmon/wm831x-hwmon.c
@@ -209,17 +209,7 @@
 	},
 };
 
-static int __init wm831x_hwmon_init(void)
-{
-	return platform_driver_register(&wm831x_hwmon_driver);
-}
-module_init(wm831x_hwmon_init);
-
-static void __exit wm831x_hwmon_exit(void)
-{
-	platform_driver_unregister(&wm831x_hwmon_driver);
-}
-module_exit(wm831x_hwmon_exit);
+module_platform_driver(wm831x_hwmon_driver);
 
 MODULE_AUTHOR("Mark Brown <broonie@opensource.wolfsonmicro.com>");
 MODULE_DESCRIPTION("WM831x Hardware Monitoring");
diff --git a/drivers/hwmon/wm8350-hwmon.c b/drivers/hwmon/wm8350-hwmon.c
index 1329059..3ff67ed 100644
--- a/drivers/hwmon/wm8350-hwmon.c
+++ b/drivers/hwmon/wm8350-hwmon.c
@@ -133,17 +133,7 @@
 	},
 };
 
-static int __init wm8350_hwmon_init(void)
-{
-	return platform_driver_register(&wm8350_hwmon_driver);
-}
-module_init(wm8350_hwmon_init);
-
-static void __exit wm8350_hwmon_exit(void)
-{
-	platform_driver_unregister(&wm8350_hwmon_driver);
-}
-module_exit(wm8350_hwmon_exit);
+module_platform_driver(wm8350_hwmon_driver);
 
 MODULE_AUTHOR("Mark Brown <broonie@opensource.wolfsonmicro.com>");
 MODULE_DESCRIPTION("WM8350 Hardware Monitoring");
diff --git a/drivers/hwspinlock/u8500_hsem.c b/drivers/hwspinlock/u8500_hsem.c
index 143461a..86980fe 100644
--- a/drivers/hwspinlock/u8500_hsem.c
+++ b/drivers/hwspinlock/u8500_hsem.c
@@ -21,6 +21,7 @@
  * General Public License for more details.
  */
 
+#include <linux/module.h>
 #include <linux/delay.h>
 #include <linux/io.h>
 #include <linux/pm_runtime.h>
@@ -108,10 +109,8 @@
 		return -ENODEV;
 
 	io_base = ioremap(res->start, resource_size(res));
-	if (!io_base) {
-		ret = -ENOMEM;
-		goto free_state;
-	}
+	if (!io_base)
+		return -ENOMEM;
 
 	/* make sure protocol 1 is selected */
 	val = readl(io_base + HSEM_CTRL_REG);
diff --git a/drivers/i2c/algos/i2c-algo-bit.c b/drivers/i2c/algos/i2c-algo-bit.c
index 85584a5..525c734 100644
--- a/drivers/i2c/algos/i2c-algo-bit.c
+++ b/drivers/i2c/algos/i2c-algo-bit.c
@@ -488,7 +488,7 @@
 
 	if (flags & I2C_M_TEN) {
 		/* a ten bit address */
-		addr = 0xf0 | ((msg->addr >> 7) & 0x03);
+		addr = 0xf0 | ((msg->addr >> 7) & 0x06);
 		bit_dbg(2, &i2c_adap->dev, "addr0: %d\n", addr);
 		/* try extended address code...*/
 		ret = try_address(i2c_adap, addr, retries);
@@ -498,7 +498,7 @@
 			return -ENXIO;
 		}
 		/* the remaining 8 bit address */
-		ret = i2c_outb(i2c_adap, msg->addr & 0x7f);
+		ret = i2c_outb(i2c_adap, msg->addr & 0xff);
 		if ((ret != 1) && !nak_ok) {
 			/* the chip did not ack / xmission error occurred */
 			dev_err(&i2c_adap->dev, "died at 2nd address code\n");
diff --git a/drivers/i2c/busses/i2c-eg20t.c b/drivers/i2c/busses/i2c-eg20t.c
index 8cebef4..18936ac 100644
--- a/drivers/i2c/busses/i2c-eg20t.c
+++ b/drivers/i2c/busses/i2c-eg20t.c
@@ -893,6 +893,13 @@
 	/* Set the number of I2C channel instance */
 	adap_info->ch_num = id->driver_data;
 
+	ret = request_irq(pdev->irq, pch_i2c_handler, IRQF_SHARED,
+		  KBUILD_MODNAME, adap_info);
+	if (ret) {
+		pch_pci_err(pdev, "request_irq FAILED\n");
+		goto err_request_irq;
+	}
+
 	for (i = 0; i < adap_info->ch_num; i++) {
 		pch_adap = &adap_info->pch_data[i].pch_adapter;
 		adap_info->pch_i2c_suspended = false;
@@ -910,28 +917,23 @@
 
 		pch_adap->dev.parent = &pdev->dev;
 
+		pch_i2c_init(&adap_info->pch_data[i]);
 		ret = i2c_add_adapter(pch_adap);
 		if (ret) {
 			pch_pci_err(pdev, "i2c_add_adapter[ch:%d] FAILED\n", i);
-			goto err_i2c_add_adapter;
+			goto err_add_adapter;
 		}
-
-		pch_i2c_init(&adap_info->pch_data[i]);
-	}
-	ret = request_irq(pdev->irq, pch_i2c_handler, IRQF_SHARED,
-		  KBUILD_MODNAME, adap_info);
-	if (ret) {
-		pch_pci_err(pdev, "request_irq FAILED\n");
-		goto err_i2c_add_adapter;
 	}
 
 	pci_set_drvdata(pdev, adap_info);
 	pch_pci_dbg(pdev, "returns %d.\n", ret);
 	return 0;
 
-err_i2c_add_adapter:
+err_add_adapter:
 	for (j = 0; j < i; j++)
 		i2c_del_adapter(&adap_info->pch_data[j].pch_adapter);
+	free_irq(pdev->irq, adap_info);
+err_request_irq:
 	pci_iounmap(pdev, base_addr);
 err_pci_iomap:
 	pci_release_regions(pdev);
diff --git a/drivers/i2c/busses/i2c-nuc900.c b/drivers/i2c/busses/i2c-nuc900.c
index 835e47b..03b6157 100644
--- a/drivers/i2c/busses/i2c-nuc900.c
+++ b/drivers/i2c/busses/i2c-nuc900.c
@@ -593,7 +593,7 @@
 	i2c->adap.algo_data = i2c;
 	i2c->adap.dev.parent = &pdev->dev;
 
-	mfp_set_groupg(&pdev->dev);
+	mfp_set_groupg(&pdev->dev, NULL);
 
 	clk_get_rate(i2c->clk);
 
diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c
index a43d002..fa23faa 100644
--- a/drivers/i2c/busses/i2c-omap.c
+++ b/drivers/i2c/busses/i2c-omap.c
@@ -1047,13 +1047,14 @@
 		 * size. This is to ensure that we can handle the status on int
 		 * call back latencies.
 		 */
-		if (dev->rev >= OMAP_I2C_REV_ON_3530_4430) {
-			dev->fifo_size = 0;
+
+		dev->fifo_size = (dev->fifo_size / 2);
+
+		if (dev->rev >= OMAP_I2C_REV_ON_3530_4430)
 			dev->b_hw = 0; /* Disable hardware fixes */
-		} else {
-			dev->fifo_size = (dev->fifo_size / 2);
+		else
 			dev->b_hw = 1; /* Enable hardware fixes */
-		}
+
 		/* calculate wakeup latency constraint for MPU */
 		if (dev->set_mpu_wkup_lat != NULL)
 			dev->latency = (1000000 * dev->fifo_size) /
diff --git a/drivers/i2c/busses/i2c-s3c2410.c b/drivers/i2c/busses/i2c-s3c2410.c
index 2754cef8..4c17180 100644
--- a/drivers/i2c/busses/i2c-s3c2410.c
+++ b/drivers/i2c/busses/i2c-s3c2410.c
@@ -534,6 +534,7 @@
 
 	/* first, try busy waiting briefly */
 	do {
+		cpu_relax();
 		iicstat = readl(i2c->regs + S3C2410_IICSTAT);
 	} while ((iicstat & S3C2410_IICSTAT_START) && --spins);
 
@@ -786,7 +787,7 @@
 #else
 static int s3c24xx_i2c_parse_dt_gpio(struct s3c24xx_i2c *i2c)
 {
-	return -EINVAL;
+	return 0;
 }
 
 static void s3c24xx_i2c_dt_gpio_free(struct s3c24xx_i2c *i2c)
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 131079a..1e56061 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -539,8 +539,10 @@
 	client->dev.type = &i2c_client_type;
 	client->dev.of_node = info->of_node;
 
+	/* For 10-bit clients, add an arbitrary offset to avoid collisions */
 	dev_set_name(&client->dev, "%d-%04x", i2c_adapter_id(adap),
-		     client->addr);
+		     client->addr | ((client->flags & I2C_CLIENT_TEN)
+				     ? 0xa000 : 0));
 	status = device_register(&client->dev);
 	if (status)
 		goto out_err;
diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index c90ce50..57a45ce8 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -579,7 +579,7 @@
 	return 0;
 }
 
-int i2cdev_notifier_call(struct notifier_block *nb, unsigned long action,
+static int i2cdev_notifier_call(struct notifier_block *nb, unsigned long action,
 			 void *data)
 {
 	struct device *dev = data;
diff --git a/drivers/ide/cy82c693.c b/drivers/ide/cy82c693.c
index 67cbcfa..847553f 100644
--- a/drivers/ide/cy82c693.c
+++ b/drivers/ide/cy82c693.c
@@ -1,7 +1,7 @@
 /*
  *  Copyright (C) 1998-2000 Andreas S. Krebs (akrebs@altavista.net), Maintainer
  *  Copyright (C) 1998-2002 Andre Hedrick <andre@linux-ide.org>, Integrator
- *  Copyright (C) 2007-2010 Bartlomiej Zolnierkiewicz
+ *  Copyright (C) 2007-2011 Bartlomiej Zolnierkiewicz
  *
  * CYPRESS CY82C693 chipset IDE controller
  *
@@ -90,7 +90,7 @@
 	u8 time_16, time_8;
 
 	/* select primary or secondary channel */
-	if (hwif->index > 0) {  /* drive is on the secondary channel */
+	if (drive->dn > 1) {  /* drive is on the secondary channel */
 		dev = pci_get_slot(dev->bus, dev->devfn+1);
 		if (!dev) {
 			printk(KERN_ERR "%s: tune_drive: "
@@ -141,7 +141,7 @@
 		pci_write_config_byte(dev, CY82_IDE_SLAVE_IOW, time_16);
 		pci_write_config_byte(dev, CY82_IDE_SLAVE_8BIT, time_8);
 	}
-	if (hwif->index > 0)
+	if (drive->dn > 1)
 		pci_dev_put(dev);
 }
 
diff --git a/drivers/ide/icside.c b/drivers/ide/icside.c
index 4a697a2..8716066 100644
--- a/drivers/ide/icside.c
+++ b/drivers/ide/icside.c
@@ -521,8 +521,8 @@
 	if (ec->dma != NO_DMA && !request_dma(ec->dma, DRV_NAME)) {
 		d.init_dma = icside_dma_init;
 		d.port_ops = &icside_v6_port_ops;
+	} else
 		d.dma_ops = NULL;
-	}
 
 	ret = ide_host_register(host, &d, hws);
 	if (ret)
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 04b0956..8126824 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -43,7 +43,6 @@
 /* For SCSI -> ATAPI command conversion */
 #include <scsi/scsi.h>
 
-#include <linux/irq.h>
 #include <linux/io.h>
 #include <asm/byteorder.h>
 #include <linux/uaccess.h>
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 61fdf544..3d42043 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -35,7 +35,6 @@
 #include <scsi/scsi_ioctl.h>
 
 #include <asm/byteorder.h>
-#include <linux/irq.h>
 #include <linux/uaccess.h>
 #include <linux/io.h>
 #include <asm/unaligned.h>
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 7ecb1ad..ce8237d 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -41,7 +41,6 @@
 #include <scsi/scsi.h>
 
 #include <asm/byteorder.h>
-#include <linux/irq.h>
 #include <linux/uaccess.h>
 #include <linux/io.h>
 #include <asm/unaligned.h>
diff --git a/drivers/ide/piix.c b/drivers/ide/piix.c
index b59d04c..1892e81 100644
--- a/drivers/ide/piix.c
+++ b/drivers/ide/piix.c
@@ -331,7 +331,7 @@
 		.udma_mask	= udma,			\
 	}
 
-#define DECLARE_ICH_DEV(udma) \
+#define DECLARE_ICH_DEV(mwdma, udma) \
 	{ \
 		.name		= DRV_NAME, \
 		.init_chipset	= init_chipset_ich, \
@@ -340,7 +340,7 @@
 		.port_ops	= &ich_port_ops, \
 		.pio_mask	= ATA_PIO4, \
 		.swdma_mask	= ATA_SWDMA2_ONLY, \
-		.mwdma_mask	= ATA_MWDMA12_ONLY, \
+		.mwdma_mask	= mwdma, \
 		.udma_mask	= udma, \
 	}
 
@@ -362,13 +362,15 @@
 	/* 2: PIIX4 */
 	DECLARE_PIIX_DEV(ATA_UDMA2),
 	/* 3: ICH0 */
-	DECLARE_ICH_DEV(ATA_UDMA2),
+	DECLARE_ICH_DEV(ATA_MWDMA12_ONLY, ATA_UDMA2),
 	/* 4: ICH */
-	DECLARE_ICH_DEV(ATA_UDMA4),
+	DECLARE_ICH_DEV(ATA_MWDMA12_ONLY, ATA_UDMA4),
 	/* 5: PIIX4 */
 	DECLARE_PIIX_DEV(ATA_UDMA4),
-	/* 6: ICH[2-7]/ICH[2-3]M/C-ICH/ICH5-SATA/ESB2/ICH8M */
-	DECLARE_ICH_DEV(ATA_UDMA5),
+	/* 6: ICH[2-6]/ICH[2-3]M/C-ICH/ICH5-SATA/ESB2/ICH8M */
+	DECLARE_ICH_DEV(ATA_MWDMA12_ONLY, ATA_UDMA5),
+	/* 7: ICH7/7-R, no MWDMA1 */
+	DECLARE_ICH_DEV(ATA_MWDMA2_ONLY, ATA_UDMA5),
 };
 
 /**
@@ -438,9 +440,9 @@
 #endif
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB_2),      6 },
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH6_19),    6 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH7_21),    6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH7_21),    7 },
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801DB_1),  6 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB2_18),    6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB2_18),    7 },
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH8_6),     6 },
 	{ 0, },
 };
diff --git a/drivers/ide/triflex.c b/drivers/ide/triflex.c
index e53a1b7..281c914 100644
--- a/drivers/ide/triflex.c
+++ b/drivers/ide/triflex.c
@@ -113,12 +113,26 @@
 };
 MODULE_DEVICE_TABLE(pci, triflex_pci_tbl);
 
+#ifdef CONFIG_PM
+static int triflex_ide_pci_suspend(struct pci_dev *dev, pm_message_t state)
+{
+	/*
+	 * We must not disable or powerdown the device.
+	 * APM bios refuses to suspend if IDE is not accessible.
+	 */
+	pci_save_state(dev);
+	return 0;
+}
+#else
+#define triflex_ide_pci_suspend NULL
+#endif
+
 static struct pci_driver triflex_pci_driver = {
 	.name		= "TRIFLEX_IDE",
 	.id_table	= triflex_pci_tbl,
 	.probe		= triflex_init_one,
 	.remove		= ide_pci_remove,
-	.suspend	= ide_pci_suspend,
+	.suspend	= triflex_ide_pci_suspend,
 	.resume		= ide_pci_resume,
 };
 
diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index 691276ba..e9cf51b 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -216,7 +216,9 @@
 
 	neigh = neigh_lookup(&arp_tbl, &rt->rt_gateway, rt->dst.dev);
 	if (!neigh || !(neigh->nud_state & NUD_VALID)) {
+		rcu_read_lock();
 		neigh_event_send(dst_get_neighbour(&rt->dst), NULL);
+		rcu_read_unlock();
 		ret = -ENODATA;
 		if (neigh)
 			goto release;
@@ -274,15 +276,16 @@
 		goto put;
 	}
 
+	rcu_read_lock();
 	neigh = dst_get_neighbour(dst);
 	if (!neigh || !(neigh->nud_state & NUD_VALID)) {
 		if (neigh)
 			neigh_event_send(neigh, NULL);
 		ret = -ENODATA;
-		goto put;
+	} else {
+		ret = rdma_copy_addr(addr, dst->dev, neigh->ha);
 	}
-
-	ret = rdma_copy_addr(addr, dst->dev, neigh->ha);
+	rcu_read_unlock();
 put:
 	dst_release(dst);
 	return ret;
diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c
index de6d077..c88b12b 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_cm.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c
@@ -1375,8 +1375,10 @@
 		goto reject;
 	}
 	dst = &rt->dst;
+	rcu_read_lock();
 	neigh = dst_get_neighbour(dst);
 	l2t = t3_l2t_get(tdev, neigh, neigh->dev);
+	rcu_read_unlock();
 	if (!l2t) {
 		printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n",
 		       __func__);
@@ -1946,10 +1948,12 @@
 	}
 	ep->dst = &rt->dst;
 
+	rcu_read_lock();
 	neigh = dst_get_neighbour(ep->dst);
 
 	/* get a l2t entry */
 	ep->l2t = t3_l2t_get(ep->com.tdev, neigh, neigh->dev);
+	rcu_read_unlock();
 	if (!ep->l2t) {
 		printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__);
 		err = -ENOMEM;
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index b36cdac..0747004 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -542,8 +542,10 @@
 		     (mpa_rev_to_use == 2 ? MPA_ENHANCED_RDMA_CONN : 0);
 	mpa->private_data_size = htons(ep->plen);
 	mpa->revision = mpa_rev_to_use;
-	if (mpa_rev_to_use == 1)
+	if (mpa_rev_to_use == 1) {
 		ep->tried_with_mpa_v1 = 1;
+		ep->retry_with_mpa_v1 = 0;
+	}
 
 	if (mpa_rev_to_use == 2) {
 		mpa->private_data_size +=
@@ -1594,6 +1596,7 @@
 		goto reject;
 	}
 	dst = &rt->dst;
+	rcu_read_lock();
 	neigh = dst_get_neighbour(dst);
 	if (neigh->dev->flags & IFF_LOOPBACK) {
 		pdev = ip_dev_find(&init_net, peer_ip);
@@ -1620,6 +1623,7 @@
 		rss_qid = dev->rdev.lldi.rxq_ids[
 			  cxgb4_port_idx(neigh->dev) * step];
 	}
+	rcu_read_unlock();
 	if (!l2t) {
 		printk(KERN_ERR MOD "%s - failed to allocate l2t entry!\n",
 		       __func__);
@@ -1820,6 +1824,7 @@
 	}
 	ep->dst = &rt->dst;
 
+	rcu_read_lock();
 	neigh = dst_get_neighbour(ep->dst);
 
 	/* get a l2t entry */
@@ -1856,6 +1861,7 @@
 		ep->rss_qid = ep->com.dev->rdev.lldi.rxq_ids[
 			cxgb4_port_idx(neigh->dev) * step];
 	}
+	rcu_read_unlock();
 	if (!ep->l2t) {
 		printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__);
 		err = -ENOMEM;
@@ -2301,6 +2307,7 @@
 	}
 	ep->dst = &rt->dst;
 
+	rcu_read_lock();
 	neigh = dst_get_neighbour(ep->dst);
 
 	/* get a l2t entry */
@@ -2339,6 +2346,7 @@
 		ep->retry_with_mpa_v1 = 0;
 		ep->tried_with_mpa_v1 = 0;
 	}
+	rcu_read_unlock();
 	if (!ep->l2t) {
 		printk(KERN_ERR MOD "%s - cannot alloc l2e.\n", __func__);
 		err = -ENOMEM;
diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c
index f35a935..0f1607c 100644
--- a/drivers/infiniband/hw/cxgb4/cq.c
+++ b/drivers/infiniband/hw/cxgb4/cq.c
@@ -311,7 +311,7 @@
 	while (ptr != cq->sw_pidx) {
 		cqe = &cq->sw_queue[ptr];
 		if (RQ_TYPE(cqe) && (CQE_OPCODE(cqe) != FW_RI_READ_RESP) &&
-		    (CQE_QPID(cqe) == wq->rq.qid) && cqe_completes_wr(cqe, wq))
+		    (CQE_QPID(cqe) == wq->sq.qid) && cqe_completes_wr(cqe, wq))
 			(*count)++;
 		if (++ptr == cq->size)
 			ptr = 0;
diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c
index dfce9ea..0a52d72 100644
--- a/drivers/infiniband/hw/nes/nes_cm.c
+++ b/drivers/infiniband/hw/nes/nes_cm.c
@@ -1377,9 +1377,11 @@
 		neigh_release(neigh);
 	}
 
-	if ((neigh == NULL) || (!(neigh->nud_state & NUD_VALID)))
+	if ((neigh == NULL) || (!(neigh->nud_state & NUD_VALID))) {
+		rcu_read_lock();
 		neigh_event_send(dst_get_neighbour(&rt->dst), NULL);
-
+		rcu_read_unlock();
+	}
 	ip_rt_put(rt);
 	return rc;
 }
diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c
index 5bd2162..1d58959 100644
--- a/drivers/infiniband/hw/qib/qib_iba7322.c
+++ b/drivers/infiniband/hw/qib/qib_iba7322.c
@@ -2307,19 +2307,11 @@
 		SYM_LSB(IBCCtrlA_0, MaxPktLen);
 	ppd->cpspec->ibcctrl_a = ibc; /* without linkcmd or linkinitcmd! */
 
-	/* initially come up waiting for TS1, without sending anything. */
-	val = ppd->cpspec->ibcctrl_a | (QLOGIC_IB_IBCC_LINKINITCMD_DISABLE <<
-		QLOGIC_IB_IBCC_LINKINITCMD_SHIFT);
-
-	ppd->cpspec->ibcctrl_a = val;
 	/*
 	 * Reset the PCS interface to the serdes (and also ibc, which is still
 	 * in reset from above).  Writes new value of ibcctrl_a as last step.
 	 */
 	qib_7322_mini_pcs_reset(ppd);
-	qib_write_kreg(dd, kr_scratch, 0ULL);
-	/* clear the linkinit cmds */
-	ppd->cpspec->ibcctrl_a &= ~SYM_MASK(IBCCtrlA_0, LinkInitCmd);
 
 	if (!ppd->cpspec->ibcctrl_b) {
 		unsigned lse = ppd->link_speed_enabled;
@@ -2385,6 +2377,14 @@
 	ppd->cpspec->ibcctrl_a |= SYM_MASK(IBCCtrlA_0, IBLinkEn);
 	set_vls(ppd);
 
+	/* initially come up DISABLED, without sending anything. */
+	val = ppd->cpspec->ibcctrl_a | (QLOGIC_IB_IBCC_LINKINITCMD_DISABLE <<
+					QLOGIC_IB_IBCC_LINKINITCMD_SHIFT);
+	qib_write_kreg_port(ppd, krp_ibcctrl_a, val);
+	qib_write_kreg(dd, kr_scratch, 0ULL);
+	/* clear the linkinit cmds */
+	ppd->cpspec->ibcctrl_a = val & ~SYM_MASK(IBCCtrlA_0, LinkInitCmd);
+
 	/* be paranoid against later code motion, etc. */
 	spin_lock_irqsave(&dd->cspec->rcvmod_lock, flags);
 	ppd->p_rcvctrl |= SYM_MASK(RcvCtrl_0, RcvIBPortEnable);
@@ -5241,7 +5241,7 @@
 			   off */
 			if (ppd->dd->flags & QIB_HAS_QSFP) {
 				qd->t_insert = get_jiffies_64();
-				schedule_work(&qd->work);
+				queue_work(ib_wq, &qd->work);
 			}
 			spin_lock_irqsave(&ppd->sdma_lock, flags);
 			if (__qib_sdma_running(ppd))
diff --git a/drivers/infiniband/hw/qib/qib_qsfp.c b/drivers/infiniband/hw/qib/qib_qsfp.c
index e06c4ed..fa71b1e 100644
--- a/drivers/infiniband/hw/qib/qib_qsfp.c
+++ b/drivers/infiniband/hw/qib/qib_qsfp.c
@@ -480,18 +480,6 @@
 	udelay(20); /* Generous RST dwell */
 
 	dd->f_gpio_mod(dd, mask, mask, mask);
-	/* Spec says module can take up to two seconds! */
-	mask = QSFP_GPIO_MOD_PRS_N;
-	if (qd->ppd->hw_pidx)
-		mask <<= QSFP_GPIO_PORT2_SHIFT;
-
-	/* Do not try to wait here. Better to let event handle it */
-	if (!qib_qsfp_mod_present(qd->ppd))
-		goto bail;
-	/* We see a module, but it may be unwise to look yet. Just schedule */
-	qd->t_insert = get_jiffies_64();
-	queue_work(ib_wq, &qd->work);
-bail:
 	return;
 }
 
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 0ef9af9..4115be54 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -57,21 +57,24 @@
 				 struct ib_pd *pd, struct ib_ah_attr *attr)
 {
 	struct ipoib_ah *ah;
+	struct ib_ah *vah;
 
 	ah = kmalloc(sizeof *ah, GFP_KERNEL);
 	if (!ah)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	ah->dev       = dev;
 	ah->last_send = 0;
 	kref_init(&ah->ref);
 
-	ah->ah = ib_create_ah(pd, attr);
-	if (IS_ERR(ah->ah)) {
+	vah = ib_create_ah(pd, attr);
+	if (IS_ERR(vah)) {
 		kfree(ah);
-		ah = NULL;
-	} else
+		ah = (struct ipoib_ah *)vah;
+	} else {
+		ah->ah = vah;
 		ipoib_dbg(netdev_priv(dev), "Created ah %p\n", ah->ah);
+	}
 
 	return ah;
 }
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 7567b60..83695b4 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -432,7 +432,7 @@
 
 	spin_lock_irqsave(&priv->lock, flags);
 
-	if (ah) {
+	if (!IS_ERR_OR_NULL(ah)) {
 		path->pathrec = *pathrec;
 
 		old_ah   = path->ah;
@@ -555,6 +555,7 @@
 	return 0;
 }
 
+/* called with rcu_read_lock */
 static void neigh_add_path(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
@@ -636,6 +637,7 @@
 	spin_unlock_irqrestore(&priv->lock, flags);
 }
 
+/* called with rcu_read_lock */
 static void ipoib_path_lookup(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(skb->dev);
@@ -720,13 +722,14 @@
 	struct neighbour *n = NULL;
 	unsigned long flags;
 
+	rcu_read_lock();
 	if (likely(skb_dst(skb)))
 		n = dst_get_neighbour(skb_dst(skb));
 
 	if (likely(n)) {
 		if (unlikely(!*to_ipoib_neigh(n))) {
 			ipoib_path_lookup(skb, dev);
-			return NETDEV_TX_OK;
+			goto unlock;
 		}
 
 		neigh = *to_ipoib_neigh(n);
@@ -749,17 +752,17 @@
 			ipoib_neigh_free(dev, neigh);
 			spin_unlock_irqrestore(&priv->lock, flags);
 			ipoib_path_lookup(skb, dev);
-			return NETDEV_TX_OK;
+			goto unlock;
 		}
 
 		if (ipoib_cm_get(neigh)) {
 			if (ipoib_cm_up(neigh)) {
 				ipoib_cm_send(dev, skb, ipoib_cm_get(neigh));
-				return NETDEV_TX_OK;
+				goto unlock;
 			}
 		} else if (neigh->ah) {
 			ipoib_send(dev, skb, neigh->ah, IPOIB_QPN(n->ha));
-			return NETDEV_TX_OK;
+			goto unlock;
 		}
 
 		if (skb_queue_len(&neigh->queue) < IPOIB_MAX_PATH_REC_QUEUE) {
@@ -793,13 +796,14 @@
 					   phdr->hwaddr + 4);
 				dev_kfree_skb_any(skb);
 				++dev->stats.tx_dropped;
-				return NETDEV_TX_OK;
+				goto unlock;
 			}
 
 			unicast_arp_send(skb, dev, phdr);
 		}
 	}
-
+unlock:
+	rcu_read_unlock();
 	return NETDEV_TX_OK;
 }
 
@@ -837,7 +841,7 @@
 	dst = skb_dst(skb);
 	n = NULL;
 	if (dst)
-		n = dst_get_neighbour(dst);
+		n = dst_get_neighbour_raw(dst);
 	if ((!dst || !n) && daddr) {
 		struct ipoib_pseudoheader *phdr =
 			(struct ipoib_pseudoheader *) skb_push(skb, sizeof *phdr);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 1b7a976..873bff9 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -240,8 +240,11 @@
 		av.grh.dgid = mcast->mcmember.mgid;
 
 		ah = ipoib_create_ah(dev, priv->pd, &av);
-		if (!ah) {
-			ipoib_warn(priv, "ib_address_create failed\n");
+		if (IS_ERR(ah)) {
+			ipoib_warn(priv, "ib_address_create failed %ld\n",
+				-PTR_ERR(ah));
+			/* use original error */
+			return PTR_ERR(ah);
 		} else {
 			spin_lock_irq(&priv->lock);
 			mcast->ah = ah;
@@ -266,7 +269,7 @@
 
 		skb->dev = dev;
 		if (dst)
-			n = dst_get_neighbour(dst);
+			n = dst_get_neighbour_raw(dst);
 		if (!dst || !n) {
 			/* put pseudoheader back on for next time */
 			skb_push(skb, sizeof (struct ipoib_pseudoheader));
@@ -722,6 +725,8 @@
 	if (mcast && mcast->ah) {
 		struct dst_entry *dst = skb_dst(skb);
 		struct neighbour *n = NULL;
+
+		rcu_read_lock();
 		if (dst)
 			n = dst_get_neighbour(dst);
 		if (n && !*to_ipoib_neigh(n)) {
@@ -734,7 +739,7 @@
 				list_add_tail(&neigh->list, &mcast->neigh_list);
 			}
 		}
-
+		rcu_read_unlock();
 		spin_unlock_irqrestore(&priv->lock, flags);
 		ipoib_send(dev, skb, mcast->ah, IB_MULTICAST_QPN);
 		return;
diff --git a/drivers/input/mouse/elantech.c b/drivers/input/mouse/elantech.c
index 09b93b1..e2a9867 100644
--- a/drivers/input/mouse/elantech.c
+++ b/drivers/input/mouse/elantech.c
@@ -1210,18 +1210,28 @@
  */
 static int elantech_set_properties(struct elantech_data *etd)
 {
+	/* This represents the version of IC body. */
 	int ver = (etd->fw_version & 0x0f0000) >> 16;
 
+	/* Early version of Elan touchpads doesn't obey the rule. */
 	if (etd->fw_version < 0x020030 || etd->fw_version == 0x020600)
 		etd->hw_version = 1;
-	else if (etd->fw_version < 0x150600)
-		etd->hw_version = 2;
-	else if (ver == 5)
-		etd->hw_version = 3;
-	else if (ver == 6)
-		etd->hw_version = 4;
-	else
-		return -1;
+	else {
+		switch (ver) {
+		case 2:
+		case 4:
+			etd->hw_version = 2;
+			break;
+		case 5:
+			etd->hw_version = 3;
+			break;
+		case 6:
+			etd->hw_version = 4;
+			break;
+		default:
+			return -1;
+		}
+	}
 
 	/*
 	 * Turn on packet checking by default.
diff --git a/drivers/input/serio/ams_delta_serio.c b/drivers/input/serio/ams_delta_serio.c
index 4b2a42f..d4d08bd 100644
--- a/drivers/input/serio/ams_delta_serio.c
+++ b/drivers/input/serio/ams_delta_serio.c
@@ -24,6 +24,7 @@
 #include <linux/irq.h>
 #include <linux/serio.h>
 #include <linux/slab.h>
+#include <linux/module.h>
 
 #include <asm/mach-types.h>
 #include <plat/board-ams-delta.h>
diff --git a/drivers/input/serio/i8042-x86ia64io.h b/drivers/input/serio/i8042-x86ia64io.h
index bb9f5d3..b4cfc6c 100644
--- a/drivers/input/serio/i8042-x86ia64io.h
+++ b/drivers/input/serio/i8042-x86ia64io.h
@@ -431,6 +431,13 @@
 			DMI_MATCH(DMI_PRODUCT_NAME, "Vostro V13"),
 		},
 	},
+	{
+		/* Newer HP Pavilion dv4 models */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP Pavilion dv4 Notebook PC"),
+		},
+	},
 	{ }
 };
 
@@ -560,6 +567,13 @@
 			DMI_MATCH(DMI_PRODUCT_NAME, "Vostro V13"),
 		},
 	},
+	{
+		/* Newer HP Pavilion dv4 models */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP Pavilion dv4 Notebook PC"),
+		},
+	},
 	{ }
 };
 
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index c0c7820..bdc447f 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -405,6 +405,9 @@
 int dmar_disabled = 1;
 #endif /*CONFIG_INTEL_IOMMU_DEFAULT_ON*/
 
+int intel_iommu_enabled = 0;
+EXPORT_SYMBOL_GPL(intel_iommu_enabled);
+
 static int dmar_map_gfx = 1;
 static int dmar_forcedac;
 static int intel_iommu_strict;
@@ -3524,7 +3527,7 @@
 	return 0;
 }
 
-int dmar_parse_rmrr_atsr_dev(void)
+int __init dmar_parse_rmrr_atsr_dev(void)
 {
 	struct dmar_rmrr_unit *rmrr, *rmrr_n;
 	struct dmar_atsr_unit *atsr, *atsr_n;
@@ -3647,6 +3650,8 @@
 
 	bus_register_notifier(&pci_bus_type, &device_nb);
 
+	intel_iommu_enabled = 1;
+
 	return 0;
 }
 
diff --git a/drivers/iommu/intr_remapping.c b/drivers/iommu/intr_remapping.c
index 07c9f18..6777ca0 100644
--- a/drivers/iommu/intr_remapping.c
+++ b/drivers/iommu/intr_remapping.c
@@ -773,7 +773,7 @@
 	return ir_supported;
 }
 
-int ir_dev_scope_init(void)
+int __init ir_dev_scope_init(void)
 {
 	if (!intr_remapping_enabled)
 		return 0;
diff --git a/drivers/iommu/omap-iommu-debug.c b/drivers/iommu/omap-iommu-debug.c
index 9c192e7..288da5c 100644
--- a/drivers/iommu/omap-iommu-debug.c
+++ b/drivers/iommu/omap-iommu-debug.c
@@ -10,6 +10,7 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/module.h>
 #include <linux/err.h>
 #include <linux/clk.h>
 #include <linux/io.h>
diff --git a/drivers/iommu/omap-iovmm.c b/drivers/iommu/omap-iovmm.c
index e8fdb88..46be456 100644
--- a/drivers/iommu/omap-iovmm.c
+++ b/drivers/iommu/omap-iovmm.c
@@ -10,6 +10,7 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/module.h>
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
diff --git a/drivers/isdn/divert/divert_procfs.c b/drivers/isdn/divert/divert_procfs.c
index 33ec9e4..9021182 100644
--- a/drivers/isdn/divert/divert_procfs.c
+++ b/drivers/isdn/divert/divert_procfs.c
@@ -242,6 +242,12 @@
 		case IIOCDOCFINT:
 			if (!divert_if.drv_to_name(dioctl.cf_ctrl.drvid))
 				return (-EINVAL);	/* invalid driver */
+			if (strnlen(dioctl.cf_ctrl.msn, sizeof(dioctl.cf_ctrl.msn)) ==
+					sizeof(dioctl.cf_ctrl.msn))
+				return -EINVAL;
+			if (strnlen(dioctl.cf_ctrl.fwd_nr, sizeof(dioctl.cf_ctrl.fwd_nr)) ==
+					sizeof(dioctl.cf_ctrl.fwd_nr))
+				return -EINVAL;
 			if ((i = cf_command(dioctl.cf_ctrl.drvid,
 					    (cmd == IIOCDOCFACT) ? 1 : (cmd == IIOCDOCFDIS) ? 0 : 2,
 					    dioctl.cf_ctrl.cfproc,
diff --git a/drivers/isdn/i4l/isdn_net.c b/drivers/isdn/i4l/isdn_net.c
index 1f73d7f..2339d73 100644
--- a/drivers/isdn/i4l/isdn_net.c
+++ b/drivers/isdn/i4l/isdn_net.c
@@ -2756,6 +2756,9 @@
 			char *c,
 			*e;
 
+			if (strnlen(cfg->drvid, sizeof(cfg->drvid)) ==
+					sizeof(cfg->drvid))
+				return -EINVAL;
 			drvidx = -1;
 			chidx = -1;
 			strcpy(drvid, cfg->drvid);
diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c
index 661b692..6d5628b 100644
--- a/drivers/leds/led-class.c
+++ b/drivers/leds/led-class.c
@@ -270,11 +270,8 @@
 	del_timer_sync(&led_cdev->blink_timer);
 
 	if (led_cdev->blink_set &&
-	    !led_cdev->blink_set(led_cdev, delay_on, delay_off)) {
-		led_cdev->blink_delay_on = *delay_on;
-		led_cdev->blink_delay_off = *delay_off;
+	    !led_cdev->blink_set(led_cdev, delay_on, delay_off))
 		return;
-	}
 
 	/* blink with 1 Hz as default if nothing specified */
 	if (!*delay_on && !*delay_off)
diff --git a/drivers/macintosh/via-macii.c b/drivers/macintosh/via-macii.c
index 817f37a..c9570fc 100644
--- a/drivers/macintosh/via-macii.c
+++ b/drivers/macintosh/via-macii.c
@@ -159,7 +159,7 @@
 	err = macii_init_via();
 	if (err) goto out;
 
-	err = request_irq(IRQ_MAC_ADB, macii_interrupt, IRQ_FLG_LOCK, "ADB",
+	err = request_irq(IRQ_MAC_ADB, macii_interrupt, 0, "ADB",
 			  macii_interrupt);
 	if (err) goto out;
 
diff --git a/drivers/macintosh/via-maciisi.c b/drivers/macintosh/via-maciisi.c
index 9ab5b0c..34d02a9 100644
--- a/drivers/macintosh/via-maciisi.c
+++ b/drivers/macintosh/via-maciisi.c
@@ -122,8 +122,8 @@
 		return err;
 	}
 
-	if (request_irq(IRQ_MAC_ADB, maciisi_interrupt, IRQ_FLG_LOCK | IRQ_FLG_FAST, 
-			"ADB", maciisi_interrupt)) {
+	if (request_irq(IRQ_MAC_ADB, maciisi_interrupt, 0, "ADB",
+			maciisi_interrupt)) {
 		printk(KERN_ERR "maciisi_init: can't get irq %d\n", IRQ_MAC_ADB);
 		return -EAGAIN;
 	}
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 7878712..b690711 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -1106,10 +1106,12 @@
 	 */
 	int i;
 
+	spin_lock_irq(&bitmap->lock);
 	for (i = 0; i < bitmap->file_pages; i++)
 		set_page_attr(bitmap, bitmap->filemap[i],
 			      BITMAP_PAGE_NEEDWRITE);
 	bitmap->allclean = 0;
+	spin_unlock_irq(&bitmap->lock);
 }
 
 static void bitmap_count_page(struct bitmap *bitmap, sector_t offset, int inc)
@@ -1605,7 +1607,9 @@
 	for (chunk = s; chunk <= e; chunk++) {
 		sector_t sec = (sector_t)chunk << CHUNK_BLOCK_SHIFT(bitmap);
 		bitmap_set_memory_bits(bitmap, sec, 1);
+		spin_lock_irq(&bitmap->lock);
 		bitmap_file_set_bit(bitmap, sec);
+		spin_unlock_irq(&bitmap->lock);
 		if (sec < bitmap->mddev->recovery_cp)
 			/* We are asserting that the array is dirty,
 			 * so move the recovery_cp address back so
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 84acfe7..ee98173 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -570,7 +570,7 @@
 	    mddev->ctime == 0 && !mddev->hold_active) {
 		/* Array is not configured at all, and not held active,
 		 * so destroy it */
-		list_del(&mddev->all_mddevs);
+		list_del_init(&mddev->all_mddevs);
 		bs = mddev->bio_set;
 		mddev->bio_set = NULL;
 		if (mddev->gendisk) {
@@ -2546,7 +2546,8 @@
 		sep = ",";
 	}
 	if (test_bit(Blocked, &rdev->flags) ||
-	    rdev->badblocks.unacked_exist) {
+	    (rdev->badblocks.unacked_exist
+	     && !test_bit(Faulty, &rdev->flags))) {
 		len += sprintf(page+len, "%sblocked", sep);
 		sep = ",";
 	}
@@ -3788,6 +3789,8 @@
 	if (err)
 		return err;
 	else {
+		if (mddev->hold_active == UNTIL_IOCTL)
+			mddev->hold_active = 0;
 		sysfs_notify_dirent_safe(mddev->sysfs_state);
 		return len;
 	}
@@ -4487,11 +4490,20 @@
 
 	if (!entry->show)
 		return -EIO;
+	spin_lock(&all_mddevs_lock);
+	if (list_empty(&mddev->all_mddevs)) {
+		spin_unlock(&all_mddevs_lock);
+		return -EBUSY;
+	}
+	mddev_get(mddev);
+	spin_unlock(&all_mddevs_lock);
+
 	rv = mddev_lock(mddev);
 	if (!rv) {
 		rv = entry->show(mddev, page);
 		mddev_unlock(mddev);
 	}
+	mddev_put(mddev);
 	return rv;
 }
 
@@ -4507,13 +4519,19 @@
 		return -EIO;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
+	spin_lock(&all_mddevs_lock);
+	if (list_empty(&mddev->all_mddevs)) {
+		spin_unlock(&all_mddevs_lock);
+		return -EBUSY;
+	}
+	mddev_get(mddev);
+	spin_unlock(&all_mddevs_lock);
 	rv = mddev_lock(mddev);
-	if (mddev->hold_active == UNTIL_IOCTL)
-		mddev->hold_active = 0;
 	if (!rv) {
 		rv = entry->store(mddev, page, length);
 		mddev_unlock(mddev);
 	}
+	mddev_put(mddev);
 	return rv;
 }
 
@@ -7840,6 +7858,7 @@
 				  s + rdev->data_offset, sectors, acknowledged);
 	if (rv) {
 		/* Make sure they get written out promptly */
+		sysfs_notify_dirent_safe(rdev->sysfs_state);
 		set_bit(MD_CHANGE_CLEAN, &rdev->mddev->flags);
 		md_wakeup_thread(rdev->mddev->thread);
 	}
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 472aedf..31670f8 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3036,6 +3036,8 @@
 		if (dev->written)
 			s->written++;
 		rdev = rcu_dereference(conf->disks[i].rdev);
+		if (rdev && test_bit(Faulty, &rdev->flags))
+			rdev = NULL;
 		if (rdev) {
 			is_bad = is_badblock(rdev, sh->sector, STRIPE_SECTORS,
 					     &first_bad, &bad_sectors);
@@ -3063,12 +3065,12 @@
 			}
 		} else if (test_bit(In_sync, &rdev->flags))
 			set_bit(R5_Insync, &dev->flags);
-		else if (!test_bit(Faulty, &rdev->flags)) {
+		else {
 			/* in sync if before recovery_offset */
 			if (sh->sector + STRIPE_SECTORS <= rdev->recovery_offset)
 				set_bit(R5_Insync, &dev->flags);
 		}
-		if (test_bit(R5_WriteError, &dev->flags)) {
+		if (rdev && test_bit(R5_WriteError, &dev->flags)) {
 			clear_bit(R5_Insync, &dev->flags);
 			if (!test_bit(Faulty, &rdev->flags)) {
 				s->handle_bad_blocks = 1;
@@ -3076,7 +3078,7 @@
 			} else
 				clear_bit(R5_WriteError, &dev->flags);
 		}
-		if (test_bit(R5_MadeGood, &dev->flags)) {
+		if (rdev && test_bit(R5_MadeGood, &dev->flags)) {
 			if (!test_bit(Faulty, &rdev->flags)) {
 				s->handle_bad_blocks = 1;
 				atomic_inc(&rdev->nr_pending);
@@ -3110,7 +3112,7 @@
 	struct r5dev *pdev, *qdev;
 
 	clear_bit(STRIPE_HANDLE, &sh->state);
-	if (test_and_set_bit(STRIPE_ACTIVE, &sh->state)) {
+	if (test_and_set_bit_lock(STRIPE_ACTIVE, &sh->state)) {
 		/* already being handled, ensure it gets handled
 		 * again when current action finishes */
 		set_bit(STRIPE_HANDLE, &sh->state);
@@ -3159,10 +3161,14 @@
 	/* check if the array has lost more than max_degraded devices and,
 	 * if so, some requests might need to be failed.
 	 */
-	if (s.failed > conf->max_degraded && s.to_read+s.to_write+s.written)
-		handle_failed_stripe(conf, sh, &s, disks, &s.return_bi);
-	if (s.failed > conf->max_degraded && s.syncing)
-		handle_failed_sync(conf, sh, &s);
+	if (s.failed > conf->max_degraded) {
+		sh->check_state = 0;
+		sh->reconstruct_state = 0;
+		if (s.to_read+s.to_write+s.written)
+			handle_failed_stripe(conf, sh, &s, disks, &s.return_bi);
+		if (s.syncing)
+			handle_failed_sync(conf, sh, &s);
+	}
 
 	/*
 	 * might be able to return some write requests if the parity blocks
@@ -3371,7 +3377,7 @@
 
 	return_io(s.return_bi);
 
-	clear_bit(STRIPE_ACTIVE, &sh->state);
+	clear_bit_unlock(STRIPE_ACTIVE, &sh->state);
 }
 
 static void raid5_activate_delayed(struct r5conf *conf)
diff --git a/drivers/media/dvb/dvb-usb/mxl111sf-i2c.c b/drivers/media/dvb/dvb-usb/mxl111sf-i2c.c
index 2e8c288..3443455 100644
--- a/drivers/media/dvb/dvb-usb/mxl111sf-i2c.c
+++ b/drivers/media/dvb/dvb-usb/mxl111sf-i2c.c
@@ -398,7 +398,6 @@
 	u8 i2c_r_data[24];
 	u8 i = 0;
 	u8 fifo_status = 0;
-	int ret;
 	int status = 0;
 
 	mxl_i2c("read %d bytes", count);
@@ -418,7 +417,7 @@
 		i2c_w_data[4+(i*3)] = 0x00;
 	}
 
-	ret = mxl111sf_i2c_get_data(state, 0, i2c_w_data, i2c_r_data);
+	mxl111sf_i2c_get_data(state, 0, i2c_w_data, i2c_r_data);
 
 	/* Check for I2C NACK status */
 	if (mxl111sf_i2c_check_status(state) == 1) {
diff --git a/drivers/media/dvb/dvb-usb/mxl111sf-phy.c b/drivers/media/dvb/dvb-usb/mxl111sf-phy.c
index 91dc1fc..b741b3a 100644
--- a/drivers/media/dvb/dvb-usb/mxl111sf-phy.c
+++ b/drivers/media/dvb/dvb-usb/mxl111sf-phy.c
@@ -296,8 +296,7 @@
 		goto fail;
 
 	ret = mxl111sf_write_reg(state, 0x00, 0x00);
-	if (mxl_fail(ret))
-		goto fail;
+	mxl_fail(ret);
 fail:
 	return ret;
 }
@@ -328,11 +327,13 @@
 		/* set hysteresis value  reg: 0x0B<5:0> */
 		ret = mxl111sf_write_reg(state, V6_IDAC_HYSTERESIS_REG,
 					 (hysteresis_value & 0x3F));
+		mxl_fail(ret);
 	}
 
 	ret = mxl111sf_write_reg(state, V6_IDAC_SETTINGS_REG, val);
+	mxl_fail(ret);
 
-	return val;
+	return ret;
 }
 
 /*
diff --git a/drivers/media/video/s5k6aa.c b/drivers/media/video/s5k6aa.c
index 2446736..0df7f2a 100644
--- a/drivers/media/video/s5k6aa.c
+++ b/drivers/media/video/s5k6aa.c
@@ -19,6 +19,7 @@
 #include <linux/gpio.h>
 #include <linux/i2c.h>
 #include <linux/media.h>
+#include <linux/module.h>
 #include <linux/regulator/consumer.h>
 #include <linux/slab.h>
 
diff --git a/drivers/media/video/s5p-mfc/s5p_mfc_dec.c b/drivers/media/video/s5p-mfc/s5p_mfc_dec.c
index 725634d..844a4d7 100644
--- a/drivers/media/video/s5p-mfc/s5p_mfc_dec.c
+++ b/drivers/media/video/s5p-mfc/s5p_mfc_dec.c
@@ -220,8 +220,8 @@
 	strncpy(cap->card, dev->plat_dev->name, sizeof(cap->card) - 1);
 	cap->bus_info[0] = 0;
 	cap->version = KERNEL_VERSION(1, 0, 0);
-	cap->capabilities = V4L2_CAP_VIDEO_CAPTURE | V4L2_CAP_VIDEO_OUTPUT
-						    | V4L2_CAP_STREAMING;
+	cap->capabilities = V4L2_CAP_VIDEO_CAPTURE_MPLANE |
+			V4L2_CAP_VIDEO_OUTPUT_MPLANE | V4L2_CAP_STREAMING;
 	return 0;
 }
 
diff --git a/drivers/media/video/s5p-mfc/s5p_mfc_enc.c b/drivers/media/video/s5p-mfc/s5p_mfc_enc.c
index ecef127..1e8cdb7 100644
--- a/drivers/media/video/s5p-mfc/s5p_mfc_enc.c
+++ b/drivers/media/video/s5p-mfc/s5p_mfc_enc.c
@@ -785,8 +785,8 @@
 	strncpy(cap->card, dev->plat_dev->name, sizeof(cap->card) - 1);
 	cap->bus_info[0] = 0;
 	cap->version = KERNEL_VERSION(1, 0, 0);
-	cap->capabilities = V4L2_CAP_VIDEO_CAPTURE
-			  | V4L2_CAP_VIDEO_OUTPUT
+	cap->capabilities = V4L2_CAP_VIDEO_CAPTURE_MPLANE
+			  | V4L2_CAP_VIDEO_OUTPUT_MPLANE
 			  | V4L2_CAP_STREAMING;
 	return 0;
 }
diff --git a/drivers/media/video/uvc/uvc_ctrl.c b/drivers/media/video/uvc/uvc_ctrl.c
index 10c2364..254d326 100644
--- a/drivers/media/video/uvc/uvc_ctrl.c
+++ b/drivers/media/video/uvc/uvc_ctrl.c
@@ -1016,7 +1016,8 @@
 
 	menu_info = &mapping->menu_info[query_menu->index];
 
-	if (ctrl->info.flags & UVC_CTRL_FLAG_GET_RES) {
+	if (mapping->data_type == UVC_CTRL_DATA_TYPE_BITMASK &&
+	    (ctrl->info.flags & UVC_CTRL_FLAG_GET_RES)) {
 		s32 bitmap;
 
 		if (!ctrl->cached) {
@@ -1225,7 +1226,8 @@
 		/* Valid menu indices are reported by the GET_RES request for
 		 * UVC controls that support it.
 		 */
-		if (ctrl->info.flags & UVC_CTRL_FLAG_GET_RES) {
+		if (mapping->data_type == UVC_CTRL_DATA_TYPE_BITMASK &&
+		    (ctrl->info.flags & UVC_CTRL_FLAG_GET_RES)) {
 			if (!ctrl->cached) {
 				ret = uvc_ctrl_populate_cache(chain, ctrl);
 				if (ret < 0)
diff --git a/drivers/media/video/v4l2-ctrls.c b/drivers/media/video/v4l2-ctrls.c
index f17f92b..0f415da 100644
--- a/drivers/media/video/v4l2-ctrls.c
+++ b/drivers/media/video/v4l2-ctrls.c
@@ -821,8 +821,8 @@
 	fill_event(&ev, ctrl, changes);
 
 	list_for_each_entry(sev, &ctrl->ev_subs, node)
-		if (sev->fh && (sev->fh != fh ||
-				(sev->flags & V4L2_EVENT_SUB_FL_ALLOW_FEEDBACK)))
+		if (sev->fh != fh ||
+		    (sev->flags & V4L2_EVENT_SUB_FL_ALLOW_FEEDBACK))
 			v4l2_event_queue_fh(sev->fh, &ev);
 }
 
@@ -947,6 +947,7 @@
 			if (ctrl->cluster[0]->has_volatiles)
 				ctrl->flags |= V4L2_CTRL_FLAG_VOLATILE;
 		}
+		fh = NULL;
 	}
 	if (changed || update_inactive) {
 		/* If a control was changed that was not one of the controls
diff --git a/drivers/media/video/v4l2-event.c b/drivers/media/video/v4l2-event.c
index 46037f22..c26ad96 100644
--- a/drivers/media/video/v4l2-event.c
+++ b/drivers/media/video/v4l2-event.c
@@ -216,6 +216,9 @@
 	unsigned long flags;
 	unsigned i;
 
+	if (sub->type == V4L2_EVENT_ALL)
+		return -EINVAL;
+
 	if (elems < 1)
 		elems = 1;
 	if (sub->type == V4L2_EVENT_CTRL) {
@@ -283,6 +286,7 @@
 {
 	struct v4l2_subscribed_event *sev;
 	unsigned long flags;
+	int i;
 
 	if (sub->type == V4L2_EVENT_ALL) {
 		v4l2_event_unsubscribe_all(fh);
@@ -293,8 +297,12 @@
 
 	sev = v4l2_event_subscribed(fh, sub->type, sub->id);
 	if (sev != NULL) {
+		/* Remove any pending events for this subscription */
+		for (i = 0; i < sev->in_use; i++) {
+			list_del(&sev->events[sev_pos(sev, i)].list);
+			fh->navailable--;
+		}
 		list_del(&sev->list);
-		sev->fh = NULL;
 	}
 
 	spin_unlock_irqrestore(&fh->vdev->fh_lock, flags);
diff --git a/drivers/media/video/videobuf2-core.c b/drivers/media/video/videobuf2-core.c
index 979e544..95a3f5e 100644
--- a/drivers/media/video/videobuf2-core.c
+++ b/drivers/media/video/videobuf2-core.c
@@ -131,6 +131,7 @@
 			continue;
 
 		for (plane = 0; plane < vb->num_planes; ++plane) {
+			vb->v4l2_planes[plane].length = q->plane_sizes[plane];
 			vb->v4l2_planes[plane].m.mem_offset = off;
 
 			dprintk(3, "Buffer %d, plane %d offset 0x%08lx\n",
@@ -264,6 +265,7 @@
 	q->num_buffers -= buffers;
 	if (!q->num_buffers)
 		q->memory = 0;
+	INIT_LIST_HEAD(&q->queued_list);
 }
 
 /**
@@ -296,14 +298,14 @@
 {
 	unsigned int plane;
 	for (plane = 0; plane < vb->num_planes; ++plane) {
+		void *mem_priv = vb->planes[plane].mem_priv;
 		/*
 		 * If num_users() has not been provided, call_memop
 		 * will return 0, apparently nobody cares about this
 		 * case anyway. If num_users() returns more than 1,
 		 * we are not the only user of the plane's memory.
 		 */
-		if (call_memop(q, plane, num_users,
-				vb->planes[plane].mem_priv) > 1)
+		if (mem_priv && call_memop(q, plane, num_users, mem_priv) > 1)
 			return true;
 	}
 	return false;
diff --git a/drivers/mfd/ab5500-core.c b/drivers/mfd/ab5500-core.c
index 4175544..ec10629 100644
--- a/drivers/mfd/ab5500-core.c
+++ b/drivers/mfd/ab5500-core.c
@@ -13,6 +13,7 @@
  * TODO: Event handling with irq_chip. Waiting for PRCMU fw support.
  */
 
+#include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/err.h>
 #include <linux/platform_device.h>
diff --git a/drivers/mfd/ab5500-debugfs.c b/drivers/mfd/ab5500-debugfs.c
index 6be1fe6..43c0ebb 100644
--- a/drivers/mfd/ab5500-debugfs.c
+++ b/drivers/mfd/ab5500-debugfs.c
@@ -4,6 +4,7 @@
  * Debugfs support for the AB5500 MFD driver
  */
 
+#include <linux/export.h>
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
 #include <linux/mfd/ab5500/ab5500.h>
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index d593878..5664696 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -472,7 +472,7 @@
 	  module will be called bmp085.
 
 config PCH_PHUB
-	tristate "Intel EG20T PCH / OKI SEMICONDUCTOR IOH(ML7213/ML7223) PHUB"
+	tristate "Intel EG20T PCH/LAPIS Semicon IOH(ML7213/ML7223/ML7831) PHUB"
 	depends on PCI
 	help
 	  This driver is for PCH(Platform controller Hub) PHUB(Packet Hub) of
@@ -480,12 +480,13 @@
 	  processor. The Topcliff has MAC address and Option ROM data in SROM.
 	  This driver can access MAC address and Option ROM data in SROM.
 
-	  This driver also can be used for OKI SEMICONDUCTOR IOH(Input/
-	  Output Hub), ML7213 and ML7223.
-	  ML7213 IOH is for IVI(In-Vehicle Infotainment) use and ML7223 IOH is
-	  for MP(Media Phone) use.
-	  ML7213/ML7223 is companion chip for Intel Atom E6xx series.
-	  ML7213/ML7223 is completely compatible for Intel EG20T PCH.
+	  This driver also can be used for LAPIS Semiconductor's IOH,
+	  ML7213/ML7223/ML7831.
+	  ML7213 which is for IVI(In-Vehicle Infotainment) use.
+	  ML7223 IOH is for MP(Media Phone) use.
+	  ML7831 IOH is for general purpose use.
+	  ML7213/ML7223/ML7831 is companion chip for Intel Atom E6xx series.
+	  ML7213/ML7223/ML7831 is completely compatible for Intel EG20T PCH.
 
 	  To compile this driver as a module, choose M here: the module will
 	  be called pch_phub.
diff --git a/drivers/misc/ad525x_dpot.h b/drivers/misc/ad525x_dpot.h
index a662f59..82b2cb7 100644
--- a/drivers/misc/ad525x_dpot.h
+++ b/drivers/misc/ad525x_dpot.h
@@ -100,7 +100,7 @@
 	AD5293_ID = DPOT_CONF(F_RDACS_RW | F_SPI_16BIT, BRDAC0, 10, 27),
 	AD7376_ID = DPOT_CONF(F_RDACS_WONLY | F_AD_APPDATA | F_SPI_8BIT,
 			BRDAC0, 7, 28),
-	AD8400_ID = DPOT_CONF(F_RDACS_WONLY | F_AD_APPDATA | F_SPI_8BIT,
+	AD8400_ID = DPOT_CONF(F_RDACS_WONLY | F_AD_APPDATA | F_SPI_16BIT,
 			BRDAC0, 8, 29),
 	AD8402_ID = DPOT_CONF(F_RDACS_WONLY | F_AD_APPDATA | F_SPI_16BIT,
 			BRDAC0 | BRDAC1, 8, 30),
diff --git a/drivers/misc/carma/carma-fpga-program.c b/drivers/misc/carma/carma-fpga-program.c
index 7ce6065..eb5cd28 100644
--- a/drivers/misc/carma/carma-fpga-program.c
+++ b/drivers/misc/carma/carma-fpga-program.c
@@ -945,8 +945,7 @@
 /* CTL-CPLD Version Register */
 #define CTL_CPLD_VERSION	0x2000
 
-static int fpga_of_probe(struct platform_device *op,
-			 const struct of_device_id *match)
+static int fpga_of_probe(struct platform_device *op)
 {
 	struct device_node *of_node = op->dev.of_node;
 	struct device *this_device;
@@ -1107,7 +1106,7 @@
 	{},
 };
 
-static struct of_platform_driver fpga_of_driver = {
+static struct platform_driver fpga_of_driver = {
 	.probe		= fpga_of_probe,
 	.remove		= fpga_of_remove,
 	.driver		= {
@@ -1124,12 +1123,12 @@
 static int __init fpga_init(void)
 {
 	led_trigger_register_simple("fpga", &ledtrig_fpga);
-	return of_register_platform_driver(&fpga_of_driver);
+	return platform_driver_register(&fpga_of_driver);
 }
 
 static void __exit fpga_exit(void)
 {
-	of_unregister_platform_driver(&fpga_of_driver);
+	platform_driver_unregister(&fpga_of_driver);
 	led_trigger_unregister_simple(ledtrig_fpga);
 }
 
diff --git a/drivers/misc/carma/carma-fpga.c b/drivers/misc/carma/carma-fpga.c
index 3965821..14e974b2 100644
--- a/drivers/misc/carma/carma-fpga.c
+++ b/drivers/misc/carma/carma-fpga.c
@@ -1249,8 +1249,7 @@
 	return true;
 }
 
-static int data_of_probe(struct platform_device *op,
-			 const struct of_device_id *match)
+static int data_of_probe(struct platform_device *op)
 {
 	struct device_node *of_node = op->dev.of_node;
 	struct device *this_device;
@@ -1401,7 +1400,7 @@
 	{},
 };
 
-static struct of_platform_driver data_of_driver = {
+static struct platform_driver data_of_driver = {
 	.probe		= data_of_probe,
 	.remove		= data_of_remove,
 	.driver		= {
@@ -1417,12 +1416,12 @@
 
 static int __init data_init(void)
 {
-	return of_register_platform_driver(&data_of_driver);
+	return platform_driver_register(&data_of_driver);
 }
 
 static void __exit data_exit(void)
 {
-	of_unregister_platform_driver(&data_of_driver);
+	platform_driver_unregister(&data_of_driver);
 }
 
 MODULE_AUTHOR("Ira W. Snyder <iws@ovro.caltech.edu>");
diff --git a/drivers/misc/eeprom/Kconfig b/drivers/misc/eeprom/Kconfig
index 26cf12c..701edf6 100644
--- a/drivers/misc/eeprom/Kconfig
+++ b/drivers/misc/eeprom/Kconfig
@@ -85,7 +85,7 @@
 
 config EEPROM_DIGSY_MTC_CFG
 	bool "DigsyMTC display configuration EEPROMs device"
-	depends on PPC_MPC5200_GPIO && GPIOLIB && SPI_GPIO
+	depends on GPIO_MPC5200 && SPI_GPIO
 	help
 	  This option enables access to display configuration EEPROMs
 	  on digsy_mtc board. You have to additionally select Microwire
diff --git a/drivers/misc/pch_phub.c b/drivers/misc/pch_phub.c
index dee33ad..10fc478 100644
--- a/drivers/misc/pch_phub.c
+++ b/drivers/misc/pch_phub.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2010 OKI SEMICONDUCTOR CO., LTD.
+ * Copyright (C) 2011 LAPIS Semiconductor Co., Ltd.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -41,10 +41,10 @@
 #define PCH_PHUB_ROM_START_ADDR_EG20T 0x80 /* ROM data area start address offset
 					      (Intel EG20T PCH)*/
 #define PCH_PHUB_ROM_START_ADDR_ML7213 0x400 /* ROM data area start address
-						offset(OKI SEMICONDUCTOR ML7213)
+						offset(LAPIS Semicon ML7213)
 					      */
 #define PCH_PHUB_ROM_START_ADDR_ML7223 0x400 /* ROM data area start address
-						offset(OKI SEMICONDUCTOR ML7223)
+						offset(LAPIS Semicon ML7223)
 					      */
 
 /* MAX number of INT_REDUCE_CONTROL registers */
@@ -73,6 +73,9 @@
 #define PCI_DEVICE_ID_ROHM_ML7223_mPHUB	0x8012 /* for Bus-m */
 #define PCI_DEVICE_ID_ROHM_ML7223_nPHUB	0x8002 /* for Bus-n */
 
+/* Macros for ML7831 */
+#define PCI_DEVICE_ID_ROHM_ML7831_PHUB 0x8801
+
 /* SROM ACCESS Macro */
 #define PCH_WORD_ADDR_MASK (~((1 << 2) - 1))
 
@@ -115,6 +118,7 @@
  * @pch_mac_start_address:		MAC address area start address
  * @pch_opt_rom_start_address:		Option ROM start address
  * @ioh_type:				Save IOH type
+ * @pdev:				pointer to pci device struct
  */
 struct pch_phub_reg {
 	u32 phub_id_reg;
@@ -136,6 +140,7 @@
 	u32 pch_mac_start_address;
 	u32 pch_opt_rom_start_address;
 	int ioh_type;
+	struct pci_dev *pdev;
 };
 
 /* SROM SPEC for MAC address assignment offset */
@@ -471,7 +476,7 @@
 	int retval;
 	int i;
 
-	if (chip->ioh_type == 1) /* EG20T */
+	if ((chip->ioh_type == 1) || (chip->ioh_type == 5)) /* EG20T or ML7831*/
 		retval = pch_phub_gbe_serial_rom_conf(chip);
 	else	/* ML7223 */
 		retval = pch_phub_gbe_serial_rom_conf_mp(chip);
@@ -498,6 +503,7 @@
 	unsigned int orom_size;
 	int ret;
 	int err;
+	ssize_t rom_size;
 
 	struct pch_phub_reg *chip =
 		dev_get_drvdata(container_of(kobj, struct device, kobj));
@@ -509,6 +515,10 @@
 	}
 
 	/* Get Rom signature */
+	chip->pch_phub_extrom_base_address = pci_map_rom(chip->pdev, &rom_size);
+	if (!chip->pch_phub_extrom_base_address)
+		goto exrom_map_err;
+
 	pch_phub_read_serial_rom(chip, chip->pch_opt_rom_start_address,
 				(unsigned char *)&rom_signature);
 	rom_signature &= 0xff;
@@ -539,10 +549,13 @@
 		goto return_err;
 	}
 return_ok:
+	pci_unmap_rom(chip->pdev, chip->pch_phub_extrom_base_address);
 	mutex_unlock(&pch_phub_mutex);
 	return addr_offset;
 
 return_err:
+	pci_unmap_rom(chip->pdev, chip->pch_phub_extrom_base_address);
+exrom_map_err:
 	mutex_unlock(&pch_phub_mutex);
 return_err_nomutex:
 	return err;
@@ -555,6 +568,7 @@
 	int err;
 	unsigned int addr_offset;
 	int ret;
+	ssize_t rom_size;
 	struct pch_phub_reg *chip =
 		dev_get_drvdata(container_of(kobj, struct device, kobj));
 
@@ -571,6 +585,12 @@
 		goto return_ok;
 	}
 
+	chip->pch_phub_extrom_base_address = pci_map_rom(chip->pdev, &rom_size);
+	if (!chip->pch_phub_extrom_base_address) {
+		err = -ENOMEM;
+		goto exrom_map_err;
+	}
+
 	for (addr_offset = 0; addr_offset < count; addr_offset++) {
 		if (PCH_PHUB_OROM_SIZE < off + addr_offset)
 			goto return_ok;
@@ -585,10 +605,14 @@
 	}
 
 return_ok:
+	pci_unmap_rom(chip->pdev, chip->pch_phub_extrom_base_address);
 	mutex_unlock(&pch_phub_mutex);
 	return addr_offset;
 
 return_err:
+	pci_unmap_rom(chip->pdev, chip->pch_phub_extrom_base_address);
+
+exrom_map_err:
 	mutex_unlock(&pch_phub_mutex);
 	return err;
 }
@@ -598,8 +622,14 @@
 {
 	u8 mac[8];
 	struct pch_phub_reg *chip = dev_get_drvdata(dev);
+	ssize_t rom_size;
+
+	chip->pch_phub_extrom_base_address = pci_map_rom(chip->pdev, &rom_size);
+	if (!chip->pch_phub_extrom_base_address)
+		return -ENOMEM;
 
 	pch_phub_read_gbe_mac_addr(chip, mac);
+	pci_unmap_rom(chip->pdev, chip->pch_phub_extrom_base_address);
 
 	return sprintf(buf, "%pM\n", mac);
 }
@@ -608,6 +638,7 @@
 			     const char *buf, size_t count)
 {
 	u8 mac[6];
+	ssize_t rom_size;
 	struct pch_phub_reg *chip = dev_get_drvdata(dev);
 
 	if (count != 18)
@@ -617,7 +648,12 @@
 		(u32 *)&mac[0], (u32 *)&mac[1], (u32 *)&mac[2], (u32 *)&mac[3],
 		(u32 *)&mac[4], (u32 *)&mac[5]);
 
+	chip->pch_phub_extrom_base_address = pci_map_rom(chip->pdev, &rom_size);
+	if (!chip->pch_phub_extrom_base_address)
+		return -ENOMEM;
+
 	pch_phub_write_gbe_mac_addr(chip, mac);
+	pci_unmap_rom(chip->pdev, chip->pch_phub_extrom_base_address);
 
 	return count;
 }
@@ -640,7 +676,6 @@
 	int retval;
 
 	int ret;
-	ssize_t rom_size;
 	struct pch_phub_reg *chip;
 
 	chip = kzalloc(sizeof(struct pch_phub_reg), GFP_KERNEL);
@@ -677,19 +712,7 @@
 		"in pch_phub_base_address variable is %p\n", __func__,
 		chip->pch_phub_base_address);
 
-	if (id->driver_data != 3) {
-		chip->pch_phub_extrom_base_address =\
-						   pci_map_rom(pdev, &rom_size);
-		if (chip->pch_phub_extrom_base_address == 0) {
-			dev_err(&pdev->dev, "%s: pci_map_rom FAILED", __func__);
-			ret = -ENOMEM;
-			goto err_pci_map;
-		}
-		dev_dbg(&pdev->dev, "%s : "
-			"pci_map_rom SUCCESS and value in "
-			"pch_phub_extrom_base_address variable is %p\n",
-			__func__, chip->pch_phub_extrom_base_address);
-	}
+	chip->pdev = pdev; /* Save pci device struct */
 
 	if (id->driver_data == 1) { /* EG20T PCH */
 		const char *board_name;
@@ -763,6 +786,22 @@
 		chip->pch_opt_rom_start_address =\
 						 PCH_PHUB_ROM_START_ADDR_ML7223;
 		chip->pch_mac_start_address = PCH_PHUB_MAC_START_ADDR_ML7223;
+	} else if (id->driver_data == 5) { /* ML7831 */
+		retval = sysfs_create_file(&pdev->dev.kobj,
+					   &dev_attr_pch_mac.attr);
+		if (retval)
+			goto err_sysfs_create;
+
+		retval = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr);
+		if (retval)
+			goto exit_bin_attr;
+
+		/* set the prefech value */
+		iowrite32(0x000affaa, chip->pch_phub_base_address + 0x14);
+		/* set the interrupt delay value */
+		iowrite32(0x25, chip->pch_phub_base_address + 0x44);
+		chip->pch_opt_rom_start_address = PCH_PHUB_ROM_START_ADDR_EG20T;
+		chip->pch_mac_start_address = PCH_PHUB_MAC_START_ADDR_EG20T;
 	}
 
 	chip->ioh_type = id->driver_data;
@@ -773,8 +812,6 @@
 	sysfs_remove_file(&pdev->dev.kobj, &dev_attr_pch_mac.attr);
 
 err_sysfs_create:
-	pci_unmap_rom(pdev, chip->pch_phub_extrom_base_address);
-err_pci_map:
 	pci_iounmap(pdev, chip->pch_phub_base_address);
 err_pci_iomap:
 	pci_release_regions(pdev);
@@ -792,7 +829,6 @@
 
 	sysfs_remove_file(&pdev->dev.kobj, &dev_attr_pch_mac.attr);
 	sysfs_remove_bin_file(&pdev->dev.kobj, &pch_bin_attr);
-	pci_unmap_rom(pdev, chip->pch_phub_extrom_base_address);
 	pci_iounmap(pdev, chip->pch_phub_base_address);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
@@ -847,6 +883,7 @@
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ROHM_ML7213_PHUB), 2,  },
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ROHM_ML7223_mPHUB), 3,  },
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ROHM_ML7223_nPHUB), 4,  },
+	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ROHM_ML7831_PHUB), 5,  },
 	{ }
 };
 MODULE_DEVICE_TABLE(pci, pch_phub_pcidev_id);
@@ -873,5 +910,5 @@
 module_init(pch_phub_pci_init);
 module_exit(pch_phub_pci_exit);
 
-MODULE_DESCRIPTION("Intel EG20T PCH/OKI SEMICONDUCTOR IOH(ML7213/ML7223) PHUB");
+MODULE_DESCRIPTION("Intel EG20T PCH/LAPIS Semiconductor IOH(ML7213/ML7223) PHUB");
 MODULE_LICENSE("GPL");
diff --git a/drivers/misc/spear13xx_pcie_gadget.c b/drivers/misc/spear13xx_pcie_gadget.c
index cfbddbe..43d073b 100644
--- a/drivers/misc/spear13xx_pcie_gadget.c
+++ b/drivers/misc/spear13xx_pcie_gadget.c
@@ -903,6 +903,6 @@
 }
 module_exit(spear_pcie_gadget_exit);
 
-MODULE_ALIAS("pcie-gadget-spear");
+MODULE_ALIAS("platform:pcie-gadget-spear");
 MODULE_AUTHOR("Pratyush Anand");
 MODULE_LICENSE("GPL");
diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
index a1cb21f..1e0e27c 100644
--- a/drivers/mmc/card/block.c
+++ b/drivers/mmc/card/block.c
@@ -1606,6 +1606,14 @@
 		  MMC_QUIRK_BLK_NO_CMD23),
 	MMC_FIXUP("MMC32G", 0x11, CID_OEMID_ANY, add_quirk_mmc,
 		  MMC_QUIRK_BLK_NO_CMD23),
+
+	/*
+	 * Some Micron MMC cards needs longer data read timeout than
+	 * indicated in CSD.
+	 */
+	MMC_FIXUP(CID_NAME_ANY, 0x13, 0x200, add_quirk_mmc,
+		  MMC_QUIRK_LONG_READ_TIME),
+
 	END_FIXUP
 };
 
diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 5278ffb..950b97d 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -529,6 +529,18 @@
 			data->timeout_clks = 0;
 		}
 	}
+
+	/*
+	 * Some cards require longer data read timeout than indicated in CSD.
+	 * Address this by setting the read timeout to a "reasonably high"
+	 * value. For the cards tested, 300ms has proven enough. If necessary,
+	 * this value can be increased if other problematic cards require this.
+	 */
+	if (mmc_card_long_read_time(card) && data->flags & MMC_DATA_READ) {
+		data->timeout_ns = 300000000;
+		data->timeout_clks = 0;
+	}
+
 	/*
 	 * Some cards need very high timeouts if driven in SPI mode.
 	 * The worst observed timeout was 900ms after writing a
@@ -1213,6 +1225,46 @@
 	mmc_host_clk_release(host);
 }
 
+static void mmc_poweroff_notify(struct mmc_host *host)
+{
+	struct mmc_card *card;
+	unsigned int timeout;
+	unsigned int notify_type = EXT_CSD_NO_POWER_NOTIFICATION;
+	int err = 0;
+
+	card = host->card;
+
+	/*
+	 * Send power notify command only if card
+	 * is mmc and notify state is powered ON
+	 */
+	if (card && mmc_card_mmc(card) &&
+	    (card->poweroff_notify_state == MMC_POWERED_ON)) {
+
+		if (host->power_notify_type == MMC_HOST_PW_NOTIFY_SHORT) {
+			notify_type = EXT_CSD_POWER_OFF_SHORT;
+			timeout = card->ext_csd.generic_cmd6_time;
+			card->poweroff_notify_state = MMC_POWEROFF_SHORT;
+		} else {
+			notify_type = EXT_CSD_POWER_OFF_LONG;
+			timeout = card->ext_csd.power_off_longtime;
+			card->poweroff_notify_state = MMC_POWEROFF_LONG;
+		}
+
+		err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
+				 EXT_CSD_POWER_OFF_NOTIFICATION,
+				 notify_type, timeout);
+
+		if (err && err != -EBADMSG)
+			pr_err("Device failed to respond within %d poweroff "
+			       "time. Forcefully powering down the device\n",
+			       timeout);
+
+		/* Set the card state to no notification after the poweroff */
+		card->poweroff_notify_state = MMC_NO_POWER_NOTIFICATION;
+	}
+}
+
 /*
  * Apply power to the MMC stack.  This is a two-stage process.
  * First, we enable power to the card without the clock running.
@@ -1269,42 +1321,12 @@
 
 void mmc_power_off(struct mmc_host *host)
 {
-	struct mmc_card *card;
-	unsigned int notify_type;
-	unsigned int timeout;
-	int err;
-
 	mmc_host_clk_hold(host);
 
-	card = host->card;
 	host->ios.clock = 0;
 	host->ios.vdd = 0;
 
-	if (card && mmc_card_mmc(card) &&
-	    (card->poweroff_notify_state == MMC_POWERED_ON)) {
-
-		if (host->power_notify_type == MMC_HOST_PW_NOTIFY_SHORT) {
-			notify_type = EXT_CSD_POWER_OFF_SHORT;
-			timeout = card->ext_csd.generic_cmd6_time;
-			card->poweroff_notify_state = MMC_POWEROFF_SHORT;
-		} else {
-			notify_type = EXT_CSD_POWER_OFF_LONG;
-			timeout = card->ext_csd.power_off_longtime;
-			card->poweroff_notify_state = MMC_POWEROFF_LONG;
-		}
-
-		err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-				 EXT_CSD_POWER_OFF_NOTIFICATION,
-				 notify_type, timeout);
-
-		if (err && err != -EBADMSG)
-			pr_err("Device failed to respond within %d poweroff "
-			       "time. Forcefully powering down the device\n",
-			       timeout);
-
-		/* Set the card state to no notification after the poweroff */
-		card->poweroff_notify_state = MMC_NO_POWER_NOTIFICATION;
-	}
+	mmc_poweroff_notify(host);
 
 	/*
 	 * Reset ocr mask to be the highest possible voltage supported for
@@ -2196,7 +2218,7 @@
 
 	mmc_bus_get(host);
 
-	if (host->bus_ops && !host->bus_dead && host->bus_ops->awake)
+	if (host->bus_ops && !host->bus_dead && host->bus_ops->sleep)
 		err = host->bus_ops->sleep(host);
 
 	mmc_bus_put(host);
@@ -2302,8 +2324,17 @@
 		 * pre-claim the host.
 		 */
 		if (mmc_try_claim_host(host)) {
-			if (host->bus_ops->suspend)
+			if (host->bus_ops->suspend) {
+				/*
+				 * For eMMC 4.5 device send notify command
+				 * before sleep, because in sleep state eMMC 4.5
+				 * devices respond to only RESET and AWAKE cmd
+				 */
+				mmc_poweroff_notify(host);
 				err = host->bus_ops->suspend(host);
+			}
+			mmc_do_release_host(host);
+
 			if (err == -ENOSYS || !host->bus_ops->resume) {
 				/*
 				 * We simply "remove" the card in this case.
@@ -2318,7 +2349,6 @@
 				host->pm_flags = 0;
 				err = 0;
 			}
-			mmc_do_release_host(host);
 		} else {
 			err = -EBUSY;
 		}
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index dbf421a..d240427 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -876,17 +876,21 @@
 	 * set the notification byte in the ext_csd register of device
 	 */
 	if ((host->caps2 & MMC_CAP2_POWEROFF_NOTIFY) &&
-	    (card->poweroff_notify_state == MMC_NO_POWER_NOTIFICATION)) {
+	    (card->ext_csd.rev >= 6)) {
 		err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 				 EXT_CSD_POWER_OFF_NOTIFICATION,
 				 EXT_CSD_POWER_ON,
 				 card->ext_csd.generic_cmd6_time);
 		if (err && err != -EBADMSG)
 			goto free_card;
-	}
 
-	if (!err)
-		card->poweroff_notify_state = MMC_POWERED_ON;
+		/*
+		 * The err can be -EBADMSG or 0,
+		 * so check for success and update the flag
+		 */
+		if (!err)
+			card->poweroff_notify_state = MMC_POWERED_ON;
+	}
 
 	/*
 	 * Activate high speed (if supported)
diff --git a/drivers/mmc/host/mxcmmc.c b/drivers/mmc/host/mxcmmc.c
index 325ea61..8e0fbe9 100644
--- a/drivers/mmc/host/mxcmmc.c
+++ b/drivers/mmc/host/mxcmmc.c
@@ -732,6 +732,7 @@
 				"failed to config DMA channel. Falling back to PIO\n");
 			dma_release_channel(host->dma);
 			host->do_dma = 0;
+			host->dma = NULL;
 		}
 	}
 
diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
index 101cd31..d5fe43d 100644
--- a/drivers/mmc/host/omap_hsmmc.c
+++ b/drivers/mmc/host/omap_hsmmc.c
@@ -1010,6 +1010,7 @@
 			host->data->sg_len,
 			omap_hsmmc_get_dma_dir(host, host->data));
 		omap_free_dma(dma_ch);
+		host->data->host_cookie = 0;
 	}
 	host->data = NULL;
 }
@@ -1575,8 +1576,10 @@
 	struct mmc_data *data = mrq->data;
 
 	if (host->use_dma) {
-		dma_unmap_sg(mmc_dev(host->mmc), data->sg, data->sg_len,
-			     omap_hsmmc_get_dma_dir(host, data));
+		if (data->host_cookie)
+			dma_unmap_sg(mmc_dev(host->mmc), data->sg,
+				     data->sg_len,
+				     omap_hsmmc_get_dma_dir(host, data));
 		data->host_cookie = 0;
 	}
 }
diff --git a/drivers/mmc/host/sdhci-cns3xxx.c b/drivers/mmc/host/sdhci-cns3xxx.c
index 4b920b7..87b6f07 100644
--- a/drivers/mmc/host/sdhci-cns3xxx.c
+++ b/drivers/mmc/host/sdhci-cns3xxx.c
@@ -15,6 +15,7 @@
 #include <linux/delay.h>
 #include <linux/device.h>
 #include <linux/mmc/host.h>
+#include <linux/module.h>
 #include <mach/cns3xxx.h>
 #include "sdhci-pltfm.h"
 
diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c
index ae57769..4b976f0 100644
--- a/drivers/mmc/host/sdhci-esdhc-imx.c
+++ b/drivers/mmc/host/sdhci-esdhc-imx.c
@@ -32,6 +32,7 @@
 /* VENDOR SPEC register */
 #define SDHCI_VENDOR_SPEC		0xC0
 #define  SDHCI_VENDOR_SPEC_SDIO_QUIRK	0x00000002
+#define SDHCI_WTMK_LVL			0x44
 #define SDHCI_MIX_CTRL			0x48
 
 /*
@@ -476,6 +477,13 @@
 	if (is_imx53_esdhc(imx_data))
 		imx_data->flags |= ESDHC_FLAG_MULTIBLK_NO_INT;
 
+	/*
+	 * The imx6q ROM code will change the default watermark level setting
+	 * to something insane.  Change it back here.
+	 */
+	if (is_imx6q_usdhc(imx_data))
+		writel(0x08100810, host->ioaddr + SDHCI_WTMK_LVL);
+
 	boarddata = &imx_data->boarddata;
 	if (sdhci_esdhc_imx_probe_dt(pdev, boarddata) < 0) {
 		if (!host->mmc->parent->platform_data) {
diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
index 3d00e72..cb60c41 100644
--- a/drivers/mmc/host/sdhci-s3c.c
+++ b/drivers/mmc/host/sdhci-s3c.c
@@ -644,8 +644,6 @@
 static struct platform_driver sdhci_s3c_driver = {
 	.probe		= sdhci_s3c_probe,
 	.remove		= __devexit_p(sdhci_s3c_remove),
-	.suspend	= sdhci_s3c_suspend,
-	.resume	        = sdhci_s3c_resume,
 	.driver		= {
 		.owner	= THIS_MODULE,
 		.name	= "s3c-sdhci",
diff --git a/drivers/mmc/host/sh_mmcif.c b/drivers/mmc/host/sh_mmcif.c
index 369366c..d5505f3 100644
--- a/drivers/mmc/host/sh_mmcif.c
+++ b/drivers/mmc/host/sh_mmcif.c
@@ -908,7 +908,7 @@
 		if (host->power) {
 			pm_runtime_put(&host->pd->dev);
 			host->power = false;
-			if (p->down_pwr)
+			if (p->down_pwr && ios->power_mode == MMC_POWER_OFF)
 				p->down_pwr(host->pd);
 		}
 		host->state = STATE_IDLE;
diff --git a/drivers/mmc/host/tmio_mmc_pio.c b/drivers/mmc/host/tmio_mmc_pio.c
index d85a60c..4208b39 100644
--- a/drivers/mmc/host/tmio_mmc_pio.c
+++ b/drivers/mmc/host/tmio_mmc_pio.c
@@ -798,7 +798,7 @@
 		/* start bus clock */
 		tmio_mmc_clk_start(host);
 	} else if (ios->power_mode != MMC_POWER_UP) {
-		if (host->set_pwr)
+		if (host->set_pwr && ios->power_mode == MMC_POWER_OFF)
 			host->set_pwr(host->pdev, 0);
 		if ((pdata->flags & TMIO_MMC_HAS_COLD_CD) &&
 		    pdata->power) {
diff --git a/drivers/mtd/maps/bcm963xx-flash.c b/drivers/mtd/maps/bcm963xx-flash.c
index 608967f..736ca10 100644
--- a/drivers/mtd/maps/bcm963xx-flash.c
+++ b/drivers/mtd/maps/bcm963xx-flash.c
@@ -21,6 +21,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/slab.h>
+#include <linux/module.h>
 #include <linux/mtd/map.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 583f66c..654a5e9 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -245,6 +245,8 @@
 
 source "drivers/net/fddi/Kconfig"
 
+source "drivers/net/hippi/Kconfig"
+
 config NET_SB1000
 	tristate "General Instruments Surfboard 1000"
 	depends on PNP
diff --git a/drivers/net/arcnet/Kconfig b/drivers/net/arcnet/Kconfig
index a73d9dc..84fb634 100644
--- a/drivers/net/arcnet/Kconfig
+++ b/drivers/net/arcnet/Kconfig
@@ -4,7 +4,7 @@
 
 menuconfig ARCNET
 	depends on NETDEVICES && (ISA || PCI || PCMCIA)
-	bool "ARCnet support"
+	tristate "ARCnet support"
 	---help---
 	  If you have a network card of this type, say Y and check out the
 	  (arguably) beautiful poetry in
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index b0c5772..7f87568 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2553,30 +2553,6 @@
 	}
 }
 
-static __be32 bond_glean_dev_ip(struct net_device *dev)
-{
-	struct in_device *idev;
-	struct in_ifaddr *ifa;
-	__be32 addr = 0;
-
-	if (!dev)
-		return 0;
-
-	rcu_read_lock();
-	idev = __in_dev_get_rcu(dev);
-	if (!idev)
-		goto out;
-
-	ifa = idev->ifa_list;
-	if (!ifa)
-		goto out;
-
-	addr = ifa->ifa_local;
-out:
-	rcu_read_unlock();
-	return addr;
-}
-
 static int bond_has_this_ip(struct bonding *bond, __be32 ip)
 {
 	struct vlan_entry *vlan;
@@ -3322,6 +3298,10 @@
 	struct bonding *bond;
 	struct vlan_entry *vlan;
 
+	/* we only care about primary address */
+	if(ifa->ifa_flags & IFA_F_SECONDARY)
+		return NOTIFY_DONE;
+
 	list_for_each_entry(bond, &bn->dev_list, bond_list) {
 		if (bond->dev == event_dev) {
 			switch (event) {
@@ -3329,7 +3309,7 @@
 				bond->master_ip = ifa->ifa_local;
 				return NOTIFY_OK;
 			case NETDEV_DOWN:
-				bond->master_ip = bond_glean_dev_ip(bond->dev);
+				bond->master_ip = 0;
 				return NOTIFY_OK;
 			default:
 				return NOTIFY_DONE;
@@ -3345,8 +3325,7 @@
 					vlan->vlan_ip = ifa->ifa_local;
 					return NOTIFY_OK;
 				case NETDEV_DOWN:
-					vlan->vlan_ip =
-						bond_glean_dev_ip(vlan_dev);
+					vlan->vlan_ip = 0;
 					return NOTIFY_OK;
 				default:
 					return NOTIFY_DONE;
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index 5a20804..4ef7e2f 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -319,6 +319,13 @@
 		goto out;
 	}
 
+	if (bond->slave_cnt > 0) {
+		pr_err("unable to update mode of %s because it has slaves.\n",
+			bond->dev->name);
+		ret = -EPERM;
+		goto out;
+	}
+
 	new_value = bond_parse_parm(buf, bond_mode_tbl);
 	if (new_value < 0)  {
 		pr_err("%s: Ignoring invalid mode value %.*s.\n",
diff --git a/drivers/net/can/sja1000/peak_pci.c b/drivers/net/can/sja1000/peak_pci.c
index 905bce0..2c7f503 100644
--- a/drivers/net/can/sja1000/peak_pci.c
+++ b/drivers/net/can/sja1000/peak_pci.c
@@ -20,7 +20,6 @@
  */
 
 #include <linux/kernel.h>
-#include <linux/version.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
 #include <linux/netdevice.h>
diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c
index 4cf835d..3fb66d0 100644
--- a/drivers/net/ethernet/broadcom/b44.c
+++ b/drivers/net/ethernet/broadcom/b44.c
@@ -608,7 +608,7 @@
 				 skb->len,
 				 DMA_TO_DEVICE);
 		rp->skb = NULL;
-		dev_kfree_skb(skb);
+		dev_kfree_skb_irq(skb);
 	}
 
 	bp->tx_cons = cons;
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
index bce203f..882f48f 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
@@ -10327,6 +10327,43 @@
 	return 0;
 }
 
+
+static void bnx2x_5461x_set_link_led(struct bnx2x_phy *phy,
+				       struct link_params *params, u8 mode)
+{
+	struct bnx2x *bp = params->bp;
+	u16 temp;
+
+	bnx2x_cl22_write(bp, phy,
+		MDIO_REG_GPHY_SHADOW,
+		MDIO_REG_GPHY_SHADOW_LED_SEL1);
+	bnx2x_cl22_read(bp, phy,
+		MDIO_REG_GPHY_SHADOW,
+		&temp);
+	temp &= 0xff00;
+
+	DP(NETIF_MSG_LINK, "54618x set link led (mode=%x)\n", mode);
+	switch (mode) {
+	case LED_MODE_FRONT_PANEL_OFF:
+	case LED_MODE_OFF:
+		temp |= 0x00ee;
+		break;
+	case LED_MODE_OPER:
+		temp |= 0x0001;
+		break;
+	case LED_MODE_ON:
+		temp |= 0x00ff;
+		break;
+	default:
+		break;
+	}
+	bnx2x_cl22_write(bp, phy,
+		MDIO_REG_GPHY_SHADOW,
+		MDIO_REG_GPHY_SHADOW_WR_ENA | temp);
+	return;
+}
+
+
 static void bnx2x_54618se_link_reset(struct bnx2x_phy *phy,
 				     struct link_params *params)
 {
@@ -11103,7 +11140,7 @@
 	.config_loopback = (config_loopback_t)bnx2x_54618se_config_loopback,
 	.format_fw_ver	= (format_fw_ver_t)NULL,
 	.hw_reset	= (hw_reset_t)NULL,
-	.set_link_led	= (set_link_led_t)NULL,
+	.set_link_led	= (set_link_led_t)bnx2x_5461x_set_link_led,
 	.phy_specific_func = (phy_specific_func_t)NULL
 };
 /*****************************************************************/
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 6486ab8..2f6361e 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -10548,33 +10548,38 @@
 
 int bnx2x_init_firmware(struct bnx2x *bp)
 {
-	const char *fw_file_name;
 	struct bnx2x_fw_file_hdr *fw_hdr;
 	int rc;
 
-	if (CHIP_IS_E1(bp))
-		fw_file_name = FW_FILE_NAME_E1;
-	else if (CHIP_IS_E1H(bp))
-		fw_file_name = FW_FILE_NAME_E1H;
-	else if (!CHIP_IS_E1x(bp))
-		fw_file_name = FW_FILE_NAME_E2;
-	else {
-		BNX2X_ERR("Unsupported chip revision\n");
-		return -EINVAL;
-	}
 
-	BNX2X_DEV_INFO("Loading %s\n", fw_file_name);
+	if (!bp->firmware) {
+		const char *fw_file_name;
 
-	rc = request_firmware(&bp->firmware, fw_file_name, &bp->pdev->dev);
-	if (rc) {
-		BNX2X_ERR("Can't load firmware file %s\n", fw_file_name);
-		goto request_firmware_exit;
-	}
+		if (CHIP_IS_E1(bp))
+			fw_file_name = FW_FILE_NAME_E1;
+		else if (CHIP_IS_E1H(bp))
+			fw_file_name = FW_FILE_NAME_E1H;
+		else if (!CHIP_IS_E1x(bp))
+			fw_file_name = FW_FILE_NAME_E2;
+		else {
+			BNX2X_ERR("Unsupported chip revision\n");
+			return -EINVAL;
+		}
+		BNX2X_DEV_INFO("Loading %s\n", fw_file_name);
 
-	rc = bnx2x_check_firmware(bp);
-	if (rc) {
-		BNX2X_ERR("Corrupt firmware file %s\n", fw_file_name);
-		goto request_firmware_exit;
+		rc = request_firmware(&bp->firmware, fw_file_name,
+				      &bp->pdev->dev);
+		if (rc) {
+			BNX2X_ERR("Can't load firmware file %s\n",
+				  fw_file_name);
+			goto request_firmware_exit;
+		}
+
+		rc = bnx2x_check_firmware(bp);
+		if (rc) {
+			BNX2X_ERR("Corrupt firmware file %s\n", fw_file_name);
+			goto request_firmware_exit;
+		}
 	}
 
 	fw_hdr = (struct bnx2x_fw_file_hdr *)bp->firmware->data;
@@ -10630,6 +10635,7 @@
 	kfree(bp->init_ops);
 	kfree(bp->init_data);
 	release_firmware(bp->firmware);
+	bp->firmware = NULL;
 }
 
 
@@ -10925,6 +10931,8 @@
 	if (bp->doorbells)
 		iounmap(bp->doorbells);
 
+	bnx2x_release_firmware(bp);
+
 	bnx2x_free_mem_bp(bp);
 
 	free_netdev(dev);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
index fc7bd0f..e58073e 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
@@ -6990,6 +6990,7 @@
 #define MDIO_REG_INTR_MASK				0x1b
 #define MDIO_REG_INTR_MASK_LINK_STATUS			(0x1 << 1)
 #define MDIO_REG_GPHY_SHADOW				0x1c
+#define MDIO_REG_GPHY_SHADOW_LED_SEL1			(0x0d << 10)
 #define MDIO_REG_GPHY_SHADOW_LED_SEL2			(0x0e << 10)
 #define MDIO_REG_GPHY_SHADOW_WR_ENA			(0x1 << 15)
 #define MDIO_REG_GPHY_SHADOW_AUTO_DET_MED		(0x1e << 10)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
index 0440425..1451769 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sp.c
@@ -5380,7 +5380,7 @@
 	rc = drv->init_fw(bp);
 	if (rc) {
 		BNX2X_ERR("Error loading firmware\n");
-		goto fw_init_err;
+		goto init_err;
 	}
 
 	/* Handle the beginning of COMMON_XXX pases separatelly... */
@@ -5388,25 +5388,25 @@
 	case FW_MSG_CODE_DRV_LOAD_COMMON_CHIP:
 		rc = bnx2x_func_init_cmn_chip(bp, drv);
 		if (rc)
-			goto init_hw_err;
+			goto init_err;
 
 		break;
 	case FW_MSG_CODE_DRV_LOAD_COMMON:
 		rc = bnx2x_func_init_cmn(bp, drv);
 		if (rc)
-			goto init_hw_err;
+			goto init_err;
 
 		break;
 	case FW_MSG_CODE_DRV_LOAD_PORT:
 		rc = bnx2x_func_init_port(bp, drv);
 		if (rc)
-			goto init_hw_err;
+			goto init_err;
 
 		break;
 	case FW_MSG_CODE_DRV_LOAD_FUNCTION:
 		rc = bnx2x_func_init_func(bp, drv);
 		if (rc)
-			goto init_hw_err;
+			goto init_err;
 
 		break;
 	default:
@@ -5414,10 +5414,7 @@
 		rc = -EINVAL;
 	}
 
-init_hw_err:
-	drv->release_fw(bp);
-
-fw_init_err:
+init_err:
 	drv->gunzip_end(bp);
 
 	/* In case of success, complete the comand immediatelly: no ramrods
diff --git a/drivers/net/ethernet/cadence/Kconfig b/drivers/net/ethernet/cadence/Kconfig
index 98849a1..b48378a 100644
--- a/drivers/net/ethernet/cadence/Kconfig
+++ b/drivers/net/ethernet/cadence/Kconfig
@@ -7,6 +7,7 @@
 
 config NET_ATMEL
 	bool "Atmel devices"
+	default y
 	depends on HAVE_NET_MACB || (ARM && ARCH_AT91RM9200)
 	---help---
 	  If you have a network (Ethernet) card belonging to this class, say Y.
diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c
index 438f458..2a22f52 100644
--- a/drivers/net/ethernet/davicom/dm9000.c
+++ b/drivers/net/ethernet/davicom/dm9000.c
@@ -613,7 +613,7 @@
 
 		if (!dm->wake_state)
 			irq_set_irq_wake(dm->irq_wake, 1);
-		else if (dm->wake_state & !opts)
+		else if (dm->wake_state && !opts)
 			irq_set_irq_wake(dm->irq_wake, 0);
 	}
 
diff --git a/drivers/net/ethernet/freescale/Kconfig b/drivers/net/ethernet/freescale/Kconfig
index c520cfd..5272f9d 100644
--- a/drivers/net/ethernet/freescale/Kconfig
+++ b/drivers/net/ethernet/freescale/Kconfig
@@ -24,6 +24,7 @@
 	bool "FEC ethernet controller (of ColdFire and some i.MX CPUs)"
 	depends on (M523x || M527x || M5272 || M528x || M520x || M532x || \
 		   ARCH_MXC || ARCH_MXS)
+	default ARCH_MXC || ARCH_MXS if ARM
 	select PHYLIB
 	---help---
 	  Say Y here if you want to use the built-in 10/100 Fast ethernet
diff --git a/drivers/net/ethernet/freescale/fec.c b/drivers/net/ethernet/freescale/fec.c
index 1124ce0..c136230 100644
--- a/drivers/net/ethernet/freescale/fec.c
+++ b/drivers/net/ethernet/freescale/fec.c
@@ -232,6 +232,7 @@
 	struct	platform_device *pdev;
 
 	int	opened;
+	int	dev_id;
 
 	/* Phylib and MDIO interface */
 	struct	mii_bus *mii_bus;
@@ -837,7 +838,7 @@
 
 	/* Adjust MAC if using macaddr */
 	if (iap == macaddr)
-		 ndev->dev_addr[ETH_ALEN-1] = macaddr[ETH_ALEN-1] + fep->pdev->id;
+		 ndev->dev_addr[ETH_ALEN-1] = macaddr[ETH_ALEN-1] + fep->dev_id;
 }
 
 /* ------------------------------------------------------------------------- */
@@ -953,7 +954,7 @@
 	char mdio_bus_id[MII_BUS_ID_SIZE];
 	char phy_name[MII_BUS_ID_SIZE + 3];
 	int phy_id;
-	int dev_id = fep->pdev->id;
+	int dev_id = fep->dev_id;
 
 	fep->phy_dev = NULL;
 
@@ -1031,7 +1032,7 @@
 	 * mdio interface in board design, and need to be configured by
 	 * fec0 mii_bus.
 	 */
-	if ((id_entry->driver_data & FEC_QUIRK_ENET_MAC) && pdev->id > 0) {
+	if ((id_entry->driver_data & FEC_QUIRK_ENET_MAC) && fep->dev_id > 0) {
 		/* fec1 uses fec0 mii_bus */
 		fep->mii_bus = fec0_mii_bus;
 		return 0;
@@ -1063,7 +1064,7 @@
 	fep->mii_bus->read = fec_enet_mdio_read;
 	fep->mii_bus->write = fec_enet_mdio_write;
 	fep->mii_bus->reset = fec_enet_mdio_reset;
-	snprintf(fep->mii_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id + 1);
+	snprintf(fep->mii_bus->id, MII_BUS_ID_SIZE, "%x", fep->dev_id + 1);
 	fep->mii_bus->priv = fep;
 	fep->mii_bus->parent = &pdev->dev;
 
@@ -1521,6 +1522,7 @@
 	int i, irq, ret = 0;
 	struct resource *r;
 	const struct of_device_id *of_id;
+	static int dev_id;
 
 	of_id = of_match_device(fec_dt_ids, &pdev->dev);
 	if (of_id)
@@ -1548,6 +1550,7 @@
 
 	fep->hwp = ioremap(r->start, resource_size(r));
 	fep->pdev = pdev;
+	fep->dev_id = dev_id++;
 
 	if (!fep->hwp) {
 		ret = -ENOMEM;
diff --git a/drivers/net/ethernet/freescale/fsl_pq_mdio.c b/drivers/net/ethernet/freescale/fsl_pq_mdio.c
index 52f4e8a..4d9f84b 100644
--- a/drivers/net/ethernet/freescale/fsl_pq_mdio.c
+++ b/drivers/net/ethernet/freescale/fsl_pq_mdio.c
@@ -183,28 +183,10 @@
 }
 EXPORT_SYMBOL_GPL(fsl_pq_mdio_bus_name);
 
-/* Scan the bus in reverse, looking for an empty spot */
-static int fsl_pq_mdio_find_free(struct mii_bus *new_bus)
-{
-	int i;
 
-	for (i = PHY_MAX_ADDR; i > 0; i--) {
-		u32 phy_id;
-
-		if (get_phy_id(new_bus, i, &phy_id))
-			return -1;
-
-		if (phy_id == 0xffffffff)
-			break;
-	}
-
-	return i;
-}
-
-
-#if defined(CONFIG_GIANFAR) || defined(CONFIG_GIANFAR_MODULE)
 static u32 __iomem *get_gfar_tbipa(struct fsl_pq_mdio __iomem *regs, struct device_node *np)
 {
+#if defined(CONFIG_GIANFAR) || defined(CONFIG_GIANFAR_MODULE)
 	struct gfar __iomem *enet_regs;
 
 	/*
@@ -220,15 +202,15 @@
 	} else if (of_device_is_compatible(np, "fsl,etsec2-mdio") ||
 			of_device_is_compatible(np, "fsl,etsec2-tbi")) {
 		return of_iomap(np, 1);
-	} else
-		return NULL;
-}
+	}
 #endif
+	return NULL;
+}
 
 
-#if defined(CONFIG_UCC_GETH) || defined(CONFIG_UCC_GETH_MODULE)
 static int get_ucc_id_for_range(u64 start, u64 end, u32 *ucc_id)
 {
+#if defined(CONFIG_UCC_GETH) || defined(CONFIG_UCC_GETH_MODULE)
 	struct device_node *np = NULL;
 	int err = 0;
 
@@ -261,9 +243,10 @@
 		return err;
 	else
 		return -EINVAL;
-}
+#else
+	return -ENODEV;
 #endif
-
+}
 
 static int fsl_pq_mdio_probe(struct platform_device *ofdev)
 {
@@ -339,19 +322,13 @@
 			of_device_is_compatible(np, "fsl,etsec2-mdio") ||
 			of_device_is_compatible(np, "fsl,etsec2-tbi") ||
 			of_device_is_compatible(np, "gianfar")) {
-#if defined(CONFIG_GIANFAR) || defined(CONFIG_GIANFAR_MODULE)
 		tbipa = get_gfar_tbipa(regs, np);
 		if (!tbipa) {
 			err = -EINVAL;
 			goto err_free_irqs;
 		}
-#else
-		err = -ENODEV;
-		goto err_free_irqs;
-#endif
 	} else if (of_device_is_compatible(np, "fsl,ucc-mdio") ||
 			of_device_is_compatible(np, "ucc_geth_phy")) {
-#if defined(CONFIG_UCC_GETH) || defined(CONFIG_UCC_GETH_MODULE)
 		u32 id;
 		static u32 mii_mng_master;
 
@@ -364,10 +341,6 @@
 			mii_mng_master = id;
 			ucc_set_qe_mux_mii_mng(id - 1);
 		}
-#else
-		err = -ENODEV;
-		goto err_free_irqs;
-#endif
 	} else {
 		err = -ENODEV;
 		goto err_free_irqs;
@@ -386,16 +359,6 @@
 	}
 
 	if (tbiaddr == -1) {
-		out_be32(tbipa, 0);
-
-		tbiaddr = fsl_pq_mdio_find_free(new_bus);
-	}
-
-	/*
-	 * We define TBIPA at 0 to be illegal, opting to fail for boards that
-	 * have PHYs at 1-31, rather than change tbipa and rescan.
-	 */
-	if (tbiaddr == 0) {
 		err = -EBUSY;
 
 		goto err_free_irqs;
diff --git a/drivers/net/ethernet/ibm/ehea/ehea.h b/drivers/net/ethernet/ibm/ehea/ehea.h
index 410d6a1..6650068 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea.h
+++ b/drivers/net/ethernet/ibm/ehea/ehea.h
@@ -61,9 +61,9 @@
 #ifdef EHEA_SMALL_QUEUES
 #define EHEA_MAX_CQE_COUNT      1023
 #define EHEA_DEF_ENTRIES_SQ     1023
-#define EHEA_DEF_ENTRIES_RQ1    4095
+#define EHEA_DEF_ENTRIES_RQ1    1023
 #define EHEA_DEF_ENTRIES_RQ2    1023
-#define EHEA_DEF_ENTRIES_RQ3    1023
+#define EHEA_DEF_ENTRIES_RQ3    511
 #else
 #define EHEA_MAX_CQE_COUNT      4080
 #define EHEA_DEF_ENTRIES_SQ     4080
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_main.c b/drivers/net/ethernet/ibm/ehea/ehea_main.c
index 37b70f7..bfeccbf 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_main.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_main.c
@@ -371,7 +371,8 @@
 out_herr:
 	free_page((unsigned long)cb2);
 resched:
-	schedule_delayed_work(&port->stats_work, msecs_to_jiffies(1000));
+	schedule_delayed_work(&port->stats_work,
+			      round_jiffies_relative(msecs_to_jiffies(1000)));
 }
 
 static void ehea_refill_rq1(struct ehea_port_res *pr, int index, int nr_of_wqes)
@@ -2434,7 +2435,8 @@
 	}
 
 	mutex_unlock(&port->port_lock);
-	schedule_delayed_work(&port->stats_work, msecs_to_jiffies(1000));
+	schedule_delayed_work(&port->stats_work,
+			      round_jiffies_relative(msecs_to_jiffies(1000)));
 
 	return ret;
 }
diff --git a/drivers/net/ethernet/ibm/iseries_veth.c b/drivers/net/ethernet/ibm/iseries_veth.c
index 4326681..acc31af 100644
--- a/drivers/net/ethernet/ibm/iseries_veth.c
+++ b/drivers/net/ethernet/ibm/iseries_veth.c
@@ -1421,7 +1421,7 @@
 
 		/* FIXME: do we need this? */
 		memset(local_list, 0, sizeof(local_list));
-		memset(remote_list, 0, sizeof(VETH_MAX_FRAMES_PER_MSG));
+		memset(remote_list, 0, sizeof(remote_list));
 
 		/* a 0 address marks the end of the valid entries */
 		if (senddata->addr[startchunk] == 0)
diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c
index 7becff1..76b8457 100644
--- a/drivers/net/ethernet/jme.c
+++ b/drivers/net/ethernet/jme.c
@@ -1745,6 +1745,112 @@
 }
 
 static int
+jme_phy_specreg_read(struct jme_adapter *jme, u32 specreg)
+{
+	u32 phy_addr;
+
+	phy_addr = JM_PHY_SPEC_REG_READ | specreg;
+	jme_mdio_write(jme->dev, jme->mii_if.phy_id, JM_PHY_SPEC_ADDR_REG,
+			phy_addr);
+	return jme_mdio_read(jme->dev, jme->mii_if.phy_id,
+			JM_PHY_SPEC_DATA_REG);
+}
+
+static void
+jme_phy_specreg_write(struct jme_adapter *jme, u32 ext_reg, u32 phy_data)
+{
+	u32 phy_addr;
+
+	phy_addr = JM_PHY_SPEC_REG_WRITE | ext_reg;
+	jme_mdio_write(jme->dev, jme->mii_if.phy_id, JM_PHY_SPEC_DATA_REG,
+			phy_data);
+	jme_mdio_write(jme->dev, jme->mii_if.phy_id, JM_PHY_SPEC_ADDR_REG,
+			phy_addr);
+}
+
+static int
+jme_phy_calibration(struct jme_adapter *jme)
+{
+	u32 ctrl1000, phy_data;
+
+	jme_phy_off(jme);
+	jme_phy_on(jme);
+	/*  Enabel PHY test mode 1 */
+	ctrl1000 = jme_mdio_read(jme->dev, jme->mii_if.phy_id, MII_CTRL1000);
+	ctrl1000 &= ~PHY_GAD_TEST_MODE_MSK;
+	ctrl1000 |= PHY_GAD_TEST_MODE_1;
+	jme_mdio_write(jme->dev, jme->mii_if.phy_id, MII_CTRL1000, ctrl1000);
+
+	phy_data = jme_phy_specreg_read(jme, JM_PHY_EXT_COMM_2_REG);
+	phy_data &= ~JM_PHY_EXT_COMM_2_CALI_MODE_0;
+	phy_data |= JM_PHY_EXT_COMM_2_CALI_LATCH |
+			JM_PHY_EXT_COMM_2_CALI_ENABLE;
+	jme_phy_specreg_write(jme, JM_PHY_EXT_COMM_2_REG, phy_data);
+	msleep(20);
+	phy_data = jme_phy_specreg_read(jme, JM_PHY_EXT_COMM_2_REG);
+	phy_data &= ~(JM_PHY_EXT_COMM_2_CALI_ENABLE |
+			JM_PHY_EXT_COMM_2_CALI_MODE_0 |
+			JM_PHY_EXT_COMM_2_CALI_LATCH);
+	jme_phy_specreg_write(jme, JM_PHY_EXT_COMM_2_REG, phy_data);
+
+	/*  Disable PHY test mode */
+	ctrl1000 = jme_mdio_read(jme->dev, jme->mii_if.phy_id, MII_CTRL1000);
+	ctrl1000 &= ~PHY_GAD_TEST_MODE_MSK;
+	jme_mdio_write(jme->dev, jme->mii_if.phy_id, MII_CTRL1000, ctrl1000);
+	return 0;
+}
+
+static int
+jme_phy_setEA(struct jme_adapter *jme)
+{
+	u32 phy_comm0 = 0, phy_comm1 = 0;
+	u8 nic_ctrl;
+
+	pci_read_config_byte(jme->pdev, PCI_PRIV_SHARE_NICCTRL, &nic_ctrl);
+	if ((nic_ctrl & 0x3) == JME_FLAG_PHYEA_ENABLE)
+		return 0;
+
+	switch (jme->pdev->device) {
+	case PCI_DEVICE_ID_JMICRON_JMC250:
+		if (((jme->chip_main_rev == 5) &&
+			((jme->chip_sub_rev == 0) || (jme->chip_sub_rev == 1) ||
+			(jme->chip_sub_rev == 3))) ||
+			(jme->chip_main_rev >= 6)) {
+			phy_comm0 = 0x008A;
+			phy_comm1 = 0x4109;
+		}
+		if ((jme->chip_main_rev == 3) &&
+			((jme->chip_sub_rev == 1) || (jme->chip_sub_rev == 2)))
+			phy_comm0 = 0xE088;
+		break;
+	case PCI_DEVICE_ID_JMICRON_JMC260:
+		if (((jme->chip_main_rev == 5) &&
+			((jme->chip_sub_rev == 0) || (jme->chip_sub_rev == 1) ||
+			(jme->chip_sub_rev == 3))) ||
+			(jme->chip_main_rev >= 6)) {
+			phy_comm0 = 0x008A;
+			phy_comm1 = 0x4109;
+		}
+		if ((jme->chip_main_rev == 3) &&
+			((jme->chip_sub_rev == 1) || (jme->chip_sub_rev == 2)))
+			phy_comm0 = 0xE088;
+		if ((jme->chip_main_rev == 2) && (jme->chip_sub_rev == 0))
+			phy_comm0 = 0x608A;
+		if ((jme->chip_main_rev == 2) && (jme->chip_sub_rev == 2))
+			phy_comm0 = 0x408A;
+		break;
+	default:
+		return -ENODEV;
+	}
+	if (phy_comm0)
+		jme_phy_specreg_write(jme, JM_PHY_EXT_COMM_0_REG, phy_comm0);
+	if (phy_comm1)
+		jme_phy_specreg_write(jme, JM_PHY_EXT_COMM_1_REG, phy_comm1);
+
+	return 0;
+}
+
+static int
 jme_open(struct net_device *netdev)
 {
 	struct jme_adapter *jme = netdev_priv(netdev);
@@ -1769,7 +1875,8 @@
 		jme_set_settings(netdev, &jme->old_ecmd);
 	else
 		jme_reset_phy_processor(jme);
-
+	jme_phy_calibration(jme);
+	jme_phy_setEA(jme);
 	jme_reset_link(jme);
 
 	return 0;
@@ -3184,7 +3291,8 @@
 		jme_set_settings(netdev, &jme->old_ecmd);
 	else
 		jme_reset_phy_processor(jme);
-
+	jme_phy_calibration(jme);
+	jme_phy_setEA(jme);
 	jme_start_irq(jme);
 	netif_device_attach(netdev);
 
@@ -3239,4 +3347,3 @@
 MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_VERSION);
 MODULE_DEVICE_TABLE(pci, jme_pci_tbl);
-
diff --git a/drivers/net/ethernet/jme.h b/drivers/net/ethernet/jme.h
index 02ea27c..4304072b 100644
--- a/drivers/net/ethernet/jme.h
+++ b/drivers/net/ethernet/jme.h
@@ -760,6 +760,25 @@
 				  RXMCS_CHECKSUM,
 };
 
+/*	Extern PHY common register 2	*/
+
+#define PHY_GAD_TEST_MODE_1			0x00002000
+#define PHY_GAD_TEST_MODE_MSK			0x0000E000
+#define JM_PHY_SPEC_REG_READ			0x00004000
+#define JM_PHY_SPEC_REG_WRITE			0x00008000
+#define PHY_CALIBRATION_DELAY			20
+#define JM_PHY_SPEC_ADDR_REG			0x1E
+#define JM_PHY_SPEC_DATA_REG			0x1F
+
+#define JM_PHY_EXT_COMM_0_REG			0x30
+#define JM_PHY_EXT_COMM_1_REG			0x31
+#define JM_PHY_EXT_COMM_2_REG			0x32
+#define JM_PHY_EXT_COMM_2_CALI_ENABLE		0x01
+#define JM_PHY_EXT_COMM_2_CALI_MODE_0		0x02
+#define JM_PHY_EXT_COMM_2_CALI_LATCH		0x10
+#define PCI_PRIV_SHARE_NICCTRL			0xF5
+#define JME_FLAG_PHYEA_ENABLE			0x2
+
 /*
  * Wakeup Frame setup interface registers
  */
diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c
index 6bb2b95..0b3567a 100644
--- a/drivers/net/ethernet/lantiq_etop.c
+++ b/drivers/net/ethernet/lantiq_etop.c
@@ -34,6 +34,8 @@
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/io.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
 
 #include <asm/checksum.h>
 
diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c
index fdc6c39..7803efa 100644
--- a/drivers/net/ethernet/marvell/sky2.c
+++ b/drivers/net/ethernet/marvell/sky2.c
@@ -50,7 +50,7 @@
 #include "sky2.h"
 
 #define DRV_NAME		"sky2"
-#define DRV_VERSION		"1.29"
+#define DRV_VERSION		"1.30"
 
 /*
  * The Yukon II chipset takes 64 bit command blocks (called list elements)
@@ -68,7 +68,7 @@
 #define MAX_SKB_TX_LE	(2 + (sizeof(dma_addr_t)/sizeof(u32))*(MAX_SKB_FRAGS+1))
 #define TX_MIN_PENDING		(MAX_SKB_TX_LE+1)
 #define TX_MAX_PENDING		1024
-#define TX_DEF_PENDING		127
+#define TX_DEF_PENDING		63
 
 #define TX_WATCHDOG		(5 * HZ)
 #define NAPI_WEIGHT		64
@@ -869,6 +869,7 @@
 
 	/* block receiver */
 	sky2_write8(hw, SK_REG(port, RX_GMF_CTRL_T), GMF_RST_SET);
+	sky2_read32(hw, B0_CTST);
 }
 
 static void sky2_set_tx_stfwd(struct sky2_hw *hw, unsigned port)
@@ -1274,6 +1275,14 @@
 		     ? BMU_ENA_RX_CHKSUM : BMU_DIS_RX_CHKSUM);
 }
 
+/*
+ * Fixed initial key as seed to RSS.
+ */
+static const uint32_t rss_init_key[10] = {
+	0x7c3351da, 0x51c5cf4e,	0x44adbdd1, 0xe8d38d18,	0x48897c43,
+	0xb1d60e7e, 0x6a3dd760, 0x01a2e453, 0x16f46f13, 0x1a0e7b30
+};
+
 /* Enable/disable receive hash calculation (RSS) */
 static void rx_set_rss(struct net_device *dev, u32 features)
 {
@@ -1289,12 +1298,9 @@
 
 	/* Program RSS initial values */
 	if (features & NETIF_F_RXHASH) {
-		u32 key[nkeys];
-
-		get_random_bytes(key, nkeys * sizeof(u32));
 		for (i = 0; i < nkeys; i++)
 			sky2_write32(hw, SK_REG(sky2->port, RSS_KEY + i * 4),
-				     key[i]);
+				     rss_init_key[i]);
 
 		/* Need to turn on (undocumented) flag to make hashing work  */
 		sky2_write32(hw, SK_REG(sky2->port, RX_GMF_CTRL_T),
@@ -1717,6 +1723,8 @@
 	if (err)
 		dev_err(&pdev->dev, "cannot assign irq %d\n", pdev->irq);
 	else {
+		hw->flags |= SKY2_HW_IRQ_SETUP;
+
 		napi_enable(&hw->napi);
 		sky2_write32(hw, B0_IMSK, Y2_IS_BASE);
 		sky2_read32(hw, B0_IMSK);
@@ -1727,7 +1735,7 @@
 
 
 /* Bring up network interface. */
-static int sky2_up(struct net_device *dev)
+static int sky2_open(struct net_device *dev)
 {
 	struct sky2_port *sky2 = netdev_priv(dev);
 	struct sky2_hw *hw = sky2->hw;
@@ -1747,6 +1755,11 @@
 
 	sky2_hw_up(sky2);
 
+	if (hw->chip_id == CHIP_ID_YUKON_OPT ||
+	    hw->chip_id == CHIP_ID_YUKON_PRM ||
+	    hw->chip_id == CHIP_ID_YUKON_OP_2)
+		imask |= Y2_IS_PHY_QLNK;	/* enable PHY Quick Link */
+
 	/* Enable interrupts from phy/mac for port */
 	imask = sky2_read32(hw, B0_IMSK);
 	imask |= portirq_msk[port];
@@ -2040,6 +2053,8 @@
 
 	sky2_write32(hw, RB_ADDR(txqaddr[port], RB_CTRL), RB_RST_SET);
 	sky2_write8(hw, SK_REG(port, TX_GMF_CTRL_T), GMF_RST_SET);
+
+	sky2_read32(hw, B0_CTST);
 }
 
 static void sky2_hw_down(struct sky2_port *sky2)
@@ -2090,7 +2105,7 @@
 }
 
 /* Network shutdown */
-static int sky2_down(struct net_device *dev)
+static int sky2_close(struct net_device *dev)
 {
 	struct sky2_port *sky2 = netdev_priv(dev);
 	struct sky2_hw *hw = sky2->hw;
@@ -2101,15 +2116,22 @@
 
 	netif_info(sky2, ifdown, dev, "disabling interface\n");
 
-	/* Disable port IRQ */
-	sky2_write32(hw, B0_IMSK,
-		     sky2_read32(hw, B0_IMSK) & ~portirq_msk[sky2->port]);
-	sky2_read32(hw, B0_IMSK);
-
 	if (hw->ports == 1) {
+		sky2_write32(hw, B0_IMSK, 0);
+		sky2_read32(hw, B0_IMSK);
+
 		napi_disable(&hw->napi);
 		free_irq(hw->pdev->irq, hw);
+		hw->flags &= ~SKY2_HW_IRQ_SETUP;
 	} else {
+		u32 imask;
+
+		/* Disable port IRQ */
+		imask  = sky2_read32(hw, B0_IMSK);
+		imask &= ~portirq_msk[sky2->port];
+		sky2_write32(hw, B0_IMSK, imask);
+		sky2_read32(hw, B0_IMSK);
+
 		synchronize_irq(hw->pdev->irq);
 		napi_synchronize(&hw->napi);
 	}
@@ -2587,7 +2609,7 @@
 	if (netif_running(dev)) {
 		sky2_tx_complete(sky2, last);
 
-		/* Wake unless it's detached, and called e.g. from sky2_down() */
+		/* Wake unless it's detached, and called e.g. from sky2_close() */
 		if (tx_avail(sky2) > MAX_SKB_TX_LE + 4)
 			netif_wake_queue(dev);
 	}
@@ -3258,7 +3280,6 @@
 	    hw->chip_id == CHIP_ID_YUKON_PRM ||
 	    hw->chip_id == CHIP_ID_YUKON_OP_2) {
 		u16 reg;
-		u32 msk;
 
 		if (hw->chip_id == CHIP_ID_YUKON_OPT && hw->chip_rev == 0) {
 			/* disable PCI-E PHY power down (set PHY reg 0x80, bit 7 */
@@ -3281,11 +3302,6 @@
 		sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
 		sky2_pci_write16(hw, PSM_CONFIG_REG4, reg);
 
-		/* enable PHY Quick Link */
-		msk = sky2_read32(hw, B0_IMSK);
-		msk |= Y2_IS_PHY_QLNK;
-		sky2_write32(hw, B0_IMSK, msk);
-
 		/* check if PSMv2 was running before */
 		reg = sky2_pci_read16(hw, PSM_CONFIG_REG3);
 		if (reg & PCI_EXP_LNKCTL_ASPMC)
@@ -3383,7 +3399,7 @@
 		netif_tx_lock(dev);
 		netif_device_detach(dev);	/* stop txq */
 		netif_tx_unlock(dev);
-		sky2_down(dev);
+		sky2_close(dev);
 	}
 }
 
@@ -3393,7 +3409,7 @@
 	int err = 0;
 
 	if (netif_running(dev)) {
-		err = sky2_up(dev);
+		err = sky2_open(dev);
 		if (err) {
 			netdev_info(dev, "could not restart %d\n", err);
 			dev_close(dev);
@@ -3410,10 +3426,13 @@
 {
 	int i;
 
-	sky2_read32(hw, B0_IMSK);
-	sky2_write32(hw, B0_IMSK, 0);
-	synchronize_irq(hw->pdev->irq);
-	napi_disable(&hw->napi);
+	if (hw->flags & SKY2_HW_IRQ_SETUP) {
+		sky2_read32(hw, B0_IMSK);
+		sky2_write32(hw, B0_IMSK, 0);
+
+		synchronize_irq(hw->pdev->irq);
+		napi_disable(&hw->napi);
+	}
 
 	for (i = 0; i < hw->ports; i++) {
 		struct net_device *dev = hw->dev[i];
@@ -3446,11 +3465,12 @@
 		netif_wake_queue(dev);
 	}
 
-	sky2_write32(hw, B0_IMSK, imask);
-	sky2_read32(hw, B0_IMSK);
-
-	sky2_read32(hw, B0_Y2_SP_LISR);
-	napi_enable(&hw->napi);
+	if (hw->flags & SKY2_HW_IRQ_SETUP) {
+		sky2_write32(hw, B0_IMSK, imask);
+		sky2_read32(hw, B0_IMSK);
+		sky2_read32(hw, B0_Y2_SP_LISR);
+		napi_enable(&hw->napi);
+	}
 }
 
 static void sky2_restart(struct work_struct *work)
@@ -4071,6 +4091,16 @@
 	return 0;
 }
 
+/*
+ * Hardware is limited to min of 128 and max of 2048 for ring size
+ * and  rounded up to next power of two
+ * to avoid division in modulus calclation
+ */
+static unsigned long roundup_ring_size(unsigned long pending)
+{
+	return max(128ul, roundup_pow_of_two(pending+1));
+}
+
 static void sky2_get_ringparam(struct net_device *dev,
 			       struct ethtool_ringparam *ering)
 {
@@ -4098,7 +4128,7 @@
 
 	sky2->rx_pending = ering->rx_pending;
 	sky2->tx_pending = ering->tx_pending;
-	sky2->tx_ring_size = roundup_pow_of_two(sky2->tx_pending+1);
+	sky2->tx_ring_size = roundup_ring_size(sky2->tx_pending);
 
 	return sky2_reattach(dev);
 }
@@ -4556,7 +4586,7 @@
 	struct net_device *dev = ptr;
 	struct sky2_port *sky2 = netdev_priv(dev);
 
-	if (dev->netdev_ops->ndo_open != sky2_up || !sky2_debug)
+	if (dev->netdev_ops->ndo_open != sky2_open || !sky2_debug)
 		return NOTIFY_DONE;
 
 	switch (event) {
@@ -4621,8 +4651,8 @@
    not allowing netpoll on second port */
 static const struct net_device_ops sky2_netdev_ops[2] = {
   {
-	.ndo_open		= sky2_up,
-	.ndo_stop		= sky2_down,
+	.ndo_open		= sky2_open,
+	.ndo_stop		= sky2_close,
 	.ndo_start_xmit		= sky2_xmit_frame,
 	.ndo_do_ioctl		= sky2_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
@@ -4638,8 +4668,8 @@
 #endif
   },
   {
-	.ndo_open		= sky2_up,
-	.ndo_stop		= sky2_down,
+	.ndo_open		= sky2_open,
+	.ndo_stop		= sky2_close,
 	.ndo_start_xmit		= sky2_xmit_frame,
 	.ndo_do_ioctl		= sky2_ioctl,
 	.ndo_validate_addr	= eth_validate_addr,
@@ -4692,7 +4722,7 @@
 	spin_lock_init(&sky2->phy_lock);
 
 	sky2->tx_pending = TX_DEF_PENDING;
-	sky2->tx_ring_size = roundup_pow_of_two(TX_DEF_PENDING+1);
+	sky2->tx_ring_size = roundup_ring_size(TX_DEF_PENDING);
 	sky2->rx_pending = RX_DEF_PENDING;
 
 	hw->dev[port] = dev;
diff --git a/drivers/net/ethernet/marvell/sky2.h b/drivers/net/ethernet/marvell/sky2.h
index 0af31b8..ff6f58b 100644
--- a/drivers/net/ethernet/marvell/sky2.h
+++ b/drivers/net/ethernet/marvell/sky2.h
@@ -2287,6 +2287,7 @@
 #define SKY2_HW_RSS_BROKEN	0x00000100
 #define SKY2_HW_VLAN_BROKEN     0x00000200
 #define SKY2_HW_RSS_CHKSUM	0x00000400	/* RSS requires chksum */
+#define SKY2_HW_IRQ_SETUP	0x00000800
 
 	u8	     	     chip_id;
 	u8		     chip_rev;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index b89c36d..c2df6c3 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -581,6 +581,7 @@
 		 * Packet is OK - process it.
 		 */
 		length = be32_to_cpu(cqe->byte_cnt);
+		length -= ring->fcs_del;
 		ring->bytes += length;
 		ring->packets++;
 
@@ -813,8 +814,11 @@
 	context->db_rec_addr = cpu_to_be64(ring->wqres.db.dma);
 
 	/* Cancel FCS removal if FW allows */
-	if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP)
+	if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP) {
 		context->param3 |= cpu_to_be32(1 << 29);
+		ring->fcs_del = ETH_FCS_LEN;
+	} else
+		ring->fcs_del = 0;
 
 	err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, context, qp, state);
 	if (err) {
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index 8fda331c..207b5ad 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -272,6 +272,7 @@
 	u32 prod;
 	u32 cons;
 	u32 buf_size;
+	u8  fcs_del;
 	void *buf;
 	void *rx_info;
 	unsigned long bytes;
diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c
index 1dca570..1c61d36 100644
--- a/drivers/net/ethernet/nvidia/forcedeth.c
+++ b/drivers/net/ethernet/nvidia/forcedeth.c
@@ -609,7 +609,7 @@
 };
 
 static const struct nv_ethtool_str nv_estats_str[] = {
-	{ "tx_bytes" },
+	{ "tx_bytes" }, /* includes Ethernet FCS CRC */
 	{ "tx_zero_rexmt" },
 	{ "tx_one_rexmt" },
 	{ "tx_many_rexmt" },
@@ -637,7 +637,7 @@
 	/* version 2 stats */
 	{ "tx_deferral" },
 	{ "tx_packets" },
-	{ "rx_bytes" },
+	{ "rx_bytes" }, /* includes Ethernet FCS CRC */
 	{ "tx_pause" },
 	{ "rx_pause" },
 	{ "rx_drop_frame" },
@@ -649,7 +649,7 @@
 };
 
 struct nv_ethtool_stats {
-	u64 tx_bytes;
+	u64 tx_bytes; /* should be ifconfig->tx_bytes + 4*tx_packets */
 	u64 tx_zero_rexmt;
 	u64 tx_one_rexmt;
 	u64 tx_many_rexmt;
@@ -670,14 +670,14 @@
 	u64 rx_unicast;
 	u64 rx_multicast;
 	u64 rx_broadcast;
-	u64 rx_packets;
+	u64 rx_packets; /* should be ifconfig->rx_packets */
 	u64 rx_errors_total;
 	u64 tx_errors_total;
 
 	/* version 2 stats */
 	u64 tx_deferral;
-	u64 tx_packets;
-	u64 rx_bytes;
+	u64 tx_packets; /* should be ifconfig->tx_packets */
+	u64 rx_bytes;   /* should be ifconfig->rx_bytes + 4*rx_packets */
 	u64 tx_pause;
 	u64 rx_pause;
 	u64 rx_drop_frame;
@@ -1706,10 +1706,17 @@
 	if (np->driver_data & (DEV_HAS_STATISTICS_V1|DEV_HAS_STATISTICS_V2|DEV_HAS_STATISTICS_V3)) {
 		nv_get_hw_stats(dev);
 
+		/*
+		 * Note: because HW stats are not always available and
+		 * for consistency reasons, the following ifconfig
+		 * stats are managed by software: rx_bytes, tx_bytes,
+		 * rx_packets and tx_packets. The related hardware
+		 * stats reported by ethtool should be equivalent to
+		 * these ifconfig stats, with 4 additional bytes per
+		 * packet (Ethernet FCS CRC).
+		 */
+
 		/* copy to net_device stats */
-		dev->stats.tx_packets = np->estats.tx_packets;
-		dev->stats.rx_bytes = np->estats.rx_bytes;
-		dev->stats.tx_bytes = np->estats.tx_bytes;
 		dev->stats.tx_fifo_errors = np->estats.tx_fifo_errors;
 		dev->stats.tx_carrier_errors = np->estats.tx_carrier_errors;
 		dev->stats.rx_crc_errors = np->estats.rx_crc_errors;
@@ -2380,6 +2387,9 @@
 				if (flags & NV_TX_ERROR) {
 					if ((flags & NV_TX_RETRYERROR) && !(flags & NV_TX_RETRYCOUNT_MASK))
 						nv_legacybackoff_reseed(dev);
+				} else {
+					dev->stats.tx_packets++;
+					dev->stats.tx_bytes += np->get_tx_ctx->skb->len;
 				}
 				dev_kfree_skb_any(np->get_tx_ctx->skb);
 				np->get_tx_ctx->skb = NULL;
@@ -2390,6 +2400,9 @@
 				if (flags & NV_TX2_ERROR) {
 					if ((flags & NV_TX2_RETRYERROR) && !(flags & NV_TX2_RETRYCOUNT_MASK))
 						nv_legacybackoff_reseed(dev);
+				} else {
+					dev->stats.tx_packets++;
+					dev->stats.tx_bytes += np->get_tx_ctx->skb->len;
 				}
 				dev_kfree_skb_any(np->get_tx_ctx->skb);
 				np->get_tx_ctx->skb = NULL;
@@ -2429,6 +2442,9 @@
 					else
 						nv_legacybackoff_reseed(dev);
 				}
+			} else {
+				dev->stats.tx_packets++;
+				dev->stats.tx_bytes += np->get_tx_ctx->skb->len;
 			}
 
 			dev_kfree_skb_any(np->get_tx_ctx->skb);
@@ -2678,6 +2694,7 @@
 		skb->protocol = eth_type_trans(skb, dev);
 		napi_gro_receive(&np->napi, skb);
 		dev->stats.rx_packets++;
+		dev->stats.rx_bytes += len;
 next_pkt:
 		if (unlikely(np->get_rx.orig++ == np->last_rx.orig))
 			np->get_rx.orig = np->first_rx.orig;
@@ -2761,6 +2778,7 @@
 			}
 			napi_gro_receive(&np->napi, skb);
 			dev->stats.rx_packets++;
+			dev->stats.rx_bytes += len;
 		} else {
 			dev_kfree_skb(skb);
 		}
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c
index 9c075ea..9cb5f91 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_param.c
@@ -18,8 +18,8 @@
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307, USA.
  */
 
-#include <linux/module.h>	/* for __MODULE_STRING */
 #include "pch_gbe.h"
+#include <linux/module.h>	/* for __MODULE_STRING */
 
 #define OPTION_UNSET   -1
 #define OPTION_DISABLED 0
diff --git a/drivers/net/ethernet/pasemi/Makefile b/drivers/net/ethernet/pasemi/Makefile
index 05db543..90497ff 100644
--- a/drivers/net/ethernet/pasemi/Makefile
+++ b/drivers/net/ethernet/pasemi/Makefile
@@ -2,4 +2,5 @@
 # Makefile for the A Semi network device drivers.
 #
 
-obj-$(CONFIG_PASEMI_MAC) += pasemi_mac.o pasemi_mac_ethtool.o
+obj-$(CONFIG_PASEMI_MAC) += pasemi_mac_driver.o
+pasemi_mac_driver-objs := pasemi_mac.o pasemi_mac_ethtool.o
diff --git a/drivers/net/ethernet/qlogic/qlge/qlge.h b/drivers/net/ethernet/qlogic/qlge/qlge.h
index 8731f79..b8478aa 100644
--- a/drivers/net/ethernet/qlogic/qlge/qlge.h
+++ b/drivers/net/ethernet/qlogic/qlge/qlge.h
@@ -58,10 +58,8 @@
 
 
 #define TX_DESC_PER_IOCB 8
-/* The maximum number of frags we handle is based
- * on PAGE_SIZE...
- */
-#if (PAGE_SHIFT == 12) || (PAGE_SHIFT == 13)	/* 4k & 8k pages */
+
+#if ((MAX_SKB_FRAGS - TX_DESC_PER_IOCB) + 2) > 0
 #define TX_DESC_PER_OAL ((MAX_SKB_FRAGS - TX_DESC_PER_IOCB) + 2)
 #else /* all other page sizes */
 #define TX_DESC_PER_OAL 0
@@ -1353,7 +1351,7 @@
 	struct ob_mac_iocb_req *queue_entry;
 	u32 index;
 	struct oal oal;
-	struct map_list map[MAX_SKB_FRAGS + 1];
+	struct map_list map[MAX_SKB_FRAGS + 2];
 	int map_cnt;
 	struct tx_ring_desc *next;
 };
diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c
index 1fc01ca..4bf68cf 100644
--- a/drivers/net/ethernet/rdc/r6040.c
+++ b/drivers/net/ethernet/rdc/r6040.c
@@ -940,7 +940,7 @@
 	iowrite16(lp->mcr0, ioaddr + MCR0);
 
 	/* Fill the MAC hash tables with their values */
-	if (lp->mcr0 && MCR0_HASH_EN) {
+	if (lp->mcr0 & MCR0_HASH_EN) {
 		iowrite16(hash_table[0], ioaddr + MAR0);
 		iowrite16(hash_table[1], ioaddr + MAR1);
 		iowrite16(hash_table[2], ioaddr + MAR2);
diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index 92b45f0..67bf078 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -1183,11 +1183,13 @@
 	return value;
 }
 
-static void rtl8169_irq_mask_and_ack(void __iomem *ioaddr)
+static void rtl8169_irq_mask_and_ack(struct rtl8169_private *tp)
 {
-	RTL_W16(IntrMask, 0x0000);
+	void __iomem *ioaddr = tp->mmio_addr;
 
-	RTL_W16(IntrStatus, 0xffff);
+	RTL_W16(IntrMask, 0x0000);
+	RTL_W16(IntrStatus, tp->intr_event);
+	RTL_R8(ChipCmd);
 }
 
 static unsigned int rtl8169_tbi_reset_pending(struct rtl8169_private *tp)
@@ -1292,7 +1294,7 @@
 		netif_carrier_off(dev);
 		netif_info(tp, ifdown, dev, "link down\n");
 		if (pm)
-			pm_schedule_suspend(&tp->pci_dev->dev, 100);
+			pm_schedule_suspend(&tp->pci_dev->dev, 5000);
 	}
 	spin_unlock_irqrestore(&tp->lock, flags);
 }
@@ -3933,8 +3935,6 @@
 			break;
 		udelay(100);
 	}
-
-	rtl8169_init_ring_indexes(tp);
 }
 
 static int __devinit
@@ -4339,7 +4339,7 @@
 	void __iomem *ioaddr = tp->mmio_addr;
 
 	/* Disable interrupts */
-	rtl8169_irq_mask_and_ack(ioaddr);
+	rtl8169_irq_mask_and_ack(tp);
 
 	rtl_rx_close(tp);
 
@@ -4885,8 +4885,7 @@
 	RTL_W16(IntrMitigate, 0x5151);
 
 	/* Work around for RxFIFO overflow. */
-	if (tp->mac_version == RTL_GIGA_MAC_VER_11 ||
-	    tp->mac_version == RTL_GIGA_MAC_VER_22) {
+	if (tp->mac_version == RTL_GIGA_MAC_VER_11) {
 		tp->intr_event |= RxFIFOOver | PCSTimeout;
 		tp->intr_event &= ~RxOverflow;
 	}
@@ -5076,6 +5075,11 @@
 	void __iomem *ioaddr = tp->mmio_addr;
 	struct pci_dev *pdev = tp->pci_dev;
 
+	if (tp->mac_version >= RTL_GIGA_MAC_VER_30) {
+		tp->intr_event &= ~RxFIFOOver;
+		tp->napi_event &= ~RxFIFOOver;
+	}
+
 	if (tp->mac_version == RTL_GIGA_MAC_VER_13 ||
 	    tp->mac_version == RTL_GIGA_MAC_VER_16) {
 		int cap = pci_pcie_cap(pdev);
@@ -5342,7 +5346,7 @@
 	/* Wait for any pending NAPI task to complete */
 	napi_disable(&tp->napi);
 
-	rtl8169_irq_mask_and_ack(ioaddr);
+	rtl8169_irq_mask_and_ack(tp);
 
 	tp->intr_mask = 0xffff;
 	RTL_W16(IntrMask, tp->intr_event);
@@ -5389,14 +5393,16 @@
 	if (!netif_running(dev))
 		goto out_unlock;
 
+	rtl8169_hw_reset(tp);
+
 	rtl8169_wait_for_quiescence(dev);
 
 	for (i = 0; i < NUM_RX_DESC; i++)
 		rtl8169_mark_to_asic(tp->RxDescArray + i, rx_buf_sz);
 
 	rtl8169_tx_clear(tp);
+	rtl8169_init_ring_indexes(tp);
 
-	rtl8169_hw_reset(tp);
 	rtl_hw_start(dev);
 	netif_wake_queue(dev);
 	rtl8169_check_link_status(dev, tp, tp->mmio_addr);
@@ -5407,11 +5413,6 @@
 
 static void rtl8169_tx_timeout(struct net_device *dev)
 {
-	struct rtl8169_private *tp = netdev_priv(dev);
-
-	rtl8169_hw_reset(tp);
-
-	/* Let's wait a bit while any (async) irq lands on */
 	rtl8169_schedule_work(dev, rtl8169_reset_task);
 }
 
@@ -5804,6 +5805,10 @@
 	 */
 	status = RTL_R16(IntrStatus);
 	while (status && status != 0xffff) {
+		status &= tp->intr_event;
+		if (!status)
+			break;
+
 		handled = 1;
 
 		/* Handle all of the error cases first. These will reset
@@ -5818,27 +5823,9 @@
 			switch (tp->mac_version) {
 			/* Work around for rx fifo overflow */
 			case RTL_GIGA_MAC_VER_11:
-			case RTL_GIGA_MAC_VER_22:
-			case RTL_GIGA_MAC_VER_26:
 				netif_stop_queue(dev);
 				rtl8169_tx_timeout(dev);
 				goto done;
-			/* Testers needed. */
-			case RTL_GIGA_MAC_VER_17:
-			case RTL_GIGA_MAC_VER_19:
-			case RTL_GIGA_MAC_VER_20:
-			case RTL_GIGA_MAC_VER_21:
-			case RTL_GIGA_MAC_VER_23:
-			case RTL_GIGA_MAC_VER_24:
-			case RTL_GIGA_MAC_VER_27:
-			case RTL_GIGA_MAC_VER_28:
-			case RTL_GIGA_MAC_VER_31:
-			/* Experimental science. Pktgen proof. */
-			case RTL_GIGA_MAC_VER_12:
-			case RTL_GIGA_MAC_VER_25:
-				if (status == RxFIFOOver)
-					goto done;
-				break;
 			default:
 				break;
 			}
diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c
index d2be42a..8843071 100644
--- a/drivers/net/ethernet/smsc/smsc911x.c
+++ b/drivers/net/ethernet/smsc/smsc911x.c
@@ -1937,6 +1937,7 @@
 {
 	struct smsc911x_data *pdata = netdev_priv(dev);
 	unsigned int byte_test;
+	unsigned int to = 100;
 
 	SMSC_TRACE(pdata, probe, "Driver Parameters:");
 	SMSC_TRACE(pdata, probe, "LAN base: 0x%08lX",
@@ -1952,6 +1953,17 @@
 		return -ENODEV;
 	}
 
+	/*
+	 * poll the READY bit in PMT_CTRL. Any other access to the device is
+	 * forbidden while this bit isn't set. Try for 100ms
+	 */
+	while (!(smsc911x_reg_read(pdata, PMT_CTRL) & PMT_CTRL_READY_) && --to)
+		udelay(1000);
+	if (to == 0) {
+		pr_err("Device not READY in 100ms aborting\n");
+		return -ENODEV;
+	}
+
 	/* Check byte ordering */
 	byte_test = smsc911x_reg_read(pdata, BYTE_TEST);
 	SMSC_TRACE(pdata, probe, "BYTE_TEST: 0x%08X", byte_test);
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
index da66ac5..4d5402a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
@@ -39,10 +39,11 @@
 	/* DMA SW reset */
 	value |= DMA_BUS_MODE_SFT_RESET;
 	writel(value, ioaddr + DMA_BUS_MODE);
-	limit = 15000;
+	limit = 10;
 	while (limit--) {
 		if (!(readl(ioaddr + DMA_BUS_MODE) & DMA_BUS_MODE_SFT_RESET))
 			break;
+		mdelay(10);
 	}
 	if (limit < 0)
 		return -EBUSY;
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
index 627f656..bc17fd0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
@@ -41,10 +41,11 @@
 	/* DMA SW reset */
 	value |= DMA_BUS_MODE_SFT_RESET;
 	writel(value, ioaddr + DMA_BUS_MODE);
-	limit = 15000;
+	limit = 10;
 	while (limit--) {
 		if (!(readl(ioaddr + DMA_BUS_MODE) & DMA_BUS_MODE_SFT_RESET))
 			break;
+		mdelay(10);
 	}
 	if (limit < 0)
 		return -EBUSY;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 9bafa6c..a140a8f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -72,7 +72,6 @@
 	spinlock_t lock;
 	spinlock_t tx_lock;
 	int wolopts;
-	int wolenabled;
 	int wol_irq;
 #ifdef CONFIG_STMMAC_TIMER
 	struct stmmac_timer *tm;
@@ -80,6 +79,7 @@
 	struct plat_stmmacenet_data *plat;
 	struct stmmac_counters mmc;
 	struct dma_features dma_cap;
+	int hw_cap_support;
 };
 
 extern int stmmac_mdio_unregister(struct net_device *ndev);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index e8eff09..0395f9e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -430,6 +430,12 @@
 	struct stmmac_priv *priv = netdev_priv(dev);
 	u32 support = WAKE_MAGIC | WAKE_UCAST;
 
+	/* By default almost all GMAC devices support the WoL via
+	 * magic frame but we can disable it if the HW capability
+	 * register shows no support for pmt_magic_frame. */
+	if ((priv->hw_cap_support) && (!priv->dma_cap.pmt_magic_frame))
+		wol->wolopts &= ~WAKE_MAGIC;
+
 	if (!device_can_wakeup(priv->device))
 		return -EINVAL;
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 20546bb..72cd190 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -321,12 +321,10 @@
 	}
 
 	/* Stop Advertising 1000BASE Capability if interface is not GMII */
-	if ((interface) && ((interface == PHY_INTERFACE_MODE_MII) ||
-	    (interface == PHY_INTERFACE_MODE_RMII))) {
-		phydev->supported &= (PHY_BASIC_FEATURES | SUPPORTED_Pause |
-				      SUPPORTED_Asym_Pause);
-		phydev->advertising = phydev->supported;
-	}
+	if ((interface == PHY_INTERFACE_MODE_MII) ||
+	    (interface == PHY_INTERFACE_MODE_RMII))
+		phydev->advertising &= ~(SUPPORTED_1000baseT_Half |
+					 SUPPORTED_1000baseT_Full);
 
 	/*
 	 * Broken HW is sometimes missing the pull-up resistor on the
@@ -783,10 +781,15 @@
 	unsigned int mode = MMC_CNTRL_RESET_ON_READ | MMC_CNTRL_COUNTER_RESET |
 			    MMC_CNTRL_PRESET | MMC_CNTRL_FULL_HALF_PRESET;
 
-	/* Do not manage MMC IRQ (FIXME) */
+	/* Mask MMC irq, counters are managed in SW and registers
+	 * are cleared on each READ eventually. */
 	dwmac_mmc_intr_all_mask(priv->ioaddr);
-	dwmac_mmc_ctrl(priv->ioaddr, mode);
-	memset(&priv->mmc, 0, sizeof(struct stmmac_counters));
+
+	if (priv->dma_cap.rmon) {
+		dwmac_mmc_ctrl(priv->ioaddr, mode);
+		memset(&priv->mmc, 0, sizeof(struct stmmac_counters));
+	} else
+		pr_info(" No MAC Management Counters available");
 }
 
 static u32 stmmac_get_synopsys_id(struct stmmac_priv *priv)
@@ -807,8 +810,29 @@
 	return 0;
 }
 
-/* New GMAC chips support a new register to indicate the
- * presence of the optional feature/functions.
+/**
+ * stmmac_selec_desc_mode
+ * @dev : device pointer
+ * Description: select the Enhanced/Alternate or Normal descriptors */
+static void stmmac_selec_desc_mode(struct stmmac_priv *priv)
+{
+	if (priv->plat->enh_desc) {
+		pr_info(" Enhanced/Alternate descriptors\n");
+		priv->hw->desc = &enh_desc_ops;
+	} else {
+		pr_info(" Normal descriptors\n");
+		priv->hw->desc = &ndesc_ops;
+	}
+}
+
+/**
+ * stmmac_get_hw_features
+ * @priv : private device pointer
+ * Description:
+ *  new GMAC chip generations have a new register to indicate the
+ *  presence of the optional feature/functions.
+ *  This can be also used to override the value passed through the
+ *  platform and necessary for old MAC10/100 and GMAC chips.
  */
 static int stmmac_get_hw_features(struct stmmac_priv *priv)
 {
@@ -829,7 +853,7 @@
 			(hw_cap & DMA_HW_FEAT_RWKSEL) >> 9;
 		priv->dma_cap.pmt_magic_frame =
 			(hw_cap & DMA_HW_FEAT_MGKSEL) >> 10;
-		/*MMC*/
+		/* MMC */
 		priv->dma_cap.rmon = (hw_cap & DMA_HW_FEAT_MMCSEL) >> 11;
 		/* IEEE 1588-2002*/
 		priv->dma_cap.time_stamp =
@@ -857,8 +881,7 @@
 		priv->dma_cap.enh_desc =
 			(hw_cap & DMA_HW_FEAT_ENHDESSEL) >> 24;
 
-	} else
-		pr_debug("\tNo HW DMA feature register supported");
+	}
 
 	return hw_cap;
 }
@@ -913,6 +936,44 @@
 		goto open_error;
 	}
 
+	stmmac_get_synopsys_id(priv);
+
+	priv->hw_cap_support = stmmac_get_hw_features(priv);
+
+	if (priv->hw_cap_support) {
+		pr_info(" Support DMA HW capability register");
+
+		/* We can override some gmac/dma configuration fields: e.g.
+		 * enh_desc, tx_coe (e.g. that are passed through the
+		 * platform) with the values from the HW capability
+		 * register (if supported).
+		 */
+		priv->plat->enh_desc = priv->dma_cap.enh_desc;
+		priv->plat->tx_coe = priv->dma_cap.tx_coe;
+		priv->plat->pmt = priv->dma_cap.pmt_remote_wake_up;
+
+		/* By default disable wol on magic frame if not supported */
+		if (!priv->dma_cap.pmt_magic_frame)
+			priv->wolopts &= ~WAKE_MAGIC;
+
+	} else
+		pr_info(" No HW DMA feature register supported");
+
+	/* Select the enhnaced/normal descriptor structures */
+	stmmac_selec_desc_mode(priv);
+
+	/* PMT module is not integrated in all the MAC devices. */
+	if (priv->plat->pmt) {
+		pr_info(" Remote wake-up capable\n");
+		device_set_wakeup_capable(priv->device, 1);
+	}
+
+	priv->rx_coe = priv->hw->mac->rx_coe(priv->ioaddr);
+	if (priv->rx_coe)
+		pr_info(" Checksum Offload Engine supported\n");
+	if (priv->plat->tx_coe)
+		pr_info(" Checksum insertion supported\n");
+
 	/* Create and initialize the TX/RX descriptors chains. */
 	priv->dma_tx_size = STMMAC_ALIGN(dma_txsize);
 	priv->dma_rx_size = STMMAC_ALIGN(dma_rxsize);
@@ -935,15 +996,6 @@
 	/* Initialize the MAC Core */
 	priv->hw->mac->core_init(priv->ioaddr);
 
-	stmmac_get_synopsys_id(priv);
-
-	stmmac_get_hw_features(priv);
-
-	priv->rx_coe = priv->hw->mac->rx_coe(priv->ioaddr);
-	if (priv->rx_coe)
-		pr_info("stmmac: Rx Checksum Offload Engine supported\n");
-	if (priv->plat->tx_coe)
-		pr_info("\tTX Checksum insertion supported\n");
 	netdev_update_features(dev);
 
 	/* Request the IRQ lines */
@@ -965,8 +1017,7 @@
 	memset(&priv->xstats, 0, sizeof(struct stmmac_extra_stats));
 	priv->xstats.threshold = tc;
 
-	if (priv->dma_cap.rmon)
-		stmmac_mmc_setup(priv);
+	stmmac_mmc_setup(priv);
 
 	/* Start the ball rolling... */
 	DBG(probe, DEBUG, "%s: DMA RX/TX processes started...\n", dev->name);
@@ -1489,9 +1540,7 @@
 	if (!priv->phydev)
 		return -EINVAL;
 
-	spin_lock(&priv->lock);
 	ret = phy_mii_ioctl(priv->phydev, rq, cmd);
-	spin_unlock(&priv->lock);
 
 	return ret;
 }
@@ -1558,7 +1607,7 @@
 	struct net_device *dev = seq->private;
 	struct stmmac_priv *priv = netdev_priv(dev);
 
-	if (!stmmac_get_hw_features(priv)) {
+	if (!priv->hw_cap_support) {
 		seq_printf(seq, "DMA HW features not supported\n");
 		return 0;
 	}
@@ -1766,12 +1815,6 @@
 	if (!device)
 		return -ENOMEM;
 
-	if (priv->plat->enh_desc) {
-		device->desc = &enh_desc_ops;
-		pr_info("\tEnhanced descriptor structure\n");
-	} else
-		device->desc = &ndesc_ops;
-
 	priv->hw = device;
 	priv->hw->ring = &ring_mode_ops;
 
@@ -1845,11 +1888,6 @@
 
 	priv->ioaddr = addr;
 
-	/* PMT module is not integrated in all the MAC devices. */
-	if (plat_dat->pmt) {
-		pr_info("\tPMT module supported\n");
-		device_set_wakeup_capable(&pdev->dev, 1);
-	}
 	/*
 	 * On some platforms e.g. SPEAr the wake up irq differs from the mac irq
 	 * The external wake up irq can be passed through the platform code
@@ -1862,7 +1900,6 @@
 	if (priv->wol_irq == -ENXIO)
 		priv->wol_irq = ndev->irq;
 
-
 	platform_set_drvdata(pdev, ndev);
 
 	/* Set the I/O base addr */
@@ -1875,7 +1912,7 @@
 			goto out_free_ndev;
 	}
 
-	/* MAC HW revice detection */
+	/* MAC HW device detection */
 	ret = stmmac_mac_device_setup(ndev);
 	if (ret < 0)
 		goto out_plat_exit;
@@ -1978,12 +2015,13 @@
 	if (!ndev || !netif_running(ndev))
 		return 0;
 
+	if (priv->phydev)
+		phy_stop(priv->phydev);
+
 	spin_lock(&priv->lock);
 
 	netif_device_detach(ndev);
 	netif_stop_queue(ndev);
-	if (priv->phydev)
-		phy_stop(priv->phydev);
 
 #ifdef CONFIG_STMMAC_TIMER
 	priv->tm->timer_stop();
@@ -2041,12 +2079,13 @@
 #endif
 	napi_enable(&priv->napi);
 
-	if (priv->phydev)
-		phy_start(priv->phydev);
-
 	netif_start_queue(ndev);
 
 	spin_unlock(&priv->lock);
+
+	if (priv->phydev)
+		phy_start(priv->phydev);
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c
index c517dac..cf14ab9 100644
--- a/drivers/net/ethernet/sun/sunhme.c
+++ b/drivers/net/ethernet/sun/sunhme.c
@@ -2637,7 +2637,7 @@
 	sbus_dp = op->dev.parent->of_node;
 
 	/* We can match PCI devices too, do not accept those here. */
-	if (strcmp(sbus_dp->name, "sbus"))
+	if (strcmp(sbus_dp->name, "sbus") && strcmp(sbus_dp->name, "sbi"))
 		return err;
 
 	if (is_qfe) {
diff --git a/drivers/net/ethernet/tile/tilepro.c b/drivers/net/ethernet/tile/tilepro.c
index 10826d8..1187a11 100644
--- a/drivers/net/ethernet/tile/tilepro.c
+++ b/drivers/net/ethernet/tile/tilepro.c
@@ -926,7 +926,7 @@
 		goto done;
 
 	/* Re-enable the ingress interrupt. */
-	enable_percpu_irq(priv->intr_id);
+	enable_percpu_irq(priv->intr_id, 0);
 
 	/* HACK: Avoid the "rotting packet" problem (see above). */
 	if (qup->__packet_receive_read !=
@@ -1296,7 +1296,7 @@
 	info->napi_enabled = true;
 
 	/* Enable the ingress interrupt. */
-	enable_percpu_irq(priv->intr_id);
+	enable_percpu_irq(priv->intr_id, 0);
 }
 
 
@@ -1697,7 +1697,7 @@
 	for (i = 0; i < sh->nr_frags; i++) {
 
 		skb_frag_t *f = &sh->frags[i];
-		unsigned long pfn = page_to_pfn(f->page);
+		unsigned long pfn = page_to_pfn(skb_frag_page(f));
 
 		/* FIXME: Compute "hash_for_home" properly. */
 		/* ISSUE: The hypervisor checks CHIP_HAS_REV1_DMA_PACKETS(). */
@@ -1706,7 +1706,7 @@
 		/* FIXME: Hmmm. */
 		if (!hash_default) {
 			void *va = pfn_to_kaddr(pfn) + f->page_offset;
-			BUG_ON(PageHighMem(f->page));
+			BUG_ON(PageHighMem(skb_frag_page(f)));
 			finv_buffer_remote(va, f->size, 0);
 		}
 
diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c
index caf3659..2681b53 100644
--- a/drivers/net/ethernet/xilinx/ll_temac_main.c
+++ b/drivers/net/ethernet/xilinx/ll_temac_main.c
@@ -114,6 +114,7 @@
 		return;
 	temac_iow(lp, XTE_LSW0_OFFSET, value);
 	temac_iow(lp, XTE_CTL0_OFFSET, CNTLREG_WRITE_ENABLE_MASK | reg);
+	temac_indirect_busywait(lp);
 }
 
 /**
@@ -203,6 +204,9 @@
 	struct temac_local *lp = netdev_priv(ndev);
 	int i;
 
+	/* Reset Local Link (DMA) */
+	lp->dma_out(lp, DMA_CONTROL_REG, DMA_CONTROL_RST);
+
 	for (i = 0; i < RX_BD_NUM; i++) {
 		if (!lp->rx_skb[i])
 			break;
@@ -860,6 +864,8 @@
 		phy_start(lp->phy_dev);
 	}
 
+	temac_device_reset(ndev);
+
 	rc = request_irq(lp->tx_irq, ll_temac_tx_irq, 0, ndev->name, ndev);
 	if (rc)
 		goto err_tx_irq;
@@ -867,7 +873,6 @@
 	if (rc)
 		goto err_rx_irq;
 
-	temac_device_reset(ndev);
 	return 0;
 
  err_rx_irq:
diff --git a/drivers/net/hippi/Kconfig b/drivers/net/hippi/Kconfig
index 7393eb7..95eb34f 100644
--- a/drivers/net/hippi/Kconfig
+++ b/drivers/net/hippi/Kconfig
@@ -36,4 +36,4 @@
 	  kernel code or by user space programs. Say Y here only if you have
 	  the memory.
 
-endif /* HIPPI */
+endif # HIPPI
diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index bb88e12..a702443 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -3,7 +3,7 @@
 #
 
 menuconfig PHYLIB
-	bool "PHY Device support and infrastructure"
+	tristate "PHY Device support and infrastructure"
 	depends on !S390
 	depends on NETDEVICES
 	help
diff --git a/drivers/net/ppp/pptp.c b/drivers/net/ppp/pptp.c
index 89f829f..f8a6853 100644
--- a/drivers/net/ppp/pptp.c
+++ b/drivers/net/ppp/pptp.c
@@ -423,10 +423,8 @@
 	lock_sock(sk);
 
 	opt->src_addr = sp->sa_addr.pptp;
-	if (add_chan(po)) {
-		release_sock(sk);
+	if (add_chan(po))
 		error = -EBUSY;
-	}
 
 	release_sock(sk);
 	return error;
diff --git a/drivers/net/usb/asix.c b/drivers/net/usb/asix.c
index e81e22e..e6fed4d 100644
--- a/drivers/net/usb/asix.c
+++ b/drivers/net/usb/asix.c
@@ -36,7 +36,7 @@
 #include <linux/usb/usbnet.h>
 #include <linux/slab.h>
 
-#define DRIVER_VERSION "26-Sep-2011"
+#define DRIVER_VERSION "08-Nov-2011"
 #define DRIVER_NAME "asix"
 
 /* ASIX AX8817X based USB 2.0 Ethernet Devices */
@@ -163,7 +163,7 @@
 #define MARVELL_CTRL_TXDELAY	0x0002
 #define MARVELL_CTRL_RXDELAY	0x0080
 
-#define	PHY_MODE_RTL8211CL	0x0004
+#define	PHY_MODE_RTL8211CL	0x000C
 
 /* This structure cannot exceed sizeof(unsigned long [5]) AKA 20 bytes */
 struct asix_data {
@@ -652,9 +652,17 @@
 {
 	int phy_reg;
 	u32 phy_id;
+	int i;
 
-	phy_reg = asix_mdio_read(dev->net, dev->mii.phy_id, MII_PHYSID1);
-	if (phy_reg < 0)
+	/* Poll for the rare case the FW or phy isn't ready yet.  */
+	for (i = 0; i < 100; i++) {
+		phy_reg = asix_mdio_read(dev->net, dev->mii.phy_id, MII_PHYSID1);
+		if (phy_reg != 0 && phy_reg != 0xFFFF)
+			break;
+		mdelay(1);
+	}
+
+	if (phy_reg <= 0 || phy_reg == 0xFFFF)
 		return 0;
 
 	phy_id = (phy_reg & 0xffff) << 16;
@@ -1075,7 +1083,7 @@
 
 static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf)
 {
-	int ret;
+	int ret, embd_phy;
 	struct asix_data *data = (struct asix_data *)&dev->data;
 	u8 buf[ETH_ALEN];
 	u32 phyid;
@@ -1100,16 +1108,36 @@
 	dev->mii.reg_num_mask = 0x1f;
 	dev->mii.phy_id = asix_get_phy_addr(dev);
 
-	phyid = asix_get_phyid(dev);
-	dbg("PHYID=0x%08x", phyid);
-
 	dev->net->netdev_ops = &ax88772_netdev_ops;
 	dev->net->ethtool_ops = &ax88772_ethtool_ops;
 
-	ret = ax88772_reset(dev);
+	embd_phy = ((dev->mii.phy_id & 0x1f) == 0x10 ? 1 : 0);
+
+	/* Reset the PHY to normal operation mode */
+	ret = asix_write_cmd(dev, AX_CMD_SW_PHY_SELECT, embd_phy, 0, 0, NULL);
+	if (ret < 0) {
+		dbg("Select PHY #1 failed: %d", ret);
+		return ret;
+	}
+
+	ret = asix_sw_reset(dev, AX_SWRESET_IPPD | AX_SWRESET_PRL);
 	if (ret < 0)
 		return ret;
 
+	msleep(150);
+
+	ret = asix_sw_reset(dev, AX_SWRESET_CLEAR);
+	if (ret < 0)
+		return ret;
+
+	msleep(150);
+
+	ret = asix_sw_reset(dev, embd_phy ? AX_SWRESET_IPRL : AX_SWRESET_PRTE);
+
+	/* Read PHYID register *AFTER* the PHY was reset properly */
+	phyid = asix_get_phyid(dev);
+	dbg("PHYID=0x%08x", phyid);
+
 	/* Asix framing packs multiple eth frames into a 2K usb bulk transfer */
 	if (dev->driver_info->flags & FLAG_FRAMING_AX) {
 		/* hard_mtu  is still the default - the device does not support
@@ -1220,6 +1248,7 @@
 	__le16 eeprom;
 	u8 status;
 	int gpio0 = 0;
+	u32 phyid;
 
 	asix_read_cmd(dev, AX_CMD_READ_GPIOS, 0, 0, 1, &status);
 	dbg("GPIO Status: 0x%04x", status);
@@ -1235,12 +1264,13 @@
 		data->ledmode = 0;
 		gpio0 = 1;
 	} else {
-		data->phymode = le16_to_cpu(eeprom) & 7;
+		data->phymode = le16_to_cpu(eeprom) & 0x7F;
 		data->ledmode = le16_to_cpu(eeprom) >> 8;
 		gpio0 = (le16_to_cpu(eeprom) & 0x80) ? 0 : 1;
 	}
 	dbg("GPIO0: %d, PhyMode: %d", gpio0, data->phymode);
 
+	/* Power up external GigaPHY through AX88178 GPIO pin */
 	asix_write_gpio(dev, AX_GPIO_RSE | AX_GPIO_GPO_1 | AX_GPIO_GPO1EN, 40);
 	if ((le16_to_cpu(eeprom) >> 8) != 1) {
 		asix_write_gpio(dev, 0x003c, 30);
@@ -1252,6 +1282,13 @@
 		asix_write_gpio(dev, AX_GPIO_GPO1EN | AX_GPIO_GPO_1, 30);
 	}
 
+	/* Read PHYID register *AFTER* powering up PHY */
+	phyid = asix_get_phyid(dev);
+	dbg("PHYID=0x%08x", phyid);
+
+	/* Set AX88178 to enable MII/GMII/RGMII interface for external PHY */
+	asix_write_cmd(dev, AX_CMD_SW_PHY_SELECT, 0, 0, 0, NULL);
+
 	asix_sw_reset(dev, 0);
 	msleep(150);
 
@@ -1396,7 +1433,6 @@
 {
 	int ret;
 	u8 buf[ETH_ALEN];
-	u32 phyid;
 	struct asix_data *data = (struct asix_data *)&dev->data;
 
 	data->eeprom_len = AX88772_EEPROM_LEN;
@@ -1423,12 +1459,12 @@
 	dev->net->netdev_ops = &ax88178_netdev_ops;
 	dev->net->ethtool_ops = &ax88178_ethtool_ops;
 
-	phyid = asix_get_phyid(dev);
-	dbg("PHYID=0x%08x", phyid);
+	/* Blink LEDS so users know driver saw dongle */
+	asix_sw_reset(dev, 0);
+	msleep(150);
 
-	ret = ax88178_reset(dev);
-	if (ret < 0)
-		return ret;
+	asix_sw_reset(dev, AX_SWRESET_PRL | AX_SWRESET_IPPD);
+	msleep(150);
 
 	/* Asix framing packs multiple eth frames into a 2K usb bulk transfer */
 	if (dev->driver_info->flags & FLAG_FRAMING_AX) {
diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c
index c924ea2..99ed6eb 100644
--- a/drivers/net/usb/cdc_ether.c
+++ b/drivers/net/usb/cdc_ether.c
@@ -567,7 +567,7 @@
 {
 	USB_DEVICE_AND_INTERFACE_INFO(0x1004, 0x61aa, USB_CLASS_COMM,
 			USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE),
-	.driver_info = (unsigned long)&wwan_info,
+	.driver_info = 0,
 },
 
 /*
diff --git a/drivers/net/usb/lg-vl600.c b/drivers/net/usb/lg-vl600.c
index d43db32..9c26c63 100644
--- a/drivers/net/usb/lg-vl600.c
+++ b/drivers/net/usb/lg-vl600.c
@@ -144,10 +144,11 @@
 	}
 
 	frame = (struct vl600_frame_hdr *) buf->data;
-	/* NOTE: Should check that frame->magic == 0x53544448?
-	 * Otherwise if we receive garbage at the beginning of the frame
-	 * we may end up allocating a huge buffer and saving all the
-	 * future incoming data into it.  */
+	/* Yes, check that frame->magic == 0x53544448 (or 0x44544d48),
+	 * otherwise we may run out of memory w/a bad packet */
+	if (ntohl(frame->magic) != 0x53544448 &&
+			ntohl(frame->magic) != 0x44544d48)
+		goto error;
 
 	if (buf->len < sizeof(*frame) ||
 			buf->len != le32_to_cpup(&frame->len)) {
@@ -296,6 +297,11 @@
 	 * overwrite the remaining fields.
 	 */
 	packet = (struct vl600_pkt_hdr *) skb->data;
+	/* The VL600 wants IPv6 packets to have an IPv4 ethertype
+	 * Since this modem only supports IPv4 and IPv6, just set all
+	 * frames to 0x0800 (ETH_P_IP)
+	 */
+	packet->h_proto = htons(ETH_P_IP);
 	memset(&packet->dummy, 0, sizeof(packet->dummy));
 	packet->len = cpu_to_le32(orig_len);
 
@@ -308,21 +314,12 @@
 	if (skb->len < full_len) /* Pad */
 		skb_put(skb, full_len - skb->len);
 
-	/* The VL600 wants IPv6 packets to have an IPv4 ethertype
-	 * Check if this is an IPv6 packet, and set the ethertype
-	 * to 0x800
-	 */
-	if ((skb->data[sizeof(struct vl600_pkt_hdr *) + 0x22] & 0xf0) == 0x60) {
-		skb->data[sizeof(struct vl600_pkt_hdr *) + 0x20] = 0x08;
-		skb->data[sizeof(struct vl600_pkt_hdr *) + 0x21] = 0;
-	}
-
 	return skb;
 }
 
 static const struct driver_info	vl600_info = {
 	.description	= "LG VL600 modem",
-	.flags		= FLAG_ETHER | FLAG_RX_ASSEMBLE,
+	.flags		= FLAG_RX_ASSEMBLE | FLAG_WWAN,
 	.bind		= vl600_bind,
 	.unbind		= vl600_unbind,
 	.status		= usbnet_cdc_status,
diff --git a/drivers/net/usb/smsc75xx.c b/drivers/net/usb/smsc75xx.c
index 22a7cf9..a5b9b12 100644
--- a/drivers/net/usb/smsc75xx.c
+++ b/drivers/net/usb/smsc75xx.c
@@ -51,6 +51,7 @@
 #define USB_VENDOR_ID_SMSC		(0x0424)
 #define USB_PRODUCT_ID_LAN7500		(0x7500)
 #define USB_PRODUCT_ID_LAN7505		(0x7505)
+#define RXW_PADDING			2
 
 #define check_warn(ret, fmt, args...) \
 	({ if (ret < 0) netdev_warn(dev->net, fmt, ##args); })
@@ -1088,13 +1089,13 @@
 
 		memcpy(&rx_cmd_b, skb->data, sizeof(rx_cmd_b));
 		le32_to_cpus(&rx_cmd_b);
-		skb_pull(skb, 4 + NET_IP_ALIGN);
+		skb_pull(skb, 4 + RXW_PADDING);
 
 		packet = skb->data;
 
 		/* get the packet length */
-		size = (rx_cmd_a & RX_CMD_A_LEN) - NET_IP_ALIGN;
-		align_count = (4 - ((size + NET_IP_ALIGN) % 4)) % 4;
+		size = (rx_cmd_a & RX_CMD_A_LEN) - RXW_PADDING;
+		align_count = (4 - ((size + RXW_PADDING) % 4)) % 4;
 
 		if (unlikely(rx_cmd_a & RX_CMD_A_RED)) {
 			netif_dbg(dev, rx_err, dev->net,
diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c
index 2f91acc..8873c6e 100644
--- a/drivers/net/wireless/ath/ath9k/hw.c
+++ b/drivers/net/wireless/ath/ath9k/hw.c
@@ -1827,7 +1827,8 @@
 	}
 
 	/* Clear Bit 14 of AR_WA after putting chip into Full Sleep mode. */
-	REG_WRITE(ah, AR_WA, ah->WARegVal & ~AR_WA_D3_L1_DISABLE);
+	if (AR_SREV_9300_20_OR_LATER(ah))
+		REG_WRITE(ah, AR_WA, ah->WARegVal & ~AR_WA_D3_L1_DISABLE);
 }
 
 /*
diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index 93fbe6f..d2348a5 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -286,7 +286,7 @@
 			ath_start_ani(common);
 	}
 
-	if (ath9k_hw_ops(ah)->antdiv_comb_conf_get && sc->ant_rx != 3) {
+	if ((ah->caps.hw_caps & ATH9K_HW_CAP_ANT_DIV_COMB) && sc->ant_rx != 3) {
 		struct ath_hw_antcomb_conf div_ant_conf;
 		u8 lna_conf;
 
diff --git a/drivers/net/wireless/ath/regd.c b/drivers/net/wireless/ath/regd.c
index 85fa9cc..65ecb5b 100644
--- a/drivers/net/wireless/ath/regd.c
+++ b/drivers/net/wireless/ath/regd.c
@@ -254,6 +254,8 @@
 	int r;
 
 	sband = wiphy->bands[IEEE80211_BAND_2GHZ];
+	if (!sband)
+		return;
 
 	/*
 	 * If no country IE has been received always enable active scan
diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c
index 58ea0e5..5f77cbe 100644
--- a/drivers/net/wireless/b43/xmit.c
+++ b/drivers/net/wireless/b43/xmit.c
@@ -175,6 +175,7 @@
 	}
 }
 
+/* TODO: verify if needed for SSLPN or LCN  */
 static u16 b43_generate_tx_phy_ctl1(struct b43_wldev *dev, u8 bitrate)
 {
 	const struct b43_phy *phy = &dev->phy;
@@ -256,6 +257,9 @@
 	unsigned int plcp_fragment_len;
 	u32 mac_ctl = 0;
 	u16 phy_ctl = 0;
+	bool fill_phy_ctl1 = (phy->type == B43_PHYTYPE_LP ||
+			      phy->type == B43_PHYTYPE_N ||
+			      phy->type == B43_PHYTYPE_HT);
 	u8 extra_ft = 0;
 	struct ieee80211_rate *txrate;
 	struct ieee80211_tx_rate *rates;
@@ -531,7 +535,7 @@
 			extra_ft |= B43_TXH_EFT_RTSFB_CCK;
 
 		if (rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS &&
-		    phy->type == B43_PHYTYPE_N) {
+		    fill_phy_ctl1) {
 			txhdr->phy_ctl1_rts = cpu_to_le16(
 				b43_generate_tx_phy_ctl1(dev, rts_rate));
 			txhdr->phy_ctl1_rts_fb = cpu_to_le16(
@@ -552,7 +556,7 @@
 		break;
 	}
 
-	if (phy->type == B43_PHYTYPE_N) {
+	if (fill_phy_ctl1) {
 		txhdr->phy_ctl1 =
 			cpu_to_le16(b43_generate_tx_phy_ctl1(dev, rate));
 		txhdr->phy_ctl1_fb =
@@ -736,7 +740,14 @@
 
 	/* Link quality statistics */
 	switch (chanstat & B43_RX_CHAN_PHYTYPE) {
+	case B43_PHYTYPE_HT:
+		/* TODO: is max the right choice? */
+		status.signal = max_t(__s8,
+			max(rxhdr->phy_ht_power0, rxhdr->phy_ht_power1),
+			rxhdr->phy_ht_power2);
+		break;
 	case B43_PHYTYPE_N:
+		/* Broadcom has code for min and avg, but always uses max */
 		if (rxhdr->power0 == 16 || rxhdr->power0 == 32)
 			status.signal = max(rxhdr->power1, rxhdr->power2);
 		else
diff --git a/drivers/net/wireless/b43/xmit.h b/drivers/net/wireless/b43/xmit.h
index 16c514d..98d9074 100644
--- a/drivers/net/wireless/b43/xmit.h
+++ b/drivers/net/wireless/b43/xmit.h
@@ -249,6 +249,12 @@
 		} __packed;
 	} __packed;
 	union {
+		/* HT-PHY */
+		struct {
+			PAD_BYTES(1);
+			__s8 phy_ht_power0;
+		} __packed;
+
 		/* RSSI for N-PHYs */
 		struct {
 			__s8 power2;
@@ -257,7 +263,15 @@
 
 		__le16 phy_status2;	/* PHY RX Status 2 */
 	} __packed;
-	__le16 phy_status3;	/* PHY RX Status 3 */
+	union {
+		/* HT-PHY */
+		struct {
+			__s8 phy_ht_power1;
+			__s8 phy_ht_power2;
+		} __packed;
+
+		__le16 phy_status3;	/* PHY RX Status 3 */
+	} __packed;
 	union {
 		/* Tested with 598.314, 644.1001 and 666.2 */
 		struct {
diff --git a/drivers/net/wireless/brcm80211/brcmsmac/dma.c b/drivers/net/wireless/brcm80211/brcmsmac/dma.c
index b56a302..6ebec8f 100644
--- a/drivers/net/wireless/brcm80211/brcmsmac/dma.c
+++ b/drivers/net/wireless/brcm80211/brcmsmac/dma.c
@@ -358,13 +358,14 @@
 
 static uint _dma_ctrlflags(struct dma_info *di, uint mask, uint flags)
 {
-	uint dmactrlflags = di->dma.dmactrlflags;
+	uint dmactrlflags;
 
 	if (di == NULL) {
-		DMA_ERROR(("%s: _dma_ctrlflags: NULL dma handle\n", di->name));
+		DMA_ERROR(("_dma_ctrlflags: NULL dma handle\n"));
 		return 0;
 	}
 
+	dmactrlflags = di->dma.dmactrlflags;
 	dmactrlflags &= ~mask;
 	dmactrlflags |= flags;
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-1000.c b/drivers/net/wireless/iwlwifi/iwl-1000.c
index e12b48c..dd008b0 100644
--- a/drivers/net/wireless/iwlwifi/iwl-1000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-1000.c
@@ -191,6 +191,7 @@
 	.chain_noise_scale = 1000,
 	.wd_timeout = IWL_DEF_WD_TIMEOUT,
 	.max_event_log_size = 128,
+	.wd_disable = true,
 };
 static struct iwl_ht_params iwl1000_ht_params = {
 	.ht_greenfield_support = true,
diff --git a/drivers/net/wireless/iwlwifi/iwl-5000.c b/drivers/net/wireless/iwlwifi/iwl-5000.c
index c511c98..f55fb2d 100644
--- a/drivers/net/wireless/iwlwifi/iwl-5000.c
+++ b/drivers/net/wireless/iwlwifi/iwl-5000.c
@@ -364,6 +364,7 @@
 	.wd_timeout = IWL_LONG_WD_TIMEOUT,
 	.max_event_log_size = 512,
 	.no_idle_support = true,
+	.wd_disable = true,
 };
 static struct iwl_ht_params iwl5000_ht_params = {
 	.ht_greenfield_support = true,
diff --git a/drivers/net/wireless/iwlwifi/iwl-agn-rxon.c b/drivers/net/wireless/iwlwifi/iwl-agn-rxon.c
index 58a381c..a7a6def 100644
--- a/drivers/net/wireless/iwlwifi/iwl-agn-rxon.c
+++ b/drivers/net/wireless/iwlwifi/iwl-agn-rxon.c
@@ -528,6 +528,24 @@
 	return 0;
 }
 
+void iwlagn_config_ht40(struct ieee80211_conf *conf,
+	struct iwl_rxon_context *ctx)
+{
+	if (conf_is_ht40_minus(conf)) {
+		ctx->ht.extension_chan_offset =
+			IEEE80211_HT_PARAM_CHA_SEC_BELOW;
+		ctx->ht.is_40mhz = true;
+	} else if (conf_is_ht40_plus(conf)) {
+		ctx->ht.extension_chan_offset =
+			IEEE80211_HT_PARAM_CHA_SEC_ABOVE;
+		ctx->ht.is_40mhz = true;
+	} else {
+		ctx->ht.extension_chan_offset =
+			IEEE80211_HT_PARAM_CHA_SEC_NONE;
+		ctx->ht.is_40mhz = false;
+	}
+}
+
 int iwlagn_mac_config(struct ieee80211_hw *hw, u32 changed)
 {
 	struct iwl_priv *priv = hw->priv;
@@ -586,19 +604,11 @@
 				ctx->ht.enabled = conf_is_ht(conf);
 
 			if (ctx->ht.enabled) {
-				if (conf_is_ht40_minus(conf)) {
-					ctx->ht.extension_chan_offset =
-						IEEE80211_HT_PARAM_CHA_SEC_BELOW;
-					ctx->ht.is_40mhz = true;
-				} else if (conf_is_ht40_plus(conf)) {
-					ctx->ht.extension_chan_offset =
-						IEEE80211_HT_PARAM_CHA_SEC_ABOVE;
-					ctx->ht.is_40mhz = true;
-				} else {
-					ctx->ht.extension_chan_offset =
-						IEEE80211_HT_PARAM_CHA_SEC_NONE;
-					ctx->ht.is_40mhz = false;
-				}
+				/* if HT40 is used, it should not change
+				 * after associated except channel switch */
+				if (iwl_is_associated_ctx(ctx) &&
+				     !ctx->ht.is_40mhz)
+					iwlagn_config_ht40(conf, ctx);
 			} else
 				ctx->ht.is_40mhz = false;
 
diff --git a/drivers/net/wireless/iwlwifi/iwl-agn-sta.c b/drivers/net/wireless/iwlwifi/iwl-agn-sta.c
index ed62836..4b2aa1d 100644
--- a/drivers/net/wireless/iwlwifi/iwl-agn-sta.c
+++ b/drivers/net/wireless/iwlwifi/iwl-agn-sta.c
@@ -1268,9 +1268,6 @@
 
 	switch (keyconf->cipher) {
 	case WLAN_CIPHER_SUITE_TKIP:
-		keyconf->flags |= IEEE80211_KEY_FLAG_GENERATE_MMIC;
-		keyconf->flags |= IEEE80211_KEY_FLAG_GENERATE_IV;
-
 		if (sta)
 			addr = sta->addr;
 		else /* station mode case only */
@@ -1283,8 +1280,6 @@
 					  seq.tkip.iv32, p1k, CMD_SYNC);
 		break;
 	case WLAN_CIPHER_SUITE_CCMP:
-		keyconf->flags |= IEEE80211_KEY_FLAG_GENERATE_IV;
-		/* fall through */
 	case WLAN_CIPHER_SUITE_WEP40:
 	case WLAN_CIPHER_SUITE_WEP104:
 		ret = iwlagn_send_sta_key(priv, keyconf, sta_id,
diff --git a/drivers/net/wireless/iwlwifi/iwl-agn.c b/drivers/net/wireless/iwlwifi/iwl-agn.c
index ccba69b..bacc06c 100644
--- a/drivers/net/wireless/iwlwifi/iwl-agn.c
+++ b/drivers/net/wireless/iwlwifi/iwl-agn.c
@@ -2316,6 +2316,17 @@
 		return -EOPNOTSUPP;
 	}
 
+	switch (key->cipher) {
+	case WLAN_CIPHER_SUITE_TKIP:
+		key->flags |= IEEE80211_KEY_FLAG_GENERATE_MMIC;
+		/* fall through */
+	case WLAN_CIPHER_SUITE_CCMP:
+		key->flags |= IEEE80211_KEY_FLAG_GENERATE_IV;
+		break;
+	default:
+		break;
+	}
+
 	/*
 	 * We could program these keys into the hardware as well, but we
 	 * don't expect much multicast traffic in IBSS and having keys
@@ -2599,21 +2610,9 @@
 
 	/* Configure HT40 channels */
 	ctx->ht.enabled = conf_is_ht(conf);
-	if (ctx->ht.enabled) {
-		if (conf_is_ht40_minus(conf)) {
-			ctx->ht.extension_chan_offset =
-				IEEE80211_HT_PARAM_CHA_SEC_BELOW;
-			ctx->ht.is_40mhz = true;
-		} else if (conf_is_ht40_plus(conf)) {
-			ctx->ht.extension_chan_offset =
-				IEEE80211_HT_PARAM_CHA_SEC_ABOVE;
-			ctx->ht.is_40mhz = true;
-		} else {
-			ctx->ht.extension_chan_offset =
-				IEEE80211_HT_PARAM_CHA_SEC_NONE;
-			ctx->ht.is_40mhz = false;
-		}
-	} else
+	if (ctx->ht.enabled)
+		iwlagn_config_ht40(conf, ctx);
+	else
 		ctx->ht.is_40mhz = false;
 
 	if ((le16_to_cpu(ctx->staging.channel) != ch))
@@ -3499,9 +3498,10 @@
 module_param_named(ack_check, iwlagn_mod_params.ack_check, bool, S_IRUGO);
 MODULE_PARM_DESC(ack_check, "Check ack health (default: 0 [disabled])");
 
-module_param_named(wd_disable, iwlagn_mod_params.wd_disable, bool, S_IRUGO);
+module_param_named(wd_disable, iwlagn_mod_params.wd_disable, int, S_IRUGO);
 MODULE_PARM_DESC(wd_disable,
-		"Disable stuck queue watchdog timer (default: 0 [enabled])");
+		"Disable stuck queue watchdog timer 0=system default, "
+		"1=disable, 2=enable (default: 0)");
 
 /*
  * set bt_coex_active to true, uCode will do kill/defer
diff --git a/drivers/net/wireless/iwlwifi/iwl-agn.h b/drivers/net/wireless/iwlwifi/iwl-agn.h
index 5b936ec..3856aba 100644
--- a/drivers/net/wireless/iwlwifi/iwl-agn.h
+++ b/drivers/net/wireless/iwlwifi/iwl-agn.h
@@ -86,6 +86,8 @@
 			     struct ieee80211_vif *vif,
 			     struct ieee80211_bss_conf *bss_conf,
 			     u32 changes);
+void iwlagn_config_ht40(struct ieee80211_conf *conf,
+			struct iwl_rxon_context *ctx);
 
 /* uCode */
 int iwlagn_rx_calib_result(struct iwl_priv *priv,
diff --git a/drivers/net/wireless/iwlwifi/iwl-core.c b/drivers/net/wireless/iwlwifi/iwl-core.c
index 001fdf1..fcf5416 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.c
+++ b/drivers/net/wireless/iwlwifi/iwl-core.c
@@ -1810,11 +1810,23 @@
 {
 	unsigned int timeout = priv->cfg->base_params->wd_timeout;
 
-	if (timeout && !iwlagn_mod_params.wd_disable)
-		mod_timer(&priv->watchdog,
-			  jiffies + msecs_to_jiffies(IWL_WD_TICK(timeout)));
-	else
-		del_timer(&priv->watchdog);
+	if (!iwlagn_mod_params.wd_disable) {
+		/* use system default */
+		if (timeout && !priv->cfg->base_params->wd_disable)
+			mod_timer(&priv->watchdog,
+				jiffies +
+				msecs_to_jiffies(IWL_WD_TICK(timeout)));
+		else
+			del_timer(&priv->watchdog);
+	} else {
+		/* module parameter overwrite default configuration */
+		if (timeout && iwlagn_mod_params.wd_disable == 2)
+			mod_timer(&priv->watchdog,
+				jiffies +
+				msecs_to_jiffies(IWL_WD_TICK(timeout)));
+		else
+			del_timer(&priv->watchdog);
+	}
 }
 
 /**
diff --git a/drivers/net/wireless/iwlwifi/iwl-core.h b/drivers/net/wireless/iwlwifi/iwl-core.h
index 137da33..f2fc288 100644
--- a/drivers/net/wireless/iwlwifi/iwl-core.h
+++ b/drivers/net/wireless/iwlwifi/iwl-core.h
@@ -113,6 +113,7 @@
  * @shadow_reg_enable: HW shadhow register bit
  * @no_idle_support: do not support idle mode
  * @hd_v2: v2 of enhanced sensitivity value, used for 2000 series and up
+ * wd_disable: disable watchdog timer
  */
 struct iwl_base_params {
 	int eeprom_size;
@@ -134,6 +135,7 @@
 	const bool shadow_reg_enable;
 	const bool no_idle_support;
 	const bool hd_v2;
+	const bool wd_disable;
 };
 /*
  * @advanced_bt_coexist: support advanced bt coexist
diff --git a/drivers/net/wireless/iwlwifi/iwl-shared.h b/drivers/net/wireless/iwlwifi/iwl-shared.h
index 1f7a93c..14eaf37 100644
--- a/drivers/net/wireless/iwlwifi/iwl-shared.h
+++ b/drivers/net/wireless/iwlwifi/iwl-shared.h
@@ -120,7 +120,7 @@
  * @restart_fw: restart firmware, default = 1
  * @plcp_check: enable plcp health check, default = true
  * @ack_check: disable ack health check, default = false
- * @wd_disable: enable stuck queue check, default = false
+ * @wd_disable: enable stuck queue check, default = 0
  * @bt_coex_active: enable bt coex, default = true
  * @led_mode: system default, default = 0
  * @no_sleep_autoadjust: disable autoadjust, default = true
@@ -141,7 +141,7 @@
 	int restart_fw;
 	bool plcp_check;
 	bool ack_check;
-	bool wd_disable;
+	int  wd_disable;
 	bool bt_coex_active;
 	int led_mode;
 	bool no_sleep_autoadjust;
diff --git a/drivers/net/wireless/iwlwifi/iwl-trans-pcie.c b/drivers/net/wireless/iwlwifi/iwl-trans-pcie.c
index da34110..ce91898 100644
--- a/drivers/net/wireless/iwlwifi/iwl-trans-pcie.c
+++ b/drivers/net/wireless/iwlwifi/iwl-trans-pcie.c
@@ -990,29 +990,16 @@
 	return 0;
 }
 
-static void iwl_trans_pcie_disable_sync_irq(struct iwl_trans *trans)
+static void iwl_trans_pcie_stop_device(struct iwl_trans *trans)
 {
 	unsigned long flags;
-	struct iwl_trans_pcie *trans_pcie =
-		IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
+	/* tell the device to stop sending interrupts */
 	spin_lock_irqsave(&trans->shrd->lock, flags);
 	iwl_disable_interrupts(trans);
 	spin_unlock_irqrestore(&trans->shrd->lock, flags);
 
-	/* wait to make sure we flush pending tasklet*/
-	synchronize_irq(bus(trans)->irq);
-	tasklet_kill(&trans_pcie->irq_tasklet);
-}
-
-static void iwl_trans_pcie_stop_device(struct iwl_trans *trans)
-{
-	/* stop and reset the on-board processor */
-	iwl_write32(bus(trans), CSR_RESET, CSR_RESET_REG_FLAG_NEVO_RESET);
-
-	/* tell the device to stop sending interrupts */
-	iwl_trans_pcie_disable_sync_irq(trans);
-
 	/* device going down, Stop using ICT table */
 	iwl_disable_ict(trans);
 
@@ -1039,6 +1026,20 @@
 
 	/* Stop the device, and put it in low power state */
 	iwl_apm_stop(priv(trans));
+
+	/* Upon stop, the APM issues an interrupt if HW RF kill is set.
+	 * Clean again the interrupt here
+	 */
+	spin_lock_irqsave(&trans->shrd->lock, flags);
+	iwl_disable_interrupts(trans);
+	spin_unlock_irqrestore(&trans->shrd->lock, flags);
+
+	/* wait to make sure we flush pending tasklet*/
+	synchronize_irq(bus(trans)->irq);
+	tasklet_kill(&trans_pcie->irq_tasklet);
+
+	/* stop and reset the on-board processor */
+	iwl_write32(bus(trans), CSR_RESET, CSR_RESET_REG_FLAG_NEVO_RESET);
 }
 
 static int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
diff --git a/drivers/net/wireless/libertas/cfg.c b/drivers/net/wireless/libertas/cfg.c
index 4fcd653..a7f1ab2 100644
--- a/drivers/net/wireless/libertas/cfg.c
+++ b/drivers/net/wireless/libertas/cfg.c
@@ -634,7 +634,7 @@
 			if (channel &&
 			    !(channel->flags & IEEE80211_CHAN_DISABLED))
 				cfg80211_inform_bss(wiphy, channel,
-					bssid, le64_to_cpu(*(__le64 *)tsfdesc),
+					bssid, get_unaligned_le64(tsfdesc),
 					capa, intvl, ie, ielen,
 					LBS_SCAN_RSSI_TO_MBM(rssi),
 					GFP_KERNEL);
diff --git a/drivers/net/wireless/libertas/if_spi.c b/drivers/net/wireless/libertas/if_spi.c
index 11b69b3..728baa4 100644
--- a/drivers/net/wireless/libertas/if_spi.c
+++ b/drivers/net/wireless/libertas/if_spi.c
@@ -995,6 +995,7 @@
 		spin_unlock_irqrestore(&card->buffer_lock, flags);
 		break;
 	default:
+		kfree(packet);
 		netdev_err(priv->dev, "can't transfer buffer of type %d\n",
 			   type);
 		err = -EINVAL;
diff --git a/drivers/net/wireless/mwifiex/scan.c b/drivers/net/wireless/mwifiex/scan.c
index dae8dbb..8d3ab37 100644
--- a/drivers/net/wireless/mwifiex/scan.c
+++ b/drivers/net/wireless/mwifiex/scan.c
@@ -819,8 +819,10 @@
 			wildcard_ssid_tlv->header.len = cpu_to_le16(
 				(u16) (ssid_len + sizeof(wildcard_ssid_tlv->
 							 max_ssid_length)));
-			wildcard_ssid_tlv->max_ssid_length =
-				user_scan_in->ssid_list[ssid_idx].max_len;
+
+			/* max_ssid_length = 0 tells firmware to perform
+			   specific scan for the SSID filled */
+			wildcard_ssid_tlv->max_ssid_length = 0;
 
 			memcpy(wildcard_ssid_tlv->ssid,
 			       user_scan_in->ssid_list[ssid_idx].ssid,
@@ -1469,7 +1471,7 @@
 			       s32 rssi, const u8 *ie_buf, size_t ie_len,
 			       u16 beacon_period, u16 cap_info_bitmap, u8 band)
 {
-	struct mwifiex_bssdescriptor *bss_desc = NULL;
+	struct mwifiex_bssdescriptor *bss_desc;
 	int ret;
 	unsigned long flags;
 	u8 *beacon_ie;
@@ -1484,6 +1486,7 @@
 
 	beacon_ie = kmemdup(ie_buf, ie_len, GFP_KERNEL);
 	if (!beacon_ie) {
+		kfree(bss_desc);
 		dev_err(priv->adapter->dev, " failed to alloc beacon_ie\n");
 		return -ENOMEM;
 	}
diff --git a/drivers/net/wireless/p54/p54spi.c b/drivers/net/wireless/p54/p54spi.c
index f18df82..78d0d69 100644
--- a/drivers/net/wireless/p54/p54spi.c
+++ b/drivers/net/wireless/p54/p54spi.c
@@ -588,8 +588,6 @@
 
 	WARN_ON(priv->fw_state != FW_STATE_READY);
 
-	cancel_work_sync(&priv->work);
-
 	p54spi_power_off(priv);
 	spin_lock_irqsave(&priv->tx_lock, flags);
 	INIT_LIST_HEAD(&priv->tx_pending);
@@ -597,6 +595,8 @@
 
 	priv->fw_state = FW_STATE_OFF;
 	mutex_unlock(&priv->mutex);
+
+	cancel_work_sync(&priv->work);
 }
 
 static int __devinit p54spi_probe(struct spi_device *spi)
@@ -656,6 +656,7 @@
 	init_completion(&priv->fw_comp);
 	INIT_LIST_HEAD(&priv->tx_pending);
 	mutex_init(&priv->mutex);
+	spin_lock_init(&priv->tx_lock);
 	SET_IEEE80211_DEV(hw, &spi->dev);
 	priv->common.open = p54spi_op_start;
 	priv->common.stop = p54spi_op_stop;
diff --git a/drivers/net/wireless/prism54/isl_ioctl.c b/drivers/net/wireless/prism54/isl_ioctl.c
index d97a2caf..bc2ba80 100644
--- a/drivers/net/wireless/prism54/isl_ioctl.c
+++ b/drivers/net/wireless/prism54/isl_ioctl.c
@@ -778,7 +778,7 @@
 		dwrq->flags = 0;
 		dwrq->length = 0;
 	}
-	essid->octets[essid->length] = '\0';
+	essid->octets[dwrq->length] = '\0';
 	memcpy(extra, essid->octets, dwrq->length);
 	kfree(essid);
 
diff --git a/drivers/net/wireless/rt2x00/rt2800lib.c b/drivers/net/wireless/rt2x00/rt2800lib.c
index 3f183a1..1ba079d 100644
--- a/drivers/net/wireless/rt2x00/rt2800lib.c
+++ b/drivers/net/wireless/rt2x00/rt2800lib.c
@@ -3771,7 +3771,7 @@
 	/* Apparently the data is read from end to start */
 	rt2800_register_read_lock(rt2x00dev, EFUSE_DATA3, &reg);
 	/* The returned value is in CPU order, but eeprom is le */
-	rt2x00dev->eeprom[i] = cpu_to_le32(reg);
+	*(u32 *)&rt2x00dev->eeprom[i] = cpu_to_le32(reg);
 	rt2800_register_read_lock(rt2x00dev, EFUSE_DATA2, &reg);
 	*(u32 *)&rt2x00dev->eeprom[i + 2] = cpu_to_le32(reg);
 	rt2800_register_read_lock(rt2x00dev, EFUSE_DATA1, &reg);
diff --git a/drivers/net/wireless/rt2x00/rt2800usb.c b/drivers/net/wireless/rt2x00/rt2800usb.c
index f156579..3778763 100644
--- a/drivers/net/wireless/rt2x00/rt2800usb.c
+++ b/drivers/net/wireless/rt2x00/rt2800usb.c
@@ -919,6 +919,7 @@
 	{ USB_DEVICE(0x050d, 0x935b) },
 	/* Buffalo */
 	{ USB_DEVICE(0x0411, 0x00e8) },
+	{ USB_DEVICE(0x0411, 0x0158) },
 	{ USB_DEVICE(0x0411, 0x016f) },
 	{ USB_DEVICE(0x0411, 0x01a2) },
 	/* Corega */
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index 2ec5c00..99ff12d 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -943,6 +943,7 @@
 	 * Powersaving work
 	 */
 	struct delayed_work autowakeup_work;
+	struct work_struct sleep_work;
 
 	/*
 	 * Data queue arrays for RX, TX, Beacon and ATIM.
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index e1fb2a8..edd317f 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -465,6 +465,23 @@
 	return NULL;
 }
 
+static void rt2x00lib_sleep(struct work_struct *work)
+{
+	struct rt2x00_dev *rt2x00dev =
+	    container_of(work, struct rt2x00_dev, sleep_work);
+
+	if (!test_bit(DEVICE_STATE_PRESENT, &rt2x00dev->flags))
+		return;
+
+	/*
+	 * Check again is powersaving is enabled, to prevent races from delayed
+	 * work execution.
+	 */
+	if (!test_bit(CONFIG_POWERSAVING, &rt2x00dev->flags))
+		rt2x00lib_config(rt2x00dev, &rt2x00dev->hw->conf,
+				 IEEE80211_CONF_CHANGE_PS);
+}
+
 static void rt2x00lib_rxdone_check_ps(struct rt2x00_dev *rt2x00dev,
 				      struct sk_buff *skb,
 				      struct rxdone_entry_desc *rxdesc)
@@ -512,8 +529,7 @@
 	cam |= (tim_ie->bitmap_ctrl & 0x01);
 
 	if (!cam && !test_bit(CONFIG_POWERSAVING, &rt2x00dev->flags))
-		rt2x00lib_config(rt2x00dev, &rt2x00dev->hw->conf,
-				 IEEE80211_CONF_CHANGE_PS);
+		queue_work(rt2x00dev->workqueue, &rt2x00dev->sleep_work);
 }
 
 static int rt2x00lib_rxdone_read_signal(struct rt2x00_dev *rt2x00dev,
@@ -1141,6 +1157,7 @@
 
 	INIT_WORK(&rt2x00dev->intf_work, rt2x00lib_intf_scheduled);
 	INIT_DELAYED_WORK(&rt2x00dev->autowakeup_work, rt2x00lib_autowakeup);
+	INIT_WORK(&rt2x00dev->sleep_work, rt2x00lib_sleep);
 
 	/*
 	 * Let the driver probe the device to detect the capabilities.
@@ -1197,6 +1214,7 @@
 	 */
 	cancel_work_sync(&rt2x00dev->intf_work);
 	cancel_delayed_work_sync(&rt2x00dev->autowakeup_work);
+	cancel_work_sync(&rt2x00dev->sleep_work);
 	if (rt2x00_is_usb(rt2x00dev)) {
 		del_timer_sync(&rt2x00dev->txstatus_timer);
 		cancel_work_sync(&rt2x00dev->rxdone_work);
diff --git a/drivers/net/wireless/rtlwifi/ps.c b/drivers/net/wireless/rtlwifi/ps.c
index db52628..55c8e50 100644
--- a/drivers/net/wireless/rtlwifi/ps.c
+++ b/drivers/net/wireless/rtlwifi/ps.c
@@ -395,7 +395,7 @@
 	if (mac->link_state != MAC80211_LINKED)
 		return;
 
-	spin_lock(&rtlpriv->locks.lps_lock);
+	spin_lock_irq(&rtlpriv->locks.lps_lock);
 
 	/* Idle for a while if we connect to AP a while ago. */
 	if (mac->cnt_after_linked >= 2) {
@@ -407,7 +407,7 @@
 		}
 	}
 
-	spin_unlock(&rtlpriv->locks.lps_lock);
+	spin_unlock_irq(&rtlpriv->locks.lps_lock);
 }
 
 /*Leave the leisure power save mode.*/
@@ -416,8 +416,9 @@
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_ps_ctl *ppsc = rtl_psc(rtl_priv(hw));
 	struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw));
+	unsigned long flags;
 
-	spin_lock(&rtlpriv->locks.lps_lock);
+	spin_lock_irqsave(&rtlpriv->locks.lps_lock, flags);
 
 	if (ppsc->fwctrl_lps) {
 		if (ppsc->dot11_psmode != EACTIVE) {
@@ -438,7 +439,7 @@
 			rtl_lps_set_psmode(hw, EACTIVE);
 		}
 	}
-	spin_unlock(&rtlpriv->locks.lps_lock);
+	spin_unlock_irqrestore(&rtlpriv->locks.lps_lock, flags);
 }
 
 /* For sw LPS*/
@@ -539,9 +540,9 @@
 		RT_CLEAR_PS_LEVEL(ppsc, RT_PS_LEVEL_ASPM);
 	}
 
-	spin_lock(&rtlpriv->locks.lps_lock);
+	spin_lock_irq(&rtlpriv->locks.lps_lock);
 	rtl_ps_set_rf_state(hw, ERFON, RF_CHANGE_BY_PS);
-	spin_unlock(&rtlpriv->locks.lps_lock);
+	spin_unlock_irq(&rtlpriv->locks.lps_lock);
 }
 
 void rtl_swlps_rfon_wq_callback(void *data)
@@ -574,9 +575,9 @@
 	if (rtlpriv->link_info.busytraffic)
 		return;
 
-	spin_lock(&rtlpriv->locks.lps_lock);
+	spin_lock_irq(&rtlpriv->locks.lps_lock);
 	rtl_ps_set_rf_state(hw, ERFSLEEP, RF_CHANGE_BY_PS);
-	spin_unlock(&rtlpriv->locks.lps_lock);
+	spin_unlock_irq(&rtlpriv->locks.lps_lock);
 
 	if (ppsc->reg_rfps_level & RT_RF_OFF_LEVL_ASPM &&
 		!RT_IN_PS_LEVEL(ppsc, RT_PS_LEVEL_ASPM)) {
diff --git a/drivers/net/wireless/rtlwifi/rtl8192ce/phy.c b/drivers/net/wireless/rtlwifi/rtl8192ce/phy.c
index 592a10a..3b585aa 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192ce/phy.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192ce/phy.c
@@ -569,7 +569,7 @@
 		}
 	case ERFSLEEP:{
 			if (ppsc->rfpwr_state == ERFOFF)
-				break;
+				return false;
 			for (queue_id = 0, i = 0;
 			     queue_id < RTL_PCI_MAX_TX_QUEUE_COUNT;) {
 				ring = &pcipriv->dev.tx_ring[queue_id];
diff --git a/drivers/net/wireless/rtlwifi/rtl8192cu/phy.c b/drivers/net/wireless/rtlwifi/rtl8192cu/phy.c
index 7285290..e49cf22 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192cu/phy.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192cu/phy.c
@@ -548,7 +548,7 @@
 		break;
 	case ERFSLEEP:
 		if (ppsc->rfpwr_state == ERFOFF)
-			break;
+			return false;
 		for (queue_id = 0, i = 0;
 		     queue_id < RTL_PCI_MAX_TX_QUEUE_COUNT;) {
 			ring = &pcipriv->dev.tx_ring[queue_id];
diff --git a/drivers/net/wireless/rtlwifi/rtl8192de/phy.c b/drivers/net/wireless/rtlwifi/rtl8192de/phy.c
index 3ac7af1..0883349 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192de/phy.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192de/phy.c
@@ -3374,7 +3374,7 @@
 		break;
 	case ERFSLEEP:
 		if (ppsc->rfpwr_state == ERFOFF)
-			break;
+			return false;
 
 		for (queue_id = 0, i = 0;
 		     queue_id < RTL_PCI_MAX_TX_QUEUE_COUNT;) {
diff --git a/drivers/net/wireless/rtlwifi/rtl8192se/phy.c b/drivers/net/wireless/rtlwifi/rtl8192se/phy.c
index f27171a..f10ac1a 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192se/phy.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192se/phy.c
@@ -602,7 +602,7 @@
 		}
 	case ERFSLEEP:
 			if (ppsc->rfpwr_state == ERFOFF)
-				break;
+				return false;
 
 			for (queue_id = 0, i = 0;
 			     queue_id < RTL_PCI_MAX_TX_QUEUE_COUNT;) {
diff --git a/drivers/net/wireless/wl12xx/scan.c b/drivers/net/wireless/wl12xx/scan.c
index 128ccb7..fc29c67 100644
--- a/drivers/net/wireless/wl12xx/scan.c
+++ b/drivers/net/wireless/wl12xx/scan.c
@@ -559,7 +559,7 @@
 						break;
 					}
 				/* Fail if SSID isn't present in the filters */
-				if (j == req->n_ssids) {
+				if (j == cmd->n_ssids) {
 					ret = -EINVAL;
 					goto out_free;
 				}
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 0cb594c..15e332d 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -1021,7 +1021,7 @@
 		pending_idx = *((u16 *)skb->data);
 		xen_netbk_idx_release(netbk, pending_idx);
 		for (j = start; j < i; j++) {
-			pending_idx = frag_get_pending_idx(&shinfo->frags[i]);
+			pending_idx = frag_get_pending_idx(&shinfo->frags[j]);
 			xen_netbk_idx_release(netbk, pending_idx);
 		}
 
@@ -1668,7 +1668,7 @@
 					     "netback/%u", group);
 
 		if (IS_ERR(netbk->task)) {
-			printk(KERN_ALERT "kthread_run() fails at netback\n");
+			printk(KERN_ALERT "kthread_create() fails at netback\n");
 			del_timer(&netbk->net_timer);
 			rc = PTR_ERR(netbk->task);
 			goto failed_init;
diff --git a/drivers/of/irq.c b/drivers/of/irq.c
index 6d3dd39..0f0cfa3 100644
--- a/drivers/of/irq.c
+++ b/drivers/of/irq.c
@@ -26,11 +26,6 @@
 #include <linux/string.h>
 #include <linux/slab.h>
 
-/* For archs that don't support NO_IRQ (such as x86), provide a dummy value */
-#ifndef NO_IRQ
-#define NO_IRQ 0
-#endif
-
 /**
  * irq_of_parse_and_map - Parse and map an interrupt into linux virq space
  * @device: Device node of the device whose interrupt is to be mapped
@@ -44,7 +39,7 @@
 	struct of_irq oirq;
 
 	if (of_irq_map_one(dev, index, &oirq))
-		return NO_IRQ;
+		return 0;
 
 	return irq_create_of_mapping(oirq.controller, oirq.specifier,
 				     oirq.size);
@@ -60,27 +55,27 @@
  */
 struct device_node *of_irq_find_parent(struct device_node *child)
 {
-	struct device_node *p, *c = child;
+	struct device_node *p;
 	const __be32 *parp;
 
-	if (!of_node_get(c))
+	if (!of_node_get(child))
 		return NULL;
 
 	do {
-		parp = of_get_property(c, "interrupt-parent", NULL);
+		parp = of_get_property(child, "interrupt-parent", NULL);
 		if (parp == NULL)
-			p = of_get_parent(c);
+			p = of_get_parent(child);
 		else {
 			if (of_irq_workarounds & OF_IMAP_NO_PHANDLE)
 				p = of_node_get(of_irq_dflt_pic);
 			else
 				p = of_find_node_by_phandle(be32_to_cpup(parp));
 		}
-		of_node_put(c);
-		c = p;
+		of_node_put(child);
+		child = p;
 	} while (p && of_get_property(p, "#interrupt-cells", NULL) == NULL);
 
-	return (p == child) ? NULL : p;
+	return p;
 }
 
 /**
@@ -345,7 +340,7 @@
 
 	/* Only dereference the resource if both the
 	 * resource and the irq are valid. */
-	if (r && irq != NO_IRQ) {
+	if (r && irq) {
 		r->start = r->end = irq;
 		r->flags = IORESOURCE_IRQ;
 		r->name = dev->full_name;
@@ -363,7 +358,7 @@
 {
 	int nr = 0;
 
-	while (of_irq_to_resource(dev, nr, NULL) != NO_IRQ)
+	while (of_irq_to_resource(dev, nr, NULL))
 		nr++;
 
 	return nr;
@@ -383,7 +378,7 @@
 	int i;
 
 	for (i = 0; i < nr_irqs; i++, res++)
-		if (of_irq_to_resource(dev, i, res) == NO_IRQ)
+		if (!of_irq_to_resource(dev, i, res))
 			break;
 
 	return i;
@@ -424,6 +419,8 @@
 
 		desc->dev = np;
 		desc->interrupt_parent = of_irq_find_parent(np);
+		if (desc->interrupt_parent == np)
+			desc->interrupt_parent = NULL;
 		list_add_tail(&desc->list, &intc_desc_list);
 	}
 
diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c
index dccd863..f8c752e 100644
--- a/drivers/oprofile/oprof.c
+++ b/drivers/oprofile/oprof.c
@@ -239,26 +239,45 @@
 	return err;
 }
 
+static int timer_mode;
+
 static int __init oprofile_init(void)
 {
 	int err;
 
+	/* always init architecture to setup backtrace support */
 	err = oprofile_arch_init(&oprofile_ops);
-	if (err < 0 || timer) {
-		printk(KERN_INFO "oprofile: using timer interrupt.\n");
+
+	timer_mode = err || timer;	/* fall back to timer mode on errors */
+	if (timer_mode) {
+		if (!err)
+			oprofile_arch_exit();
 		err = oprofile_timer_init(&oprofile_ops);
 		if (err)
 			return err;
 	}
-	return oprofilefs_register();
+
+	err = oprofilefs_register();
+	if (!err)
+		return 0;
+
+	/* failed */
+	if (timer_mode)
+		oprofile_timer_exit();
+	else
+		oprofile_arch_exit();
+
+	return err;
 }
 
 
 static void __exit oprofile_exit(void)
 {
-	oprofile_timer_exit();
 	oprofilefs_unregister();
-	oprofile_arch_exit();
+	if (timer_mode)
+		oprofile_timer_exit();
+	else
+		oprofile_arch_exit();
 }
 
 
diff --git a/drivers/oprofile/timer_int.c b/drivers/oprofile/timer_int.c
index 3ef4462..878fba1 100644
--- a/drivers/oprofile/timer_int.c
+++ b/drivers/oprofile/timer_int.c
@@ -110,6 +110,7 @@
 	ops->start = oprofile_hrtimer_start;
 	ops->stop = oprofile_hrtimer_stop;
 	ops->cpu_type = "timer";
+	printk(KERN_INFO "oprofile: using timer interrupt.\n");
 	return 0;
 }
 
diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig
index b6f9749..f02b523 100644
--- a/drivers/pci/Kconfig
+++ b/drivers/pci/Kconfig
@@ -76,6 +76,7 @@
 
 config PCI_PRI
 	bool "PCI PRI support"
+	depends on PCI
 	select PCI_ATS
 	help
 	  PRI is the PCI Page Request Interface. It allows PCI devices that are
diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c
index 7ec56fb..b0dd08e 100644
--- a/drivers/pci/ats.c
+++ b/drivers/pci/ats.c
@@ -13,6 +13,7 @@
 #include <linux/export.h>
 #include <linux/pci-ats.h>
 #include <linux/pci.h>
+#include <linux/slab.h>
 
 #include "pci.h"
 
diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index 596172b..9ddf69e 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -132,6 +132,18 @@
 	if (!acpi_pci_check_ejectable(pbus, handle) && !is_dock_device(handle))
 		return AE_OK;
 
+	pdev = pbus->self;
+	if (pdev && pci_is_pcie(pdev)) {
+		tmp = acpi_find_root_bridge_handle(pdev);
+		if (tmp) {
+			struct acpi_pci_root *root = acpi_pci_find_root(tmp);
+
+			if (root && (root->osc_control_set &
+					OSC_PCI_EXPRESS_NATIVE_HP_CONTROL))
+				return AE_OK;
+		}
+	}
+
 	acpi_evaluate_integer(handle, "_ADR", NULL, &adr);
 	device = (adr >> 16) & 0xffff;
 	function = adr & 0xffff;
@@ -213,7 +225,6 @@
 
 	pdev = pci_get_slot(pbus, PCI_DEVFN(device, function));
 	if (pdev) {
-		pdev->current_state = PCI_D0;
 		slot->flags |= (SLOT_ENABLED | SLOT_POWEREDON);
 		pci_dev_put(pdev);
 	}
@@ -1378,11 +1389,13 @@
 {
 	int *count = (int *)context;
 
-	if (acpi_is_root_bridge(handle)) {
-		acpi_install_notify_handler(handle, ACPI_SYSTEM_NOTIFY,
-				handle_hotplug_event_bridge, NULL);
-			(*count)++;
-	}
+	if (!acpi_is_root_bridge(handle))
+		return AE_OK;
+
+	(*count)++;
+	acpi_install_notify_handler(handle, ACPI_SYSTEM_NOTIFY,
+				    handle_hotplug_event_bridge, NULL);
+
 	return AE_OK ;
 }
 
diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c
index 1e9c9aa..085dbb5 100644
--- a/drivers/pci/hotplug/pciehp_ctrl.c
+++ b/drivers/pci/hotplug/pciehp_ctrl.c
@@ -213,9 +213,6 @@
 		goto err_exit;
 	}
 
-	/* Wait for 1 second after checking link training status */
-	msleep(1000);
-
 	/* Check for a power fault */
 	if (ctrl->power_fault_detected || pciehp_query_power_fault(p_slot)) {
 		ctrl_err(ctrl, "Power fault on slot %s\n", slot_name(p_slot));
diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c
index 96dc473..7b14148 100644
--- a/drivers/pci/hotplug/pciehp_hpc.c
+++ b/drivers/pci/hotplug/pciehp_hpc.c
@@ -280,6 +280,14 @@
         else
                 msleep(1000);
 
+	/*
+	 * Need to wait for 1000 ms after Data Link Layer Link Active
+	 * (DLLLA) bit reads 1b before sending configuration request.
+	 * We need it before checking Link Training (LT) bit becuase
+	 * LT is still set even after DLLLA bit is set on some platform.
+	 */
+	msleep(1000);
+
 	retval = pciehp_readw(ctrl, PCI_EXP_LNKSTA, &lnk_status);
 	if (retval) {
 		ctrl_err(ctrl, "Cannot read LNKSTATUS register\n");
@@ -294,6 +302,16 @@
 		return retval;
 	}
 
+	/*
+	 * If the port supports Link speeds greater than 5.0 GT/s, we
+	 * must wait for 100 ms after Link training completes before
+	 * sending configuration request.
+	 */
+	if (ctrl->pcie->port->subordinate->max_bus_speed > PCIE_SPEED_5_0GT)
+		msleep(100);
+
+	pcie_update_link_speed(ctrl->pcie->port->subordinate, lnk_status);
+
 	return retval;
 }
 
@@ -484,7 +502,6 @@
 	u16 slot_cmd;
 	u16 cmd_mask;
 	u16 slot_status;
-	u16 lnk_status;
 	int retval = 0;
 
 	/* Clear sticky power-fault bit from previous power failures */
@@ -516,14 +533,6 @@
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n", __func__,
 		 pci_pcie_cap(ctrl->pcie->port) + PCI_EXP_SLTCTL, slot_cmd);
 
-	retval = pciehp_readw(ctrl, PCI_EXP_LNKSTA, &lnk_status);
-	if (retval) {
-		ctrl_err(ctrl, "%s: Cannot read LNKSTA register\n",
-				__func__);
-		return retval;
-	}
-	pcie_update_link_speed(ctrl->pcie->port->subordinate, lnk_status);
-
 	return retval;
 }
 
diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c
index aca972b..dd7e0c5 100644
--- a/drivers/pci/hotplug/shpchp_core.c
+++ b/drivers/pci/hotplug/shpchp_core.c
@@ -278,8 +278,8 @@
 
 static int is_shpc_capable(struct pci_dev *dev)
 {
-	if ((dev->vendor == PCI_VENDOR_ID_AMD) || (dev->device ==
-						PCI_DEVICE_ID_AMD_GOLAM_7450))
+	if (dev->vendor == PCI_VENDOR_ID_AMD &&
+	    dev->device == PCI_DEVICE_ID_AMD_GOLAM_7450)
 		return 1;
 	if (!pci_find_capability(dev, PCI_CAP_ID_SHPC))
 		return 0;
diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c
index 36547f0..75ba231 100644
--- a/drivers/pci/hotplug/shpchp_hpc.c
+++ b/drivers/pci/hotplug/shpchp_hpc.c
@@ -944,8 +944,8 @@
 	ctrl->pci_dev = pdev;  /* pci_dev of the P2P bridge */
 	ctrl_dbg(ctrl, "Hotplug Controller:\n");
 
-	if ((pdev->vendor == PCI_VENDOR_ID_AMD) || (pdev->device ==
-				PCI_DEVICE_ID_AMD_GOLAM_7450)) {
+	if (pdev->vendor == PCI_VENDOR_ID_AMD &&
+	    pdev->device == PCI_DEVICE_ID_AMD_GOLAM_7450) {
 		/* amd shpc driver doesn't use Base Offset; assume 0 */
 		ctrl->mmio_base = pci_resource_start(pdev, 0);
 		ctrl->mmio_size = pci_resource_len(pdev, 0);
diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
index b82c155..1969a3e 100644
--- a/drivers/pci/iov.c
+++ b/drivers/pci/iov.c
@@ -283,6 +283,7 @@
 	struct resource *res;
 	struct pci_dev *pdev;
 	struct pci_sriov *iov = dev->sriov;
+	int bars = 0;
 
 	if (!nr_virtfn)
 		return 0;
@@ -307,6 +308,7 @@
 
 	nres = 0;
 	for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) {
+		bars |= (1 << (i + PCI_IOV_RESOURCES));
 		res = dev->resource + PCI_IOV_RESOURCES + i;
 		if (res->parent)
 			nres++;
@@ -324,6 +326,11 @@
 		return -ENOMEM;
 	}
 
+	if (pci_enable_resources(dev, bars)) {
+		dev_err(&dev->dev, "SR-IOV: IOV BARS not allocated\n");
+		return -ENOMEM;
+	}
+
 	if (iov->link != dev->devfn) {
 		pdev = pci_get_slot(dev->bus, iov->link);
 		if (!pdev)
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 6f45a73..6d4a531 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -664,6 +664,9 @@
 		error = platform_pci_set_power_state(dev, state);
 		if (!error)
 			pci_update_current_state(dev, state);
+		/* Fall back to PCI_D0 if native PM is not supported */
+		if (!dev->pm_cap)
+			dev->current_state = PCI_D0;
 	} else {
 		error = -ENODEV;
 		/* Fall back to PCI_D0 if native PM is not supported */
@@ -1126,7 +1129,11 @@
 	if (atomic_add_return(1, &dev->enable_cnt) > 1)
 		return 0;		/* already enabled */
 
-	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++)
+	/* only skip sriov related */
+	for (i = 0; i <= PCI_ROM_RESOURCE; i++)
+		if (dev->resource[i].flags & flags)
+			bars |= (1 << i);
+	for (i = PCI_BRIDGE_RESOURCES; i < DEVICE_COUNT_RESOURCE; i++)
 		if (dev->resource[i].flags & flags)
 			bars |= (1 << i);
 
diff --git a/drivers/pinctrl/Kconfig b/drivers/pinctrl/Kconfig
index ef56644..e17e2f8 100644
--- a/drivers/pinctrl/Kconfig
+++ b/drivers/pinctrl/Kconfig
@@ -2,23 +2,17 @@
 # PINCTRL infrastructure and drivers
 #
 
-menuconfig PINCTRL
-	bool "PINCTRL Support"
+config PINCTRL
+	bool
 	depends on EXPERIMENTAL
-	help
-	  This enables the PINCTRL subsystem for controlling pins
-	  on chip packages, for example multiplexing pins on primarily
-	  PGA and BGA packages for systems on chip.
-
-	  If unsure, say N.
 
 if PINCTRL
 
+menu "Pin controllers"
+	depends on PINCTRL
+
 config PINMUX
 	bool "Support pinmux controllers"
-	help
-	  Say Y here if you want the pincontrol subsystem to handle pin
-	  multiplexing drivers.
 
 config DEBUG_PINCTRL
 	bool "Debug PINCTRL calls"
@@ -30,14 +24,12 @@
 	bool "CSR SiRFprimaII pinmux driver"
 	depends on ARCH_PRIMA2
 	select PINMUX
-	help
-	  Say Y here to enable the SiRFprimaII pinmux driver
 
 config PINMUX_U300
 	bool "U300 pinmux driver"
 	depends on ARCH_U300
 	select PINMUX
-	help
-	  Say Y here to enable the U300 pinmux driver
+
+endmenu
 
 endif
diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index f4e3d82..7f43cf8 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -83,8 +83,10 @@
 	depends on EXPERIMENTAL
 	depends on BACKLIGHT_CLASS_DEVICE
 	depends on RFKILL || RFKILL = n
-	depends on POWER_SUPPLY
 	depends on SERIO_I8042
+	select POWER_SUPPLY
+	select LEDS_CLASS
+	select NEW_LEDS
 	default n
 	---help---
 	This driver adds support for rfkill and backlight control to Dell
diff --git a/drivers/platform/x86/dell-laptop.c b/drivers/platform/x86/dell-laptop.c
index a43cfd9..d93e962 100644
--- a/drivers/platform/x86/dell-laptop.c
+++ b/drivers/platform/x86/dell-laptop.c
@@ -589,14 +589,14 @@
 	.update_status  = dell_send_intensity,
 };
 
-static void touchpad_led_on()
+static void touchpad_led_on(void)
 {
 	int command = 0x97;
 	char data = 1;
 	i8042_command(&data, command | 1 << 12);
 }
 
-static void touchpad_led_off()
+static void touchpad_led_off(void)
 {
 	int command = 0x97;
 	char data = 2;
diff --git a/drivers/platform/x86/toshiba_acpi.c b/drivers/platform/x86/toshiba_acpi.c
index 13ef8c3..dcdc1f4 100644
--- a/drivers/platform/x86/toshiba_acpi.c
+++ b/drivers/platform/x86/toshiba_acpi.c
@@ -121,6 +121,7 @@
 	int illumination_supported:1;
 	int video_supported:1;
 	int fan_supported:1;
+	int system_event_supported:1;
 
 	struct mutex mutex;
 };
@@ -724,7 +725,7 @@
 	u32 hci_result;
 	u32 value;
 
-	if (!dev->key_event_valid) {
+	if (!dev->key_event_valid && dev->system_event_supported) {
 		hci_read1(dev, HCI_SYSTEM_EVENT, &value, &hci_result);
 		if (hci_result == HCI_SUCCESS) {
 			dev->key_event_valid = 1;
@@ -964,6 +965,8 @@
 
 	/* enable event fifo */
 	hci_write1(dev, HCI_SYSTEM_EVENT, 1, &hci_result);
+	if (hci_result == HCI_SUCCESS)
+		dev->system_event_supported = 1;
 
 	props.type = BACKLIGHT_PLATFORM;
 	props.max_brightness = HCI_LCD_BRIGHTNESS_LEVELS - 1;
@@ -1032,12 +1035,15 @@
 {
 	struct toshiba_acpi_dev *dev = acpi_driver_data(acpi_dev);
 	u32 hci_result, value;
+	int retries = 3;
 
-	if (event != 0x80)
+	if (!dev->system_event_supported || event != 0x80)
 		return;
+
 	do {
 		hci_read1(dev, HCI_SYSTEM_EVENT, &value, &hci_result);
-		if (hci_result == HCI_SUCCESS) {
+		switch (hci_result) {
+		case HCI_SUCCESS:
 			if (value == 0x100)
 				continue;
 			/* act on key press; ignore key release */
@@ -1049,14 +1055,19 @@
 				pr_info("Unknown key %x\n",
 				       value);
 			}
-		} else if (hci_result == HCI_NOT_SUPPORTED) {
+			break;
+		case HCI_NOT_SUPPORTED:
 			/* This is a workaround for an unresolved issue on
 			 * some machines where system events sporadically
 			 * become disabled. */
 			hci_write1(dev, HCI_SYSTEM_EVENT, 1, &hci_result);
 			pr_notice("Re-enabled hotkeys\n");
+			/* fall through */
+		default:
+			retries--;
+			break;
 		}
-	} while (hci_result != HCI_EMPTY);
+	} while (retries && hci_result != HCI_EMPTY);
 }
 
 
diff --git a/drivers/power/intel_mid_battery.c b/drivers/power/intel_mid_battery.c
index cffcb7c..01fa671 100644
--- a/drivers/power/intel_mid_battery.c
+++ b/drivers/power/intel_mid_battery.c
@@ -61,7 +61,8 @@
 #define PMIC_BATT_CHR_SBATDET_MASK	(1 << 5)
 #define PMIC_BATT_CHR_SDCLMT_MASK	(1 << 6)
 #define PMIC_BATT_CHR_SUSBOVP_MASK	(1 << 7)
-#define PMIC_BATT_CHR_EXCPT_MASK	0xC6
+#define PMIC_BATT_CHR_EXCPT_MASK	0x86
+
 #define PMIC_BATT_ADC_ACCCHRG_MASK	(1 << 31)
 #define PMIC_BATT_ADC_ACCCHRGVAL_MASK	0x7FFFFFFF
 
@@ -304,11 +305,6 @@
 			pbi->batt_status = POWER_SUPPLY_STATUS_NOT_CHARGING;
 			pmic_battery_log_event(BATT_EVENT_BATOVP_EXCPT);
 			batt_exception = 1;
-		} else if (r8 & PMIC_BATT_CHR_SDCLMT_MASK) {
-			pbi->batt_health = POWER_SUPPLY_HEALTH_OVERVOLTAGE;
-			pbi->batt_status = POWER_SUPPLY_STATUS_NOT_CHARGING;
-			pmic_battery_log_event(BATT_EVENT_DCLMT_EXCPT);
-			batt_exception = 1;
 		} else if (r8 & PMIC_BATT_CHR_STEMP_MASK) {
 			pbi->batt_health = POWER_SUPPLY_HEALTH_OVERHEAT;
 			pbi->batt_status = POWER_SUPPLY_STATUS_NOT_CHARGING;
@@ -316,6 +312,10 @@
 			batt_exception = 1;
 		} else {
 			pbi->batt_health = POWER_SUPPLY_HEALTH_GOOD;
+			if (r8 & PMIC_BATT_CHR_SDCLMT_MASK) {
+				/* PMIC will change charging current automatically */
+				pmic_battery_log_event(BATT_EVENT_DCLMT_EXCPT);
+			}
 		}
 	}
 
diff --git a/drivers/ps3/ps3-vuart.c b/drivers/ps3/ps3-vuart.c
index d9fb729..fb73008 100644
--- a/drivers/ps3/ps3-vuart.c
+++ b/drivers/ps3/ps3-vuart.c
@@ -952,7 +952,7 @@
 	}
 
 	result = request_irq(vuart_bus_priv.virq, ps3_vuart_irq_handler,
-		IRQF_DISABLED, "vuart", &vuart_bus_priv);
+		0, "vuart", &vuart_bus_priv);
 
 	if (result) {
 		pr_debug("%s:%d: request_irq failed (%d)\n",
diff --git a/drivers/ps3/ps3stor_lib.c b/drivers/ps3/ps3stor_lib.c
index cc328dec..8c3f5ad 100644
--- a/drivers/ps3/ps3stor_lib.c
+++ b/drivers/ps3/ps3stor_lib.c
@@ -167,7 +167,7 @@
 		goto fail_close_device;
 	}
 
-	error = request_irq(dev->irq, handler, IRQF_DISABLED,
+	error = request_irq(dev->irq, handler, 0,
 			    dev->sbd.core.driver->name, dev);
 	if (error) {
 		dev_err(&dev->sbd.core, "%s:%u: request_irq failed %d\n",
diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c
index cf3f999..10451a1 100644
--- a/drivers/ptp/ptp_clock.c
+++ b/drivers/ptp/ptp_clock.c
@@ -101,7 +101,9 @@
 
 static int ptp_clock_getres(struct posix_clock *pc, struct timespec *tp)
 {
-	return 1; /* always round timer functions to one nanosecond */
+	tp->tv_sec = 0;
+	tp->tv_nsec = 1;
+	return 0;
 }
 
 static int ptp_clock_settime(struct posix_clock *pc, const struct timespec *tp)
diff --git a/drivers/rapidio/devices/tsi721.c b/drivers/rapidio/devices/tsi721.c
index 5225930..691b1ab 100644
--- a/drivers/rapidio/devices/tsi721.c
+++ b/drivers/rapidio/devices/tsi721.c
@@ -851,14 +851,12 @@
 	INIT_WORK(&priv->idb_work, tsi721_db_dpc);
 
 	/* Allocate buffer for inbound doorbells queue */
-	priv->idb_base = dma_alloc_coherent(&priv->pdev->dev,
+	priv->idb_base = dma_zalloc_coherent(&priv->pdev->dev,
 				IDB_QSIZE * TSI721_IDB_ENTRY_SIZE,
 				&priv->idb_dma, GFP_KERNEL);
 	if (!priv->idb_base)
 		return -ENOMEM;
 
-	memset(priv->idb_base, 0, IDB_QSIZE * TSI721_IDB_ENTRY_SIZE);
-
 	dev_dbg(&priv->pdev->dev, "Allocated IDB buffer @ %p (phys = %llx)\n",
 		priv->idb_base, (unsigned long long)priv->idb_dma);
 
@@ -904,7 +902,7 @@
 	 */
 
 	/* Allocate space for DMA descriptors */
-	bd_ptr = dma_alloc_coherent(&priv->pdev->dev,
+	bd_ptr = dma_zalloc_coherent(&priv->pdev->dev,
 					bd_num * sizeof(struct tsi721_dma_desc),
 					&bd_phys, GFP_KERNEL);
 	if (!bd_ptr)
@@ -913,8 +911,6 @@
 	priv->bdma[chnum].bd_phys = bd_phys;
 	priv->bdma[chnum].bd_base = bd_ptr;
 
-	memset(bd_ptr, 0, bd_num * sizeof(struct tsi721_dma_desc));
-
 	dev_dbg(&priv->pdev->dev, "DMA descriptors @ %p (phys = %llx)\n",
 		bd_ptr, (unsigned long long)bd_phys);
 
@@ -922,7 +918,7 @@
 	sts_size = (bd_num >= TSI721_DMA_MINSTSSZ) ?
 					bd_num : TSI721_DMA_MINSTSSZ;
 	sts_size = roundup_pow_of_two(sts_size);
-	sts_ptr = dma_alloc_coherent(&priv->pdev->dev,
+	sts_ptr = dma_zalloc_coherent(&priv->pdev->dev,
 				     sts_size * sizeof(struct tsi721_dma_sts),
 				     &sts_phys, GFP_KERNEL);
 	if (!sts_ptr) {
@@ -938,8 +934,6 @@
 	priv->bdma[chnum].sts_base = sts_ptr;
 	priv->bdma[chnum].sts_size = sts_size;
 
-	memset(sts_ptr, 0, sts_size);
-
 	dev_dbg(&priv->pdev->dev,
 		"desc status FIFO @ %p (phys = %llx) size=0x%x\n",
 		sts_ptr, (unsigned long long)sts_phys, sts_size);
@@ -1400,7 +1394,7 @@
 
 	/* Outbound message descriptor status FIFO allocation */
 	priv->omsg_ring[mbox].sts_size = roundup_pow_of_two(entries + 1);
-	priv->omsg_ring[mbox].sts_base = dma_alloc_coherent(&priv->pdev->dev,
+	priv->omsg_ring[mbox].sts_base = dma_zalloc_coherent(&priv->pdev->dev,
 			priv->omsg_ring[mbox].sts_size *
 						sizeof(struct tsi721_dma_sts),
 			&priv->omsg_ring[mbox].sts_phys, GFP_KERNEL);
@@ -1412,9 +1406,6 @@
 		goto out_desc;
 	}
 
-	memset(priv->omsg_ring[mbox].sts_base, 0,
-		entries * sizeof(struct tsi721_dma_sts));
-
 	/*
 	 * Configure Outbound Messaging Engine
 	 */
@@ -2116,8 +2107,8 @@
 	INIT_LIST_HEAD(&mport->dbells);
 
 	rio_init_dbell_res(&mport->riores[RIO_DOORBELL_RESOURCE], 0, 0xffff);
-	rio_init_mbox_res(&mport->riores[RIO_INB_MBOX_RESOURCE], 0, 0);
-	rio_init_mbox_res(&mport->riores[RIO_OUTB_MBOX_RESOURCE], 0, 0);
+	rio_init_mbox_res(&mport->riores[RIO_INB_MBOX_RESOURCE], 0, 3);
+	rio_init_mbox_res(&mport->riores[RIO_OUTB_MBOX_RESOURCE], 0, 3);
 	strcpy(mport->name, "Tsi721 mport");
 
 	/* Hook up interrupt handler */
@@ -2163,7 +2154,7 @@
 				  const struct pci_device_id *id)
 {
 	struct tsi721_device *priv;
-	int i;
+	int i, cap;
 	int err;
 	u32 regval;
 
@@ -2271,10 +2262,20 @@
 			dev_info(&pdev->dev, "Unable to set consistent DMA mask\n");
 	}
 
-	/* Clear "no snoop" and "relaxed ordering" bits. */
-	pci_read_config_dword(pdev, 0x40 + PCI_EXP_DEVCTL, &regval);
-	regval &= ~(PCI_EXP_DEVCTL_RELAX_EN | PCI_EXP_DEVCTL_NOSNOOP_EN);
-	pci_write_config_dword(pdev, 0x40 + PCI_EXP_DEVCTL, regval);
+	cap = pci_pcie_cap(pdev);
+	BUG_ON(cap == 0);
+
+	/* Clear "no snoop" and "relaxed ordering" bits, use default MRRS. */
+	pci_read_config_dword(pdev, cap + PCI_EXP_DEVCTL, &regval);
+	regval &= ~(PCI_EXP_DEVCTL_READRQ | PCI_EXP_DEVCTL_RELAX_EN |
+		    PCI_EXP_DEVCTL_NOSNOOP_EN);
+	regval |= 0x2 << MAX_READ_REQUEST_SZ_SHIFT;
+	pci_write_config_dword(pdev, cap + PCI_EXP_DEVCTL, regval);
+
+	/* Adjust PCIe completion timeout. */
+	pci_read_config_dword(pdev, cap + PCI_EXP_DEVCTL2, &regval);
+	regval &= ~(0x0f);
+	pci_write_config_dword(pdev, cap + PCI_EXP_DEVCTL2, regval | 0x2);
 
 	/*
 	 * FIXUP: correct offsets of MSI-X tables in the MSI-X Capability Block
diff --git a/drivers/rapidio/devices/tsi721.h b/drivers/rapidio/devices/tsi721.h
index 58be4de..822e54c3 100644
--- a/drivers/rapidio/devices/tsi721.h
+++ b/drivers/rapidio/devices/tsi721.h
@@ -72,6 +72,8 @@
 #define TSI721_MSIXPBA_OFFSET	0x2a000
 #define TSI721_PCIECFG_EPCTL	0x400
 
+#define MAX_READ_REQUEST_SZ_SHIFT	12
+
 /*
  * Event Management Registers
  */
diff --git a/drivers/regulator/aat2870-regulator.c b/drivers/regulator/aat2870-regulator.c
index 5abeb3a..298c6c6 100644
--- a/drivers/regulator/aat2870-regulator.c
+++ b/drivers/regulator/aat2870-regulator.c
@@ -160,7 +160,7 @@
 			break;
 	}
 
-	if (!ri)
+	if (i == ARRAY_SIZE(aat2870_regulators))
 		return NULL;
 
 	ri->enable_addr = AAT2870_LDO_EN;
diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index 669d021..938398f 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -2799,8 +2799,8 @@
 	list_del(&rdev->list);
 	if (rdev->supply)
 		regulator_put(rdev->supply);
-	device_unregister(&rdev->dev);
 	kfree(rdev->constraints);
+	device_unregister(&rdev->dev);
 	mutex_unlock(&regulator_list_mutex);
 }
 EXPORT_SYMBOL_GPL(regulator_unregister);
diff --git a/drivers/regulator/tps65910-regulator.c b/drivers/regulator/tps65910-regulator.c
index 66d2d60..b552aae 100644
--- a/drivers/regulator/tps65910-regulator.c
+++ b/drivers/regulator/tps65910-regulator.c
@@ -664,10 +664,10 @@
 
 	switch (id) {
 	case TPS65910_REG_VDD1:
-		dcdc_mult = (selector / VDD1_2_NUM_VOLTS) + 1;
+		dcdc_mult = (selector / VDD1_2_NUM_VOLT_FINE) + 1;
 		if (dcdc_mult == 1)
 			dcdc_mult--;
-		vsel = (selector % VDD1_2_NUM_VOLTS) + 3;
+		vsel = (selector % VDD1_2_NUM_VOLT_FINE) + 3;
 
 		tps65910_modify_bits(pmic, TPS65910_VDD1,
 				(dcdc_mult << VDD1_VGAIN_SEL_SHIFT),
@@ -675,10 +675,10 @@
 		tps65910_reg_write(pmic, TPS65910_VDD1_OP, vsel);
 		break;
 	case TPS65910_REG_VDD2:
-		dcdc_mult = (selector / VDD1_2_NUM_VOLTS) + 1;
+		dcdc_mult = (selector / VDD1_2_NUM_VOLT_FINE) + 1;
 		if (dcdc_mult == 1)
 			dcdc_mult--;
-		vsel = (selector % VDD1_2_NUM_VOLTS) + 3;
+		vsel = (selector % VDD1_2_NUM_VOLT_FINE) + 3;
 
 		tps65910_modify_bits(pmic, TPS65910_VDD2,
 				(dcdc_mult << VDD2_VGAIN_SEL_SHIFT),
@@ -756,9 +756,9 @@
 	switch (id) {
 	case TPS65910_REG_VDD1:
 	case TPS65910_REG_VDD2:
-		mult = (selector / VDD1_2_NUM_VOLTS) + 1;
+		mult = (selector / VDD1_2_NUM_VOLT_FINE) + 1;
 		volt = VDD1_2_MIN_VOLT +
-				(selector % VDD1_2_NUM_VOLTS) * VDD1_2_OFFSET;
+				(selector % VDD1_2_NUM_VOLT_FINE) * VDD1_2_OFFSET;
 		break;
 	case TPS65911_REG_VDDCTRL:
 		volt = VDDCTRL_MIN_VOLT + (selector * VDDCTRL_OFFSET);
@@ -947,6 +947,8 @@
 
 		if (i == TPS65910_REG_VDD1 || i == TPS65910_REG_VDD2) {
 			pmic->desc[i].ops = &tps65910_ops_dcdc;
+			pmic->desc[i].n_voltages = VDD1_2_NUM_VOLT_FINE *
+							VDD1_2_NUM_VOLT_COARSE;
 		} else if (i == TPS65910_REG_VDD3) {
 			if (tps65910_chip_id(tps65910) == TPS65910)
 				pmic->desc[i].ops = &tps65910_ops_vdd3;
diff --git a/drivers/regulator/twl-regulator.c b/drivers/regulator/twl-regulator.c
index ee8747f..11cc308 100644
--- a/drivers/regulator/twl-regulator.c
+++ b/drivers/regulator/twl-regulator.c
@@ -71,6 +71,7 @@
 #define VREG_TYPE		1
 #define VREG_REMAP		2
 #define VREG_DEDICATED		3	/* LDO control */
+#define VREG_VOLTAGE_SMPS_4030	9
 /* TWL6030 register offsets */
 #define VREG_TRANS		1
 #define VREG_STATE		2
@@ -514,6 +515,32 @@
 	.get_status	= twl4030reg_get_status,
 };
 
+static int
+twl4030smps_set_voltage(struct regulator_dev *rdev, int min_uV, int max_uV,
+			unsigned *selector)
+{
+	struct twlreg_info *info = rdev_get_drvdata(rdev);
+	int vsel = DIV_ROUND_UP(min_uV - 600000, 12500);
+
+	twlreg_write(info, TWL_MODULE_PM_RECEIVER, VREG_VOLTAGE_SMPS_4030,
+		vsel);
+	return 0;
+}
+
+static int twl4030smps_get_voltage(struct regulator_dev *rdev)
+{
+	struct twlreg_info *info = rdev_get_drvdata(rdev);
+	int vsel = twlreg_read(info, TWL_MODULE_PM_RECEIVER,
+		VREG_VOLTAGE_SMPS_4030);
+
+	return vsel * 12500 + 600000;
+}
+
+static struct regulator_ops twl4030smps_ops = {
+	.set_voltage	= twl4030smps_set_voltage,
+	.get_voltage	= twl4030smps_get_voltage,
+};
+
 static int twl6030ldo_list_voltage(struct regulator_dev *rdev, unsigned index)
 {
 	struct twlreg_info	*info = rdev_get_drvdata(rdev);
@@ -856,6 +883,21 @@
 		}, \
 	}
 
+#define TWL4030_ADJUSTABLE_SMPS(label, offset, num, turnon_delay, remap_conf) \
+	{ \
+	.base = offset, \
+	.id = num, \
+	.delay = turnon_delay, \
+	.remap = remap_conf, \
+	.desc = { \
+		.name = #label, \
+		.id = TWL4030_REG_##label, \
+		.ops = &twl4030smps_ops, \
+		.type = REGULATOR_VOLTAGE, \
+		.owner = THIS_MODULE, \
+		}, \
+	}
+
 #define TWL6030_ADJUSTABLE_LDO(label, offset, min_mVolts, max_mVolts) { \
 	.base = offset, \
 	.min_mV = min_mVolts, \
@@ -947,8 +989,8 @@
 	TWL4030_ADJUSTABLE_LDO(VINTANA2, 0x43, 12, 100, 0x08),
 	TWL4030_FIXED_LDO(VINTDIG, 0x47, 1500, 13, 100, 0x08),
 	TWL4030_ADJUSTABLE_LDO(VIO, 0x4b, 14, 1000, 0x08),
-	TWL4030_ADJUSTABLE_LDO(VDD1, 0x55, 15, 1000, 0x08),
-	TWL4030_ADJUSTABLE_LDO(VDD2, 0x63, 16, 1000, 0x08),
+	TWL4030_ADJUSTABLE_SMPS(VDD1, 0x55, 15, 1000, 0x08),
+	TWL4030_ADJUSTABLE_SMPS(VDD2, 0x63, 16, 1000, 0x08),
 	TWL4030_FIXED_LDO(VUSB1V5, 0x71, 1500, 17, 100, 0x08),
 	TWL4030_FIXED_LDO(VUSB1V8, 0x74, 1800, 18, 100, 0x08),
 	TWL4030_FIXED_LDO(VUSB3V1, 0x77, 3100, 19, 150, 0x08),
diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c
index e8326f2..dc4c274 100644
--- a/drivers/rtc/class.c
+++ b/drivers/rtc/class.c
@@ -63,7 +63,7 @@
 	 */
 	delta = timespec_sub(old_system, old_rtc);
 	delta_delta = timespec_sub(delta, old_delta);
-	if (abs(delta_delta.tv_sec)  >= 2) {
+	if (delta_delta.tv_sec < -2 || delta_delta.tv_sec >= 2) {
 		/*
 		 * if delta_delta is too large, assume time correction
 		 * has occured and set old_delta to the current delta.
@@ -97,9 +97,8 @@
 	rtc_tm_to_time(&tm, &new_rtc.tv_sec);
 	new_rtc.tv_nsec = 0;
 
-	if (new_rtc.tv_sec <= old_rtc.tv_sec) {
-		if (new_rtc.tv_sec < old_rtc.tv_sec)
-			pr_debug("%s:  time travel!\n", dev_name(&rtc->dev));
+	if (new_rtc.tv_sec < old_rtc.tv_sec) {
+		pr_debug("%s:  time travel!\n", dev_name(&rtc->dev));
 		return 0;
 	}
 
@@ -116,7 +115,8 @@
 	sleep_time = timespec_sub(sleep_time,
 			timespec_sub(new_system, old_system));
 
-	timekeeping_inject_sleeptime(&sleep_time);
+	if (sleep_time.tv_sec >= 0)
+		timekeeping_inject_sleeptime(&sleep_time);
 	return 0;
 }
 
diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index 8e28625..fa4d9f3 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -319,6 +319,20 @@
 }
 EXPORT_SYMBOL_GPL(rtc_read_alarm);
 
+static int ___rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
+{
+	int err;
+
+	if (!rtc->ops)
+		err = -ENODEV;
+	else if (!rtc->ops->set_alarm)
+		err = -EINVAL;
+	else
+		err = rtc->ops->set_alarm(rtc->dev.parent, alarm);
+
+	return err;
+}
+
 static int __rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 {
 	struct rtc_time tm;
@@ -342,14 +356,7 @@
 	 * over right here, before we set the alarm.
 	 */
 
-	if (!rtc->ops)
-		err = -ENODEV;
-	else if (!rtc->ops->set_alarm)
-		err = -EINVAL;
-	else
-		err = rtc->ops->set_alarm(rtc->dev.parent, alarm);
-
-	return err;
+	return ___rtc_set_alarm(rtc, alarm);
 }
 
 int rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
@@ -763,6 +770,20 @@
 	return 0;
 }
 
+static void rtc_alarm_disable(struct rtc_device *rtc)
+{
+	struct rtc_wkalrm alarm;
+	struct rtc_time tm;
+
+	__rtc_read_time(rtc, &tm);
+
+	alarm.time = rtc_ktime_to_tm(ktime_add(rtc_tm_to_ktime(tm),
+				     ktime_set(300, 0)));
+	alarm.enabled = 0;
+
+	___rtc_set_alarm(rtc, &alarm);
+}
+
 /**
  * rtc_timer_remove - Removes a rtc_timer from the rtc_device timerqueue
  * @rtc rtc device
@@ -784,8 +805,10 @@
 		struct rtc_wkalrm alarm;
 		int err;
 		next = timerqueue_getnext(&rtc->timerqueue);
-		if (!next)
+		if (!next) {
+			rtc_alarm_disable(rtc);
 			return;
+		}
 		alarm.time = rtc_ktime_to_tm(next->expires);
 		alarm.enabled = 1;
 		err = __rtc_set_alarm(rtc, &alarm);
@@ -847,7 +870,8 @@
 		err = __rtc_set_alarm(rtc, &alarm);
 		if (err == -ETIME)
 			goto again;
-	}
+	} else
+		rtc_alarm_disable(rtc);
 
 	mutex_unlock(&rtc->ops_lock);
 }
diff --git a/drivers/rtc/rtc-mrst.c b/drivers/rtc/rtc-mrst.c
index d335448..bb21f44 100644
--- a/drivers/rtc/rtc-mrst.c
+++ b/drivers/rtc/rtc-mrst.c
@@ -76,12 +76,15 @@
 /*
  * rtc_time's year contains the increment over 1900, but vRTC's YEAR
  * register can't be programmed to value larger than 0x64, so vRTC
- * driver chose to use 1960 (1970 is UNIX time start point) as the base,
+ * driver chose to use 1972 (1970 is UNIX time start point) as the base,
  * and does the translation at read/write time.
  *
- * Why not just use 1970 as the offset? it's because using 1960 will
+ * Why not just use 1970 as the offset? it's because using 1972 will
  * make it consistent in leap year setting for both vrtc and low-level
- * physical rtc devices.
+ * physical rtc devices. Then why not use 1960 as the offset? If we use
+ * 1960, for a device's first use, its YEAR register is 0 and the system
+ * year will be parsed as 1960 which is not a valid UNIX time and will
+ * cause many applications to fail mysteriously.
  */
 static int mrst_read_time(struct device *dev, struct rtc_time *time)
 {
@@ -99,10 +102,10 @@
 	time->tm_year = vrtc_cmos_read(RTC_YEAR);
 	spin_unlock_irqrestore(&rtc_lock, flags);
 
-	/* Adjust for the 1960/1900 */
-	time->tm_year += 60;
+	/* Adjust for the 1972/1900 */
+	time->tm_year += 72;
 	time->tm_mon--;
-	return RTC_24H;
+	return rtc_valid_tm(time);
 }
 
 static int mrst_set_time(struct device *dev, struct rtc_time *time)
@@ -119,9 +122,9 @@
 	min = time->tm_min;
 	sec = time->tm_sec;
 
-	if (yrs < 70 || yrs > 138)
+	if (yrs < 72 || yrs > 138)
 		return -EINVAL;
-	yrs -= 60;
+	yrs -= 72;
 
 	spin_lock_irqsave(&rtc_lock, flags);
 
diff --git a/drivers/rtc/rtc-puv3.c b/drivers/rtc/rtc-puv3.c
index b3eba3c..e4b6880 100644
--- a/drivers/rtc/rtc-puv3.c
+++ b/drivers/rtc/rtc-puv3.c
@@ -220,7 +220,7 @@
 	}
 }
 
-static int puv3_rtc_remove(struct platform_device *dev)
+static int __devexit puv3_rtc_remove(struct platform_device *dev)
 {
 	struct rtc_device *rtc = platform_get_drvdata(dev);
 
@@ -236,7 +236,7 @@
 	return 0;
 }
 
-static int puv3_rtc_probe(struct platform_device *pdev)
+static int __devinit puv3_rtc_probe(struct platform_device *pdev)
 {
 	struct rtc_device *rtc;
 	struct resource *res;
diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c
index 7639ab9..5b979d9 100644
--- a/drivers/rtc/rtc-s3c.c
+++ b/drivers/rtc/rtc-s3c.c
@@ -202,7 +202,6 @@
 	void __iomem *base = s3c_rtc_base;
 	int year = tm->tm_year - 100;
 
-	clk_enable(rtc_clk);
 	pr_debug("set time %04d.%02d.%02d %02d:%02d:%02d\n",
 		 1900 + tm->tm_year, tm->tm_mon, tm->tm_mday,
 		 tm->tm_hour, tm->tm_min, tm->tm_sec);
@@ -214,6 +213,7 @@
 		return -EINVAL;
 	}
 
+	clk_enable(rtc_clk);
 	writeb(bin2bcd(tm->tm_sec),  base + S3C2410_RTCSEC);
 	writeb(bin2bcd(tm->tm_min),  base + S3C2410_RTCMIN);
 	writeb(bin2bcd(tm->tm_hour), base + S3C2410_RTCHOUR);
diff --git a/drivers/s390/char/zcore.c b/drivers/s390/char/zcore.c
index 43068fb..1b6d924 100644
--- a/drivers/s390/char/zcore.c
+++ b/drivers/s390/char/zcore.c
@@ -641,6 +641,8 @@
 
 	if (ipl_info.type != IPL_TYPE_FCP_DUMP)
 		return -ENODATA;
+	if (OLDMEM_BASE)
+		return -ENODATA;
 
 	zcore_dbf = debug_register("zcore", 4, 1, 4 * sizeof(long));
 	debug_register_view(zcore_dbf, &debug_sprintf_view);
diff --git a/drivers/s390/cio/chsc.c b/drivers/s390/cio/chsc.c
index 75c3f1f..a84631a 100644
--- a/drivers/s390/cio/chsc.c
+++ b/drivers/s390/cio/chsc.c
@@ -529,10 +529,7 @@
 int chsc_chp_vary(struct chp_id chpid, int on)
 {
 	struct channel_path *chp = chpid_to_chp(chpid);
-	struct chp_link link;
 
-	memset(&link, 0, sizeof(struct chp_link));
-	link.chpid = chpid;
 	/* Wait until previous actions have settled. */
 	css_wait_for_slow_path();
 	/*
@@ -542,10 +539,10 @@
 		/* Try to update the channel path descritor. */
 		chsc_determine_base_channel_path_desc(chpid, &chp->desc);
 		for_each_subchannel_staged(s390_subchannel_vary_chpid_on,
-					   __s390_vary_chpid_on, &link);
+					   __s390_vary_chpid_on, &chpid);
 	} else
 		for_each_subchannel_staged(s390_subchannel_vary_chpid_off,
-					   NULL, &link);
+					   NULL, &chpid);
 
 	return 0;
 }
diff --git a/drivers/s390/cio/cio.h b/drivers/s390/cio/cio.h
index 155a82b..4a1ff5c 100644
--- a/drivers/s390/cio/cio.h
+++ b/drivers/s390/cio/cio.h
@@ -68,8 +68,13 @@
 	__u8 mda[4];		 /* model dependent area */
 } __attribute__ ((packed,aligned(4)));
 
+/*
+ * When rescheduled, todo's with higher values will overwrite those
+ * with lower values.
+ */
 enum sch_todo {
 	SCH_TODO_NOTHING,
+	SCH_TODO_EVAL,
 	SCH_TODO_UNREG,
 };
 
diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c
index 92d7324..21908e6 100644
--- a/drivers/s390/cio/css.c
+++ b/drivers/s390/cio/css.c
@@ -195,51 +195,6 @@
 }
 EXPORT_SYMBOL_GPL(css_sch_device_unregister);
 
-static void css_sch_todo(struct work_struct *work)
-{
-	struct subchannel *sch;
-	enum sch_todo todo;
-
-	sch = container_of(work, struct subchannel, todo_work);
-	/* Find out todo. */
-	spin_lock_irq(sch->lock);
-	todo = sch->todo;
-	CIO_MSG_EVENT(4, "sch_todo: sch=0.%x.%04x, todo=%d\n", sch->schid.ssid,
-		      sch->schid.sch_no, todo);
-	sch->todo = SCH_TODO_NOTHING;
-	spin_unlock_irq(sch->lock);
-	/* Perform todo. */
-	if (todo == SCH_TODO_UNREG)
-		css_sch_device_unregister(sch);
-	/* Release workqueue ref. */
-	put_device(&sch->dev);
-}
-
-/**
- * css_sched_sch_todo - schedule a subchannel operation
- * @sch: subchannel
- * @todo: todo
- *
- * Schedule the operation identified by @todo to be performed on the slow path
- * workqueue. Do nothing if another operation with higher priority is already
- * scheduled. Needs to be called with subchannel lock held.
- */
-void css_sched_sch_todo(struct subchannel *sch, enum sch_todo todo)
-{
-	CIO_MSG_EVENT(4, "sch_todo: sched sch=0.%x.%04x todo=%d\n",
-		      sch->schid.ssid, sch->schid.sch_no, todo);
-	if (sch->todo >= todo)
-		return;
-	/* Get workqueue ref. */
-	if (!get_device(&sch->dev))
-		return;
-	sch->todo = todo;
-	if (!queue_work(cio_work_q, &sch->todo_work)) {
-		/* Already queued, release workqueue ref. */
-		put_device(&sch->dev);
-	}
-}
-
 static void ssd_from_pmcw(struct chsc_ssd_info *ssd, struct pmcw *pmcw)
 {
 	int i;
@@ -466,6 +421,65 @@
 		css_schedule_eval(schid);
 }
 
+/**
+ * css_sched_sch_todo - schedule a subchannel operation
+ * @sch: subchannel
+ * @todo: todo
+ *
+ * Schedule the operation identified by @todo to be performed on the slow path
+ * workqueue. Do nothing if another operation with higher priority is already
+ * scheduled. Needs to be called with subchannel lock held.
+ */
+void css_sched_sch_todo(struct subchannel *sch, enum sch_todo todo)
+{
+	CIO_MSG_EVENT(4, "sch_todo: sched sch=0.%x.%04x todo=%d\n",
+		      sch->schid.ssid, sch->schid.sch_no, todo);
+	if (sch->todo >= todo)
+		return;
+	/* Get workqueue ref. */
+	if (!get_device(&sch->dev))
+		return;
+	sch->todo = todo;
+	if (!queue_work(cio_work_q, &sch->todo_work)) {
+		/* Already queued, release workqueue ref. */
+		put_device(&sch->dev);
+	}
+}
+
+static void css_sch_todo(struct work_struct *work)
+{
+	struct subchannel *sch;
+	enum sch_todo todo;
+	int ret;
+
+	sch = container_of(work, struct subchannel, todo_work);
+	/* Find out todo. */
+	spin_lock_irq(sch->lock);
+	todo = sch->todo;
+	CIO_MSG_EVENT(4, "sch_todo: sch=0.%x.%04x, todo=%d\n", sch->schid.ssid,
+		      sch->schid.sch_no, todo);
+	sch->todo = SCH_TODO_NOTHING;
+	spin_unlock_irq(sch->lock);
+	/* Perform todo. */
+	switch (todo) {
+	case SCH_TODO_NOTHING:
+		break;
+	case SCH_TODO_EVAL:
+		ret = css_evaluate_known_subchannel(sch, 1);
+		if (ret == -EAGAIN) {
+			spin_lock_irq(sch->lock);
+			css_sched_sch_todo(sch, todo);
+			spin_unlock_irq(sch->lock);
+		}
+		break;
+	case SCH_TODO_UNREG:
+		css_sch_device_unregister(sch);
+		break;
+	}
+	/* Release workqueue ref. */
+	put_device(&sch->dev);
+}
+
 static struct idset *slow_subchannel_set;
 static spinlock_t slow_subchannel_lock;
 static wait_queue_head_t css_eval_wq;
diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c
index d734f4a..4726985 100644
--- a/drivers/s390/cio/device.c
+++ b/drivers/s390/cio/device.c
@@ -1868,9 +1868,9 @@
 	 */
 	cdev->private->flags.resuming = 1;
 	cdev->private->path_new_mask = LPM_ANYPATH;
-	css_schedule_eval(sch->schid);
+	css_sched_sch_todo(sch, SCH_TODO_EVAL);
 	spin_unlock_irq(sch->lock);
-	css_complete_work();
+	css_wait_for_slow_path();
 
 	/* cdev may have been moved to a different subchannel. */
 	sch = to_subchannel(cdev->dev.parent);
diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c
index 52c233f..1b85351 100644
--- a/drivers/s390/cio/device_fsm.c
+++ b/drivers/s390/cio/device_fsm.c
@@ -496,8 +496,26 @@
 	cdev->private->pgid_reset_mask = 0;
 }
 
-void
-ccw_device_verify_done(struct ccw_device *cdev, int err)
+static void create_fake_irb(struct irb *irb, int type)
+{
+	memset(irb, 0, sizeof(*irb));
+	if (type == FAKE_CMD_IRB) {
+		struct cmd_scsw *scsw = &irb->scsw.cmd;
+		scsw->cc = 1;
+		scsw->fctl = SCSW_FCTL_START_FUNC;
+		scsw->actl = SCSW_ACTL_START_PEND;
+		scsw->stctl = SCSW_STCTL_STATUS_PEND;
+	} else if (type == FAKE_TM_IRB) {
+		struct tm_scsw *scsw = &irb->scsw.tm;
+		scsw->x = 1;
+		scsw->cc = 1;
+		scsw->fctl = SCSW_FCTL_START_FUNC;
+		scsw->actl = SCSW_ACTL_START_PEND;
+		scsw->stctl = SCSW_STCTL_STATUS_PEND;
+	}
+}
+
+void ccw_device_verify_done(struct ccw_device *cdev, int err)
 {
 	struct subchannel *sch;
 
@@ -520,12 +538,8 @@
 		ccw_device_done(cdev, DEV_STATE_ONLINE);
 		/* Deliver fake irb to device driver, if needed. */
 		if (cdev->private->flags.fake_irb) {
-			memset(&cdev->private->irb, 0, sizeof(struct irb));
-			cdev->private->irb.scsw.cmd.cc = 1;
-			cdev->private->irb.scsw.cmd.fctl = SCSW_FCTL_START_FUNC;
-			cdev->private->irb.scsw.cmd.actl = SCSW_ACTL_START_PEND;
-			cdev->private->irb.scsw.cmd.stctl =
-				SCSW_STCTL_STATUS_PEND;
+			create_fake_irb(&cdev->private->irb,
+					cdev->private->flags.fake_irb);
 			cdev->private->flags.fake_irb = 0;
 			if (cdev->handler)
 				cdev->handler(cdev, cdev->private->intparm,
diff --git a/drivers/s390/cio/device_ops.c b/drivers/s390/cio/device_ops.c
index f98698d..ec7fb6d 100644
--- a/drivers/s390/cio/device_ops.c
+++ b/drivers/s390/cio/device_ops.c
@@ -198,7 +198,7 @@
 	if (cdev->private->state == DEV_STATE_VERIFY) {
 		/* Remember to fake irb when finished. */
 		if (!cdev->private->flags.fake_irb) {
-			cdev->private->flags.fake_irb = 1;
+			cdev->private->flags.fake_irb = FAKE_CMD_IRB;
 			cdev->private->intparm = intparm;
 			return 0;
 		} else
@@ -213,9 +213,9 @@
 	ret = cio_set_options (sch, flags);
 	if (ret)
 		return ret;
-	/* Adjust requested path mask to excluded varied off paths. */
+	/* Adjust requested path mask to exclude unusable paths. */
 	if (lpm) {
-		lpm &= sch->opm;
+		lpm &= sch->lpm;
 		if (lpm == 0)
 			return -EACCES;
 	}
@@ -605,11 +605,21 @@
 	sch = to_subchannel(cdev->dev.parent);
 	if (!sch->schib.pmcw.ena)
 		return -EINVAL;
+	if (cdev->private->state == DEV_STATE_VERIFY) {
+		/* Remember to fake irb when finished. */
+		if (!cdev->private->flags.fake_irb) {
+			cdev->private->flags.fake_irb = FAKE_TM_IRB;
+			cdev->private->intparm = intparm;
+			return 0;
+		} else
+			/* There's already a fake I/O around. */
+			return -EBUSY;
+	}
 	if (cdev->private->state != DEV_STATE_ONLINE)
 		return -EIO;
-	/* Adjust requested path mask to excluded varied off paths. */
+	/* Adjust requested path mask to exclude unusable paths. */
 	if (lpm) {
-		lpm &= sch->opm;
+		lpm &= sch->lpm;
 		if (lpm == 0)
 			return -EACCES;
 	}
diff --git a/drivers/s390/cio/io_sch.h b/drivers/s390/cio/io_sch.h
index 2ebb492..76253df 100644
--- a/drivers/s390/cio/io_sch.h
+++ b/drivers/s390/cio/io_sch.h
@@ -111,6 +111,9 @@
 	CDEV_TODO_UNREG_EVAL,
 };
 
+#define FAKE_CMD_IRB	1
+#define FAKE_TM_IRB	2
+
 struct ccw_device_private {
 	struct ccw_device *cdev;
 	struct subchannel *sch;
@@ -138,7 +141,7 @@
 		unsigned int doverify:1;    /* delayed path verification */
 		unsigned int donotify:1;    /* call notify function */
 		unsigned int recog_done:1;  /* dev. recog. complete */
-		unsigned int fake_irb:1;    /* deliver faked irb */
+		unsigned int fake_irb:2;    /* deliver faked irb */
 		unsigned int resuming:1;    /* recognition while resume */
 		unsigned int pgroup:1;	    /* pathgroup is set up */
 		unsigned int mpath:1;	    /* multipathing is set up */
diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c
index b77ae51..96bbe9d 100644
--- a/drivers/s390/crypto/ap_bus.c
+++ b/drivers/s390/crypto/ap_bus.c
@@ -1271,18 +1271,16 @@
 }
 
 /**
- * ap_schedule_poll_timer(): Schedule poll timer.
+ * __ap_schedule_poll_timer(): Schedule poll timer.
  *
  * Set up the timer to run the poll tasklet
  */
-static inline void ap_schedule_poll_timer(void)
+static inline void __ap_schedule_poll_timer(void)
 {
 	ktime_t hr_time;
 
 	spin_lock_bh(&ap_poll_timer_lock);
-	if (ap_using_interrupts() || ap_suspend_flag)
-		goto out;
-	if (hrtimer_is_queued(&ap_poll_timer))
+	if (hrtimer_is_queued(&ap_poll_timer) || ap_suspend_flag)
 		goto out;
 	if (ktime_to_ns(hrtimer_expires_remaining(&ap_poll_timer)) <= 0) {
 		hr_time = ktime_set(0, poll_timeout);
@@ -1294,6 +1292,18 @@
 }
 
 /**
+ * ap_schedule_poll_timer(): Schedule poll timer.
+ *
+ * Set up the timer to run the poll tasklet
+ */
+static inline void ap_schedule_poll_timer(void)
+{
+	if (ap_using_interrupts())
+		return;
+	__ap_schedule_poll_timer();
+}
+
+/**
  * ap_poll_read(): Receive pending reply messages from an AP device.
  * @ap_dev: pointer to the AP device
  * @flags: pointer to control flags, bit 2^0 is set if another poll is
@@ -1374,8 +1384,9 @@
 			*flags |= 1;
 		*flags |= 2;
 		break;
-	case AP_RESPONSE_Q_FULL:
 	case AP_RESPONSE_RESET_IN_PROGRESS:
+		__ap_schedule_poll_timer();
+	case AP_RESPONSE_Q_FULL:
 		*flags |= 2;
 		break;
 	case AP_RESPONSE_MESSAGE_TOO_BIG:
@@ -1541,6 +1552,8 @@
 	rc = ap_init_queue(ap_dev->qid);
 	if (rc == -ENODEV)
 		ap_dev->unregistered = 1;
+	else
+		__ap_schedule_poll_timer();
 }
 
 static int __ap_poll_device(struct ap_device *ap_dev, unsigned long *flags)
diff --git a/drivers/s390/net/Kconfig b/drivers/s390/net/Kconfig
index fa80ba1..9b66d2d 100644
--- a/drivers/s390/net/Kconfig
+++ b/drivers/s390/net/Kconfig
@@ -4,7 +4,7 @@
 config LCS
 	def_tristate m
 	prompt "Lan Channel Station Interface"
-	depends on CCW && NETDEVICES && (NET_ETHERNET || TR || FDDI)
+	depends on CCW && NETDEVICES && (ETHERNET || TR || FDDI)
 	help
 	   Select this option if you want to use LCS networking on IBM System z.
 	   This device driver supports Token Ring (IEEE 802.5),
diff --git a/drivers/s390/net/lcs.c b/drivers/s390/net/lcs.c
index c28713d..863fc21 100644
--- a/drivers/s390/net/lcs.c
+++ b/drivers/s390/net/lcs.c
@@ -50,7 +50,7 @@
 #include "lcs.h"
 
 
-#if !defined(CONFIG_NET_ETHERNET) && \
+#if !defined(CONFIG_ETHERNET) && \
     !defined(CONFIG_TR) && !defined(CONFIG_FDDI)
 #error Cannot compile lcs.c without some net devices switched on.
 #endif
@@ -1634,7 +1634,7 @@
 	int rc;
 
 	LCS_DBF_TEXT(2, trace, "strtauto");
-#ifdef CONFIG_NET_ETHERNET
+#ifdef CONFIG_ETHERNET
 	card->lan_type = LCS_FRAME_TYPE_ENET;
 	rc = lcs_send_startlan(card, LCS_INITIATOR_TCPIP);
 	if (rc == 0)
@@ -2166,7 +2166,7 @@
 		goto netdev_out;
 	}
 	switch (card->lan_type) {
-#ifdef CONFIG_NET_ETHERNET
+#ifdef CONFIG_ETHERNET
 	case LCS_FRAME_TYPE_ENET:
 		card->lan_type_trans = eth_type_trans;
 		dev = alloc_etherdev(0);
diff --git a/drivers/s390/net/netiucv.c b/drivers/s390/net/netiucv.c
index 3251333..b6a6356 100644
--- a/drivers/s390/net/netiucv.c
+++ b/drivers/s390/net/netiucv.c
@@ -1994,6 +1994,8 @@
 			   netiucv_setup_netdevice);
 	if (!dev)
 		return NULL;
+	if (dev_alloc_name(dev, dev->name) < 0)
+		goto out_netdev;
 
 	privptr = netdev_priv(dev);
 	privptr->fsm = init_fsm("netiucvdev", dev_state_names,
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index b77c65e..4abc79d 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -236,8 +236,7 @@
 #define QETH_IN_BUF_COUNT_MAX 128
 #define QETH_MAX_BUFFER_ELEMENTS(card) ((card)->qdio.in_buf_size >> 12)
 #define QETH_IN_BUF_REQUEUE_THRESHOLD(card) \
-		((card)->ssqd.qdioac1 & AC1_SIGA_INPUT_NEEDED ? 1 : \
-		 ((card)->qdio.in_buf_pool.buf_count / 2))
+		 ((card)->qdio.in_buf_pool.buf_count / 2)
 
 /* buffers we have to be behind before we get a PCI */
 #define QETH_PCI_THRESHOLD_A(card) ((card)->qdio.in_buf_pool.buf_count+1)
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 8153443..fff57de 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -881,7 +881,6 @@
 void qeth_schedule_recovery(struct qeth_card *card)
 {
 	QETH_CARD_TEXT(card, 2, "startrec");
-	WARN_ON(1);
 	if (qeth_set_thread_start_bit(card, QETH_RECOVER_THREAD) == 0)
 		schedule_work(&card->kernel_thread_starter);
 }
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index e4c1176..4d5307d 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2756,11 +2756,13 @@
 	struct neighbour *n = NULL;
 	struct dst_entry *dst;
 
+	rcu_read_lock();
 	dst = skb_dst(skb);
 	if (dst)
 		n = dst_get_neighbour(dst);
 	if (n) {
 		cast_type = n->type;
+		rcu_read_unlock();
 		if ((cast_type == RTN_BROADCAST) ||
 		    (cast_type == RTN_MULTICAST) ||
 		    (cast_type == RTN_ANYCAST))
@@ -2768,6 +2770,8 @@
 		else
 			return RTN_UNSPEC;
 	}
+	rcu_read_unlock();
+
 	/* try something else */
 	if (skb->protocol == ETH_P_IPV6)
 		return (skb_network_header(skb)[24] == 0xff) ?
@@ -2847,6 +2851,8 @@
 	}
 
 	hdr->hdr.l3.length = skb->len - sizeof(struct qeth_hdr);
+
+	rcu_read_lock();
 	dst = skb_dst(skb);
 	if (dst)
 		n = dst_get_neighbour(dst);
@@ -2893,6 +2899,7 @@
 				QETH_CAST_UNICAST | QETH_HDR_PASSTHRU;
 		}
 	}
+	rcu_read_unlock();
 }
 
 static inline void qeth_l3_hdr_csum(struct qeth_card *card,
diff --git a/drivers/s390/net/qeth_l3_sys.c b/drivers/s390/net/qeth_l3_sys.c
index 0ea2fbf..d979bb2 100644
--- a/drivers/s390/net/qeth_l3_sys.c
+++ b/drivers/s390/net/qeth_l3_sys.c
@@ -335,10 +335,10 @@
 					QETH_IN_BUF_COUNT_MAX)
 				qeth_realloc_buffer_pool(card,
 					QETH_IN_BUF_COUNT_MAX);
-			break;
 		} else
 			rc = -EPERM;
-	default:   /* fall through */
+		break;
+	default:
 		rc = -EINVAL;
 	}
 out:
diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c
index 11f07f8..b79576b 100644
--- a/drivers/s390/scsi/zfcp_scsi.c
+++ b/drivers/s390/scsi/zfcp_scsi.c
@@ -55,6 +55,10 @@
 {
 	struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
 
+	/* if previous slave_alloc returned early, there is nothing to do */
+	if (!zfcp_sdev->port)
+		return;
+
 	zfcp_erp_lun_shutdown_wait(sdev, "scssd_1");
 	put_device(&zfcp_sdev->port->dev);
 }
diff --git a/drivers/sbus/char/bbc_i2c.c b/drivers/sbus/char/bbc_i2c.c
index 5f94d22..5426682 100644
--- a/drivers/sbus/char/bbc_i2c.c
+++ b/drivers/sbus/char/bbc_i2c.c
@@ -233,13 +233,9 @@
 	int ret = 0;
 
 	while (len > 0) {
-		int err = bbc_i2c_writeb(client, *buf, off);
-
-		if (err < 0) {
-			ret = err;
+		ret = bbc_i2c_writeb(client, *buf, off);
+		if (ret < 0)
 			break;
-		}
-
 		len--;
 		buf++;
 		off++;
@@ -253,11 +249,9 @@
 	int ret = 0;
 
 	while (len > 0) {
-		int err = bbc_i2c_readb(client, buf, off);
-		if (err < 0) {
-			ret = err;
+		ret = bbc_i2c_readb(client, buf, off);
+		if (ret < 0)
 			break;
-		}
 		len--;
 		buf++;
 		off++;
@@ -422,17 +416,6 @@
 	.remove		= __devexit_p(bbc_i2c_remove),
 };
 
-static int __init bbc_i2c_init(void)
-{
-	return platform_driver_register(&bbc_i2c_driver);
-}
-
-static void __exit bbc_i2c_exit(void)
-{
-	platform_driver_unregister(&bbc_i2c_driver);
-}
-
-module_init(bbc_i2c_init);
-module_exit(bbc_i2c_exit);
+module_platform_driver(bbc_i2c_driver);
 
 MODULE_LICENSE("GPL");
diff --git a/drivers/sbus/char/display7seg.c b/drivers/sbus/char/display7seg.c
index 965a1fc..4b99397 100644
--- a/drivers/sbus/char/display7seg.c
+++ b/drivers/sbus/char/display7seg.c
@@ -275,15 +275,4 @@
 	.remove		= __devexit_p(d7s_remove),
 };
 
-static int __init d7s_init(void)
-{
-	return platform_driver_register(&d7s_driver);
-}
-
-static void __exit d7s_exit(void)
-{
-	platform_driver_unregister(&d7s_driver);
-}
-
-module_init(d7s_init);
-module_exit(d7s_exit);
+module_platform_driver(d7s_driver);
diff --git a/drivers/sbus/char/envctrl.c b/drivers/sbus/char/envctrl.c
index be7b4e5..339fd6f 100644
--- a/drivers/sbus/char/envctrl.c
+++ b/drivers/sbus/char/envctrl.c
@@ -1138,16 +1138,6 @@
 	.remove		= __devexit_p(envctrl_remove),
 };
 
-static int __init envctrl_init(void)
-{
-	return platform_driver_register(&envctrl_driver);
-}
+module_platform_driver(envctrl_driver);
 
-static void __exit envctrl_exit(void)
-{
-	platform_driver_unregister(&envctrl_driver);
-}
-
-module_init(envctrl_init);
-module_exit(envctrl_exit);
 MODULE_LICENSE("GPL");
diff --git a/drivers/sbus/char/flash.c b/drivers/sbus/char/flash.c
index 73dd4e7..826157f 100644
--- a/drivers/sbus/char/flash.c
+++ b/drivers/sbus/char/flash.c
@@ -216,16 +216,6 @@
 	.remove		= __devexit_p(flash_remove),
 };
 
-static int __init flash_init(void)
-{
-	return platform_driver_register(&flash_driver);
-}
+module_platform_driver(flash_driver);
 
-static void __exit flash_cleanup(void)
-{
-	platform_driver_unregister(&flash_driver);
-}
-
-module_init(flash_init);
-module_exit(flash_cleanup);
 MODULE_LICENSE("GPL");
diff --git a/drivers/sbus/char/uctrl.c b/drivers/sbus/char/uctrl.c
index ebce963..0b31658 100644
--- a/drivers/sbus/char/uctrl.c
+++ b/drivers/sbus/char/uctrl.c
@@ -435,16 +435,6 @@
 };
 
 
-static int __init uctrl_init(void)
-{
-	return platform_driver_register(&uctrl_driver);
-}
+module_platform_driver(uctrl_driver);
 
-static void __exit uctrl_exit(void)
-{
-	platform_driver_unregister(&uctrl_driver);
-}
-
-module_init(uctrl_init);
-module_exit(uctrl_exit);
 MODULE_LICENSE("GPL");
diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c
index 4aa76d6..705e13e 100644
--- a/drivers/scsi/aacraid/linit.c
+++ b/drivers/scsi/aacraid/linit.c
@@ -38,6 +38,7 @@
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/pci.h>
+#include <linux/pci-aspm.h>
 #include <linux/slab.h>
 #include <linux/mutex.h>
 #include <linux/spinlock.h>
@@ -1109,6 +1110,9 @@
 		unique_id++;
 	}
 
+	pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1 |
+			       PCIE_LINK_STATE_CLKPM);
+
 	error = pci_enable_device(pdev);
 	if (error)
 		goto out;
diff --git a/drivers/scsi/bnx2i/bnx2i_hwi.c b/drivers/scsi/bnx2i/bnx2i_hwi.c
index dba72a4..1ad0b82 100644
--- a/drivers/scsi/bnx2i/bnx2i_hwi.c
+++ b/drivers/scsi/bnx2i/bnx2i_hwi.c
@@ -1906,18 +1906,19 @@
 	spin_lock(&session->lock);
 	task = iscsi_itt_to_task(bnx2i_conn->cls_conn->dd_data,
 				 cqe->itt & ISCSI_CMD_RESPONSE_INDEX);
-	if (!task) {
+	if (!task || !task->sc) {
 		spin_unlock(&session->lock);
 		return -EINVAL;
 	}
 	sc = task->sc;
-	spin_unlock(&session->lock);
 
 	if (!blk_rq_cpu_valid(sc->request))
 		cpu = smp_processor_id();
 	else
 		cpu = sc->request->cpu;
 
+	spin_unlock(&session->lock);
+
 	p = &per_cpu(bnx2i_percpu, cpu);
 	spin_lock(&p->p_work_lock);
 	if (unlikely(!p->iothread)) {
diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c
index cefbe44..8d67467 100644
--- a/drivers/scsi/fcoe/fcoe.c
+++ b/drivers/scsi/fcoe/fcoe.c
@@ -31,6 +31,8 @@
 #include <linux/sysfs.h>
 #include <linux/ctype.h>
 #include <linux/workqueue.h>
+#include <net/dcbnl.h>
+#include <net/dcbevent.h>
 #include <scsi/scsi_tcq.h>
 #include <scsi/scsicam.h>
 #include <scsi/scsi_transport.h>
@@ -101,6 +103,8 @@
 static int fcoe_ddp_target(struct fc_lport *, u16, struct scatterlist *,
 			   unsigned int);
 static int fcoe_cpu_callback(struct notifier_block *, unsigned long, void *);
+static int fcoe_dcb_app_notification(struct notifier_block *notifier,
+				     ulong event, void *ptr);
 
 static bool fcoe_match(struct net_device *netdev);
 static int fcoe_create(struct net_device *netdev, enum fip_state fip_mode);
@@ -129,6 +133,11 @@
 	.notifier_call = fcoe_cpu_callback,
 };
 
+/* notification function for DCB events */
+static struct notifier_block dcb_notifier = {
+	.notifier_call = fcoe_dcb_app_notification,
+};
+
 static struct scsi_transport_template *fcoe_nport_scsi_transport;
 static struct scsi_transport_template *fcoe_vport_scsi_transport;
 
@@ -1522,6 +1531,8 @@
 	skb_reset_network_header(skb);
 	skb->mac_len = elen;
 	skb->protocol = htons(ETH_P_FCOE);
+	skb->priority = port->priority;
+
 	if (fcoe->netdev->priv_flags & IFF_802_1Q_VLAN &&
 	    fcoe->realdev->features & NETIF_F_HW_VLAN_TX) {
 		skb->vlan_tci = VLAN_TAG_PRESENT |
@@ -1624,6 +1635,7 @@
 	stats->InvalidCRCCount++;
 	if (stats->InvalidCRCCount < 5)
 		printk(KERN_WARNING "fcoe: dropping frame with CRC error\n");
+	put_cpu();
 	return -EINVAL;
 }
 
@@ -1746,6 +1758,7 @@
  */
 static void fcoe_dev_setup(void)
 {
+	register_dcbevent_notifier(&dcb_notifier);
 	register_netdevice_notifier(&fcoe_notifier);
 }
 
@@ -1754,9 +1767,69 @@
  */
 static void fcoe_dev_cleanup(void)
 {
+	unregister_dcbevent_notifier(&dcb_notifier);
 	unregister_netdevice_notifier(&fcoe_notifier);
 }
 
+static struct fcoe_interface *
+fcoe_hostlist_lookup_realdev_port(struct net_device *netdev)
+{
+	struct fcoe_interface *fcoe;
+	struct net_device *real_dev;
+
+	list_for_each_entry(fcoe, &fcoe_hostlist, list) {
+		if (fcoe->netdev->priv_flags & IFF_802_1Q_VLAN)
+			real_dev = vlan_dev_real_dev(fcoe->netdev);
+		else
+			real_dev = fcoe->netdev;
+
+		if (netdev == real_dev)
+			return fcoe;
+	}
+	return NULL;
+}
+
+static int fcoe_dcb_app_notification(struct notifier_block *notifier,
+				     ulong event, void *ptr)
+{
+	struct dcb_app_type *entry = ptr;
+	struct fcoe_interface *fcoe;
+	struct net_device *netdev;
+	struct fcoe_port *port;
+	int prio;
+
+	if (entry->app.selector != DCB_APP_IDTYPE_ETHTYPE)
+		return NOTIFY_OK;
+
+	netdev = dev_get_by_index(&init_net, entry->ifindex);
+	if (!netdev)
+		return NOTIFY_OK;
+
+	fcoe = fcoe_hostlist_lookup_realdev_port(netdev);
+	dev_put(netdev);
+	if (!fcoe)
+		return NOTIFY_OK;
+
+	if (entry->dcbx & DCB_CAP_DCBX_VER_CEE)
+		prio = ffs(entry->app.priority) - 1;
+	else
+		prio = entry->app.priority;
+
+	if (prio < 0)
+		return NOTIFY_OK;
+
+	if (entry->app.protocol == ETH_P_FIP ||
+	    entry->app.protocol == ETH_P_FCOE)
+		fcoe->ctlr.priority = prio;
+
+	if (entry->app.protocol == ETH_P_FCOE) {
+		port = lport_priv(fcoe->ctlr.lp);
+		port->priority = prio;
+	}
+
+	return NOTIFY_OK;
+}
+
 /**
  * fcoe_device_notification() - Handler for net device events
  * @notifier: The context of the notification
@@ -1965,6 +2038,46 @@
 }
 
 /**
+ * fcoe_dcb_create() - Initialize DCB attributes and hooks
+ * @netdev: The net_device object of the L2 link that should be queried
+ * @port: The fcoe_port to bind FCoE APP priority with
+ * @
+ */
+static void fcoe_dcb_create(struct fcoe_interface *fcoe)
+{
+#ifdef CONFIG_DCB
+	int dcbx;
+	u8 fup, up;
+	struct net_device *netdev = fcoe->realdev;
+	struct fcoe_port *port = lport_priv(fcoe->ctlr.lp);
+	struct dcb_app app = {
+				.priority = 0,
+				.protocol = ETH_P_FCOE
+			     };
+
+	/* setup DCB priority attributes. */
+	if (netdev && netdev->dcbnl_ops && netdev->dcbnl_ops->getdcbx) {
+		dcbx = netdev->dcbnl_ops->getdcbx(netdev);
+
+		if (dcbx & DCB_CAP_DCBX_VER_IEEE) {
+			app.selector = IEEE_8021QAZ_APP_SEL_ETHERTYPE;
+			up = dcb_ieee_getapp_mask(netdev, &app);
+			app.protocol = ETH_P_FIP;
+			fup = dcb_ieee_getapp_mask(netdev, &app);
+		} else {
+			app.selector = DCB_APP_IDTYPE_ETHTYPE;
+			up = dcb_getapp(netdev, &app);
+			app.protocol = ETH_P_FIP;
+			fup = dcb_getapp(netdev, &app);
+		}
+
+		port->priority = ffs(up) ? ffs(up) - 1 : 0;
+		fcoe->ctlr.priority = ffs(fup) ? ffs(fup) - 1 : port->priority;
+	}
+#endif
+}
+
+/**
  * fcoe_create() - Create a fcoe interface
  * @netdev  : The net_device object the Ethernet interface to create on
  * @fip_mode: The FIP mode for this creation
@@ -2007,6 +2120,9 @@
 	/* Make this the "master" N_Port */
 	fcoe->ctlr.lp = lport;
 
+	/* setup DCB priority attributes. */
+	fcoe_dcb_create(fcoe);
+
 	/* add to lports list */
 	fcoe_hostlist_add(lport);
 
diff --git a/drivers/scsi/fcoe/fcoe_ctlr.c b/drivers/scsi/fcoe/fcoe_ctlr.c
index c74c4b8..e7522dc 100644
--- a/drivers/scsi/fcoe/fcoe_ctlr.c
+++ b/drivers/scsi/fcoe/fcoe_ctlr.c
@@ -320,6 +320,7 @@
 
 	skb_put(skb, sizeof(*sol));
 	skb->protocol = htons(ETH_P_FIP);
+	skb->priority = fip->priority;
 	skb_reset_mac_header(skb);
 	skb_reset_network_header(skb);
 	fip->send(fip, skb);
@@ -474,6 +475,7 @@
 	}
 	skb_put(skb, len);
 	skb->protocol = htons(ETH_P_FIP);
+	skb->priority = fip->priority;
 	skb_reset_mac_header(skb);
 	skb_reset_network_header(skb);
 	fip->send(fip, skb);
@@ -566,6 +568,7 @@
 	cap->fip.fip_dl_len = htons(dlen / FIP_BPW);
 
 	skb->protocol = htons(ETH_P_FIP);
+	skb->priority = fip->priority;
 	skb_reset_mac_header(skb);
 	skb_reset_network_header(skb);
 	return 0;
@@ -1911,6 +1914,7 @@
 
 	skb_put(skb, len);
 	skb->protocol = htons(ETH_P_FIP);
+	skb->priority = fip->priority;
 	skb_reset_mac_header(skb);
 	skb_reset_network_header(skb);
 
diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index e76107b..865d452 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -23,6 +23,7 @@
 #include <linux/interrupt.h>
 #include <linux/types.h>
 #include <linux/pci.h>
+#include <linux/pci-aspm.h>
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/delay.h>
@@ -3922,6 +3923,10 @@
 		dev_warn(&h->pdev->dev, "controller appears to be disabled\n");
 		return -ENODEV;
 	}
+
+	pci_disable_link_state(h->pdev, PCIE_LINK_STATE_L0S |
+			       PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_CLKPM);
+
 	err = pci_enable_device(h->pdev);
 	if (err) {
 		dev_warn(&h->pdev->dev, "unable to enable PCI device\n");
diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c
index 8889b1b..d570573 100644
--- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c
+++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c
@@ -2802,6 +2802,11 @@
 
 	if (ioc->is_driver_loading)
 		return;
+
+	fw_event = kzalloc(sizeof(struct fw_event_work), GFP_ATOMIC);
+	if (!fw_event)
+		return;
+
 	fw_event->event = MPT2SAS_REMOVE_UNRESPONDING_DEVICES;
 	fw_event->ioc = ioc;
 	_scsih_fw_event_add(ioc, fw_event);
@@ -4330,7 +4335,7 @@
 	/* insert into event log */
 	sz = offsetof(Mpi2EventNotificationReply_t, EventData) +
 	     sizeof(Mpi2EventDataSasDeviceStatusChange_t);
-	event_reply = kzalloc(sz, GFP_KERNEL);
+	event_reply = kzalloc(sz, GFP_ATOMIC);
 	if (!event_reply) {
 		printk(MPT2SAS_ERR_FMT "failure at %s:%d/%s()!\n",
 		    ioc->name, __FILE__, __LINE__, __func__);
diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index ac326c4..6465dae 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -1762,12 +1762,31 @@
 	scsi_qla_host_t *vha = shost_priv(shost);
 	struct scsi_qla_host *base_vha = pci_get_drvdata(vha->hw->pdev);
 
-	if (!base_vha->flags.online)
+	if (!base_vha->flags.online) {
 		fc_host_port_state(shost) = FC_PORTSTATE_OFFLINE;
-	else if (atomic_read(&base_vha->loop_state) == LOOP_TIMEOUT)
-		fc_host_port_state(shost) = FC_PORTSTATE_UNKNOWN;
-	else
+		return;
+	}
+
+	switch (atomic_read(&base_vha->loop_state)) {
+	case LOOP_UPDATE:
+		fc_host_port_state(shost) = FC_PORTSTATE_DIAGNOSTICS;
+		break;
+	case LOOP_DOWN:
+		if (test_bit(LOOP_RESYNC_NEEDED, &base_vha->dpc_flags))
+			fc_host_port_state(shost) = FC_PORTSTATE_DIAGNOSTICS;
+		else
+			fc_host_port_state(shost) = FC_PORTSTATE_LINKDOWN;
+		break;
+	case LOOP_DEAD:
+		fc_host_port_state(shost) = FC_PORTSTATE_LINKDOWN;
+		break;
+	case LOOP_READY:
 		fc_host_port_state(shost) = FC_PORTSTATE_ONLINE;
+		break;
+	default:
+		fc_host_port_state(shost) = FC_PORTSTATE_UNKNOWN;
+		break;
+	}
 }
 
 static int
diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c
index 9df4787..f3cddd5 100644
--- a/drivers/scsi/qla2xxx/qla_dbg.c
+++ b/drivers/scsi/qla2xxx/qla_dbg.c
@@ -12,17 +12,17 @@
  * |             Level            |   Last Value Used  |     Holes	|
  * ----------------------------------------------------------------------
  * | Module Init and Probe        |       0x0116       |  		|
- * | Mailbox commands             |       0x1129       |		|
+ * | Mailbox commands             |       0x112b       |		|
  * | Device Discovery             |       0x2083       |		|
  * | Queue Command and IO tracing |       0x302e       |     0x3008     |
  * | DPC Thread                   |       0x401c       |		|
  * | Async Events                 |       0x5059       |		|
- * | Timer Routines               |       0x600d       |		|
+ * | Timer Routines               |       0x6010       | 0x600e,0x600f  |
  * | User Space Interactions      |       0x709d       |		|
- * | Task Management              |       0x8041       |    		|
+ * | Task Management              |       0x8041       | 0x800b         |
  * | AER/EEH                      |       0x900f       |		|
  * | Virtual Port                 |       0xa007       |		|
- * | ISP82XX Specific             |       0xb051       |    		|
+ * | ISP82XX Specific             |       0xb052       |    		|
  * | MultiQ                       |       0xc00b       |		|
  * | Misc                         |       0xd00b       |		|
  * ----------------------------------------------------------------------
diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h
index ce32d81..c0c11af 100644
--- a/drivers/scsi/qla2xxx/qla_gbl.h
+++ b/drivers/scsi/qla2xxx/qla_gbl.h
@@ -578,6 +578,7 @@
 extern void qla82xx_chip_reset_cleanup(scsi_qla_host_t *);
 extern int qla82xx_mbx_beacon_ctl(scsi_qla_host_t *, int);
 extern char *qdev_state(uint32_t);
+extern void qla82xx_clear_pending_mbx(scsi_qla_host_t *);
 
 /* BSG related functions */
 extern int qla24xx_bsg_request(struct fc_bsg_job *);
diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index f03e915f..54ea68c 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -1509,7 +1509,8 @@
 				    &ha->fw_xcb_count, NULL, NULL,
 				    &ha->max_npiv_vports, NULL);
 
-				if (!fw_major_version && ql2xallocfwdump)
+				if (!fw_major_version && ql2xallocfwdump
+				    && !IS_QLA82XX(ha))
 					qla2x00_alloc_fw_dump(vha);
 			}
 		} else {
diff --git a/drivers/scsi/qla2xxx/qla_iocb.c b/drivers/scsi/qla2xxx/qla_iocb.c
index dbec896..a4b267e 100644
--- a/drivers/scsi/qla2xxx/qla_iocb.c
+++ b/drivers/scsi/qla2xxx/qla_iocb.c
@@ -120,11 +120,10 @@
  * Returns a pointer to the continuation type 1 IOCB packet.
  */
 static inline cont_a64_entry_t *
-qla2x00_prep_cont_type1_iocb(scsi_qla_host_t *vha)
+qla2x00_prep_cont_type1_iocb(scsi_qla_host_t *vha, struct req_que *req)
 {
 	cont_a64_entry_t *cont_pkt;
 
-	struct req_que *req = vha->req;
 	/* Adjust ring index. */
 	req->ring_index++;
 	if (req->ring_index == req->length) {
@@ -292,7 +291,7 @@
 			 * Five DSDs are available in the Continuation
 			 * Type 1 IOCB.
 			 */
-			cont_pkt = qla2x00_prep_cont_type1_iocb(vha);
+			cont_pkt = qla2x00_prep_cont_type1_iocb(vha, vha->req);
 			cur_dsd = (uint32_t *)cont_pkt->dseg_0_address;
 			avail_dsds = 5;
 		}
@@ -684,7 +683,7 @@
 			 * Five DSDs are available in the Continuation
 			 * Type 1 IOCB.
 			 */
-			cont_pkt = qla2x00_prep_cont_type1_iocb(vha);
+			cont_pkt = qla2x00_prep_cont_type1_iocb(vha, vha->req);
 			cur_dsd = (uint32_t *)cont_pkt->dseg_0_address;
 			avail_dsds = 5;
 		}
@@ -2070,7 +2069,8 @@
 			* Five DSDs are available in the Cont.
 			* Type 1 IOCB.
 			       */
-			cont_pkt = qla2x00_prep_cont_type1_iocb(vha);
+			cont_pkt = qla2x00_prep_cont_type1_iocb(vha,
+			    vha->hw->req_q_map[0]);
 			cur_dsd = (uint32_t *) cont_pkt->dseg_0_address;
 			avail_dsds = 5;
 			cont_iocb_prsnt = 1;
@@ -2096,6 +2096,7 @@
 	int index;
 	uint16_t tot_dsds;
         scsi_qla_host_t *vha = sp->fcport->vha;
+	struct qla_hw_data *ha = vha->hw;
 	struct fc_bsg_job *bsg_job = ((struct srb_ctx *)sp->ctx)->u.bsg_job;
 	int loop_iterartion = 0;
 	int cont_iocb_prsnt = 0;
@@ -2141,7 +2142,8 @@
 			* Five DSDs are available in the Cont.
 			* Type 1 IOCB.
 			       */
-			cont_pkt = qla2x00_prep_cont_type1_iocb(vha);
+			cont_pkt = qla2x00_prep_cont_type1_iocb(vha,
+			    ha->req_q_map[0]);
 			cur_dsd = (uint32_t *) cont_pkt->dseg_0_address;
 			avail_dsds = 5;
 			cont_iocb_prsnt = 1;
diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
index 2516adf..7b91b29 100644
--- a/drivers/scsi/qla2xxx/qla_isr.c
+++ b/drivers/scsi/qla2xxx/qla_isr.c
@@ -1741,7 +1741,7 @@
 				    resid, scsi_bufflen(cp));
 
 				cp->result = DID_ERROR << 16 | lscsi_status;
-				break;
+				goto check_scsi_status;
 			}
 
 			if (!lscsi_status &&
diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
index 3b3cec9..82a3353 100644
--- a/drivers/scsi/qla2xxx/qla_mbx.c
+++ b/drivers/scsi/qla2xxx/qla_mbx.c
@@ -79,8 +79,7 @@
 		mcp->mb[0] = MBS_LINK_DOWN_ERROR;
 		ql_log(ql_log_warn, base_vha, 0x1004,
 		    "FW hung = %d.\n", ha->flags.isp82xx_fw_hung);
-		rval = QLA_FUNCTION_FAILED;
-		goto premature_exit;
+		return QLA_FUNCTION_TIMEOUT;
 	}
 
 	/*
@@ -163,6 +162,7 @@
 				HINT_MBX_INT_PENDING) {
 				spin_unlock_irqrestore(&ha->hardware_lock,
 					flags);
+				ha->flags.mbox_busy = 0;
 				ql_dbg(ql_dbg_mbx, base_vha, 0x1010,
 				    "Pending mailbox timeout, exiting.\n");
 				rval = QLA_FUNCTION_TIMEOUT;
@@ -188,6 +188,7 @@
 				HINT_MBX_INT_PENDING) {
 				spin_unlock_irqrestore(&ha->hardware_lock,
 					flags);
+				ha->flags.mbox_busy = 0;
 				ql_dbg(ql_dbg_mbx, base_vha, 0x1012,
 				    "Pending mailbox timeout, exiting.\n");
 				rval = QLA_FUNCTION_TIMEOUT;
@@ -302,7 +303,15 @@
 			if (!test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags) &&
 			    !test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags) &&
 			    !test_bit(ISP_ABORT_RETRY, &vha->dpc_flags)) {
-
+				if (IS_QLA82XX(ha)) {
+					ql_dbg(ql_dbg_mbx, vha, 0x112a,
+					    "disabling pause transmit on port "
+					    "0 & 1.\n");
+					qla82xx_wr_32(ha,
+					    QLA82XX_CRB_NIU + 0x98,
+					    CRB_NIU_XG_PAUSE_CTL_P0|
+					    CRB_NIU_XG_PAUSE_CTL_P1);
+				}
 				ql_log(ql_log_info, base_vha, 0x101c,
 				    "Mailbox cmd timeout occured. "
 				    "Scheduling ISP abort eeh_busy=0x%x.\n",
@@ -318,7 +327,15 @@
 			if (!test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags) &&
 			    !test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags) &&
 			    !test_bit(ISP_ABORT_RETRY, &vha->dpc_flags)) {
-
+				if (IS_QLA82XX(ha)) {
+					ql_dbg(ql_dbg_mbx, vha, 0x112b,
+					    "disabling pause transmit on port "
+					    "0 & 1.\n");
+					qla82xx_wr_32(ha,
+					    QLA82XX_CRB_NIU + 0x98,
+					    CRB_NIU_XG_PAUSE_CTL_P0|
+					    CRB_NIU_XG_PAUSE_CTL_P1);
+				}
 				ql_log(ql_log_info, base_vha, 0x101e,
 				    "Mailbox cmd timeout occured. "
 				    "Scheduling ISP abort.\n");
diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c
index 94bded5..0355493 100644
--- a/drivers/scsi/qla2xxx/qla_nx.c
+++ b/drivers/scsi/qla2xxx/qla_nx.c
@@ -3817,6 +3817,20 @@
 	return rval;
 }
 
+void qla82xx_clear_pending_mbx(scsi_qla_host_t *vha)
+{
+	struct qla_hw_data *ha = vha->hw;
+
+	if (ha->flags.mbox_busy) {
+		ha->flags.mbox_int = 1;
+		ha->flags.mbox_busy = 0;
+		ql_log(ql_log_warn, vha, 0x6010,
+		    "Doing premature completion of mbx command.\n");
+		if (test_bit(MBX_INTR_WAIT, &ha->mbx_cmd_flags))
+			complete(&ha->mbx_intr_comp);
+	}
+}
+
 void qla82xx_watchdog(scsi_qla_host_t *vha)
 {
 	uint32_t dev_state, halt_status;
@@ -3839,9 +3853,13 @@
 			qla2xxx_wake_dpc(vha);
 		} else {
 			if (qla82xx_check_fw_alive(vha)) {
+				ql_dbg(ql_dbg_timer, vha, 0x6011,
+				    "disabling pause transmit on port 0 & 1.\n");
+				qla82xx_wr_32(ha, QLA82XX_CRB_NIU + 0x98,
+				    CRB_NIU_XG_PAUSE_CTL_P0|CRB_NIU_XG_PAUSE_CTL_P1);
 				halt_status = qla82xx_rd_32(ha,
 				    QLA82XX_PEG_HALT_STATUS1);
-				ql_dbg(ql_dbg_timer, vha, 0x6005,
+				ql_log(ql_log_info, vha, 0x6005,
 				    "dumping hw/fw registers:.\n "
 				    " PEG_HALT_STATUS1: 0x%x, PEG_HALT_STATUS2: 0x%x,.\n "
 				    " PEG_NET_0_PC: 0x%x, PEG_NET_1_PC: 0x%x,.\n "
@@ -3858,6 +3876,11 @@
 					    QLA82XX_CRB_PEG_NET_3 + 0x3c),
 				    qla82xx_rd_32(ha,
 					    QLA82XX_CRB_PEG_NET_4 + 0x3c));
+				if (LSW(MSB(halt_status)) == 0x67)
+					ql_log(ql_log_warn, vha, 0xb052,
+					    "Firmware aborted with "
+					    "error code 0x00006700. Device is "
+					    "being reset.\n");
 				if (halt_status & HALT_STATUS_UNRECOVERABLE) {
 					set_bit(ISP_UNRECOVERABLE,
 					    &vha->dpc_flags);
@@ -3869,16 +3892,8 @@
 				}
 				qla2xxx_wake_dpc(vha);
 				ha->flags.isp82xx_fw_hung = 1;
-				if (ha->flags.mbox_busy) {
-					ha->flags.mbox_int = 1;
-					ql_log(ql_log_warn, vha, 0x6007,
-					    "Due to FW hung, doing "
-					    "premature completion of mbx "
-					    "command.\n");
-					if (test_bit(MBX_INTR_WAIT,
-					    &ha->mbx_cmd_flags))
-						complete(&ha->mbx_intr_comp);
-				}
+				ql_log(ql_log_warn, vha, 0x6007, "Firmware hung.\n");
+				qla82xx_clear_pending_mbx(vha);
 			}
 		}
 	}
@@ -4073,10 +4088,7 @@
 			msleep(1000);
 			if (qla82xx_check_fw_alive(vha)) {
 				ha->flags.isp82xx_fw_hung = 1;
-				if (ha->flags.mbox_busy) {
-					ha->flags.mbox_int = 1;
-					complete(&ha->mbx_intr_comp);
-				}
+				qla82xx_clear_pending_mbx(vha);
 				break;
 			}
 		}
diff --git a/drivers/scsi/qla2xxx/qla_nx.h b/drivers/scsi/qla2xxx/qla_nx.h
index 57820c1..57a226b 100644
--- a/drivers/scsi/qla2xxx/qla_nx.h
+++ b/drivers/scsi/qla2xxx/qla_nx.h
@@ -1173,4 +1173,8 @@
 
 static const int MD_MIU_TEST_AGT_RDDATA[] = { 0x410000A8, 0x410000AC,
 	0x410000B8, 0x410000BC };
+
+#define CRB_NIU_XG_PAUSE_CTL_P0        0x1
+#define CRB_NIU_XG_PAUSE_CTL_P1        0x8
+
 #endif
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index fd14c7b..f9e5b85 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -201,12 +201,12 @@
 		"Set the Minidump driver capture mask level. "
 		"Default is 0x7F - Can be set to 0x3, 0x7, 0xF, 0x1F, 0x7F.");
 
-int ql2xmdenable;
+int ql2xmdenable = 1;
 module_param(ql2xmdenable, int, S_IRUGO);
 MODULE_PARM_DESC(ql2xmdenable,
 		"Enable/disable MiniDump. "
-		"0 (Default) - MiniDump disabled. "
-		"1 - MiniDump enabled.");
+		"0 - MiniDump disabled. "
+		"1 (Default) - MiniDump enabled.");
 
 /*
  * SCSI host template entry points
@@ -423,6 +423,7 @@
 	qla25xx_delete_queues(vha);
 	destroy_workqueue(ha->wq);
 	ha->wq = NULL;
+	vha->req = ha->req_q_map[0];
 fail:
 	ha->mqenable = 0;
 	kfree(ha->req_q_map);
@@ -814,49 +815,6 @@
 	return return_status;
 }
 
-/*
- * qla2x00_wait_for_loop_ready
- *    Wait for MAX_LOOP_TIMEOUT(5 min) value for loop
- *    to be in LOOP_READY state.
- * Input:
- *     ha - pointer to host adapter structure
- *
- * Note:
- *    Does context switching-Release SPIN_LOCK
- *    (if any) before calling this routine.
- *
- *
- * Return:
- *    Success (LOOP_READY) : 0
- *    Failed  (LOOP_NOT_READY) : 1
- */
-static inline int
-qla2x00_wait_for_loop_ready(scsi_qla_host_t *vha)
-{
-	int 	 return_status = QLA_SUCCESS;
-	unsigned long loop_timeout ;
-	struct qla_hw_data *ha = vha->hw;
-	scsi_qla_host_t *base_vha = pci_get_drvdata(ha->pdev);
-
-	/* wait for 5 min at the max for loop to be ready */
-	loop_timeout = jiffies + (MAX_LOOP_TIMEOUT * HZ);
-
-	while ((!atomic_read(&base_vha->loop_down_timer) &&
-	    atomic_read(&base_vha->loop_state) == LOOP_DOWN) ||
-	    atomic_read(&base_vha->loop_state) != LOOP_READY) {
-		if (atomic_read(&base_vha->loop_state) == LOOP_DEAD) {
-			return_status = QLA_FUNCTION_FAILED;
-			break;
-		}
-		msleep(1000);
-		if (time_after_eq(jiffies, loop_timeout)) {
-			return_status = QLA_FUNCTION_FAILED;
-			break;
-		}
-	}
-	return (return_status);
-}
-
 static void
 sp_get(struct srb *sp)
 {
@@ -1035,12 +993,6 @@
 		    "Wait for hba online failed for cmd=%p.\n", cmd);
 		goto eh_reset_failed;
 	}
-	err = 1;
-	if (qla2x00_wait_for_loop_ready(vha) != QLA_SUCCESS) {
-		ql_log(ql_log_warn, vha, 0x800b,
-		    "Wait for loop ready failed for cmd=%p.\n", cmd);
-		goto eh_reset_failed;
-	}
 	err = 2;
 	if (do_reset(fcport, cmd->device->lun, cmd->request->cpu + 1)
 		!= QLA_SUCCESS) {
@@ -1137,10 +1089,9 @@
 		goto eh_bus_reset_done;
 	}
 
-	if (qla2x00_wait_for_loop_ready(vha) == QLA_SUCCESS) {
-		if (qla2x00_loop_reset(vha) == QLA_SUCCESS)
-			ret = SUCCESS;
-	}
+	if (qla2x00_loop_reset(vha) == QLA_SUCCESS)
+		ret = SUCCESS;
+
 	if (ret == FAILED)
 		goto eh_bus_reset_done;
 
@@ -1206,15 +1157,6 @@
 	if (qla2x00_wait_for_reset_ready(vha) != QLA_SUCCESS)
 		goto eh_host_reset_lock;
 
-	/*
-	 * Fixme-may be dpc thread is active and processing
-	 * loop_resync,so wait a while for it to
-	 * be completed and then issue big hammer.Otherwise
-	 * it may cause I/O failure as big hammer marks the
-	 * devices as lost kicking of the port_down_timer
-	 * while dpc is stuck for the mailbox to complete.
-	 */
-	qla2x00_wait_for_loop_ready(vha);
 	if (vha != base_vha) {
 		if (qla2x00_vp_abort_isp(vha))
 			goto eh_host_reset_lock;
@@ -1297,16 +1239,13 @@
 		atomic_set(&vha->loop_state, LOOP_DOWN);
 		atomic_set(&vha->loop_down_timer, LOOP_DOWN_TIME);
 		qla2x00_mark_all_devices_lost(vha, 0);
-		qla2x00_wait_for_loop_ready(vha);
 	}
 
 	if (ha->flags.enable_lip_reset) {
 		ret = qla2x00_lip_reset(vha);
-		if (ret != QLA_SUCCESS) {
+		if (ret != QLA_SUCCESS)
 			ql_dbg(ql_dbg_taskm, vha, 0x802e,
 			    "lip_reset failed (%d).\n", ret);
-		} else
-			qla2x00_wait_for_loop_ready(vha);
 	}
 
 	/* Issue marker command only when we are going to start the I/O */
@@ -4070,13 +4009,8 @@
 		/* For ISP82XX complete any pending mailbox cmd */
 		if (IS_QLA82XX(ha)) {
 			ha->flags.isp82xx_fw_hung = 1;
-			if (ha->flags.mbox_busy) {
-				ha->flags.mbox_int = 1;
-				ql_dbg(ql_dbg_aer, vha, 0x9001,
-				    "Due to pci channel io frozen, doing premature "
-				    "completion of mbx command.\n");
-				complete(&ha->mbx_intr_comp);
-			}
+			ql_dbg(ql_dbg_aer, vha, 0x9001, "Pci channel io frozen\n");
+			qla82xx_clear_pending_mbx(vha);
 		}
 		qla2x00_free_irqs(vha);
 		pci_disable_device(pdev);
diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h
index 13b6357..23f33a6 100644
--- a/drivers/scsi/qla2xxx/qla_version.h
+++ b/drivers/scsi/qla2xxx/qla_version.h
@@ -7,7 +7,7 @@
 /*
  * Driver version
  */
-#define QLA2XXX_VERSION      "8.03.07.07-k"
+#define QLA2XXX_VERSION      "8.03.07.12-k"
 
 #define QLA_DRIVER_MAJOR_VER	8
 #define QLA_DRIVER_MINOR_VER	3
diff --git a/drivers/scsi/qla4xxx/ql4_def.h b/drivers/scsi/qla4xxx/ql4_def.h
index ace637b..fd5edc6 100644
--- a/drivers/scsi/qla4xxx/ql4_def.h
+++ b/drivers/scsi/qla4xxx/ql4_def.h
@@ -147,7 +147,7 @@
 #define ISCSI_ALIAS_SIZE		32	/* ISCSI Alias name size */
 #define ISCSI_NAME_SIZE			0xE0	/* ISCSI Name size */
 
-#define QL4_SESS_RECOVERY_TMO		30	/* iSCSI session */
+#define QL4_SESS_RECOVERY_TMO		120	/* iSCSI session */
 						/* recovery timeout */
 
 #define LSDW(x) ((u32)((u64)(x)))
@@ -173,6 +173,8 @@
 #define ISNS_DEREG_TOV			5
 #define HBA_ONLINE_TOV			30
 #define DISABLE_ACB_TOV			30
+#define IP_CONFIG_TOV			30
+#define LOGIN_TOV			12
 
 #define MAX_RESET_HA_RETRIES		2
 
@@ -240,6 +242,45 @@
 
 	uint16_t fw_ddb_index;	/* DDB firmware index */
 	uint32_t fw_ddb_device_state; /* F/W Device State  -- see ql4_fw.h */
+	uint16_t ddb_type;
+#define FLASH_DDB 0x01
+
+	struct dev_db_entry fw_ddb_entry;
+	int (*unblock_sess)(struct iscsi_cls_session *cls_session);
+	int (*ddb_change)(struct scsi_qla_host *ha, uint32_t fw_ddb_index,
+			  struct ddb_entry *ddb_entry, uint32_t state);
+
+	/* Driver Re-login  */
+	unsigned long flags;		  /* DDB Flags */
+	uint16_t default_relogin_timeout; /*  Max time to wait for
+					   *  relogin to complete */
+	atomic_t retry_relogin_timer;	  /* Min Time between relogins
+					   * (4000 only) */
+	atomic_t relogin_timer;		  /* Max Time to wait for
+					   * relogin to complete */
+	atomic_t relogin_retry_count;	  /* Num of times relogin has been
+					   * retried */
+	uint32_t default_time2wait;	  /* Default Min time between
+					   * relogins (+aens) */
+
+};
+
+struct qla_ddb_index {
+	struct list_head list;
+	uint16_t fw_ddb_idx;
+	struct dev_db_entry fw_ddb;
+};
+
+#define DDB_IPADDR_LEN 64
+
+struct ql4_tuple_ddb {
+	int port;
+	int tpgt;
+	char ip_addr[DDB_IPADDR_LEN];
+	char iscsi_name[ISCSI_NAME_SIZE];
+	uint16_t options;
+#define DDB_OPT_IPV6 0x0e0e
+#define DDB_OPT_IPV4 0x0f0f
 };
 
 /*
@@ -411,7 +452,7 @@
 #define AF_FW_RECOVERY			19 /* 0x00080000 */
 #define AF_EEH_BUSY			20 /* 0x00100000 */
 #define AF_PCI_CHANNEL_IO_PERM_FAILURE	21 /* 0x00200000 */
-
+#define AF_BUILD_DDB_LIST		22 /* 0x00400000 */
 	unsigned long dpc_flags;
 
 #define DPC_RESET_HA			1 /* 0x00000002 */
@@ -604,6 +645,7 @@
 	uint16_t bootload_minor;
 	uint16_t bootload_patch;
 	uint16_t bootload_build;
+	uint16_t def_timeout; /* Default login timeout */
 
 	uint32_t flash_state;
 #define	QLFLASH_WAITING		0
@@ -623,6 +665,11 @@
 	uint16_t iscsi_pci_func_cnt;
 	uint8_t model_name[16];
 	struct completion disable_acb_comp;
+	struct dma_pool *fw_ddb_dma_pool;
+#define DDB_DMA_BLOCK_SIZE 512
+	uint16_t pri_ddb_idx;
+	uint16_t sec_ddb_idx;
+	int is_reset;
 };
 
 struct ql4_task_data {
@@ -835,6 +882,10 @@
 /*---------------------------------------------------------------------------*/
 
 /* Defines for qla4xxx_initialize_adapter() and qla4xxx_recover_adapter() */
+
+#define INIT_ADAPTER    0
+#define RESET_ADAPTER   1
+
 #define PRESERVE_DDB_LIST	0
 #define REBUILD_DDB_LIST	1
 
diff --git a/drivers/scsi/qla4xxx/ql4_fw.h b/drivers/scsi/qla4xxx/ql4_fw.h
index cbd5a20..4ac07f8 100644
--- a/drivers/scsi/qla4xxx/ql4_fw.h
+++ b/drivers/scsi/qla4xxx/ql4_fw.h
@@ -12,6 +12,7 @@
 #define MAX_PRST_DEV_DB_ENTRIES		64
 #define MIN_DISC_DEV_DB_ENTRY		MAX_PRST_DEV_DB_ENTRIES
 #define MAX_DEV_DB_ENTRIES		512
+#define MAX_DEV_DB_ENTRIES_40XX		256
 
 /*************************************************************************
  *
@@ -604,6 +605,13 @@
 	uint8_t res14[140];	/* 274-2FF */
 };
 
+#define IP_ADDR_COUNT	4 /* Total 4 IP address supported in one interface
+			   * One IPv4, one IPv6 link local and 2 IPv6
+			   */
+
+#define IP_STATE_MASK	0x0F000000
+#define IP_STATE_SHIFT	24
+
 struct init_fw_ctrl_blk {
 	struct addr_ctrl_blk pri;
 /*	struct addr_ctrl_blk sec;*/
diff --git a/drivers/scsi/qla4xxx/ql4_glbl.h b/drivers/scsi/qla4xxx/ql4_glbl.h
index 160db9d..d0dd4b3 100644
--- a/drivers/scsi/qla4xxx/ql4_glbl.h
+++ b/drivers/scsi/qla4xxx/ql4_glbl.h
@@ -13,7 +13,7 @@
 int qla4xxx_hw_reset(struct scsi_qla_host *ha);
 int ql4xxx_lock_drvr_wait(struct scsi_qla_host *a);
 int qla4xxx_send_command_to_isp(struct scsi_qla_host *ha, struct srb *srb);
-int qla4xxx_initialize_adapter(struct scsi_qla_host *ha);
+int qla4xxx_initialize_adapter(struct scsi_qla_host *ha, int is_reset);
 int qla4xxx_soft_reset(struct scsi_qla_host *ha);
 irqreturn_t qla4xxx_intr_handler(int irq, void *dev_id);
 
@@ -153,10 +153,13 @@
 			  uint32_t *mbx_sts);
 int qla4xxx_clear_ddb_entry(struct scsi_qla_host *ha, uint32_t fw_ddb_index);
 int qla4xxx_send_passthru0(struct iscsi_task *task);
+void qla4xxx_free_ddb_index(struct scsi_qla_host *ha);
 int qla4xxx_get_mgmt_data(struct scsi_qla_host *ha, uint16_t fw_ddb_index,
 			  uint16_t stats_size, dma_addr_t stats_dma);
 void qla4xxx_update_session_conn_param(struct scsi_qla_host *ha,
 				       struct ddb_entry *ddb_entry);
+void qla4xxx_update_session_conn_fwddb_param(struct scsi_qla_host *ha,
+					     struct ddb_entry *ddb_entry);
 int qla4xxx_bootdb_by_index(struct scsi_qla_host *ha,
 			    struct dev_db_entry *fw_ddb_entry,
 			    dma_addr_t fw_ddb_entry_dma, uint16_t ddb_index);
@@ -169,11 +172,22 @@
 int qla4xxx_restore_factory_defaults(struct scsi_qla_host *ha,
 				     uint32_t region, uint32_t field0,
 				     uint32_t field1);
+int qla4xxx_get_ddb_index(struct scsi_qla_host *ha, uint16_t *ddb_index);
+void qla4xxx_login_flash_ddb(struct iscsi_cls_session *cls_session);
+int qla4xxx_unblock_ddb(struct iscsi_cls_session *cls_session);
+int qla4xxx_unblock_flash_ddb(struct iscsi_cls_session *cls_session);
+int qla4xxx_flash_ddb_change(struct scsi_qla_host *ha, uint32_t fw_ddb_index,
+			     struct ddb_entry *ddb_entry, uint32_t state);
+int qla4xxx_ddb_change(struct scsi_qla_host *ha, uint32_t fw_ddb_index,
+		       struct ddb_entry *ddb_entry, uint32_t state);
+void qla4xxx_build_ddb_list(struct scsi_qla_host *ha, int is_reset);
 
 /* BSG Functions */
 int qla4xxx_bsg_request(struct bsg_job *bsg_job);
 int qla4xxx_process_vendor_specific(struct bsg_job *bsg_job);
 
+void qla4xxx_arm_relogin_timer(struct ddb_entry *ddb_entry);
+
 extern int ql4xextended_error_logging;
 extern int ql4xdontresethba;
 extern int ql4xenablemsix;
diff --git a/drivers/scsi/qla4xxx/ql4_init.c b/drivers/scsi/qla4xxx/ql4_init.c
index 3075fba..1bdfa81 100644
--- a/drivers/scsi/qla4xxx/ql4_init.c
+++ b/drivers/scsi/qla4xxx/ql4_init.c
@@ -773,22 +773,24 @@
  * be freed so that when login happens from user space there are free DDB
  * indices available.
  **/
-static void qla4xxx_free_ddb_index(struct scsi_qla_host *ha)
+void qla4xxx_free_ddb_index(struct scsi_qla_host *ha)
 {
 	int max_ddbs;
 	int ret;
 	uint32_t idx = 0, next_idx = 0;
 	uint32_t state = 0, conn_err = 0;
 
-	max_ddbs =  is_qla40XX(ha) ? MAX_PRST_DEV_DB_ENTRIES :
+	max_ddbs =  is_qla40XX(ha) ? MAX_DEV_DB_ENTRIES_40XX :
 				     MAX_DEV_DB_ENTRIES;
 
 	for (idx = 0; idx < max_ddbs; idx = next_idx) {
 		ret = qla4xxx_get_fwddb_entry(ha, idx, NULL, 0, NULL,
 					      &next_idx, &state, &conn_err,
 						NULL, NULL);
-		if (ret == QLA_ERROR)
+		if (ret == QLA_ERROR) {
+			next_idx++;
 			continue;
+		}
 		if (state == DDB_DS_NO_CONNECTION_ACTIVE ||
 		    state == DDB_DS_SESSION_FAILED) {
 			DEBUG2(ql4_printk(KERN_INFO, ha,
@@ -804,7 +806,6 @@
 	}
 }
 
-
 /**
  * qla4xxx_initialize_adapter - initiailizes hba
  * @ha: Pointer to host adapter structure.
@@ -812,7 +813,7 @@
  * This routine parforms all of the steps necessary to initialize the adapter.
  *
  **/
-int qla4xxx_initialize_adapter(struct scsi_qla_host *ha)
+int qla4xxx_initialize_adapter(struct scsi_qla_host *ha, int is_reset)
 {
 	int status = QLA_ERROR;
 
@@ -840,7 +841,8 @@
 	if (status == QLA_ERROR)
 		goto exit_init_hba;
 
-	qla4xxx_free_ddb_index(ha);
+	if (is_reset == RESET_ADAPTER)
+		qla4xxx_build_ddb_list(ha, is_reset);
 
 	set_bit(AF_ONLINE, &ha->flags);
 exit_init_hba:
@@ -855,38 +857,12 @@
 	return status;
 }
 
-/**
- * qla4xxx_process_ddb_changed - process ddb state change
- * @ha - Pointer to host adapter structure.
- * @fw_ddb_index - Firmware's device database index
- * @state - Device state
- *
- * This routine processes a Decive Database Changed AEN Event.
- **/
-int qla4xxx_process_ddb_changed(struct scsi_qla_host *ha, uint32_t fw_ddb_index,
-		uint32_t state, uint32_t conn_err)
+int qla4xxx_ddb_change(struct scsi_qla_host *ha, uint32_t fw_ddb_index,
+		       struct ddb_entry *ddb_entry, uint32_t state)
 {
-	struct ddb_entry * ddb_entry;
 	uint32_t old_fw_ddb_device_state;
 	int status = QLA_ERROR;
 
-	/* check for out of range index */
-	if (fw_ddb_index >= MAX_DDB_ENTRIES)
-		goto exit_ddb_event;
-
-	/* Get the corresponging ddb entry */
-	ddb_entry = qla4xxx_lookup_ddb_by_fw_index(ha, fw_ddb_index);
-	/* Device does not currently exist in our database. */
-	if (ddb_entry == NULL) {
-		ql4_printk(KERN_ERR, ha, "%s: No ddb_entry at FW index [%d]\n",
-			   __func__, fw_ddb_index);
-
-		if (state == DDB_DS_NO_CONNECTION_ACTIVE)
-			clear_bit(fw_ddb_index, ha->ddb_idx_map);
-
-		goto exit_ddb_event;
-	}
-
 	old_fw_ddb_device_state = ddb_entry->fw_ddb_device_state;
 	DEBUG2(ql4_printk(KERN_INFO, ha,
 			  "%s: DDB - old state = 0x%x, new state = 0x%x for "
@@ -900,9 +876,7 @@
 		switch (state) {
 		case DDB_DS_SESSION_ACTIVE:
 		case DDB_DS_DISCOVERY:
-			iscsi_conn_start(ddb_entry->conn);
-			iscsi_conn_login_event(ddb_entry->conn,
-					       ISCSI_CONN_STATE_LOGGED_IN);
+			ddb_entry->unblock_sess(ddb_entry->sess);
 			qla4xxx_update_session_conn_param(ha, ddb_entry);
 			status = QLA_SUCCESS;
 			break;
@@ -936,9 +910,7 @@
 		switch (state) {
 		case DDB_DS_SESSION_ACTIVE:
 		case DDB_DS_DISCOVERY:
-			iscsi_conn_start(ddb_entry->conn);
-			iscsi_conn_login_event(ddb_entry->conn,
-					       ISCSI_CONN_STATE_LOGGED_IN);
+			ddb_entry->unblock_sess(ddb_entry->sess);
 			qla4xxx_update_session_conn_param(ha, ddb_entry);
 			status = QLA_SUCCESS;
 			break;
@@ -954,7 +926,198 @@
 				__func__));
 		break;
 	}
+	return status;
+}
+
+void qla4xxx_arm_relogin_timer(struct ddb_entry *ddb_entry)
+{
+	/*
+	 * This triggers a relogin.  After the relogin_timer
+	 * expires, the relogin gets scheduled.  We must wait a
+	 * minimum amount of time since receiving an 0x8014 AEN
+	 * with failed device_state or a logout response before
+	 * we can issue another relogin.
+	 *
+	 * Firmware pads this timeout: (time2wait +1).
+	 * Driver retry to login should be longer than F/W.
+	 * Otherwise F/W will fail
+	 * set_ddb() mbx cmd with 0x4005 since it still
+	 * counting down its time2wait.
+	 */
+	atomic_set(&ddb_entry->relogin_timer, 0);
+	atomic_set(&ddb_entry->retry_relogin_timer,
+		   ddb_entry->default_time2wait + 4);
+
+}
+
+int qla4xxx_flash_ddb_change(struct scsi_qla_host *ha, uint32_t fw_ddb_index,
+			     struct ddb_entry *ddb_entry, uint32_t state)
+{
+	uint32_t old_fw_ddb_device_state;
+	int status = QLA_ERROR;
+
+	old_fw_ddb_device_state = ddb_entry->fw_ddb_device_state;
+	DEBUG2(ql4_printk(KERN_INFO, ha,
+			  "%s: DDB - old state = 0x%x, new state = 0x%x for "
+			  "index [%d]\n", __func__,
+			  ddb_entry->fw_ddb_device_state, state, fw_ddb_index));
+
+	ddb_entry->fw_ddb_device_state = state;
+
+	switch (old_fw_ddb_device_state) {
+	case DDB_DS_LOGIN_IN_PROCESS:
+	case DDB_DS_NO_CONNECTION_ACTIVE:
+		switch (state) {
+		case DDB_DS_SESSION_ACTIVE:
+			ddb_entry->unblock_sess(ddb_entry->sess);
+			qla4xxx_update_session_conn_fwddb_param(ha, ddb_entry);
+			status = QLA_SUCCESS;
+			break;
+		case DDB_DS_SESSION_FAILED:
+			iscsi_block_session(ddb_entry->sess);
+			if (!test_bit(DF_RELOGIN, &ddb_entry->flags))
+				qla4xxx_arm_relogin_timer(ddb_entry);
+			status = QLA_SUCCESS;
+			break;
+		}
+		break;
+	case DDB_DS_SESSION_ACTIVE:
+		switch (state) {
+		case DDB_DS_SESSION_FAILED:
+			iscsi_block_session(ddb_entry->sess);
+			if (!test_bit(DF_RELOGIN, &ddb_entry->flags))
+				qla4xxx_arm_relogin_timer(ddb_entry);
+			status = QLA_SUCCESS;
+			break;
+		}
+		break;
+	case DDB_DS_SESSION_FAILED:
+		switch (state) {
+		case DDB_DS_SESSION_ACTIVE:
+			ddb_entry->unblock_sess(ddb_entry->sess);
+			qla4xxx_update_session_conn_fwddb_param(ha, ddb_entry);
+			status = QLA_SUCCESS;
+			break;
+		case DDB_DS_SESSION_FAILED:
+			if (!test_bit(DF_RELOGIN, &ddb_entry->flags))
+				qla4xxx_arm_relogin_timer(ddb_entry);
+			status = QLA_SUCCESS;
+			break;
+		}
+		break;
+	default:
+		DEBUG2(ql4_printk(KERN_INFO, ha, "%s: Unknown Event\n",
+				  __func__));
+		break;
+	}
+	return status;
+}
+
+/**
+ * qla4xxx_process_ddb_changed - process ddb state change
+ * @ha - Pointer to host adapter structure.
+ * @fw_ddb_index - Firmware's device database index
+ * @state - Device state
+ *
+ * This routine processes a Decive Database Changed AEN Event.
+ **/
+int qla4xxx_process_ddb_changed(struct scsi_qla_host *ha,
+				uint32_t fw_ddb_index,
+				uint32_t state, uint32_t conn_err)
+{
+	struct ddb_entry *ddb_entry;
+	int status = QLA_ERROR;
+
+	/* check for out of range index */
+	if (fw_ddb_index >= MAX_DDB_ENTRIES)
+		goto exit_ddb_event;
+
+	/* Get the corresponging ddb entry */
+	ddb_entry = qla4xxx_lookup_ddb_by_fw_index(ha, fw_ddb_index);
+	/* Device does not currently exist in our database. */
+	if (ddb_entry == NULL) {
+		ql4_printk(KERN_ERR, ha, "%s: No ddb_entry at FW index [%d]\n",
+			   __func__, fw_ddb_index);
+
+		if (state == DDB_DS_NO_CONNECTION_ACTIVE)
+			clear_bit(fw_ddb_index, ha->ddb_idx_map);
+
+		goto exit_ddb_event;
+	}
+
+	ddb_entry->ddb_change(ha, fw_ddb_index, ddb_entry, state);
 
 exit_ddb_event:
 	return status;
 }
+
+/**
+ * qla4xxx_login_flash_ddb - Login to target (DDB)
+ * @cls_session: Pointer to the session to login
+ *
+ * This routine logins to the target.
+ * Issues setddb and conn open mbx
+ **/
+void qla4xxx_login_flash_ddb(struct iscsi_cls_session *cls_session)
+{
+	struct iscsi_session *sess;
+	struct ddb_entry *ddb_entry;
+	struct scsi_qla_host *ha;
+	struct dev_db_entry *fw_ddb_entry = NULL;
+	dma_addr_t fw_ddb_dma;
+	uint32_t mbx_sts = 0;
+	int ret;
+
+	sess = cls_session->dd_data;
+	ddb_entry = sess->dd_data;
+	ha =  ddb_entry->ha;
+
+	if (!test_bit(AF_LINK_UP, &ha->flags))
+		return;
+
+	if (ddb_entry->ddb_type != FLASH_DDB) {
+		DEBUG2(ql4_printk(KERN_INFO, ha,
+				  "Skipping login to non FLASH DB"));
+		goto exit_login;
+	}
+
+	fw_ddb_entry = dma_pool_alloc(ha->fw_ddb_dma_pool, GFP_KERNEL,
+				      &fw_ddb_dma);
+	if (fw_ddb_entry == NULL) {
+		DEBUG2(ql4_printk(KERN_ERR, ha, "Out of memory\n"));
+		goto exit_login;
+	}
+
+	if (ddb_entry->fw_ddb_index == INVALID_ENTRY) {
+		ret = qla4xxx_get_ddb_index(ha, &ddb_entry->fw_ddb_index);
+		if (ret == QLA_ERROR)
+			goto exit_login;
+
+		ha->fw_ddb_index_map[ddb_entry->fw_ddb_index] = ddb_entry;
+		ha->tot_ddbs++;
+	}
+
+	memcpy(fw_ddb_entry, &ddb_entry->fw_ddb_entry,
+	       sizeof(struct dev_db_entry));
+	ddb_entry->sess->target_id = ddb_entry->fw_ddb_index;
+
+	ret = qla4xxx_set_ddb_entry(ha, ddb_entry->fw_ddb_index,
+				    fw_ddb_dma, &mbx_sts);
+	if (ret == QLA_ERROR) {
+		DEBUG2(ql4_printk(KERN_ERR, ha, "Set DDB failed\n"));
+		goto exit_login;
+	}
+
+	ddb_entry->fw_ddb_device_state = DDB_DS_LOGIN_IN_PROCESS;
+	ret = qla4xxx_conn_open(ha, ddb_entry->fw_ddb_index);
+	if (ret == QLA_ERROR) {
+		ql4_printk(KERN_ERR, ha, "%s: Login failed: %s\n", __func__,
+			   sess->targetname);
+		goto exit_login;
+	}
+
+exit_login:
+	if (fw_ddb_entry)
+		dma_pool_free(ha->fw_ddb_dma_pool, fw_ddb_entry, fw_ddb_dma);
+}
+
diff --git a/drivers/scsi/qla4xxx/ql4_mbx.c b/drivers/scsi/qla4xxx/ql4_mbx.c
index 4c2b848..c259378 100644
--- a/drivers/scsi/qla4xxx/ql4_mbx.c
+++ b/drivers/scsi/qla4xxx/ql4_mbx.c
@@ -41,6 +41,16 @@
 		return status;
 	}
 
+	if (is_qla40XX(ha)) {
+		if (test_bit(AF_HA_REMOVAL, &ha->flags)) {
+			DEBUG2(ql4_printk(KERN_WARNING, ha, "scsi%ld: %s: "
+					  "prematurely completing mbx cmd as "
+					  "adapter removal detected\n",
+					  ha->host_no, __func__));
+			return status;
+		}
+	}
+
 	if (is_qla8022(ha)) {
 		if (test_bit(AF_FW_RECOVERY, &ha->flags)) {
 			DEBUG2(ql4_printk(KERN_WARNING, ha, "scsi%ld: %s: "
@@ -413,6 +423,7 @@
 	memcpy(ha->name_string, init_fw_cb->iscsi_name,
 		min(sizeof(ha->name_string),
 		sizeof(init_fw_cb->iscsi_name)));
+	ha->def_timeout = le16_to_cpu(init_fw_cb->def_timeout);
 	/*memcpy(ha->alias, init_fw_cb->Alias,
 	       min(sizeof(ha->alias), sizeof(init_fw_cb->Alias)));*/
 
diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c
index 30f31b1..4169c8b 100644
--- a/drivers/scsi/qla4xxx/ql4_os.c
+++ b/drivers/scsi/qla4xxx/ql4_os.c
@@ -8,6 +8,7 @@
 #include <linux/slab.h>
 #include <linux/blkdev.h>
 #include <linux/iscsi_boot_sysfs.h>
+#include <linux/inet.h>
 
 #include <scsi/scsi_tcq.h>
 #include <scsi/scsicam.h>
@@ -31,6 +32,13 @@
 /*
  * Module parameter information and variables
  */
+int ql4xdisablesysfsboot = 1;
+module_param(ql4xdisablesysfsboot, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(ql4xdisablesysfsboot,
+		"Set to disable exporting boot targets to sysfs\n"
+		" 0 - Export boot targets\n"
+		" 1 - Do not export boot targets (Default)");
+
 int ql4xdontresethba = 0;
 module_param(ql4xdontresethba, int, S_IRUGO | S_IWUSR);
 MODULE_PARM_DESC(ql4xdontresethba,
@@ -63,7 +71,7 @@
 module_param(ql4xsess_recovery_tmo, int, S_IRUGO);
 MODULE_PARM_DESC(ql4xsess_recovery_tmo,
 		"Target Session Recovery Timeout.\n"
-		" Default: 30 sec.");
+		" Default: 120 sec.");
 
 static int qla4xxx_wait_for_hba_online(struct scsi_qla_host *ha);
 /*
@@ -415,7 +423,7 @@
 	qla_ep = ep->dd_data;
 	ha = to_qla_host(qla_ep->host);
 
-	if (adapter_up(ha))
+	if (adapter_up(ha) && !test_bit(AF_BUILD_DDB_LIST, &ha->flags))
 		ret = 1;
 
 	return ret;
@@ -975,6 +983,150 @@
 
 }
 
+int qla4xxx_get_ddb_index(struct scsi_qla_host *ha, uint16_t *ddb_index)
+{
+	uint32_t mbx_sts = 0;
+	uint16_t tmp_ddb_index;
+	int ret;
+
+get_ddb_index:
+	tmp_ddb_index = find_first_zero_bit(ha->ddb_idx_map, MAX_DDB_ENTRIES);
+
+	if (tmp_ddb_index >= MAX_DDB_ENTRIES) {
+		DEBUG2(ql4_printk(KERN_INFO, ha,
+				  "Free DDB index not available\n"));
+		ret = QLA_ERROR;
+		goto exit_get_ddb_index;
+	}
+
+	if (test_and_set_bit(tmp_ddb_index, ha->ddb_idx_map))
+		goto get_ddb_index;
+
+	DEBUG2(ql4_printk(KERN_INFO, ha,
+			  "Found a free DDB index at %d\n", tmp_ddb_index));
+	ret = qla4xxx_req_ddb_entry(ha, tmp_ddb_index, &mbx_sts);
+	if (ret == QLA_ERROR) {
+		if (mbx_sts == MBOX_STS_COMMAND_ERROR) {
+			ql4_printk(KERN_INFO, ha,
+				   "DDB index = %d not available trying next\n",
+				   tmp_ddb_index);
+			goto get_ddb_index;
+		}
+		DEBUG2(ql4_printk(KERN_INFO, ha,
+				  "Free FW DDB not available\n"));
+	}
+
+	*ddb_index = tmp_ddb_index;
+
+exit_get_ddb_index:
+	return ret;
+}
+
+static int qla4xxx_match_ipaddress(struct scsi_qla_host *ha,
+				   struct ddb_entry *ddb_entry,
+				   char *existing_ipaddr,
+				   char *user_ipaddr)
+{
+	uint8_t dst_ipaddr[IPv6_ADDR_LEN];
+	char formatted_ipaddr[DDB_IPADDR_LEN];
+	int status = QLA_SUCCESS, ret = 0;
+
+	if (ddb_entry->fw_ddb_entry.options & DDB_OPT_IPV6_DEVICE) {
+		ret = in6_pton(user_ipaddr, strlen(user_ipaddr), dst_ipaddr,
+			       '\0', NULL);
+		if (ret == 0) {
+			status = QLA_ERROR;
+			goto out_match;
+		}
+		ret = sprintf(formatted_ipaddr, "%pI6", dst_ipaddr);
+	} else {
+		ret = in4_pton(user_ipaddr, strlen(user_ipaddr), dst_ipaddr,
+			       '\0', NULL);
+		if (ret == 0) {
+			status = QLA_ERROR;
+			goto out_match;
+		}
+		ret = sprintf(formatted_ipaddr, "%pI4", dst_ipaddr);
+	}
+
+	if (strcmp(existing_ipaddr, formatted_ipaddr))
+		status = QLA_ERROR;
+
+out_match:
+	return status;
+}
+
+static int qla4xxx_match_fwdb_session(struct scsi_qla_host *ha,
+				      struct iscsi_cls_conn *cls_conn)
+{
+	int idx = 0, max_ddbs, rval;
+	struct iscsi_cls_session *cls_sess = iscsi_conn_to_session(cls_conn);
+	struct iscsi_session *sess, *existing_sess;
+	struct iscsi_conn *conn, *existing_conn;
+	struct ddb_entry *ddb_entry;
+
+	sess = cls_sess->dd_data;
+	conn = cls_conn->dd_data;
+
+	if (sess->targetname == NULL ||
+	    conn->persistent_address == NULL ||
+	    conn->persistent_port == 0)
+		return QLA_ERROR;
+
+	max_ddbs =  is_qla40XX(ha) ? MAX_DEV_DB_ENTRIES_40XX :
+				     MAX_DEV_DB_ENTRIES;
+
+	for (idx = 0; idx < max_ddbs; idx++) {
+		ddb_entry = qla4xxx_lookup_ddb_by_fw_index(ha, idx);
+		if (ddb_entry == NULL)
+			continue;
+
+		if (ddb_entry->ddb_type != FLASH_DDB)
+			continue;
+
+		existing_sess = ddb_entry->sess->dd_data;
+		existing_conn = ddb_entry->conn->dd_data;
+
+		if (existing_sess->targetname == NULL ||
+		    existing_conn->persistent_address == NULL ||
+		    existing_conn->persistent_port == 0)
+			continue;
+
+		DEBUG2(ql4_printk(KERN_INFO, ha,
+				  "IQN = %s User IQN = %s\n",
+				  existing_sess->targetname,
+				  sess->targetname));
+
+		DEBUG2(ql4_printk(KERN_INFO, ha,
+				  "IP = %s User IP = %s\n",
+				  existing_conn->persistent_address,
+				  conn->persistent_address));
+
+		DEBUG2(ql4_printk(KERN_INFO, ha,
+				  "Port = %d User Port = %d\n",
+				  existing_conn->persistent_port,
+				  conn->persistent_port));
+
+		if (strcmp(existing_sess->targetname, sess->targetname))
+			continue;
+		rval = qla4xxx_match_ipaddress(ha, ddb_entry,
+					existing_conn->persistent_address,
+					conn->persistent_address);
+		if (rval == QLA_ERROR)
+			continue;
+		if (existing_conn->persistent_port != conn->persistent_port)
+			continue;
+		break;
+	}
+
+	if (idx == max_ddbs)
+		return QLA_ERROR;
+
+	DEBUG2(ql4_printk(KERN_INFO, ha,
+			  "Match found in fwdb sessions\n"));
+	return QLA_SUCCESS;
+}
+
 static struct iscsi_cls_session *
 qla4xxx_session_create(struct iscsi_endpoint *ep,
 			uint16_t cmds_max, uint16_t qdepth,
@@ -984,8 +1136,7 @@
 	struct scsi_qla_host *ha;
 	struct qla_endpoint *qla_ep;
 	struct ddb_entry *ddb_entry;
-	uint32_t ddb_index;
-	uint32_t mbx_sts = 0;
+	uint16_t ddb_index;
 	struct iscsi_session *sess;
 	struct sockaddr *dst_addr;
 	int ret;
@@ -1000,32 +1151,9 @@
 	dst_addr = (struct sockaddr *)&qla_ep->dst_addr;
 	ha = to_qla_host(qla_ep->host);
 
-get_ddb_index:
-	ddb_index = find_first_zero_bit(ha->ddb_idx_map, MAX_DDB_ENTRIES);
-
-	if (ddb_index >= MAX_DDB_ENTRIES) {
-		DEBUG2(ql4_printk(KERN_INFO, ha,
-				  "Free DDB index not available\n"));
+	ret = qla4xxx_get_ddb_index(ha, &ddb_index);
+	if (ret == QLA_ERROR)
 		return NULL;
-	}
-
-	if (test_and_set_bit(ddb_index, ha->ddb_idx_map))
-		goto get_ddb_index;
-
-	DEBUG2(ql4_printk(KERN_INFO, ha,
-			  "Found a free DDB index at %d\n", ddb_index));
-	ret = qla4xxx_req_ddb_entry(ha, ddb_index, &mbx_sts);
-	if (ret == QLA_ERROR) {
-		if (mbx_sts == MBOX_STS_COMMAND_ERROR) {
-			ql4_printk(KERN_INFO, ha,
-				   "DDB index = %d not available trying next\n",
-				   ddb_index);
-			goto get_ddb_index;
-		}
-		DEBUG2(ql4_printk(KERN_INFO, ha,
-				  "Free FW DDB not available\n"));
-		return NULL;
-	}
 
 	cls_sess = iscsi_session_setup(&qla4xxx_iscsi_transport, qla_ep->host,
 				       cmds_max, sizeof(struct ddb_entry),
@@ -1040,6 +1168,8 @@
 	ddb_entry->fw_ddb_device_state = DDB_DS_NO_CONNECTION_ACTIVE;
 	ddb_entry->ha = ha;
 	ddb_entry->sess = cls_sess;
+	ddb_entry->unblock_sess = qla4xxx_unblock_ddb;
+	ddb_entry->ddb_change = qla4xxx_ddb_change;
 	cls_sess->recovery_tmo = ql4xsess_recovery_tmo;
 	ha->fw_ddb_index_map[ddb_entry->fw_ddb_index] = ddb_entry;
 	ha->tot_ddbs++;
@@ -1077,6 +1207,9 @@
 	DEBUG2(printk(KERN_INFO "Func: %s\n", __func__));
 	cls_conn = iscsi_conn_setup(cls_sess, sizeof(struct qla_conn),
 				    conn_idx);
+	if (!cls_conn)
+		return NULL;
+
 	sess = cls_sess->dd_data;
 	ddb_entry = sess->dd_data;
 	ddb_entry->conn = cls_conn;
@@ -1109,7 +1242,7 @@
 	struct iscsi_session *sess;
 	struct ddb_entry *ddb_entry;
 	struct scsi_qla_host *ha;
-	struct dev_db_entry *fw_ddb_entry;
+	struct dev_db_entry *fw_ddb_entry = NULL;
 	dma_addr_t fw_ddb_entry_dma;
 	uint32_t mbx_sts = 0;
 	int ret = 0;
@@ -1120,12 +1253,25 @@
 	ddb_entry = sess->dd_data;
 	ha = ddb_entry->ha;
 
+	/* Check if we have  matching FW DDB, if yes then do not
+	 * login to this target. This could cause target to logout previous
+	 * connection
+	 */
+	ret = qla4xxx_match_fwdb_session(ha, cls_conn);
+	if (ret == QLA_SUCCESS) {
+		ql4_printk(KERN_INFO, ha,
+			   "Session already exist in FW.\n");
+		ret = -EEXIST;
+		goto exit_conn_start;
+	}
+
 	fw_ddb_entry = dma_alloc_coherent(&ha->pdev->dev, sizeof(*fw_ddb_entry),
 					  &fw_ddb_entry_dma, GFP_KERNEL);
 	if (!fw_ddb_entry) {
 		ql4_printk(KERN_ERR, ha,
 			   "%s: Unable to allocate dma buffer\n", __func__);
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto exit_conn_start;
 	}
 
 	ret = qla4xxx_set_param_ddbentry(ha, ddb_entry, cls_conn, &mbx_sts);
@@ -1138,9 +1284,7 @@
 		if (mbx_sts)
 			if (ddb_entry->fw_ddb_device_state ==
 						DDB_DS_SESSION_ACTIVE) {
-				iscsi_conn_start(ddb_entry->conn);
-				iscsi_conn_login_event(ddb_entry->conn,
-						ISCSI_CONN_STATE_LOGGED_IN);
+				ddb_entry->unblock_sess(ddb_entry->sess);
 				goto exit_set_param;
 			}
 
@@ -1167,8 +1311,9 @@
 	ret = 0;
 
 exit_conn_start:
-	dma_free_coherent(&ha->pdev->dev, sizeof(*fw_ddb_entry),
-			  fw_ddb_entry, fw_ddb_entry_dma);
+	if (fw_ddb_entry)
+		dma_free_coherent(&ha->pdev->dev, sizeof(*fw_ddb_entry),
+				  fw_ddb_entry, fw_ddb_entry_dma);
 	return ret;
 }
 
@@ -1344,6 +1489,101 @@
 	return -ENOSYS;
 }
 
+static void qla4xxx_copy_fwddb_param(struct scsi_qla_host *ha,
+				     struct dev_db_entry *fw_ddb_entry,
+				     struct iscsi_cls_session *cls_sess,
+				     struct iscsi_cls_conn *cls_conn)
+{
+	int buflen = 0;
+	struct iscsi_session *sess;
+	struct iscsi_conn *conn;
+	char ip_addr[DDB_IPADDR_LEN];
+	uint16_t options = 0;
+
+	sess = cls_sess->dd_data;
+	conn = cls_conn->dd_data;
+
+	conn->max_recv_dlength = BYTE_UNITS *
+			  le16_to_cpu(fw_ddb_entry->iscsi_max_rcv_data_seg_len);
+
+	conn->max_xmit_dlength = BYTE_UNITS *
+			  le16_to_cpu(fw_ddb_entry->iscsi_max_snd_data_seg_len);
+
+	sess->initial_r2t_en =
+			    (BIT_10 & le16_to_cpu(fw_ddb_entry->iscsi_options));
+
+	sess->max_r2t = le16_to_cpu(fw_ddb_entry->iscsi_max_outsnd_r2t);
+
+	sess->imm_data_en = (BIT_11 & le16_to_cpu(fw_ddb_entry->iscsi_options));
+
+	sess->first_burst = BYTE_UNITS *
+			       le16_to_cpu(fw_ddb_entry->iscsi_first_burst_len);
+
+	sess->max_burst = BYTE_UNITS *
+				 le16_to_cpu(fw_ddb_entry->iscsi_max_burst_len);
+
+	sess->time2wait = le16_to_cpu(fw_ddb_entry->iscsi_def_time2wait);
+
+	sess->time2retain = le16_to_cpu(fw_ddb_entry->iscsi_def_time2retain);
+
+	conn->persistent_port = le16_to_cpu(fw_ddb_entry->port);
+
+	sess->tpgt = le32_to_cpu(fw_ddb_entry->tgt_portal_grp);
+
+	options = le16_to_cpu(fw_ddb_entry->options);
+	if (options & DDB_OPT_IPV6_DEVICE)
+		sprintf(ip_addr, "%pI6", fw_ddb_entry->ip_addr);
+	else
+		sprintf(ip_addr, "%pI4", fw_ddb_entry->ip_addr);
+
+	iscsi_set_param(cls_conn, ISCSI_PARAM_TARGET_NAME,
+			(char *)fw_ddb_entry->iscsi_name, buflen);
+	iscsi_set_param(cls_conn, ISCSI_PARAM_INITIATOR_NAME,
+			(char *)ha->name_string, buflen);
+	iscsi_set_param(cls_conn, ISCSI_PARAM_PERSISTENT_ADDRESS,
+			(char *)ip_addr, buflen);
+}
+
+void qla4xxx_update_session_conn_fwddb_param(struct scsi_qla_host *ha,
+					     struct ddb_entry *ddb_entry)
+{
+	struct iscsi_cls_session *cls_sess;
+	struct iscsi_cls_conn *cls_conn;
+	uint32_t ddb_state;
+	dma_addr_t fw_ddb_entry_dma;
+	struct dev_db_entry *fw_ddb_entry;
+
+	fw_ddb_entry = dma_alloc_coherent(&ha->pdev->dev, sizeof(*fw_ddb_entry),
+					  &fw_ddb_entry_dma, GFP_KERNEL);
+	if (!fw_ddb_entry) {
+		ql4_printk(KERN_ERR, ha,
+			   "%s: Unable to allocate dma buffer\n", __func__);
+		goto exit_session_conn_fwddb_param;
+	}
+
+	if (qla4xxx_get_fwddb_entry(ha, ddb_entry->fw_ddb_index, fw_ddb_entry,
+				    fw_ddb_entry_dma, NULL, NULL, &ddb_state,
+				    NULL, NULL, NULL) == QLA_ERROR) {
+		DEBUG2(ql4_printk(KERN_ERR, ha, "scsi%ld: %s: failed "
+				  "get_ddb_entry for fw_ddb_index %d\n",
+				  ha->host_no, __func__,
+				  ddb_entry->fw_ddb_index));
+		goto exit_session_conn_fwddb_param;
+	}
+
+	cls_sess = ddb_entry->sess;
+
+	cls_conn = ddb_entry->conn;
+
+	/* Update params */
+	qla4xxx_copy_fwddb_param(ha, fw_ddb_entry, cls_sess, cls_conn);
+
+exit_session_conn_fwddb_param:
+	if (fw_ddb_entry)
+		dma_free_coherent(&ha->pdev->dev, sizeof(*fw_ddb_entry),
+				  fw_ddb_entry, fw_ddb_entry_dma);
+}
+
 void qla4xxx_update_session_conn_param(struct scsi_qla_host *ha,
 				       struct ddb_entry *ddb_entry)
 {
@@ -1360,7 +1600,7 @@
 	if (!fw_ddb_entry) {
 		ql4_printk(KERN_ERR, ha,
 			   "%s: Unable to allocate dma buffer\n", __func__);
-		return;
+		goto exit_session_conn_param;
 	}
 
 	if (qla4xxx_get_fwddb_entry(ha, ddb_entry->fw_ddb_index, fw_ddb_entry,
@@ -1370,7 +1610,7 @@
 				  "get_ddb_entry for fw_ddb_index %d\n",
 				  ha->host_no, __func__,
 				  ddb_entry->fw_ddb_index));
-		return;
+		goto exit_session_conn_param;
 	}
 
 	cls_sess = ddb_entry->sess;
@@ -1379,6 +1619,12 @@
 	cls_conn = ddb_entry->conn;
 	conn = cls_conn->dd_data;
 
+	/* Update timers after login */
+	ddb_entry->default_relogin_timeout =
+				le16_to_cpu(fw_ddb_entry->def_timeout);
+	ddb_entry->default_time2wait =
+				le16_to_cpu(fw_ddb_entry->iscsi_def_time2wait);
+
 	/* Update params */
 	conn->max_recv_dlength = BYTE_UNITS *
 			  le16_to_cpu(fw_ddb_entry->iscsi_max_rcv_data_seg_len);
@@ -1407,6 +1653,11 @@
 
 	memcpy(sess->initiatorname, ha->name_string,
 	       min(sizeof(ha->name_string), sizeof(sess->initiatorname)));
+
+exit_session_conn_param:
+	if (fw_ddb_entry)
+		dma_free_coherent(&ha->pdev->dev, sizeof(*fw_ddb_entry),
+				  fw_ddb_entry, fw_ddb_entry_dma);
 }
 
 /*
@@ -1607,6 +1858,9 @@
 		vfree(ha->chap_list);
 	ha->chap_list = NULL;
 
+	if (ha->fw_ddb_dma_pool)
+		dma_pool_destroy(ha->fw_ddb_dma_pool);
+
 	/* release io space registers  */
 	if (is_qla8022(ha)) {
 		if (ha->nx_pcibase)
@@ -1689,6 +1943,16 @@
 		goto mem_alloc_error_exit;
 	}
 
+	ha->fw_ddb_dma_pool = dma_pool_create("ql4_fw_ddb", &ha->pdev->dev,
+					      DDB_DMA_BLOCK_SIZE, 8, 0);
+
+	if (ha->fw_ddb_dma_pool == NULL) {
+		ql4_printk(KERN_WARNING, ha,
+			   "%s: fw_ddb_dma_pool allocation failed..\n",
+			   __func__);
+		goto mem_alloc_error_exit;
+	}
+
 	return QLA_SUCCESS;
 
 mem_alloc_error_exit:
@@ -1800,6 +2064,60 @@
 	}
 }
 
+void qla4xxx_check_relogin_flash_ddb(struct iscsi_cls_session *cls_sess)
+{
+	struct iscsi_session *sess;
+	struct ddb_entry *ddb_entry;
+	struct scsi_qla_host *ha;
+
+	sess = cls_sess->dd_data;
+	ddb_entry = sess->dd_data;
+	ha = ddb_entry->ha;
+
+	if (!(ddb_entry->ddb_type == FLASH_DDB))
+		return;
+
+	if (adapter_up(ha) && !test_bit(DF_RELOGIN, &ddb_entry->flags) &&
+	    !iscsi_is_session_online(cls_sess)) {
+		if (atomic_read(&ddb_entry->retry_relogin_timer) !=
+		    INVALID_ENTRY) {
+			if (atomic_read(&ddb_entry->retry_relogin_timer) ==
+					0) {
+				atomic_set(&ddb_entry->retry_relogin_timer,
+					   INVALID_ENTRY);
+				set_bit(DPC_RELOGIN_DEVICE, &ha->dpc_flags);
+				set_bit(DF_RELOGIN, &ddb_entry->flags);
+				DEBUG2(ql4_printk(KERN_INFO, ha,
+				       "%s: index [%d] login device\n",
+					__func__, ddb_entry->fw_ddb_index));
+			} else
+				atomic_dec(&ddb_entry->retry_relogin_timer);
+		}
+	}
+
+	/* Wait for relogin to timeout */
+	if (atomic_read(&ddb_entry->relogin_timer) &&
+	    (atomic_dec_and_test(&ddb_entry->relogin_timer) != 0)) {
+		/*
+		 * If the relogin times out and the device is
+		 * still NOT ONLINE then try and relogin again.
+		 */
+		if (!iscsi_is_session_online(cls_sess)) {
+			/* Reset retry relogin timer */
+			atomic_inc(&ddb_entry->relogin_retry_count);
+			DEBUG2(ql4_printk(KERN_INFO, ha,
+				"%s: index[%d] relogin timed out-retrying"
+				" relogin (%d), retry (%d)\n", __func__,
+				ddb_entry->fw_ddb_index,
+				atomic_read(&ddb_entry->relogin_retry_count),
+				ddb_entry->default_time2wait + 4));
+			set_bit(DPC_RELOGIN_DEVICE, &ha->dpc_flags);
+			atomic_set(&ddb_entry->retry_relogin_timer,
+				   ddb_entry->default_time2wait + 4);
+		}
+	}
+}
+
 /**
  * qla4xxx_timer - checks every second for work to do.
  * @ha: Pointer to host adapter structure.
@@ -1809,6 +2127,8 @@
 	int start_dpc = 0;
 	uint16_t w;
 
+	iscsi_host_for_each_session(ha->host, qla4xxx_check_relogin_flash_ddb);
+
 	/* If we are in the middle of AER/EEH processing
 	 * skip any processing and reschedule the timer
 	 */
@@ -2078,7 +2398,12 @@
 	sess = cls_session->dd_data;
 	ddb_entry = sess->dd_data;
 	ddb_entry->fw_ddb_device_state = DDB_DS_SESSION_FAILED;
-	iscsi_session_failure(cls_session->dd_data, ISCSI_ERR_CONN_FAILED);
+
+	if (ddb_entry->ddb_type == FLASH_DDB)
+		iscsi_block_session(ddb_entry->sess);
+	else
+		iscsi_session_failure(cls_session->dd_data,
+				      ISCSI_ERR_CONN_FAILED);
 }
 
 /**
@@ -2163,7 +2488,7 @@
 
 		/* NOTE: AF_ONLINE flag set upon successful completion of
 		 *       qla4xxx_initialize_adapter */
-		status = qla4xxx_initialize_adapter(ha);
+		status = qla4xxx_initialize_adapter(ha, RESET_ADAPTER);
 	}
 
 	/* Retry failed adapter initialization, if necessary
@@ -2245,17 +2570,108 @@
 			iscsi_unblock_session(ddb_entry->sess);
 		} else {
 			/* Trigger relogin */
-			iscsi_session_failure(cls_session->dd_data,
-					      ISCSI_ERR_CONN_FAILED);
+			if (ddb_entry->ddb_type == FLASH_DDB) {
+				if (!test_bit(DF_RELOGIN, &ddb_entry->flags))
+					qla4xxx_arm_relogin_timer(ddb_entry);
+			} else
+				iscsi_session_failure(cls_session->dd_data,
+						      ISCSI_ERR_CONN_FAILED);
 		}
 	}
 }
 
+int qla4xxx_unblock_flash_ddb(struct iscsi_cls_session *cls_session)
+{
+	struct iscsi_session *sess;
+	struct ddb_entry *ddb_entry;
+	struct scsi_qla_host *ha;
+
+	sess = cls_session->dd_data;
+	ddb_entry = sess->dd_data;
+	ha = ddb_entry->ha;
+	ql4_printk(KERN_INFO, ha, "scsi%ld: %s: ddb[%d]"
+		   " unblock session\n", ha->host_no, __func__,
+		   ddb_entry->fw_ddb_index);
+
+	iscsi_unblock_session(ddb_entry->sess);
+
+	/* Start scan target */
+	if (test_bit(AF_ONLINE, &ha->flags)) {
+		ql4_printk(KERN_INFO, ha, "scsi%ld: %s: ddb[%d]"
+			   " start scan\n", ha->host_no, __func__,
+			   ddb_entry->fw_ddb_index);
+		scsi_queue_work(ha->host, &ddb_entry->sess->scan_work);
+	}
+	return QLA_SUCCESS;
+}
+
+int qla4xxx_unblock_ddb(struct iscsi_cls_session *cls_session)
+{
+	struct iscsi_session *sess;
+	struct ddb_entry *ddb_entry;
+	struct scsi_qla_host *ha;
+
+	sess = cls_session->dd_data;
+	ddb_entry = sess->dd_data;
+	ha = ddb_entry->ha;
+	ql4_printk(KERN_INFO, ha, "scsi%ld: %s: ddb[%d]"
+		   " unblock user space session\n", ha->host_no, __func__,
+		   ddb_entry->fw_ddb_index);
+	iscsi_conn_start(ddb_entry->conn);
+	iscsi_conn_login_event(ddb_entry->conn,
+			       ISCSI_CONN_STATE_LOGGED_IN);
+
+	return QLA_SUCCESS;
+}
+
 static void qla4xxx_relogin_all_devices(struct scsi_qla_host *ha)
 {
 	iscsi_host_for_each_session(ha->host, qla4xxx_relogin_devices);
 }
 
+static void qla4xxx_relogin_flash_ddb(struct iscsi_cls_session *cls_sess)
+{
+	uint16_t relogin_timer;
+	struct iscsi_session *sess;
+	struct ddb_entry *ddb_entry;
+	struct scsi_qla_host *ha;
+
+	sess = cls_sess->dd_data;
+	ddb_entry = sess->dd_data;
+	ha = ddb_entry->ha;
+
+	relogin_timer = max(ddb_entry->default_relogin_timeout,
+			    (uint16_t)RELOGIN_TOV);
+	atomic_set(&ddb_entry->relogin_timer, relogin_timer);
+
+	DEBUG2(ql4_printk(KERN_INFO, ha,
+			  "scsi%ld: Relogin index [%d]. TOV=%d\n", ha->host_no,
+			  ddb_entry->fw_ddb_index, relogin_timer));
+
+	qla4xxx_login_flash_ddb(cls_sess);
+}
+
+static void qla4xxx_dpc_relogin(struct iscsi_cls_session *cls_sess)
+{
+	struct iscsi_session *sess;
+	struct ddb_entry *ddb_entry;
+	struct scsi_qla_host *ha;
+
+	sess = cls_sess->dd_data;
+	ddb_entry = sess->dd_data;
+	ha = ddb_entry->ha;
+
+	if (!(ddb_entry->ddb_type == FLASH_DDB))
+		return;
+
+	if (test_and_clear_bit(DF_RELOGIN, &ddb_entry->flags) &&
+	    !iscsi_is_session_online(cls_sess)) {
+		DEBUG2(ql4_printk(KERN_INFO, ha,
+				  "relogin issued\n"));
+		qla4xxx_relogin_flash_ddb(cls_sess);
+	}
+}
+
 void qla4xxx_wake_dpc(struct scsi_qla_host *ha)
 {
 	if (ha->dpc_thread)
@@ -2356,6 +2772,12 @@
 	if (test_and_clear_bit(DPC_GET_DHCP_IP_ADDR, &ha->dpc_flags))
 		qla4xxx_get_dhcp_ip_address(ha);
 
+	/* ---- relogin device? --- */
+	if (adapter_up(ha) &&
+	    test_and_clear_bit(DPC_RELOGIN_DEVICE, &ha->dpc_flags)) {
+		iscsi_host_for_each_session(ha->host, qla4xxx_dpc_relogin);
+	}
+
 	/* ---- link change? --- */
 	if (test_and_clear_bit(DPC_LINK_CHANGED, &ha->dpc_flags)) {
 		if (!test_bit(AF_LINK_UP, &ha->flags)) {
@@ -2368,8 +2790,12 @@
 			 * fatal error recovery.  Therefore, the driver must
 			 * manually relogin to devices when recovering from
 			 * connection failures, logouts, expired KATO, etc. */
-
-			qla4xxx_relogin_all_devices(ha);
+			if (test_and_clear_bit(AF_BUILD_DDB_LIST, &ha->flags)) {
+				qla4xxx_build_ddb_list(ha, ha->is_reset);
+				iscsi_host_for_each_session(ha->host,
+						qla4xxx_login_flash_ddb);
+			} else
+				qla4xxx_relogin_all_devices(ha);
 		}
 	}
 }
@@ -2867,6 +3293,9 @@
 			  " target ID %d\n", __func__, ddb_index[0],
 			  ddb_index[1]));
 
+	ha->pri_ddb_idx = ddb_index[0];
+	ha->sec_ddb_idx = ddb_index[1];
+
 exit_boot_info_free:
 	dma_free_coherent(&ha->pdev->dev, size, buf, buf_dma);
 exit_boot_info:
@@ -3034,6 +3463,9 @@
 		return ret;
 	}
 
+	if (ql4xdisablesysfsboot)
+		return QLA_SUCCESS;
+
 	if (ddb_index[0] == 0xffff)
 		goto sec_target;
 
@@ -3066,7 +3498,15 @@
 	struct iscsi_boot_kobj *boot_kobj;
 
 	if (qla4xxx_get_boot_info(ha) != QLA_SUCCESS)
-		return 0;
+		return QLA_ERROR;
+
+	if (ql4xdisablesysfsboot) {
+		ql4_printk(KERN_INFO, ha,
+			   "%s: syfsboot disabled - driver will trigger login"
+			   "and publish session for discovery .\n", __func__);
+		return QLA_SUCCESS;
+	}
+
 
 	ha->boot_kset = iscsi_boot_create_host_kset(ha->host->host_no);
 	if (!ha->boot_kset)
@@ -3108,7 +3548,7 @@
 	if (!boot_kobj)
 		goto put_host;
 
-	return 0;
+	return QLA_SUCCESS;
 
 put_host:
 	scsi_host_put(ha->host);
@@ -3174,9 +3614,507 @@
 exit_chap_list:
 	dma_free_coherent(&ha->pdev->dev, chap_size,
 			chap_flash_data, chap_dma);
-	return;
 }
 
+static void qla4xxx_get_param_ddb(struct ddb_entry *ddb_entry,
+				  struct ql4_tuple_ddb *tddb)
+{
+	struct scsi_qla_host *ha;
+	struct iscsi_cls_session *cls_sess;
+	struct iscsi_cls_conn *cls_conn;
+	struct iscsi_session *sess;
+	struct iscsi_conn *conn;
+
+	DEBUG2(printk(KERN_INFO "Func: %s\n", __func__));
+	ha = ddb_entry->ha;
+	cls_sess = ddb_entry->sess;
+	sess = cls_sess->dd_data;
+	cls_conn = ddb_entry->conn;
+	conn = cls_conn->dd_data;
+
+	tddb->tpgt = sess->tpgt;
+	tddb->port = conn->persistent_port;
+	strncpy(tddb->iscsi_name, sess->targetname, ISCSI_NAME_SIZE);
+	strncpy(tddb->ip_addr, conn->persistent_address, DDB_IPADDR_LEN);
+}
+
+static void qla4xxx_convert_param_ddb(struct dev_db_entry *fw_ddb_entry,
+				      struct ql4_tuple_ddb *tddb)
+{
+	uint16_t options = 0;
+
+	tddb->tpgt = le32_to_cpu(fw_ddb_entry->tgt_portal_grp);
+	memcpy(&tddb->iscsi_name[0], &fw_ddb_entry->iscsi_name[0],
+	       min(sizeof(tddb->iscsi_name), sizeof(fw_ddb_entry->iscsi_name)));
+
+	options = le16_to_cpu(fw_ddb_entry->options);
+	if (options & DDB_OPT_IPV6_DEVICE)
+		sprintf(tddb->ip_addr, "%pI6", fw_ddb_entry->ip_addr);
+	else
+		sprintf(tddb->ip_addr, "%pI4", fw_ddb_entry->ip_addr);
+
+	tddb->port = le16_to_cpu(fw_ddb_entry->port);
+}
+
+static int qla4xxx_compare_tuple_ddb(struct scsi_qla_host *ha,
+				     struct ql4_tuple_ddb *old_tddb,
+				     struct ql4_tuple_ddb *new_tddb)
+{
+	if (strcmp(old_tddb->iscsi_name, new_tddb->iscsi_name))
+		return QLA_ERROR;
+
+	if (strcmp(old_tddb->ip_addr, new_tddb->ip_addr))
+		return QLA_ERROR;
+
+	if (old_tddb->port != new_tddb->port)
+		return QLA_ERROR;
+
+	DEBUG2(ql4_printk(KERN_INFO, ha,
+			  "Match Found, fw[%d,%d,%s,%s], [%d,%d,%s,%s]",
+			  old_tddb->port, old_tddb->tpgt, old_tddb->ip_addr,
+			  old_tddb->iscsi_name, new_tddb->port, new_tddb->tpgt,
+			  new_tddb->ip_addr, new_tddb->iscsi_name));
+
+	return QLA_SUCCESS;
+}
+
+static int qla4xxx_is_session_exists(struct scsi_qla_host *ha,
+				     struct dev_db_entry *fw_ddb_entry)
+{
+	struct ddb_entry *ddb_entry;
+	struct ql4_tuple_ddb *fw_tddb = NULL;
+	struct ql4_tuple_ddb *tmp_tddb = NULL;
+	int idx;
+	int ret = QLA_ERROR;
+
+	fw_tddb = vzalloc(sizeof(*fw_tddb));
+	if (!fw_tddb) {
+		DEBUG2(ql4_printk(KERN_WARNING, ha,
+				  "Memory Allocation failed.\n"));
+		ret = QLA_SUCCESS;
+		goto exit_check;
+	}
+
+	tmp_tddb = vzalloc(sizeof(*tmp_tddb));
+	if (!tmp_tddb) {
+		DEBUG2(ql4_printk(KERN_WARNING, ha,
+				  "Memory Allocation failed.\n"));
+		ret = QLA_SUCCESS;
+		goto exit_check;
+	}
+
+	qla4xxx_convert_param_ddb(fw_ddb_entry, fw_tddb);
+
+	for (idx = 0; idx < MAX_DDB_ENTRIES; idx++) {
+		ddb_entry = qla4xxx_lookup_ddb_by_fw_index(ha, idx);
+		if (ddb_entry == NULL)
+			continue;
+
+		qla4xxx_get_param_ddb(ddb_entry, tmp_tddb);
+		if (!qla4xxx_compare_tuple_ddb(ha, fw_tddb, tmp_tddb)) {
+			ret = QLA_SUCCESS; /* found */
+			goto exit_check;
+		}
+	}
+
+exit_check:
+	if (fw_tddb)
+		vfree(fw_tddb);
+	if (tmp_tddb)
+		vfree(tmp_tddb);
+	return ret;
+}
+
+static int qla4xxx_is_flash_ddb_exists(struct scsi_qla_host *ha,
+				       struct list_head *list_nt,
+				       struct dev_db_entry *fw_ddb_entry)
+{
+	struct qla_ddb_index  *nt_ddb_idx, *nt_ddb_idx_tmp;
+	struct ql4_tuple_ddb *fw_tddb = NULL;
+	struct ql4_tuple_ddb *tmp_tddb = NULL;
+	int ret = QLA_ERROR;
+
+	fw_tddb = vzalloc(sizeof(*fw_tddb));
+	if (!fw_tddb) {
+		DEBUG2(ql4_printk(KERN_WARNING, ha,
+				  "Memory Allocation failed.\n"));
+		ret = QLA_SUCCESS;
+		goto exit_check;
+	}
+
+	tmp_tddb = vzalloc(sizeof(*tmp_tddb));
+	if (!tmp_tddb) {
+		DEBUG2(ql4_printk(KERN_WARNING, ha,
+				  "Memory Allocation failed.\n"));
+		ret = QLA_SUCCESS;
+		goto exit_check;
+	}
+
+	qla4xxx_convert_param_ddb(fw_ddb_entry, fw_tddb);
+
+	list_for_each_entry_safe(nt_ddb_idx, nt_ddb_idx_tmp, list_nt, list) {
+		qla4xxx_convert_param_ddb(&nt_ddb_idx->fw_ddb, tmp_tddb);
+		if (!qla4xxx_compare_tuple_ddb(ha, fw_tddb, tmp_tddb)) {
+			ret = QLA_SUCCESS; /* found */
+			goto exit_check;
+		}
+	}
+
+exit_check:
+	if (fw_tddb)
+		vfree(fw_tddb);
+	if (tmp_tddb)
+		vfree(tmp_tddb);
+	return ret;
+}
+
+static void qla4xxx_free_nt_list(struct list_head *list_nt)
+{
+	struct qla_ddb_index  *nt_ddb_idx, *nt_ddb_idx_tmp;
+
+	/* Free up the normaltargets list */
+	list_for_each_entry_safe(nt_ddb_idx, nt_ddb_idx_tmp, list_nt, list) {
+		list_del_init(&nt_ddb_idx->list);
+		vfree(nt_ddb_idx);
+	}
+
+}
+
+static struct iscsi_endpoint *qla4xxx_get_ep_fwdb(struct scsi_qla_host *ha,
+					struct dev_db_entry *fw_ddb_entry)
+{
+	struct iscsi_endpoint *ep;
+	struct sockaddr_in *addr;
+	struct sockaddr_in6 *addr6;
+	struct sockaddr *dst_addr;
+	char *ip;
+
+	/* TODO: need to destroy on unload iscsi_endpoint*/
+	dst_addr = vmalloc(sizeof(*dst_addr));
+	if (!dst_addr)
+		return NULL;
+
+	if (fw_ddb_entry->options & DDB_OPT_IPV6_DEVICE) {
+		dst_addr->sa_family = AF_INET6;
+		addr6 = (struct sockaddr_in6 *)dst_addr;
+		ip = (char *)&addr6->sin6_addr;
+		memcpy(ip, fw_ddb_entry->ip_addr, IPv6_ADDR_LEN);
+		addr6->sin6_port = htons(le16_to_cpu(fw_ddb_entry->port));
+
+	} else {
+		dst_addr->sa_family = AF_INET;
+		addr = (struct sockaddr_in *)dst_addr;
+		ip = (char *)&addr->sin_addr;
+		memcpy(ip, fw_ddb_entry->ip_addr, IP_ADDR_LEN);
+		addr->sin_port = htons(le16_to_cpu(fw_ddb_entry->port));
+	}
+
+	ep = qla4xxx_ep_connect(ha->host, dst_addr, 0);
+	vfree(dst_addr);
+	return ep;
+}
+
+static int qla4xxx_verify_boot_idx(struct scsi_qla_host *ha, uint16_t idx)
+{
+	if (ql4xdisablesysfsboot)
+		return QLA_SUCCESS;
+	if (idx == ha->pri_ddb_idx || idx == ha->sec_ddb_idx)
+		return QLA_ERROR;
+	return QLA_SUCCESS;
+}
+
+static void qla4xxx_setup_flash_ddb_entry(struct scsi_qla_host *ha,
+					  struct ddb_entry *ddb_entry)
+{
+	ddb_entry->ddb_type = FLASH_DDB;
+	ddb_entry->fw_ddb_index = INVALID_ENTRY;
+	ddb_entry->fw_ddb_device_state = DDB_DS_NO_CONNECTION_ACTIVE;
+	ddb_entry->ha = ha;
+	ddb_entry->unblock_sess = qla4xxx_unblock_flash_ddb;
+	ddb_entry->ddb_change = qla4xxx_flash_ddb_change;
+
+	atomic_set(&ddb_entry->retry_relogin_timer, INVALID_ENTRY);
+	atomic_set(&ddb_entry->relogin_timer, 0);
+	atomic_set(&ddb_entry->relogin_retry_count, 0);
+
+	ddb_entry->default_relogin_timeout =
+		le16_to_cpu(ddb_entry->fw_ddb_entry.def_timeout);
+	ddb_entry->default_time2wait =
+		le16_to_cpu(ddb_entry->fw_ddb_entry.iscsi_def_time2wait);
+}
+
+static void qla4xxx_wait_for_ip_configuration(struct scsi_qla_host *ha)
+{
+	uint32_t idx = 0;
+	uint32_t ip_idx[IP_ADDR_COUNT] = {0, 1, 2, 3}; /* 4 IP interfaces */
+	uint32_t sts[MBOX_REG_COUNT];
+	uint32_t ip_state;
+	unsigned long wtime;
+	int ret;
+
+	wtime = jiffies + (HZ * IP_CONFIG_TOV);
+	do {
+		for (idx = 0; idx < IP_ADDR_COUNT; idx++) {
+			if (ip_idx[idx] == -1)
+				continue;
+
+			ret = qla4xxx_get_ip_state(ha, 0, ip_idx[idx], sts);
+
+			if (ret == QLA_ERROR) {
+				ip_idx[idx] = -1;
+				continue;
+			}
+
+			ip_state = (sts[1] & IP_STATE_MASK) >> IP_STATE_SHIFT;
+
+			DEBUG2(ql4_printk(KERN_INFO, ha,
+					  "Waiting for IP state for idx = %d, state = 0x%x\n",
+					  ip_idx[idx], ip_state));
+			if (ip_state == IP_ADDRSTATE_UNCONFIGURED ||
+			    ip_state == IP_ADDRSTATE_INVALID ||
+			    ip_state == IP_ADDRSTATE_PREFERRED ||
+			    ip_state == IP_ADDRSTATE_DEPRICATED ||
+			    ip_state == IP_ADDRSTATE_DISABLING)
+				ip_idx[idx] = -1;
+
+		}
+
+		/* Break if all IP states checked */
+		if ((ip_idx[0] == -1) &&
+		    (ip_idx[1] == -1) &&
+		    (ip_idx[2] == -1) &&
+		    (ip_idx[3] == -1))
+			break;
+		schedule_timeout_uninterruptible(HZ);
+	} while (time_after(wtime, jiffies));
+}
+
+void qla4xxx_build_ddb_list(struct scsi_qla_host *ha, int is_reset)
+{
+	int max_ddbs;
+	int ret;
+	uint32_t idx = 0, next_idx = 0;
+	uint32_t state = 0, conn_err = 0;
+	uint16_t conn_id;
+	struct dev_db_entry *fw_ddb_entry;
+	struct ddb_entry *ddb_entry = NULL;
+	dma_addr_t fw_ddb_dma;
+	struct iscsi_cls_session *cls_sess;
+	struct iscsi_session *sess;
+	struct iscsi_cls_conn *cls_conn;
+	struct iscsi_endpoint *ep;
+	uint16_t cmds_max = 32, tmo = 0;
+	uint32_t initial_cmdsn = 0;
+	struct list_head list_st, list_nt; /* List of sendtargets */
+	struct qla_ddb_index  *st_ddb_idx, *st_ddb_idx_tmp;
+	int fw_idx_size;
+	unsigned long wtime;
+	struct qla_ddb_index  *nt_ddb_idx;
+
+	if (!test_bit(AF_LINK_UP, &ha->flags)) {
+		set_bit(AF_BUILD_DDB_LIST, &ha->flags);
+		ha->is_reset = is_reset;
+		return;
+	}
+	max_ddbs =  is_qla40XX(ha) ? MAX_DEV_DB_ENTRIES_40XX :
+				     MAX_DEV_DB_ENTRIES;
+
+	fw_ddb_entry = dma_pool_alloc(ha->fw_ddb_dma_pool, GFP_KERNEL,
+				      &fw_ddb_dma);
+	if (fw_ddb_entry == NULL) {
+		DEBUG2(ql4_printk(KERN_ERR, ha, "Out of memory\n"));
+		goto exit_ddb_list;
+	}
+
+	INIT_LIST_HEAD(&list_st);
+	INIT_LIST_HEAD(&list_nt);
+	fw_idx_size = sizeof(struct qla_ddb_index);
+
+	for (idx = 0; idx < max_ddbs; idx = next_idx) {
+		ret = qla4xxx_get_fwddb_entry(ha, idx, fw_ddb_entry,
+					      fw_ddb_dma, NULL,
+					      &next_idx, &state, &conn_err,
+					      NULL, &conn_id);
+		if (ret == QLA_ERROR)
+			break;
+
+		if (qla4xxx_verify_boot_idx(ha, idx) != QLA_SUCCESS)
+			goto continue_next_st;
+
+		/* Check if ST, add to the list_st */
+		if (strlen((char *) fw_ddb_entry->iscsi_name) != 0)
+			goto continue_next_st;
+
+		st_ddb_idx = vzalloc(fw_idx_size);
+		if (!st_ddb_idx)
+			break;
+
+		st_ddb_idx->fw_ddb_idx = idx;
+
+		list_add_tail(&st_ddb_idx->list, &list_st);
+continue_next_st:
+		if (next_idx == 0)
+			break;
+	}
+
+	/* Before issuing conn open mbox, ensure all IPs states are configured
+	 * Note, conn open fails if IPs are not configured
+	 */
+	qla4xxx_wait_for_ip_configuration(ha);
+
+	/* Go thru the STs and fire the sendtargets by issuing conn open mbx */
+	list_for_each_entry_safe(st_ddb_idx, st_ddb_idx_tmp, &list_st, list) {
+		qla4xxx_conn_open(ha, st_ddb_idx->fw_ddb_idx);
+	}
+
+	/* Wait to ensure all sendtargets are done for min 12 sec wait */
+	tmo = ((ha->def_timeout < LOGIN_TOV) ? LOGIN_TOV : ha->def_timeout);
+	DEBUG2(ql4_printk(KERN_INFO, ha,
+			  "Default time to wait for build ddb %d\n", tmo));
+
+	wtime = jiffies + (HZ * tmo);
+	do {
+		list_for_each_entry_safe(st_ddb_idx, st_ddb_idx_tmp, &list_st,
+					 list) {
+			ret = qla4xxx_get_fwddb_entry(ha,
+						      st_ddb_idx->fw_ddb_idx,
+						      NULL, 0, NULL, &next_idx,
+						      &state, &conn_err, NULL,
+						      NULL);
+			if (ret == QLA_ERROR)
+				continue;
+
+			if (state == DDB_DS_NO_CONNECTION_ACTIVE ||
+			    state == DDB_DS_SESSION_FAILED) {
+				list_del_init(&st_ddb_idx->list);
+				vfree(st_ddb_idx);
+			}
+		}
+		schedule_timeout_uninterruptible(HZ / 10);
+	} while (time_after(wtime, jiffies));
+
+	/* Free up the sendtargets list */
+	list_for_each_entry_safe(st_ddb_idx, st_ddb_idx_tmp, &list_st, list) {
+		list_del_init(&st_ddb_idx->list);
+		vfree(st_ddb_idx);
+	}
+
+	for (idx = 0; idx < max_ddbs; idx = next_idx) {
+		ret = qla4xxx_get_fwddb_entry(ha, idx, fw_ddb_entry,
+					      fw_ddb_dma, NULL,
+					      &next_idx, &state, &conn_err,
+					      NULL, &conn_id);
+		if (ret == QLA_ERROR)
+			break;
+
+		if (qla4xxx_verify_boot_idx(ha, idx) != QLA_SUCCESS)
+			goto continue_next_nt;
+
+		/* Check if NT, then add to list it */
+		if (strlen((char *) fw_ddb_entry->iscsi_name) == 0)
+			goto continue_next_nt;
+
+		if (state == DDB_DS_NO_CONNECTION_ACTIVE ||
+		    state == DDB_DS_SESSION_FAILED) {
+			DEBUG2(ql4_printk(KERN_INFO, ha,
+					  "Adding  DDB to session = 0x%x\n",
+					  idx));
+			if (is_reset == INIT_ADAPTER) {
+				nt_ddb_idx = vmalloc(fw_idx_size);
+				if (!nt_ddb_idx)
+					break;
+
+				nt_ddb_idx->fw_ddb_idx = idx;
+
+				memcpy(&nt_ddb_idx->fw_ddb, fw_ddb_entry,
+				       sizeof(struct dev_db_entry));
+
+				if (qla4xxx_is_flash_ddb_exists(ha, &list_nt,
+						fw_ddb_entry) == QLA_SUCCESS) {
+					vfree(nt_ddb_idx);
+					goto continue_next_nt;
+				}
+				list_add_tail(&nt_ddb_idx->list, &list_nt);
+			} else if (is_reset == RESET_ADAPTER) {
+				if (qla4xxx_is_session_exists(ha,
+						   fw_ddb_entry) == QLA_SUCCESS)
+					goto continue_next_nt;
+			}
+
+			/* Create session object, with INVALID_ENTRY,
+			 * the targer_id would get set when we issue the login
+			 */
+			cls_sess = iscsi_session_setup(&qla4xxx_iscsi_transport,
+						ha->host, cmds_max,
+						sizeof(struct ddb_entry),
+						sizeof(struct ql4_task_data),
+						initial_cmdsn, INVALID_ENTRY);
+			if (!cls_sess)
+				goto exit_ddb_list;
+
+			/*
+			 * iscsi_session_setup increments the driver reference
+			 * count which wouldn't let the driver to be unloaded.
+			 * so calling module_put function to decrement the
+			 * reference count.
+			 **/
+			module_put(qla4xxx_iscsi_transport.owner);
+			sess = cls_sess->dd_data;
+			ddb_entry = sess->dd_data;
+			ddb_entry->sess = cls_sess;
+
+			cls_sess->recovery_tmo = ql4xsess_recovery_tmo;
+			memcpy(&ddb_entry->fw_ddb_entry, fw_ddb_entry,
+			       sizeof(struct dev_db_entry));
+
+			qla4xxx_setup_flash_ddb_entry(ha, ddb_entry);
+
+			cls_conn = iscsi_conn_setup(cls_sess,
+						    sizeof(struct qla_conn),
+						    conn_id);
+			if (!cls_conn)
+				goto exit_ddb_list;
+
+			ddb_entry->conn = cls_conn;
+
+			/* Setup ep, for displaying attributes in sysfs */
+			ep = qla4xxx_get_ep_fwdb(ha, fw_ddb_entry);
+			if (ep) {
+				ep->conn = cls_conn;
+				cls_conn->ep = ep;
+			} else {
+				DEBUG2(ql4_printk(KERN_ERR, ha,
+						  "Unable to get ep\n"));
+			}
+
+			/* Update sess/conn params */
+			qla4xxx_copy_fwddb_param(ha, fw_ddb_entry, cls_sess,
+						 cls_conn);
+
+			if (is_reset == RESET_ADAPTER) {
+				iscsi_block_session(cls_sess);
+				/* Use the relogin path to discover new devices
+				 *  by short-circuting the logic of setting
+				 *  timer to relogin - instead set the flags
+				 *  to initiate login right away.
+				 */
+				set_bit(DPC_RELOGIN_DEVICE, &ha->dpc_flags);
+				set_bit(DF_RELOGIN, &ddb_entry->flags);
+			}
+		}
+continue_next_nt:
+		if (next_idx == 0)
+			break;
+	}
+exit_ddb_list:
+	qla4xxx_free_nt_list(&list_nt);
+	if (fw_ddb_entry)
+		dma_pool_free(ha->fw_ddb_dma_pool, fw_ddb_entry, fw_ddb_dma);
+
+	qla4xxx_free_ddb_index(ha);
+}
+
+
 /**
  * qla4xxx_probe_adapter - callback function to probe HBA
  * @pdev: pointer to pci_dev structure
@@ -3298,7 +4236,7 @@
 	 * firmware
 	 * NOTE: interrupts enabled upon successful completion
 	 */
-	status = qla4xxx_initialize_adapter(ha);
+	status = qla4xxx_initialize_adapter(ha, INIT_ADAPTER);
 	while ((!test_bit(AF_ONLINE, &ha->flags)) &&
 	    init_retry_count++ < MAX_INIT_RETRIES) {
 
@@ -3319,7 +4257,7 @@
 		if (ha->isp_ops->reset_chip(ha) == QLA_ERROR)
 			continue;
 
-		status = qla4xxx_initialize_adapter(ha);
+		status = qla4xxx_initialize_adapter(ha, INIT_ADAPTER);
 	}
 
 	if (!test_bit(AF_ONLINE, &ha->flags)) {
@@ -3386,12 +4324,16 @@
 	       ha->host_no, ha->firmware_version[0], ha->firmware_version[1],
 	       ha->patch_number, ha->build_number);
 
-	qla4xxx_create_chap_list(ha);
-
 	if (qla4xxx_setup_boot_info(ha))
 		ql4_printk(KERN_ERR, ha, "%s:ISCSI boot info setup failed\n",
 			   __func__);
 
+		/* Perform the build ddb list and login to each */
+	qla4xxx_build_ddb_list(ha, INIT_ADAPTER);
+	iscsi_host_for_each_session(ha->host, qla4xxx_login_flash_ddb);
+
+	qla4xxx_create_chap_list(ha);
+
 	qla4xxx_create_ifaces(ha);
 	return 0;
 
@@ -3449,6 +4391,38 @@
 	}
 }
 
+static void qla4xxx_destroy_fw_ddb_session(struct scsi_qla_host *ha)
+{
+	struct ddb_entry *ddb_entry;
+	int options;
+	int idx;
+
+	for (idx = 0; idx < MAX_DDB_ENTRIES; idx++) {
+
+		ddb_entry = qla4xxx_lookup_ddb_by_fw_index(ha, idx);
+		if ((ddb_entry != NULL) &&
+		    (ddb_entry->ddb_type == FLASH_DDB)) {
+
+			options = LOGOUT_OPTION_CLOSE_SESSION;
+			if (qla4xxx_session_logout_ddb(ha, ddb_entry, options)
+			    == QLA_ERROR)
+				ql4_printk(KERN_ERR, ha, "%s: Logout failed\n",
+					   __func__);
+
+			qla4xxx_clear_ddb_entry(ha, ddb_entry->fw_ddb_index);
+			/*
+			 * we have decremented the reference count of the driver
+			 * when we setup the session to have the driver unload
+			 * to be seamless without actually destroying the
+			 * session
+			 **/
+			try_module_get(qla4xxx_iscsi_transport.owner);
+			iscsi_destroy_endpoint(ddb_entry->conn->ep);
+			qla4xxx_free_ddb(ha, ddb_entry);
+			iscsi_session_teardown(ddb_entry->sess);
+		}
+	}
+}
 /**
  * qla4xxx_remove_adapter - calback function to remove adapter.
  * @pci_dev: PCI device pointer
@@ -3465,9 +4439,11 @@
 	/* destroy iface from sysfs */
 	qla4xxx_destroy_ifaces(ha);
 
-	if (ha->boot_kset)
+	if ((!ql4xdisablesysfsboot) && ha->boot_kset)
 		iscsi_boot_destroy_kset(ha->boot_kset);
 
+	qla4xxx_destroy_fw_ddb_session(ha);
+
 	scsi_remove_host(ha->host);
 
 	qla4xxx_free_adapter(ha);
@@ -4115,7 +5091,7 @@
 
 		qla4_8xxx_idc_unlock(ha);
 		clear_bit(AF_FW_RECOVERY, &ha->flags);
-		rval = qla4xxx_initialize_adapter(ha);
+		rval = qla4xxx_initialize_adapter(ha, RESET_ADAPTER);
 		qla4_8xxx_idc_lock(ha);
 
 		if (rval != QLA_SUCCESS) {
@@ -4151,7 +5127,7 @@
 		if ((qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE) ==
 		    QLA82XX_DEV_READY)) {
 			clear_bit(AF_FW_RECOVERY, &ha->flags);
-			rval = qla4xxx_initialize_adapter(ha);
+			rval = qla4xxx_initialize_adapter(ha, RESET_ADAPTER);
 			if (rval == QLA_SUCCESS) {
 				ret = qla4xxx_request_irqs(ha);
 				if (ret) {
diff --git a/drivers/scsi/qla4xxx/ql4_version.h b/drivers/scsi/qla4xxx/ql4_version.h
index c15347d..5254e57 100644
--- a/drivers/scsi/qla4xxx/ql4_version.h
+++ b/drivers/scsi/qla4xxx/ql4_version.h
@@ -5,4 +5,4 @@
  * See LICENSE.qla4xxx for copyright and licensing details.
  */
 
-#define QLA4XXX_DRIVER_VERSION	"5.02.00-k8"
+#define QLA4XXX_DRIVER_VERSION	"5.02.00-k9"
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 06bc2655..f85cfa6 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1409,6 +1409,8 @@
 
 	blk_start_request(req);
 
+	scmd_printk(KERN_INFO, cmd, "killing request\n");
+
 	sdev = cmd->device;
 	starget = scsi_target(sdev);
 	shost = sdev->host;
@@ -1490,7 +1492,6 @@
 	struct request *req;
 
 	if (!sdev) {
-		printk("scsi: killing requests for dead queue\n");
 		while ((req = blk_peek_request(q)) != NULL)
 			scsi_kill_request(req, q);
 		return;
diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index 72273a0..b3c6d95 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -319,11 +319,7 @@
 	return sdev;
 
 out_device_destroy:
-	scsi_device_set_state(sdev, SDEV_DEL);
-	transport_destroy_device(&sdev->sdev_gendev);
-	put_device(&sdev->sdev_dev);
-	scsi_free_queue(sdev->request_queue);
-	put_device(&sdev->sdev_gendev);
+	__scsi_remove_device(sdev);
 out:
 	if (display_failure_msg)
 		printk(ALLOC_FAILURE_MSG, __func__);
diff --git a/drivers/sh/Makefile b/drivers/sh/Makefile
index 24e6cec..67e272a 100644
--- a/drivers/sh/Makefile
+++ b/drivers/sh/Makefile
@@ -7,3 +7,11 @@
 obj-$(CONFIG_MAPLE)		+= maple/
 obj-$(CONFIG_SUPERHYWAY)	+= superhyway/
 obj-$(CONFIG_GENERIC_GPIO)	+= pfc.o
+
+#
+# For the moment we only use this framework for ARM-based SH/R-Mobile
+# platforms and generic SH. SH-based SH-Mobile platforms are still using
+# an older framework that is pending up-porting, at which point this
+# special casing can go away.
+#
+obj-$(CONFIG_SUPERH)$(CONFIG_ARCH_SHMOBILE)	+= pm_runtime.o
diff --git a/drivers/sh/clk/core.c b/drivers/sh/clk/core.c
index dc8d022..db257a3 100644
--- a/drivers/sh/clk/core.c
+++ b/drivers/sh/clk/core.c
@@ -25,7 +25,6 @@
 #include <linux/seq_file.h>
 #include <linux/err.h>
 #include <linux/io.h>
-#include <linux/debugfs.h>
 #include <linux/cpufreq.h>
 #include <linux/clk.h>
 #include <linux/sh_clk.h>
@@ -173,6 +172,26 @@
 	return clk_rate_round_helper(&div_range_round);
 }
 
+static long clk_rate_mult_range_iter(unsigned int pos,
+				      struct clk_rate_round_data *rounder)
+{
+	return clk_get_rate(rounder->arg) * pos;
+}
+
+long clk_rate_mult_range_round(struct clk *clk, unsigned int mult_min,
+			       unsigned int mult_max, unsigned long rate)
+{
+	struct clk_rate_round_data mult_range_round = {
+		.min	= mult_min,
+		.max	= mult_max,
+		.func	= clk_rate_mult_range_iter,
+		.arg	= clk_get_parent(clk),
+		.rate	= rate,
+	};
+
+	return clk_rate_round_helper(&mult_range_round);
+}
+
 int clk_rate_table_find(struct clk *clk,
 			struct cpufreq_frequency_table *freq_table,
 			unsigned long rate)
@@ -205,9 +224,6 @@
 		list_add(&child->sibling, &parent->children);
 	child->parent = parent;
 
-	/* now do the debugfs renaming to reattach the child
-	   to the proper parent */
-
 	return 0;
 }
 
@@ -665,89 +681,6 @@
 subsys_initcall(clk_syscore_init);
 #endif
 
-/*
- *	debugfs support to trace clock tree hierarchy and attributes
- */
-static struct dentry *clk_debugfs_root;
-
-static int clk_debugfs_register_one(struct clk *c)
-{
-	int err;
-	struct dentry *d;
-	struct clk *pa = c->parent;
-	char s[255];
-	char *p = s;
-
-	p += sprintf(p, "%p", c);
-	d = debugfs_create_dir(s, pa ? pa->dentry : clk_debugfs_root);
-	if (!d)
-		return -ENOMEM;
-	c->dentry = d;
-
-	d = debugfs_create_u8("usecount", S_IRUGO, c->dentry, (u8 *)&c->usecount);
-	if (!d) {
-		err = -ENOMEM;
-		goto err_out;
-	}
-	d = debugfs_create_u32("rate", S_IRUGO, c->dentry, (u32 *)&c->rate);
-	if (!d) {
-		err = -ENOMEM;
-		goto err_out;
-	}
-	d = debugfs_create_x32("flags", S_IRUGO, c->dentry, (u32 *)&c->flags);
-	if (!d) {
-		err = -ENOMEM;
-		goto err_out;
-	}
-	return 0;
-
-err_out:
-	debugfs_remove_recursive(c->dentry);
-	return err;
-}
-
-static int clk_debugfs_register(struct clk *c)
-{
-	int err;
-	struct clk *pa = c->parent;
-
-	if (pa && !pa->dentry) {
-		err = clk_debugfs_register(pa);
-		if (err)
-			return err;
-	}
-
-	if (!c->dentry) {
-		err = clk_debugfs_register_one(c);
-		if (err)
-			return err;
-	}
-	return 0;
-}
-
-static int __init clk_debugfs_init(void)
-{
-	struct clk *c;
-	struct dentry *d;
-	int err;
-
-	d = debugfs_create_dir("clock", NULL);
-	if (!d)
-		return -ENOMEM;
-	clk_debugfs_root = d;
-
-	list_for_each_entry(c, &clock_list, node) {
-		err = clk_debugfs_register(c);
-		if (err)
-			goto err_out;
-	}
-	return 0;
-err_out:
-	debugfs_remove_recursive(clk_debugfs_root);
-	return err;
-}
-late_initcall(clk_debugfs_init);
-
 static int __init clk_late_init(void)
 {
 	unsigned long flags;
diff --git a/arch/arm/mach-shmobile/pm_runtime.c b/drivers/sh/pm_runtime.c
similarity index 93%
rename from arch/arm/mach-shmobile/pm_runtime.c
rename to drivers/sh/pm_runtime.c
index bd5c6a3..afe9282 100644
--- a/arch/arm/mach-shmobile/pm_runtime.c
+++ b/drivers/sh/pm_runtime.c
@@ -1,7 +1,5 @@
 /*
- * arch/arm/mach-shmobile/pm_runtime.c
- *
- * Runtime PM support code for SuperH Mobile ARM
+ * Runtime PM support code
  *
  *  Copyright (C) 2009-2010 Magnus Damm
  *
diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index a1fd73d..8ba4510 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -199,7 +199,7 @@
 	depends on FSL_SOC
 
 config SPI_FSL_SPI
-	tristate "Freescale SPI controller"
+	bool "Freescale SPI controller"
 	depends on FSL_SOC
 	select SPI_FSL_LIB
 	help
@@ -208,7 +208,7 @@
 	  MPC8569 uses the controller in QE mode, MPC8610 in cpu mode.
 
 config SPI_FSL_ESPI
-	tristate "Freescale eSPI controller"
+	bool "Freescale eSPI controller"
 	depends on FSL_SOC
 	select SPI_FSL_LIB
 	help
diff --git a/drivers/spi/spi-ath79.c b/drivers/spi/spi-ath79.c
index 024b48a..acc88b4 100644
--- a/drivers/spi/spi-ath79.c
+++ b/drivers/spi/spi-ath79.c
@@ -13,6 +13,7 @@
  */
 
 #include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/spinlock.h>
diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c
index 79665e2..16d6a83 100644
--- a/drivers/spi/spi-atmel.c
+++ b/drivers/spi/spi-atmel.c
@@ -907,7 +907,7 @@
 
 /*-------------------------------------------------------------------------*/
 
-static int __init atmel_spi_probe(struct platform_device *pdev)
+static int __devinit atmel_spi_probe(struct platform_device *pdev)
 {
 	struct resource		*regs;
 	int			irq;
@@ -1003,7 +1003,7 @@
 	return ret;
 }
 
-static int __exit atmel_spi_remove(struct platform_device *pdev)
+static int __devexit atmel_spi_remove(struct platform_device *pdev)
 {
 	struct spi_master	*master = platform_get_drvdata(pdev);
 	struct atmel_spi	*as = spi_master_get_devdata(master);
@@ -1072,6 +1072,7 @@
 	},
 	.suspend	= atmel_spi_suspend,
 	.resume		= atmel_spi_resume,
+	.probe		= atmel_spi_probe,
 	.remove		= __exit_p(atmel_spi_remove),
 };
 module_platform_driver(atmel_spi_driver);
diff --git a/drivers/spi/spi-gpio.c b/drivers/spi/spi-gpio.c
index e093d3e..0094c645 100644
--- a/drivers/spi/spi-gpio.c
+++ b/drivers/spi/spi-gpio.c
@@ -256,7 +256,7 @@
 	spi_bitbang_cleanup(spi);
 }
 
-static int __init spi_gpio_alloc(unsigned pin, const char *label, bool is_in)
+static int __devinit spi_gpio_alloc(unsigned pin, const char *label, bool is_in)
 {
 	int value;
 
@@ -270,7 +270,7 @@
 	return value;
 }
 
-static int __init
+static int __devinit
 spi_gpio_request(struct spi_gpio_platform_data *pdata, const char *label,
 	u16 *res_flags)
 {
diff --git a/drivers/spi/spi-nuc900.c b/drivers/spi/spi-nuc900.c
index e763254..182e9c8 100644
--- a/drivers/spi/spi-nuc900.c
+++ b/drivers/spi/spi-nuc900.c
@@ -8,6 +8,7 @@
  *
  */
 
+#include <linux/module.h>
 #include <linux/init.h>
 #include <linux/spinlock.h>
 #include <linux/workqueue.h>
@@ -426,7 +427,7 @@
 		goto err_clk;
 	}
 
-	mfp_set_groupg(&pdev->dev);
+	mfp_set_groupg(&pdev->dev, NULL);
 	nuc900_init_spi(hw);
 
 	err = spi_bitbang_start(&hw->bitbang);
diff --git a/drivers/spi/spi-pl022.c b/drivers/spi/spi-pl022.c
index f103e47..5559b22 100644
--- a/drivers/spi/spi-pl022.c
+++ b/drivers/spi/spi-pl022.c
@@ -2184,6 +2184,12 @@
 		goto  err_clk_prep;
 	}
 
+	status = clk_enable(pl022->clk);
+	if (status) {
+		dev_err(&adev->dev, "could not enable SSP/SPI bus clock\n");
+		goto err_no_clk_en;
+	}
+
 	/* Disable SSP */
 	writew((readw(SSP_CR1(pl022->virtbase)) & (~SSP_CR1_MASK_SSE)),
 	       SSP_CR1(pl022->virtbase));
@@ -2237,6 +2243,8 @@
 
 	free_irq(adev->irq[0], pl022);
  err_no_irq:
+	clk_disable(pl022->clk);
+ err_no_clk_en:
 	clk_unprepare(pl022->clk);
  err_clk_prep:
 	clk_put(pl022->clk);
diff --git a/drivers/ssb/driver_pcicore.c b/drivers/ssb/driver_pcicore.c
index 84c934c..520e828 100644
--- a/drivers/ssb/driver_pcicore.c
+++ b/drivers/ssb/driver_pcicore.c
@@ -517,10 +517,14 @@
 
 static void __devinit ssb_pcicore_init_clientmode(struct ssb_pcicore *pc)
 {
-	ssb_pcicore_fix_sprom_core_index(pc);
+	struct ssb_device *pdev = pc->dev;
+	struct ssb_bus *bus = pdev->bus;
+
+	if (bus->bustype == SSB_BUSTYPE_PCI)
+		ssb_pcicore_fix_sprom_core_index(pc);
 
 	/* Disable PCI interrupts. */
-	ssb_write32(pc->dev, SSB_INTVEC, 0);
+	ssb_write32(pdev, SSB_INTVEC, 0);
 
 	/* Additional PCIe always once-executed workarounds */
 	if (pc->dev->id.coreid == SSB_DEV_PCIE) {
diff --git a/drivers/staging/comedi/comedi_fops.c b/drivers/staging/comedi/comedi_fops.c
index 21d8c1c..5e78c77 100644
--- a/drivers/staging/comedi/comedi_fops.c
+++ b/drivers/staging/comedi/comedi_fops.c
@@ -671,7 +671,7 @@
 	}
 
 	insns =
-	    kmalloc(sizeof(struct comedi_insn) * insnlist.n_insns, GFP_KERNEL);
+	    kcalloc(insnlist.n_insns, sizeof(struct comedi_insn), GFP_KERNEL);
 	if (!insns) {
 		DPRINTK("kmalloc failed\n");
 		ret = -ENOMEM;
@@ -1432,7 +1432,21 @@
 	return ret;
 }
 
-static void comedi_unmap(struct vm_area_struct *area)
+
+static void comedi_vm_open(struct vm_area_struct *area)
+{
+	struct comedi_async *async;
+	struct comedi_device *dev;
+
+	async = area->vm_private_data;
+	dev = async->subdevice->device;
+
+	mutex_lock(&dev->mutex);
+	async->mmap_count++;
+	mutex_unlock(&dev->mutex);
+}
+
+static void comedi_vm_close(struct vm_area_struct *area)
 {
 	struct comedi_async *async;
 	struct comedi_device *dev;
@@ -1446,15 +1460,13 @@
 }
 
 static struct vm_operations_struct comedi_vm_ops = {
-	.close = comedi_unmap,
+	.open = comedi_vm_open,
+	.close = comedi_vm_close,
 };
 
 static int comedi_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	const unsigned minor = iminor(file->f_dentry->d_inode);
-	struct comedi_device_file_info *dev_file_info =
-	    comedi_get_device_file_info(minor);
-	struct comedi_device *dev = dev_file_info->device;
 	struct comedi_async *async = NULL;
 	unsigned long start = vma->vm_start;
 	unsigned long size;
@@ -1462,6 +1474,15 @@
 	int i;
 	int retval;
 	struct comedi_subdevice *s;
+	struct comedi_device_file_info *dev_file_info;
+	struct comedi_device *dev;
+
+	dev_file_info = comedi_get_device_file_info(minor);
+	if (dev_file_info == NULL)
+	        return -ENODEV;
+	dev = dev_file_info->device;
+	if (dev == NULL)
+	        return -ENODEV;
 
 	mutex_lock(&dev->mutex);
 	if (!dev->attached) {
@@ -1528,11 +1549,17 @@
 {
 	unsigned int mask = 0;
 	const unsigned minor = iminor(file->f_dentry->d_inode);
-	struct comedi_device_file_info *dev_file_info =
-	    comedi_get_device_file_info(minor);
-	struct comedi_device *dev = dev_file_info->device;
 	struct comedi_subdevice *read_subdev;
 	struct comedi_subdevice *write_subdev;
+	struct comedi_device_file_info *dev_file_info;
+	struct comedi_device *dev;
+	dev_file_info = comedi_get_device_file_info(minor);
+
+	if (dev_file_info == NULL)
+	        return -ENODEV;
+	dev = dev_file_info->device;
+	if (dev == NULL)
+	        return -ENODEV;
 
 	mutex_lock(&dev->mutex);
 	if (!dev->attached) {
@@ -1578,9 +1605,15 @@
 	int n, m, count = 0, retval = 0;
 	DECLARE_WAITQUEUE(wait, current);
 	const unsigned minor = iminor(file->f_dentry->d_inode);
-	struct comedi_device_file_info *dev_file_info =
-	    comedi_get_device_file_info(minor);
-	struct comedi_device *dev = dev_file_info->device;
+	struct comedi_device_file_info *dev_file_info;
+	struct comedi_device *dev;
+	dev_file_info = comedi_get_device_file_info(minor);
+
+	if (dev_file_info == NULL)
+	        return -ENODEV;
+	dev = dev_file_info->device;
+	if (dev == NULL)
+	        return -ENODEV;
 
 	if (!dev->attached) {
 		DPRINTK("no driver configured on comedi%i\n", dev->minor);
@@ -1640,11 +1673,11 @@
 				retval = -EAGAIN;
 				break;
 			}
+			schedule();
 			if (signal_pending(current)) {
 				retval = -ERESTARTSYS;
 				break;
 			}
-			schedule();
 			if (!s->busy)
 				break;
 			if (s->busy != file) {
@@ -1683,9 +1716,15 @@
 	int n, m, count = 0, retval = 0;
 	DECLARE_WAITQUEUE(wait, current);
 	const unsigned minor = iminor(file->f_dentry->d_inode);
-	struct comedi_device_file_info *dev_file_info =
-	    comedi_get_device_file_info(minor);
-	struct comedi_device *dev = dev_file_info->device;
+	struct comedi_device_file_info *dev_file_info;
+	struct comedi_device *dev;
+	dev_file_info = comedi_get_device_file_info(minor);
+
+	if (dev_file_info == NULL)
+	        return -ENODEV;
+	dev = dev_file_info->device;
+	if (dev == NULL)
+	        return -ENODEV;
 
 	if (!dev->attached) {
 		DPRINTK("no driver configured on comedi%i\n", dev->minor);
@@ -1741,11 +1780,11 @@
 				retval = -EAGAIN;
 				break;
 			}
+			schedule();
 			if (signal_pending(current)) {
 				retval = -ERESTARTSYS;
 				break;
 			}
-			schedule();
 			if (!s->busy) {
 				retval = 0;
 				break;
@@ -1885,11 +1924,17 @@
 static int comedi_close(struct inode *inode, struct file *file)
 {
 	const unsigned minor = iminor(inode);
-	struct comedi_device_file_info *dev_file_info =
-	    comedi_get_device_file_info(minor);
-	struct comedi_device *dev = dev_file_info->device;
 	struct comedi_subdevice *s = NULL;
 	int i;
+	struct comedi_device_file_info *dev_file_info;
+	struct comedi_device *dev;
+	dev_file_info = comedi_get_device_file_info(minor);
+
+	if (dev_file_info == NULL)
+	        return -ENODEV;
+	dev = dev_file_info->device;
+	if (dev == NULL)
+	        return -ENODEV;
 
 	mutex_lock(&dev->mutex);
 
@@ -1923,10 +1968,15 @@
 static int comedi_fasync(int fd, struct file *file, int on)
 {
 	const unsigned minor = iminor(file->f_dentry->d_inode);
-	struct comedi_device_file_info *dev_file_info =
-	    comedi_get_device_file_info(minor);
+	struct comedi_device_file_info *dev_file_info;
+	struct comedi_device *dev;
+	dev_file_info = comedi_get_device_file_info(minor);
 
-	struct comedi_device *dev = dev_file_info->device;
+	if (dev_file_info == NULL)
+	        return -ENODEV;
+	dev = dev_file_info->device;
+	if (dev == NULL)
+	        return -ENODEV;
 
 	return fasync_helper(fd, file, on, &dev->async_queue);
 }
diff --git a/drivers/staging/comedi/drivers/usbduxsigma.c b/drivers/staging/comedi/drivers/usbduxsigma.c
index a8fea9a91..6144afb 100644
--- a/drivers/staging/comedi/drivers/usbduxsigma.c
+++ b/drivers/staging/comedi/drivers/usbduxsigma.c
@@ -1,4 +1,4 @@
-#define DRIVER_VERSION "v0.5"
+#define DRIVER_VERSION "v0.6"
 #define DRIVER_AUTHOR "Bernd Porr, BerndPorr@f2s.com"
 #define DRIVER_DESC "Stirling/ITL USB-DUX SIGMA -- Bernd.Porr@f2s.com"
 /*
@@ -25,7 +25,7 @@
 Description: University of Stirling USB DAQ & INCITE Technology Limited
 Devices: [ITL] USB-DUX (usbduxsigma.o)
 Author: Bernd Porr <BerndPorr@f2s.com>
-Updated: 21 Jul 2011
+Updated: 8 Nov 2011
 Status: testing
 */
 /*
@@ -44,6 +44,7 @@
  *   0.3: proper vendor ID and driver name
  *   0.4: fixed D/A voltage range
  *   0.5: various bug fixes, health check at startup
+ *   0.6: corrected wrong input range
  */
 
 /* generates loads of debug info */
@@ -175,7 +176,7 @@
 /* comedi constants */
 static const struct comedi_lrange range_usbdux_ai_range = { 1, {
 								BIP_RANGE
-								(2.65)
+								(2.65/2.0)
 								}
 };
 
diff --git a/drivers/staging/et131x/Kconfig b/drivers/staging/et131x/Kconfig
index 9e1864c..8190f2a 100644
--- a/drivers/staging/et131x/Kconfig
+++ b/drivers/staging/et131x/Kconfig
@@ -1,6 +1,7 @@
 config ET131X
 	tristate "Agere ET-1310 Gigabit Ethernet support"
-	depends on PCI
+	depends on PCI && NET && NETDEVICES
+	select PHYLIB
 	default n
 	---help---
 	  This driver supports Agere ET-1310 ethernet adapters.
diff --git a/drivers/staging/et131x/et131x.c b/drivers/staging/et131x/et131x.c
index f5f44a0..0c1c6ca 100644
--- a/drivers/staging/et131x/et131x.c
+++ b/drivers/staging/et131x/et131x.c
@@ -4469,6 +4469,12 @@
 	return 0;
 }
 
+static SIMPLE_DEV_PM_OPS(et131x_pm_ops, et131x_suspend, et131x_resume);
+#define ET131X_PM_OPS (&et131x_pm_ops)
+#else
+#define ET131X_PM_OPS NULL
+#endif
+
 /* ISR functions */
 
 /**
@@ -5470,12 +5476,6 @@
 	return result;
 }
 
-static SIMPLE_DEV_PM_OPS(et131x_pm_ops, et131x_suspend, et131x_resume);
-#define ET131X_PM_OPS (&et131x_pm_ops)
-#else
-#define ET131X_PM_OPS NULL
-#endif
-
 static DEFINE_PCI_DEVICE_TABLE(et131x_pci_table) = {
 	{ PCI_VDEVICE(ATT, ET131X_PCI_DEVICE_ID_GIG), 0UL},
 	{ PCI_VDEVICE(ATT, ET131X_PCI_DEVICE_ID_FAST), 0UL},
diff --git a/drivers/staging/gma500/accel_2d.c b/drivers/staging/gma500/accel_2d.c
index 114b99a..b8f78eb 100644
--- a/drivers/staging/gma500/accel_2d.c
+++ b/drivers/staging/gma500/accel_2d.c
@@ -253,7 +253,7 @@
 		return;
 
 	offset = psbfb->gtt->offset;
-	stride = fb->pitch;
+	stride = fb->pitches[0];
 
 	switch (fb->depth) {
 	case 8:
diff --git a/drivers/staging/gma500/cdv_intel_display.c b/drivers/staging/gma500/cdv_intel_display.c
index 7b97c60..c63a327 100644
--- a/drivers/staging/gma500/cdv_intel_display.c
+++ b/drivers/staging/gma500/cdv_intel_display.c
@@ -507,9 +507,9 @@
 	if (ret < 0)
 		goto psb_intel_pipe_set_base_exit;
 	start = psbfb->gtt->offset;
-	offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8);
+	offset = y * crtc->fb->pitches[0] + x * (crtc->fb->bits_per_pixel / 8);
 
-	REG_WRITE(dspstride, crtc->fb->pitch);
+	REG_WRITE(dspstride, crtc->fb->pitches[0]);
 
 	dspcntr = REG_READ(dspcntr_reg);
 	dspcntr &= ~DISPPLANE_PIXFORMAT_MASK;
diff --git a/drivers/staging/gma500/framebuffer.c b/drivers/staging/gma500/framebuffer.c
index 7aee4d2..b00761c 100644
--- a/drivers/staging/gma500/framebuffer.c
+++ b/drivers/staging/gma500/framebuffer.c
@@ -511,7 +511,7 @@
 		info->apertures->ranges[0].size = dev_priv->gtt.stolen_size;
 	}
 
-	drm_fb_helper_fill_fix(info, fb->pitch, fb->depth);
+	drm_fb_helper_fill_fix(info, fb->pitches[0], fb->depth);
 	drm_fb_helper_fill_var(info, &fbdev->psb_fb_helper,
 				sizes->fb_width, sizes->fb_height);
 
diff --git a/drivers/staging/gma500/mdfld_intel_display.c b/drivers/staging/gma500/mdfld_intel_display.c
index 8eb827e..0b37b7b 100644
--- a/drivers/staging/gma500/mdfld_intel_display.c
+++ b/drivers/staging/gma500/mdfld_intel_display.c
@@ -390,9 +390,9 @@
 	        goto psb_intel_pipe_set_base_exit;
 
 	start = psbfb->gtt->offset;
-	offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8);
+	offset = y * crtc->fb->pitches[0] + x * (crtc->fb->bits_per_pixel / 8);
 
-	REG_WRITE(dspstride, crtc->fb->pitch);
+	REG_WRITE(dspstride, crtc->fb->pitches[0]);
 	dspcntr = REG_READ(dspcntr_reg);
 	dspcntr &= ~DISPPLANE_PIXFORMAT_MASK;
 
diff --git a/drivers/staging/gma500/mrst_crtc.c b/drivers/staging/gma500/mrst_crtc.c
index c9311a5..980837e 100644
--- a/drivers/staging/gma500/mrst_crtc.c
+++ b/drivers/staging/gma500/mrst_crtc.c
@@ -543,9 +543,9 @@
 		return 0;
 
 	start = psbfb->gtt->offset;
-	offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8);
+	offset = y * crtc->fb->pitches[0] + x * (crtc->fb->bits_per_pixel / 8);
 
-	REG_WRITE(dspstride, crtc->fb->pitch);
+	REG_WRITE(dspstride, crtc->fb->pitches[0]);
 
 	dspcntr = REG_READ(dspcntr_reg);
 	dspcntr &= ~DISPPLANE_PIXFORMAT_MASK;
diff --git a/drivers/staging/gma500/power.c b/drivers/staging/gma500/power.c
index 436fe97..4082570 100644
--- a/drivers/staging/gma500/power.c
+++ b/drivers/staging/gma500/power.c
@@ -266,7 +266,7 @@
 	ret = gma_resume_pci(dev->pdev);
 	if (ret == 0) {
 		/* FIXME: we want to defer this for Medfield/Oaktrail */
-		gma_resume_display(dev);
+		gma_resume_display(dev->pdev);
 		psb_irq_preinstall(dev);
 		psb_irq_postinstall(dev);
 		pm_runtime_get(&dev->pdev->dev);
diff --git a/drivers/staging/gma500/psb_intel_display.c b/drivers/staging/gma500/psb_intel_display.c
index caa9d86..8565961 100644
--- a/drivers/staging/gma500/psb_intel_display.c
+++ b/drivers/staging/gma500/psb_intel_display.c
@@ -367,9 +367,9 @@
 		goto psb_intel_pipe_set_base_exit;
 	start = psbfb->gtt->offset;
 
-	offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8);
+	offset = y * crtc->fb->pitches[0] + x * (crtc->fb->bits_per_pixel / 8);
 
-	REG_WRITE(dspstride, crtc->fb->pitch);
+	REG_WRITE(dspstride, crtc->fb->pitches[0]);
 
 	dspcntr = REG_READ(dspcntr_reg);
 	dspcntr &= ~DISPPLANE_PIXFORMAT_MASK;
diff --git a/drivers/staging/iio/industrialio-core.c b/drivers/staging/iio/industrialio-core.c
index 326e967..aec9311 100644
--- a/drivers/staging/iio/industrialio-core.c
+++ b/drivers/staging/iio/industrialio-core.c
@@ -242,19 +242,26 @@
 
 static int iio_event_getfd(struct iio_dev *indio_dev)
 {
-	if (indio_dev->event_interface == NULL)
+	struct iio_event_interface *ev_int = indio_dev->event_interface;
+	int fd;
+
+	if (ev_int == NULL)
 		return -ENODEV;
 
-	mutex_lock(&indio_dev->event_interface->event_list_lock);
-	if (test_and_set_bit(IIO_BUSY_BIT_POS,
-			     &indio_dev->event_interface->flags)) {
-		mutex_unlock(&indio_dev->event_interface->event_list_lock);
+	mutex_lock(&ev_int->event_list_lock);
+	if (test_and_set_bit(IIO_BUSY_BIT_POS, &ev_int->flags)) {
+		mutex_unlock(&ev_int->event_list_lock);
 		return -EBUSY;
 	}
-	mutex_unlock(&indio_dev->event_interface->event_list_lock);
-	return anon_inode_getfd("iio:event",
-				&iio_event_chrdev_fileops,
-				indio_dev->event_interface, O_RDONLY);
+	mutex_unlock(&ev_int->event_list_lock);
+	fd = anon_inode_getfd("iio:event",
+				&iio_event_chrdev_fileops, ev_int, O_RDONLY);
+	if (fd < 0) {
+		mutex_lock(&ev_int->event_list_lock);
+		clear_bit(IIO_BUSY_BIT_POS, &ev_int->flags);
+		mutex_unlock(&ev_int->event_list_lock);
+	}
+	return fd;
 }
 
 static int __init iio_init(void)
diff --git a/drivers/staging/media/as102/as102_drv.c b/drivers/staging/media/as102/as102_drv.c
index d335c7d..828526d 100644
--- a/drivers/staging/media/as102/as102_drv.c
+++ b/drivers/staging/media/as102/as102_drv.c
@@ -32,8 +32,8 @@
 #include "as102_fw.h"
 #include "dvbdev.h"
 
-int debug;
-module_param_named(debug, debug, int, 0644);
+int as102_debug;
+module_param_named(debug, as102_debug, int, 0644);
 MODULE_PARM_DESC(debug, "Turn on/off debugging (default: off)");
 
 int dual_tuner;
diff --git a/drivers/staging/media/as102/as102_drv.h b/drivers/staging/media/as102/as102_drv.h
index bcda635..fd33f5a 100644
--- a/drivers/staging/media/as102/as102_drv.h
+++ b/drivers/staging/media/as102/as102_drv.h
@@ -37,7 +37,8 @@
 #define DRIVER_FULL_NAME "Abilis Systems as10x usb driver"
 #define DRIVER_NAME "as10x_usb"
 
-extern int debug;
+extern int as102_debug;
+#define debug	as102_debug
 
 #define dprintk(debug, args...) \
 	do { if (debug) {	\
diff --git a/drivers/staging/octeon/ethernet-tx.c b/drivers/staging/octeon/ethernet-tx.c
index b445cd6..2542c37 100644
--- a/drivers/staging/octeon/ethernet-tx.c
+++ b/drivers/staging/octeon/ethernet-tx.c
@@ -275,7 +275,7 @@
 		CVM_OCT_SKB_CB(skb)[0] = hw_buffer.u64;
 		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 			struct skb_frag_struct *fs = skb_shinfo(skb)->frags + i;
-			hw_buffer.s.addr = XKPHYS_TO_PHYS((u64)(page_address(fs->page) + fs->page_offset));
+			hw_buffer.s.addr = XKPHYS_TO_PHYS((u64)(page_address(fs->page.p) + fs->page_offset));
 			hw_buffer.s.size = fs->size;
 			CVM_OCT_SKB_CB(skb)[i + 1] = hw_buffer.u64;
 		}
diff --git a/drivers/staging/rtl8712/usb_intf.c b/drivers/staging/rtl8712/usb_intf.c
index fb2e89c..5385da2 100644
--- a/drivers/staging/rtl8712/usb_intf.c
+++ b/drivers/staging/rtl8712/usb_intf.c
@@ -89,6 +89,7 @@
 	{USB_DEVICE(0x0DF6, 0x0045)},
 	{USB_DEVICE(0x0DF6, 0x0059)}, /* 11n mode disable */
 	{USB_DEVICE(0x0DF6, 0x004B)},
+	{USB_DEVICE(0x0DF6, 0x005D)},
 	{USB_DEVICE(0x0DF6, 0x0063)},
 	/* Sweex */
 	{USB_DEVICE(0x177F, 0x0154)},
diff --git a/drivers/staging/rts_pstor/rtsx.c b/drivers/staging/rts_pstor/rtsx.c
index 480b0ed..115635f 100644
--- a/drivers/staging/rts_pstor/rtsx.c
+++ b/drivers/staging/rts_pstor/rtsx.c
@@ -1021,6 +1021,7 @@
 	th = kthread_create(rtsx_scan_thread, dev, "rtsx-scan");
 	if (IS_ERR(th)) {
 		printk(KERN_ERR "Unable to start the device-scanning thread\n");
+		complete(&dev->scanning_done);
 		quiesce_and_remove_host(dev);
 		err = PTR_ERR(th);
 		goto errout;
diff --git a/drivers/staging/slicoss/Kconfig b/drivers/staging/slicoss/Kconfig
index 5cde96b..5c2a15b 100644
--- a/drivers/staging/slicoss/Kconfig
+++ b/drivers/staging/slicoss/Kconfig
@@ -1,6 +1,6 @@
 config SLICOSS
 	tristate "Alacritech Gigabit IS-NIC support"
-	depends on PCI && X86
+	depends on PCI && X86 && NET
 	default n
 	help
 	  This driver supports Alacritech's IS-NIC gigabit ethernet cards.
diff --git a/drivers/staging/tidspbridge/core/dsp-clock.c b/drivers/staging/tidspbridge/core/dsp-clock.c
index 3d1279c..7eb5617 100644
--- a/drivers/staging/tidspbridge/core/dsp-clock.c
+++ b/drivers/staging/tidspbridge/core/dsp-clock.c
@@ -54,6 +54,7 @@
 
 /* Bridge GPT id (1 - 4), DM Timer id (5 - 8) */
 #define DMT_ID(id) ((id) + 4)
+#define DM_TIMER_CLOCKS		4
 
 /* Bridge MCBSP id (6 - 10), OMAP Mcbsp id (0 - 4) */
 #define MCBSP_ID(id) ((id) - 6)
@@ -114,8 +115,13 @@
  */
 void dsp_clk_exit(void)
 {
+	int i;
+
 	dsp_clock_disable_all(dsp_clocks);
 
+	for (i = 0; i < DM_TIMER_CLOCKS; i++)
+		omap_dm_timer_free(timer[i]);
+
 	clk_put(iva2_clk);
 	clk_put(ssi.sst_fck);
 	clk_put(ssi.ssr_fck);
@@ -130,9 +136,13 @@
 void dsp_clk_init(void)
 {
 	static struct platform_device dspbridge_device;
+	int i, id;
 
 	dspbridge_device.dev.bus = &platform_bus_type;
 
+	for (i = 0, id = 5; i < DM_TIMER_CLOCKS; i++, id++)
+		timer[i] = omap_dm_timer_request_specific(id);
+
 	iva2_clk = clk_get(&dspbridge_device.dev, "iva2_ck");
 	if (IS_ERR(iva2_clk))
 		dev_err(bridge, "failed to get iva2 clock %p\n", iva2_clk);
@@ -204,8 +214,7 @@
 		clk_enable(iva2_clk);
 		break;
 	case GPT_CLK:
-		timer[clk_id - 1] =
-				omap_dm_timer_request_specific(DMT_ID(clk_id));
+		status = omap_dm_timer_start(timer[clk_id - 1]);
 		break;
 #ifdef CONFIG_OMAP_MCBSP
 	case MCBSP_CLK:
@@ -281,7 +290,7 @@
 		clk_disable(iva2_clk);
 		break;
 	case GPT_CLK:
-		omap_dm_timer_free(timer[clk_id - 1]);
+		status = omap_dm_timer_stop(timer[clk_id - 1]);
 		break;
 #ifdef CONFIG_OMAP_MCBSP
 	case MCBSP_CLK:
diff --git a/drivers/staging/tidspbridge/rmgr/drv_interface.c b/drivers/staging/tidspbridge/rmgr/drv_interface.c
index c43c7e3..76cfc6e 100644
--- a/drivers/staging/tidspbridge/rmgr/drv_interface.c
+++ b/drivers/staging/tidspbridge/rmgr/drv_interface.c
@@ -24,11 +24,7 @@
 #include <linux/types.h>
 #include <linux/platform_device.h>
 #include <linux/pm.h>
-
-#ifdef MODULE
 #include <linux/module.h>
-#endif
-
 #include <linux/device.h>
 #include <linux/init.h>
 #include <linux/moduleparam.h>
diff --git a/drivers/staging/usbip/vhci_rx.c b/drivers/staging/usbip/vhci_rx.c
index 09c44ab..3872b8c 100644
--- a/drivers/staging/usbip/vhci_rx.c
+++ b/drivers/staging/usbip/vhci_rx.c
@@ -68,6 +68,7 @@
 {
 	struct usbip_device *ud = &vdev->ud;
 	struct urb *urb;
+	unsigned long flags;
 
 	spin_lock(&vdev->priv_lock);
 	urb = pickup_urb_and_free_priv(vdev, pdu->base.seqnum);
@@ -101,9 +102,9 @@
 
 	usbip_dbg_vhci_rx("now giveback urb %p\n", urb);
 
-	spin_lock(&the_controller->lock);
+	spin_lock_irqsave(&the_controller->lock, flags);
 	usb_hcd_unlink_urb_from_ep(vhci_to_hcd(the_controller), urb);
-	spin_unlock(&the_controller->lock);
+	spin_unlock_irqrestore(&the_controller->lock, flags);
 
 	usb_hcd_giveback_urb(vhci_to_hcd(the_controller), urb, urb->status);
 
@@ -141,6 +142,7 @@
 {
 	struct vhci_unlink *unlink;
 	struct urb *urb;
+	unsigned long flags;
 
 	usbip_dump_header(pdu);
 
@@ -170,9 +172,9 @@
 		urb->status = pdu->u.ret_unlink.status;
 		pr_info("urb->status %d\n", urb->status);
 
-		spin_lock(&the_controller->lock);
+		spin_lock_irqsave(&the_controller->lock, flags);
 		usb_hcd_unlink_urb_from_ep(vhci_to_hcd(the_controller), urb);
-		spin_unlock(&the_controller->lock);
+		spin_unlock_irqrestore(&the_controller->lock, flags);
 
 		usb_hcd_giveback_urb(vhci_to_hcd(the_controller), urb,
 				     urb->status);
diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index 0fd96c1..8599545 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -614,13 +614,12 @@
 	hdr	= (struct iscsi_reject *) cmd->pdu;
 	hdr->reason = reason;
 
-	cmd->buf_ptr = kzalloc(ISCSI_HDR_LEN, GFP_KERNEL);
+	cmd->buf_ptr = kmemdup(buf, ISCSI_HDR_LEN, GFP_KERNEL);
 	if (!cmd->buf_ptr) {
 		pr_err("Unable to allocate memory for cmd->buf_ptr\n");
 		iscsit_release_cmd(cmd);
 		return -1;
 	}
-	memcpy(cmd->buf_ptr, buf, ISCSI_HDR_LEN);
 
 	spin_lock_bh(&conn->cmd_lock);
 	list_add_tail(&cmd->i_list, &conn->conn_cmd_list);
@@ -661,13 +660,12 @@
 	hdr	= (struct iscsi_reject *) cmd->pdu;
 	hdr->reason = reason;
 
-	cmd->buf_ptr = kzalloc(ISCSI_HDR_LEN, GFP_KERNEL);
+	cmd->buf_ptr = kmemdup(buf, ISCSI_HDR_LEN, GFP_KERNEL);
 	if (!cmd->buf_ptr) {
 		pr_err("Unable to allocate memory for cmd->buf_ptr\n");
 		iscsit_release_cmd(cmd);
 		return -1;
 	}
-	memcpy(cmd->buf_ptr, buf, ISCSI_HDR_LEN);
 
 	if (add_to_conn) {
 		spin_lock_bh(&conn->cmd_lock);
@@ -1017,11 +1015,6 @@
 				" non-existent or non-exported iSCSI LUN:"
 				" 0x%016Lx\n", get_unaligned_le64(&hdr->lun));
 		}
-		if (ret == PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES)
-			return iscsit_add_reject_from_cmd(
-					ISCSI_REASON_BOOKMARK_NO_RESOURCES,
-					1, 1, buf, cmd);
-
 		send_check_condition = 1;
 		goto attach_cmd;
 	}
@@ -1044,6 +1037,8 @@
 		 */
 		send_check_condition = 1;
 	} else {
+		cmd->data_length = cmd->se_cmd.data_length;
+
 		if (iscsit_decide_list_to_build(cmd, payload_length) < 0)
 			return iscsit_add_reject_from_cmd(
 				ISCSI_REASON_BOOKMARK_NO_RESOURCES,
@@ -1123,7 +1118,7 @@
 	 * the backend memory allocation.
 	 */
 	ret = transport_generic_new_cmd(&cmd->se_cmd);
-	if ((ret < 0) || (cmd->se_cmd.se_cmd_flags & SCF_SE_CMD_FAILED)) {
+	if (ret < 0) {
 		immed_ret = IMMEDIATE_DATA_NORMAL_OPERATION;
 		dump_immediate_data = 1;
 		goto after_immediate_data;
@@ -1341,7 +1336,7 @@
 
 		spin_lock_irqsave(&se_cmd->t_state_lock, flags);
 		if (!(se_cmd->se_cmd_flags & SCF_SUPPORTED_SAM_OPCODE) ||
-		     (se_cmd->se_cmd_flags & SCF_SE_CMD_FAILED))
+		     (se_cmd->se_cmd_flags & SCF_SCSI_CDB_EXCEPTION))
 			dump_unsolicited_data = 1;
 		spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
 
@@ -2513,10 +2508,10 @@
 	if (hdr->flags & ISCSI_FLAG_DATA_STATUS) {
 		if (cmd->se_cmd.se_cmd_flags & SCF_OVERFLOW_BIT) {
 			hdr->flags |= ISCSI_FLAG_DATA_OVERFLOW;
-			hdr->residual_count = cpu_to_be32(cmd->residual_count);
+			hdr->residual_count = cpu_to_be32(cmd->se_cmd.residual_count);
 		} else if (cmd->se_cmd.se_cmd_flags & SCF_UNDERFLOW_BIT) {
 			hdr->flags |= ISCSI_FLAG_DATA_UNDERFLOW;
-			hdr->residual_count = cpu_to_be32(cmd->residual_count);
+			hdr->residual_count = cpu_to_be32(cmd->se_cmd.residual_count);
 		}
 	}
 	hton24(hdr->dlength, datain.length);
@@ -3018,10 +3013,10 @@
 	hdr->flags		|= ISCSI_FLAG_CMD_FINAL;
 	if (cmd->se_cmd.se_cmd_flags & SCF_OVERFLOW_BIT) {
 		hdr->flags |= ISCSI_FLAG_CMD_OVERFLOW;
-		hdr->residual_count = cpu_to_be32(cmd->residual_count);
+		hdr->residual_count = cpu_to_be32(cmd->se_cmd.residual_count);
 	} else if (cmd->se_cmd.se_cmd_flags & SCF_UNDERFLOW_BIT) {
 		hdr->flags |= ISCSI_FLAG_CMD_UNDERFLOW;
-		hdr->residual_count = cpu_to_be32(cmd->residual_count);
+		hdr->residual_count = cpu_to_be32(cmd->se_cmd.residual_count);
 	}
 	hdr->response		= cmd->iscsi_response;
 	hdr->cmd_status		= cmd->se_cmd.scsi_status;
@@ -3133,6 +3128,7 @@
 	hdr			= (struct iscsi_tm_rsp *) cmd->pdu;
 	memset(hdr, 0, ISCSI_HDR_LEN);
 	hdr->opcode		= ISCSI_OP_SCSI_TMFUNC_RSP;
+	hdr->flags		= ISCSI_FLAG_CMD_FINAL;
 	hdr->response		= iscsit_convert_tcm_tmr_rsp(se_tmr);
 	hdr->itt		= cpu_to_be32(cmd->init_task_tag);
 	cmd->stat_sn		= conn->stat_sn++;
diff --git a/drivers/target/iscsi/iscsi_target_auth.c b/drivers/target/iscsi/iscsi_target_auth.c
index beb3946..1cd6ce3 100644
--- a/drivers/target/iscsi/iscsi_target_auth.c
+++ b/drivers/target/iscsi/iscsi_target_auth.c
@@ -30,9 +30,11 @@
 
 static int chap_string_to_hex(unsigned char *dst, unsigned char *src, int len)
 {
-	int j = DIV_ROUND_UP(len, 2);
+	int j = DIV_ROUND_UP(len, 2), rc;
 
-	hex2bin(dst, src, j);
+	rc = hex2bin(dst, src, j);
+	if (rc < 0)
+		pr_debug("CHAP string contains non hex digit symbols\n");
 
 	dst[j] = '\0';
 	return j;
diff --git a/drivers/target/iscsi/iscsi_target_core.h b/drivers/target/iscsi/iscsi_target_core.h
index 3723d90..f1a02da 100644
--- a/drivers/target/iscsi/iscsi_target_core.h
+++ b/drivers/target/iscsi/iscsi_target_core.h
@@ -398,7 +398,6 @@
 	u32			pdu_send_order;
 	/* Current struct iscsi_pdu in struct iscsi_cmd->pdu_list */
 	u32			pdu_start;
-	u32			residual_count;
 	/* Next struct iscsi_seq to send in struct iscsi_cmd->seq_list */
 	u32			seq_send_order;
 	/* Number of struct iscsi_seq in struct iscsi_cmd->seq_list */
@@ -535,7 +534,6 @@
 	atomic_t		connection_exit;
 	atomic_t		connection_recovery;
 	atomic_t		connection_reinstatement;
-	atomic_t		connection_wait;
 	atomic_t		connection_wait_rcfr;
 	atomic_t		sleep_on_conn_wait_comp;
 	atomic_t		transport_failed;
@@ -643,7 +641,6 @@
 	atomic_t		session_reinstatement;
 	atomic_t		session_stop_active;
 	atomic_t		sleep_on_sess_wait_comp;
-	atomic_t		transport_wait_cmds;
 	/* connection list */
 	struct list_head	sess_conn_list;
 	struct list_head	cr_active_list;
diff --git a/drivers/target/iscsi/iscsi_target_erl1.c b/drivers/target/iscsi/iscsi_target_erl1.c
index c4c68da..101b1be 100644
--- a/drivers/target/iscsi/iscsi_target_erl1.c
+++ b/drivers/target/iscsi/iscsi_target_erl1.c
@@ -938,8 +938,7 @@
 		 * handle the SCF_SCSI_RESERVATION_CONFLICT case here as well.
 		 */
 		if (se_cmd->se_cmd_flags & SCF_SCSI_CDB_EXCEPTION) {
-			if (se_cmd->se_cmd_flags &
-					SCF_SCSI_RESERVATION_CONFLICT) {
+			if (se_cmd->scsi_sense_reason == TCM_RESERVATION_CONFLICT) {
 				cmd->i_state = ISTATE_SEND_STATUS;
 				spin_unlock_bh(&cmd->istate_lock);
 				iscsit_add_cmd_to_response_queue(cmd, cmd->conn,
diff --git a/drivers/target/iscsi/iscsi_target_login.c b/drivers/target/iscsi/iscsi_target_login.c
index daad362..d734bde 100644
--- a/drivers/target/iscsi/iscsi_target_login.c
+++ b/drivers/target/iscsi/iscsi_target_login.c
@@ -224,7 +224,7 @@
 		iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
 				ISCSI_LOGIN_STATUS_NO_RESOURCES);
 		pr_err("Could not allocate memory for session\n");
-		return -1;
+		return -ENOMEM;
 	}
 
 	iscsi_login_set_conn_values(sess, conn, pdu->cid);
@@ -250,7 +250,8 @@
 		pr_err("idr_pre_get() for sess_idr failed\n");
 		iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
 				ISCSI_LOGIN_STATUS_NO_RESOURCES);
-		return -1;
+		kfree(sess);
+		return -ENOMEM;
 	}
 	spin_lock(&sess_idr_lock);
 	idr_get_new(&sess_idr, NULL, &sess->session_index);
@@ -270,14 +271,16 @@
 				ISCSI_LOGIN_STATUS_NO_RESOURCES);
 		pr_err("Unable to allocate memory for"
 				" struct iscsi_sess_ops.\n");
-		return -1;
+		kfree(sess);
+		return -ENOMEM;
 	}
 
 	sess->se_sess = transport_init_session();
-	if (!sess->se_sess) {
+	if (IS_ERR(sess->se_sess)) {
 		iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
 				ISCSI_LOGIN_STATUS_NO_RESOURCES);
-		return -1;
+		kfree(sess);
+		return -ENOMEM;
 	}
 
 	return 0;
diff --git a/drivers/target/iscsi/iscsi_target_nego.c b/drivers/target/iscsi/iscsi_target_nego.c
index 426cd4b..98936cb 100644
--- a/drivers/target/iscsi/iscsi_target_nego.c
+++ b/drivers/target/iscsi/iscsi_target_nego.c
@@ -981,14 +981,13 @@
 		return NULL;
 	}
 
-	login->req = kzalloc(ISCSI_HDR_LEN, GFP_KERNEL);
+	login->req = kmemdup(login_pdu, ISCSI_HDR_LEN, GFP_KERNEL);
 	if (!login->req) {
 		pr_err("Unable to allocate memory for Login Request.\n");
 		iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR,
 				ISCSI_LOGIN_STATUS_NO_RESOURCES);
 		goto out;
 	}
-	memcpy(login->req, login_pdu, ISCSI_HDR_LEN);
 
 	login->req_buf = kzalloc(MAX_KEY_VALUE_PAIRS, GFP_KERNEL);
 	if (!login->req_buf) {
diff --git a/drivers/target/loopback/tcm_loop.c b/drivers/target/loopback/tcm_loop.c
index 3df1c9b..81d5832 100644
--- a/drivers/target/loopback/tcm_loop.c
+++ b/drivers/target/loopback/tcm_loop.c
@@ -113,11 +113,9 @@
 			scsi_bufflen(sc), sc->sc_data_direction, sam_task_attr,
 			&tl_cmd->tl_sense_buf[0]);
 
-	/*
-	 * Signal BIDI usage with T_TASK(cmd)->t_tasks_bidi
-	 */
 	if (scsi_bidi_cmnd(sc))
-		se_cmd->t_tasks_bidi = 1;
+		se_cmd->se_cmd_flags |= SCF_BIDI;
+
 	/*
 	 * Locate the struct se_lun pointer and attach it to struct se_cmd
 	 */
@@ -148,27 +146,13 @@
 	 * Allocate the necessary tasks to complete the received CDB+data
 	 */
 	ret = transport_generic_allocate_tasks(se_cmd, sc->cmnd);
-	if (ret == -ENOMEM) {
-		/* Out of Resources */
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
-	} else if (ret == -EINVAL) {
-		/*
-		 * Handle case for SAM_STAT_RESERVATION_CONFLICT
-		 */
-		if (se_cmd->se_cmd_flags & SCF_SCSI_RESERVATION_CONFLICT)
-			return PYX_TRANSPORT_RESERVATION_CONFLICT;
-		/*
-		 * Otherwise, return SAM_STAT_CHECK_CONDITION and return
-		 * sense data.
-		 */
-		return PYX_TRANSPORT_USE_SENSE_REASON;
-	}
-
+	if (ret != 0)
+		return ret;
 	/*
 	 * For BIDI commands, pass in the extra READ buffer
 	 * to transport_generic_map_mem_to_cmd() below..
 	 */
-	if (se_cmd->t_tasks_bidi) {
+	if (se_cmd->se_cmd_flags & SCF_BIDI) {
 		struct scsi_data_buffer *sdb = scsi_in(sc);
 
 		sgl_bidi = sdb->table.sgl;
@@ -194,12 +178,8 @@
 	}
 
 	/* Tell the core about our preallocated memory */
-	ret = transport_generic_map_mem_to_cmd(se_cmd, scsi_sglist(sc),
+	return transport_generic_map_mem_to_cmd(se_cmd, scsi_sglist(sc),
 			scsi_sg_count(sc), sgl_bidi, sgl_bidi_count);
-	if (ret < 0)
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
-
-	return 0;
 }
 
 /*
@@ -1360,17 +1340,16 @@
 {
 	struct tcm_loop_hba *tl_hba = container_of(wwn,
 				struct tcm_loop_hba, tl_hba_wwn);
-	int host_no = tl_hba->sh->host_no;
+
+	pr_debug("TCM_Loop_ConfigFS: Deallocating emulated Target"
+		" SAS Address: %s at Linux/SCSI Host ID: %d\n",
+		tl_hba->tl_wwn_address, tl_hba->sh->host_no);
 	/*
 	 * Call device_unregister() on the original tl_hba->dev.
 	 * tcm_loop_fabric_scsi.c:tcm_loop_release_adapter() will
 	 * release *tl_hba;
 	 */
 	device_unregister(&tl_hba->dev);
-
-	pr_debug("TCM_Loop_ConfigFS: Deallocated emulated Target"
-		" SAS Address: %s at Linux/SCSI Host ID: %d\n",
-		config_item_name(&wwn->wwn_group.cg_item), host_no);
 }
 
 /* Start items for tcm_loop_cit */
diff --git a/drivers/target/target_core_alua.c b/drivers/target/target_core_alua.c
index 88f2ad4..1dcbef4 100644
--- a/drivers/target/target_core_alua.c
+++ b/drivers/target/target_core_alua.c
@@ -191,9 +191,10 @@
 	int alua_access_state, primary = 0, rc;
 	u16 tg_pt_id, rtpi;
 
-	if (!l_port)
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
-
+	if (!l_port) {
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
+	}
 	buf = transport_kmap_first_data_page(cmd);
 
 	/*
@@ -203,7 +204,8 @@
 	l_tg_pt_gp_mem = l_port->sep_alua_tg_pt_gp_mem;
 	if (!l_tg_pt_gp_mem) {
 		pr_err("Unable to access l_port->sep_alua_tg_pt_gp_mem\n");
-		rc = PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+		rc = -EINVAL;
 		goto out;
 	}
 	spin_lock(&l_tg_pt_gp_mem->tg_pt_gp_mem_lock);
@@ -211,7 +213,8 @@
 	if (!l_tg_pt_gp) {
 		spin_unlock(&l_tg_pt_gp_mem->tg_pt_gp_mem_lock);
 		pr_err("Unable to access *l_tg_pt_gp_mem->tg_pt_gp\n");
-		rc = PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+		rc = -EINVAL;
 		goto out;
 	}
 	rc = (l_tg_pt_gp->tg_pt_gp_alua_access_type & TPGS_EXPLICT_ALUA);
@@ -220,7 +223,8 @@
 	if (!rc) {
 		pr_debug("Unable to process SET_TARGET_PORT_GROUPS"
 				" while TPGS_EXPLICT_ALUA is disabled\n");
-		rc = PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+		rc = -EINVAL;
 		goto out;
 	}
 
@@ -245,7 +249,8 @@
 			 * REQUEST, and the additional sense code set to INVALID
 			 * FIELD IN PARAMETER LIST.
 			 */
-			rc = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+			cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+			rc = -EINVAL;
 			goto out;
 		}
 		rc = -1;
@@ -298,7 +303,8 @@
 			 * throw an exception with ASCQ: INVALID_PARAMETER_LIST
 			 */
 			if (rc != 0) {
-				rc = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+				cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+				rc = -EINVAL;
 				goto out;
 			}
 		} else {
@@ -335,7 +341,8 @@
 			 * INVALID_PARAMETER_LIST
 			 */
 			if (rc != 0) {
-				rc = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+				cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+				rc = -EINVAL;
 				goto out;
 			}
 		}
@@ -1184,7 +1191,6 @@
 	 * struct t10_alua_lu_gp.
 	 */
 	spin_lock(&lu_gps_lock);
-	atomic_set(&lu_gp->lu_gp_shutdown, 1);
 	list_del(&lu_gp->lu_gp_node);
 	alua_lu_gps_count--;
 	spin_unlock(&lu_gps_lock);
@@ -1438,7 +1444,6 @@
 
 	tg_pt_gp_mem->tg_pt = port;
 	port->sep_alua_tg_pt_gp_mem = tg_pt_gp_mem;
-	atomic_set(&port->sep_tg_pt_gp_active, 1);
 
 	return tg_pt_gp_mem;
 }
diff --git a/drivers/target/target_core_cdb.c b/drivers/target/target_core_cdb.c
index 683ba02..831468b 100644
--- a/drivers/target/target_core_cdb.c
+++ b/drivers/target/target_core_cdb.c
@@ -478,7 +478,7 @@
 	if (cmd->data_length < 60)
 		return 0;
 
-	buf[2] = 0x3c;
+	buf[3] = 0x3c;
 	/* Set HEADSUP, ORDSUP, SIMPSUP */
 	buf[5] = 0x07;
 
@@ -703,6 +703,7 @@
 	if (cmd->data_length < 4) {
 		pr_err("SCSI Inquiry payload length: %u"
 			" too small for EVPD=1\n", cmd->data_length);
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
 		return -EINVAL;
 	}
 
@@ -719,6 +720,7 @@
 	}
 
 	pr_err("Unknown VPD Code: 0x%02x\n", cdb[2]);
+	cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
 	ret = -EINVAL;
 
 out_unmap:
@@ -969,7 +971,8 @@
 	default:
 		pr_err("MODE SENSE: unimplemented page/subpage: 0x%02x/0x%02x\n",
 		       cdb[2] & 0x3f, cdb[3]);
-		return PYX_TRANSPORT_UNKNOWN_MODE_PAGE;
+		cmd->scsi_sense_reason = TCM_UNKNOWN_MODE_PAGE;
+		return -EINVAL;
 	}
 	offset += length;
 
@@ -1027,7 +1030,8 @@
 	if (cdb[1] & 0x01) {
 		pr_err("REQUEST_SENSE description emulation not"
 			" supported\n");
-		return PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -ENOSYS;
 	}
 
 	buf = transport_kmap_first_data_page(cmd);
@@ -1100,7 +1104,8 @@
 	if (!dev->transport->do_discard) {
 		pr_err("UNMAP emulation not supported for: %s\n",
 				dev->transport->name);
-		return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+		return -ENOSYS;
 	}
 
 	/* First UNMAP block descriptor starts at 8 byte offset */
@@ -1157,7 +1162,8 @@
 	if (!dev->transport->do_discard) {
 		pr_err("WRITE_SAME emulation not supported"
 				" for: %s\n", dev->transport->name);
-		return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+		return -ENOSYS;
 	}
 
 	if (cmd->t_task_cdb[0] == WRITE_SAME)
@@ -1193,11 +1199,13 @@
 int target_emulate_synchronize_cache(struct se_task *task)
 {
 	struct se_device *dev = task->task_se_cmd->se_dev;
+	struct se_cmd *cmd = task->task_se_cmd;
 
 	if (!dev->transport->do_sync_cache) {
 		pr_err("SYNCHRONIZE_CACHE emulation not supported"
 			" for: %s\n", dev->transport->name);
-		return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+		return -ENOSYS;
 	}
 
 	dev->transport->do_sync_cache(task);
diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c
index e0c1e8a..93d4f6a 100644
--- a/drivers/target/target_core_configfs.c
+++ b/drivers/target/target_core_configfs.c
@@ -67,9 +67,6 @@
 static struct config_group alua_group;
 static struct config_group alua_lu_gps_group;
 
-static DEFINE_SPINLOCK(se_device_lock);
-static LIST_HEAD(se_dev_list);
-
 static inline struct se_hba *
 item_to_hba(struct config_item *item)
 {
@@ -2741,7 +2738,6 @@
 				" struct se_subsystem_dev\n");
 		goto unlock;
 	}
-	INIT_LIST_HEAD(&se_dev->se_dev_node);
 	INIT_LIST_HEAD(&se_dev->t10_wwn.t10_vpd_list);
 	spin_lock_init(&se_dev->t10_wwn.t10_vpd_lock);
 	INIT_LIST_HEAD(&se_dev->t10_pr.registration_list);
@@ -2777,9 +2773,6 @@
 			" from allocate_virtdevice()\n");
 		goto out;
 	}
-	spin_lock(&se_device_lock);
-	list_add_tail(&se_dev->se_dev_node, &se_dev_list);
-	spin_unlock(&se_device_lock);
 
 	config_group_init_type_name(&se_dev->se_dev_group, name,
 			&target_core_dev_cit);
@@ -2874,10 +2867,6 @@
 	mutex_lock(&hba->hba_access_mutex);
 	t = hba->transport;
 
-	spin_lock(&se_device_lock);
-	list_del(&se_dev->se_dev_node);
-	spin_unlock(&se_device_lock);
-
 	dev_stat_grp = &se_dev->dev_stat_grps.stat_group;
 	for (i = 0; dev_stat_grp->default_groups[i]; i++) {
 		df_item = &dev_stat_grp->default_groups[i]->cg_item;
diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
index ba5edec..9b86394 100644
--- a/drivers/target/target_core_device.c
+++ b/drivers/target/target_core_device.c
@@ -104,7 +104,6 @@
 		se_cmd->se_lun = deve->se_lun;
 		se_cmd->pr_res_key = deve->pr_res_key;
 		se_cmd->orig_fe_lun = unpacked_lun;
-		se_cmd->se_orig_obj_ptr = se_cmd->se_lun->lun_se_dev;
 		se_cmd->se_cmd_flags |= SCF_SE_LUN_CMD;
 	}
 	spin_unlock_irqrestore(&se_sess->se_node_acl->device_list_lock, flags);
@@ -137,7 +136,6 @@
 		se_lun = &se_sess->se_tpg->tpg_virt_lun0;
 		se_cmd->se_lun = &se_sess->se_tpg->tpg_virt_lun0;
 		se_cmd->orig_fe_lun = 0;
-		se_cmd->se_orig_obj_ptr = se_cmd->se_lun->lun_se_dev;
 		se_cmd->se_cmd_flags |= SCF_SE_LUN_CMD;
 	}
 	/*
@@ -200,7 +198,6 @@
 		se_lun = deve->se_lun;
 		se_cmd->pr_res_key = deve->pr_res_key;
 		se_cmd->orig_fe_lun = unpacked_lun;
-		se_cmd->se_orig_obj_ptr = se_cmd->se_dev;
 	}
 	spin_unlock_irqrestore(&se_sess->se_node_acl->device_list_lock, flags);
 
@@ -708,7 +705,7 @@
 
 	se_task->task_scsi_status = GOOD;
 	transport_complete_task(se_task, 1);
-	return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+	return 0;
 }
 
 /*	se_release_device_for_hba():
@@ -957,8 +954,12 @@
 		return -EINVAL;
 	}
 
-	pr_err("dpo_emulated not supported\n");
-	return -EINVAL;
+	if (flag) {
+		pr_err("dpo_emulated not supported\n");
+		return -EINVAL;
+	}
+
+	return 0;
 }
 
 int se_dev_set_emulate_fua_write(struct se_device *dev, int flag)
@@ -968,7 +969,7 @@
 		return -EINVAL;
 	}
 
-	if (dev->transport->fua_write_emulated == 0) {
+	if (flag && dev->transport->fua_write_emulated == 0) {
 		pr_err("fua_write_emulated not supported\n");
 		return -EINVAL;
 	}
@@ -985,8 +986,12 @@
 		return -EINVAL;
 	}
 
-	pr_err("ua read emulated not supported\n");
-	return -EINVAL;
+	if (flag) {
+		pr_err("ua read emulated not supported\n");
+		return -EINVAL;
+	}
+
+	return 0;
 }
 
 int se_dev_set_emulate_write_cache(struct se_device *dev, int flag)
@@ -995,7 +1000,7 @@
 		pr_err("Illegal value %d\n", flag);
 		return -EINVAL;
 	}
-	if (dev->transport->write_cache_emulated == 0) {
+	if (flag && dev->transport->write_cache_emulated == 0) {
 		pr_err("write_cache_emulated not supported\n");
 		return -EINVAL;
 	}
@@ -1056,7 +1061,7 @@
 	 * We expect this value to be non-zero when generic Block Layer
 	 * Discard supported is detected iblock_create_virtdevice().
 	 */
-	if (!dev->se_sub_dev->se_dev_attrib.max_unmap_block_desc_count) {
+	if (flag && !dev->se_sub_dev->se_dev_attrib.max_unmap_block_desc_count) {
 		pr_err("Generic Block Discard not supported\n");
 		return -ENOSYS;
 	}
@@ -1077,7 +1082,7 @@
 	 * We expect this value to be non-zero when generic Block Layer
 	 * Discard supported is detected iblock_create_virtdevice().
 	 */
-	if (!dev->se_sub_dev->se_dev_attrib.max_unmap_block_desc_count) {
+	if (flag && !dev->se_sub_dev->se_dev_attrib.max_unmap_block_desc_count) {
 		pr_err("Generic Block Discard not supported\n");
 		return -ENOSYS;
 	}
@@ -1587,7 +1592,6 @@
 		ret = -ENOMEM;
 		goto out;
 	}
-	INIT_LIST_HEAD(&se_dev->se_dev_node);
 	INIT_LIST_HEAD(&se_dev->t10_wwn.t10_vpd_list);
 	spin_lock_init(&se_dev->t10_wwn.t10_vpd_lock);
 	INIT_LIST_HEAD(&se_dev->t10_pr.registration_list);
diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c
index 67cd6fe..b4864fb 100644
--- a/drivers/target/target_core_file.c
+++ b/drivers/target/target_core_file.c
@@ -289,9 +289,9 @@
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < task->task_sg_nents; i++) {
-		iov[i].iov_len = sg[i].length;
-		iov[i].iov_base = sg_virt(&sg[i]);
+	for_each_sg(task->task_sg, sg, task->task_sg_nents, i) {
+		iov[i].iov_len = sg->length;
+		iov[i].iov_base = sg_virt(sg);
 	}
 
 	old_fs = get_fs();
@@ -342,9 +342,9 @@
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < task->task_sg_nents; i++) {
-		iov[i].iov_len = sg[i].length;
-		iov[i].iov_base = sg_virt(&sg[i]);
+	for_each_sg(task->task_sg, sg, task->task_sg_nents, i) {
+		iov[i].iov_len = sg->length;
+		iov[i].iov_base = sg_virt(sg);
 	}
 
 	old_fs = get_fs();
@@ -438,7 +438,7 @@
 		if (ret > 0 &&
 		    dev->se_sub_dev->se_dev_attrib.emulate_write_cache > 0 &&
 		    dev->se_sub_dev->se_dev_attrib.emulate_fua_write > 0 &&
-		    cmd->t_tasks_fua) {
+		    (cmd->se_cmd_flags & SCF_FUA)) {
 			/*
 			 * We might need to be a bit smarter here
 			 * and return some sense data to let the initiator
@@ -449,13 +449,15 @@
 
 	}
 
-	if (ret < 0)
+	if (ret < 0) {
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
 		return ret;
+	}
 	if (ret) {
 		task->task_scsi_status = GOOD;
 		transport_complete_task(task, 1);
 	}
-	return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+	return 0;
 }
 
 /*	fd_free_task(): (Part of se_subsystem_api_t template)
diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c
index 7698efe..4aa9922 100644
--- a/drivers/target/target_core_iblock.c
+++ b/drivers/target/target_core_iblock.c
@@ -531,7 +531,7 @@
 		 */
 		if (dev->se_sub_dev->se_dev_attrib.emulate_write_cache == 0 ||
 		    (dev->se_sub_dev->se_dev_attrib.emulate_fua_write > 0 &&
-		     task->task_se_cmd->t_tasks_fua))
+		     (cmd->se_cmd_flags & SCF_FUA)))
 			rw = WRITE_FUA;
 		else
 			rw = WRITE;
@@ -554,12 +554,15 @@
 	else {
 		pr_err("Unsupported SCSI -> BLOCK LBA conversion:"
 				" %u\n", dev->se_sub_dev->se_dev_attrib.block_size);
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -ENOSYS;
 	}
 
 	bio = iblock_get_bio(task, block_lba, sg_num);
-	if (!bio)
-		return PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+	if (!bio) {
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -ENOMEM;
+	}
 
 	bio_list_init(&list);
 	bio_list_add(&list, bio);
@@ -588,12 +591,13 @@
 		submit_bio(rw, bio);
 	blk_finish_plug(&plug);
 
-	return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+	return 0;
 
 fail:
 	while ((bio = bio_list_pop(&list)))
 		bio_put(bio);
-	return PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+	cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+	return -ENOMEM;
 }
 
 static u32 iblock_get_device_rev(struct se_device *dev)
diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c
index 5a4ebfc..95dee70 100644
--- a/drivers/target/target_core_pr.c
+++ b/drivers/target/target_core_pr.c
@@ -191,7 +191,7 @@
 		pr_err("Received legacy SPC-2 RESERVE/RELEASE"
 			" while active SPC-3 registrations exist,"
 			" returning RESERVATION_CONFLICT\n");
-		*ret = PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
 		return true;
 	}
 
@@ -252,7 +252,8 @@
 	    (cmd->t_task_cdb[1] & 0x02)) {
 		pr_err("LongIO and Obselete Bits set, returning"
 				" ILLEGAL_REQUEST\n");
-		ret = PYX_TRANSPORT_ILLEGAL_REQUEST;
+		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+		ret = -EINVAL;
 		goto out;
 	}
 	/*
@@ -277,7 +278,8 @@
 			" from %s \n", cmd->se_lun->unpacked_lun,
 			cmd->se_deve->mapped_lun,
 			sess->se_node_acl->initiatorname);
-		ret = PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		ret = -EINVAL;
 		goto out_unlock;
 	}
 
@@ -1510,7 +1512,8 @@
 	tidh_new = kzalloc(sizeof(struct pr_transport_id_holder), GFP_KERNEL);
 	if (!tidh_new) {
 		pr_err("Unable to allocate tidh_new\n");
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	INIT_LIST_HEAD(&tidh_new->dest_list);
 	tidh_new->dest_tpg = tpg;
@@ -1522,7 +1525,8 @@
 				sa_res_key, all_tg_pt, aptpl);
 	if (!local_pr_reg) {
 		kfree(tidh_new);
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -ENOMEM;
 	}
 	tidh_new->dest_pr_reg = local_pr_reg;
 	/*
@@ -1548,7 +1552,8 @@
 		pr_err("SPC-3 PR: Illegal tpdl: %u + 28 byte header"
 			" does not equal CDB data_length: %u\n", tpdl,
 			cmd->data_length);
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 	/*
@@ -1598,7 +1603,9 @@
 					" for tmp_tpg\n");
 				atomic_dec(&tmp_tpg->tpg_pr_ref_count);
 				smp_mb__after_atomic_dec();
-				ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+				cmd->scsi_sense_reason =
+					TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+				ret = -EINVAL;
 				goto out;
 			}
 			/*
@@ -1628,7 +1635,9 @@
 				atomic_dec(&dest_node_acl->acl_pr_ref_count);
 				smp_mb__after_atomic_dec();
 				core_scsi3_tpg_undepend_item(tmp_tpg);
-				ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+				cmd->scsi_sense_reason =
+					TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+				ret = -EINVAL;
 				goto out;
 			}
 
@@ -1646,7 +1655,8 @@
 		if (!dest_tpg) {
 			pr_err("SPC-3 PR SPEC_I_PT: Unable to locate"
 					" dest_tpg\n");
-			ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+			cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+			ret = -EINVAL;
 			goto out;
 		}
 #if 0
@@ -1660,7 +1670,8 @@
 				" %u for Transport ID: %s\n", tid_len, ptr);
 			core_scsi3_nodeacl_undepend_item(dest_node_acl);
 			core_scsi3_tpg_undepend_item(dest_tpg);
-			ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+			cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+			ret = -EINVAL;
 			goto out;
 		}
 		/*
@@ -1678,7 +1689,8 @@
 
 			core_scsi3_nodeacl_undepend_item(dest_node_acl);
 			core_scsi3_tpg_undepend_item(dest_tpg);
-			ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+			cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+			ret = -EINVAL;
 			goto out;
 		}
 
@@ -1690,7 +1702,9 @@
 			smp_mb__after_atomic_dec();
 			core_scsi3_nodeacl_undepend_item(dest_node_acl);
 			core_scsi3_tpg_undepend_item(dest_tpg);
-			ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+			cmd->scsi_sense_reason =
+				TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+			ret = -EINVAL;
 			goto out;
 		}
 #if 0
@@ -1727,7 +1741,9 @@
 			core_scsi3_lunacl_undepend_item(dest_se_deve);
 			core_scsi3_nodeacl_undepend_item(dest_node_acl);
 			core_scsi3_tpg_undepend_item(dest_tpg);
-			ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+			cmd->scsi_sense_reason =
+				TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+			ret = -ENOMEM;
 			goto out;
 		}
 		INIT_LIST_HEAD(&tidh_new->dest_list);
@@ -1759,7 +1775,8 @@
 			core_scsi3_nodeacl_undepend_item(dest_node_acl);
 			core_scsi3_tpg_undepend_item(dest_tpg);
 			kfree(tidh_new);
-			ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+			cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+			ret = -EINVAL;
 			goto out;
 		}
 		tidh_new->dest_pr_reg = dest_pr_reg;
@@ -2098,7 +2115,8 @@
 
 	if (!se_sess || !se_lun) {
 		pr_err("SPC-3 PR: se_sess || struct se_lun is NULL!\n");
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	se_tpg = se_sess->se_tpg;
 	se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
@@ -2117,13 +2135,14 @@
 		if (res_key) {
 			pr_warn("SPC-3 PR: Reservation Key non-zero"
 				" for SA REGISTER, returning CONFLICT\n");
-			return PYX_TRANSPORT_RESERVATION_CONFLICT;
+			cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+			return -EINVAL;
 		}
 		/*
 		 * Do nothing but return GOOD status.
 		 */
 		if (!sa_res_key)
-			return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+			return 0;
 
 		if (!spec_i_pt) {
 			/*
@@ -2138,7 +2157,8 @@
 			if (ret != 0) {
 				pr_err("Unable to allocate"
 					" struct t10_pr_registration\n");
-				return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+				cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+				return -EINVAL;
 			}
 		} else {
 			/*
@@ -2197,14 +2217,16 @@
 					" 0x%016Lx\n", res_key,
 					pr_reg->pr_res_key);
 				core_scsi3_put_pr_reg(pr_reg);
-				return PYX_TRANSPORT_RESERVATION_CONFLICT;
+				cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+				return -EINVAL;
 			}
 		}
 		if (spec_i_pt) {
 			pr_err("SPC-3 PR UNREGISTER: SPEC_I_PT"
 				" set while sa_res_key=0\n");
 			core_scsi3_put_pr_reg(pr_reg);
-			return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+			cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+			return -EINVAL;
 		}
 		/*
 		 * An existing ALL_TG_PT=1 registration being released
@@ -2215,7 +2237,8 @@
 				" registration exists, but ALL_TG_PT=1 bit not"
 				" present in received PROUT\n");
 			core_scsi3_put_pr_reg(pr_reg);
-			return PYX_TRANSPORT_INVALID_CDB_FIELD;
+			cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+			return -EINVAL;
 		}
 		/*
 		 * Allocate APTPL metadata buffer used for UNREGISTER ops
@@ -2227,7 +2250,9 @@
 				pr_err("Unable to allocate"
 					" pr_aptpl_buf\n");
 				core_scsi3_put_pr_reg(pr_reg);
-				return PYX_TRANSPORT_LU_COMM_FAILURE;
+				cmd->scsi_sense_reason =
+					TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+				return -EINVAL;
 			}
 		}
 		/*
@@ -2241,7 +2266,8 @@
 			if (pr_holder < 0) {
 				kfree(pr_aptpl_buf);
 				core_scsi3_put_pr_reg(pr_reg);
-				return PYX_TRANSPORT_RESERVATION_CONFLICT;
+				cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+				return -EINVAL;
 			}
 
 			spin_lock(&pr_tmpl->registration_lock);
@@ -2405,7 +2431,8 @@
 
 	if (!se_sess || !se_lun) {
 		pr_err("SPC-3 PR: se_sess || struct se_lun is NULL!\n");
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	se_tpg = se_sess->se_tpg;
 	se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
@@ -2417,7 +2444,8 @@
 	if (!pr_reg) {
 		pr_err("SPC-3 PR: Unable to locate"
 			" PR_REGISTERED *pr_reg for RESERVE\n");
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	/*
 	 * From spc4r17 Section 5.7.9: Reserving:
@@ -2433,7 +2461,8 @@
 			" does not match existing SA REGISTER res_key:"
 			" 0x%016Lx\n", res_key, pr_reg->pr_res_key);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		return -EINVAL;
 	}
 	/*
 	 * From spc4r17 Section 5.7.9: Reserving:
@@ -2448,7 +2477,8 @@
 	if (scope != PR_SCOPE_LU_SCOPE) {
 		pr_err("SPC-3 PR: Illegal SCOPE: 0x%02x\n", scope);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		return -EINVAL;
 	}
 	/*
 	 * See if we have an existing PR reservation holder pointer at
@@ -2480,7 +2510,8 @@
 
 			spin_unlock(&dev->dev_reservation_lock);
 			core_scsi3_put_pr_reg(pr_reg);
-			return PYX_TRANSPORT_RESERVATION_CONFLICT;
+			cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+			return -EINVAL;
 		}
 		/*
 		 * From spc4r17 Section 5.7.9: Reserving:
@@ -2503,7 +2534,8 @@
 
 			spin_unlock(&dev->dev_reservation_lock);
 			core_scsi3_put_pr_reg(pr_reg);
-			return PYX_TRANSPORT_RESERVATION_CONFLICT;
+			cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+			return -EINVAL;
 		}
 		/*
 		 * From spc4r17 Section 5.7.9: Reserving:
@@ -2517,7 +2549,7 @@
 		 */
 		spin_unlock(&dev->dev_reservation_lock);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+		return 0;
 	}
 	/*
 	 * Otherwise, our *pr_reg becomes the PR reservation holder for said
@@ -2574,7 +2606,8 @@
 	default:
 		pr_err("SPC-3 PR: Unknown Service Action RESERVE Type:"
 			" 0x%02x\n", type);
-		return PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -EINVAL;
 	}
 
 	return ret;
@@ -2630,7 +2663,8 @@
 
 	if (!se_sess || !se_lun) {
 		pr_err("SPC-3 PR: se_sess || struct se_lun is NULL!\n");
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	/*
 	 * Locate the existing *pr_reg via struct se_node_acl pointers
@@ -2639,7 +2673,8 @@
 	if (!pr_reg) {
 		pr_err("SPC-3 PR: Unable to locate"
 			" PR_REGISTERED *pr_reg for RELEASE\n");
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	/*
 	 * From spc4r17 Section 5.7.11.2 Releasing:
@@ -2661,7 +2696,7 @@
 		 */
 		spin_unlock(&dev->dev_reservation_lock);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+		return 0;
 	}
 	if ((pr_res_holder->pr_res_type == PR_TYPE_WRITE_EXCLUSIVE_ALLREG) ||
 	    (pr_res_holder->pr_res_type == PR_TYPE_EXCLUSIVE_ACCESS_ALLREG))
@@ -2675,7 +2710,7 @@
 		 */
 		spin_unlock(&dev->dev_reservation_lock);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+		return 0;
 	}
 	/*
 	 * From spc4r17 Section 5.7.11.2 Releasing:
@@ -2697,7 +2732,8 @@
 			" 0x%016Lx\n", res_key, pr_reg->pr_res_key);
 		spin_unlock(&dev->dev_reservation_lock);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		return -EINVAL;
 	}
 	/*
 	 * From spc4r17 Section 5.7.11.2 Releasing and above:
@@ -2719,7 +2755,8 @@
 
 		spin_unlock(&dev->dev_reservation_lock);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		return -EINVAL;
 	}
 	/*
 	 * In response to a persistent reservation release request from the
@@ -2802,7 +2839,8 @@
 	if (!pr_reg_n) {
 		pr_err("SPC-3 PR: Unable to locate"
 			" PR_REGISTERED *pr_reg for CLEAR\n");
-			return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	/*
 	 * From spc4r17 section 5.7.11.6, Clearing:
@@ -2821,7 +2859,8 @@
 			" existing SA REGISTER res_key:"
 			" 0x%016Lx\n", res_key, pr_reg_n->pr_res_key);
 		core_scsi3_put_pr_reg(pr_reg_n);
-		return PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		return -EINVAL;
 	}
 	/*
 	 * a) Release the persistent reservation, if any;
@@ -2979,8 +3018,10 @@
 	int all_reg = 0, calling_it_nexus = 0, released_regs = 0;
 	int prh_type = 0, prh_scope = 0, ret;
 
-	if (!se_sess)
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+	if (!se_sess) {
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
+	}
 
 	se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
 	pr_reg_n = core_scsi3_locate_pr_reg(cmd->se_dev, se_sess->se_node_acl,
@@ -2989,16 +3030,19 @@
 		pr_err("SPC-3 PR: Unable to locate"
 			" PR_REGISTERED *pr_reg for PREEMPT%s\n",
 			(abort) ? "_AND_ABORT" : "");
-		return PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		return -EINVAL;
 	}
 	if (pr_reg_n->pr_res_key != res_key) {
 		core_scsi3_put_pr_reg(pr_reg_n);
-		return PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		return -EINVAL;
 	}
 	if (scope != PR_SCOPE_LU_SCOPE) {
 		pr_err("SPC-3 PR: Illegal SCOPE: 0x%02x\n", scope);
 		core_scsi3_put_pr_reg(pr_reg_n);
-		return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		return -EINVAL;
 	}
 	INIT_LIST_HEAD(&preempt_and_abort_list);
 
@@ -3012,7 +3056,8 @@
 	if (!all_reg && !sa_res_key) {
 		spin_unlock(&dev->dev_reservation_lock);
 		core_scsi3_put_pr_reg(pr_reg_n);
-		return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		return -EINVAL;
 	}
 	/*
 	 * From spc4r17, section 5.7.11.4.4 Removing Registrations:
@@ -3106,7 +3151,8 @@
 		if (!released_regs) {
 			spin_unlock(&dev->dev_reservation_lock);
 			core_scsi3_put_pr_reg(pr_reg_n);
-			return PYX_TRANSPORT_RESERVATION_CONFLICT;
+			cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+			return -EINVAL;
 		}
 		/*
 		 * For an existing all registrants type reservation
@@ -3297,7 +3343,8 @@
 	default:
 		pr_err("SPC-3 PR: Unknown Service Action PREEMPT%s"
 			" Type: 0x%02x\n", (abort) ? "_AND_ABORT" : "", type);
-		return PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -EINVAL;
 	}
 
 	return ret;
@@ -3331,7 +3378,8 @@
 
 	if (!se_sess || !se_lun) {
 		pr_err("SPC-3 PR: se_sess || struct se_lun is NULL!\n");
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	memset(dest_iport, 0, 64);
 	memset(i_buf, 0, PR_REG_ISID_ID_LEN);
@@ -3349,7 +3397,8 @@
 	if (!pr_reg) {
 		pr_err("SPC-3 PR: Unable to locate PR_REGISTERED"
 			" *pr_reg for REGISTER_AND_MOVE\n");
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
 	}
 	/*
 	 * The provided reservation key much match the existing reservation key
@@ -3360,7 +3409,8 @@
 			" res_key: 0x%016Lx does not match existing SA REGISTER"
 			" res_key: 0x%016Lx\n", res_key, pr_reg->pr_res_key);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		return -EINVAL;
 	}
 	/*
 	 * The service active reservation key needs to be non zero
@@ -3369,7 +3419,8 @@
 		pr_warn("SPC-3 PR REGISTER_AND_MOVE: Received zero"
 			" sa_res_key\n");
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		return -EINVAL;
 	}
 
 	/*
@@ -3392,7 +3443,8 @@
 			" does not equal CDB data_length: %u\n", tid_len,
 			cmd->data_length);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		return -EINVAL;
 	}
 
 	spin_lock(&dev->se_port_lock);
@@ -3417,7 +3469,8 @@
 			atomic_dec(&dest_se_tpg->tpg_pr_ref_count);
 			smp_mb__after_atomic_dec();
 			core_scsi3_put_pr_reg(pr_reg);
-			return PYX_TRANSPORT_LU_COMM_FAILURE;
+			cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+			return -EINVAL;
 		}
 
 		spin_lock(&dev->se_port_lock);
@@ -3430,7 +3483,8 @@
 			" fabric ops from Relative Target Port Identifier:"
 			" %hu\n", rtpi);
 		core_scsi3_put_pr_reg(pr_reg);
-		return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		return -EINVAL;
 	}
 
 	buf = transport_kmap_first_data_page(cmd);
@@ -3445,14 +3499,16 @@
 			" from fabric: %s\n", proto_ident,
 			dest_tf_ops->get_fabric_proto_ident(dest_se_tpg),
 			dest_tf_ops->get_fabric_name());
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 	if (dest_tf_ops->tpg_parse_pr_out_transport_id == NULL) {
 		pr_err("SPC-3 PR REGISTER_AND_MOVE: Fabric does not"
 			" containg a valid tpg_parse_pr_out_transport_id"
 			" function pointer\n");
-		ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		ret = -EINVAL;
 		goto out;
 	}
 	initiator_str = dest_tf_ops->tpg_parse_pr_out_transport_id(dest_se_tpg,
@@ -3460,7 +3516,8 @@
 	if (!initiator_str) {
 		pr_err("SPC-3 PR REGISTER_AND_MOVE: Unable to locate"
 			" initiator_str from Transport ID\n");
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 
@@ -3489,7 +3546,8 @@
 		pr_err("SPC-3 PR REGISTER_AND_MOVE: TransportID: %s"
 			" matches: %s on received I_T Nexus\n", initiator_str,
 			pr_reg_nacl->initiatorname);
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 	if (!strcmp(iport_ptr, pr_reg->pr_reg_isid)) {
@@ -3497,7 +3555,8 @@
 			" matches: %s %s on received I_T Nexus\n",
 			initiator_str, iport_ptr, pr_reg_nacl->initiatorname,
 			pr_reg->pr_reg_isid);
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 after_iport_check:
@@ -3517,7 +3576,8 @@
 		pr_err("Unable to locate %s dest_node_acl for"
 			" TransportID%s\n", dest_tf_ops->get_fabric_name(),
 			initiator_str);
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 	ret = core_scsi3_nodeacl_depend_item(dest_node_acl);
@@ -3527,7 +3587,8 @@
 		atomic_dec(&dest_node_acl->acl_pr_ref_count);
 		smp_mb__after_atomic_dec();
 		dest_node_acl = NULL;
-		ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 #if 0
@@ -3543,7 +3604,8 @@
 	if (!dest_se_deve) {
 		pr_err("Unable to locate %s dest_se_deve from RTPI:"
 			" %hu\n",  dest_tf_ops->get_fabric_name(), rtpi);
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 
@@ -3553,7 +3615,8 @@
 		atomic_dec(&dest_se_deve->pr_ref_count);
 		smp_mb__after_atomic_dec();
 		dest_se_deve = NULL;
-		ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		ret = -EINVAL;
 		goto out;
 	}
 #if 0
@@ -3572,7 +3635,8 @@
 		pr_warn("SPC-3 PR REGISTER_AND_MOVE: No reservation"
 			" currently held\n");
 		spin_unlock(&dev->dev_reservation_lock);
-		ret = PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		ret = -EINVAL;
 		goto out;
 	}
 	/*
@@ -3585,7 +3649,8 @@
 		pr_warn("SPC-3 PR REGISTER_AND_MOVE: Calling I_T"
 			" Nexus is not reservation holder\n");
 		spin_unlock(&dev->dev_reservation_lock);
-		ret = PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		ret = -EINVAL;
 		goto out;
 	}
 	/*
@@ -3603,7 +3668,8 @@
 			" reservation for type: %s\n",
 			core_scsi3_pr_dump_type(pr_res_holder->pr_res_type));
 		spin_unlock(&dev->dev_reservation_lock);
-		ret = PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		ret = -EINVAL;
 		goto out;
 	}
 	pr_res_nacl = pr_res_holder->pr_reg_nacl;
@@ -3640,7 +3706,8 @@
 				sa_res_key, 0, aptpl, 2, 1);
 		if (ret != 0) {
 			spin_unlock(&dev->dev_reservation_lock);
-			ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+			cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+			ret = -EINVAL;
 			goto out;
 		}
 		dest_pr_reg = __core_scsi3_locate_pr_reg(dev, dest_node_acl,
@@ -3771,7 +3838,8 @@
 		pr_err("Received PERSISTENT_RESERVE CDB while legacy"
 			" SPC-2 reservation is held, returning"
 			" RESERVATION_CONFLICT\n");
-		ret = PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		ret = EINVAL;
 		goto out;
 	}
 
@@ -3779,13 +3847,16 @@
 	 * FIXME: A NULL struct se_session pointer means an this is not coming from
 	 * a $FABRIC_MOD's nexus, but from internal passthrough ops.
 	 */
-	if (!cmd->se_sess)
-		return PYX_TRANSPORT_LU_COMM_FAILURE;
+	if (!cmd->se_sess) {
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		return -EINVAL;
+	}
 
 	if (cmd->data_length < 24) {
 		pr_warn("SPC-PR: Received PR OUT parameter list"
 			" length too small: %u\n", cmd->data_length);
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 	/*
@@ -3820,7 +3891,8 @@
 	 * SPEC_I_PT=1 is only valid for Service action: REGISTER
 	 */
 	if (spec_i_pt && ((cdb[1] & 0x1f) != PRO_REGISTER)) {
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 
@@ -3837,7 +3909,8 @@
 	    (cmd->data_length != 24)) {
 		pr_warn("SPC-PR: Received PR OUT illegal parameter"
 			" list length: %u\n", cmd->data_length);
-		ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
+		ret = -EINVAL;
 		goto out;
 	}
 	/*
@@ -3878,7 +3951,8 @@
 	default:
 		pr_err("Unknown PERSISTENT_RESERVE_OUT service"
 			" action: 0x%02x\n", cdb[1] & 0x1f);
-		ret = PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		ret = -EINVAL;
 		break;
 	}
 
@@ -3906,7 +3980,8 @@
 	if (cmd->data_length < 8) {
 		pr_err("PRIN SA READ_KEYS SCSI Data Length: %u"
 			" too small\n", cmd->data_length);
-		return PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -EINVAL;
 	}
 
 	buf = transport_kmap_first_data_page(cmd);
@@ -3965,7 +4040,8 @@
 	if (cmd->data_length < 8) {
 		pr_err("PRIN SA READ_RESERVATIONS SCSI Data Length: %u"
 			" too small\n", cmd->data_length);
-		return PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -EINVAL;
 	}
 
 	buf = transport_kmap_first_data_page(cmd);
@@ -4047,7 +4123,8 @@
 	if (cmd->data_length < 6) {
 		pr_err("PRIN SA REPORT_CAPABILITIES SCSI Data Length:"
 			" %u too small\n", cmd->data_length);
-		return PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -EINVAL;
 	}
 
 	buf = transport_kmap_first_data_page(cmd);
@@ -4108,7 +4185,8 @@
 	if (cmd->data_length < 8) {
 		pr_err("PRIN SA READ_FULL_STATUS SCSI Data Length: %u"
 			" too small\n", cmd->data_length);
-		return PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		return -EINVAL;
 	}
 
 	buf = transport_kmap_first_data_page(cmd);
@@ -4255,7 +4333,8 @@
 		pr_err("Received PERSISTENT_RESERVE CDB while legacy"
 			" SPC-2 reservation is held, returning"
 			" RESERVATION_CONFLICT\n");
-		return PYX_TRANSPORT_RESERVATION_CONFLICT;
+		cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+		return -EINVAL;
 	}
 
 	switch (cmd->t_task_cdb[1] & 0x1f) {
@@ -4274,7 +4353,8 @@
 	default:
 		pr_err("Unknown PERSISTENT_RESERVE_IN service"
 			" action: 0x%02x\n", cmd->t_task_cdb[1] & 0x1f);
-		ret = PYX_TRANSPORT_INVALID_CDB_FIELD;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+		ret = -EINVAL;
 		break;
 	}
 
diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
index ed32e1e..8b15e56 100644
--- a/drivers/target/target_core_pscsi.c
+++ b/drivers/target/target_core_pscsi.c
@@ -963,6 +963,7 @@
 static int pscsi_map_sg(struct se_task *task, struct scatterlist *task_sg,
 		struct bio **hbio)
 {
+	struct se_cmd *cmd = task->task_se_cmd;
 	struct pscsi_dev_virt *pdv = task->task_se_cmd->se_dev->dev_ptr;
 	u32 task_sg_num = task->task_sg_nents;
 	struct bio *bio = NULL, *tbio = NULL;
@@ -971,7 +972,7 @@
 	u32 data_len = task->task_size, i, len, bytes, off;
 	int nr_pages = (task->task_size + task_sg[0].offset +
 			PAGE_SIZE - 1) >> PAGE_SHIFT;
-	int nr_vecs = 0, rc, ret = PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+	int nr_vecs = 0, rc;
 	int rw = (task->task_data_direction == DMA_TO_DEVICE);
 
 	*hbio = NULL;
@@ -1058,11 +1059,13 @@
 		bio->bi_next = NULL;
 		bio_endio(bio, 0);	/* XXX: should be error */
 	}
-	return ret;
+	cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+	return -ENOMEM;
 }
 
 static int pscsi_do_task(struct se_task *task)
 {
+	struct se_cmd *cmd = task->task_se_cmd;
 	struct pscsi_dev_virt *pdv = task->task_se_cmd->se_dev->dev_ptr;
 	struct pscsi_plugin_task *pt = PSCSI_TASK(task);
 	struct request *req;
@@ -1078,7 +1081,9 @@
 		if (!req || IS_ERR(req)) {
 			pr_err("PSCSI: blk_get_request() failed: %ld\n",
 					req ? IS_ERR(req) : -ENOMEM);
-			return PYX_TRANSPORT_LU_COMM_FAILURE;
+			cmd->scsi_sense_reason =
+				TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+			return -ENODEV;
 		}
 	} else {
 		BUG_ON(!task->task_size);
@@ -1087,8 +1092,11 @@
 		 * Setup the main struct request for the task->task_sg[] payload
 		 */
 		ret = pscsi_map_sg(task, task->task_sg, &hbio);
-		if (ret < 0)
-			return PYX_TRANSPORT_LU_COMM_FAILURE;
+		if (ret < 0) {
+			cmd->scsi_sense_reason =
+				TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+			return ret;
+		}
 
 		req = blk_make_request(pdv->pdv_sd->request_queue, hbio,
 				       GFP_KERNEL);
@@ -1115,7 +1123,7 @@
 			(task->task_se_cmd->sam_task_attr == MSG_HEAD_TAG),
 			pscsi_req_done);
 
-	return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+	return 0;
 
 fail:
 	while (hbio) {
@@ -1124,7 +1132,8 @@
 		bio->bi_next = NULL;
 		bio_endio(bio, 0);	/* XXX: should be error */
 	}
-	return PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+	cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+	return -ENOMEM;
 }
 
 /*	pscsi_get_sense_buffer():
@@ -1198,9 +1207,8 @@
 			" 0x%02x Result: 0x%08x\n", task, pt->pscsi_cdb[0],
 			pt->pscsi_result);
 		task->task_scsi_status = SAM_STAT_CHECK_CONDITION;
-		task->task_error_status = PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
-		task->task_se_cmd->transport_error_status =
-					PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+		task->task_se_cmd->scsi_sense_reason =
+					TCM_UNSUPPORTED_SCSI_OPCODE;
 		transport_complete_task(task, 0);
 		break;
 	}
diff --git a/drivers/target/target_core_rd.c b/drivers/target/target_core_rd.c
index 5158d384..02e51fa 100644
--- a/drivers/target/target_core_rd.c
+++ b/drivers/target/target_core_rd.c
@@ -343,235 +343,74 @@
 	return NULL;
 }
 
-/*	rd_MEMCPY_read():
- *
- *
- */
-static int rd_MEMCPY_read(struct rd_request *req)
+static int rd_MEMCPY(struct rd_request *req, u32 read_rd)
 {
 	struct se_task *task = &req->rd_task;
 	struct rd_dev *dev = req->rd_task.task_se_cmd->se_dev->dev_ptr;
 	struct rd_dev_sg_table *table;
-	struct scatterlist *sg_d, *sg_s;
-	void *dst, *src;
-	u32 i = 0, j = 0, dst_offset = 0, src_offset = 0;
-	u32 length, page_end = 0, table_sg_end;
+	struct scatterlist *rd_sg;
+	struct sg_mapping_iter m;
 	u32 rd_offset = req->rd_offset;
+	u32 src_len;
 
 	table = rd_get_sg_table(dev, req->rd_page);
 	if (!table)
 		return -EINVAL;
 
-	table_sg_end = (table->page_end_offset - req->rd_page);
-	sg_d = task->task_sg;
-	sg_s = &table->sg_table[req->rd_page - table->page_start_offset];
+	rd_sg = &table->sg_table[req->rd_page - table->page_start_offset];
 
-	pr_debug("RD[%u]: Read LBA: %llu, Size: %u Page: %u, Offset:"
-		" %u\n", dev->rd_dev_id, task->task_lba, req->rd_size,
-		req->rd_page, req->rd_offset);
+	pr_debug("RD[%u]: %s LBA: %llu, Size: %u Page: %u, Offset: %u\n",
+			dev->rd_dev_id, read_rd ? "Read" : "Write",
+			task->task_lba, req->rd_size, req->rd_page,
+			rd_offset);
 
-	src_offset = rd_offset;
-
+	src_len = PAGE_SIZE - rd_offset;
+	sg_miter_start(&m, task->task_sg, task->task_sg_nents,
+			read_rd ? SG_MITER_TO_SG : SG_MITER_FROM_SG);
 	while (req->rd_size) {
-		if ((sg_d[i].length - dst_offset) <
-		    (sg_s[j].length - src_offset)) {
-			length = (sg_d[i].length - dst_offset);
+		u32 len;
+		void *rd_addr;
 
-			pr_debug("Step 1 - sg_d[%d]: %p length: %d"
-				" offset: %u sg_s[%d].length: %u\n", i,
-				&sg_d[i], sg_d[i].length, sg_d[i].offset, j,
-				sg_s[j].length);
-			pr_debug("Step 1 - length: %u dst_offset: %u"
-				" src_offset: %u\n", length, dst_offset,
-				src_offset);
+		sg_miter_next(&m);
+		len = min((u32)m.length, src_len);
+		m.consumed = len;
 
-			if (length > req->rd_size)
-				length = req->rd_size;
+		rd_addr = sg_virt(rd_sg) + rd_offset;
 
-			dst = sg_virt(&sg_d[i++]) + dst_offset;
-			BUG_ON(!dst);
+		if (read_rd)
+			memcpy(m.addr, rd_addr, len);
+		else
+			memcpy(rd_addr, m.addr, len);
 
-			src = sg_virt(&sg_s[j]) + src_offset;
-			BUG_ON(!src);
-
-			dst_offset = 0;
-			src_offset = length;
-			page_end = 0;
-		} else {
-			length = (sg_s[j].length - src_offset);
-
-			pr_debug("Step 2 - sg_d[%d]: %p length: %d"
-				" offset: %u sg_s[%d].length: %u\n", i,
-				&sg_d[i], sg_d[i].length, sg_d[i].offset,
-				j, sg_s[j].length);
-			pr_debug("Step 2 - length: %u dst_offset: %u"
-				" src_offset: %u\n", length, dst_offset,
-				src_offset);
-
-			if (length > req->rd_size)
-				length = req->rd_size;
-
-			dst = sg_virt(&sg_d[i]) + dst_offset;
-			BUG_ON(!dst);
-
-			if (sg_d[i].length == length) {
-				i++;
-				dst_offset = 0;
-			} else
-				dst_offset = length;
-
-			src = sg_virt(&sg_s[j++]) + src_offset;
-			BUG_ON(!src);
-
-			src_offset = 0;
-			page_end = 1;
-		}
-
-		memcpy(dst, src, length);
-
-		pr_debug("page: %u, remaining size: %u, length: %u,"
-			" i: %u, j: %u\n", req->rd_page,
-			(req->rd_size - length), length, i, j);
-
-		req->rd_size -= length;
+		req->rd_size -= len;
 		if (!req->rd_size)
-			return 0;
-
-		if (!page_end)
 			continue;
 
-		if (++req->rd_page <= table->page_end_offset) {
-			pr_debug("page: %u in same page table\n",
-				req->rd_page);
+		src_len -= len;
+		if (src_len) {
+			rd_offset += len;
 			continue;
 		}
 
-		pr_debug("getting new page table for page: %u\n",
-				req->rd_page);
+		/* rd page completed, next one please */
+		req->rd_page++;
+		rd_offset = 0;
+		src_len = PAGE_SIZE;
+		if (req->rd_page <= table->page_end_offset) {
+			rd_sg++;
+			continue;
+		}
 
 		table = rd_get_sg_table(dev, req->rd_page);
-		if (!table)
+		if (!table) {
+			sg_miter_stop(&m);
 			return -EINVAL;
-
-		sg_s = &table->sg_table[j = 0];
-	}
-
-	return 0;
-}
-
-/*	rd_MEMCPY_write():
- *
- *
- */
-static int rd_MEMCPY_write(struct rd_request *req)
-{
-	struct se_task *task = &req->rd_task;
-	struct rd_dev *dev = req->rd_task.task_se_cmd->se_dev->dev_ptr;
-	struct rd_dev_sg_table *table;
-	struct scatterlist *sg_d, *sg_s;
-	void *dst, *src;
-	u32 i = 0, j = 0, dst_offset = 0, src_offset = 0;
-	u32 length, page_end = 0, table_sg_end;
-	u32 rd_offset = req->rd_offset;
-
-	table = rd_get_sg_table(dev, req->rd_page);
-	if (!table)
-		return -EINVAL;
-
-	table_sg_end = (table->page_end_offset - req->rd_page);
-	sg_d = &table->sg_table[req->rd_page - table->page_start_offset];
-	sg_s = task->task_sg;
-
-	pr_debug("RD[%d] Write LBA: %llu, Size: %u, Page: %u,"
-		" Offset: %u\n", dev->rd_dev_id, task->task_lba, req->rd_size,
-		req->rd_page, req->rd_offset);
-
-	dst_offset = rd_offset;
-
-	while (req->rd_size) {
-		if ((sg_s[i].length - src_offset) <
-		    (sg_d[j].length - dst_offset)) {
-			length = (sg_s[i].length - src_offset);
-
-			pr_debug("Step 1 - sg_s[%d]: %p length: %d"
-				" offset: %d sg_d[%d].length: %u\n", i,
-				&sg_s[i], sg_s[i].length, sg_s[i].offset,
-				j, sg_d[j].length);
-			pr_debug("Step 1 - length: %u src_offset: %u"
-				" dst_offset: %u\n", length, src_offset,
-				dst_offset);
-
-			if (length > req->rd_size)
-				length = req->rd_size;
-
-			src = sg_virt(&sg_s[i++]) + src_offset;
-			BUG_ON(!src);
-
-			dst = sg_virt(&sg_d[j]) + dst_offset;
-			BUG_ON(!dst);
-
-			src_offset = 0;
-			dst_offset = length;
-			page_end = 0;
-		} else {
-			length = (sg_d[j].length - dst_offset);
-
-			pr_debug("Step 2 - sg_s[%d]: %p length: %d"
-				" offset: %d sg_d[%d].length: %u\n", i,
-				&sg_s[i], sg_s[i].length, sg_s[i].offset,
-				j, sg_d[j].length);
-			pr_debug("Step 2 - length: %u src_offset: %u"
-				" dst_offset: %u\n", length, src_offset,
-				dst_offset);
-
-			if (length > req->rd_size)
-				length = req->rd_size;
-
-			src = sg_virt(&sg_s[i]) + src_offset;
-			BUG_ON(!src);
-
-			if (sg_s[i].length == length) {
-				i++;
-				src_offset = 0;
-			} else
-				src_offset = length;
-
-			dst = sg_virt(&sg_d[j++]) + dst_offset;
-			BUG_ON(!dst);
-
-			dst_offset = 0;
-			page_end = 1;
 		}
 
-		memcpy(dst, src, length);
-
-		pr_debug("page: %u, remaining size: %u, length: %u,"
-			" i: %u, j: %u\n", req->rd_page,
-			(req->rd_size - length), length, i, j);
-
-		req->rd_size -= length;
-		if (!req->rd_size)
-			return 0;
-
-		if (!page_end)
-			continue;
-
-		if (++req->rd_page <= table->page_end_offset) {
-			pr_debug("page: %u in same page table\n",
-				req->rd_page);
-			continue;
-		}
-
-		pr_debug("getting new page table for page: %u\n",
-				req->rd_page);
-
-		table = rd_get_sg_table(dev, req->rd_page);
-		if (!table)
-			return -EINVAL;
-
-		sg_d = &table->sg_table[j = 0];
+		/* since we increment, the first sg entry is correct */
+		rd_sg = table->sg_table;
 	}
-
+	sg_miter_stop(&m);
 	return 0;
 }
 
@@ -583,28 +422,21 @@
 {
 	struct se_device *dev = task->task_se_cmd->se_dev;
 	struct rd_request *req = RD_REQ(task);
-	unsigned long long lba;
+	u64 tmp;
 	int ret;
 
-	req->rd_page = (task->task_lba * dev->se_sub_dev->se_dev_attrib.block_size) / PAGE_SIZE;
-	lba = task->task_lba;
-	req->rd_offset = (do_div(lba,
-			  (PAGE_SIZE / dev->se_sub_dev->se_dev_attrib.block_size))) *
-			   dev->se_sub_dev->se_dev_attrib.block_size;
+	tmp = task->task_lba * dev->se_sub_dev->se_dev_attrib.block_size;
+	req->rd_offset = do_div(tmp, PAGE_SIZE);
+	req->rd_page = tmp;
 	req->rd_size = task->task_size;
 
-	if (task->task_data_direction == DMA_FROM_DEVICE)
-		ret = rd_MEMCPY_read(req);
-	else
-		ret = rd_MEMCPY_write(req);
-
+	ret = rd_MEMCPY(req, task->task_data_direction == DMA_FROM_DEVICE);
 	if (ret != 0)
 		return ret;
 
 	task->task_scsi_status = GOOD;
 	transport_complete_task(task, 1);
-
-	return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+	return 0;
 }
 
 /*	rd_free_task(): (Part of se_subsystem_api_t template)
diff --git a/drivers/target/target_core_tmr.c b/drivers/target/target_core_tmr.c
index 217e29d..6845228 100644
--- a/drivers/target/target_core_tmr.c
+++ b/drivers/target/target_core_tmr.c
@@ -345,10 +345,6 @@
 			" %d t_fe_count: %d\n", (preempt_and_abort_list) ?
 			"Preempt" : "", cmd, cmd->t_state,
 			atomic_read(&cmd->t_fe_count));
-		/*
-		 * Signal that the command has failed via cmd->se_cmd_flags,
-		 */
-		transport_new_cmd_failure(cmd);
 
 		core_tmr_handle_tas_abort(tmr_nacl, cmd, tas,
 				atomic_read(&cmd->t_fe_count));
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index 3400ae6..0257658 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -61,7 +61,6 @@
 static int sub_api_initialized;
 
 static struct workqueue_struct *target_completion_wq;
-static struct kmem_cache *se_cmd_cache;
 static struct kmem_cache *se_sess_cache;
 struct kmem_cache *se_tmr_req_cache;
 struct kmem_cache *se_ua_cache;
@@ -82,24 +81,18 @@
 static void transport_put_cmd(struct se_cmd *cmd);
 static void transport_remove_cmd_from_queue(struct se_cmd *cmd);
 static int transport_set_sense_codes(struct se_cmd *cmd, u8 asc, u8 ascq);
-static void transport_generic_request_failure(struct se_cmd *, int, int);
+static void transport_generic_request_failure(struct se_cmd *);
 static void target_complete_ok_work(struct work_struct *work);
 
 int init_se_kmem_caches(void)
 {
-	se_cmd_cache = kmem_cache_create("se_cmd_cache",
-			sizeof(struct se_cmd), __alignof__(struct se_cmd), 0, NULL);
-	if (!se_cmd_cache) {
-		pr_err("kmem_cache_create for struct se_cmd failed\n");
-		goto out;
-	}
 	se_tmr_req_cache = kmem_cache_create("se_tmr_cache",
 			sizeof(struct se_tmr_req), __alignof__(struct se_tmr_req),
 			0, NULL);
 	if (!se_tmr_req_cache) {
 		pr_err("kmem_cache_create() for struct se_tmr_req"
 				" failed\n");
-		goto out_free_cmd_cache;
+		goto out;
 	}
 	se_sess_cache = kmem_cache_create("se_sess_cache",
 			sizeof(struct se_session), __alignof__(struct se_session),
@@ -182,8 +175,6 @@
 	kmem_cache_destroy(se_sess_cache);
 out_free_tmr_req_cache:
 	kmem_cache_destroy(se_tmr_req_cache);
-out_free_cmd_cache:
-	kmem_cache_destroy(se_cmd_cache);
 out:
 	return -ENOMEM;
 }
@@ -191,7 +182,6 @@
 void release_se_kmem_caches(void)
 {
 	destroy_workqueue(target_completion_wq);
-	kmem_cache_destroy(se_cmd_cache);
 	kmem_cache_destroy(se_tmr_req_cache);
 	kmem_cache_destroy(se_sess_cache);
 	kmem_cache_destroy(se_ua_cache);
@@ -680,9 +670,9 @@
 		task->task_scsi_status = GOOD;
 	} else {
 		task->task_scsi_status = SAM_STAT_CHECK_CONDITION;
-		task->task_error_status = PYX_TRANSPORT_ILLEGAL_REQUEST;
-		task->task_se_cmd->transport_error_status =
-					PYX_TRANSPORT_ILLEGAL_REQUEST;
+		task->task_se_cmd->scsi_sense_reason =
+				TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+
 	}
 
 	transport_complete_task(task, good);
@@ -693,7 +683,7 @@
 {
 	struct se_cmd *cmd = container_of(work, struct se_cmd, work);
 
-	transport_generic_request_failure(cmd, 1, 1);
+	transport_generic_request_failure(cmd);
 }
 
 /*	transport_complete_task():
@@ -755,10 +745,11 @@
 	if (cmd->t_tasks_failed) {
 		if (!task->task_error_status) {
 			task->task_error_status =
-				PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
-			cmd->transport_error_status =
-				PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+				TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+			cmd->scsi_sense_reason =
+				TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
 		}
+
 		INIT_WORK(&cmd->work, target_complete_failure_work);
 	} else {
 		atomic_set(&cmd->t_transport_complete, 1);
@@ -1335,23 +1326,17 @@
 	dev->se_hba		= hba;
 	dev->se_sub_dev		= se_dev;
 	dev->transport		= transport;
-	atomic_set(&dev->active_cmds, 0);
 	INIT_LIST_HEAD(&dev->dev_list);
 	INIT_LIST_HEAD(&dev->dev_sep_list);
 	INIT_LIST_HEAD(&dev->dev_tmr_list);
 	INIT_LIST_HEAD(&dev->execute_task_list);
 	INIT_LIST_HEAD(&dev->delayed_cmd_list);
-	INIT_LIST_HEAD(&dev->ordered_cmd_list);
 	INIT_LIST_HEAD(&dev->state_task_list);
 	INIT_LIST_HEAD(&dev->qf_cmd_list);
 	spin_lock_init(&dev->execute_task_lock);
 	spin_lock_init(&dev->delayed_cmd_lock);
-	spin_lock_init(&dev->ordered_cmd_lock);
-	spin_lock_init(&dev->state_task_lock);
-	spin_lock_init(&dev->dev_alua_lock);
 	spin_lock_init(&dev->dev_reservation_lock);
 	spin_lock_init(&dev->dev_status_lock);
-	spin_lock_init(&dev->dev_status_thr_lock);
 	spin_lock_init(&dev->se_port_lock);
 	spin_lock_init(&dev->se_tmr_lock);
 	spin_lock_init(&dev->qf_cmd_lock);
@@ -1507,7 +1492,6 @@
 {
 	INIT_LIST_HEAD(&cmd->se_lun_node);
 	INIT_LIST_HEAD(&cmd->se_delayed_node);
-	INIT_LIST_HEAD(&cmd->se_ordered_node);
 	INIT_LIST_HEAD(&cmd->se_qf_node);
 	INIT_LIST_HEAD(&cmd->se_queue_node);
 	INIT_LIST_HEAD(&cmd->se_cmd_list);
@@ -1573,6 +1557,8 @@
 		pr_err("Received SCSI CDB with command_size: %d that"
 			" exceeds SCSI_MAX_VARLEN_CDB_SIZE: %d\n",
 			scsi_command_size(cdb), SCSI_MAX_VARLEN_CDB_SIZE);
+		cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
 		return -EINVAL;
 	}
 	/*
@@ -1588,6 +1574,9 @@
 				" %u > sizeof(cmd->__t_task_cdb): %lu ops\n",
 				scsi_command_size(cdb),
 				(unsigned long)sizeof(cmd->__t_task_cdb));
+			cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+			cmd->scsi_sense_reason =
+					TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
 			return -ENOMEM;
 		}
 	} else
@@ -1658,11 +1647,9 @@
 	 * and call transport_generic_request_failure() if necessary..
 	 */
 	ret = transport_generic_new_cmd(cmd);
-	if (ret < 0) {
-		cmd->transport_error_status = ret;
-		transport_generic_request_failure(cmd, 0,
-				(cmd->data_direction != DMA_TO_DEVICE));
-	}
+	if (ret < 0)
+		transport_generic_request_failure(cmd);
+
 	return 0;
 }
 EXPORT_SYMBOL(transport_handle_cdb_direct);
@@ -1798,20 +1785,16 @@
 /*
  * Handle SAM-esque emulation for generic transport request failures.
  */
-static void transport_generic_request_failure(
-	struct se_cmd *cmd,
-	int complete,
-	int sc)
+static void transport_generic_request_failure(struct se_cmd *cmd)
 {
 	int ret = 0;
 
 	pr_debug("-----[ Storage Engine Exception for cmd: %p ITT: 0x%08x"
 		" CDB: 0x%02x\n", cmd, cmd->se_tfo->get_task_tag(cmd),
 		cmd->t_task_cdb[0]);
-	pr_debug("-----[ i_state: %d t_state: %d transport_error_status: %d\n",
+	pr_debug("-----[ i_state: %d t_state: %d scsi_sense_reason: %d\n",
 		cmd->se_tfo->get_cmd_state(cmd),
-		cmd->t_state,
-		cmd->transport_error_status);
+		cmd->t_state, cmd->scsi_sense_reason);
 	pr_debug("-----[ t_tasks: %d t_task_cdbs_left: %d"
 		" t_task_cdbs_sent: %d t_task_cdbs_ex_left: %d --"
 		" t_transport_active: %d t_transport_stop: %d"
@@ -1829,46 +1812,19 @@
 	if (cmd->se_dev->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
 		transport_complete_task_attr(cmd);
 
-	if (complete) {
-		cmd->transport_error_status = PYX_TRANSPORT_LU_COMM_FAILURE;
-	}
-
-	switch (cmd->transport_error_status) {
-	case PYX_TRANSPORT_UNKNOWN_SAM_OPCODE:
-		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+	switch (cmd->scsi_sense_reason) {
+	case TCM_NON_EXISTENT_LUN:
+	case TCM_UNSUPPORTED_SCSI_OPCODE:
+	case TCM_INVALID_CDB_FIELD:
+	case TCM_INVALID_PARAMETER_LIST:
+	case TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE:
+	case TCM_UNKNOWN_MODE_PAGE:
+	case TCM_WRITE_PROTECTED:
+	case TCM_CHECK_CONDITION_ABORT_CMD:
+	case TCM_CHECK_CONDITION_UNIT_ATTENTION:
+	case TCM_CHECK_CONDITION_NOT_READY:
 		break;
-	case PYX_TRANSPORT_REQ_TOO_MANY_SECTORS:
-		cmd->scsi_sense_reason = TCM_SECTOR_COUNT_TOO_MANY;
-		break;
-	case PYX_TRANSPORT_INVALID_CDB_FIELD:
-		cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
-		break;
-	case PYX_TRANSPORT_INVALID_PARAMETER_LIST:
-		cmd->scsi_sense_reason = TCM_INVALID_PARAMETER_LIST;
-		break;
-	case PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES:
-		if (!sc)
-			transport_new_cmd_failure(cmd);
-		/*
-		 * Currently for PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES,
-		 * we force this session to fall back to session
-		 * recovery.
-		 */
-		cmd->se_tfo->fall_back_to_erl0(cmd->se_sess);
-		cmd->se_tfo->stop_session(cmd->se_sess, 0, 0);
-
-		goto check_stop;
-	case PYX_TRANSPORT_LU_COMM_FAILURE:
-	case PYX_TRANSPORT_ILLEGAL_REQUEST:
-		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
-		break;
-	case PYX_TRANSPORT_UNKNOWN_MODE_PAGE:
-		cmd->scsi_sense_reason = TCM_UNKNOWN_MODE_PAGE;
-		break;
-	case PYX_TRANSPORT_WRITE_PROTECTED:
-		cmd->scsi_sense_reason = TCM_WRITE_PROTECTED;
-		break;
-	case PYX_TRANSPORT_RESERVATION_CONFLICT:
+	case TCM_RESERVATION_CONFLICT:
 		/*
 		 * No SENSE Data payload for this case, set SCSI Status
 		 * and queue the response to $FABRIC_MOD.
@@ -1893,15 +1849,9 @@
 		if (ret == -EAGAIN || ret == -ENOMEM)
 			goto queue_full;
 		goto check_stop;
-	case PYX_TRANSPORT_USE_SENSE_REASON:
-		/*
-		 * struct se_cmd->scsi_sense_reason already set
-		 */
-		break;
 	default:
 		pr_err("Unknown transport error for CDB 0x%02x: %d\n",
-			cmd->t_task_cdb[0],
-			cmd->transport_error_status);
+			cmd->t_task_cdb[0], cmd->scsi_sense_reason);
 		cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
 		break;
 	}
@@ -1912,14 +1862,10 @@
 	 * transport_send_check_condition_and_sense() after handling
 	 * possible unsoliticied write data payloads.
 	 */
-	if (!sc && !cmd->se_tfo->new_cmd_map)
-		transport_new_cmd_failure(cmd);
-	else {
-		ret = transport_send_check_condition_and_sense(cmd,
-				cmd->scsi_sense_reason, 0);
-		if (ret == -EAGAIN || ret == -ENOMEM)
-			goto queue_full;
-	}
+	ret = transport_send_check_condition_and_sense(cmd,
+			cmd->scsi_sense_reason, 0);
+	if (ret == -EAGAIN || ret == -ENOMEM)
+		goto queue_full;
 
 check_stop:
 	transport_lun_remove_cmd(cmd);
@@ -2002,19 +1948,12 @@
 	 * to allow the passed struct se_cmd list of tasks to the front of the list.
 	 */
 	 if (cmd->sam_task_attr == MSG_HEAD_TAG) {
-		atomic_inc(&cmd->se_dev->dev_hoq_count);
-		smp_mb__after_atomic_inc();
 		pr_debug("Added HEAD_OF_QUEUE for CDB:"
 			" 0x%02x, se_ordered_id: %u\n",
 			cmd->t_task_cdb[0],
 			cmd->se_ordered_id);
 		return 1;
 	} else if (cmd->sam_task_attr == MSG_ORDERED_TAG) {
-		spin_lock(&cmd->se_dev->ordered_cmd_lock);
-		list_add_tail(&cmd->se_ordered_node,
-				&cmd->se_dev->ordered_cmd_list);
-		spin_unlock(&cmd->se_dev->ordered_cmd_lock);
-
 		atomic_inc(&cmd->se_dev->dev_ordered_sync);
 		smp_mb__after_atomic_inc();
 
@@ -2076,9 +2015,9 @@
 {
 	int add_tasks;
 
-	if (se_dev_check_online(cmd->se_orig_obj_ptr) != 0) {
-		cmd->transport_error_status = PYX_TRANSPORT_LU_COMM_FAILURE;
-		transport_generic_request_failure(cmd, 0, 1);
+	if (se_dev_check_online(cmd->se_dev) != 0) {
+		cmd->scsi_sense_reason = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+		transport_generic_request_failure(cmd);
 		return 0;
 	}
 
@@ -2163,14 +2102,13 @@
 	else
 		error = dev->transport->do_task(task);
 	if (error != 0) {
-		cmd->transport_error_status = error;
 		spin_lock_irqsave(&cmd->t_state_lock, flags);
 		task->task_flags &= ~TF_ACTIVE;
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 		atomic_set(&cmd->t_transport_sent, 0);
 		transport_stop_tasks_for_cmd(cmd);
 		atomic_inc(&dev->depth_left);
-		transport_generic_request_failure(cmd, 0, 1);
+		transport_generic_request_failure(cmd);
 	}
 
 	goto check_depth;
@@ -2178,19 +2116,6 @@
 	return 0;
 }
 
-void transport_new_cmd_failure(struct se_cmd *se_cmd)
-{
-	unsigned long flags;
-	/*
-	 * Any unsolicited data will get dumped for failed command inside of
-	 * the fabric plugin
-	 */
-	spin_lock_irqsave(&se_cmd->t_state_lock, flags);
-	se_cmd->se_cmd_flags |= SCF_SE_CMD_FAILED;
-	se_cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
-	spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
-}
-
 static inline u32 transport_get_sectors_6(
 	unsigned char *cdb,
 	struct se_cmd *cmd,
@@ -2213,10 +2138,15 @@
 
 	/*
 	 * Everything else assume TYPE_DISK Sector CDB location.
-	 * Use 8-bit sector value.
+	 * Use 8-bit sector value.  SBC-3 says:
+	 *
+	 *   A TRANSFER LENGTH field set to zero specifies that 256
+	 *   logical blocks shall be written.  Any other value
+	 *   specifies the number of logical blocks that shall be
+	 *   written.
 	 */
 type_disk:
-	return (u32)cdb[4];
+	return cdb[4] ? : 256;
 }
 
 static inline u32 transport_get_sectors_10(
@@ -2460,27 +2390,6 @@
 	return -1;
 }
 
-static int
-transport_handle_reservation_conflict(struct se_cmd *cmd)
-{
-	cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
-	cmd->se_cmd_flags |= SCF_SCSI_RESERVATION_CONFLICT;
-	cmd->scsi_status = SAM_STAT_RESERVATION_CONFLICT;
-	/*
-	 * For UA Interlock Code 11b, a RESERVATION CONFLICT will
-	 * establish a UNIT ATTENTION with PREVIOUS RESERVATION
-	 * CONFLICT STATUS.
-	 *
-	 * See spc4r17, section 7.4.6 Control Mode Page, Table 349
-	 */
-	if (cmd->se_sess &&
-	    cmd->se_dev->se_sub_dev->se_dev_attrib.emulate_ua_intlck_ctrl == 2)
-		core_scsi3_ua_allocate(cmd->se_sess->se_node_acl,
-			cmd->orig_fe_lun, 0x2C,
-			ASCQ_2CH_PREVIOUS_RESERVATION_CONFLICT_STATUS);
-	return -EINVAL;
-}
-
 static inline long long transport_dev_end_lba(struct se_device *dev)
 {
 	return dev->transport->get_blocks(dev) + 1;
@@ -2595,8 +2504,12 @@
 	 */
 	if (su_dev->t10_pr.pr_ops.t10_reservation_check(cmd, &pr_reg_type) != 0) {
 		if (su_dev->t10_pr.pr_ops.t10_seq_non_holder(
-					cmd, cdb, pr_reg_type) != 0)
-			return transport_handle_reservation_conflict(cmd);
+					cmd, cdb, pr_reg_type) != 0) {
+			cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+			cmd->se_cmd_flags |= SCF_SCSI_RESERVATION_CONFLICT;
+			cmd->scsi_sense_reason = TCM_RESERVATION_CONFLICT;
+			return -EBUSY;
+		}
 		/*
 		 * This means the CDB is allowed for the SCSI Initiator port
 		 * when said port is *NOT* holding the legacy SPC-2 or
@@ -2658,7 +2571,8 @@
 			goto out_unsupported_cdb;
 		size = transport_get_size(sectors, cdb, cmd);
 		cmd->t_task_lba = transport_lba_32(cdb);
-		cmd->t_tasks_fua = (cdb[1] & 0x8);
+		if (cdb[1] & 0x8)
+			cmd->se_cmd_flags |= SCF_FUA;
 		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
 		break;
 	case WRITE_12:
@@ -2667,7 +2581,8 @@
 			goto out_unsupported_cdb;
 		size = transport_get_size(sectors, cdb, cmd);
 		cmd->t_task_lba = transport_lba_32(cdb);
-		cmd->t_tasks_fua = (cdb[1] & 0x8);
+		if (cdb[1] & 0x8)
+			cmd->se_cmd_flags |= SCF_FUA;
 		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
 		break;
 	case WRITE_16:
@@ -2676,12 +2591,13 @@
 			goto out_unsupported_cdb;
 		size = transport_get_size(sectors, cdb, cmd);
 		cmd->t_task_lba = transport_lba_64(cdb);
-		cmd->t_tasks_fua = (cdb[1] & 0x8);
+		if (cdb[1] & 0x8)
+			cmd->se_cmd_flags |= SCF_FUA;
 		cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
 		break;
 	case XDWRITEREAD_10:
 		if ((cmd->data_direction != DMA_TO_DEVICE) ||
-		    !(cmd->t_tasks_bidi))
+		    !(cmd->se_cmd_flags & SCF_BIDI))
 			goto out_invalid_cdb_field;
 		sectors = transport_get_sectors_10(cdb, cmd, &sector_ret);
 		if (sector_ret)
@@ -2700,7 +2616,8 @@
 		 * Setup BIDI XOR callback to be run after I/O completion.
 		 */
 		cmd->transport_complete_callback = &transport_xor_callback;
-		cmd->t_tasks_fua = (cdb[1] & 0x8);
+		if (cdb[1] & 0x8)
+			cmd->se_cmd_flags |= SCF_FUA;
 		break;
 	case VARIABLE_LENGTH_CMD:
 		service_action = get_unaligned_be16(&cdb[8]);
@@ -2728,7 +2645,8 @@
 			 * completion.
 			 */
 			cmd->transport_complete_callback = &transport_xor_callback;
-			cmd->t_tasks_fua = (cdb[10] & 0x8);
+			if (cdb[1] & 0x8)
+				cmd->se_cmd_flags |= SCF_FUA;
 			break;
 		case WRITE_SAME_32:
 			sectors = transport_get_sectors_32(cdb, cmd, &sector_ret);
@@ -3171,18 +3089,13 @@
 			" SIMPLE: %u\n", dev->dev_cur_ordered_id,
 			cmd->se_ordered_id);
 	} else if (cmd->sam_task_attr == MSG_HEAD_TAG) {
-		atomic_dec(&dev->dev_hoq_count);
-		smp_mb__after_atomic_dec();
 		dev->dev_cur_ordered_id++;
 		pr_debug("Incremented dev_cur_ordered_id: %u for"
 			" HEAD_OF_QUEUE: %u\n", dev->dev_cur_ordered_id,
 			cmd->se_ordered_id);
 	} else if (cmd->sam_task_attr == MSG_ORDERED_TAG) {
-		spin_lock(&dev->ordered_cmd_lock);
-		list_del(&cmd->se_ordered_node);
 		atomic_dec(&dev->dev_ordered_sync);
 		smp_mb__after_atomic_dec();
-		spin_unlock(&dev->ordered_cmd_lock);
 
 		dev->dev_cur_ordered_id++;
 		pr_debug("Incremented dev_cur_ordered_id: %u for ORDERED:"
@@ -3495,6 +3408,18 @@
 
 	if ((cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB) ||
 	    (cmd->se_cmd_flags & SCF_SCSI_CONTROL_SG_IO_CDB)) {
+		/*
+		 * Reject SCSI data overflow with map_mem_to_cmd() as incoming
+		 * scatterlists already have been set to follow what the fabric
+		 * passes for the original expected data transfer length.
+		 */
+		if (cmd->se_cmd_flags & SCF_OVERFLOW_BIT) {
+			pr_warn("Rejecting SCSI DATA overflow for fabric using"
+				" SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC\n");
+			cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+			cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+			return -EINVAL;
+		}
 
 		cmd->t_data_sg = sgl;
 		cmd->t_data_nents = sgl_count;
@@ -3813,7 +3738,7 @@
 	    cmd->data_length) {
 		ret = transport_generic_get_mem(cmd);
 		if (ret < 0)
-			return ret;
+			goto out_fail;
 	}
 
 	/*
@@ -3842,8 +3767,15 @@
 		task_cdbs = transport_allocate_control_task(cmd);
 	}
 
-	if (task_cdbs <= 0)
+	if (task_cdbs < 0)
 		goto out_fail;
+	else if (!task_cdbs && (cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB)) {
+		cmd->t_state = TRANSPORT_COMPLETE;
+		atomic_set(&cmd->t_transport_active, 1);
+		INIT_WORK(&cmd->work, target_complete_ok_work);
+		queue_work(target_completion_wq, &cmd->work);
+		return 0;
+	}
 
 	if (set_counts) {
 		atomic_inc(&cmd->t_fe_count);
@@ -3929,7 +3861,7 @@
 	else if (ret < 0)
 		return ret;
 
-	return PYX_TRANSPORT_WRITE_PENDING;
+	return 1;
 
 queue_full:
 	pr_debug("Handling write_pending QUEUE__FULL: se_cmd: %p\n", cmd);
@@ -4602,9 +4534,6 @@
 		if (cmd->se_tfo->write_pending_status(cmd) != 0) {
 			atomic_inc(&cmd->t_transport_aborted);
 			smp_mb__after_atomic_inc();
-			cmd->scsi_status = SAM_STAT_TASK_ABORTED;
-			transport_new_cmd_failure(cmd);
-			return;
 		}
 	}
 	cmd->scsi_status = SAM_STAT_TASK_ABORTED;
@@ -4670,8 +4599,6 @@
 	struct se_cmd *cmd;
 	struct se_device *dev = (struct se_device *) param;
 
-	set_user_nice(current, -20);
-
 	while (!kthread_should_stop()) {
 		ret = wait_event_interruptible(dev->dev_queue_obj.thread_wq,
 				atomic_read(&dev->dev_queue_obj.queue_cnt) ||
@@ -4698,18 +4625,13 @@
 			}
 			ret = cmd->se_tfo->new_cmd_map(cmd);
 			if (ret < 0) {
-				cmd->transport_error_status = ret;
-				transport_generic_request_failure(cmd,
-						0, (cmd->data_direction !=
-						    DMA_TO_DEVICE));
+				transport_generic_request_failure(cmd);
 				break;
 			}
 			ret = transport_generic_new_cmd(cmd);
 			if (ret < 0) {
-				cmd->transport_error_status = ret;
-				transport_generic_request_failure(cmd,
-					0, (cmd->data_direction !=
-					 DMA_TO_DEVICE));
+				transport_generic_request_failure(cmd);
+				break;
 			}
 			break;
 		case TRANSPORT_PROCESS_WRITE:
diff --git a/drivers/target/tcm_fc/tfc_cmd.c b/drivers/target/tcm_fc/tfc_cmd.c
index 4fac37c..71fc9ce 100644
--- a/drivers/target/tcm_fc/tfc_cmd.c
+++ b/drivers/target/tcm_fc/tfc_cmd.c
@@ -200,7 +200,7 @@
 	lport = ep->lp;
 	fp = fc_frame_alloc(lport, sizeof(*txrdy));
 	if (!fp)
-		return PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+		return -ENOMEM; /* Signal QUEUE_FULL */
 
 	txrdy = fc_frame_payload_get(fp, sizeof(*txrdy));
 	memset(txrdy, 0, sizeof(*txrdy));
diff --git a/drivers/target/tcm_fc/tfc_conf.c b/drivers/target/tcm_fc/tfc_conf.c
index 5f77041..9402b73 100644
--- a/drivers/target/tcm_fc/tfc_conf.c
+++ b/drivers/target/tcm_fc/tfc_conf.c
@@ -436,8 +436,7 @@
 	struct ft_lport_acl *lacl = container_of(wwn,
 				struct ft_lport_acl, fc_lport_wwn);
 
-	pr_debug("del lport %s\n",
-			config_item_name(&wwn->wwn_group.cg_item));
+	pr_debug("del lport %s\n", lacl->name);
 	mutex_lock(&ft_lport_lock);
 	list_del(&lacl->list);
 	mutex_unlock(&ft_lport_lock);
diff --git a/drivers/tty/hvc/hvc_dcc.c b/drivers/tty/hvc/hvc_dcc.c
index 435f6fa..44fbeba 100644
--- a/drivers/tty/hvc/hvc_dcc.c
+++ b/drivers/tty/hvc/hvc_dcc.c
@@ -46,6 +46,7 @@
 
 	asm volatile("mrc p14, 0, %0, c0, c5, 0	@ read comms data reg"
 		: "=r" (__c));
+	isb();
 
 	return __c;
 }
@@ -55,6 +56,7 @@
 	asm volatile("mcr p14, 0, %0, c0, c5, 0	@ write a char"
 		: /* no output register */
 		: "r" (c));
+	isb();
 }
 
 static int hvc_dcc_put_chars(uint32_t vt, const char *buf, int count)
diff --git a/drivers/tty/n_gsm.c b/drivers/tty/n_gsm.c
index 4cb0d0a..fc7bbba 100644
--- a/drivers/tty/n_gsm.c
+++ b/drivers/tty/n_gsm.c
@@ -66,14 +66,16 @@
 static int debug;
 module_param(debug, int, 0600);
 
-#define T1	(HZ/10)
-#define T2	(HZ/3)
-#define N2	3
+/* Defaults: these are from the specification */
+
+#define T1	10		/* 100mS */
+#define T2	34		/* 333mS */
+#define N2	3		/* Retry 3 times */
 
 /* Use long timers for testing at low speed with debug on */
 #ifdef DEBUG_TIMING
-#define T1	HZ
-#define T2	(2 * HZ)
+#define T1	100
+#define T2	200
 #endif
 
 /*
diff --git a/drivers/tty/serial/Kconfig b/drivers/tty/serial/Kconfig
index 5f479da..925a1e5 100644
--- a/drivers/tty/serial/Kconfig
+++ b/drivers/tty/serial/Kconfig
@@ -1560,7 +1560,7 @@
 	  Support for the IFX6x60 modem devices on Intel MID platforms.
 
 config SERIAL_PCH_UART
-	tristate "Intel EG20T PCH / OKI SEMICONDUCTOR IOH(ML7213/ML7223) UART"
+	tristate "Intel EG20T PCH/LAPIS Semicon IOH(ML7213/ML7223/ML7831) UART"
 	depends on PCI
 	select SERIAL_CORE
 	help
@@ -1568,12 +1568,12 @@
 	  which is an IOH(Input/Output Hub) for x86 embedded processor.
 	  Enabling PCH_DMA, this PCH UART works as DMA mode.
 
-	  This driver also can be used for OKI SEMICONDUCTOR IOH(Input/
-	  Output Hub), ML7213 and ML7223.
-	  ML7213 IOH is for IVI(In-Vehicle Infotainment) use and ML7223 IOH is
-	  for MP(Media Phone) use.
-	  ML7213/ML7223 is companion chip for Intel Atom E6xx series.
-	  ML7213/ML7223 is completely compatible for Intel EG20T PCH.
+	  This driver also can be used for LAPIS Semiconductor IOH(Input/
+	  Output Hub), ML7213, ML7223 and ML7831.
+	  ML7213 IOH is for IVI(In-Vehicle Infotainment) use, ML7223 IOH is
+	  for MP(Media Phone) use and ML7831 IOH is for general purpose use.
+	  ML7213/ML7223/ML7831 is companion chip for Intel Atom E6xx series.
+	  ML7213/ML7223/ML7831 is completely compatible for Intel EG20T PCH.
 
 config SERIAL_MSM_SMD
 	bool "Enable tty device interface for some SMD ports"
diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
index 4a0f86f..4c823f3 100644
--- a/drivers/tty/serial/atmel_serial.c
+++ b/drivers/tty/serial/atmel_serial.c
@@ -228,7 +228,7 @@
 	if (rs485conf->flags & SER_RS485_ENABLED) {
 		dev_dbg(port->dev, "Setting UART to RS485\n");
 		atmel_port->tx_done_mask = ATMEL_US_TXEMPTY;
-		if (rs485conf->flags & SER_RS485_RTS_AFTER_SEND)
+		if ((rs485conf->delay_rts_after_send) > 0)
 			UART_PUT_TTGR(port, rs485conf->delay_rts_after_send);
 		mode |= ATMEL_US_USMODE_RS485;
 	} else {
@@ -304,7 +304,7 @@
 
 	if (atmel_port->rs485.flags & SER_RS485_ENABLED) {
 		dev_dbg(port->dev, "Setting UART to RS485\n");
-		if (atmel_port->rs485.flags & SER_RS485_RTS_AFTER_SEND)
+		if ((atmel_port->rs485.delay_rts_after_send) > 0)
 			UART_PUT_TTGR(port,
 					atmel_port->rs485.delay_rts_after_send);
 		mode |= ATMEL_US_USMODE_RS485;
@@ -1228,7 +1228,7 @@
 
 	if (atmel_port->rs485.flags & SER_RS485_ENABLED) {
 		dev_dbg(port->dev, "Setting UART to RS485\n");
-		if (atmel_port->rs485.flags & SER_RS485_RTS_AFTER_SEND)
+		if ((atmel_port->rs485.delay_rts_after_send) > 0)
 			UART_PUT_TTGR(port,
 					atmel_port->rs485.delay_rts_after_send);
 		mode |= ATMEL_US_USMODE_RS485;
@@ -1447,16 +1447,6 @@
 		rs485conf->delay_rts_after_send = rs485_delay[1];
 		rs485conf->flags = 0;
 
-		if (rs485conf->delay_rts_before_send == 0 &&
-		    rs485conf->delay_rts_after_send == 0) {
-			rs485conf->flags |= SER_RS485_RTS_ON_SEND;
-		} else {
-			if (rs485conf->delay_rts_before_send)
-				rs485conf->flags |= SER_RS485_RTS_BEFORE_SEND;
-			if (rs485conf->delay_rts_after_send)
-				rs485conf->flags |= SER_RS485_RTS_AFTER_SEND;
-		}
-
 		if (of_get_property(np, "rs485-rx-during-tx", NULL))
 			rs485conf->flags |= SER_RS485_RX_DURING_TX;
 
diff --git a/drivers/tty/serial/crisv10.c b/drivers/tty/serial/crisv10.c
index b743504..1dfba7b 100644
--- a/drivers/tty/serial/crisv10.c
+++ b/drivers/tty/serial/crisv10.c
@@ -3234,9 +3234,8 @@
 		e100_disable_rx(info);
 		e100_enable_rx_irq(info);
 #endif
-		if ((info->rs485.flags & SER_RS485_RTS_BEFORE_SEND) &&
-			(info->rs485.delay_rts_before_send > 0))
-				msleep(info->rs485.delay_rts_before_send);
+		if (info->rs485.delay_rts_before_send > 0)
+			msleep(info->rs485.delay_rts_before_send);
 	}
 #endif /* CONFIG_ETRAX_RS485 */
 
@@ -3693,10 +3692,6 @@
 
 		rs485data.delay_rts_before_send = rs485ctrl.delay_rts_before_send;
 		rs485data.flags = 0;
-		if (rs485data.delay_rts_before_send != 0)
-			rs485data.flags |= SER_RS485_RTS_BEFORE_SEND;
-		else
-			rs485data.flags &= ~(SER_RS485_RTS_BEFORE_SEND);
 
 		if (rs485ctrl.enabled)
 			rs485data.flags |= SER_RS485_ENABLED;
@@ -4531,7 +4526,6 @@
 		/* Set sane defaults */
 		info->rs485.flags &= ~(SER_RS485_RTS_ON_SEND);
 		info->rs485.flags |= SER_RS485_RTS_AFTER_SEND;
-		info->rs485.flags &= ~(SER_RS485_RTS_BEFORE_SEND);
 		info->rs485.delay_rts_before_send = 0;
 		info->rs485.flags &= ~(SER_RS485_ENABLED);
 #endif
diff --git a/drivers/tty/serial/mfd.c b/drivers/tty/serial/mfd.c
index 286c386..e272d39 100644
--- a/drivers/tty/serial/mfd.c
+++ b/drivers/tty/serial/mfd.c
@@ -884,7 +884,6 @@
 {
 	struct uart_hsu_port *up =
 			container_of(port, struct uart_hsu_port, port);
-	struct tty_struct *tty = port->state->port.tty;
 	unsigned char cval, fcr = 0;
 	unsigned long flags;
 	unsigned int baud, quot;
@@ -907,8 +906,7 @@
 	}
 
 	/* CMSPAR isn't supported by this driver */
-	if (tty)
-		tty->termios->c_cflag &= ~CMSPAR;
+	termios->c_cflag &= ~CMSPAR;
 
 	if (termios->c_cflag & CSTOPB)
 		cval |= UART_LCR_STOP;
diff --git a/drivers/tty/serial/pch_uart.c b/drivers/tty/serial/pch_uart.c
index 21febef..d6aba8c 100644
--- a/drivers/tty/serial/pch_uart.c
+++ b/drivers/tty/serial/pch_uart.c
@@ -1,5 +1,5 @@
 /*
- *Copyright (C) 2010 OKI SEMICONDUCTOR CO., LTD.
+ *Copyright (C) 2011 LAPIS Semiconductor Co., Ltd.
  *
  *This program is free software; you can redistribute it and/or modify
  *it under the terms of the GNU General Public License as published by
@@ -46,8 +46,8 @@
 
 /* Set the max number of UART port
  * Intel EG20T PCH: 4 port
- * OKI SEMICONDUCTOR ML7213 IOH: 3 port
- * OKI SEMICONDUCTOR ML7223 IOH: 2 port
+ * LAPIS Semiconductor ML7213 IOH: 3 port
+ * LAPIS Semiconductor ML7223 IOH: 2 port
 */
 #define PCH_UART_NR	4
 
@@ -258,6 +258,8 @@
 	pch_ml7213_uart2,
 	pch_ml7223_uart0,
 	pch_ml7223_uart1,
+	pch_ml7831_uart0,
+	pch_ml7831_uart1,
 };
 
 static struct pch_uart_driver_data drv_dat[] = {
@@ -270,6 +272,8 @@
 	[pch_ml7213_uart2] = {PCH_UART_2LINE, 2},
 	[pch_ml7223_uart0] = {PCH_UART_8LINE, 0},
 	[pch_ml7223_uart1] = {PCH_UART_2LINE, 1},
+	[pch_ml7831_uart0] = {PCH_UART_8LINE, 0},
+	[pch_ml7831_uart1] = {PCH_UART_2LINE, 1},
 };
 
 static unsigned int default_baud = 9600;
@@ -628,6 +632,7 @@
 		dev_err(priv->port.dev, "%s:dma_request_channel FAILS(Rx)\n",
 			__func__);
 		dma_release_channel(priv->chan_tx);
+		priv->chan_tx = NULL;
 		return;
 	}
 
@@ -1215,8 +1220,7 @@
 		dev_err(priv->port.dev,
 			"pch_uart_hal_set_fifo Failed(ret=%d)\n", ret);
 
-	if (priv->use_dma_flag)
-		pch_free_dma(port);
+	pch_free_dma(port);
 
 	free_irq(priv->port.irq, priv);
 }
@@ -1280,6 +1284,7 @@
 	if (rtn)
 		goto out;
 
+	pch_uart_set_mctrl(&priv->port, priv->port.mctrl);
 	/* Don't rewrite B0 */
 	if (tty_termios_baud_rate(termios))
 		tty_termios_encode_baud_rate(termios, baud, baud);
@@ -1552,6 +1557,10 @@
 	 .driver_data = pch_ml7223_uart0},
 	{PCI_DEVICE(PCI_VENDOR_ID_ROHM, 0x800D),
 	 .driver_data = pch_ml7223_uart1},
+	{PCI_DEVICE(PCI_VENDOR_ID_ROHM, 0x8811),
+	 .driver_data = pch_ml7831_uart0},
+	{PCI_DEVICE(PCI_VENDOR_ID_ROHM, 0x8812),
+	 .driver_data = pch_ml7831_uart1},
 	{0,},
 };
 
diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c
index 1945c70..aff9d61 100644
--- a/drivers/tty/serial/sh-sci.c
+++ b/drivers/tty/serial/sh-sci.c
@@ -207,6 +207,25 @@
 	},
 
 	/*
+	 * Common SH-2(A) SCIF definitions for ports with FIFO data
+	 * count registers.
+	 */
+	[SCIx_SH2_SCIF_FIFODATA_REGTYPE] = {
+		[SCSMR]		= { 0x00, 16 },
+		[SCBRR]		= { 0x04,  8 },
+		[SCSCR]		= { 0x08, 16 },
+		[SCxTDR]	= { 0x0c,  8 },
+		[SCxSR]		= { 0x10, 16 },
+		[SCxRDR]	= { 0x14,  8 },
+		[SCFCR]		= { 0x18, 16 },
+		[SCFDR]		= { 0x1c, 16 },
+		[SCTFDR]	= sci_reg_invalid,
+		[SCRFDR]	= sci_reg_invalid,
+		[SCSPTR]	= { 0x20, 16 },
+		[SCLSR]		= { 0x24, 16 },
+	},
+
+	/*
 	 * Common SH-3 SCIF definitions.
 	 */
 	[SCIx_SH3_SCIF_REGTYPE] = {
diff --git a/drivers/tty/tty_ldisc.c b/drivers/tty/tty_ldisc.c
index 512c49f..8e0924f 100644
--- a/drivers/tty/tty_ldisc.c
+++ b/drivers/tty/tty_ldisc.c
@@ -36,6 +36,7 @@
 
 #include <linux/kmod.h>
 #include <linux/nsproxy.h>
+#include <linux/ratelimit.h>
 
 /*
  *	This guards the refcounted line discipline lists. The lock
@@ -547,15 +548,16 @@
 /**
  *	tty_ldisc_wait_idle	-	wait for the ldisc to become idle
  *	@tty: tty to wait for
+ *	@timeout: for how long to wait at most
  *
  *	Wait for the line discipline to become idle. The discipline must
  *	have been halted for this to guarantee it remains idle.
  */
-static int tty_ldisc_wait_idle(struct tty_struct *tty)
+static int tty_ldisc_wait_idle(struct tty_struct *tty, long timeout)
 {
-	int ret;
+	long ret;
 	ret = wait_event_timeout(tty_ldisc_idle,
-			atomic_read(&tty->ldisc->users) == 1, 5 * HZ);
+			atomic_read(&tty->ldisc->users) == 1, timeout);
 	if (ret < 0)
 		return ret;
 	return ret > 0 ? 0 : -EBUSY;
@@ -665,7 +667,7 @@
 
 	tty_ldisc_flush_works(tty);
 
-	retval = tty_ldisc_wait_idle(tty);
+	retval = tty_ldisc_wait_idle(tty, 5 * HZ);
 
 	tty_lock();
 	mutex_lock(&tty->ldisc_mutex);
@@ -762,8 +764,6 @@
 	if (IS_ERR(ld))
 		return -1;
 
-	WARN_ON_ONCE(tty_ldisc_wait_idle(tty));
-
 	tty_ldisc_close(tty, tty->ldisc);
 	tty_ldisc_put(tty->ldisc);
 	tty->ldisc = NULL;
@@ -838,7 +838,7 @@
 	tty_unlock();
 	cancel_work_sync(&tty->buf.work);
 	mutex_unlock(&tty->ldisc_mutex);
-
+retry:
 	tty_lock();
 	mutex_lock(&tty->ldisc_mutex);
 
@@ -847,6 +847,22 @@
 	   it means auditing a lot of other paths so this is
 	   a FIXME */
 	if (tty->ldisc) {	/* Not yet closed */
+		if (atomic_read(&tty->ldisc->users) != 1) {
+			char cur_n[TASK_COMM_LEN], tty_n[64];
+			long timeout = 3 * HZ;
+			tty_unlock();
+
+			while (tty_ldisc_wait_idle(tty, timeout) == -EBUSY) {
+				timeout = MAX_SCHEDULE_TIMEOUT;
+				printk_ratelimited(KERN_WARNING
+					"%s: waiting (%s) for %s took too long, but we keep waiting...\n",
+					__func__, get_task_comm(cur_n, current),
+					tty_name(tty, tty_n));
+			}
+			mutex_unlock(&tty->ldisc_mutex);
+			goto retry;
+		}
+
 		if (reset == 0) {
 
 			if (!tty_ldisc_reinit(tty, tty->termios->c_line))
diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c
index 6960715..a8078d0 100644
--- a/drivers/usb/class/cdc-acm.c
+++ b/drivers/usb/class/cdc-acm.c
@@ -539,7 +539,6 @@
 {
 	int i;
 
-	mutex_lock(&open_mutex);
 	if (acm->dev) {
 		usb_autopm_get_interface(acm->control);
 		acm_set_control(acm, acm->ctrlout = 0);
@@ -551,14 +550,15 @@
 		acm->control->needs_remote_wakeup = 0;
 		usb_autopm_put_interface(acm->control);
 	}
-	mutex_unlock(&open_mutex);
 }
 
 static void acm_tty_hangup(struct tty_struct *tty)
 {
 	struct acm *acm = tty->driver_data;
 	tty_port_hangup(&acm->port);
+	mutex_lock(&open_mutex);
 	acm_port_down(acm);
+	mutex_unlock(&open_mutex);
 }
 
 static void acm_tty_close(struct tty_struct *tty, struct file *filp)
@@ -569,8 +569,9 @@
 	   shutdown */
 	if (!acm)
 		return;
+
+	mutex_lock(&open_mutex);
 	if (tty_port_close_start(&acm->port, tty, filp) == 0) {
-		mutex_lock(&open_mutex);
 		if (!acm->dev) {
 			tty_port_tty_set(&acm->port, NULL);
 			acm_tty_unregister(acm);
@@ -582,6 +583,7 @@
 	acm_port_down(acm);
 	tty_port_close_end(&acm->port, tty);
 	tty_port_tty_set(&acm->port, NULL);
+	mutex_unlock(&open_mutex);
 }
 
 static int acm_tty_write(struct tty_struct *tty,
@@ -1456,6 +1458,16 @@
 	},
 	{ USB_DEVICE(0x22b8, 0x6425), /* Motorola MOTOMAGX phones */
 	},
+	/* Motorola H24 HSPA module: */
+	{ USB_DEVICE(0x22b8, 0x2d91) }, /* modem                                */
+	{ USB_DEVICE(0x22b8, 0x2d92) }, /* modem           + diagnostics        */
+	{ USB_DEVICE(0x22b8, 0x2d93) }, /* modem + AT port                      */
+	{ USB_DEVICE(0x22b8, 0x2d95) }, /* modem + AT port + diagnostics        */
+	{ USB_DEVICE(0x22b8, 0x2d96) }, /* modem                         + NMEA */
+	{ USB_DEVICE(0x22b8, 0x2d97) }, /* modem           + diagnostics + NMEA */
+	{ USB_DEVICE(0x22b8, 0x2d99) }, /* modem + AT port               + NMEA */
+	{ USB_DEVICE(0x22b8, 0x2d9a) }, /* modem + AT port + diagnostics + NMEA */
+
 	{ USB_DEVICE(0x0572, 0x1329), /* Hummingbird huc56s (Conexant) */
 	.driver_info = NO_UNION_NORMAL, /* union descriptor misplaced on
 					   data interface instead of
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 96f05b2..7978146 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -813,6 +813,12 @@
 					USB_PORT_FEAT_C_PORT_LINK_STATE);
 		}
 
+		if ((portchange & USB_PORT_STAT_C_BH_RESET) &&
+				hub_is_superspeed(hub->hdev)) {
+			need_debounce_delay = true;
+			clear_port_feature(hub->hdev, port1,
+					USB_PORT_FEAT_C_BH_PORT_RESET);
+		}
 		/* We can forget about a "removed" device when there's a
 		 * physical disconnect or the connect status changes.
 		 */
diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c
index d6a8d82..ecf12e1 100644
--- a/drivers/usb/core/quirks.c
+++ b/drivers/usb/core/quirks.c
@@ -50,15 +50,42 @@
 	/* Logitech Webcam B/C500 */
 	{ USB_DEVICE(0x046d, 0x0807), .driver_info = USB_QUIRK_RESET_RESUME },
 
+	/* Logitech Webcam C600 */
+	{ USB_DEVICE(0x046d, 0x0808), .driver_info = USB_QUIRK_RESET_RESUME },
+
 	/* Logitech Webcam Pro 9000 */
 	{ USB_DEVICE(0x046d, 0x0809), .driver_info = USB_QUIRK_RESET_RESUME },
 
+	/* Logitech Webcam C905 */
+	{ USB_DEVICE(0x046d, 0x080a), .driver_info = USB_QUIRK_RESET_RESUME },
+
+	/* Logitech Webcam C210 */
+	{ USB_DEVICE(0x046d, 0x0819), .driver_info = USB_QUIRK_RESET_RESUME },
+
+	/* Logitech Webcam C260 */
+	{ USB_DEVICE(0x046d, 0x081a), .driver_info = USB_QUIRK_RESET_RESUME },
+
 	/* Logitech Webcam C310 */
 	{ USB_DEVICE(0x046d, 0x081b), .driver_info = USB_QUIRK_RESET_RESUME },
 
+	/* Logitech Webcam C910 */
+	{ USB_DEVICE(0x046d, 0x0821), .driver_info = USB_QUIRK_RESET_RESUME },
+
+	/* Logitech Webcam C160 */
+	{ USB_DEVICE(0x046d, 0x0824), .driver_info = USB_QUIRK_RESET_RESUME },
+
 	/* Logitech Webcam C270 */
 	{ USB_DEVICE(0x046d, 0x0825), .driver_info = USB_QUIRK_RESET_RESUME },
 
+	/* Logitech Quickcam Pro 9000 */
+	{ USB_DEVICE(0x046d, 0x0990), .driver_info = USB_QUIRK_RESET_RESUME },
+
+	/* Logitech Quickcam E3500 */
+	{ USB_DEVICE(0x046d, 0x09a4), .driver_info = USB_QUIRK_RESET_RESUME },
+
+	/* Logitech Quickcam Vision Pro */
+	{ USB_DEVICE(0x046d, 0x09a6), .driver_info = USB_QUIRK_RESET_RESUME },
+
 	/* Logitech Harmony 700-series */
 	{ USB_DEVICE(0x046d, 0xc122), .driver_info = USB_QUIRK_DELAY_INIT },
 
diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
index fa824cf..25dbd86 100644
--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -1284,6 +1284,7 @@
 			int		ret;
 
 			dep->endpoint.maxpacket = 1024;
+			dep->endpoint.max_streams = 15;
 			dep->endpoint.ops = &dwc3_gadget_ep_ops;
 			list_add_tail(&dep->endpoint.ep_list,
 					&dwc->gadget.ep_list);
diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig
index b21cd37..23a4473 100644
--- a/drivers/usb/gadget/Kconfig
+++ b/drivers/usb/gadget/Kconfig
@@ -469,7 +469,7 @@
 	   gadget drivers to also be dynamically linked.
 
 config USB_EG20T
-	tristate "Intel EG20T PCH/OKI SEMICONDUCTOR ML7213 IOH UDC"
+	tristate "Intel EG20T PCH/LAPIS Semiconductor IOH(ML7213/ML7831) UDC"
 	depends on PCI
 	select USB_GADGET_DUALSPEED
 	help
@@ -485,10 +485,11 @@
 	  This driver dose not support interrupt transfer or isochronous
 	  transfer modes.
 
-	  This driver also can be used for OKI SEMICONDUCTOR's ML7213 which is
+	  This driver also can be used for LAPIS Semiconductor's ML7213 which is
 	  for IVI(In-Vehicle Infotainment) use.
-	  ML7213 is companion chip for Intel Atom E6xx series.
-	  ML7213 is completely compatible for Intel EG20T PCH.
+	  ML7831 is for general purpose use.
+	  ML7213/ML7831 is companion chip for Intel Atom E6xx series.
+	  ML7213/ML7831 is completely compatible for Intel EG20T PCH.
 
 config USB_CI13XXX_MSM
 	tristate "MIPS USB CI13xxx for MSM"
diff --git a/drivers/usb/gadget/amd5536udc.c b/drivers/usb/gadget/amd5536udc.c
index 4730016..45f422a 100644
--- a/drivers/usb/gadget/amd5536udc.c
+++ b/drivers/usb/gadget/amd5536udc.c
@@ -1959,7 +1959,7 @@
 	u32 tmp;
 
 	if (!driver || !bind || !driver->setup
-			|| driver->speed != USB_SPEED_HIGH)
+			|| driver->speed < USB_SPEED_HIGH)
 		return -EINVAL;
 	if (!dev)
 		return -ENODEV;
diff --git a/drivers/usb/gadget/ci13xxx_msm.c b/drivers/usb/gadget/ci13xxx_msm.c
index 4eedfe5..1fc6129 100644
--- a/drivers/usb/gadget/ci13xxx_msm.c
+++ b/drivers/usb/gadget/ci13xxx_msm.c
@@ -122,3 +122,5 @@
 	return platform_driver_register(&ci13xxx_msm_driver);
 }
 module_init(ci13xxx_msm_init);
+
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/usb/gadget/ci13xxx_udc.c b/drivers/usb/gadget/ci13xxx_udc.c
index 83428f5..9a0c397 100644
--- a/drivers/usb/gadget/ci13xxx_udc.c
+++ b/drivers/usb/gadget/ci13xxx_udc.c
@@ -71,6 +71,9 @@
 /******************************************************************************
  * DEFINE
  *****************************************************************************/
+
+#define DMA_ADDR_INVALID	(~(dma_addr_t)0)
+
 /* ctrl register bank access */
 static DEFINE_SPINLOCK(udc_lock);
 
@@ -1434,7 +1437,7 @@
 		return -EALREADY;
 
 	mReq->req.status = -EALREADY;
-	if (length && !mReq->req.dma) {
+	if (length && mReq->req.dma == DMA_ADDR_INVALID) {
 		mReq->req.dma = \
 			dma_map_single(mEp->device, mReq->req.buf,
 				       length, mEp->dir ? DMA_TO_DEVICE :
@@ -1453,7 +1456,7 @@
 				dma_unmap_single(mEp->device, mReq->req.dma,
 					length, mEp->dir ? DMA_TO_DEVICE :
 					DMA_FROM_DEVICE);
-				mReq->req.dma = 0;
+				mReq->req.dma = DMA_ADDR_INVALID;
 				mReq->map     = 0;
 			}
 			return -ENOMEM;
@@ -1549,7 +1552,7 @@
 	if (mReq->map) {
 		dma_unmap_single(mEp->device, mReq->req.dma, mReq->req.length,
 				 mEp->dir ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
-		mReq->req.dma = 0;
+		mReq->req.dma = DMA_ADDR_INVALID;
 		mReq->map     = 0;
 	}
 
@@ -1610,7 +1613,6 @@
  * @gadget: gadget
  *
  * This function returns an error code
- * Caller must hold lock
  */
 static int _gadget_stop_activity(struct usb_gadget *gadget)
 {
@@ -2189,6 +2191,7 @@
 	mReq = kzalloc(sizeof(struct ci13xxx_req), gfp_flags);
 	if (mReq != NULL) {
 		INIT_LIST_HEAD(&mReq->queue);
+		mReq->req.dma = DMA_ADDR_INVALID;
 
 		mReq->ptr = dma_pool_alloc(mEp->td_pool, gfp_flags,
 					   &mReq->dma);
@@ -2328,7 +2331,7 @@
 	if (mReq->map) {
 		dma_unmap_single(mEp->device, mReq->req.dma, mReq->req.length,
 				 mEp->dir ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
-		mReq->req.dma = 0;
+		mReq->req.dma = DMA_ADDR_INVALID;
 		mReq->map     = 0;
 	}
 	req->status = -ECONNRESET;
@@ -2500,12 +2503,12 @@
 	spin_lock_irqsave(udc->lock, flags);
 	if (!udc->remote_wakeup) {
 		ret = -EOPNOTSUPP;
-		dbg_trace("remote wakeup feature is not enabled\n");
+		trace("remote wakeup feature is not enabled\n");
 		goto out;
 	}
 	if (!hw_cread(CAP_PORTSC, PORTSC_SUSP)) {
 		ret = -EINVAL;
-		dbg_trace("port is not suspended\n");
+		trace("port is not suspended\n");
 		goto out;
 	}
 	hw_cwrite(CAP_PORTSC, PORTSC_FPR, PORTSC_FPR);
@@ -2703,7 +2706,9 @@
 		if (udc->udc_driver->notify_event)
 			udc->udc_driver->notify_event(udc,
 			CI13XXX_CONTROLLER_STOPPED_EVENT);
+		spin_unlock_irqrestore(udc->lock, flags);
 		_gadget_stop_activity(&udc->gadget);
+		spin_lock_irqsave(udc->lock, flags);
 		pm_runtime_put(&udc->gadget.dev);
 	}
 
@@ -2850,7 +2855,7 @@
 	struct ci13xxx *udc;
 	int retval = 0;
 
-	trace("%p, %p, %p", dev, regs, name);
+	trace("%p, %p, %p", dev, regs, driver->name);
 
 	if (dev == NULL || regs == NULL || driver == NULL ||
 			driver->name == NULL)
diff --git a/drivers/usb/gadget/f_mass_storage.c b/drivers/usb/gadget/f_mass_storage.c
index 52583a2..1a6f415 100644
--- a/drivers/usb/gadget/f_mass_storage.c
+++ b/drivers/usb/gadget/f_mass_storage.c
@@ -624,7 +624,8 @@
 		if (ctrl->bRequestType !=
 		    (USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_INTERFACE))
 			break;
-		if (w_index != fsg->interface_number || w_value != 0)
+		if (w_index != fsg->interface_number || w_value != 0 ||
+				w_length != 0)
 			return -EDOM;
 
 		/*
@@ -639,7 +640,8 @@
 		if (ctrl->bRequestType !=
 		    (USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE))
 			break;
-		if (w_index != fsg->interface_number || w_value != 0)
+		if (w_index != fsg->interface_number || w_value != 0 ||
+				w_length != 1)
 			return -EDOM;
 		VDBG(fsg, "get max LUN\n");
 		*(u8 *)req->buf = fsg->common->nluns - 1;
@@ -2973,6 +2975,7 @@
 	fsg_common_put(common);
 	usb_free_descriptors(fsg->function.descriptors);
 	usb_free_descriptors(fsg->function.hs_descriptors);
+	usb_free_descriptors(fsg->function.ss_descriptors);
 	kfree(fsg);
 }
 
diff --git a/drivers/usb/gadget/f_midi.c b/drivers/usb/gadget/f_midi.c
index 67b2229..3797b3d 100644
--- a/drivers/usb/gadget/f_midi.c
+++ b/drivers/usb/gadget/f_midi.c
@@ -95,7 +95,6 @@
 
 DECLARE_UAC_AC_HEADER_DESCRIPTOR(1);
 DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(1);
-DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(16);
 DECLARE_USB_MS_ENDPOINT_DESCRIPTOR(16);
 
 /* B.3.1  Standard AC Interface Descriptor */
@@ -140,26 +139,6 @@
 	/* .wTotalLength =	DYNAMIC */
 };
 
-/* B.4.3  Embedded MIDI IN Jack Descriptor */
-static struct usb_midi_in_jack_descriptor jack_in_emb_desc = {
-	.bLength =	      USB_DT_MIDI_IN_SIZE,
-	.bDescriptorType =      USB_DT_CS_INTERFACE,
-	.bDescriptorSubtype =   USB_MS_MIDI_IN_JACK,
-	.bJackType =	    USB_MS_EMBEDDED,
-	/* .bJackID =		DYNAMIC */
-};
-
-/* B.4.4  Embedded MIDI OUT Jack Descriptor */
-static struct usb_midi_out_jack_descriptor_16 jack_out_emb_desc = {
-	/* .bLength =		DYNAMIC */
-	.bDescriptorType =	USB_DT_CS_INTERFACE,
-	.bDescriptorSubtype =	USB_MS_MIDI_OUT_JACK,
-	.bJackType =		USB_MS_EMBEDDED,
-	/* .bJackID =		DYNAMIC */
-	/* .bNrInputPins =	DYNAMIC */
-	/* .pins =		DYNAMIC */
-};
-
 /* B.5.1  Standard Bulk OUT Endpoint Descriptor */
 static struct usb_endpoint_descriptor bulk_out_desc = {
 	.bLength =		USB_DT_ENDPOINT_AUDIO_SIZE,
@@ -758,9 +737,11 @@
 static int __init
 f_midi_bind(struct usb_configuration *c, struct usb_function *f)
 {
-	struct usb_descriptor_header *midi_function[(MAX_PORTS * 2) + 12];
+	struct usb_descriptor_header **midi_function;
 	struct usb_midi_in_jack_descriptor jack_in_ext_desc[MAX_PORTS];
+	struct usb_midi_in_jack_descriptor jack_in_emb_desc[MAX_PORTS];
 	struct usb_midi_out_jack_descriptor_1 jack_out_ext_desc[MAX_PORTS];
+	struct usb_midi_out_jack_descriptor_1 jack_out_emb_desc[MAX_PORTS];
 	struct usb_composite_dev *cdev = c->cdev;
 	struct f_midi *midi = func_to_midi(f);
 	int status, n, jack = 1, i = 0;
@@ -798,6 +779,14 @@
 		goto fail;
 	midi->out_ep->driver_data = cdev;	/* claim */
 
+	/* allocate temporary function list */
+	midi_function = kcalloc((MAX_PORTS * 4) + 9, sizeof(midi_function),
+				GFP_KERNEL);
+	if (!midi_function) {
+		status = -ENOMEM;
+		goto fail;
+	}
+
 	/*
 	 * construct the function's descriptor set. As the number of
 	 * input and output MIDI ports is configurable, we have to do
@@ -811,73 +800,74 @@
 
 	/* calculate the header's wTotalLength */
 	n = USB_DT_MS_HEADER_SIZE
-		+ (1 + midi->in_ports) * USB_DT_MIDI_IN_SIZE
-		+ (1 + midi->out_ports) * USB_DT_MIDI_OUT_SIZE(1);
+		+ (midi->in_ports + midi->out_ports) *
+			(USB_DT_MIDI_IN_SIZE + USB_DT_MIDI_OUT_SIZE(1));
 	ms_header_desc.wTotalLength = cpu_to_le16(n);
 
 	midi_function[i++] = (struct usb_descriptor_header *) &ms_header_desc;
 
-	/* we have one embedded IN jack */
-	jack_in_emb_desc.bJackID = jack++;
-	midi_function[i++] = (struct usb_descriptor_header *) &jack_in_emb_desc;
-
-	/* and a dynamic amount of external IN jacks */
+	/* configure the external IN jacks, each linked to an embedded OUT jack */
 	for (n = 0; n < midi->in_ports; n++) {
-		struct usb_midi_in_jack_descriptor *ext = &jack_in_ext_desc[n];
+		struct usb_midi_in_jack_descriptor *in_ext = &jack_in_ext_desc[n];
+		struct usb_midi_out_jack_descriptor_1 *out_emb = &jack_out_emb_desc[n];
 
-		ext->bLength =			USB_DT_MIDI_IN_SIZE;
-		ext->bDescriptorType =		USB_DT_CS_INTERFACE;
-		ext->bDescriptorSubtype =	USB_MS_MIDI_IN_JACK;
-		ext->bJackType =		USB_MS_EXTERNAL;
-		ext->bJackID =			jack++;
-		ext->iJack =			0;
+		in_ext->bLength			= USB_DT_MIDI_IN_SIZE;
+		in_ext->bDescriptorType		= USB_DT_CS_INTERFACE;
+		in_ext->bDescriptorSubtype	= USB_MS_MIDI_IN_JACK;
+		in_ext->bJackType		= USB_MS_EXTERNAL;
+		in_ext->bJackID			= jack++;
+		in_ext->iJack			= 0;
+		midi_function[i++] = (struct usb_descriptor_header *) in_ext;
 
-		midi_function[i++] = (struct usb_descriptor_header *) ext;
+		out_emb->bLength		= USB_DT_MIDI_OUT_SIZE(1);
+		out_emb->bDescriptorType	= USB_DT_CS_INTERFACE;
+		out_emb->bDescriptorSubtype	= USB_MS_MIDI_OUT_JACK;
+		out_emb->bJackType		= USB_MS_EMBEDDED;
+		out_emb->bJackID		= jack++;
+		out_emb->bNrInputPins		= 1;
+		out_emb->pins[0].baSourcePin	= 1;
+		out_emb->pins[0].baSourceID	= in_ext->bJackID;
+		out_emb->iJack			= 0;
+		midi_function[i++] = (struct usb_descriptor_header *) out_emb;
+
+		/* link it to the endpoint */
+		ms_in_desc.baAssocJackID[n] = out_emb->bJackID;
 	}
 
-	/* one embedded OUT jack ... */
-	jack_out_emb_desc.bLength = USB_DT_MIDI_OUT_SIZE(midi->in_ports);
-	jack_out_emb_desc.bJackID = jack++;
-	jack_out_emb_desc.bNrInputPins = midi->in_ports;
-	/* ... which referencess all external IN jacks */
-	for (n = 0; n < midi->in_ports; n++) {
-		jack_out_emb_desc.pins[n].baSourceID = jack_in_ext_desc[n].bJackID;
-		jack_out_emb_desc.pins[n].baSourcePin =	1;
-	}
-
-	midi_function[i++] = (struct usb_descriptor_header *) &jack_out_emb_desc;
-
-	/* and multiple external OUT jacks ... */
+	/* configure the external OUT jacks, each linked to an embedded IN jack */
 	for (n = 0; n < midi->out_ports; n++) {
-		struct usb_midi_out_jack_descriptor_1 *ext = &jack_out_ext_desc[n];
-		int m;
+		struct usb_midi_in_jack_descriptor *in_emb = &jack_in_emb_desc[n];
+		struct usb_midi_out_jack_descriptor_1 *out_ext = &jack_out_ext_desc[n];
 
-		ext->bLength =			USB_DT_MIDI_OUT_SIZE(1);
-		ext->bDescriptorType =		USB_DT_CS_INTERFACE;
-		ext->bDescriptorSubtype =	USB_MS_MIDI_OUT_JACK;
-		ext->bJackType =		USB_MS_EXTERNAL;
-		ext->bJackID =			jack++;
-		ext->bNrInputPins =		1;
-		ext->iJack =			0;
-		/* ... which all reference the same embedded IN jack */
-		for (m = 0; m < midi->out_ports; m++) {
-			ext->pins[m].baSourceID =	jack_in_emb_desc.bJackID;
-			ext->pins[m].baSourcePin =	1;
-		}
+		in_emb->bLength			= USB_DT_MIDI_IN_SIZE;
+		in_emb->bDescriptorType		= USB_DT_CS_INTERFACE;
+		in_emb->bDescriptorSubtype	= USB_MS_MIDI_IN_JACK;
+		in_emb->bJackType		= USB_MS_EMBEDDED;
+		in_emb->bJackID			= jack++;
+		in_emb->iJack			= 0;
+		midi_function[i++] = (struct usb_descriptor_header *) in_emb;
 
-		midi_function[i++] = (struct usb_descriptor_header *) ext;
+		out_ext->bLength =		USB_DT_MIDI_OUT_SIZE(1);
+		out_ext->bDescriptorType =	USB_DT_CS_INTERFACE;
+		out_ext->bDescriptorSubtype =	USB_MS_MIDI_OUT_JACK;
+		out_ext->bJackType =		USB_MS_EXTERNAL;
+		out_ext->bJackID =		jack++;
+		out_ext->bNrInputPins =		1;
+		out_ext->iJack =		0;
+		out_ext->pins[0].baSourceID =	in_emb->bJackID;
+		out_ext->pins[0].baSourcePin =	1;
+		midi_function[i++] = (struct usb_descriptor_header *) out_ext;
+
+		/* link it to the endpoint */
+		ms_out_desc.baAssocJackID[n] = in_emb->bJackID;
 	}
 
 	/* configure the endpoint descriptors ... */
 	ms_out_desc.bLength = USB_DT_MS_ENDPOINT_SIZE(midi->in_ports);
 	ms_out_desc.bNumEmbMIDIJack = midi->in_ports;
-	for (n = 0; n < midi->in_ports; n++)
-		ms_out_desc.baAssocJackID[n] = jack_in_emb_desc.bJackID;
 
 	ms_in_desc.bLength = USB_DT_MS_ENDPOINT_SIZE(midi->out_ports);
 	ms_in_desc.bNumEmbMIDIJack = midi->out_ports;
-	for (n = 0; n < midi->out_ports; n++)
-		ms_in_desc.baAssocJackID[n] = jack_out_emb_desc.bJackID;
 
 	/* ... and add them to the list */
 	midi_function[i++] = (struct usb_descriptor_header *) &bulk_out_desc;
@@ -901,6 +891,8 @@
 		f->descriptors = usb_copy_descriptors(midi_function);
 	}
 
+	kfree(midi_function);
+
 	return 0;
 
 fail:
diff --git a/drivers/usb/gadget/f_phonet.c b/drivers/usb/gadget/f_phonet.c
index 3490770..16a509a 100644
--- a/drivers/usb/gadget/f_phonet.c
+++ b/drivers/usb/gadget/f_phonet.c
@@ -346,7 +346,7 @@
 		}
 
 		skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
-				skb->len == 0, req->actual);
+				skb->len <= 1, req->actual);
 		page = NULL;
 
 		if (req->actual < req->length) { /* Last fragment */
diff --git a/drivers/usb/gadget/f_serial.c b/drivers/usb/gadget/f_serial.c
index 91fdf79..cf33a8d 100644
--- a/drivers/usb/gadget/f_serial.c
+++ b/drivers/usb/gadget/f_serial.c
@@ -131,8 +131,8 @@
 	}
 	if (!gser->port.in->desc || !gser->port.out->desc) {
 		DBG(cdev, "activate generic ttyGS%d\n", gser->port_num);
-		if (!config_ep_by_speed(cdev->gadget, f, gser->port.in) ||
-		    !config_ep_by_speed(cdev->gadget, f, gser->port.out)) {
+		if (config_ep_by_speed(cdev->gadget, f, gser->port.in) ||
+		    config_ep_by_speed(cdev->gadget, f, gser->port.out)) {
 			gser->port.in->desc = NULL;
 			gser->port.out->desc = NULL;
 			return -EINVAL;
diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c
index f7e39b0..11b5196 100644
--- a/drivers/usb/gadget/file_storage.c
+++ b/drivers/usb/gadget/file_storage.c
@@ -859,7 +859,7 @@
 			if (ctrl->bRequestType != (USB_DIR_OUT |
 					USB_TYPE_CLASS | USB_RECIP_INTERFACE))
 				break;
-			if (w_index != 0 || w_value != 0) {
+			if (w_index != 0 || w_value != 0 || w_length != 0) {
 				value = -EDOM;
 				break;
 			}
@@ -875,7 +875,7 @@
 			if (ctrl->bRequestType != (USB_DIR_IN |
 					USB_TYPE_CLASS | USB_RECIP_INTERFACE))
 				break;
-			if (w_index != 0 || w_value != 0) {
+			if (w_index != 0 || w_value != 0 || w_length != 1) {
 				value = -EDOM;
 				break;
 			}
diff --git a/drivers/usb/gadget/fsl_mxc_udc.c b/drivers/usb/gadget/fsl_mxc_udc.c
index 43a49ec..dcbc0a2 100644
--- a/drivers/usb/gadget/fsl_mxc_udc.c
+++ b/drivers/usb/gadget/fsl_mxc_udc.c
@@ -16,6 +16,7 @@
 #include <linux/err.h>
 #include <linux/fsl_devices.h>
 #include <linux/platform_device.h>
+#include <linux/io.h>
 
 #include <mach/hardware.h>
 
@@ -88,7 +89,6 @@
 void fsl_udc_clk_finalize(struct platform_device *pdev)
 {
 	struct fsl_usb2_platform_data *pdata = pdev->dev.platform_data;
-#if defined(CONFIG_SOC_IMX35)
 	if (cpu_is_mx35()) {
 		unsigned int v;
 
@@ -101,7 +101,6 @@
 					USBPHYCTRL_OTGBASE_OFFSET));
 		}
 	}
-#endif
 
 	/* ULPI transceivers don't need usbpll */
 	if (pdata->phy_mode == FSL_USB2_PHY_ULPI) {
diff --git a/drivers/usb/gadget/fsl_qe_udc.c b/drivers/usb/gadget/fsl_qe_udc.c
index 2a03e4d..e00cf92 100644
--- a/drivers/usb/gadget/fsl_qe_udc.c
+++ b/drivers/usb/gadget/fsl_qe_udc.c
@@ -2336,8 +2336,7 @@
 	if (!udc_controller)
 		return -ENODEV;
 
-	if (!driver || (driver->speed != USB_SPEED_FULL
-			&& driver->speed != USB_SPEED_HIGH)
+	if (!driver || driver->speed < USB_SPEED_FULL
 			|| !bind || !driver->disconnect || !driver->setup)
 		return -EINVAL;
 
diff --git a/drivers/usb/gadget/fsl_udc_core.c b/drivers/usb/gadget/fsl_udc_core.c
index b2c44e1..dd28ef3 100644
--- a/drivers/usb/gadget/fsl_udc_core.c
+++ b/drivers/usb/gadget/fsl_udc_core.c
@@ -696,12 +696,31 @@
 		kfree(req);
 }
 
-/*-------------------------------------------------------------------------*/
+/* Actually add a dTD chain to an empty dQH and let go */
+static void fsl_prime_ep(struct fsl_ep *ep, struct ep_td_struct *td)
+{
+	struct ep_queue_head *qh = get_qh_by_ep(ep);
+
+	/* Write dQH next pointer and terminate bit to 0 */
+	qh->next_dtd_ptr = cpu_to_hc32(td->td_dma
+			& EP_QUEUE_HEAD_NEXT_POINTER_MASK);
+
+	/* Clear active and halt bit */
+	qh->size_ioc_int_sts &= cpu_to_hc32(~(EP_QUEUE_HEAD_STATUS_ACTIVE
+					| EP_QUEUE_HEAD_STATUS_HALT));
+
+	/* Ensure that updates to the QH will occur before priming. */
+	wmb();
+
+	/* Prime endpoint by writing correct bit to ENDPTPRIME */
+	fsl_writel(ep_is_in(ep) ? (1 << (ep_index(ep) + 16))
+			: (1 << (ep_index(ep))), &dr_regs->endpointprime);
+}
+
+/* Add dTD chain to the dQH of an EP */
 static void fsl_queue_td(struct fsl_ep *ep, struct fsl_req *req)
 {
-	int i = ep_index(ep) * 2 + ep_is_in(ep);
 	u32 temp, bitmask, tmp_stat;
-	struct ep_queue_head *dQH = &ep->udc->ep_qh[i];
 
 	/* VDBG("QH addr Register 0x%8x", dr_regs->endpointlistaddr);
 	VDBG("ep_qh[%d] addr is 0x%8x", i, (u32)&(ep->udc->ep_qh[i])); */
@@ -719,7 +738,7 @@
 			cpu_to_hc32(req->head->td_dma & DTD_ADDR_MASK);
 		/* Read prime bit, if 1 goto done */
 		if (fsl_readl(&dr_regs->endpointprime) & bitmask)
-			goto out;
+			return;
 
 		do {
 			/* Set ATDTW bit in USBCMD */
@@ -736,28 +755,10 @@
 		fsl_writel(temp & ~USB_CMD_ATDTW, &dr_regs->usbcmd);
 
 		if (tmp_stat)
-			goto out;
+			return;
 	}
 
-	/* Write dQH next pointer and terminate bit to 0 */
-	temp = req->head->td_dma & EP_QUEUE_HEAD_NEXT_POINTER_MASK;
-	dQH->next_dtd_ptr = cpu_to_hc32(temp);
-
-	/* Clear active and halt bit */
-	temp = cpu_to_hc32(~(EP_QUEUE_HEAD_STATUS_ACTIVE
-			| EP_QUEUE_HEAD_STATUS_HALT));
-	dQH->size_ioc_int_sts &= temp;
-
-	/* Ensure that updates to the QH will occur before priming. */
-	wmb();
-
-	/* Prime endpoint by writing 1 to ENDPTPRIME */
-	temp = ep_is_in(ep)
-		? (1 << (ep_index(ep) + 16))
-		: (1 << (ep_index(ep)));
-	fsl_writel(temp, &dr_regs->endpointprime);
-out:
-	return;
+	fsl_prime_ep(ep, req->head);
 }
 
 /* Fill in the dTD structure
@@ -877,7 +878,7 @@
 		VDBG("%s, bad ep", __func__);
 		return -EINVAL;
 	}
-	if (ep->desc->bmAttributes == USB_ENDPOINT_XFER_ISOC) {
+	if (usb_endpoint_xfer_isoc(ep->desc)) {
 		if (req->req.length > ep->ep.maxpacket)
 			return -EMSGSIZE;
 	}
@@ -973,25 +974,20 @@
 
 		/* The request isn't the last request in this ep queue */
 		if (req->queue.next != &ep->queue) {
-			struct ep_queue_head *qh;
 			struct fsl_req *next_req;
 
-			qh = ep->qh;
 			next_req = list_entry(req->queue.next, struct fsl_req,
 					queue);
 
-			/* Point the QH to the first TD of next request */
-			fsl_writel((u32) next_req->head, &qh->curr_dtd_ptr);
+			/* prime with dTD of next request */
+			fsl_prime_ep(ep, next_req->head);
 		}
-
-		/* The request hasn't been processed, patch up the TD chain */
+	/* The request hasn't been processed, patch up the TD chain */
 	} else {
 		struct fsl_req *prev_req;
 
 		prev_req = list_entry(req->queue.prev, struct fsl_req, queue);
-		fsl_writel(fsl_readl(&req->tail->next_td_ptr),
-				&prev_req->tail->next_td_ptr);
-
+		prev_req->tail->next_td_ptr = req->tail->next_td_ptr;
 	}
 
 	done(ep, req, -ECONNRESET);
@@ -1032,7 +1028,7 @@
 		goto out;
 	}
 
-	if (ep->desc->bmAttributes == USB_ENDPOINT_XFER_ISOC) {
+	if (usb_endpoint_xfer_isoc(ep->desc)) {
 		status = -EOPNOTSUPP;
 		goto out;
 	}
@@ -1068,7 +1064,7 @@
 	struct fsl_udc *udc;
 	int size = 0;
 	u32 bitmask;
-	struct ep_queue_head *d_qh;
+	struct ep_queue_head *qh;
 
 	ep = container_of(_ep, struct fsl_ep, ep);
 	if (!_ep || (!ep->desc && ep_index(ep) != 0))
@@ -1079,13 +1075,13 @@
 	if (!udc->driver || udc->gadget.speed == USB_SPEED_UNKNOWN)
 		return -ESHUTDOWN;
 
-	d_qh = &ep->udc->ep_qh[ep_index(ep) * 2 + ep_is_in(ep)];
+	qh = get_qh_by_ep(ep);
 
 	bitmask = (ep_is_in(ep)) ? (1 << (ep_index(ep) + 16)) :
 	    (1 << (ep_index(ep)));
 
 	if (fsl_readl(&dr_regs->endptstatus) & bitmask)
-		size = (d_qh->size_ioc_int_sts & DTD_PACKET_SIZE)
+		size = (qh->size_ioc_int_sts & DTD_PACKET_SIZE)
 		    >> DTD_LENGTH_BIT_POS;
 
 	pr_debug("%s %u\n", __func__, size);
@@ -1717,7 +1713,7 @@
 
 static inline enum usb_device_speed portscx_device_speed(u32 reg)
 {
-	switch (speed & PORTSCX_PORT_SPEED_MASK) {
+	switch (reg & PORTSCX_PORT_SPEED_MASK) {
 	case PORTSCX_PORT_SPEED_HIGH:
 		return USB_SPEED_HIGH;
 	case PORTSCX_PORT_SPEED_FULL:
@@ -1938,8 +1934,7 @@
 	if (!udc_controller)
 		return -ENODEV;
 
-	if (!driver || (driver->speed != USB_SPEED_FULL
-				&& driver->speed != USB_SPEED_HIGH)
+	if (!driver || driver->speed < USB_SPEED_FULL
 			|| !bind || !driver->disconnect || !driver->setup)
 		return -EINVAL;
 
@@ -2480,8 +2475,7 @@
 
 #ifndef CONFIG_ARCH_MXC
 	if (pdata->have_sysif_regs)
-		usb_sys_regs = (struct usb_sys_interface *)
-				((u32)dr_regs + USB_DR_SYS_OFFSET);
+		usb_sys_regs = (void *)dr_regs + USB_DR_SYS_OFFSET;
 #endif
 
 	/* Initialize USB clocks */
diff --git a/drivers/usb/gadget/fsl_usb2_udc.h b/drivers/usb/gadget/fsl_usb2_udc.h
index 1d51be8..f781f5d 100644
--- a/drivers/usb/gadget/fsl_usb2_udc.h
+++ b/drivers/usb/gadget/fsl_usb2_udc.h
@@ -569,6 +569,16 @@
 					* 2 + ((windex & USB_DIR_IN) ? 1 : 0))
 #define get_pipe_by_ep(EP)	(ep_index(EP) * 2 + ep_is_in(EP))
 
+static inline struct ep_queue_head *get_qh_by_ep(struct fsl_ep *ep)
+{
+	/* we only have one ep0 structure but two queue heads */
+	if (ep_index(ep) != 0)
+		return ep->qh;
+	else
+		return &ep->udc->ep_qh[(ep->udc->ep0_dir ==
+				USB_DIR_IN) ? 1 : 0];
+}
+
 struct platform_device;
 #ifdef CONFIG_ARCH_MXC
 int fsl_udc_clk_init(struct platform_device *pdev);
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index a392ec0..6ccae27 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -1730,8 +1730,9 @@
 gadgetfs_disconnect (struct usb_gadget *gadget)
 {
 	struct dev_data		*dev = get_gadget_data (gadget);
+	unsigned long		flags;
 
-	spin_lock (&dev->lock);
+	spin_lock_irqsave (&dev->lock, flags);
 	if (dev->state == STATE_DEV_UNCONNECTED)
 		goto exit;
 	dev->state = STATE_DEV_UNCONNECTED;
@@ -1740,7 +1741,7 @@
 	next_event (dev, GADGETFS_DISCONNECT);
 	ep0_readable (dev);
 exit:
-	spin_unlock (&dev->lock);
+	spin_unlock_irqrestore (&dev->lock, flags);
 }
 
 static void
diff --git a/drivers/usb/gadget/m66592-udc.c b/drivers/usb/gadget/m66592-udc.c
index 91d0af2..9aa1cbb 100644
--- a/drivers/usb/gadget/m66592-udc.c
+++ b/drivers/usb/gadget/m66592-udc.c
@@ -1472,7 +1472,7 @@
 	int retval;
 
 	if (!driver
-			|| driver->speed != USB_SPEED_HIGH
+			|| driver->speed < USB_SPEED_HIGH
 			|| !bind
 			|| !driver->setup)
 		return -EINVAL;
diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c
index 7f1bc9a..da2b9d0 100644
--- a/drivers/usb/gadget/net2280.c
+++ b/drivers/usb/gadget/net2280.c
@@ -1881,7 +1881,7 @@
 	 * (dev->usb->xcvrdiag & FORCE_FULL_SPEED_MODE)
 	 * "must not be used in normal operation"
 	 */
-	if (!driver || driver->speed != USB_SPEED_HIGH
+	if (!driver || driver->speed < USB_SPEED_HIGH
 			|| !driver->setup)
 		return -EINVAL;
 
diff --git a/drivers/usb/gadget/pch_udc.c b/drivers/usb/gadget/pch_udc.c
index 550d6dc..5048a0c 100644
--- a/drivers/usb/gadget/pch_udc.c
+++ b/drivers/usb/gadget/pch_udc.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2010 OKI SEMICONDUCTOR CO., LTD.
+ * Copyright (C) 2011 LAPIS Semiconductor Co., Ltd.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -354,6 +354,7 @@
 #define PCI_DEVICE_ID_INTEL_EG20T_UDC	0x8808
 #define PCI_VENDOR_ID_ROHM		0x10DB
 #define PCI_DEVICE_ID_ML7213_IOH_UDC	0x801D
+#define PCI_DEVICE_ID_ML7831_IOH_UDC	0x8808
 
 static const char	ep0_string[] = "ep0in";
 static DEFINE_SPINLOCK(udc_stall_spinlock);	/* stall spin lock */
@@ -2970,6 +2971,11 @@
 		.class = (PCI_CLASS_SERIAL_USB << 8) | 0xfe,
 		.class_mask = 0xffffffff,
 	},
+	{
+		PCI_DEVICE(PCI_VENDOR_ID_ROHM, PCI_DEVICE_ID_ML7831_IOH_UDC),
+		.class = (PCI_CLASS_SERIAL_USB << 8) | 0xfe,
+		.class_mask = 0xffffffff,
+	},
 	{ 0 },
 };
 
@@ -2999,5 +3005,5 @@
 module_exit(pch_udc_pci_exit);
 
 MODULE_DESCRIPTION("Intel EG20T USB Device Controller");
-MODULE_AUTHOR("OKI SEMICONDUCTOR, <toshiharu-linux@dsn.okisemi.com>");
+MODULE_AUTHOR("LAPIS Semiconductor, <tomoya-linux@dsn.lapis-semi.com>");
 MODULE_LICENSE("GPL");
diff --git a/drivers/usb/gadget/r8a66597-udc.c b/drivers/usb/gadget/r8a66597-udc.c
index 68a826a..fc719a3 100644
--- a/drivers/usb/gadget/r8a66597-udc.c
+++ b/drivers/usb/gadget/r8a66597-udc.c
@@ -1718,6 +1718,8 @@
 	if (list_empty(&ep->queue) && !ep->busy) {
 		pipe_stop(ep->r8a66597, ep->pipenum);
 		r8a66597_bclr(ep->r8a66597, BCLR, ep->fifoctr);
+		r8a66597_write(ep->r8a66597, ACLRM, ep->pipectr);
+		r8a66597_write(ep->r8a66597, 0, ep->pipectr);
 	}
 	spin_unlock_irqrestore(&ep->r8a66597->lock, flags);
 }
@@ -1742,26 +1744,16 @@
 		struct usb_gadget_driver *driver)
 {
 	struct r8a66597 *r8a66597 = gadget_to_r8a66597(gadget);
-	int retval;
 
 	if (!driver
-			|| driver->speed != USB_SPEED_HIGH
+			|| driver->speed < USB_SPEED_HIGH
 			|| !driver->setup)
 		return -EINVAL;
 	if (!r8a66597)
 		return -ENODEV;
 
 	/* hook up the driver */
-	driver->driver.bus = NULL;
 	r8a66597->driver = driver;
-	r8a66597->gadget.dev.driver = &driver->driver;
-
-	retval = device_add(&r8a66597->gadget.dev);
-	if (retval) {
-		dev_err(r8a66597_to_dev(r8a66597), "device_add error (%d)\n",
-			retval);
-		goto error;
-	}
 
 	init_controller(r8a66597);
 	r8a66597_bset(r8a66597, VBSE, INTENB0);
@@ -1775,12 +1767,6 @@
 	}
 
 	return 0;
-
-error:
-	r8a66597->driver = NULL;
-	r8a66597->gadget.dev.driver = NULL;
-
-	return retval;
 }
 
 static int r8a66597_stop(struct usb_gadget *gadget,
@@ -1794,7 +1780,6 @@
 	disable_controller(r8a66597);
 	spin_unlock_irqrestore(&r8a66597->lock, flags);
 
-	device_del(&r8a66597->gadget.dev);
 	r8a66597->driver = NULL;
 	return 0;
 }
@@ -1845,6 +1830,7 @@
 		clk_put(r8a66597->clk);
 	}
 #endif
+	device_unregister(&r8a66597->gadget.dev);
 	kfree(r8a66597);
 	return 0;
 }
@@ -1924,13 +1910,17 @@
 	r8a66597->irq_sense_low = irq_trigger == IRQF_TRIGGER_LOW;
 
 	r8a66597->gadget.ops = &r8a66597_gadget_ops;
-	device_initialize(&r8a66597->gadget.dev);
 	dev_set_name(&r8a66597->gadget.dev, "gadget");
 	r8a66597->gadget.is_dualspeed = 1;
 	r8a66597->gadget.dev.parent = &pdev->dev;
 	r8a66597->gadget.dev.dma_mask = pdev->dev.dma_mask;
 	r8a66597->gadget.dev.release = pdev->dev.release;
 	r8a66597->gadget.name = udc_name;
+	ret = device_register(&r8a66597->gadget.dev);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "device_register failed\n");
+		goto clean_up;
+	}
 
 	init_timer(&r8a66597->timer);
 	r8a66597->timer.function = r8a66597_timer;
@@ -1945,7 +1935,7 @@
 			dev_err(&pdev->dev, "cannot get clock \"%s\"\n",
 				clk_name);
 			ret = PTR_ERR(r8a66597->clk);
-			goto clean_up;
+			goto clean_up_dev;
 		}
 		clk_enable(r8a66597->clk);
 	}
@@ -2014,7 +2004,9 @@
 		clk_disable(r8a66597->clk);
 		clk_put(r8a66597->clk);
 	}
+clean_up_dev:
 #endif
+	device_unregister(&r8a66597->gadget.dev);
 clean_up:
 	if (r8a66597) {
 		if (r8a66597->sudmac_reg)
diff --git a/drivers/usb/gadget/s3c-hsotg.c b/drivers/usb/gadget/s3c-hsotg.c
index a552453..b314482 100644
--- a/drivers/usb/gadget/s3c-hsotg.c
+++ b/drivers/usb/gadget/s3c-hsotg.c
@@ -2586,10 +2586,8 @@
 		return -EINVAL;
 	}
 
-	if (driver->speed != USB_SPEED_HIGH &&
-	    driver->speed != USB_SPEED_FULL) {
+	if (driver->speed < USB_SPEED_FULL)
 		dev_err(hsotg->dev, "%s: bad speed\n", __func__);
-	}
 
 	if (!bind || !driver->setup) {
 		dev_err(hsotg->dev, "%s: missing entry points\n", __func__);
diff --git a/drivers/usb/gadget/s3c-hsudc.c b/drivers/usb/gadget/s3c-hsudc.c
index 8d54f89..20a553b 100644
--- a/drivers/usb/gadget/s3c-hsudc.c
+++ b/drivers/usb/gadget/s3c-hsudc.c
@@ -1142,8 +1142,7 @@
 	int ret;
 
 	if (!driver
-		|| (driver->speed != USB_SPEED_FULL &&
-			driver->speed != USB_SPEED_HIGH)
+		|| driver->speed < USB_SPEED_FULL
 		|| !bind
 		|| !driver->unbind || !driver->disconnect || !driver->setup)
 		return -EINVAL;
diff --git a/drivers/usb/gadget/udc-core.c b/drivers/usb/gadget/udc-core.c
index 022baec..6939e17 100644
--- a/drivers/usb/gadget/udc-core.c
+++ b/drivers/usb/gadget/udc-core.c
@@ -210,10 +210,10 @@
 	kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE);
 
 	if (udc_is_newstyle(udc)) {
-		usb_gadget_disconnect(udc->gadget);
+		udc->driver->disconnect(udc->gadget);
 		udc->driver->unbind(udc->gadget);
 		usb_gadget_udc_stop(udc->gadget, udc->driver);
-
+		usb_gadget_disconnect(udc->gadget);
 	} else {
 		usb_gadget_stop(udc->gadget, udc->driver);
 	}
@@ -344,7 +344,7 @@
 static ssize_t usb_udc_srp_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t n)
 {
-	struct usb_udc		*udc = dev_get_drvdata(dev);
+	struct usb_udc		*udc = container_of(dev, struct usb_udc, dev);
 
 	if (sysfs_streq(buf, "1"))
 		usb_gadget_wakeup(udc->gadget);
@@ -378,7 +378,7 @@
 	return snprintf(buf, PAGE_SIZE, "%s\n",
 			usb_speed_string(udc->gadget->speed));
 }
-static DEVICE_ATTR(speed, S_IRUSR, usb_udc_speed_show, NULL);
+static DEVICE_ATTR(speed, S_IRUGO, usb_udc_speed_show, NULL);
 
 #define USB_UDC_ATTR(name)					\
 ssize_t usb_udc_##name##_show(struct device *dev,		\
@@ -389,7 +389,7 @@
 								\
 	return snprintf(buf, PAGE_SIZE, "%d\n", gadget->name);	\
 }								\
-static DEVICE_ATTR(name, S_IRUSR, usb_udc_##name##_show, NULL)
+static DEVICE_ATTR(name, S_IRUGO, usb_udc_##name##_show, NULL)
 
 static USB_UDC_ATTR(is_dualspeed);
 static USB_UDC_ATTR(is_otg);
diff --git a/drivers/usb/host/ehci-sched.c b/drivers/usb/host/ehci-sched.c
index 2e829fa..a60679c 100644
--- a/drivers/usb/host/ehci-sched.c
+++ b/drivers/usb/host/ehci-sched.c
@@ -1475,30 +1475,36 @@
 	 * jump until after the queue is primed.
 	 */
 	else {
+		int done = 0;
 		start = SCHEDULE_SLOP + (now & ~0x07);
 
 		/* NOTE:  assumes URB_ISO_ASAP, to limit complexity/bugs */
 
-		/* find a uframe slot with enough bandwidth */
-		next = start + period;
-		for (; start < next; start++) {
-
+		/* find a uframe slot with enough bandwidth.
+		 * Early uframes are more precious because full-speed
+		 * iso IN transfers can't use late uframes,
+		 * and therefore they should be allocated last.
+		 */
+		next = start;
+		start += period;
+		do {
+			start--;
 			/* check schedule: enough space? */
 			if (stream->highspeed) {
 				if (itd_slot_ok(ehci, mod, start,
 						stream->usecs, period))
-					break;
+					done = 1;
 			} else {
 				if ((start % 8) >= 6)
 					continue;
 				if (sitd_slot_ok(ehci, mod, stream,
 						start, sched, period))
-					break;
+					done = 1;
 			}
-		}
+		} while (start > next && !done);
 
 		/* no room in the schedule */
-		if (start == next) {
+		if (!done) {
 			ehci_dbg(ehci, "iso resched full %p (now %d max %d)\n",
 				urb, now, now + mod);
 			status = -ENOSPC;
diff --git a/drivers/usb/host/ehci-xls.c b/drivers/usb/host/ehci-xls.c
index fe74bd6..b4fb511 100644
--- a/drivers/usb/host/ehci-xls.c
+++ b/drivers/usb/host/ehci-xls.c
@@ -19,7 +19,7 @@
 
 	ehci->caps = hcd->regs;
 	ehci->regs = hcd->regs +
-		HC_LENGTH(ehci_readl(ehci, &ehci->caps->hc_capbase));
+		HC_LENGTH(ehci, ehci_readl(ehci, &ehci->caps->hc_capbase));
 	dbg_hcs_params(ehci, "reset");
 	dbg_hcc_params(ehci, "reset");
 
diff --git a/drivers/usb/host/ohci-at91.c b/drivers/usb/host/ohci-at91.c
index ba3a46b..95a9fec 100644
--- a/drivers/usb/host/ohci-at91.c
+++ b/drivers/usb/host/ohci-at91.c
@@ -223,6 +223,9 @@
 	if (port < 0 || port >= 2)
 		return;
 
+	if (pdata->vbus_pin[port] <= 0)
+		return;
+
 	gpio_set_value(pdata->vbus_pin[port], !pdata->vbus_pin_inverted ^ enable);
 }
 
@@ -231,6 +234,9 @@
 	if (port < 0 || port >= 2)
 		return -EINVAL;
 
+	if (pdata->vbus_pin[port] <= 0)
+		return -EINVAL;
+
 	return gpio_get_value(pdata->vbus_pin[port]) ^ !pdata->vbus_pin_inverted;
 }
 
diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index 34efd47..b263919 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -389,17 +389,14 @@
 	struct ohci_hcd *ohci;
 
 	ohci = hcd_to_ohci (hcd);
-	ohci_writel (ohci, OHCI_INTR_MIE, &ohci->regs->intrdisable);
-	ohci->hc_control = ohci_readl(ohci, &ohci->regs->control);
+	ohci_writel(ohci, (u32) ~0, &ohci->regs->intrdisable);
 
-	/* If the SHUTDOWN quirk is set, don't put the controller in RESET */
-	ohci->hc_control &= (ohci->flags & OHCI_QUIRK_SHUTDOWN ?
-			OHCI_CTRL_RWC | OHCI_CTRL_HCFS :
-			OHCI_CTRL_RWC);
-	ohci_writel(ohci, ohci->hc_control, &ohci->regs->control);
+	/* Software reset, after which the controller goes into SUSPEND */
+	ohci_writel(ohci, OHCI_HCR, &ohci->regs->cmdstatus);
+	ohci_readl(ohci, &ohci->regs->cmdstatus);	/* flush the writes */
+	udelay(10);
 
-	/* flush the writes */
-	(void) ohci_readl (ohci, &ohci->regs->control);
+	ohci_writel(ohci, ohci->fminterval, &ohci->regs->fminterval);
 }
 
 static int check_ed(struct ohci_hcd *ohci, struct ed *ed)
diff --git a/drivers/usb/host/ohci-pci.c b/drivers/usb/host/ohci-pci.c
index ad8166c..bc01b06 100644
--- a/drivers/usb/host/ohci-pci.c
+++ b/drivers/usb/host/ohci-pci.c
@@ -175,28 +175,6 @@
 	return 0;
 }
 
-/* nVidia controllers continue to drive Reset signalling on the bus
- * even after system shutdown, wasting power.  This flag tells the
- * shutdown routine to leave the controller OPERATIONAL instead of RESET.
- */
-static int ohci_quirk_nvidia_shutdown(struct usb_hcd *hcd)
-{
-	struct pci_dev *pdev = to_pci_dev(hcd->self.controller);
-	struct ohci_hcd	*ohci = hcd_to_ohci(hcd);
-
-	/* Evidently nVidia fixed their later hardware; this is a guess at
-	 * the changeover point.
-	 */
-#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_USB		0x026d
-
-	if (pdev->device < PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_USB) {
-		ohci->flags |= OHCI_QUIRK_SHUTDOWN;
-		ohci_dbg(ohci, "enabled nVidia shutdown quirk\n");
-	}
-
-	return 0;
-}
-
 static void sb800_prefetch(struct ohci_hcd *ohci, int on)
 {
 	struct pci_dev *pdev;
@@ -260,10 +238,6 @@
 		PCI_DEVICE(PCI_VENDOR_ID_ATI, 0x4399),
 		.driver_data = (unsigned long)ohci_quirk_amd700,
 	},
-	{
-		PCI_DEVICE(PCI_VENDOR_ID_NVIDIA, PCI_ANY_ID),
-		.driver_data = (unsigned long) ohci_quirk_nvidia_shutdown,
-	},
 
 	/* FIXME for some of the early AMD 760 southbridges, OHCI
 	 * won't work at all.  blacklist them.
diff --git a/drivers/usb/host/ohci.h b/drivers/usb/host/ohci.h
index 35e5fd6..0795b93 100644
--- a/drivers/usb/host/ohci.h
+++ b/drivers/usb/host/ohci.h
@@ -403,7 +403,6 @@
 #define	OHCI_QUIRK_HUB_POWER	0x100			/* distrust firmware power/oc setup */
 #define	OHCI_QUIRK_AMD_PLL	0x200			/* AMD PLL quirk*/
 #define	OHCI_QUIRK_AMD_PREFETCH	0x400			/* pre-fetch for ISO transfer */
-#define	OHCI_QUIRK_SHUTDOWN	0x800			/* nVidia power bug */
 	// there are also chip quirks/bugs in init logic
 
 	struct work_struct	nec_work;	/* Worker for NEC quirk */
diff --git a/drivers/usb/host/pci-quirks.c b/drivers/usb/host/pci-quirks.c
index 27a3dec..caf8742 100644
--- a/drivers/usb/host/pci-quirks.c
+++ b/drivers/usb/host/pci-quirks.c
@@ -37,6 +37,7 @@
 #define OHCI_INTRENABLE		0x10
 #define OHCI_INTRDISABLE	0x14
 #define OHCI_FMINTERVAL		0x34
+#define OHCI_HCFS		(3 << 6)	/* hc functional state */
 #define OHCI_HCR		(1 << 0)	/* host controller reset */
 #define OHCI_OCR		(1 << 3)	/* ownership change request */
 #define OHCI_CTRL_RWC		(1 << 9)	/* remote wakeup connected */
@@ -466,6 +467,8 @@
 {
 	void __iomem *base;
 	u32 control;
+	u32 fminterval;
+	int cnt;
 
 	if (!mmio_resource_enabled(pdev, 0))
 		return;
@@ -498,41 +501,32 @@
 	}
 #endif
 
-	/* reset controller, preserving RWC (and possibly IR) */
-	writel(control & OHCI_CTRL_MASK, base + OHCI_CONTROL);
-	readl(base + OHCI_CONTROL);
+	/* disable interrupts */
+	writel((u32) ~0, base + OHCI_INTRDISABLE);
 
-	/* Some NVIDIA controllers stop working if kept in RESET for too long */
-	if (pdev->vendor == PCI_VENDOR_ID_NVIDIA) {
-		u32 fminterval;
-		int cnt;
+	/* Reset the USB bus, if the controller isn't already in RESET */
+	if (control & OHCI_HCFS) {
+		/* Go into RESET, preserving RWC (and possibly IR) */
+		writel(control & OHCI_CTRL_MASK, base + OHCI_CONTROL);
+		readl(base + OHCI_CONTROL);
 
-		/* drive reset for at least 50 ms (7.1.7.5) */
+		/* drive bus reset for at least 50 ms (7.1.7.5) */
 		msleep(50);
-
-		/* software reset of the controller, preserving HcFmInterval */
-		fminterval = readl(base + OHCI_FMINTERVAL);
-		writel(OHCI_HCR, base + OHCI_CMDSTATUS);
-
-		/* reset requires max 10 us delay */
-		for (cnt = 30; cnt > 0; --cnt) {	/* ... allow extra time */
-			if ((readl(base + OHCI_CMDSTATUS) & OHCI_HCR) == 0)
-				break;
-			udelay(1);
-		}
-		writel(fminterval, base + OHCI_FMINTERVAL);
-
-		/* Now we're in the SUSPEND state with all devices reset
-		 * and wakeups and interrupts disabled
-		 */
 	}
 
-	/*
-	 * disable interrupts
-	 */
-	writel(~(u32)0, base + OHCI_INTRDISABLE);
-	writel(~(u32)0, base + OHCI_INTRSTATUS);
+	/* software reset of the controller, preserving HcFmInterval */
+	fminterval = readl(base + OHCI_FMINTERVAL);
+	writel(OHCI_HCR, base + OHCI_CMDSTATUS);
 
+	/* reset requires max 10 us delay */
+	for (cnt = 30; cnt > 0; --cnt) {	/* ... allow extra time */
+		if ((readl(base + OHCI_CMDSTATUS) & OHCI_HCR) == 0)
+			break;
+		udelay(1);
+	}
+	writel(fminterval, base + OHCI_FMINTERVAL);
+
+	/* Now the controller is safely in SUSPEND and nothing can wake it up */
 	iounmap(base);
 }
 
@@ -627,7 +621,7 @@
 	void __iomem *base, *op_reg_base;
 	u32	hcc_params, cap, val;
 	u8	offset, cap_length;
-	int	wait_time, delta, count = 256/4;
+	int	wait_time, count = 256/4;
 
 	if (!mmio_resource_enabled(pdev, 0))
 		return;
@@ -673,11 +667,10 @@
 		writel(val, op_reg_base + EHCI_USBCMD);
 
 		wait_time = 2000;
-		delta = 100;
 		do {
 			writel(0x3f, op_reg_base + EHCI_USBSTS);
-			udelay(delta);
-			wait_time -= delta;
+			udelay(100);
+			wait_time -= 100;
 			val = readl(op_reg_base + EHCI_USBSTS);
 			if ((val == ~(u32)0) || (val & EHCI_USBSTS_HALTED)) {
 				break;
diff --git a/drivers/usb/host/whci/qset.c b/drivers/usb/host/whci/qset.c
index d6e1754..a403b53 100644
--- a/drivers/usb/host/whci/qset.c
+++ b/drivers/usb/host/whci/qset.c
@@ -124,7 +124,7 @@
 {
 	qset->td_start = qset->td_end = qset->ntds = 0;
 
-	qset->qh.link = cpu_to_le32(QH_LINK_NTDS(8) | QH_LINK_T);
+	qset->qh.link = cpu_to_le64(QH_LINK_NTDS(8) | QH_LINK_T);
 	qset->qh.status = qset->qh.status & QH_STATUS_SEQ_MASK;
 	qset->qh.err_count = 0;
 	qset->qh.scratch[0] = 0;
diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c
index 42a22b8..0e4b25f 100644
--- a/drivers/usb/host/xhci-mem.c
+++ b/drivers/usb/host/xhci-mem.c
@@ -982,7 +982,6 @@
 	struct xhci_virt_device *dev;
 	struct xhci_ep_ctx	*ep0_ctx;
 	struct xhci_slot_ctx    *slot_ctx;
-	struct xhci_input_control_ctx *ctrl_ctx;
 	u32			port_num;
 	struct usb_device *top_dev;
 
@@ -994,12 +993,8 @@
 		return -EINVAL;
 	}
 	ep0_ctx = xhci_get_ep_ctx(xhci, dev->in_ctx, 0);
-	ctrl_ctx = xhci_get_input_control_ctx(xhci, dev->in_ctx);
 	slot_ctx = xhci_get_slot_ctx(xhci, dev->in_ctx);
 
-	/* 2) New slot context and endpoint 0 context are valid*/
-	ctrl_ctx->add_flags = cpu_to_le32(SLOT_FLAG | EP0_FLAG);
-
 	/* 3) Only the control endpoint is valid - one endpoint context */
 	slot_ctx->dev_info |= cpu_to_le32(LAST_CTX(1) | udev->route);
 	switch (udev->speed) {
diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c
index 940321b..9f1d4b1 100644
--- a/drivers/usb/host/xhci-ring.c
+++ b/drivers/usb/host/xhci-ring.c
@@ -816,23 +816,24 @@
 	struct xhci_ring *ring;
 	struct xhci_td *cur_td;
 	int ret, i, j;
+	unsigned long flags;
 
 	ep = (struct xhci_virt_ep *) arg;
 	xhci = ep->xhci;
 
-	spin_lock(&xhci->lock);
+	spin_lock_irqsave(&xhci->lock, flags);
 
 	ep->stop_cmds_pending--;
 	if (xhci->xhc_state & XHCI_STATE_DYING) {
 		xhci_dbg(xhci, "Stop EP timer ran, but another timer marked "
 				"xHCI as DYING, exiting.\n");
-		spin_unlock(&xhci->lock);
+		spin_unlock_irqrestore(&xhci->lock, flags);
 		return;
 	}
 	if (!(ep->stop_cmds_pending == 0 && (ep->ep_state & EP_HALT_PENDING))) {
 		xhci_dbg(xhci, "Stop EP timer ran, but no command pending, "
 				"exiting.\n");
-		spin_unlock(&xhci->lock);
+		spin_unlock_irqrestore(&xhci->lock, flags);
 		return;
 	}
 
@@ -844,11 +845,11 @@
 	xhci->xhc_state |= XHCI_STATE_DYING;
 	/* Disable interrupts from the host controller and start halting it */
 	xhci_quiesce(xhci);
-	spin_unlock(&xhci->lock);
+	spin_unlock_irqrestore(&xhci->lock, flags);
 
 	ret = xhci_halt(xhci);
 
-	spin_lock(&xhci->lock);
+	spin_lock_irqsave(&xhci->lock, flags);
 	if (ret < 0) {
 		/* This is bad; the host is not responding to commands and it's
 		 * not allowing itself to be halted.  At least interrupts are
@@ -896,7 +897,7 @@
 			}
 		}
 	}
-	spin_unlock(&xhci->lock);
+	spin_unlock_irqrestore(&xhci->lock, flags);
 	xhci_dbg(xhci, "Calling usb_hc_died()\n");
 	usb_hc_died(xhci_to_hcd(xhci)->primary_hcd);
 	xhci_dbg(xhci, "xHCI host controller is dead.\n");
diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index 1ff95a0..a1afb7c 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -711,7 +711,10 @@
 	ring = xhci->cmd_ring;
 	seg = ring->deq_seg;
 	do {
-		memset(seg->trbs, 0, SEGMENT_SIZE);
+		memset(seg->trbs, 0,
+			sizeof(union xhci_trb) * (TRBS_PER_SEGMENT - 1));
+		seg->trbs[TRBS_PER_SEGMENT - 1].link.control &=
+			cpu_to_le32(~TRB_CYCLE);
 		seg = seg->next;
 	} while (seg != ring->deq_seg);
 
@@ -799,7 +802,7 @@
 	u32			command, temp = 0;
 	struct usb_hcd		*hcd = xhci_to_hcd(xhci);
 	struct usb_hcd		*secondary_hcd;
-	int			retval;
+	int			retval = 0;
 
 	/* Wait a bit if either of the roothubs need to settle from the
 	 * transition into bus suspend.
@@ -809,6 +812,9 @@
 				xhci->bus_state[1].next_statechange))
 		msleep(100);
 
+	set_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
+	set_bit(HCD_FLAG_HW_ACCESSIBLE, &xhci->shared_hcd->flags);
+
 	spin_lock_irq(&xhci->lock);
 	if (xhci->quirks & XHCI_RESET_ON_RESUME)
 		hibernated = true;
@@ -878,20 +884,13 @@
 			return retval;
 		xhci_dbg(xhci, "Start the primary HCD\n");
 		retval = xhci_run(hcd->primary_hcd);
-		if (retval)
-			goto failed_restart;
-
-		xhci_dbg(xhci, "Start the secondary HCD\n");
-		retval = xhci_run(secondary_hcd);
 		if (!retval) {
-			set_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
-			set_bit(HCD_FLAG_HW_ACCESSIBLE,
-					&xhci->shared_hcd->flags);
+			xhci_dbg(xhci, "Start the secondary HCD\n");
+			retval = xhci_run(secondary_hcd);
 		}
-failed_restart:
 		hcd->state = HC_STATE_SUSPENDED;
 		xhci->shared_hcd->state = HC_STATE_SUSPENDED;
-		return retval;
+		goto done;
 	}
 
 	/* step 4: set Run/Stop bit */
@@ -910,11 +909,14 @@
 	 * Running endpoints by ringing their doorbells
 	 */
 
-	set_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
-	set_bit(HCD_FLAG_HW_ACCESSIBLE, &xhci->shared_hcd->flags);
-
 	spin_unlock_irq(&xhci->lock);
-	return 0;
+
+ done:
+	if (retval == 0) {
+		usb_hcd_resume_root_hub(hcd);
+		usb_hcd_resume_root_hub(xhci->shared_hcd);
+	}
+	return retval;
 }
 #endif	/* CONFIG_PM */
 
@@ -3504,6 +3506,10 @@
 	/* Otherwise, update the control endpoint ring enqueue pointer. */
 	else
 		xhci_copy_ep0_dequeue_into_input_ctx(xhci, udev);
+	ctrl_ctx = xhci_get_input_control_ctx(xhci, virt_dev->in_ctx);
+	ctrl_ctx->add_flags = cpu_to_le32(SLOT_FLAG | EP0_FLAG);
+	ctrl_ctx->drop_flags = 0;
+
 	xhci_dbg(xhci, "Slot ID %d Input Context:\n", udev->slot_id);
 	xhci_dbg_ctx(xhci, virt_dev->in_ctx, 2);
 
@@ -3585,7 +3591,6 @@
 	virt_dev->address = (le32_to_cpu(slot_ctx->dev_state) & DEV_ADDR_MASK)
 		+ 1;
 	/* Zero the input context control for later use */
-	ctrl_ctx = xhci_get_input_control_ctx(xhci, virt_dev->in_ctx);
 	ctrl_ctx->add_flags = 0;
 	ctrl_ctx->drop_flags = 0;
 
diff --git a/drivers/usb/musb/Kconfig b/drivers/usb/musb/Kconfig
index fc34b8b..07a0346 100644
--- a/drivers/usb/musb/Kconfig
+++ b/drivers/usb/musb/Kconfig
@@ -11,6 +11,7 @@
 	select TWL4030_USB if MACH_OMAP_3430SDP
 	select TWL6030_USB if MACH_OMAP_4430SDP || MACH_OMAP4_PANDA
 	select USB_OTG_UTILS
+	select USB_GADGET_DUALSPEED
 	tristate 'Inventra Highspeed Dual Role Controller (TI, ADI, ...)'
 	help
 	  Say Y here if your system has a dual role high speed USB
@@ -60,7 +61,7 @@
 
 config USB_MUSB_UX500
 	tristate "U8500 and U5500"
-	depends on (ARCH_U8500 && AB8500_USB) || (ARCH_U5500)
+	depends on (ARCH_U8500 && AB8500_USB)
 
 endchoice
 
diff --git a/drivers/usb/musb/am35x.c b/drivers/usb/musb/am35x.c
index 08f1d0b6..e233d2b 100644
--- a/drivers/usb/musb/am35x.c
+++ b/drivers/usb/musb/am35x.c
@@ -27,6 +27,7 @@
  */
 
 #include <linux/init.h>
+#include <linux/module.h>
 #include <linux/clk.h>
 #include <linux/io.h>
 #include <linux/platform_device.h>
diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c
index 4da7492..2613bfd 100644
--- a/drivers/usb/musb/da8xx.c
+++ b/drivers/usb/musb/da8xx.c
@@ -27,6 +27,7 @@
  */
 
 #include <linux/init.h>
+#include <linux/module.h>
 #include <linux/clk.h>
 #include <linux/io.h>
 #include <linux/platform_device.h>
diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c
index 20a2873..b63ab15 100644
--- a/drivers/usb/musb/musb_core.c
+++ b/drivers/usb/musb/musb_core.c
@@ -1477,8 +1477,7 @@
 /*-------------------------------------------------------------------------*/
 
 #if defined(CONFIG_SOC_OMAP2430) || defined(CONFIG_SOC_OMAP3430) || \
-	defined(CONFIG_ARCH_OMAP4) || defined(CONFIG_ARCH_U8500) || \
-	defined(CONFIG_ARCH_U5500)
+	defined(CONFIG_ARCH_OMAP4) || defined(CONFIG_ARCH_U8500)
 
 static irqreturn_t generic_interrupt(int irq, void *__hci)
 {
@@ -2302,18 +2301,12 @@
 		 */
 	}
 
-	musb_save_context(musb);
-
 	spin_unlock_irqrestore(&musb->lock, flags);
 	return 0;
 }
 
 static int musb_resume_noirq(struct device *dev)
 {
-	struct musb	*musb = dev_to_musb(dev);
-
-	musb_restore_context(musb);
-
 	/* for static cmos like DaVinci, register values were preserved
 	 * unless for some reason the whole soc powered down or the USB
 	 * module got reset through the PSC (vs just being disabled).
diff --git a/drivers/usb/musb/musb_gadget.c b/drivers/usb/musb/musb_gadget.c
index ae4a20a..922148f 100644
--- a/drivers/usb/musb/musb_gadget.c
+++ b/drivers/usb/musb/musb_gadget.c
@@ -1903,7 +1903,7 @@
 	unsigned long		flags;
 	int			retval = -EINVAL;
 
-	if (driver->speed != USB_SPEED_HIGH)
+	if (driver->speed < USB_SPEED_HIGH)
 		goto err0;
 
 	pm_runtime_get_sync(musb->controller);
@@ -1999,10 +1999,6 @@
 					nuke(&hw_ep->ep_out, -ESHUTDOWN);
 			}
 		}
-
-		spin_unlock(&musb->lock);
-		driver->disconnect(&musb->g);
-		spin_lock(&musb->lock);
 	}
 }
 
diff --git a/drivers/usb/renesas_usbhs/common.c b/drivers/usb/renesas_usbhs/common.c
index d2e2efa..08c679c 100644
--- a/drivers/usb/renesas_usbhs/common.c
+++ b/drivers/usb/renesas_usbhs/common.c
@@ -405,7 +405,7 @@
 /*
  *		platform functions
  */
-static int __devinit usbhs_probe(struct platform_device *pdev)
+static int usbhs_probe(struct platform_device *pdev)
 {
 	struct renesas_usbhs_platform_info *info = pdev->dev.platform_data;
 	struct renesas_usbhs_driver_callback *dfunc;
diff --git a/drivers/usb/renesas_usbhs/fifo.c b/drivers/usb/renesas_usbhs/fifo.c
index 8da685e..ffdf5d1 100644
--- a/drivers/usb/renesas_usbhs/fifo.c
+++ b/drivers/usb/renesas_usbhs/fifo.c
@@ -820,7 +820,7 @@
 	if (len % 4) /* 32bit alignment */
 		goto usbhsf_pio_prepare_push;
 
-	if ((*(u32 *) pkt->buf + pkt->actual) & 0x7) /* 8byte alignment */
+	if ((uintptr_t)(pkt->buf + pkt->actual) & 0x7) /* 8byte alignment */
 		goto usbhsf_pio_prepare_push;
 
 	/* get enable DMA fifo */
@@ -897,7 +897,7 @@
 	if (!fifo)
 		goto usbhsf_pio_prepare_pop;
 
-	if ((*(u32 *) pkt->buf + pkt->actual) & 0x7) /* 8byte alignment */
+	if ((uintptr_t)(pkt->buf + pkt->actual) & 0x7) /* 8byte alignment */
 		goto usbhsf_pio_prepare_pop;
 
 	ret = usbhsf_fifo_select(pipe, fifo, 0);
diff --git a/drivers/usb/renesas_usbhs/mod.c b/drivers/usb/renesas_usbhs/mod.c
index 053f86d..ad96a38 100644
--- a/drivers/usb/renesas_usbhs/mod.c
+++ b/drivers/usb/renesas_usbhs/mod.c
@@ -349,7 +349,7 @@
 		if (mod->irq_attch)
 			intenb1 |= ATTCHE;
 
-		if (mod->irq_attch)
+		if (mod->irq_dtch)
 			intenb1 |= DTCHE;
 
 		if (mod->irq_sign)
diff --git a/drivers/usb/renesas_usbhs/mod.h b/drivers/usb/renesas_usbhs/mod.h
index 8ae3733..6c68755 100644
--- a/drivers/usb/renesas_usbhs/mod.h
+++ b/drivers/usb/renesas_usbhs/mod.h
@@ -143,8 +143,8 @@
  */
 #if	defined(CONFIG_USB_RENESAS_USBHS_HCD) || \
 	defined(CONFIG_USB_RENESAS_USBHS_HCD_MODULE)
-extern int __devinit usbhs_mod_host_probe(struct usbhs_priv *priv);
-extern int __devexit usbhs_mod_host_remove(struct usbhs_priv *priv);
+extern int usbhs_mod_host_probe(struct usbhs_priv *priv);
+extern int usbhs_mod_host_remove(struct usbhs_priv *priv);
 #else
 static inline int usbhs_mod_host_probe(struct usbhs_priv *priv)
 {
@@ -157,8 +157,8 @@
 
 #if	defined(CONFIG_USB_RENESAS_USBHS_UDC) || \
 	defined(CONFIG_USB_RENESAS_USBHS_UDC_MODULE)
-extern int __devinit usbhs_mod_gadget_probe(struct usbhs_priv *priv);
-extern void __devexit usbhs_mod_gadget_remove(struct usbhs_priv *priv);
+extern int usbhs_mod_gadget_probe(struct usbhs_priv *priv);
+extern void usbhs_mod_gadget_remove(struct usbhs_priv *priv);
 #else
 static inline int usbhs_mod_gadget_probe(struct usbhs_priv *priv)
 {
diff --git a/drivers/usb/renesas_usbhs/mod_gadget.c b/drivers/usb/renesas_usbhs/mod_gadget.c
index 4cc7ee0..7f4e803 100644
--- a/drivers/usb/renesas_usbhs/mod_gadget.c
+++ b/drivers/usb/renesas_usbhs/mod_gadget.c
@@ -751,53 +751,32 @@
 		struct usb_gadget_driver *driver)
 {
 	struct usbhsg_gpriv *gpriv = usbhsg_gadget_to_gpriv(gadget);
-	struct usbhs_priv *priv;
-	struct device *dev;
-	int ret;
+	struct usbhs_priv *priv = usbhsg_gpriv_to_priv(gpriv);
 
 	if (!driver		||
 	    !driver->setup	||
-	    driver->speed != USB_SPEED_HIGH)
+	    driver->speed < USB_SPEED_FULL)
 		return -EINVAL;
 
-	dev  = usbhsg_gpriv_to_dev(gpriv);
-	priv = usbhsg_gpriv_to_priv(gpriv);
-
 	/* first hook up the driver ... */
 	gpriv->driver = driver;
 	gpriv->gadget.dev.driver = &driver->driver;
 
-	ret = device_add(&gpriv->gadget.dev);
-	if (ret) {
-		dev_err(dev, "device_add error %d\n", ret);
-		goto add_fail;
-	}
-
 	return usbhsg_try_start(priv, USBHSG_STATUS_REGISTERD);
-
-add_fail:
-	gpriv->driver = NULL;
-	gpriv->gadget.dev.driver = NULL;
-
-	return ret;
 }
 
 static int usbhsg_gadget_stop(struct usb_gadget *gadget,
 		struct usb_gadget_driver *driver)
 {
 	struct usbhsg_gpriv *gpriv = usbhsg_gadget_to_gpriv(gadget);
-	struct usbhs_priv *priv;
-	struct device *dev;
+	struct usbhs_priv *priv = usbhsg_gpriv_to_priv(gpriv);
 
 	if (!driver		||
 	    !driver->unbind)
 		return -EINVAL;
 
-	dev  = usbhsg_gpriv_to_dev(gpriv);
-	priv = usbhsg_gpriv_to_priv(gpriv);
-
 	usbhsg_try_stop(priv, USBHSG_STATUS_REGISTERD);
-	device_del(&gpriv->gadget.dev);
+	gpriv->gadget.dev.driver = NULL;
 	gpriv->driver = NULL;
 
 	return 0;
@@ -827,10 +806,17 @@
 
 static int usbhsg_stop(struct usbhs_priv *priv)
 {
+	struct usbhsg_gpriv *gpriv = usbhsg_priv_to_gpriv(priv);
+
+	/* cable disconnect */
+	if (gpriv->driver &&
+	    gpriv->driver->disconnect)
+		gpriv->driver->disconnect(&gpriv->gadget);
+
 	return usbhsg_try_stop(priv, USBHSG_STATUS_STARTED);
 }
 
-int __devinit usbhs_mod_gadget_probe(struct usbhs_priv *priv)
+int usbhs_mod_gadget_probe(struct usbhs_priv *priv)
 {
 	struct usbhsg_gpriv *gpriv;
 	struct usbhsg_uep *uep;
@@ -876,12 +862,14 @@
 	/*
 	 * init gadget
 	 */
-	device_initialize(&gpriv->gadget.dev);
 	dev_set_name(&gpriv->gadget.dev, "gadget");
 	gpriv->gadget.dev.parent	= dev;
 	gpriv->gadget.name		= "renesas_usbhs_udc";
 	gpriv->gadget.ops		= &usbhsg_gadget_ops;
 	gpriv->gadget.is_dualspeed	= 1;
+	ret = device_register(&gpriv->gadget.dev);
+	if (ret < 0)
+		goto err_add_udc;
 
 	INIT_LIST_HEAD(&gpriv->gadget.ep_list);
 
@@ -912,12 +900,15 @@
 
 	ret = usb_add_gadget_udc(dev, &gpriv->gadget);
 	if (ret)
-		goto err_add_udc;
+		goto err_register;
 
 
 	dev_info(dev, "gadget probed\n");
 
 	return 0;
+
+err_register:
+	device_unregister(&gpriv->gadget.dev);
 err_add_udc:
 	kfree(gpriv->uep);
 
@@ -927,12 +918,14 @@
 	return ret;
 }
 
-void __devexit usbhs_mod_gadget_remove(struct usbhs_priv *priv)
+void usbhs_mod_gadget_remove(struct usbhs_priv *priv)
 {
 	struct usbhsg_gpriv *gpriv = usbhsg_priv_to_gpriv(priv);
 
 	usb_del_gadget_udc(&gpriv->gadget);
 
+	device_unregister(&gpriv->gadget.dev);
+
 	usbhsg_controller_unregister(gpriv);
 
 	kfree(gpriv->uep);
diff --git a/drivers/usb/renesas_usbhs/mod_host.c b/drivers/usb/renesas_usbhs/mod_host.c
index 1a7208a..7955de5 100644
--- a/drivers/usb/renesas_usbhs/mod_host.c
+++ b/drivers/usb/renesas_usbhs/mod_host.c
@@ -103,7 +103,7 @@
 
 	u32	port_stat;	/* USB_PORT_STAT_xxx */
 
-	struct completion	*done;
+	struct completion	setup_ack_done;
 
 	/* see usbhsh_req_alloc/free */
 	struct list_head	ureq_link_active;
@@ -355,6 +355,7 @@
 struct usbhsh_ep *usbhsh_endpoint_alloc(struct usbhsh_hpriv *hpriv,
 					struct usbhsh_device *udev,
 					struct usb_host_endpoint *ep,
+					int dir_in_req,
 					gfp_t mem_flags)
 {
 	struct usbhs_priv *priv = usbhsh_hpriv_to_priv(hpriv);
@@ -364,27 +365,38 @@
 	struct usbhs_pipe *pipe, *best_pipe;
 	struct device *dev = usbhsh_hcd_to_dev(hcd);
 	struct usb_endpoint_descriptor *desc = &ep->desc;
-	int type, i;
+	int type, i, dir_in;
 	unsigned int min_usr;
 
+	dir_in_req = !!dir_in_req;
+
 	uep = kzalloc(sizeof(struct usbhsh_ep), mem_flags);
 	if (!uep) {
 		dev_err(dev, "usbhsh_ep alloc fail\n");
 		return NULL;
 	}
-	type = usb_endpoint_type(desc);
+
+	if (usb_endpoint_xfer_control(desc)) {
+		best_pipe = usbhsh_hpriv_to_dcp(hpriv);
+		goto usbhsh_endpoint_alloc_find_pipe;
+	}
 
 	/*
 	 * find best pipe for endpoint
 	 * see
 	 *	HARDWARE LIMITATION
 	 */
+	type = usb_endpoint_type(desc);
 	min_usr = ~0;
 	best_pipe = NULL;
-	usbhs_for_each_pipe_with_dcp(pipe, priv, i) {
+	usbhs_for_each_pipe(pipe, priv, i) {
 		if (!usbhs_pipe_type_is(pipe, type))
 			continue;
 
+		dir_in = !!usbhs_pipe_is_dir_in(pipe);
+		if (0 != (dir_in - dir_in_req))
+			continue;
+
 		info = usbhsh_pipe_info(pipe);
 
 		if (min_usr > info->usr_cnt) {
@@ -398,7 +410,7 @@
 		kfree(uep);
 		return NULL;
 	}
-
+usbhsh_endpoint_alloc_find_pipe:
 	/*
 	 * init uep
 	 */
@@ -423,6 +435,7 @@
 	 * see
 	 *  DCPMAXP/PIPEMAXP
 	 */
+	usbhs_pipe_sequence_data0(uep->pipe);
 	usbhs_pipe_config_update(uep->pipe,
 				 usbhsh_device_number(hpriv, udev),
 				 usb_endpoint_num(desc),
@@ -430,7 +443,7 @@
 
 	dev_dbg(dev, "%s [%d-%s](%p)\n", __func__,
 		usbhsh_device_number(hpriv, udev),
-		usbhs_pipe_name(pipe), uep);
+		usbhs_pipe_name(uep->pipe), uep);
 
 	return uep;
 }
@@ -549,8 +562,7 @@
 	 *	usbhsh_irq_setup_ack()
 	 *	usbhsh_irq_setup_err()
 	 */
-	DECLARE_COMPLETION(done);
-	hpriv->done = &done;
+	init_completion(&hpriv->setup_ack_done);
 
 	/* copy original request */
 	memcpy(&req, urb->setup_packet, sizeof(struct usb_ctrlrequest));
@@ -572,8 +584,7 @@
 	/*
 	 * wait setup packet ACK
 	 */
-	wait_for_completion(&done);
-	hpriv->done = NULL;
+	wait_for_completion(&hpriv->setup_ack_done);
 
 	dev_dbg(dev, "%s done\n", __func__);
 }
@@ -724,11 +735,11 @@
 	struct usbhsh_device *udev, *new_udev = NULL;
 	struct usbhs_pipe *pipe;
 	struct usbhsh_ep *uep;
+	int is_dir_in = usb_pipein(urb->pipe);
 
 	int ret;
 
-	dev_dbg(dev, "%s (%s)\n",
-		__func__, usb_pipein(urb->pipe) ? "in" : "out");
+	dev_dbg(dev, "%s (%s)\n", __func__, is_dir_in ? "in" : "out");
 
 	ret = usb_hcd_link_urb_to_ep(hcd, urb);
 	if (ret)
@@ -751,7 +762,8 @@
 	 */
 	uep = usbhsh_ep_to_uep(ep);
 	if (!uep) {
-		uep = usbhsh_endpoint_alloc(hpriv, udev, ep, mem_flags);
+		uep = usbhsh_endpoint_alloc(hpriv, udev, ep,
+					    is_dir_in, mem_flags);
 		if (!uep)
 			goto usbhsh_urb_enqueue_error_free_device;
 	}
@@ -1095,10 +1107,7 @@
 
 	dev_dbg(dev, "setup packet OK\n");
 
-	if (unlikely(!hpriv->done))
-		dev_err(dev, "setup ack happen without necessary data\n");
-	else
-		complete(hpriv->done); /* see usbhsh_urb_enqueue() */
+	complete(&hpriv->setup_ack_done); /* see usbhsh_urb_enqueue() */
 
 	return 0;
 }
@@ -1111,10 +1120,7 @@
 
 	dev_dbg(dev, "setup packet Err\n");
 
-	if (unlikely(!hpriv->done))
-		dev_err(dev, "setup err happen without necessary data\n");
-	else
-		complete(hpriv->done); /* see usbhsh_urb_enqueue() */
+	complete(&hpriv->setup_ack_done); /* see usbhsh_urb_enqueue() */
 
 	return 0;
 }
@@ -1221,8 +1227,18 @@
 {
 	struct usbhsh_hpriv *hpriv = usbhsh_priv_to_hpriv(priv);
 	struct usb_hcd *hcd = usbhsh_hpriv_to_hcd(hpriv);
+	struct usbhs_mod *mod = usbhs_mod_get_current(priv);
 	struct device *dev = usbhs_priv_to_dev(priv);
 
+	/*
+	 * disable irq callback
+	 */
+	mod->irq_attch	= NULL;
+	mod->irq_dtch	= NULL;
+	mod->irq_sack	= NULL;
+	mod->irq_sign	= NULL;
+	usbhs_irq_callback_update(priv, mod);
+
 	usb_remove_hcd(hcd);
 
 	/* disable sys */
@@ -1235,7 +1251,7 @@
 	return 0;
 }
 
-int __devinit usbhs_mod_host_probe(struct usbhs_priv *priv)
+int usbhs_mod_host_probe(struct usbhs_priv *priv)
 {
 	struct usbhsh_hpriv *hpriv;
 	struct usb_hcd *hcd;
@@ -1251,6 +1267,7 @@
 		dev_err(dev, "Failed to create hcd\n");
 		return -ENOMEM;
 	}
+	hcd->has_tt = 1; /* for low/full speed */
 
 	pipe_info = kzalloc(sizeof(*pipe_info) * pipe_size, GFP_KERNEL);
 	if (!pipe_info) {
@@ -1279,7 +1296,6 @@
 	hpriv->mod.stop		= usbhsh_stop;
 	hpriv->pipe_info	= pipe_info;
 	hpriv->pipe_size	= pipe_size;
-	hpriv->done		= NULL;
 	usbhsh_req_list_init(hpriv);
 	usbhsh_port_stat_init(hpriv);
 
@@ -1299,7 +1315,7 @@
 	return -ENOMEM;
 }
 
-int __devexit usbhs_mod_host_remove(struct usbhs_priv *priv)
+int usbhs_mod_host_remove(struct usbhs_priv *priv)
 {
 	struct usbhsh_hpriv *hpriv = usbhsh_priv_to_hpriv(priv);
 	struct usb_hcd *hcd = usbhsh_hpriv_to_hcd(hpriv);
diff --git a/drivers/usb/serial/ark3116.c b/drivers/usb/serial/ark3116.c
index 5cdb9d9..18e875b 100644
--- a/drivers/usb/serial/ark3116.c
+++ b/drivers/usb/serial/ark3116.c
@@ -42,7 +42,7 @@
  * Version information
  */
 
-#define DRIVER_VERSION "v0.6"
+#define DRIVER_VERSION "v0.7"
 #define DRIVER_AUTHOR "Bart Hartgers <bart.hartgers+ark3116@gmail.com>"
 #define DRIVER_DESC "USB ARK3116 serial/IrDA driver"
 #define DRIVER_DEV_DESC "ARK3116 RS232/IrDA"
@@ -380,10 +380,6 @@
 		goto err_out;
 	}
 
-	/* setup termios */
-	if (tty)
-		ark3116_set_termios(tty, port, NULL);
-
 	/* remove any data still left: also clears error state */
 	ark3116_read_reg(serial, UART_RX, buf);
 
@@ -406,6 +402,10 @@
 	/* enable DMA */
 	ark3116_write_reg(port->serial, UART_FCR, UART_FCR_DMA_SELECT);
 
+	/* setup termios */
+	if (tty)
+		ark3116_set_termios(tty, port, NULL);
+
 err_out:
 	kfree(buf);
 	return result;
diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index 8fe034d..ff3db5d 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -736,6 +736,7 @@
 	{ USB_DEVICE(TML_VID, TML_USB_SERIAL_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_ELSTER_UNICOM_PID) },
 	{ USB_DEVICE(FTDI_VID, FTDI_PROPOX_JTAGCABLEII_PID) },
+	{ USB_DEVICE(FTDI_VID, FTDI_PROPOX_ISPCABLEIII_PID) },
 	{ USB_DEVICE(OLIMEX_VID, OLIMEX_ARM_USB_OCD_PID),
 		.driver_info = (kernel_ulong_t)&ftdi_jtag_quirk },
 	{ USB_DEVICE(OLIMEX_VID, OLIMEX_ARM_USB_OCD_H_PID),
@@ -2104,13 +2105,19 @@
 
 	cflag = termios->c_cflag;
 
-	/* FIXME -For this cut I don't care if the line is really changing or
-	   not  - so just do the change regardless  - should be able to
-	   compare old_termios and tty->termios */
+	if (old_termios->c_cflag == termios->c_cflag
+	    && old_termios->c_ispeed == termios->c_ispeed
+	    && old_termios->c_ospeed == termios->c_ospeed)
+		goto no_c_cflag_changes;
+
 	/* NOTE These routines can get interrupted by
 	   ftdi_sio_read_bulk_callback  - need to examine what this means -
 	   don't see any problems yet */
 
+	if ((old_termios->c_cflag & (CSIZE|PARODD|PARENB|CMSPAR|CSTOPB)) ==
+	    (termios->c_cflag & (CSIZE|PARODD|PARENB|CMSPAR|CSTOPB)))
+		goto no_data_parity_stop_changes;
+
 	/* Set number of data bits, parity, stop bits */
 
 	urb_value = 0;
@@ -2151,6 +2158,7 @@
 	}
 
 	/* Now do the baudrate */
+no_data_parity_stop_changes:
 	if ((cflag & CBAUD) == B0) {
 		/* Disable flow control */
 		if (usb_control_msg(dev, usb_sndctrlpipe(dev, 0),
@@ -2178,6 +2186,7 @@
 
 	/* Set flow control */
 	/* Note device also supports DTR/CD (ugh) and Xon/Xoff in hardware */
+no_c_cflag_changes:
 	if (cflag & CRTSCTS) {
 		dbg("%s Setting to CRTSCTS flow control", __func__);
 		if (usb_control_msg(dev,
diff --git a/drivers/usb/serial/ftdi_sio_ids.h b/drivers/usb/serial/ftdi_sio_ids.h
index 571fa96..055b64e 100644
--- a/drivers/usb/serial/ftdi_sio_ids.h
+++ b/drivers/usb/serial/ftdi_sio_ids.h
@@ -112,6 +112,7 @@
 
 /* Propox devices */
 #define FTDI_PROPOX_JTAGCABLEII_PID	0xD738
+#define FTDI_PROPOX_ISPCABLEIII_PID	0xD739
 
 /* Lenz LI-USB Computer Interface. */
 #define FTDI_LENZ_LIUSB_PID	0xD780
diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c
index 89ae1f6..6dd6453 100644
--- a/drivers/usb/serial/option.c
+++ b/drivers/usb/serial/option.c
@@ -156,6 +156,7 @@
 #define HUAWEI_PRODUCT_K4511			0x14CC
 #define HUAWEI_PRODUCT_ETS1220			0x1803
 #define HUAWEI_PRODUCT_E353			0x1506
+#define HUAWEI_PRODUCT_E173S			0x1C05
 
 #define QUANTA_VENDOR_ID			0x0408
 #define QUANTA_PRODUCT_Q101			0xEA02
@@ -316,6 +317,9 @@
 #define ZTE_PRODUCT_AC8710			0xfff1
 #define ZTE_PRODUCT_AC2726			0xfff5
 #define ZTE_PRODUCT_AC8710T			0xffff
+#define ZTE_PRODUCT_MC2718			0xffe8
+#define ZTE_PRODUCT_AD3812			0xffeb
+#define ZTE_PRODUCT_MC2716			0xffed
 
 #define BENQ_VENDOR_ID				0x04a5
 #define BENQ_PRODUCT_H10			0x4068
@@ -468,6 +472,10 @@
 #define YUGA_PRODUCT_CLU528			0x260D
 #define YUGA_PRODUCT_CLU526			0x260F
 
+/* Viettel products */
+#define VIETTEL_VENDOR_ID			0x2262
+#define VIETTEL_PRODUCT_VT1000			0x0002
+
 /* some devices interfaces need special handling due to a number of reasons */
 enum option_blacklist_reason {
 		OPTION_BLACKLIST_NONE = 0,
@@ -500,6 +508,18 @@
 	.reserved = BIT(4),
 };
 
+static const struct option_blacklist_info zte_ad3812_z_blacklist = {
+	.sendsetup = BIT(0) | BIT(1) | BIT(2),
+};
+
+static const struct option_blacklist_info zte_mc2718_z_blacklist = {
+	.sendsetup = BIT(1) | BIT(2) | BIT(3) | BIT(4),
+};
+
+static const struct option_blacklist_info zte_mc2716_z_blacklist = {
+	.sendsetup = BIT(1) | BIT(2) | BIT(3),
+};
+
 static const struct option_blacklist_info huawei_cdc12_blacklist = {
 	.reserved = BIT(1) | BIT(2),
 };
@@ -622,6 +642,7 @@
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143D, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143E, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E143F, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E173S, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_K4505, 0xff, 0xff, 0xff),
 		.driver_info = (kernel_ulong_t) &huawei_cdc12_blacklist },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_K3765, 0xff, 0xff, 0xff),
@@ -640,6 +661,14 @@
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_K4511, 0xff, 0x01, 0x31) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_K4511, 0xff, 0x01, 0x32) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x01, 0x01) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x01, 0x02) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x01, 0x03) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x01, 0x10) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x01, 0x12) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x01, 0x13) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x02, 0x01) },  /* E398 3G Modem */
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x02, 0x02) },  /* E398 3G PC UI Interface */
+	{ USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E353, 0xff, 0x02, 0x03) },  /* E398 3G Application Interface */
 	{ USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V640) },
 	{ USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V620) },
 	{ USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V740) },
@@ -726,6 +755,7 @@
 	{ USB_DEVICE(KYOCERA_VENDOR_ID, KYOCERA_PRODUCT_KPC680) },
 	{ USB_DEVICE(QUALCOMM_VENDOR_ID, 0x6000)}, /* ZTE AC8700 */
 	{ USB_DEVICE(QUALCOMM_VENDOR_ID, 0x6613)}, /* Onda H600/ZTE MF330 */
+	{ USB_DEVICE(QUALCOMM_VENDOR_ID, 0x9000)}, /* SIMCom SIM5218 */
 	{ USB_DEVICE(CMOTECH_VENDOR_ID, CMOTECH_PRODUCT_6280) }, /* BP3-USB & BP3-EXT HSDPA */
 	{ USB_DEVICE(CMOTECH_VENDOR_ID, CMOTECH_PRODUCT_6008) },
 	{ USB_DEVICE(TELIT_VENDOR_ID, TELIT_PRODUCT_UC864E) },
@@ -1043,6 +1073,12 @@
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_AC8710, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_AC2726, 0xff, 0xff, 0xff) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_AC8710T, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_MC2718, 0xff, 0xff, 0xff),
+	 .driver_info = (kernel_ulong_t)&zte_mc2718_z_blacklist },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_AD3812, 0xff, 0xff, 0xff),
+	 .driver_info = (kernel_ulong_t)&zte_ad3812_z_blacklist },
+	{ USB_DEVICE_AND_INTERFACE_INFO(ZTE_VENDOR_ID, ZTE_PRODUCT_MC2716, 0xff, 0xff, 0xff),
+	 .driver_info = (kernel_ulong_t)&zte_mc2716_z_blacklist },
 	{ USB_DEVICE(BENQ_VENDOR_ID, BENQ_PRODUCT_H10) },
 	{ USB_DEVICE(DLINK_VENDOR_ID, DLINK_PRODUCT_DWM_652) },
 	{ USB_DEVICE(ALINK_VENDOR_ID, DLINK_PRODUCT_DWM_652_U5) }, /* Yes, ALINK_VENDOR_ID */
@@ -1141,6 +1177,7 @@
 	{ USB_DEVICE(YUGA_VENDOR_ID, YUGA_PRODUCT_CLU516) },
 	{ USB_DEVICE(YUGA_VENDOR_ID, YUGA_PRODUCT_CLU528) },
 	{ USB_DEVICE(YUGA_VENDOR_ID, YUGA_PRODUCT_CLU526) },
+	{ USB_DEVICE_AND_INTERFACE_INFO(VIETTEL_VENDOR_ID, VIETTEL_PRODUCT_VT1000, 0xff, 0xff, 0xff) },
 	{ } /* Terminating entry */
 };
 MODULE_DEVICE_TABLE(usb, option_ids);
diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index 9083d1e..fc2d66f 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -91,7 +91,6 @@
 	{ USB_DEVICE(SONY_VENDOR_ID, SONY_QN3USB_PRODUCT_ID) },
 	{ USB_DEVICE(SANWA_VENDOR_ID, SANWA_PRODUCT_ID) },
 	{ USB_DEVICE(ADLINK_VENDOR_ID, ADLINK_ND6530_PRODUCT_ID) },
-	{ USB_DEVICE(WINCHIPHEAD_VENDOR_ID, WINCHIPHEAD_USBSER_PRODUCT_ID) },
 	{ USB_DEVICE(SMART_VENDOR_ID, SMART_PRODUCT_ID) },
 	{ }					/* Terminating entry */
 };
diff --git a/drivers/usb/serial/pl2303.h b/drivers/usb/serial/pl2303.h
index 3d10d7f..c38b8c0 100644
--- a/drivers/usb/serial/pl2303.h
+++ b/drivers/usb/serial/pl2303.h
@@ -145,10 +145,6 @@
 #define ADLINK_VENDOR_ID		0x0b63
 #define ADLINK_ND6530_PRODUCT_ID	0x6530
 
-/* WinChipHead USB->RS 232 adapter */
-#define WINCHIPHEAD_VENDOR_ID		0x4348
-#define WINCHIPHEAD_USBSER_PRODUCT_ID	0x5523
-
 /* SMART USB Serial Adapter */
 #define SMART_VENDOR_ID	0x0b8c
 #define SMART_PRODUCT_ID	0x2303
diff --git a/drivers/usb/storage/ene_ub6250.c b/drivers/usb/storage/ene_ub6250.c
index 4dca3ef..9fbe742 100644
--- a/drivers/usb/storage/ene_ub6250.c
+++ b/drivers/usb/storage/ene_ub6250.c
@@ -1762,10 +1762,9 @@
 		result = ene_send_scsi_cmd(us, FDIR_WRITE, scsi_sglist(srb), 1);
 	} else {
 		void *buf;
-		int offset;
+		int offset = 0;
 		u16 PhyBlockAddr;
 		u8 PageNum;
-		u32 result;
 		u16 len, oldphy, newphy;
 
 		buf = kmalloc(blenByte, GFP_KERNEL);
diff --git a/drivers/usb/storage/protocol.c b/drivers/usb/storage/protocol.c
index 93c1a4d..82dd834 100644
--- a/drivers/usb/storage/protocol.c
+++ b/drivers/usb/storage/protocol.c
@@ -59,7 +59,9 @@
 
 void usb_stor_pad12_command(struct scsi_cmnd *srb, struct us_data *us)
 {
-	/* Pad the SCSI command with zeros out to 12 bytes
+	/*
+	 * Pad the SCSI command with zeros out to 12 bytes.  If the
+	 * command already is 12 bytes or longer, leave it alone.
 	 *
 	 * NOTE: This only works because a scsi_cmnd struct field contains
 	 * a unsigned char cmnd[16], so we know we have storage available
@@ -67,9 +69,6 @@
 	for (; srb->cmd_len<12; srb->cmd_len++)
 		srb->cmnd[srb->cmd_len] = 0;
 
-	/* set command length to 12 bytes */
-	srb->cmd_len = 12;
-
 	/* send the command to the transport layer */
 	usb_stor_invoke_transport(srb, us);
 }
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 3041a97..24caba7 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -1854,6 +1854,13 @@
 		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
+/* Reported by Qinglin Ye <yestyle@gmail.com> */
+UNUSUAL_DEV(  0x13fe, 0x3600, 0x0100, 0x0100,
+		"Kingston",
+		"DT 101 G2",
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
+		US_FL_BULK_IGNORE_TAG ),
+
 /* Reported by Francesco Foresti <frafore@tiscali.it> */
 UNUSUAL_DEV(  0x14cd, 0x6600, 0x0201, 0x0201,
 		"Super Top",
diff --git a/drivers/video/da8xx-fb.c b/drivers/video/da8xx-fb.c
index 55f91d9..29577bf 100644
--- a/drivers/video/da8xx-fb.c
+++ b/drivers/video/da8xx-fb.c
@@ -116,6 +116,7 @@
 /* Clock registers available only on Version 2 */
 #define  LCD_CLK_ENABLE_REG			0x6c
 #define  LCD_CLK_RESET_REG			0x70
+#define  LCD_CLK_MAIN_RESET			BIT(3)
 
 #define LCD_NUM_BUFFERS	2
 
@@ -244,6 +245,10 @@
 {
 	u32 reg;
 
+	/* Bring LCDC out of reset */
+	if (lcd_revision == LCD_VERSION_2)
+		lcdc_write(0, LCD_CLK_RESET_REG);
+
 	reg = lcdc_read(LCD_RASTER_CTRL_REG);
 	if (!(reg & LCD_RASTER_ENABLE))
 		lcdc_write(reg | LCD_RASTER_ENABLE, LCD_RASTER_CTRL_REG);
@@ -257,6 +262,10 @@
 	reg = lcdc_read(LCD_RASTER_CTRL_REG);
 	if (reg & LCD_RASTER_ENABLE)
 		lcdc_write(reg & ~LCD_RASTER_ENABLE, LCD_RASTER_CTRL_REG);
+
+	if (lcd_revision == LCD_VERSION_2)
+		/* Write 1 to reset LCDC */
+		lcdc_write(LCD_CLK_MAIN_RESET, LCD_CLK_RESET_REG);
 }
 
 static void lcd_blit(int load_mode, struct da8xx_fb_par *par)
@@ -584,8 +593,12 @@
 	lcdc_write(0, LCD_DMA_CTRL_REG);
 	lcdc_write(0, LCD_RASTER_CTRL_REG);
 
-	if (lcd_revision == LCD_VERSION_2)
+	if (lcd_revision == LCD_VERSION_2) {
 		lcdc_write(0, LCD_INT_ENABLE_SET_REG);
+		/* Write 1 to reset */
+		lcdc_write(LCD_CLK_MAIN_RESET, LCD_CLK_RESET_REG);
+		lcdc_write(0, LCD_CLK_RESET_REG);
+	}
 }
 
 static void lcd_calc_clk_divider(struct da8xx_fb_par *par)
diff --git a/drivers/video/omap/dispc.c b/drivers/video/omap/dispc.c
index 0ccd7ad..6f61e78 100644
--- a/drivers/video/omap/dispc.c
+++ b/drivers/video/omap/dispc.c
@@ -19,6 +19,7 @@
  * 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 #include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/dma-mapping.h>
 #include <linux/mm.h>
 #include <linux/vmalloc.h>
diff --git a/drivers/video/omap2/dss/dispc.c b/drivers/video/omap2/dss/dispc.c
index 3532782..5c81533 100644
--- a/drivers/video/omap2/dss/dispc.c
+++ b/drivers/video/omap2/dss/dispc.c
@@ -1720,12 +1720,11 @@
 	const int maxdownscale = dss_feat_get_param_max(FEAT_PARAM_DOWNSCALE);
 	unsigned long fclk = 0;
 
-	if ((ovl->caps & OMAP_DSS_OVL_CAP_SCALE) == 0) {
-		if (width != out_width || height != out_height)
-			return -EINVAL;
-		else
-			return 0;
-	}
+	if (width == out_width && height == out_height)
+		return 0;
+
+	if ((ovl->caps & OMAP_DSS_OVL_CAP_SCALE) == 0)
+		return -EINVAL;
 
 	if (out_width < width / maxdownscale ||
 			out_width > width * 8)
diff --git a/drivers/video/omap2/dss/hdmi.c b/drivers/video/omap2/dss/hdmi.c
index 3262f0f..c56378c 100644
--- a/drivers/video/omap2/dss/hdmi.c
+++ b/drivers/video/omap2/dss/hdmi.c
@@ -269,7 +269,7 @@
 unsigned long hdmi_get_pixel_clock(void)
 {
 	/* HDMI Pixel Clock in Mhz */
-	return hdmi.ip_data.cfg.timings.timings.pixel_clock * 10000;
+	return hdmi.ip_data.cfg.timings.timings.pixel_clock * 1000;
 }
 
 static void hdmi_compute_pll(struct omap_dss_device *dssdev, int phy,
diff --git a/drivers/video/via/share.h b/drivers/video/via/share.h
index 69d882c..c01c1c1 100644
--- a/drivers/video/via/share.h
+++ b/drivers/video/via/share.h
@@ -559,8 +559,8 @@
 #define M1200X720_R60_VSP       POSITIVE
 
 /* 1200x900@60 Sync Polarity (DCON) */
-#define M1200X900_R60_HSP       NEGATIVE
-#define M1200X900_R60_VSP       NEGATIVE
+#define M1200X900_R60_HSP       POSITIVE
+#define M1200X900_R60_VSP       POSITIVE
 
 /* 1280x600@60 Sync Polarity (GTF Mode) */
 #define M1280x600_R60_HSP       NEGATIVE
diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
index 816ed08..1a61939 100644
--- a/drivers/virtio/Kconfig
+++ b/drivers/virtio/Kconfig
@@ -37,7 +37,7 @@
 
  config VIRTIO_MMIO
  	tristate "Platform bus driver for memory mapped virtio devices (EXPERIMENTAL)"
- 	depends on EXPERIMENTAL
+ 	depends on HAS_IOMEM && EXPERIMENTAL
  	select VIRTIO
  	select VIRTIO_RING
  	---help---
diff --git a/drivers/virtio/virtio_mmio.c b/drivers/virtio/virtio_mmio.c
index acc5e43..7317dc2e 100644
--- a/drivers/virtio/virtio_mmio.c
+++ b/drivers/virtio/virtio_mmio.c
@@ -118,7 +118,7 @@
 	vring_transport_features(vdev);
 
 	for (i = 0; i < ARRAY_SIZE(vdev->features); i++) {
-		writel(i, vm_dev->base + VIRTIO_MMIO_GUEST_FEATURES_SET);
+		writel(i, vm_dev->base + VIRTIO_MMIO_GUEST_FEATURES_SEL);
 		writel(vdev->features[i],
 				vm_dev->base + VIRTIO_MMIO_GUEST_FEATURES);
 	}
diff --git a/drivers/virtio/virtio_pci.c b/drivers/virtio/virtio_pci.c
index 79a31e5..03d1984 100644
--- a/drivers/virtio/virtio_pci.c
+++ b/drivers/virtio/virtio_pci.c
@@ -169,11 +169,29 @@
 	iowrite8(status, vp_dev->ioaddr + VIRTIO_PCI_STATUS);
 }
 
+/* wait for pending irq handlers */
+static void vp_synchronize_vectors(struct virtio_device *vdev)
+{
+	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
+	int i;
+
+	if (vp_dev->intx_enabled)
+		synchronize_irq(vp_dev->pci_dev->irq);
+
+	for (i = 0; i < vp_dev->msix_vectors; ++i)
+		synchronize_irq(vp_dev->msix_entries[i].vector);
+}
+
 static void vp_reset(struct virtio_device *vdev)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
 	/* 0 status means a reset. */
 	iowrite8(0, vp_dev->ioaddr + VIRTIO_PCI_STATUS);
+	/* Flush out the status write, and flush in device writes,
+	 * including MSi-X interrupts, if any. */
+	ioread8(vp_dev->ioaddr + VIRTIO_PCI_STATUS);
+	/* Flush pending VQ/configuration callbacks. */
+	vp_synchronize_vectors(vdev);
 }
 
 /* the notify function used when creating a virt queue */
@@ -594,11 +612,11 @@
 
 static void virtio_pci_release_dev(struct device *_d)
 {
-	struct virtio_device *dev = container_of(_d, struct virtio_device,
-						 dev);
-	struct virtio_pci_device *vp_dev = to_vp_device(dev);
-
-	kfree(vp_dev);
+	/*
+	 * No need for a release method as we allocate/free
+	 * all devices together with the pci devices.
+	 * Provide an empty one to avoid getting a warning from core.
+	 */
 }
 
 /* the PCI probing function */
@@ -686,6 +704,7 @@
 	pci_iounmap(pci_dev, vp_dev->ioaddr);
 	pci_release_regions(pci_dev);
 	pci_disable_device(pci_dev);
+	kfree(vp_dev);
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 6285867..79fd606 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -314,13 +314,6 @@
 	  To compile this driver as a module, choose M here: the
 	  module will be called nuc900_wdt.
 
-config ADX_WATCHDOG
-	tristate "Avionic Design Xanthos watchdog"
-	depends on ARCH_PXA_ADX
-	help
-	  Say Y here if you want support for the watchdog timer on Avionic
-	  Design Xanthos boards.
-
 config TS72XX_WATCHDOG
 	tristate "TS-72XX SBC Watchdog"
 	depends on MACH_TS72XX
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index 55bd574..fe893e9 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -51,7 +51,6 @@
 obj-$(CONFIG_COH901327_WATCHDOG) += coh901327_wdt.o
 obj-$(CONFIG_STMP3XXX_WATCHDOG) += stmp3xxx_wdt.o
 obj-$(CONFIG_NUC900_WATCHDOG) += nuc900_wdt.o
-obj-$(CONFIG_ADX_WATCHDOG) += adx_wdt.o
 obj-$(CONFIG_TS72XX_WATCHDOG) += ts72xx_wdt.o
 obj-$(CONFIG_IMX2_WDT) += imx2_wdt.o
 
diff --git a/drivers/watchdog/adx_wdt.c b/drivers/watchdog/adx_wdt.c
deleted file mode 100644
index af6e6b1..0000000
--- a/drivers/watchdog/adx_wdt.c
+++ /dev/null
@@ -1,355 +0,0 @@
-/*
- * Copyright (C) 2008-2009 Avionic Design GmbH
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <linux/fs.h>
-#include <linux/gfp.h>
-#include <linux/io.h>
-#include <linux/miscdevice.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/types.h>
-#include <linux/uaccess.h>
-#include <linux/watchdog.h>
-
-#define WATCHDOG_NAME "adx-wdt"
-
-/* register offsets */
-#define	ADX_WDT_CONTROL		0x00
-#define	ADX_WDT_CONTROL_ENABLE	(1 << 0)
-#define	ADX_WDT_CONTROL_nRESET	(1 << 1)
-#define	ADX_WDT_TIMEOUT		0x08
-
-static struct platform_device *adx_wdt_dev;
-static unsigned long driver_open;
-
-#define	WDT_STATE_STOP	0
-#define	WDT_STATE_START	1
-
-struct adx_wdt {
-	void __iomem *base;
-	unsigned long timeout;
-	unsigned int state;
-	unsigned int wake;
-	spinlock_t lock;
-};
-
-static const struct watchdog_info adx_wdt_info = {
-	.identity = "Avionic Design Xanthos Watchdog",
-	.options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING,
-};
-
-static void adx_wdt_start_locked(struct adx_wdt *wdt)
-{
-	u32 ctrl;
-
-	ctrl = readl(wdt->base + ADX_WDT_CONTROL);
-	ctrl |= ADX_WDT_CONTROL_ENABLE;
-	writel(ctrl, wdt->base + ADX_WDT_CONTROL);
-	wdt->state = WDT_STATE_START;
-}
-
-static void adx_wdt_start(struct adx_wdt *wdt)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&wdt->lock, flags);
-	adx_wdt_start_locked(wdt);
-	spin_unlock_irqrestore(&wdt->lock, flags);
-}
-
-static void adx_wdt_stop_locked(struct adx_wdt *wdt)
-{
-	u32 ctrl;
-
-	ctrl = readl(wdt->base + ADX_WDT_CONTROL);
-	ctrl &= ~ADX_WDT_CONTROL_ENABLE;
-	writel(ctrl, wdt->base + ADX_WDT_CONTROL);
-	wdt->state = WDT_STATE_STOP;
-}
-
-static void adx_wdt_stop(struct adx_wdt *wdt)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&wdt->lock, flags);
-	adx_wdt_stop_locked(wdt);
-	spin_unlock_irqrestore(&wdt->lock, flags);
-}
-
-static void adx_wdt_set_timeout(struct adx_wdt *wdt, unsigned long seconds)
-{
-	unsigned long timeout = seconds * 1000;
-	unsigned long flags;
-	unsigned int state;
-
-	spin_lock_irqsave(&wdt->lock, flags);
-	state = wdt->state;
-	adx_wdt_stop_locked(wdt);
-	writel(timeout, wdt->base + ADX_WDT_TIMEOUT);
-
-	if (state == WDT_STATE_START)
-		adx_wdt_start_locked(wdt);
-
-	wdt->timeout = timeout;
-	spin_unlock_irqrestore(&wdt->lock, flags);
-}
-
-static void adx_wdt_get_timeout(struct adx_wdt *wdt, unsigned long *seconds)
-{
-	*seconds = wdt->timeout / 1000;
-}
-
-static void adx_wdt_keepalive(struct adx_wdt *wdt)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&wdt->lock, flags);
-	writel(wdt->timeout, wdt->base + ADX_WDT_TIMEOUT);
-	spin_unlock_irqrestore(&wdt->lock, flags);
-}
-
-static int adx_wdt_open(struct inode *inode, struct file *file)
-{
-	struct adx_wdt *wdt = platform_get_drvdata(adx_wdt_dev);
-
-	if (test_and_set_bit(0, &driver_open))
-		return -EBUSY;
-
-	file->private_data = wdt;
-	adx_wdt_set_timeout(wdt, 30);
-	adx_wdt_start(wdt);
-
-	return nonseekable_open(inode, file);
-}
-
-static int adx_wdt_release(struct inode *inode, struct file *file)
-{
-	struct adx_wdt *wdt = file->private_data;
-
-	adx_wdt_stop(wdt);
-	clear_bit(0, &driver_open);
-
-	return 0;
-}
-
-static long adx_wdt_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
-{
-	struct adx_wdt *wdt = file->private_data;
-	void __user *argp = (void __user *)arg;
-	unsigned long __user *p = argp;
-	unsigned long seconds = 0;
-	unsigned int options;
-	long ret = -EINVAL;
-
-	switch (cmd) {
-	case WDIOC_GETSUPPORT:
-		if (copy_to_user(argp, &adx_wdt_info, sizeof(adx_wdt_info)))
-			return -EFAULT;
-		else
-			return 0;
-
-	case WDIOC_GETSTATUS:
-	case WDIOC_GETBOOTSTATUS:
-		return put_user(0, p);
-
-	case WDIOC_KEEPALIVE:
-		adx_wdt_keepalive(wdt);
-		return 0;
-
-	case WDIOC_SETTIMEOUT:
-		if (get_user(seconds, p))
-			return -EFAULT;
-
-		adx_wdt_set_timeout(wdt, seconds);
-
-		/* fallthrough */
-	case WDIOC_GETTIMEOUT:
-		adx_wdt_get_timeout(wdt, &seconds);
-		return put_user(seconds, p);
-
-	case WDIOC_SETOPTIONS:
-		if (copy_from_user(&options, argp, sizeof(options)))
-			return -EFAULT;
-
-		if (options & WDIOS_DISABLECARD) {
-			adx_wdt_stop(wdt);
-			ret = 0;
-		}
-
-		if (options & WDIOS_ENABLECARD) {
-			adx_wdt_start(wdt);
-			ret = 0;
-		}
-
-		return ret;
-
-	default:
-		break;
-	}
-
-	return -ENOTTY;
-}
-
-static ssize_t adx_wdt_write(struct file *file, const char __user *data,
-		size_t len, loff_t *ppos)
-{
-	struct adx_wdt *wdt = file->private_data;
-
-	if (len)
-		adx_wdt_keepalive(wdt);
-
-	return len;
-}
-
-static const struct file_operations adx_wdt_fops = {
-	.owner = THIS_MODULE,
-	.llseek = no_llseek,
-	.open = adx_wdt_open,
-	.release = adx_wdt_release,
-	.unlocked_ioctl = adx_wdt_ioctl,
-	.write = adx_wdt_write,
-};
-
-static struct miscdevice adx_wdt_miscdev = {
-	.minor = WATCHDOG_MINOR,
-	.name = "watchdog",
-	.fops = &adx_wdt_fops,
-};
-
-static int __devinit adx_wdt_probe(struct platform_device *pdev)
-{
-	struct resource *res;
-	struct adx_wdt *wdt;
-	int ret = 0;
-	u32 ctrl;
-
-	wdt = devm_kzalloc(&pdev->dev, sizeof(*wdt), GFP_KERNEL);
-	if (!wdt) {
-		dev_err(&pdev->dev, "cannot allocate WDT structure\n");
-		return -ENOMEM;
-	}
-
-	spin_lock_init(&wdt->lock);
-
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(&pdev->dev, "cannot obtain I/O memory region\n");
-		return -ENXIO;
-	}
-
-	res = devm_request_mem_region(&pdev->dev, res->start,
-			resource_size(res), res->name);
-	if (!res) {
-		dev_err(&pdev->dev, "cannot request I/O memory region\n");
-		return -ENXIO;
-	}
-
-	wdt->base = devm_ioremap_nocache(&pdev->dev, res->start,
-			resource_size(res));
-	if (!wdt->base) {
-		dev_err(&pdev->dev, "cannot remap I/O memory region\n");
-		return -ENXIO;
-	}
-
-	/* disable watchdog and reboot on timeout */
-	ctrl = readl(wdt->base + ADX_WDT_CONTROL);
-	ctrl &= ~ADX_WDT_CONTROL_ENABLE;
-	ctrl &= ~ADX_WDT_CONTROL_nRESET;
-	writel(ctrl, wdt->base + ADX_WDT_CONTROL);
-
-	platform_set_drvdata(pdev, wdt);
-	adx_wdt_dev = pdev;
-
-	ret = misc_register(&adx_wdt_miscdev);
-	if (ret) {
-		dev_err(&pdev->dev, "cannot register miscdev on minor %d "
-				"(err=%d)\n", WATCHDOG_MINOR, ret);
-		return ret;
-	}
-
-	return 0;
-}
-
-static int __devexit adx_wdt_remove(struct platform_device *pdev)
-{
-	struct adx_wdt *wdt = platform_get_drvdata(pdev);
-
-	misc_deregister(&adx_wdt_miscdev);
-	adx_wdt_stop(wdt);
-	platform_set_drvdata(pdev, NULL);
-
-	return 0;
-}
-
-static void adx_wdt_shutdown(struct platform_device *pdev)
-{
-	struct adx_wdt *wdt = platform_get_drvdata(pdev);
-	adx_wdt_stop(wdt);
-}
-
-#ifdef CONFIG_PM
-static int adx_wdt_suspend(struct device *dev)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct adx_wdt *wdt = platform_get_drvdata(pdev);
-
-	wdt->wake = (wdt->state == WDT_STATE_START) ? 1 : 0;
-	adx_wdt_stop(wdt);
-
-	return 0;
-}
-
-static int adx_wdt_resume(struct device *dev)
-{
-	struct platform_device *pdev = to_platform_device(dev);
-	struct adx_wdt *wdt = platform_get_drvdata(pdev);
-
-	if (wdt->wake)
-		adx_wdt_start(wdt);
-
-	return 0;
-}
-
-static const struct dev_pm_ops adx_wdt_pm_ops = {
-	.suspend = adx_wdt_suspend,
-	.resume = adx_wdt_resume,
-};
-
-#  define ADX_WDT_PM_OPS	(&adx_wdt_pm_ops)
-#else
-#  define ADX_WDT_PM_OPS	NULL
-#endif
-
-static struct platform_driver adx_wdt_driver = {
-	.probe = adx_wdt_probe,
-	.remove = __devexit_p(adx_wdt_remove),
-	.shutdown = adx_wdt_shutdown,
-	.driver = {
-		.name = WATCHDOG_NAME,
-		.owner = THIS_MODULE,
-		.pm = ADX_WDT_PM_OPS,
-	},
-};
-
-static int __init adx_wdt_init(void)
-{
-	return platform_driver_register(&adx_wdt_driver);
-}
-
-static void __exit adx_wdt_exit(void)
-{
-	platform_driver_unregister(&adx_wdt_driver);
-}
-
-module_init(adx_wdt_init);
-module_exit(adx_wdt_exit);
-
-MODULE_DESCRIPTION("Avionic Design Xanthos Watchdog Driver");
-MODULE_LICENSE("GPL v2");
-MODULE_AUTHOR("Thierry Reding <thierry.reding@avionic-design.de>");
-MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/s3c2410_wdt.c b/drivers/watchdog/s3c2410_wdt.c
index 5de7e4f..a79e384 100644
--- a/drivers/watchdog/s3c2410_wdt.c
+++ b/drivers/watchdog/s3c2410_wdt.c
@@ -401,8 +401,8 @@
 
 	dev_info(dev, "watchdog %sactive, reset %sabled, irq %sabled\n",
 		 (wtcon & S3C2410_WTCON_ENABLE) ?  "" : "in",
-		 (wtcon & S3C2410_WTCON_RSTEN) ? "" : "dis",
-		 (wtcon & S3C2410_WTCON_INTEN) ? "" : "en");
+		 (wtcon & S3C2410_WTCON_RSTEN) ? "en" : "dis",
+		 (wtcon & S3C2410_WTCON_INTEN) ? "en" : "dis");
 
 	return 0;
 
diff --git a/drivers/watchdog/wm831x_wdt.c b/drivers/watchdog/wm831x_wdt.c
index 7be3855..e789a47 100644
--- a/drivers/watchdog/wm831x_wdt.c
+++ b/drivers/watchdog/wm831x_wdt.c
@@ -150,7 +150,7 @@
 		if (wm831x_wdt_cfgs[i].time == timeout)
 			break;
 	if (i == ARRAY_SIZE(wm831x_wdt_cfgs))
-		ret = -EINVAL;
+		return -EINVAL;
 
 	ret = wm831x_reg_unlock(wm831x);
 	if (ret == 0) {
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index a767884..31ab82f 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -501,7 +501,7 @@
  * alloc_xenballooned_pages - get pages that have been ballooned out
  * @nr_pages: Number of pages to get
  * @pages: pages returned
- * @highmem: highmem or lowmem pages
+ * @highmem: allow highmem pages
  * @return 0 on success, error otherwise
  */
 int alloc_xenballooned_pages(int nr_pages, struct page **pages, bool highmem)
@@ -511,7 +511,7 @@
 	mutex_lock(&balloon_mutex);
 	while (pgno < nr_pages) {
 		page = balloon_retrieve(highmem);
-		if (page && PageHighMem(page) == highmem) {
+		if (page && (highmem || !PageHighMem(page))) {
 			pages[pgno++] = page;
 		} else {
 			enum bp_state st;
diff --git a/drivers/xen/gntalloc.c b/drivers/xen/gntalloc.c
index f6832f4..e1c4c6e 100644
--- a/drivers/xen/gntalloc.c
+++ b/drivers/xen/gntalloc.c
@@ -135,7 +135,7 @@
 		/* Grant foreign access to the page. */
 		gref->gref_id = gnttab_grant_foreign_access(op->domid,
 			pfn_to_mfn(page_to_pfn(gref->page)), readonly);
-		if (gref->gref_id < 0) {
+		if ((int)gref->gref_id < 0) {
 			rc = gref->gref_id;
 			goto undo;
 		}
@@ -280,7 +280,7 @@
 		goto out;
 	}
 
-	gref_ids = kzalloc(sizeof(gref_ids[0]) * op.count, GFP_TEMPORARY);
+	gref_ids = kcalloc(op.count, sizeof(gref_ids[0]), GFP_TEMPORARY);
 	if (!gref_ids) {
 		rc = -ENOMEM;
 		goto out;
diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
index 3987132..afca14d 100644
--- a/drivers/xen/gntdev.c
+++ b/drivers/xen/gntdev.c
@@ -114,11 +114,11 @@
 	if (NULL == add)
 		return NULL;
 
-	add->grants    = kzalloc(sizeof(add->grants[0])    * count, GFP_KERNEL);
-	add->map_ops   = kzalloc(sizeof(add->map_ops[0])   * count, GFP_KERNEL);
-	add->unmap_ops = kzalloc(sizeof(add->unmap_ops[0]) * count, GFP_KERNEL);
-	add->kmap_ops  = kzalloc(sizeof(add->kmap_ops[0])  * count, GFP_KERNEL);
-	add->pages     = kzalloc(sizeof(add->pages[0])     * count, GFP_KERNEL);
+	add->grants    = kcalloc(count, sizeof(add->grants[0]), GFP_KERNEL);
+	add->map_ops   = kcalloc(count, sizeof(add->map_ops[0]), GFP_KERNEL);
+	add->unmap_ops = kcalloc(count, sizeof(add->unmap_ops[0]), GFP_KERNEL);
+	add->kmap_ops  = kcalloc(count, sizeof(add->kmap_ops[0]), GFP_KERNEL);
+	add->pages     = kcalloc(count, sizeof(add->pages[0]), GFP_KERNEL);
 	if (NULL == add->grants    ||
 	    NULL == add->map_ops   ||
 	    NULL == add->unmap_ops ||
diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c
index 4864e5d..19e6a20 100644
--- a/drivers/xen/swiotlb-xen.c
+++ b/drivers/xen/swiotlb-xen.c
@@ -166,7 +166,7 @@
 	/*
 	 * Get IO TLB memory from any location.
 	 */
-	xen_io_tlb_start = alloc_bootmem(bytes);
+	xen_io_tlb_start = alloc_bootmem_pages(PAGE_ALIGN(bytes));
 	if (!xen_io_tlb_start) {
 		m = "Cannot allocate Xen-SWIOTLB buffer!\n";
 		goto error;
@@ -179,7 +179,7 @@
 			       bytes,
 			       xen_io_tlb_nslabs);
 	if (rc) {
-		free_bootmem(__pa(xen_io_tlb_start), bytes);
+		free_bootmem(__pa(xen_io_tlb_start), PAGE_ALIGN(bytes));
 		m = "Failed to get contiguous memory for DMA from Xen!\n"\
 		    "You either: don't have the permissions, do not have"\
 		    " enough free memory under 4GB, or the hypervisor memory"\
diff --git a/drivers/xen/xenbus/xenbus_client.c b/drivers/xen/xenbus/xenbus_client.c
index 81c3ce6..1906125 100644
--- a/drivers/xen/xenbus/xenbus_client.c
+++ b/drivers/xen/xenbus/xenbus_client.c
@@ -35,6 +35,7 @@
 #include <linux/vmalloc.h>
 #include <linux/export.h>
 #include <asm/xen/hypervisor.h>
+#include <asm/xen/page.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/event_channel.h>
 #include <xen/events.h>
@@ -436,19 +437,20 @@
 int xenbus_map_ring_valloc(struct xenbus_device *dev, int gnt_ref, void **vaddr)
 {
 	struct gnttab_map_grant_ref op = {
-		.flags = GNTMAP_host_map,
+		.flags = GNTMAP_host_map | GNTMAP_contains_pte,
 		.ref   = gnt_ref,
 		.dom   = dev->otherend_id,
 	};
 	struct vm_struct *area;
+	pte_t *pte;
 
 	*vaddr = NULL;
 
-	area = alloc_vm_area(PAGE_SIZE);
+	area = alloc_vm_area(PAGE_SIZE, &pte);
 	if (!area)
 		return -ENOMEM;
 
-	op.host_addr = (unsigned long)area->addr;
+	op.host_addr = arbitrary_virt_to_machine(pte).maddr;
 
 	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
 		BUG();
@@ -527,6 +529,7 @@
 	struct gnttab_unmap_grant_ref op = {
 		.host_addr = (unsigned long)vaddr,
 	};
+	unsigned int level;
 
 	/* It'd be nice if linux/vmalloc.h provided a find_vm_area(void *addr)
 	 * method so that we don't have to muck with vmalloc internals here.
@@ -548,6 +551,8 @@
 	}
 
 	op.handle = (grant_handle_t)area->phys_addr;
+	op.host_addr = arbitrary_virt_to_machine(
+		lookup_address((unsigned long)vaddr, &level)).maddr;
 
 	if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
 		BUG();
diff --git a/fs/bio.c b/fs/bio.c
index 41c93c7..b1fe82c 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -337,7 +337,7 @@
  *	RETURNS:
  *	Pointer to new bio on success, NULL on failure.
  */
-struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs)
+struct bio *bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs)
 {
 	struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set);
 
@@ -365,7 +365,7 @@
  *   %__GFP_WAIT, the allocation is guaranteed to succeed.
  *
  **/
-struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs)
+struct bio *bio_kmalloc(gfp_t gfp_mask, unsigned int nr_iovecs)
 {
 	struct bio *bio;
 
@@ -696,7 +696,8 @@
 	kfree(bmd);
 }
 
-static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count,
+static struct bio_map_data *bio_alloc_map_data(int nr_segs,
+					       unsigned int iov_count,
 					       gfp_t gfp_mask)
 {
 	struct bio_map_data *bmd;
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 7ec1409..cb97174 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -64,6 +64,8 @@
 	int idle;
 };
 
+static int __btrfs_start_workers(struct btrfs_workers *workers);
+
 /*
  * btrfs_start_workers uses kthread_run, which can block waiting for memory
  * for a very long time.  It will actually throttle on page writeback,
@@ -88,27 +90,10 @@
 {
 	struct worker_start *start;
 	start = container_of(work, struct worker_start, work);
-	btrfs_start_workers(start->queue, 1);
+	__btrfs_start_workers(start->queue);
 	kfree(start);
 }
 
-static int start_new_worker(struct btrfs_workers *queue)
-{
-	struct worker_start *start;
-	int ret;
-
-	start = kzalloc(sizeof(*start), GFP_NOFS);
-	if (!start)
-		return -ENOMEM;
-
-	start->work.func = start_new_worker_func;
-	start->queue = queue;
-	ret = btrfs_queue_worker(queue->atomic_worker_start, &start->work);
-	if (ret)
-		kfree(start);
-	return ret;
-}
-
 /*
  * helper function to move a thread onto the idle list after it
  * has finished some requests.
@@ -153,12 +138,20 @@
 static void check_pending_worker_creates(struct btrfs_worker_thread *worker)
 {
 	struct btrfs_workers *workers = worker->workers;
+	struct worker_start *start;
 	unsigned long flags;
 
 	rmb();
 	if (!workers->atomic_start_pending)
 		return;
 
+	start = kzalloc(sizeof(*start), GFP_NOFS);
+	if (!start)
+		return;
+
+	start->work.func = start_new_worker_func;
+	start->queue = workers;
+
 	spin_lock_irqsave(&workers->lock, flags);
 	if (!workers->atomic_start_pending)
 		goto out;
@@ -170,10 +163,11 @@
 
 	workers->num_workers_starting += 1;
 	spin_unlock_irqrestore(&workers->lock, flags);
-	start_new_worker(workers);
+	btrfs_queue_worker(workers->atomic_worker_start, &start->work);
 	return;
 
 out:
+	kfree(start);
 	spin_unlock_irqrestore(&workers->lock, flags);
 }
 
@@ -331,7 +325,7 @@
 			run_ordered_completions(worker->workers, work);
 
 			check_pending_worker_creates(worker);
-
+			cond_resched();
 		}
 
 		spin_lock_irq(&worker->lock);
@@ -462,56 +456,55 @@
  * starts new worker threads.  This does not enforce the max worker
  * count in case you need to temporarily go past it.
  */
-static int __btrfs_start_workers(struct btrfs_workers *workers,
-				 int num_workers)
+static int __btrfs_start_workers(struct btrfs_workers *workers)
 {
 	struct btrfs_worker_thread *worker;
 	int ret = 0;
-	int i;
 
-	for (i = 0; i < num_workers; i++) {
-		worker = kzalloc(sizeof(*worker), GFP_NOFS);
-		if (!worker) {
-			ret = -ENOMEM;
-			goto fail;
-		}
-
-		INIT_LIST_HEAD(&worker->pending);
-		INIT_LIST_HEAD(&worker->prio_pending);
-		INIT_LIST_HEAD(&worker->worker_list);
-		spin_lock_init(&worker->lock);
-
-		atomic_set(&worker->num_pending, 0);
-		atomic_set(&worker->refs, 1);
-		worker->workers = workers;
-		worker->task = kthread_run(worker_loop, worker,
-					   "btrfs-%s-%d", workers->name,
-					   workers->num_workers + i);
-		if (IS_ERR(worker->task)) {
-			ret = PTR_ERR(worker->task);
-			kfree(worker);
-			goto fail;
-		}
-		spin_lock_irq(&workers->lock);
-		list_add_tail(&worker->worker_list, &workers->idle_list);
-		worker->idle = 1;
-		workers->num_workers++;
-		workers->num_workers_starting--;
-		WARN_ON(workers->num_workers_starting < 0);
-		spin_unlock_irq(&workers->lock);
+	worker = kzalloc(sizeof(*worker), GFP_NOFS);
+	if (!worker) {
+		ret = -ENOMEM;
+		goto fail;
 	}
+
+	INIT_LIST_HEAD(&worker->pending);
+	INIT_LIST_HEAD(&worker->prio_pending);
+	INIT_LIST_HEAD(&worker->worker_list);
+	spin_lock_init(&worker->lock);
+
+	atomic_set(&worker->num_pending, 0);
+	atomic_set(&worker->refs, 1);
+	worker->workers = workers;
+	worker->task = kthread_run(worker_loop, worker,
+				   "btrfs-%s-%d", workers->name,
+				   workers->num_workers + 1);
+	if (IS_ERR(worker->task)) {
+		ret = PTR_ERR(worker->task);
+		kfree(worker);
+		goto fail;
+	}
+	spin_lock_irq(&workers->lock);
+	list_add_tail(&worker->worker_list, &workers->idle_list);
+	worker->idle = 1;
+	workers->num_workers++;
+	workers->num_workers_starting--;
+	WARN_ON(workers->num_workers_starting < 0);
+	spin_unlock_irq(&workers->lock);
+
 	return 0;
 fail:
-	btrfs_stop_workers(workers);
+	spin_lock_irq(&workers->lock);
+	workers->num_workers_starting--;
+	spin_unlock_irq(&workers->lock);
 	return ret;
 }
 
-int btrfs_start_workers(struct btrfs_workers *workers, int num_workers)
+int btrfs_start_workers(struct btrfs_workers *workers)
 {
 	spin_lock_irq(&workers->lock);
-	workers->num_workers_starting += num_workers;
+	workers->num_workers_starting++;
 	spin_unlock_irq(&workers->lock);
-	return __btrfs_start_workers(workers, num_workers);
+	return __btrfs_start_workers(workers);
 }
 
 /*
@@ -568,6 +561,7 @@
 	struct btrfs_worker_thread *worker;
 	unsigned long flags;
 	struct list_head *fallback;
+	int ret;
 
 again:
 	spin_lock_irqsave(&workers->lock, flags);
@@ -584,7 +578,9 @@
 			workers->num_workers_starting++;
 			spin_unlock_irqrestore(&workers->lock, flags);
 			/* we're below the limit, start another worker */
-			__btrfs_start_workers(workers, 1);
+			ret = __btrfs_start_workers(workers);
+			if (ret)
+				goto fallback;
 			goto again;
 		}
 	}
@@ -665,7 +661,7 @@
 /*
  * places a struct btrfs_work into the pending queue of one of the kthreads
  */
-int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work)
+void btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work)
 {
 	struct btrfs_worker_thread *worker;
 	unsigned long flags;
@@ -673,7 +669,7 @@
 
 	/* don't requeue something already on a list */
 	if (test_and_set_bit(WORK_QUEUED_BIT, &work->flags))
-		goto out;
+		return;
 
 	worker = find_worker(workers);
 	if (workers->ordered) {
@@ -712,7 +708,4 @@
 	if (wake)
 		wake_up_process(worker->task);
 	spin_unlock_irqrestore(&worker->lock, flags);
-
-out:
-	return 0;
 }
diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index 5077746..f34cc31 100644
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -109,8 +109,8 @@
 	char *name;
 };
 
-int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work);
-int btrfs_start_workers(struct btrfs_workers *workers, int num_workers);
+void btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work);
+int btrfs_start_workers(struct btrfs_workers *workers);
 int btrfs_stop_workers(struct btrfs_workers *workers);
 void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max,
 			struct btrfs_workers *async_starter);
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 8855aad..22c64ff 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -683,7 +683,7 @@
 		return PTR_ERR(fspath);
 
 	if (fspath > fspath_min) {
-		ipath->fspath->val[i] = (u64)fspath;
+		ipath->fspath->val[i] = (u64)(unsigned long)fspath;
 		++ipath->fspath->elem_cnt;
 		ipath->fspath->bytes_left = fspath - fspath_min;
 	} else {
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 5a5d325..634608d2 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -147,14 +147,12 @@
 	 * the btrfs file release call will add this inode to the
 	 * ordered operations list so that we make sure to flush out any
 	 * new data the application may have written before commit.
-	 *
-	 * yes, its silly to have a single bitflag, but we might grow more
-	 * of these.
 	 */
 	unsigned ordered_data_close:1;
 	unsigned orphan_meta_reserved:1;
 	unsigned dummy_inode:1;
 	unsigned in_defrag:1;
+	unsigned delalloc_meta_reserved:1;
 
 	/*
 	 * always compress this one file
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 0fe615e..dede441 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -514,10 +514,25 @@
 				   struct btrfs_root *root,
 				   struct extent_buffer *buf)
 {
+	/* ensure we can see the force_cow */
+	smp_rmb();
+
+	/*
+	 * We do not need to cow a block if
+	 * 1) this block is not created or changed in this transaction;
+	 * 2) this block does not belong to TREE_RELOC tree;
+	 * 3) the root is not forced COW.
+	 *
+	 * What is forced COW:
+	 *    when we create snapshot during commiting the transaction,
+	 *    after we've finished coping src root, we must COW the shared
+	 *    block to ensure the metadata consistency.
+	 */
 	if (btrfs_header_generation(buf) == trans->transid &&
 	    !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN) &&
 	    !(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID &&
-	      btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC)))
+	      btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC)) &&
+	    !root->force_cow)
 		return 0;
 	return 1;
 }
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index b9ba59f..6738503 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -848,7 +848,8 @@
 enum btrfs_caching_type {
 	BTRFS_CACHE_NO		= 0,
 	BTRFS_CACHE_STARTED	= 1,
-	BTRFS_CACHE_FINISHED	= 2,
+	BTRFS_CACHE_FAST	= 2,
+	BTRFS_CACHE_FINISHED	= 3,
 };
 
 enum btrfs_disk_cache_state {
@@ -1271,6 +1272,8 @@
 	 * for stat.  It may be used for more later
 	 */
 	dev_t anon_dev;
+
+	int force_cow;
 };
 
 struct btrfs_ioctl_defrag_range_args {
@@ -2366,6 +2369,9 @@
 int btrfs_block_rsv_refill(struct btrfs_root *root,
 			  struct btrfs_block_rsv *block_rsv,
 			  u64 min_reserved);
+int btrfs_block_rsv_refill_noflush(struct btrfs_root *root,
+				   struct btrfs_block_rsv *block_rsv,
+				   u64 min_reserved);
 int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv,
 			    struct btrfs_block_rsv *dst_rsv,
 			    u64 num_bytes);
@@ -2686,7 +2692,8 @@
 int btrfs_readpage(struct file *file, struct page *page);
 void btrfs_evict_inode(struct inode *inode);
 int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc);
-void btrfs_dirty_inode(struct inode *inode, int flags);
+int btrfs_dirty_inode(struct inode *inode);
+int btrfs_update_time(struct file *file);
 struct inode *btrfs_alloc_inode(struct super_block *sb);
 void btrfs_destroy_inode(struct inode *inode);
 int btrfs_drop_inode(struct inode *inode);
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 3a1b939..9c1eccc 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -617,12 +617,14 @@
 static int btrfs_delayed_inode_reserve_metadata(
 					struct btrfs_trans_handle *trans,
 					struct btrfs_root *root,
+					struct inode *inode,
 					struct btrfs_delayed_node *node)
 {
 	struct btrfs_block_rsv *src_rsv;
 	struct btrfs_block_rsv *dst_rsv;
 	u64 num_bytes;
 	int ret;
+	int release = false;
 
 	src_rsv = trans->block_rsv;
 	dst_rsv = &root->fs_info->delayed_block_rsv;
@@ -638,8 +640,8 @@
 	 * Now if src_rsv == delalloc_block_rsv we'll let it just steal since
 	 * we're accounted for.
 	 */
-	if (!trans->bytes_reserved &&
-	    src_rsv != &root->fs_info->delalloc_block_rsv) {
+	if (!src_rsv || (!trans->bytes_reserved &&
+	    src_rsv != &root->fs_info->delalloc_block_rsv)) {
 		ret = btrfs_block_rsv_add_noflush(root, dst_rsv, num_bytes);
 		/*
 		 * Since we're under a transaction reserve_metadata_bytes could
@@ -652,12 +654,65 @@
 		if (!ret)
 			node->bytes_reserved = num_bytes;
 		return ret;
+	} else if (src_rsv == &root->fs_info->delalloc_block_rsv) {
+		spin_lock(&BTRFS_I(inode)->lock);
+		if (BTRFS_I(inode)->delalloc_meta_reserved) {
+			BTRFS_I(inode)->delalloc_meta_reserved = 0;
+			spin_unlock(&BTRFS_I(inode)->lock);
+			release = true;
+			goto migrate;
+		}
+		spin_unlock(&BTRFS_I(inode)->lock);
+
+		/* Ok we didn't have space pre-reserved.  This shouldn't happen
+		 * too often but it can happen if we do delalloc to an existing
+		 * inode which gets dirtied because of the time update, and then
+		 * isn't touched again until after the transaction commits and
+		 * then we try to write out the data.  First try to be nice and
+		 * reserve something strictly for us.  If not be a pain and try
+		 * to steal from the delalloc block rsv.
+		 */
+		ret = btrfs_block_rsv_add_noflush(root, dst_rsv, num_bytes);
+		if (!ret)
+			goto out;
+
+		ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
+		if (!ret)
+			goto out;
+
+		/*
+		 * Ok this is a problem, let's just steal from the global rsv
+		 * since this really shouldn't happen that often.
+		 */
+		WARN_ON(1);
+		ret = btrfs_block_rsv_migrate(&root->fs_info->global_block_rsv,
+					      dst_rsv, num_bytes);
+		goto out;
 	}
 
+migrate:
 	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
+
+out:
+	/*
+	 * Migrate only takes a reservation, it doesn't touch the size of the
+	 * block_rsv.  This is to simplify people who don't normally have things
+	 * migrated from their block rsv.  If they go to release their
+	 * reservation, that will decrease the size as well, so if migrate
+	 * reduced size we'd end up with a negative size.  But for the
+	 * delalloc_meta_reserved stuff we will only know to drop 1 reservation,
+	 * but we could in fact do this reserve/migrate dance several times
+	 * between the time we did the original reservation and we'd clean it
+	 * up.  So to take care of this, release the space for the meta
+	 * reservation here.  I think it may be time for a documentation page on
+	 * how block rsvs. work.
+	 */
 	if (!ret)
 		node->bytes_reserved = num_bytes;
 
+	if (release)
+		btrfs_block_rsv_release(root, src_rsv, num_bytes);
+
 	return ret;
 }
 
@@ -1708,7 +1763,8 @@
 		goto release_node;
 	}
 
-	ret = btrfs_delayed_inode_reserve_metadata(trans, root, delayed_node);
+	ret = btrfs_delayed_inode_reserve_metadata(trans, root, inode,
+						   delayed_node);
 	if (ret)
 		goto release_node;
 
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 102c176..f44b392 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -620,7 +620,7 @@
 
 static int btree_io_failed_hook(struct bio *failed_bio,
 			 struct page *page, u64 start, u64 end,
-			 u64 mirror_num, struct extent_state *state)
+			 int mirror_num, struct extent_state *state)
 {
 	struct extent_io_tree *tree;
 	unsigned long len;
@@ -1890,31 +1890,32 @@
 	u64 features;
 	struct btrfs_key location;
 	struct buffer_head *bh;
-	struct btrfs_root *extent_root = kzalloc(sizeof(struct btrfs_root),
-						 GFP_NOFS);
-	struct btrfs_root *csum_root = kzalloc(sizeof(struct btrfs_root),
-						 GFP_NOFS);
+	struct btrfs_super_block *disk_super;
 	struct btrfs_root *tree_root = btrfs_sb(sb);
-	struct btrfs_fs_info *fs_info = NULL;
-	struct btrfs_root *chunk_root = kzalloc(sizeof(struct btrfs_root),
-						GFP_NOFS);
-	struct btrfs_root *dev_root = kzalloc(sizeof(struct btrfs_root),
-					      GFP_NOFS);
+	struct btrfs_fs_info *fs_info = tree_root->fs_info;
+	struct btrfs_root *extent_root;
+	struct btrfs_root *csum_root;
+	struct btrfs_root *chunk_root;
+	struct btrfs_root *dev_root;
 	struct btrfs_root *log_tree_root;
-
 	int ret;
 	int err = -EINVAL;
 	int num_backups_tried = 0;
 	int backup_index = 0;
 
-	struct btrfs_super_block *disk_super;
+	extent_root = fs_info->extent_root =
+		kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
+	csum_root = fs_info->csum_root =
+		kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
+	chunk_root = fs_info->chunk_root =
+		kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
+	dev_root = fs_info->dev_root =
+		kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
 
-	if (!extent_root || !tree_root || !tree_root->fs_info ||
-	    !chunk_root || !dev_root || !csum_root) {
+	if (!extent_root || !csum_root || !chunk_root || !dev_root) {
 		err = -ENOMEM;
 		goto fail;
 	}
-	fs_info = tree_root->fs_info;
 
 	ret = init_srcu_struct(&fs_info->subvol_srcu);
 	if (ret) {
@@ -1954,12 +1955,6 @@
 	mutex_init(&fs_info->reloc_mutex);
 
 	init_completion(&fs_info->kobj_unregister);
-	fs_info->tree_root = tree_root;
-	fs_info->extent_root = extent_root;
-	fs_info->csum_root = csum_root;
-	fs_info->chunk_root = chunk_root;
-	fs_info->dev_root = dev_root;
-	fs_info->fs_devices = fs_devices;
 	INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
 	INIT_LIST_HEAD(&fs_info->space_info);
 	btrfs_mapping_init(&fs_info->mapping_tree);
@@ -2199,19 +2194,27 @@
 	fs_info->endio_meta_write_workers.idle_thresh = 2;
 	fs_info->readahead_workers.idle_thresh = 2;
 
-	btrfs_start_workers(&fs_info->workers, 1);
-	btrfs_start_workers(&fs_info->generic_worker, 1);
-	btrfs_start_workers(&fs_info->submit_workers, 1);
-	btrfs_start_workers(&fs_info->delalloc_workers, 1);
-	btrfs_start_workers(&fs_info->fixup_workers, 1);
-	btrfs_start_workers(&fs_info->endio_workers, 1);
-	btrfs_start_workers(&fs_info->endio_meta_workers, 1);
-	btrfs_start_workers(&fs_info->endio_meta_write_workers, 1);
-	btrfs_start_workers(&fs_info->endio_write_workers, 1);
-	btrfs_start_workers(&fs_info->endio_freespace_worker, 1);
-	btrfs_start_workers(&fs_info->delayed_workers, 1);
-	btrfs_start_workers(&fs_info->caching_workers, 1);
-	btrfs_start_workers(&fs_info->readahead_workers, 1);
+	/*
+	 * btrfs_start_workers can really only fail because of ENOMEM so just
+	 * return -ENOMEM if any of these fail.
+	 */
+	ret = btrfs_start_workers(&fs_info->workers);
+	ret |= btrfs_start_workers(&fs_info->generic_worker);
+	ret |= btrfs_start_workers(&fs_info->submit_workers);
+	ret |= btrfs_start_workers(&fs_info->delalloc_workers);
+	ret |= btrfs_start_workers(&fs_info->fixup_workers);
+	ret |= btrfs_start_workers(&fs_info->endio_workers);
+	ret |= btrfs_start_workers(&fs_info->endio_meta_workers);
+	ret |= btrfs_start_workers(&fs_info->endio_meta_write_workers);
+	ret |= btrfs_start_workers(&fs_info->endio_write_workers);
+	ret |= btrfs_start_workers(&fs_info->endio_freespace_worker);
+	ret |= btrfs_start_workers(&fs_info->delayed_workers);
+	ret |= btrfs_start_workers(&fs_info->caching_workers);
+	ret |= btrfs_start_workers(&fs_info->readahead_workers);
+	if (ret) {
+		ret = -ENOMEM;
+		goto fail_sb_buffer;
+	}
 
 	fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super);
 	fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages,
@@ -2465,21 +2468,20 @@
 	btrfs_stop_workers(&fs_info->caching_workers);
 fail_alloc:
 fail_iput:
+	btrfs_mapping_tree_free(&fs_info->mapping_tree);
+
 	invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
 	iput(fs_info->btree_inode);
-
-	btrfs_close_devices(fs_info->fs_devices);
-	btrfs_mapping_tree_free(&fs_info->mapping_tree);
 fail_bdi:
 	bdi_destroy(&fs_info->bdi);
 fail_srcu:
 	cleanup_srcu_struct(&fs_info->subvol_srcu);
 fail:
+	btrfs_close_devices(fs_info->fs_devices);
 	free_fs_info(fs_info);
 	return ERR_PTR(err);
 
 recovery_tree_root:
-
 	if (!btrfs_test_opt(tree_root, RECOVERY))
 		goto fail_tree_roots;
 
@@ -2579,22 +2581,10 @@
 	int errors = 0;
 	u32 crc;
 	u64 bytenr;
-	int last_barrier = 0;
 
 	if (max_mirrors == 0)
 		max_mirrors = BTRFS_SUPER_MIRROR_MAX;
 
-	/* make sure only the last submit_bh does a barrier */
-	if (do_barriers) {
-		for (i = 0; i < max_mirrors; i++) {
-			bytenr = btrfs_sb_offset(i);
-			if (bytenr + BTRFS_SUPER_INFO_SIZE >=
-			    device->total_bytes)
-				break;
-			last_barrier = i;
-		}
-	}
-
 	for (i = 0; i < max_mirrors; i++) {
 		bytenr = btrfs_sb_offset(i);
 		if (bytenr + BTRFS_SUPER_INFO_SIZE >= device->total_bytes)
@@ -2640,17 +2630,136 @@
 			bh->b_end_io = btrfs_end_buffer_write_sync;
 		}
 
-		if (i == last_barrier && do_barriers)
-			ret = submit_bh(WRITE_FLUSH_FUA, bh);
-		else
-			ret = submit_bh(WRITE_SYNC, bh);
-
+		/*
+		 * we fua the first super.  The others we allow
+		 * to go down lazy.
+		 */
+		ret = submit_bh(WRITE_FUA, bh);
 		if (ret)
 			errors++;
 	}
 	return errors < i ? 0 : -1;
 }
 
+/*
+ * endio for the write_dev_flush, this will wake anyone waiting
+ * for the barrier when it is done
+ */
+static void btrfs_end_empty_barrier(struct bio *bio, int err)
+{
+	if (err) {
+		if (err == -EOPNOTSUPP)
+			set_bit(BIO_EOPNOTSUPP, &bio->bi_flags);
+		clear_bit(BIO_UPTODATE, &bio->bi_flags);
+	}
+	if (bio->bi_private)
+		complete(bio->bi_private);
+	bio_put(bio);
+}
+
+/*
+ * trigger flushes for one the devices.  If you pass wait == 0, the flushes are
+ * sent down.  With wait == 1, it waits for the previous flush.
+ *
+ * any device where the flush fails with eopnotsupp are flagged as not-barrier
+ * capable
+ */
+static int write_dev_flush(struct btrfs_device *device, int wait)
+{
+	struct bio *bio;
+	int ret = 0;
+
+	if (device->nobarriers)
+		return 0;
+
+	if (wait) {
+		bio = device->flush_bio;
+		if (!bio)
+			return 0;
+
+		wait_for_completion(&device->flush_wait);
+
+		if (bio_flagged(bio, BIO_EOPNOTSUPP)) {
+			printk("btrfs: disabling barriers on dev %s\n",
+			       device->name);
+			device->nobarriers = 1;
+		}
+		if (!bio_flagged(bio, BIO_UPTODATE)) {
+			ret = -EIO;
+		}
+
+		/* drop the reference from the wait == 0 run */
+		bio_put(bio);
+		device->flush_bio = NULL;
+
+		return ret;
+	}
+
+	/*
+	 * one reference for us, and we leave it for the
+	 * caller
+	 */
+	device->flush_bio = NULL;;
+	bio = bio_alloc(GFP_NOFS, 0);
+	if (!bio)
+		return -ENOMEM;
+
+	bio->bi_end_io = btrfs_end_empty_barrier;
+	bio->bi_bdev = device->bdev;
+	init_completion(&device->flush_wait);
+	bio->bi_private = &device->flush_wait;
+	device->flush_bio = bio;
+
+	bio_get(bio);
+	submit_bio(WRITE_FLUSH, bio);
+
+	return 0;
+}
+
+/*
+ * send an empty flush down to each device in parallel,
+ * then wait for them
+ */
+static int barrier_all_devices(struct btrfs_fs_info *info)
+{
+	struct list_head *head;
+	struct btrfs_device *dev;
+	int errors = 0;
+	int ret;
+
+	/* send down all the barriers */
+	head = &info->fs_devices->devices;
+	list_for_each_entry_rcu(dev, head, dev_list) {
+		if (!dev->bdev) {
+			errors++;
+			continue;
+		}
+		if (!dev->in_fs_metadata || !dev->writeable)
+			continue;
+
+		ret = write_dev_flush(dev, 0);
+		if (ret)
+			errors++;
+	}
+
+	/* wait for all the barriers */
+	list_for_each_entry_rcu(dev, head, dev_list) {
+		if (!dev->bdev) {
+			errors++;
+			continue;
+		}
+		if (!dev->in_fs_metadata || !dev->writeable)
+			continue;
+
+		ret = write_dev_flush(dev, 1);
+		if (ret)
+			errors++;
+	}
+	if (errors)
+		return -EIO;
+	return 0;
+}
+
 int write_all_supers(struct btrfs_root *root, int max_mirrors)
 {
 	struct list_head *head;
@@ -2672,6 +2781,10 @@
 
 	mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
 	head = &root->fs_info->fs_devices->devices;
+
+	if (do_barriers)
+		barrier_all_devices(root->fs_info);
+
 	list_for_each_entry_rcu(dev, head, dev_list) {
 		if (!dev->bdev) {
 			total_errors++;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 9879bd4..f5fbe57 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -467,13 +467,59 @@
 			     struct btrfs_root *root,
 			     int load_cache_only)
 {
+	DEFINE_WAIT(wait);
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	struct btrfs_caching_control *caching_ctl;
 	int ret = 0;
 
-	smp_mb();
-	if (cache->cached != BTRFS_CACHE_NO)
+	caching_ctl = kzalloc(sizeof(*caching_ctl), GFP_NOFS);
+	BUG_ON(!caching_ctl);
+
+	INIT_LIST_HEAD(&caching_ctl->list);
+	mutex_init(&caching_ctl->mutex);
+	init_waitqueue_head(&caching_ctl->wait);
+	caching_ctl->block_group = cache;
+	caching_ctl->progress = cache->key.objectid;
+	atomic_set(&caching_ctl->count, 1);
+	caching_ctl->work.func = caching_thread;
+
+	spin_lock(&cache->lock);
+	/*
+	 * This should be a rare occasion, but this could happen I think in the
+	 * case where one thread starts to load the space cache info, and then
+	 * some other thread starts a transaction commit which tries to do an
+	 * allocation while the other thread is still loading the space cache
+	 * info.  The previous loop should have kept us from choosing this block
+	 * group, but if we've moved to the state where we will wait on caching
+	 * block groups we need to first check if we're doing a fast load here,
+	 * so we can wait for it to finish, otherwise we could end up allocating
+	 * from a block group who's cache gets evicted for one reason or
+	 * another.
+	 */
+	while (cache->cached == BTRFS_CACHE_FAST) {
+		struct btrfs_caching_control *ctl;
+
+		ctl = cache->caching_ctl;
+		atomic_inc(&ctl->count);
+		prepare_to_wait(&ctl->wait, &wait, TASK_UNINTERRUPTIBLE);
+		spin_unlock(&cache->lock);
+
+		schedule();
+
+		finish_wait(&ctl->wait, &wait);
+		put_caching_control(ctl);
+		spin_lock(&cache->lock);
+	}
+
+	if (cache->cached != BTRFS_CACHE_NO) {
+		spin_unlock(&cache->lock);
+		kfree(caching_ctl);
 		return 0;
+	}
+	WARN_ON(cache->caching_ctl);
+	cache->caching_ctl = caching_ctl;
+	cache->cached = BTRFS_CACHE_FAST;
+	spin_unlock(&cache->lock);
 
 	/*
 	 * We can't do the read from on-disk cache during a commit since we need
@@ -484,56 +530,51 @@
 	if (trans && (!trans->transaction->in_commit) &&
 	    (root && root != root->fs_info->tree_root) &&
 	    btrfs_test_opt(root, SPACE_CACHE)) {
-		spin_lock(&cache->lock);
-		if (cache->cached != BTRFS_CACHE_NO) {
-			spin_unlock(&cache->lock);
-			return 0;
-		}
-		cache->cached = BTRFS_CACHE_STARTED;
-		spin_unlock(&cache->lock);
-
 		ret = load_free_space_cache(fs_info, cache);
 
 		spin_lock(&cache->lock);
 		if (ret == 1) {
+			cache->caching_ctl = NULL;
 			cache->cached = BTRFS_CACHE_FINISHED;
 			cache->last_byte_to_unpin = (u64)-1;
 		} else {
-			cache->cached = BTRFS_CACHE_NO;
+			if (load_cache_only) {
+				cache->caching_ctl = NULL;
+				cache->cached = BTRFS_CACHE_NO;
+			} else {
+				cache->cached = BTRFS_CACHE_STARTED;
+			}
 		}
 		spin_unlock(&cache->lock);
+		wake_up(&caching_ctl->wait);
 		if (ret == 1) {
+			put_caching_control(caching_ctl);
 			free_excluded_extents(fs_info->extent_root, cache);
 			return 0;
 		}
-	}
-
-	if (load_cache_only)
-		return 0;
-
-	caching_ctl = kzalloc(sizeof(*caching_ctl), GFP_NOFS);
-	BUG_ON(!caching_ctl);
-
-	INIT_LIST_HEAD(&caching_ctl->list);
-	mutex_init(&caching_ctl->mutex);
-	init_waitqueue_head(&caching_ctl->wait);
-	caching_ctl->block_group = cache;
-	caching_ctl->progress = cache->key.objectid;
-	/* one for caching kthread, one for caching block group list */
-	atomic_set(&caching_ctl->count, 2);
-	caching_ctl->work.func = caching_thread;
-
-	spin_lock(&cache->lock);
-	if (cache->cached != BTRFS_CACHE_NO) {
+	} else {
+		/*
+		 * We are not going to do the fast caching, set cached to the
+		 * appropriate value and wakeup any waiters.
+		 */
+		spin_lock(&cache->lock);
+		if (load_cache_only) {
+			cache->caching_ctl = NULL;
+			cache->cached = BTRFS_CACHE_NO;
+		} else {
+			cache->cached = BTRFS_CACHE_STARTED;
+		}
 		spin_unlock(&cache->lock);
-		kfree(caching_ctl);
+		wake_up(&caching_ctl->wait);
+	}
+
+	if (load_cache_only) {
+		put_caching_control(caching_ctl);
 		return 0;
 	}
-	cache->caching_ctl = caching_ctl;
-	cache->cached = BTRFS_CACHE_STARTED;
-	spin_unlock(&cache->lock);
 
 	down_write(&fs_info->extent_commit_sem);
+	atomic_inc(&caching_ctl->count);
 	list_add_tail(&caching_ctl->list, &fs_info->caching_block_groups);
 	up_write(&fs_info->extent_commit_sem);
 
@@ -2781,7 +2822,7 @@
 	btrfs_release_path(path);
 out:
 	spin_lock(&block_group->lock);
-	if (!ret)
+	if (!ret && dcs == BTRFS_DC_SETUP)
 		block_group->cache_generation = trans->transid;
 	block_group->disk_cache_state = dcs;
 	spin_unlock(&block_group->lock);
@@ -3797,16 +3838,16 @@
 	kfree(rsv);
 }
 
-int btrfs_block_rsv_add(struct btrfs_root *root,
-			struct btrfs_block_rsv *block_rsv,
-			u64 num_bytes)
+static inline int __block_rsv_add(struct btrfs_root *root,
+				  struct btrfs_block_rsv *block_rsv,
+				  u64 num_bytes, int flush)
 {
 	int ret;
 
 	if (num_bytes == 0)
 		return 0;
 
-	ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 1);
+	ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush);
 	if (!ret) {
 		block_rsv_add_bytes(block_rsv, num_bytes, 1);
 		return 0;
@@ -3815,22 +3856,18 @@
 	return ret;
 }
 
+int btrfs_block_rsv_add(struct btrfs_root *root,
+			struct btrfs_block_rsv *block_rsv,
+			u64 num_bytes)
+{
+	return __block_rsv_add(root, block_rsv, num_bytes, 1);
+}
+
 int btrfs_block_rsv_add_noflush(struct btrfs_root *root,
 				struct btrfs_block_rsv *block_rsv,
 				u64 num_bytes)
 {
-	int ret;
-
-	if (num_bytes == 0)
-		return 0;
-
-	ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 0);
-	if (!ret) {
-		block_rsv_add_bytes(block_rsv, num_bytes, 1);
-		return 0;
-	}
-
-	return ret;
+	return __block_rsv_add(root, block_rsv, num_bytes, 0);
 }
 
 int btrfs_block_rsv_check(struct btrfs_root *root,
@@ -3851,9 +3888,9 @@
 	return ret;
 }
 
-int btrfs_block_rsv_refill(struct btrfs_root *root,
-			  struct btrfs_block_rsv *block_rsv,
-			  u64 min_reserved)
+static inline int __btrfs_block_rsv_refill(struct btrfs_root *root,
+					   struct btrfs_block_rsv *block_rsv,
+					   u64 min_reserved, int flush)
 {
 	u64 num_bytes = 0;
 	int ret = -ENOSPC;
@@ -3872,7 +3909,7 @@
 	if (!ret)
 		return 0;
 
-	ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 1);
+	ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush);
 	if (!ret) {
 		block_rsv_add_bytes(block_rsv, num_bytes, 0);
 		return 0;
@@ -3881,6 +3918,20 @@
 	return ret;
 }
 
+int btrfs_block_rsv_refill(struct btrfs_root *root,
+			   struct btrfs_block_rsv *block_rsv,
+			   u64 min_reserved)
+{
+	return __btrfs_block_rsv_refill(root, block_rsv, min_reserved, 1);
+}
+
+int btrfs_block_rsv_refill_noflush(struct btrfs_root *root,
+				   struct btrfs_block_rsv *block_rsv,
+				   u64 min_reserved)
+{
+	return __btrfs_block_rsv_refill(root, block_rsv, min_reserved, 0);
+}
+
 int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv,
 			    struct btrfs_block_rsv *dst_rsv,
 			    u64 num_bytes)
@@ -4064,23 +4115,30 @@
  */
 static unsigned drop_outstanding_extent(struct inode *inode)
 {
+	unsigned drop_inode_space = 0;
 	unsigned dropped_extents = 0;
 
 	BUG_ON(!BTRFS_I(inode)->outstanding_extents);
 	BTRFS_I(inode)->outstanding_extents--;
 
+	if (BTRFS_I(inode)->outstanding_extents == 0 &&
+	    BTRFS_I(inode)->delalloc_meta_reserved) {
+		drop_inode_space = 1;
+		BTRFS_I(inode)->delalloc_meta_reserved = 0;
+	}
+
 	/*
 	 * If we have more or the same amount of outsanding extents than we have
 	 * reserved then we need to leave the reserved extents count alone.
 	 */
 	if (BTRFS_I(inode)->outstanding_extents >=
 	    BTRFS_I(inode)->reserved_extents)
-		return 0;
+		return drop_inode_space;
 
 	dropped_extents = BTRFS_I(inode)->reserved_extents -
 		BTRFS_I(inode)->outstanding_extents;
 	BTRFS_I(inode)->reserved_extents -= dropped_extents;
-	return dropped_extents;
+	return dropped_extents + drop_inode_space;
 }
 
 /**
@@ -4146,12 +4204,17 @@
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_block_rsv *block_rsv = &root->fs_info->delalloc_block_rsv;
 	u64 to_reserve = 0;
+	u64 csum_bytes;
 	unsigned nr_extents = 0;
+	int extra_reserve = 0;
 	int flush = 1;
 	int ret;
 
+	/* Need to be holding the i_mutex here if we aren't free space cache */
 	if (btrfs_is_free_space_inode(root, inode))
 		flush = 0;
+	else
+		WARN_ON(!mutex_is_locked(&inode->i_mutex));
 
 	if (flush && btrfs_transaction_in_commit(root->fs_info))
 		schedule_timeout(1);
@@ -4162,14 +4225,22 @@
 	BTRFS_I(inode)->outstanding_extents++;
 
 	if (BTRFS_I(inode)->outstanding_extents >
-	    BTRFS_I(inode)->reserved_extents) {
+	    BTRFS_I(inode)->reserved_extents)
 		nr_extents = BTRFS_I(inode)->outstanding_extents -
 			BTRFS_I(inode)->reserved_extents;
-		BTRFS_I(inode)->reserved_extents += nr_extents;
 
-		to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents);
+	/*
+	 * Add an item to reserve for updating the inode when we complete the
+	 * delalloc io.
+	 */
+	if (!BTRFS_I(inode)->delalloc_meta_reserved) {
+		nr_extents++;
+		extra_reserve = 1;
 	}
+
+	to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents);
 	to_reserve += calc_csum_metadata_size(inode, num_bytes, 1);
+	csum_bytes = BTRFS_I(inode)->csum_bytes;
 	spin_unlock(&BTRFS_I(inode)->lock);
 
 	ret = reserve_metadata_bytes(root, block_rsv, to_reserve, flush);
@@ -4179,22 +4250,35 @@
 
 		spin_lock(&BTRFS_I(inode)->lock);
 		dropped = drop_outstanding_extent(inode);
-		to_free = calc_csum_metadata_size(inode, num_bytes, 0);
-		spin_unlock(&BTRFS_I(inode)->lock);
-		to_free += btrfs_calc_trans_metadata_size(root, dropped);
-
 		/*
-		 * Somebody could have come in and twiddled with the
-		 * reservation, so if we have to free more than we would have
-		 * reserved from this reservation go ahead and release those
-		 * bytes.
+		 * If the inodes csum_bytes is the same as the original
+		 * csum_bytes then we know we haven't raced with any free()ers
+		 * so we can just reduce our inodes csum bytes and carry on.
+		 * Otherwise we have to do the normal free thing to account for
+		 * the case that the free side didn't free up its reserve
+		 * because of this outstanding reservation.
 		 */
-		to_free -= to_reserve;
+		if (BTRFS_I(inode)->csum_bytes == csum_bytes)
+			calc_csum_metadata_size(inode, num_bytes, 0);
+		else
+			to_free = calc_csum_metadata_size(inode, num_bytes, 0);
+		spin_unlock(&BTRFS_I(inode)->lock);
+		if (dropped)
+			to_free += btrfs_calc_trans_metadata_size(root, dropped);
+
 		if (to_free)
 			btrfs_block_rsv_release(root, block_rsv, to_free);
 		return ret;
 	}
 
+	spin_lock(&BTRFS_I(inode)->lock);
+	if (extra_reserve) {
+		BTRFS_I(inode)->delalloc_meta_reserved = 1;
+		nr_extents--;
+	}
+	BTRFS_I(inode)->reserved_extents += nr_extents;
+	spin_unlock(&BTRFS_I(inode)->lock);
+
 	block_rsv_add_bytes(block_rsv, to_reserve, 1);
 
 	return 0;
@@ -5040,11 +5124,11 @@
 	struct btrfs_root *root = orig_root->fs_info->extent_root;
 	struct btrfs_free_cluster *last_ptr = NULL;
 	struct btrfs_block_group_cache *block_group = NULL;
+	struct btrfs_block_group_cache *used_block_group;
 	int empty_cluster = 2 * 1024 * 1024;
 	int allowed_chunk_alloc = 0;
 	int done_chunk_alloc = 0;
 	struct btrfs_space_info *space_info;
-	int last_ptr_loop = 0;
 	int loop = 0;
 	int index = 0;
 	int alloc_type = (data & BTRFS_BLOCK_GROUP_DATA) ?
@@ -5106,6 +5190,7 @@
 ideal_cache:
 		block_group = btrfs_lookup_block_group(root->fs_info,
 						       search_start);
+		used_block_group = block_group;
 		/*
 		 * we don't want to use the block group if it doesn't match our
 		 * allocation bits, or if its not cached.
@@ -5143,6 +5228,7 @@
 		u64 offset;
 		int cached;
 
+		used_block_group = block_group;
 		btrfs_get_block_group(block_group);
 		search_start = block_group->key.objectid;
 
@@ -5166,13 +5252,15 @@
 		}
 
 have_block_group:
-		if (unlikely(block_group->cached == BTRFS_CACHE_NO)) {
+		cached = block_group_cache_done(block_group);
+		if (unlikely(!cached)) {
 			u64 free_percent;
 
+			found_uncached_bg = true;
 			ret = cache_block_group(block_group, trans,
 						orig_root, 1);
 			if (block_group->cached == BTRFS_CACHE_FINISHED)
-				goto have_block_group;
+				goto alloc;
 
 			free_percent = btrfs_block_group_used(&block_group->item);
 			free_percent *= 100;
@@ -5194,7 +5282,6 @@
 							orig_root, 0);
 				BUG_ON(ret);
 			}
-			found_uncached_bg = true;
 
 			/*
 			 * If loop is set for cached only, try the next block
@@ -5204,94 +5291,80 @@
 				goto loop;
 		}
 
-		cached = block_group_cache_done(block_group);
-		if (unlikely(!cached))
-			found_uncached_bg = true;
-
+alloc:
 		if (unlikely(block_group->ro))
 			goto loop;
 
 		spin_lock(&block_group->free_space_ctl->tree_lock);
 		if (cached &&
 		    block_group->free_space_ctl->free_space <
-		    num_bytes + empty_size) {
+		    num_bytes + empty_cluster + empty_size) {
 			spin_unlock(&block_group->free_space_ctl->tree_lock);
 			goto loop;
 		}
 		spin_unlock(&block_group->free_space_ctl->tree_lock);
 
 		/*
-		 * Ok we want to try and use the cluster allocator, so lets look
-		 * there, unless we are on LOOP_NO_EMPTY_SIZE, since we will
-		 * have tried the cluster allocator plenty of times at this
-		 * point and not have found anything, so we are likely way too
-		 * fragmented for the clustering stuff to find anything, so lets
-		 * just skip it and let the allocator find whatever block it can
-		 * find
+		 * Ok we want to try and use the cluster allocator, so
+		 * lets look there
 		 */
-		if (last_ptr && loop < LOOP_NO_EMPTY_SIZE) {
+		if (last_ptr) {
 			/*
 			 * the refill lock keeps out other
 			 * people trying to start a new cluster
 			 */
 			spin_lock(&last_ptr->refill_lock);
-			if (last_ptr->block_group &&
-			    (last_ptr->block_group->ro ||
-			    !block_group_bits(last_ptr->block_group, data))) {
-				offset = 0;
+			used_block_group = last_ptr->block_group;
+			if (used_block_group != block_group &&
+			    (!used_block_group ||
+			     used_block_group->ro ||
+			     !block_group_bits(used_block_group, data))) {
+				used_block_group = block_group;
 				goto refill_cluster;
 			}
 
-			offset = btrfs_alloc_from_cluster(block_group, last_ptr,
-						 num_bytes, search_start);
+			if (used_block_group != block_group)
+				btrfs_get_block_group(used_block_group);
+
+			offset = btrfs_alloc_from_cluster(used_block_group,
+			  last_ptr, num_bytes, used_block_group->key.objectid);
 			if (offset) {
 				/* we have a block, we're done */
 				spin_unlock(&last_ptr->refill_lock);
 				goto checks;
 			}
 
-			spin_lock(&last_ptr->lock);
-			/*
-			 * whoops, this cluster doesn't actually point to
-			 * this block group.  Get a ref on the block
-			 * group is does point to and try again
-			 */
-			if (!last_ptr_loop && last_ptr->block_group &&
-			    last_ptr->block_group != block_group &&
-			    index <=
-				 get_block_group_index(last_ptr->block_group)) {
-
-				btrfs_put_block_group(block_group);
-				block_group = last_ptr->block_group;
-				btrfs_get_block_group(block_group);
-				spin_unlock(&last_ptr->lock);
-				spin_unlock(&last_ptr->refill_lock);
-
-				last_ptr_loop = 1;
-				search_start = block_group->key.objectid;
-				/*
-				 * we know this block group is properly
-				 * in the list because
-				 * btrfs_remove_block_group, drops the
-				 * cluster before it removes the block
-				 * group from the list
-				 */
-				goto have_block_group;
+			WARN_ON(last_ptr->block_group != used_block_group);
+			if (used_block_group != block_group) {
+				btrfs_put_block_group(used_block_group);
+				used_block_group = block_group;
 			}
-			spin_unlock(&last_ptr->lock);
 refill_cluster:
+			BUG_ON(used_block_group != block_group);
+			/* If we are on LOOP_NO_EMPTY_SIZE, we can't
+			 * set up a new clusters, so lets just skip it
+			 * and let the allocator find whatever block
+			 * it can find.  If we reach this point, we
+			 * will have tried the cluster allocator
+			 * plenty of times and not have found
+			 * anything, so we are likely way too
+			 * fragmented for the clustering stuff to find
+			 * anything.  */
+			if (loop >= LOOP_NO_EMPTY_SIZE) {
+				spin_unlock(&last_ptr->refill_lock);
+				goto unclustered_alloc;
+			}
+
 			/*
 			 * this cluster didn't work out, free it and
 			 * start over
 			 */
 			btrfs_return_cluster_to_free_space(NULL, last_ptr);
 
-			last_ptr_loop = 0;
-
 			/* allocate a cluster in this block group */
 			ret = btrfs_find_space_cluster(trans, root,
 					       block_group, last_ptr,
-					       offset, num_bytes,
+					       search_start, num_bytes,
 					       empty_cluster + empty_size);
 			if (ret == 0) {
 				/*
@@ -5327,6 +5400,7 @@
 			goto loop;
 		}
 
+unclustered_alloc:
 		offset = btrfs_find_space_for_alloc(block_group, search_start,
 						    num_bytes, empty_size);
 		/*
@@ -5353,14 +5427,14 @@
 		search_start = stripe_align(root, offset);
 		/* move on to the next group */
 		if (search_start + num_bytes >= search_end) {
-			btrfs_add_free_space(block_group, offset, num_bytes);
+			btrfs_add_free_space(used_block_group, offset, num_bytes);
 			goto loop;
 		}
 
 		/* move on to the next group */
 		if (search_start + num_bytes >
-		    block_group->key.objectid + block_group->key.offset) {
-			btrfs_add_free_space(block_group, offset, num_bytes);
+		    used_block_group->key.objectid + used_block_group->key.offset) {
+			btrfs_add_free_space(used_block_group, offset, num_bytes);
 			goto loop;
 		}
 
@@ -5368,14 +5442,14 @@
 		ins->offset = num_bytes;
 
 		if (offset < search_start)
-			btrfs_add_free_space(block_group, offset,
+			btrfs_add_free_space(used_block_group, offset,
 					     search_start - offset);
 		BUG_ON(offset > search_start);
 
-		ret = btrfs_update_reserved_bytes(block_group, num_bytes,
+		ret = btrfs_update_reserved_bytes(used_block_group, num_bytes,
 						  alloc_type);
 		if (ret == -EAGAIN) {
-			btrfs_add_free_space(block_group, offset, num_bytes);
+			btrfs_add_free_space(used_block_group, offset, num_bytes);
 			goto loop;
 		}
 
@@ -5384,15 +5458,19 @@
 		ins->offset = num_bytes;
 
 		if (offset < search_start)
-			btrfs_add_free_space(block_group, offset,
+			btrfs_add_free_space(used_block_group, offset,
 					     search_start - offset);
 		BUG_ON(offset > search_start);
+		if (used_block_group != block_group)
+			btrfs_put_block_group(used_block_group);
 		btrfs_put_block_group(block_group);
 		break;
 loop:
 		failed_cluster_refill = false;
 		failed_alloc = false;
 		BUG_ON(index != get_block_group_index(block_group));
+		if (used_block_group != block_group)
+			btrfs_put_block_group(used_block_group);
 		btrfs_put_block_group(block_group);
 	}
 	up_read(&space_info->groups_sem);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 1f87c4d..49f3c9d 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -935,8 +935,10 @@
 	node = tree_search(tree, start);
 	if (!node) {
 		prealloc = alloc_extent_state_atomic(prealloc);
-		if (!prealloc)
-			return -ENOMEM;
+		if (!prealloc) {
+			err = -ENOMEM;
+			goto out;
+		}
 		err = insert_state(tree, prealloc, start, end, &bits);
 		prealloc = NULL;
 		BUG_ON(err == -EEXIST);
@@ -992,8 +994,10 @@
 	 */
 	if (state->start < start) {
 		prealloc = alloc_extent_state_atomic(prealloc);
-		if (!prealloc)
-			return -ENOMEM;
+		if (!prealloc) {
+			err = -ENOMEM;
+			goto out;
+		}
 		err = split_state(tree, state, prealloc, start);
 		BUG_ON(err == -EEXIST);
 		prealloc = NULL;
@@ -1024,8 +1028,10 @@
 			this_end = last_start - 1;
 
 		prealloc = alloc_extent_state_atomic(prealloc);
-		if (!prealloc)
-			return -ENOMEM;
+		if (!prealloc) {
+			err = -ENOMEM;
+			goto out;
+		}
 
 		/*
 		 * Avoid to free 'prealloc' if it can be merged with
@@ -1051,8 +1057,10 @@
 	 */
 	if (state->start <= end && state->end > end) {
 		prealloc = alloc_extent_state_atomic(prealloc);
-		if (!prealloc)
-			return -ENOMEM;
+		if (!prealloc) {
+			err = -ENOMEM;
+			goto out;
+		}
 
 		err = split_state(tree, state, prealloc, end + 1);
 		BUG_ON(err == -EEXIST);
@@ -2285,16 +2293,22 @@
 				clean_io_failure(start, page);
 		}
 		if (!uptodate) {
-			u64 failed_mirror;
-			failed_mirror = (u64)bio->bi_bdev;
-			if (tree->ops && tree->ops->readpage_io_failed_hook)
-				ret = tree->ops->readpage_io_failed_hook(
-						bio, page, start, end,
-						failed_mirror, state);
-			else
-				ret = bio_readpage_error(bio, page, start, end,
-							 failed_mirror, NULL);
+			int failed_mirror;
+			failed_mirror = (int)(unsigned long)bio->bi_bdev;
+			/*
+			 * The generic bio_readpage_error handles errors the
+			 * following way: If possible, new read requests are
+			 * created and submitted and will end up in
+			 * end_bio_extent_readpage as well (if we're lucky, not
+			 * in the !uptodate case). In that case it returns 0 and
+			 * we just go on with the next page in our bio. If it
+			 * can't handle the error it will return -EIO and we
+			 * remain responsible for that page.
+			 */
+			ret = bio_readpage_error(bio, page, start, end,
+							failed_mirror, NULL);
 			if (ret == 0) {
+error_handled:
 				uptodate =
 					test_bit(BIO_UPTODATE, &bio->bi_flags);
 				if (err)
@@ -2302,6 +2316,13 @@
 				uncache_state(&cached);
 				continue;
 			}
+			if (tree->ops && tree->ops->readpage_io_failed_hook) {
+				ret = tree->ops->readpage_io_failed_hook(
+							bio, page, start, end,
+							failed_mirror, state);
+				if (ret == 0)
+					goto error_handled;
+			}
 		}
 
 		if (uptodate) {
@@ -3366,6 +3387,9 @@
 		return -ENOMEM;
 	path->leave_spinning = 1;
 
+	start = ALIGN(start, BTRFS_I(inode)->root->sectorsize);
+	len = ALIGN(len, BTRFS_I(inode)->root->sectorsize);
+
 	/*
 	 * lookup the last file extent.  We're not using i_size here
 	 * because there might be preallocation past i_size
@@ -3413,7 +3437,7 @@
 	lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len, 0,
 			 &cached_state, GFP_NOFS);
 
-	em = get_extent_skip_holes(inode, off, last_for_get_extent,
+	em = get_extent_skip_holes(inode, start, last_for_get_extent,
 				   get_extent);
 	if (!em)
 		goto out;
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index feb9be0..7604c30 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -70,7 +70,7 @@
 			      unsigned long bio_flags);
 	int (*readpage_io_hook)(struct page *page, u64 start, u64 end);
 	int (*readpage_io_failed_hook)(struct bio *bio, struct page *page,
-				       u64 start, u64 end, u64 failed_mirror,
+				       u64 start, u64 end, int failed_mirror,
 				       struct extent_state *state);
 	int (*writepage_io_failed_hook)(struct bio *bio, struct page *page,
 					u64 start, u64 end,
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index dafdfa0..97fbe93 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1167,6 +1167,8 @@
 	nrptrs = min((iov_iter_count(i) + PAGE_CACHE_SIZE - 1) /
 		     PAGE_CACHE_SIZE, PAGE_CACHE_SIZE /
 		     (sizeof(struct page *)));
+	nrptrs = min(nrptrs, current->nr_dirtied_pause - current->nr_dirtied);
+	nrptrs = max(nrptrs, 8);
 	pages = kmalloc(nrptrs * sizeof(struct page *), GFP_KERNEL);
 	if (!pages)
 		return -ENOMEM;
@@ -1387,7 +1389,11 @@
 		goto out;
 	}
 
-	file_update_time(file);
+	err = btrfs_update_time(file);
+	if (err) {
+		mutex_unlock(&inode->i_mutex);
+		goto out;
+	}
 	BTRFS_I(inode)->sequence++;
 
 	start_pos = round_down(pos, root->sectorsize);
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 7a15fcf..ec23d43 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -351,6 +351,11 @@
 		}
 	}
 
+	for (i = 0; i < io_ctl->num_pages; i++) {
+		clear_page_dirty_for_io(io_ctl->pages[i]);
+		set_page_extent_mapped(io_ctl->pages[i]);
+	}
+
 	return 0;
 }
 
@@ -537,6 +542,13 @@
 			    struct btrfs_free_space *entry, u8 *type)
 {
 	struct btrfs_free_space_entry *e;
+	int ret;
+
+	if (!io_ctl->cur) {
+		ret = io_ctl_check_crc(io_ctl, io_ctl->index);
+		if (ret)
+			return ret;
+	}
 
 	e = io_ctl->cur;
 	entry->offset = le64_to_cpu(e->offset);
@@ -550,10 +562,7 @@
 
 	io_ctl_unmap_page(io_ctl);
 
-	if (io_ctl->index >= io_ctl->num_pages)
-		return 0;
-
-	return io_ctl_check_crc(io_ctl, io_ctl->index);
+	return 0;
 }
 
 static int io_ctl_read_bitmap(struct io_ctl *io_ctl,
@@ -561,9 +570,6 @@
 {
 	int ret;
 
-	if (io_ctl->cur && io_ctl->cur != io_ctl->orig)
-		io_ctl_unmap_page(io_ctl);
-
 	ret = io_ctl_check_crc(io_ctl, io_ctl->index);
 	if (ret)
 		return ret;
@@ -699,6 +705,8 @@
 		num_entries--;
 	}
 
+	io_ctl_unmap_page(&io_ctl);
+
 	/*
 	 * We add the bitmaps at the end of the entries in order that
 	 * the bitmap entries are added to the cache.
@@ -1462,6 +1470,7 @@
 {
 	info->offset = offset_to_bitmap(ctl, offset);
 	info->bytes = 0;
+	INIT_LIST_HEAD(&info->list);
 	link_free_space(ctl, info);
 	ctl->total_bitmaps++;
 
@@ -1841,7 +1850,13 @@
 		info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset),
 					  1, 0);
 		if (!info) {
-			WARN_ON(1);
+			/* the tree logging code might be calling us before we
+			 * have fully loaded the free space rbtree for this
+			 * block group.  So it is possible the entry won't
+			 * be in the rbtree yet at all.  The caching code
+			 * will make sure not to put it in the rbtree if
+			 * the logging code has pinned it.
+			 */
 			goto out_lock;
 		}
 	}
@@ -2305,6 +2320,7 @@
 
 	if (!found) {
 		start = i;
+		cluster->max_size = 0;
 		found = true;
 	}
 
@@ -2448,16 +2464,23 @@
 {
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	struct btrfs_free_space *entry;
-	struct rb_node *node;
 	int ret = -ENOSPC;
+	u64 bitmap_offset = offset_to_bitmap(ctl, offset);
 
 	if (ctl->total_bitmaps == 0)
 		return -ENOSPC;
 
 	/*
-	 * First check our cached list of bitmaps and see if there is an entry
-	 * here that will work.
+	 * The bitmap that covers offset won't be in the list unless offset
+	 * is just its start offset.
 	 */
+	entry = list_first_entry(bitmaps, struct btrfs_free_space, list);
+	if (entry->offset != bitmap_offset) {
+		entry = tree_search_offset(ctl, bitmap_offset, 1, 0);
+		if (entry && list_empty(&entry->list))
+			list_add(&entry->list, bitmaps);
+	}
+
 	list_for_each_entry(entry, bitmaps, list) {
 		if (entry->bytes < min_bytes)
 			continue;
@@ -2468,38 +2491,10 @@
 	}
 
 	/*
-	 * If we do have entries on our list and we are here then we didn't find
-	 * anything, so go ahead and get the next entry after the last entry in
-	 * this list and start the search from there.
+	 * The bitmaps list has all the bitmaps that record free space
+	 * starting after offset, so no more search is required.
 	 */
-	if (!list_empty(bitmaps)) {
-		entry = list_entry(bitmaps->prev, struct btrfs_free_space,
-				   list);
-		node = rb_next(&entry->offset_index);
-		if (!node)
-			return -ENOSPC;
-		entry = rb_entry(node, struct btrfs_free_space, offset_index);
-		goto search;
-	}
-
-	entry = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), 0, 1);
-	if (!entry)
-		return -ENOSPC;
-
-search:
-	node = &entry->offset_index;
-	do {
-		entry = rb_entry(node, struct btrfs_free_space, offset_index);
-		node = rb_next(&entry->offset_index);
-		if (!entry->bitmap)
-			continue;
-		if (entry->bytes < min_bytes)
-			continue;
-		ret = btrfs_bitmap_cluster(block_group, entry, cluster, offset,
-					   bytes, min_bytes);
-	} while (ret && node);
-
-	return ret;
+	return -ENOSPC;
 }
 
 /*
@@ -2517,8 +2512,8 @@
 			     u64 offset, u64 bytes, u64 empty_size)
 {
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
-	struct list_head bitmaps;
 	struct btrfs_free_space *entry, *tmp;
+	LIST_HEAD(bitmaps);
 	u64 min_bytes;
 	int ret;
 
@@ -2557,7 +2552,6 @@
 		goto out;
 	}
 
-	INIT_LIST_HEAD(&bitmaps);
 	ret = setup_cluster_no_bitmap(block_group, cluster, &bitmaps, offset,
 				      bytes, min_bytes);
 	if (ret)
diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c
index 53dcbdf..f8962a9 100644
--- a/fs/btrfs/inode-map.c
+++ b/fs/btrfs/inode-map.c
@@ -398,6 +398,8 @@
 	struct btrfs_free_space_ctl *ctl = root->free_ino_ctl;
 	struct btrfs_path *path;
 	struct inode *inode;
+	struct btrfs_block_rsv *rsv;
+	u64 num_bytes;
 	u64 alloc_hint = 0;
 	int ret;
 	int prealloc;
@@ -421,11 +423,26 @@
 	if (!path)
 		return -ENOMEM;
 
+	rsv = trans->block_rsv;
+	trans->block_rsv = &root->fs_info->trans_block_rsv;
+
+	num_bytes = trans->bytes_reserved;
+	/*
+	 * 1 item for inode item insertion if need
+	 * 3 items for inode item update (in the worst case)
+	 * 1 item for free space object
+	 * 3 items for pre-allocation
+	 */
+	trans->bytes_reserved = btrfs_calc_trans_metadata_size(root, 8);
+	ret = btrfs_block_rsv_add_noflush(root, trans->block_rsv,
+					  trans->bytes_reserved);
+	if (ret)
+		goto out;
 again:
 	inode = lookup_free_ino_inode(root, path);
 	if (IS_ERR(inode) && PTR_ERR(inode) != -ENOENT) {
 		ret = PTR_ERR(inode);
-		goto out;
+		goto out_release;
 	}
 
 	if (IS_ERR(inode)) {
@@ -434,7 +451,7 @@
 
 		ret = create_free_ino_inode(root, trans, path);
 		if (ret)
-			goto out;
+			goto out_release;
 		goto again;
 	}
 
@@ -477,11 +494,14 @@
 	}
 	btrfs_free_reserved_data_space(inode, prealloc);
 
+	ret = btrfs_write_out_ino_cache(root, trans, path);
 out_put:
 	iput(inode);
+out_release:
+	btrfs_block_rsv_release(root, trans->block_rsv, trans->bytes_reserved);
 out:
-	if (ret == 0)
-		ret = btrfs_write_out_ino_cache(root, trans, path);
+	trans->block_rsv = rsv;
+	trans->bytes_reserved = num_bytes;
 
 	btrfs_free_path(path);
 	return ret;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 966ddcc..0a6b928 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -38,6 +38,7 @@
 #include <linux/falloc.h>
 #include <linux/slab.h>
 #include <linux/ratelimit.h>
+#include <linux/mount.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -93,6 +94,8 @@
 				   struct page *locked_page,
 				   u64 start, u64 end, int *page_started,
 				   unsigned long *nr_written, int unlock);
+static noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
+				struct btrfs_root *root, struct inode *inode);
 
 static int btrfs_init_inode_security(struct btrfs_trans_handle *trans,
 				     struct inode *inode,  struct inode *dir,
@@ -1741,7 +1744,7 @@
 				trans = btrfs_join_transaction(root);
 			BUG_ON(IS_ERR(trans));
 			trans->block_rsv = &root->fs_info->delalloc_block_rsv;
-			ret = btrfs_update_inode(trans, root, inode);
+			ret = btrfs_update_inode_fallback(trans, root, inode);
 			BUG_ON(ret);
 		}
 		goto out;
@@ -1791,7 +1794,7 @@
 
 	ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent);
 	if (!ret || !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
-		ret = btrfs_update_inode(trans, root, inode);
+		ret = btrfs_update_inode_fallback(trans, root, inode);
 		BUG_ON(ret);
 	}
 	ret = 0;
@@ -2029,7 +2032,7 @@
 	/* insert an orphan item to track this unlinked/truncated file */
 	if (insert >= 1) {
 		ret = btrfs_insert_orphan_item(trans, root, btrfs_ino(inode));
-		BUG_ON(ret);
+		BUG_ON(ret && ret != -EEXIST);
 	}
 
 	/* insert an orphan item to track subvolume contains orphan files */
@@ -2156,6 +2159,38 @@
 		if (ret && ret != -ESTALE)
 			goto out;
 
+		if (ret == -ESTALE && root == root->fs_info->tree_root) {
+			struct btrfs_root *dead_root;
+			struct btrfs_fs_info *fs_info = root->fs_info;
+			int is_dead_root = 0;
+
+			/*
+			 * this is an orphan in the tree root. Currently these
+			 * could come from 2 sources:
+			 *  a) a snapshot deletion in progress
+			 *  b) a free space cache inode
+			 * We need to distinguish those two, as the snapshot
+			 * orphan must not get deleted.
+			 * find_dead_roots already ran before us, so if this
+			 * is a snapshot deletion, we should find the root
+			 * in the dead_roots list
+			 */
+			spin_lock(&fs_info->trans_lock);
+			list_for_each_entry(dead_root, &fs_info->dead_roots,
+					    root_list) {
+				if (dead_root->root_key.objectid ==
+				    found_key.objectid) {
+					is_dead_root = 1;
+					break;
+				}
+			}
+			spin_unlock(&fs_info->trans_lock);
+			if (is_dead_root) {
+				/* prevent this orphan from being found again */
+				key.offset = found_key.objectid - 1;
+				continue;
+			}
+		}
 		/*
 		 * Inode is already gone but the orphan item is still there,
 		 * kill the orphan item.
@@ -2189,7 +2224,14 @@
 				continue;
 			}
 			nr_truncate++;
+			/*
+			 * Need to hold the imutex for reservation purposes, not
+			 * a huge deal here but I have a WARN_ON in
+			 * btrfs_delalloc_reserve_space to catch offenders.
+			 */
+			mutex_lock(&inode->i_mutex);
 			ret = btrfs_truncate(inode);
+			mutex_unlock(&inode->i_mutex);
 		} else {
 			nr_unlink++;
 		}
@@ -2199,6 +2241,9 @@
 		if (ret)
 			goto out;
 	}
+	/* release the path since we're done with it */
+	btrfs_release_path(path);
+
 	root->orphan_cleanup_state = ORPHAN_CLEANUP_DONE;
 
 	if (root->orphan_block_rsv)
@@ -2426,7 +2471,7 @@
 /*
  * copy everything in the in-memory inode into the btree.
  */
-noinline int btrfs_update_inode(struct btrfs_trans_handle *trans,
+static noinline int btrfs_update_inode_item(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root, struct inode *inode)
 {
 	struct btrfs_inode_item *inode_item;
@@ -2434,21 +2479,6 @@
 	struct extent_buffer *leaf;
 	int ret;
 
-	/*
-	 * If the inode is a free space inode, we can deadlock during commit
-	 * if we put it into the delayed code.
-	 *
-	 * The data relocation inode should also be directly updated
-	 * without delay
-	 */
-	if (!btrfs_is_free_space_inode(root, inode)
-	    && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) {
-		ret = btrfs_delayed_update_inode(trans, root, inode);
-		if (!ret)
-			btrfs_set_inode_last_trans(trans, inode);
-		return ret;
-	}
-
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
@@ -2477,6 +2507,43 @@
 }
 
 /*
+ * copy everything in the in-memory inode into the btree.
+ */
+noinline int btrfs_update_inode(struct btrfs_trans_handle *trans,
+				struct btrfs_root *root, struct inode *inode)
+{
+	int ret;
+
+	/*
+	 * If the inode is a free space inode, we can deadlock during commit
+	 * if we put it into the delayed code.
+	 *
+	 * The data relocation inode should also be directly updated
+	 * without delay
+	 */
+	if (!btrfs_is_free_space_inode(root, inode)
+	    && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) {
+		ret = btrfs_delayed_update_inode(trans, root, inode);
+		if (!ret)
+			btrfs_set_inode_last_trans(trans, inode);
+		return ret;
+	}
+
+	return btrfs_update_inode_item(trans, root, inode);
+}
+
+static noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
+				struct btrfs_root *root, struct inode *inode)
+{
+	int ret;
+
+	ret = btrfs_update_inode(trans, root, inode);
+	if (ret == -ENOSPC)
+		return btrfs_update_inode_item(trans, root, inode);
+	return ret;
+}
+
+/*
  * unlink helper that gets used here in inode.c and in the tree logging
  * recovery code.  It remove a link in a directory with a given name, and
  * also drops the back refs in the inode to the directory
@@ -3300,7 +3367,7 @@
 			u64 hint_byte = 0;
 			hole_size = last_byte - cur_offset;
 
-			trans = btrfs_start_transaction(root, 2);
+			trans = btrfs_start_transaction(root, 3);
 			if (IS_ERR(trans)) {
 				err = PTR_ERR(trans);
 				break;
@@ -3310,6 +3377,7 @@
 						 cur_offset + hole_size,
 						 &hint_byte, 1);
 			if (err) {
+				btrfs_update_inode(trans, root, inode);
 				btrfs_end_transaction(trans, root);
 				break;
 			}
@@ -3319,6 +3387,7 @@
 					0, hole_size, 0, hole_size,
 					0, 0, 0);
 			if (err) {
+				btrfs_update_inode(trans, root, inode);
 				btrfs_end_transaction(trans, root);
 				break;
 			}
@@ -3326,6 +3395,7 @@
 			btrfs_drop_extent_cache(inode, hole_start,
 					last_byte - 1, 0);
 
+			btrfs_update_inode(trans, root, inode);
 			btrfs_end_transaction(trans, root);
 		}
 		free_extent_map(em);
@@ -3343,6 +3413,8 @@
 
 static int btrfs_setsize(struct inode *inode, loff_t newsize)
 {
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_trans_handle *trans;
 	loff_t oldsize = i_size_read(inode);
 	int ret;
 
@@ -3350,16 +3422,19 @@
 		return 0;
 
 	if (newsize > oldsize) {
-		i_size_write(inode, newsize);
-		btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL);
 		truncate_pagecache(inode, oldsize, newsize);
 		ret = btrfs_cont_expand(inode, oldsize, newsize);
-		if (ret) {
-			btrfs_setsize(inode, oldsize);
+		if (ret)
 			return ret;
-		}
 
-		mark_inode_dirty(inode);
+		trans = btrfs_start_transaction(root, 1);
+		if (IS_ERR(trans))
+			return PTR_ERR(trans);
+
+		i_size_write(inode, newsize);
+		btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL);
+		ret = btrfs_update_inode(trans, root, inode);
+		btrfs_end_transaction_throttle(trans, root);
 	} else {
 
 		/*
@@ -3399,9 +3474,9 @@
 
 	if (attr->ia_valid) {
 		setattr_copy(inode, attr);
-		mark_inode_dirty(inode);
+		err = btrfs_dirty_inode(inode);
 
-		if (attr->ia_valid & ATTR_MODE)
+		if (!err && attr->ia_valid & ATTR_MODE)
 			err = btrfs_acl_chmod(inode);
 	}
 
@@ -3463,7 +3538,7 @@
 	 * doing the truncate.
 	 */
 	while (1) {
-		ret = btrfs_block_rsv_refill(root, rsv, min_size);
+		ret = btrfs_block_rsv_refill_noflush(root, rsv, min_size);
 
 		/*
 		 * Try and steal from the global reserve since we will
@@ -4177,42 +4252,80 @@
  * FIXME, needs more benchmarking...there are no reasons other than performance
  * to keep or drop this code.
  */
-void btrfs_dirty_inode(struct inode *inode, int flags)
+int btrfs_dirty_inode(struct inode *inode)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_trans_handle *trans;
 	int ret;
 
 	if (BTRFS_I(inode)->dummy_inode)
-		return;
+		return 0;
 
 	trans = btrfs_join_transaction(root);
-	BUG_ON(IS_ERR(trans));
+	if (IS_ERR(trans))
+		return PTR_ERR(trans);
 
 	ret = btrfs_update_inode(trans, root, inode);
 	if (ret && ret == -ENOSPC) {
 		/* whoops, lets try again with the full transaction */
 		btrfs_end_transaction(trans, root);
 		trans = btrfs_start_transaction(root, 1);
-		if (IS_ERR(trans)) {
-			printk_ratelimited(KERN_ERR "btrfs: fail to "
-				       "dirty  inode %llu error %ld\n",
-				       (unsigned long long)btrfs_ino(inode),
-				       PTR_ERR(trans));
-			return;
-		}
+		if (IS_ERR(trans))
+			return PTR_ERR(trans);
 
 		ret = btrfs_update_inode(trans, root, inode);
-		if (ret) {
-			printk_ratelimited(KERN_ERR "btrfs: fail to "
-				       "dirty  inode %llu error %d\n",
-				       (unsigned long long)btrfs_ino(inode),
-				       ret);
-		}
 	}
 	btrfs_end_transaction(trans, root);
 	if (BTRFS_I(inode)->delayed_node)
 		btrfs_balance_delayed_items(root);
+
+	return ret;
+}
+
+/*
+ * This is a copy of file_update_time.  We need this so we can return error on
+ * ENOSPC for updating the inode in the case of file write and mmap writes.
+ */
+int btrfs_update_time(struct file *file)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct timespec now;
+	int ret;
+	enum { S_MTIME = 1, S_CTIME = 2, S_VERSION = 4 } sync_it = 0;
+
+	/* First try to exhaust all avenues to not sync */
+	if (IS_NOCMTIME(inode))
+		return 0;
+
+	now = current_fs_time(inode->i_sb);
+	if (!timespec_equal(&inode->i_mtime, &now))
+		sync_it = S_MTIME;
+
+	if (!timespec_equal(&inode->i_ctime, &now))
+		sync_it |= S_CTIME;
+
+	if (IS_I_VERSION(inode))
+		sync_it |= S_VERSION;
+
+	if (!sync_it)
+		return 0;
+
+	/* Finally allowed to write? Takes lock. */
+	if (mnt_want_write_file(file))
+		return 0;
+
+	/* Only change inode inside the lock region */
+	if (sync_it & S_VERSION)
+		inode_inc_iversion(inode);
+	if (sync_it & S_CTIME)
+		inode->i_ctime = now;
+	if (sync_it & S_MTIME)
+		inode->i_mtime = now;
+	ret = btrfs_dirty_inode(inode);
+	if (!ret)
+		mark_inode_dirty_sync(inode);
+	mnt_drop_write(file->f_path.mnt);
+	return ret;
 }
 
 /*
@@ -4528,11 +4641,18 @@
 		goto out_unlock;
 	}
 
+	/*
+	* If the active LSM wants to access the inode during
+	* d_instantiate it needs these. Smack checks to see
+	* if the filesystem supports xattrs by looking at the
+	* ops vector.
+	*/
+
+	inode->i_op = &btrfs_special_inode_operations;
 	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
 	if (err)
 		drop_inode = 1;
 	else {
-		inode->i_op = &btrfs_special_inode_operations;
 		init_special_inode(inode, inode->i_mode, rdev);
 		btrfs_update_inode(trans, root, inode);
 	}
@@ -4586,14 +4706,21 @@
 		goto out_unlock;
 	}
 
+	/*
+	* If the active LSM wants to access the inode during
+	* d_instantiate it needs these. Smack checks to see
+	* if the filesystem supports xattrs by looking at the
+	* ops vector.
+	*/
+	inode->i_fop = &btrfs_file_operations;
+	inode->i_op = &btrfs_file_inode_operations;
+
 	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
 	if (err)
 		drop_inode = 1;
 	else {
 		inode->i_mapping->a_ops = &btrfs_aops;
 		inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
-		inode->i_fop = &btrfs_file_operations;
-		inode->i_op = &btrfs_file_inode_operations;
 		BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
 	}
 out_unlock:
@@ -5632,7 +5759,7 @@
 	if (test_bit(BTRFS_ORDERED_NOCOW, &ordered->flags)) {
 		ret = btrfs_ordered_update_i_size(inode, 0, ordered);
 		if (!ret)
-			err = btrfs_update_inode(trans, root, inode);
+			err = btrfs_update_inode_fallback(trans, root, inode);
 		goto out;
 	}
 
@@ -5670,7 +5797,7 @@
 	add_pending_csums(trans, inode, ordered->file_offset, &ordered->list);
 	ret = btrfs_ordered_update_i_size(inode, 0, ordered);
 	if (!ret || !test_bit(BTRFS_ORDERED_PREALLOC, &ordered->flags))
-		btrfs_update_inode(trans, root, inode);
+		btrfs_update_inode_fallback(trans, root, inode);
 	ret = 0;
 out_unlock:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, ordered->file_offset,
@@ -6276,7 +6403,12 @@
 	u64 page_start;
 	u64 page_end;
 
+	/* Need this to keep space reservations serialized */
+	mutex_lock(&inode->i_mutex);
 	ret  = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE);
+	mutex_unlock(&inode->i_mutex);
+	if (!ret)
+		ret = btrfs_update_time(vma->vm_file);
 	if (ret) {
 		if (ret == -ENOMEM)
 			ret = VM_FAULT_OOM;
@@ -6488,8 +6620,9 @@
 			/* Just need the 1 for updating the inode */
 			trans = btrfs_start_transaction(root, 1);
 			if (IS_ERR(trans)) {
-				err = PTR_ERR(trans);
-				goto out;
+				ret = err = PTR_ERR(trans);
+				trans = NULL;
+				break;
 			}
 		}
 
@@ -6529,14 +6662,16 @@
 		ret = btrfs_orphan_del(NULL, inode);
 	}
 
-	trans->block_rsv = &root->fs_info->trans_block_rsv;
-	ret = btrfs_update_inode(trans, root, inode);
-	if (ret && !err)
-		err = ret;
+	if (trans) {
+		trans->block_rsv = &root->fs_info->trans_block_rsv;
+		ret = btrfs_update_inode(trans, root, inode);
+		if (ret && !err)
+			err = ret;
 
-	nr = trans->blocks_used;
-	ret = btrfs_end_transaction_throttle(trans, root);
-	btrfs_btree_balance_dirty(root, nr);
+		nr = trans->blocks_used;
+		ret = btrfs_end_transaction_throttle(trans, root);
+		btrfs_btree_balance_dirty(root, nr);
+	}
 
 out:
 	btrfs_free_block_rsv(root, rsv);
@@ -6605,6 +6740,7 @@
 	ei->orphan_meta_reserved = 0;
 	ei->dummy_inode = 0;
 	ei->in_defrag = 0;
+	ei->delalloc_meta_reserved = 0;
 	ei->force_compress = BTRFS_COMPRESS_NONE;
 
 	ei->delayed_node = NULL;
@@ -6764,11 +6900,13 @@
 			 struct dentry *dentry, struct kstat *stat)
 {
 	struct inode *inode = dentry->d_inode;
+	u32 blocksize = inode->i_sb->s_blocksize;
+
 	generic_fillattr(inode, stat);
 	stat->dev = BTRFS_I(inode)->root->anon_dev;
 	stat->blksize = PAGE_CACHE_SIZE;
-	stat->blocks = (inode_get_bytes(inode) +
-			BTRFS_I(inode)->delalloc_bytes) >> 9;
+	stat->blocks = (ALIGN(inode_get_bytes(inode), blocksize) +
+		ALIGN(BTRFS_I(inode)->delalloc_bytes, blocksize)) >> 9;
 	return 0;
 }
 
@@ -7044,14 +7182,21 @@
 		goto out_unlock;
 	}
 
+	/*
+	* If the active LSM wants to access the inode during
+	* d_instantiate it needs these. Smack checks to see
+	* if the filesystem supports xattrs by looking at the
+	* ops vector.
+	*/
+	inode->i_fop = &btrfs_file_operations;
+	inode->i_op = &btrfs_file_inode_operations;
+
 	err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index);
 	if (err)
 		drop_inode = 1;
 	else {
 		inode->i_mapping->a_ops = &btrfs_aops;
 		inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
-		inode->i_fop = &btrfs_file_operations;
-		inode->i_op = &btrfs_file_inode_operations;
 		BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
 	}
 	if (drop_inode)
@@ -7321,6 +7466,7 @@
 	.follow_link	= page_follow_link_light,
 	.put_link	= page_put_link,
 	.getattr	= btrfs_getattr,
+	.setattr	= btrfs_setattr,
 	.permission	= btrfs_permission,
 	.setxattr	= btrfs_setxattr,
 	.getxattr	= btrfs_getxattr,
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 4a34c47..c04f02c 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -252,11 +252,11 @@
 	trans = btrfs_join_transaction(root);
 	BUG_ON(IS_ERR(trans));
 
+	btrfs_update_iflags(inode);
+	inode->i_ctime = CURRENT_TIME;
 	ret = btrfs_update_inode(trans, root, inode);
 	BUG_ON(ret);
 
-	btrfs_update_iflags(inode);
-	inode->i_ctime = CURRENT_TIME;
 	btrfs_end_transaction(trans, root);
 
 	mnt_drop_write(file->f_path.mnt);
@@ -858,8 +858,10 @@
 		return 0;
 	file_end = (isize - 1) >> PAGE_CACHE_SHIFT;
 
+	mutex_lock(&inode->i_mutex);
 	ret = btrfs_delalloc_reserve_space(inode,
 					   num_pages << PAGE_CACHE_SHIFT);
+	mutex_unlock(&inode->i_mutex);
 	if (ret)
 		return ret;
 again:
@@ -1216,12 +1218,12 @@
 		*devstr = '\0';
 		devstr = vol_args->name;
 		devid = simple_strtoull(devstr, &end, 10);
-		printk(KERN_INFO "resizing devid %llu\n",
+		printk(KERN_INFO "btrfs: resizing devid %llu\n",
 		       (unsigned long long)devid);
 	}
 	device = btrfs_find_device(root, devid, NULL, NULL);
 	if (!device) {
-		printk(KERN_INFO "resizer unable to find device %llu\n",
+		printk(KERN_INFO "btrfs: resizer unable to find device %llu\n",
 		       (unsigned long long)devid);
 		ret = -EINVAL;
 		goto out_unlock;
@@ -1267,7 +1269,7 @@
 	do_div(new_size, root->sectorsize);
 	new_size *= root->sectorsize;
 
-	printk(KERN_INFO "new size for %s is %llu\n",
+	printk(KERN_INFO "btrfs: new size for %s is %llu\n",
 		device->name, (unsigned long long)new_size);
 
 	if (new_size > old_size) {
@@ -1278,7 +1280,7 @@
 		}
 		ret = btrfs_grow_device(trans, device, new_size);
 		btrfs_commit_transaction(trans, root);
-	} else {
+	} else if (new_size < old_size) {
 		ret = btrfs_shrink_device(device, new_size);
 	}
 
@@ -2930,11 +2932,13 @@
 		goto out;
 
 	for (i = 0; i < ipath->fspath->elem_cnt; ++i) {
-		rel_ptr = ipath->fspath->val[i] - (u64)ipath->fspath->val;
+		rel_ptr = ipath->fspath->val[i] -
+			  (u64)(unsigned long)ipath->fspath->val;
 		ipath->fspath->val[i] = rel_ptr;
 	}
 
-	ret = copy_to_user((void *)ipa->fspath, (void *)ipath->fspath, size);
+	ret = copy_to_user((void *)(unsigned long)ipa->fspath,
+			   (void *)(unsigned long)ipath->fspath, size);
 	if (ret) {
 		ret = -EFAULT;
 		goto out;
@@ -3017,7 +3021,8 @@
 	if (ret < 0)
 		goto out;
 
-	ret = copy_to_user((void *)loi->inodes, (void *)inodes, size);
+	ret = copy_to_user((void *)(unsigned long)loi->inodes,
+			   (void *)(unsigned long)inodes, size);
 	if (ret)
 		ret = -EFAULT;
 
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 24d654c..cfb5543 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1174,6 +1174,8 @@
 			list_add_tail(&new_edge->list[UPPER],
 				      &new_node->lower);
 		}
+	} else {
+		list_add_tail(&new_node->lower, &cache->leaves);
 	}
 
 	rb_node = tree_insert(&cache->rb_root, new_node->bytenr,
@@ -2945,7 +2947,9 @@
 	index = (cluster->start - offset) >> PAGE_CACHE_SHIFT;
 	last_index = (cluster->end - offset) >> PAGE_CACHE_SHIFT;
 	while (index <= last_index) {
+		mutex_lock(&inode->i_mutex);
 		ret = btrfs_delalloc_reserve_metadata(inode, PAGE_CACHE_SIZE);
+		mutex_unlock(&inode->i_mutex);
 		if (ret)
 			goto out;
 
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index ed11d38..ddf2c90 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -256,6 +256,11 @@
 	btrfs_release_path(swarn->path);
 
 	ipath = init_ipath(4096, local_root, swarn->path);
+	if (IS_ERR(ipath)) {
+		ret = PTR_ERR(ipath);
+		ipath = NULL;
+		goto err;
+	}
 	ret = paths_from_inode(inum, ipath);
 
 	if (ret < 0)
@@ -272,7 +277,7 @@
 			swarn->logical, swarn->dev->name,
 			(unsigned long long)swarn->sector, root, inum, offset,
 			min(isize - offset, (u64)PAGE_SIZE), nlink,
-			(char *)ipath->fspath->val[i]);
+			(char *)(unsigned long)ipath->fspath->val[i]);
 
 	free_ipath(ipath);
 	return 0;
@@ -944,50 +949,18 @@
 static int scrub_submit(struct scrub_dev *sdev)
 {
 	struct scrub_bio *sbio;
-	struct bio *bio;
-	int i;
 
 	if (sdev->curr == -1)
 		return 0;
 
 	sbio = sdev->bios[sdev->curr];
-
-	bio = bio_alloc(GFP_NOFS, sbio->count);
-	if (!bio)
-		goto nomem;
-
-	bio->bi_private = sbio;
-	bio->bi_end_io = scrub_bio_end_io;
-	bio->bi_bdev = sdev->dev->bdev;
-	bio->bi_sector = sbio->physical >> 9;
-
-	for (i = 0; i < sbio->count; ++i) {
-		struct page *page;
-		int ret;
-
-		page = alloc_page(GFP_NOFS);
-		if (!page)
-			goto nomem;
-
-		ret = bio_add_page(bio, page, PAGE_SIZE, 0);
-		if (!ret) {
-			__free_page(page);
-			goto nomem;
-		}
-	}
-
 	sbio->err = 0;
 	sdev->curr = -1;
 	atomic_inc(&sdev->in_flight);
 
-	submit_bio(READ, bio);
+	submit_bio(READ, sbio->bio);
 
 	return 0;
-
-nomem:
-	scrub_free_bio(bio);
-
-	return -ENOMEM;
 }
 
 static int scrub_page(struct scrub_dev *sdev, u64 logical, u64 len,
@@ -995,6 +968,8 @@
 		      u8 *csum, int force)
 {
 	struct scrub_bio *sbio;
+	struct page *page;
+	int ret;
 
 again:
 	/*
@@ -1015,12 +990,22 @@
 	}
 	sbio = sdev->bios[sdev->curr];
 	if (sbio->count == 0) {
+		struct bio *bio;
+
 		sbio->physical = physical;
 		sbio->logical = logical;
+		bio = bio_alloc(GFP_NOFS, SCRUB_PAGES_PER_BIO);
+		if (!bio)
+			return -ENOMEM;
+
+		bio->bi_private = sbio;
+		bio->bi_end_io = scrub_bio_end_io;
+		bio->bi_bdev = sdev->dev->bdev;
+		bio->bi_sector = sbio->physical >> 9;
+		sbio->err = 0;
+		sbio->bio = bio;
 	} else if (sbio->physical + sbio->count * PAGE_SIZE != physical ||
 		   sbio->logical + sbio->count * PAGE_SIZE != logical) {
-		int ret;
-
 		ret = scrub_submit(sdev);
 		if (ret)
 			return ret;
@@ -1030,6 +1015,20 @@
 	sbio->spag[sbio->count].generation = gen;
 	sbio->spag[sbio->count].have_csum = 0;
 	sbio->spag[sbio->count].mirror_num = mirror_num;
+
+	page = alloc_page(GFP_NOFS);
+	if (!page)
+		return -ENOMEM;
+
+	ret = bio_add_page(sbio->bio, page, PAGE_SIZE, 0);
+	if (!ret) {
+		__free_page(page);
+		ret = scrub_submit(sdev);
+		if (ret)
+			return ret;
+		goto again;
+	}
+
 	if (csum) {
 		sbio->spag[sbio->count].have_csum = 1;
 		memcpy(sbio->spag[sbio->count].csum, csum, sdev->csum_size);
@@ -1536,18 +1535,22 @@
 static noinline_for_stack int scrub_workers_get(struct btrfs_root *root)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
+	int ret = 0;
 
 	mutex_lock(&fs_info->scrub_lock);
 	if (fs_info->scrub_workers_refcnt == 0) {
 		btrfs_init_workers(&fs_info->scrub_workers, "scrub",
 			   fs_info->thread_pool_size, &fs_info->generic_worker);
 		fs_info->scrub_workers.idle_thresh = 4;
-		btrfs_start_workers(&fs_info->scrub_workers, 1);
+		ret = btrfs_start_workers(&fs_info->scrub_workers);
+		if (ret)
+			goto out;
 	}
 	++fs_info->scrub_workers_refcnt;
+out:
 	mutex_unlock(&fs_info->scrub_lock);
 
-	return 0;
+	return ret;
 }
 
 static noinline_for_stack void scrub_workers_put(struct btrfs_root *root)
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 57080df..200f63b 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -41,6 +41,7 @@
 #include <linux/slab.h>
 #include <linux/cleancache.h>
 #include <linux/mnt_namespace.h>
+#include <linux/ratelimit.h>
 #include "compat.h"
 #include "delayed-inode.h"
 #include "ctree.h"
@@ -197,7 +198,7 @@
 	{Opt_subvolrootid, "subvolrootid=%d"},
 	{Opt_defrag, "autodefrag"},
 	{Opt_inode_cache, "inode_cache"},
-	{Opt_no_space_cache, "no_space_cache"},
+	{Opt_no_space_cache, "nospace_cache"},
 	{Opt_recovery, "recovery"},
 	{Opt_err, NULL},
 };
@@ -448,6 +449,7 @@
 		token = match_token(p, tokens, args);
 		switch (token) {
 		case Opt_subvol:
+			kfree(*subvol_name);
 			*subvol_name = match_strdup(&args[0]);
 			break;
 		case Opt_subvolid:
@@ -710,7 +712,7 @@
 	if (btrfs_test_opt(root, SPACE_CACHE))
 		seq_puts(seq, ",space_cache");
 	else
-		seq_puts(seq, ",no_space_cache");
+		seq_puts(seq, ",nospace_cache");
 	if (btrfs_test_opt(root, CLEAR_CACHE))
 		seq_puts(seq, ",clear_cache");
 	if (btrfs_test_opt(root, USER_SUBVOL_RM_ALLOWED))
@@ -824,13 +826,9 @@
 static struct dentry *mount_subvol(const char *subvol_name, int flags,
 				   const char *device_name, char *data)
 {
-	struct super_block *s;
 	struct dentry *root;
 	struct vfsmount *mnt;
-	struct mnt_namespace *ns_private;
 	char *newargs;
-	struct path path;
-	int error;
 
 	newargs = setup_root_args(data);
 	if (!newargs)
@@ -841,39 +839,17 @@
 	if (IS_ERR(mnt))
 		return ERR_CAST(mnt);
 
-	ns_private = create_mnt_ns(mnt);
-	if (IS_ERR(ns_private)) {
-		mntput(mnt);
-		return ERR_CAST(ns_private);
-	}
+	root = mount_subtree(mnt, subvol_name);
 
-	/*
-	 * This will trigger the automount of the subvol so we can just
-	 * drop the mnt we have here and return the dentry that we
-	 * found.
-	 */
-	error = vfs_path_lookup(mnt->mnt_root, mnt, subvol_name,
-				LOOKUP_FOLLOW, &path);
-	put_mnt_ns(ns_private);
-	if (error)
-		return ERR_PTR(error);
-
-	if (!is_subvolume_inode(path.dentry->d_inode)) {
-		path_put(&path);
-		mntput(mnt);
-		error = -EINVAL;
+	if (!IS_ERR(root) && !is_subvolume_inode(root->d_inode)) {
+		struct super_block *s = root->d_sb;
+		dput(root);
+		root = ERR_PTR(-EINVAL);
+		deactivate_locked_super(s);
 		printk(KERN_ERR "btrfs: '%s' is not a valid subvolume\n",
 				subvol_name);
-		return ERR_PTR(-EINVAL);
 	}
 
-	/* Get a ref to the sb and the dentry we found and return it */
-	s = path.mnt->mnt_sb;
-	atomic_inc(&s->s_active);
-	root = dget(path.dentry);
-	path_put(&path);
-	down_write(&s->s_umount);
-
 	return root;
 }
 
@@ -890,7 +866,6 @@
 	struct super_block *s;
 	struct dentry *root;
 	struct btrfs_fs_devices *fs_devices = NULL;
-	struct btrfs_root *tree_root = NULL;
 	struct btrfs_fs_info *fs_info = NULL;
 	fmode_t mode = FMODE_READ;
 	char *subvol_name = NULL;
@@ -904,8 +879,10 @@
 	error = btrfs_parse_early_options(data, mode, fs_type,
 					  &subvol_name, &subvol_objectid,
 					  &subvol_rootid, &fs_devices);
-	if (error)
+	if (error) {
+		kfree(subvol_name);
 		return ERR_PTR(error);
+	}
 
 	if (subvol_name) {
 		root = mount_subvol(subvol_name, flags, device_name, data);
@@ -917,15 +894,6 @@
 	if (error)
 		return ERR_PTR(error);
 
-	error = btrfs_open_devices(fs_devices, mode, fs_type);
-	if (error)
-		return ERR_PTR(error);
-
-	if (!(flags & MS_RDONLY) && fs_devices->rw_devices == 0) {
-		error = -EACCES;
-		goto error_close_devices;
-	}
-
 	/*
 	 * Setup a dummy root and fs_info for test/set super.  This is because
 	 * we don't actually fill this stuff out until open_ctree, but we need
@@ -933,24 +901,36 @@
 	 * then open_ctree will properly initialize everything later.
 	 */
 	fs_info = kzalloc(sizeof(struct btrfs_fs_info), GFP_NOFS);
-	tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
-	if (!fs_info || !tree_root) {
+	if (!fs_info)
+		return ERR_PTR(-ENOMEM);
+
+	fs_info->tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
+	if (!fs_info->tree_root) {
 		error = -ENOMEM;
-		goto error_close_devices;
+		goto error_fs_info;
 	}
-	fs_info->tree_root = tree_root;
+	fs_info->tree_root->fs_info = fs_info;
 	fs_info->fs_devices = fs_devices;
-	tree_root->fs_info = fs_info;
 
 	fs_info->super_copy = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_NOFS);
 	fs_info->super_for_commit = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_NOFS);
 	if (!fs_info->super_copy || !fs_info->super_for_commit) {
 		error = -ENOMEM;
+		goto error_fs_info;
+	}
+
+	error = btrfs_open_devices(fs_devices, mode, fs_type);
+	if (error)
+		goto error_fs_info;
+
+	if (!(flags & MS_RDONLY) && fs_devices->rw_devices == 0) {
+		error = -EACCES;
 		goto error_close_devices;
 	}
 
 	bdev = fs_devices->latest_bdev;
-	s = sget(fs_type, btrfs_test_super, btrfs_set_super, tree_root);
+	s = sget(fs_type, btrfs_test_super, btrfs_set_super,
+		 fs_info->tree_root);
 	if (IS_ERR(s)) {
 		error = PTR_ERR(s);
 		goto error_close_devices;
@@ -959,12 +939,12 @@
 	if (s->s_root) {
 		if ((flags ^ s->s_flags) & MS_RDONLY) {
 			deactivate_locked_super(s);
-			return ERR_PTR(-EBUSY);
+			error = -EBUSY;
+			goto error_close_devices;
 		}
 
 		btrfs_close_devices(fs_devices);
 		free_fs_info(fs_info);
-		kfree(tree_root);
 	} else {
 		char b[BDEVNAME_SIZE];
 
@@ -991,8 +971,8 @@
 
 error_close_devices:
 	btrfs_close_devices(fs_devices);
+error_fs_info:
 	free_fs_info(fs_info);
-	kfree(tree_root);
 	return ERR_PTR(error);
 }
 
@@ -1074,11 +1054,11 @@
 	u64 avail_space;
 	u64 used_space;
 	u64 min_stripe_size;
-	int min_stripes = 1;
+	int min_stripes = 1, num_stripes = 1;
 	int i = 0, nr_devices;
 	int ret;
 
-	nr_devices = fs_info->fs_devices->rw_devices;
+	nr_devices = fs_info->fs_devices->open_devices;
 	BUG_ON(!nr_devices);
 
 	devices_info = kmalloc(sizeof(*devices_info) * nr_devices,
@@ -1088,20 +1068,24 @@
 
 	/* calc min stripe number for data space alloction */
 	type = btrfs_get_alloc_profile(root, 1);
-	if (type & BTRFS_BLOCK_GROUP_RAID0)
+	if (type & BTRFS_BLOCK_GROUP_RAID0) {
 		min_stripes = 2;
-	else if (type & BTRFS_BLOCK_GROUP_RAID1)
+		num_stripes = nr_devices;
+	} else if (type & BTRFS_BLOCK_GROUP_RAID1) {
 		min_stripes = 2;
-	else if (type & BTRFS_BLOCK_GROUP_RAID10)
+		num_stripes = 2;
+	} else if (type & BTRFS_BLOCK_GROUP_RAID10) {
 		min_stripes = 4;
+		num_stripes = 4;
+	}
 
 	if (type & BTRFS_BLOCK_GROUP_DUP)
 		min_stripe_size = 2 * BTRFS_STRIPE_LEN;
 	else
 		min_stripe_size = BTRFS_STRIPE_LEN;
 
-	list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) {
-		if (!device->in_fs_metadata)
+	list_for_each_entry(device, &fs_devices->devices, dev_list) {
+		if (!device->in_fs_metadata || !device->bdev)
 			continue;
 
 		avail_space = device->total_bytes - device->bytes_used;
@@ -1162,13 +1146,16 @@
 	i = nr_devices - 1;
 	avail_space = 0;
 	while (nr_devices >= min_stripes) {
+		if (num_stripes > nr_devices)
+			num_stripes = nr_devices;
+
 		if (devices_info[i].max_avail >= min_stripe_size) {
 			int j;
 			u64 alloc_size;
 
-			avail_space += devices_info[i].max_avail * min_stripes;
+			avail_space += devices_info[i].max_avail * num_stripes;
 			alloc_size = devices_info[i].max_avail;
-			for (j = i + 1 - min_stripes; j <= i; j++)
+			for (j = i + 1 - num_stripes; j <= i; j++)
 				devices_info[j].max_avail -= alloc_size;
 		}
 		i--;
@@ -1285,6 +1272,16 @@
 	return 0;
 }
 
+static void btrfs_fs_dirty_inode(struct inode *inode, int flags)
+{
+	int ret;
+
+	ret = btrfs_dirty_inode(inode);
+	if (ret)
+		printk_ratelimited(KERN_ERR "btrfs: fail to dirty inode %Lu "
+				   "error %d\n", btrfs_ino(inode), ret);
+}
+
 static const struct super_operations btrfs_super_ops = {
 	.drop_inode	= btrfs_drop_inode,
 	.evict_inode	= btrfs_evict_inode,
@@ -1292,7 +1289,7 @@
 	.sync_fs	= btrfs_sync_fs,
 	.show_options	= btrfs_show_options,
 	.write_inode	= btrfs_write_inode,
-	.dirty_inode	= btrfs_dirty_inode,
+	.dirty_inode	= btrfs_fs_dirty_inode,
 	.alloc_inode	= btrfs_alloc_inode,
 	.destroy_inode	= btrfs_destroy_inode,
 	.statfs		= btrfs_statfs,
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 960835e..81376d9 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -785,6 +785,10 @@
 
 			btrfs_save_ino_cache(root, trans);
 
+			/* see comments in should_cow_block() */
+			root->force_cow = 0;
+			smp_wmb();
+
 			if (root->commit_root != root->node) {
 				mutex_lock(&root->fs_commit_mutex);
 				switch_commit_root(root);
@@ -882,8 +886,8 @@
 	btrfs_reloc_pre_snapshot(trans, pending, &to_reserve);
 
 	if (to_reserve > 0) {
-		ret = btrfs_block_rsv_add(root, &pending->block_rsv,
-					  to_reserve);
+		ret = btrfs_block_rsv_add_noflush(root, &pending->block_rsv,
+						  to_reserve);
 		if (ret) {
 			pending->error = ret;
 			goto fail;
@@ -947,6 +951,10 @@
 	btrfs_tree_unlock(old);
 	free_extent_buffer(old);
 
+	/* see comments in should_cow_block() */
+	root->force_cow = 1;
+	smp_wmb();
+
 	btrfs_set_root_node(new_root_item, tmp);
 	/* record when the snapshot was created in key.offset */
 	key.offset = trans->transid;
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index f8e29431..f4b839f 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -295,6 +295,12 @@
 			btrfs_requeue_work(&device->work);
 			goto done;
 		}
+		/* unplug every 64 requests just for good measure */
+		if (batch_run % 64 == 0) {
+			blk_finish_plug(&plug);
+			blk_start_plug(&plug);
+			sync_pending = 0;
+		}
 	}
 
 	cond_resched();
@@ -999,7 +1005,7 @@
 	key.objectid = device->devid;
 	key.offset = start;
 	key.type = BTRFS_DEV_EXTENT_KEY;
-
+again:
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 	if (ret > 0) {
 		ret = btrfs_previous_item(root, path, key.objectid,
@@ -1012,6 +1018,9 @@
 					struct btrfs_dev_extent);
 		BUG_ON(found_key.offset > start || found_key.offset +
 		       btrfs_dev_extent_length(leaf, extent) < start);
+		key = found_key;
+		btrfs_release_path(path);
+		goto again;
 	} else if (ret == 0) {
 		leaf = path->nodes[0];
 		extent = btrfs_item_ptr(leaf, path->slots[0],
@@ -1608,7 +1617,7 @@
 	if ((sb->s_flags & MS_RDONLY) && !root->fs_info->fs_devices->seeding)
 		return -EINVAL;
 
-	bdev = blkdev_get_by_path(device_path, FMODE_EXCL,
+	bdev = blkdev_get_by_path(device_path, FMODE_WRITE | FMODE_EXCL,
 				  root->fs_info->bdev_holder);
 	if (IS_ERR(bdev))
 		return PTR_ERR(bdev);
@@ -3255,7 +3264,7 @@
 		 */
 		if (atomic_read(&bbio->error) > bbio->max_errors) {
 			err = -EIO;
-		} else if (err) {
+		} else {
 			/*
 			 * this bio is actually up to date, we didn't
 			 * go over the max number of errors
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index ab5b1c4..78f2d4d 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -100,6 +100,12 @@
 	struct reada_zone *reada_curr_zone;
 	struct radix_tree_root reada_zones;
 	struct radix_tree_root reada_extents;
+
+	/* for sending down flush barriers */
+	struct bio *flush_bio;
+	struct completion flush_wait;
+	int nobarriers;
+
 };
 
 struct btrfs_fs_devices {
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 4144caf..173b1d2 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -87,7 +87,7 @@
 	snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context);
 
 	/* dirty the head */
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (ci->i_head_snapc == NULL)
 		ci->i_head_snapc = ceph_get_snap_context(snapc);
 	++ci->i_wrbuffer_ref_head;
@@ -100,7 +100,7 @@
 	     ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref_head-1,
 	     ci->i_wrbuffer_ref, ci->i_wrbuffer_ref_head,
 	     snapc, snapc->seq, snapc->num_snaps);
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	/* now adjust page */
 	spin_lock_irq(&mapping->tree_lock);
@@ -391,7 +391,7 @@
 	struct ceph_snap_context *snapc = NULL;
 	struct ceph_cap_snap *capsnap = NULL;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
 		dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap,
 		     capsnap->context, capsnap->dirty_pages);
@@ -407,7 +407,7 @@
 		dout(" head snapc %p has %d dirty pages\n",
 		     snapc, ci->i_wrbuffer_ref_head);
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return snapc;
 }
 
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 0f327c6..8b53193 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -309,7 +309,7 @@
 /*
  * Find ceph_cap for given mds, if any.
  *
- * Called with i_lock held.
+ * Called with i_ceph_lock held.
  */
 static struct ceph_cap *__get_cap_for_mds(struct ceph_inode_info *ci, int mds)
 {
@@ -332,9 +332,9 @@
 {
 	struct ceph_cap *cap;
 
-	spin_lock(&ci->vfs_inode.i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	cap = __get_cap_for_mds(ci, mds);
-	spin_unlock(&ci->vfs_inode.i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return cap;
 }
 
@@ -361,15 +361,16 @@
 
 int ceph_get_cap_mds(struct inode *inode)
 {
+	struct ceph_inode_info *ci = ceph_inode(inode);
 	int mds;
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	mds = __ceph_get_cap_mds(ceph_inode(inode));
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return mds;
 }
 
 /*
- * Called under i_lock.
+ * Called under i_ceph_lock.
  */
 static void __insert_cap_node(struct ceph_inode_info *ci,
 			      struct ceph_cap *new)
@@ -415,7 +416,7 @@
  *
  * If I_FLUSH is set, leave the inode at the front of the list.
  *
- * Caller holds i_lock
+ * Caller holds i_ceph_lock
  *    -> we take mdsc->cap_delay_lock
  */
 static void __cap_delay_requeue(struct ceph_mds_client *mdsc,
@@ -457,7 +458,7 @@
 /*
  * Cancel delayed work on cap.
  *
- * Caller must hold i_lock.
+ * Caller must hold i_ceph_lock.
  */
 static void __cap_delay_cancel(struct ceph_mds_client *mdsc,
 			       struct ceph_inode_info *ci)
@@ -532,14 +533,14 @@
 		wanted |= ceph_caps_for_mode(fmode);
 
 retry:
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	cap = __get_cap_for_mds(ci, mds);
 	if (!cap) {
 		if (new_cap) {
 			cap = new_cap;
 			new_cap = NULL;
 		} else {
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 			new_cap = get_cap(mdsc, caps_reservation);
 			if (new_cap == NULL)
 				return -ENOMEM;
@@ -625,7 +626,7 @@
 
 	if (fmode >= 0)
 		__ceph_get_fmode(ci, fmode);
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	wake_up_all(&ci->i_cap_wq);
 	return 0;
 }
@@ -792,7 +793,7 @@
 	struct rb_node *p;
 	int ret = 0;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	for (p = rb_first(&ci->i_caps); p; p = rb_next(p)) {
 		cap = rb_entry(p, struct ceph_cap, ci_node);
 		if (__cap_is_valid(cap) &&
@@ -801,7 +802,7 @@
 			break;
 		}
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	dout("ceph_caps_revoking %p %s = %d\n", inode,
 	     ceph_cap_string(mask), ret);
 	return ret;
@@ -855,7 +856,7 @@
 }
 
 /*
- * called under i_lock
+ * called under i_ceph_lock
  */
 static int __ceph_is_any_caps(struct ceph_inode_info *ci)
 {
@@ -865,7 +866,7 @@
 /*
  * Remove a cap.  Take steps to deal with a racing iterate_session_caps.
  *
- * caller should hold i_lock.
+ * caller should hold i_ceph_lock.
  * caller will not hold session s_mutex if called from destroy_inode.
  */
 void __ceph_remove_cap(struct ceph_cap *cap)
@@ -1028,7 +1029,7 @@
 
 /*
  * Queue cap releases when an inode is dropped from our cache.  Since
- * inode is about to be destroyed, there is no need for i_lock.
+ * inode is about to be destroyed, there is no need for i_ceph_lock.
  */
 void ceph_queue_caps_release(struct inode *inode)
 {
@@ -1049,7 +1050,7 @@
 
 /*
  * Send a cap msg on the given inode.  Update our caps state, then
- * drop i_lock and send the message.
+ * drop i_ceph_lock and send the message.
  *
  * Make note of max_size reported/requested from mds, revoked caps
  * that have now been implemented.
@@ -1061,13 +1062,13 @@
  * Return non-zero if delayed release, or we experienced an error
  * such that the caller should requeue + retry later.
  *
- * called with i_lock, then drops it.
+ * called with i_ceph_lock, then drops it.
  * caller should hold snap_rwsem (read), s_mutex.
  */
 static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
 		      int op, int used, int want, int retain, int flushing,
 		      unsigned *pflush_tid)
-	__releases(cap->ci->vfs_inode->i_lock)
+	__releases(cap->ci->i_ceph_lock)
 {
 	struct ceph_inode_info *ci = cap->ci;
 	struct inode *inode = &ci->vfs_inode;
@@ -1170,7 +1171,7 @@
 		xattr_version = ci->i_xattrs.version;
 	}
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	ret = send_cap_msg(session, ceph_vino(inode).ino, cap_id,
 		op, keep, want, flushing, seq, flush_tid, issue_seq, mseq,
@@ -1198,13 +1199,13 @@
  * Unless @again is true, skip cap_snaps that were already sent to
  * the MDS (i.e., during this session).
  *
- * Called under i_lock.  Takes s_mutex as needed.
+ * Called under i_ceph_lock.  Takes s_mutex as needed.
  */
 void __ceph_flush_snaps(struct ceph_inode_info *ci,
 			struct ceph_mds_session **psession,
 			int again)
-		__releases(ci->vfs_inode->i_lock)
-		__acquires(ci->vfs_inode->i_lock)
+		__releases(ci->i_ceph_lock)
+		__acquires(ci->i_ceph_lock)
 {
 	struct inode *inode = &ci->vfs_inode;
 	int mds;
@@ -1261,7 +1262,7 @@
 			session = NULL;
 		}
 		if (!session) {
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 			mutex_lock(&mdsc->mutex);
 			session = __ceph_lookup_mds_session(mdsc, mds);
 			mutex_unlock(&mdsc->mutex);
@@ -1275,7 +1276,7 @@
 			 * deletion or migration.  retry, and we'll
 			 * get a better @mds value next time.
 			 */
-			spin_lock(&inode->i_lock);
+			spin_lock(&ci->i_ceph_lock);
 			goto retry;
 		}
 
@@ -1285,7 +1286,7 @@
 			list_del_init(&capsnap->flushing_item);
 		list_add_tail(&capsnap->flushing_item,
 			      &session->s_cap_snaps_flushing);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 
 		dout("flush_snaps %p cap_snap %p follows %lld tid %llu\n",
 		     inode, capsnap, capsnap->follows, capsnap->flush_tid);
@@ -1302,7 +1303,7 @@
 		next_follows = capsnap->follows + 1;
 		ceph_put_cap_snap(capsnap);
 
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		goto retry;
 	}
 
@@ -1322,11 +1323,9 @@
 
 static void ceph_flush_snaps(struct ceph_inode_info *ci)
 {
-	struct inode *inode = &ci->vfs_inode;
-
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	__ceph_flush_snaps(ci, NULL, 0);
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 }
 
 /*
@@ -1373,7 +1372,7 @@
  * Add dirty inode to the flushing list.  Assigned a seq number so we
  * can wait for caps to flush without starving.
  *
- * Called under i_lock.
+ * Called under i_ceph_lock.
  */
 static int __mark_caps_flushing(struct inode *inode,
 				 struct ceph_mds_session *session)
@@ -1421,9 +1420,9 @@
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	u32 invalidating_gen = ci->i_rdcache_gen;
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	invalidate_mapping_pages(&inode->i_data, 0, -1);
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 
 	if (inode->i_data.nrpages == 0 &&
 	    invalidating_gen == ci->i_rdcache_gen) {
@@ -1470,7 +1469,7 @@
 	if (mdsc->stopping)
 		is_delayed = 1;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 
 	if (ci->i_ceph_flags & CEPH_I_FLUSH)
 		flags |= CHECK_CAPS_FLUSH;
@@ -1480,7 +1479,7 @@
 		__ceph_flush_snaps(ci, &session, 0);
 	goto retry_locked;
 retry:
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 retry_locked:
 	file_wanted = __ceph_caps_file_wanted(ci);
 	used = __ceph_caps_used(ci);
@@ -1634,7 +1633,7 @@
 			if (mutex_trylock(&session->s_mutex) == 0) {
 				dout("inverting session/ino locks on %p\n",
 				     session);
-				spin_unlock(&inode->i_lock);
+				spin_unlock(&ci->i_ceph_lock);
 				if (took_snap_rwsem) {
 					up_read(&mdsc->snap_rwsem);
 					took_snap_rwsem = 0;
@@ -1648,7 +1647,7 @@
 			if (down_read_trylock(&mdsc->snap_rwsem) == 0) {
 				dout("inverting snap/in locks on %p\n",
 				     inode);
-				spin_unlock(&inode->i_lock);
+				spin_unlock(&ci->i_ceph_lock);
 				down_read(&mdsc->snap_rwsem);
 				took_snap_rwsem = 1;
 				goto retry;
@@ -1664,10 +1663,10 @@
 		mds = cap->mds;  /* remember mds, so we don't repeat */
 		sent++;
 
-		/* __send_cap drops i_lock */
+		/* __send_cap drops i_ceph_lock */
 		delayed += __send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, used, want,
 				      retain, flushing, NULL);
-		goto retry; /* retake i_lock and restart our cap scan. */
+		goto retry; /* retake i_ceph_lock and restart our cap scan. */
 	}
 
 	/*
@@ -1681,7 +1680,7 @@
 	else if (!is_delayed || force_requeue)
 		__cap_delay_requeue(mdsc, ci);
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (queue_invalidate)
 		ceph_queue_invalidate(inode);
@@ -1704,7 +1703,7 @@
 	int flushing = 0;
 
 retry:
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (ci->i_ceph_flags & CEPH_I_NOFLUSH) {
 		dout("try_flush_caps skipping %p I_NOFLUSH set\n", inode);
 		goto out;
@@ -1716,7 +1715,7 @@
 		int delayed;
 
 		if (!session) {
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 			session = cap->session;
 			mutex_lock(&session->s_mutex);
 			goto retry;
@@ -1727,18 +1726,18 @@
 
 		flushing = __mark_caps_flushing(inode, session);
 
-		/* __send_cap drops i_lock */
+		/* __send_cap drops i_ceph_lock */
 		delayed = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH, used, want,
 				     cap->issued | cap->implemented, flushing,
 				     flush_tid);
 		if (!delayed)
 			goto out_unlocked;
 
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		__cap_delay_requeue(mdsc, ci);
 	}
 out:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 out_unlocked:
 	if (session && unlock_session)
 		mutex_unlock(&session->s_mutex);
@@ -1753,7 +1752,7 @@
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int i, ret = 1;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	for (i = 0; i < CEPH_CAP_BITS; i++)
 		if ((ci->i_flushing_caps & (1 << i)) &&
 		    ci->i_cap_flush_tid[i] <= tid) {
@@ -1761,7 +1760,7 @@
 			ret = 0;
 			break;
 		}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return ret;
 }
 
@@ -1868,10 +1867,10 @@
 		struct ceph_mds_client *mdsc =
 			ceph_sb_to_client(inode->i_sb)->mdsc;
 
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		if (__ceph_caps_dirty(ci))
 			__cap_delay_requeue_front(mdsc, ci);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 	}
 	return err;
 }
@@ -1894,7 +1893,7 @@
 		struct inode *inode = &ci->vfs_inode;
 		struct ceph_cap *cap;
 
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		cap = ci->i_auth_cap;
 		if (cap && cap->session == session) {
 			dout("kick_flushing_caps %p cap %p capsnap %p\n", inode,
@@ -1904,7 +1903,7 @@
 			pr_err("%p auth cap %p not mds%d ???\n", inode,
 			       cap, session->s_mds);
 		}
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 	}
 }
 
@@ -1921,7 +1920,7 @@
 		struct ceph_cap *cap;
 		int delayed = 0;
 
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		cap = ci->i_auth_cap;
 		if (cap && cap->session == session) {
 			dout("kick_flushing_caps %p cap %p %s\n", inode,
@@ -1932,14 +1931,14 @@
 					     cap->issued | cap->implemented,
 					     ci->i_flushing_caps, NULL);
 			if (delayed) {
-				spin_lock(&inode->i_lock);
+				spin_lock(&ci->i_ceph_lock);
 				__cap_delay_requeue(mdsc, ci);
-				spin_unlock(&inode->i_lock);
+				spin_unlock(&ci->i_ceph_lock);
 			}
 		} else {
 			pr_err("%p auth cap %p not mds%d ???\n", inode,
 			       cap, session->s_mds);
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 		}
 	}
 }
@@ -1952,7 +1951,7 @@
 	struct ceph_cap *cap;
 	int delayed = 0;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	cap = ci->i_auth_cap;
 	dout("kick_flushing_inode_caps %p flushing %s flush_seq %lld\n", inode,
 	     ceph_cap_string(ci->i_flushing_caps), ci->i_cap_flush_seq);
@@ -1964,12 +1963,12 @@
 				     cap->issued | cap->implemented,
 				     ci->i_flushing_caps, NULL);
 		if (delayed) {
-			spin_lock(&inode->i_lock);
+			spin_lock(&ci->i_ceph_lock);
 			__cap_delay_requeue(mdsc, ci);
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 		}
 	} else {
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 	}
 }
 
@@ -1978,7 +1977,7 @@
  * Take references to capabilities we hold, so that we don't release
  * them to the MDS prematurely.
  *
- * Protected by i_lock.
+ * Protected by i_ceph_lock.
  */
 static void __take_cap_refs(struct ceph_inode_info *ci, int got)
 {
@@ -2016,7 +2015,7 @@
 
 	dout("get_cap_refs %p need %s want %s\n", inode,
 	     ceph_cap_string(need), ceph_cap_string(want));
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 
 	/* make sure file is actually open */
 	file_wanted = __ceph_caps_file_wanted(ci);
@@ -2077,7 +2076,7 @@
 		     ceph_cap_string(have), ceph_cap_string(need));
 	}
 out:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	dout("get_cap_refs %p ret %d got %s\n", inode,
 	     ret, ceph_cap_string(*got));
 	return ret;
@@ -2094,7 +2093,7 @@
 	int check = 0;
 
 	/* do we need to explicitly request a larger max_size? */
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if ((endoff >= ci->i_max_size ||
 	     endoff > (inode->i_size << 1)) &&
 	    endoff > ci->i_wanted_max_size) {
@@ -2103,7 +2102,7 @@
 		ci->i_wanted_max_size = endoff;
 		check = 1;
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	if (check)
 		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
 }
@@ -2140,9 +2139,9 @@
  */
 void ceph_get_cap_refs(struct ceph_inode_info *ci, int caps)
 {
-	spin_lock(&ci->vfs_inode.i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	__take_cap_refs(ci, caps);
-	spin_unlock(&ci->vfs_inode.i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 }
 
 /*
@@ -2160,7 +2159,7 @@
 	int last = 0, put = 0, flushsnaps = 0, wake = 0;
 	struct ceph_cap_snap *capsnap;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (had & CEPH_CAP_PIN)
 		--ci->i_pin_ref;
 	if (had & CEPH_CAP_FILE_RD)
@@ -2193,7 +2192,7 @@
 				}
 			}
 		}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had),
 	     last ? " last" : "", put ? " put" : "");
@@ -2225,7 +2224,7 @@
 	int found = 0;
 	struct ceph_cap_snap *capsnap = NULL;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	ci->i_wrbuffer_ref -= nr;
 	last = !ci->i_wrbuffer_ref;
 
@@ -2274,7 +2273,7 @@
 		}
 	}
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (last) {
 		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
@@ -2291,7 +2290,7 @@
  * Handle a cap GRANT message from the MDS.  (Note that a GRANT may
  * actually be a revocation if it specifies a smaller cap set.)
  *
- * caller holds s_mutex and i_lock, we drop both.
+ * caller holds s_mutex and i_ceph_lock, we drop both.
  *
  * return value:
  *  0 - ok
@@ -2302,7 +2301,7 @@
 			     struct ceph_mds_session *session,
 			     struct ceph_cap *cap,
 			     struct ceph_buffer *xattr_buf)
-		__releases(inode->i_lock)
+		__releases(ci->i_ceph_lock)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int mds = session->s_mds;
@@ -2453,7 +2452,7 @@
 	}
 	BUG_ON(cap->issued & ~cap->implemented);
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	if (writeback)
 		/*
 		 * queue inode for writeback: we can't actually call
@@ -2483,7 +2482,7 @@
 				 struct ceph_mds_caps *m,
 				 struct ceph_mds_session *session,
 				 struct ceph_cap *cap)
-	__releases(inode->i_lock)
+	__releases(ci->i_ceph_lock)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
@@ -2539,7 +2538,7 @@
 	wake_up_all(&ci->i_cap_wq);
 
 out:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	if (drop)
 		iput(inode);
 }
@@ -2562,7 +2561,7 @@
 	dout("handle_cap_flushsnap_ack inode %p ci %p mds%d follows %lld\n",
 	     inode, ci, session->s_mds, follows);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
 		if (capsnap->follows == follows) {
 			if (capsnap->flush_tid != flush_tid) {
@@ -2585,7 +2584,7 @@
 			     capsnap, capsnap->follows);
 		}
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	if (drop)
 		iput(inode);
 }
@@ -2598,7 +2597,7 @@
 static void handle_cap_trunc(struct inode *inode,
 			     struct ceph_mds_caps *trunc,
 			     struct ceph_mds_session *session)
-	__releases(inode->i_lock)
+	__releases(ci->i_ceph_lock)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int mds = session->s_mds;
@@ -2617,7 +2616,7 @@
 	     inode, mds, seq, truncate_size, truncate_seq);
 	queue_trunc = ceph_fill_file_size(inode, issued,
 					  truncate_seq, truncate_size, size);
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (queue_trunc)
 		ceph_queue_vmtruncate(inode);
@@ -2646,7 +2645,7 @@
 	dout("handle_cap_export inode %p ci %p mds%d mseq %d\n",
 	     inode, ci, mds, mseq);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 
 	/* make sure we haven't seen a higher mseq */
 	for (p = rb_first(&ci->i_caps); p; p = rb_next(p)) {
@@ -2690,7 +2689,7 @@
 	}
 	/* else, we already released it */
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 }
 
 /*
@@ -2745,9 +2744,9 @@
 	up_read(&mdsc->snap_rwsem);
 
 	/* make sure we re-request max_size, if necessary */
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	ci->i_requested_max_size = 0;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 }
 
 /*
@@ -2762,6 +2761,7 @@
 	struct ceph_mds_client *mdsc = session->s_mdsc;
 	struct super_block *sb = mdsc->fsc->sb;
 	struct inode *inode;
+	struct ceph_inode_info *ci;
 	struct ceph_cap *cap;
 	struct ceph_mds_caps *h;
 	int mds = session->s_mds;
@@ -2815,6 +2815,7 @@
 
 	/* lookup ino */
 	inode = ceph_find_inode(sb, vino);
+	ci = ceph_inode(inode);
 	dout(" op %s ino %llx.%llx inode %p\n", ceph_cap_op_name(op), vino.ino,
 	     vino.snap, inode);
 	if (!inode) {
@@ -2844,16 +2845,16 @@
 	}
 
 	/* the rest require a cap */
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	cap = __get_cap_for_mds(ceph_inode(inode), mds);
 	if (!cap) {
 		dout(" no cap on %p ino %llx.%llx from mds%d\n",
 		     inode, ceph_ino(inode), ceph_snap(inode), mds);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		goto flush_cap_releases;
 	}
 
-	/* note that each of these drops i_lock for us */
+	/* note that each of these drops i_ceph_lock for us */
 	switch (op) {
 	case CEPH_CAP_OP_REVOKE:
 	case CEPH_CAP_OP_GRANT:
@@ -2869,7 +2870,7 @@
 		break;
 
 	default:
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		pr_err("ceph_handle_caps: unknown cap op %d %s\n", op,
 		       ceph_cap_op_name(op));
 	}
@@ -2962,13 +2963,13 @@
 	struct inode *inode = &ci->vfs_inode;
 	int last = 0;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	dout("put_fmode %p fmode %d %d -> %d\n", inode, fmode,
 	     ci->i_nr_by_mode[fmode], ci->i_nr_by_mode[fmode]-1);
 	BUG_ON(ci->i_nr_by_mode[fmode] == 0);
 	if (--ci->i_nr_by_mode[fmode] == 0)
 		last++;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (last && ci->i_vino.snap == CEPH_NOSNAP)
 		ceph_check_caps(ci, 0, NULL);
@@ -2991,7 +2992,7 @@
 	int used, dirty;
 	int ret = 0;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	used = __ceph_caps_used(ci);
 	dirty = __ceph_caps_dirty(ci);
 
@@ -3046,7 +3047,7 @@
 			     inode, cap, ceph_cap_string(cap->issued));
 		}
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return ret;
 }
 
@@ -3061,7 +3062,7 @@
 
 	/*
 	 * force an record for the directory caps if we have a dentry lease.
-	 * this is racy (can't take i_lock and d_lock together), but it
+	 * this is racy (can't take i_ceph_lock and d_lock together), but it
 	 * doesn't have to be perfect; the mds will revoke anything we don't
 	 * release.
 	 */
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 2abd0df..3eeb976 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -281,18 +281,18 @@
 	}
 
 	/* can we use the dcache? */
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if ((filp->f_pos == 2 || fi->dentry) &&
 	    !ceph_test_mount_opt(fsc, NOASYNCREADDIR) &&
 	    ceph_snap(inode) != CEPH_SNAPDIR &&
 	    ceph_dir_test_complete(inode) &&
 	    __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) {
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		err = __dcache_readdir(filp, dirent, filldir);
 		if (err != -EAGAIN)
 			return err;
 	} else {
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 	}
 	if (fi->dentry) {
 		err = note_last_dentry(fi, fi->dentry->d_name.name,
@@ -428,12 +428,12 @@
 	 * were released during the whole readdir, and we should have
 	 * the complete dir contents in our cache.
 	 */
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (ci->i_release_count == fi->dir_release_count) {
 		ceph_dir_set_complete(inode);
 		ci->i_max_offset = filp->f_pos;
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	dout("readdir %p filp %p done.\n", inode, filp);
 	return 0;
@@ -607,7 +607,7 @@
 		struct ceph_inode_info *ci = ceph_inode(dir);
 		struct ceph_dentry_info *di = ceph_dentry(dentry);
 
-		spin_lock(&dir->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		dout(" dir %p flags are %d\n", dir, ci->i_ceph_flags);
 		if (strncmp(dentry->d_name.name,
 			    fsc->mount_options->snapdir_name,
@@ -615,13 +615,13 @@
 		    !is_root_ceph_dentry(dir, dentry) &&
 		    ceph_dir_test_complete(dir) &&
 		    (__ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1))) {
-			spin_unlock(&dir->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 			dout(" dir %p complete, -ENOENT\n", dir);
 			d_add(dentry, NULL);
 			di->lease_shared_gen = ci->i_shared_gen;
 			return NULL;
 		}
-		spin_unlock(&dir->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 	}
 
 	op = ceph_snap(dir) == CEPH_SNAPDIR ?
@@ -841,12 +841,12 @@
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int drop = CEPH_CAP_LINK_SHARED | CEPH_CAP_LINK_EXCL;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (inode->i_nlink == 1) {
 		drop |= ~(__ceph_caps_wanted(ci) | CEPH_CAP_PIN);
 		ci->i_ceph_flags |= CEPH_I_NODELAY;
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return drop;
 }
 
@@ -1015,10 +1015,10 @@
 	struct ceph_dentry_info *di = ceph_dentry(dentry);
 	int valid = 0;
 
-	spin_lock(&dir->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (ci->i_shared_gen == di->lease_shared_gen)
 		valid = __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1);
-	spin_unlock(&dir->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	dout("dir_lease_is_valid dir %p v%u dentry %p v%u = %d\n",
 	     dir, (unsigned)ci->i_shared_gen, dentry,
 	     (unsigned)di->lease_shared_gen, valid);
@@ -1143,7 +1143,7 @@
 {
 	struct ceph_dentry_info *di;
 
-	dout("d_release %p\n", dentry);
+	dout("ceph_d_prune %p\n", dentry);
 
 	/* do we have a valid parent? */
 	if (!dentry->d_parent || IS_ROOT(dentry))
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index ce549d3..ed72428 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -147,9 +147,9 @@
 
 	/* trivially open snapdir */
 	if (ceph_snap(inode) == CEPH_SNAPDIR) {
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		__ceph_get_fmode(ci, fmode);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		return ceph_init_file(inode, file, fmode);
 	}
 
@@ -158,7 +158,7 @@
 	 * write) or any MDS (for read).  Update wanted set
 	 * asynchronously.
 	 */
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (__ceph_is_any_real_caps(ci) &&
 	    (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) {
 		int mds_wanted = __ceph_caps_mds_wanted(ci);
@@ -168,7 +168,7 @@
 		     inode, fmode, ceph_cap_string(wanted),
 		     ceph_cap_string(issued));
 		__ceph_get_fmode(ci, fmode);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 
 		/* adjust wanted? */
 		if ((issued & wanted) != wanted &&
@@ -180,10 +180,10 @@
 	} else if (ceph_snap(inode) != CEPH_NOSNAP &&
 		   (ci->i_snap_caps & wanted) == wanted) {
 		__ceph_get_fmode(ci, fmode);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		return ceph_init_file(inode, file, fmode);
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	dout("open fmode %d wants %s\n", fmode, ceph_cap_string(wanted));
 	req = prepare_open_request(inode->i_sb, flags, 0);
@@ -743,9 +743,9 @@
 		 */
 		int dirty;
 
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		ceph_put_cap_refs(ci, got);
 
 		ret = generic_file_aio_write(iocb, iov, nr_segs, pos);
@@ -764,9 +764,9 @@
 
 	if (ret >= 0) {
 		int dirty;
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		if (dirty)
 			__mark_inode_dirty(inode, dirty);
 	}
@@ -797,7 +797,8 @@
 
 	mutex_lock(&inode->i_mutex);
 	__ceph_do_pending_vmtruncate(inode);
-	if (origin != SEEK_CUR || origin != SEEK_SET) {
+
+	if (origin == SEEK_END || origin == SEEK_DATA || origin == SEEK_HOLE) {
 		ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE);
 		if (ret < 0) {
 			offset = ret;
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index e392bfc..87fb132 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -297,6 +297,8 @@
 
 	dout("alloc_inode %p\n", &ci->vfs_inode);
 
+	spin_lock_init(&ci->i_ceph_lock);
+
 	ci->i_version = 0;
 	ci->i_time_warp_seq = 0;
 	ci->i_ceph_flags = 0;
@@ -583,7 +585,7 @@
 			       iinfo->xattr_len);
 	}
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 
 	/*
 	 * provided version will be odd if inode value is projected,
@@ -680,7 +682,7 @@
 			char *sym;
 
 			BUG_ON(symlen != inode->i_size);
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 
 			err = -ENOMEM;
 			sym = kmalloc(symlen+1, GFP_NOFS);
@@ -689,7 +691,7 @@
 			memcpy(sym, iinfo->symlink, symlen);
 			sym[symlen] = 0;
 
-			spin_lock(&inode->i_lock);
+			spin_lock(&ci->i_ceph_lock);
 			if (!ci->i_symlink)
 				ci->i_symlink = sym;
 			else
@@ -715,7 +717,7 @@
 	}
 
 no_change:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	/* queue truncate if we saw i_size decrease */
 	if (queue_trunc)
@@ -750,13 +752,13 @@
 				     info->cap.flags,
 				     caps_reservation);
 		} else {
-			spin_lock(&inode->i_lock);
+			spin_lock(&ci->i_ceph_lock);
 			dout(" %p got snap_caps %s\n", inode,
 			     ceph_cap_string(le32_to_cpu(info->cap.caps)));
 			ci->i_snap_caps |= le32_to_cpu(info->cap.caps);
 			if (cap_fmode >= 0)
 				__ceph_get_fmode(ci, cap_fmode);
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 		}
 	} else if (cap_fmode >= 0) {
 		pr_warning("mds issued no caps on %llx.%llx\n",
@@ -849,19 +851,20 @@
 {
 	struct dentry *dir = dn->d_parent;
 	struct inode *inode = dir->d_inode;
+	struct ceph_inode_info *ci = ceph_inode(inode);
 	struct ceph_dentry_info *di;
 
 	BUG_ON(!inode);
 
 	di = ceph_dentry(dn);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (!ceph_dir_test_complete(inode)) {
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		return;
 	}
 	di->offset = ceph_inode(inode)->i_max_offset++;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	spin_lock(&dir->d_lock);
 	spin_lock_nested(&dn->d_lock, DENTRY_D_LOCK_NESTED);
@@ -1308,7 +1311,7 @@
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int ret = 0;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	dout("set_size %p %llu -> %llu\n", inode, inode->i_size, size);
 	inode->i_size = size;
 	inode->i_blocks = (size + (1 << 9) - 1) >> 9;
@@ -1318,7 +1321,7 @@
 	    (ci->i_reported_size << 1) < ci->i_max_size)
 		ret = 1;
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return ret;
 }
 
@@ -1328,12 +1331,13 @@
  */
 void ceph_queue_writeback(struct inode *inode)
 {
+	ihold(inode);
 	if (queue_work(ceph_inode_to_client(inode)->wb_wq,
 		       &ceph_inode(inode)->i_wb_work)) {
 		dout("ceph_queue_writeback %p\n", inode);
-		ihold(inode);
 	} else {
 		dout("ceph_queue_writeback %p failed\n", inode);
+		iput(inode);
 	}
 }
 
@@ -1353,12 +1357,13 @@
  */
 void ceph_queue_invalidate(struct inode *inode)
 {
+	ihold(inode);
 	if (queue_work(ceph_inode_to_client(inode)->pg_inv_wq,
 		       &ceph_inode(inode)->i_pg_inv_work)) {
 		dout("ceph_queue_invalidate %p\n", inode);
-		ihold(inode);
 	} else {
 		dout("ceph_queue_invalidate %p failed\n", inode);
+		iput(inode);
 	}
 }
 
@@ -1374,20 +1379,20 @@
 	u32 orig_gen;
 	int check = 0;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	dout("invalidate_pages %p gen %d revoking %d\n", inode,
 	     ci->i_rdcache_gen, ci->i_rdcache_revoking);
 	if (ci->i_rdcache_revoking != ci->i_rdcache_gen) {
 		/* nevermind! */
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		goto out;
 	}
 	orig_gen = ci->i_rdcache_gen;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	truncate_inode_pages(&inode->i_data, 0);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (orig_gen == ci->i_rdcache_gen &&
 	    orig_gen == ci->i_rdcache_revoking) {
 		dout("invalidate_pages %p gen %d successful\n", inode,
@@ -1399,7 +1404,7 @@
 		     inode, orig_gen, ci->i_rdcache_gen,
 		     ci->i_rdcache_revoking);
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (check)
 		ceph_check_caps(ci, 0, NULL);
@@ -1434,13 +1439,14 @@
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 
+	ihold(inode);
 	if (queue_work(ceph_sb_to_client(inode->i_sb)->trunc_wq,
 		       &ci->i_vmtruncate_work)) {
 		dout("ceph_queue_vmtruncate %p\n", inode);
-		ihold(inode);
 	} else {
 		dout("ceph_queue_vmtruncate %p failed, pending=%d\n",
 		     inode, ci->i_truncate_pending);
+		iput(inode);
 	}
 }
 
@@ -1457,10 +1463,10 @@
 	int wrbuffer_refs, wake = 0;
 
 retry:
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	if (ci->i_truncate_pending == 0) {
 		dout("__do_pending_vmtruncate %p none pending\n", inode);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		return;
 	}
 
@@ -1471,7 +1477,7 @@
 	if (ci->i_wrbuffer_ref_head < ci->i_wrbuffer_ref) {
 		dout("__do_pending_vmtruncate %p flushing snaps first\n",
 		     inode);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		filemap_write_and_wait_range(&inode->i_data, 0,
 					     inode->i_sb->s_maxbytes);
 		goto retry;
@@ -1481,15 +1487,15 @@
 	wrbuffer_refs = ci->i_wrbuffer_ref;
 	dout("__do_pending_vmtruncate %p (%d) to %lld\n", inode,
 	     ci->i_truncate_pending, to);
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	truncate_inode_pages(inode->i_mapping, to);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	ci->i_truncate_pending--;
 	if (ci->i_truncate_pending == 0)
 		wake = 1;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (wrbuffer_refs == 0)
 		ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
@@ -1544,7 +1550,7 @@
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	issued = __ceph_caps_issued(ci, NULL);
 	dout("setattr %p issued %s\n", inode, ceph_cap_string(issued));
 
@@ -1692,7 +1698,7 @@
 	}
 
 	release &= issued;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (inode_dirty_flags)
 		__mark_inode_dirty(inode, inode_dirty_flags);
@@ -1714,7 +1720,7 @@
 	__ceph_do_pending_vmtruncate(inode);
 	return err;
 out:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	ceph_mdsc_put_request(req);
 	return err;
 }
diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 5a14c29..790914a59 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -241,11 +241,11 @@
 	struct ceph_inode_info *ci = ceph_inode(inode);
 
 	if ((fi->fmode & CEPH_FILE_MODE_LAZY) == 0) {
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		ci->i_nr_by_mode[fi->fmode]--;
 		fi->fmode |= CEPH_FILE_MODE_LAZY;
 		ci->i_nr_by_mode[fi->fmode]++;
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		dout("ioctl_layzio: file %p marked lazy\n", file);
 
 		ceph_check_caps(ci, 0, NULL);
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 264ab70..6203d80 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -732,21 +732,21 @@
 		}
 	}
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	cap = NULL;
 	if (mode == USE_AUTH_MDS)
 		cap = ci->i_auth_cap;
 	if (!cap && !RB_EMPTY_ROOT(&ci->i_caps))
 		cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node);
 	if (!cap) {
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		goto random;
 	}
 	mds = cap->session->s_mds;
 	dout("choose_mds %p %llx.%llx mds%d (%scap %p)\n",
 	     inode, ceph_vinop(inode), mds,
 	     cap == ci->i_auth_cap ? "auth " : "", cap);
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return mds;
 
 random:
@@ -951,7 +951,7 @@
 
 	dout("removing cap %p, ci is %p, inode is %p\n",
 	     cap, ci, &ci->vfs_inode);
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	__ceph_remove_cap(cap);
 	if (!__ceph_is_any_real_caps(ci)) {
 		struct ceph_mds_client *mdsc =
@@ -984,7 +984,7 @@
 		}
 		spin_unlock(&mdsc->cap_dirty_lock);
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	while (drop--)
 		iput(inode);
 	return 0;
@@ -1015,10 +1015,10 @@
 
 	wake_up_all(&ci->i_cap_wq);
 	if (arg) {
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		ci->i_wanted_max_size = 0;
 		ci->i_requested_max_size = 0;
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 	}
 	return 0;
 }
@@ -1151,7 +1151,7 @@
 	if (session->s_trim_caps <= 0)
 		return -1;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	mine = cap->issued | cap->implemented;
 	used = __ceph_caps_used(ci);
 	oissued = __ceph_caps_issued_other(ci, cap);
@@ -1170,7 +1170,7 @@
 		__ceph_remove_cap(cap);
 	} else {
 		/* try to drop referring dentries */
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		d_prune_aliases(inode);
 		dout("trim_caps_cb %p cap %p  pruned, count now %d\n",
 		     inode, cap, atomic_read(&inode->i_count));
@@ -1178,7 +1178,7 @@
 	}
 
 out:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return 0;
 }
 
@@ -1296,7 +1296,7 @@
 					   i_flushing_item);
 			struct inode *inode = &ci->vfs_inode;
 
-			spin_lock(&inode->i_lock);
+			spin_lock(&ci->i_ceph_lock);
 			if (ci->i_cap_flush_seq <= want_flush_seq) {
 				dout("check_cap_flush still flushing %p "
 				     "seq %lld <= %lld to mds%d\n", inode,
@@ -1304,7 +1304,7 @@
 				     session->s_mds);
 				ret = 0;
 			}
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 		}
 		mutex_unlock(&session->s_mutex);
 		ceph_put_mds_session(session);
@@ -1495,6 +1495,7 @@
 			     pos, temp);
 		} else if (stop_on_nosnap && inode &&
 			   ceph_snap(inode) == CEPH_NOSNAP) {
+			spin_unlock(&temp->d_lock);
 			break;
 		} else {
 			pos -= temp->d_name.len;
@@ -2011,10 +2012,10 @@
 	struct ceph_inode_info *ci = ceph_inode(inode);
 
 	dout("invalidate_dir_request %p (D_COMPLETE, lease(s))\n", inode);
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	ceph_dir_clear_complete(inode);
 	ci->i_release_count++;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (req->r_dentry)
 		ceph_invalidate_dentry_lease(req->r_dentry);
@@ -2422,7 +2423,7 @@
 	if (err)
 		goto out_free;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	cap->seq = 0;        /* reset cap seq */
 	cap->issue_seq = 0;  /* and issue_seq */
 
@@ -2445,7 +2446,7 @@
 		rec.v1.pathbase = cpu_to_le64(pathbase);
 		reclen = sizeof(rec.v1);
 	}
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 
 	if (recon_state->flock) {
 		int num_fcntl_locks, num_flock_locks;
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index 4bb2399..a50ca0e 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -20,7 +20,7 @@
  *
  *         mdsc->snap_rwsem
  *
- *         inode->i_lock
+ *         ci->i_ceph_lock
  *                 mdsc->snap_flush_lock
  *                 mdsc->cap_delay_lock
  *
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index e264371..a559c80 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -446,7 +446,7 @@
 		return;
 	}
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	used = __ceph_caps_used(ci);
 	dirty = __ceph_caps_dirty(ci);
 
@@ -528,7 +528,7 @@
 		kfree(capsnap);
 	}
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 }
 
 /*
@@ -537,7 +537,7 @@
  *
  * If capsnap can now be flushed, add to snap_flush list, and return 1.
  *
- * Caller must hold i_lock.
+ * Caller must hold i_ceph_lock.
  */
 int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
 			    struct ceph_cap_snap *capsnap)
@@ -739,9 +739,9 @@
 		inode = &ci->vfs_inode;
 		ihold(inode);
 		spin_unlock(&mdsc->snap_flush_lock);
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		__ceph_flush_snaps(ci, &session, 0);
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		iput(inode);
 		spin_lock(&mdsc->snap_flush_lock);
 	}
@@ -847,7 +847,7 @@
 				continue;
 			ci = ceph_inode(inode);
 
-			spin_lock(&inode->i_lock);
+			spin_lock(&ci->i_ceph_lock);
 			if (!ci->i_snap_realm)
 				goto skip_inode;
 			/*
@@ -876,7 +876,7 @@
 			oldrealm = ci->i_snap_realm;
 			ci->i_snap_realm = realm;
 			spin_unlock(&realm->inodes_with_caps_lock);
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 
 			ceph_get_snap_realm(mdsc, realm);
 			ceph_put_snap_realm(mdsc, oldrealm);
@@ -885,7 +885,7 @@
 			continue;
 
 skip_inode:
-			spin_unlock(&inode->i_lock);
+			spin_unlock(&ci->i_ceph_lock);
 			iput(inode);
 		}
 
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index a90846f..b48f15f 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -383,7 +383,7 @@
 	if (fsopt->rsize != CEPH_RSIZE_DEFAULT)
 		seq_printf(m, ",rsize=%d", fsopt->rsize);
 	if (fsopt->rasize != CEPH_RASIZE_DEFAULT)
-		seq_printf(m, ",rasize=%d", fsopt->rsize);
+		seq_printf(m, ",rasize=%d", fsopt->rasize);
 	if (fsopt->congestion_kb != default_congestion_kb())
 		seq_printf(m, ",write_congestion_kb=%d", fsopt->congestion_kb);
 	if (fsopt->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT)
@@ -638,10 +638,12 @@
 	if (err == 0) {
 		dout("open_root_inode success\n");
 		if (ceph_ino(req->r_target_inode) == CEPH_INO_ROOT &&
-		    fsc->sb->s_root == NULL)
+		    fsc->sb->s_root == NULL) {
 			root = d_alloc_root(req->r_target_inode);
-		else
+			ceph_init_dentry(root);
+		} else {
 			root = d_obtain_alias(req->r_target_inode);
+		}
 		req->r_target_inode = NULL;
 		dout("open_root_inode success, root dentry is %p\n", root);
 	} else {
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 01bf189..edcbf37 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -220,7 +220,7 @@
  * The locking for D_COMPLETE is a bit odd:
  *  - we can clear it at almost any time (see ceph_d_prune)
  *  - it is only meaningful if:
- *    - we hold dir inode i_lock
+ *    - we hold dir inode i_ceph_lock
  *    - we hold dir FILE_SHARED caps
  *    - the dentry D_COMPLETE is set
  */
@@ -250,6 +250,8 @@
 struct ceph_inode_info {
 	struct ceph_vino i_vino;   /* ceph ino + snap */
 
+	spinlock_t i_ceph_lock;
+
 	u64 i_version;
 	u32 i_time_warp_seq;
 
@@ -271,7 +273,7 @@
 
 	struct ceph_inode_xattrs_info i_xattrs;
 
-	/* capabilities.  protected _both_ by i_lock and cap->session's
+	/* capabilities.  protected _both_ by i_ceph_lock and cap->session's
 	 * s_mutex. */
 	struct rb_root i_caps;           /* cap list */
 	struct ceph_cap *i_auth_cap;     /* authoritative cap, if any */
@@ -437,18 +439,18 @@
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	ci->i_ceph_flags &= ~mask;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 }
 
 static inline void ceph_i_set(struct inode *inode, unsigned mask)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	ci->i_ceph_flags |= mask;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 }
 
 static inline bool ceph_i_test(struct inode *inode, unsigned mask)
@@ -456,9 +458,9 @@
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	bool r;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	r = (ci->i_ceph_flags & mask) == mask;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return r;
 }
 
@@ -508,9 +510,9 @@
 static inline int ceph_caps_issued(struct ceph_inode_info *ci)
 {
 	int issued;
-	spin_lock(&ci->vfs_inode.i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	issued = __ceph_caps_issued(ci, NULL);
-	spin_unlock(&ci->vfs_inode.i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return issued;
 }
 
@@ -518,9 +520,9 @@
 					int touch)
 {
 	int r;
-	spin_lock(&ci->vfs_inode.i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	r = __ceph_caps_issued_mask(ci, mask, touch);
-	spin_unlock(&ci->vfs_inode.i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return r;
 }
 
@@ -743,10 +745,9 @@
 extern void __ceph_remove_cap(struct ceph_cap *cap);
 static inline void ceph_remove_cap(struct ceph_cap *cap)
 {
-	struct inode *inode = &cap->ci->vfs_inode;
-	spin_lock(&inode->i_lock);
+	spin_lock(&cap->ci->i_ceph_lock);
 	__ceph_remove_cap(cap);
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&cap->ci->i_ceph_lock);
 }
 extern void ceph_put_cap(struct ceph_mds_client *mdsc,
 			 struct ceph_cap *cap);
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index 96c6739..a5e36e4 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -343,8 +343,8 @@
 }
 
 static int __build_xattrs(struct inode *inode)
-	__releases(inode->i_lock)
-	__acquires(inode->i_lock)
+	__releases(ci->i_ceph_lock)
+	__acquires(ci->i_ceph_lock)
 {
 	u32 namelen;
 	u32 numattr = 0;
@@ -372,7 +372,7 @@
 		end = p + ci->i_xattrs.blob->vec.iov_len;
 		ceph_decode_32_safe(&p, end, numattr, bad);
 		xattr_version = ci->i_xattrs.version;
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 
 		xattrs = kcalloc(numattr, sizeof(struct ceph_xattr *),
 				 GFP_NOFS);
@@ -387,7 +387,7 @@
 				goto bad_lock;
 		}
 
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		if (ci->i_xattrs.version != xattr_version) {
 			/* lost a race, retry */
 			for (i = 0; i < numattr; i++)
@@ -418,7 +418,7 @@
 
 	return err;
 bad_lock:
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 bad:
 	if (xattrs) {
 		for (i = 0; i < numattr; i++)
@@ -512,7 +512,7 @@
 	if (vxattrs)
 		vxattr = ceph_match_vxattr(vxattrs, name);
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	dout("getxattr %p ver=%lld index_ver=%lld\n", inode,
 	     ci->i_xattrs.version, ci->i_xattrs.index_version);
 
@@ -520,14 +520,14 @@
 	    (ci->i_xattrs.index_version >= ci->i_xattrs.version)) {
 		goto get_xattr;
 	} else {
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		/* get xattrs from mds (if we don't already have them) */
 		err = ceph_do_getattr(inode, CEPH_STAT_CAP_XATTR);
 		if (err)
 			return err;
 	}
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 
 	if (vxattr && vxattr->readonly) {
 		err = vxattr->getxattr_cb(ci, value, size);
@@ -558,7 +558,7 @@
 	memcpy(value, xattr->val, xattr->val_len);
 
 out:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return err;
 }
 
@@ -573,7 +573,7 @@
 	u32 len;
 	int i;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	dout("listxattr %p ver=%lld index_ver=%lld\n", inode,
 	     ci->i_xattrs.version, ci->i_xattrs.index_version);
 
@@ -581,13 +581,13 @@
 	    (ci->i_xattrs.index_version >= ci->i_xattrs.version)) {
 		goto list_xattr;
 	} else {
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		err = ceph_do_getattr(inode, CEPH_STAT_CAP_XATTR);
 		if (err)
 			return err;
 	}
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 
 	err = __build_xattrs(inode);
 	if (err < 0)
@@ -619,7 +619,7 @@
 		}
 
 out:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	return err;
 }
 
@@ -739,7 +739,7 @@
 	if (!xattr)
 		goto out;
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 retry:
 	issued = __ceph_caps_issued(ci, NULL);
 	if (!(issued & CEPH_CAP_XATTR_EXCL))
@@ -752,12 +752,12 @@
 	    required_blob_size > ci->i_xattrs.prealloc_blob->alloc_len) {
 		struct ceph_buffer *blob = NULL;
 
-		spin_unlock(&inode->i_lock);
+		spin_unlock(&ci->i_ceph_lock);
 		dout(" preaallocating new blob size=%d\n", required_blob_size);
 		blob = ceph_buffer_new(required_blob_size, GFP_NOFS);
 		if (!blob)
 			goto out;
-		spin_lock(&inode->i_lock);
+		spin_lock(&ci->i_ceph_lock);
 		if (ci->i_xattrs.prealloc_blob)
 			ceph_buffer_put(ci->i_xattrs.prealloc_blob);
 		ci->i_xattrs.prealloc_blob = blob;
@@ -770,13 +770,13 @@
 	dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
 	ci->i_xattrs.dirty = true;
 	inode->i_ctime = CURRENT_TIME;
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	if (dirty)
 		__mark_inode_dirty(inode, dirty);
 	return err;
 
 do_sync:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	err = ceph_sync_setxattr(dentry, name, value, size, flags);
 out:
 	kfree(newname);
@@ -833,7 +833,7 @@
 			return -EOPNOTSUPP;
 	}
 
-	spin_lock(&inode->i_lock);
+	spin_lock(&ci->i_ceph_lock);
 	__build_xattrs(inode);
 	issued = __ceph_caps_issued(ci, NULL);
 	dout("removexattr %p issued %s\n", inode, ceph_cap_string(issued));
@@ -846,12 +846,12 @@
 	ci->i_xattrs.dirty = true;
 	inode->i_ctime = CURRENT_TIME;
 
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	if (dirty)
 		__mark_inode_dirty(inode, dirty);
 	return err;
 do_sync:
-	spin_unlock(&inode->i_lock);
+	spin_unlock(&ci->i_ceph_lock);
 	err = ceph_send_removexattr(dentry, name);
 	return err;
 }
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index d6a972d..8cd4b52 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -441,6 +441,8 @@
 	smb_msg.msg_controllen = 0;
 
 	for (total_read = 0; to_read; total_read += length, to_read -= length) {
+		try_to_freeze();
+
 		if (server_unresponsive(server)) {
 			total_read = -EAGAIN;
 			break;
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index cf0b153..4dd9283 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -702,6 +702,13 @@
 					 lock->type, lock->netfid, conf_lock);
 }
 
+/*
+ * Check if there is another lock that prevents us to set the lock (mandatory
+ * style). If such a lock exists, update the flock structure with its
+ * properties. Otherwise, set the flock type to F_UNLCK if we can cache brlocks
+ * or leave it the same if we can't. Returns 0 if we don't need to request to
+ * the server or 1 otherwise.
+ */
 static int
 cifs_lock_test(struct cifsInodeInfo *cinode, __u64 offset, __u64 length,
 	       __u8 type, __u16 netfid, struct file_lock *flock)
@@ -739,6 +746,12 @@
 	mutex_unlock(&cinode->lock_mutex);
 }
 
+/*
+ * Set the byte-range lock (mandatory style). Returns:
+ * 1) 0, if we set the lock and don't need to request to the server;
+ * 2) 1, if no locks prevent us but we need to request to the server;
+ * 3) -EACCESS, if there is a lock that prevents us and wait is false.
+ */
 static int
 cifs_lock_add_if(struct cifsInodeInfo *cinode, struct cifsLockInfo *lock,
 		 bool wait)
@@ -778,6 +791,13 @@
 	return rc;
 }
 
+/*
+ * Check if there is another lock that prevents us to set the lock (posix
+ * style). If such a lock exists, update the flock structure with its
+ * properties. Otherwise, set the flock type to F_UNLCK if we can cache brlocks
+ * or leave it the same if we can't. Returns 0 if we don't need to request to
+ * the server or 1 otherwise.
+ */
 static int
 cifs_posix_lock_test(struct file *file, struct file_lock *flock)
 {
@@ -800,6 +820,12 @@
 	return rc;
 }
 
+/*
+ * Set the byte-range lock (posix style). Returns:
+ * 1) 0, if we set the lock and don't need to request to the server;
+ * 2) 1, if we need to request to the server;
+ * 3) <0, if the error occurs while setting the lock.
+ */
 static int
 cifs_posix_lock_set(struct file *file, struct file_lock *flock)
 {
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 5de03ec..a090bbe 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -554,7 +554,10 @@
 				 rc);
 			return rc;
 		}
-		cifs_save_resume_key(cifsFile->srch_inf.last_entry, cifsFile);
+		/* FindFirst/Next set last_entry to NULL on malformed reply */
+		if (cifsFile->srch_inf.last_entry)
+			cifs_save_resume_key(cifsFile->srch_inf.last_entry,
+						cifsFile);
 	}
 
 	while ((index_to_find >= cifsFile->srch_inf.index_of_last_entry) &&
@@ -562,7 +565,10 @@
 		cFYI(1, "calling findnext2");
 		rc = CIFSFindNext(xid, pTcon, cifsFile->netfid,
 				  &cifsFile->srch_inf);
-		cifs_save_resume_key(cifsFile->srch_inf.last_entry, cifsFile);
+		/* FindFirst/Next set last_entry to NULL on malformed reply */
+		if (cifsFile->srch_inf.last_entry)
+			cifs_save_resume_key(cifsFile->srch_inf.last_entry,
+						cifsFile);
 		if (rc)
 			return -ENOENT;
 	}
diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c
index 7cacba1..80d8508 100644
--- a/fs/cifs/smbencrypt.c
+++ b/fs/cifs/smbencrypt.c
@@ -209,7 +209,7 @@
 {
 	int rc;
 	int len;
-	__u16 wpwd[129];
+	__le16 wpwd[129];
 
 	/* Password cannot be longer than 128 characters */
 	if (passwd) /* Password must be converted to NT unicode */
@@ -219,8 +219,8 @@
 		*wpwd = 0; /* Ensure string is null terminated */
 	}
 
-	rc = mdfour(p16, (unsigned char *) wpwd, len * sizeof(__u16));
-	memset(wpwd, 0, 129 * sizeof(__u16));
+	rc = mdfour(p16, (unsigned char *) wpwd, len * sizeof(__le16));
+	memset(wpwd, 0, 129 * sizeof(__le16));
 
 	return rc;
 }
diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c
index ca418aa..9d8715c 100644
--- a/fs/configfs/inode.c
+++ b/fs/configfs/inode.c
@@ -292,7 +292,7 @@
 	return bdi_init(&configfs_backing_dev_info);
 }
 
-void __exit configfs_inode_exit(void)
+void configfs_inode_exit(void)
 {
 	bdi_destroy(&configfs_backing_dev_info);
 }
diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c
index ecc6217..276e15c 100644
--- a/fs/configfs/mount.c
+++ b/fs/configfs/mount.c
@@ -143,28 +143,26 @@
 		goto out;
 
 	config_kobj = kobject_create_and_add("config", kernel_kobj);
-	if (!config_kobj) {
-		kmem_cache_destroy(configfs_dir_cachep);
-		configfs_dir_cachep = NULL;
-		goto out;
-	}
-
-	err = register_filesystem(&configfs_fs_type);
-	if (err) {
-		printk(KERN_ERR "configfs: Unable to register filesystem!\n");
-		kobject_put(config_kobj);
-		kmem_cache_destroy(configfs_dir_cachep);
-		configfs_dir_cachep = NULL;
-		goto out;
-	}
+	if (!config_kobj)
+		goto out2;
 
 	err = configfs_inode_init();
-	if (err) {
-		unregister_filesystem(&configfs_fs_type);
-		kobject_put(config_kobj);
-		kmem_cache_destroy(configfs_dir_cachep);
-		configfs_dir_cachep = NULL;
-	}
+	if (err)
+		goto out3;
+
+	err = register_filesystem(&configfs_fs_type);
+	if (err)
+		goto out4;
+
+	return 0;
+out4:
+	printk(KERN_ERR "configfs: Unable to register filesystem!\n");
+	configfs_inode_exit();
+out3:
+	kobject_put(config_kobj);
+out2:
+	kmem_cache_destroy(configfs_dir_cachep);
+	configfs_dir_cachep = NULL;
 out:
 	return err;
 }
diff --git a/fs/dcache.c b/fs/dcache.c
index a901c69..89509b5 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -36,6 +36,7 @@
 #include <linux/bit_spinlock.h>
 #include <linux/rculist_bl.h>
 #include <linux/prefetch.h>
+#include <linux/ratelimit.h>
 #include "internal.h"
 
 /*
@@ -2383,8 +2384,16 @@
 				actual = __d_unalias(inode, dentry, alias);
 			}
 			write_sequnlock(&rename_lock);
-			if (IS_ERR(actual))
+			if (IS_ERR(actual)) {
+				if (PTR_ERR(actual) == -ELOOP)
+					pr_warn_ratelimited(
+						"VFS: Lookup of '%s' in %s %s"
+						" would have caused loop\n",
+						dentry->d_name.name,
+						inode->i_sb->s_type->name,
+						inode->i_sb->s_id);
 				dput(alias);
+			}
 			goto out_nolock;
 		}
 	}
@@ -2430,16 +2439,14 @@
 /**
  * prepend_path - Prepend path string to a buffer
  * @path: the dentry/vfsmount to report
- * @root: root vfsmnt/dentry (may be modified by this function)
+ * @root: root vfsmnt/dentry
  * @buffer: pointer to the end of the buffer
  * @buflen: pointer to buffer length
  *
  * Caller holds the rename_lock.
- *
- * If path is not reachable from the supplied root, then the value of
- * root is changed (without modifying refcounts).
  */
-static int prepend_path(const struct path *path, struct path *root,
+static int prepend_path(const struct path *path,
+			const struct path *root,
 			char **buffer, int *buflen)
 {
 	struct dentry *dentry = path->dentry;
@@ -2474,10 +2481,10 @@
 		dentry = parent;
 	}
 
-out:
 	if (!error && !slash)
 		error = prepend(buffer, buflen, "/", 1);
 
+out:
 	br_read_unlock(vfsmount_lock);
 	return error;
 
@@ -2491,15 +2498,17 @@
 		WARN(1, "Root dentry has weird name <%.*s>\n",
 		     (int) dentry->d_name.len, dentry->d_name.name);
 	}
-	root->mnt = vfsmnt;
-	root->dentry = dentry;
+	if (!slash)
+		error = prepend(buffer, buflen, "/", 1);
+	if (!error)
+		error = vfsmnt->mnt_ns ? 1 : 2;
 	goto out;
 }
 
 /**
  * __d_path - return the path of a dentry
  * @path: the dentry/vfsmount to report
- * @root: root vfsmnt/dentry (may be modified by this function)
+ * @root: root vfsmnt/dentry
  * @buf: buffer to return value in
  * @buflen: buffer length
  *
@@ -2510,10 +2519,10 @@
  *
  * "buflen" should be positive.
  *
- * If path is not reachable from the supplied root, then the value of
- * root is changed (without modifying refcounts).
+ * If the path is not reachable from the supplied root, return %NULL.
  */
-char *__d_path(const struct path *path, struct path *root,
+char *__d_path(const struct path *path,
+	       const struct path *root,
 	       char *buf, int buflen)
 {
 	char *res = buf + buflen;
@@ -2524,7 +2533,28 @@
 	error = prepend_path(path, root, &res, &buflen);
 	write_sequnlock(&rename_lock);
 
-	if (error)
+	if (error < 0)
+		return ERR_PTR(error);
+	if (error > 0)
+		return NULL;
+	return res;
+}
+
+char *d_absolute_path(const struct path *path,
+	       char *buf, int buflen)
+{
+	struct path root = {};
+	char *res = buf + buflen;
+	int error;
+
+	prepend(&res, &buflen, "\0", 1);
+	write_seqlock(&rename_lock);
+	error = prepend_path(path, &root, &res, &buflen);
+	write_sequnlock(&rename_lock);
+
+	if (error > 1)
+		error = -EINVAL;
+	if (error < 0)
 		return ERR_PTR(error);
 	return res;
 }
@@ -2532,8 +2562,9 @@
 /*
  * same as __d_path but appends "(deleted)" for unlinked files.
  */
-static int path_with_deleted(const struct path *path, struct path *root,
-				 char **buf, int *buflen)
+static int path_with_deleted(const struct path *path,
+			     const struct path *root,
+			     char **buf, int *buflen)
 {
 	prepend(buf, buflen, "\0", 1);
 	if (d_unlinked(path->dentry)) {
@@ -2570,7 +2601,6 @@
 {
 	char *res = buf + buflen;
 	struct path root;
-	struct path tmp;
 	int error;
 
 	/*
@@ -2585,9 +2615,8 @@
 
 	get_fs_root(current->fs, &root);
 	write_seqlock(&rename_lock);
-	tmp = root;
-	error = path_with_deleted(path, &tmp, &res, &buflen);
-	if (error)
+	error = path_with_deleted(path, &root, &res, &buflen);
+	if (error < 0)
 		res = ERR_PTR(error);
 	write_sequnlock(&rename_lock);
 	path_put(&root);
@@ -2608,7 +2637,6 @@
 {
 	char *res = buf + buflen;
 	struct path root;
-	struct path tmp;
 	int error;
 
 	if (path->dentry->d_op && path->dentry->d_op->d_dname)
@@ -2616,9 +2644,8 @@
 
 	get_fs_root(current->fs, &root);
 	write_seqlock(&rename_lock);
-	tmp = root;
-	error = path_with_deleted(path, &tmp, &res, &buflen);
-	if (!error && !path_equal(&tmp, &root))
+	error = path_with_deleted(path, &root, &res, &buflen);
+	if (error > 0)
 		error = prepend_unreachable(&res, &buflen);
 	write_sequnlock(&rename_lock);
 	path_put(&root);
@@ -2749,19 +2776,18 @@
 	write_seqlock(&rename_lock);
 	if (!d_unlinked(pwd.dentry)) {
 		unsigned long len;
-		struct path tmp = root;
 		char *cwd = page + PAGE_SIZE;
 		int buflen = PAGE_SIZE;
 
 		prepend(&cwd, &buflen, "\0", 1);
-		error = prepend_path(&pwd, &tmp, &cwd, &buflen);
+		error = prepend_path(&pwd, &root, &cwd, &buflen);
 		write_sequnlock(&rename_lock);
 
-		if (error)
+		if (error < 0)
 			goto out;
 
 		/* Unreachable from current root */
-		if (!path_equal(&tmp, &root)) {
+		if (error > 0) {
 			error = prepend_unreachable(&cwd, &buflen);
 			if (error)
 				goto out;
diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index 58609bd..2a83425 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -967,7 +967,7 @@
 
 /**
  * ecryptfs_new_file_context
- * @ecryptfs_dentry: The eCryptfs dentry
+ * @ecryptfs_inode: The eCryptfs inode
  *
  * If the crypto context for the file has not yet been established,
  * this is where we do that.  Establishing a new crypto context
@@ -984,13 +984,13 @@
  *
  * Returns zero on success; non-zero otherwise
  */
-int ecryptfs_new_file_context(struct dentry *ecryptfs_dentry)
+int ecryptfs_new_file_context(struct inode *ecryptfs_inode)
 {
 	struct ecryptfs_crypt_stat *crypt_stat =
-	    &ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat;
+	    &ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat;
 	struct ecryptfs_mount_crypt_stat *mount_crypt_stat =
 	    &ecryptfs_superblock_to_private(
-		    ecryptfs_dentry->d_sb)->mount_crypt_stat;
+		    ecryptfs_inode->i_sb)->mount_crypt_stat;
 	int cipher_name_len;
 	int rc = 0;
 
@@ -1299,12 +1299,12 @@
 }
 
 static int
-ecryptfs_write_metadata_to_contents(struct dentry *ecryptfs_dentry,
+ecryptfs_write_metadata_to_contents(struct inode *ecryptfs_inode,
 				    char *virt, size_t virt_len)
 {
 	int rc;
 
-	rc = ecryptfs_write_lower(ecryptfs_dentry->d_inode, virt,
+	rc = ecryptfs_write_lower(ecryptfs_inode, virt,
 				  0, virt_len);
 	if (rc < 0)
 		printk(KERN_ERR "%s: Error attempting to write header "
@@ -1338,7 +1338,8 @@
 
 /**
  * ecryptfs_write_metadata
- * @ecryptfs_dentry: The eCryptfs dentry
+ * @ecryptfs_dentry: The eCryptfs dentry, which should be negative
+ * @ecryptfs_inode: The newly created eCryptfs inode
  *
  * Write the file headers out.  This will likely involve a userspace
  * callout, in which the session key is encrypted with one or more
@@ -1348,10 +1349,11 @@
  *
  * Returns zero on success; non-zero on error
  */
-int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry)
+int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry,
+			    struct inode *ecryptfs_inode)
 {
 	struct ecryptfs_crypt_stat *crypt_stat =
-		&ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat;
+		&ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat;
 	unsigned int order;
 	char *virt;
 	size_t virt_len;
@@ -1391,7 +1393,7 @@
 		rc = ecryptfs_write_metadata_to_xattr(ecryptfs_dentry, virt,
 						      size);
 	else
-		rc = ecryptfs_write_metadata_to_contents(ecryptfs_dentry, virt,
+		rc = ecryptfs_write_metadata_to_contents(ecryptfs_inode, virt,
 							 virt_len);
 	if (rc) {
 		printk(KERN_ERR "%s: Error writing metadata out to lower file; "
@@ -1943,7 +1945,7 @@
 
 /* We could either offset on every reverse map or just pad some 0x00's
  * at the front here */
-static const unsigned char filename_rev_map[] = {
+static const unsigned char filename_rev_map[256] = {
 	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 7 */
 	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 15 */
 	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 23 */
@@ -1959,7 +1961,7 @@
 	0x00, 0x26, 0x27, 0x28, 0x29, 0x2A, 0x2B, 0x2C, /* 103 */
 	0x2D, 0x2E, 0x2F, 0x30, 0x31, 0x32, 0x33, 0x34, /* 111 */
 	0x35, 0x36, 0x37, 0x38, 0x39, 0x3A, 0x3B, 0x3C, /* 119 */
-	0x3D, 0x3E, 0x3F
+	0x3D, 0x3E, 0x3F /* 123 - 255 initialized to 0x00 */
 };
 
 /**
diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h
index 54481a3..a9f29b1 100644
--- a/fs/ecryptfs/ecryptfs_kernel.h
+++ b/fs/ecryptfs/ecryptfs_kernel.h
@@ -584,9 +584,10 @@
 int ecryptfs_write_inode_size_to_metadata(struct inode *ecryptfs_inode);
 int ecryptfs_encrypt_page(struct page *page);
 int ecryptfs_decrypt_page(struct page *page);
-int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry);
+int ecryptfs_write_metadata(struct dentry *ecryptfs_dentry,
+			    struct inode *ecryptfs_inode);
 int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry);
-int ecryptfs_new_file_context(struct dentry *ecryptfs_dentry);
+int ecryptfs_new_file_context(struct inode *ecryptfs_inode);
 void ecryptfs_write_crypt_stat_flags(char *page_virt,
 				     struct ecryptfs_crypt_stat *crypt_stat,
 				     size_t *written);
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index c6ac98c..d3f95f9 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -139,6 +139,27 @@
 	return rc;
 }
 
+static void ecryptfs_vma_close(struct vm_area_struct *vma)
+{
+	filemap_write_and_wait(vma->vm_file->f_mapping);
+}
+
+static const struct vm_operations_struct ecryptfs_file_vm_ops = {
+	.close		= ecryptfs_vma_close,
+	.fault		= filemap_fault,
+};
+
+static int ecryptfs_file_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	int rc;
+
+	rc = generic_file_mmap(file, vma);
+	if (!rc)
+		vma->vm_ops = &ecryptfs_file_vm_ops;
+
+	return rc;
+}
+
 struct kmem_cache *ecryptfs_file_info_cache;
 
 /**
@@ -349,7 +370,7 @@
 #ifdef CONFIG_COMPAT
 	.compat_ioctl = ecryptfs_compat_ioctl,
 #endif
-	.mmap = generic_file_mmap,
+	.mmap = ecryptfs_file_mmap,
 	.open = ecryptfs_open,
 	.flush = ecryptfs_flush,
 	.release = ecryptfs_release,
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index a36d327..32f90a3a 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -172,22 +172,23 @@
  * it. It will also update the eCryptfs directory inode to mimic the
  * stat of the lower directory inode.
  *
- * Returns zero on success; non-zero on error condition
+ * Returns the new eCryptfs inode on success; an ERR_PTR on error condition
  */
-static int
+static struct inode *
 ecryptfs_do_create(struct inode *directory_inode,
 		   struct dentry *ecryptfs_dentry, int mode)
 {
 	int rc;
 	struct dentry *lower_dentry;
 	struct dentry *lower_dir_dentry;
+	struct inode *inode;
 
 	lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry);
 	lower_dir_dentry = lock_parent(lower_dentry);
 	if (IS_ERR(lower_dir_dentry)) {
 		ecryptfs_printk(KERN_ERR, "Error locking directory of "
 				"dentry\n");
-		rc = PTR_ERR(lower_dir_dentry);
+		inode = ERR_CAST(lower_dir_dentry);
 		goto out;
 	}
 	rc = ecryptfs_create_underlying_file(lower_dir_dentry->d_inode,
@@ -195,20 +196,19 @@
 	if (rc) {
 		printk(KERN_ERR "%s: Failure to create dentry in lower fs; "
 		       "rc = [%d]\n", __func__, rc);
+		inode = ERR_PTR(rc);
 		goto out_lock;
 	}
-	rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry,
-				directory_inode->i_sb);
-	if (rc) {
-		ecryptfs_printk(KERN_ERR, "Failure in ecryptfs_interpose\n");
+	inode = __ecryptfs_get_inode(lower_dentry->d_inode,
+				     directory_inode->i_sb);
+	if (IS_ERR(inode))
 		goto out_lock;
-	}
 	fsstack_copy_attr_times(directory_inode, lower_dir_dentry->d_inode);
 	fsstack_copy_inode_size(directory_inode, lower_dir_dentry->d_inode);
 out_lock:
 	unlock_dir(lower_dir_dentry);
 out:
-	return rc;
+	return inode;
 }
 
 /**
@@ -219,26 +219,26 @@
  *
  * Returns zero on success
  */
-static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry)
+static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry,
+				    struct inode *ecryptfs_inode)
 {
 	struct ecryptfs_crypt_stat *crypt_stat =
-		&ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat;
+		&ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat;
 	int rc = 0;
 
-	if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) {
+	if (S_ISDIR(ecryptfs_inode->i_mode)) {
 		ecryptfs_printk(KERN_DEBUG, "This is a directory\n");
 		crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED);
 		goto out;
 	}
 	ecryptfs_printk(KERN_DEBUG, "Initializing crypto context\n");
-	rc = ecryptfs_new_file_context(ecryptfs_dentry);
+	rc = ecryptfs_new_file_context(ecryptfs_inode);
 	if (rc) {
 		ecryptfs_printk(KERN_ERR, "Error creating new file "
 				"context; rc = [%d]\n", rc);
 		goto out;
 	}
-	rc = ecryptfs_get_lower_file(ecryptfs_dentry,
-				     ecryptfs_dentry->d_inode);
+	rc = ecryptfs_get_lower_file(ecryptfs_dentry, ecryptfs_inode);
 	if (rc) {
 		printk(KERN_ERR "%s: Error attempting to initialize "
 			"the lower file for the dentry with name "
@@ -246,10 +246,10 @@
 			ecryptfs_dentry->d_name.name, rc);
 		goto out;
 	}
-	rc = ecryptfs_write_metadata(ecryptfs_dentry);
+	rc = ecryptfs_write_metadata(ecryptfs_dentry, ecryptfs_inode);
 	if (rc)
 		printk(KERN_ERR "Error writing headers; rc = [%d]\n", rc);
-	ecryptfs_put_lower_file(ecryptfs_dentry->d_inode);
+	ecryptfs_put_lower_file(ecryptfs_inode);
 out:
 	return rc;
 }
@@ -269,18 +269,28 @@
 ecryptfs_create(struct inode *directory_inode, struct dentry *ecryptfs_dentry,
 		int mode, struct nameidata *nd)
 {
+	struct inode *ecryptfs_inode;
 	int rc;
 
-	/* ecryptfs_do_create() calls ecryptfs_interpose() */
-	rc = ecryptfs_do_create(directory_inode, ecryptfs_dentry, mode);
-	if (unlikely(rc)) {
+	ecryptfs_inode = ecryptfs_do_create(directory_inode, ecryptfs_dentry,
+					    mode);
+	if (unlikely(IS_ERR(ecryptfs_inode))) {
 		ecryptfs_printk(KERN_WARNING, "Failed to create file in"
 				"lower filesystem\n");
+		rc = PTR_ERR(ecryptfs_inode);
 		goto out;
 	}
 	/* At this point, a file exists on "disk"; we need to make sure
 	 * that this on disk file is prepared to be an ecryptfs file */
-	rc = ecryptfs_initialize_file(ecryptfs_dentry);
+	rc = ecryptfs_initialize_file(ecryptfs_dentry, ecryptfs_inode);
+	if (rc) {
+		drop_nlink(ecryptfs_inode);
+		unlock_new_inode(ecryptfs_inode);
+		iput(ecryptfs_inode);
+		goto out;
+	}
+	d_instantiate(ecryptfs_dentry, ecryptfs_inode);
+	unlock_new_inode(ecryptfs_inode);
 out:
 	return rc;
 }
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index f6dba45..12ccacd 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -565,7 +565,7 @@
 	brelse(bitmap_bh);
 	printk(KERN_DEBUG "ext4_count_free_clusters: stored = %llu"
 	       ", computed = %llu, %llu\n",
-	       EXT4_B2C(sbi, ext4_free_blocks_count(es)),
+	       EXT4_B2C(EXT4_SB(sb), ext4_free_blocks_count(es)),
 	       desc_count, bitmap_count);
 	return bitmap_count;
 #else
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 61fa9e1..607b155 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -1095,7 +1095,7 @@
 		  le32_to_cpu(EXT_FIRST_INDEX(neh)->ei_block),
 		  ext4_idx_pblock(EXT_FIRST_INDEX(neh)));
 
-	neh->eh_depth = cpu_to_le16(neh->eh_depth + 1);
+	neh->eh_depth = cpu_to_le16(le16_to_cpu(neh->eh_depth) + 1);
 	ext4_mark_inode_dirty(handle, inode);
 out:
 	brelse(bh);
@@ -2955,7 +2955,6 @@
 	/* Pre-conditions */
 	BUG_ON(!ext4_ext_is_uninitialized(ex));
 	BUG_ON(!in_range(map->m_lblk, ee_block, ee_len));
-	BUG_ON(map->m_lblk + map->m_len > ee_block + ee_len);
 
 	/*
 	 * Attempt to transfer newly initialized blocks from the currently
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 240f6e2..92655fd 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -1339,8 +1339,11 @@
 					clear_buffer_unwritten(bh);
 				}
 
-				/* skip page if block allocation undone */
-				if (buffer_delay(bh) || buffer_unwritten(bh))
+				/*
+				 * skip page if block allocation undone and
+				 * block is dirty
+				 */
+				if (ext4_bh_delay_or_unwritten(NULL, bh))
 					skip_page = 1;
 				bh = bh->b_this_page;
 				block_start += bh->b_size;
@@ -2270,6 +2273,7 @@
 			ext4_msg(inode->i_sb, KERN_CRIT, "%s: jbd2_start: "
 			       "%ld pages, ino %lu; err %d", __func__,
 				wbc->nr_to_write, inode->i_ino, ret);
+			blk_finish_plug(&plug);
 			goto out_writepages;
 		}
 
@@ -2386,7 +2390,6 @@
 	pgoff_t index;
 	struct inode *inode = mapping->host;
 	handle_t *handle;
-	loff_t page_len;
 
 	index = pos >> PAGE_CACHE_SHIFT;
 
@@ -2433,13 +2436,6 @@
 		 */
 		if (pos + len > inode->i_size)
 			ext4_truncate_failed_write(inode);
-	} else {
-		page_len = pos & (PAGE_CACHE_SIZE - 1);
-		if (page_len > 0) {
-			ret = ext4_discard_partial_page_buffers_no_lock(handle,
-				inode, page, pos - page_len, page_len,
-				EXT4_DISCARD_PARTIAL_PG_ZERO_UNMAPPED);
-		}
 	}
 
 	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
@@ -2482,7 +2478,6 @@
 	loff_t new_i_size;
 	unsigned long start, end;
 	int write_mode = (int)(unsigned long)fsdata;
-	loff_t page_len;
 
 	if (write_mode == FALL_BACK_TO_NONDELALLOC) {
 		if (ext4_should_order_data(inode)) {
@@ -2507,7 +2502,7 @@
 	 */
 
 	new_i_size = pos + copied;
-	if (new_i_size > EXT4_I(inode)->i_disksize) {
+	if (copied && new_i_size > EXT4_I(inode)->i_disksize) {
 		if (ext4_da_should_update_i_disksize(page, end)) {
 			down_write(&EXT4_I(inode)->i_data_sem);
 			if (new_i_size > EXT4_I(inode)->i_disksize) {
@@ -2531,16 +2526,6 @@
 	}
 	ret2 = generic_write_end(file, mapping, pos, len, copied,
 							page, fsdata);
-
-	page_len = PAGE_CACHE_SIZE -
-			((pos + copied - 1) & (PAGE_CACHE_SIZE - 1));
-
-	if (page_len > 0) {
-		ret = ext4_discard_partial_page_buffers_no_lock(handle,
-			inode, page, pos + copied - 1, page_len,
-			EXT4_DISCARD_PARTIAL_PG_ZERO_UNMAPPED);
-	}
-
 	copied = ret2;
 	if (ret2 < 0)
 		ret = ret2;
@@ -2780,10 +2765,11 @@
  		  iocb->private, io_end->inode->i_ino, iocb, offset,
 		  size);
 
+	iocb->private = NULL;
+
 	/* if not aio dio with unwritten extents, just free io and return */
 	if (!(io_end->flag & EXT4_IO_END_UNWRITTEN)) {
 		ext4_free_io_end(io_end);
-		iocb->private = NULL;
 out:
 		if (is_async)
 			aio_complete(iocb, ret, 0);
@@ -2807,7 +2793,6 @@
 
 	/* queue the work to convert unwritten extents to written */
 	queue_work(wq, &io_end->work);
-	iocb->private = NULL;
 
 	/* XXX: probably should move into the real I/O completion handler */
 	inode_dio_done(inode);
@@ -3202,26 +3187,8 @@
 
 	iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits);
 
-	if (!page_has_buffers(page)) {
-		/*
-		 * If the range to be discarded covers a partial block
-		 * we need to get the page buffers.  This is because
-		 * partial blocks cannot be released and the page needs
-		 * to be updated with the contents of the block before
-		 * we write the zeros on top of it.
-		 */
-		if ((from & (blocksize - 1)) ||
-		    ((from + length) & (blocksize - 1))) {
-			create_empty_buffers(page, blocksize, 0);
-		} else {
-			/*
-			 * If there are no partial blocks,
-			 * there is nothing to update,
-			 * so we can return now
-			 */
-			return 0;
-		}
-	}
+	if (!page_has_buffers(page))
+		create_empty_buffers(page, blocksize, 0);
 
 	/* Find the buffer that contains "offset" */
 	bh = page_buffers(page);
diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c
index 7ce1d0b..7e106c8 100644
--- a/fs/ext4/page-io.c
+++ b/fs/ext4/page-io.c
@@ -385,6 +385,18 @@
 
 		block_end = block_start + blocksize;
 		if (block_start >= len) {
+			/*
+			 * Comments copied from block_write_full_page_endio:
+			 *
+			 * The page straddles i_size.  It must be zeroed out on
+			 * each and every writepage invocation because it may
+			 * be mmapped.  "A file is mapped in multiples of the
+			 * page size.  For a file that is not a multiple of
+			 * the  page size, the remaining memory is zeroed when
+			 * mapped, and writes to that region are not written
+			 * out to the file."
+			 */
+			zero_user_segment(page, block_start, block_end);
 			clear_buffer_dirty(bh);
 			set_buffer_uptodate(bh);
 			continue;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 9953d80..3e1329e 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1155,9 +1155,9 @@
 		seq_puts(seq, ",block_validity");
 
 	if (!test_opt(sb, INIT_INODE_TABLE))
-		seq_puts(seq, ",noinit_inode_table");
+		seq_puts(seq, ",noinit_itable");
 	else if (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT)
-		seq_printf(seq, ",init_inode_table=%u",
+		seq_printf(seq, ",init_itable=%u",
 			   (unsigned) sbi->s_li_wait_mult);
 
 	ext4_show_quota_options(seq, sb);
@@ -1333,8 +1333,7 @@
 	Opt_nomblk_io_submit, Opt_block_validity, Opt_noblock_validity,
 	Opt_inode_readahead_blks, Opt_journal_ioprio,
 	Opt_dioread_nolock, Opt_dioread_lock,
-	Opt_discard, Opt_nodiscard,
-	Opt_init_inode_table, Opt_noinit_inode_table,
+	Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable,
 };
 
 static const match_table_t tokens = {
@@ -1407,9 +1406,9 @@
 	{Opt_dioread_lock, "dioread_lock"},
 	{Opt_discard, "discard"},
 	{Opt_nodiscard, "nodiscard"},
-	{Opt_init_inode_table, "init_itable=%u"},
-	{Opt_init_inode_table, "init_itable"},
-	{Opt_noinit_inode_table, "noinit_itable"},
+	{Opt_init_itable, "init_itable=%u"},
+	{Opt_init_itable, "init_itable"},
+	{Opt_noinit_itable, "noinit_itable"},
 	{Opt_err, NULL},
 };
 
@@ -1683,7 +1682,9 @@
 			data_opt = EXT4_MOUNT_WRITEBACK_DATA;
 		datacheck:
 			if (is_remount) {
-				if (test_opt(sb, DATA_FLAGS) != data_opt) {
+				if (!sbi->s_journal)
+					ext4_msg(sb, KERN_WARNING, "Remounting file system with no journal so ignoring journalled data option");
+				else if (test_opt(sb, DATA_FLAGS) != data_opt) {
 					ext4_msg(sb, KERN_ERR,
 						"Cannot change data mode on remount");
 					return 0;
@@ -1890,7 +1891,7 @@
 		case Opt_dioread_lock:
 			clear_opt(sb, DIOREAD_NOLOCK);
 			break;
-		case Opt_init_inode_table:
+		case Opt_init_itable:
 			set_opt(sb, INIT_INODE_TABLE);
 			if (args[0].from) {
 				if (match_int(&args[0], &option))
@@ -1901,7 +1902,7 @@
 				return 0;
 			sbi->s_li_wait_mult = option;
 			break;
-		case Opt_noinit_inode_table:
+		case Opt_noinit_itable:
 			clear_opt(sb, INIT_INODE_TABLE);
 			break;
 		default:
@@ -3099,8 +3100,6 @@
 }
 
 static int ext4_fill_super(struct super_block *sb, void *data, int silent)
-				__releases(kernel_lock)
-				__acquires(kernel_lock)
 {
 	char *orig_data = kstrdup(data, GFP_KERNEL);
 	struct buffer_head *bh;
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 73c3992..ac86f8b 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -156,6 +156,7 @@
  * bdi_start_writeback - start writeback
  * @bdi: the backing device to write from
  * @nr_pages: the number of pages to write
+ * @reason: reason why some writeback work was initiated
  *
  * Description:
  *   This does WB_SYNC_NONE opportunistic writeback. The IO is only
@@ -1223,6 +1224,7 @@
  * writeback_inodes_sb_nr -	writeback dirty inodes from given super_block
  * @sb: the superblock
  * @nr: the number of pages to write
+ * @reason: reason why some writeback work initiated
  *
  * Start writeback on some inodes on this super_block. No guarantees are made
  * on how many (if any) will be written, and this function does not wait
@@ -1251,6 +1253,7 @@
 /**
  * writeback_inodes_sb	-	writeback dirty inodes from given super_block
  * @sb: the superblock
+ * @reason: reason why some writeback work was initiated
  *
  * Start writeback on some inodes on this super_block. No guarantees are made
  * on how many (if any) will be written, and this function does not wait
@@ -1265,6 +1268,7 @@
 /**
  * writeback_inodes_sb_if_idle	-	start writeback if none underway
  * @sb: the superblock
+ * @reason: reason why some writeback work was initiated
  *
  * Invoke writeback_inodes_sb if no writeback is currently underway.
  * Returns 1 if writeback was started, 0 if not.
@@ -1285,6 +1289,7 @@
  * writeback_inodes_sb_if_idle	-	start writeback if none underway
  * @sb: the superblock
  * @nr: the number of pages to write
+ * @reason: reason why some writeback work was initiated
  *
  * Invoke writeback_inodes_sb if no writeback is currently underway.
  * Returns 1 if writeback was started, 0 if not.
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 5cb8614..2aaf3ea 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -1512,7 +1512,7 @@
 	else if (outarg->offset + num > file_size)
 		num = file_size - outarg->offset;
 
-	while (num) {
+	while (num && req->num_pages < FUSE_MAX_PAGES_PER_REQ) {
 		struct page *page;
 		unsigned int this_num;
 
@@ -1526,6 +1526,7 @@
 
 		num -= this_num;
 		total_len += this_num;
+		index++;
 	}
 	req->misc.retrieve_in.offset = outarg->offset;
 	req->misc.retrieve_in.size = total_len;
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 594f07a..0c84100 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1556,7 +1556,7 @@
 	struct inode *inode = file->f_path.dentry->d_inode;
 
 	mutex_lock(&inode->i_mutex);
-	if (origin != SEEK_CUR || origin != SEEK_SET) {
+	if (origin != SEEK_CUR && origin != SEEK_SET) {
 		retval = fuse_update_attributes(inode, NULL, file, NULL);
 		if (retval)
 			goto exit;
@@ -1567,6 +1567,10 @@
 		offset += i_size_read(inode);
 		break;
 	case SEEK_CUR:
+		if (offset == 0) {
+			retval = file->f_pos;
+			goto exit;
+		}
 		offset += file->f_pos;
 		break;
 	case SEEK_DATA:
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 3e6d727..aa83109 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -1138,28 +1138,28 @@
 {
 	int err;
 
-	err = register_filesystem(&fuse_fs_type);
-	if (err)
-		goto out;
-
-	err = register_fuseblk();
-	if (err)
-		goto out_unreg;
-
 	fuse_inode_cachep = kmem_cache_create("fuse_inode",
 					      sizeof(struct fuse_inode),
 					      0, SLAB_HWCACHE_ALIGN,
 					      fuse_inode_init_once);
 	err = -ENOMEM;
 	if (!fuse_inode_cachep)
-		goto out_unreg2;
+		goto out;
+
+	err = register_fuseblk();
+	if (err)
+		goto out2;
+
+	err = register_filesystem(&fuse_fs_type);
+	if (err)
+		goto out3;
 
 	return 0;
 
- out_unreg2:
+ out3:
 	unregister_fuseblk();
- out_unreg:
-	unregister_filesystem(&fuse_fs_type);
+ out2:
+	kmem_cache_destroy(fuse_inode_cachep);
  out:
 	return err;
 }
diff --git a/fs/hfs/trans.c b/fs/hfs/trans.c
index e673a88..b1ce4c7 100644
--- a/fs/hfs/trans.c
+++ b/fs/hfs/trans.c
@@ -40,6 +40,8 @@
 
 	src = in->name;
 	srclen = in->len;
+	if (srclen > HFS_NAMELEN)
+		srclen = HFS_NAMELEN;
 	dst = out;
 	dstlen = HFS_MAX_NAMELEN;
 	if (nls_io) {
diff --git a/fs/minix/bitmap.c b/fs/minix/bitmap.c
index 3f32bcb..ef175cb 100644
--- a/fs/minix/bitmap.c
+++ b/fs/minix/bitmap.c
@@ -16,38 +16,26 @@
 #include <linux/bitops.h>
 #include <linux/sched.h>
 
-static const int nibblemap[] = { 4,3,3,2,3,2,2,1,3,2,2,1,2,1,1,0 };
-
 static DEFINE_SPINLOCK(bitmap_lock);
 
-static unsigned long count_free(struct buffer_head *map[], unsigned numblocks, __u32 numbits)
+/*
+ * bitmap consists of blocks filled with 16bit words
+ * bit set == busy, bit clear == free
+ * endianness is a mess, but for counting zero bits it really doesn't matter...
+ */
+static __u32 count_free(struct buffer_head *map[], unsigned blocksize, __u32 numbits)
 {
-	unsigned i, j, sum = 0;
-	struct buffer_head *bh;
-  
-	for (i=0; i<numblocks-1; i++) {
-		if (!(bh=map[i])) 
-			return(0);
-		for (j=0; j<bh->b_size; j++)
-			sum += nibblemap[bh->b_data[j] & 0xf]
-				+ nibblemap[(bh->b_data[j]>>4) & 0xf];
+	__u32 sum = 0;
+	unsigned blocks = DIV_ROUND_UP(numbits, blocksize * 8);
+
+	while (blocks--) {
+		unsigned words = blocksize / 2;
+		__u16 *p = (__u16 *)(*map++)->b_data;
+		while (words--)
+			sum += 16 - hweight16(*p++);
 	}
 
-	if (numblocks==0 || !(bh=map[numblocks-1]))
-		return(0);
-	i = ((numbits - (numblocks-1) * bh->b_size * 8) / 16) * 2;
-	for (j=0; j<i; j++) {
-		sum += nibblemap[bh->b_data[j] & 0xf]
-			+ nibblemap[(bh->b_data[j]>>4) & 0xf];
-	}
-
-	i = numbits%16;
-	if (i!=0) {
-		i = *(__u16 *)(&bh->b_data[j]) | ~((1<<i) - 1);
-		sum += nibblemap[i & 0xf] + nibblemap[(i>>4) & 0xf];
-		sum += nibblemap[(i>>8) & 0xf] + nibblemap[(i>>12) & 0xf];
-	}
-	return(sum);
+	return sum;
 }
 
 void minix_free_block(struct inode *inode, unsigned long block)
@@ -105,10 +93,12 @@
 	return 0;
 }
 
-unsigned long minix_count_free_blocks(struct minix_sb_info *sbi)
+unsigned long minix_count_free_blocks(struct super_block *sb)
 {
-	return (count_free(sbi->s_zmap, sbi->s_zmap_blocks,
-		sbi->s_nzones - sbi->s_firstdatazone + 1)
+	struct minix_sb_info *sbi = minix_sb(sb);
+	u32 bits = sbi->s_nzones - (sbi->s_firstdatazone + 1);
+
+	return (count_free(sbi->s_zmap, sb->s_blocksize, bits)
 		<< sbi->s_log_zone_size);
 }
 
@@ -273,7 +263,10 @@
 	return inode;
 }
 
-unsigned long minix_count_free_inodes(struct minix_sb_info *sbi)
+unsigned long minix_count_free_inodes(struct super_block *sb)
 {
-	return count_free(sbi->s_imap, sbi->s_imap_blocks, sbi->s_ninodes + 1);
+	struct minix_sb_info *sbi = minix_sb(sb);
+	u32 bits = sbi->s_ninodes + 1;
+
+	return count_free(sbi->s_imap, sb->s_blocksize, bits);
 }
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index 64cdcd6..1d9e339 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -279,6 +279,27 @@
  	else if (sbi->s_mount_state & MINIX_ERROR_FS)
 		printk("MINIX-fs: mounting file system with errors, "
 			"running fsck is recommended\n");
+
+	/* Apparently minix can create filesystems that allocate more blocks for
+	 * the bitmaps than needed.  We simply ignore that, but verify it didn't
+	 * create one with not enough blocks and bail out if so.
+	 */
+	block = minix_blocks_needed(sbi->s_ninodes, s->s_blocksize);
+	if (sbi->s_imap_blocks < block) {
+		printk("MINIX-fs: file system does not have enough "
+				"imap blocks allocated.  Refusing to mount\n");
+		goto out_iput;
+	}
+
+	block = minix_blocks_needed(
+			(sbi->s_nzones - (sbi->s_firstdatazone + 1)),
+			s->s_blocksize);
+	if (sbi->s_zmap_blocks < block) {
+		printk("MINIX-fs: file system does not have enough "
+				"zmap blocks allocated.  Refusing to mount.\n");
+		goto out_iput;
+	}
+
 	return 0;
 
 out_iput:
@@ -339,10 +360,10 @@
 	buf->f_type = sb->s_magic;
 	buf->f_bsize = sb->s_blocksize;
 	buf->f_blocks = (sbi->s_nzones - sbi->s_firstdatazone) << sbi->s_log_zone_size;
-	buf->f_bfree = minix_count_free_blocks(sbi);
+	buf->f_bfree = minix_count_free_blocks(sb);
 	buf->f_bavail = buf->f_bfree;
 	buf->f_files = sbi->s_ninodes;
-	buf->f_ffree = minix_count_free_inodes(sbi);
+	buf->f_ffree = minix_count_free_inodes(sb);
 	buf->f_namelen = sbi->s_namelen;
 	buf->f_fsid.val[0] = (u32)id;
 	buf->f_fsid.val[1] = (u32)(id >> 32);
diff --git a/fs/minix/minix.h b/fs/minix/minix.h
index 341e212..26bbd55 100644
--- a/fs/minix/minix.h
+++ b/fs/minix/minix.h
@@ -48,10 +48,10 @@
 extern struct minix2_inode * minix_V2_raw_inode(struct super_block *, ino_t, struct buffer_head **);
 extern struct inode * minix_new_inode(const struct inode *, int, int *);
 extern void minix_free_inode(struct inode * inode);
-extern unsigned long minix_count_free_inodes(struct minix_sb_info *sbi);
+extern unsigned long minix_count_free_inodes(struct super_block *sb);
 extern int minix_new_block(struct inode * inode);
 extern void minix_free_block(struct inode *inode, unsigned long block);
-extern unsigned long minix_count_free_blocks(struct minix_sb_info *sbi);
+extern unsigned long minix_count_free_blocks(struct super_block *sb);
 extern int minix_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 extern int minix_prepare_chunk(struct page *page, loff_t pos, unsigned len);
 
@@ -88,6 +88,11 @@
 	return list_entry(inode, struct minix_inode_info, vfs_inode);
 }
 
+static inline unsigned minix_blocks_needed(unsigned bits, unsigned blocksize)
+{
+	return DIV_ROUND_UP(bits, blocksize * 8);
+}
+
 #if defined(CONFIG_MINIX_FS_NATIVE_ENDIAN) && \
 	defined(CONFIG_MINIX_FS_BIG_ENDIAN_16BIT_INDEXED)
 
@@ -125,7 +130,7 @@
 	if (!size)
 		return 0;
 
-	size = (size >> 4) + ((size & 15) > 0);
+	size >>= 4;
 	while (*p++ == 0xffff) {
 		if (--size == 0)
 			return (p - addr) << 4;
diff --git a/fs/namespace.c b/fs/namespace.c
index e5e1c7d..cfc6d44 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1048,15 +1048,12 @@
 	if (err)
 		goto out;
 	seq_putc(m, ' ');
-	seq_path_root(m, &mnt_path, &root, " \t\n\\");
-	if (root.mnt != p->root.mnt || root.dentry != p->root.dentry) {
-		/*
-		 * Mountpoint is outside root, discard that one.  Ugly,
-		 * but less so than trying to do that in iterator in a
-		 * race-free way (due to renames).
-		 */
-		return SEQ_SKIP;
-	}
+
+	/* mountpoints outside of chroot jail will give SEQ_SKIP on this */
+	err = seq_path_root(m, &mnt_path, &root, " \t\n\\");
+	if (err)
+		goto out;
+
 	seq_puts(m, mnt->mnt_flags & MNT_READONLY ? " ro" : " rw");
 	show_mnt_opts(m, mnt);
 
@@ -2483,11 +2480,43 @@
 		__mnt_make_longterm(mnt);
 		new_ns->root = mnt;
 		list_add(&new_ns->list, &new_ns->root->mnt_list);
+	} else {
+		mntput(mnt);
 	}
 	return new_ns;
 }
 EXPORT_SYMBOL(create_mnt_ns);
 
+struct dentry *mount_subtree(struct vfsmount *mnt, const char *name)
+{
+	struct mnt_namespace *ns;
+	struct super_block *s;
+	struct path path;
+	int err;
+
+	ns = create_mnt_ns(mnt);
+	if (IS_ERR(ns))
+		return ERR_CAST(ns);
+
+	err = vfs_path_lookup(mnt->mnt_root, mnt,
+			name, LOOKUP_FOLLOW|LOOKUP_AUTOMOUNT, &path);
+
+	put_mnt_ns(ns);
+
+	if (err)
+		return ERR_PTR(err);
+
+	/* trade a vfsmount reference for active sb one */
+	s = path.mnt->mnt_sb;
+	atomic_inc(&s->s_active);
+	mntput(path.mnt);
+	/* lock the sucker */
+	down_write(&s->s_umount);
+	/* ... and return the root of (sub)tree on it */
+	return path.dentry;
+}
+EXPORT_SYMBOL(mount_subtree);
+
 SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name,
 		char __user *, type, unsigned long, flags, void __user *, data)
 {
@@ -2744,3 +2773,8 @@
 	}
 }
 EXPORT_SYMBOL(kern_unmount);
+
+bool our_mnt(struct vfsmount *mnt)
+{
+	return check_mnt(mnt);
+}
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index 5b5fa33..cbd1a61 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -548,7 +548,7 @@
 
 	error = bdi_setup_and_register(&server->bdi, "ncpfs", BDI_CAP_MAP_COPY);
 	if (error)
-		goto out_bdi;
+		goto out_fput;
 
 	server->ncp_filp = ncp_filp;
 	server->ncp_sock = sock;
@@ -559,7 +559,7 @@
 		error = -EBADF;
 		server->info_filp = fget(data.info_fd);
 		if (!server->info_filp)
-			goto out_fput;
+			goto out_bdi;
 		error = -ENOTSOCK;
 		sock_inode = server->info_filp->f_path.dentry->d_inode;
 		if (!S_ISSOCK(sock_inode->i_mode))
@@ -746,9 +746,9 @@
 out_fput2:
 	if (server->info_filp)
 		fput(server->info_filp);
-out_fput:
-	bdi_destroy(&server->bdi);
 out_bdi:
+	bdi_destroy(&server->bdi);
+out_fput:
 	/* 23/12/1998 Marcin Dalecki <dalecki@cs.net.pl>:
 	 * 
 	 * The previously used put_filp(ncp_filp); was bogus, since
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index b238d95..ac28990 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1468,12 +1468,12 @@
 				res = NULL;
 				goto out;
 			/* This turned out not to be a regular file */
+			case -EISDIR:
 			case -ENOTDIR:
 				goto no_open;
 			case -ELOOP:
 				if (!(nd->intent.open.flags & O_NOFOLLOW))
 					goto no_open;
-			/* case -EISDIR: */
 			/* case -EINVAL: */
 			default:
 				res = ERR_CAST(inode);
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 0a1f831..eca56d4 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -40,48 +40,8 @@
 
 #define NFSDBG_FACILITY		NFSDBG_FILE
 
-static int nfs_file_open(struct inode *, struct file *);
-static int nfs_file_release(struct inode *, struct file *);
-static loff_t nfs_file_llseek(struct file *file, loff_t offset, int origin);
-static int  nfs_file_mmap(struct file *, struct vm_area_struct *);
-static ssize_t nfs_file_splice_read(struct file *filp, loff_t *ppos,
-					struct pipe_inode_info *pipe,
-					size_t count, unsigned int flags);
-static ssize_t nfs_file_read(struct kiocb *, const struct iovec *iov,
-				unsigned long nr_segs, loff_t pos);
-static ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe,
-					struct file *filp, loff_t *ppos,
-					size_t count, unsigned int flags);
-static ssize_t nfs_file_write(struct kiocb *, const struct iovec *iov,
-				unsigned long nr_segs, loff_t pos);
-static int  nfs_file_flush(struct file *, fl_owner_t id);
-static int  nfs_file_fsync(struct file *, loff_t, loff_t, int datasync);
-static int nfs_check_flags(int flags);
-static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl);
-static int nfs_flock(struct file *filp, int cmd, struct file_lock *fl);
-static int nfs_setlease(struct file *file, long arg, struct file_lock **fl);
-
 static const struct vm_operations_struct nfs_file_vm_ops;
 
-const struct file_operations nfs_file_operations = {
-	.llseek		= nfs_file_llseek,
-	.read		= do_sync_read,
-	.write		= do_sync_write,
-	.aio_read	= nfs_file_read,
-	.aio_write	= nfs_file_write,
-	.mmap		= nfs_file_mmap,
-	.open		= nfs_file_open,
-	.flush		= nfs_file_flush,
-	.release	= nfs_file_release,
-	.fsync		= nfs_file_fsync,
-	.lock		= nfs_lock,
-	.flock		= nfs_flock,
-	.splice_read	= nfs_file_splice_read,
-	.splice_write	= nfs_file_splice_write,
-	.check_flags	= nfs_check_flags,
-	.setlease	= nfs_setlease,
-};
-
 const struct inode_operations nfs_file_inode_operations = {
 	.permission	= nfs_permission,
 	.getattr	= nfs_getattr,
@@ -886,3 +846,54 @@
 			file->f_path.dentry->d_name.name, arg);
 	return -EINVAL;
 }
+
+const struct file_operations nfs_file_operations = {
+	.llseek		= nfs_file_llseek,
+	.read		= do_sync_read,
+	.write		= do_sync_write,
+	.aio_read	= nfs_file_read,
+	.aio_write	= nfs_file_write,
+	.mmap		= nfs_file_mmap,
+	.open		= nfs_file_open,
+	.flush		= nfs_file_flush,
+	.release	= nfs_file_release,
+	.fsync		= nfs_file_fsync,
+	.lock		= nfs_lock,
+	.flock		= nfs_flock,
+	.splice_read	= nfs_file_splice_read,
+	.splice_write	= nfs_file_splice_write,
+	.check_flags	= nfs_check_flags,
+	.setlease	= nfs_setlease,
+};
+
+#ifdef CONFIG_NFS_V4
+static int
+nfs4_file_open(struct inode *inode, struct file *filp)
+{
+	/*
+	 * NFSv4 opens are handled in d_lookup and d_revalidate. If we get to
+	 * this point, then something is very wrong
+	 */
+	dprintk("NFS: %s called! inode=%p filp=%p\n", __func__, inode, filp);
+	return -ENOTDIR;
+}
+
+const struct file_operations nfs4_file_operations = {
+	.llseek		= nfs_file_llseek,
+	.read		= do_sync_read,
+	.write		= do_sync_write,
+	.aio_read	= nfs_file_read,
+	.aio_write	= nfs_file_write,
+	.mmap		= nfs_file_mmap,
+	.open		= nfs4_file_open,
+	.flush		= nfs_file_flush,
+	.release	= nfs_file_release,
+	.fsync		= nfs_file_fsync,
+	.lock		= nfs_lock,
+	.flock		= nfs_flock,
+	.splice_read	= nfs_file_splice_read,
+	.splice_write	= nfs_file_splice_write,
+	.check_flags	= nfs_check_flags,
+	.setlease	= nfs_setlease,
+};
+#endif /* CONFIG_NFS_V4 */
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index c07a55a..50a15fa 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -291,7 +291,7 @@
 		 */
 		inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->file_inode_ops;
 		if (S_ISREG(inode->i_mode)) {
-			inode->i_fop = &nfs_file_operations;
+			inode->i_fop = NFS_SB(sb)->nfs_client->rpc_ops->file_ops;
 			inode->i_data.a_ops = &nfs_file_aops;
 			inode->i_data.backing_dev_info = &NFS_SB(sb)->backing_dev_info;
 		} else if (S_ISDIR(inode->i_mode)) {
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index c1a1bd8..3f4d957 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -299,6 +299,8 @@
 extern int nfs_generic_pagein(struct nfs_pageio_descriptor *desc,
 		struct list_head *head);
 
+extern void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio,
+		struct inode *inode);
 extern void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio);
 extern void nfs_readdata_release(struct nfs_read_data *rdata);
 
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 85f1690..d4bc9ed9 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -853,6 +853,7 @@
 	.dentry_ops	= &nfs_dentry_operations,
 	.dir_inode_ops	= &nfs3_dir_inode_operations,
 	.file_inode_ops	= &nfs3_file_inode_operations,
+	.file_ops	= &nfs_file_operations,
 	.getroot	= nfs3_proc_get_root,
 	.getattr	= nfs3_proc_getattr,
 	.setattr	= nfs3_proc_setattr,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b60fddf..be2bbac 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2464,8 +2464,7 @@
 		case -NFS4ERR_BADNAME:
 			return -ENOENT;
 		case -NFS4ERR_MOVED:
-			err = nfs4_get_referral(dir, name, fattr, fhandle);
-			break;
+			return nfs4_get_referral(dir, name, fattr, fhandle);
 		case -NFS4ERR_WRONGSEC:
 			nfs_fixup_secinfo_attributes(fattr, fhandle);
 		}
@@ -6253,6 +6252,7 @@
 	.dentry_ops	= &nfs4_dentry_operations,
 	.dir_inode_ops	= &nfs4_dir_inode_operations,
 	.file_inode_ops	= &nfs4_file_inode_operations,
+	.file_ops	= &nfs4_file_operations,
 	.getroot	= nfs4_proc_get_root,
 	.getattr	= nfs4_proc_getattr,
 	.setattr	= nfs4_proc_setattr,
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index baf7353..8e672a2 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -1260,6 +1260,25 @@
 }
 EXPORT_SYMBOL_GPL(pnfs_generic_pg_writepages);
 
+static void pnfs_ld_handle_read_error(struct nfs_read_data *data)
+{
+	struct nfs_pageio_descriptor pgio;
+
+	put_lseg(data->lseg);
+	data->lseg = NULL;
+	dprintk("pnfs write error = %d\n", data->pnfs_error);
+
+	nfs_pageio_init_read_mds(&pgio, data->inode);
+
+	while (!list_empty(&data->pages)) {
+		struct nfs_page *req = nfs_list_entry(data->pages.next);
+
+		nfs_list_remove_request(req);
+		nfs_pageio_add_request(&pgio, req);
+	}
+	nfs_pageio_complete(&pgio);
+}
+
 /*
  * Called by non rpc-based layout drivers
  */
@@ -1268,11 +1287,8 @@
 	if (likely(!data->pnfs_error)) {
 		__nfs4_read_done_cb(data);
 		data->mds_ops->rpc_call_done(&data->task, data);
-	} else {
-		put_lseg(data->lseg);
-		data->lseg = NULL;
-		dprintk("pnfs write error = %d\n", data->pnfs_error);
-	}
+	} else
+		pnfs_ld_handle_read_error(data);
 	data->mds_ops->rpc_release(data);
 }
 EXPORT_SYMBOL_GPL(pnfs_ld_read_done);
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index ac40b85..f48125d 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -710,6 +710,7 @@
 	.dentry_ops	= &nfs_dentry_operations,
 	.dir_inode_ops	= &nfs_dir_inode_operations,
 	.file_inode_ops	= &nfs_file_inode_operations,
+	.file_ops	= &nfs_file_operations,
 	.getroot	= nfs_proc_get_root,
 	.getattr	= nfs_proc_getattr,
 	.setattr	= nfs_proc_setattr,
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 8b48ec6..cfa175c 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -109,7 +109,7 @@
 	}
 }
 
-static void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio,
+void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio,
 		struct inode *inode)
 {
 	nfs_pageio_init(pgio, inode, &nfs_pageio_read_ops,
@@ -534,23 +534,13 @@
 static void nfs_readpage_release_full(void *calldata)
 {
 	struct nfs_read_data *data = calldata;
-	struct nfs_pageio_descriptor pgio;
 
-	if (data->pnfs_error) {
-		nfs_pageio_init_read_mds(&pgio, data->inode);
-		pgio.pg_recoalesce = 1;
-	}
 	while (!list_empty(&data->pages)) {
 		struct nfs_page *req = nfs_list_entry(data->pages.next);
 
 		nfs_list_remove_request(req);
-		if (!data->pnfs_error)
-			nfs_readpage_release(req);
-		else
-			nfs_pageio_add_request(&pgio, req);
+		nfs_readpage_release(req);
 	}
-	if (data->pnfs_error)
-		nfs_pageio_complete(&pgio);
 	nfs_readdata_release(calldata);
 }
 
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 480b3b6..1347774 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -2787,43 +2787,18 @@
 static struct dentry *nfs_follow_remote_path(struct vfsmount *root_mnt,
 		const char *export_path)
 {
-	struct mnt_namespace *ns_private;
-	struct super_block *s;
 	struct dentry *dentry;
-	struct path path;
-	int ret;
+	int ret = nfs_referral_loop_protect();
 
-	ns_private = create_mnt_ns(root_mnt);
-	ret = PTR_ERR(ns_private);
-	if (IS_ERR(ns_private))
-		goto out_mntput;
+	if (ret) {
+		mntput(root_mnt);
+		return ERR_PTR(ret);
+	}
 
-	ret = nfs_referral_loop_protect();
-	if (ret != 0)
-		goto out_put_mnt_ns;
-
-	ret = vfs_path_lookup(root_mnt->mnt_root, root_mnt,
-			export_path, LOOKUP_FOLLOW|LOOKUP_AUTOMOUNT, &path);
-
+	dentry = mount_subtree(root_mnt, export_path);
 	nfs_referral_loop_unprotect();
-	put_mnt_ns(ns_private);
 
-	if (ret != 0)
-		goto out_err;
-
-	s = path.mnt->mnt_sb;
-	atomic_inc(&s->s_active);
-	dentry = dget(path.dentry);
-
-	path_put(&path);
-	down_write(&s->s_umount);
 	return dentry;
-out_put_mnt_ns:
-	put_mnt_ns(ns_private);
-out_mntput:
-	mntput(root_mnt);
-out_err:
-	return ERR_PTR(ret);
 }
 
 static struct dentry *nfs4_try_mount(int flags, const char *dev_name,
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index ed553c6..3165aeb 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -5699,7 +5699,7 @@
 					   OCFS2_JOURNAL_ACCESS_WRITE);
 	if (ret) {
 		mlog_errno(ret);
-		goto out;
+		goto out_commit;
 	}
 
 	dquot_free_space_nodirty(inode,
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index c1efe93..78b68af 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -290,7 +290,15 @@
 	}
 
 	if (down_read_trylock(&oi->ip_alloc_sem) == 0) {
+		/*
+		 * Unlock the page and cycle ip_alloc_sem so that we don't
+		 * busyloop waiting for ip_alloc_sem to unlock
+		 */
 		ret = AOP_TRUNCATED_PAGE;
+		unlock_page(page);
+		unlock = 0;
+		down_read(&oi->ip_alloc_sem);
+		up_read(&oi->ip_alloc_sem);
 		goto out_inode_unlock;
 	}
 
@@ -563,6 +571,7 @@
 {
 	struct inode *inode = iocb->ki_filp->f_path.dentry->d_inode;
 	int level;
+	wait_queue_head_t *wq = ocfs2_ioend_wq(inode);
 
 	/* this io's submitter should not have unlocked this before we could */
 	BUG_ON(!ocfs2_iocb_is_rw_locked(iocb));
@@ -570,6 +579,15 @@
 	if (ocfs2_iocb_is_sem_locked(iocb))
 		ocfs2_iocb_clear_sem_locked(iocb);
 
+	if (ocfs2_iocb_is_unaligned_aio(iocb)) {
+		ocfs2_iocb_clear_unaligned_aio(iocb);
+
+		if (atomic_dec_and_test(&OCFS2_I(inode)->ip_unaligned_aio) &&
+		    waitqueue_active(wq)) {
+			wake_up_all(wq);
+		}
+	}
+
 	ocfs2_iocb_clear_rw_locked(iocb);
 
 	level = ocfs2_iocb_rw_locked_level(iocb);
@@ -863,6 +881,12 @@
 	struct page			*w_target_page;
 
 	/*
+	 * w_target_locked is used for page_mkwrite path indicating no unlocking
+	 * against w_target_page in ocfs2_write_end_nolock.
+	 */
+	unsigned int			w_target_locked:1;
+
+	/*
 	 * ocfs2_write_end() uses this to know what the real range to
 	 * write in the target should be.
 	 */
@@ -895,6 +919,24 @@
 
 static void ocfs2_free_write_ctxt(struct ocfs2_write_ctxt *wc)
 {
+	int i;
+
+	/*
+	 * w_target_locked is only set to true in the page_mkwrite() case.
+	 * The intent is to allow us to lock the target page from write_begin()
+	 * to write_end(). The caller must hold a ref on w_target_page.
+	 */
+	if (wc->w_target_locked) {
+		BUG_ON(!wc->w_target_page);
+		for (i = 0; i < wc->w_num_pages; i++) {
+			if (wc->w_target_page == wc->w_pages[i]) {
+				wc->w_pages[i] = NULL;
+				break;
+			}
+		}
+		mark_page_accessed(wc->w_target_page);
+		page_cache_release(wc->w_target_page);
+	}
 	ocfs2_unlock_and_free_pages(wc->w_pages, wc->w_num_pages);
 
 	brelse(wc->w_di_bh);
@@ -1132,20 +1174,17 @@
 			 */
 			lock_page(mmap_page);
 
+			/* Exit and let the caller retry */
 			if (mmap_page->mapping != mapping) {
+				WARN_ON(mmap_page->mapping);
 				unlock_page(mmap_page);
-				/*
-				 * Sanity check - the locking in
-				 * ocfs2_pagemkwrite() should ensure
-				 * that this code doesn't trigger.
-				 */
-				ret = -EINVAL;
-				mlog_errno(ret);
+				ret = -EAGAIN;
 				goto out;
 			}
 
 			page_cache_get(mmap_page);
 			wc->w_pages[i] = mmap_page;
+			wc->w_target_locked = true;
 		} else {
 			wc->w_pages[i] = find_or_create_page(mapping, index,
 							     GFP_NOFS);
@@ -1160,6 +1199,8 @@
 			wc->w_target_page = wc->w_pages[i];
 	}
 out:
+	if (ret)
+		wc->w_target_locked = false;
 	return ret;
 }
 
@@ -1817,11 +1858,23 @@
 	 */
 	ret = ocfs2_grab_pages_for_write(mapping, wc, wc->w_cpos, pos, len,
 					 cluster_of_pages, mmap_page);
-	if (ret) {
+	if (ret && ret != -EAGAIN) {
 		mlog_errno(ret);
 		goto out_quota;
 	}
 
+	/*
+	 * ocfs2_grab_pages_for_write() returns -EAGAIN if it could not lock
+	 * the target page. In this case, we exit with no error and no target
+	 * page. This will trigger the caller, page_mkwrite(), to re-try
+	 * the operation.
+	 */
+	if (ret == -EAGAIN) {
+		BUG_ON(wc->w_target_page);
+		ret = 0;
+		goto out_quota;
+	}
+
 	ret = ocfs2_write_cluster_by_desc(mapping, data_ac, meta_ac, wc, pos,
 					  len);
 	if (ret) {
diff --git a/fs/ocfs2/aops.h b/fs/ocfs2/aops.h
index 75cf3ad..ffb2da3 100644
--- a/fs/ocfs2/aops.h
+++ b/fs/ocfs2/aops.h
@@ -78,6 +78,7 @@
 	OCFS2_IOCB_RW_LOCK = 0,
 	OCFS2_IOCB_RW_LOCK_LEVEL,
 	OCFS2_IOCB_SEM,
+	OCFS2_IOCB_UNALIGNED_IO,
 	OCFS2_IOCB_NUM_LOCKS
 };
 
@@ -91,4 +92,17 @@
 	clear_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private)
 #define ocfs2_iocb_is_sem_locked(iocb) \
 	test_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private)
+
+#define ocfs2_iocb_set_unaligned_aio(iocb) \
+	set_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private)
+#define ocfs2_iocb_clear_unaligned_aio(iocb) \
+	clear_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private)
+#define ocfs2_iocb_is_unaligned_aio(iocb) \
+	test_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private)
+
+#define OCFS2_IOEND_WQ_HASH_SZ	37
+#define ocfs2_ioend_wq(v)   (&ocfs2__ioend_wq[((unsigned long)(v)) %\
+					    OCFS2_IOEND_WQ_HASH_SZ])
+extern wait_queue_head_t ocfs2__ioend_wq[OCFS2_IOEND_WQ_HASH_SZ];
+
 #endif /* OCFS2_FILE_H */
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index 9a3e6bb..a4e855e 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -216,6 +216,7 @@
 
 	struct list_head	hr_all_item;
 	unsigned		hr_unclean_stop:1,
+				hr_aborted_start:1,
 				hr_item_pinned:1,
 				hr_item_dropped:1;
 
@@ -254,6 +255,10 @@
 	 * a more complete api that doesn't lead to this sort of fragility. */
 	atomic_t		hr_steady_iterations;
 
+	/* terminate o2hb thread if it does not reach steady state
+	 * (hr_steady_iterations == 0) within hr_unsteady_iterations */
+	atomic_t		hr_unsteady_iterations;
+
 	char			hr_dev_name[BDEVNAME_SIZE];
 
 	unsigned int		hr_timeout_ms;
@@ -324,6 +329,10 @@
 
 static void o2hb_arm_write_timeout(struct o2hb_region *reg)
 {
+	/* Arm writeout only after thread reaches steady state */
+	if (atomic_read(&reg->hr_steady_iterations) != 0)
+		return;
+
 	mlog(ML_HEARTBEAT, "Queue write timeout for %u ms\n",
 	     O2HB_MAX_WRITE_TIMEOUT_MS);
 
@@ -537,9 +546,14 @@
 	return read == computed;
 }
 
-/* We want to make sure that nobody is heartbeating on top of us --
- * this will help detect an invalid configuration. */
-static void o2hb_check_last_timestamp(struct o2hb_region *reg)
+/*
+ * Compare the slot data with what we wrote in the last iteration.
+ * If the match fails, print an appropriate error message. This is to
+ * detect errors like... another node hearting on the same slot,
+ * flaky device that is losing writes, etc.
+ * Returns 1 if check succeeds, 0 otherwise.
+ */
+static int o2hb_check_own_slot(struct o2hb_region *reg)
 {
 	struct o2hb_disk_slot *slot;
 	struct o2hb_disk_heartbeat_block *hb_block;
@@ -548,13 +562,13 @@
 	slot = &reg->hr_slots[o2nm_this_node()];
 	/* Don't check on our 1st timestamp */
 	if (!slot->ds_last_time)
-		return;
+		return 0;
 
 	hb_block = slot->ds_raw_block;
 	if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time &&
 	    le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation &&
 	    hb_block->hb_node == slot->ds_node_num)
-		return;
+		return 1;
 
 #define ERRSTR1		"Another node is heartbeating on device"
 #define ERRSTR2		"Heartbeat generation mismatch on device"
@@ -574,6 +588,8 @@
 	     (unsigned long long)slot->ds_last_time, hb_block->hb_node,
 	     (unsigned long long)le64_to_cpu(hb_block->hb_generation),
 	     (unsigned long long)le64_to_cpu(hb_block->hb_seq));
+
+	return 0;
 }
 
 static inline void o2hb_prepare_block(struct o2hb_region *reg,
@@ -719,17 +735,24 @@
 	o2nm_node_put(node);
 }
 
-static void o2hb_set_quorum_device(struct o2hb_region *reg,
-				   struct o2hb_disk_slot *slot)
+static void o2hb_set_quorum_device(struct o2hb_region *reg)
 {
-	assert_spin_locked(&o2hb_live_lock);
-
 	if (!o2hb_global_heartbeat_active())
 		return;
 
-	if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap))
+	/* Prevent race with o2hb_heartbeat_group_drop_item() */
+	if (kthread_should_stop())
 		return;
 
+	/* Tag region as quorum only after thread reaches steady state */
+	if (atomic_read(&reg->hr_steady_iterations) != 0)
+		return;
+
+	spin_lock(&o2hb_live_lock);
+
+	if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap))
+		goto unlock;
+
 	/*
 	 * A region can be added to the quorum only when it sees all
 	 * live nodes heartbeat on it. In other words, the region has been
@@ -737,13 +760,10 @@
 	 */
 	if (memcmp(reg->hr_live_node_bitmap, o2hb_live_node_bitmap,
 		   sizeof(o2hb_live_node_bitmap)))
-		return;
+		goto unlock;
 
-	if (slot->ds_changed_samples < O2HB_LIVE_THRESHOLD)
-		return;
-
-	printk(KERN_NOTICE "o2hb: Region %s is now a quorum device\n",
-	       config_item_name(&reg->hr_item));
+	printk(KERN_NOTICE "o2hb: Region %s (%s) is now a quorum device\n",
+	       config_item_name(&reg->hr_item), reg->hr_dev_name);
 
 	set_bit(reg->hr_region_num, o2hb_quorum_region_bitmap);
 
@@ -754,6 +774,8 @@
 	if (o2hb_pop_count(&o2hb_quorum_region_bitmap,
 			   O2NM_MAX_REGIONS) > O2HB_PIN_CUT_OFF)
 		o2hb_region_unpin(NULL);
+unlock:
+	spin_unlock(&o2hb_live_lock);
 }
 
 static int o2hb_check_slot(struct o2hb_region *reg,
@@ -925,8 +947,6 @@
 		slot->ds_equal_samples = 0;
 	}
 out:
-	o2hb_set_quorum_device(reg, slot);
-
 	spin_unlock(&o2hb_live_lock);
 
 	o2hb_run_event_list(&event);
@@ -957,7 +977,8 @@
 
 static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 {
-	int i, ret, highest_node, change = 0;
+	int i, ret, highest_node;
+	int membership_change = 0, own_slot_ok = 0;
 	unsigned long configured_nodes[BITS_TO_LONGS(O2NM_MAX_NODES)];
 	unsigned long live_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
 	struct o2hb_bio_wait_ctxt write_wc;
@@ -966,7 +987,7 @@
 				       sizeof(configured_nodes));
 	if (ret) {
 		mlog_errno(ret);
-		return ret;
+		goto bail;
 	}
 
 	/*
@@ -982,8 +1003,9 @@
 
 	highest_node = o2hb_highest_node(configured_nodes, O2NM_MAX_NODES);
 	if (highest_node >= O2NM_MAX_NODES) {
-		mlog(ML_NOTICE, "ocfs2_heartbeat: no configured nodes found!\n");
-		return -EINVAL;
+		mlog(ML_NOTICE, "o2hb: No configured nodes found!\n");
+		ret = -EINVAL;
+		goto bail;
 	}
 
 	/* No sense in reading the slots of nodes that don't exist
@@ -993,29 +1015,27 @@
 	ret = o2hb_read_slots(reg, highest_node + 1);
 	if (ret < 0) {
 		mlog_errno(ret);
-		return ret;
+		goto bail;
 	}
 
 	/* With an up to date view of the slots, we can check that no
 	 * other node has been improperly configured to heartbeat in
 	 * our slot. */
-	o2hb_check_last_timestamp(reg);
+	own_slot_ok = o2hb_check_own_slot(reg);
 
 	/* fill in the proper info for our next heartbeat */
 	o2hb_prepare_block(reg, reg->hr_generation);
 
-	/* And fire off the write. Note that we don't wait on this I/O
-	 * until later. */
 	ret = o2hb_issue_node_write(reg, &write_wc);
 	if (ret < 0) {
 		mlog_errno(ret);
-		return ret;
+		goto bail;
 	}
 
 	i = -1;
 	while((i = find_next_bit(configured_nodes,
 				 O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) {
-		change |= o2hb_check_slot(reg, &reg->hr_slots[i]);
+		membership_change |= o2hb_check_slot(reg, &reg->hr_slots[i]);
 	}
 
 	/*
@@ -1030,18 +1050,39 @@
 		 * disk */
 		mlog(ML_ERROR, "Write error %d on device \"%s\"\n",
 		     write_wc.wc_error, reg->hr_dev_name);
-		return write_wc.wc_error;
+		ret = write_wc.wc_error;
+		goto bail;
 	}
 
-	o2hb_arm_write_timeout(reg);
+	/* Skip disarming the timeout if own slot has stale/bad data */
+	if (own_slot_ok) {
+		o2hb_set_quorum_device(reg);
+		o2hb_arm_write_timeout(reg);
+	}
 
+bail:
 	/* let the person who launched us know when things are steady */
-	if (!change && (atomic_read(&reg->hr_steady_iterations) != 0)) {
-		if (atomic_dec_and_test(&reg->hr_steady_iterations))
-			wake_up(&o2hb_steady_queue);
+	if (atomic_read(&reg->hr_steady_iterations) != 0) {
+		if (!ret && own_slot_ok && !membership_change) {
+			if (atomic_dec_and_test(&reg->hr_steady_iterations))
+				wake_up(&o2hb_steady_queue);
+		}
 	}
 
-	return 0;
+	if (atomic_read(&reg->hr_steady_iterations) != 0) {
+		if (atomic_dec_and_test(&reg->hr_unsteady_iterations)) {
+			printk(KERN_NOTICE "o2hb: Unable to stabilize "
+			       "heartbeart on region %s (%s)\n",
+			       config_item_name(&reg->hr_item),
+			       reg->hr_dev_name);
+			atomic_set(&reg->hr_steady_iterations, 0);
+			reg->hr_aborted_start = 1;
+			wake_up(&o2hb_steady_queue);
+			ret = -EIO;
+		}
+	}
+
+	return ret;
 }
 
 /* Subtract b from a, storing the result in a. a *must* have a larger
@@ -1095,7 +1136,8 @@
 	/* Pin node */
 	o2nm_depend_this_node();
 
-	while (!kthread_should_stop() && !reg->hr_unclean_stop) {
+	while (!kthread_should_stop() &&
+	       !reg->hr_unclean_stop && !reg->hr_aborted_start) {
 		/* We track the time spent inside
 		 * o2hb_do_disk_heartbeat so that we avoid more than
 		 * hr_timeout_ms between disk writes. On busy systems
@@ -1103,10 +1145,7 @@
 		 * likely to time itself out. */
 		do_gettimeofday(&before_hb);
 
-		i = 0;
-		do {
-			ret = o2hb_do_disk_heartbeat(reg);
-		} while (ret && ++i < 2);
+		ret = o2hb_do_disk_heartbeat(reg);
 
 		do_gettimeofday(&after_hb);
 		elapsed_msec = o2hb_elapsed_msecs(&before_hb, &after_hb);
@@ -1117,7 +1156,8 @@
 		     after_hb.tv_sec, (unsigned long) after_hb.tv_usec,
 		     elapsed_msec);
 
-		if (elapsed_msec < reg->hr_timeout_ms) {
+		if (!kthread_should_stop() &&
+		    elapsed_msec < reg->hr_timeout_ms) {
 			/* the kthread api has blocked signals for us so no
 			 * need to record the return value. */
 			msleep_interruptible(reg->hr_timeout_ms - elapsed_msec);
@@ -1134,20 +1174,20 @@
 	 * to timeout on this region when we could just as easily
 	 * write a clear generation - thus indicating to them that
 	 * this node has left this region.
-	 *
-	 * XXX: Should we skip this on unclean_stop? */
-	o2hb_prepare_block(reg, 0);
-	ret = o2hb_issue_node_write(reg, &write_wc);
-	if (ret == 0) {
-		o2hb_wait_on_io(reg, &write_wc);
-	} else {
-		mlog_errno(ret);
+	 */
+	if (!reg->hr_unclean_stop && !reg->hr_aborted_start) {
+		o2hb_prepare_block(reg, 0);
+		ret = o2hb_issue_node_write(reg, &write_wc);
+		if (ret == 0)
+			o2hb_wait_on_io(reg, &write_wc);
+		else
+			mlog_errno(ret);
 	}
 
 	/* Unpin node */
 	o2nm_undepend_this_node();
 
-	mlog(ML_HEARTBEAT|ML_KTHREAD, "hb thread exiting\n");
+	mlog(ML_HEARTBEAT|ML_KTHREAD, "o2hb thread exiting\n");
 
 	return 0;
 }
@@ -1158,6 +1198,7 @@
 	struct o2hb_debug_buf *db = inode->i_private;
 	struct o2hb_region *reg;
 	unsigned long map[BITS_TO_LONGS(O2NM_MAX_NODES)];
+	unsigned long lts;
 	char *buf = NULL;
 	int i = -1;
 	int out = 0;
@@ -1194,9 +1235,11 @@
 
 	case O2HB_DB_TYPE_REGION_ELAPSED_TIME:
 		reg = (struct o2hb_region *)db->db_data;
-		out += snprintf(buf + out, PAGE_SIZE - out, "%u\n",
-				jiffies_to_msecs(jiffies -
-						 reg->hr_last_timeout_start));
+		lts = reg->hr_last_timeout_start;
+		/* If 0, it has never been set before */
+		if (lts)
+			lts = jiffies_to_msecs(jiffies - lts);
+		out += snprintf(buf + out, PAGE_SIZE - out, "%lu\n", lts);
 		goto done;
 
 	case O2HB_DB_TYPE_REGION_PINNED:
@@ -1426,6 +1469,8 @@
 	struct page *page;
 	struct o2hb_region *reg = to_o2hb_region(item);
 
+	mlog(ML_HEARTBEAT, "hb region release (%s)\n", reg->hr_dev_name);
+
 	if (reg->hr_tmp_block)
 		kfree(reg->hr_tmp_block);
 
@@ -1792,7 +1837,10 @@
 			live_threshold <<= 1;
 		spin_unlock(&o2hb_live_lock);
 	}
-	atomic_set(&reg->hr_steady_iterations, live_threshold + 1);
+	++live_threshold;
+	atomic_set(&reg->hr_steady_iterations, live_threshold);
+	/* unsteady_iterations is double the steady_iterations */
+	atomic_set(&reg->hr_unsteady_iterations, (live_threshold << 1));
 
 	hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s",
 			      reg->hr_item.ci_name);
@@ -1809,14 +1857,12 @@
 	ret = wait_event_interruptible(o2hb_steady_queue,
 				atomic_read(&reg->hr_steady_iterations) == 0);
 	if (ret) {
-		/* We got interrupted (hello ptrace!).  Clean up */
-		spin_lock(&o2hb_live_lock);
-		hb_task = reg->hr_task;
-		reg->hr_task = NULL;
-		spin_unlock(&o2hb_live_lock);
+		atomic_set(&reg->hr_steady_iterations, 0);
+		reg->hr_aborted_start = 1;
+	}
 
-		if (hb_task)
-			kthread_stop(hb_task);
+	if (reg->hr_aborted_start) {
+		ret = -EIO;
 		goto out;
 	}
 
@@ -1833,8 +1879,8 @@
 		ret = -EIO;
 
 	if (hb_task && o2hb_global_heartbeat_active())
-		printk(KERN_NOTICE "o2hb: Heartbeat started on region %s\n",
-		       config_item_name(&reg->hr_item));
+		printk(KERN_NOTICE "o2hb: Heartbeat started on region %s (%s)\n",
+		       config_item_name(&reg->hr_item), reg->hr_dev_name);
 
 out:
 	if (filp)
@@ -2092,13 +2138,6 @@
 
 	/* stop the thread when the user removes the region dir */
 	spin_lock(&o2hb_live_lock);
-	if (o2hb_global_heartbeat_active()) {
-		clear_bit(reg->hr_region_num, o2hb_region_bitmap);
-		clear_bit(reg->hr_region_num, o2hb_live_region_bitmap);
-		if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap))
-			quorum_region = 1;
-		clear_bit(reg->hr_region_num, o2hb_quorum_region_bitmap);
-	}
 	hb_task = reg->hr_task;
 	reg->hr_task = NULL;
 	reg->hr_item_dropped = 1;
@@ -2107,19 +2146,30 @@
 	if (hb_task)
 		kthread_stop(hb_task);
 
+	if (o2hb_global_heartbeat_active()) {
+		spin_lock(&o2hb_live_lock);
+		clear_bit(reg->hr_region_num, o2hb_region_bitmap);
+		clear_bit(reg->hr_region_num, o2hb_live_region_bitmap);
+		if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap))
+			quorum_region = 1;
+		clear_bit(reg->hr_region_num, o2hb_quorum_region_bitmap);
+		spin_unlock(&o2hb_live_lock);
+		printk(KERN_NOTICE "o2hb: Heartbeat %s on region %s (%s)\n",
+		       ((atomic_read(&reg->hr_steady_iterations) == 0) ?
+			"stopped" : "start aborted"), config_item_name(item),
+		       reg->hr_dev_name);
+	}
+
 	/*
 	 * If we're racing a dev_write(), we need to wake them.  They will
 	 * check reg->hr_task
 	 */
 	if (atomic_read(&reg->hr_steady_iterations) != 0) {
+		reg->hr_aborted_start = 1;
 		atomic_set(&reg->hr_steady_iterations, 0);
 		wake_up(&o2hb_steady_queue);
 	}
 
-	if (o2hb_global_heartbeat_active())
-		printk(KERN_NOTICE "o2hb: Heartbeat stopped on region %s\n",
-		       config_item_name(&reg->hr_item));
-
 	config_item_put(item);
 
 	if (!o2hb_global_heartbeat_active() || !quorum_region)
diff --git a/fs/ocfs2/cluster/netdebug.c b/fs/ocfs2/cluster/netdebug.c
index 3a583590..dc45deb 100644
--- a/fs/ocfs2/cluster/netdebug.c
+++ b/fs/ocfs2/cluster/netdebug.c
@@ -47,6 +47,7 @@
 #define SC_DEBUG_NAME		"sock_containers"
 #define NST_DEBUG_NAME		"send_tracking"
 #define STATS_DEBUG_NAME	"stats"
+#define NODES_DEBUG_NAME	"connected_nodes"
 
 #define SHOW_SOCK_CONTAINERS	0
 #define SHOW_SOCK_STATS		1
@@ -55,6 +56,7 @@
 static struct dentry *sc_dentry;
 static struct dentry *nst_dentry;
 static struct dentry *stats_dentry;
+static struct dentry *nodes_dentry;
 
 static DEFINE_SPINLOCK(o2net_debug_lock);
 
@@ -491,53 +493,87 @@
 	.release = sc_fop_release,
 };
 
-int o2net_debugfs_init(void)
+static int o2net_fill_bitmap(char *buf, int len)
 {
-	o2net_dentry = debugfs_create_dir(O2NET_DEBUG_DIR, NULL);
-	if (!o2net_dentry) {
-		mlog_errno(-ENOMEM);
-		goto bail;
-	}
+	unsigned long map[BITS_TO_LONGS(O2NM_MAX_NODES)];
+	int i = -1, out = 0;
 
-	nst_dentry = debugfs_create_file(NST_DEBUG_NAME, S_IFREG|S_IRUSR,
-					 o2net_dentry, NULL,
-					 &nst_seq_fops);
-	if (!nst_dentry) {
-		mlog_errno(-ENOMEM);
-		goto bail;
-	}
+	o2net_fill_node_map(map, sizeof(map));
 
-	sc_dentry = debugfs_create_file(SC_DEBUG_NAME, S_IFREG|S_IRUSR,
-					o2net_dentry, NULL,
-					&sc_seq_fops);
-	if (!sc_dentry) {
-		mlog_errno(-ENOMEM);
-		goto bail;
-	}
+	while ((i = find_next_bit(map, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES)
+		out += snprintf(buf + out, PAGE_SIZE - out, "%d ", i);
+	out += snprintf(buf + out, PAGE_SIZE - out, "\n");
 
-	stats_dentry = debugfs_create_file(STATS_DEBUG_NAME, S_IFREG|S_IRUSR,
-					   o2net_dentry, NULL,
-					   &stats_seq_fops);
-	if (!stats_dentry) {
-		mlog_errno(-ENOMEM);
-		goto bail;
-	}
+	return out;
+}
+
+static int nodes_fop_open(struct inode *inode, struct file *file)
+{
+	char *buf;
+
+	buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	i_size_write(inode, o2net_fill_bitmap(buf, PAGE_SIZE));
+
+	file->private_data = buf;
 
 	return 0;
-bail:
-	debugfs_remove(stats_dentry);
-	debugfs_remove(sc_dentry);
-	debugfs_remove(nst_dentry);
-	debugfs_remove(o2net_dentry);
-	return -ENOMEM;
 }
 
+static int o2net_debug_release(struct inode *inode, struct file *file)
+{
+	kfree(file->private_data);
+	return 0;
+}
+
+static ssize_t o2net_debug_read(struct file *file, char __user *buf,
+				size_t nbytes, loff_t *ppos)
+{
+	return simple_read_from_buffer(buf, nbytes, ppos, file->private_data,
+				       i_size_read(file->f_mapping->host));
+}
+
+static const struct file_operations nodes_fops = {
+	.open		= nodes_fop_open,
+	.release	= o2net_debug_release,
+	.read		= o2net_debug_read,
+	.llseek		= generic_file_llseek,
+};
+
 void o2net_debugfs_exit(void)
 {
+	debugfs_remove(nodes_dentry);
 	debugfs_remove(stats_dentry);
 	debugfs_remove(sc_dentry);
 	debugfs_remove(nst_dentry);
 	debugfs_remove(o2net_dentry);
 }
 
+int o2net_debugfs_init(void)
+{
+	mode_t mode = S_IFREG|S_IRUSR;
+
+	o2net_dentry = debugfs_create_dir(O2NET_DEBUG_DIR, NULL);
+	if (o2net_dentry)
+		nst_dentry = debugfs_create_file(NST_DEBUG_NAME, mode,
+					o2net_dentry, NULL, &nst_seq_fops);
+	if (nst_dentry)
+		sc_dentry = debugfs_create_file(SC_DEBUG_NAME, mode,
+					o2net_dentry, NULL, &sc_seq_fops);
+	if (sc_dentry)
+		stats_dentry = debugfs_create_file(STATS_DEBUG_NAME, mode,
+					o2net_dentry, NULL, &stats_seq_fops);
+	if (stats_dentry)
+		nodes_dentry = debugfs_create_file(NODES_DEBUG_NAME, mode,
+					o2net_dentry, NULL, &nodes_fops);
+	if (nodes_dentry)
+		return 0;
+
+	o2net_debugfs_exit();
+	mlog_errno(-ENOMEM);
+	return -ENOMEM;
+}
+
 #endif	/* CONFIG_DEBUG_FS */
diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c
index ad7d0c1..044e7b5 100644
--- a/fs/ocfs2/cluster/tcp.c
+++ b/fs/ocfs2/cluster/tcp.c
@@ -546,7 +546,7 @@
 	}
 
 	if (was_valid && !valid) {
-		printk(KERN_NOTICE "o2net: no longer connected to "
+		printk(KERN_NOTICE "o2net: No longer connected to "
 		       SC_NODEF_FMT "\n", SC_NODEF_ARGS(old_sc));
 		o2net_complete_nodes_nsw(nn);
 	}
@@ -556,7 +556,7 @@
 		cancel_delayed_work(&nn->nn_connect_expired);
 		printk(KERN_NOTICE "o2net: %s " SC_NODEF_FMT "\n",
 		       o2nm_this_node() > sc->sc_node->nd_num ?
-		       		"connected to" : "accepted connection from",
+		       "Connected to" : "Accepted connection from",
 		       SC_NODEF_ARGS(sc));
 	}
 
@@ -644,7 +644,7 @@
 			o2net_sc_queue_work(sc, &sc->sc_connect_work);
 			break;
 		default:
-			printk(KERN_INFO "o2net: connection to " SC_NODEF_FMT
+			printk(KERN_INFO "o2net: Connection to " SC_NODEF_FMT
 			      " shutdown, state %d\n",
 			      SC_NODEF_ARGS(sc), sk->sk_state);
 			o2net_sc_queue_work(sc, &sc->sc_shutdown_work);
@@ -1035,6 +1035,25 @@
 	return ret;
 }
 
+/* Get a map of all nodes to which this node is currently connected to */
+void o2net_fill_node_map(unsigned long *map, unsigned bytes)
+{
+	struct o2net_sock_container *sc;
+	int node, ret;
+
+	BUG_ON(bytes < (BITS_TO_LONGS(O2NM_MAX_NODES) * sizeof(unsigned long)));
+
+	memset(map, 0, bytes);
+	for (node = 0; node < O2NM_MAX_NODES; ++node) {
+		o2net_tx_can_proceed(o2net_nn_from_num(node), &sc, &ret);
+		if (!ret) {
+			set_bit(node, map);
+			sc_put(sc);
+		}
+	}
+}
+EXPORT_SYMBOL_GPL(o2net_fill_node_map);
+
 int o2net_send_message_vec(u32 msg_type, u32 key, struct kvec *caller_vec,
 			   size_t caller_veclen, u8 target_node, int *status)
 {
@@ -1285,11 +1304,11 @@
 	struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num);
 
 	if (hand->protocol_version != cpu_to_be64(O2NET_PROTOCOL_VERSION)) {
-		mlog(ML_NOTICE, SC_NODEF_FMT " advertised net protocol "
-		     "version %llu but %llu is required, disconnecting\n",
-		     SC_NODEF_ARGS(sc),
-		     (unsigned long long)be64_to_cpu(hand->protocol_version),
-		     O2NET_PROTOCOL_VERSION);
+		printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " Advertised net "
+		       "protocol version %llu but %llu is required. "
+		       "Disconnecting.\n", SC_NODEF_ARGS(sc),
+		       (unsigned long long)be64_to_cpu(hand->protocol_version),
+		       O2NET_PROTOCOL_VERSION);
 
 		/* don't bother reconnecting if its the wrong version. */
 		o2net_ensure_shutdown(nn, sc, -ENOTCONN);
@@ -1303,33 +1322,33 @@
 	 */
 	if (be32_to_cpu(hand->o2net_idle_timeout_ms) !=
 				o2net_idle_timeout()) {
-		mlog(ML_NOTICE, SC_NODEF_FMT " uses a network idle timeout of "
-		     "%u ms, but we use %u ms locally.  disconnecting\n",
-		     SC_NODEF_ARGS(sc),
-		     be32_to_cpu(hand->o2net_idle_timeout_ms),
-		     o2net_idle_timeout());
+		printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a network "
+		       "idle timeout of %u ms, but we use %u ms locally. "
+		       "Disconnecting.\n", SC_NODEF_ARGS(sc),
+		       be32_to_cpu(hand->o2net_idle_timeout_ms),
+		       o2net_idle_timeout());
 		o2net_ensure_shutdown(nn, sc, -ENOTCONN);
 		return -1;
 	}
 
 	if (be32_to_cpu(hand->o2net_keepalive_delay_ms) !=
 			o2net_keepalive_delay()) {
-		mlog(ML_NOTICE, SC_NODEF_FMT " uses a keepalive delay of "
-		     "%u ms, but we use %u ms locally.  disconnecting\n",
-		     SC_NODEF_ARGS(sc),
-		     be32_to_cpu(hand->o2net_keepalive_delay_ms),
-		     o2net_keepalive_delay());
+		printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a keepalive "
+		       "delay of %u ms, but we use %u ms locally. "
+		       "Disconnecting.\n", SC_NODEF_ARGS(sc),
+		       be32_to_cpu(hand->o2net_keepalive_delay_ms),
+		       o2net_keepalive_delay());
 		o2net_ensure_shutdown(nn, sc, -ENOTCONN);
 		return -1;
 	}
 
 	if (be32_to_cpu(hand->o2hb_heartbeat_timeout_ms) !=
 			O2HB_MAX_WRITE_TIMEOUT_MS) {
-		mlog(ML_NOTICE, SC_NODEF_FMT " uses a heartbeat timeout of "
-		     "%u ms, but we use %u ms locally.  disconnecting\n",
-		     SC_NODEF_ARGS(sc),
-		     be32_to_cpu(hand->o2hb_heartbeat_timeout_ms),
-		     O2HB_MAX_WRITE_TIMEOUT_MS);
+		printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a heartbeat "
+		       "timeout of %u ms, but we use %u ms locally. "
+		       "Disconnecting.\n", SC_NODEF_ARGS(sc),
+		       be32_to_cpu(hand->o2hb_heartbeat_timeout_ms),
+		       O2HB_MAX_WRITE_TIMEOUT_MS);
 		o2net_ensure_shutdown(nn, sc, -ENOTCONN);
 		return -1;
 	}
@@ -1540,28 +1559,16 @@
 {
 	struct o2net_sock_container *sc = (struct o2net_sock_container *)data;
 	struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num);
-
 #ifdef CONFIG_DEBUG_FS
-	ktime_t now = ktime_get();
+	unsigned long msecs = ktime_to_ms(ktime_get()) -
+		ktime_to_ms(sc->sc_tv_timer);
+#else
+	unsigned long msecs = o2net_idle_timeout();
 #endif
 
-	printk(KERN_NOTICE "o2net: connection to " SC_NODEF_FMT " has been idle for %u.%u "
-	     "seconds, shutting it down.\n", SC_NODEF_ARGS(sc),
-		     o2net_idle_timeout() / 1000,
-		     o2net_idle_timeout() % 1000);
-
-#ifdef CONFIG_DEBUG_FS
-	mlog(ML_NOTICE, "Here are some times that might help debug the "
-	     "situation: (Timer: %lld, Now %lld, DataReady %lld, Advance %lld-%lld, "
-	     "Key 0x%08x, Func %u, FuncTime %lld-%lld)\n",
-	     (long long)ktime_to_us(sc->sc_tv_timer), (long long)ktime_to_us(now),
-	     (long long)ktime_to_us(sc->sc_tv_data_ready),
-	     (long long)ktime_to_us(sc->sc_tv_advance_start),
-	     (long long)ktime_to_us(sc->sc_tv_advance_stop),
-	     sc->sc_msg_key, sc->sc_msg_type,
-	     (long long)ktime_to_us(sc->sc_tv_func_start),
-	     (long long)ktime_to_us(sc->sc_tv_func_stop));
-#endif
+	printk(KERN_NOTICE "o2net: Connection to " SC_NODEF_FMT " has been "
+	       "idle for %lu.%lu secs, shutting it down.\n", SC_NODEF_ARGS(sc),
+	       msecs / 1000, msecs % 1000);
 
 	/*
 	 * Initialize the nn_timeout so that the next connection attempt
@@ -1694,8 +1701,8 @@
 
 out:
 	if (ret) {
-		mlog(ML_NOTICE, "connect attempt to " SC_NODEF_FMT " failed "
-		     "with errno %d\n", SC_NODEF_ARGS(sc), ret);
+		printk(KERN_NOTICE "o2net: Connect attempt to " SC_NODEF_FMT
+		       " failed with errno %d\n", SC_NODEF_ARGS(sc), ret);
 		/* 0 err so that another will be queued and attempted
 		 * from set_nn_state */
 		if (sc)
@@ -1718,8 +1725,8 @@
 
 	spin_lock(&nn->nn_lock);
 	if (!nn->nn_sc_valid) {
-		mlog(ML_ERROR, "no connection established with node %u after "
-		     "%u.%u seconds, giving up and returning errors.\n",
+		printk(KERN_NOTICE "o2net: No connection established with "
+		       "node %u after %u.%u seconds, giving up.\n",
 		     o2net_num_from_nn(nn),
 		     o2net_idle_timeout() / 1000,
 		     o2net_idle_timeout() % 1000);
@@ -1862,21 +1869,21 @@
 
 	node = o2nm_get_node_by_ip(sin.sin_addr.s_addr);
 	if (node == NULL) {
-		mlog(ML_NOTICE, "attempt to connect from unknown node at %pI4:%d\n",
-		     &sin.sin_addr.s_addr, ntohs(sin.sin_port));
+		printk(KERN_NOTICE "o2net: Attempt to connect from unknown "
+		       "node at %pI4:%d\n", &sin.sin_addr.s_addr,
+		       ntohs(sin.sin_port));
 		ret = -EINVAL;
 		goto out;
 	}
 
 	if (o2nm_this_node() >= node->nd_num) {
 		local_node = o2nm_get_node_by_num(o2nm_this_node());
-		mlog(ML_NOTICE, "unexpected connect attempt seen at node '%s' ("
-		     "%u, %pI4:%d) from node '%s' (%u, %pI4:%d)\n",
-		     local_node->nd_name, local_node->nd_num,
-		     &(local_node->nd_ipv4_address),
-		     ntohs(local_node->nd_ipv4_port),
-		     node->nd_name, node->nd_num, &sin.sin_addr.s_addr,
-		     ntohs(sin.sin_port));
+		printk(KERN_NOTICE "o2net: Unexpected connect attempt seen "
+		       "at node '%s' (%u, %pI4:%d) from node '%s' (%u, "
+		       "%pI4:%d)\n", local_node->nd_name, local_node->nd_num,
+		       &(local_node->nd_ipv4_address),
+		       ntohs(local_node->nd_ipv4_port), node->nd_name,
+		       node->nd_num, &sin.sin_addr.s_addr, ntohs(sin.sin_port));
 		ret = -EINVAL;
 		goto out;
 	}
@@ -1901,10 +1908,10 @@
 		ret = 0;
 	spin_unlock(&nn->nn_lock);
 	if (ret) {
-		mlog(ML_NOTICE, "attempt to connect from node '%s' at "
-		     "%pI4:%d but it already has an open connection\n",
-		     node->nd_name, &sin.sin_addr.s_addr,
-		     ntohs(sin.sin_port));
+		printk(KERN_NOTICE "o2net: Attempt to connect from node '%s' "
+		       "at %pI4:%d but it already has an open connection\n",
+		       node->nd_name, &sin.sin_addr.s_addr,
+		       ntohs(sin.sin_port));
 		goto out;
 	}
 
@@ -1984,7 +1991,7 @@
 
 	ret = sock_create(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock);
 	if (ret < 0) {
-		mlog(ML_ERROR, "unable to create socket, ret=%d\n", ret);
+		printk(KERN_ERR "o2net: Error %d while creating socket\n", ret);
 		goto out;
 	}
 
@@ -2001,16 +2008,15 @@
 	sock->sk->sk_reuse = 1;
 	ret = sock->ops->bind(sock, (struct sockaddr *)&sin, sizeof(sin));
 	if (ret < 0) {
-		mlog(ML_ERROR, "unable to bind socket at %pI4:%u, "
-		     "ret=%d\n", &addr, ntohs(port), ret);
+		printk(KERN_ERR "o2net: Error %d while binding socket at "
+		       "%pI4:%u\n", ret, &addr, ntohs(port)); 
 		goto out;
 	}
 
 	ret = sock->ops->listen(sock, 64);
-	if (ret < 0) {
-		mlog(ML_ERROR, "unable to listen on %pI4:%u, ret=%d\n",
-		     &addr, ntohs(port), ret);
-	}
+	if (ret < 0)
+		printk(KERN_ERR "o2net: Error %d while listening on %pI4:%u\n",
+		       ret, &addr, ntohs(port));
 
 out:
 	if (ret) {
diff --git a/fs/ocfs2/cluster/tcp.h b/fs/ocfs2/cluster/tcp.h
index fd6179e..5bada2a 100644
--- a/fs/ocfs2/cluster/tcp.h
+++ b/fs/ocfs2/cluster/tcp.h
@@ -106,6 +106,8 @@
 			   struct list_head *unreg_list);
 void o2net_unregister_handler_list(struct list_head *list);
 
+void o2net_fill_node_map(unsigned long *map, unsigned bytes);
+
 struct o2nm_node;
 int o2net_register_hb_callbacks(void);
 void o2net_unregister_hb_callbacks(void);
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index e2878b5..8fe4e28 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -1184,8 +1184,7 @@
 			if (pde)
 				le16_add_cpu(&pde->rec_len,
 						le16_to_cpu(de->rec_len));
-			else
-				de->inode = 0;
+			de->inode = 0;
 			dir->i_version++;
 			ocfs2_journal_dirty(handle, bh);
 			goto bail;
diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h
index d602abb..a5952ce 100644
--- a/fs/ocfs2/dlm/dlmcommon.h
+++ b/fs/ocfs2/dlm/dlmcommon.h
@@ -859,8 +859,8 @@
 void dlm_wait_for_recovery(struct dlm_ctxt *dlm);
 void dlm_kick_recovery_thread(struct dlm_ctxt *dlm);
 int dlm_is_node_dead(struct dlm_ctxt *dlm, u8 node);
-int dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout);
-int dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout);
+void dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout);
+void dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout);
 
 void dlm_put(struct dlm_ctxt *dlm);
 struct dlm_ctxt *dlm_grab(struct dlm_ctxt *dlm);
@@ -877,9 +877,8 @@
 	kref_get(&res->refs);
 }
 void dlm_lockres_put(struct dlm_lock_resource *res);
-void __dlm_unhash_lockres(struct dlm_lock_resource *res);
-void __dlm_insert_lockres(struct dlm_ctxt *dlm,
-			  struct dlm_lock_resource *res);
+void __dlm_unhash_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res);
+void __dlm_insert_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res);
 struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm,
 						     const char *name,
 						     unsigned int len,
@@ -902,46 +901,15 @@
 					  const char *name,
 					  unsigned int namelen);
 
-#define dlm_lockres_set_refmap_bit(bit,res)  \
-	__dlm_lockres_set_refmap_bit(bit,res,__FILE__,__LINE__)
-#define dlm_lockres_clear_refmap_bit(bit,res)  \
-	__dlm_lockres_clear_refmap_bit(bit,res,__FILE__,__LINE__)
+void dlm_lockres_set_refmap_bit(struct dlm_ctxt *dlm,
+				struct dlm_lock_resource *res, int bit);
+void dlm_lockres_clear_refmap_bit(struct dlm_ctxt *dlm,
+				  struct dlm_lock_resource *res, int bit);
 
-static inline void __dlm_lockres_set_refmap_bit(int bit,
-						struct dlm_lock_resource *res,
-						const char *file,
-						int line)
-{
-	//printk("%s:%d:%.*s: setting bit %d\n", file, line,
-	//     res->lockname.len, res->lockname.name, bit);
-	set_bit(bit, res->refmap);
-}
-
-static inline void __dlm_lockres_clear_refmap_bit(int bit,
-						  struct dlm_lock_resource *res,
-						  const char *file,
-						  int line)
-{
-	//printk("%s:%d:%.*s: clearing bit %d\n", file, line,
-	//     res->lockname.len, res->lockname.name, bit);
-	clear_bit(bit, res->refmap);
-}
-
-void __dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm,
-				   struct dlm_lock_resource *res,
-				   const char *file,
-				   int line);
-void __dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm,
-				   struct dlm_lock_resource *res,
-				   int new_lockres,
-				   const char *file,
-				   int line);
-#define dlm_lockres_drop_inflight_ref(d,r)  \
-	__dlm_lockres_drop_inflight_ref(d,r,__FILE__,__LINE__)
-#define dlm_lockres_grab_inflight_ref(d,r)  \
-	__dlm_lockres_grab_inflight_ref(d,r,0,__FILE__,__LINE__)
-#define dlm_lockres_grab_inflight_ref_new(d,r)  \
-	__dlm_lockres_grab_inflight_ref(d,r,1,__FILE__,__LINE__)
+void dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm,
+				   struct dlm_lock_resource *res);
+void dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm,
+				   struct dlm_lock_resource *res);
 
 void dlm_queue_ast(struct dlm_ctxt *dlm, struct dlm_lock *lock);
 void dlm_queue_bast(struct dlm_ctxt *dlm, struct dlm_lock *lock);
diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index 6ed6b95..92f2ead 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -157,16 +157,18 @@
 
 static void dlm_unregister_domain_handlers(struct dlm_ctxt *dlm);
 
-void __dlm_unhash_lockres(struct dlm_lock_resource *lockres)
+void __dlm_unhash_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res)
 {
-	if (!hlist_unhashed(&lockres->hash_node)) {
-		hlist_del_init(&lockres->hash_node);
-		dlm_lockres_put(lockres);
-	}
+	if (hlist_unhashed(&res->hash_node))
+		return;
+
+	mlog(0, "%s: Unhash res %.*s\n", dlm->name, res->lockname.len,
+	     res->lockname.name);
+	hlist_del_init(&res->hash_node);
+	dlm_lockres_put(res);
 }
 
-void __dlm_insert_lockres(struct dlm_ctxt *dlm,
-		       struct dlm_lock_resource *res)
+void __dlm_insert_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res)
 {
 	struct hlist_head *bucket;
 	struct qstr *q;
@@ -180,6 +182,9 @@
 	dlm_lockres_get(res);
 
 	hlist_add_head(&res->hash_node, bucket);
+
+	mlog(0, "%s: Hash res %.*s\n", dlm->name, res->lockname.len,
+	     res->lockname.name);
 }
 
 struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm,
@@ -539,17 +544,17 @@
 
 static void __dlm_print_nodes(struct dlm_ctxt *dlm)
 {
-	int node = -1;
+	int node = -1, num = 0;
 
 	assert_spin_locked(&dlm->spinlock);
 
-	printk(KERN_NOTICE "o2dlm: Nodes in domain %s: ", dlm->name);
-
+	printk("( ");
 	while ((node = find_next_bit(dlm->domain_map, O2NM_MAX_NODES,
 				     node + 1)) < O2NM_MAX_NODES) {
 		printk("%d ", node);
+		++num;
 	}
-	printk("\n");
+	printk(") %u nodes\n", num);
 }
 
 static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data,
@@ -566,11 +571,10 @@
 
 	node = exit_msg->node_idx;
 
-	printk(KERN_NOTICE "o2dlm: Node %u leaves domain %s\n", node, dlm->name);
-
 	spin_lock(&dlm->spinlock);
 	clear_bit(node, dlm->domain_map);
 	clear_bit(node, dlm->exit_domain_map);
+	printk(KERN_NOTICE "o2dlm: Node %u leaves domain %s ", node, dlm->name);
 	__dlm_print_nodes(dlm);
 
 	/* notify anything attached to the heartbeat events */
@@ -755,6 +759,7 @@
 
 		dlm_mark_domain_leaving(dlm);
 		dlm_leave_domain(dlm);
+		printk(KERN_NOTICE "o2dlm: Leaving domain %s\n", dlm->name);
 		dlm_force_free_mles(dlm);
 		dlm_complete_dlm_shutdown(dlm);
 	}
@@ -970,7 +975,7 @@
 		clear_bit(assert->node_idx, dlm->exit_domain_map);
 		__dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN);
 
-		printk(KERN_NOTICE "o2dlm: Node %u joins domain %s\n",
+		printk(KERN_NOTICE "o2dlm: Node %u joins domain %s ",
 		       assert->node_idx, dlm->name);
 		__dlm_print_nodes(dlm);
 
@@ -1701,8 +1706,10 @@
 bail:
 	spin_lock(&dlm->spinlock);
 	__dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN);
-	if (!status)
+	if (!status) {
+		printk(KERN_NOTICE "o2dlm: Joining domain %s ", dlm->name);
 		__dlm_print_nodes(dlm);
+	}
 	spin_unlock(&dlm->spinlock);
 
 	if (ctxt) {
@@ -2131,13 +2138,6 @@
 		goto leave;
 	}
 
-	if (!o2hb_check_local_node_heartbeating()) {
-		mlog(ML_ERROR, "the local node has not been configured, or is "
-		     "not heartbeating\n");
-		ret = -EPROTO;
-		goto leave;
-	}
-
 	mlog(0, "register called for domain \"%s\"\n", domain);
 
 retry:
diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c
index 8d39e0fd6..975810b 100644
--- a/fs/ocfs2/dlm/dlmlock.c
+++ b/fs/ocfs2/dlm/dlmlock.c
@@ -183,10 +183,6 @@
 			kick_thread = 1;
 		}
 	}
-	/* reduce the inflight count, this may result in the lockres
-	 * being purged below during calc_usage */
-	if (lock->ml.node == dlm->node_num)
-		dlm_lockres_drop_inflight_ref(dlm, res);
 
 	spin_unlock(&res->spinlock);
 	wake_up(&res->wq);
@@ -231,10 +227,16 @@
 	     lock->ml.type, res->lockname.len,
 	     res->lockname.name, flags);
 
+	/*
+	 * Wait if resource is getting recovered, remastered, etc.
+	 * If the resource was remastered and new owner is self, then exit.
+	 */
 	spin_lock(&res->spinlock);
-
-	/* will exit this call with spinlock held */
 	__dlm_wait_on_lockres(res);
+	if (res->owner == dlm->node_num) {
+		spin_unlock(&res->spinlock);
+		return DLM_RECOVERING;
+	}
 	res->state |= DLM_LOCK_RES_IN_PROGRESS;
 
 	/* add lock to local (secondary) queue */
@@ -319,27 +321,23 @@
 	tmpret = o2net_send_message(DLM_CREATE_LOCK_MSG, dlm->key, &create,
 				    sizeof(create), res->owner, &status);
 	if (tmpret >= 0) {
-		// successfully sent and received
-		ret = status;  // this is already a dlm_status
+		ret = status;
 		if (ret == DLM_REJECTED) {
-			mlog(ML_ERROR, "%s:%.*s: BUG.  this is a stale lockres "
-			     "no longer owned by %u.  that node is coming back "
-			     "up currently.\n", dlm->name, create.namelen,
+			mlog(ML_ERROR, "%s: res %.*s, Stale lockres no longer "
+			     "owned by node %u. That node is coming back up "
+			     "currently.\n", dlm->name, create.namelen,
 			     create.name, res->owner);
 			dlm_print_one_lock_resource(res);
 			BUG();
 		}
 	} else {
-		mlog(ML_ERROR, "Error %d when sending message %u (key 0x%x) to "
-		     "node %u\n", tmpret, DLM_CREATE_LOCK_MSG, dlm->key,
-		     res->owner);
-		if (dlm_is_host_down(tmpret)) {
+		mlog(ML_ERROR, "%s: res %.*s, Error %d send CREATE LOCK to "
+		     "node %u\n", dlm->name, create.namelen, create.name,
+		     tmpret, res->owner);
+		if (dlm_is_host_down(tmpret))
 			ret = DLM_RECOVERING;
-			mlog(0, "node %u died so returning DLM_RECOVERING "
-			     "from lock message!\n", res->owner);
-		} else {
+		else
 			ret = dlm_err_to_dlm_status(tmpret);
-		}
 	}
 
 	return ret;
@@ -440,7 +438,7 @@
 		/* zero memory only if kernel-allocated */
 		lksb = kzalloc(sizeof(*lksb), GFP_NOFS);
 		if (!lksb) {
-			kfree(lock);
+			kmem_cache_free(dlm_lock_cache, lock);
 			return NULL;
 		}
 		kernel_allocated = 1;
@@ -718,18 +716,10 @@
 
 		if (status == DLM_RECOVERING || status == DLM_MIGRATING ||
 		    status == DLM_FORWARD) {
-			mlog(0, "retrying lock with migration/"
-			     "recovery/in progress\n");
 			msleep(100);
-			/* no waiting for dlm_reco_thread */
 			if (recovery) {
 				if (status != DLM_RECOVERING)
 					goto retry_lock;
-
-				mlog(0, "%s: got RECOVERING "
-				     "for $RECOVERY lock, master "
-				     "was %u\n", dlm->name,
-				     res->owner);
 				/* wait to see the node go down, then
 				 * drop down and allow the lockres to
 				 * get cleaned up.  need to remaster. */
@@ -741,6 +731,14 @@
 			}
 		}
 
+		/* Inflight taken in dlm_get_lock_resource() is dropped here */
+		spin_lock(&res->spinlock);
+		dlm_lockres_drop_inflight_ref(dlm, res);
+		spin_unlock(&res->spinlock);
+
+		dlm_lockres_calc_usage(dlm, res);
+		dlm_kick_thread(dlm, res);
+
 		if (status != DLM_NORMAL) {
 			lock->lksb->flags &= ~DLM_LKSB_GET_LVB;
 			if (status != DLM_NOTQUEUED)
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index 11eefb8..005261c3 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -631,39 +631,54 @@
 	return NULL;
 }
 
-void __dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm,
-				   struct dlm_lock_resource *res,
-				   int new_lockres,
-				   const char *file,
-				   int line)
+void dlm_lockres_set_refmap_bit(struct dlm_ctxt *dlm,
+				struct dlm_lock_resource *res, int bit)
 {
-	if (!new_lockres)
-		assert_spin_locked(&res->spinlock);
+	assert_spin_locked(&res->spinlock);
 
-	if (!test_bit(dlm->node_num, res->refmap)) {
-		BUG_ON(res->inflight_locks != 0);
-		dlm_lockres_set_refmap_bit(dlm->node_num, res);
-	}
-	res->inflight_locks++;
-	mlog(0, "%s:%.*s: inflight++: now %u\n",
-	     dlm->name, res->lockname.len, res->lockname.name,
-	     res->inflight_locks);
+	mlog(0, "res %.*s, set node %u, %ps()\n", res->lockname.len,
+	     res->lockname.name, bit, __builtin_return_address(0));
+
+	set_bit(bit, res->refmap);
 }
 
-void __dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm,
-				   struct dlm_lock_resource *res,
-				   const char *file,
-				   int line)
+void dlm_lockres_clear_refmap_bit(struct dlm_ctxt *dlm,
+				  struct dlm_lock_resource *res, int bit)
+{
+	assert_spin_locked(&res->spinlock);
+
+	mlog(0, "res %.*s, clr node %u, %ps()\n", res->lockname.len,
+	     res->lockname.name, bit, __builtin_return_address(0));
+
+	clear_bit(bit, res->refmap);
+}
+
+
+void dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm,
+				   struct dlm_lock_resource *res)
+{
+	assert_spin_locked(&res->spinlock);
+
+	res->inflight_locks++;
+
+	mlog(0, "%s: res %.*s, inflight++: now %u, %ps()\n", dlm->name,
+	     res->lockname.len, res->lockname.name, res->inflight_locks,
+	     __builtin_return_address(0));
+}
+
+void dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm,
+				   struct dlm_lock_resource *res)
 {
 	assert_spin_locked(&res->spinlock);
 
 	BUG_ON(res->inflight_locks == 0);
+
 	res->inflight_locks--;
-	mlog(0, "%s:%.*s: inflight--: now %u\n",
-	     dlm->name, res->lockname.len, res->lockname.name,
-	     res->inflight_locks);
-	if (res->inflight_locks == 0)
-		dlm_lockres_clear_refmap_bit(dlm->node_num, res);
+
+	mlog(0, "%s: res %.*s, inflight--: now %u, %ps()\n", dlm->name,
+	     res->lockname.len, res->lockname.name, res->inflight_locks,
+	     __builtin_return_address(0));
+
 	wake_up(&res->wq);
 }
 
@@ -697,7 +712,6 @@
 	unsigned int hash;
 	int tries = 0;
 	int bit, wait_on_recovery = 0;
-	int drop_inflight_if_nonlocal = 0;
 
 	BUG_ON(!lockid);
 
@@ -709,36 +723,33 @@
 	spin_lock(&dlm->spinlock);
 	tmpres = __dlm_lookup_lockres_full(dlm, lockid, namelen, hash);
 	if (tmpres) {
-		int dropping_ref = 0;
-
 		spin_unlock(&dlm->spinlock);
-
 		spin_lock(&tmpres->spinlock);
-		/* We wait for the other thread that is mastering the resource */
+		/* Wait on the thread that is mastering the resource */
 		if (tmpres->owner == DLM_LOCK_RES_OWNER_UNKNOWN) {
 			__dlm_wait_on_lockres(tmpres);
 			BUG_ON(tmpres->owner == DLM_LOCK_RES_OWNER_UNKNOWN);
-		}
-
-		if (tmpres->owner == dlm->node_num) {
-			BUG_ON(tmpres->state & DLM_LOCK_RES_DROPPING_REF);
-			dlm_lockres_grab_inflight_ref(dlm, tmpres);
-		} else if (tmpres->state & DLM_LOCK_RES_DROPPING_REF)
-			dropping_ref = 1;
-		spin_unlock(&tmpres->spinlock);
-
-		/* wait until done messaging the master, drop our ref to allow
-		 * the lockres to be purged, start over. */
-		if (dropping_ref) {
-			spin_lock(&tmpres->spinlock);
-			__dlm_wait_on_lockres_flags(tmpres, DLM_LOCK_RES_DROPPING_REF);
 			spin_unlock(&tmpres->spinlock);
 			dlm_lockres_put(tmpres);
 			tmpres = NULL;
 			goto lookup;
 		}
 
-		mlog(0, "found in hash!\n");
+		/* Wait on the resource purge to complete before continuing */
+		if (tmpres->state & DLM_LOCK_RES_DROPPING_REF) {
+			BUG_ON(tmpres->owner == dlm->node_num);
+			__dlm_wait_on_lockres_flags(tmpres,
+						    DLM_LOCK_RES_DROPPING_REF);
+			spin_unlock(&tmpres->spinlock);
+			dlm_lockres_put(tmpres);
+			tmpres = NULL;
+			goto lookup;
+		}
+
+		/* Grab inflight ref to pin the resource */
+		dlm_lockres_grab_inflight_ref(dlm, tmpres);
+
+		spin_unlock(&tmpres->spinlock);
 		if (res)
 			dlm_lockres_put(res);
 		res = tmpres;
@@ -829,8 +840,8 @@
 		 * but they might own this lockres.  wait on them. */
 		bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0);
 		if (bit < O2NM_MAX_NODES) {
-			mlog(ML_NOTICE, "%s:%.*s: at least one node (%d) to "
-			     "recover before lock mastery can begin\n",
+			mlog(0, "%s: res %.*s, At least one node (%d) "
+			     "to recover before lock mastery can begin\n",
 			     dlm->name, namelen, (char *)lockid, bit);
 			wait_on_recovery = 1;
 		}
@@ -843,12 +854,11 @@
 
 	/* finally add the lockres to its hash bucket */
 	__dlm_insert_lockres(dlm, res);
-	/* since this lockres is new it doesn't not require the spinlock */
-	dlm_lockres_grab_inflight_ref_new(dlm, res);
 
-	/* if this node does not become the master make sure to drop
-	 * this inflight reference below */
-	drop_inflight_if_nonlocal = 1;
+	/* Grab inflight ref to pin the resource */
+	spin_lock(&res->spinlock);
+	dlm_lockres_grab_inflight_ref(dlm, res);
+	spin_unlock(&res->spinlock);
 
 	/* get an extra ref on the mle in case this is a BLOCK
 	 * if so, the creator of the BLOCK may try to put the last
@@ -864,8 +874,8 @@
 		 * dlm spinlock would be detectable be a change on the mle,
 		 * so we only need to clear out the recovery map once. */
 		if (dlm_is_recovery_lock(lockid, namelen)) {
-			mlog(ML_NOTICE, "%s: recovery map is not empty, but "
-			     "must master $RECOVERY lock now\n", dlm->name);
+			mlog(0, "%s: Recovery map is not empty, but must "
+			     "master $RECOVERY lock now\n", dlm->name);
 			if (!dlm_pre_master_reco_lockres(dlm, res))
 				wait_on_recovery = 0;
 			else {
@@ -883,8 +893,8 @@
 		spin_lock(&dlm->spinlock);
 		bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0);
 		if (bit < O2NM_MAX_NODES) {
-			mlog(ML_NOTICE, "%s:%.*s: at least one node (%d) to "
-			     "recover before lock mastery can begin\n",
+			mlog(0, "%s: res %.*s, At least one node (%d) "
+			     "to recover before lock mastery can begin\n",
 			     dlm->name, namelen, (char *)lockid, bit);
 			wait_on_recovery = 1;
 		} else
@@ -913,8 +923,8 @@
 			 * yet, keep going until it does.  this is how the
 			 * master will know that asserts are needed back to
 			 * the lower nodes. */
-			mlog(0, "%s:%.*s: requests only up to %u but master "
-			     "is %u, keep going\n", dlm->name, namelen,
+			mlog(0, "%s: res %.*s, Requests only up to %u but "
+			     "master is %u, keep going\n", dlm->name, namelen,
 			     lockid, nodenum, mle->master);
 		}
 	}
@@ -924,13 +934,12 @@
 	ret = dlm_wait_for_lock_mastery(dlm, res, mle, &blocked);
 	if (ret < 0) {
 		wait_on_recovery = 1;
-		mlog(0, "%s:%.*s: node map changed, redo the "
-		     "master request now, blocked=%d\n",
-		     dlm->name, res->lockname.len,
+		mlog(0, "%s: res %.*s, Node map changed, redo the master "
+		     "request now, blocked=%d\n", dlm->name, res->lockname.len,
 		     res->lockname.name, blocked);
 		if (++tries > 20) {
-			mlog(ML_ERROR, "%s:%.*s: spinning on "
-			     "dlm_wait_for_lock_mastery, blocked=%d\n",
+			mlog(ML_ERROR, "%s: res %.*s, Spinning on "
+			     "dlm_wait_for_lock_mastery, blocked = %d\n",
 			     dlm->name, res->lockname.len,
 			     res->lockname.name, blocked);
 			dlm_print_one_lock_resource(res);
@@ -940,7 +949,8 @@
 		goto redo_request;
 	}
 
-	mlog(0, "lockres mastered by %u\n", res->owner);
+	mlog(0, "%s: res %.*s, Mastered by %u\n", dlm->name, res->lockname.len,
+	     res->lockname.name, res->owner);
 	/* make sure we never continue without this */
 	BUG_ON(res->owner == O2NM_MAX_NODES);
 
@@ -952,8 +962,6 @@
 
 wake_waiters:
 	spin_lock(&res->spinlock);
-	if (res->owner != dlm->node_num && drop_inflight_if_nonlocal)
-		dlm_lockres_drop_inflight_ref(dlm, res);
 	res->state &= ~DLM_LOCK_RES_IN_PROGRESS;
 	spin_unlock(&res->spinlock);
 	wake_up(&res->wq);
@@ -1426,9 +1434,7 @@
 		}
 
 		if (res->owner == dlm->node_num) {
-			mlog(0, "%s:%.*s: setting bit %u in refmap\n",
-			     dlm->name, namelen, name, request->node_idx);
-			dlm_lockres_set_refmap_bit(request->node_idx, res);
+			dlm_lockres_set_refmap_bit(dlm, res, request->node_idx);
 			spin_unlock(&res->spinlock);
 			response = DLM_MASTER_RESP_YES;
 			if (mle)
@@ -1493,10 +1499,8 @@
 				 * go back and clean the mles on any
 				 * other nodes */
 				dispatch_assert = 1;
-				dlm_lockres_set_refmap_bit(request->node_idx, res);
-				mlog(0, "%s:%.*s: setting bit %u in refmap\n",
-				     dlm->name, namelen, name,
-				     request->node_idx);
+				dlm_lockres_set_refmap_bit(dlm, res,
+							   request->node_idx);
 			} else
 				response = DLM_MASTER_RESP_NO;
 		} else {
@@ -1702,7 +1706,7 @@
 			     "lockres, set the bit in the refmap\n",
 			     namelen, lockname, to);
 			spin_lock(&res->spinlock);
-			dlm_lockres_set_refmap_bit(to, res);
+			dlm_lockres_set_refmap_bit(dlm, res, to);
 			spin_unlock(&res->spinlock);
 		}
 	}
@@ -2187,8 +2191,6 @@
 	namelen = res->lockname.len;
 	BUG_ON(namelen > O2NM_MAX_NAME_LEN);
 
-	mlog(0, "%s:%.*s: sending deref to %d\n",
-	     dlm->name, namelen, lockname, res->owner);
 	memset(&deref, 0, sizeof(deref));
 	deref.node_idx = dlm->node_num;
 	deref.namelen = namelen;
@@ -2197,14 +2199,12 @@
 	ret = o2net_send_message(DLM_DEREF_LOCKRES_MSG, dlm->key,
 				 &deref, sizeof(deref), res->owner, &r);
 	if (ret < 0)
-		mlog(ML_ERROR, "Error %d when sending message %u (key 0x%x) to "
-		     "node %u\n", ret, DLM_DEREF_LOCKRES_MSG, dlm->key,
-		     res->owner);
+		mlog(ML_ERROR, "%s: res %.*s, error %d send DEREF to node %u\n",
+		     dlm->name, namelen, lockname, ret, res->owner);
 	else if (r < 0) {
 		/* BAD.  other node says I did not have a ref. */
-		mlog(ML_ERROR,"while dropping ref on %s:%.*s "
-		    "(master=%u) got %d.\n", dlm->name, namelen,
-		    lockname, res->owner, r);
+		mlog(ML_ERROR, "%s: res %.*s, DEREF to node %u got %d\n",
+		     dlm->name, namelen, lockname, res->owner, r);
 		dlm_print_one_lock_resource(res);
 		BUG();
 	}
@@ -2260,7 +2260,7 @@
 	else {
 		BUG_ON(res->state & DLM_LOCK_RES_DROPPING_REF);
 		if (test_bit(node, res->refmap)) {
-			dlm_lockres_clear_refmap_bit(node, res);
+			dlm_lockres_clear_refmap_bit(dlm, res, node);
 			cleared = 1;
 		}
 	}
@@ -2320,7 +2320,7 @@
 	BUG_ON(res->state & DLM_LOCK_RES_DROPPING_REF);
 	if (test_bit(node, res->refmap)) {
 		__dlm_wait_on_lockres_flags(res, DLM_LOCK_RES_SETREF_INPROG);
-		dlm_lockres_clear_refmap_bit(node, res);
+		dlm_lockres_clear_refmap_bit(dlm, res, node);
 		cleared = 1;
 	}
 	spin_unlock(&res->spinlock);
@@ -2802,7 +2802,8 @@
 				BUG_ON(!list_empty(&lock->bast_list));
 				BUG_ON(lock->ast_pending);
 				BUG_ON(lock->bast_pending);
-				dlm_lockres_clear_refmap_bit(lock->ml.node, res);
+				dlm_lockres_clear_refmap_bit(dlm, res,
+							     lock->ml.node);
 				list_del_init(&lock->list);
 				dlm_lock_put(lock);
 				/* In a normal unlock, we would have added a
@@ -2823,7 +2824,7 @@
 			mlog(0, "%s:%.*s: node %u had a ref to this "
 			     "migrating lockres, clearing\n", dlm->name,
 			     res->lockname.len, res->lockname.name, bit);
-			dlm_lockres_clear_refmap_bit(bit, res);
+			dlm_lockres_clear_refmap_bit(dlm, res, bit);
 		}
 		bit++;
 	}
@@ -2916,9 +2917,9 @@
 					 &migrate, sizeof(migrate), nodenum,
 					 &status);
 		if (ret < 0) {
-			mlog(ML_ERROR, "Error %d when sending message %u (key "
-			     "0x%x) to node %u\n", ret, DLM_MIGRATE_REQUEST_MSG,
-			     dlm->key, nodenum);
+			mlog(ML_ERROR, "%s: res %.*s, Error %d send "
+			     "MIGRATE_REQUEST to node %u\n", dlm->name,
+			     migrate.namelen, migrate.name, ret, nodenum);
 			if (!dlm_is_host_down(ret)) {
 				mlog(ML_ERROR, "unhandled error=%d!\n", ret);
 				BUG();
@@ -2937,7 +2938,7 @@
 			     dlm->name, res->lockname.len, res->lockname.name,
 			     nodenum);
 			spin_lock(&res->spinlock);
-			dlm_lockres_set_refmap_bit(nodenum, res);
+			dlm_lockres_set_refmap_bit(dlm, res, nodenum);
 			spin_unlock(&res->spinlock);
 		}
 	}
@@ -3271,7 +3272,7 @@
 	 * mastery reference here since old_master will briefly have
 	 * a reference after the migration completes */
 	spin_lock(&res->spinlock);
-	dlm_lockres_set_refmap_bit(old_master, res);
+	dlm_lockres_set_refmap_bit(dlm, res, old_master);
 	spin_unlock(&res->spinlock);
 
 	mlog(0, "now time to do a migrate request to other nodes\n");
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index 7efab6d..01ebfd0 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -362,40 +362,38 @@
 }
 
 
-int dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout)
+void dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout)
 {
-	if (timeout) {
-		mlog(ML_NOTICE, "%s: waiting %dms for notification of "
-		     "death of node %u\n", dlm->name, timeout, node);
+	if (dlm_is_node_dead(dlm, node))
+		return;
+
+	printk(KERN_NOTICE "o2dlm: Waiting on the death of node %u in "
+	       "domain %s\n", node, dlm->name);
+
+	if (timeout)
 		wait_event_timeout(dlm->dlm_reco_thread_wq,
-			   dlm_is_node_dead(dlm, node),
-			   msecs_to_jiffies(timeout));
-	} else {
-		mlog(ML_NOTICE, "%s: waiting indefinitely for notification "
-		     "of death of node %u\n", dlm->name, node);
+				   dlm_is_node_dead(dlm, node),
+				   msecs_to_jiffies(timeout));
+	else
 		wait_event(dlm->dlm_reco_thread_wq,
 			   dlm_is_node_dead(dlm, node));
-	}
-	/* for now, return 0 */
-	return 0;
 }
 
-int dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout)
+void dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout)
 {
-	if (timeout) {
-		mlog(0, "%s: waiting %dms for notification of "
-		     "recovery of node %u\n", dlm->name, timeout, node);
+	if (dlm_is_node_recovered(dlm, node))
+		return;
+
+	printk(KERN_NOTICE "o2dlm: Waiting on the recovery of node %u in "
+	       "domain %s\n", node, dlm->name);
+
+	if (timeout)
 		wait_event_timeout(dlm->dlm_reco_thread_wq,
-			   dlm_is_node_recovered(dlm, node),
-			   msecs_to_jiffies(timeout));
-	} else {
-		mlog(0, "%s: waiting indefinitely for notification "
-		     "of recovery of node %u\n", dlm->name, node);
+				   dlm_is_node_recovered(dlm, node),
+				   msecs_to_jiffies(timeout));
+	else
 		wait_event(dlm->dlm_reco_thread_wq,
 			   dlm_is_node_recovered(dlm, node));
-	}
-	/* for now, return 0 */
-	return 0;
 }
 
 /* callers of the top-level api calls (dlmlock/dlmunlock) should
@@ -430,6 +428,8 @@
 {
 	spin_lock(&dlm->spinlock);
 	BUG_ON(dlm->reco.state & DLM_RECO_STATE_ACTIVE);
+	printk(KERN_NOTICE "o2dlm: Begin recovery on domain %s for node %u\n",
+	       dlm->name, dlm->reco.dead_node);
 	dlm->reco.state |= DLM_RECO_STATE_ACTIVE;
 	spin_unlock(&dlm->spinlock);
 }
@@ -440,9 +440,18 @@
 	BUG_ON(!(dlm->reco.state & DLM_RECO_STATE_ACTIVE));
 	dlm->reco.state &= ~DLM_RECO_STATE_ACTIVE;
 	spin_unlock(&dlm->spinlock);
+	printk(KERN_NOTICE "o2dlm: End recovery on domain %s\n", dlm->name);
 	wake_up(&dlm->reco.event);
 }
 
+static void dlm_print_recovery_master(struct dlm_ctxt *dlm)
+{
+	printk(KERN_NOTICE "o2dlm: Node %u (%s) is the Recovery Master for the "
+	       "dead node %u in domain %s\n", dlm->reco.new_master,
+	       (dlm->node_num == dlm->reco.new_master ? "me" : "he"),
+	       dlm->reco.dead_node, dlm->name);
+}
+
 static int dlm_do_recovery(struct dlm_ctxt *dlm)
 {
 	int status = 0;
@@ -505,9 +514,8 @@
 		}
 		mlog(0, "another node will master this recovery session.\n");
 	}
-	mlog(0, "dlm=%s (%d), new_master=%u, this node=%u, dead_node=%u\n",
-	     dlm->name, task_pid_nr(dlm->dlm_reco_thread_task), dlm->reco.new_master,
-	     dlm->node_num, dlm->reco.dead_node);
+
+	dlm_print_recovery_master(dlm);
 
 	/* it is safe to start everything back up here
 	 * because all of the dead node's lock resources
@@ -518,15 +526,13 @@
 	return 0;
 
 master_here:
-	mlog(ML_NOTICE, "(%d) Node %u is the Recovery Master for the Dead Node "
-	     "%u for Domain %s\n", task_pid_nr(dlm->dlm_reco_thread_task),
-	     dlm->node_num, dlm->reco.dead_node, dlm->name);
+	dlm_print_recovery_master(dlm);
 
 	status = dlm_remaster_locks(dlm, dlm->reco.dead_node);
 	if (status < 0) {
 		/* we should never hit this anymore */
-		mlog(ML_ERROR, "error %d remastering locks for node %u, "
-		     "retrying.\n", status, dlm->reco.dead_node);
+		mlog(ML_ERROR, "%s: Error %d remastering locks for node %u, "
+		     "retrying.\n", dlm->name, status, dlm->reco.dead_node);
 		/* yield a bit to allow any final network messages
 		 * to get handled on remaining nodes */
 		msleep(100);
@@ -567,7 +573,7 @@
 		BUG_ON(ndata->state != DLM_RECO_NODE_DATA_INIT);
 		ndata->state = DLM_RECO_NODE_DATA_REQUESTING;
 
-		mlog(0, "requesting lock info from node %u\n",
+		mlog(0, "%s: Requesting lock info from node %u\n", dlm->name,
 		     ndata->node_num);
 
 		if (ndata->node_num == dlm->node_num) {
@@ -640,7 +646,7 @@
 		spin_unlock(&dlm_reco_state_lock);
 	}
 
-	mlog(0, "done requesting all lock info\n");
+	mlog(0, "%s: Done requesting all lock info\n", dlm->name);
 
 	/* nodes should be sending reco data now
 	 * just need to wait */
@@ -802,10 +808,9 @@
 
 	/* negative status is handled by caller */
 	if (ret < 0)
-		mlog(ML_ERROR, "Error %d when sending message %u (key "
-		     "0x%x) to node %u\n", ret, DLM_LOCK_REQUEST_MSG,
-		     dlm->key, request_from);
-
+		mlog(ML_ERROR, "%s: Error %d send LOCK_REQUEST to node %u "
+		     "to recover dead node %u\n", dlm->name, ret,
+		     request_from, dead_node);
 	// return from here, then
 	// sleep until all received or error
 	return ret;
@@ -956,9 +961,9 @@
 	ret = o2net_send_message(DLM_RECO_DATA_DONE_MSG, dlm->key, &done_msg,
 				 sizeof(done_msg), send_to, &tmpret);
 	if (ret < 0) {
-		mlog(ML_ERROR, "Error %d when sending message %u (key "
-		     "0x%x) to node %u\n", ret, DLM_RECO_DATA_DONE_MSG,
-		     dlm->key, send_to);
+		mlog(ML_ERROR, "%s: Error %d send RECO_DATA_DONE to node %u "
+		     "to recover dead node %u\n", dlm->name, ret, send_to,
+		     dead_node);
 		if (!dlm_is_host_down(ret)) {
 			BUG();
 		}
@@ -1127,9 +1132,11 @@
 	if (ret < 0) {
 		/* XXX: negative status is not handled.
 		 * this will end up killing this node. */
-		mlog(ML_ERROR, "Error %d when sending message %u (key "
-		     "0x%x) to node %u\n", ret, DLM_MIG_LOCKRES_MSG,
-		     dlm->key, send_to);
+		mlog(ML_ERROR, "%s: res %.*s, Error %d send MIG_LOCKRES to "
+		     "node %u (%s)\n", dlm->name, mres->lockname_len,
+		     mres->lockname, ret, send_to,
+		     (orig_flags & DLM_MRES_MIGRATION ?
+		      "migration" : "recovery"));
 	} else {
 		/* might get an -ENOMEM back here */
 		ret = status;
@@ -1767,7 +1774,7 @@
 			     dlm->name, mres->lockname_len, mres->lockname,
 			     from);
 			spin_lock(&res->spinlock);
-			dlm_lockres_set_refmap_bit(from, res);
+			dlm_lockres_set_refmap_bit(dlm, res, from);
 			spin_unlock(&res->spinlock);
 			added++;
 			break;
@@ -1965,7 +1972,7 @@
 			mlog(0, "%s:%.*s: added lock for node %u, "
 			     "setting refmap bit\n", dlm->name,
 			     res->lockname.len, res->lockname.name, ml->node);
-			dlm_lockres_set_refmap_bit(ml->node, res);
+			dlm_lockres_set_refmap_bit(dlm, res, ml->node);
 			added++;
 		}
 		spin_unlock(&res->spinlock);
@@ -2084,6 +2091,9 @@
 
 	list_for_each_entry_safe(res, next, &dlm->reco.resources, recovering) {
 		if (res->owner == dead_node) {
+			mlog(0, "%s: res %.*s, Changing owner from %u to %u\n",
+			     dlm->name, res->lockname.len, res->lockname.name,
+			     res->owner, new_master);
 			list_del_init(&res->recovering);
 			spin_lock(&res->spinlock);
 			/* new_master has our reference from
@@ -2105,40 +2115,30 @@
 	for (i = 0; i < DLM_HASH_BUCKETS; i++) {
 		bucket = dlm_lockres_hash(dlm, i);
 		hlist_for_each_entry(res, hash_iter, bucket, hash_node) {
-			if (res->state & DLM_LOCK_RES_RECOVERING) {
-				if (res->owner == dead_node) {
-					mlog(0, "(this=%u) res %.*s owner=%u "
-					     "was not on recovering list, but "
-					     "clearing state anyway\n",
-					     dlm->node_num, res->lockname.len,
-					     res->lockname.name, new_master);
-				} else if (res->owner == dlm->node_num) {
-					mlog(0, "(this=%u) res %.*s owner=%u "
-					     "was not on recovering list, "
-					     "owner is THIS node, clearing\n",
-					     dlm->node_num, res->lockname.len,
-					     res->lockname.name, new_master);
-				} else
-					continue;
+			if (!(res->state & DLM_LOCK_RES_RECOVERING))
+				continue;
 
-				if (!list_empty(&res->recovering)) {
-					mlog(0, "%s:%.*s: lockres was "
-					     "marked RECOVERING, owner=%u\n",
-					     dlm->name, res->lockname.len,
-					     res->lockname.name, res->owner);
-					list_del_init(&res->recovering);
-					dlm_lockres_put(res);
-				}
-				spin_lock(&res->spinlock);
-				/* new_master has our reference from
-				 * the lock state sent during recovery */
-				dlm_change_lockres_owner(dlm, res, new_master);
-				res->state &= ~DLM_LOCK_RES_RECOVERING;
-				if (__dlm_lockres_has_locks(res))
-					__dlm_dirty_lockres(dlm, res);
-				spin_unlock(&res->spinlock);
-				wake_up(&res->wq);
+			if (res->owner != dead_node &&
+			    res->owner != dlm->node_num)
+				continue;
+
+			if (!list_empty(&res->recovering)) {
+				list_del_init(&res->recovering);
+				dlm_lockres_put(res);
 			}
+
+			/* new_master has our reference from
+			 * the lock state sent during recovery */
+			mlog(0, "%s: res %.*s, Changing owner from %u to %u\n",
+			     dlm->name, res->lockname.len, res->lockname.name,
+			     res->owner, new_master);
+			spin_lock(&res->spinlock);
+			dlm_change_lockres_owner(dlm, res, new_master);
+			res->state &= ~DLM_LOCK_RES_RECOVERING;
+			if (__dlm_lockres_has_locks(res))
+				__dlm_dirty_lockres(dlm, res);
+			spin_unlock(&res->spinlock);
+			wake_up(&res->wq);
 		}
 	}
 }
@@ -2252,12 +2252,12 @@
 			     res->lockname.len, res->lockname.name, freed, dead_node);
 			__dlm_print_one_lock_resource(res);
 		}
-		dlm_lockres_clear_refmap_bit(dead_node, res);
+		dlm_lockres_clear_refmap_bit(dlm, res, dead_node);
 	} else if (test_bit(dead_node, res->refmap)) {
 		mlog(0, "%s:%.*s: dead node %u had a ref, but had "
 		     "no locks and had not purged before dying\n", dlm->name,
 		     res->lockname.len, res->lockname.name, dead_node);
-		dlm_lockres_clear_refmap_bit(dead_node, res);
+		dlm_lockres_clear_refmap_bit(dlm, res, dead_node);
 	}
 
 	/* do not kick thread yet */
@@ -2324,9 +2324,9 @@
 			dlm_revalidate_lvb(dlm, res, dead_node);
 			if (res->owner == dead_node) {
 				if (res->state & DLM_LOCK_RES_DROPPING_REF) {
-					mlog(ML_NOTICE, "Ignore %.*s for "
+					mlog(ML_NOTICE, "%s: res %.*s, Skip "
 					     "recovery as it is being freed\n",
-					     res->lockname.len,
+					     dlm->name, res->lockname.len,
 					     res->lockname.name);
 				} else
 					dlm_move_lockres_to_recovery_list(dlm,
diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c
index 1d6d1d2..e73c833 100644
--- a/fs/ocfs2/dlm/dlmthread.c
+++ b/fs/ocfs2/dlm/dlmthread.c
@@ -94,24 +94,26 @@
 {
 	int bit;
 
+	assert_spin_locked(&res->spinlock);
+
 	if (__dlm_lockres_has_locks(res))
 		return 0;
 
+	/* Locks are in the process of being created */
+	if (res->inflight_locks)
+		return 0;
+
 	if (!list_empty(&res->dirty) || res->state & DLM_LOCK_RES_DIRTY)
 		return 0;
 
 	if (res->state & DLM_LOCK_RES_RECOVERING)
 		return 0;
 
+	/* Another node has this resource with this node as the master */
 	bit = find_next_bit(res->refmap, O2NM_MAX_NODES, 0);
 	if (bit < O2NM_MAX_NODES)
 		return 0;
 
-	/*
-	 * since the bit for dlm->node_num is not set, inflight_locks better
-	 * be zero
-	 */
-	BUG_ON(res->inflight_locks != 0);
 	return 1;
 }
 
@@ -185,8 +187,6 @@
 		/* clear our bit from the master's refmap, ignore errors */
 		ret = dlm_drop_lockres_ref(dlm, res);
 		if (ret < 0) {
-			mlog(ML_ERROR, "%s: deref %.*s failed %d\n", dlm->name,
-			     res->lockname.len, res->lockname.name, ret);
 			if (!dlm_is_host_down(ret))
 				BUG();
 		}
@@ -209,7 +209,7 @@
 		BUG();
 	}
 
-	__dlm_unhash_lockres(res);
+	__dlm_unhash_lockres(dlm, res);
 
 	/* lockres is not in the hash now.  drop the flag and wake up
 	 * any processes waiting in dlm_get_lock_resource. */
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index e1ed5e5..81a4cd2 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -1692,7 +1692,7 @@
 	mlog(0, "inode %llu take PRMODE open lock\n",
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno);
 
-	if (ocfs2_mount_local(osb))
+	if (ocfs2_is_hard_readonly(osb) || ocfs2_mount_local(osb))
 		goto out;
 
 	lockres = &OCFS2_I(inode)->ip_open_lockres;
@@ -1718,6 +1718,12 @@
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
 	     write ? "EXMODE" : "PRMODE");
 
+	if (ocfs2_is_hard_readonly(osb)) {
+		if (write)
+			status = -EROFS;
+		goto out;
+	}
+
 	if (ocfs2_mount_local(osb))
 		goto out;
 
@@ -2298,7 +2304,7 @@
 	if (ocfs2_is_hard_readonly(osb)) {
 		if (ex)
 			status = -EROFS;
-		goto bail;
+		goto getbh;
 	}
 
 	if (ocfs2_mount_local(osb))
@@ -2356,7 +2362,7 @@
 			mlog_errno(status);
 		goto bail;
 	}
-
+getbh:
 	if (ret_bh) {
 		status = ocfs2_assign_bh(inode, ret_bh, local_bh);
 		if (status < 0) {
@@ -2628,8 +2634,11 @@
 
 	BUG_ON(!dl);
 
-	if (ocfs2_is_hard_readonly(osb))
-		return -EROFS;
+	if (ocfs2_is_hard_readonly(osb)) {
+		if (ex)
+			return -EROFS;
+		return 0;
+	}
 
 	if (ocfs2_mount_local(osb))
 		return 0;
@@ -2647,7 +2656,7 @@
 	struct ocfs2_dentry_lock *dl = dentry->d_fsdata;
 	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
 
-	if (!ocfs2_mount_local(osb))
+	if (!ocfs2_is_hard_readonly(osb) && !ocfs2_mount_local(osb))
 		ocfs2_cluster_unlock(osb, &dl->dl_lockres, level);
 }
 
diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c
index 23457b4..2f5b92e 100644
--- a/fs/ocfs2/extent_map.c
+++ b/fs/ocfs2/extent_map.c
@@ -832,6 +832,102 @@
 	return ret;
 }
 
+int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int origin)
+{
+	struct inode *inode = file->f_mapping->host;
+	int ret;
+	unsigned int is_last = 0, is_data = 0;
+	u16 cs_bits = OCFS2_SB(inode->i_sb)->s_clustersize_bits;
+	u32 cpos, cend, clen, hole_size;
+	u64 extoff, extlen;
+	struct buffer_head *di_bh = NULL;
+	struct ocfs2_extent_rec rec;
+
+	BUG_ON(origin != SEEK_DATA && origin != SEEK_HOLE);
+
+	ret = ocfs2_inode_lock(inode, &di_bh, 0);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	down_read(&OCFS2_I(inode)->ip_alloc_sem);
+
+	if (*offset >= inode->i_size) {
+		ret = -ENXIO;
+		goto out_unlock;
+	}
+
+	if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) {
+		if (origin == SEEK_HOLE)
+			*offset = inode->i_size;
+		goto out_unlock;
+	}
+
+	clen = 0;
+	cpos = *offset >> cs_bits;
+	cend = ocfs2_clusters_for_bytes(inode->i_sb, inode->i_size);
+
+	while (cpos < cend && !is_last) {
+		ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos, &hole_size,
+						 &rec, &is_last);
+		if (ret) {
+			mlog_errno(ret);
+			goto out_unlock;
+		}
+
+		extoff = cpos;
+		extoff <<= cs_bits;
+
+		if (rec.e_blkno == 0ULL) {
+			clen = hole_size;
+			is_data = 0;
+		} else {
+			clen = le16_to_cpu(rec.e_leaf_clusters) -
+				(cpos - le32_to_cpu(rec.e_cpos));
+			is_data = (rec.e_flags & OCFS2_EXT_UNWRITTEN) ?  0 : 1;
+		}
+
+		if ((!is_data && origin == SEEK_HOLE) ||
+		    (is_data && origin == SEEK_DATA)) {
+			if (extoff > *offset)
+				*offset = extoff;
+			goto out_unlock;
+		}
+
+		if (!is_last)
+			cpos += clen;
+	}
+
+	if (origin == SEEK_HOLE) {
+		extoff = cpos;
+		extoff <<= cs_bits;
+		extlen = clen;
+		extlen <<=  cs_bits;
+
+		if ((extoff + extlen) > inode->i_size)
+			extlen = inode->i_size - extoff;
+		extoff += extlen;
+		if (extoff > *offset)
+			*offset = extoff;
+		goto out_unlock;
+	}
+
+	ret = -ENXIO;
+
+out_unlock:
+
+	brelse(di_bh);
+
+	up_read(&OCFS2_I(inode)->ip_alloc_sem);
+
+	ocfs2_inode_unlock(inode, 0);
+out:
+	if (ret && ret != -ENXIO)
+		ret = -ENXIO;
+	return ret;
+}
+
 int ocfs2_read_virt_blocks(struct inode *inode, u64 v_block, int nr,
 			   struct buffer_head *bhs[], int flags,
 			   int (*validate)(struct super_block *sb,
diff --git a/fs/ocfs2/extent_map.h b/fs/ocfs2/extent_map.h
index e79d41c..67ea57d 100644
--- a/fs/ocfs2/extent_map.h
+++ b/fs/ocfs2/extent_map.h
@@ -53,6 +53,8 @@
 int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		 u64 map_start, u64 map_len);
 
+int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int origin);
+
 int ocfs2_xattr_get_clusters(struct inode *inode, u32 v_cluster,
 			     u32 *p_cluster, u32 *num_clusters,
 			     struct ocfs2_extent_list *el,
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index de4ea1a..6e39668 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1950,6 +1950,9 @@
 	if (ret < 0)
 		mlog_errno(ret);
 
+	if (file->f_flags & O_SYNC)
+		handle->h_sync = 1;
+
 	ocfs2_commit_trans(osb, handle);
 
 out_inode_unlock:
@@ -2052,6 +2055,23 @@
 	return ret;
 }
 
+static void ocfs2_aiodio_wait(struct inode *inode)
+{
+	wait_queue_head_t *wq = ocfs2_ioend_wq(inode);
+
+	wait_event(*wq, (atomic_read(&OCFS2_I(inode)->ip_unaligned_aio) == 0));
+}
+
+static int ocfs2_is_io_unaligned(struct inode *inode, size_t count, loff_t pos)
+{
+	int blockmask = inode->i_sb->s_blocksize - 1;
+	loff_t final_size = pos + count;
+
+	if ((pos & blockmask) || (final_size & blockmask))
+		return 1;
+	return 0;
+}
+
 static int ocfs2_prepare_inode_for_refcount(struct inode *inode,
 					    struct file *file,
 					    loff_t pos, size_t count,
@@ -2230,6 +2250,7 @@
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	int full_coherency = !(osb->s_mount_opt &
 			       OCFS2_MOUNT_COHERENCY_BUFFERED);
+	int unaligned_dio = 0;
 
 	trace_ocfs2_file_aio_write(inode, file, file->f_path.dentry,
 		(unsigned long long)OCFS2_I(inode)->ip_blkno,
@@ -2297,6 +2318,10 @@
 		goto out;
 	}
 
+	if (direct_io && !is_sync_kiocb(iocb))
+		unaligned_dio = ocfs2_is_io_unaligned(inode, iocb->ki_left,
+						      *ppos);
+
 	/*
 	 * We can't complete the direct I/O as requested, fall back to
 	 * buffered I/O.
@@ -2311,6 +2336,18 @@
 		goto relock;
 	}
 
+	if (unaligned_dio) {
+		/*
+		 * Wait on previous unaligned aio to complete before
+		 * proceeding.
+		 */
+		ocfs2_aiodio_wait(inode);
+
+		/* Mark the iocb as needing a decrement in ocfs2_dio_end_io */
+		atomic_inc(&OCFS2_I(inode)->ip_unaligned_aio);
+		ocfs2_iocb_set_unaligned_aio(iocb);
+	}
+
 	/*
 	 * To later detect whether a journal commit for sync writes is
 	 * necessary, we sample i_size, and cluster count here.
@@ -2382,8 +2419,12 @@
 	if ((ret == -EIOCBQUEUED) || (!ocfs2_iocb_is_rw_locked(iocb))) {
 		rw_level = -1;
 		have_alloc_sem = 0;
+		unaligned_dio = 0;
 	}
 
+	if (unaligned_dio)
+		atomic_dec(&OCFS2_I(inode)->ip_unaligned_aio);
+
 out:
 	if (rw_level != -1)
 		ocfs2_rw_unlock(inode, rw_level);
@@ -2591,6 +2632,57 @@
 	return ret;
 }
 
+/* Refer generic_file_llseek_unlocked() */
+static loff_t ocfs2_file_llseek(struct file *file, loff_t offset, int origin)
+{
+	struct inode *inode = file->f_mapping->host;
+	int ret = 0;
+
+	mutex_lock(&inode->i_mutex);
+
+	switch (origin) {
+	case SEEK_SET:
+		break;
+	case SEEK_END:
+		offset += inode->i_size;
+		break;
+	case SEEK_CUR:
+		if (offset == 0) {
+			offset = file->f_pos;
+			goto out;
+		}
+		offset += file->f_pos;
+		break;
+	case SEEK_DATA:
+	case SEEK_HOLE:
+		ret = ocfs2_seek_data_hole_offset(file, &offset, origin);
+		if (ret)
+			goto out;
+		break;
+	default:
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET))
+		ret = -EINVAL;
+	if (!ret && offset > inode->i_sb->s_maxbytes)
+		ret = -EINVAL;
+	if (ret)
+		goto out;
+
+	if (offset != file->f_pos) {
+		file->f_pos = offset;
+		file->f_version = 0;
+	}
+
+out:
+	mutex_unlock(&inode->i_mutex);
+	if (ret)
+		return ret;
+	return offset;
+}
+
 const struct inode_operations ocfs2_file_iops = {
 	.setattr	= ocfs2_setattr,
 	.getattr	= ocfs2_getattr,
@@ -2615,7 +2707,7 @@
  * ocfs2_fops_no_plocks and ocfs2_dops_no_plocks!
  */
 const struct file_operations ocfs2_fops = {
-	.llseek		= generic_file_llseek,
+	.llseek		= ocfs2_file_llseek,
 	.read		= do_sync_read,
 	.write		= do_sync_write,
 	.mmap		= ocfs2_mmap,
@@ -2663,7 +2755,7 @@
  * the cluster.
  */
 const struct file_operations ocfs2_fops_no_plocks = {
-	.llseek		= generic_file_llseek,
+	.llseek		= ocfs2_file_llseek,
 	.read		= do_sync_read,
 	.write		= do_sync_write,
 	.mmap		= ocfs2_mmap,
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index a22d2c0..17454a9 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -951,7 +951,7 @@
 	trace_ocfs2_cleanup_delete_inode(
 		(unsigned long long)OCFS2_I(inode)->ip_blkno, sync_data);
 	if (sync_data)
-		write_inode_now(inode, 1);
+		filemap_write_and_wait(inode->i_mapping);
 	truncate_inode_pages(&inode->i_data, 0);
 }
 
diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h
index 1c508b1..88924a3 100644
--- a/fs/ocfs2/inode.h
+++ b/fs/ocfs2/inode.h
@@ -43,6 +43,9 @@
 	/* protects extended attribute changes on this inode */
 	struct rw_semaphore		ip_xattr_sem;
 
+	/* Number of outstanding AIO's which are not page aligned */
+	atomic_t			ip_unaligned_aio;
+
 	/* These fields are protected by ip_lock */
 	spinlock_t			ip_lock;
 	u32				ip_open_count;
diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c
index bc91072..726ff26 100644
--- a/fs/ocfs2/ioctl.c
+++ b/fs/ocfs2/ioctl.c
@@ -122,7 +122,7 @@
 	if ((oldflags & OCFS2_IMMUTABLE_FL) || ((flags ^ oldflags) &
 		(OCFS2_APPEND_FL | OCFS2_IMMUTABLE_FL))) {
 		if (!capable(CAP_LINUX_IMMUTABLE))
-			goto bail_unlock;
+			goto bail_commit;
 	}
 
 	ocfs2_inode->ip_attr = flags;
@@ -132,6 +132,7 @@
 	if (status < 0)
 		mlog_errno(status);
 
+bail_commit:
 	ocfs2_commit_trans(osb, handle);
 bail_unlock:
 	ocfs2_inode_unlock(inode, 1);
@@ -381,7 +382,7 @@
 	if (!oifi) {
 		status = -ENOMEM;
 		mlog_errno(status);
-		goto bail;
+		goto out_err;
 	}
 
 	if (o2info_from_user(*oifi, req))
@@ -431,7 +432,7 @@
 		o2info_set_request_error(&oifi->ifi_req, req);
 
 	kfree(oifi);
-
+out_err:
 	return status;
 }
 
@@ -666,7 +667,7 @@
 	if (!oiff) {
 		status = -ENOMEM;
 		mlog_errno(status);
-		goto bail;
+		goto out_err;
 	}
 
 	if (o2info_from_user(*oiff, req))
@@ -716,7 +717,7 @@
 		o2info_set_request_error(&oiff->iff_req, req);
 
 	kfree(oiff);
-
+out_err:
 	return status;
 }
 
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 295d564..0a42ae9 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1544,9 +1544,9 @@
 	/* we need to run complete recovery for offline orphan slots */
 	ocfs2_replay_map_set_state(osb, REPLAY_NEEDED);
 
-	mlog(ML_NOTICE, "Recovering node %d from slot %d on device (%u,%u)\n",
-	     node_num, slot_num,
-	     MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev));
+	printk(KERN_NOTICE "ocfs2: Begin replay journal (node %d, slot %d) on "\
+	       "device (%u,%u)\n", node_num, slot_num, MAJOR(osb->sb->s_dev),
+	       MINOR(osb->sb->s_dev));
 
 	OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
 
@@ -1601,6 +1601,9 @@
 
 	jbd2_journal_destroy(journal);
 
+	printk(KERN_NOTICE "ocfs2: End replay journal (node %d, slot %d) on "\
+	       "device (%u,%u)\n", node_num, slot_num, MAJOR(osb->sb->s_dev),
+	       MINOR(osb->sb->s_dev));
 done:
 	/* drop the lock on this nodes journal */
 	if (got_lock)
@@ -1808,6 +1811,20 @@
  * every slot, queuing a recovery of the slot on the ocfs2_wq thread. This
  * is done to catch any orphans that are left over in orphan directories.
  *
+ * It scans all slots, even ones that are in use. It does so to handle the
+ * case described below:
+ *
+ *   Node 1 has an inode it was using. The dentry went away due to memory
+ *   pressure.  Node 1 closes the inode, but it's on the free list. The node
+ *   has the open lock.
+ *   Node 2 unlinks the inode. It grabs the dentry lock to notify others,
+ *   but node 1 has no dentry and doesn't get the message. It trylocks the
+ *   open lock, sees that another node has a PR, and does nothing.
+ *   Later node 2 runs its orphan dir. It igets the inode, trylocks the
+ *   open lock, sees the PR still, and does nothing.
+ *   Basically, we have to trigger an orphan iput on node 1. The only way
+ *   for this to happen is if node 1 runs node 2's orphan dir.
+ *
  * ocfs2_queue_orphan_scan gets called every ORPHAN_SCAN_SCHEDULE_TIMEOUT
  * seconds.  It gets an EX lock on os_lockres and checks sequence number
  * stored in LVB. If the sequence number has changed, it means some other
diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h
index 68cf2f6..a3385b6 100644
--- a/fs/ocfs2/journal.h
+++ b/fs/ocfs2/journal.h
@@ -441,10 +441,11 @@
 #define OCFS2_SIMPLE_DIR_EXTEND_CREDITS (2)
 
 /* file update (nlink, etc) + directory mtime/ctime + dir entry block + quota
- * update on dir + index leaf + dx root update for free list */
+ * update on dir + index leaf + dx root update for free list +
+ * previous dirblock update in the free list */
 static inline int ocfs2_link_credits(struct super_block *sb)
 {
-	return 2*OCFS2_INODE_UPDATE_CREDITS + 3 +
+	return 2*OCFS2_INODE_UPDATE_CREDITS + 4 +
 	       ocfs2_quota_trans_credits(sb);
 }
 
diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c
index 3e9393c..9cd4108 100644
--- a/fs/ocfs2/mmap.c
+++ b/fs/ocfs2/mmap.c
@@ -61,7 +61,7 @@
 static int __ocfs2_page_mkwrite(struct file *file, struct buffer_head *di_bh,
 				struct page *page)
 {
-	int ret;
+	int ret = VM_FAULT_NOPAGE;
 	struct inode *inode = file->f_path.dentry->d_inode;
 	struct address_space *mapping = inode->i_mapping;
 	loff_t pos = page_offset(page);
@@ -71,32 +71,25 @@
 	void *fsdata;
 	loff_t size = i_size_read(inode);
 
-	/*
-	 * Another node might have truncated while we were waiting on
-	 * cluster locks.
-	 * We don't check size == 0 before the shift. This is borrowed
-	 * from do_generic_file_read.
-	 */
 	last_index = (size - 1) >> PAGE_CACHE_SHIFT;
-	if (unlikely(!size || page->index > last_index)) {
-		ret = -EINVAL;
-		goto out;
-	}
 
 	/*
-	 * The i_size check above doesn't catch the case where nodes
-	 * truncated and then re-extended the file. We'll re-check the
-	 * page mapping after taking the page lock inside of
-	 * ocfs2_write_begin_nolock().
+	 * There are cases that lead to the page no longer bebongs to the
+	 * mapping.
+	 * 1) pagecache truncates locally due to memory pressure.
+	 * 2) pagecache truncates when another is taking EX lock against 
+	 * inode lock. see ocfs2_data_convert_worker.
+	 * 
+	 * The i_size check doesn't catch the case where nodes truncated and
+	 * then re-extended the file. We'll re-check the page mapping after
+	 * taking the page lock inside of ocfs2_write_begin_nolock().
+	 *
+	 * Let VM retry with these cases.
 	 */
-	if (!PageUptodate(page) || page->mapping != inode->i_mapping) {
-		/*
-		 * the page has been umapped in ocfs2_data_downconvert_worker.
-		 * So return 0 here and let VFS retry.
-		 */
-		ret = 0;
+	if ((page->mapping != inode->i_mapping) ||
+	    (!PageUptodate(page)) ||
+	    (page_offset(page) >= size))
 		goto out;
-	}
 
 	/*
 	 * Call ocfs2_write_begin() and ocfs2_write_end() to take
@@ -116,17 +109,21 @@
 	if (ret) {
 		if (ret != -ENOSPC)
 			mlog_errno(ret);
+		if (ret == -ENOMEM)
+			ret = VM_FAULT_OOM;
+		else
+			ret = VM_FAULT_SIGBUS;
 		goto out;
 	}
 
-	ret = ocfs2_write_end_nolock(mapping, pos, len, len, locked_page,
-				     fsdata);
-	if (ret < 0) {
-		mlog_errno(ret);
+	if (!locked_page) {
+		ret = VM_FAULT_NOPAGE;
 		goto out;
 	}
+	ret = ocfs2_write_end_nolock(mapping, pos, len, len, locked_page,
+				     fsdata);
 	BUG_ON(ret != len);
-	ret = 0;
+	ret = VM_FAULT_LOCKED;
 out:
 	return ret;
 }
@@ -168,8 +165,6 @@
 
 out:
 	ocfs2_unblock_signals(&oldset);
-	if (ret)
-		ret = VM_FAULT_SIGBUS;
 	return ret;
 }
 
diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
index d53cb70..184c76b 100644
--- a/fs/ocfs2/move_extents.c
+++ b/fs/ocfs2/move_extents.c
@@ -745,7 +745,7 @@
 	 */
 	ocfs2_probe_alloc_group(inode, gd_bh, &goal_bit, len, move_max_hop,
 				new_phys_cpos);
-	if (!new_phys_cpos) {
+	if (!*new_phys_cpos) {
 		ret = -ENOSPC;
 		goto out_commit;
 	}
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index 4092858..d355e6e 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -836,18 +836,65 @@
 
 static inline void _ocfs2_set_bit(unsigned int bit, unsigned long *bitmap)
 {
-	__test_and_set_bit_le(bit, bitmap);
+	__set_bit_le(bit, bitmap);
 }
 #define ocfs2_set_bit(bit, addr) _ocfs2_set_bit((bit), (unsigned long *)(addr))
 
 static inline void _ocfs2_clear_bit(unsigned int bit, unsigned long *bitmap)
 {
-	__test_and_clear_bit_le(bit, bitmap);
+	__clear_bit_le(bit, bitmap);
 }
 #define ocfs2_clear_bit(bit, addr) _ocfs2_clear_bit((bit), (unsigned long *)(addr))
 
 #define ocfs2_test_bit test_bit_le
 #define ocfs2_find_next_zero_bit find_next_zero_bit_le
 #define ocfs2_find_next_bit find_next_bit_le
+
+static inline void *correct_addr_and_bit_unaligned(int *bit, void *addr)
+{
+#if BITS_PER_LONG == 64
+	*bit += ((unsigned long) addr & 7UL) << 3;
+	addr = (void *) ((unsigned long) addr & ~7UL);
+#elif BITS_PER_LONG == 32
+	*bit += ((unsigned long) addr & 3UL) << 3;
+	addr = (void *) ((unsigned long) addr & ~3UL);
+#else
+#error "how many bits you are?!"
+#endif
+	return addr;
+}
+
+static inline void ocfs2_set_bit_unaligned(int bit, void *bitmap)
+{
+	bitmap = correct_addr_and_bit_unaligned(&bit, bitmap);
+	ocfs2_set_bit(bit, bitmap);
+}
+
+static inline void ocfs2_clear_bit_unaligned(int bit, void *bitmap)
+{
+	bitmap = correct_addr_and_bit_unaligned(&bit, bitmap);
+	ocfs2_clear_bit(bit, bitmap);
+}
+
+static inline int ocfs2_test_bit_unaligned(int bit, void *bitmap)
+{
+	bitmap = correct_addr_and_bit_unaligned(&bit, bitmap);
+	return ocfs2_test_bit(bit, bitmap);
+}
+
+static inline int ocfs2_find_next_zero_bit_unaligned(void *bitmap, int max,
+							int start)
+{
+	int fix = 0, ret, tmpmax;
+	bitmap = correct_addr_and_bit_unaligned(&fix, bitmap);
+	tmpmax = max + fix;
+	start += fix;
+
+	ret = ocfs2_find_next_zero_bit(bitmap, tmpmax, start) - fix;
+	if (ret > max)
+		return max;
+	return ret;
+}
+
 #endif  /* OCFS2_H */
 
diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c
index dc8007f..f100bf7 100644
--- a/fs/ocfs2/quota_local.c
+++ b/fs/ocfs2/quota_local.c
@@ -404,7 +404,9 @@
 	int status = 0;
 	struct ocfs2_quota_recovery *rec;
 
-	mlog(ML_NOTICE, "Beginning quota recovery in slot %u\n", slot_num);
+	printk(KERN_NOTICE "ocfs2: Beginning quota recovery on device (%s) for "
+	       "slot %u\n", osb->dev_str, slot_num);
+
 	rec = ocfs2_alloc_quota_recovery();
 	if (!rec)
 		return ERR_PTR(-ENOMEM);
@@ -549,8 +551,8 @@
 				goto out_commit;
 			}
 			lock_buffer(qbh);
-			WARN_ON(!ocfs2_test_bit(bit, dchunk->dqc_bitmap));
-			ocfs2_clear_bit(bit, dchunk->dqc_bitmap);
+			WARN_ON(!ocfs2_test_bit_unaligned(bit, dchunk->dqc_bitmap));
+			ocfs2_clear_bit_unaligned(bit, dchunk->dqc_bitmap);
 			le32_add_cpu(&dchunk->dqc_free, 1);
 			unlock_buffer(qbh);
 			ocfs2_journal_dirty(handle, qbh);
@@ -596,7 +598,9 @@
 	struct inode *lqinode;
 	unsigned int flags;
 
-	mlog(ML_NOTICE, "Finishing quota recovery in slot %u\n", slot_num);
+	printk(KERN_NOTICE "ocfs2: Finishing quota recovery on device (%s) for "
+	       "slot %u\n", osb->dev_str, slot_num);
+
 	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
 	for (type = 0; type < MAXQUOTAS; type++) {
 		if (list_empty(&(rec->r_list[type])))
@@ -612,8 +616,9 @@
 		/* Someone else is holding the lock? Then he must be
 		 * doing the recovery. Just skip the file... */
 		if (status == -EAGAIN) {
-			mlog(ML_NOTICE, "skipping quota recovery for slot %d "
-			     "because quota file is locked.\n", slot_num);
+			printk(KERN_NOTICE "ocfs2: Skipping quota recovery on "
+			       "device (%s) for slot %d because quota file is "
+			       "locked.\n", osb->dev_str, slot_num);
 			status = 0;
 			goto out_put;
 		} else if (status < 0) {
@@ -944,7 +949,7 @@
 		      * ol_quota_entries_per_block(sb);
 	}
 
-	found = ocfs2_find_next_zero_bit(dchunk->dqc_bitmap, len, 0);
+	found = ocfs2_find_next_zero_bit_unaligned(dchunk->dqc_bitmap, len, 0);
 	/* We failed? */
 	if (found == len) {
 		mlog(ML_ERROR, "Did not find empty entry in chunk %d with %u"
@@ -1208,7 +1213,7 @@
 	struct ocfs2_local_disk_chunk *dchunk;
 
 	dchunk = (struct ocfs2_local_disk_chunk *)bh->b_data;
-	ocfs2_set_bit(*offset, dchunk->dqc_bitmap);
+	ocfs2_set_bit_unaligned(*offset, dchunk->dqc_bitmap);
 	le32_add_cpu(&dchunk->dqc_free, -1);
 }
 
@@ -1289,7 +1294,7 @@
 			(od->dq_chunk->qc_headerbh->b_data);
 	/* Mark structure as freed */
 	lock_buffer(od->dq_chunk->qc_headerbh);
-	ocfs2_clear_bit(offset, dchunk->dqc_bitmap);
+	ocfs2_clear_bit_unaligned(offset, dchunk->dqc_bitmap);
 	le32_add_cpu(&dchunk->dqc_free, 1);
 	unlock_buffer(od->dq_chunk->qc_headerbh);
 	ocfs2_journal_dirty(handle, od->dq_chunk->qc_headerbh);
diff --git a/fs/ocfs2/slot_map.c b/fs/ocfs2/slot_map.c
index 26fc001..1424c15 100644
--- a/fs/ocfs2/slot_map.c
+++ b/fs/ocfs2/slot_map.c
@@ -493,8 +493,8 @@
 			goto bail;
 		}
 	} else
-		mlog(ML_NOTICE, "slot %d is already allocated to this node!\n",
-		     slot);
+		printk(KERN_INFO "ocfs2: Slot %d on device (%s) was already "
+		       "allocated to this node!\n", slot, osb->dev_str);
 
 	ocfs2_set_slot(si, slot, osb->node_num);
 	osb->slot_num = slot;
diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c
index 19965b0..9436801 100644
--- a/fs/ocfs2/stack_o2cb.c
+++ b/fs/ocfs2/stack_o2cb.c
@@ -28,6 +28,7 @@
 #include "cluster/masklog.h"
 #include "cluster/nodemanager.h"
 #include "cluster/heartbeat.h"
+#include "cluster/tcp.h"
 
 #include "stackglue.h"
 
@@ -256,6 +257,61 @@
 }
 
 /*
+ * Check if this node is heartbeating and is connected to all other
+ * heartbeating nodes.
+ */
+static int o2cb_cluster_check(void)
+{
+	u8 node_num;
+	int i;
+	unsigned long hbmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
+	unsigned long netmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
+
+	node_num = o2nm_this_node();
+	if (node_num == O2NM_MAX_NODES) {
+		printk(KERN_ERR "o2cb: This node has not been configured.\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * o2dlm expects o2net sockets to be created. If not, then
+	 * dlm_join_domain() fails with a stack of errors which are both cryptic
+	 * and incomplete. The idea here is to detect upfront whether we have
+	 * managed to connect to all nodes or not. If not, then list the nodes
+	 * to allow the user to check the configuration (incorrect IP, firewall,
+	 * etc.) Yes, this is racy. But its not the end of the world.
+	 */
+#define	O2CB_MAP_STABILIZE_COUNT	60
+	for (i = 0; i < O2CB_MAP_STABILIZE_COUNT; ++i) {
+		o2hb_fill_node_map(hbmap, sizeof(hbmap));
+		if (!test_bit(node_num, hbmap)) {
+			printk(KERN_ERR "o2cb: %s heartbeat has not been "
+			       "started.\n", (o2hb_global_heartbeat_active() ?
+					      "Global" : "Local"));
+			return -EINVAL;
+		}
+		o2net_fill_node_map(netmap, sizeof(netmap));
+		/* Force set the current node to allow easy compare */
+		set_bit(node_num, netmap);
+		if (!memcmp(hbmap, netmap, sizeof(hbmap)))
+			return 0;
+		if (i < O2CB_MAP_STABILIZE_COUNT)
+			msleep(1000);
+	}
+
+	printk(KERN_ERR "o2cb: This node could not connect to nodes:");
+	i = -1;
+	while ((i = find_next_bit(hbmap, O2NM_MAX_NODES,
+				  i + 1)) < O2NM_MAX_NODES) {
+		if (!test_bit(i, netmap))
+			printk(" %u", i);
+	}
+	printk(".\n");
+
+	return -ENOTCONN;
+}
+
+/*
  * Called from the dlm when it's about to evict a node. This is how the
  * classic stack signals node death.
  */
@@ -263,8 +319,8 @@
 {
 	struct ocfs2_cluster_connection *conn = data;
 
-	mlog(ML_NOTICE, "o2dlm has evicted node %d from group %.*s\n",
-	     node_num, conn->cc_namelen, conn->cc_name);
+	printk(KERN_NOTICE "o2cb: o2dlm has evicted node %d from domain %.*s\n",
+	       node_num, conn->cc_namelen, conn->cc_name);
 
 	conn->cc_recovery_handler(node_num, conn->cc_recovery_data);
 }
@@ -280,12 +336,11 @@
 	BUG_ON(conn == NULL);
 	BUG_ON(conn->cc_proto == NULL);
 
-	/* for now we only have one cluster/node, make sure we see it
-	 * in the heartbeat universe */
-	if (!o2hb_check_local_node_heartbeating()) {
-		if (o2hb_global_heartbeat_active())
-			mlog(ML_ERROR, "Global heartbeat not started\n");
-		rc = -EINVAL;
+	/* Ensure cluster stack is up and all nodes are connected */
+	rc = o2cb_cluster_check();
+	if (rc) {
+		printk(KERN_ERR "o2cb: Cluster check failed. Fix errors "
+		       "before retrying.\n");
 		goto out;
 	}
 
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 56f6102..4994f8b 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -54,6 +54,7 @@
 #include "ocfs1_fs_compat.h"
 
 #include "alloc.h"
+#include "aops.h"
 #include "blockcheck.h"
 #include "dlmglue.h"
 #include "export.h"
@@ -1107,9 +1108,9 @@
 
 		ocfs2_set_ro_flag(osb, 1);
 
-		printk(KERN_NOTICE "Readonly device detected. No cluster "
-		       "services will be utilized for this mount. Recovery "
-		       "will be skipped.\n");
+		printk(KERN_NOTICE "ocfs2: Readonly device (%s) detected. "
+		       "Cluster services will not be used for this mount. "
+		       "Recovery will be skipped.\n", osb->dev_str);
 	}
 
 	if (!ocfs2_is_hard_readonly(osb)) {
@@ -1616,12 +1617,17 @@
 	return 0;
 }
 
+wait_queue_head_t ocfs2__ioend_wq[OCFS2_IOEND_WQ_HASH_SZ];
+
 static int __init ocfs2_init(void)
 {
-	int status;
+	int status, i;
 
 	ocfs2_print_version();
 
+	for (i = 0; i < OCFS2_IOEND_WQ_HASH_SZ; i++)
+		init_waitqueue_head(&ocfs2__ioend_wq[i]);
+
 	status = init_ocfs2_uptodate_cache();
 	if (status < 0) {
 		mlog_errno(status);
@@ -1760,7 +1766,7 @@
 	ocfs2_extent_map_init(&oi->vfs_inode);
 	INIT_LIST_HEAD(&oi->ip_io_markers);
 	oi->ip_dir_start_lookup = 0;
-
+	atomic_set(&oi->ip_unaligned_aio, 0);
 	init_rwsem(&oi->ip_alloc_sem);
 	init_rwsem(&oi->ip_xattr_sem);
 	mutex_init(&oi->ip_io_mutex);
@@ -1974,7 +1980,8 @@
 	 * If we failed before we got a uuid_str yet, we can't stop
 	 * heartbeat.  Otherwise, do it.
 	 */
-	if (!mnt_err && !ocfs2_mount_local(osb) && osb->uuid_str)
+	if (!mnt_err && !ocfs2_mount_local(osb) && osb->uuid_str &&
+	    !ocfs2_is_hard_readonly(osb))
 		hangup_needed = 1;
 
 	if (osb->cconn)
@@ -2353,7 +2360,7 @@
 		mlog_errno(status);
 		goto bail;
 	}
-	cleancache_init_shared_fs((char *)&uuid_net_key, sb);
+	cleancache_init_shared_fs((char *)&di->id2.i_super.s_uuid, sb);
 
 bail:
 	return status;
@@ -2462,8 +2469,8 @@
 			goto finally;
 		}
 	} else {
-		mlog(ML_NOTICE, "File system was not unmounted cleanly, "
-		     "recovering volume.\n");
+		printk(KERN_NOTICE "ocfs2: File system on device (%s) was not "
+		       "unmounted cleanly, recovering it.\n", osb->dev_str);
 	}
 
 	local = ocfs2_mount_local(osb);
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index 194fb22..aa9e877 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -2376,16 +2376,18 @@
 		}
 
 		ret = ocfs2_xattr_value_truncate(inode, vb, 0, &ctxt);
-		if (ret < 0) {
-			mlog_errno(ret);
-			break;
-		}
 
 		ocfs2_commit_trans(osb, ctxt.handle);
 		if (ctxt.meta_ac) {
 			ocfs2_free_alloc_context(ctxt.meta_ac);
 			ctxt.meta_ac = NULL;
 		}
+
+		if (ret < 0) {
+			mlog_errno(ret);
+			break;
+		}
+
 	}
 
 	if (ctxt.meta_ac)
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 2db1bd3..851ba3d 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1652,46 +1652,12 @@
 	return error;
 }
 
-static int proc_pid_fd_link_getattr(struct vfsmount *mnt, struct dentry *dentry,
-		struct kstat *stat)
-{
-	struct inode *inode = dentry->d_inode;
-	struct task_struct *task = get_proc_task(inode);
-	int rc;
-
-	if (task == NULL)
-		return -ESRCH;
-
-	rc = -EACCES;
-	if (lock_trace(task))
-		goto out_task;
-
-	generic_fillattr(inode, stat);
-	unlock_trace(task);
-	rc = 0;
-out_task:
-	put_task_struct(task);
-	return rc;
-}
-
 static const struct inode_operations proc_pid_link_inode_operations = {
 	.readlink	= proc_pid_readlink,
 	.follow_link	= proc_pid_follow_link,
 	.setattr	= proc_setattr,
 };
 
-static const struct inode_operations proc_fdinfo_link_inode_operations = {
-	.setattr	= proc_setattr,
-	.getattr	= proc_pid_fd_link_getattr,
-};
-
-static const struct inode_operations proc_fd_link_inode_operations = {
-	.readlink	= proc_pid_readlink,
-	.follow_link	= proc_pid_follow_link,
-	.setattr	= proc_setattr,
-	.getattr	= proc_pid_fd_link_getattr,
-};
-
 
 /* building an inode */
 
@@ -1923,61 +1889,49 @@
 
 static int proc_fd_info(struct inode *inode, struct path *path, char *info)
 {
-	struct task_struct *task;
-	struct files_struct *files;
+	struct task_struct *task = get_proc_task(inode);
+	struct files_struct *files = NULL;
 	struct file *file;
 	int fd = proc_fd(inode);
-	int rc;
 
-	task = get_proc_task(inode);
-	if (!task)
-		return -ENOENT;
+	if (task) {
+		files = get_files_struct(task);
+		put_task_struct(task);
+	}
+	if (files) {
+		/*
+		 * We are not taking a ref to the file structure, so we must
+		 * hold ->file_lock.
+		 */
+		spin_lock(&files->file_lock);
+		file = fcheck_files(files, fd);
+		if (file) {
+			unsigned int f_flags;
+			struct fdtable *fdt;
 
-	rc = -EACCES;
-	if (lock_trace(task))
-		goto out_task;
+			fdt = files_fdtable(files);
+			f_flags = file->f_flags & ~O_CLOEXEC;
+			if (FD_ISSET(fd, fdt->close_on_exec))
+				f_flags |= O_CLOEXEC;
 
-	rc = -ENOENT;
-	files = get_files_struct(task);
-	if (files == NULL)
-		goto out_unlock;
-
-	/*
-	 * We are not taking a ref to the file structure, so we must
-	 * hold ->file_lock.
-	 */
-	spin_lock(&files->file_lock);
-	file = fcheck_files(files, fd);
-	if (file) {
-		unsigned int f_flags;
-		struct fdtable *fdt;
-
-		fdt = files_fdtable(files);
-		f_flags = file->f_flags & ~O_CLOEXEC;
-		if (FD_ISSET(fd, fdt->close_on_exec))
-			f_flags |= O_CLOEXEC;
-
-		if (path) {
-			*path = file->f_path;
-			path_get(&file->f_path);
+			if (path) {
+				*path = file->f_path;
+				path_get(&file->f_path);
+			}
+			if (info)
+				snprintf(info, PROC_FDINFO_MAX,
+					 "pos:\t%lli\n"
+					 "flags:\t0%o\n",
+					 (long long) file->f_pos,
+					 f_flags);
+			spin_unlock(&files->file_lock);
+			put_files_struct(files);
+			return 0;
 		}
-		if (info)
-			snprintf(info, PROC_FDINFO_MAX,
-				 "pos:\t%lli\n"
-				 "flags:\t0%o\n",
-				 (long long) file->f_pos,
-				 f_flags);
-		rc = 0;
-	} else
-		rc = -ENOENT;
-	spin_unlock(&files->file_lock);
-	put_files_struct(files);
-
-out_unlock:
-	unlock_trace(task);
-out_task:
-	put_task_struct(task);
-	return rc;
+		spin_unlock(&files->file_lock);
+		put_files_struct(files);
+	}
+	return -ENOENT;
 }
 
 static int proc_fd_link(struct inode *inode, struct path *path)
@@ -2072,7 +2026,7 @@
 	spin_unlock(&files->file_lock);
 	put_files_struct(files);
 
-	inode->i_op = &proc_fd_link_inode_operations;
+	inode->i_op = &proc_pid_link_inode_operations;
 	inode->i_size = 64;
 	ei->op.proc_get_link = proc_fd_link;
 	d_set_d_op(dentry, &tid_fd_dentry_operations);
@@ -2104,12 +2058,7 @@
 	if (fd == ~0U)
 		goto out;
 
-	result = ERR_PTR(-EACCES);
-	if (lock_trace(task))
-		goto out;
-
 	result = instantiate(dir, dentry, task, &fd);
-	unlock_trace(task);
 out:
 	put_task_struct(task);
 out_no_task:
@@ -2129,28 +2078,23 @@
 	retval = -ENOENT;
 	if (!p)
 		goto out_no_task;
-
-	retval = -EACCES;
-	if (lock_trace(p))
-		goto out;
-
 	retval = 0;
 
 	fd = filp->f_pos;
 	switch (fd) {
 		case 0:
 			if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0)
-				goto out_unlock;
+				goto out;
 			filp->f_pos++;
 		case 1:
 			ino = parent_ino(dentry);
 			if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0)
-				goto out_unlock;
+				goto out;
 			filp->f_pos++;
 		default:
 			files = get_files_struct(p);
 			if (!files)
-				goto out_unlock;
+				goto out;
 			rcu_read_lock();
 			for (fd = filp->f_pos-2;
 			     fd < files_fdtable(files)->max_fds;
@@ -2174,9 +2118,6 @@
 			rcu_read_unlock();
 			put_files_struct(files);
 	}
-
-out_unlock:
-	unlock_trace(p);
 out:
 	put_task_struct(p);
 out_no_task:
@@ -2254,7 +2195,6 @@
 	ei->fd = fd;
 	inode->i_mode = S_IFREG | S_IRUSR;
 	inode->i_fop = &proc_fdinfo_file_operations;
-	inode->i_op = &proc_fdinfo_link_inode_operations;
 	d_set_d_op(dentry, &tid_fd_dentry_operations);
 	d_add(dentry, inode);
 	/* Close the race of the process dying before we return the dentry */
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index 5861741..80e4645 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -131,12 +131,13 @@
 		K(i.freeswap),
 		K(global_page_state(NR_FILE_DIRTY)),
 		K(global_page_state(NR_WRITEBACK)),
-		K(global_page_state(NR_ANON_PAGES)
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
+		K(global_page_state(NR_ANON_PAGES)
 		  + global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) *
-		  HPAGE_PMD_NR
+		  HPAGE_PMD_NR),
+#else
+		K(global_page_state(NR_ANON_PAGES)),
 #endif
-		  ),
 		K(global_page_state(NR_FILE_MAPPED)),
 		K(global_page_state(NR_SHMEM)),
 		K(global_page_state(NR_SLAB_RECLAIMABLE) +
diff --git a/fs/proc/root.c b/fs/proc/root.c
index 9a8a2b7..03102d9 100644
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -91,20 +91,18 @@
 
 void __init proc_root_init(void)
 {
-	struct vfsmount *mnt;
 	int err;
 
 	proc_init_inodecache();
 	err = register_filesystem(&proc_fs_type);
 	if (err)
 		return;
-	mnt = kern_mount_data(&proc_fs_type, &init_pid_ns);
-	if (IS_ERR(mnt)) {
+	err = pid_ns_prepare_proc(&init_pid_ns);
+	if (err) {
 		unregister_filesystem(&proc_fs_type);
 		return;
 	}
 
-	init_pid_ns.proc_mnt = mnt;
 	proc_symlink("mounts", NULL, "self/mounts");
 
 	proc_net_init();
@@ -209,5 +207,5 @@
 
 void pid_ns_release_proc(struct pid_namespace *ns)
 {
-	mntput(ns->proc_mnt);
+	kern_unmount(ns->proc_mnt);
 }
diff --git a/fs/proc/stat.c b/fs/proc/stat.c
index 42b274d..2a30d67 100644
--- a/fs/proc/stat.c
+++ b/fs/proc/stat.c
@@ -32,7 +32,7 @@
 		idle = kstat_cpu(cpu).cpustat.idle;
 		idle = cputime64_add(idle, arch_idle_time(cpu));
 	} else
-		idle = usecs_to_cputime(idle_time);
+		idle = nsecs_to_jiffies64(1000 * idle_time);
 
 	return idle;
 }
@@ -46,7 +46,7 @@
 		/* !NO_HZ so we can rely on cpustat.iowait */
 		iowait = kstat_cpu(cpu).cpustat.iowait;
 	else
-		iowait = usecs_to_cputime(iowait_time);
+		iowait = nsecs_to_jiffies64(1000 * iowait_time);
 
 	return iowait;
 }
diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c
index 2bd620f..57bbf90 100644
--- a/fs/pstore/platform.c
+++ b/fs/pstore/platform.c
@@ -167,6 +167,7 @@
 	}
 
 	psinfo = psi;
+	mutex_init(&psinfo->read_mutex);
 	spin_unlock(&pstore_lock);
 
 	if (owner && !try_module_get(owner)) {
@@ -195,30 +196,32 @@
 void pstore_get_records(int quiet)
 {
 	struct pstore_info *psi = psinfo;
+	char			*buf = NULL;
 	ssize_t			size;
 	u64			id;
 	enum pstore_type_id	type;
 	struct timespec		time;
 	int			failed = 0, rc;
-	unsigned long		flags;
 
 	if (!psi)
 		return;
 
-	spin_lock_irqsave(&psinfo->buf_lock, flags);
+	mutex_lock(&psi->read_mutex);
 	rc = psi->open(psi);
 	if (rc)
 		goto out;
 
-	while ((size = psi->read(&id, &type, &time, psi)) > 0) {
-		rc = pstore_mkfile(type, psi->name, id, psi->buf, (size_t)size,
+	while ((size = psi->read(&id, &type, &time, &buf, psi)) > 0) {
+		rc = pstore_mkfile(type, psi->name, id, buf, (size_t)size,
 				  time, psi);
+		kfree(buf);
+		buf = NULL;
 		if (rc && (rc != -EEXIST || !quiet))
 			failed++;
 	}
 	psi->close(psi);
 out:
-	spin_unlock_irqrestore(&psinfo->buf_lock, flags);
+	mutex_unlock(&psi->read_mutex);
 
 	if (failed)
 		printk(KERN_WARNING "pstore: failed to load %d record(s) from '%s'\n",
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 05d6b0e..dba43c3 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -449,8 +449,6 @@
 
 /*
  * Same as seq_path, but relative to supplied root.
- *
- * root may be changed, see __d_path().
  */
 int seq_path_root(struct seq_file *m, struct path *path, struct path *root,
 		  char *esc)
@@ -463,6 +461,8 @@
 		char *p;
 
 		p = __d_path(path, root, buf, size);
+		if (!p)
+			return SEQ_SKIP;
 		res = PTR_ERR(p);
 		if (!IS_ERR(p)) {
 			char *end = mangle_path(buf, p, esc);
@@ -474,7 +474,7 @@
 	}
 	seq_commit(m, res);
 
-	return res < 0 ? res : 0;
+	return res < 0 && res != -ENAMETOOLONG ? res : 0;
 }
 
 /*
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 20403dc..ae0e76b 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -2264,19 +2264,12 @@
 		return -EINVAL;
 	}
 
-	err = register_filesystem(&ubifs_fs_type);
-	if (err) {
-		ubifs_err("cannot register file system, error %d", err);
-		return err;
-	}
-
-	err = -ENOMEM;
 	ubifs_inode_slab = kmem_cache_create("ubifs_inode_slab",
 				sizeof(struct ubifs_inode), 0,
 				SLAB_MEM_SPREAD | SLAB_RECLAIM_ACCOUNT,
 				&inode_slab_ctor);
 	if (!ubifs_inode_slab)
-		goto out_reg;
+		return -ENOMEM;
 
 	register_shrinker(&ubifs_shrinker_info);
 
@@ -2288,15 +2281,20 @@
 	if (err)
 		goto out_compr;
 
+	err = register_filesystem(&ubifs_fs_type);
+	if (err) {
+		ubifs_err("cannot register file system, error %d", err);
+		goto out_dbg;
+	}
 	return 0;
 
+out_dbg:
+	dbg_debugfs_exit();
 out_compr:
 	ubifs_compressors_exit();
 out_shrinker:
 	unregister_shrinker(&ubifs_shrinker_info);
 	kmem_cache_destroy(ubifs_inode_slab);
-out_reg:
-	unregister_filesystem(&ubifs_fs_type);
 	return err;
 }
 /* late_initcall to let compressors initialize first */
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index b6c4b37..76e4266 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -42,6 +42,8 @@
 	int count, i;
 
 	count = be32_to_cpu(aclp->acl_cnt);
+	if (count > XFS_ACL_MAX_ENTRIES)
+		return ERR_PTR(-EFSCORRUPTED);
 
 	acl = posix_acl_alloc(count, GFP_KERNEL);
 	if (!acl)
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 33b1331..574d4ee 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -189,7 +189,7 @@
 	int		error = 0;
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-		error = -EIO;
+		ioend->io_error = -EIO;
 		goto done;
 	}
 	if (ioend->io_error)
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c
index d4906e7..c1b55e5 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/xfs_attr_leaf.c
@@ -110,6 +110,7 @@
 /*
  * Query whether the requested number of additional bytes of extended
  * attribute space will be able to fit inline.
+ *
  * Returns zero if not, else the di_forkoff fork offset to be used in the
  * literal area for attribute data once the new bytes have been added.
  *
@@ -122,7 +123,7 @@
 	int offset;
 	int minforkoff;	/* lower limit on valid forkoff locations */
 	int maxforkoff;	/* upper limit on valid forkoff locations */
-	int dsize;	
+	int dsize;
 	xfs_mount_t *mp = dp->i_mount;
 
 	offset = (XFS_LITINO(mp) - bytes) >> 3; /* rounded down */
@@ -136,47 +137,60 @@
 		return (offset >= minforkoff) ? minforkoff : 0;
 	}
 
-	if (!(mp->m_flags & XFS_MOUNT_ATTR2)) {
-		if (bytes <= XFS_IFORK_ASIZE(dp))
-			return dp->i_d.di_forkoff;
+	/*
+	 * If the requested numbers of bytes is smaller or equal to the
+	 * current attribute fork size we can always proceed.
+	 *
+	 * Note that if_bytes in the data fork might actually be larger than
+	 * the current data fork size is due to delalloc extents. In that
+	 * case either the extent count will go down when they are converted
+	 * to real extents, or the delalloc conversion will take care of the
+	 * literal area rebalancing.
+	 */
+	if (bytes <= XFS_IFORK_ASIZE(dp))
+		return dp->i_d.di_forkoff;
+
+	/*
+	 * For attr2 we can try to move the forkoff if there is space in the
+	 * literal area, but for the old format we are done if there is no
+	 * space in the fixed attribute fork.
+	 */
+	if (!(mp->m_flags & XFS_MOUNT_ATTR2))
 		return 0;
-	}
 
 	dsize = dp->i_df.if_bytes;
-	
+
 	switch (dp->i_d.di_format) {
 	case XFS_DINODE_FMT_EXTENTS:
-		/* 
+		/*
 		 * If there is no attr fork and the data fork is extents, 
-		 * determine if creating the default attr fork will result 
-		 * in the extents form migrating to btree. If so, the 
-		 * minimum offset only needs to be the space required for 
+		 * determine if creating the default attr fork will result
+		 * in the extents form migrating to btree. If so, the
+		 * minimum offset only needs to be the space required for
 		 * the btree root.
-		 */ 
+		 */
 		if (!dp->i_d.di_forkoff && dp->i_df.if_bytes >
 		    xfs_default_attroffset(dp))
 			dsize = XFS_BMDR_SPACE_CALC(MINDBTPTRS);
 		break;
-		
 	case XFS_DINODE_FMT_BTREE:
 		/*
-		 * If have data btree then keep forkoff if we have one,
-		 * otherwise we are adding a new attr, so then we set 
-		 * minforkoff to where the btree root can finish so we have 
+		 * If we have a data btree then keep forkoff if we have one,
+		 * otherwise we are adding a new attr, so then we set
+		 * minforkoff to where the btree root can finish so we have
 		 * plenty of room for attrs
 		 */
 		if (dp->i_d.di_forkoff) {
-			if (offset < dp->i_d.di_forkoff) 
+			if (offset < dp->i_d.di_forkoff)
 				return 0;
-			else 
-				return dp->i_d.di_forkoff;
-		} else
-			dsize = XFS_BMAP_BROOT_SPACE(dp->i_df.if_broot);
+			return dp->i_d.di_forkoff;
+		}
+		dsize = XFS_BMAP_BROOT_SPACE(dp->i_df.if_broot);
 		break;
 	}
-	
-	/* 
-	 * A data fork btree root must have space for at least 
+
+	/*
+	 * A data fork btree root must have space for at least
 	 * MINDBTPTRS key/ptr pairs if the data fork is small or empty.
 	 */
 	minforkoff = MAX(dsize, XFS_BMDR_SPACE_CALC(MINDBTPTRS));
@@ -186,10 +200,10 @@
 	maxforkoff = XFS_LITINO(mp) - XFS_BMDR_SPACE_CALC(MINABTPTRS);
 	maxforkoff = maxforkoff >> 3;	/* rounded down */
 
-	if (offset >= minforkoff && offset < maxforkoff)
-		return offset;
 	if (offset >= maxforkoff)
 		return maxforkoff;
+	if (offset >= minforkoff)
+		return offset;
 	return 0;
 }
 
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index c68baeb..d0ab788 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -2383,6 +2383,8 @@
 	int		tryagain;
 	int		error;
 
+	ASSERT(ap->length);
+
 	mp = ap->ip->i_mount;
 	align = ap->userdata ? xfs_get_extsz_hint(ap->ip) : 0;
 	if (unlikely(align)) {
@@ -4629,6 +4631,8 @@
 	int			error;
 	int			rt;
 
+	ASSERT(bma->length > 0);
+
 	rt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(bma->ip);
 
 	/*
@@ -4849,6 +4853,7 @@
 	ASSERT(*nmap <= XFS_BMAP_MAX_NMAP);
 	ASSERT(!(flags & XFS_BMAPI_IGSTATE));
 	ASSERT(tp != NULL);
+	ASSERT(len > 0);
 
 	whichfork = (flags & XFS_BMAPI_ATTRFORK) ?
 		XFS_ATTR_FORK : XFS_DATA_FORK;
@@ -4918,9 +4923,22 @@
 			bma.eof = eof;
 			bma.conv = !!(flags & XFS_BMAPI_CONVERT);
 			bma.wasdel = wasdelay;
-			bma.length = len;
 			bma.offset = bno;
 
+			/*
+			 * There's a 32/64 bit type mismatch between the
+			 * allocation length request (which can be 64 bits in
+			 * length) and the bma length request, which is
+			 * xfs_extlen_t and therefore 32 bits. Hence we have to
+			 * check for 32-bit overflows and handle them here.
+			 */
+			if (len > (xfs_filblks_t)MAXEXTLEN)
+				bma.length = MAXEXTLEN;
+			else
+				bma.length = len;
+
+			ASSERT(len > 0);
+			ASSERT(bma.length > 0);
 			error = xfs_bmapi_allocate(&bma, flags);
 			if (error)
 				goto error0;
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 1a35138..eac97ef 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -656,7 +656,7 @@
 /*
  * This is the ops vector shared by all buf log items.
  */
-static struct xfs_item_ops xfs_buf_item_ops = {
+static const struct xfs_item_ops xfs_buf_item_ops = {
 	.iop_size	= xfs_buf_item_size,
 	.iop_format	= xfs_buf_item_format,
 	.iop_pin	= xfs_buf_item_pin,
diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c
index bb3f71d..0dee0b7 100644
--- a/fs/xfs/xfs_dquot_item.c
+++ b/fs/xfs/xfs_dquot_item.c
@@ -295,7 +295,7 @@
 /*
  * This is the ops vector for dquots
  */
-static struct xfs_item_ops xfs_dquot_item_ops = {
+static const struct xfs_item_ops xfs_dquot_item_ops = {
 	.iop_size	= xfs_qm_dquot_logitem_size,
 	.iop_format	= xfs_qm_dquot_logitem_format,
 	.iop_pin	= xfs_qm_dquot_logitem_pin,
@@ -483,7 +483,7 @@
 {
 }
 
-static struct xfs_item_ops xfs_qm_qoffend_logitem_ops = {
+static const struct xfs_item_ops xfs_qm_qoffend_logitem_ops = {
 	.iop_size	= xfs_qm_qoff_logitem_size,
 	.iop_format	= xfs_qm_qoff_logitem_format,
 	.iop_pin	= xfs_qm_qoff_logitem_pin,
@@ -498,7 +498,7 @@
 /*
  * This is the ops vector shared by all quotaoff-start log items.
  */
-static struct xfs_item_ops xfs_qm_qoff_logitem_ops = {
+static const struct xfs_item_ops xfs_qm_qoff_logitem_ops = {
 	.iop_size	= xfs_qm_qoff_logitem_size,
 	.iop_format	= xfs_qm_qoff_logitem_format,
 	.iop_pin	= xfs_qm_qoff_logitem_pin,
diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c
index da10897..558910f 100644
--- a/fs/xfs/xfs_export.c
+++ b/fs/xfs/xfs_export.c
@@ -98,22 +98,22 @@
 	switch (fileid_type) {
 	case FILEID_INO32_GEN_PARENT:
 		spin_lock(&dentry->d_lock);
-		fid->i32.parent_ino = dentry->d_parent->d_inode->i_ino;
+		fid->i32.parent_ino = XFS_I(dentry->d_parent->d_inode)->i_ino;
 		fid->i32.parent_gen = dentry->d_parent->d_inode->i_generation;
 		spin_unlock(&dentry->d_lock);
 		/*FALLTHRU*/
 	case FILEID_INO32_GEN:
-		fid->i32.ino = inode->i_ino;
+		fid->i32.ino = XFS_I(inode)->i_ino;
 		fid->i32.gen = inode->i_generation;
 		break;
 	case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG:
 		spin_lock(&dentry->d_lock);
-		fid64->parent_ino = dentry->d_parent->d_inode->i_ino;
+		fid64->parent_ino = XFS_I(dentry->d_parent->d_inode)->i_ino;
 		fid64->parent_gen = dentry->d_parent->d_inode->i_generation;
 		spin_unlock(&dentry->d_lock);
 		/*FALLTHRU*/
 	case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG:
-		fid64->ino = inode->i_ino;
+		fid64->ino = XFS_I(inode)->i_ino;
 		fid64->gen = inode->i_generation;
 		break;
 	}
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index d22e626..35c2aff 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -217,7 +217,7 @@
 /*
  * This is the ops vector shared by all efi log items.
  */
-static struct xfs_item_ops xfs_efi_item_ops = {
+static const struct xfs_item_ops xfs_efi_item_ops = {
 	.iop_size	= xfs_efi_item_size,
 	.iop_format	= xfs_efi_item_format,
 	.iop_pin	= xfs_efi_item_pin,
@@ -477,7 +477,7 @@
 /*
  * This is the ops vector shared by all efd log items.
  */
-static struct xfs_item_ops xfs_efd_item_ops = {
+static const struct xfs_item_ops xfs_efd_item_ops = {
 	.iop_size	= xfs_efd_item_size,
 	.iop_format	= xfs_efd_item_format,
 	.iop_pin	= xfs_efd_item_pin,
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index c0237c6..755ee81 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -2835,6 +2835,27 @@
 	return XFS_ERROR(EFSCORRUPTED);
 }
 
+void
+xfs_promote_inode(
+	struct xfs_inode	*ip)
+{
+	struct xfs_buf		*bp;
+
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
+
+	bp = xfs_incore(ip->i_mount->m_ddev_targp, ip->i_imap.im_blkno,
+			ip->i_imap.im_len, XBF_TRYLOCK);
+	if (!bp)
+		return;
+
+	if (XFS_BUF_ISDELAYWRITE(bp)) {
+		xfs_buf_delwri_promote(bp);
+		wake_up_process(ip->i_mount->m_ddev_targp->bt_task);
+	}
+
+	xfs_buf_relse(bp);
+}
+
 /*
  * Return a pointer to the extent record at file index idx.
  */
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 760140d..b4cd473 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -498,6 +498,7 @@
 void		xfs_iext_realloc(xfs_inode_t *, int, int);
 void		xfs_iunpin_wait(xfs_inode_t *);
 int		xfs_iflush(xfs_inode_t *, uint);
+void		xfs_promote_inode(struct xfs_inode *);
 void		xfs_lock_inodes(xfs_inode_t **, int, uint);
 void		xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint);
 
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index b7cf21b..abaafdb 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -795,7 +795,7 @@
 /*
  * This is the ops vector shared by all buf log items.
  */
-static struct xfs_item_ops xfs_inode_item_ops = {
+static const struct xfs_item_ops xfs_inode_item_ops = {
 	.iop_size	= xfs_inode_item_size,
 	.iop_format	= xfs_inode_item_format,
 	.iop_pin	= xfs_inode_item_pin,
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 2758a62..34817ad 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -150,6 +150,117 @@
 	} while (head_val != old);
 }
 
+STATIC bool
+xlog_reserveq_wake(
+	struct log		*log,
+	int			*free_bytes)
+{
+	struct xlog_ticket	*tic;
+	int			need_bytes;
+
+	list_for_each_entry(tic, &log->l_reserveq, t_queue) {
+		if (tic->t_flags & XLOG_TIC_PERM_RESERV)
+			need_bytes = tic->t_unit_res * tic->t_cnt;
+		else
+			need_bytes = tic->t_unit_res;
+
+		if (*free_bytes < need_bytes)
+			return false;
+		*free_bytes -= need_bytes;
+
+		trace_xfs_log_grant_wake_up(log, tic);
+		wake_up(&tic->t_wait);
+	}
+
+	return true;
+}
+
+STATIC bool
+xlog_writeq_wake(
+	struct log		*log,
+	int			*free_bytes)
+{
+	struct xlog_ticket	*tic;
+	int			need_bytes;
+
+	list_for_each_entry(tic, &log->l_writeq, t_queue) {
+		ASSERT(tic->t_flags & XLOG_TIC_PERM_RESERV);
+
+		need_bytes = tic->t_unit_res;
+
+		if (*free_bytes < need_bytes)
+			return false;
+		*free_bytes -= need_bytes;
+
+		trace_xfs_log_regrant_write_wake_up(log, tic);
+		wake_up(&tic->t_wait);
+	}
+
+	return true;
+}
+
+STATIC int
+xlog_reserveq_wait(
+	struct log		*log,
+	struct xlog_ticket	*tic,
+	int			need_bytes)
+{
+	list_add_tail(&tic->t_queue, &log->l_reserveq);
+
+	do {
+		if (XLOG_FORCED_SHUTDOWN(log))
+			goto shutdown;
+		xlog_grant_push_ail(log, need_bytes);
+
+		XFS_STATS_INC(xs_sleep_logspace);
+		trace_xfs_log_grant_sleep(log, tic);
+
+		xlog_wait(&tic->t_wait, &log->l_grant_reserve_lock);
+		trace_xfs_log_grant_wake(log, tic);
+
+		spin_lock(&log->l_grant_reserve_lock);
+		if (XLOG_FORCED_SHUTDOWN(log))
+			goto shutdown;
+	} while (xlog_space_left(log, &log->l_grant_reserve_head) < need_bytes);
+
+	list_del_init(&tic->t_queue);
+	return 0;
+shutdown:
+	list_del_init(&tic->t_queue);
+	return XFS_ERROR(EIO);
+}
+
+STATIC int
+xlog_writeq_wait(
+	struct log		*log,
+	struct xlog_ticket	*tic,
+	int			need_bytes)
+{
+	list_add_tail(&tic->t_queue, &log->l_writeq);
+
+	do {
+		if (XLOG_FORCED_SHUTDOWN(log))
+			goto shutdown;
+		xlog_grant_push_ail(log, need_bytes);
+
+		XFS_STATS_INC(xs_sleep_logspace);
+		trace_xfs_log_regrant_write_sleep(log, tic);
+
+		xlog_wait(&tic->t_wait, &log->l_grant_write_lock);
+		trace_xfs_log_regrant_write_wake(log, tic);
+
+		spin_lock(&log->l_grant_write_lock);
+		if (XLOG_FORCED_SHUTDOWN(log))
+			goto shutdown;
+	} while (xlog_space_left(log, &log->l_grant_write_head) < need_bytes);
+
+	list_del_init(&tic->t_queue);
+	return 0;
+shutdown:
+	list_del_init(&tic->t_queue);
+	return XFS_ERROR(EIO);
+}
+
 static void
 xlog_tic_reset_res(xlog_ticket_t *tic)
 {
@@ -350,8 +461,19 @@
 		retval = xlog_grant_log_space(log, internal_ticket);
 	}
 
+	if (unlikely(retval)) {
+		/*
+		 * If we are failing, make sure the ticket doesn't have any
+		 * current reservations.  We don't want to add this back
+		 * when the ticket/ transaction gets cancelled.
+		 */
+		internal_ticket->t_curr_res = 0;
+		/* ungrant will give back unit_res * t_cnt. */
+		internal_ticket->t_cnt = 0;
+	}
+
 	return retval;
-}	/* xfs_log_reserve */
+}
 
 
 /*
@@ -626,7 +748,7 @@
 	struct xfs_mount	*mp,
 	struct xfs_log_item	*item,
 	int			type,
-	struct xfs_item_ops	*ops)
+	const struct xfs_item_ops *ops)
 {
 	item->li_mountp = mp;
 	item->li_ailp = mp->m_ail;
@@ -2481,8 +2603,8 @@
 /*
  * Atomically get the log space required for a log ticket.
  *
- * Once a ticket gets put onto the reserveq, it will only return after
- * the needed reservation is satisfied.
+ * Once a ticket gets put onto the reserveq, it will only return after the
+ * needed reservation is satisfied.
  *
  * This function is structured so that it has a lock free fast path. This is
  * necessary because every new transaction reservation will come through this
@@ -2490,113 +2612,53 @@
  * every pass.
  *
  * As tickets are only ever moved on and off the reserveq under the
- * l_grant_reserve_lock, we only need to take that lock if we are going
- * to add the ticket to the queue and sleep. We can avoid taking the lock if the
- * ticket was never added to the reserveq because the t_queue list head will be
- * empty and we hold the only reference to it so it can safely be checked
- * unlocked.
+ * l_grant_reserve_lock, we only need to take that lock if we are going to add
+ * the ticket to the queue and sleep. We can avoid taking the lock if the ticket
+ * was never added to the reserveq because the t_queue list head will be empty
+ * and we hold the only reference to it so it can safely be checked unlocked.
  */
 STATIC int
-xlog_grant_log_space(xlog_t	   *log,
-		     xlog_ticket_t *tic)
+xlog_grant_log_space(
+	struct log		*log,
+	struct xlog_ticket	*tic)
 {
-	int		 free_bytes;
-	int		 need_bytes;
+	int			free_bytes, need_bytes;
+	int			error = 0;
 
-#ifdef DEBUG
-	if (log->l_flags & XLOG_ACTIVE_RECOVERY)
-		panic("grant Recovery problem");
-#endif
+	ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY));
 
 	trace_xfs_log_grant_enter(log, tic);
 
+	/*
+	 * If there are other waiters on the queue then give them a chance at
+	 * logspace before us.  Wake up the first waiters, if we do not wake
+	 * up all the waiters then go to sleep waiting for more free space,
+	 * otherwise try to get some space for this transaction.
+	 */
 	need_bytes = tic->t_unit_res;
 	if (tic->t_flags & XFS_LOG_PERM_RESERV)
 		need_bytes *= tic->t_ocnt;
-
-	/* something is already sleeping; insert new transaction at end */
+	free_bytes = xlog_space_left(log, &log->l_grant_reserve_head);
 	if (!list_empty_careful(&log->l_reserveq)) {
 		spin_lock(&log->l_grant_reserve_lock);
-		/* recheck the queue now we are locked */
-		if (list_empty(&log->l_reserveq)) {
-			spin_unlock(&log->l_grant_reserve_lock);
-			goto redo;
-		}
-		list_add_tail(&tic->t_queue, &log->l_reserveq);
-
-		trace_xfs_log_grant_sleep1(log, tic);
-
-		/*
-		 * Gotta check this before going to sleep, while we're
-		 * holding the grant lock.
-		 */
-		if (XLOG_FORCED_SHUTDOWN(log))
-			goto error_return;
-
-		XFS_STATS_INC(xs_sleep_logspace);
-		xlog_wait(&tic->t_wait, &log->l_grant_reserve_lock);
-
-		/*
-		 * If we got an error, and the filesystem is shutting down,
-		 * we'll catch it down below. So just continue...
-		 */
-		trace_xfs_log_grant_wake1(log, tic);
-	}
-
-redo:
-	if (XLOG_FORCED_SHUTDOWN(log))
-		goto error_return_unlocked;
-
-	free_bytes = xlog_space_left(log, &log->l_grant_reserve_head);
-	if (free_bytes < need_bytes) {
+		if (!xlog_reserveq_wake(log, &free_bytes) ||
+		    free_bytes < need_bytes)
+			error = xlog_reserveq_wait(log, tic, need_bytes);
+		spin_unlock(&log->l_grant_reserve_lock);
+	} else if (free_bytes < need_bytes) {
 		spin_lock(&log->l_grant_reserve_lock);
-		if (list_empty(&tic->t_queue))
-			list_add_tail(&tic->t_queue, &log->l_reserveq);
-
-		trace_xfs_log_grant_sleep2(log, tic);
-
-		if (XLOG_FORCED_SHUTDOWN(log))
-			goto error_return;
-
-		xlog_grant_push_ail(log, need_bytes);
-
-		XFS_STATS_INC(xs_sleep_logspace);
-		xlog_wait(&tic->t_wait, &log->l_grant_reserve_lock);
-
-		trace_xfs_log_grant_wake2(log, tic);
-		goto redo;
-	}
-
-	if (!list_empty(&tic->t_queue)) {
-		spin_lock(&log->l_grant_reserve_lock);
-		list_del_init(&tic->t_queue);
+		error = xlog_reserveq_wait(log, tic, need_bytes);
 		spin_unlock(&log->l_grant_reserve_lock);
 	}
+	if (error)
+		return error;
 
-	/* we've got enough space */
 	xlog_grant_add_space(log, &log->l_grant_reserve_head, need_bytes);
 	xlog_grant_add_space(log, &log->l_grant_write_head, need_bytes);
 	trace_xfs_log_grant_exit(log, tic);
 	xlog_verify_grant_tail(log);
 	return 0;
-
-error_return_unlocked:
-	spin_lock(&log->l_grant_reserve_lock);
-error_return:
-	list_del_init(&tic->t_queue);
-	spin_unlock(&log->l_grant_reserve_lock);
-	trace_xfs_log_grant_error(log, tic);
-
-	/*
-	 * If we are failing, make sure the ticket doesn't have any
-	 * current reservations. We don't want to add this back when
-	 * the ticket/transaction gets cancelled.
-	 */
-	tic->t_curr_res = 0;
-	tic->t_cnt = 0; /* ungrant will give back unit_res * t_cnt. */
-	return XFS_ERROR(EIO);
-}	/* xlog_grant_log_space */
-
+}
 
 /*
  * Replenish the byte reservation required by moving the grant write head.
@@ -2605,10 +2667,12 @@
  * free fast path.
  */
 STATIC int
-xlog_regrant_write_log_space(xlog_t	   *log,
-			     xlog_ticket_t *tic)
+xlog_regrant_write_log_space(
+	struct log		*log,
+	struct xlog_ticket	*tic)
 {
-	int		free_bytes, need_bytes;
+	int			free_bytes, need_bytes;
+	int			error = 0;
 
 	tic->t_curr_res = tic->t_unit_res;
 	xlog_tic_reset_res(tic);
@@ -2616,104 +2680,38 @@
 	if (tic->t_cnt > 0)
 		return 0;
 
-#ifdef DEBUG
-	if (log->l_flags & XLOG_ACTIVE_RECOVERY)
-		panic("regrant Recovery problem");
-#endif
+	ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY));
 
 	trace_xfs_log_regrant_write_enter(log, tic);
-	if (XLOG_FORCED_SHUTDOWN(log))
-		goto error_return_unlocked;
 
-	/* If there are other waiters on the queue then give them a
-	 * chance at logspace before us. Wake up the first waiters,
-	 * if we do not wake up all the waiters then go to sleep waiting
-	 * for more free space, otherwise try to get some space for
-	 * this transaction.
+	/*
+	 * If there are other waiters on the queue then give them a chance at
+	 * logspace before us.  Wake up the first waiters, if we do not wake
+	 * up all the waiters then go to sleep waiting for more free space,
+	 * otherwise try to get some space for this transaction.
 	 */
 	need_bytes = tic->t_unit_res;
-	if (!list_empty_careful(&log->l_writeq)) {
-		struct xlog_ticket *ntic;
-
-		spin_lock(&log->l_grant_write_lock);
-		free_bytes = xlog_space_left(log, &log->l_grant_write_head);
-		list_for_each_entry(ntic, &log->l_writeq, t_queue) {
-			ASSERT(ntic->t_flags & XLOG_TIC_PERM_RESERV);
-
-			if (free_bytes < ntic->t_unit_res)
-				break;
-			free_bytes -= ntic->t_unit_res;
-			wake_up(&ntic->t_wait);
-		}
-
-		if (ntic != list_first_entry(&log->l_writeq,
-						struct xlog_ticket, t_queue)) {
-			if (list_empty(&tic->t_queue))
-				list_add_tail(&tic->t_queue, &log->l_writeq);
-			trace_xfs_log_regrant_write_sleep1(log, tic);
-
-			xlog_grant_push_ail(log, need_bytes);
-
-			XFS_STATS_INC(xs_sleep_logspace);
-			xlog_wait(&tic->t_wait, &log->l_grant_write_lock);
-			trace_xfs_log_regrant_write_wake1(log, tic);
-		} else
-			spin_unlock(&log->l_grant_write_lock);
-	}
-
-redo:
-	if (XLOG_FORCED_SHUTDOWN(log))
-		goto error_return_unlocked;
-
 	free_bytes = xlog_space_left(log, &log->l_grant_write_head);
-	if (free_bytes < need_bytes) {
+	if (!list_empty_careful(&log->l_writeq)) {
 		spin_lock(&log->l_grant_write_lock);
-		if (list_empty(&tic->t_queue))
-			list_add_tail(&tic->t_queue, &log->l_writeq);
-
-		if (XLOG_FORCED_SHUTDOWN(log))
-			goto error_return;
-
-		xlog_grant_push_ail(log, need_bytes);
-
-		XFS_STATS_INC(xs_sleep_logspace);
-		trace_xfs_log_regrant_write_sleep2(log, tic);
-		xlog_wait(&tic->t_wait, &log->l_grant_write_lock);
-
-		trace_xfs_log_regrant_write_wake2(log, tic);
-		goto redo;
-	}
-
-	if (!list_empty(&tic->t_queue)) {
+		if (!xlog_writeq_wake(log, &free_bytes) ||
+		    free_bytes < need_bytes)
+			error = xlog_writeq_wait(log, tic, need_bytes);
+		spin_unlock(&log->l_grant_write_lock);
+	} else if (free_bytes < need_bytes) {
 		spin_lock(&log->l_grant_write_lock);
-		list_del_init(&tic->t_queue);
+		error = xlog_writeq_wait(log, tic, need_bytes);
 		spin_unlock(&log->l_grant_write_lock);
 	}
 
-	/* we've got enough space */
+	if (error)
+		return error;
+
 	xlog_grant_add_space(log, &log->l_grant_write_head, need_bytes);
 	trace_xfs_log_regrant_write_exit(log, tic);
 	xlog_verify_grant_tail(log);
 	return 0;
-
-
- error_return_unlocked:
-	spin_lock(&log->l_grant_write_lock);
- error_return:
-	list_del_init(&tic->t_queue);
-	spin_unlock(&log->l_grant_write_lock);
-	trace_xfs_log_regrant_write_error(log, tic);
-
-	/*
-	 * If we are failing, make sure the ticket doesn't have any
-	 * current reservations. We don't want to add this back when
-	 * the ticket/transaction gets cancelled.
-	 */
-	tic->t_curr_res = 0;
-	tic->t_cnt = 0; /* ungrant will give back unit_res * t_cnt. */
-	return XFS_ERROR(EIO);
-}	/* xlog_regrant_write_log_space */
-
+}
 
 /* The first cnt-1 times through here we don't need to
  * move the grant write head because the permanent
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 78c9039..3f7bf45 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -137,7 +137,7 @@
 void	xfs_log_item_init(struct xfs_mount	*mp,
 			struct xfs_log_item	*item,
 			int			type,
-			struct xfs_item_ops	*ops);
+			const struct xfs_item_ops *ops);
 
 xfs_lsn_t xfs_log_done(struct xfs_mount *mp,
 		       struct xlog_ticket *ticket,
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 5cff443..0bbb1a4 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -674,7 +674,8 @@
 	 * disk and we didn't ask it to allocate;
 	 * ESRCH if quotas got turned off suddenly.
 	 */
-	error = xfs_qm_dqget(ip->i_mount, ip, id, type, XFS_QMOPT_DOWARN, &dqp);
+	error = xfs_qm_dqget(ip->i_mount, ip, id, type,
+			     doalloc | XFS_QMOPT_DOWARN, &dqp);
 	if (error)
 		return error;
 
diff --git a/fs/xfs/xfs_sync.c b/fs/xfs/xfs_sync.c
index aa3dc1a..be5c51d 100644
--- a/fs/xfs/xfs_sync.c
+++ b/fs/xfs/xfs_sync.c
@@ -770,6 +770,17 @@
 	if (!xfs_iflock_nowait(ip)) {
 		if (!(sync_mode & SYNC_WAIT))
 			goto out;
+
+		/*
+		 * If we only have a single dirty inode in a cluster there is
+		 * a fair chance that the AIL push may have pushed it into
+		 * the buffer, but xfsbufd won't touch it until 30 seconds
+		 * from now, and thus we will lock up here.
+		 *
+		 * Promote the inode buffer to the front of the delwri list
+		 * and wake up xfsbufd now.
+		 */
+		xfs_promote_inode(ip);
 		xfs_iflock(ip);
 	}
 
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index f1d2802..4940357 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -834,18 +834,14 @@
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_enter);
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_exit);
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_error);
-DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep1);
-DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake1);
-DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep2);
-DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake2);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake);
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake_up);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_enter);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_exit);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_error);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep1);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake1);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep2);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake2);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake_up);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_enter);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_exit);
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index 603f3eb..3ae713c 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -326,7 +326,7 @@
 						 struct xfs_log_item *);
 							/* buffer item iodone */
 							/* callback func */
-	struct xfs_item_ops		*li_ops;	/* function list */
+	const struct xfs_item_ops	*li_ops;	/* function list */
 
 	/* delayed logging */
 	struct list_head		li_cil;		/* CIL pointers */
@@ -341,7 +341,7 @@
 	{ XFS_LI_IN_AIL,	"IN_AIL" }, \
 	{ XFS_LI_ABORTED,	"ABORTED" }
 
-typedef struct xfs_item_ops {
+struct xfs_item_ops {
 	uint (*iop_size)(xfs_log_item_t *);
 	void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *);
 	void (*iop_pin)(xfs_log_item_t *);
@@ -352,7 +352,7 @@
 	void (*iop_push)(xfs_log_item_t *);
 	bool (*iop_pushbuf)(xfs_log_item_t *);
 	void (*iop_committing)(xfs_log_item_t *, xfs_lsn_t);
-} xfs_item_ops_t;
+};
 
 #define IOP_SIZE(ip)		(*(ip)->li_ops->iop_size)(ip)
 #define IOP_FORMAT(ip,vp)	(*(ip)->li_ops->iop_format)(ip, vp)
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index 4ecf2a5..ce9268a 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -112,7 +112,7 @@
 	char		*link)
 {
 	xfs_mount_t	*mp = ip->i_mount;
-	int		pathlen;
+	xfs_fsize_t	pathlen;
 	int		error = 0;
 
 	trace_xfs_readlink(ip);
@@ -122,13 +122,19 @@
 
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
 
-	ASSERT(S_ISLNK(ip->i_d.di_mode));
-	ASSERT(ip->i_d.di_size <= MAXPATHLEN);
-
 	pathlen = ip->i_d.di_size;
 	if (!pathlen)
 		goto out;
 
+	if (pathlen < 0 || pathlen > MAXPATHLEN) {
+		xfs_alert(mp, "%s: inode (%llu) bad symlink length (%lld)",
+			 __func__, (unsigned long long) ip->i_ino,
+			 (long long) pathlen);
+		ASSERT(0);
+		return XFS_ERROR(EFSCORRUPTED);
+	}
+
+
 	if (ip->i_df.if_flags & XFS_IFINLINE) {
 		memcpy(link, ip->i_df.if_u1.if_data, pathlen);
 		link[pathlen] = '\0';
diff --git a/include/asm-generic/unistd.h b/include/asm-generic/unistd.h
index f4c38d8c..2292d1a 100644
--- a/include/asm-generic/unistd.h
+++ b/include/asm-generic/unistd.h
@@ -685,9 +685,15 @@
 __SYSCALL(__NR_setns, sys_setns)
 #define __NR_sendmmsg 269
 __SC_COMP(__NR_sendmmsg, sys_sendmmsg, compat_sys_sendmmsg)
+#define __NR_process_vm_readv 270
+__SC_COMP(__NR_process_vm_readv, sys_process_vm_readv, \
+          compat_sys_process_vm_readv)
+#define __NR_process_vm_writev 271
+__SC_COMP(__NR_process_vm_writev, sys_process_vm_writev, \
+          compat_sys_process_vm_writev)
 
 #undef __NR_syscalls
-#define __NR_syscalls 270
+#define __NR_syscalls 272
 
 /*
  * All syscalls below here should go away really,
diff --git a/include/drm/Kbuild b/include/drm/Kbuild
index 3a60ac8..a5c0e10 100644
--- a/include/drm/Kbuild
+++ b/include/drm/Kbuild
@@ -1,4 +1,5 @@
 header-y += drm.h
+header-y += drm_fourcc.h
 header-y += drm_mode.h
 header-y += drm_sarea.h
 header-y += i810_drm.h
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 1f630a5..ecd5984 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -990,7 +990,9 @@
 	struct proc_dir_entry *proc_root;  /**< proc directory entry */
 	struct drm_info_node proc_nodes;
 	struct dentry *debugfs_root;
-	struct drm_info_node debugfs_nodes;
+
+	struct list_head debugfs_list;
+	struct mutex debugfs_lock; /* Protects debugfs_list. */
 
 	struct drm_master *master; /* currently active master for this node */
 	struct list_head master_list;
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index dd55727..2deb6f9 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -239,7 +239,8 @@
 	struct list_head head;
 	struct drm_mode_object base;
 	const struct drm_framebuffer_funcs *funcs;
-	unsigned int pitch;
+	unsigned int pitches[4];
+	unsigned int offsets[4];
 	unsigned int width;
 	unsigned int height;
 	/* depth can be 15 or 16 */
@@ -828,7 +829,7 @@
 			  struct drm_plane *plane,
 			  unsigned long possible_crtcs,
 			  const struct drm_plane_funcs *funcs,
-			  uint32_t *formats, uint32_t format_count);
+			  const uint32_t *formats, uint32_t format_count);
 extern void drm_plane_cleanup(struct drm_plane *plane);
 
 extern void drm_encoder_cleanup(struct drm_encoder *encoder);
diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h
index e88b7d7..37515d1 100644
--- a/include/drm/drm_crtc_helper.h
+++ b/include/drm/drm_crtc_helper.h
@@ -144,4 +144,7 @@
 
 extern void drm_kms_helper_poll_disable(struct drm_device *dev);
 extern void drm_kms_helper_poll_enable(struct drm_device *dev);
+
+extern int drm_format_num_planes(uint32_t format);
+
 #endif
diff --git a/include/drm/drm_fourcc.h b/include/drm/drm_fourcc.h
index bb75249..bdf0152 100644
--- a/include/drm/drm_fourcc.h
+++ b/include/drm/drm_fourcc.h
@@ -26,8 +26,8 @@
 
 #include <linux/types.h>
 
-#define fourcc_code(a,b,c,d) ((u32)(a) | ((u32)(b) << 8) | \
-			      ((u32)(c) << 16) | ((u32)(d) << 24))
+#define fourcc_code(a, b, c, d) ((__u32)(a) | ((__u32)(b) << 8) | \
+				 ((__u32)(c) << 16) | ((__u32)(d) << 24))
 
 #define DRM_FORMAT_BIG_ENDIAN (1<<31) /* format is big endian instead of little endian */
 
@@ -107,6 +107,10 @@
 #define DRM_FORMAT_NV16		fourcc_code('N', 'V', '1', '6') /* 2x1 subsampled Cr:Cb plane */
 #define DRM_FORMAT_NV61		fourcc_code('N', 'V', '6', '1') /* 2x1 subsampled Cb:Cr plane */
 
+/* 2 non contiguous plane YCbCr */
+#define DRM_FORMAT_NV12M	fourcc_code('N', 'M', '1', '2') /* 2x2 subsampled Cr:Cb plane */
+#define DRM_FORMAT_NV12MT	fourcc_code('T', 'M', '1', '2') /* 2x2 subsampled Cr:Cb plane 64x32 macroblocks */
+
 /*
  * 3 plane YCbCr
  * index 0: Y plane, [7:0] Y
@@ -127,4 +131,7 @@
 #define DRM_FORMAT_YUV444	fourcc_code('Y', 'U', '2', '4') /* non-subsampled Cb (1) and Cr (2) planes */
 #define DRM_FORMAT_YVU444	fourcc_code('Y', 'V', '2', '4') /* non-subsampled Cr (1) and Cb (2) planes */
 
+/* 3 non contiguous plane YCbCr */
+#define DRM_FORMAT_YUV420M	fourcc_code('Y', 'M', '1', '2') /* 2x2 subsampled Cb (1) and Cr (2) planes */
+
 #endif /* DRM_FOURCC_H */
diff --git a/include/drm/drm_mode.h b/include/drm/drm_mode.h
index 95d7aad..2a2acda 100644
--- a/include/drm/drm_mode.h
+++ b/include/drm/drm_mode.h
@@ -268,7 +268,7 @@
 	__u32 handle;
 };
 
-#define DRM_MODE_FB_INTERLACED	(1<<0 /* for interlaced framebuffers */
+#define DRM_MODE_FB_INTERLACED	(1<<0) /* for interlaced framebuffers */
 
 struct drm_mode_fb_cmd2 {
 	__u32 fb_id;
@@ -299,6 +299,8 @@
 #define DRM_MODE_FB_DIRTY_ANNOTATE_FILL 0x02
 #define DRM_MODE_FB_DIRTY_FLAGS         0x03
 
+#define DRM_MODE_FB_DIRTY_MAX_CLIPS     256
+
 /*
  * Mark a region of a framebuffer as dirty.
  *
diff --git a/include/drm/drm_pciids.h b/include/drm/drm_pciids.h
index 3d53efd..14b6cd0 100644
--- a/include/drm/drm_pciids.h
+++ b/include/drm/drm_pciids.h
@@ -4,6 +4,7 @@
 */
 #define radeon_PCI_IDS \
 	{0x1002, 0x3150, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV380|RADEON_IS_MOBILITY}, \
+	{0x1002, 0x3151, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV380|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x3152, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV380|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x3154, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV380|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x3155, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV380|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
@@ -55,6 +56,7 @@
 	{0x1002, 0x4C64, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV250|RADEON_IS_MOBILITY}, \
 	{0x1002, 0x4C66, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV250|RADEON_IS_MOBILITY}, \
 	{0x1002, 0x4C67, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV250|RADEON_IS_MOBILITY}, \
+	{0x1002, 0x4C6E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV280|RADEON_IS_MOBILITY}, \
 	{0x1002, 0x4E44, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_R300}, \
 	{0x1002, 0x4E45, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_R300}, \
 	{0x1002, 0x4E46, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_R300}, \
@@ -180,8 +182,11 @@
 	{0x1002, 0x6748, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6749, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6750, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6751, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6758, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6759, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x675B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x675D, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x675F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6760, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6761, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
@@ -193,8 +198,18 @@
 	{0x1002, 0x6767, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6768, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6770, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6772, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6778, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6779, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x677B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6840, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6841, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6842, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6843, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6849, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6850, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6858, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x6859, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6880, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYPRESS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6888, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYPRESS|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x6889, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYPRESS|RADEON_NEW_MEMMAP}, \
@@ -236,6 +251,7 @@
 	{0x1002, 0x68f2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CEDAR|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x68f8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CEDAR|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x68f9, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CEDAR|RADEON_NEW_MEMMAP}, \
+	{0x1002, 0x68fa, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CEDAR|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x68fe, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CEDAR|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x7100, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_R520|RADEON_NEW_MEMMAP}, \
 	{0x1002, 0x7101, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_R520|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \
@@ -478,6 +494,8 @@
 	{0x1002, 0x9647, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_SUMO|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP|RADEON_IS_IGP},\
 	{0x1002, 0x9648, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_SUMO|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP|RADEON_IS_IGP},\
 	{0x1002, 0x964a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_SUMO|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
+	{0x1002, 0x964b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_SUMO|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
+	{0x1002, 0x964c, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_SUMO|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
 	{0x1002, 0x964e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_SUMO|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP|RADEON_IS_IGP},\
 	{0x1002, 0x964f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_SUMO|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP|RADEON_IS_IGP},\
 	{0x1002, 0x9710, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RS880|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
@@ -492,6 +510,8 @@
 	{0x1002, 0x9805, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PALM|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
 	{0x1002, 0x9806, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PALM|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
 	{0x1002, 0x9807, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PALM|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
+	{0x1002, 0x9808, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PALM|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
+	{0x1002, 0x9809, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_PALM|RADEON_NEW_MEMMAP|RADEON_IS_IGP}, \
 	{0, 0, 0}
 
 #define r128_PCI_IDS \
diff --git a/include/drm/drm_sman.h b/include/drm/drm_sman.h
deleted file mode 100644
index 08ecf83..0000000
--- a/include/drm/drm_sman.h
+++ /dev/null
@@ -1,176 +0,0 @@
-/**************************************************************************
- *
- * Copyright 2006 Tungsten Graphics, Inc., Bismarck, ND., USA.
- * All Rights Reserved.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
- *
- *
- **************************************************************************/
-/*
- * Simple memory MANager interface that keeps track on allocate regions on a
- * per "owner" basis. All regions associated with an "owner" can be released
- * with a simple call. Typically if the "owner" exists. The owner is any
- * "unsigned long" identifier. Can typically be a pointer to a file private
- * struct or a context identifier.
- *
- * Authors:
- * Thomas Hellström <thomas-at-tungstengraphics-dot-com>
- */
-
-#ifndef DRM_SMAN_H
-#define DRM_SMAN_H
-
-#include "drmP.h"
-#include "drm_hashtab.h"
-
-/*
- * A class that is an abstration of a simple memory allocator.
- * The sman implementation provides a default such allocator
- * using the drm_mm.c implementation. But the user can replace it.
- * See the SiS implementation, which may use the SiS FB kernel module
- * for memory management.
- */
-
-struct drm_sman_mm {
-	/* private info. If allocated, needs to be destroyed by the destroy
-	   function */
-	void *private;
-
-	/* Allocate a memory block with given size and alignment.
-	   Return an opaque reference to the memory block */
-
-	void *(*allocate) (void *private, unsigned long size,
-			   unsigned alignment);
-
-	/* Free a memory block. "ref" is the opaque reference that we got from
-	   the "alloc" function */
-
-	void (*free) (void *private, void *ref);
-
-	/* Free all resources associated with this allocator */
-
-	void (*destroy) (void *private);
-
-	/* Return a memory offset from the opaque reference returned from the
-	   "alloc" function */
-
-	unsigned long (*offset) (void *private, void *ref);
-};
-
-struct drm_memblock_item {
-	struct list_head owner_list;
-	struct drm_hash_item user_hash;
-	void *mm_info;
-	struct drm_sman_mm *mm;
-	struct drm_sman *sman;
-};
-
-struct drm_sman {
-	struct drm_sman_mm *mm;
-	int num_managers;
-	struct drm_open_hash owner_hash_tab;
-	struct drm_open_hash user_hash_tab;
-	struct list_head owner_items;
-};
-
-/*
- * Take down a memory manager. This function should only be called after a
- * successful init and after a call to drm_sman_cleanup.
- */
-
-extern void drm_sman_takedown(struct drm_sman * sman);
-
-/*
- * Allocate structures for a manager.
- * num_managers are the number of memory pools to manage. (VRAM, AGP, ....)
- * user_order is the log2 of the number of buckets in the user hash table.
- *	    set this to approximately log2 of the max number of memory regions
- *	    that will be allocated for _all_ pools together.
- * owner_order is the log2 of the number of buckets in the owner hash table.
- *	    set this to approximately log2 of
- *	    the number of client file connections that will
- *	    be using the manager.
- *
- */
-
-extern int drm_sman_init(struct drm_sman * sman, unsigned int num_managers,
-			 unsigned int user_order, unsigned int owner_order);
-
-/*
- * Initialize a drm_mm.c allocator. Should be called only once for each
- * manager unless a customized allogator is used.
- */
-
-extern int drm_sman_set_range(struct drm_sman * sman, unsigned int manager,
-			      unsigned long start, unsigned long size);
-
-/*
- * Initialize a customized allocator for one of the managers.
- * (See the SiS module). The object pointed to by "allocator" is copied,
- * so it can be destroyed after this call.
- */
-
-extern int drm_sman_set_manager(struct drm_sman * sman, unsigned int mananger,
-				struct drm_sman_mm * allocator);
-
-/*
- * Allocate a memory block. Aligment is not implemented yet.
- */
-
-extern struct drm_memblock_item *drm_sman_alloc(struct drm_sman * sman,
-						unsigned int manager,
-						unsigned long size,
-						unsigned alignment,
-						unsigned long owner);
-/*
- * Free a memory block identified by its user hash key.
- */
-
-extern int drm_sman_free_key(struct drm_sman * sman, unsigned int key);
-
-/*
- * returns 1 iff there are no stale memory blocks associated with this owner.
- * Typically called to determine if we need to idle the hardware and call
- * drm_sman_owner_cleanup. If there are no stale memory blocks, it removes all
- * resources associated with owner.
- */
-
-extern int drm_sman_owner_clean(struct drm_sman * sman, unsigned long owner);
-
-/*
- * Frees all stale memory blocks associated with this owner. Note that this
- * requires that the hardware is finished with all blocks, so the graphics engine
- * should be idled before this call is made. This function also frees
- * any resources associated with "owner" and should be called when owner
- * is not going to be referenced anymore.
- */
-
-extern void drm_sman_owner_cleanup(struct drm_sman * sman, unsigned long owner);
-
-/*
- * Frees all stale memory blocks associated with the memory manager.
- * See idling above.
- */
-
-extern void drm_sman_cleanup(struct drm_sman * sman);
-
-#endif
diff --git a/include/drm/exynos_drm.h b/include/drm/exynos_drm.h
index 874c4d2..5e120f1 100644
--- a/include/drm/exynos_drm.h
+++ b/include/drm/exynos_drm.h
@@ -32,13 +32,14 @@
 /**
  * User-desired buffer creation information structure.
  *
- * @size: requested size for the object.
+ * @size: user-desired memory allocation size.
  *	- this size value would be page-aligned internally.
  * @flags: user request for setting memory type or cache attributes.
- * @handle: returned handle for the object.
+ * @handle: returned a handle to created gem object.
+ *	- this handle will be set by gem module of kernel side.
  */
 struct drm_exynos_gem_create {
-	unsigned int size;
+	uint64_t size;
 	unsigned int flags;
 	unsigned int handle;
 };
@@ -73,9 +74,16 @@
 	uint64_t mapped;
 };
 
+struct drm_exynos_plane_set_zpos {
+	__u32 plane_id;
+	__s32 zpos;
+};
+
 #define DRM_EXYNOS_GEM_CREATE		0x00
 #define DRM_EXYNOS_GEM_MAP_OFFSET	0x01
 #define DRM_EXYNOS_GEM_MMAP		0x02
+/* Reserved 0x03 ~ 0x05 for exynos specific gem ioctl */
+#define DRM_EXYNOS_PLANE_SET_ZPOS	0x06
 
 #define DRM_IOCTL_EXYNOS_GEM_CREATE		DRM_IOWR(DRM_COMMAND_BASE + \
 		DRM_EXYNOS_GEM_CREATE, struct drm_exynos_gem_create)
@@ -86,6 +94,9 @@
 #define DRM_IOCTL_EXYNOS_GEM_MMAP	DRM_IOWR(DRM_COMMAND_BASE + \
 		DRM_EXYNOS_GEM_MMAP, struct drm_exynos_gem_mmap)
 
+#define DRM_IOCTL_EXYNOS_PLANE_SET_ZPOS	DRM_IOWR(DRM_COMMAND_BASE + \
+		DRM_EXYNOS_PLANE_SET_ZPOS, struct drm_exynos_plane_set_zpos)
+
 /**
  * Platform Specific Structure for DRM based FIMD.
  *
@@ -101,4 +112,31 @@
 	unsigned int			bpp;
 };
 
+/**
+ * Platform Specific Structure for DRM based HDMI.
+ *
+ * @hdmi_dev: device point to specific hdmi driver.
+ * @mixer_dev: device point to specific mixer driver.
+ *
+ * this structure is used for common hdmi driver and each device object
+ * would be used to access specific device driver(hdmi or mixer driver)
+ */
+struct exynos_drm_common_hdmi_pd {
+	struct device *hdmi_dev;
+	struct device *mixer_dev;
+};
+
+/**
+ * Platform Specific Structure for DRM based HDMI core.
+ *
+ * @timing: default video mode for initializing
+ * @default_win: default window layer number to be used for UI.
+ * @bpp: default bit per pixel.
+ */
+struct exynos_drm_hdmi_pdata {
+	struct fb_videomode		timing;
+	unsigned int			default_win;
+	unsigned int			bpp;
+};
+
 #endif
diff --git a/include/drm/radeon_drm.h b/include/drm/radeon_drm.h
index b65be60..be94be6 100644
--- a/include/drm/radeon_drm.h
+++ b/include/drm/radeon_drm.h
@@ -874,6 +874,10 @@
 
 #define RADEON_CHUNK_ID_RELOCS	0x01
 #define RADEON_CHUNK_ID_IB	0x02
+#define RADEON_CHUNK_ID_FLAGS	0x03
+
+/* The first dword of RADEON_CHUNK_ID_FLAGS is a uint32 of these flags: */
+#define RADEON_CS_KEEP_TILING_FLAGS 0x01
 
 struct drm_radeon_cs_chunk {
 	uint32_t		chunk_id;
diff --git a/include/drm/sis_drm.h b/include/drm/sis_drm.h
index 30f7b38..035b804 100644
--- a/include/drm/sis_drm.h
+++ b/include/drm/sis_drm.h
@@ -64,4 +64,8 @@
 	unsigned int offset, size;
 } drm_sis_fb_t;
 
+struct sis_file_private {
+	struct list_head obj_list;
+};
+
 #endif				/* __SIS_DRM_H__ */
diff --git a/include/drm/via_drm.h b/include/drm/via_drm.h
index fd11a5b..79b3b6e 100644
--- a/include/drm/via_drm.h
+++ b/include/drm/via_drm.h
@@ -274,4 +274,8 @@
 	drm_via_blitsync_t sync;
 } drm_via_dmablit_t;
 
+struct via_file_private {
+	struct list_head obj_list;
+};
+
 #endif				/* _VIA_DRM_H_ */
diff --git a/include/linux/bio.h b/include/linux/bio.h
index a3c071c..847994a 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -211,8 +211,8 @@
 extern struct bio_set *bioset_create(unsigned int, unsigned int);
 extern void bioset_free(struct bio_set *);
 
-extern struct bio *bio_alloc(gfp_t, int);
-extern struct bio *bio_kmalloc(gfp_t, int);
+extern struct bio *bio_alloc(gfp_t, unsigned int);
+extern struct bio *bio_kmalloc(gfp_t, unsigned int);
 extern struct bio *bio_alloc_bioset(gfp_t, int, struct bio_set *);
 extern void bio_put(struct bio *);
 extern void bio_free(struct bio *, struct bio_set *);
@@ -519,7 +519,11 @@
 #define bioset_integrity_create(a, b)	(0)
 #define bio_integrity_prep(a)		(0)
 #define bio_integrity_enabled(a)	(0)
-#define bio_integrity_clone(a, b, c, d)	(0)
+static inline int bio_integrity_clone(struct bio *bio, struct bio *bio_src,
+				      gfp_t gfp_mask, struct bio_set *bs)
+{
+	return 0;
+}
 #define bioset_integrity_free(a)	do { } while (0)
 #define bio_integrity_free(a, b)	do { } while (0)
 #define bio_integrity_endio(a, b)	do { } while (0)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index c7a6d3b..94acd81 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -805,9 +805,6 @@
  */
 extern struct request_queue *blk_init_queue_node(request_fn_proc *rfn,
 					spinlock_t *lock, int node_id);
-extern struct request_queue *blk_init_allocated_queue_node(struct request_queue *,
-							   request_fn_proc *,
-							   spinlock_t *, int node_id);
 extern struct request_queue *blk_init_queue(request_fn_proc *, spinlock_t *);
 extern struct request_queue *blk_init_allocated_queue(struct request_queue *,
 						      request_fn_proc *, spinlock_t *);
diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h
index f88eacb..7c05ac2 100644
--- a/include/linux/ceph/osd_client.h
+++ b/include/linux/ceph/osd_client.h
@@ -10,6 +10,12 @@
 #include "osdmap.h"
 #include "messenger.h"
 
+/* 
+ * Maximum object name size 
+ * (must be at least as big as RBD_MAX_MD_NAME_LEN -- currently 100) 
+ */
+#define MAX_OBJ_NAME_SIZE 100
+
 struct ceph_msg;
 struct ceph_snap_context;
 struct ceph_osd_request;
@@ -75,7 +81,7 @@
 	struct inode *r_inode;         	      /* for use by callbacks */
 	void *r_priv;			      /* ditto */
 
-	char              r_oid[40];          /* object name */
+	char              r_oid[MAX_OBJ_NAME_SIZE];          /* object name */
 	int               r_oid_len;
 	unsigned long     r_stamp;            /* send OR check time */
 
diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h
index 139c4db..c86c940 100644
--- a/include/linux/clocksource.h
+++ b/include/linux/clocksource.h
@@ -156,6 +156,7 @@
  * @mult:		cycle to nanosecond multiplier
  * @shift:		cycle to nanosecond divisor (power of two)
  * @max_idle_ns:	max idle time permitted by the clocksource (nsecs)
+ * @maxadj		maximum adjustment value to mult (~11%)
  * @flags:		flags describing special properties
  * @archdata:		arch-specific data
  * @suspend:		suspend function for the clocksource, if necessary
@@ -172,7 +173,7 @@
 	u32 mult;
 	u32 shift;
 	u64 max_idle_ns;
-
+	u32 maxadj;
 #ifdef CONFIG_ARCH_CLOCKSOURCE_DATA
 	struct arch_clocksource_data archdata;
 #endif
diff --git a/include/linux/compat.h b/include/linux/compat.h
index 154bf56..66ed067 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -552,5 +552,14 @@
 
 extern void __user *compat_alloc_user_space(unsigned long len);
 
+asmlinkage ssize_t compat_sys_process_vm_readv(compat_pid_t pid,
+		const struct compat_iovec __user *lvec,
+		unsigned long liovcnt, const struct compat_iovec __user *rvec,
+		unsigned long riovcnt, unsigned long flags);
+asmlinkage ssize_t compat_sys_process_vm_writev(compat_pid_t pid,
+		const struct compat_iovec __user *lvec,
+		unsigned long liovcnt, const struct compat_iovec __user *rvec,
+		unsigned long riovcnt, unsigned long flags);
+
 #endif /* CONFIG_COMPAT */
 #endif /* _LINUX_COMPAT_H */
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index 4df9261..ed9f74f 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -339,7 +339,8 @@
  */
 extern char *dynamic_dname(struct dentry *, char *, int, const char *, ...);
 
-extern char *__d_path(const struct path *path, struct path *root, char *, int);
+extern char *__d_path(const struct path *, const struct path *, char *, int);
+extern char *d_absolute_path(const struct path *, char *, int);
 extern char *d_path(const struct path *, char *, int);
 extern char *d_path_with_unreachable(const struct path *, char *, int);
 extern char *dentry_path_raw(struct dentry *, char *, int);
diff --git a/include/linux/devfreq.h b/include/linux/devfreq.h
index afb9458..98ce812 100644
--- a/include/linux/devfreq.h
+++ b/include/linux/devfreq.h
@@ -41,7 +41,7 @@
 	unsigned long total_time;
 	unsigned long busy_time;
 	unsigned long current_frequency;
-	void *private_date;
+	void *private_data;
 };
 
 /**
diff --git a/include/linux/device.h b/include/linux/device.h
index ffbcf95..3136ede 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -69,7 +69,7 @@
  * @resume:	Called to bring a device on this bus out of sleep mode.
  * @pm:		Power management operations of this bus, callback the specific
  *		device driver's pm-ops.
- * @iommu_ops   IOMMU specific operations for this bus, used to attach IOMMU
+ * @iommu_ops:  IOMMU specific operations for this bus, used to attach IOMMU
  *              driver implementations to a bus and allow the driver to do
  *              bus-specific setup
  * @p:		The private data of the driver core, only the driver core can
@@ -682,6 +682,11 @@
 	return !!dev->power.async_suspend;
 }
 
+static inline void pm_suspend_ignore_children(struct device *dev, bool enable)
+{
+	dev->power.ignore_children = enable;
+}
+
 static inline void device_lock(struct device *dev)
 {
 	mutex_lock(&dev->mutex);
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index ef90cbd..57c9a8a 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -31,6 +31,7 @@
 extern int iommu_calculate_agaw(struct intel_iommu *iommu);
 extern int iommu_calculate_max_sagaw(struct intel_iommu *iommu);
 extern int dmar_disabled;
+extern int intel_iommu_enabled;
 #else
 static inline int iommu_calculate_agaw(struct intel_iommu *iommu)
 {
@@ -44,6 +45,7 @@
 {
 }
 #define dmar_disabled	(1)
+#define intel_iommu_enabled (0)
 #endif
 
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 0c4df26..e0bc4ff 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -393,8 +393,8 @@
 #include <linux/semaphore.h>
 #include <linux/fiemap.h>
 #include <linux/rculist_bl.h>
-#include <linux/shrinker.h>
 #include <linux/atomic.h>
+#include <linux/shrinker.h>
 
 #include <asm/byteorder.h>
 
@@ -1886,6 +1886,7 @@
 extern struct dentry *mount_nodev(struct file_system_type *fs_type,
 	int flags, void *data,
 	int (*fill_super)(struct super_block *, void *, int));
+extern struct dentry *mount_subtree(struct vfsmount *mnt, const char *path);
 void generic_shutdown_super(struct super_block *sb);
 void kill_block_super(struct super_block *sb);
 void kill_anon_super(struct super_block *sb);
@@ -1941,6 +1942,7 @@
 extern int statfs_by_dentry(struct dentry *, struct kstatfs *);
 extern int freeze_super(struct super_block *super);
 extern int thaw_super(struct super_block *super);
+extern bool our_mnt(struct vfsmount *mnt);
 
 extern int current_umask(void);
 
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 96efa679..c3da42d 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -172,6 +172,7 @@
 	TRACE_EVENT_FL_FILTERED_BIT,
 	TRACE_EVENT_FL_RECORDED_CMD_BIT,
 	TRACE_EVENT_FL_CAP_ANY_BIT,
+	TRACE_EVENT_FL_NO_SET_FILTER_BIT,
 };
 
 enum {
@@ -179,6 +180,7 @@
 	TRACE_EVENT_FL_FILTERED		= (1 << TRACE_EVENT_FL_FILTERED_BIT),
 	TRACE_EVENT_FL_RECORDED_CMD	= (1 << TRACE_EVENT_FL_RECORDED_CMD_BIT),
 	TRACE_EVENT_FL_CAP_ANY		= (1 << TRACE_EVENT_FL_CAP_ANY_BIT),
+	TRACE_EVENT_FL_NO_SET_FILTER	= (1 << TRACE_EVENT_FL_NO_SET_FILTER_BIT),
 };
 
 struct ftrace_event_call {
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 9de31bc..6d18f35 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -21,8 +21,6 @@
 #define dev_to_part(device)	container_of((device), struct hd_struct, __dev)
 #define disk_to_dev(disk)	(&(disk)->part0.__dev)
 #define part_to_dev(part)	(&((part)->__dev))
-#define alias_name(disk)	((disk)->alias ? (disk)->alias : \
-						 (disk)->disk_name)
 
 extern struct device_type part_type;
 extern struct kobject *block_depr;
@@ -60,7 +58,6 @@
 
 #define DISK_MAX_PARTS			256
 #define DISK_NAME_LEN			32
-#define ALIAS_LEN			256
 
 #include <linux/major.h>
 #include <linux/device.h>
@@ -166,7 +163,6 @@
                                          * disks that can't be partitioned. */
 
 	char disk_name[DISK_NAME_LEN];	/* name of major driver */
-	char *alias;			/* alias name of disk */
 	char *(*devnode)(struct gendisk *gd, mode_t *mode);
 
 	unsigned int events;		/* supported events */
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 19644e0..d9d6c86 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -110,11 +110,6 @@
 
 #define hugetlb_change_protection(vma, address, end, newprot)
 
-#ifndef HPAGE_MASK
-#define HPAGE_MASK	PAGE_MASK		/* Keep the compiler happy */
-#define HPAGE_SIZE	PAGE_SIZE
-#endif
-
 #endif /* !CONFIG_HUGETLB_PAGE */
 
 #define HUGETLB_ANON_FILE "anon_hugepage"
diff --git a/include/linux/hwspinlock.h b/include/linux/hwspinlock.h
index 08a2fee..aad6bd4 100644
--- a/include/linux/hwspinlock.h
+++ b/include/linux/hwspinlock.h
@@ -118,7 +118,6 @@
 static inline
 void __hwspin_unlock(struct hwspinlock *hwlock, int mode, unsigned long *flags)
 {
-	return 0;
 }
 
 static inline int hwspin_lock_get_id(struct hwspinlock *hwlock)
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index a81bf6d..07d103a 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -432,9 +432,6 @@
 /* Internal numbers to terminate lists */
 #define I2C_CLIENT_END		0xfffeU
 
-/* The numbers to use to set I2C bus address */
-#define ANY_I2C_BUS		0xffff
-
 /* Construct an I2C_CLIENT_END-terminated array of i2c addresses */
 #define I2C_ADDRS(addr, addrs...) \
 	((const unsigned short []){ addr, ## addrs, I2C_CLIENT_END })
diff --git a/include/linux/inet_diag.h b/include/linux/inet_diag.h
index 80b480c..abf5028 100644
--- a/include/linux/inet_diag.h
+++ b/include/linux/inet_diag.h
@@ -98,9 +98,10 @@
 	INET_DIAG_VEGASINFO,
 	INET_DIAG_CONG,
 	INET_DIAG_TOS,
+	INET_DIAG_TCLASS,
 };
 
-#define INET_DIAG_MAX INET_DIAG_TOS
+#define INET_DIAG_MAX INET_DIAG_TCLASS
 
 
 /* INET_DIAG_MEM */
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 08ffab0..32574ee 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -126,6 +126,8 @@
 # define INIT_PERF_EVENTS(tsk)
 #endif
 
+#define INIT_TASK_COMM "swapper"
+
 /*
  *  INIT_TASK is used to set up the first task table, touch at
  * your own risk!. Base=0, limit=0x1fffff (=2MB)
@@ -162,7 +164,7 @@
 	.group_leader	= &tsk,						\
 	RCU_INIT_POINTER(.real_cred, &init_cred),			\
 	RCU_INIT_POINTER(.cred, &init_cred),				\
-	.comm		= "swapper",					\
+	.comm		= INIT_TASK_COMM,				\
 	.thread		= INIT_THREAD,					\
 	.fs		= &init_fs,					\
 	.files		= &init_files,					\
@@ -184,7 +186,6 @@
 		[PIDTYPE_SID]  = INIT_PID_LINK(PIDTYPE_SID),		\
 	},								\
 	.thread_group	= LIST_HEAD_INIT(tsk.thread_group),		\
-	.dirties = INIT_PROP_LOCAL_SINGLE(dirties),			\
 	INIT_IDS							\
 	INIT_PERF_EVENTS(tsk)						\
 	INIT_TRACE_IRQFLAGS						\
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index f47fcd3..c3892fc 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -555,7 +555,6 @@
 #define KVM_CAP_PPC_SMT 64
 #define KVM_CAP_PPC_RMA	65
 #define KVM_CAP_MAX_VCPUS 66       /* returns max vcpus per vm */
-#define KVM_CAP_PPC_HIOR 67
 #define KVM_CAP_PPC_PAPR 68
 #define KVM_CAP_S390_GMAP 71
 
diff --git a/include/linux/log2.h b/include/linux/log2.h
index 25b8086..fd7ff3d 100644
--- a/include/linux/log2.h
+++ b/include/linux/log2.h
@@ -185,7 +185,6 @@
 #define rounddown_pow_of_two(n)			\
 (						\
 	__builtin_constant_p(n) ? (		\
-		(n == 1) ? 0 :			\
 		(1UL << ilog2(n))) :		\
 	__rounddown_pow_of_two(n)		\
  )
diff --git a/include/linux/mfd/tps65910.h b/include/linux/mfd/tps65910.h
index 82b4c88..8bf2cb9 100644
--- a/include/linux/mfd/tps65910.h
+++ b/include/linux/mfd/tps65910.h
@@ -243,7 +243,8 @@
 
 
 /*Registers VDD1, VDD2 voltage values definitions */
-#define VDD1_2_NUM_VOLTS				73
+#define VDD1_2_NUM_VOLT_FINE				73
+#define VDD1_2_NUM_VOLT_COARSE				3
 #define VDD1_2_MIN_VOLT					6000
 #define VDD1_2_OFFSET					125
 
diff --git a/include/linux/mfd/wm8994/registers.h b/include/linux/mfd/wm8994/registers.h
index fae2950..83a9cae 100644
--- a/include/linux/mfd/wm8994/registers.h
+++ b/include/linux/mfd/wm8994/registers.h
@@ -1963,6 +1963,21 @@
 #define WM8958_MICB2_DISCH_WIDTH                     1  /* MICB2_DISCH */
 
 /*
+ * R210 (0xD2) - Mic Detect 3
+ */
+#define WM8958_MICD_LVL_MASK                    0x07FC  /* MICD_LVL - [10:2] */
+#define WM8958_MICD_LVL_SHIFT                        2  /* MICD_LVL - [10:2] */
+#define WM8958_MICD_LVL_WIDTH                        9  /* MICD_LVL - [10:2] */
+#define WM8958_MICD_VALID                       0x0002  /* MICD_VALID */
+#define WM8958_MICD_VALID_MASK                  0x0002  /* MICD_VALID */
+#define WM8958_MICD_VALID_SHIFT                      1  /* MICD_VALID */
+#define WM8958_MICD_VALID_WIDTH                      1  /* MICD_VALID */
+#define WM8958_MICD_STS                         0x0001  /* MICD_STS */
+#define WM8958_MICD_STS_MASK                    0x0001  /* MICD_STS */
+#define WM8958_MICD_STS_SHIFT                        0  /* MICD_STS */
+#define WM8958_MICD_STS_WIDTH                        1  /* MICD_STS */
+
+/*
  * R76 (0x4C) - Charge Pump (1)
  */
 #define WM8994_CP_ENA                           0x8000  /* CP_ENA */
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 3dc3a8c..4baadd1 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -10,6 +10,7 @@
 #include <linux/mmzone.h>
 #include <linux/rbtree.h>
 #include <linux/prio_tree.h>
+#include <linux/atomic.h>
 #include <linux/debug_locks.h>
 #include <linux/mm_types.h>
 #include <linux/range.h>
diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h
index 415f2db..c8ef9bc 100644
--- a/include/linux/mmc/card.h
+++ b/include/linux/mmc/card.h
@@ -218,6 +218,7 @@
 #define MMC_QUIRK_INAND_CMD38	(1<<6)		/* iNAND devices have broken CMD38 */
 #define MMC_QUIRK_BLK_NO_CMD23	(1<<7)		/* Avoid CMD23 for regular multiblock */
 #define MMC_QUIRK_BROKEN_BYTE_MODE_512 (1<<8)	/* Avoid sending 512 bytes in */
+#define MMC_QUIRK_LONG_READ_TIME (1<<9)		/* Data read time > CSD says */
 						/* byte mode */
 	unsigned int    poweroff_notify_state;	/* eMMC4.5 notify feature */
 #define MMC_NO_POWER_NOTIFICATION	0
@@ -433,6 +434,11 @@
 	return c->quirks & MMC_QUIRK_BROKEN_BYTE_MODE_512;
 }
 
+static inline int mmc_card_long_read_time(const struct mmc_card *c)
+{
+	return c->quirks & MMC_QUIRK_LONG_READ_TIME;
+}
+
 #define mmc_card_name(c)	((c)->cid.prod_name)
 #define mmc_card_id(c)		(dev_name(&(c)->dev))
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index cbeb586..a82ad4d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2536,6 +2536,8 @@
 extern void *dev_seq_start(struct seq_file *seq, loff_t *pos);
 extern void *dev_seq_next(struct seq_file *seq, void *v, loff_t *pos);
 extern void dev_seq_stop(struct seq_file *seq, void *v);
+extern int dev_seq_open_ops(struct inode *inode, struct file *file,
+			    const struct seq_operations *ops);
 #endif
 
 extern int netdev_class_create_file(struct class_attribute *class_attr);
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index ab2c634..92ecf55 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -410,6 +410,9 @@
 extern const struct inode_operations nfs3_file_inode_operations;
 #endif /* CONFIG_NFS_V3 */
 extern const struct file_operations nfs_file_operations;
+#ifdef CONFIG_NFS_V4
+extern const struct file_operations nfs4_file_operations;
+#endif /* CONFIG_NFS_V4 */
 extern const struct address_space_operations nfs_file_aops;
 extern const struct address_space_operations nfs_dir_aops;
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index c74595b..2a7c533 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -1192,6 +1192,7 @@
 	const struct dentry_operations *dentry_ops;
 	const struct inode_operations *dir_inode_ops;
 	const struct inode_operations *file_inode_ops;
+	const struct file_operations *file_ops;
 
 	int	(*getroot) (struct nfs_server *, struct nfs_fh *,
 			    struct nfs_fsinfo *);
diff --git a/include/linux/pci-ats.h b/include/linux/pci-ats.h
index e3d0b38..7ef6872 100644
--- a/include/linux/pci-ats.h
+++ b/include/linux/pci-ats.h
@@ -12,7 +12,7 @@
 	unsigned int is_enabled:1;      /* Enable bit is set */
 };
 
-#ifdef CONFIG_PCI_IOV
+#ifdef CONFIG_PCI_ATS
 
 extern int pci_enable_ats(struct pci_dev *dev, int ps);
 extern void pci_disable_ats(struct pci_dev *dev);
@@ -29,7 +29,7 @@
 	return dev->ats && dev->ats->is_enabled;
 }
 
-#else /* CONFIG_PCI_IOV */
+#else /* CONFIG_PCI_ATS */
 
 static inline int pci_enable_ats(struct pci_dev *dev, int ps)
 {
@@ -50,7 +50,7 @@
 	return 0;
 }
 
-#endif /* CONFIG_PCI_IOV */
+#endif /* CONFIG_PCI_ATS */
 
 #ifdef CONFIG_PCI_PRI
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 337df0d5..7cda65b 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -338,7 +338,7 @@
 	struct list_head msi_list;
 #endif
 	struct pci_vpd *vpd;
-#ifdef CONFIG_PCI_IOV
+#ifdef CONFIG_PCI_ATS
 	union {
 		struct pci_sriov *sriov;	/* SR-IOV capability related */
 		struct pci_dev *physfn;	/* the PF this VF is associated with */
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 3fdf251..2aaee0c 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -517,8 +517,12 @@
 #define PCI_DEVICE_ID_AMD_11H_NB_DRAM	0x1302
 #define PCI_DEVICE_ID_AMD_11H_NB_MISC	0x1303
 #define PCI_DEVICE_ID_AMD_11H_NB_LINK	0x1304
+#define PCI_DEVICE_ID_AMD_15H_NB_F0	0x1600
+#define PCI_DEVICE_ID_AMD_15H_NB_F1	0x1601
+#define PCI_DEVICE_ID_AMD_15H_NB_F2	0x1602
 #define PCI_DEVICE_ID_AMD_15H_NB_F3	0x1603
 #define PCI_DEVICE_ID_AMD_15H_NB_F4	0x1604
+#define PCI_DEVICE_ID_AMD_15H_NB_F5	0x1605
 #define PCI_DEVICE_ID_AMD_CNB17H_F3	0x1703
 #define PCI_DEVICE_ID_AMD_LANCE		0x2000
 #define PCI_DEVICE_ID_AMD_LANCE_HOME	0x2001
@@ -2405,6 +2409,8 @@
 
 #define PCI_VENDOR_ID_AZWAVE		0x1a3b
 
+#define PCI_VENDOR_ID_ASMEDIA		0x1b21
+
 #define PCI_VENDOR_ID_TEKRAM		0x1de1
 #define PCI_DEVICE_ID_TEKRAM_DC290	0xdc29
 
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 1e9ebe5..b1f8912 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -822,6 +822,7 @@
 	int				mmap_locked;
 	struct user_struct		*mmap_user;
 	struct ring_buffer		*rb;
+	struct list_head		rb_entry;
 
 	/* poll related */
 	wait_queue_head_t		waitq;
diff --git a/include/linux/pinctrl/pinctrl.h b/include/linux/pinctrl/pinctrl.h
index 3605e94..04c0110 100644
--- a/include/linux/pinctrl/pinctrl.h
+++ b/include/linux/pinctrl/pinctrl.h
@@ -121,6 +121,7 @@
 extern void *pinctrl_dev_get_drvdata(struct pinctrl_dev *pctldev);
 #else
 
+struct pinctrl_dev;
 
 /* Sufficiently stupid default function when pinctrl is not in use */
 static inline bool pin_is_valid(struct pinctrl_dev *pctldev, int pin)
diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index c533670..7281d5a 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -30,7 +30,7 @@
  */
 
 struct tc_stats {
-	__u64	bytes;			/* NUmber of enqueues bytes */
+	__u64	bytes;			/* Number of enqueued bytes */
 	__u32	packets;		/* Number of enqueued packets	*/
 	__u32	drops;			/* Packets dropped because of lack of resources */
 	__u32	overlimits;		/* Number of throttle events when this
@@ -297,7 +297,7 @@
 	__u32 debug;		/* debug flags */
 
 	/* stats */
-	__u32 direct_pkts; /* count of non shapped packets */
+	__u32 direct_pkts; /* count of non shaped packets */
 };
 enum {
 	TCA_HTB_UNSPEC,
@@ -503,7 +503,7 @@
 };
 #define NETEM_LOSS_MAX (__NETEM_LOSS_MAX - 1)
 
-/* State transition probablities for 4 state model */
+/* State transition probabilities for 4 state model */
 struct tc_netem_gimodel {
 	__u32	p13;
 	__u32	p31;
diff --git a/include/linux/pm.h b/include/linux/pm.h
index f15acb6..3f3ed83 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -54,118 +54,145 @@
 /**
  * struct dev_pm_ops - device PM callbacks
  *
- * Several driver power state transitions are externally visible, affecting
+ * Several device power state transitions are externally visible, affecting
  * the state of pending I/O queues and (for drivers that touch hardware)
  * interrupts, wakeups, DMA, and other hardware state.  There may also be
- * internal transitions to various low power modes, which are transparent
+ * internal transitions to various low-power modes which are transparent
  * to the rest of the driver stack (such as a driver that's ON gating off
  * clocks which are not in active use).
  *
- * The externally visible transitions are handled with the help of the following
- * callbacks included in this structure:
+ * The externally visible transitions are handled with the help of callbacks
+ * included in this structure in such a way that two levels of callbacks are
+ * involved.  First, the PM core executes callbacks provided by PM domains,
+ * device types, classes and bus types.  They are the subsystem-level callbacks
+ * supposed to execute callbacks provided by device drivers, although they may
+ * choose not to do that.  If the driver callbacks are executed, they have to
+ * collaborate with the subsystem-level callbacks to achieve the goals
+ * appropriate for the given system transition, given transition phase and the
+ * subsystem the device belongs to.
  *
- * @prepare: Prepare the device for the upcoming transition, but do NOT change
- *	its hardware state.  Prevent new children of the device from being
- *	registered after @prepare() returns (the driver's subsystem and
- *	generally the rest of the kernel is supposed to prevent new calls to the
- *	probe method from being made too once @prepare() has succeeded).  If
- *	@prepare() detects a situation it cannot handle (e.g. registration of a
- *	child already in progress), it may return -EAGAIN, so that the PM core
- *	can execute it once again (e.g. after the new child has been registered)
- *	to recover from the race condition.  This method is executed for all
- *	kinds of suspend transitions and is followed by one of the suspend
- *	callbacks: @suspend(), @freeze(), or @poweroff().
- *	The PM core executes @prepare() for all devices before starting to
- *	execute suspend callbacks for any of them, so drivers may assume all of
- *	the other devices to be present and functional while @prepare() is being
- *	executed.  In particular, it is safe to make GFP_KERNEL memory
- *	allocations from within @prepare().  However, drivers may NOT assume
- *	anything about the availability of the user space at that time and it
- *	is not correct to request firmware from within @prepare() (it's too
- *	late to do that).  [To work around this limitation, drivers may
- *	register suspend and hibernation notifiers that are executed before the
- *	freezing of tasks.]
+ * @prepare: The principal role of this callback is to prevent new children of
+ *	the device from being registered after it has returned (the driver's
+ *	subsystem and generally the rest of the kernel is supposed to prevent
+ *	new calls to the probe method from being made too once @prepare() has
+ *	succeeded).  If @prepare() detects a situation it cannot handle (e.g.
+ *	registration of a child already in progress), it may return -EAGAIN, so
+ *	that the PM core can execute it once again (e.g. after a new child has
+ *	been registered) to recover from the race condition.
+ *	This method is executed for all kinds of suspend transitions and is
+ *	followed by one of the suspend callbacks: @suspend(), @freeze(), or
+ *	@poweroff().  The PM core executes subsystem-level @prepare() for all
+ *	devices before starting to invoke suspend callbacks for any of them, so
+ *	generally devices may be assumed to be functional or to respond to
+ *	runtime resume requests while @prepare() is being executed.  However,
+ *	device drivers may NOT assume anything about the availability of user
+ *	space at that time and it is NOT valid to request firmware from within
+ *	@prepare() (it's too late to do that).  It also is NOT valid to allocate
+ *	substantial amounts of memory from @prepare() in the GFP_KERNEL mode.
+ *	[To work around these limitations, drivers may register suspend and
+ *	hibernation notifiers to be executed before the freezing of tasks.]
  *
  * @complete: Undo the changes made by @prepare().  This method is executed for
  *	all kinds of resume transitions, following one of the resume callbacks:
  *	@resume(), @thaw(), @restore().  Also called if the state transition
- *	fails before the driver's suspend callback (@suspend(), @freeze(),
- *	@poweroff()) can be executed (e.g. if the suspend callback fails for one
+ *	fails before the driver's suspend callback: @suspend(), @freeze() or
+ *	@poweroff(), can be executed (e.g. if the suspend callback fails for one
  *	of the other devices that the PM core has unsuccessfully attempted to
  *	suspend earlier).
- *	The PM core executes @complete() after it has executed the appropriate
- *	resume callback for all devices.
+ *	The PM core executes subsystem-level @complete() after it has executed
+ *	the appropriate resume callbacks for all devices.
  *
  * @suspend: Executed before putting the system into a sleep state in which the
- *	contents of main memory are preserved.  Quiesce the device, put it into
- *	a low power state appropriate for the upcoming system state (such as
- *	PCI_D3hot), and enable wakeup events as appropriate.
+ *	contents of main memory are preserved.  The exact action to perform
+ *	depends on the device's subsystem (PM domain, device type, class or bus
+ *	type), but generally the device must be quiescent after subsystem-level
+ *	@suspend() has returned, so that it doesn't do any I/O or DMA.
+ *	Subsystem-level @suspend() is executed for all devices after invoking
+ *	subsystem-level @prepare() for all of them.
  *
  * @resume: Executed after waking the system up from a sleep state in which the
- *	contents of main memory were preserved.  Put the device into the
- *	appropriate state, according to the information saved in memory by the
- *	preceding @suspend().  The driver starts working again, responding to
- *	hardware events and software requests.  The hardware may have gone
- *	through a power-off reset, or it may have maintained state from the
- *	previous suspend() which the driver may rely on while resuming.  On most
- *	platforms, there are no restrictions on availability of resources like
- *	clocks during @resume().
+ *	contents of main memory were preserved.  The exact action to perform
+ *	depends on the device's subsystem, but generally the driver is expected
+ *	to start working again, responding to hardware events and software
+ *	requests (the device itself may be left in a low-power state, waiting
+ *	for a runtime resume to occur).  The state of the device at the time its
+ *	driver's @resume() callback is run depends on the platform and subsystem
+ *	the device belongs to.  On most platforms, there are no restrictions on
+ *	availability of resources like clocks during @resume().
+ *	Subsystem-level @resume() is executed for all devices after invoking
+ *	subsystem-level @resume_noirq() for all of them.
  *
  * @freeze: Hibernation-specific, executed before creating a hibernation image.
- *	Quiesce operations so that a consistent image can be created, but do NOT
- *	otherwise put the device into a low power device state and do NOT emit
- *	system wakeup events.  Save in main memory the device settings to be
- *	used by @restore() during the subsequent resume from hibernation or by
- *	the subsequent @thaw(), if the creation of the image or the restoration
- *	of main memory contents from it fails.
+ *	Analogous to @suspend(), but it should not enable the device to signal
+ *	wakeup events or change its power state.  The majority of subsystems
+ *	(with the notable exception of the PCI bus type) expect the driver-level
+ *	@freeze() to save the device settings in memory to be used by @restore()
+ *	during the subsequent resume from hibernation.
+ *	Subsystem-level @freeze() is executed for all devices after invoking
+ *	subsystem-level @prepare() for all of them.
  *
  * @thaw: Hibernation-specific, executed after creating a hibernation image OR
- *	if the creation of the image fails.  Also executed after a failing
+ *	if the creation of an image has failed.  Also executed after a failing
  *	attempt to restore the contents of main memory from such an image.
  *	Undo the changes made by the preceding @freeze(), so the device can be
  *	operated in the same way as immediately before the call to @freeze().
+ *	Subsystem-level @thaw() is executed for all devices after invoking
+ *	subsystem-level @thaw_noirq() for all of them.  It also may be executed
+ *	directly after @freeze() in case of a transition error.
  *
  * @poweroff: Hibernation-specific, executed after saving a hibernation image.
- *	Quiesce the device, put it into a low power state appropriate for the
- *	upcoming system state (such as PCI_D3hot), and enable wakeup events as
- *	appropriate.
+ *	Analogous to @suspend(), but it need not save the device's settings in
+ *	memory.
+ *	Subsystem-level @poweroff() is executed for all devices after invoking
+ *	subsystem-level @prepare() for all of them.
  *
  * @restore: Hibernation-specific, executed after restoring the contents of main
- *	memory from a hibernation image.  Driver starts working again,
- *	responding to hardware events and software requests.  Drivers may NOT
- *	make ANY assumptions about the hardware state right prior to @restore().
- *	On most platforms, there are no restrictions on availability of
- *	resources like clocks during @restore().
+ *	memory from a hibernation image, analogous to @resume().
  *
- * @suspend_noirq: Complete the operations of ->suspend() by carrying out any
- *	actions required for suspending the device that need interrupts to be
- *	disabled
+ * @suspend_noirq: Complete the actions started by @suspend().  Carry out any
+ *	additional operations required for suspending the device that might be
+ *	racing with its driver's interrupt handler, which is guaranteed not to
+ *	run while @suspend_noirq() is being executed.
+ *	It generally is expected that the device will be in a low-power state
+ *	(appropriate for the target system sleep state) after subsystem-level
+ *	@suspend_noirq() has returned successfully.  If the device can generate
+ *	system wakeup signals and is enabled to wake up the system, it should be
+ *	configured to do so at that time.  However, depending on the platform
+ *	and device's subsystem, @suspend() may be allowed to put the device into
+ *	the low-power state and configure it to generate wakeup signals, in
+ *	which case it generally is not necessary to define @suspend_noirq().
  *
- * @resume_noirq: Prepare for the execution of ->resume() by carrying out any
- *	actions required for resuming the device that need interrupts to be
- *	disabled
+ * @resume_noirq: Prepare for the execution of @resume() by carrying out any
+ *	operations required for resuming the device that might be racing with
+ *	its driver's interrupt handler, which is guaranteed not to run while
+ *	@resume_noirq() is being executed.
  *
- * @freeze_noirq: Complete the operations of ->freeze() by carrying out any
- *	actions required for freezing the device that need interrupts to be
- *	disabled
+ * @freeze_noirq: Complete the actions started by @freeze().  Carry out any
+ *	additional operations required for freezing the device that might be
+ *	racing with its driver's interrupt handler, which is guaranteed not to
+ *	run while @freeze_noirq() is being executed.
+ *	The power state of the device should not be changed by either @freeze()
+ *	or @freeze_noirq() and it should not be configured to signal system
+ *	wakeup by any of these callbacks.
  *
- * @thaw_noirq: Prepare for the execution of ->thaw() by carrying out any
- *	actions required for thawing the device that need interrupts to be
- *	disabled
+ * @thaw_noirq: Prepare for the execution of @thaw() by carrying out any
+ *	operations required for thawing the device that might be racing with its
+ *	driver's interrupt handler, which is guaranteed not to run while
+ *	@thaw_noirq() is being executed.
  *
- * @poweroff_noirq: Complete the operations of ->poweroff() by carrying out any
- *	actions required for handling the device that need interrupts to be
- *	disabled
+ * @poweroff_noirq: Complete the actions started by @poweroff().  Analogous to
+ *	@suspend_noirq(), but it need not save the device's settings in memory.
  *
- * @restore_noirq: Prepare for the execution of ->restore() by carrying out any
- *	actions required for restoring the operations of the device that need
- *	interrupts to be disabled
+ * @restore_noirq: Prepare for the execution of @restore() by carrying out any
+ *	operations required for thawing the device that might be racing with its
+ *	driver's interrupt handler, which is guaranteed not to run while
+ *	@restore_noirq() is being executed.  Analogous to @resume_noirq().
  *
  * All of the above callbacks, except for @complete(), return error codes.
  * However, the error codes returned by the resume operations, @resume(),
- * @thaw(), @restore(), @resume_noirq(), @thaw_noirq(), and @restore_noirq() do
+ * @thaw(), @restore(), @resume_noirq(), @thaw_noirq(), and @restore_noirq(), do
  * not cause the PM core to abort the resume transition during which they are
- * returned.  The error codes returned in that cases are only printed by the PM
+ * returned.  The error codes returned in those cases are only printed by the PM
  * core to the system logs for debugging purposes.  Still, it is recommended
  * that drivers only return error codes from their resume methods in case of an
  * unrecoverable failure (i.e. when the device being handled refuses to resume
@@ -174,31 +201,43 @@
  * their children.
  *
  * It is allowed to unregister devices while the above callbacks are being
- * executed.  However, it is not allowed to unregister a device from within any
- * of its own callbacks.
+ * executed.  However, a callback routine must NOT try to unregister the device
+ * it was called for, although it may unregister children of that device (for
+ * example, if it detects that a child was unplugged while the system was
+ * asleep).
  *
- * There also are the following callbacks related to run-time power management
- * of devices:
+ * Refer to Documentation/power/devices.txt for more information about the role
+ * of the above callbacks in the system suspend process.
+ *
+ * There also are callbacks related to runtime power management of devices.
+ * Again, these callbacks are executed by the PM core only for subsystems
+ * (PM domains, device types, classes and bus types) and the subsystem-level
+ * callbacks are supposed to invoke the driver callbacks.  Moreover, the exact
+ * actions to be performed by a device driver's callbacks generally depend on
+ * the platform and subsystem the device belongs to.
  *
  * @runtime_suspend: Prepare the device for a condition in which it won't be
  *	able to communicate with the CPU(s) and RAM due to power management.
- *	This need not mean that the device should be put into a low power state.
+ *	This need not mean that the device should be put into a low-power state.
  *	For example, if the device is behind a link which is about to be turned
  *	off, the device may remain at full power.  If the device does go to low
- *	power and is capable of generating run-time wake-up events, remote
- *	wake-up (i.e., a hardware mechanism allowing the device to request a
- *	change of its power state via a wake-up event, such as PCI PME) should
- *	be enabled for it.
+ *	power and is capable of generating runtime wakeup events, remote wakeup
+ *	(i.e., a hardware mechanism allowing the device to request a change of
+ *	its power state via an interrupt) should be enabled for it.
  *
  * @runtime_resume: Put the device into the fully active state in response to a
- *	wake-up event generated by hardware or at the request of software.  If
- *	necessary, put the device into the full power state and restore its
+ *	wakeup event generated by hardware or at the request of software.  If
+ *	necessary, put the device into the full-power state and restore its
  *	registers, so that it is fully operational.
  *
- * @runtime_idle: Device appears to be inactive and it might be put into a low
- *	power state if all of the necessary conditions are satisfied.  Check
+ * @runtime_idle: Device appears to be inactive and it might be put into a
+ *	low-power state if all of the necessary conditions are satisfied.  Check
  *	these conditions and handle the device as appropriate, possibly queueing
  *	a suspend request for it.  The return value is ignored by the PM core.
+ *
+ * Refer to Documentation/power/runtime_pm.txt for more information about the
+ * role of the above callbacks in device runtime power management.
+ *
  */
 
 struct dev_pm_ops {
@@ -447,6 +486,7 @@
 	unsigned int		async_suspend:1;
 	bool			is_prepared:1;	/* Owned by the PM core */
 	bool			is_suspended:1;	/* Ditto */
+	bool			ignore_children:1;
 	spinlock_t		lock;
 #ifdef CONFIG_PM_SLEEP
 	struct list_head	entry;
@@ -464,7 +504,6 @@
 	atomic_t		usage_count;
 	atomic_t		child_count;
 	unsigned int		disable_depth:3;
-	unsigned int		ignore_children:1;
 	unsigned int		idle_notification:1;
 	unsigned int		request_pending:1;
 	unsigned int		deferred_resume:1;
diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h
index d8d9036..d3085e7 100644
--- a/include/linux/pm_runtime.h
+++ b/include/linux/pm_runtime.h
@@ -52,11 +52,6 @@
 		|| !atomic_read(&dev->power.child_count);
 }
 
-static inline void pm_suspend_ignore_children(struct device *dev, bool enable)
-{
-	dev->power.ignore_children = enable;
-}
-
 static inline void pm_runtime_get_noresume(struct device *dev)
 {
 	atomic_inc(&dev->power.usage_count);
@@ -130,7 +125,6 @@
 static inline void pm_runtime_forbid(struct device *dev) {}
 
 static inline bool pm_children_suspended(struct device *dev) { return false; }
-static inline void pm_suspend_ignore_children(struct device *dev, bool en) {}
 static inline void pm_runtime_get_noresume(struct device *dev) {}
 static inline void pm_runtime_put_noidle(struct device *dev) {}
 static inline bool device_run_wake(struct device *dev) { return false; }
diff --git a/include/linux/pstore.h b/include/linux/pstore.h
index ea56732..2ca8cde 100644
--- a/include/linux/pstore.h
+++ b/include/linux/pstore.h
@@ -35,10 +35,12 @@
 	spinlock_t	buf_lock;	/* serialize access to 'buf' */
 	char		*buf;
 	size_t		bufsize;
+	struct mutex	read_mutex;	/* serialize open/read/close */
 	int		(*open)(struct pstore_info *psi);
 	int		(*close)(struct pstore_info *psi);
 	ssize_t		(*read)(u64 *id, enum pstore_type_id *type,
-			struct timespec *time, struct pstore_info *psi);
+			struct timespec *time, char **buf,
+			struct pstore_info *psi);
 	int		(*write)(enum pstore_type_id type, u64 *id,
 			unsigned int part, size_t size, struct pstore_info *psi);
 	int		(*erase)(enum pstore_type_id type, u64 id,
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 68daf4f..1c4f3e9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1521,7 +1521,6 @@
 #ifdef CONFIG_FAULT_INJECTION
 	int make_it_fail;
 #endif
-	struct prop_local_single dirties;
 	/*
 	 * when (nr_dirtied >= nr_dirtied_pause), it's time to call
 	 * balance_dirty_pages() for some dirty throttling pause
diff --git a/include/linux/serial.h b/include/linux/serial.h
index 97ff8e2..3d86517 100644
--- a/include/linux/serial.h
+++ b/include/linux/serial.h
@@ -207,13 +207,15 @@
 
 struct serial_rs485 {
 	__u32	flags;			/* RS485 feature flags */
-#define SER_RS485_ENABLED		(1 << 0)
-#define SER_RS485_RTS_ON_SEND		(1 << 1)
-#define SER_RS485_RTS_AFTER_SEND	(1 << 2)
-#define SER_RS485_RTS_BEFORE_SEND	(1 << 3)
+#define SER_RS485_ENABLED		(1 << 0)	/* If enabled */
+#define SER_RS485_RTS_ON_SEND		(1 << 1)	/* Logical level for
+							   RTS pin when
+							   sending */
+#define SER_RS485_RTS_AFTER_SEND	(1 << 2)	/* Logical level for
+							   RTS pin after sent*/
 #define SER_RS485_RX_DURING_TX		(1 << 4)
-	__u32	delay_rts_before_send;	/* Milliseconds */
-	__u32	delay_rts_after_send;	/* Milliseconds */
+	__u32	delay_rts_before_send;	/* Delay before send (milliseconds) */
+	__u32	delay_rts_after_send;	/* Delay after send (milliseconds) */
 	__u32	padding[5];		/* Memory is cheap, new structs
 					   are a royal PITA .. */
 };
diff --git a/include/linux/serial_sci.h b/include/linux/serial_sci.h
index 0efa1f1..369273a 100644
--- a/include/linux/serial_sci.h
+++ b/include/linux/serial_sci.h
@@ -67,6 +67,7 @@
 	SCIx_IRDA_REGTYPE,
 	SCIx_SCIFA_REGTYPE,
 	SCIx_SCIFB_REGTYPE,
+	SCIx_SH2_SCIF_FIFODATA_REGTYPE,
 	SCIx_SH3_SCIF_REGTYPE,
 	SCIx_SH4_SCIF_REGTYPE,
 	SCIx_SH4_SCIF_NO_SCSPTR_REGTYPE,
diff --git a/include/linux/sh_clk.h b/include/linux/sh_clk.h
index 3ccf186..a20831c 100644
--- a/include/linux/sh_clk.h
+++ b/include/linux/sh_clk.h
@@ -52,7 +52,6 @@
 
 	unsigned long		arch_flags;
 	void			*priv;
-	struct dentry		*dentry;
 	struct clk_mapping	*mapping;
 	struct cpufreq_frequency_table *freq_table;
 	unsigned int		nr_freqs;
@@ -94,6 +93,9 @@
 long clk_rate_div_range_round(struct clk *clk, unsigned int div_min,
 			      unsigned int div_max, unsigned long rate);
 
+long clk_rate_mult_range_round(struct clk *clk, unsigned int mult_min,
+			       unsigned int mult_max, unsigned long rate);
+
 long clk_round_parent(struct clk *clk, unsigned long target,
 		      unsigned long *best_freq, unsigned long *parent_freq,
 		      unsigned int div_min, unsigned int div_max);
diff --git a/include/linux/sh_pfc.h b/include/linux/sh_pfc.h
index bc8c920..8446789 100644
--- a/include/linux/sh_pfc.h
+++ b/include/linux/sh_pfc.h
@@ -104,4 +104,80 @@
 int register_pinmux(struct pinmux_info *pip);
 int unregister_pinmux(struct pinmux_info *pip);
 
+/* helper macro for port */
+#define PORT_1(fn, pfx, sfx) fn(pfx, sfx)
+
+#define PORT_10(fn, pfx, sfx) \
+	PORT_1(fn, pfx##0, sfx), PORT_1(fn, pfx##1, sfx),	\
+	PORT_1(fn, pfx##2, sfx), PORT_1(fn, pfx##3, sfx),	\
+	PORT_1(fn, pfx##4, sfx), PORT_1(fn, pfx##5, sfx),	\
+	PORT_1(fn, pfx##6, sfx), PORT_1(fn, pfx##7, sfx),	\
+	PORT_1(fn, pfx##8, sfx), PORT_1(fn, pfx##9, sfx)
+
+#define PORT_90(fn, pfx, sfx) \
+	PORT_10(fn, pfx##1, sfx), PORT_10(fn, pfx##2, sfx),	\
+	PORT_10(fn, pfx##3, sfx), PORT_10(fn, pfx##4, sfx),	\
+	PORT_10(fn, pfx##5, sfx), PORT_10(fn, pfx##6, sfx),	\
+	PORT_10(fn, pfx##7, sfx), PORT_10(fn, pfx##8, sfx),	\
+	PORT_10(fn, pfx##9, sfx)
+
+#define _PORT_ALL(pfx, sfx) pfx##_##sfx
+#define _GPIO_PORT(pfx, sfx) PINMUX_GPIO(GPIO_PORT##pfx, PORT##pfx##_DATA)
+#define PORT_ALL(str)	CPU_ALL_PORT(_PORT_ALL, PORT, str)
+#define GPIO_PORT_ALL()	CPU_ALL_PORT(_GPIO_PORT, , unused)
+#define GPIO_FN(str) PINMUX_GPIO(GPIO_FN_##str, str##_MARK)
+
+/* helper macro for pinmux_enum_t */
+#define PORT_DATA_I(nr)	\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_IN)
+
+#define PORT_DATA_I_PD(nr)	\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
+		    PORT##nr##_IN, PORT##nr##_IN_PD)
+
+#define PORT_DATA_I_PU(nr)	\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,	\
+		    PORT##nr##_IN, PORT##nr##_IN_PU)
+
+#define PORT_DATA_I_PU_PD(nr)	\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0,			\
+		    PORT##nr##_IN, PORT##nr##_IN_PD, PORT##nr##_IN_PU)
+
+#define PORT_DATA_O(nr)		\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT)
+
+#define PORT_DATA_IO(nr)	\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT,	\
+		    PORT##nr##_IN)
+
+#define PORT_DATA_IO_PD(nr)	\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT,	\
+		    PORT##nr##_IN, PORT##nr##_IN_PD)
+
+#define PORT_DATA_IO_PU(nr)	\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT,	\
+		    PORT##nr##_IN, PORT##nr##_IN_PU)
+
+#define PORT_DATA_IO_PU_PD(nr)	\
+	PINMUX_DATA(PORT##nr##_DATA, PORT##nr##_FN0, PORT##nr##_OUT,	\
+		    PORT##nr##_IN, PORT##nr##_IN_PD, PORT##nr##_IN_PU)
+
+/* helper macro for top 4 bits in PORTnCR */
+#define _PCRH(in, in_pd, in_pu, out)	\
+	0, (out), (in), 0,		\
+	0, 0, 0, 0,			\
+	0, 0, (in_pd), 0,		\
+	0, 0, (in_pu), 0
+
+#define PORTCR(nr, reg)							\
+	{								\
+		PINMUX_CFG_REG("PORT" nr "CR", reg, 8, 4) {		\
+			_PCRH(PORT##nr##_IN, PORT##nr##_IN_PD,		\
+			      PORT##nr##_IN_PU, PORT##nr##_OUT),	\
+				PORT##nr##_FN0, PORT##nr##_FN1,		\
+				PORT##nr##_FN2, PORT##nr##_FN3,		\
+				PORT##nr##_FN4, PORT##nr##_FN5,		\
+				PORT##nr##_FN6, PORT##nr##_FN7 }	\
+	}
+
 #endif /* __SH_PFC_H */
diff --git a/include/linux/shrinker.h b/include/linux/shrinker.h
index a83833a..07ceb97 100644
--- a/include/linux/shrinker.h
+++ b/include/linux/shrinker.h
@@ -35,7 +35,7 @@
 
 	/* These are for internal use */
 	struct list_head list;
-	long nr;	/* objs pending delete */
+	atomic_long_t nr_in_batch; /* objs pending delete */
 };
 #define DEFAULT_SEEKS 2 /* A good number if you don't know better. */
 extern void register_shrinker(struct shrinker *);
diff --git a/include/linux/sigma.h b/include/linux/sigma.h
index e2accb3..d0de882 100644
--- a/include/linux/sigma.h
+++ b/include/linux/sigma.h
@@ -24,7 +24,7 @@
 struct sigma_firmware_header {
 	unsigned char magic[7];
 	u8 version;
-	u32 crc;
+	__le32 crc;
 };
 
 enum {
@@ -40,19 +40,14 @@
 struct sigma_action {
 	u8 instr;
 	u8 len_hi;
-	u16 len;
-	u16 addr;
+	__le16 len;
+	__be16 addr;
 	unsigned char payload[];
 };
 
 static inline u32 sigma_action_len(struct sigma_action *sa)
 {
-	return (sa->len_hi << 16) | sa->len;
-}
-
-static inline size_t sigma_action_size(struct sigma_action *sa, u32 payload_len)
-{
-	return sizeof(*sa) + payload_len + (payload_len % 2);
+	return (sa->len_hi << 16) | le16_to_cpu(sa->len);
 }
 
 extern int process_sigma_firmware(struct i2c_client *client, const char *name);
diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h
index add4790..e9e72bd 100644
--- a/include/linux/virtio_config.h
+++ b/include/linux/virtio_config.h
@@ -85,6 +85,8 @@
  * @reset: reset the device
  *	vdev: the virtio device
  *	After this, status and feature negotiation must be done again
+ *	Device must not be reset from its vq/config callbacks, or in
+ *	parallel with being added/removed.
  * @find_vqs: find virtqueues and instantiate them.
  *	vdev: the virtio_device
  *	nvqs: the number of virtqueues to find
diff --git a/include/linux/virtio_mmio.h b/include/linux/virtio_mmio.h
index 27c7ede..5c7b6f0 100644
--- a/include/linux/virtio_mmio.h
+++ b/include/linux/virtio_mmio.h
@@ -63,7 +63,7 @@
 #define VIRTIO_MMIO_GUEST_FEATURES	0x020
 
 /* Activated features set selector - Write Only */
-#define VIRTIO_MMIO_GUEST_FEATURES_SET	0x024
+#define VIRTIO_MMIO_GUEST_FEATURES_SEL	0x024
 
 /* Guest's memory page size in bytes - Write Only */
 #define VIRTIO_MMIO_GUEST_PAGE_SIZE	0x028
diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index 687fb11..4bde182 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -119,7 +119,7 @@
 #endif
 
 /* Allocate/destroy a 'vmalloc' VM area. */
-extern struct vm_struct *alloc_vm_area(size_t size);
+extern struct vm_struct *alloc_vm_area(size_t size, pte_t **ptes);
 extern void free_vm_area(struct vm_struct *area);
 
 /* for /dev/kmem */
diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h
index ab90ae0..6cc18f3 100644
--- a/include/net/bluetooth/l2cap.h
+++ b/include/net/bluetooth/l2cap.h
@@ -39,8 +39,11 @@
 #define L2CAP_DEFAULT_ACK_TO		200
 #define L2CAP_LE_DEFAULT_MTU		23
 
-#define L2CAP_CONN_TIMEOUT	(40000) /* 40 seconds */
-#define L2CAP_INFO_TIMEOUT	(4000)  /*  4 seconds */
+#define L2CAP_DISC_TIMEOUT             (100)
+#define L2CAP_DISC_REJ_TIMEOUT         (5000)  /*  5 seconds */
+#define L2CAP_ENC_TIMEOUT              (5000)  /*  5 seconds */
+#define L2CAP_CONN_TIMEOUT             (40000) /* 40 seconds */
+#define L2CAP_INFO_TIMEOUT             (4000)  /*  4 seconds */
 
 /* L2CAP socket address */
 struct sockaddr_l2 {
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 92cf1c2..95852e3 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -456,6 +456,9 @@
  *	as the AC bitmap in the QoS info field
  * @max_sp: max Service Period. same format as the MAX_SP in the
  *	QoS info field (but already shifted down)
+ * @sta_modify_mask: bitmap indicating which parameters changed
+ *	(for those that don't have a natural "no change" value),
+ *	see &enum station_parameters_apply_mask
  */
 struct station_parameters {
 	u8 *supported_rates;
@@ -615,6 +618,7 @@
  *	user space MLME/SME implementation. The information is provided for
  *	the cfg80211_new_sta() calls to notify user space of the IEs.
  * @assoc_req_ies_len: Length of assoc_req_ies buffer in octets.
+ * @sta_flags: station flags mask & values
  */
 struct station_info {
 	u32 filled;
diff --git a/include/net/dst.h b/include/net/dst.h
index 4fb6c43..6faec1a 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -205,12 +205,7 @@
 
 static inline u32 dst_mtu(const struct dst_entry *dst)
 {
-	u32 mtu = dst_metric_raw(dst, RTAX_MTU);
-
-	if (!mtu)
-		mtu = dst->ops->default_mtu(dst);
-
-	return mtu;
+	return dst->ops->mtu(dst);
 }
 
 /* RTT metrics are stored in milliseconds for user ABI, but used as jiffies */
diff --git a/include/net/dst_ops.h b/include/net/dst_ops.h
index 9adb998..e1c2ee0 100644
--- a/include/net/dst_ops.h
+++ b/include/net/dst_ops.h
@@ -17,7 +17,7 @@
 	int			(*gc)(struct dst_ops *ops);
 	struct dst_entry *	(*check)(struct dst_entry *, __u32 cookie);
 	unsigned int		(*default_advmss)(const struct dst_entry *);
-	unsigned int		(*default_mtu)(const struct dst_entry *);
+	unsigned int		(*mtu)(const struct dst_entry *);
 	u32 *			(*cow_metrics)(struct dst_entry *, unsigned long);
 	void			(*destroy)(struct dst_entry *);
 	void			(*ifdown)(struct dst_entry *,
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index b897d6e..f941964 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -31,6 +31,7 @@
 /** struct ip_options - IP Options
  *
  * @faddr - Saved first hop address
+ * @nexthop - Saved nexthop address in LSRR and SSRR
  * @is_data - Options in __data, rather than skb
  * @is_strictroute - Strict source route
  * @srr_is_hit - Packet destination addr was our one
@@ -41,6 +42,7 @@
  */
 struct ip_options {
 	__be32		faddr;
+	__be32		nexthop;
 	unsigned char	optlen;
 	unsigned char	srr;
 	unsigned char	rr;
diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h
index 78c83e6..e9ff3fc 100644
--- a/include/net/inetpeer.h
+++ b/include/net/inetpeer.h
@@ -35,6 +35,7 @@
 
 	u32			metrics[RTAX_MAX];
 	u32			rate_tokens;	/* rate limiting for ICMP */
+	int			redirect_genid;
 	unsigned long		rate_last;
 	unsigned long		pmtu_expires;
 	u32			pmtu_orig;
diff --git a/include/net/netfilter/nf_conntrack_ecache.h b/include/net/netfilter/nf_conntrack_ecache.h
index 4283508..a88fb69 100644
--- a/include/net/netfilter/nf_conntrack_ecache.h
+++ b/include/net/netfilter/nf_conntrack_ecache.h
@@ -67,18 +67,18 @@
 	int (*fcn)(unsigned int events, struct nf_ct_event *item);
 };
 
-extern struct nf_ct_event_notifier __rcu *nf_conntrack_event_cb;
-extern int nf_conntrack_register_notifier(struct nf_ct_event_notifier *nb);
-extern void nf_conntrack_unregister_notifier(struct nf_ct_event_notifier *nb);
+extern int nf_conntrack_register_notifier(struct net *net, struct nf_ct_event_notifier *nb);
+extern void nf_conntrack_unregister_notifier(struct net *net, struct nf_ct_event_notifier *nb);
 
 extern void nf_ct_deliver_cached_events(struct nf_conn *ct);
 
 static inline void
 nf_conntrack_event_cache(enum ip_conntrack_events event, struct nf_conn *ct)
 {
+	struct net *net = nf_ct_net(ct);
 	struct nf_conntrack_ecache *e;
 
-	if (nf_conntrack_event_cb == NULL)
+	if (net->ct.nf_conntrack_event_cb == NULL)
 		return;
 
 	e = nf_ct_ecache_find(ct);
@@ -95,11 +95,12 @@
 			      int report)
 {
 	int ret = 0;
+	struct net *net = nf_ct_net(ct);
 	struct nf_ct_event_notifier *notify;
 	struct nf_conntrack_ecache *e;
 
 	rcu_read_lock();
-	notify = rcu_dereference(nf_conntrack_event_cb);
+	notify = rcu_dereference(net->ct.nf_conntrack_event_cb);
 	if (notify == NULL)
 		goto out_unlock;
 
@@ -164,9 +165,8 @@
 	int (*fcn)(unsigned int events, struct nf_exp_event *item);
 };
 
-extern struct nf_exp_event_notifier __rcu *nf_expect_event_cb;
-extern int nf_ct_expect_register_notifier(struct nf_exp_event_notifier *nb);
-extern void nf_ct_expect_unregister_notifier(struct nf_exp_event_notifier *nb);
+extern int nf_ct_expect_register_notifier(struct net *net, struct nf_exp_event_notifier *nb);
+extern void nf_ct_expect_unregister_notifier(struct net *net, struct nf_exp_event_notifier *nb);
 
 static inline void
 nf_ct_expect_event_report(enum ip_conntrack_expect_events event,
@@ -174,11 +174,12 @@
 			  u32 pid,
 			  int report)
 {
+	struct net *net = nf_ct_exp_net(exp);
 	struct nf_exp_event_notifier *notify;
 	struct nf_conntrack_ecache *e;
 
 	rcu_read_lock();
-	notify = rcu_dereference(nf_expect_event_cb);
+	notify = rcu_dereference(net->ct.nf_expect_event_cb);
 	if (notify == NULL)
 		goto out_unlock;
 
diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h
index 0249399..7a911ec 100644
--- a/include/net/netns/conntrack.h
+++ b/include/net/netns/conntrack.h
@@ -18,6 +18,8 @@
 	struct hlist_nulls_head	unconfirmed;
 	struct hlist_nulls_head	dying;
 	struct ip_conntrack_stat __percpu *stat;
+	struct nf_ct_event_notifier __rcu *nf_conntrack_event_cb;
+	struct nf_exp_event_notifier __rcu *nf_expect_event_cb;
 	int			sysctl_events;
 	unsigned int		sysctl_events_retry_timeout;
 	int			sysctl_acct;
diff --git a/include/net/red.h b/include/net/red.h
index 3319f16..b72a3b8 100644
--- a/include/net/red.h
+++ b/include/net/red.h
@@ -116,7 +116,7 @@
 	u32		qR;		/* Cached random number */
 
 	unsigned long	qavg;		/* Average queue length: A scaled */
-	psched_time_t	qidlestart;	/* Start of current idle period */
+	ktime_t		qidlestart;	/* Start of current idle period */
 };
 
 static inline u32 red_rmask(u8 Plog)
@@ -148,17 +148,17 @@
 
 static inline int red_is_idling(struct red_parms *p)
 {
-	return p->qidlestart != PSCHED_PASTPERFECT;
+	return p->qidlestart.tv64 != 0;
 }
 
 static inline void red_start_of_idle_period(struct red_parms *p)
 {
-	p->qidlestart = psched_get_time();
+	p->qidlestart = ktime_get();
 }
 
 static inline void red_end_of_idle_period(struct red_parms *p)
 {
-	p->qidlestart = PSCHED_PASTPERFECT;
+	p->qidlestart.tv64 = 0;
 }
 
 static inline void red_restart(struct red_parms *p)
@@ -170,13 +170,10 @@
 
 static inline unsigned long red_calc_qavg_from_idle_time(struct red_parms *p)
 {
-	psched_time_t now;
-	long us_idle;
+	s64 delta = ktime_us_delta(ktime_get(), p->qidlestart);
+	long us_idle = min_t(s64, delta, p->Scell_max);
 	int  shift;
 
-	now = psched_get_time();
-	us_idle = psched_tdiff_bounded(now, p->qidlestart, p->Scell_max);
-
 	/*
 	 * The problem: ideally, average length queue recalcultion should
 	 * be done over constant clock intervals. This is too expensive, so
diff --git a/include/net/route.h b/include/net/route.h
index db7b343..91855d1 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -71,12 +71,12 @@
 	struct fib_info		*fi; /* for client ref to shared metrics */
 };
 
-static inline bool rt_is_input_route(struct rtable *rt)
+static inline bool rt_is_input_route(const struct rtable *rt)
 {
 	return rt->rt_route_iif != 0;
 }
 
-static inline bool rt_is_output_route(struct rtable *rt)
+static inline bool rt_is_output_route(const struct rtable *rt)
 {
 	return rt->rt_route_iif == 0;
 }
diff --git a/include/scsi/libfcoe.h b/include/scsi/libfcoe.h
index d1e95c6..5a35a2a 100644
--- a/include/scsi/libfcoe.h
+++ b/include/scsi/libfcoe.h
@@ -147,6 +147,7 @@
 	u8 map_dest;
 	u8 spma;
 	u8 probe_tries;
+	u8 priority;
 	u8 dest_addr[ETH_ALEN];
 	u8 ctl_src_addr[ETH_ALEN];
 
@@ -301,6 +302,7 @@
  * @lport:		       The associated local port
  * @fcoe_pending_queue:	       The pending Rx queue of skbs
  * @fcoe_pending_queue_active: Indicates if the pending queue is active
+ * @priority:		       Packet priority (DCB)
  * @max_queue_depth:	       Max queue depth of pending queue
  * @min_queue_depth:	       Min queue depth of pending queue
  * @timer:		       The queue timer
@@ -316,6 +318,7 @@
 	struct fc_lport	      *lport;
 	struct sk_buff_head   fcoe_pending_queue;
 	u8		      fcoe_pending_queue_active;
+	u8		      priority;
 	u32		      max_queue_depth;
 	u32		      min_queue_depth;
 	struct timer_list     timer;
diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h
index 7f5fed3..6873c7d 100644
--- a/include/target/target_core_base.h
+++ b/include/target/target_core_base.h
@@ -103,9 +103,10 @@
 	SCF_SCSI_NON_DATA_CDB		= 0x00000040,
 	SCF_SCSI_CDB_EXCEPTION		= 0x00000080,
 	SCF_SCSI_RESERVATION_CONFLICT	= 0x00000100,
-	SCF_SE_CMD_FAILED		= 0x00000400,
+	SCF_FUA				= 0x00000200,
 	SCF_SE_LUN_CMD			= 0x00000800,
 	SCF_SE_ALLOW_EOO		= 0x00001000,
+	SCF_BIDI			= 0x00002000,
 	SCF_SENT_CHECK_CONDITION	= 0x00004000,
 	SCF_OVERFLOW_BIT		= 0x00008000,
 	SCF_UNDERFLOW_BIT		= 0x00010000,
@@ -154,6 +155,7 @@
 	TCM_CHECK_CONDITION_ABORT_CMD		= 0x0d,
 	TCM_CHECK_CONDITION_UNIT_ATTENTION	= 0x0e,
 	TCM_CHECK_CONDITION_NOT_READY		= 0x0f,
+	TCM_RESERVATION_CONFLICT		= 0x10,
 };
 
 struct se_obj {
@@ -211,7 +213,6 @@
 	u16	lu_gp_id;
 	int	lu_gp_valid_id;
 	u32	lu_gp_members;
-	atomic_t lu_gp_shutdown;
 	atomic_t lu_gp_ref_cnt;
 	spinlock_t lu_gp_lock;
 	struct config_group lu_gp_group;
@@ -422,11 +423,9 @@
 	int			sam_task_attr;
 	/* Transport protocol dependent state, see transport_state_table */
 	enum transport_state_table t_state;
-	/* Transport specific error status */
-	int			transport_error_status;
 	/* Used to signal cmd->se_tfo->check_release_cmd() usage per cmd */
-	int			check_release:1;
-	int			cmd_wait_set:1;
+	unsigned		check_release:1;
+	unsigned		cmd_wait_set:1;
 	/* See se_cmd_flags_table */
 	u32			se_cmd_flags;
 	u32			se_ordered_id;
@@ -441,13 +440,10 @@
 	/* Used for sense data */
 	void			*sense_buffer;
 	struct list_head	se_delayed_node;
-	struct list_head	se_ordered_node;
 	struct list_head	se_lun_node;
 	struct list_head	se_qf_node;
 	struct se_device      *se_dev;
 	struct se_dev_entry   *se_deve;
-	struct se_device	*se_obj_ptr;
-	struct se_device	*se_orig_obj_ptr;
 	struct se_lun		*se_lun;
 	/* Only used for internal passthrough and legacy TCM fabric modules */
 	struct se_session	*se_sess;
@@ -463,8 +459,6 @@
 	unsigned char		__t_task_cdb[TCM_MAX_COMMAND_SIZE];
 	unsigned long long	t_task_lba;
 	int			t_tasks_failed;
-	int			t_tasks_fua;
-	bool			t_tasks_bidi;
 	u32			t_tasks_sg_chained_no;
 	atomic_t		t_fe_count;
 	atomic_t		t_se_count;
@@ -489,14 +483,6 @@
 
 	struct work_struct	work;
 
-	/*
-	 * Used for pre-registered fabric SGL passthrough WRITE and READ
-	 * with the special SCF_PASSTHROUGH_CONTIG_TO_SG case for TCM_Loop
-	 * and other HW target mode fabric modules.
-	 */
-	struct scatterlist	*t_task_pt_sgl;
-	u32			t_task_pt_sgl_num;
-
 	struct scatterlist	*t_data_sg;
 	unsigned int		t_data_nents;
 	struct scatterlist	*t_bidi_data_sg;
@@ -562,7 +548,7 @@
 } ____cacheline_aligned;
 
 struct se_session {
-	int			sess_tearing_down:1;
+	unsigned		sess_tearing_down:1;
 	u64			sess_bin_isid;
 	struct se_node_acl	*se_node_acl;
 	struct se_portal_group *se_tpg;
@@ -683,7 +669,6 @@
 	struct t10_reservation t10_pr;
 	spinlock_t      se_dev_lock;
 	void            *se_dev_su_ptr;
-	struct list_head se_dev_node;
 	struct config_group se_dev_group;
 	/* For T10 Reservations */
 	struct config_group se_dev_pr_group;
@@ -692,9 +677,6 @@
 } ____cacheline_aligned;
 
 struct se_device {
-	/* Set to 1 if thread is NOT sleeping on thread_sem */
-	u8			thread_active;
-	u8			dev_status_timer_flags;
 	/* RELATIVE TARGET PORT IDENTIFER Counter */
 	u16			dev_rpti_counter;
 	/* Used for SAM Task Attribute ordering */
@@ -719,14 +701,10 @@
 	u64			write_bytes;
 	spinlock_t		stats_lock;
 	/* Active commands on this virtual SE device */
-	atomic_t		active_cmds;
 	atomic_t		simple_cmds;
 	atomic_t		depth_left;
 	atomic_t		dev_ordered_id;
-	atomic_t		dev_tur_active;
 	atomic_t		execute_tasks;
-	atomic_t		dev_status_thr_count;
-	atomic_t		dev_hoq_count;
 	atomic_t		dev_ordered_sync;
 	atomic_t		dev_qf_count;
 	struct se_obj		dev_obj;
@@ -734,14 +712,9 @@
 	struct se_obj		dev_export_obj;
 	struct se_queue_obj	dev_queue_obj;
 	spinlock_t		delayed_cmd_lock;
-	spinlock_t		ordered_cmd_lock;
 	spinlock_t		execute_task_lock;
-	spinlock_t		state_task_lock;
-	spinlock_t		dev_alua_lock;
 	spinlock_t		dev_reservation_lock;
-	spinlock_t		dev_state_lock;
 	spinlock_t		dev_status_lock;
-	spinlock_t		dev_status_thr_lock;
 	spinlock_t		se_port_lock;
 	spinlock_t		se_tmr_lock;
 	spinlock_t		qf_cmd_lock;
@@ -753,14 +726,10 @@
 	struct t10_pr_registration *dev_pr_res_holder;
 	struct list_head	dev_sep_list;
 	struct list_head	dev_tmr_list;
-	struct timer_list	dev_status_timer;
 	/* Pointer to descriptor for processing thread */
 	struct task_struct	*process_thread;
-	pid_t			process_thread_pid;
-	struct task_struct		*dev_mgmt_thread;
 	struct work_struct	qf_work_queue;
 	struct list_head	delayed_cmd_list;
-	struct list_head	ordered_cmd_list;
 	struct list_head	execute_task_list;
 	struct list_head	state_task_list;
 	struct list_head	qf_cmd_list;
@@ -771,8 +740,6 @@
 	struct se_subsystem_api *transport;
 	/* Linked list for struct se_hba struct se_device list */
 	struct list_head	dev_list;
-	/* Linked list for struct se_global->g_se_dev_list */
-	struct list_head	g_se_dev_list;
 }  ____cacheline_aligned;
 
 struct se_hba {
@@ -834,7 +801,6 @@
 	u32		sep_index;
 	struct scsi_port_stats sep_stats;
 	/* Used for ALUA Target Port Groups membership */
-	atomic_t	sep_tg_pt_gp_active;
 	atomic_t	sep_tg_pt_secondary_offline;
 	/* Used for PR ALL_TG_PT=1 */
 	atomic_t	sep_tg_pt_ref_cnt;
diff --git a/include/target/target_core_transport.h b/include/target/target_core_transport.h
index c16e943..dac4f2d 100644
--- a/include/target/target_core_transport.h
+++ b/include/target/target_core_transport.h
@@ -10,29 +10,6 @@
 
 #define PYX_TRANSPORT_STATUS_INTERVAL		5 /* In seconds */
 
-#define PYX_TRANSPORT_SENT_TO_TRANSPORT		0
-#define PYX_TRANSPORT_WRITE_PENDING		1
-
-#define PYX_TRANSPORT_UNKNOWN_SAM_OPCODE	-1
-#define PYX_TRANSPORT_HBA_QUEUE_FULL		-2
-#define PYX_TRANSPORT_REQ_TOO_MANY_SECTORS	-3
-#define PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES	-4
-#define PYX_TRANSPORT_INVALID_CDB_FIELD		-5
-#define PYX_TRANSPORT_INVALID_PARAMETER_LIST	-6
-#define PYX_TRANSPORT_LU_COMM_FAILURE		-7
-#define PYX_TRANSPORT_UNKNOWN_MODE_PAGE		-8
-#define PYX_TRANSPORT_WRITE_PROTECTED		-9
-#define PYX_TRANSPORT_RESERVATION_CONFLICT	-10
-#define PYX_TRANSPORT_ILLEGAL_REQUEST		-11
-#define PYX_TRANSPORT_USE_SENSE_REASON		-12
-
-#ifndef SAM_STAT_RESERVATION_CONFLICT
-#define SAM_STAT_RESERVATION_CONFLICT		0x18
-#endif
-
-#define TRANSPORT_PLUGIN_FREE			0
-#define TRANSPORT_PLUGIN_REGISTERED		1
-
 #define TRANSPORT_PLUGIN_PHBA_PDEV		1
 #define TRANSPORT_PLUGIN_VHBA_PDEV		2
 #define TRANSPORT_PLUGIN_VHBA_VDEV		3
@@ -158,7 +135,6 @@
 extern int transport_handle_cdb_direct(struct se_cmd *);
 extern int transport_generic_handle_cdb_map(struct se_cmd *);
 extern int transport_generic_handle_data(struct se_cmd *);
-extern void transport_new_cmd_failure(struct se_cmd *);
 extern int transport_generic_handle_tmr(struct se_cmd *);
 extern bool target_stop_task(struct se_task *task, unsigned long *flags);
 extern int transport_generic_map_mem_to_cmd(struct se_cmd *cmd, struct scatterlist *, u32,
diff --git a/include/video/omapdss.h b/include/video/omapdss.h
index b66ebb2..378c7ed 100644
--- a/include/video/omapdss.h
+++ b/include/video/omapdss.h
@@ -307,15 +307,8 @@
 	void (*dsi_disable_pads)(int dsi_id, unsigned lane_mask);
 };
 
-#if defined(CONFIG_OMAP2_DSS_MODULE) || defined(CONFIG_OMAP2_DSS)
 /* Init with the board info */
 extern int omap_display_init(struct omap_dss_board_info *board_data);
-#else
-static inline int omap_display_init(struct omap_dss_board_info *board_data)
-{
-	return 0;
-}
-#endif
 
 struct omap_display_platform_data {
 	struct omap_dss_board_info *board_data;
diff --git a/include/xen/platform_pci.h b/include/xen/platform_pci.h
index a785a3b..438c256 100644
--- a/include/xen/platform_pci.h
+++ b/include/xen/platform_pci.h
@@ -29,8 +29,7 @@
 static inline int xen_must_unplug_nics(void) {
 #if (defined(CONFIG_XEN_NETDEV_FRONTEND) || \
 		defined(CONFIG_XEN_NETDEV_FRONTEND_MODULE)) && \
-		(defined(CONFIG_XEN_PLATFORM_PCI) || \
-		 defined(CONFIG_XEN_PLATFORM_PCI_MODULE))
+		defined(CONFIG_XEN_PVHVM)
         return 1;
 #else
         return 0;
@@ -40,8 +39,7 @@
 static inline int xen_must_unplug_disks(void) {
 #if (defined(CONFIG_XEN_BLKDEV_FRONTEND) || \
 		defined(CONFIG_XEN_BLKDEV_FRONTEND_MODULE)) && \
-		(defined(CONFIG_XEN_PLATFORM_PCI) || \
-		 defined(CONFIG_XEN_PLATFORM_PCI_MODULE))
+		defined(CONFIG_XEN_PVHVM)
         return 1;
 #else
         return 0;
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 2e0ecfc..5b4293d 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -1269,7 +1269,7 @@
 
 void mq_put_mnt(struct ipc_namespace *ns)
 {
-	mntput(ns->mq_mnt);
+	kern_unmount(ns->mq_mnt);
 }
 
 static int __init init_mqueue_fs(void)
@@ -1291,11 +1291,9 @@
 
 	spin_lock_init(&mq_lock);
 
-	init_ipc_ns.mq_mnt = kern_mount_data(&mqueue_fs_type, &init_ipc_ns);
-	if (IS_ERR(init_ipc_ns.mq_mnt)) {
-		error = PTR_ERR(init_ipc_ns.mq_mnt);
+	error = mq_init_ns(&init_ipc_ns);
+	if (error)
 		goto out_filesystem;
-	}
 
 	return 0;
 
diff --git a/ipc/msgutil.c b/ipc/msgutil.c
index 8b5ce5d3..5652101 100644
--- a/ipc/msgutil.c
+++ b/ipc/msgutil.c
@@ -27,11 +27,6 @@
  */
 struct ipc_namespace init_ipc_ns = {
 	.count		= ATOMIC_INIT(1),
-#ifdef CONFIG_POSIX_MQUEUE
-	.mq_queues_max   = DFLT_QUEUESMAX,
-	.mq_msg_max      = DFLT_MSGMAX,
-	.mq_msgsize_max  = DFLT_MSGSIZEMAX,
-#endif
 	.user_ns = &init_user_ns,
 };
 
diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c
index 5e828a2..213c035 100644
--- a/kernel/cgroup_freezer.c
+++ b/kernel/cgroup_freezer.c
@@ -153,6 +153,13 @@
 	kfree(cgroup_freezer(cgroup));
 }
 
+/* task is frozen or will freeze immediately when next it gets woken */
+static bool is_task_frozen_enough(struct task_struct *task)
+{
+	return frozen(task) ||
+		(task_is_stopped_or_traced(task) && freezing(task));
+}
+
 /*
  * The call to cgroup_lock() in the freezer.state write method prevents
  * a write to that file racing against an attach, and hence the
@@ -231,7 +238,7 @@
 	cgroup_iter_start(cgroup, &it);
 	while ((task = cgroup_iter_next(cgroup, &it))) {
 		ntotal++;
-		if (frozen(task))
+		if (is_task_frozen_enough(task))
 			nfrozen++;
 	}
 
@@ -284,7 +291,7 @@
 	while ((task = cgroup_iter_next(cgroup, &it))) {
 		if (!freeze_task(task, true))
 			continue;
-		if (frozen(task))
+		if (is_task_frozen_enough(task))
 			continue;
 		if (!freezing(task) && !freezer_should_skip(task))
 			num_cant_freeze_now++;
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 0e8457d..58690af 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -185,6 +185,9 @@
 static void update_context_time(struct perf_event_context *ctx);
 static u64 perf_event_time(struct perf_event *event);
 
+static void ring_buffer_attach(struct perf_event *event,
+			       struct ring_buffer *rb);
+
 void __weak perf_event_print_debug(void)	{ }
 
 extern __weak const char *perf_pmu_name(void)
@@ -2171,9 +2174,10 @@
 	 */
 	cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 
-	perf_event_sched_in(cpuctx, ctx, task);
+	if (ctx->nr_events)
+		cpuctx->task_ctx = ctx;
 
-	cpuctx->task_ctx = ctx;
+	perf_event_sched_in(cpuctx, cpuctx->task_ctx, task);
 
 	perf_pmu_enable(ctx->pmu);
 	perf_ctx_unlock(cpuctx, ctx);
@@ -3190,12 +3194,33 @@
 	struct ring_buffer *rb;
 	unsigned int events = POLL_HUP;
 
+	/*
+	 * Race between perf_event_set_output() and perf_poll(): perf_poll()
+	 * grabs the rb reference but perf_event_set_output() overrides it.
+	 * Here is the timeline for two threads T1, T2:
+	 * t0: T1, rb = rcu_dereference(event->rb)
+	 * t1: T2, old_rb = event->rb
+	 * t2: T2, event->rb = new rb
+	 * t3: T2, ring_buffer_detach(old_rb)
+	 * t4: T1, ring_buffer_attach(rb1)
+	 * t5: T1, poll_wait(event->waitq)
+	 *
+	 * To avoid this problem, we grab mmap_mutex in perf_poll()
+	 * thereby ensuring that the assignment of the new ring buffer
+	 * and the detachment of the old buffer appear atomic to perf_poll()
+	 */
+	mutex_lock(&event->mmap_mutex);
+
 	rcu_read_lock();
 	rb = rcu_dereference(event->rb);
-	if (rb)
+	if (rb) {
+		ring_buffer_attach(event, rb);
 		events = atomic_xchg(&rb->poll, 0);
+	}
 	rcu_read_unlock();
 
+	mutex_unlock(&event->mmap_mutex);
+
 	poll_wait(file, &event->waitq, wait);
 
 	return events;
@@ -3496,6 +3521,53 @@
 	return ret;
 }
 
+static void ring_buffer_attach(struct perf_event *event,
+			       struct ring_buffer *rb)
+{
+	unsigned long flags;
+
+	if (!list_empty(&event->rb_entry))
+		return;
+
+	spin_lock_irqsave(&rb->event_lock, flags);
+	if (!list_empty(&event->rb_entry))
+		goto unlock;
+
+	list_add(&event->rb_entry, &rb->event_list);
+unlock:
+	spin_unlock_irqrestore(&rb->event_lock, flags);
+}
+
+static void ring_buffer_detach(struct perf_event *event,
+			       struct ring_buffer *rb)
+{
+	unsigned long flags;
+
+	if (list_empty(&event->rb_entry))
+		return;
+
+	spin_lock_irqsave(&rb->event_lock, flags);
+	list_del_init(&event->rb_entry);
+	wake_up_all(&event->waitq);
+	spin_unlock_irqrestore(&rb->event_lock, flags);
+}
+
+static void ring_buffer_wakeup(struct perf_event *event)
+{
+	struct ring_buffer *rb;
+
+	rcu_read_lock();
+	rb = rcu_dereference(event->rb);
+	if (!rb)
+		goto unlock;
+
+	list_for_each_entry_rcu(event, &rb->event_list, rb_entry)
+		wake_up_all(&event->waitq);
+
+unlock:
+	rcu_read_unlock();
+}
+
 static void rb_free_rcu(struct rcu_head *rcu_head)
 {
 	struct ring_buffer *rb;
@@ -3521,9 +3593,19 @@
 
 static void ring_buffer_put(struct ring_buffer *rb)
 {
+	struct perf_event *event, *n;
+	unsigned long flags;
+
 	if (!atomic_dec_and_test(&rb->refcount))
 		return;
 
+	spin_lock_irqsave(&rb->event_lock, flags);
+	list_for_each_entry_safe(event, n, &rb->event_list, rb_entry) {
+		list_del_init(&event->rb_entry);
+		wake_up_all(&event->waitq);
+	}
+	spin_unlock_irqrestore(&rb->event_lock, flags);
+
 	call_rcu(&rb->rcu_head, rb_free_rcu);
 }
 
@@ -3546,6 +3628,7 @@
 		atomic_long_sub((size >> PAGE_SHIFT) + 1, &user->locked_vm);
 		vma->vm_mm->pinned_vm -= event->mmap_locked;
 		rcu_assign_pointer(event->rb, NULL);
+		ring_buffer_detach(event, rb);
 		mutex_unlock(&event->mmap_mutex);
 
 		ring_buffer_put(rb);
@@ -3700,7 +3783,7 @@
 
 void perf_event_wakeup(struct perf_event *event)
 {
-	wake_up_all(&event->waitq);
+	ring_buffer_wakeup(event);
 
 	if (event->pending_kill) {
 		kill_fasync(&event->fasync, SIGIO, event->pending_kill);
@@ -5822,6 +5905,8 @@
 	INIT_LIST_HEAD(&event->group_entry);
 	INIT_LIST_HEAD(&event->event_entry);
 	INIT_LIST_HEAD(&event->sibling_list);
+	INIT_LIST_HEAD(&event->rb_entry);
+
 	init_waitqueue_head(&event->waitq);
 	init_irq_work(&event->pending, perf_pending_event);
 
@@ -6028,6 +6113,8 @@
 
 	old_rb = event->rb;
 	rcu_assign_pointer(event->rb, rb);
+	if (old_rb)
+		ring_buffer_detach(event, old_rb);
 	ret = 0;
 unlock:
 	mutex_unlock(&event->mmap_mutex);
diff --git a/kernel/events/internal.h b/kernel/events/internal.h
index 09097dd..64568a6 100644
--- a/kernel/events/internal.h
+++ b/kernel/events/internal.h
@@ -22,6 +22,9 @@
 	local_t				lost;		/* nr records lost   */
 
 	long				watermark;	/* wakeup watermark  */
+	/* poll crap */
+	spinlock_t			event_lock;
+	struct list_head		event_list;
 
 	struct perf_event_mmap_page	*user_page;
 	void				*data_pages[0];
diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
index a2a2920..7f3011c 100644
--- a/kernel/events/ring_buffer.c
+++ b/kernel/events/ring_buffer.c
@@ -209,6 +209,9 @@
 		rb->writable = 1;
 
 	atomic_set(&rb->refcount, 1);
+
+	INIT_LIST_HEAD(&rb->event_list);
+	spin_lock_init(&rb->event_lock);
 }
 
 #ifndef CONFIG_PERF_USE_VMALLOC
diff --git a/kernel/fork.c b/kernel/fork.c
index ba0d172..da4a6a1 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -162,7 +162,6 @@
 
 void free_task(struct task_struct *tsk)
 {
-	prop_local_destroy_single(&tsk->dirties);
 	account_kernel_stack(tsk->stack, -1);
 	free_thread_info(tsk->stack);
 	rt_mutex_debug_task_free(tsk);
@@ -274,10 +273,6 @@
 
 	tsk->stack = ti;
 
-	err = prop_local_init_single(&tsk->dirties);
-	if (err)
-		goto out;
-
 	setup_thread_stack(tsk, orig);
 	clear_user_return_notifier(tsk);
 	clear_tsk_need_resched(tsk);
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 422e567..ae34bf5 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -885,10 +885,13 @@
 			     struct hrtimer_clock_base *base,
 			     unsigned long newstate, int reprogram)
 {
+	struct timerqueue_node *next_timer;
 	if (!(timer->state & HRTIMER_STATE_ENQUEUED))
 		goto out;
 
-	if (&timer->node == timerqueue_getnext(&base->active)) {
+	next_timer = timerqueue_getnext(&base->active);
+	timerqueue_del(&base->active, &timer->node);
+	if (&timer->node == next_timer) {
 #ifdef CONFIG_HIGH_RES_TIMERS
 		/* Reprogram the clock event device. if enabled */
 		if (reprogram && hrtimer_hres_active()) {
@@ -901,7 +904,6 @@
 		}
 #endif
 	}
-	timerqueue_del(&base->active, &timer->node);
 	if (!timerqueue_getnext(&base->active))
 		base->cpu_base->active_bases &= ~(1 << base->index);
 out:
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 67ce837..1da999f 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -623,8 +623,9 @@
 
 static int irq_wait_for_interrupt(struct irqaction *action)
 {
+	set_current_state(TASK_INTERRUPTIBLE);
+
 	while (!kthread_should_stop()) {
-		set_current_state(TASK_INTERRUPTIBLE);
 
 		if (test_and_clear_bit(IRQTF_RUNTHREAD,
 				       &action->thread_flags)) {
@@ -632,7 +633,9 @@
 			return 0;
 		}
 		schedule();
+		set_current_state(TASK_INTERRUPTIBLE);
 	}
+	__set_current_state(TASK_RUNNING);
 	return -1;
 }
 
@@ -1596,7 +1599,7 @@
 		return -ENOMEM;
 
 	action->handler = handler;
-	action->flags = IRQF_PERCPU;
+	action->flags = IRQF_PERCPU | IRQF_NO_SUSPEND;
 	action->name = devname;
 	action->percpu_dev_id = dev_id;
 
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index aa57d5d..dc813a9 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -84,7 +84,9 @@
 	 */
 	action = desc->action;
 	if (!action || !(action->flags & IRQF_SHARED) ||
-	    (action->flags & __IRQF_TIMER) || !action->next)
+	    (action->flags & __IRQF_TIMER) ||
+	    (action->handler(irq, action->dev_id) == IRQ_HANDLED) ||
+	    !action->next)
 		goto out;
 
 	/* Already running on another processor */
@@ -115,7 +117,7 @@
 	struct irq_desc *desc;
 	int i, ok = 0;
 
-	if (atomic_inc_return(&irq_poll_active) == 1)
+	if (atomic_inc_return(&irq_poll_active) != 1)
 		goto out;
 
 	irq_poll_cpu = smp_processor_id();
diff --git a/kernel/jump_label.c b/kernel/jump_label.c
index bbdfe2a..66ff710 100644
--- a/kernel/jump_label.c
+++ b/kernel/jump_label.c
@@ -66,8 +66,9 @@
 		return;
 
 	jump_label_lock();
-	if (atomic_add_return(1, &key->enabled) == 1)
+	if (atomic_read(&key->enabled) == 0)
 		jump_label_update(key, JUMP_LABEL_ENABLE);
+	atomic_inc(&key->enabled);
 	jump_label_unlock();
 }
 
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index e69434b..b2e08c9 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -44,6 +44,7 @@
 #include <linux/stringify.h>
 #include <linux/bitops.h>
 #include <linux/gfp.h>
+#include <linux/kmemcheck.h>
 
 #include <asm/sections.h>
 
@@ -2948,7 +2949,12 @@
 void lockdep_init_map(struct lockdep_map *lock, const char *name,
 		      struct lock_class_key *key, int subclass)
 {
-	memset(lock, 0, sizeof(*lock));
+	int i;
+
+	kmemcheck_mark_initialized(lock, sizeof(*lock));
+
+	for (i = 0; i < NR_LOCKDEP_CACHING_CLASSES; i++)
+		lock->class_cache[i] = NULL;
 
 #ifdef CONFIG_LOCK_STAT
 	lock->cpu = raw_smp_processor_id();
diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index b4511b6..a6b0503 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -55,6 +55,8 @@
 
 static int hibernation_mode = HIBERNATION_SHUTDOWN;
 
+static bool freezer_test_done;
+
 static const struct platform_hibernation_ops *hibernation_ops;
 
 /**
@@ -345,11 +347,24 @@
 
 	error = freeze_kernel_threads();
 	if (error)
-		goto Close;
+		goto Cleanup;
+
+	if (hibernation_test(TEST_FREEZER) ||
+		hibernation_testmode(HIBERNATION_TESTPROC)) {
+
+		/*
+		 * Indicate to the caller that we are returning due to a
+		 * successful freezer test.
+		 */
+		freezer_test_done = true;
+		goto Cleanup;
+	}
 
 	error = dpm_prepare(PMSG_FREEZE);
-	if (error)
-		goto Complete_devices;
+	if (error) {
+		dpm_complete(msg);
+		goto Cleanup;
+	}
 
 	suspend_console();
 	pm_restrict_gfp_mask();
@@ -378,8 +393,6 @@
 		pm_restore_gfp_mask();
 
 	resume_console();
-
- Complete_devices:
 	dpm_complete(msg);
 
  Close:
@@ -389,6 +402,10 @@
  Recover_platform:
 	platform_recover(platform_mode);
 	goto Resume_devices;
+
+ Cleanup:
+	swsusp_free();
+	goto Close;
 }
 
 /**
@@ -641,15 +658,13 @@
 	if (error)
 		goto Finish;
 
-	if (hibernation_test(TEST_FREEZER))
-		goto Thaw;
-
-	if (hibernation_testmode(HIBERNATION_TESTPROC))
-		goto Thaw;
-
 	error = hibernation_snapshot(hibernation_mode == HIBERNATION_PLATFORM);
 	if (error)
 		goto Thaw;
+	if (freezer_test_done) {
+		freezer_test_done = false;
+		goto Thaw;
+	}
 
 	if (in_suspend) {
 		unsigned int flags = 0;
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 71f49fe..36e0f09 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -290,13 +290,14 @@
 		if (*s && len == strlen(*s) && !strncmp(buf, *s, len))
 			break;
 	}
-	if (state < PM_SUSPEND_MAX && *s)
+	if (state < PM_SUSPEND_MAX && *s) {
 		error = enter_state(state);
 		if (error) {
 			suspend_stats.fail++;
 			dpm_save_failed_errno(error);
 		} else
 			suspend_stats.success++;
+	}
 #endif
 
  Exit:
diff --git a/kernel/power/qos.c b/kernel/power/qos.c
index 56db751..995e3bd 100644
--- a/kernel/power/qos.c
+++ b/kernel/power/qos.c
@@ -70,6 +70,7 @@
 };
 static struct pm_qos_object cpu_dma_pm_qos = {
 	.constraints = &cpu_dma_constraints,
+	.name = "cpu_dma_latency",
 };
 
 static BLOCKING_NOTIFIER_HEAD(network_lat_notifier);
diff --git a/kernel/printk.c b/kernel/printk.c
index 1455a0d..7982a0a 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -1293,10 +1293,11 @@
 	raw_spin_lock(&logbuf_lock);
 	if (con_start != log_end)
 		retry = 1;
+	raw_spin_unlock_irqrestore(&logbuf_lock, flags);
+
 	if (retry && console_trylock())
 		goto again;
 
-	raw_spin_unlock_irqrestore(&logbuf_lock, flags);
 	if (wake_klogd)
 		wake_up_klogd();
 }
diff --git a/kernel/sched.c b/kernel/sched.c
index 0e9344a..d6b149c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -71,6 +71,7 @@
 #include <linux/ctype.h>
 #include <linux/ftrace.h>
 #include <linux/slab.h>
+#include <linux/init_task.h>
 
 #include <asm/tlb.h>
 #include <asm/irq_regs.h>
@@ -4810,6 +4811,9 @@
  * This waits for either a completion of a specific task to be signaled or for a
  * specified timeout to expire. The timeout is in jiffies. It is not
  * interruptible.
+ *
+ * The return value is 0 if timed out, and positive (at least 1, or number of
+ * jiffies left till timeout) if completed.
  */
 unsigned long __sched
 wait_for_completion_timeout(struct completion *x, unsigned long timeout)
@@ -4824,6 +4828,8 @@
  *
  * This waits for completion of a specific task to be signaled. It is
  * interruptible.
+ *
+ * The return value is -ERESTARTSYS if interrupted, 0 if completed.
  */
 int __sched wait_for_completion_interruptible(struct completion *x)
 {
@@ -4841,6 +4847,9 @@
  *
  * This waits for either a completion of a specific task to be signaled or for a
  * specified timeout to expire. It is interruptible. The timeout is in jiffies.
+ *
+ * The return value is -ERESTARTSYS if interrupted, 0 if timed out,
+ * positive (at least 1, or number of jiffies left till timeout) if completed.
  */
 long __sched
 wait_for_completion_interruptible_timeout(struct completion *x,
@@ -4856,6 +4865,8 @@
  *
  * This waits to be signaled for completion of a specific task. It can be
  * interrupted by a kill signal.
+ *
+ * The return value is -ERESTARTSYS if interrupted, 0 if completed.
  */
 int __sched wait_for_completion_killable(struct completion *x)
 {
@@ -4874,6 +4885,9 @@
  * This waits for either a completion of a specific task to be
  * signaled or for a specified timeout to expire. It can be
  * interrupted by a kill signal. The timeout is in jiffies.
+ *
+ * The return value is -ERESTARTSYS if interrupted, 0 if timed out,
+ * positive (at least 1, or number of jiffies left till timeout) if completed.
  */
 long __sched
 wait_for_completion_killable_timeout(struct completion *x,
@@ -6099,6 +6113,9 @@
 	 */
 	idle->sched_class = &idle_sched_class;
 	ftrace_graph_init_idle_task(idle, cpu);
+#if defined(CONFIG_SMP)
+	sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu);
+#endif
 }
 
 /*
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 5c9e679..8a39fa3 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -772,19 +772,32 @@
 		list_del_leaf_cfs_rq(cfs_rq);
 }
 
+static inline long calc_tg_weight(struct task_group *tg, struct cfs_rq *cfs_rq)
+{
+	long tg_weight;
+
+	/*
+	 * Use this CPU's actual weight instead of the last load_contribution
+	 * to gain a more accurate current total weight. See
+	 * update_cfs_rq_load_contribution().
+	 */
+	tg_weight = atomic_read(&tg->load_weight);
+	tg_weight -= cfs_rq->load_contribution;
+	tg_weight += cfs_rq->load.weight;
+
+	return tg_weight;
+}
+
 static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
 {
-	long load_weight, load, shares;
+	long tg_weight, load, shares;
 
+	tg_weight = calc_tg_weight(tg, cfs_rq);
 	load = cfs_rq->load.weight;
 
-	load_weight = atomic_read(&tg->load_weight);
-	load_weight += load;
-	load_weight -= cfs_rq->load_contribution;
-
 	shares = (tg->shares * load);
-	if (load_weight)
-		shares /= load_weight;
+	if (tg_weight)
+		shares /= tg_weight;
 
 	if (shares < MIN_SHARES)
 		shares = MIN_SHARES;
@@ -1743,7 +1756,7 @@
 
 static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq)
 {
-	if (!cfs_rq->runtime_enabled || !cfs_rq->nr_running)
+	if (!cfs_rq->runtime_enabled || cfs_rq->nr_running)
 		return;
 
 	__return_cfs_rq_runtime(cfs_rq);
@@ -2036,36 +2049,100 @@
  * Adding load to a group doesn't make a group heavier, but can cause movement
  * of group shares between cpus. Assuming the shares were perfectly aligned one
  * can calculate the shift in shares.
+ *
+ * Calculate the effective load difference if @wl is added (subtracted) to @tg
+ * on this @cpu and results in a total addition (subtraction) of @wg to the
+ * total group weight.
+ *
+ * Given a runqueue weight distribution (rw_i) we can compute a shares
+ * distribution (s_i) using:
+ *
+ *   s_i = rw_i / \Sum rw_j						(1)
+ *
+ * Suppose we have 4 CPUs and our @tg is a direct child of the root group and
+ * has 7 equal weight tasks, distributed as below (rw_i), with the resulting
+ * shares distribution (s_i):
+ *
+ *   rw_i = {   2,   4,   1,   0 }
+ *   s_i  = { 2/7, 4/7, 1/7,   0 }
+ *
+ * As per wake_affine() we're interested in the load of two CPUs (the CPU the
+ * task used to run on and the CPU the waker is running on), we need to
+ * compute the effect of waking a task on either CPU and, in case of a sync
+ * wakeup, compute the effect of the current task going to sleep.
+ *
+ * So for a change of @wl to the local @cpu with an overall group weight change
+ * of @wl we can compute the new shares distribution (s'_i) using:
+ *
+ *   s'_i = (rw_i + @wl) / (@wg + \Sum rw_j)				(2)
+ *
+ * Suppose we're interested in CPUs 0 and 1, and want to compute the load
+ * differences in waking a task to CPU 0. The additional task changes the
+ * weight and shares distributions like:
+ *
+ *   rw'_i = {   3,   4,   1,   0 }
+ *   s'_i  = { 3/8, 4/8, 1/8,   0 }
+ *
+ * We can then compute the difference in effective weight by using:
+ *
+ *   dw_i = S * (s'_i - s_i)						(3)
+ *
+ * Where 'S' is the group weight as seen by its parent.
+ *
+ * Therefore the effective change in loads on CPU 0 would be 5/56 (3/8 - 2/7)
+ * times the weight of the group. The effect on CPU 1 would be -4/56 (4/8 -
+ * 4/7) times the weight of the group.
  */
 static long effective_load(struct task_group *tg, int cpu, long wl, long wg)
 {
 	struct sched_entity *se = tg->se[cpu];
 
-	if (!tg->parent)
+	if (!tg->parent)	/* the trivial, non-cgroup case */
 		return wl;
 
 	for_each_sched_entity(se) {
-		long lw, w;
+		long w, W;
 
 		tg = se->my_q->tg;
-		w = se->my_q->load.weight;
 
-		/* use this cpu's instantaneous contribution */
-		lw = atomic_read(&tg->load_weight);
-		lw -= se->my_q->load_contribution;
-		lw += w + wg;
+		/*
+		 * W = @wg + \Sum rw_j
+		 */
+		W = wg + calc_tg_weight(tg, se->my_q);
 
-		wl += w;
+		/*
+		 * w = rw_i + @wl
+		 */
+		w = se->my_q->load.weight + wl;
 
-		if (lw > 0 && wl < lw)
-			wl = (wl * tg->shares) / lw;
+		/*
+		 * wl = S * s'_i; see (2)
+		 */
+		if (W > 0 && w < W)
+			wl = (w * tg->shares) / W;
 		else
 			wl = tg->shares;
 
-		/* zero point is MIN_SHARES */
+		/*
+		 * Per the above, wl is the new se->load.weight value; since
+		 * those are clipped to [MIN_SHARES, ...) do so now. See
+		 * calc_cfs_shares().
+		 */
 		if (wl < MIN_SHARES)
 			wl = MIN_SHARES;
+
+		/*
+		 * wl = dw_i = S * (s'_i - s_i); see (3)
+		 */
 		wl -= se->load.weight;
+
+		/*
+		 * Recursively apply this logic to all parent groups to compute
+		 * the final effective load change on the root group. Since
+		 * only the @tg group gets extra weight, all parent groups can
+		 * only redistribute existing shares. @wl is the shift in shares
+		 * resulting from this level per the above.
+		 */
 		wg = 0;
 	}
 
@@ -2249,7 +2326,8 @@
 	int cpu = smp_processor_id();
 	int prev_cpu = task_cpu(p);
 	struct sched_domain *sd;
-	int i;
+	struct sched_group *sg;
+	int i, smt = 0;
 
 	/*
 	 * If the task is going to be woken-up on this cpu and if it is
@@ -2269,25 +2347,40 @@
 	 * Otherwise, iterate the domains and find an elegible idle cpu.
 	 */
 	rcu_read_lock();
+again:
 	for_each_domain(target, sd) {
+		if (!smt && (sd->flags & SD_SHARE_CPUPOWER))
+			continue;
+
+		if (smt && !(sd->flags & SD_SHARE_CPUPOWER))
+			break;
+
 		if (!(sd->flags & SD_SHARE_PKG_RESOURCES))
 			break;
 
-		for_each_cpu_and(i, sched_domain_span(sd), tsk_cpus_allowed(p)) {
-			if (idle_cpu(i)) {
-				target = i;
-				break;
-			}
-		}
+		sg = sd->groups;
+		do {
+			if (!cpumask_intersects(sched_group_cpus(sg),
+						tsk_cpus_allowed(p)))
+				goto next;
 
-		/*
-		 * Lets stop looking for an idle sibling when we reached
-		 * the domain that spans the current cpu and prev_cpu.
-		 */
-		if (cpumask_test_cpu(cpu, sched_domain_span(sd)) &&
-		    cpumask_test_cpu(prev_cpu, sched_domain_span(sd)))
-			break;
+			for_each_cpu(i, sched_group_cpus(sg)) {
+				if (!idle_cpu(i))
+					goto next;
+			}
+
+			target = cpumask_first_and(sched_group_cpus(sg),
+					tsk_cpus_allowed(p));
+			goto done;
+next:
+			sg = sg->next;
+		} while (sg != sd->groups);
 	}
+	if (!smt) {
+		smt = 1;
+		goto again;
+	}
+done:
 	rcu_read_unlock();
 
 	return target;
@@ -3511,7 +3604,7 @@
 }
 
 /**
- * update_sd_lb_stats - Update sched_group's statistics for load balancing.
+ * update_sd_lb_stats - Update sched_domain's statistics for load balancing.
  * @sd: sched_domain whose statistics are to be updated.
  * @this_cpu: Cpu for which load balance is currently performed.
  * @idle: Idle status of this_cpu
diff --git a/kernel/sched_features.h b/kernel/sched_features.h
index efa0a7b..8480224 100644
--- a/kernel/sched_features.h
+++ b/kernel/sched_features.h
@@ -67,3 +67,4 @@
 SCHED_FEAT(TTWU_QUEUE, 1)
 
 SCHED_FEAT(FORCE_SD_OVERLAP, 0)
+SCHED_FEAT(RT_RUNTIME_SHARE, 1)
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 056cbd2..583a136 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -560,6 +560,9 @@
 {
 	int more = 0;
 
+	if (!sched_feat(RT_RUNTIME_SHARE))
+		return more;
+
 	if (rt_rq->rt_time > rt_rq->rt_runtime) {
 		raw_spin_unlock(&rt_rq->rt_runtime_lock);
 		more = do_balance_runtime(rt_rq);
diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c
index c436e79..8a46f5d 100644
--- a/kernel/time/alarmtimer.c
+++ b/kernel/time/alarmtimer.c
@@ -195,7 +195,7 @@
 		struct alarm *alarm;
 		ktime_t expired = next->expires;
 
-		if (expired.tv64 >= now.tv64)
+		if (expired.tv64 > now.tv64)
 			break;
 
 		alarm = container_of(next, struct alarm, node);
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 1ecd6ba..c4eb71c 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -387,6 +387,7 @@
 	 * released list and do a notify add later.
 	 */
 	if (old) {
+		old->event_handler = clockevents_handle_noop;
 		clockevents_set_mode(old, CLOCK_EVT_MODE_UNUSED);
 		list_del(&old->list);
 		list_add(&old->list, &clockevents_released);
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index cf52fda..da2f760 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -492,6 +492,22 @@
 }
 
 /**
+ * clocksource_max_adjustment- Returns max adjustment amount
+ * @cs:         Pointer to clocksource
+ *
+ */
+static u32 clocksource_max_adjustment(struct clocksource *cs)
+{
+	u64 ret;
+	/*
+	 * We won't try to correct for more then 11% adjustments (110,000 ppm),
+	 */
+	ret = (u64)cs->mult * 11;
+	do_div(ret,100);
+	return (u32)ret;
+}
+
+/**
  * clocksource_max_deferment - Returns max time the clocksource can be deferred
  * @cs:         Pointer to clocksource
  *
@@ -503,25 +519,28 @@
 	/*
 	 * Calculate the maximum number of cycles that we can pass to the
 	 * cyc2ns function without overflowing a 64-bit signed result. The
-	 * maximum number of cycles is equal to ULLONG_MAX/cs->mult which
-	 * is equivalent to the below.
-	 * max_cycles < (2^63)/cs->mult
-	 * max_cycles < 2^(log2((2^63)/cs->mult))
-	 * max_cycles < 2^(log2(2^63) - log2(cs->mult))
-	 * max_cycles < 2^(63 - log2(cs->mult))
-	 * max_cycles < 1 << (63 - log2(cs->mult))
+	 * maximum number of cycles is equal to ULLONG_MAX/(cs->mult+cs->maxadj)
+	 * which is equivalent to the below.
+	 * max_cycles < (2^63)/(cs->mult + cs->maxadj)
+	 * max_cycles < 2^(log2((2^63)/(cs->mult + cs->maxadj)))
+	 * max_cycles < 2^(log2(2^63) - log2(cs->mult + cs->maxadj))
+	 * max_cycles < 2^(63 - log2(cs->mult + cs->maxadj))
+	 * max_cycles < 1 << (63 - log2(cs->mult + cs->maxadj))
 	 * Please note that we add 1 to the result of the log2 to account for
 	 * any rounding errors, ensure the above inequality is satisfied and
 	 * no overflow will occur.
 	 */
-	max_cycles = 1ULL << (63 - (ilog2(cs->mult) + 1));
+	max_cycles = 1ULL << (63 - (ilog2(cs->mult + cs->maxadj) + 1));
 
 	/*
 	 * The actual maximum number of cycles we can defer the clocksource is
 	 * determined by the minimum of max_cycles and cs->mask.
+	 * Note: Here we subtract the maxadj to make sure we don't sleep for
+	 * too long if there's a large negative adjustment.
 	 */
 	max_cycles = min_t(u64, max_cycles, (u64) cs->mask);
-	max_nsecs = clocksource_cyc2ns(max_cycles, cs->mult, cs->shift);
+	max_nsecs = clocksource_cyc2ns(max_cycles, cs->mult - cs->maxadj,
+					cs->shift);
 
 	/*
 	 * To ensure that the clocksource does not wrap whilst we are idle,
@@ -529,7 +548,7 @@
 	 * note a margin of 12.5% is used because this can be computed with
 	 * a shift, versus say 10% which would require division.
 	 */
-	return max_nsecs - (max_nsecs >> 5);
+	return max_nsecs - (max_nsecs >> 3);
 }
 
 #ifndef CONFIG_ARCH_USES_GETTIMEOFFSET
@@ -640,7 +659,6 @@
 void __clocksource_updatefreq_scale(struct clocksource *cs, u32 scale, u32 freq)
 {
 	u64 sec;
-
 	/*
 	 * Calc the maximum number of seconds which we can run before
 	 * wrapping around. For clocksources which have a mask > 32bit
@@ -651,7 +669,7 @@
 	 * ~ 0.06ppm granularity for NTP. We apply the same 12.5%
 	 * margin as we do in clocksource_max_deferment()
 	 */
-	sec = (cs->mask - (cs->mask >> 5));
+	sec = (cs->mask - (cs->mask >> 3));
 	do_div(sec, freq);
 	do_div(sec, scale);
 	if (!sec)
@@ -661,6 +679,20 @@
 
 	clocks_calc_mult_shift(&cs->mult, &cs->shift, freq,
 			       NSEC_PER_SEC / scale, sec * scale);
+
+	/*
+	 * for clocksources that have large mults, to avoid overflow.
+	 * Since mult may be adjusted by ntp, add an safety extra margin
+	 *
+	 */
+	cs->maxadj = clocksource_max_adjustment(cs);
+	while ((cs->mult + cs->maxadj < cs->mult)
+		|| (cs->mult - cs->maxadj > cs->mult)) {
+		cs->mult >>= 1;
+		cs->shift--;
+		cs->maxadj = clocksource_max_adjustment(cs);
+	}
+
 	cs->max_idle_ns = clocksource_max_deferment(cs);
 }
 EXPORT_SYMBOL_GPL(__clocksource_updatefreq_scale);
@@ -701,6 +733,12 @@
  */
 int clocksource_register(struct clocksource *cs)
 {
+	/* calculate max adjustment for given mult/shift */
+	cs->maxadj = clocksource_max_adjustment(cs);
+	WARN_ONCE(cs->mult + cs->maxadj < cs->mult,
+		"Clocksource %s might overflow on 11%% adjustment\n",
+		cs->name);
+
 	/* calculate max idle time permitted for this clocksource */
 	cs->max_idle_ns = clocksource_max_deferment(cs);
 
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index f954282..fd4a7b1 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -71,7 +71,7 @@
 	     (dev->features & CLOCK_EVT_FEAT_C3STOP))
 		return 0;
 
-	clockevents_exchange_device(NULL, dev);
+	clockevents_exchange_device(tick_broadcast_device.evtdev, dev);
 	tick_broadcast_device.evtdev = dev;
 	if (!cpumask_empty(tick_get_broadcast_mask()))
 		tick_broadcast_start_periodic(dev);
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 2b021b0e..2378413 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -249,6 +249,8 @@
 		secs = xtime.tv_sec + wall_to_monotonic.tv_sec;
 		nsecs = xtime.tv_nsec + wall_to_monotonic.tv_nsec;
 		nsecs += timekeeping_get_ns();
+		/* If arch requires, add in gettimeoffset() */
+		nsecs += arch_gettimeoffset();
 
 	} while (read_seqretry(&xtime_lock, seq));
 	/*
@@ -280,6 +282,8 @@
 		*ts = xtime;
 		tomono = wall_to_monotonic;
 		nsecs = timekeeping_get_ns();
+		/* If arch requires, add in gettimeoffset() */
+		nsecs += arch_gettimeoffset();
 
 	} while (read_seqretry(&xtime_lock, seq));
 
@@ -802,14 +806,44 @@
 	s64 error, interval = timekeeper.cycle_interval;
 	int adj;
 
+	/*
+	 * The point of this is to check if the error is greater then half
+	 * an interval.
+	 *
+	 * First we shift it down from NTP_SHIFT to clocksource->shifted nsecs.
+	 *
+	 * Note we subtract one in the shift, so that error is really error*2.
+	 * This "saves" dividing(shifting) intererval twice, but keeps the
+	 * (error > interval) comparision as still measuring if error is
+	 * larger then half an interval.
+	 *
+	 * Note: It does not "save" on aggrivation when reading the code.
+	 */
 	error = timekeeper.ntp_error >> (timekeeper.ntp_error_shift - 1);
 	if (error > interval) {
+		/*
+		 * We now divide error by 4(via shift), which checks if
+		 * the error is greater then twice the interval.
+		 * If it is greater, we need a bigadjust, if its smaller,
+		 * we can adjust by 1.
+		 */
 		error >>= 2;
+		/*
+		 * XXX - In update_wall_time, we round up to the next
+		 * nanosecond, and store the amount rounded up into
+		 * the error. This causes the likely below to be unlikely.
+		 *
+		 * The properfix is to avoid rounding up by using
+		 * the high precision timekeeper.xtime_nsec instead of
+		 * xtime.tv_nsec everywhere. Fixing this will take some
+		 * time.
+		 */
 		if (likely(error <= interval))
 			adj = 1;
 		else
 			adj = timekeeping_bigadjust(error, &interval, &offset);
 	} else if (error < -interval) {
+		/* See comment above, this is just switched for the negative */
 		error >>= 2;
 		if (likely(error >= -interval)) {
 			adj = -1;
@@ -817,9 +851,65 @@
 			offset = -offset;
 		} else
 			adj = timekeeping_bigadjust(error, &interval, &offset);
-	} else
+	} else /* No adjustment needed */
 		return;
 
+	WARN_ONCE(timekeeper.clock->maxadj &&
+			(timekeeper.mult + adj > timekeeper.clock->mult +
+						timekeeper.clock->maxadj),
+			"Adjusting %s more then 11%% (%ld vs %ld)\n",
+			timekeeper.clock->name, (long)timekeeper.mult + adj,
+			(long)timekeeper.clock->mult +
+				timekeeper.clock->maxadj);
+	/*
+	 * So the following can be confusing.
+	 *
+	 * To keep things simple, lets assume adj == 1 for now.
+	 *
+	 * When adj != 1, remember that the interval and offset values
+	 * have been appropriately scaled so the math is the same.
+	 *
+	 * The basic idea here is that we're increasing the multiplier
+	 * by one, this causes the xtime_interval to be incremented by
+	 * one cycle_interval. This is because:
+	 *	xtime_interval = cycle_interval * mult
+	 * So if mult is being incremented by one:
+	 *	xtime_interval = cycle_interval * (mult + 1)
+	 * Its the same as:
+	 *	xtime_interval = (cycle_interval * mult) + cycle_interval
+	 * Which can be shortened to:
+	 *	xtime_interval += cycle_interval
+	 *
+	 * So offset stores the non-accumulated cycles. Thus the current
+	 * time (in shifted nanoseconds) is:
+	 *	now = (offset * adj) + xtime_nsec
+	 * Now, even though we're adjusting the clock frequency, we have
+	 * to keep time consistent. In other words, we can't jump back
+	 * in time, and we also want to avoid jumping forward in time.
+	 *
+	 * So given the same offset value, we need the time to be the same
+	 * both before and after the freq adjustment.
+	 *	now = (offset * adj_1) + xtime_nsec_1
+	 *	now = (offset * adj_2) + xtime_nsec_2
+	 * So:
+	 *	(offset * adj_1) + xtime_nsec_1 =
+	 *		(offset * adj_2) + xtime_nsec_2
+	 * And we know:
+	 *	adj_2 = adj_1 + 1
+	 * So:
+	 *	(offset * adj_1) + xtime_nsec_1 =
+	 *		(offset * (adj_1+1)) + xtime_nsec_2
+	 *	(offset * adj_1) + xtime_nsec_1 =
+	 *		(offset * adj_1) + offset + xtime_nsec_2
+	 * Canceling the sides:
+	 *	xtime_nsec_1 = offset + xtime_nsec_2
+	 * Which gives us:
+	 *	xtime_nsec_2 = xtime_nsec_1 - offset
+	 * Which simplfies to:
+	 *	xtime_nsec -= offset
+	 *
+	 * XXX - TODO: Doc ntp_error calculation.
+	 */
 	timekeeper.mult += adj;
 	timekeeper.xtime_interval += interval;
 	timekeeper.xtime_nsec -= offset;
diff --git a/kernel/timer.c b/kernel/timer.c
index dbaa624..9c3c62b 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -1368,7 +1368,7 @@
 	int pid;
 
 	rcu_read_lock();
-	pid = task_tgid_vnr(current->real_parent);
+	pid = task_tgid_vnr(rcu_dereference(current->real_parent));
 	rcu_read_unlock();
 
 	return pid;
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 900b409..b1e8943 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -152,7 +152,6 @@
 	ftrace_pid_function = ftrace_stub;
 }
 
-#undef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST
 #ifndef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST
 /*
  * For those archs that do not test ftrace_trace_stop in their
@@ -1212,7 +1211,9 @@
 	if (!src->count) {
 		free_ftrace_hash_rcu(*dst);
 		rcu_assign_pointer(*dst, EMPTY_HASH);
-		return 0;
+		/* still need to update the function records */
+		ret = 0;
+		goto out;
 	}
 
 	/*
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 581876f..c212a7f 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -1078,7 +1078,6 @@
 	/* First see if we did not already create this dir */
 	list_for_each_entry(system, &event_subsystems, list) {
 		if (strcmp(system->name, name) == 0) {
-			__get_system(system);
 			system->nr_events++;
 			return system->entry;
 		}
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index 816d3d0..95dc31e 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -1649,7 +1649,9 @@
 		 */
 		err = replace_preds(call, NULL, ps, filter_string, true);
 		if (err)
-			goto fail;
+			call->flags |= TRACE_EVENT_FL_NO_SET_FILTER;
+		else
+			call->flags &= ~TRACE_EVENT_FL_NO_SET_FILTER;
 	}
 
 	list_for_each_entry(call, &ftrace_events, list) {
@@ -1658,6 +1660,9 @@
 		if (strcmp(call->class->system, system->name) != 0)
 			continue;
 
+		if (call->flags & TRACE_EVENT_FL_NO_SET_FILTER)
+			continue;
+
 		filter_item = kzalloc(sizeof(*filter_item), GFP_KERNEL);
 		if (!filter_item)
 			goto fail_mem;
@@ -1686,7 +1691,7 @@
 		 * replace the filter for the call.
 		 */
 		filter = call->filter;
-		call->filter = filter_item->filter;
+		rcu_assign_pointer(call->filter, filter_item->filter);
 		filter_item->filter = filter;
 
 		fail = false;
@@ -1741,7 +1746,7 @@
 		filter = call->filter;
 		if (!filter)
 			goto out_unlock;
-		call->filter = NULL;
+		RCU_INIT_POINTER(call->filter, NULL);
 		/* Make sure the filter is not being used */
 		synchronize_sched();
 		__free_filter(filter);
@@ -1782,7 +1787,7 @@
 	 * string
 	 */
 	tmp = call->filter;
-	call->filter = filter;
+	rcu_assign_pointer(call->filter, filter);
 	if (tmp) {
 		/* Make sure the call is done with the filter */
 		synchronize_sched();
diff --git a/lib/dma-debug.c b/lib/dma-debug.c
index 74c6c7f..fea790a 100644
--- a/lib/dma-debug.c
+++ b/lib/dma-debug.c
@@ -245,7 +245,7 @@
 
 static bool exact_match(struct dma_debug_entry *a, struct dma_debug_entry *b)
 {
-	return ((a->dev_addr == a->dev_addr) &&
+	return ((a->dev_addr == b->dev_addr) &&
 		(a->dev == b->dev)) ? true : false;
 }
 
diff --git a/mm/backing-dev.c b/mm/backing-dev.c
index a086064..71034f4 100644
--- a/mm/backing-dev.c
+++ b/mm/backing-dev.c
@@ -724,6 +724,14 @@
 
 	bdi_unregister(bdi);
 
+	/*
+	 * If bdi_unregister() had already been called earlier, the
+	 * wakeup_timer could still be armed because bdi_prune_sb()
+	 * can race with the bdi_wakeup_thread_delayed() calls from
+	 * __mark_inode_dirty().
+	 */
+	del_timer_sync(&bdi->wb.wakeup_timer);
+
 	for (i = 0; i < NR_BDI_STAT_ITEMS; i++)
 		percpu_counter_destroy(&bdi->bdi_stat[i]);
 
diff --git a/mm/filemap.c b/mm/filemap.c
index c0018f2..c106d3b 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2407,7 +2407,6 @@
 						iov_iter_count(i));
 
 again:
-
 		/*
 		 * Bring in the user page that we will copy from _first_.
 		 * Otherwise there's a nasty deadlock on copying from the
@@ -2463,7 +2462,10 @@
 		written += copied;
 
 		balance_dirty_pages_ratelimited(mapping);
-
+		if (fatal_signal_pending(current)) {
+			status = -EINTR;
+			break;
+		}
 	} while (iov_iter_count(i));
 
 	return written ? written : status;
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 4298aba..36b3d98 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2259,12 +2259,8 @@
 
 static void khugepaged_alloc_sleep(void)
 {
-	DEFINE_WAIT(wait);
-	add_wait_queue(&khugepaged_wait, &wait);
-	schedule_timeout_interruptible(
-		msecs_to_jiffies(
-			khugepaged_alloc_sleep_millisecs));
-	remove_wait_queue(&khugepaged_wait, &wait);
+	wait_event_freezable_timeout(khugepaged_wait, false,
+			msecs_to_jiffies(khugepaged_alloc_sleep_millisecs));
 }
 
 #ifndef CONFIG_NUMA
@@ -2313,14 +2309,10 @@
 		if (unlikely(kthread_should_stop()))
 			break;
 		if (khugepaged_has_work()) {
-			DEFINE_WAIT(wait);
 			if (!khugepaged_scan_sleep_millisecs)
 				continue;
-			add_wait_queue(&khugepaged_wait, &wait);
-			schedule_timeout_interruptible(
-				msecs_to_jiffies(
-					khugepaged_scan_sleep_millisecs));
-			remove_wait_queue(&khugepaged_wait, &wait);
+			wait_event_freezable_timeout(khugepaged_wait, false,
+			    msecs_to_jiffies(khugepaged_scan_sleep_millisecs));
 		} else if (khugepaged_enabled())
 			wait_event_freezable(khugepaged_wait,
 					     khugepaged_wait_event());
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index dae27ba..73f17c0 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -576,6 +576,7 @@
 	__SetPageHead(page);
 	for (i = 1; i < nr_pages; i++, p = mem_map_next(p, page, i)) {
 		__SetPageTail(p);
+		set_page_count(p, 0);
 		p->first_page = page;
 	}
 }
@@ -2422,6 +2423,8 @@
 	 * anon_vma prepared.
 	 */
 	if (unlikely(anon_vma_prepare(vma))) {
+		page_cache_release(new_page);
+		page_cache_release(old_page);
 		/* Caller expects lock to be held */
 		spin_lock(&mm->page_table_lock);
 		return VM_FAULT_OOM;
diff --git a/mm/migrate.c b/mm/migrate.c
index 578e291..177aca4 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -871,9 +871,9 @@
 
 	if (anon_vma)
 		put_anon_vma(anon_vma);
-out:
 	unlock_page(hpage);
 
+out:
 	if (rc != -EAGAIN) {
 		list_del(&hpage->lru);
 		put_page(hpage);
diff --git a/mm/nommu.c b/mm/nommu.c
index 73419c5..b982290 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -454,7 +454,7 @@
  *	between processes, it syncs the pagetable across all
  *	processes.
  */
-struct vm_struct *alloc_vm_area(size_t size)
+struct vm_struct *alloc_vm_area(size_t size, pte_t **ptes)
 {
 	BUG();
 	return NULL;
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 471dedb..76f2c5a 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -185,6 +185,11 @@
 	if (!p)
 		return 0;
 
+	if (p->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) {
+		task_unlock(p);
+		return 0;
+	}
+
 	/*
 	 * The memory controller may have a limit of 0 bytes, so avoid a divide
 	 * by zero, if necessary.
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index a3278f0..50f0824 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -128,7 +128,6 @@
  *
  */
 static struct prop_descriptor vm_completions;
-static struct prop_descriptor vm_dirties;
 
 /*
  * couple the period to the dirty_ratio:
@@ -154,7 +153,6 @@
 {
 	int shift = calc_period_shift();
 	prop_change_shift(&vm_completions, shift);
-	prop_change_shift(&vm_dirties, shift);
 
 	writeback_set_ratelimit();
 }
@@ -235,11 +233,6 @@
 }
 EXPORT_SYMBOL_GPL(bdi_writeout_inc);
 
-void task_dirty_inc(struct task_struct *tsk)
-{
-	prop_inc_single(&vm_dirties, &tsk->dirties);
-}
-
 /*
  * Obtain an accurate fraction of the BDI's portion.
  */
@@ -418,8 +411,13 @@
  *
  * Returns @bdi's dirty limit in pages. The term "dirty" in the context of
  * dirty balancing includes all PG_dirty, PG_writeback and NFS unstable pages.
- * And the "limit" in the name is not seriously taken as hard limit in
- * balance_dirty_pages().
+ *
+ * Note that balance_dirty_pages() will only seriously take it as a hard limit
+ * when sleeping max_pause per page is not enough to keep the dirty pages under
+ * control. For example, when the device is completely stalled due to some error
+ * conditions, or when there are 1000 dd tasks writing to a slow 10MB/s USB key.
+ * In the other normal situations, it acts more gently by throttling the tasks
+ * more (rather than completely block them) when the bdi dirty pages go high.
  *
  * It allocates high/low dirty limits to fast/slow devices, in order to prevent
  * - starving fast devices
@@ -601,6 +599,13 @@
 	 */
 	if (unlikely(bdi_thresh > thresh))
 		bdi_thresh = thresh;
+	/*
+	 * It's very possible that bdi_thresh is close to 0 not because the
+	 * device is slow, but that it has remained inactive for long time.
+	 * Honour such devices a reasonable good (hopefully IO efficient)
+	 * threshold, so that the occasional writes won't be blocked and active
+	 * writes can rampup the threshold quickly.
+	 */
 	bdi_thresh = max(bdi_thresh, (limit - dirty) / 8);
 	/*
 	 * scale global setpoint to bdi's:
@@ -984,8 +989,7 @@
 	 *
 	 * 8 serves as the safety ratio.
 	 */
-	if (bdi_dirty)
-		t = min(t, bdi_dirty * HZ / (8 * bw + 1));
+	t = min(t, bdi_dirty * HZ / (8 * bw + 1));
 
 	/*
 	 * The pause time will be settled within range (max_pause/4, max_pause).
@@ -1133,17 +1137,30 @@
 					  pages_dirtied,
 					  pause,
 					  start_time);
-		__set_current_state(TASK_UNINTERRUPTIBLE);
+		__set_current_state(TASK_KILLABLE);
 		io_schedule_timeout(pause);
 
-		dirty_thresh = hard_dirty_limit(dirty_thresh);
 		/*
-		 * max-pause area. If dirty exceeded but still within this
-		 * area, no need to sleep for more than 200ms: (a) 8 pages per
-		 * 200ms is typically more than enough to curb heavy dirtiers;
-		 * (b) the pause time limit makes the dirtiers more responsive.
+		 * This is typically equal to (nr_dirty < dirty_thresh) and can
+		 * also keep "1000+ dd on a slow USB stick" under control.
 		 */
-		if (nr_dirty < dirty_thresh)
+		if (task_ratelimit)
+			break;
+
+		/*
+		 * In the case of an unresponding NFS server and the NFS dirty
+		 * pages exceeds dirty_thresh, give the other good bdi's a pipe
+		 * to go through, so that tasks on them still remain responsive.
+		 *
+		 * In theory 1 page is enough to keep the comsumer-producer
+		 * pipe going: the flusher cleans 1 page => the task dirties 1
+		 * more page. However bdi_dirty has accounting errors.  So use
+		 * the larger and more IO friendly bdi_stat_error.
+		 */
+		if (bdi_dirty <= bdi_stat_error(bdi))
+			break;
+
+		if (fatal_signal_pending(current))
 			break;
 	}
 
@@ -1395,7 +1412,6 @@
 
 	shift = calc_period_shift();
 	prop_descriptor_init(&vm_completions, shift);
-	prop_descriptor_init(&vm_dirties, shift);
 }
 
 /**
@@ -1724,7 +1740,6 @@
 		__inc_zone_page_state(page, NR_DIRTIED);
 		__inc_bdi_stat(mapping->backing_dev_info, BDI_RECLAIMABLE);
 		__inc_bdi_stat(mapping->backing_dev_info, BDI_DIRTIED);
-		task_dirty_inc(current);
 		task_io_account_write(PAGE_CACHE_SIZE);
 	}
 }
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9dd443d..2b8ba3a 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -356,8 +356,8 @@
 	__SetPageHead(page);
 	for (i = 1; i < nr_pages; i++) {
 		struct page *p = page + i;
-
 		__SetPageTail(p);
+		set_page_count(p, 0);
 		p->first_page = page;
 	}
 }
@@ -3377,9 +3377,15 @@
 	unsigned long block_migratetype;
 	int reserve;
 
-	/* Get the start pfn, end pfn and the number of blocks to reserve */
+	/*
+	 * Get the start pfn, end pfn and the number of blocks to reserve
+	 * We have to be careful to be aligned to pageblock_nr_pages to
+	 * make sure that we always check pfn_valid for the first page in
+	 * the block.
+	 */
 	start_pfn = zone->zone_start_pfn;
 	end_pfn = start_pfn + zone->spanned_pages;
+	start_pfn = roundup(start_pfn, pageblock_nr_pages);
 	reserve = roundup(min_wmark_pages(zone), pageblock_nr_pages) >>
 							pageblock_order;
 
diff --git a/mm/percpu-vm.c b/mm/percpu-vm.c
index ea53496..12a48a88 100644
--- a/mm/percpu-vm.c
+++ b/mm/percpu-vm.c
@@ -50,14 +50,13 @@
 
 	if (!pages || !bitmap) {
 		if (may_alloc && !pages)
-			pages = pcpu_mem_alloc(pages_size);
+			pages = pcpu_mem_zalloc(pages_size);
 		if (may_alloc && !bitmap)
-			bitmap = pcpu_mem_alloc(bitmap_size);
+			bitmap = pcpu_mem_zalloc(bitmap_size);
 		if (!pages || !bitmap)
 			return NULL;
 	}
 
-	memset(pages, 0, pages_size);
 	bitmap_copy(bitmap, chunk->populated, pcpu_unit_pages);
 
 	*bitmapp = bitmap;
@@ -143,8 +142,8 @@
 				 int page_start, int page_end)
 {
 	flush_cache_vunmap(
-		pcpu_chunk_addr(chunk, pcpu_first_unit_cpu, page_start),
-		pcpu_chunk_addr(chunk, pcpu_last_unit_cpu, page_end));
+		pcpu_chunk_addr(chunk, pcpu_low_unit_cpu, page_start),
+		pcpu_chunk_addr(chunk, pcpu_high_unit_cpu, page_end));
 }
 
 static void __pcpu_unmap_pages(unsigned long addr, int nr_pages)
@@ -206,8 +205,8 @@
 				      int page_start, int page_end)
 {
 	flush_tlb_kernel_range(
-		pcpu_chunk_addr(chunk, pcpu_first_unit_cpu, page_start),
-		pcpu_chunk_addr(chunk, pcpu_last_unit_cpu, page_end));
+		pcpu_chunk_addr(chunk, pcpu_low_unit_cpu, page_start),
+		pcpu_chunk_addr(chunk, pcpu_high_unit_cpu, page_end));
 }
 
 static int __pcpu_map_pages(unsigned long addr, struct page **pages,
@@ -284,8 +283,8 @@
 				int page_start, int page_end)
 {
 	flush_cache_vmap(
-		pcpu_chunk_addr(chunk, pcpu_first_unit_cpu, page_start),
-		pcpu_chunk_addr(chunk, pcpu_last_unit_cpu, page_end));
+		pcpu_chunk_addr(chunk, pcpu_low_unit_cpu, page_start),
+		pcpu_chunk_addr(chunk, pcpu_high_unit_cpu, page_end));
 }
 
 /**
diff --git a/mm/percpu.c b/mm/percpu.c
index bf80e55..3bb810a 100644
--- a/mm/percpu.c
+++ b/mm/percpu.c
@@ -116,9 +116,9 @@
 static int pcpu_nr_slots __read_mostly;
 static size_t pcpu_chunk_struct_size __read_mostly;
 
-/* cpus with the lowest and highest unit numbers */
-static unsigned int pcpu_first_unit_cpu __read_mostly;
-static unsigned int pcpu_last_unit_cpu __read_mostly;
+/* cpus with the lowest and highest unit addresses */
+static unsigned int pcpu_low_unit_cpu __read_mostly;
+static unsigned int pcpu_high_unit_cpu __read_mostly;
 
 /* the address of the first chunk which starts with the kernel static area */
 void *pcpu_base_addr __read_mostly;
@@ -273,11 +273,11 @@
 	     (rs) = (re) + 1, pcpu_next_pop((chunk), &(rs), &(re), (end)))
 
 /**
- * pcpu_mem_alloc - allocate memory
+ * pcpu_mem_zalloc - allocate memory
  * @size: bytes to allocate
  *
  * Allocate @size bytes.  If @size is smaller than PAGE_SIZE,
- * kzalloc() is used; otherwise, vmalloc() is used.  The returned
+ * kzalloc() is used; otherwise, vzalloc() is used.  The returned
  * memory is always zeroed.
  *
  * CONTEXT:
@@ -286,7 +286,7 @@
  * RETURNS:
  * Pointer to the allocated area on success, NULL on failure.
  */
-static void *pcpu_mem_alloc(size_t size)
+static void *pcpu_mem_zalloc(size_t size)
 {
 	if (WARN_ON_ONCE(!slab_is_available()))
 		return NULL;
@@ -302,7 +302,7 @@
  * @ptr: memory to free
  * @size: size of the area
  *
- * Free @ptr.  @ptr should have been allocated using pcpu_mem_alloc().
+ * Free @ptr.  @ptr should have been allocated using pcpu_mem_zalloc().
  */
 static void pcpu_mem_free(void *ptr, size_t size)
 {
@@ -384,7 +384,7 @@
 	size_t old_size = 0, new_size = new_alloc * sizeof(new[0]);
 	unsigned long flags;
 
-	new = pcpu_mem_alloc(new_size);
+	new = pcpu_mem_zalloc(new_size);
 	if (!new)
 		return -ENOMEM;
 
@@ -604,11 +604,12 @@
 {
 	struct pcpu_chunk *chunk;
 
-	chunk = pcpu_mem_alloc(pcpu_chunk_struct_size);
+	chunk = pcpu_mem_zalloc(pcpu_chunk_struct_size);
 	if (!chunk)
 		return NULL;
 
-	chunk->map = pcpu_mem_alloc(PCPU_DFL_MAP_ALLOC * sizeof(chunk->map[0]));
+	chunk->map = pcpu_mem_zalloc(PCPU_DFL_MAP_ALLOC *
+						sizeof(chunk->map[0]));
 	if (!chunk->map) {
 		kfree(chunk);
 		return NULL;
@@ -977,6 +978,17 @@
  * address.  The caller is responsible for ensuring @addr stays valid
  * until this function finishes.
  *
+ * percpu allocator has special setup for the first chunk, which currently
+ * supports either embedding in linear address space or vmalloc mapping,
+ * and, from the second one, the backing allocator (currently either vm or
+ * km) provides translation.
+ *
+ * The addr can be tranlated simply without checking if it falls into the
+ * first chunk. But the current code reflects better how percpu allocator
+ * actually works, and the verification can discover both bugs in percpu
+ * allocator itself and per_cpu_ptr_to_phys() callers. So we keep current
+ * code.
+ *
  * RETURNS:
  * The physical address for @addr.
  */
@@ -984,19 +996,19 @@
 {
 	void __percpu *base = __addr_to_pcpu_ptr(pcpu_base_addr);
 	bool in_first_chunk = false;
-	unsigned long first_start, first_end;
+	unsigned long first_low, first_high;
 	unsigned int cpu;
 
 	/*
-	 * The following test on first_start/end isn't strictly
+	 * The following test on unit_low/high isn't strictly
 	 * necessary but will speed up lookups of addresses which
 	 * aren't in the first chunk.
 	 */
-	first_start = pcpu_chunk_addr(pcpu_first_chunk, pcpu_first_unit_cpu, 0);
-	first_end = pcpu_chunk_addr(pcpu_first_chunk, pcpu_last_unit_cpu,
-				    pcpu_unit_pages);
-	if ((unsigned long)addr >= first_start &&
-	    (unsigned long)addr < first_end) {
+	first_low = pcpu_chunk_addr(pcpu_first_chunk, pcpu_low_unit_cpu, 0);
+	first_high = pcpu_chunk_addr(pcpu_first_chunk, pcpu_high_unit_cpu,
+				     pcpu_unit_pages);
+	if ((unsigned long)addr >= first_low &&
+	    (unsigned long)addr < first_high) {
 		for_each_possible_cpu(cpu) {
 			void *start = per_cpu_ptr(base, cpu);
 
@@ -1233,7 +1245,9 @@
 
 	for (cpu = 0; cpu < nr_cpu_ids; cpu++)
 		unit_map[cpu] = UINT_MAX;
-	pcpu_first_unit_cpu = NR_CPUS;
+
+	pcpu_low_unit_cpu = NR_CPUS;
+	pcpu_high_unit_cpu = NR_CPUS;
 
 	for (group = 0, unit = 0; group < ai->nr_groups; group++, unit += i) {
 		const struct pcpu_group_info *gi = &ai->groups[group];
@@ -1253,9 +1267,13 @@
 			unit_map[cpu] = unit + i;
 			unit_off[cpu] = gi->base_offset + i * ai->unit_size;
 
-			if (pcpu_first_unit_cpu == NR_CPUS)
-				pcpu_first_unit_cpu = cpu;
-			pcpu_last_unit_cpu = cpu;
+			/* determine low/high unit_cpu */
+			if (pcpu_low_unit_cpu == NR_CPUS ||
+			    unit_off[cpu] < unit_off[pcpu_low_unit_cpu])
+				pcpu_low_unit_cpu = cpu;
+			if (pcpu_high_unit_cpu == NR_CPUS ||
+			    unit_off[cpu] > unit_off[pcpu_high_unit_cpu])
+				pcpu_high_unit_cpu = cpu;
 		}
 	}
 	pcpu_nr_units = unit;
@@ -1889,7 +1907,7 @@
 
 		BUILD_BUG_ON(size > PAGE_SIZE);
 
-		map = pcpu_mem_alloc(size);
+		map = pcpu_mem_zalloc(size);
 		BUG_ON(!map);
 
 		spin_lock_irqsave(&pcpu_lock, flags);
diff --git a/mm/slab.c b/mm/slab.c
index 708efe8..83311c9a 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -595,6 +595,7 @@
 	PARTIAL_AC,
 	PARTIAL_L3,
 	EARLY,
+	LATE,
 	FULL
 } g_cpucache_up;
 
@@ -671,7 +672,7 @@
 {
 	struct cache_sizes *s = malloc_sizes;
 
-	if (g_cpucache_up != FULL)
+	if (g_cpucache_up < LATE)
 		return;
 
 	for (s = malloc_sizes; s->cs_size != ULONG_MAX; s++) {
@@ -1666,6 +1667,8 @@
 {
 	struct kmem_cache *cachep;
 
+	g_cpucache_up = LATE;
+
 	/* Annotate slab for lockdep -- annotate the malloc caches */
 	init_lock_keys();
 
diff --git a/mm/slub.c b/mm/slub.c
index 7d2a996..ed3334d 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1862,7 +1862,7 @@
 {
 	struct kmem_cache_node *n = NULL;
 	struct kmem_cache_cpu *c = this_cpu_ptr(s->cpu_slab);
-	struct page *page;
+	struct page *page, *discard_page = NULL;
 
 	while ((page = c->partial)) {
 		enum slab_modes { M_PARTIAL, M_FREE };
@@ -1904,7 +1904,8 @@
 				if (l == M_PARTIAL)
 					remove_partial(n, page);
 				else
-					add_partial(n, page, 1);
+					add_partial(n, page,
+						DEACTIVATE_TO_TAIL);
 
 				l = m;
 			}
@@ -1915,14 +1916,22 @@
 				"unfreezing slab"));
 
 		if (m == M_FREE) {
-			stat(s, DEACTIVATE_EMPTY);
-			discard_slab(s, page);
-			stat(s, FREE_SLAB);
+			page->next = discard_page;
+			discard_page = page;
 		}
 	}
 
 	if (n)
 		spin_unlock(&n->list_lock);
+
+	while (discard_page) {
+		page = discard_page;
+		discard_page = discard_page->next;
+
+		stat(s, DEACTIVATE_EMPTY);
+		discard_slab(s, page);
+		stat(s, FREE_SLAB);
+	}
 }
 
 /*
@@ -1969,7 +1978,7 @@
 		page->pobjects = pobjects;
 		page->next = oldpage;
 
-	} while (this_cpu_cmpxchg(s->cpu_slab->partial, oldpage, page) != oldpage);
+	} while (irqsafe_cpu_cmpxchg(s->cpu_slab->partial, oldpage, page) != oldpage);
 	stat(s, CPU_PARTIAL_FREE);
 	return pobjects;
 }
@@ -4435,30 +4444,31 @@
 
 		for_each_possible_cpu(cpu) {
 			struct kmem_cache_cpu *c = per_cpu_ptr(s->cpu_slab, cpu);
+			int node = ACCESS_ONCE(c->node);
 			struct page *page;
 
-			if (!c || c->node < 0)
+			if (node < 0)
 				continue;
-
-			if (c->page) {
-					if (flags & SO_TOTAL)
-						x = c->page->objects;
+			page = ACCESS_ONCE(c->page);
+			if (page) {
+				if (flags & SO_TOTAL)
+					x = page->objects;
 				else if (flags & SO_OBJECTS)
-					x = c->page->inuse;
+					x = page->inuse;
 				else
 					x = 1;
 
 				total += x;
-				nodes[c->node] += x;
+				nodes[node] += x;
 			}
 			page = c->partial;
 
 			if (page) {
 				x = page->pobjects;
-                                total += x;
-                                nodes[c->node] += x;
+				total += x;
+				nodes[node] += x;
 			}
-			per_cpu[c->node]++;
+			per_cpu[node]++;
 		}
 	}
 
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index b669aa6..1d8b32f 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -1633,6 +1633,8 @@
 		goto fail;
 
 	addr = __vmalloc_area_node(area, gfp_mask, prot, node, caller);
+	if (!addr)
+		return NULL;
 
 	/*
 	 * In this function, newly allocated vm_struct is not added
@@ -2141,23 +2143,30 @@
 
 static int f(pte_t *pte, pgtable_t table, unsigned long addr, void *data)
 {
-	/* apply_to_page_range() does all the hard work. */
+	pte_t ***p = data;
+
+	if (p) {
+		*(*p) = pte;
+		(*p)++;
+	}
 	return 0;
 }
 
 /**
  *	alloc_vm_area - allocate a range of kernel address space
  *	@size:		size of the area
+ *	@ptes:		returns the PTEs for the address space
  *
  *	Returns:	NULL on failure, vm_struct on success
  *
  *	This function reserves a range of kernel address space, and
  *	allocates pagetables to map that range.  No actual mappings
- *	are created.  If the kernel address space is not shared
- *	between processes, it syncs the pagetable across all
- *	processes.
+ *	are created.
+ *
+ *	If @ptes is non-NULL, pointers to the PTEs (in init_mm)
+ *	allocated for the VM area are returned.
  */
-struct vm_struct *alloc_vm_area(size_t size)
+struct vm_struct *alloc_vm_area(size_t size, pte_t **ptes)
 {
 	struct vm_struct *area;
 
@@ -2171,19 +2180,11 @@
 	 * of kernel virtual address space and mapped into init_mm.
 	 */
 	if (apply_to_page_range(&init_mm, (unsigned long)area->addr,
-				area->size, f, NULL)) {
+				size, f, ptes ? &ptes : NULL)) {
 		free_vm_area(area);
 		return NULL;
 	}
 
-	/*
-	 * If the allocated address space is passed to a hypercall
-	 * before being used then we cannot rely on a page fault to
-	 * trigger an update of the page tables.  So sync all the page
-	 * tables here.
-	 */
-	vmalloc_sync_all();
-
 	return area;
 }
 EXPORT_SYMBOL_GPL(alloc_vm_area);
diff --git a/mm/vmscan.c b/mm/vmscan.c
index a1893c0..f54a05b 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -183,7 +183,7 @@
  */
 void register_shrinker(struct shrinker *shrinker)
 {
-	shrinker->nr = 0;
+	atomic_long_set(&shrinker->nr_in_batch, 0);
 	down_write(&shrinker_rwsem);
 	list_add_tail(&shrinker->list, &shrinker_list);
 	up_write(&shrinker_rwsem);
@@ -247,25 +247,26 @@
 
 	list_for_each_entry(shrinker, &shrinker_list, list) {
 		unsigned long long delta;
-		unsigned long total_scan;
-		unsigned long max_pass;
+		long total_scan;
+		long max_pass;
 		int shrink_ret = 0;
 		long nr;
 		long new_nr;
 		long batch_size = shrinker->batch ? shrinker->batch
 						  : SHRINK_BATCH;
 
+		max_pass = do_shrinker_shrink(shrinker, shrink, 0);
+		if (max_pass <= 0)
+			continue;
+
 		/*
 		 * copy the current shrinker scan count into a local variable
 		 * and zero it so that other concurrent shrinker invocations
 		 * don't also do this scanning work.
 		 */
-		do {
-			nr = shrinker->nr;
-		} while (cmpxchg(&shrinker->nr, nr, 0) != nr);
+		nr = atomic_long_xchg(&shrinker->nr_in_batch, 0);
 
 		total_scan = nr;
-		max_pass = do_shrinker_shrink(shrinker, shrink, 0);
 		delta = (4 * nr_pages_scanned) / shrinker->seeks;
 		delta *= max_pass;
 		do_div(delta, lru_pages + 1);
@@ -325,12 +326,11 @@
 		 * manner that handles concurrent updates. If we exhausted the
 		 * scan, there is no need to do an update.
 		 */
-		do {
-			nr = shrinker->nr;
-			new_nr = total_scan + nr;
-			if (total_scan <= 0)
-				break;
-		} while (cmpxchg(&shrinker->nr, nr, new_nr) != nr);
+		if (total_scan > 0)
+			new_nr = atomic_long_add_return(total_scan,
+					&shrinker->nr_in_batch);
+		else
+			new_nr = atomic_long_read(&shrinker->nr_in_batch);
 
 		trace_mm_shrink_slab_end(shrinker, shrink_ret, nr, new_nr);
 	}
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index c7aafc7..5f09a57 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -245,9 +245,11 @@
 	if (tt_global_entry) {
 		/* This node is probably going to update its tt table */
 		tt_global_entry->orig_node->tt_poss_change = true;
-		/* The global entry has to be marked as PENDING and has to be
+		/* The global entry has to be marked as ROAMING and has to be
 		 * kept for consistency purpose */
-		tt_global_entry->flags |= TT_CLIENT_PENDING;
+		tt_global_entry->flags |= TT_CLIENT_ROAM;
+		tt_global_entry->roam_at = jiffies;
+
 		send_roam_adv(bat_priv, tt_global_entry->addr,
 			      tt_global_entry->orig_node);
 	}
@@ -694,6 +696,7 @@
 		   const char *message, bool roaming)
 {
 	struct tt_global_entry *tt_global_entry = NULL;
+	struct tt_local_entry *tt_local_entry = NULL;
 
 	tt_global_entry = tt_global_hash_find(bat_priv, addr);
 	if (!tt_global_entry)
@@ -701,15 +704,29 @@
 
 	if (tt_global_entry->orig_node == orig_node) {
 		if (roaming) {
-			tt_global_entry->flags |= TT_CLIENT_ROAM;
-			tt_global_entry->roam_at = jiffies;
-			goto out;
+			/* if we are deleting a global entry due to a roam
+			 * event, there are two possibilities:
+			 * 1) the client roamed from node A to node B => we mark
+			 *    it with TT_CLIENT_ROAM, we start a timer and we
+			 *    wait for node B to claim it. In case of timeout
+			 *    the entry is purged.
+			 * 2) the client roamed to us => we can directly delete
+			 *    the global entry, since it is useless now. */
+			tt_local_entry = tt_local_hash_find(bat_priv,
+							tt_global_entry->addr);
+			if (!tt_local_entry) {
+				tt_global_entry->flags |= TT_CLIENT_ROAM;
+				tt_global_entry->roam_at = jiffies;
+				goto out;
+			}
 		}
 		_tt_global_del(bat_priv, tt_global_entry, message);
 	}
 out:
 	if (tt_global_entry)
 		tt_global_entry_free_ref(tt_global_entry);
+	if (tt_local_entry)
+		tt_local_entry_free_ref(tt_local_entry);
 }
 
 void tt_global_del_orig(struct bat_priv *bat_priv,
diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c
index 91bcd3a..1eea820 100644
--- a/net/bluetooth/bnep/core.c
+++ b/net/bluetooth/bnep/core.c
@@ -79,17 +79,12 @@
 
 static void __bnep_link_session(struct bnep_session *s)
 {
-	/* It's safe to call __module_get() here because sessions are added
-	   by the socket layer which has to hold the reference to this module.
-	 */
-	__module_get(THIS_MODULE);
 	list_add(&s->list, &bnep_session_list);
 }
 
 static void __bnep_unlink_session(struct bnep_session *s)
 {
 	list_del(&s->list);
-	module_put(THIS_MODULE);
 }
 
 static int bnep_send(struct bnep_session *s, void *data, size_t len)
@@ -530,6 +525,7 @@
 
 	up_write(&bnep_session_sem);
 	free_netdev(dev);
+	module_put_and_exit(0);
 	return 0;
 }
 
@@ -616,9 +612,11 @@
 
 	__bnep_link_session(s);
 
+	__module_get(THIS_MODULE);
 	s->task = kthread_run(bnep_session, s, "kbnepd %s", dev->name);
 	if (IS_ERR(s->task)) {
 		/* Session thread start failed, gotta cleanup. */
+		module_put(THIS_MODULE);
 		unregister_netdev(dev);
 		__bnep_unlink_session(s);
 		err = PTR_ERR(s->task);
diff --git a/net/bluetooth/cmtp/core.c b/net/bluetooth/cmtp/core.c
index 7d00ddf..5a6e634 100644
--- a/net/bluetooth/cmtp/core.c
+++ b/net/bluetooth/cmtp/core.c
@@ -67,14 +67,12 @@
 
 static void __cmtp_link_session(struct cmtp_session *session)
 {
-	__module_get(THIS_MODULE);
 	list_add(&session->list, &cmtp_session_list);
 }
 
 static void __cmtp_unlink_session(struct cmtp_session *session)
 {
 	list_del(&session->list);
-	module_put(THIS_MODULE);
 }
 
 static void __cmtp_copy_session(struct cmtp_session *session, struct cmtp_conninfo *ci)
@@ -327,6 +325,7 @@
 	up_write(&cmtp_session_sem);
 
 	kfree(session);
+	module_put_and_exit(0);
 	return 0;
 }
 
@@ -376,9 +375,11 @@
 
 	__cmtp_link_session(session);
 
+	__module_get(THIS_MODULE);
 	session->task = kthread_run(cmtp_session, session, "kcmtpd_ctr_%d",
 								session->num);
 	if (IS_ERR(session->task)) {
+		module_put(THIS_MODULE);
 		err = PTR_ERR(session->task);
 		goto unlink;
 	}
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index c1c597e..e0af723 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -673,7 +673,7 @@
 		goto encrypt;
 
 auth:
-	if (test_bit(HCI_CONN_ENCRYPT_PEND, &conn->pend))
+	if (test_and_set_bit(HCI_CONN_ENCRYPT_PEND, &conn->pend))
 		return 0;
 
 	if (!hci_conn_auth(conn, sec_level, auth_type))
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index d7d96b6..643a41b 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -545,7 +545,7 @@
 {
 	hci_setup_event_mask(hdev);
 
-	if (hdev->lmp_ver > 1)
+	if (hdev->hci_ver > 1)
 		hci_send_cmd(hdev, HCI_OP_READ_LOCAL_COMMANDS, 0, NULL);
 
 	if (hdev->features[6] & LMP_SIMPLE_PAIR) {
diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index 8cd1291..5ea94a1 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -251,7 +251,7 @@
 
 	if (sock_owned_by_user(sk)) {
 		/* sk is owned by user. Try again later */
-		__set_chan_timer(chan, HZ / 5);
+		__set_chan_timer(chan, L2CAP_DISC_TIMEOUT);
 		bh_unlock_sock(sk);
 		chan_put(chan);
 		return;
@@ -2488,7 +2488,7 @@
 		if (sock_owned_by_user(sk)) {
 			l2cap_state_change(chan, BT_DISCONN);
 			__clear_chan_timer(chan);
-			__set_chan_timer(chan, HZ / 5);
+			__set_chan_timer(chan, L2CAP_DISC_TIMEOUT);
 			break;
 		}
 
@@ -2661,7 +2661,7 @@
 
 	default:
 		sk->sk_err = ECONNRESET;
-		__set_chan_timer(chan, HZ * 5);
+		__set_chan_timer(chan, L2CAP_DISC_REJ_TIMEOUT);
 		l2cap_send_disconn_req(conn, chan, ECONNRESET);
 		goto done;
 	}
@@ -2718,7 +2718,7 @@
 	if (sock_owned_by_user(sk)) {
 		l2cap_state_change(chan, BT_DISCONN);
 		__clear_chan_timer(chan);
-		__set_chan_timer(chan, HZ / 5);
+		__set_chan_timer(chan, L2CAP_DISC_TIMEOUT);
 		bh_unlock_sock(sk);
 		return 0;
 	}
@@ -2752,7 +2752,7 @@
 	if (sock_owned_by_user(sk)) {
 		l2cap_state_change(chan,BT_DISCONN);
 		__clear_chan_timer(chan);
-		__set_chan_timer(chan, HZ / 5);
+		__set_chan_timer(chan, L2CAP_DISC_TIMEOUT);
 		bh_unlock_sock(sk);
 		return 0;
 	}
@@ -3998,7 +3998,7 @@
 	if (encrypt == 0x00) {
 		if (chan->sec_level == BT_SECURITY_MEDIUM) {
 			__clear_chan_timer(chan);
-			__set_chan_timer(chan, HZ * 5);
+			__set_chan_timer(chan, L2CAP_ENC_TIMEOUT);
 		} else if (chan->sec_level == BT_SECURITY_HIGH)
 			l2cap_chan_close(chan, ECONNREFUSED);
 	} else {
@@ -4066,7 +4066,7 @@
 					L2CAP_CONN_REQ, sizeof(req), &req);
 			} else {
 				__clear_chan_timer(chan);
-				__set_chan_timer(chan, HZ / 10);
+				__set_chan_timer(chan, L2CAP_DISC_TIMEOUT);
 			}
 		} else if (chan->state == BT_CONNECT2) {
 			struct l2cap_conn_rsp rsp;
@@ -4086,7 +4086,7 @@
 				}
 			} else {
 				l2cap_state_change(chan, BT_DISCONN);
-				__set_chan_timer(chan, HZ / 10);
+				__set_chan_timer(chan, L2CAP_DISC_TIMEOUT);
 				res = L2CAP_CR_SEC_BLOCK;
 				stat = L2CAP_CS_NO_INFO;
 			}
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 995cbe0..a5f4e57 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -1501,6 +1501,8 @@
 
 	__skb_pull(skb2, offset);
 	skb_reset_transport_header(skb2);
+	skb_postpull_rcsum(skb2, skb_network_header(skb2),
+			   skb_network_header_len(skb2));
 
 	icmp6_type = icmp6_hdr(skb2)->icmp6_type;
 
@@ -1770,7 +1772,7 @@
 	int err = 0;
 	struct net_bridge_mdb_htable *mdb;
 
-	spin_lock(&br->multicast_lock);
+	spin_lock_bh(&br->multicast_lock);
 	if (br->multicast_disabled == !val)
 		goto unlock;
 
@@ -1806,7 +1808,7 @@
 	}
 
 unlock:
-	spin_unlock(&br->multicast_lock);
+	spin_unlock_bh(&br->multicast_lock);
 
 	return err;
 }
diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c
index e5f9ece3..a1daf82 100644
--- a/net/bridge/br_netlink.c
+++ b/net/bridge/br_netlink.c
@@ -18,6 +18,7 @@
 #include <net/sock.h>
 
 #include "br_private.h"
+#include "br_private_stp.h"
 
 static inline size_t br_nlmsg_size(void)
 {
@@ -188,6 +189,11 @@
 
 	p->state = new_state;
 	br_log_state(p);
+
+	spin_lock_bh(&p->br->lock);
+	br_port_state_selection(p->br);
+	spin_unlock_bh(&p->br->lock);
+
 	br_ifinfo_notify(RTM_NEWLINK, p);
 
 	return 0;
diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c
index ad0a3f7..dd147d7 100644
--- a/net/bridge/br_stp.c
+++ b/net/bridge/br_stp.c
@@ -399,25 +399,24 @@
 	struct net_bridge_port *p;
 	unsigned int liveports = 0;
 
-	/* Don't change port states if userspace is handling STP */
-	if (br->stp_enabled == BR_USER_STP)
-		return;
-
 	list_for_each_entry(p, &br->port_list, list) {
 		if (p->state == BR_STATE_DISABLED)
 			continue;
 
-		if (p->port_no == br->root_port) {
-			p->config_pending = 0;
-			p->topology_change_ack = 0;
-			br_make_forwarding(p);
-		} else if (br_is_designated_port(p)) {
-			del_timer(&p->message_age_timer);
-			br_make_forwarding(p);
-		} else {
-			p->config_pending = 0;
-			p->topology_change_ack = 0;
-			br_make_blocking(p);
+		/* Don't change port states if userspace is handling STP */
+		if (br->stp_enabled != BR_USER_STP) {
+			if (p->port_no == br->root_port) {
+				p->config_pending = 0;
+				p->topology_change_ack = 0;
+				br_make_forwarding(p);
+			} else if (br_is_designated_port(p)) {
+				del_timer(&p->message_age_timer);
+				br_make_forwarding(p);
+			} else {
+				p->config_pending = 0;
+				p->topology_change_ack = 0;
+				br_make_blocking(p);
+			}
 		}
 
 		if (p->state == BR_STATE_FORWARDING)
diff --git a/net/caif/cffrml.c b/net/caif/cffrml.c
index f399211..d3ca87b 100644
--- a/net/caif/cffrml.c
+++ b/net/caif/cffrml.c
@@ -136,20 +136,21 @@
 
 static int cffrml_transmit(struct cflayer *layr, struct cfpkt *pkt)
 {
-	int tmp;
 	u16 chks;
 	u16 len;
+	__le16 data;
+
 	struct cffrml *this = container_obj(layr);
 	if (this->dofcs) {
 		chks = cfpkt_iterate(pkt, cffrml_checksum, 0xffff);
-		tmp = cpu_to_le16(chks);
-		cfpkt_add_trail(pkt, &tmp, 2);
+		data = cpu_to_le16(chks);
+		cfpkt_add_trail(pkt, &data, 2);
 	} else {
 		cfpkt_pad_trail(pkt, 2);
 	}
 	len = cfpkt_getlen(pkt);
-	tmp = cpu_to_le16(len);
-	cfpkt_add_head(pkt, &tmp, 2);
+	data = cpu_to_le16(len);
+	cfpkt_add_head(pkt, &data, 2);
 	cfpkt_info(pkt)->hdr_len += 2;
 	if (cfpkt_erroneous(pkt)) {
 		pr_err("Packet is erroneous!\n");
diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c
index 42599e3..3a94eae 100644
--- a/net/ceph/crush/mapper.c
+++ b/net/ceph/crush/mapper.c
@@ -477,7 +477,6 @@
 	int i, j;
 	int numrep;
 	int firstn;
-	int rc = -1;
 
 	BUG_ON(ruleno >= map->max_rules);
 
@@ -491,23 +490,18 @@
 	 * that this may or may not correspond to the specific types
 	 * referenced by the crush rule.
 	 */
-	if (force >= 0) {
-		if (force >= map->max_devices ||
-		    map->device_parents[force] == 0) {
-			/*dprintk("CRUSH: forcefed device dne\n");*/
-			rc = -1;  /* force fed device dne */
-			goto out;
-		}
-		if (!is_out(map, weight, force, x)) {
-			while (1) {
-				force_context[++force_pos] = force;
-				if (force >= 0)
-					force = map->device_parents[force];
-				else
-					force = map->bucket_parents[-1-force];
-				if (force == 0)
-					break;
-			}
+	if (force >= 0 &&
+	    force < map->max_devices &&
+	    map->device_parents[force] != 0 &&
+	    !is_out(map, weight, force, x)) {
+		while (1) {
+			force_context[++force_pos] = force;
+			if (force >= 0)
+				force = map->device_parents[force];
+			else
+				force = map->bucket_parents[-1-force];
+			if (force == 0)
+				break;
 		}
 	}
 
@@ -600,10 +594,7 @@
 			BUG_ON(1);
 		}
 	}
-	rc = result_len;
-
-out:
-	return rc;
+	return result_len;
 }
 
 
diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
index 733e4600..f4f3f58 100644
--- a/net/ceph/osd_client.c
+++ b/net/ceph/osd_client.c
@@ -244,7 +244,7 @@
 		ceph_pagelist_init(req->r_trail);
 	}
 	/* create request message; allow space for oid */
-	msg_size += 40;
+	msg_size += MAX_OBJ_NAME_SIZE;
 	if (snapc)
 		msg_size += sizeof(u64) * snapc->num_snaps;
 	if (use_mempool)
diff --git a/net/core/dev.c b/net/core/dev.c
index 6ba50a1..5a13edf 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1396,7 +1396,7 @@
 	for_each_net(net) {
 		for_each_netdev(net, dev) {
 			if (dev == last)
-				break;
+				goto outroll;
 
 			if (dev->flags & IFF_UP) {
 				nb->notifier_call(nb, NETDEV_GOING_DOWN, dev);
@@ -1407,6 +1407,7 @@
 		}
 	}
 
+outroll:
 	raw_notifier_chain_unregister(&netdev_chain, nb);
 	goto unlock;
 }
@@ -4282,6 +4283,12 @@
 			    sizeof(struct dev_iter_state));
 }
 
+int dev_seq_open_ops(struct inode *inode, struct file *file,
+		     const struct seq_operations *ops)
+{
+	return seq_open_net(inode, file, ops, sizeof(struct dev_iter_state));
+}
+
 static const struct file_operations dev_seq_fops = {
 	.owner	 = THIS_MODULE,
 	.open    = dev_seq_open,
diff --git a/net/core/dev_addr_lists.c b/net/core/dev_addr_lists.c
index 277faef..febba51 100644
--- a/net/core/dev_addr_lists.c
+++ b/net/core/dev_addr_lists.c
@@ -696,8 +696,7 @@
 
 static int dev_mc_seq_open(struct inode *inode, struct file *file)
 {
-	return seq_open_net(inode, file, &dev_mc_seq_ops,
-			    sizeof(struct seq_net_private));
+	return dev_seq_open_ops(inode, file, &dev_mc_seq_ops);
 }
 
 static const struct file_operations dev_mc_seq_fops = {
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 039d51e..5ac07d3 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -2397,7 +2397,10 @@
 	struct net *net = seq_file_net(seq);
 	struct neigh_table *tbl = state->tbl;
 
-	pn = pn->next;
+	do {
+		pn = pn->next;
+	} while (pn && !net_eq(pneigh_net(pn), net));
+
 	while (!pn) {
 		if (++state->bucket > PNEIGH_HASHMASK)
 			break;
diff --git a/net/core/request_sock.c b/net/core/request_sock.c
index 182236b..9b570a6 100644
--- a/net/core/request_sock.c
+++ b/net/core/request_sock.c
@@ -26,10 +26,11 @@
  * but then some measure against one socket starving all other sockets
  * would be needed.
  *
- * It was 128 by default. Experiments with real servers show, that
+ * The minimum value of it is 128. Experiments with real servers show that
  * it is absolutely not enough even at 100conn/sec. 256 cures most
- * of problems. This value is adjusted to 128 for very small machines
- * (<=32Mb of memory) and to 1024 on normal or better ones (>=256Mb).
+ * of problems.
+ * This value is adjusted to 128 for low memory machines,
+ * and it will increase in proportion to the memory of machine.
  * Note : Dont forget somaxconn that may limit backlog too.
  */
 int sysctl_max_syn_backlog = 256;
diff --git a/net/core/secure_seq.c b/net/core/secure_seq.c
index 025233d..925991a 100644
--- a/net/core/secure_seq.c
+++ b/net/core/secure_seq.c
@@ -19,6 +19,7 @@
 }
 late_initcall(net_secret_init);
 
+#ifdef CONFIG_INET
 static u32 seq_scale(u32 seq)
 {
 	/*
@@ -33,6 +34,7 @@
 	 */
 	return seq + (ktime_to_ns(ktime_get_real()) >> 6);
 }
+#endif
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 __u32 secure_tcpv6_sequence_number(const __be32 *saddr, const __be32 *daddr,
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 18a3ceb..3c30ee4 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -2230,7 +2230,7 @@
  * @shiftlen: shift up to this many bytes
  *
  * Attempts to shift up to shiftlen worth of bytes, which may be less than
- * the length of the skb, from tgt to skb. Returns number bytes shifted.
+ * the length of the skb, from skb to tgt. Returns number bytes shifted.
  * It's up to caller to free skb if everything was shifted.
  *
  * If @tgt runs out of frags, the whole operation is aborted.
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 90a919a..3f4e541 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -111,6 +111,7 @@
 	rt = ip_route_newports(fl4, rt, orig_sport, orig_dport,
 			       inet->inet_sport, inet->inet_dport, sk);
 	if (IS_ERR(rt)) {
+		err = PTR_ERR(rt);
 		rt = NULL;
 		goto failure;
 	}
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index a77d161..94f4ec0 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -112,7 +112,7 @@
 static int dn_dst_gc(struct dst_ops *ops);
 static struct dst_entry *dn_dst_check(struct dst_entry *, __u32);
 static unsigned int dn_dst_default_advmss(const struct dst_entry *dst);
-static unsigned int dn_dst_default_mtu(const struct dst_entry *dst);
+static unsigned int dn_dst_mtu(const struct dst_entry *dst);
 static void dn_dst_destroy(struct dst_entry *);
 static struct dst_entry *dn_dst_negative_advice(struct dst_entry *);
 static void dn_dst_link_failure(struct sk_buff *);
@@ -135,7 +135,7 @@
 	.gc =			dn_dst_gc,
 	.check =		dn_dst_check,
 	.default_advmss =	dn_dst_default_advmss,
-	.default_mtu =		dn_dst_default_mtu,
+	.mtu =			dn_dst_mtu,
 	.cow_metrics =		dst_cow_metrics_generic,
 	.destroy =		dn_dst_destroy,
 	.negative_advice =	dn_dst_negative_advice,
@@ -825,9 +825,11 @@
 	return dn_mss_from_pmtu(dst->dev, dst_mtu(dst));
 }
 
-static unsigned int dn_dst_default_mtu(const struct dst_entry *dst)
+static unsigned int dn_dst_mtu(const struct dst_entry *dst)
 {
-	return dst->dev->mtu;
+	unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
+
+	return mtu ? : dst->dev->mtu;
 }
 
 static struct neighbour *dn_dst_neigh_lookup(const struct dst_entry *dst, const void *daddr)
diff --git a/net/decnet/dn_timer.c b/net/decnet/dn_timer.c
index 67f691b..d9c150c 100644
--- a/net/decnet/dn_timer.c
+++ b/net/decnet/dn_timer.c
@@ -36,16 +36,13 @@
 
 void dn_start_slow_timer(struct sock *sk)
 {
-	sk->sk_timer.expires	= jiffies + SLOW_INTERVAL;
-	sk->sk_timer.function	= dn_slow_timer;
-	sk->sk_timer.data	= (unsigned long)sk;
-
-	add_timer(&sk->sk_timer);
+	setup_timer(&sk->sk_timer, dn_slow_timer, (unsigned long)sk);
+	sk_reset_timer(sk, &sk->sk_timer, jiffies + SLOW_INTERVAL);
 }
 
 void dn_stop_slow_timer(struct sock *sk)
 {
-	del_timer(&sk->sk_timer);
+	sk_stop_timer(sk, &sk->sk_timer);
 }
 
 static void dn_slow_timer(unsigned long arg)
@@ -53,12 +50,10 @@
 	struct sock *sk = (struct sock *)arg;
 	struct dn_scp *scp = DN_SK(sk);
 
-	sock_hold(sk);
 	bh_lock_sock(sk);
 
 	if (sock_owned_by_user(sk)) {
-		sk->sk_timer.expires = jiffies + HZ / 10;
-		add_timer(&sk->sk_timer);
+		sk_reset_timer(sk, &sk->sk_timer, jiffies + HZ / 10);
 		goto out;
 	}
 
@@ -100,9 +95,7 @@
 			scp->keepalive_fxn(sk);
 	}
 
-	sk->sk_timer.expires = jiffies + SLOW_INTERVAL;
-
-	add_timer(&sk->sk_timer);
+	sk_reset_timer(sk, &sk->sk_timer, jiffies + SLOW_INTERVAL);
 out:
 	bh_unlock_sock(sk);
 	sock_put(sk);
diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c
index c1f4154..36d1440 100644
--- a/net/ipv4/ah4.c
+++ b/net/ipv4/ah4.c
@@ -136,8 +136,6 @@
 		memcpy(top_iph+1, iph+1, top_iph->ihl*4 - sizeof(struct iphdr));
 	}
 
-	err = ah->nexthdr;
-
 	kfree(AH_SKB_CB(skb)->tmp);
 	xfrm_output_resume(skb, err);
 }
@@ -264,12 +262,12 @@
 	if (err)
 		goto out;
 
+	err = ah->nexthdr;
+
 	skb->network_header += ah_hlen;
 	memcpy(skb_network_header(skb), work_iph, ihl);
 	__skb_pull(skb, ah_hlen + ihl);
 	skb_set_transport_header(skb, -ihl);
-
-	err = ah->nexthdr;
 out:
 	kfree(AH_SKB_CB(skb)->tmp);
 	xfrm_input_resume(skb, err);
@@ -371,8 +369,6 @@
 		if (err == -EINPROGRESS)
 			goto out;
 
-		if (err == -EBUSY)
-			err = NET_XMIT_DROP;
 		goto out_free;
 	}
 
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index c6b5092..65f01dc 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1490,7 +1490,9 @@
 			     void __user *buffer,
 			     size_t *lenp, loff_t *ppos)
 {
+	int old_value = *(int *)ctl->data;
 	int ret = proc_dointvec(ctl, write, buffer, lenp, ppos);
+	int new_value = *(int *)ctl->data;
 
 	if (write) {
 		struct ipv4_devconf *cnf = ctl->extra1;
@@ -1501,6 +1503,9 @@
 
 		if (cnf == net->ipv4.devconf_dflt)
 			devinet_copy_dflt_conf(net, i);
+		if (i == IPV4_DEVCONF_ACCEPT_LOCAL - 1)
+			if ((new_value == 0) && (old_value != 0))
+				rt_cache_flush(net, 0);
 	}
 
 	return ret;
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index c7472ef..b2ca095 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -1716,7 +1716,8 @@
 	if (err) {
 		int j;
 
-		pmc->sfcount[sfmode]--;
+		if (!delta)
+			pmc->sfcount[sfmode]--;
 		for (j=0; j<i; j++)
 			(void) ip_mc_del1_src(pmc, sfmode, &psfsrc[j]);
 	} else if (isexclude != (pmc->sfcount[MCAST_EXCLUDE] != 0)) {
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index f5e2bda..ccee270 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -108,9 +108,6 @@
 		       icsk->icsk_ca_ops->name);
 	}
 
-	if ((ext & (1 << (INET_DIAG_TOS - 1))) && (sk->sk_family != AF_INET6))
-		RTA_PUT_U8(skb, INET_DIAG_TOS, inet->tos);
-
 	r->idiag_family = sk->sk_family;
 	r->idiag_state = sk->sk_state;
 	r->idiag_timer = 0;
@@ -125,16 +122,23 @@
 	r->id.idiag_src[0] = inet->inet_rcv_saddr;
 	r->id.idiag_dst[0] = inet->inet_daddr;
 
+	/* IPv6 dual-stack sockets use inet->tos for IPv4 connections,
+	 * hence this needs to be included regardless of socket family.
+	 */
+	if (ext & (1 << (INET_DIAG_TOS - 1)))
+		RTA_PUT_U8(skb, INET_DIAG_TOS, inet->tos);
+
 #if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
 	if (r->idiag_family == AF_INET6) {
 		const struct ipv6_pinfo *np = inet6_sk(sk);
 
+		if (ext & (1 << (INET_DIAG_TCLASS - 1)))
+			RTA_PUT_U8(skb, INET_DIAG_TCLASS, np->tclass);
+
 		ipv6_addr_copy((struct in6_addr *)r->id.idiag_src,
 			       &np->rcv_saddr);
 		ipv6_addr_copy((struct in6_addr *)r->id.idiag_dst,
 			       &np->daddr);
-		if (ext & (1 << (INET_DIAG_TOS - 1)))
-			RTA_PUT_U8(skb, INET_DIAG_TOS, np->tclass);
 	}
 #endif
 
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index 3b34d1c..29a07b6 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -84,7 +84,7 @@
 
 	rt = skb_rtable(skb);
 
-	if (opt->is_strictroute && ip_hdr(skb)->daddr != rt->rt_gateway)
+	if (opt->is_strictroute && opt->nexthop != rt->rt_gateway)
 		goto sr_failed;
 
 	if (unlikely(skb->len > dst_mtu(&rt->dst) && !skb_is_gso(skb) &&
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index ec93335..1e60f76 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -568,12 +568,13 @@
 		     ) {
 			if (srrptr + 3 > srrspace)
 				break;
-			if (memcmp(&ip_hdr(skb)->daddr, &optptr[srrptr-1], 4) == 0)
+			if (memcmp(&opt->nexthop, &optptr[srrptr-1], 4) == 0)
 				break;
 		}
 		if (srrptr + 3 <= srrspace) {
 			opt->is_changed = 1;
 			ip_rt_get_source(&optptr[srrptr-1], skb, rt);
+			ip_hdr(skb)->daddr = opt->nexthop;
 			optptr[2] = srrptr+4;
 		} else if (net_ratelimit())
 			printk(KERN_CRIT "ip_forward(): Argh! Destination lost!\n");
@@ -640,6 +641,7 @@
 	}
 	if (srrptr <= srrspace) {
 		opt->srr_is_hit = 1;
+		opt->nexthop = nexthop;
 		opt->is_changed = 1;
 	}
 	return 0;
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 065effd..0b2e732 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -285,6 +285,8 @@
 	if (register_netdevice(dev) < 0)
 		goto failed_free;
 
+	strcpy(nt->parms.name, dev->name);
+
 	dev_hold(dev);
 	ipip_tunnel_link(ipn, nt);
 	return nt;
@@ -759,7 +761,6 @@
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 
 	tunnel->dev = dev;
-	strcpy(tunnel->parms.name, dev->name);
 
 	memcpy(dev->dev_addr, &tunnel->parms.iph.saddr, 4);
 	memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4);
@@ -825,6 +826,7 @@
 static int __net_init ipip_init_net(struct net *net)
 {
 	struct ipip_net *ipn = net_generic(net, ipip_net_id);
+	struct ip_tunnel *t;
 	int err;
 
 	ipn->tunnels[0] = ipn->tunnels_wc;
@@ -848,6 +850,9 @@
 	if ((err = register_netdev(ipn->fb_tunnel_dev)))
 		goto err_reg_dev;
 
+	t = netdev_priv(ipn->fb_tunnel_dev);
+
+	strcpy(t->parms.name, ipn->fb_tunnel_dev->name);
 	return 0;
 
 err_reg_dev:
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 9899619..4f47e06 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -64,7 +64,8 @@
 	/* Change in oif may mean change in hh_len. */
 	hh_len = skb_dst(skb)->dev->hard_header_len;
 	if (skb_headroom(skb) < hh_len &&
-	    pskb_expand_head(skb, hh_len - skb_headroom(skb), 0, GFP_ATOMIC))
+	    pskb_expand_head(skb, HH_DATA_ALIGN(hh_len - skb_headroom(skb)),
+				0, GFP_ATOMIC))
 		return -1;
 
 	return 0;
diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index 1dfc18a..f19f218 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -325,7 +325,6 @@
 # raw + specific targets
 config IP_NF_RAW
 	tristate  'raw table support (required for NOTRACK/TRACE)'
-	depends on NETFILTER_ADVANCED
 	help
 	  This option adds a `raw' table to iptables. This table is the very
 	  first in the netfilter framework and hooks in at the PREROUTING
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index a06f73f..43d4c3b 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -339,7 +339,6 @@
 	sk = ping_v4_lookup(net, iph->daddr, iph->saddr,
 			    ntohs(icmph->un.echo.id), skb->dev->ifindex);
 	if (sk == NULL) {
-		ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
 		pr_debug("no socket, dropping\n");
 		return;	/* No socket for error */
 	}
@@ -679,7 +678,6 @@
 	pr_debug("ping_queue_rcv_skb(sk=%p,sk->num=%d,skb=%p)\n",
 		inet_sk(sk), inet_sk(sk)->inet_num, skb);
 	if (sock_queue_rcv_skb(sk, skb) < 0) {
-		ICMP_INC_STATS_BH(sock_net(sk), ICMP_MIB_INERRORS);
 		kfree_skb(skb);
 		pr_debug("ping_queue_rcv_skb -> failed\n");
 		return -1;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 155138d..46af623 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -112,7 +112,7 @@
 #include <net/secure_seq.h>
 
 #define RT_FL_TOS(oldflp4) \
-    ((u32)(oldflp4->flowi4_tos & (IPTOS_RT_MASK | RTO_ONLINK)))
+	((oldflp4)->flowi4_tos & (IPTOS_RT_MASK | RTO_ONLINK))
 
 #define IP_MAX_MTU	0xFFF0
 
@@ -131,6 +131,7 @@
 static int ip_rt_min_pmtu __read_mostly		= 512 + 20 + 20;
 static int ip_rt_min_advmss __read_mostly	= 256;
 static int rt_chain_length_max __read_mostly	= 20;
+static int redirect_genid;
 
 /*
  *	Interface to generic destination cache.
@@ -138,7 +139,7 @@
 
 static struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie);
 static unsigned int	 ipv4_default_advmss(const struct dst_entry *dst);
-static unsigned int	 ipv4_default_mtu(const struct dst_entry *dst);
+static unsigned int	 ipv4_mtu(const struct dst_entry *dst);
 static void		 ipv4_dst_destroy(struct dst_entry *dst);
 static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst);
 static void		 ipv4_link_failure(struct sk_buff *skb);
@@ -193,7 +194,7 @@
 	.gc =			rt_garbage_collect,
 	.check =		ipv4_dst_check,
 	.default_advmss =	ipv4_default_advmss,
-	.default_mtu =		ipv4_default_mtu,
+	.mtu =			ipv4_mtu,
 	.cow_metrics =		ipv4_cow_metrics,
 	.destroy =		ipv4_dst_destroy,
 	.ifdown =		ipv4_dst_ifdown,
@@ -416,9 +417,13 @@
 	else {
 		struct rtable *r = v;
 		struct neighbour *n;
-		int len;
+		int len, HHUptod;
 
+		rcu_read_lock();
 		n = dst_get_neighbour(&r->dst);
+		HHUptod = (n && (n->nud_state & NUD_CONNECTED)) ? 1 : 0;
+		rcu_read_unlock();
+
 		seq_printf(seq, "%s\t%08X\t%08X\t%8X\t%d\t%u\t%d\t"
 			      "%08X\t%d\t%u\t%u\t%02X\t%d\t%1d\t%08X%n",
 			r->dst.dev ? r->dst.dev->name : "*",
@@ -432,7 +437,7 @@
 			      dst_metric(&r->dst, RTAX_RTTVAR)),
 			r->rt_key_tos,
 			-1,
-			(n && (n->nud_state & NUD_CONNECTED)) ? 1 : 0,
+			HHUptod,
 			r->rt_spec_dst, &len);
 
 		seq_printf(seq, "%*s\n", 127 - len, "");
@@ -837,6 +842,7 @@
 
 	get_random_bytes(&shuffle, sizeof(shuffle));
 	atomic_add(shuffle + 1U, &net->ipv4.rt_genid);
+	redirect_genid++;
 }
 
 /*
@@ -1304,16 +1310,40 @@
 	spin_unlock_bh(rt_hash_lock_addr(hash));
 }
 
+static void check_peer_redir(struct dst_entry *dst, struct inet_peer *peer)
+{
+	struct rtable *rt = (struct rtable *) dst;
+	__be32 orig_gw = rt->rt_gateway;
+	struct neighbour *n, *old_n;
+
+	dst_confirm(&rt->dst);
+
+	rt->rt_gateway = peer->redirect_learned.a4;
+
+	n = ipv4_neigh_lookup(&rt->dst, &rt->rt_gateway);
+	if (IS_ERR(n)) {
+		rt->rt_gateway = orig_gw;
+		return;
+	}
+	old_n = xchg(&rt->dst._neighbour, n);
+	if (old_n)
+		neigh_release(old_n);
+	if (!(n->nud_state & NUD_VALID)) {
+		neigh_event_send(n, NULL);
+	} else {
+		rt->rt_flags |= RTCF_REDIRECTED;
+		call_netevent_notifiers(NETEVENT_NEIGH_UPDATE, n);
+	}
+}
+
 /* called in rcu_read_lock() section */
 void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw,
 		    __be32 saddr, struct net_device *dev)
 {
 	int s, i;
 	struct in_device *in_dev = __in_dev_get_rcu(dev);
-	struct rtable *rt;
 	__be32 skeys[2] = { saddr, 0 };
 	int    ikeys[2] = { dev->ifindex, 0 };
-	struct flowi4 fl4;
 	struct inet_peer *peer;
 	struct net *net;
 
@@ -1336,33 +1366,44 @@
 			goto reject_redirect;
 	}
 
-	memset(&fl4, 0, sizeof(fl4));
-	fl4.daddr = daddr;
 	for (s = 0; s < 2; s++) {
 		for (i = 0; i < 2; i++) {
-			fl4.flowi4_oif = ikeys[i];
-			fl4.saddr = skeys[s];
-			rt = __ip_route_output_key(net, &fl4);
-			if (IS_ERR(rt))
-				continue;
+			unsigned int hash;
+			struct rtable __rcu **rthp;
+			struct rtable *rt;
 
-			if (rt->dst.error || rt->dst.dev != dev ||
-			    rt->rt_gateway != old_gw) {
-				ip_rt_put(rt);
-				continue;
+			hash = rt_hash(daddr, skeys[s], ikeys[i], rt_genid(net));
+
+			rthp = &rt_hash_table[hash].chain;
+
+			while ((rt = rcu_dereference(*rthp)) != NULL) {
+				rthp = &rt->dst.rt_next;
+
+				if (rt->rt_key_dst != daddr ||
+				    rt->rt_key_src != skeys[s] ||
+				    rt->rt_oif != ikeys[i] ||
+				    rt_is_input_route(rt) ||
+				    rt_is_expired(rt) ||
+				    !net_eq(dev_net(rt->dst.dev), net) ||
+				    rt->dst.error ||
+				    rt->dst.dev != dev ||
+				    rt->rt_gateway != old_gw)
+					continue;
+
+				if (!rt->peer)
+					rt_bind_peer(rt, rt->rt_dst, 1);
+
+				peer = rt->peer;
+				if (peer) {
+					if (peer->redirect_learned.a4 != new_gw ||
+					    peer->redirect_genid != redirect_genid) {
+						peer->redirect_learned.a4 = new_gw;
+						peer->redirect_genid = redirect_genid;
+						atomic_inc(&__rt_peer_genid);
+					}
+					check_peer_redir(&rt->dst, peer);
+				}
 			}
-
-			if (!rt->peer)
-				rt_bind_peer(rt, rt->rt_dst, 1);
-
-			peer = rt->peer;
-			if (peer) {
-				peer->redirect_learned.a4 = new_gw;
-				atomic_inc(&__rt_peer_genid);
-			}
-
-			ip_rt_put(rt);
-			return;
 		}
 	}
 	return;
@@ -1649,40 +1690,9 @@
 	}
 }
 
-static int check_peer_redir(struct dst_entry *dst, struct inet_peer *peer)
+
+static void ipv4_validate_peer(struct rtable *rt)
 {
-	struct rtable *rt = (struct rtable *) dst;
-	__be32 orig_gw = rt->rt_gateway;
-	struct neighbour *n, *old_n;
-
-	dst_confirm(&rt->dst);
-
-	rt->rt_gateway = peer->redirect_learned.a4;
-
-	n = ipv4_neigh_lookup(&rt->dst, &rt->rt_gateway);
-	if (IS_ERR(n))
-		return PTR_ERR(n);
-	old_n = xchg(&rt->dst._neighbour, n);
-	if (old_n)
-		neigh_release(old_n);
-	if (!n || !(n->nud_state & NUD_VALID)) {
-		if (n)
-			neigh_event_send(n, NULL);
-		rt->rt_gateway = orig_gw;
-		return -EAGAIN;
-	} else {
-		rt->rt_flags |= RTCF_REDIRECTED;
-		call_netevent_notifiers(NETEVENT_NEIGH_UPDATE, n);
-	}
-	return 0;
-}
-
-static struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie)
-{
-	struct rtable *rt = (struct rtable *) dst;
-
-	if (rt_is_expired(rt))
-		return NULL;
 	if (rt->rt_peer_genid != rt_peer_genid()) {
 		struct inet_peer *peer;
 
@@ -1691,17 +1701,26 @@
 
 		peer = rt->peer;
 		if (peer) {
-			check_peer_pmtu(dst, peer);
+			check_peer_pmtu(&rt->dst, peer);
 
+			if (peer->redirect_genid != redirect_genid)
+				peer->redirect_learned.a4 = 0;
 			if (peer->redirect_learned.a4 &&
-			    peer->redirect_learned.a4 != rt->rt_gateway) {
-				if (check_peer_redir(dst, peer))
-					return NULL;
-			}
+			    peer->redirect_learned.a4 != rt->rt_gateway)
+				check_peer_redir(&rt->dst, peer);
 		}
 
 		rt->rt_peer_genid = rt_peer_genid();
 	}
+}
+
+static struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie)
+{
+	struct rtable *rt = (struct rtable *) dst;
+
+	if (rt_is_expired(rt))
+		return NULL;
+	ipv4_validate_peer(rt);
 	return dst;
 }
 
@@ -1806,12 +1825,17 @@
 	return advmss;
 }
 
-static unsigned int ipv4_default_mtu(const struct dst_entry *dst)
+static unsigned int ipv4_mtu(const struct dst_entry *dst)
 {
-	unsigned int mtu = dst->dev->mtu;
+	const struct rtable *rt = (const struct rtable *) dst;
+	unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
+
+	if (mtu && rt_is_output_route(rt))
+		return mtu;
+
+	mtu = dst->dev->mtu;
 
 	if (unlikely(dst_metric_locked(dst, RTAX_MTU))) {
-		const struct rtable *rt = (const struct rtable *) dst;
 
 		if (rt->rt_gateway != rt->rt_dst && mtu > 576)
 			mtu = 576;
@@ -1844,6 +1868,8 @@
 		dst_init_metrics(&rt->dst, peer->metrics, false);
 
 		check_peer_pmtu(&rt->dst, peer);
+		if (peer->redirect_genid != redirect_genid)
+			peer->redirect_learned.a4 = 0;
 		if (peer->redirect_learned.a4 &&
 		    peer->redirect_learned.a4 != rt->rt_gateway) {
 			rt->rt_gateway = peer->redirect_learned.a4;
@@ -2349,6 +2375,7 @@
 		    rth->rt_mark == skb->mark &&
 		    net_eq(dev_net(rth->dst.dev), net) &&
 		    !rt_is_expired(rth)) {
+			ipv4_validate_peer(rth);
 			if (noref) {
 				dst_use_noref(&rth->dst, jiffies);
 				skb_dst_set_noref(skb, &rth->dst);
@@ -2407,11 +2434,11 @@
 static struct rtable *__mkroute_output(const struct fib_result *res,
 				       const struct flowi4 *fl4,
 				       __be32 orig_daddr, __be32 orig_saddr,
-				       int orig_oif, struct net_device *dev_out,
+				       int orig_oif, __u8 orig_rtos,
+				       struct net_device *dev_out,
 				       unsigned int flags)
 {
 	struct fib_info *fi = res->fi;
-	u32 tos = RT_FL_TOS(fl4);
 	struct in_device *in_dev;
 	u16 type = res->type;
 	struct rtable *rth;
@@ -2462,7 +2489,7 @@
 	rth->rt_genid = rt_genid(dev_net(dev_out));
 	rth->rt_flags	= flags;
 	rth->rt_type	= type;
-	rth->rt_key_tos	= tos;
+	rth->rt_key_tos	= orig_rtos;
 	rth->rt_dst	= fl4->daddr;
 	rth->rt_src	= fl4->saddr;
 	rth->rt_route_iif = 0;
@@ -2512,7 +2539,7 @@
 static struct rtable *ip_route_output_slow(struct net *net, struct flowi4 *fl4)
 {
 	struct net_device *dev_out = NULL;
-	u32 tos	= RT_FL_TOS(fl4);
+	__u8 tos = RT_FL_TOS(fl4);
 	unsigned int flags = 0;
 	struct fib_result res;
 	struct rtable *rth;
@@ -2688,7 +2715,7 @@
 
 make_route:
 	rth = __mkroute_output(&res, fl4, orig_daddr, orig_saddr, orig_oif,
-			       dev_out, flags);
+			       tos, dev_out, flags);
 	if (!IS_ERR(rth)) {
 		unsigned int hash;
 
@@ -2724,6 +2751,7 @@
 			    (IPTOS_RT_MASK | RTO_ONLINK)) &&
 		    net_eq(dev_net(rth->dst.dev), net) &&
 		    !rt_is_expired(rth)) {
+			ipv4_validate_peer(rth);
 			dst_use(&rth->dst, jiffies);
 			RT_CACHE_STAT_INC(out_hit);
 			rcu_read_unlock_bh();
@@ -2747,9 +2775,11 @@
 	return NULL;
 }
 
-static unsigned int ipv4_blackhole_default_mtu(const struct dst_entry *dst)
+static unsigned int ipv4_blackhole_mtu(const struct dst_entry *dst)
 {
-	return 0;
+	unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
+
+	return mtu ? : dst->dev->mtu;
 }
 
 static void ipv4_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu)
@@ -2767,7 +2797,7 @@
 	.protocol		=	cpu_to_be16(ETH_P_IP),
 	.destroy		=	ipv4_dst_destroy,
 	.check			=	ipv4_blackhole_dst_check,
-	.default_mtu		=	ipv4_blackhole_default_mtu,
+	.mtu			=	ipv4_blackhole_mtu,
 	.default_advmss		=	ipv4_default_advmss,
 	.update_pmtu		=	ipv4_rt_blackhole_update_pmtu,
 	.cow_metrics		=	ipv4_rt_blackhole_cow_metrics,
@@ -2845,7 +2875,7 @@
 	struct rtable *rt = skb_rtable(skb);
 	struct rtmsg *r;
 	struct nlmsghdr *nlh;
-	long expires = 0;
+	unsigned long expires = 0;
 	const struct inet_peer *peer = rt->peer;
 	u32 id = 0, ts = 0, tsage = 0, error;
 
@@ -2902,8 +2932,12 @@
 			tsage = get_seconds() - peer->tcp_ts_stamp;
 		}
 		expires = ACCESS_ONCE(peer->pmtu_expires);
-		if (expires)
-			expires -= jiffies;
+		if (expires) {
+			if (time_before(jiffies, expires))
+				expires -= jiffies;
+			else
+				expires = 0;
+		}
 	}
 
 	if (rt_is_input_route(rt)) {
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index a744315..a9db4b1 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1510,6 +1510,7 @@
 	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS);
 	return NULL;
 put_and_exit:
+	tcp_clear_xmit_timers(newsk);
 	bh_unlock_sock(newsk);
 	sock_put(newsk);
 	goto exit;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 980b98f..63170e29 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1382,7 +1382,7 @@
 /* Return 0, if packet can be sent now without violation Nagle's rules:
  * 1. It is full sized.
  * 2. Or it contains FIN. (already checked by caller)
- * 3. Or TCP_NODELAY was set.
+ * 3. Or TCP_CORK is not set, and TCP_NODELAY is set.
  * 4. Or TCP_CORK is not set, and all sent packets are ACKed.
  *    With Minshall's modification: all sent small packets are ACKed.
  */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index ab0966d..5a65eea 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1164,7 +1164,7 @@
 	struct inet_sock *inet = inet_sk(sk);
 	struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
 	struct sk_buff *skb;
-	unsigned int ulen;
+	unsigned int ulen, copied;
 	int peeked;
 	int err;
 	int is_udplite = IS_UDPLITE(sk);
@@ -1186,9 +1186,10 @@
 		goto out;
 
 	ulen = skb->len - sizeof(struct udphdr);
-	if (len > ulen)
-		len = ulen;
-	else if (len < ulen)
+	copied = len;
+	if (copied > ulen)
+		copied = ulen;
+	else if (copied < ulen)
 		msg->msg_flags |= MSG_TRUNC;
 
 	/*
@@ -1197,14 +1198,14 @@
 	 * coverage checksum (UDP-Lite), do it before the copy.
 	 */
 
-	if (len < ulen || UDP_SKB_CB(skb)->partial_cov) {
+	if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
 		if (udp_lib_checksum_complete(skb))
 			goto csum_copy_err;
 	}
 
 	if (skb_csum_unnecessary(skb))
 		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
-					      msg->msg_iov, len);
+					      msg->msg_iov, copied);
 	else {
 		err = skb_copy_and_csum_datagram_iovec(skb,
 						       sizeof(struct udphdr),
@@ -1233,7 +1234,7 @@
 	if (inet->cmsg_flags)
 		ip_cmsg_recv(msg, skb);
 
-	err = len;
+	err = copied;
 	if (flags & MSG_TRUNC)
 		err = ulen;
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index cf88df8..36806de 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1805,7 +1805,8 @@
 		return ERR_PTR(-EACCES);
 
 	/* Add default multicast route */
-	addrconf_add_mroute(dev);
+	if (!(dev->flags & IFF_LOOPBACK))
+		addrconf_add_mroute(dev);
 
 	/* Add link local route */
 	addrconf_add_lroute(dev);
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index 2195ae6..4c0f894 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -324,8 +324,6 @@
 #endif
 	}
 
-	err = ah->nexthdr;
-
 	kfree(AH_SKB_CB(skb)->tmp);
 	xfrm_output_resume(skb, err);
 }
@@ -466,12 +464,12 @@
 	if (err)
 		goto out;
 
+	err = ah->nexthdr;
+
 	skb->network_header += ah_hlen;
 	memcpy(skb_network_header(skb), work_iph, hdr_len);
 	__skb_pull(skb, ah_hlen + hdr_len);
 	skb_set_transport_header(skb, -hdr_len);
-
-	err = ah->nexthdr;
 out:
 	kfree(AH_SKB_CB(skb)->tmp);
 	xfrm_input_resume(skb, err);
@@ -583,8 +581,6 @@
 		if (err == -EINPROGRESS)
 			goto out;
 
-		if (err == -EBUSY)
-			err = NET_XMIT_DROP;
 		goto out_free;
 	}
 
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index fee46d5..1567fb1 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -85,7 +85,7 @@
  * request_sock (formerly open request) hash tables.
  */
 static u32 inet6_synq_hash(const struct in6_addr *raddr, const __be16 rport,
-			   const u32 rnd, const u16 synq_hsize)
+			   const u32 rnd, const u32 synq_hsize)
 {
 	u32 c;
 
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 027c7ff..a46c64e 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -111,6 +111,14 @@
 	    ipv6_addr_loopback(&hdr->daddr))
 		goto err;
 
+	/*
+	 * RFC4291 2.7
+	 * Multicast addresses must not be used as source addresses in IPv6
+	 * packets or appear in any Routing header.
+	 */
+	if (ipv6_addr_is_multicast(&hdr->saddr))
+		goto err;
+
 	skb->transport_header = skb->network_header + sizeof(*hdr);
 	IP6CB(skb)->nhoff = offsetof(struct ipv6hdr, nexthdr);
 
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index bdc15c9..4e2e9ff 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -289,6 +289,8 @@
 	if ((err = register_netdevice(dev)) < 0)
 		goto failed_free;
 
+	strcpy(t->parms.name, dev->name);
+
 	dev_hold(dev);
 	ip6_tnl_link(ip6n, t);
 	return t;
@@ -1407,7 +1409,6 @@
 	struct ip6_tnl *t = netdev_priv(dev);
 
 	t->dev = dev;
-	strcpy(t->parms.name, dev->name);
 	dev->tstats = alloc_percpu(struct pcpu_tstats);
 	if (!dev->tstats)
 		return -ENOMEM;
@@ -1487,6 +1488,7 @@
 static int __net_init ip6_tnl_init_net(struct net *net)
 {
 	struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id);
+	struct ip6_tnl *t = NULL;
 	int err;
 
 	ip6n->tnls[0] = ip6n->tnls_wc;
@@ -1507,6 +1509,10 @@
 	err = register_netdev(ip6n->fb_tnl_dev);
 	if (err < 0)
 		goto err_register;
+
+	t = netdev_priv(ip6n->fb_tnl_dev);
+
+	strcpy(t->parms.name, ip6n->fb_tnl_dev->name);
 	return 0;
 
 err_register:
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index c99e3ee..26cb08c 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -503,7 +503,7 @@
 			goto e_inval;
 		if (val > 255 || val < -1)
 			goto e_inval;
-		np->mcast_hops = val;
+		np->mcast_hops = (val == -1 ? IPV6_DEFAULT_MCASTHOPS : val);
 		retv = 0;
 		break;
 
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 44e5b7f..0cb78d7 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1571,7 +1571,7 @@
 	}
 	if (!rt->rt6i_peer)
 		rt6_bind_peer(rt, 1);
-	if (inet_peer_xrlim_allow(rt->rt6i_peer, 1*HZ))
+	if (!inet_peer_xrlim_allow(rt->rt6i_peer, 1*HZ))
 		goto release;
 
 	if (dev->addr_len) {
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index 4484648..f792b34 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -186,7 +186,6 @@
 
 config IP6_NF_RAW
 	tristate  'raw table support (required for TRACE)'
-	depends on NETFILTER_ADVANCED
 	help
 	  This option adds a `raw' table to ip6tables. This table is the very
 	  first in the netfilter framework and hooks in at the PREROUTING
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 8473016..b582a0a 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -77,7 +77,7 @@
 				    const struct in6_addr *dest);
 static struct dst_entry	*ip6_dst_check(struct dst_entry *dst, u32 cookie);
 static unsigned int	 ip6_default_advmss(const struct dst_entry *dst);
-static unsigned int	 ip6_default_mtu(const struct dst_entry *dst);
+static unsigned int	 ip6_mtu(const struct dst_entry *dst);
 static struct dst_entry *ip6_negative_advice(struct dst_entry *);
 static void		ip6_dst_destroy(struct dst_entry *);
 static void		ip6_dst_ifdown(struct dst_entry *,
@@ -144,7 +144,7 @@
 	.gc_thresh		=	1024,
 	.check			=	ip6_dst_check,
 	.default_advmss		=	ip6_default_advmss,
-	.default_mtu		=	ip6_default_mtu,
+	.mtu			=	ip6_mtu,
 	.cow_metrics		=	ipv6_cow_metrics,
 	.destroy		=	ip6_dst_destroy,
 	.ifdown			=	ip6_dst_ifdown,
@@ -155,9 +155,11 @@
 	.neigh_lookup		=	ip6_neigh_lookup,
 };
 
-static unsigned int ip6_blackhole_default_mtu(const struct dst_entry *dst)
+static unsigned int ip6_blackhole_mtu(const struct dst_entry *dst)
 {
-	return 0;
+	unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
+
+	return mtu ? : dst->dev->mtu;
 }
 
 static void ip6_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu)
@@ -175,7 +177,7 @@
 	.protocol		=	cpu_to_be16(ETH_P_IPV6),
 	.destroy		=	ip6_dst_destroy,
 	.check			=	ip6_dst_check,
-	.default_mtu		=	ip6_blackhole_default_mtu,
+	.mtu			=	ip6_blackhole_mtu,
 	.default_advmss		=	ip6_default_advmss,
 	.update_pmtu		=	ip6_rt_blackhole_update_pmtu,
 	.cow_metrics		=	ip6_rt_blackhole_cow_metrics,
@@ -726,7 +728,7 @@
 		int attempts = !in_softirq();
 
 		if (!(rt->rt6i_flags&RTF_GATEWAY)) {
-			if (rt->rt6i_dst.plen != 128 &&
+			if (ort->rt6i_dst.plen != 128 &&
 			    ipv6_addr_equal(&ort->rt6i_dst.addr, daddr))
 				rt->rt6i_flags |= RTF_ANYCAST;
 			ipv6_addr_copy(&rt->rt6i_gateway, daddr);
@@ -1041,10 +1043,15 @@
 	return mtu;
 }
 
-static unsigned int ip6_default_mtu(const struct dst_entry *dst)
+static unsigned int ip6_mtu(const struct dst_entry *dst)
 {
-	unsigned int mtu = IPV6_MIN_MTU;
 	struct inet6_dev *idev;
+	unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
+
+	if (mtu)
+		return mtu;
+
+	mtu = IPV6_MIN_MTU;
 
 	rcu_read_lock();
 	idev = __in6_dev_get(dst->dev);
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index a7a1860..96f3623 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -263,6 +263,8 @@
 	if (register_netdevice(dev) < 0)
 		goto failed_free;
 
+	strcpy(nt->parms.name, dev->name);
+
 	dev_hold(dev);
 
 	ipip6_tunnel_link(sitn, nt);
@@ -1144,7 +1146,6 @@
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 
 	tunnel->dev = dev;
-	strcpy(tunnel->parms.name, dev->name);
 
 	memcpy(dev->dev_addr, &tunnel->parms.iph.saddr, 4);
 	memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4);
@@ -1207,6 +1208,7 @@
 static int __net_init sit_init_net(struct net *net)
 {
 	struct sit_net *sitn = net_generic(net, sit_net_id);
+	struct ip_tunnel *t;
 	int err;
 
 	sitn->tunnels[0] = sitn->tunnels_wc;
@@ -1231,6 +1233,9 @@
 	if ((err = register_netdev(sitn->fb_tunnel_dev)))
 		goto err_reg_dev;
 
+	t = netdev_priv(sitn->fb_tunnel_dev);
+
+	strcpy(t->parms.name, sitn->fb_tunnel_dev->name);
 	return 0;
 
 err_reg_dev:
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 36131d1..2dea4bb 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1255,6 +1255,13 @@
 	if (!want_cookie || tmp_opt.tstamp_ok)
 		TCP_ECN_create_request(req, tcp_hdr(skb));
 
+	treq->iif = sk->sk_bound_dev_if;
+
+	/* So that link locals have meaning */
+	if (!sk->sk_bound_dev_if &&
+	    ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL)
+		treq->iif = inet6_iif(skb);
+
 	if (!isn) {
 		struct inet_peer *peer = NULL;
 
@@ -1264,12 +1271,6 @@
 			atomic_inc(&skb->users);
 			treq->pktopts = skb;
 		}
-		treq->iif = sk->sk_bound_dev_if;
-
-		/* So that link locals have meaning */
-		if (!sk->sk_bound_dev_if &&
-		    ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL)
-			treq->iif = inet6_iif(skb);
 
 		if (want_cookie) {
 			isn = cookie_v6_init_sequence(sk, skb, &req->mss);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 846f4757..8c25419 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -340,7 +340,7 @@
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct inet_sock *inet = inet_sk(sk);
 	struct sk_buff *skb;
-	unsigned int ulen;
+	unsigned int ulen, copied;
 	int peeked;
 	int err;
 	int is_udplite = IS_UDPLITE(sk);
@@ -363,9 +363,10 @@
 		goto out;
 
 	ulen = skb->len - sizeof(struct udphdr);
-	if (len > ulen)
-		len = ulen;
-	else if (len < ulen)
+	copied = len;
+	if (copied > ulen)
+		copied = ulen;
+	else if (copied < ulen)
 		msg->msg_flags |= MSG_TRUNC;
 
 	is_udp4 = (skb->protocol == htons(ETH_P_IP));
@@ -376,14 +377,14 @@
 	 * coverage checksum (UDP-Lite), do it before the copy.
 	 */
 
-	if (len < ulen || UDP_SKB_CB(skb)->partial_cov) {
+	if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
 		if (udp_lib_checksum_complete(skb))
 			goto csum_copy_err;
 	}
 
 	if (skb_csum_unnecessary(skb))
 		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
-					      msg->msg_iov,len);
+					      msg->msg_iov, copied       );
 	else {
 		err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov);
 		if (err == -EINVAL)
@@ -432,7 +433,7 @@
 			datagram_recv_ctl(sk, msg, skb);
 	}
 
-	err = len;
+	err = copied;
 	if (flags & MSG_TRUNC)
 		err = ulen;
 
diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index bf8d50c..89ff8c6 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -756,9 +756,6 @@
 		goto error;
 	}
 
-	/* Point to L2TP header */
-	optr = ptr = skb->data;
-
 	/* Trace packet contents, if enabled */
 	if (tunnel->debug & L2TP_MSG_DATA) {
 		length = min(32u, skb->len);
@@ -769,12 +766,15 @@
 
 		offset = 0;
 		do {
-			printk(" %02X", ptr[offset]);
+			printk(" %02X", skb->data[offset]);
 		} while (++offset < length);
 
 		printk("\n");
 	}
 
+	/* Point to L2TP header */
+	optr = ptr = skb->data;
+
 	/* Get L2TP header flags */
 	hdrflags = ntohs(*(__be16 *) ptr);
 
@@ -1072,7 +1072,7 @@
 
 	/* Get routing info from the tunnel socket */
 	skb_dst_drop(skb);
-	skb_dst_set(skb, dst_clone(__sk_dst_get(sk)));
+	skb_dst_set(skb, dst_clone(__sk_dst_check(sk, 0)));
 
 	inet = inet_sk(sk);
 	fl = &inet->cork.fl;
diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
index b3f6552..2e4b961 100644
--- a/net/mac80211/agg-tx.c
+++ b/net/mac80211/agg-tx.c
@@ -161,6 +161,12 @@
 		return -ENOENT;
 	}
 
+	/* if we're already stopping ignore any new requests to stop */
+	if (test_bit(HT_AGG_STATE_STOPPING, &tid_tx->state)) {
+		spin_unlock_bh(&sta->lock);
+		return -EALREADY;
+	}
+
 	if (test_bit(HT_AGG_STATE_WANT_START, &tid_tx->state)) {
 		/* not even started yet! */
 		ieee80211_assign_tid_tx(sta, tid, NULL);
@@ -169,6 +175,8 @@
 		return 0;
 	}
 
+	set_bit(HT_AGG_STATE_STOPPING, &tid_tx->state);
+
 	spin_unlock_bh(&sta->lock);
 
 #ifdef CONFIG_MAC80211_HT_DEBUG
@@ -176,8 +184,6 @@
 	       sta->sta.addr, tid);
 #endif /* CONFIG_MAC80211_HT_DEBUG */
 
-	set_bit(HT_AGG_STATE_STOPPING, &tid_tx->state);
-
 	del_timer_sync(&tid_tx->addba_resp_timer);
 
 	/*
@@ -187,6 +193,20 @@
 	 */
 	clear_bit(HT_AGG_STATE_OPERATIONAL, &tid_tx->state);
 
+	/*
+	 * There might be a few packets being processed right now (on
+	 * another CPU) that have already gotten past the aggregation
+	 * check when it was still OPERATIONAL and consequently have
+	 * IEEE80211_TX_CTL_AMPDU set. In that case, this code might
+	 * call into the driver at the same time or even before the
+	 * TX paths calls into it, which could confuse the driver.
+	 *
+	 * Wait for all currently running TX paths to finish before
+	 * telling the driver. New packets will not go through since
+	 * the aggregation session is no longer OPERATIONAL.
+	 */
+	synchronize_net();
+
 	tid_tx->stop_initiator = initiator;
 	tid_tx->tx_stop = tx;
 
@@ -283,6 +303,38 @@
 	__release(agg_queue);
 }
 
+/*
+ * splice packets from the STA's pending to the local pending,
+ * requires a call to ieee80211_agg_splice_finish later
+ */
+static void __acquires(agg_queue)
+ieee80211_agg_splice_packets(struct ieee80211_local *local,
+			     struct tid_ampdu_tx *tid_tx, u16 tid)
+{
+	int queue = ieee80211_ac_from_tid(tid);
+	unsigned long flags;
+
+	ieee80211_stop_queue_agg(local, tid);
+
+	if (WARN(!tid_tx, "TID %d gone but expected when splicing aggregates"
+			  " from the pending queue\n", tid))
+		return;
+
+	if (!skb_queue_empty(&tid_tx->pending)) {
+		spin_lock_irqsave(&local->queue_stop_reason_lock, flags);
+		/* copy over remaining packets */
+		skb_queue_splice_tail_init(&tid_tx->pending,
+					   &local->pending[queue]);
+		spin_unlock_irqrestore(&local->queue_stop_reason_lock, flags);
+	}
+}
+
+static void __releases(agg_queue)
+ieee80211_agg_splice_finish(struct ieee80211_local *local, u16 tid)
+{
+	ieee80211_wake_queue_agg(local, tid);
+}
+
 void ieee80211_tx_ba_session_handle_start(struct sta_info *sta, int tid)
 {
 	struct tid_ampdu_tx *tid_tx;
@@ -294,19 +346,17 @@
 	tid_tx = rcu_dereference_protected_tid_tx(sta, tid);
 
 	/*
-	 * While we're asking the driver about the aggregation,
-	 * stop the AC queue so that we don't have to worry
-	 * about frames that came in while we were doing that,
-	 * which would require us to put them to the AC pending
-	 * afterwards which just makes the code more complex.
+	 * Start queuing up packets for this aggregation session.
+	 * We're going to release them once the driver is OK with
+	 * that.
 	 */
-	ieee80211_stop_queue_agg(local, tid);
-
 	clear_bit(HT_AGG_STATE_WANT_START, &tid_tx->state);
 
 	/*
-	 * make sure no packets are being processed to get
-	 * valid starting sequence number
+	 * Make sure no packets are being processed. This ensures that
+	 * we have a valid starting sequence number and that in-flight
+	 * packets have been flushed out and no packets for this TID
+	 * will go into the driver during the ampdu_action call.
 	 */
 	synchronize_net();
 
@@ -320,17 +370,15 @@
 					" tid %d\n", tid);
 #endif
 		spin_lock_bh(&sta->lock);
+		ieee80211_agg_splice_packets(local, tid_tx, tid);
 		ieee80211_assign_tid_tx(sta, tid, NULL);
+		ieee80211_agg_splice_finish(local, tid);
 		spin_unlock_bh(&sta->lock);
 
-		ieee80211_wake_queue_agg(local, tid);
 		kfree_rcu(tid_tx, rcu_head);
 		return;
 	}
 
-	/* we can take packets again now */
-	ieee80211_wake_queue_agg(local, tid);
-
 	/* activate the timer for the recipient's addBA response */
 	mod_timer(&tid_tx->addba_resp_timer, jiffies + ADDBA_RESP_INTERVAL);
 #ifdef CONFIG_MAC80211_HT_DEBUG
@@ -446,38 +494,6 @@
 }
 EXPORT_SYMBOL(ieee80211_start_tx_ba_session);
 
-/*
- * splice packets from the STA's pending to the local pending,
- * requires a call to ieee80211_agg_splice_finish later
- */
-static void __acquires(agg_queue)
-ieee80211_agg_splice_packets(struct ieee80211_local *local,
-			     struct tid_ampdu_tx *tid_tx, u16 tid)
-{
-	int queue = ieee80211_ac_from_tid(tid);
-	unsigned long flags;
-
-	ieee80211_stop_queue_agg(local, tid);
-
-	if (WARN(!tid_tx, "TID %d gone but expected when splicing aggregates"
-			  " from the pending queue\n", tid))
-		return;
-
-	if (!skb_queue_empty(&tid_tx->pending)) {
-		spin_lock_irqsave(&local->queue_stop_reason_lock, flags);
-		/* copy over remaining packets */
-		skb_queue_splice_tail_init(&tid_tx->pending,
-					   &local->pending[queue]);
-		spin_unlock_irqrestore(&local->queue_stop_reason_lock, flags);
-	}
-}
-
-static void __releases(agg_queue)
-ieee80211_agg_splice_finish(struct ieee80211_local *local, u16 tid)
-{
-	ieee80211_wake_queue_agg(local, tid);
-}
-
 static void ieee80211_agg_tx_operational(struct ieee80211_local *local,
 					 struct sta_info *sta, u16 tid)
 {
@@ -757,11 +773,27 @@
 		goto out;
 	}
 
-	del_timer(&tid_tx->addba_resp_timer);
+	del_timer_sync(&tid_tx->addba_resp_timer);
 
 #ifdef CONFIG_MAC80211_HT_DEBUG
 	printk(KERN_DEBUG "switched off addBA timer for tid %d\n", tid);
 #endif
+
+	/*
+	 * addba_resp_timer may have fired before we got here, and
+	 * caused WANT_STOP to be set. If the stop then was already
+	 * processed further, STOPPING might be set.
+	 */
+	if (test_bit(HT_AGG_STATE_WANT_STOP, &tid_tx->state) ||
+	    test_bit(HT_AGG_STATE_STOPPING, &tid_tx->state)) {
+#ifdef CONFIG_MAC80211_HT_DEBUG
+		printk(KERN_DEBUG
+		       "got addBA resp for tid %d but we already gave up\n",
+		       tid);
+#endif
+		goto out;
+	}
+
 	/*
 	 * IEEE 802.11-2007 7.3.1.14:
 	 * In an ADDBA Response frame, when the Status Code field
diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c
index c5f3417..3110cbd 100644
--- a/net/mac80211/debugfs_sta.c
+++ b/net/mac80211/debugfs_sta.c
@@ -274,9 +274,9 @@
 
 		PRINT_HT_CAP((htc->cap & BIT(10)), "HT Delayed Block Ack");
 
-		PRINT_HT_CAP((htc->cap & BIT(11)), "Max AMSDU length: "
-			     "3839 bytes");
 		PRINT_HT_CAP(!(htc->cap & BIT(11)), "Max AMSDU length: "
+			     "3839 bytes");
+		PRINT_HT_CAP((htc->cap & BIT(11)), "Max AMSDU length: "
 			     "7935 bytes");
 
 		/*
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index d999bf3..cae4435 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -757,6 +757,12 @@
 	if (!local->int_scan_req)
 		return -ENOMEM;
 
+	for (band = 0; band < IEEE80211_NUM_BANDS; band++) {
+		if (!local->hw.wiphy->bands[band])
+			continue;
+		local->int_scan_req->rates[band] = (u32) -1;
+	}
+
 	/* if low-level driver supports AP, we also support VLAN */
 	if (local->hw.wiphy->interface_modes & BIT(NL80211_IFTYPE_AP)) {
 		hw->wiphy->interface_modes |= BIT(NL80211_IFTYPE_AP_VLAN);
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 72c8bea..b1b1bb3 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -1487,6 +1487,7 @@
 	int i, j, err;
 	bool have_higher_than_11mbit = false;
 	u16 ap_ht_cap_flags;
+	int min_rate = INT_MAX, min_rate_index = -1;
 
 	/* AssocResp and ReassocResp have identical structure */
 
@@ -1553,6 +1554,10 @@
 				rates |= BIT(j);
 				if (is_basic)
 					basic_rates |= BIT(j);
+				if (rate < min_rate) {
+					min_rate = rate;
+					min_rate_index = j;
+				}
 				break;
 			}
 		}
@@ -1570,11 +1575,25 @@
 				rates |= BIT(j);
 				if (is_basic)
 					basic_rates |= BIT(j);
+				if (rate < min_rate) {
+					min_rate = rate;
+					min_rate_index = j;
+				}
 				break;
 			}
 		}
 	}
 
+	/*
+	 * some buggy APs don't advertise basic_rates. use the lowest
+	 * supported rate instead.
+	 */
+	if (unlikely(!basic_rates) && min_rate_index >= 0) {
+		printk(KERN_DEBUG "%s: No basic rates in AssocResp. "
+		       "Using min supported rate instead.\n", sdata->name);
+		basic_rates = BIT(min_rate_index);
+	}
+
 	sta->sta.supp_rates[wk->chan->band] = rates;
 	sdata->vif.bss_conf.basic_rates = basic_rates;
 
@@ -2269,6 +2288,7 @@
 
 	cancel_work_sync(&ifmgd->request_smps_work);
 
+	cancel_work_sync(&ifmgd->monitor_work);
 	cancel_work_sync(&ifmgd->beacon_connection_loss_work);
 	if (del_timer_sync(&ifmgd->timer))
 		set_bit(TMR_RUNNING_TIMER, &ifmgd->timers_running);
@@ -2277,7 +2297,6 @@
 	if (del_timer_sync(&ifmgd->chswitch_timer))
 		set_bit(TMR_RUNNING_CHANSW, &ifmgd->timers_running);
 
-	cancel_work_sync(&ifmgd->monitor_work);
 	/* these will just be re-established on connection */
 	del_timer_sync(&ifmgd->conn_mon_timer);
 	del_timer_sync(&ifmgd->bcn_mon_timer);
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index bb53726..fb123e2 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -141,8 +141,9 @@
 	pos++;
 
 	/* IEEE80211_RADIOTAP_RATE */
-	if (status->flag & RX_FLAG_HT) {
+	if (!rate || status->flag & RX_FLAG_HT) {
 		/*
+		 * Without rate information don't add it. If we have,
 		 * MCS information is a separate field in radiotap,
 		 * added below. The byte here is needed as padding
 		 * for the channel though, so initialise it to 0.
@@ -163,12 +164,14 @@
 	else if (status->flag & RX_FLAG_HT)
 		put_unaligned_le16(IEEE80211_CHAN_DYN | IEEE80211_CHAN_2GHZ,
 				   pos);
-	else if (rate->flags & IEEE80211_RATE_ERP_G)
+	else if (rate && rate->flags & IEEE80211_RATE_ERP_G)
 		put_unaligned_le16(IEEE80211_CHAN_OFDM | IEEE80211_CHAN_2GHZ,
 				   pos);
-	else
+	else if (rate)
 		put_unaligned_le16(IEEE80211_CHAN_CCK | IEEE80211_CHAN_2GHZ,
 				   pos);
+	else
+		put_unaligned_le16(IEEE80211_CHAN_2GHZ, pos);
 	pos += 2;
 
 	/* IEEE80211_RADIOTAP_DBM_ANTSIGNAL */
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index ce962d2..8eaa746 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -1354,12 +1354,12 @@
 			 * Use MoreData flag to indicate whether there are
 			 * more buffered frames for this STA
 			 */
-			if (!more_data)
-				hdr->frame_control &=
-					cpu_to_le16(~IEEE80211_FCTL_MOREDATA);
-			else
+			if (more_data || !skb_queue_empty(&frames))
 				hdr->frame_control |=
 					cpu_to_le16(IEEE80211_FCTL_MOREDATA);
+			else
+				hdr->frame_control &=
+					cpu_to_le16(~IEEE80211_FCTL_MOREDATA);
 
 			if (ieee80211_is_data_qos(hdr->frame_control) ||
 			    ieee80211_is_qos_nullfunc(hdr->frame_control))
diff --git a/net/mac80211/status.c b/net/mac80211/status.c
index 80de436..16518f3 100644
--- a/net/mac80211/status.c
+++ b/net/mac80211/status.c
@@ -260,7 +260,7 @@
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data;
 	struct ieee80211_radiotap_header *rthdr;
 	unsigned char *pos;
-	__le16 txflags;
+	u16 txflags;
 
 	rthdr = (struct ieee80211_radiotap_header *) skb_push(skb, rtap_len);
 
@@ -290,13 +290,13 @@
 	txflags = 0;
 	if (!(info->flags & IEEE80211_TX_STAT_ACK) &&
 	    !is_multicast_ether_addr(hdr->addr1))
-		txflags |= cpu_to_le16(IEEE80211_RADIOTAP_F_TX_FAIL);
+		txflags |= IEEE80211_RADIOTAP_F_TX_FAIL;
 
 	if ((info->status.rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS) ||
 	    (info->status.rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT))
-		txflags |= cpu_to_le16(IEEE80211_RADIOTAP_F_TX_CTS);
+		txflags |= IEEE80211_RADIOTAP_F_TX_CTS;
 	else if (info->status.rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS)
-		txflags |= cpu_to_le16(IEEE80211_RADIOTAP_F_TX_RTS);
+		txflags |= IEEE80211_RADIOTAP_F_TX_RTS;
 
 	put_unaligned_le16(txflags, pos);
 	pos += 2;
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 51e256c..d5230ec 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -881,6 +881,8 @@
 	skb = ieee80211_probereq_get(&local->hw, &sdata->vif,
 				     ssid, ssid_len,
 				     buf, buf_len);
+	if (!skb)
+		goto out;
 
 	if (dst) {
 		mgmt = (struct ieee80211_mgmt *) skb->data;
@@ -889,6 +891,8 @@
 	}
 
 	IEEE80211_SKB_CB(skb)->flags |= IEEE80211_TX_INTFL_DONT_ENCRYPT;
+
+ out:
 	kfree(buf);
 
 	return skb;
@@ -1035,7 +1039,6 @@
 					     struct ieee80211_sub_if_data,
 					     u.ap);
 
-			memset(&sta->sta.drv_priv, 0, hw->sta_data_size);
 			WARN_ON(drv_sta_add(local, sdata, &sta->sta));
 		}
 	}
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 8260b13..d5597b7 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -201,7 +201,6 @@
 
 config NF_CONNTRACK_NETBIOS_NS
 	tristate "NetBIOS name service protocol support"
-	depends on NETFILTER_ADVANCED
 	select NF_CONNTRACK_BROADCAST
 	help
 	  NetBIOS name service requests are sent as broadcast messages from an
@@ -542,7 +541,6 @@
 	tristate  '"NOTRACK" target support'
 	depends on IP_NF_RAW || IP6_NF_RAW
 	depends on NF_CONNTRACK
-	depends on NETFILTER_ADVANCED
 	help
 	  The NOTRACK target allows a select rule to specify
 	  which packets *not* to enter the conntrack/NAT
diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c
index 6ee10f5..37d667e 100644
--- a/net/netfilter/ipset/ip_set_hash_ipport.c
+++ b/net/netfilter/ipset/ip_set_hash_ipport.c
@@ -158,7 +158,7 @@
 	const struct ip_set_hash *h = set->data;
 	ipset_adtfn adtfn = set->variant->adt[adt];
 	struct hash_ipport4_elem data = { };
-	u32 ip, ip_to, p = 0, port, port_to;
+	u32 ip, ip_to = 0, p = 0, port, port_to;
 	u32 timeout = h->timeout;
 	bool with_ports = false;
 	int ret;
diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c
index fb90e34..e69e271 100644
--- a/net/netfilter/ipset/ip_set_hash_ipportip.c
+++ b/net/netfilter/ipset/ip_set_hash_ipportip.c
@@ -162,7 +162,7 @@
 	const struct ip_set_hash *h = set->data;
 	ipset_adtfn adtfn = set->variant->adt[adt];
 	struct hash_ipportip4_elem data = { };
-	u32 ip, ip_to, p = 0, port, port_to;
+	u32 ip, ip_to = 0, p = 0, port, port_to;
 	u32 timeout = h->timeout;
 	bool with_ports = false;
 	int ret;
diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c
index deb3e3d..64199b4 100644
--- a/net/netfilter/ipset/ip_set_hash_ipportnet.c
+++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c
@@ -184,7 +184,7 @@
 	const struct ip_set_hash *h = set->data;
 	ipset_adtfn adtfn = set->variant->adt[adt];
 	struct hash_ipportnet4_elem data = { .cidr = HOST_MASK };
-	u32 ip, ip_to, p = 0, port, port_to;
+	u32 ip, ip_to = 0, p = 0, port, port_to;
 	u32 ip2_from = 0, ip2_to, ip2_last, ip2;
 	u32 timeout = h->timeout;
 	bool with_ports = false;
diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c
index 6b368be..b62c414 100644
--- a/net/netfilter/nf_conntrack_ecache.c
+++ b/net/netfilter/nf_conntrack_ecache.c
@@ -27,22 +27,17 @@
 
 static DEFINE_MUTEX(nf_ct_ecache_mutex);
 
-struct nf_ct_event_notifier __rcu *nf_conntrack_event_cb __read_mostly;
-EXPORT_SYMBOL_GPL(nf_conntrack_event_cb);
-
-struct nf_exp_event_notifier __rcu *nf_expect_event_cb __read_mostly;
-EXPORT_SYMBOL_GPL(nf_expect_event_cb);
-
 /* deliver cached events and clear cache entry - must be called with locally
  * disabled softirqs */
 void nf_ct_deliver_cached_events(struct nf_conn *ct)
 {
+	struct net *net = nf_ct_net(ct);
 	unsigned long events;
 	struct nf_ct_event_notifier *notify;
 	struct nf_conntrack_ecache *e;
 
 	rcu_read_lock();
-	notify = rcu_dereference(nf_conntrack_event_cb);
+	notify = rcu_dereference(net->ct.nf_conntrack_event_cb);
 	if (notify == NULL)
 		goto out_unlock;
 
@@ -83,19 +78,20 @@
 }
 EXPORT_SYMBOL_GPL(nf_ct_deliver_cached_events);
 
-int nf_conntrack_register_notifier(struct nf_ct_event_notifier *new)
+int nf_conntrack_register_notifier(struct net *net,
+				   struct nf_ct_event_notifier *new)
 {
 	int ret = 0;
 	struct nf_ct_event_notifier *notify;
 
 	mutex_lock(&nf_ct_ecache_mutex);
-	notify = rcu_dereference_protected(nf_conntrack_event_cb,
+	notify = rcu_dereference_protected(net->ct.nf_conntrack_event_cb,
 					   lockdep_is_held(&nf_ct_ecache_mutex));
 	if (notify != NULL) {
 		ret = -EBUSY;
 		goto out_unlock;
 	}
-	RCU_INIT_POINTER(nf_conntrack_event_cb, new);
+	RCU_INIT_POINTER(net->ct.nf_conntrack_event_cb, new);
 	mutex_unlock(&nf_ct_ecache_mutex);
 	return ret;
 
@@ -105,32 +101,34 @@
 }
 EXPORT_SYMBOL_GPL(nf_conntrack_register_notifier);
 
-void nf_conntrack_unregister_notifier(struct nf_ct_event_notifier *new)
+void nf_conntrack_unregister_notifier(struct net *net,
+				      struct nf_ct_event_notifier *new)
 {
 	struct nf_ct_event_notifier *notify;
 
 	mutex_lock(&nf_ct_ecache_mutex);
-	notify = rcu_dereference_protected(nf_conntrack_event_cb,
+	notify = rcu_dereference_protected(net->ct.nf_conntrack_event_cb,
 					   lockdep_is_held(&nf_ct_ecache_mutex));
 	BUG_ON(notify != new);
-	RCU_INIT_POINTER(nf_conntrack_event_cb, NULL);
+	RCU_INIT_POINTER(net->ct.nf_conntrack_event_cb, NULL);
 	mutex_unlock(&nf_ct_ecache_mutex);
 }
 EXPORT_SYMBOL_GPL(nf_conntrack_unregister_notifier);
 
-int nf_ct_expect_register_notifier(struct nf_exp_event_notifier *new)
+int nf_ct_expect_register_notifier(struct net *net,
+				   struct nf_exp_event_notifier *new)
 {
 	int ret = 0;
 	struct nf_exp_event_notifier *notify;
 
 	mutex_lock(&nf_ct_ecache_mutex);
-	notify = rcu_dereference_protected(nf_expect_event_cb,
+	notify = rcu_dereference_protected(net->ct.nf_expect_event_cb,
 					   lockdep_is_held(&nf_ct_ecache_mutex));
 	if (notify != NULL) {
 		ret = -EBUSY;
 		goto out_unlock;
 	}
-	RCU_INIT_POINTER(nf_expect_event_cb, new);
+	RCU_INIT_POINTER(net->ct.nf_expect_event_cb, new);
 	mutex_unlock(&nf_ct_ecache_mutex);
 	return ret;
 
@@ -140,15 +138,16 @@
 }
 EXPORT_SYMBOL_GPL(nf_ct_expect_register_notifier);
 
-void nf_ct_expect_unregister_notifier(struct nf_exp_event_notifier *new)
+void nf_ct_expect_unregister_notifier(struct net *net,
+				      struct nf_exp_event_notifier *new)
 {
 	struct nf_exp_event_notifier *notify;
 
 	mutex_lock(&nf_ct_ecache_mutex);
-	notify = rcu_dereference_protected(nf_expect_event_cb,
+	notify = rcu_dereference_protected(net->ct.nf_expect_event_cb,
 					   lockdep_is_held(&nf_ct_ecache_mutex));
 	BUG_ON(notify != new);
-	RCU_INIT_POINTER(nf_expect_event_cb, NULL);
+	RCU_INIT_POINTER(net->ct.nf_expect_event_cb, NULL);
 	mutex_unlock(&nf_ct_ecache_mutex);
 }
 EXPORT_SYMBOL_GPL(nf_ct_expect_unregister_notifier);
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index e58aa9b..ef21b22 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -4,7 +4,7 @@
  * (C) 2001 by Jay Schulist <jschlst@samba.org>
  * (C) 2002-2006 by Harald Welte <laforge@gnumonks.org>
  * (C) 2003 by Patrick Mchardy <kaber@trash.net>
- * (C) 2005-2008 by Pablo Neira Ayuso <pablo@netfilter.org>
+ * (C) 2005-2011 by Pablo Neira Ayuso <pablo@netfilter.org>
  *
  * Initial connection tracking via netlink development funded and
  * generally made possible by Network Robots, Inc. (www.networkrobots.com)
@@ -2163,6 +2163,54 @@
 MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_CTNETLINK);
 MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_CTNETLINK_EXP);
 
+static int __net_init ctnetlink_net_init(struct net *net)
+{
+#ifdef CONFIG_NF_CONNTRACK_EVENTS
+	int ret;
+
+	ret = nf_conntrack_register_notifier(net, &ctnl_notifier);
+	if (ret < 0) {
+		pr_err("ctnetlink_init: cannot register notifier.\n");
+		goto err_out;
+	}
+
+	ret = nf_ct_expect_register_notifier(net, &ctnl_notifier_exp);
+	if (ret < 0) {
+		pr_err("ctnetlink_init: cannot expect register notifier.\n");
+		goto err_unreg_notifier;
+	}
+#endif
+	return 0;
+
+#ifdef CONFIG_NF_CONNTRACK_EVENTS
+err_unreg_notifier:
+	nf_conntrack_unregister_notifier(net, &ctnl_notifier);
+err_out:
+	return ret;
+#endif
+}
+
+static void ctnetlink_net_exit(struct net *net)
+{
+#ifdef CONFIG_NF_CONNTRACK_EVENTS
+	nf_ct_expect_unregister_notifier(net, &ctnl_notifier_exp);
+	nf_conntrack_unregister_notifier(net, &ctnl_notifier);
+#endif
+}
+
+static void __net_exit ctnetlink_net_exit_batch(struct list_head *net_exit_list)
+{
+	struct net *net;
+
+	list_for_each_entry(net, net_exit_list, exit_list)
+		ctnetlink_net_exit(net);
+}
+
+static struct pernet_operations ctnetlink_net_ops = {
+	.init		= ctnetlink_net_init,
+	.exit_batch	= ctnetlink_net_exit_batch,
+};
+
 static int __init ctnetlink_init(void)
 {
 	int ret;
@@ -2180,28 +2228,15 @@
 		goto err_unreg_subsys;
 	}
 
-#ifdef CONFIG_NF_CONNTRACK_EVENTS
-	ret = nf_conntrack_register_notifier(&ctnl_notifier);
-	if (ret < 0) {
-		pr_err("ctnetlink_init: cannot register notifier.\n");
+	if (register_pernet_subsys(&ctnetlink_net_ops)) {
+		pr_err("ctnetlink_init: cannot register pernet operations\n");
 		goto err_unreg_exp_subsys;
 	}
 
-	ret = nf_ct_expect_register_notifier(&ctnl_notifier_exp);
-	if (ret < 0) {
-		pr_err("ctnetlink_init: cannot expect register notifier.\n");
-		goto err_unreg_notifier;
-	}
-#endif
-
 	return 0;
 
-#ifdef CONFIG_NF_CONNTRACK_EVENTS
-err_unreg_notifier:
-	nf_conntrack_unregister_notifier(&ctnl_notifier);
 err_unreg_exp_subsys:
 	nfnetlink_subsys_unregister(&ctnl_exp_subsys);
-#endif
 err_unreg_subsys:
 	nfnetlink_subsys_unregister(&ctnl_subsys);
 err_out:
@@ -2213,11 +2248,7 @@
 	pr_info("ctnetlink: unregistering from nfnetlink.\n");
 
 	nf_ct_remove_userspace_expectations();
-#ifdef CONFIG_NF_CONNTRACK_EVENTS
-	nf_ct_expect_unregister_notifier(&ctnl_notifier_exp);
-	nf_conntrack_unregister_notifier(&ctnl_notifier);
-#endif
-
+	unregister_pernet_subsys(&ctnetlink_net_ops);
 	nfnetlink_subsys_unregister(&ctnl_exp_subsys);
 	nfnetlink_subsys_unregister(&ctnl_subsys);
 }
diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c
index 9c24de10..824f184 100644
--- a/net/netlabel/netlabel_kapi.c
+++ b/net/netlabel/netlabel_kapi.c
@@ -111,8 +111,6 @@
 	struct netlbl_domaddr_map *addrmap = NULL;
 	struct netlbl_domaddr4_map *map4 = NULL;
 	struct netlbl_domaddr6_map *map6 = NULL;
-	const struct in_addr *addr4, *mask4;
-	const struct in6_addr *addr6, *mask6;
 
 	entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
 	if (entry == NULL)
@@ -133,9 +131,9 @@
 		INIT_LIST_HEAD(&addrmap->list6);
 
 		switch (family) {
-		case AF_INET:
-			addr4 = addr;
-			mask4 = mask;
+		case AF_INET: {
+			const struct in_addr *addr4 = addr;
+			const struct in_addr *mask4 = mask;
 			map4 = kzalloc(sizeof(*map4), GFP_ATOMIC);
 			if (map4 == NULL)
 				goto cfg_unlbl_map_add_failure;
@@ -148,9 +146,11 @@
 			if (ret_val != 0)
 				goto cfg_unlbl_map_add_failure;
 			break;
-		case AF_INET6:
-			addr6 = addr;
-			mask6 = mask;
+			}
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+		case AF_INET6: {
+			const struct in6_addr *addr6 = addr;
+			const struct in6_addr *mask6 = mask;
 			map6 = kzalloc(sizeof(*map6), GFP_ATOMIC);
 			if (map6 == NULL)
 				goto cfg_unlbl_map_add_failure;
@@ -162,11 +162,13 @@
 			map6->list.addr.s6_addr32[3] &= mask6->s6_addr32[3];
 			ipv6_addr_copy(&map6->list.mask, mask6);
 			map6->list.valid = 1;
-			ret_val = netlbl_af4list_add(&map4->list,
-						     &addrmap->list4);
+			ret_val = netlbl_af6list_add(&map6->list,
+						     &addrmap->list6);
 			if (ret_val != 0)
 				goto cfg_unlbl_map_add_failure;
 			break;
+			}
+#endif /* IPv6 */
 		default:
 			goto cfg_unlbl_map_add_failure;
 			break;
@@ -225,9 +227,11 @@
 	case AF_INET:
 		addr_len = sizeof(struct in_addr);
 		break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 	case AF_INET6:
 		addr_len = sizeof(struct in6_addr);
 		break;
+#endif /* IPv6 */
 	default:
 		return -EPFNOSUPPORT;
 	}
@@ -266,9 +270,11 @@
 	case AF_INET:
 		addr_len = sizeof(struct in_addr);
 		break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 	case AF_INET6:
 		addr_len = sizeof(struct in6_addr);
 		break;
+#endif /* IPv6 */
 	default:
 		return -EPFNOSUPPORT;
 	}
diff --git a/net/rds/Kconfig b/net/rds/Kconfig
index 4cf6dc7..ec753b3 100644
--- a/net/rds/Kconfig
+++ b/net/rds/Kconfig
@@ -9,7 +9,6 @@
 
 config RDS_RDMA
 	tristate "RDS over Infiniband and iWARP"
-	select LLIST
 	depends on RDS && INFINIBAND && INFINIBAND_ADDR_TRANS
 	---help---
 	  Allow RDS to use Infiniband and iWARP as a transport.
diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c
index b9493a0..6cd8ddf 100644
--- a/net/sched/sch_gred.c
+++ b/net/sched/sch_gred.c
@@ -385,7 +385,7 @@
 	struct gred_sched_data *q;
 
 	if (table->tab[dp] == NULL) {
-		table->tab[dp] = kzalloc(sizeof(*q), GFP_KERNEL);
+		table->tab[dp] = kzalloc(sizeof(*q), GFP_ATOMIC);
 		if (table->tab[dp] == NULL)
 			return -ENOMEM;
 	}
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index 6649463..d617161 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -209,8 +209,8 @@
 				 ctl->Plog, ctl->Scell_log,
 				 nla_data(tb[TCA_RED_STAB]));
 
-	if (skb_queue_empty(&sch->q))
-		red_end_of_idle_period(&q->parms);
+	if (!q->qdisc->q.qlen)
+		red_start_of_idle_period(&q->parms);
 
 	sch_tree_unlock(sch);
 	return 0;
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index a3b7120..4f4c52c 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -225,11 +225,11 @@
 
 
 static int
-__teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, struct net_device *dev)
+__teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res,
+	       struct net_device *dev, struct netdev_queue *txq,
+	       struct neighbour *mn)
 {
-	struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, 0);
-	struct teql_sched_data *q = qdisc_priv(dev_queue->qdisc);
-	struct neighbour *mn = dst_get_neighbour(skb_dst(skb));
+	struct teql_sched_data *q = qdisc_priv(txq->qdisc);
 	struct neighbour *n = q->ncache;
 
 	if (mn->tbl == NULL)
@@ -262,17 +262,26 @@
 }
 
 static inline int teql_resolve(struct sk_buff *skb,
-			       struct sk_buff *skb_res, struct net_device *dev)
+			       struct sk_buff *skb_res,
+			       struct net_device *dev,
+			       struct netdev_queue *txq)
 {
-	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
+	struct dst_entry *dst = skb_dst(skb);
+	struct neighbour *mn;
+	int res;
+
 	if (txq->qdisc == &noop_qdisc)
 		return -ENODEV;
 
-	if (dev->header_ops == NULL ||
-	    skb_dst(skb) == NULL ||
-	    dst_get_neighbour(skb_dst(skb)) == NULL)
+	if (!dev->header_ops || !dst)
 		return 0;
-	return __teql_resolve(skb, skb_res, dev);
+
+	rcu_read_lock();
+	mn = dst_get_neighbour(dst);
+	res = mn ? __teql_resolve(skb, skb_res, dev, txq, mn) : 0;
+	rcu_read_unlock();
+
+	return res;
 }
 
 static netdev_tx_t teql_master_xmit(struct sk_buff *skb, struct net_device *dev)
@@ -307,7 +316,7 @@
 			continue;
 		}
 
-		switch (teql_resolve(skb, skb_res, slave)) {
+		switch (teql_resolve(skb, skb_res, slave, slave_txq)) {
 		case 0:
 			if (__netif_tx_trylock(slave_txq)) {
 				unsigned int length = qdisc_pkt_len(skb);
diff --git a/net/sctp/auth.c b/net/sctp/auth.c
index 865e68f..bf81204 100644
--- a/net/sctp/auth.c
+++ b/net/sctp/auth.c
@@ -82,7 +82,7 @@
 	struct sctp_auth_bytes *key;
 
 	/* Verify that we are not going to overflow INT_MAX */
-	if ((INT_MAX - key_len) < sizeof(struct sctp_auth_bytes))
+	if (key_len > (INT_MAX - sizeof(struct sctp_auth_bytes)))
 		return NULL;
 
 	/* Allocate the shared key */
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index d7f97ef..55472c4 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -496,7 +496,7 @@
 	struct rpc_rqst *req = task->tk_rqstp;
 	struct rpc_xprt *xprt = req->rq_xprt;
 	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
-	int ret = 0;
+	int ret = -EAGAIN;
 
 	dprintk("RPC: %5u xmit incomplete (%u left of %u)\n",
 			task->tk_pid, req->rq_slen - req->rq_bytes_sent,
@@ -508,7 +508,6 @@
 	/* Don't race with disconnect */
 	if (xprt_connected(xprt)) {
 		if (test_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags)) {
-			ret = -EAGAIN;
 			/*
 			 * Notify TCP that we're limited by the application
 			 * window size
@@ -2530,8 +2529,10 @@
 		int err;
 		err = xs_init_anyaddr(args->dstaddr->sa_family,
 					(struct sockaddr *)&new->srcaddr);
-		if (err != 0)
+		if (err != 0) {
+			xprt_free(xprt);
 			return ERR_PTR(err);
+		}
 	}
 
 	return xprt;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 466fbcc..b595a3d 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -1957,6 +1957,7 @@
 			if ((UNIXCB(skb).pid  != siocb->scm->pid) ||
 			    (UNIXCB(skb).cred != siocb->scm->cred)) {
 				skb_queue_head(&sk->sk_receive_queue, skb);
+				sk->sk_data_ready(sk, skb->len);
 				break;
 			}
 		} else {
@@ -1974,6 +1975,7 @@
 		chunk = min_t(unsigned int, skb->len, size);
 		if (memcpy_toiovec(msg->msg_iov, skb->data, chunk)) {
 			skb_queue_head(&sk->sk_receive_queue, skb);
+			sk->sk_data_ready(sk, skb->len);
 			if (copied == 0)
 				copied = -EFAULT;
 			break;
@@ -1991,6 +1993,7 @@
 			/* put the skb back if we didn't use it up.. */
 			if (skb->len) {
 				skb_queue_head(&sk->sk_receive_queue, skb);
+				sk->sk_data_ready(sk, skb->len);
 				break;
 			}
 
@@ -2006,6 +2009,7 @@
 
 			/* put message back and return */
 			skb_queue_head(&sk->sk_receive_queue, skb);
+			sk->sk_data_ready(sk, skb->len);
 			break;
 		}
 	} while (size);
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 48260c2..ffafda5 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -89,8 +89,8 @@
 	[NL80211_ATTR_IFINDEX] = { .type = NLA_U32 },
 	[NL80211_ATTR_IFNAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ-1 },
 
-	[NL80211_ATTR_MAC] = { .type = NLA_BINARY, .len = ETH_ALEN },
-	[NL80211_ATTR_PREV_BSSID] = { .type = NLA_BINARY, .len = ETH_ALEN },
+	[NL80211_ATTR_MAC] = { .len = ETH_ALEN },
+	[NL80211_ATTR_PREV_BSSID] = { .len = ETH_ALEN },
 
 	[NL80211_ATTR_KEY] = { .type = NLA_NESTED, },
 	[NL80211_ATTR_KEY_DATA] = { .type = NLA_BINARY,
@@ -132,8 +132,7 @@
 	[NL80211_ATTR_MESH_CONFIG] = { .type = NLA_NESTED },
 	[NL80211_ATTR_SUPPORT_MESH_AUTH] = { .type = NLA_FLAG },
 
-	[NL80211_ATTR_HT_CAPABILITY] = { .type = NLA_BINARY,
-					 .len = NL80211_HT_CAPABILITY_LEN },
+	[NL80211_ATTR_HT_CAPABILITY] = { .len = NL80211_HT_CAPABILITY_LEN },
 
 	[NL80211_ATTR_MGMT_SUBTYPE] = { .type = NLA_U8 },
 	[NL80211_ATTR_IE] = { .type = NLA_BINARY,
@@ -1253,6 +1252,12 @@
 			goto bad_res;
 		}
 
+		if (netdev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
+		    netdev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) {
+			result = -EINVAL;
+			goto bad_res;
+		}
+
 		nla_for_each_nested(nl_txq_params,
 				    info->attrs[NL80211_ATTR_WIPHY_TXQ_PARAMS],
 				    rem_txq_params) {
diff --git a/net/wireless/reg.c b/net/wireless/reg.c
index 6acba9d..3302c56 100644
--- a/net/wireless/reg.c
+++ b/net/wireless/reg.c
@@ -57,8 +57,17 @@
 #define REG_DBG_PRINT(args...)
 #endif
 
+static struct regulatory_request core_request_world = {
+	.initiator = NL80211_REGDOM_SET_BY_CORE,
+	.alpha2[0] = '0',
+	.alpha2[1] = '0',
+	.intersect = false,
+	.processed = true,
+	.country_ie_env = ENVIRON_ANY,
+};
+
 /* Receipt of information from last regulatory request */
-static struct regulatory_request *last_request;
+static struct regulatory_request *last_request = &core_request_world;
 
 /* To trigger userspace events */
 static struct platform_device *reg_pdev;
@@ -150,7 +159,7 @@
 module_param(ieee80211_regdom, charp, 0444);
 MODULE_PARM_DESC(ieee80211_regdom, "IEEE 802.11 regulatory domain code");
 
-static void reset_regdomains(void)
+static void reset_regdomains(bool full_reset)
 {
 	/* avoid freeing static information or freeing something twice */
 	if (cfg80211_regdomain == cfg80211_world_regdom)
@@ -165,6 +174,13 @@
 
 	cfg80211_world_regdom = &world_regdom;
 	cfg80211_regdomain = NULL;
+
+	if (!full_reset)
+		return;
+
+	if (last_request != &core_request_world)
+		kfree(last_request);
+	last_request = &core_request_world;
 }
 
 /*
@@ -175,7 +191,7 @@
 {
 	BUG_ON(!last_request);
 
-	reset_regdomains();
+	reset_regdomains(false);
 
 	cfg80211_world_regdom = rd;
 	cfg80211_regdomain = rd;
@@ -1407,7 +1423,8 @@
 	}
 
 new_request:
-	kfree(last_request);
+	if (last_request != &core_request_world)
+		kfree(last_request);
 
 	last_request = pending_request;
 	last_request->intersect = intersect;
@@ -1577,9 +1594,6 @@
 {
 	struct regulatory_request *request;
 
-	kfree(last_request);
-	last_request = NULL;
-
 	request = kzalloc(sizeof(struct regulatory_request),
 			  GFP_KERNEL);
 	if (!request)
@@ -1777,7 +1791,7 @@
 	mutex_lock(&cfg80211_mutex);
 	mutex_lock(&reg_mutex);
 
-	reset_regdomains();
+	reset_regdomains(true);
 	restore_alpha2(alpha2, reset_user);
 
 	/*
@@ -2037,12 +2051,18 @@
 	}
 
 	request_wiphy = wiphy_idx_to_wiphy(last_request->wiphy_idx);
+	if (!request_wiphy &&
+	    (last_request->initiator == NL80211_REGDOM_SET_BY_DRIVER ||
+	     last_request->initiator == NL80211_REGDOM_SET_BY_COUNTRY_IE)) {
+		schedule_delayed_work(&reg_timeout, 0);
+		return -ENODEV;
+	}
 
 	if (!last_request->intersect) {
 		int r;
 
 		if (last_request->initiator != NL80211_REGDOM_SET_BY_DRIVER) {
-			reset_regdomains();
+			reset_regdomains(false);
 			cfg80211_regdomain = rd;
 			return 0;
 		}
@@ -2063,7 +2083,7 @@
 		if (r)
 			return r;
 
-		reset_regdomains();
+		reset_regdomains(false);
 		cfg80211_regdomain = rd;
 		return 0;
 	}
@@ -2088,7 +2108,7 @@
 
 		rd = NULL;
 
-		reset_regdomains();
+		reset_regdomains(false);
 		cfg80211_regdomain = intersected_rd;
 
 		return 0;
@@ -2108,7 +2128,7 @@
 	kfree(rd);
 	rd = NULL;
 
-	reset_regdomains();
+	reset_regdomains(false);
 	cfg80211_regdomain = intersected_rd;
 
 	return 0;
@@ -2261,9 +2281,9 @@
 	mutex_lock(&cfg80211_mutex);
 	mutex_lock(&reg_mutex);
 
-	reset_regdomains();
+	reset_regdomains(true);
 
-	kfree(last_request);
+	dev_set_uevent_suppress(&reg_pdev->dev, true);
 
 	platform_device_unregister(reg_pdev);
 
diff --git a/net/wireless/scan.c b/net/wireless/scan.c
index 0fb1424..dc23b31 100644
--- a/net/wireless/scan.c
+++ b/net/wireless/scan.c
@@ -259,17 +259,20 @@
 {
 	const u8 *ie1 = cfg80211_find_ie(num, ies1, len1);
 	const u8 *ie2 = cfg80211_find_ie(num, ies2, len2);
-	int r;
 
+	/* equal if both missing */
 	if (!ie1 && !ie2)
 		return 0;
-	if (!ie1 || !ie2)
+	/* sort missing IE before (left of) present IE */
+	if (!ie1)
 		return -1;
+	if (!ie2)
+		return 1;
 
-	r = memcmp(ie1 + 2, ie2 + 2, min(ie1[1], ie2[1]));
-	if (r == 0 && ie1[1] != ie2[1])
+	/* sort by length first, then by contents */
+	if (ie1[1] != ie2[1])
 		return ie2[1] - ie1[1];
-	return r;
+	return memcmp(ie1 + 2, ie2 + 2, ie1[1]);
 }
 
 static bool is_bss(struct cfg80211_bss *a,
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 552df27..2118d64 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -2382,9 +2382,11 @@
 	return dst_metric_advmss(dst->path);
 }
 
-static unsigned int xfrm_default_mtu(const struct dst_entry *dst)
+static unsigned int xfrm_mtu(const struct dst_entry *dst)
 {
-	return dst_mtu(dst->path);
+	unsigned int mtu = dst_metric_raw(dst, RTAX_MTU);
+
+	return mtu ? : dst_mtu(dst->path);
 }
 
 static struct neighbour *xfrm_neigh_lookup(const struct dst_entry *dst, const void *daddr)
@@ -2411,8 +2413,8 @@
 			dst_ops->check = xfrm_dst_check;
 		if (likely(dst_ops->default_advmss == NULL))
 			dst_ops->default_advmss = xfrm_default_advmss;
-		if (likely(dst_ops->default_mtu == NULL))
-			dst_ops->default_mtu = xfrm_default_mtu;
+		if (likely(dst_ops->mtu == NULL))
+			dst_ops->mtu = xfrm_mtu;
 		if (likely(dst_ops->negative_advice == NULL))
 			dst_ops->negative_advice = xfrm_negative_advice;
 		if (likely(dst_ops->link_failure == NULL))
diff --git a/security/apparmor/path.c b/security/apparmor/path.c
index 36cc0cc..b566eba 100644
--- a/security/apparmor/path.c
+++ b/security/apparmor/path.c
@@ -57,23 +57,44 @@
 static int d_namespace_path(struct path *path, char *buf, int buflen,
 			    char **name, int flags)
 {
-	struct path root, tmp;
 	char *res;
-	int connected, error = 0;
+	int error = 0;
+	int connected = 1;
 
-	/* Get the root we want to resolve too, released below */
-	if (flags & PATH_CHROOT_REL) {
-		/* resolve paths relative to chroot */
-		get_fs_root(current->fs, &root);
-	} else {
-		/* resolve paths relative to namespace */
-		root.mnt = current->nsproxy->mnt_ns->root;
-		root.dentry = root.mnt->mnt_root;
-		path_get(&root);
+	if (path->mnt->mnt_flags & MNT_INTERNAL) {
+		/* it's not mounted anywhere */
+		res = dentry_path(path->dentry, buf, buflen);
+		*name = res;
+		if (IS_ERR(res)) {
+			*name = buf;
+			return PTR_ERR(res);
+		}
+		if (path->dentry->d_sb->s_magic == PROC_SUPER_MAGIC &&
+		    strncmp(*name, "/sys/", 5) == 0) {
+			/* TODO: convert over to using a per namespace
+			 * control instead of hard coded /proc
+			 */
+			return prepend(name, *name - buf, "/proc", 5);
+		}
+		return 0;
 	}
 
-	tmp = root;
-	res = __d_path(path, &tmp, buf, buflen);
+	/* resolve paths relative to chroot?*/
+	if (flags & PATH_CHROOT_REL) {
+		struct path root;
+		get_fs_root(current->fs, &root);
+		res = __d_path(path, &root, buf, buflen);
+		if (res && !IS_ERR(res)) {
+			/* everything's fine */
+			*name = res;
+			path_put(&root);
+			goto ok;
+		}
+		path_put(&root);
+		connected = 0;
+	}
+
+	res = d_absolute_path(path, buf, buflen);
 
 	*name = res;
 	/* handle error conditions - and still allow a partial path to
@@ -84,7 +105,10 @@
 		*name = buf;
 		goto out;
 	}
+	if (!our_mnt(path->mnt))
+		connected = 0;
 
+ok:
 	/* Handle two cases:
 	 * 1. A deleted dentry && profile is not allowing mediation of deleted
 	 * 2. On some filesystems, newly allocated dentries appear to the
@@ -97,10 +121,7 @@
 			goto out;
 	}
 
-	/* Determine if the path is connected to the expected root */
-	connected = tmp.dentry == root.dentry && tmp.mnt == root.mnt;
-
-	/* If the path is not connected,
+	/* If the path is not connected to the expected root,
 	 * check if it is a sysctl and handle specially else remove any
 	 * leading / that __d_path may have returned.
 	 * Unless
@@ -112,17 +133,9 @@
 	 *     namespace root.
 	 */
 	if (!connected) {
-		/* is the disconnect path a sysctl? */
-		if (tmp.dentry->d_sb->s_magic == PROC_SUPER_MAGIC &&
-		    strncmp(*name, "/sys/", 5) == 0) {
-			/* TODO: convert over to using a per namespace
-			 * control instead of hard coded /proc
-			 */
-			error = prepend(name, *name - buf, "/proc", 5);
-		} else if (!(flags & PATH_CONNECT_PATH) &&
+		if (!(flags & PATH_CONNECT_PATH) &&
 			   !(((flags & CHROOT_NSCONNECT) == CHROOT_NSCONNECT) &&
-			     (tmp.mnt == current->nsproxy->mnt_ns->root &&
-			      tmp.dentry == tmp.mnt->mnt_root))) {
+			     our_mnt(path->mnt))) {
 			/* disconnected path, don't return pathname starting
 			 * with '/'
 			 */
@@ -133,8 +146,6 @@
 	}
 
 out:
-	path_put(&root);
-
 	return error;
 }
 
diff --git a/security/keys/encrypted-keys/Makefile b/security/keys/encrypted-keys/Makefile
index 6bc7a86..d6f8433 100644
--- a/security/keys/encrypted-keys/Makefile
+++ b/security/keys/encrypted-keys/Makefile
@@ -2,5 +2,9 @@
 # Makefile for encrypted keys
 #
 
-obj-$(CONFIG_ENCRYPTED_KEYS) += encrypted.o ecryptfs_format.o
-obj-$(CONFIG_TRUSTED_KEYS) += masterkey_trusted.o
+obj-$(CONFIG_ENCRYPTED_KEYS) += encrypted-keys.o
+
+encrypted-keys-y := encrypted.o ecryptfs_format.o
+masterkey-$(CONFIG_TRUSTED_KEYS) := masterkey_trusted.o
+masterkey-$(CONFIG_TRUSTED_KEYS)-$(CONFIG_ENCRYPTED_KEYS) := masterkey_trusted.o
+encrypted-keys-y += $(masterkey-y) $(masterkey-m-m)
diff --git a/security/keys/encrypted-keys/encrypted.c b/security/keys/encrypted-keys/encrypted.c
index dcc843c..41144f7 100644
--- a/security/keys/encrypted-keys/encrypted.c
+++ b/security/keys/encrypted-keys/encrypted.c
@@ -444,7 +444,7 @@
 		goto out;
 
 	if (IS_ERR(mkey)) {
-		int ret = PTR_ERR(epayload);
+		int ret = PTR_ERR(mkey);
 
 		if (ret == -ENOTSUPP)
 			pr_info("encrypted_key: key %s not supported",
diff --git a/security/keys/encrypted-keys/encrypted.h b/security/keys/encrypted-keys/encrypted.h
index b6ade89..8136a2d 100644
--- a/security/keys/encrypted-keys/encrypted.h
+++ b/security/keys/encrypted-keys/encrypted.h
@@ -2,7 +2,8 @@
 #define __ENCRYPTED_KEY_H
 
 #define ENCRYPTED_DEBUG 0
-#ifdef CONFIG_TRUSTED_KEYS
+#if defined(CONFIG_TRUSTED_KEYS) || \
+  (defined(CONFIG_TRUSTED_KEYS_MODULE) && defined(CONFIG_ENCRYPTED_KEYS_MODULE))
 extern struct key *request_trusted_key(const char *trusted_desc,
 				       u8 **master_key, size_t *master_keylen);
 #else
diff --git a/security/keys/user_defined.c b/security/keys/user_defined.c
index 5b366d7..69ff52c 100644
--- a/security/keys/user_defined.c
+++ b/security/keys/user_defined.c
@@ -102,7 +102,8 @@
 		key->expiry = 0;
 	}
 
-	kfree_rcu(zap, rcu);
+	if (zap)
+		kfree_rcu(zap, rcu);
 
 error:
 	return ret;
diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c
index 6aceef5..5c32f36 100644
--- a/security/smack/smackfs.c
+++ b/security/smack/smackfs.c
@@ -102,9 +102,6 @@
 
 const char *smack_cipso_option = SMACK_CIPSO_OPTION;
 
-
-#define	SEQ_READ_FINISHED	((loff_t)-1)
-
 /*
  * Values for parsing cipso rules
  * SMK_DIGITLEN: Length of a digit field in a rule.
@@ -357,10 +354,12 @@
 
 	rc = count;
 	/*
+	 * If this is "load" as opposed to "load-self" and a new rule
+	 * it needs to get added for reporting.
 	 * smk_set_access returns true if there was already a rule
 	 * for the subject/object pair, and false if it was new.
 	 */
-	if (!smk_set_access(rule, rule_list, rule_lock)) {
+	if (load && !smk_set_access(rule, rule_list, rule_lock)) {
 		smlp = kzalloc(sizeof(*smlp), GFP_KERNEL);
 		if (smlp != NULL) {
 			smlp->smk_rule = rule;
@@ -377,12 +376,12 @@
 	return rc;
 }
 
-
 /*
- * Seq_file read operations for /smack/load
+ * Core logic for smackfs seq list operations.
  */
 
-static void *load_seq_start(struct seq_file *s, loff_t *pos)
+static void *smk_seq_start(struct seq_file *s, loff_t *pos,
+				struct list_head *head)
 {
 	struct list_head *list;
 
@@ -390,7 +389,7 @@
 	 * This is 0 the first time through.
 	 */
 	if (s->index == 0)
-		s->private = &smack_rule_list;
+		s->private = head;
 
 	if (s->private == NULL)
 		return NULL;
@@ -404,11 +403,12 @@
 	return list;
 }
 
-static void *load_seq_next(struct seq_file *s, void *v, loff_t *pos)
+static void *smk_seq_next(struct seq_file *s, void *v, loff_t *pos,
+				struct list_head *head)
 {
 	struct list_head *list = v;
 
-	if (list_is_last(list, &smack_rule_list)) {
+	if (list_is_last(list, head)) {
 		s->private = NULL;
 		return NULL;
 	}
@@ -416,6 +416,25 @@
 	return list->next;
 }
 
+static void smk_seq_stop(struct seq_file *s, void *v)
+{
+	/* No-op */
+}
+
+/*
+ * Seq_file read operations for /smack/load
+ */
+
+static void *load_seq_start(struct seq_file *s, loff_t *pos)
+{
+	return smk_seq_start(s, pos, &smack_rule_list);
+}
+
+static void *load_seq_next(struct seq_file *s, void *v, loff_t *pos)
+{
+	return smk_seq_next(s, v, pos, &smack_rule_list);
+}
+
 static int load_seq_show(struct seq_file *s, void *v)
 {
 	struct list_head *list = v;
@@ -446,16 +465,11 @@
 	return 0;
 }
 
-static void load_seq_stop(struct seq_file *s, void *v)
-{
-	/* No-op */
-}
-
 static const struct seq_operations load_seq_ops = {
 	.start = load_seq_start,
 	.next  = load_seq_next,
 	.show  = load_seq_show,
-	.stop  = load_seq_stop,
+	.stop  = smk_seq_stop,
 };
 
 /**
@@ -574,28 +588,12 @@
 
 static void *cipso_seq_start(struct seq_file *s, loff_t *pos)
 {
-	if (*pos == SEQ_READ_FINISHED)
-		return NULL;
-	if (list_empty(&smack_known_list))
-		return NULL;
-
-	return smack_known_list.next;
+	return smk_seq_start(s, pos, &smack_known_list);
 }
 
 static void *cipso_seq_next(struct seq_file *s, void *v, loff_t *pos)
 {
-	struct list_head  *list = v;
-
-	/*
-	 * labels with no associated cipso value wont be printed
-	 * in cipso_seq_show
-	 */
-	if (list_is_last(list, &smack_known_list)) {
-		*pos = SEQ_READ_FINISHED;
-		return NULL;
-	}
-
-	return list->next;
+	return smk_seq_next(s, v, pos, &smack_known_list);
 }
 
 /*
@@ -634,16 +632,11 @@
 	return 0;
 }
 
-static void cipso_seq_stop(struct seq_file *s, void *v)
-{
-	/* No-op */
-}
-
 static const struct seq_operations cipso_seq_ops = {
 	.start = cipso_seq_start,
-	.stop  = cipso_seq_stop,
 	.next  = cipso_seq_next,
 	.show  = cipso_seq_show,
+	.stop  = smk_seq_stop,
 };
 
 /**
@@ -788,23 +781,12 @@
 
 static void *netlbladdr_seq_start(struct seq_file *s, loff_t *pos)
 {
-	if (*pos == SEQ_READ_FINISHED)
-		return NULL;
-	if (list_empty(&smk_netlbladdr_list))
-		return NULL;
-	return smk_netlbladdr_list.next;
+	return smk_seq_start(s, pos, &smk_netlbladdr_list);
 }
 
 static void *netlbladdr_seq_next(struct seq_file *s, void *v, loff_t *pos)
 {
-	struct list_head *list = v;
-
-	if (list_is_last(list, &smk_netlbladdr_list)) {
-		*pos = SEQ_READ_FINISHED;
-		return NULL;
-	}
-
-	return list->next;
+	return smk_seq_next(s, v, pos, &smk_netlbladdr_list);
 }
 #define BEBITS	(sizeof(__be32) * 8)
 
@@ -828,16 +810,11 @@
 	return 0;
 }
 
-static void netlbladdr_seq_stop(struct seq_file *s, void *v)
-{
-	/* No-op */
-}
-
 static const struct seq_operations netlbladdr_seq_ops = {
 	.start = netlbladdr_seq_start,
-	.stop  = netlbladdr_seq_stop,
 	.next  = netlbladdr_seq_next,
 	.show  = netlbladdr_seq_show,
+	.stop  = smk_seq_stop,
 };
 
 /**
@@ -1405,23 +1382,14 @@
 {
 	struct task_smack *tsp = current_security();
 
-	if (*pos == SEQ_READ_FINISHED)
-		return NULL;
-	if (list_empty(&tsp->smk_rules))
-		return NULL;
-	return tsp->smk_rules.next;
+	return smk_seq_start(s, pos, &tsp->smk_rules);
 }
 
 static void *load_self_seq_next(struct seq_file *s, void *v, loff_t *pos)
 {
 	struct task_smack *tsp = current_security();
-	struct list_head *list = v;
 
-	if (list_is_last(list, &tsp->smk_rules)) {
-		*pos = SEQ_READ_FINISHED;
-		return NULL;
-	}
-	return list->next;
+	return smk_seq_next(s, v, pos, &tsp->smk_rules);
 }
 
 static int load_self_seq_show(struct seq_file *s, void *v)
@@ -1453,16 +1421,11 @@
 	return 0;
 }
 
-static void load_self_seq_stop(struct seq_file *s, void *v)
-{
-	/* No-op */
-}
-
 static const struct seq_operations load_self_seq_ops = {
 	.start = load_self_seq_start,
 	.next  = load_self_seq_next,
 	.show  = load_self_seq_show,
-	.stop  = load_self_seq_stop,
+	.stop  = smk_seq_stop,
 };
 
 
diff --git a/security/tomoyo/realpath.c b/security/tomoyo/realpath.c
index 738bbdf..d9f3ced 100644
--- a/security/tomoyo/realpath.c
+++ b/security/tomoyo/realpath.c
@@ -101,9 +101,8 @@
 {
 	char *pos = ERR_PTR(-ENOMEM);
 	if (buflen >= 256) {
-		struct path ns_root = { };
 		/* go to whatever namespace root we are under */
-		pos = __d_path(path, &ns_root, buffer, buflen - 1);
+		pos = d_absolute_path(path, buffer, buflen - 1);
 		if (!IS_ERR(pos) && *pos == '/' && pos[1]) {
 			struct inode *inode = path->dentry->d_inode;
 			if (inode && S_ISDIR(inode->i_mode)) {
@@ -294,8 +293,16 @@
 			pos = tomoyo_get_local_path(path->dentry, buf,
 						    buf_len - 1);
 		/* Get absolute name for the rest. */
-		else
+		else {
 			pos = tomoyo_get_absolute_path(path, buf, buf_len - 1);
+			/*
+			 * Fall back to local name if absolute name is not
+			 * available.
+			 */
+			if (pos == ERR_PTR(-EINVAL))
+				pos = tomoyo_get_local_path(path->dentry, buf,
+							    buf_len - 1);
+		}
 encode:
 		if (IS_ERR(pos))
 			continue;
diff --git a/sound/core/vmaster.c b/sound/core/vmaster.c
index 5dbab38..130cfe6 100644
--- a/sound/core/vmaster.c
+++ b/sound/core/vmaster.c
@@ -52,6 +52,7 @@
 	struct link_ctl_info info;
 	int vals[2];		/* current values */
 	unsigned int flags;
+	struct snd_kcontrol *kctl; /* original kcontrol pointer */
 	struct snd_kcontrol slave; /* the copy of original control entry */
 };
 
@@ -252,6 +253,7 @@
 		       slave->count * sizeof(*slave->vd), GFP_KERNEL);
 	if (!srec)
 		return -ENOMEM;
+	srec->kctl = slave;
 	srec->slave = *slave;
 	memcpy(srec->slave.vd, slave->vd, slave->count * sizeof(*slave->vd));
 	srec->master = master_link;
@@ -333,10 +335,18 @@
 static void master_free(struct snd_kcontrol *kcontrol)
 {
 	struct link_master *master = snd_kcontrol_chip(kcontrol);
-	struct link_slave *slave;
+	struct link_slave *slave, *n;
 
-	list_for_each_entry(slave, &master->slaves, list)
-		slave->master = NULL;
+	/* free all slave links and retore the original slave kctls */
+	list_for_each_entry_safe(slave, n, &master->slaves, list) {
+		struct snd_kcontrol *sctl = slave->kctl;
+		struct list_head olist = sctl->list;
+		memcpy(sctl, &slave->slave, sizeof(*sctl));
+		memcpy(sctl->vd, slave->slave.vd,
+		       sctl->count * sizeof(*sctl->vd));
+		sctl->list = olist; /* keep the current linked-list */
+		kfree(slave);
+	}
 	kfree(master);
 }
 
diff --git a/sound/pci/cs5535audio/cs5535audio_pcm.c b/sound/pci/cs5535audio/cs5535audio_pcm.c
index e083122..dbf94b1 100644
--- a/sound/pci/cs5535audio/cs5535audio_pcm.c
+++ b/sound/pci/cs5535audio/cs5535audio_pcm.c
@@ -148,7 +148,7 @@
 		struct cs5535audio_dma_desc *desc =
 			&((struct cs5535audio_dma_desc *) dma->desc_buf.area)[i];
 		desc->addr = cpu_to_le32(addr);
-		desc->size = cpu_to_le32(period_bytes);
+		desc->size = cpu_to_le16(period_bytes);
 		desc->ctlreserved = cpu_to_le16(PRD_EOP);
 		desc_addr += sizeof(struct cs5535audio_dma_desc);
 		addr += period_bytes;
diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c
index 916a186..4562e9d 100644
--- a/sound/pci/hda/hda_codec.c
+++ b/sound/pci/hda/hda_codec.c
@@ -2331,6 +2331,39 @@
 	return 0;
 }
 
+typedef int (*map_slave_func_t)(void *, struct snd_kcontrol *);
+
+/* apply the function to all matching slave ctls in the mixer list */
+static int map_slaves(struct hda_codec *codec, const char * const *slaves,
+		      map_slave_func_t func, void *data) 
+{
+	struct hda_nid_item *items;
+	const char * const *s;
+	int i, err;
+
+	items = codec->mixers.list;
+	for (i = 0; i < codec->mixers.used; i++) {
+		struct snd_kcontrol *sctl = items[i].kctl;
+		if (!sctl || !sctl->id.name ||
+		    sctl->id.iface != SNDRV_CTL_ELEM_IFACE_MIXER)
+			continue;
+		for (s = slaves; *s; s++) {
+			if (!strcmp(sctl->id.name, *s)) {
+				err = func(data, sctl);
+				if (err)
+					return err;
+				break;
+			}
+		}
+	}
+	return 0;
+}
+
+static int check_slave_present(void *data, struct snd_kcontrol *sctl)
+{
+	return 1;
+}
+
 /**
  * snd_hda_add_vmaster - create a virtual master control and add slaves
  * @codec: HD-audio codec
@@ -2351,12 +2384,10 @@
 			unsigned int *tlv, const char * const *slaves)
 {
 	struct snd_kcontrol *kctl;
-	const char * const *s;
 	int err;
 
-	for (s = slaves; *s && !snd_hda_find_mixer_ctl(codec, *s); s++)
-		;
-	if (!*s) {
+	err = map_slaves(codec, slaves, check_slave_present, NULL);
+	if (err != 1) {
 		snd_printdd("No slave found for %s\n", name);
 		return 0;
 	}
@@ -2367,23 +2398,10 @@
 	if (err < 0)
 		return err;
 
-	for (s = slaves; *s; s++) {
-		struct snd_kcontrol *sctl;
-		int i = 0;
-		for (;;) {
-			sctl = _snd_hda_find_mixer_ctl(codec, *s, i);
-			if (!sctl) {
-				if (!i)
-					snd_printdd("Cannot find slave %s, "
-						    "skipped\n", *s);
-				break;
-			}
-			err = snd_ctl_add_slave(kctl, sctl);
-			if (err < 0)
-				return err;
-			i++;
-		}
-	}
+	err = map_slaves(codec, slaves, (map_slave_func_t)snd_ctl_add_slave,
+			 kctl);
+	if (err < 0)
+		return err;
 	return 0;
 }
 EXPORT_SYMBOL_HDA(snd_hda_add_vmaster);
@@ -4028,9 +4046,9 @@
 
 	/* Search for codec ID */
 	for (q = tbl; q->subvendor; q++) {
-		unsigned long vendorid = (q->subdevice) | (q->subvendor << 16);
-
-		if (vendorid == codec->subsystem_id)
+		unsigned int mask = 0xffff0000 | q->subdevice_mask;
+		unsigned int id = (q->subdevice | (q->subvendor << 16)) & mask;
+		if ((codec->subsystem_id & mask) == id)
 			break;
 	}
 
@@ -4752,6 +4770,7 @@
 	memset(sequences_hp, 0, sizeof(sequences_hp));
 	assoc_line_out = 0;
 
+	codec->ignore_misc_bit = true;
 	end_nid = codec->start_nid + codec->num_nodes;
 	for (nid = codec->start_nid; nid < end_nid; nid++) {
 		unsigned int wid_caps = get_wcaps(codec, nid);
@@ -4767,6 +4786,9 @@
 			continue;
 
 		def_conf = snd_hda_codec_get_pincfg(codec, nid);
+		if (!(get_defcfg_misc(snd_hda_codec_get_pincfg(codec, nid)) &
+		      AC_DEFCFG_MISC_NO_PRESENCE))
+			codec->ignore_misc_bit = false;
 		conn = get_defcfg_connect(def_conf);
 		if (conn == AC_JACK_PORT_NONE)
 			continue;
diff --git a/sound/pci/hda/hda_codec.h b/sound/pci/hda/hda_codec.h
index 755f2b0..5644711 100644
--- a/sound/pci/hda/hda_codec.h
+++ b/sound/pci/hda/hda_codec.h
@@ -854,6 +854,7 @@
 	unsigned int no_sticky_stream:1; /* no sticky-PCM stream assignment */
 	unsigned int pins_shutup:1;	/* pins are shut up */
 	unsigned int no_trigger_sense:1; /* don't trigger at pin-sensing */
+	unsigned int ignore_misc_bit:1; /* ignore MISC_NO_PRESENCE bit */
 #ifdef CONFIG_SND_HDA_POWER_SAVE
 	unsigned int power_on :1;	/* current (global) power-state */
 	unsigned int power_transition :1; /* power-state in transition */
diff --git a/sound/pci/hda/hda_eld.c b/sound/pci/hda/hda_eld.c
index 1c8ddf5..c1da422 100644
--- a/sound/pci/hda/hda_eld.c
+++ b/sound/pci/hda/hda_eld.c
@@ -297,10 +297,18 @@
 					buf + ELD_FIXED_BYTES + mnl + 3 * i);
 	}
 
+	/*
+	 * HDMI sink's ELD info cannot always be retrieved for now, e.g.
+	 * in console or for audio devices. Assume the highest speakers
+	 * configuration, to _not_ prohibit multi-channel audio playback.
+	 */
+	if (!e->spk_alloc)
+		e->spk_alloc = 0xffff;
+
+	e->eld_valid = true;
 	return 0;
 
 out_fail:
-	e->eld_ver = 0;
 	return -EINVAL;
 }
 
@@ -323,9 +331,6 @@
 	 * ELD is valid, actual eld_size is assigned in hdmi_update_eld()
 	 */
 
-	if (!eld->eld_valid)
-		return -ENOENT;
-
 	size = snd_hdmi_get_eld_size(codec, nid);
 	if (size == 0) {
 		/* wfg: workaround for ASUS P5E-VM HDMI board */
@@ -342,18 +347,28 @@
 
 	for (i = 0; i < size; i++) {
 		unsigned int val = hdmi_get_eld_data(codec, nid, i);
+		/*
+		 * Graphics driver might be writing to ELD buffer right now.
+		 * Just abort. The caller will repoll after a while.
+		 */
 		if (!(val & AC_ELDD_ELD_VALID)) {
-			if (!i) {
-				snd_printd(KERN_INFO
-					   "HDMI: invalid ELD data\n");
-				ret = -EINVAL;
-				goto error;
-			}
 			snd_printd(KERN_INFO
 				  "HDMI: invalid ELD data byte %d\n", i);
-			val = 0;
-		} else
-			val &= AC_ELDD_ELD_DATA;
+			ret = -EINVAL;
+			goto error;
+		}
+		val &= AC_ELDD_ELD_DATA;
+		/*
+		 * The first byte cannot be zero. This can happen on some DVI
+		 * connections. Some Intel chips may also need some 250ms delay
+		 * to return non-zero ELD data, even when the graphics driver
+		 * correctly writes ELD content before setting ELD_valid bit.
+		 */
+		if (!val && !i) {
+			snd_printdd(KERN_INFO "HDMI: 0 ELD data\n");
+			ret = -EINVAL;
+			goto error;
+		}
 		buf[i] = val;
 	}
 
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index 096507d..c2f79e6 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -2507,8 +2507,8 @@
 	SND_PCI_QUIRK(0x1043, 0x813d, "ASUS P5AD2", POS_FIX_LPIB),
 	SND_PCI_QUIRK(0x1043, 0x81b3, "ASUS", POS_FIX_LPIB),
 	SND_PCI_QUIRK(0x1043, 0x81e7, "ASUS M2V", POS_FIX_LPIB),
+	SND_PCI_QUIRK(0x1043, 0x83ce, "ASUS 1101HA", POS_FIX_LPIB),
 	SND_PCI_QUIRK(0x104d, 0x9069, "Sony VPCS11V9E", POS_FIX_LPIB),
-	SND_PCI_QUIRK(0x1106, 0x3288, "ASUS M2V-MX SE", POS_FIX_LPIB),
 	SND_PCI_QUIRK(0x1297, 0x3166, "Shuttle", POS_FIX_LPIB),
 	SND_PCI_QUIRK(0x1458, 0xa022, "ga-ma770-ud3", POS_FIX_LPIB),
 	SND_PCI_QUIRK(0x1462, 0x1002, "MSI Wind U115", POS_FIX_LPIB),
@@ -2971,7 +2971,8 @@
 	/* SCH */
 	{ PCI_DEVICE(0x8086, 0x811b),
 	  .driver_data = AZX_DRIVER_SCH | AZX_DCAPS_SCH_SNOOP |
-	  AZX_DCAPS_BUFSIZE},
+	  AZX_DCAPS_BUFSIZE | AZX_DCAPS_POSFIX_LPIB }, /* Poulsbo */
+	/* ICH */
 	{ PCI_DEVICE(0x8086, 0x2668),
 	  .driver_data = AZX_DRIVER_ICH | AZX_DCAPS_OLD_SSYNC |
 	  AZX_DCAPS_BUFSIZE },  /* ICH6 */
diff --git a/sound/pci/hda/hda_local.h b/sound/pci/hda/hda_local.h
index dcbea0d..618ddad 100644
--- a/sound/pci/hda/hda_local.h
+++ b/sound/pci/hda/hda_local.h
@@ -510,13 +510,15 @@
 
 static inline bool is_jack_detectable(struct hda_codec *codec, hda_nid_t nid)
 {
-	return (snd_hda_query_pin_caps(codec, nid) & AC_PINCAP_PRES_DETECT) &&
-		/* disable MISC_NO_PRESENCE check because it may break too
-		 * many devices
-		 */
-		/*(get_defcfg_misc(snd_hda_codec_get_pincfg(codec, nid) &
-		  AC_DEFCFG_MISC_NO_PRESENCE)) &&*/
-		(get_wcaps(codec, nid) & AC_WCAP_UNSOL_CAP);
+	if (!(snd_hda_query_pin_caps(codec, nid) & AC_PINCAP_PRES_DETECT))
+		return false;
+	if (!codec->ignore_misc_bit &&
+	    (get_defcfg_misc(snd_hda_codec_get_pincfg(codec, nid)) &
+	     AC_DEFCFG_MISC_NO_PRESENCE))
+		return false;
+	if (!(get_wcaps(codec, nid) & AC_WCAP_UNSOL_CAP))
+		return false;
+	return true;
 }
 
 /* flags for hda_nid_item */
@@ -651,6 +653,9 @@
 	int	spk_alloc;
 	int	sad_count;
 	struct cea_sad sad[ELD_MAX_SAD];
+	/*
+	 * all fields above eld_buffer will be cleared before updating ELD
+	 */
 	char    eld_buffer[ELD_MAX_SIZE];
 #ifdef CONFIG_PROC_FS
 	struct snd_info_entry *proc_entry;
diff --git a/sound/pci/hda/patch_cirrus.c b/sound/pci/hda/patch_cirrus.c
index 2a2d864..70a7abd 100644
--- a/sound/pci/hda/patch_cirrus.c
+++ b/sound/pci/hda/patch_cirrus.c
@@ -58,6 +58,8 @@
 	unsigned int gpio_mask;
 	unsigned int gpio_dir;
 	unsigned int gpio_data;
+	unsigned int gpio_eapd_hp; /* EAPD GPIO bit for headphones */
+	unsigned int gpio_eapd_speaker; /* EAPD GPIO bit for speakers */
 
 	struct hda_pcm pcm_rec[2];	/* PCM information */
 
@@ -76,6 +78,7 @@
 	CS420X_MBP53,
 	CS420X_MBP55,
 	CS420X_IMAC27,
+	CS420X_APPLE,
 	CS420X_AUTO,
 	CS420X_MODELS
 };
@@ -237,6 +240,15 @@
 	return snd_hda_multi_out_dig_cleanup(codec, &spec->multiout);
 }
 
+static void cs_update_input_select(struct hda_codec *codec)
+{
+	struct cs_spec *spec = codec->spec;
+	if (spec->cur_adc)
+		snd_hda_codec_write(codec, spec->cur_adc, 0,
+				    AC_VERB_SET_CONNECT_SEL,
+				    spec->adc_idx[spec->cur_input]);
+}
+
 /*
  * Analog capture
  */
@@ -250,6 +262,7 @@
 	spec->cur_adc = spec->adc_nid[spec->cur_input];
 	spec->cur_adc_stream_tag = stream_tag;
 	spec->cur_adc_format = format;
+	cs_update_input_select(codec);
 	snd_hda_codec_setup_stream(codec, spec->cur_adc, stream_tag, 0, format);
 	return 0;
 }
@@ -689,10 +702,8 @@
 					   spec->cur_adc_stream_tag, 0,
 					   spec->cur_adc_format);
 	}
-	snd_hda_codec_write(codec, spec->cur_adc, 0,
-			    AC_VERB_SET_CONNECT_SEL,
-			    spec->adc_idx[idx]);
 	spec->cur_input = idx;
+	cs_update_input_select(codec);
 	return 1;
 }
 
@@ -920,10 +931,9 @@
 					spdif_present ? 0 : PIN_OUT);
 		}
 	}
-	if (spec->board_config == CS420X_MBP53 ||
-	    spec->board_config == CS420X_MBP55 ||
-	    spec->board_config == CS420X_IMAC27) {
-		unsigned int gpio = hp_present ? 0x02 : 0x08;
+	if (spec->gpio_eapd_hp) {
+		unsigned int gpio = hp_present ?
+			spec->gpio_eapd_hp : spec->gpio_eapd_speaker;
 		snd_hda_codec_write(codec, 0x01, 0,
 				    AC_VERB_SET_GPIO_DATA, gpio);
 	}
@@ -973,10 +983,7 @@
 		} else  {
 			spec->cur_input = spec->last_input;
 		}
-
-		snd_hda_codec_write_cache(codec, spec->cur_adc, 0,
-					AC_VERB_SET_CONNECT_SEL,
-					spec->adc_idx[spec->cur_input]);
+		cs_update_input_select(codec);
 	} else {
 		if (present)
 			change_cur_input(codec, spec->automic_idx, 0);
@@ -1073,9 +1080,7 @@
 			cs_automic(codec);
 		else  {
 			spec->cur_adc = spec->adc_nid[spec->cur_input];
-			snd_hda_codec_write(codec, spec->cur_adc, 0,
-					AC_VERB_SET_CONNECT_SEL,
-					spec->adc_idx[spec->cur_input]);
+			cs_update_input_select(codec);
 		}
 	} else {
 		change_cur_input(codec, spec->cur_input, 1);
@@ -1273,6 +1278,7 @@
 	[CS420X_MBP53] = "mbp53",
 	[CS420X_MBP55] = "mbp55",
 	[CS420X_IMAC27] = "imac27",
+	[CS420X_APPLE] = "apple",
 	[CS420X_AUTO] = "auto",
 };
 
@@ -1282,7 +1288,13 @@
 	SND_PCI_QUIRK(0x10de, 0x0d94, "MacBookAir 3,1(2)", CS420X_MBP55),
 	SND_PCI_QUIRK(0x10de, 0xcb79, "MacBookPro 5,5", CS420X_MBP55),
 	SND_PCI_QUIRK(0x10de, 0xcb89, "MacBookPro 7,1", CS420X_MBP55),
-	SND_PCI_QUIRK(0x8086, 0x7270, "IMac 27 Inch", CS420X_IMAC27),
+	/* this conflicts with too many other models */
+	/*SND_PCI_QUIRK(0x8086, 0x7270, "IMac 27 Inch", CS420X_IMAC27),*/
+	{} /* terminator */
+};
+
+static const struct snd_pci_quirk cs420x_codec_cfg_tbl[] = {
+	SND_PCI_QUIRK_VENDOR(0x106b, "Apple", CS420X_APPLE),
 	{} /* terminator */
 };
 
@@ -1364,6 +1376,10 @@
 	spec->board_config =
 		snd_hda_check_board_config(codec, CS420X_MODELS,
 					   cs420x_models, cs420x_cfg_tbl);
+	if (spec->board_config < 0)
+		spec->board_config =
+			snd_hda_check_board_codec_sid_config(codec,
+				CS420X_MODELS, NULL, cs420x_codec_cfg_tbl);
 	if (spec->board_config >= 0)
 		fix_pincfg(codec, spec->board_config, cs_pincfgs);
 
@@ -1371,10 +1387,11 @@
 	case CS420X_IMAC27:
 	case CS420X_MBP53:
 	case CS420X_MBP55:
-		/* GPIO1 = headphones */
-		/* GPIO3 = speakers */
-		spec->gpio_mask = 0x0a;
-		spec->gpio_dir = 0x0a;
+	case CS420X_APPLE:
+		spec->gpio_eapd_hp = 2; /* GPIO1 = headphones */
+		spec->gpio_eapd_speaker = 8; /* GPIO3 = speakers */
+		spec->gpio_mask = spec->gpio_dir =
+			spec->gpio_eapd_hp | spec->gpio_eapd_speaker;
 		break;
 	}
 
diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c
index 5e706e4..0de2119 100644
--- a/sound/pci/hda/patch_conexant.c
+++ b/sound/pci/hda/patch_conexant.c
@@ -3062,7 +3062,6 @@
 	SND_PCI_QUIRK(0x1043, 0x1993, "Asus U50F", CXT5066_ASUS),
 	SND_PCI_QUIRK(0x1179, 0xff1e, "Toshiba Satellite C650D", CXT5066_IDEAPAD),
 	SND_PCI_QUIRK(0x1179, 0xff50, "Toshiba Satellite P500-PSPGSC-01800T", CXT5066_OLPC_XO_1_5),
-	SND_PCI_QUIRK(0x1179, 0xffe0, "Toshiba Satellite Pro T130-15F", CXT5066_OLPC_XO_1_5),
 	SND_PCI_QUIRK(0x14f1, 0x0101, "Conexant Reference board",
 		      CXT5066_LAPTOP),
 	SND_PCI_QUIRK(0x152d, 0x0833, "OLPC XO-1.5", CXT5066_OLPC_XO_1_5),
diff --git a/sound/pci/hda/patch_hdmi.c b/sound/pci/hda/patch_hdmi.c
index 81b7b79..c505fd5 100644
--- a/sound/pci/hda/patch_hdmi.c
+++ b/sound/pci/hda/patch_hdmi.c
@@ -65,7 +65,11 @@
 	hda_nid_t pin_nid;
 	int num_mux_nids;
 	hda_nid_t mux_nids[HDA_MAX_CONNECTIONS];
+
+	struct hda_codec *codec;
 	struct hdmi_eld sink_eld;
+	struct delayed_work work;
+	int repoll_count;
 };
 
 struct hdmi_spec {
@@ -745,8 +749,7 @@
  * Unsolicited events
  */
 
-static void hdmi_present_sense(struct hda_codec *codec, hda_nid_t pin_nid,
-			       struct hdmi_eld *eld);
+static void hdmi_present_sense(struct hdmi_spec_per_pin *per_pin, int repoll);
 
 static void hdmi_intrinsic_event(struct hda_codec *codec, unsigned int res)
 {
@@ -755,7 +758,6 @@
 	int pd = !!(res & AC_UNSOL_RES_PD);
 	int eldv = !!(res & AC_UNSOL_RES_ELDV);
 	int pin_idx;
-	struct hdmi_eld *eld;
 
 	printk(KERN_INFO
 		"HDMI hot plug event: Codec=%d Pin=%d Presence_Detect=%d ELD_Valid=%d\n",
@@ -764,17 +766,8 @@
 	pin_idx = pin_nid_to_pin_index(spec, pin_nid);
 	if (pin_idx < 0)
 		return;
-	eld = &spec->pins[pin_idx].sink_eld;
 
-	hdmi_present_sense(codec, pin_nid, eld);
-
-	/*
-	 * HDMI sink's ELD info cannot always be retrieved for now, e.g.
-	 * in console or for audio devices. Assume the highest speakers
-	 * configuration, to _not_ prohibit multi-channel audio playback.
-	 */
-	if (!eld->spk_alloc)
-		eld->spk_alloc = 0xffff;
+	hdmi_present_sense(&spec->pins[pin_idx], 1);
 }
 
 static void hdmi_non_intrinsic_event(struct hda_codec *codec, unsigned int res)
@@ -968,9 +961,11 @@
 	return 0;
 }
 
-static void hdmi_present_sense(struct hda_codec *codec, hda_nid_t pin_nid,
-			       struct hdmi_eld *eld)
+static void hdmi_present_sense(struct hdmi_spec_per_pin *per_pin, int repoll)
 {
+	struct hda_codec *codec = per_pin->codec;
+	struct hdmi_eld *eld = &per_pin->sink_eld;
+	hda_nid_t pin_nid = per_pin->pin_nid;
 	/*
 	 * Always execute a GetPinSense verb here, even when called from
 	 * hdmi_intrinsic_event; for some NVIDIA HW, the unsolicited
@@ -980,26 +975,42 @@
 	 * the unsolicited response to avoid custom WARs.
 	 */
 	int present = snd_hda_pin_sense(codec, pin_nid);
+	bool eld_valid = false;
 
-	memset(eld, 0, sizeof(*eld));
+	memset(eld, 0, offsetof(struct hdmi_eld, eld_buffer));
 
 	eld->monitor_present	= !!(present & AC_PINSENSE_PRESENCE);
 	if (eld->monitor_present)
-		eld->eld_valid	= !!(present & AC_PINSENSE_ELDV);
-	else
-		eld->eld_valid	= 0;
+		eld_valid	= !!(present & AC_PINSENSE_ELDV);
 
 	printk(KERN_INFO
 		"HDMI status: Codec=%d Pin=%d Presence_Detect=%d ELD_Valid=%d\n",
-		codec->addr, pin_nid, eld->monitor_present, eld->eld_valid);
+		codec->addr, pin_nid, eld->monitor_present, eld_valid);
 
-	if (eld->eld_valid)
+	if (eld_valid) {
 		if (!snd_hdmi_get_eld(eld, codec, pin_nid))
 			snd_hdmi_show_eld(eld);
+		else if (repoll) {
+			queue_delayed_work(codec->bus->workq,
+					   &per_pin->work,
+					   msecs_to_jiffies(300));
+		}
+	}
 
 	snd_hda_input_jack_report(codec, pin_nid);
 }
 
+static void hdmi_repoll_eld(struct work_struct *work)
+{
+	struct hdmi_spec_per_pin *per_pin =
+	container_of(to_delayed_work(work), struct hdmi_spec_per_pin, work);
+
+	if (per_pin->repoll_count++ > 6)
+		per_pin->repoll_count = 0;
+
+	hdmi_present_sense(per_pin, per_pin->repoll_count);
+}
+
 static int hdmi_add_pin(struct hda_codec *codec, hda_nid_t pin_nid)
 {
 	struct hdmi_spec *spec = codec->spec;
@@ -1228,7 +1239,7 @@
 	if (err < 0)
 		return err;
 
-	hdmi_present_sense(codec, per_pin->pin_nid, &per_pin->sink_eld);
+	hdmi_present_sense(per_pin, 0);
 	return 0;
 }
 
@@ -1279,6 +1290,8 @@
 				    AC_VERB_SET_UNSOLICITED_ENABLE,
 				    AC_USRSP_EN | pin_nid);
 
+		per_pin->codec = codec;
+		INIT_DELAYED_WORK(&per_pin->work, hdmi_repoll_eld);
 		snd_hda_eld_proc_new(codec, eld, pin_idx);
 	}
 	return 0;
@@ -1293,10 +1306,12 @@
 		struct hdmi_spec_per_pin *per_pin = &spec->pins[pin_idx];
 		struct hdmi_eld *eld = &per_pin->sink_eld;
 
+		cancel_delayed_work(&per_pin->work);
 		snd_hda_eld_proc_free(codec, eld);
 	}
 	snd_hda_input_jack_free(codec);
 
+	flush_workqueue(codec->bus->workq);
 	kfree(spec);
 }
 
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index a24e068..1d07e8f 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -277,6 +277,12 @@
 	return false;
 }
 
+static inline hda_nid_t get_capsrc(struct alc_spec *spec, int idx)
+{
+	return spec->capsrc_nids ?
+		spec->capsrc_nids[idx] : spec->adc_nids[idx];
+}
+
 /* select the given imux item; either unmute exclusively or select the route */
 static int alc_mux_select(struct hda_codec *codec, unsigned int adc_idx,
 			  unsigned int idx, bool force)
@@ -284,13 +290,15 @@
 	struct alc_spec *spec = codec->spec;
 	const struct hda_input_mux *imux;
 	unsigned int mux_idx;
-	int i, type;
+	int i, type, num_conns;
 	hda_nid_t nid;
 
 	mux_idx = adc_idx >= spec->num_mux_defs ? 0 : adc_idx;
 	imux = &spec->input_mux[mux_idx];
 	if (!imux->num_items && mux_idx > 0)
 		imux = &spec->input_mux[0];
+	if (!imux->num_items)
+		return 0;
 
 	if (idx >= imux->num_items)
 		idx = imux->num_items - 1;
@@ -303,20 +311,20 @@
 		adc_idx = spec->dyn_adc_idx[idx];
 	}
 
-	nid = spec->capsrc_nids ?
-		spec->capsrc_nids[adc_idx] : spec->adc_nids[adc_idx];
+	nid = get_capsrc(spec, adc_idx);
 
 	/* no selection? */
-	if (snd_hda_get_conn_list(codec, nid, NULL) <= 1)
+	num_conns = snd_hda_get_conn_list(codec, nid, NULL);
+	if (num_conns <= 1)
 		return 1;
 
 	type = get_wcaps_type(get_wcaps(codec, nid));
 	if (type == AC_WID_AUD_MIX) {
 		/* Matrix-mixer style (e.g. ALC882) */
-		for (i = 0; i < imux->num_items; i++) {
-			unsigned int v = (i == idx) ? 0 : HDA_AMP_MUTE;
-			snd_hda_codec_amp_stereo(codec, nid, HDA_INPUT,
-						 imux->items[i].index,
+		int active = imux->items[idx].index;
+		for (i = 0; i < num_conns; i++) {
+			unsigned int v = (i == active) ? 0 : HDA_AMP_MUTE;
+			snd_hda_codec_amp_stereo(codec, nid, HDA_INPUT, i,
 						 HDA_AMP_MUTE, v);
 		}
 	} else {
@@ -1053,8 +1061,19 @@
 	spec->imux_pins[2] = spec->dock_mic_pin;
 	for (i = 0; i < 3; i++) {
 		strcpy(imux->items[i].label, texts[i]);
-		if (spec->imux_pins[i])
+		if (spec->imux_pins[i]) {
+			hda_nid_t pin = spec->imux_pins[i];
+			int c;
+			for (c = 0; c < spec->num_adc_nids; c++) {
+				hda_nid_t cap = get_capsrc(spec, c);
+				int idx = get_connection_index(codec, cap, pin);
+				if (idx >= 0) {
+					imux->items[i].index = idx;
+					break;
+				}
+			}
 			imux->num_items = i + 1;
+		}
 	}
 	spec->num_mux_defs = 1;
 	spec->input_mux = imux;
@@ -1451,7 +1470,7 @@
 		switch (fix->type) {
 		case ALC_FIXUP_SKU:
 			if (action != ALC_FIXUP_ACT_PRE_PROBE || !fix->v.sku)
-				break;;
+				break;
 			snd_printdd(KERN_INFO "hda_codec: %s: "
 				    "Apply sku override for %s\n",
 				    codec->chip_name, modelname);
@@ -1956,10 +1975,8 @@
 		if (!kctl)
 			kctl = snd_hda_find_mixer_ctl(codec, "Input Source");
 		for (i = 0; kctl && i < kctl->count; i++) {
-			const hda_nid_t *nids = spec->capsrc_nids;
-			if (!nids)
-				nids = spec->adc_nids;
-			err = snd_hda_add_nid(codec, kctl, i, nids[i]);
+			err = snd_hda_add_nid(codec, kctl, i,
+					      get_capsrc(spec, i));
 			if (err < 0)
 				return err;
 		}
@@ -2614,6 +2631,8 @@
 	case AUTO_PIN_SPEAKER_OUT:
 		if (cfg->line_outs == 1)
 			return "Speaker";
+		if (cfg->line_outs == 2)
+			return ch ? "Bass Speaker" : "Speaker";
 		break;
 	case AUTO_PIN_HP_OUT:
 		/* for multi-io case, only the primary out */
@@ -2746,8 +2765,7 @@
 		}
 
 		for (c = 0; c < num_adcs; c++) {
-			hda_nid_t cap = spec->capsrc_nids ?
-				spec->capsrc_nids[c] : spec->adc_nids[c];
+			hda_nid_t cap = get_capsrc(spec, c);
 			idx = get_connection_index(codec, cap, pin);
 			if (idx >= 0) {
 				spec->imux_pins[imux->num_items] = pin;
@@ -2888,7 +2906,7 @@
 		if (!nid)
 			continue;
 		if (found_in_nid_list(nid, spec->multiout.dac_nids,
-				      spec->multiout.num_dacs))
+				      ARRAY_SIZE(spec->private_dac_nids)))
 			continue;
 		if (found_in_nid_list(nid, spec->multiout.hp_out_nid,
 				      ARRAY_SIZE(spec->multiout.hp_out_nid)))
@@ -2909,6 +2927,7 @@
 	return 0;
 }
 
+/* return 0 if no possible DAC is found, 1 if one or more found */
 static int alc_auto_fill_extra_dacs(struct hda_codec *codec, int num_outs,
 				    const hda_nid_t *pins, hda_nid_t *dacs)
 {
@@ -2926,7 +2945,7 @@
 		if (!dacs[i])
 			dacs[i] = alc_auto_look_for_dac(codec, pins[i]);
 	}
-	return 0;
+	return 1;
 }
 
 static int alc_auto_fill_multi_ios(struct hda_codec *codec,
@@ -2936,7 +2955,7 @@
 static int alc_auto_fill_dac_nids(struct hda_codec *codec)
 {
 	struct alc_spec *spec = codec->spec;
-	const struct auto_pin_cfg *cfg = &spec->autocfg;
+	struct auto_pin_cfg *cfg = &spec->autocfg;
 	bool redone = false;
 	int i;
 
@@ -2947,6 +2966,7 @@
 	spec->multiout.extra_out_nid[0] = 0;
 	memset(spec->private_dac_nids, 0, sizeof(spec->private_dac_nids));
 	spec->multiout.dac_nids = spec->private_dac_nids;
+	spec->multi_ios = 0;
 
 	/* fill hard-wired DACs first */
 	if (!redone) {
@@ -2980,10 +3000,12 @@
 	for (i = 0; i < cfg->line_outs; i++) {
 		if (spec->private_dac_nids[i])
 			spec->multiout.num_dacs++;
-		else
+		else {
 			memmove(spec->private_dac_nids + i,
 				spec->private_dac_nids + i + 1,
 				sizeof(hda_nid_t) * (cfg->line_outs - i - 1));
+			spec->private_dac_nids[cfg->line_outs - 1] = 0;
+		}
 	}
 
 	if (cfg->line_outs == 1 && cfg->line_out_type != AUTO_PIN_SPEAKER_OUT) {
@@ -3005,9 +3027,28 @@
 	if (cfg->line_out_type != AUTO_PIN_HP_OUT)
 		alc_auto_fill_extra_dacs(codec, cfg->hp_outs, cfg->hp_pins,
 				 spec->multiout.hp_out_nid);
-	if (cfg->line_out_type != AUTO_PIN_SPEAKER_OUT)
-		alc_auto_fill_extra_dacs(codec, cfg->speaker_outs, cfg->speaker_pins,
-				 spec->multiout.extra_out_nid);
+	if (cfg->line_out_type != AUTO_PIN_SPEAKER_OUT) {
+		int err = alc_auto_fill_extra_dacs(codec, cfg->speaker_outs,
+					cfg->speaker_pins,
+					spec->multiout.extra_out_nid);
+		/* if no speaker volume is assigned, try again as the primary
+		 * output
+		 */
+		if (!err && cfg->speaker_outs > 0 &&
+		    cfg->line_out_type == AUTO_PIN_HP_OUT) {
+			cfg->hp_outs = cfg->line_outs;
+			memcpy(cfg->hp_pins, cfg->line_out_pins,
+			       sizeof(cfg->hp_pins));
+			cfg->line_outs = cfg->speaker_outs;
+			memcpy(cfg->line_out_pins, cfg->speaker_pins,
+			       sizeof(cfg->speaker_pins));
+			cfg->speaker_outs = 0;
+			memset(cfg->speaker_pins, 0, sizeof(cfg->speaker_pins));
+			cfg->line_out_type = AUTO_PIN_SPEAKER_OUT;
+			redone = false;
+			goto again;
+		}
+	}
 
 	return 0;
 }
@@ -3157,7 +3198,8 @@
 }
 
 static int alc_auto_create_extra_out(struct hda_codec *codec, hda_nid_t pin,
-				     hda_nid_t dac, const char *pfx)
+				     hda_nid_t dac, const char *pfx,
+				     int cidx)
 {
 	struct alc_spec *spec = codec->spec;
 	hda_nid_t sw, vol;
@@ -3173,15 +3215,15 @@
 		if (is_ctl_used(spec->sw_ctls, val))
 			return 0; /* already created */
 		mark_ctl_usage(spec->sw_ctls, val);
-		return add_pb_sw_ctrl(spec, ALC_CTL_WIDGET_MUTE, pfx, val);
+		return __add_pb_sw_ctrl(spec, ALC_CTL_WIDGET_MUTE, pfx, cidx, val);
 	}
 
 	sw = alc_look_for_out_mute_nid(codec, pin, dac);
 	vol = alc_look_for_out_vol_nid(codec, pin, dac);
-	err = alc_auto_add_stereo_vol(codec, pfx, 0, vol);
+	err = alc_auto_add_stereo_vol(codec, pfx, cidx, vol);
 	if (err < 0)
 		return err;
-	err = alc_auto_add_stereo_sw(codec, pfx, 0, sw);
+	err = alc_auto_add_stereo_sw(codec, pfx, cidx, sw);
 	if (err < 0)
 		return err;
 	return 0;
@@ -3222,16 +3264,21 @@
 		hda_nid_t dac = *dacs;
 		if (!dac)
 			dac = spec->multiout.dac_nids[0];
-		return alc_auto_create_extra_out(codec, *pins, dac, pfx);
+		return alc_auto_create_extra_out(codec, *pins, dac, pfx, 0);
 	}
 
 	if (dacs[num_pins - 1]) {
 		/* OK, we have a multi-output system with individual volumes */
 		for (i = 0; i < num_pins; i++) {
-			snprintf(name, sizeof(name), "%s %s",
-				 pfx, channel_name[i]);
-			err = alc_auto_create_extra_out(codec, pins[i], dacs[i],
-							name);
+			if (num_pins >= 3) {
+				snprintf(name, sizeof(name), "%s %s",
+					 pfx, channel_name[i]);
+				err = alc_auto_create_extra_out(codec, pins[i], dacs[i],
+								name, 0);
+			} else {
+				err = alc_auto_create_extra_out(codec, pins[i], dacs[i],
+								pfx, i);
+			}
 			if (err < 0)
 				return err;
 		}
@@ -3693,8 +3740,7 @@
 	if (!pin)
 		return 0;
 	for (i = 0; i < spec->num_adc_nids; i++) {
-		hda_nid_t cap = spec->capsrc_nids ?
-			spec->capsrc_nids[i] : spec->adc_nids[i];
+		hda_nid_t cap = get_capsrc(spec, i);
 		int idx;
 
 		idx = get_connection_index(codec, cap, pin);
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index 4e715fe..616678f 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -95,6 +95,7 @@
 	STAC_92HD83XXX_REF,
 	STAC_92HD83XXX_PWR_REF,
 	STAC_DELL_S14,
+	STAC_DELL_VOSTRO_3500,
 	STAC_92HD83XXX_HP,
 	STAC_92HD83XXX_HP_cNB11_INTQUAD,
 	STAC_HP_DV7_4000,
@@ -214,6 +215,7 @@
 	unsigned int gpio_mute;
 	unsigned int gpio_led;
 	unsigned int gpio_led_polarity;
+	unsigned int vref_mute_led_nid; /* pin NID for mute-LED vref control */
 	unsigned int vref_led;
 
 	/* stream */
@@ -226,7 +228,6 @@
 
 	/* power management */
 	unsigned int num_pwrs;
-	const unsigned int *pwr_mapping;
 	const hda_nid_t *pwr_nids;
 	const hda_nid_t *dac_list;
 
@@ -373,18 +374,15 @@
 
 #define STAC92HD83_DAC_COUNT 3
 
-static const hda_nid_t stac92hd83xxx_pwr_nids[4] = {
-	0xa, 0xb, 0xd, 0xe,
+static const hda_nid_t stac92hd83xxx_pwr_nids[7] = {
+	0x0a, 0x0b, 0x0c, 0xd, 0x0e,
+	0x0f, 0x10
 };
 
 static const hda_nid_t stac92hd83xxx_slave_dig_outs[2] = {
 	0x1e, 0,
 };
 
-static const unsigned int stac92hd83xxx_pwr_mapping[4] = {
-	0x03, 0x0c, 0x20, 0x40,
-};
-
 static const hda_nid_t stac92hd83xxx_dmic_nids[] = {
 		0x11, 0x20,
 };
@@ -1644,6 +1642,8 @@
 		      "Alienware M17x", STAC_ALIENWARE_M17X),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL, 0x043a,
 		      "Alienware M17x", STAC_ALIENWARE_M17X),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL, 0x0490,
+		      "Alienware M17x", STAC_ALIENWARE_M17X),
 	{} /* terminator */
 };
 
@@ -1659,6 +1659,12 @@
 	0x40f000f0, 0x40f000f0,
 };
 
+static const unsigned int dell_vostro_3500_pin_configs[10] = {
+	0x02a11020, 0x0221101f, 0x400000f0, 0x90170110,
+	0x400000f1, 0x400000f2, 0x400000f3, 0x90a60160,
+	0x400000f4, 0x400000f5,
+};
+
 static const unsigned int hp_dv7_4000_pin_configs[10] = {
 	0x03a12050, 0x0321201f, 0x40f000f0, 0x90170110,
 	0x40f000f0, 0x40f000f0, 0x90170110, 0xd5a30140,
@@ -1675,6 +1681,7 @@
 	[STAC_92HD83XXX_REF] = ref92hd83xxx_pin_configs,
 	[STAC_92HD83XXX_PWR_REF] = ref92hd83xxx_pin_configs,
 	[STAC_DELL_S14] = dell_s14_pin_configs,
+	[STAC_DELL_VOSTRO_3500] = dell_vostro_3500_pin_configs,
 	[STAC_92HD83XXX_HP_cNB11_INTQUAD] = hp_cNB11_intquad_pin_configs,
 	[STAC_HP_DV7_4000] = hp_dv7_4000_pin_configs,
 };
@@ -1684,6 +1691,7 @@
 	[STAC_92HD83XXX_REF] = "ref",
 	[STAC_92HD83XXX_PWR_REF] = "mic-ref",
 	[STAC_DELL_S14] = "dell-s14",
+	[STAC_DELL_VOSTRO_3500] = "dell-vostro-3500",
 	[STAC_92HD83XXX_HP] = "hp",
 	[STAC_92HD83XXX_HP_cNB11_INTQUAD] = "hp_cNB11_intquad",
 	[STAC_HP_DV7_4000] = "hp-dv7-4000",
@@ -1697,6 +1705,8 @@
 		      "DFI LanParty", STAC_92HD83XXX_REF),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL, 0x02ba,
 		      "unknown Dell", STAC_DELL_S14),
+	SND_PCI_QUIRK(PCI_VENDOR_ID_DELL, 0x1028,
+		      "Dell Vostro 3500", STAC_DELL_VOSTRO_3500),
 	SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_HP, 0xff00, 0x3600,
 			  "HP", STAC_92HD83XXX_HP),
 	SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x1656,
@@ -4309,12 +4319,10 @@
 		spec->eapd_switch = val;
 	get_int_hint(codec, "gpio_led_polarity", &spec->gpio_led_polarity);
 	if (get_int_hint(codec, "gpio_led", &spec->gpio_led)) {
-		if (spec->gpio_led <= 8) {
-			spec->gpio_mask |= spec->gpio_led;
-			spec->gpio_dir |= spec->gpio_led;
-			if (spec->gpio_led_polarity)
-				spec->gpio_data |= spec->gpio_led;
-		}
+		spec->gpio_mask |= spec->gpio_led;
+		spec->gpio_dir |= spec->gpio_led;
+		if (spec->gpio_led_polarity)
+			spec->gpio_data |= spec->gpio_led;
 	}
 }
 
@@ -4432,7 +4440,9 @@
 		int pinctl, def_conf;
 
 		/* power on when no jack detection is available */
-		if (!spec->hp_detect) {
+		/* or when the VREF is used for controlling LED */
+		if (!spec->hp_detect ||
+		    spec->vref_mute_led_nid == nid) {
 			stac_toggle_power_map(codec, nid, 1);
 			continue;
 		}
@@ -4459,8 +4469,12 @@
 				stac_toggle_power_map(codec, nid, 1);
 			continue;
 		}
-		if (enable_pin_detect(codec, nid, STAC_PWR_EVENT))
+		if (enable_pin_detect(codec, nid, STAC_PWR_EVENT)) {
 			stac_issue_unsol_event(codec, nid);
+			continue;
+		}
+		/* none of the above, turn the port OFF */
+		stac_toggle_power_map(codec, nid, 0);
 	}
 
 	/* sync mute LED */
@@ -4716,11 +4730,7 @@
 	if (idx >= spec->num_pwrs)
 		return;
 
-	/* several codecs have two power down bits */
-	if (spec->pwr_mapping)
-		idx = spec->pwr_mapping[idx];
-	else
-		idx = 1 << idx;
+	idx = 1 << idx;
 
 	val = snd_hda_codec_read(codec, codec->afg, 0, 0x0fec, 0x0) & 0xff;
 	if (enable)
@@ -4904,8 +4914,14 @@
 			if (sscanf(dev->name, "HP_Mute_LED_%d_%x",
 				  &spec->gpio_led_polarity,
 				  &spec->gpio_led) == 2) {
-				if (spec->gpio_led < 4)
+				unsigned int max_gpio;
+				max_gpio = snd_hda_param_read(codec, codec->afg,
+							      AC_PAR_GPIO_CAP);
+				max_gpio &= AC_GPIO_IO_COUNT;
+				if (spec->gpio_led < max_gpio)
 					spec->gpio_led = 1 << spec->gpio_led;
+				else
+					spec->vref_mute_led_nid = spec->gpio_led;
 				return 1;
 			}
 			if (sscanf(dev->name, "HP_Mute_LED_%d",
@@ -4913,6 +4929,12 @@
 				set_hp_led_gpio(codec);
 				return 1;
 			}
+			/* BIOS bug: unfilled OEM string */
+			if (strstr(dev->name, "HP_Mute_LED_P_G")) {
+				set_hp_led_gpio(codec);
+				spec->gpio_led_polarity = 1;
+				return 1;
+			}
 		}
 
 		/*
@@ -5034,29 +5056,12 @@
 	struct sigmatel_spec *spec = codec->spec;
 
 	/* sync mute LED */
-	if (spec->gpio_led) {
-		if (spec->gpio_led <= 8) {
-			stac_gpio_set(codec, spec->gpio_mask,
-					spec->gpio_dir, spec->gpio_data);
-		} else {
-			stac_vrefout_set(codec,
-					spec->gpio_led, spec->vref_led);
-		}
-	}
-	return 0;
-}
-
-static int stac92xx_post_suspend(struct hda_codec *codec)
-{
-	struct sigmatel_spec *spec = codec->spec;
-	if (spec->gpio_led > 8) {
-		/* with vref-out pin used for mute led control
-		 * codec AFG is prevented from D3 state, but on
-		 * system suspend it can (and should) be used
-		 */
-		snd_hda_codec_read(codec, codec->afg, 0,
-				AC_VERB_SET_POWER_STATE, AC_PWRST_D3);
-	}
+	if (spec->vref_mute_led_nid)
+		stac_vrefout_set(codec, spec->vref_mute_led_nid,
+				 spec->vref_led);
+	else if (spec->gpio_led)
+		stac_gpio_set(codec, spec->gpio_mask,
+			      spec->gpio_dir, spec->gpio_data);
 	return 0;
 }
 
@@ -5067,7 +5072,7 @@
 	struct sigmatel_spec *spec = codec->spec;
 
 	if (power_state == AC_PWRST_D3) {
-		if (spec->gpio_led > 8) {
+		if (spec->vref_mute_led_nid) {
 			/* with vref-out pin used for mute led control
 			 * codec AFG is prevented from D3 state
 			 */
@@ -5120,7 +5125,7 @@
 		}
 	}
 	/*polarity defines *not* muted state level*/
-	if (spec->gpio_led <= 8) {
+	if (!spec->vref_mute_led_nid) {
 		if (muted)
 			spec->gpio_data &= ~spec->gpio_led; /* orange */
 		else
@@ -5138,7 +5143,8 @@
 		muted_lvl = spec->gpio_led_polarity ?
 				AC_PINCTL_VREF_GRD : AC_PINCTL_VREF_HIZ;
 		spec->vref_led = muted ? muted_lvl : notmtd_lvl;
-		stac_vrefout_set(codec,	spec->gpio_led, spec->vref_led);
+		stac_vrefout_set(codec,	spec->vref_mute_led_nid,
+				 spec->vref_led);
 	}
 	return 0;
 }
@@ -5618,9 +5624,6 @@
 		snd_hda_codec_set_pincfg(codec, 0xf, 0x2181205e);
 	}
 
-	/* reset pin power-down; Windows may leave these bits after reboot */
-	snd_hda_codec_write_cache(codec, codec->afg, 0, 0x7EC, 0);
-	snd_hda_codec_write_cache(codec, codec->afg, 0, 0x7ED, 0);
 	codec->no_trigger_sense = 1;
 	codec->spec = spec;
 
@@ -5630,7 +5633,6 @@
 	codec->slave_dig_outs = stac92hd83xxx_slave_dig_outs;
 	spec->digbeep_nid = 0x21;
 	spec->pwr_nids = stac92hd83xxx_pwr_nids;
-	spec->pwr_mapping = stac92hd83xxx_pwr_mapping;
 	spec->num_pwrs = ARRAY_SIZE(stac92hd83xxx_pwr_nids);
 	spec->multiout.dac_nids = spec->dac_nids;
 	spec->init = stac92hd83xxx_core_init;
@@ -5647,9 +5649,6 @@
 		stac92xx_set_config_regs(codec,
 				stac92hd83xxx_brd_tbl[spec->board_config]);
 
-	if (spec->board_config != STAC_92HD83XXX_PWR_REF)
-		spec->num_pwrs = 0;
-
 	codec->patch_ops = stac92xx_patch_ops;
 
 	if (find_mute_led_gpio(codec, 0))
@@ -5659,15 +5658,13 @@
 
 #ifdef CONFIG_SND_HDA_POWER_SAVE
 	if (spec->gpio_led) {
-		if (spec->gpio_led <= 8) {
+		if (!spec->vref_mute_led_nid) {
 			spec->gpio_mask |= spec->gpio_led;
 			spec->gpio_dir |= spec->gpio_led;
 			spec->gpio_data |= spec->gpio_led;
 		} else {
 			codec->patch_ops.set_power_state =
 					stac92xx_set_power_state;
-			codec->patch_ops.post_suspend =
-					stac92xx_post_suspend;
 		}
 		codec->patch_ops.pre_resume = stac92xx_pre_resume;
 		codec->patch_ops.check_power_status =
@@ -5858,8 +5855,6 @@
 		    (codec->revision_id & 0xf) == 1)
 			spec->stream_delay = 40; /* 40 milliseconds */
 
-		/* no output amps */
-		spec->num_pwrs = 0;
 		/* disable VSW */
 		spec->init = stac92hd71bxx_core_init;
 		unmute_init++;
@@ -5874,8 +5869,6 @@
 		if ((codec->revision_id & 0xf) == 1)
 			spec->stream_delay = 40; /* 40 milliseconds */
 
-		/* no output amps */
-		spec->num_pwrs = 0;
 		/* fallthru */
 	default:
 		spec->init = stac92hd71bxx_core_init;
@@ -5978,15 +5971,13 @@
 
 #ifdef CONFIG_SND_HDA_POWER_SAVE
 	if (spec->gpio_led) {
-		if (spec->gpio_led <= 8) {
+		if (!spec->vref_mute_led_nid) {
 			spec->gpio_mask |= spec->gpio_led;
 			spec->gpio_dir |= spec->gpio_led;
 			spec->gpio_data |= spec->gpio_led;
 		} else {
 			codec->patch_ops.set_power_state =
 					stac92xx_set_power_state;
-			codec->patch_ops.post_suspend =
-					stac92xx_post_suspend;
 		}
 		codec->patch_ops.pre_resume = stac92xx_pre_resume;
 		codec->patch_ops.check_power_status =
diff --git a/sound/pci/hda/patch_via.c b/sound/pci/hda/patch_via.c
index 431c0d4..b513762 100644
--- a/sound/pci/hda/patch_via.c
+++ b/sound/pci/hda/patch_via.c
@@ -208,6 +208,7 @@
 	/* work to check hp jack state */
 	struct hda_codec *codec;
 	struct delayed_work vt1708_hp_work;
+	int hp_work_active;
 	int vt1708_jack_detect;
 	int vt1708_hp_present;
 
@@ -305,27 +306,35 @@
 static void analog_low_current_mode(struct hda_codec *codec);
 static bool is_aa_path_mute(struct hda_codec *codec);
 
-static void vt1708_start_hp_work(struct via_spec *spec)
-{
-	if (spec->codec_type != VT1708 || spec->autocfg.hp_pins[0] == 0)
-		return;
-	snd_hda_codec_write(spec->codec, 0x1, 0, 0xf81,
-			    !spec->vt1708_jack_detect);
-	if (!delayed_work_pending(&spec->vt1708_hp_work))
-		schedule_delayed_work(&spec->vt1708_hp_work,
-				      msecs_to_jiffies(100));
-}
+#define hp_detect_with_aa(codec) \
+	(snd_hda_get_bool_hint(codec, "analog_loopback_hp_detect") == 1 && \
+	 !is_aa_path_mute(codec))
 
 static void vt1708_stop_hp_work(struct via_spec *spec)
 {
 	if (spec->codec_type != VT1708 || spec->autocfg.hp_pins[0] == 0)
 		return;
-	if (snd_hda_get_bool_hint(spec->codec, "analog_loopback_hp_detect") == 1
-	    && !is_aa_path_mute(spec->codec))
+	if (spec->hp_work_active) {
+		snd_hda_codec_write(spec->codec, 0x1, 0, 0xf81, 1);
+		cancel_delayed_work_sync(&spec->vt1708_hp_work);
+		spec->hp_work_active = 0;
+	}
+}
+
+static void vt1708_update_hp_work(struct via_spec *spec)
+{
+	if (spec->codec_type != VT1708 || spec->autocfg.hp_pins[0] == 0)
 		return;
-	snd_hda_codec_write(spec->codec, 0x1, 0, 0xf81,
-			    !spec->vt1708_jack_detect);
-	cancel_delayed_work_sync(&spec->vt1708_hp_work);
+	if (spec->vt1708_jack_detect &&
+	    (spec->active_streams || hp_detect_with_aa(spec->codec))) {
+		if (!spec->hp_work_active) {
+			snd_hda_codec_write(spec->codec, 0x1, 0, 0xf81, 0);
+			schedule_delayed_work(&spec->vt1708_hp_work,
+					      msecs_to_jiffies(100));
+			spec->hp_work_active = 1;
+		}
+	} else if (!hp_detect_with_aa(spec->codec))
+		vt1708_stop_hp_work(spec);
 }
 
 static void set_widgets_power_state(struct hda_codec *codec)
@@ -343,12 +352,7 @@
 
 	set_widgets_power_state(codec);
 	analog_low_current_mode(snd_kcontrol_chip(kcontrol));
-	if (snd_hda_get_bool_hint(codec, "analog_loopback_hp_detect") == 1) {
-		if (is_aa_path_mute(codec))
-			vt1708_start_hp_work(codec->spec);
-		else
-			vt1708_stop_hp_work(codec->spec);
-	}
+	vt1708_update_hp_work(codec->spec);
 	return change;
 }
 
@@ -1154,7 +1158,7 @@
 	spec->cur_dac_stream_tag = stream_tag;
 	spec->cur_dac_format = format;
 	mutex_unlock(&spec->config_mutex);
-	vt1708_start_hp_work(spec);
+	vt1708_update_hp_work(spec);
 	return 0;
 }
 
@@ -1174,7 +1178,7 @@
 	spec->cur_hp_stream_tag = stream_tag;
 	spec->cur_hp_format = format;
 	mutex_unlock(&spec->config_mutex);
-	vt1708_start_hp_work(spec);
+	vt1708_update_hp_work(spec);
 	return 0;
 }
 
@@ -1188,7 +1192,7 @@
 	snd_hda_multi_out_analog_cleanup(codec, &spec->multiout);
 	spec->active_streams &= ~STREAM_MULTI_OUT;
 	mutex_unlock(&spec->config_mutex);
-	vt1708_stop_hp_work(spec);
+	vt1708_update_hp_work(spec);
 	return 0;
 }
 
@@ -1203,7 +1207,7 @@
 		snd_hda_codec_setup_stream(codec, spec->hp_dac_nid, 0, 0, 0);
 	spec->active_streams &= ~STREAM_INDEP_HP;
 	mutex_unlock(&spec->config_mutex);
-	vt1708_stop_hp_work(spec);
+	vt1708_update_hp_work(spec);
 	return 0;
 }
 
@@ -1645,7 +1649,8 @@
 	int nums;
 	struct via_spec *spec = codec->spec;
 
-	if (!spec->hp_independent_mode && spec->autocfg.hp_pins[0])
+	if (!spec->hp_independent_mode && spec->autocfg.hp_pins[0] &&
+	    (spec->codec_type != VT1708 || spec->vt1708_jack_detect))
 		present = snd_hda_jack_detect(codec, spec->autocfg.hp_pins[0]);
 
 	if (spec->smart51_enabled)
@@ -2612,8 +2617,6 @@
 
 	if (spec->codec_type != VT1708)
 		return 0;
-	spec->vt1708_jack_detect =
-		!((snd_hda_codec_read(codec, 0x1, 0, 0xf84, 0) >> 8) & 0x1);
 	ucontrol->value.integer.value[0] = spec->vt1708_jack_detect;
 	return 0;
 }
@@ -2623,18 +2626,22 @@
 {
 	struct hda_codec *codec = snd_kcontrol_chip(kcontrol);
 	struct via_spec *spec = codec->spec;
-	int change;
+	int val;
 
 	if (spec->codec_type != VT1708)
 		return 0;
-	spec->vt1708_jack_detect = ucontrol->value.integer.value[0];
-	change = (0x1 & (snd_hda_codec_read(codec, 0x1, 0, 0xf84, 0) >> 8))
-		== !spec->vt1708_jack_detect;
-	if (spec->vt1708_jack_detect) {
+	val = !!ucontrol->value.integer.value[0];
+	if (spec->vt1708_jack_detect == val)
+		return 0;
+	spec->vt1708_jack_detect = val;
+	if (spec->vt1708_jack_detect &&
+	    snd_hda_get_bool_hint(codec, "analog_loopback_hp_detect") != 1) {
 		mute_aa_path(codec, 1);
 		notify_aa_path_ctls(codec);
 	}
-	return change;
+	via_hp_automute(codec);
+	vt1708_update_hp_work(spec);
+	return 1;
 }
 
 static const struct snd_kcontrol_new vt1708_jack_detect_ctl = {
@@ -2771,6 +2778,7 @@
 	via_auto_init_unsol_event(codec);
 
 	via_hp_automute(codec);
+	vt1708_update_hp_work(spec);
 
 	return 0;
 }
@@ -2787,7 +2795,9 @@
 		spec->vt1708_hp_present ^= 1;
 		via_hp_automute(spec->codec);
 	}
-	vt1708_start_hp_work(spec);
+	if (spec->vt1708_jack_detect)
+		schedule_delayed_work(&spec->vt1708_hp_work,
+				      msecs_to_jiffies(100));
 }
 
 static int get_mux_nids(struct hda_codec *codec)
diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c
index 29e3125..11718b49 100644
--- a/sound/pci/intel8x0.c
+++ b/sound/pci/intel8x0.c
@@ -1077,6 +1077,13 @@
 		}
 		if (civ != igetbyte(chip, ichdev->reg_offset + ICH_REG_OFF_CIV))
 			continue;
+
+		/* IO read operation is very expensive inside virtual machine
+		 * as it is emulated. The probability that subsequent PICB read
+		 * will return different result is high enough to loop till
+		 * timeout here.
+		 * Same CIV is strict enough condition to be sure that PICB
+		 * is valid inside VM on emulated card. */
 		if (chip->inside_vm)
 			break;
 		if (ptr1 == igetword(chip, ichdev->reg_offset + ichdev->roff_picb))
@@ -2930,6 +2937,45 @@
 	ICH_PCR, ICH_SCR, ICH_SIS_TCR
 };
 
+static int __devinit snd_intel8x0_inside_vm(struct pci_dev *pci)
+{
+	int result  = inside_vm;
+	char *msg   = NULL;
+
+	/* check module parameter first (override detection) */
+	if (result >= 0) {
+		msg = result ? "enable (forced) VM" : "disable (forced) VM";
+		goto fini;
+	}
+
+	/* detect KVM and Parallels virtual environments */
+	result = kvm_para_available();
+#ifdef X86_FEATURE_HYPERVISOR
+	result = result || boot_cpu_has(X86_FEATURE_HYPERVISOR);
+#endif
+	if (!result)
+		goto fini;
+
+	/* check for known (emulated) devices */
+	if (pci->subsystem_vendor == 0x1af4 &&
+	    pci->subsystem_device == 0x1100) {
+		/* KVM emulated sound, PCI SSID: 1af4:1100 */
+		msg = "enable KVM";
+	} else if (pci->subsystem_vendor == 0x1ab8) {
+		/* Parallels VM emulated sound, PCI SSID: 1ab8:xxxx */
+		msg = "enable Parallels VM";
+	} else {
+		msg = "disable (unknown or VT-d) VM";
+		result = 0;
+	}
+
+fini:
+	if (msg != NULL)
+		printk(KERN_INFO "intel8x0: %s optimization\n", msg);
+
+	return result;
+}
+
 static int __devinit snd_intel8x0_create(struct snd_card *card,
 					 struct pci_dev *pci,
 					 unsigned long device_type,
@@ -2997,9 +3043,7 @@
 	if (xbox)
 		chip->xbox = 1;
 
-	chip->inside_vm = inside_vm;
-	if (inside_vm)
-		printk(KERN_INFO "intel8x0: enable KVM optimization\n");
+	chip->inside_vm = snd_intel8x0_inside_vm(pci);
 
 	if (pci->vendor == PCI_VENDOR_ID_INTEL &&
 	    pci->device == PCI_DEVICE_ID_INTEL_440MX)
@@ -3243,14 +3287,6 @@
 			buggy_irq = 0;
 	}
 
-	if (inside_vm < 0) {
-		/* detect KVM and Parallels virtual environments */
-		inside_vm = kvm_para_available();
-#if defined(__i386__) || defined(__x86_64__)
-		inside_vm = inside_vm || boot_cpu_has(X86_FEATURE_HYPERVISOR);
-#endif
-	}
-
 	if ((err = snd_intel8x0_create(card, pci, pci_id->driver_data,
 				       &chip)) < 0) {
 		snd_card_free(card);
diff --git a/sound/pci/lx6464es/lx_core.c b/sound/pci/lx6464es/lx_core.c
index 5c8717e..8c3e7fc 100644
--- a/sound/pci/lx6464es/lx_core.c
+++ b/sound/pci/lx6464es/lx_core.c
@@ -78,10 +78,15 @@
 	return ioread32(address);
 }
 
-void lx_dsp_reg_readbuf(struct lx6464es *chip, int port, u32 *data, u32 len)
+static void lx_dsp_reg_readbuf(struct lx6464es *chip, int port, u32 *data,
+			       u32 len)
 {
-	void __iomem *address = lx_dsp_register(chip, port);
-	memcpy_fromio(data, address, len*sizeof(u32));
+	u32 __iomem *address = lx_dsp_register(chip, port);
+	int i;
+
+	/* we cannot use memcpy_fromio */
+	for (i = 0; i != len; ++i)
+		data[i] = ioread32(address + i);
 }
 
 
@@ -91,11 +96,15 @@
 	iowrite32(data, address);
 }
 
-void lx_dsp_reg_writebuf(struct lx6464es *chip, int port, const u32 *data,
-			 u32 len)
+static void lx_dsp_reg_writebuf(struct lx6464es *chip, int port,
+				const u32 *data, u32 len)
 {
-	void __iomem *address = lx_dsp_register(chip, port);
-	memcpy_toio(address, data, len*sizeof(u32));
+	u32 __iomem *address = lx_dsp_register(chip, port);
+	int i;
+
+	/* we cannot use memcpy_to */
+	for (i = 0; i != len; ++i)
+		iowrite32(data[i], address + i);
 }
 
 
diff --git a/sound/pci/lx6464es/lx_core.h b/sound/pci/lx6464es/lx_core.h
index 1dd5629..4d7ff79 100644
--- a/sound/pci/lx6464es/lx_core.h
+++ b/sound/pci/lx6464es/lx_core.h
@@ -72,10 +72,7 @@
 };
 
 unsigned long lx_dsp_reg_read(struct lx6464es *chip, int port);
-void lx_dsp_reg_readbuf(struct lx6464es *chip, int port, u32 *data, u32 len);
 void lx_dsp_reg_write(struct lx6464es *chip, int port, unsigned data);
-void lx_dsp_reg_writebuf(struct lx6464es *chip, int port, const u32 *data,
-			 u32 len);
 
 /* plx register access */
 enum {
diff --git a/sound/pci/rme9652/hdspm.c b/sound/pci/rme9652/hdspm.c
index e760ada..19ee220 100644
--- a/sound/pci/rme9652/hdspm.c
+++ b/sound/pci/rme9652/hdspm.c
@@ -6518,7 +6518,7 @@
 			hdspm->io_type = AES32;
 			hdspm->card_name = "RME AES32";
 			hdspm->midiPorts = 2;
-		} else if ((hdspm->firmware_rev == 0xd5) ||
+		} else if ((hdspm->firmware_rev == 0xd2) ||
 			((hdspm->firmware_rev >= 0xc8)  &&
 				(hdspm->firmware_rev <= 0xcf))) {
 			hdspm->io_type = MADI;
diff --git a/sound/pci/sis7019.c b/sound/pci/sis7019.c
index a391e62..28dfafb 100644
--- a/sound/pci/sis7019.c
+++ b/sound/pci/sis7019.c
@@ -41,6 +41,7 @@
 static int index = SNDRV_DEFAULT_IDX1;	/* Index 0-MAX */
 static char *id = SNDRV_DEFAULT_STR1;	/* ID for this card */
 static int enable = 1;
+static int codecs = 1;
 
 module_param(index, int, 0444);
 MODULE_PARM_DESC(index, "Index value for SiS7019 Audio Accelerator.");
@@ -48,6 +49,8 @@
 MODULE_PARM_DESC(id, "ID string for SiS7019 Audio Accelerator.");
 module_param(enable, bool, 0444);
 MODULE_PARM_DESC(enable, "Enable SiS7019 Audio Accelerator.");
+module_param(codecs, int, 0444);
+MODULE_PARM_DESC(codecs, "Set bit to indicate that codec number is expected to be present (default 1)");
 
 static DEFINE_PCI_DEVICE_TABLE(snd_sis7019_ids) = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_SI, 0x7019) },
@@ -140,6 +143,9 @@
 	dma_addr_t silence_dma_addr;
 };
 
+/* These values are also used by the module param 'codecs' to indicate
+ * which codecs should be present.
+ */
 #define SIS_PRIMARY_CODEC_PRESENT	0x0001
 #define SIS_SECONDARY_CODEC_PRESENT	0x0002
 #define SIS_TERTIARY_CODEC_PRESENT	0x0004
@@ -1078,6 +1084,7 @@
 {
 	unsigned long io = sis->ioport;
 	void __iomem *ioaddr = sis->ioaddr;
+	unsigned long timeout;
 	u16 status;
 	int count;
 	int i;
@@ -1104,22 +1111,46 @@
 	while ((inw(io + SIS_AC97_STATUS) & SIS_AC97_STATUS_BUSY) && --count)
 		udelay(1);
 
-	/* Now that we've finished the reset, find out what's attached.
-	 */
-	status = inl(io + SIS_AC97_STATUS);
-	if (status & SIS_AC97_STATUS_CODEC_READY)
-		sis->codecs_present |= SIS_PRIMARY_CODEC_PRESENT;
-	if (status & SIS_AC97_STATUS_CODEC2_READY)
-		sis->codecs_present |= SIS_SECONDARY_CODEC_PRESENT;
-	if (status & SIS_AC97_STATUS_CODEC3_READY)
-		sis->codecs_present |= SIS_TERTIARY_CODEC_PRESENT;
-
-	/* All done, let go of the semaphore, and check for errors
+	/* Command complete, we can let go of the semaphore now.
 	 */
 	outl(SIS_AC97_SEMA_RELEASE, io + SIS_AC97_SEMA);
-	if (!sis->codecs_present || !count)
+	if (!count)
 		return -EIO;
 
+	/* Now that we've finished the reset, find out what's attached.
+	 * There are some codec/board combinations that take an extremely
+	 * long time to come up. 350+ ms has been observed in the field,
+	 * so we'll give them up to 500ms.
+	 */
+	sis->codecs_present = 0;
+	timeout = msecs_to_jiffies(500) + jiffies;
+	while (time_before_eq(jiffies, timeout)) {
+		status = inl(io + SIS_AC97_STATUS);
+		if (status & SIS_AC97_STATUS_CODEC_READY)
+			sis->codecs_present |= SIS_PRIMARY_CODEC_PRESENT;
+		if (status & SIS_AC97_STATUS_CODEC2_READY)
+			sis->codecs_present |= SIS_SECONDARY_CODEC_PRESENT;
+		if (status & SIS_AC97_STATUS_CODEC3_READY)
+			sis->codecs_present |= SIS_TERTIARY_CODEC_PRESENT;
+
+		if (sis->codecs_present == codecs)
+			break;
+
+		msleep(1);
+	}
+
+	/* All done, check for errors.
+	 */
+	if (!sis->codecs_present) {
+		printk(KERN_ERR "sis7019: could not find any codecs\n");
+		return -EIO;
+	}
+
+	if (sis->codecs_present != codecs) {
+		printk(KERN_WARNING "sis7019: missing codecs, found %0x, expected %0x\n",
+		       sis->codecs_present, codecs);
+	}
+
 	/* Let the hardware know that the audio driver is alive,
 	 * and enable PCM slots on the AC-link for L/R playback (3 & 4) and
 	 * record channels. We're going to want to use Variable Rate Audio
@@ -1390,6 +1421,17 @@
 	if (!enable)
 		goto error_out;
 
+	/* The user can specify which codecs should be present so that we
+	 * can wait for them to show up if they are slow to recover from
+	 * the AC97 cold reset. We default to a single codec, the primary.
+	 *
+	 * We assume that SIS_PRIMARY_*_PRESENT matches bits 0-2.
+	 */
+	codecs &= SIS_PRIMARY_CODEC_PRESENT | SIS_SECONDARY_CODEC_PRESENT |
+		  SIS_TERTIARY_CODEC_PRESENT;
+	if (!codecs)
+		codecs = SIS_PRIMARY_CODEC_PRESENT;
+
 	rc = snd_card_create(index, id, THIS_MODULE, sizeof(*sis), &card);
 	if (rc < 0)
 		goto error_out;
diff --git a/sound/ppc/snd_ps3.c b/sound/ppc/snd_ps3.c
index a3ce1b2..1aa52ef 100644
--- a/sound/ppc/snd_ps3.c
+++ b/sound/ppc/snd_ps3.c
@@ -876,7 +876,7 @@
 		(0x0fUL << 12) |
 		(PS3_AUDIO_IOID);
 
-	ret = lv1_gpu_attribute(0x100, 0x007, val, 0, 0);
+	ret = lv1_gpu_attribute(0x100, 0x007, val);
 	if (ret)
 		pr_info("%s: gpu_attribute failed %d\n", __func__,
 			ret);
diff --git a/sound/soc/atmel/Kconfig b/sound/soc/atmel/Kconfig
index bee3c94..d1fcc81 100644
--- a/sound/soc/atmel/Kconfig
+++ b/sound/soc/atmel/Kconfig
@@ -1,6 +1,6 @@
 config SND_ATMEL_SOC
 	tristate "SoC Audio for the Atmel System-on-Chip"
-	depends on ARCH_AT91 || AVR32
+	depends on ARCH_AT91
 	help
 	  Say Y or M if you want to add support for codecs attached to
 	  the ATMEL SSC interface. You will also need
@@ -24,25 +24,6 @@
 	  Say Y if you want to add support for SoC audio on WM8731-based
 	  AT91sam9g20 evaluation board.
 
-config SND_AT32_SOC_PLAYPAQ
-        tristate "SoC Audio support for PlayPaq with WM8510"
-        depends on SND_ATMEL_SOC && BOARD_PLAYPAQ && AT91_PROGRAMMABLE_CLOCKS
-        select SND_ATMEL_SOC_SSC
-        select SND_SOC_WM8510
-        help
-          Say Y or M here if you want to add support for SoC audio
-          on the LRS PlayPaq.
-
-config SND_AT32_SOC_PLAYPAQ_SLAVE
-        bool "Run CODEC on PlayPaq in slave mode"
-        depends on SND_AT32_SOC_PLAYPAQ
-        default n
-        help
-          Say Y if you want to run with the AT32 SSC generating the BCLK
-          and FRAME signals on the PlayPaq.  Unless you want to play
-          with the AT32 as the SSC master, you probably want to say N here,
-          as this will give you better sound quality.
-
 config SND_AT91_SOC_AFEB9260
 	tristate "SoC Audio support for AFEB9260 board"
 	depends on ARCH_AT91 && MACH_AFEB9260 && SND_ATMEL_SOC
diff --git a/sound/soc/atmel/Makefile b/sound/soc/atmel/Makefile
index e7ea56b..a5c0bf1 100644
--- a/sound/soc/atmel/Makefile
+++ b/sound/soc/atmel/Makefile
@@ -8,9 +8,5 @@
 # AT91 Machine Support
 snd-soc-sam9g20-wm8731-objs := sam9g20_wm8731.o
 
-# AT32 Machine Support
-snd-soc-playpaq-objs := playpaq_wm8510.o
-
 obj-$(CONFIG_SND_AT91_SOC_SAM9G20_WM8731) += snd-soc-sam9g20-wm8731.o
-obj-$(CONFIG_SND_AT32_SOC_PLAYPAQ) += snd-soc-playpaq.o
 obj-$(CONFIG_SND_AT91_SOC_AFEB9260) += snd-soc-afeb9260.o
diff --git a/sound/soc/atmel/playpaq_wm8510.c b/sound/soc/atmel/playpaq_wm8510.c
deleted file mode 100644
index 73ae99a..0000000
--- a/sound/soc/atmel/playpaq_wm8510.c
+++ /dev/null
@@ -1,473 +0,0 @@
-/* sound/soc/at32/playpaq_wm8510.c
- * ASoC machine driver for PlayPaq using WM8510 codec
- *
- * Copyright (C) 2008 Long Range Systems
- *    Geoffrey Wossum <gwossum@acm.org>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This code is largely inspired by sound/soc/at91/eti_b1_wm8731.c
- *
- * NOTE: If you don't have the AT32 enhanced portmux configured (which
- * isn't currently in the mainline or Atmel patched kernel), you will
- * need to set the MCLK pin (PA30) to peripheral A in your board initialization
- * code.  Something like:
- *	at32_select_periph(GPIO_PIN_PA(30), GPIO_PERIPH_A, 0);
- *
- */
-
-/* #define DEBUG */
-
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/kernel.h>
-#include <linux/errno.h>
-#include <linux/clk.h>
-#include <linux/timer.h>
-#include <linux/interrupt.h>
-#include <linux/platform_device.h>
-
-#include <sound/core.h>
-#include <sound/pcm.h>
-#include <sound/pcm_params.h>
-#include <sound/soc.h>
-
-#include <mach/at32ap700x.h>
-#include <mach/portmux.h>
-
-#include "../codecs/wm8510.h"
-#include "atmel-pcm.h"
-#include "atmel_ssc_dai.h"
-
-
-/*-------------------------------------------------------------------------*\
- * constants
-\*-------------------------------------------------------------------------*/
-#define MCLK_PIN		GPIO_PIN_PA(30)
-#define MCLK_PERIPH		GPIO_PERIPH_A
-
-
-/*-------------------------------------------------------------------------*\
- * data types
-\*-------------------------------------------------------------------------*/
-/* SSC clocking data */
-struct ssc_clock_data {
-	/* CMR div */
-	unsigned int cmr_div;
-
-	/* Frame period (as needed by xCMR.PERIOD) */
-	unsigned int period;
-
-	/* The SSC clock rate these settings where calculated for */
-	unsigned long ssc_rate;
-};
-
-
-/*-------------------------------------------------------------------------*\
- * module data
-\*-------------------------------------------------------------------------*/
-static struct clk *_gclk0;
-static struct clk *_pll0;
-
-#define CODEC_CLK (_gclk0)
-
-
-/*-------------------------------------------------------------------------*\
- * Sound SOC operations
-\*-------------------------------------------------------------------------*/
-#if defined CONFIG_SND_AT32_SOC_PLAYPAQ_SLAVE
-static struct ssc_clock_data playpaq_wm8510_calc_ssc_clock(
-	struct snd_pcm_hw_params *params,
-	struct snd_soc_dai *cpu_dai)
-{
-	struct at32_ssc_info *ssc_p = snd_soc_dai_get_drvdata(cpu_dai);
-	struct ssc_device *ssc = ssc_p->ssc;
-	struct ssc_clock_data cd;
-	unsigned int rate, width_bits, channels;
-	unsigned int bitrate, ssc_div;
-	unsigned actual_rate;
-
-
-	/*
-	 * Figure out required bitrate
-	 */
-	rate = params_rate(params);
-	channels = params_channels(params);
-	width_bits = snd_pcm_format_physical_width(params_format(params));
-	bitrate = rate * width_bits * channels;
-
-
-	/*
-	 * Figure out required SSC divider and period for required bitrate
-	 */
-	cd.ssc_rate = clk_get_rate(ssc->clk);
-	ssc_div = cd.ssc_rate / bitrate;
-	cd.cmr_div = ssc_div / 2;
-	if (ssc_div & 1) {
-		/* round cmr_div up */
-		cd.cmr_div++;
-	}
-	cd.period = width_bits - 1;
-
-
-	/*
-	 * Find actual rate, compare to requested rate
-	 */
-	actual_rate = (cd.ssc_rate / (cd.cmr_div * 2)) / (2 * (cd.period + 1));
-	pr_debug("playpaq_wm8510: Request rate = %u, actual rate = %u\n",
-		 rate, actual_rate);
-
-
-	return cd;
-}
-#endif /* CONFIG_SND_AT32_SOC_PLAYPAQ_SLAVE */
-
-
-
-static int playpaq_wm8510_hw_params(struct snd_pcm_substream *substream,
-				    struct snd_pcm_hw_params *params)
-{
-	struct snd_soc_pcm_runtime *rtd = substream->private_data;
-	struct snd_soc_dai *codec_dai = rtd->codec_dai;
-	struct snd_soc_dai *cpu_dai = rtd->cpu_dai;
-	struct at32_ssc_info *ssc_p = snd_soc_dai_get_drvdata(cpu_dai);
-	struct ssc_device *ssc = ssc_p->ssc;
-	unsigned int pll_out = 0, bclk = 0, mclk_div = 0;
-	int ret;
-
-
-	/* Due to difficulties with getting the correct clocks from the AT32's
-	 * PLL0, we're going to let the CODEC be in charge of all the clocks
-	 */
-#if !defined CONFIG_SND_AT32_SOC_PLAYPAQ_SLAVE
-	const unsigned int fmt = (SND_SOC_DAIFMT_I2S |
-				  SND_SOC_DAIFMT_NB_NF |
-				  SND_SOC_DAIFMT_CBM_CFM);
-#else
-	struct ssc_clock_data cd;
-	const unsigned int fmt = (SND_SOC_DAIFMT_I2S |
-				  SND_SOC_DAIFMT_NB_NF |
-				  SND_SOC_DAIFMT_CBS_CFS);
-#endif
-
-	if (ssc == NULL) {
-		pr_warning("playpaq_wm8510_hw_params: ssc is NULL!\n");
-		return -EINVAL;
-	}
-
-
-	/*
-	 * Figure out PLL and BCLK dividers for WM8510
-	 */
-	switch (params_rate(params)) {
-	case 48000:
-		pll_out = 24576000;
-		mclk_div = WM8510_MCLKDIV_2;
-		bclk = WM8510_BCLKDIV_8;
-		break;
-
-	case 44100:
-		pll_out = 22579200;
-		mclk_div = WM8510_MCLKDIV_2;
-		bclk = WM8510_BCLKDIV_8;
-		break;
-
-	case 22050:
-		pll_out = 22579200;
-		mclk_div = WM8510_MCLKDIV_4;
-		bclk = WM8510_BCLKDIV_8;
-		break;
-
-	case 16000:
-		pll_out = 24576000;
-		mclk_div = WM8510_MCLKDIV_6;
-		bclk = WM8510_BCLKDIV_8;
-		break;
-
-	case 11025:
-		pll_out = 22579200;
-		mclk_div = WM8510_MCLKDIV_8;
-		bclk = WM8510_BCLKDIV_8;
-		break;
-
-	case 8000:
-		pll_out = 24576000;
-		mclk_div = WM8510_MCLKDIV_12;
-		bclk = WM8510_BCLKDIV_8;
-		break;
-
-	default:
-		pr_warning("playpaq_wm8510: Unsupported sample rate %d\n",
-			   params_rate(params));
-		return -EINVAL;
-	}
-
-
-	/*
-	 * set CPU and CODEC DAI configuration
-	 */
-	ret = snd_soc_dai_set_fmt(codec_dai, fmt);
-	if (ret < 0) {
-		pr_warning("playpaq_wm8510: "
-			   "Failed to set CODEC DAI format (%d)\n",
-			   ret);
-		return ret;
-	}
-	ret = snd_soc_dai_set_fmt(cpu_dai, fmt);
-	if (ret < 0) {
-		pr_warning("playpaq_wm8510: "
-			   "Failed to set CPU DAI format (%d)\n",
-			   ret);
-		return ret;
-	}
-
-
-	/*
-	 * Set CPU clock configuration
-	 */
-#if defined CONFIG_SND_AT32_SOC_PLAYPAQ_SLAVE
-	cd = playpaq_wm8510_calc_ssc_clock(params, cpu_dai);
-	pr_debug("playpaq_wm8510: cmr_div = %d, period = %d\n",
-		 cd.cmr_div, cd.period);
-	ret = snd_soc_dai_set_clkdiv(cpu_dai, AT32_SSC_CMR_DIV, cd.cmr_div);
-	if (ret < 0) {
-		pr_warning("playpaq_wm8510: Failed to set CPU CMR_DIV (%d)\n",
-			   ret);
-		return ret;
-	}
-	ret = snd_soc_dai_set_clkdiv(cpu_dai, AT32_SSC_TCMR_PERIOD,
-					  cd.period);
-	if (ret < 0) {
-		pr_warning("playpaq_wm8510: "
-			   "Failed to set CPU transmit period (%d)\n",
-			   ret);
-		return ret;
-	}
-#endif /* CONFIG_SND_AT32_SOC_PLAYPAQ_SLAVE */
-
-
-	/*
-	 * Set CODEC clock configuration
-	 */
-	pr_debug("playpaq_wm8510: "
-		 "pll_in = %ld, pll_out = %u, bclk = %x, mclk = %x\n",
-		 clk_get_rate(CODEC_CLK), pll_out, bclk, mclk_div);
-
-
-#if !defined CONFIG_SND_AT32_SOC_PLAYPAQ_SLAVE
-	ret = snd_soc_dai_set_clkdiv(codec_dai, WM8510_BCLKDIV, bclk);
-	if (ret < 0) {
-		pr_warning
-		    ("playpaq_wm8510: Failed to set CODEC DAI BCLKDIV (%d)\n",
-		     ret);
-		return ret;
-	}
-#endif /* CONFIG_SND_AT32_SOC_PLAYPAQ_SLAVE */
-
-
-	ret = snd_soc_dai_set_pll(codec_dai, 0, 0,
-					 clk_get_rate(CODEC_CLK), pll_out);
-	if (ret < 0) {
-		pr_warning("playpaq_wm8510: Failed to set CODEC DAI PLL (%d)\n",
-			   ret);
-		return ret;
-	}
-
-
-	ret = snd_soc_dai_set_clkdiv(codec_dai, WM8510_MCLKDIV, mclk_div);
-	if (ret < 0) {
-		pr_warning("playpaq_wm8510: Failed to set CODEC MCLKDIV (%d)\n",
-			   ret);
-		return ret;
-	}
-
-
-	return 0;
-}
-
-
-
-static struct snd_soc_ops playpaq_wm8510_ops = {
-	.hw_params = playpaq_wm8510_hw_params,
-};
-
-
-
-static const struct snd_soc_dapm_widget playpaq_dapm_widgets[] = {
-	SND_SOC_DAPM_MIC("Int Mic", NULL),
-	SND_SOC_DAPM_SPK("Ext Spk", NULL),
-};
-
-
-
-static const struct snd_soc_dapm_route intercon[] = {
-	/* speaker connected to SPKOUT */
-	{"Ext Spk", NULL, "SPKOUTP"},
-	{"Ext Spk", NULL, "SPKOUTN"},
-
-	{"Mic Bias", NULL, "Int Mic"},
-	{"MICN", NULL, "Mic Bias"},
-	{"MICP", NULL, "Mic Bias"},
-};
-
-
-
-static int playpaq_wm8510_init(struct snd_soc_pcm_runtime *rtd)
-{
-	struct snd_soc_codec *codec = rtd->codec;
-	struct snd_soc_dapm_context *dapm = &codec->dapm;
-	int i;
-
-	/*
-	 * Add DAPM widgets
-	 */
-	for (i = 0; i < ARRAY_SIZE(playpaq_dapm_widgets); i++)
-		snd_soc_dapm_new_control(dapm, &playpaq_dapm_widgets[i]);
-
-
-
-	/*
-	 * Setup audio path interconnects
-	 */
-	snd_soc_dapm_add_routes(dapm, intercon, ARRAY_SIZE(intercon));
-
-
-
-	/* always connected pins */
-	snd_soc_dapm_enable_pin(dapm, "Int Mic");
-	snd_soc_dapm_enable_pin(dapm, "Ext Spk");
-
-
-
-	/* Make CSB show PLL rate */
-	snd_soc_dai_set_clkdiv(rtd->codec_dai, WM8510_OPCLKDIV,
-				       WM8510_OPCLKDIV_1 | 4);
-
-	return 0;
-}
-
-
-
-static struct snd_soc_dai_link playpaq_wm8510_dai = {
-	.name = "WM8510",
-	.stream_name = "WM8510 PCM",
-	.cpu_dai_name= "atmel-ssc-dai.0",
-	.platform_name = "atmel-pcm-audio",
-	.codec_name = "wm8510-codec.0-0x1a",
-	.codec_dai_name = "wm8510-hifi",
-	.init = playpaq_wm8510_init,
-	.ops = &playpaq_wm8510_ops,
-};
-
-
-
-static struct snd_soc_card snd_soc_playpaq = {
-	.name = "LRS_PlayPaq_WM8510",
-	.dai_link = &playpaq_wm8510_dai,
-	.num_links = 1,
-};
-
-static struct platform_device *playpaq_snd_device;
-
-
-static int __init playpaq_asoc_init(void)
-{
-	int ret = 0;
-
-	/*
-	 * Configure MCLK for WM8510
-	 */
-	_gclk0 = clk_get(NULL, "gclk0");
-	if (IS_ERR(_gclk0)) {
-		_gclk0 = NULL;
-		ret = PTR_ERR(_gclk0);
-		goto err_gclk0;
-	}
-	_pll0 = clk_get(NULL, "pll0");
-	if (IS_ERR(_pll0)) {
-		_pll0 = NULL;
-		ret = PTR_ERR(_pll0);
-		goto err_pll0;
-	}
-	ret = clk_set_parent(_gclk0, _pll0);
-	if (ret) {
-		pr_warning("snd-soc-playpaq: "
-			   "Failed to set PLL0 as parent for DAC clock\n");
-		goto err_set_clk;
-	}
-	clk_set_rate(CODEC_CLK, 12000000);
-	clk_enable(CODEC_CLK);
-
-#if defined CONFIG_AT32_ENHANCED_PORTMUX
-	at32_select_periph(MCLK_PIN, MCLK_PERIPH, 0);
-#endif
-
-
-	/*
-	 * Create and register platform device
-	 */
-	playpaq_snd_device = platform_device_alloc("soc-audio", 0);
-	if (playpaq_snd_device == NULL) {
-		ret = -ENOMEM;
-		goto err_device_alloc;
-	}
-
-	platform_set_drvdata(playpaq_snd_device, &snd_soc_playpaq);
-
-	ret = platform_device_add(playpaq_snd_device);
-	if (ret) {
-		pr_warning("playpaq_wm8510: platform_device_add failed (%d)\n",
-			   ret);
-		goto err_device_add;
-	}
-
-	return 0;
-
-
-err_device_add:
-	if (playpaq_snd_device != NULL) {
-		platform_device_put(playpaq_snd_device);
-		playpaq_snd_device = NULL;
-	}
-err_device_alloc:
-err_set_clk:
-	if (_pll0 != NULL) {
-		clk_put(_pll0);
-		_pll0 = NULL;
-	}
-err_pll0:
-	if (_gclk0 != NULL) {
-		clk_put(_gclk0);
-		_gclk0 = NULL;
-	}
-	return ret;
-}
-
-
-static void __exit playpaq_asoc_exit(void)
-{
-	if (_gclk0 != NULL) {
-		clk_put(_gclk0);
-		_gclk0 = NULL;
-	}
-	if (_pll0 != NULL) {
-		clk_put(_pll0);
-		_pll0 = NULL;
-	}
-
-#if defined CONFIG_AT32_ENHANCED_PORTMUX
-	at32_free_pin(MCLK_PIN);
-#endif
-
-	platform_device_unregister(playpaq_snd_device);
-	playpaq_snd_device = NULL;
-}
-
-module_init(playpaq_asoc_init);
-module_exit(playpaq_asoc_exit);
-
-MODULE_AUTHOR("Geoffrey Wossum <gwossum@acm.org>");
-MODULE_DESCRIPTION("ASoC machine driver for LRS PlayPaq");
-MODULE_LICENSE("GPL");
diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig
index 4584514..fa787d4 100644
--- a/sound/soc/codecs/Kconfig
+++ b/sound/soc/codecs/Kconfig
@@ -33,7 +33,7 @@
 	select SND_SOC_CX20442
 	select SND_SOC_DA7210 if I2C
 	select SND_SOC_DFBMCS320
-	select SND_SOC_JZ4740_CODEC if SOC_JZ4740
+	select SND_SOC_JZ4740_CODEC
 	select SND_SOC_LM4857 if I2C
 	select SND_SOC_MAX98088 if I2C
 	select SND_SOC_MAX98095 if I2C
diff --git a/sound/soc/codecs/ad1836.h b/sound/soc/codecs/ad1836.h
index 444747f..dd7be0d 100644
--- a/sound/soc/codecs/ad1836.h
+++ b/sound/soc/codecs/ad1836.h
@@ -34,7 +34,7 @@
 
 #define AD1836_ADC_CTRL2               13
 #define AD1836_ADC_WORD_LEN_MASK       0x30
-#define AD1836_ADC_WORD_OFFSET         5
+#define AD1836_ADC_WORD_OFFSET         4
 #define AD1836_ADC_SERFMT_MASK         (7 << 6)
 #define AD1836_ADC_SERFMT_PCK256       (0x4 << 6)
 #define AD1836_ADC_SERFMT_PCK128       (0x5 << 6)
diff --git a/sound/soc/codecs/adau1373.c b/sound/soc/codecs/adau1373.c
index 1ccf8dd..45c6302 100644
--- a/sound/soc/codecs/adau1373.c
+++ b/sound/soc/codecs/adau1373.c
@@ -245,7 +245,7 @@
 };
 
 static const unsigned int adau1373_bass_tlv[] = {
-	TLV_DB_RANGE_HEAD(4),
+	TLV_DB_RANGE_HEAD(3),
 	0, 2, TLV_DB_SCALE_ITEM(-600, 600, 1),
 	3, 4, TLV_DB_SCALE_ITEM(950, 250, 0),
 	5, 7, TLV_DB_SCALE_ITEM(1400, 150, 0),
diff --git a/sound/soc/codecs/cs4270.c b/sound/soc/codecs/cs4270.c
index f1f237e..73f46eb 100644
--- a/sound/soc/codecs/cs4270.c
+++ b/sound/soc/codecs/cs4270.c
@@ -601,7 +601,6 @@
 static int cs4270_soc_resume(struct snd_soc_codec *codec)
 {
 	struct cs4270_private *cs4270 = snd_soc_codec_get_drvdata(codec);
-	struct i2c_client *i2c_client = to_i2c_client(codec->dev);
 	int reg;
 
 	regulator_bulk_enable(ARRAY_SIZE(cs4270->supplies),
@@ -612,14 +611,7 @@
 	ndelay(500);
 
 	/* first restore the entire register cache ... */
-	for (reg = CS4270_FIRSTREG; reg <= CS4270_LASTREG; reg++) {
-		u8 val = snd_soc_read(codec, reg);
-
-		if (i2c_smbus_write_byte_data(i2c_client, reg, val)) {
-			dev_err(codec->dev, "i2c write failed\n");
-			return -EIO;
-		}
-	}
+	snd_soc_cache_sync(codec);
 
 	/* ... then disable the power-down bits */
 	reg = snd_soc_read(codec, CS4270_PWRCTL);
diff --git a/sound/soc/codecs/cs4271.c b/sound/soc/codecs/cs4271.c
index 23d1bd5..69fde15 100644
--- a/sound/soc/codecs/cs4271.c
+++ b/sound/soc/codecs/cs4271.c
@@ -434,7 +434,8 @@
 {
 	int ret;
 	/* Set power-down bit */
-	ret = snd_soc_update_bits(codec, CS4271_MODE2, 0, CS4271_MODE2_PDN);
+	ret = snd_soc_update_bits(codec, CS4271_MODE2, CS4271_MODE2_PDN,
+				  CS4271_MODE2_PDN);
 	if (ret < 0)
 		return ret;
 	return 0;
@@ -501,8 +502,9 @@
 		return ret;
 	}
 
-	ret = snd_soc_update_bits(codec, CS4271_MODE2, 0,
-		CS4271_MODE2_PDN | CS4271_MODE2_CPEN);
+	ret = snd_soc_update_bits(codec, CS4271_MODE2,
+				  CS4271_MODE2_PDN | CS4271_MODE2_CPEN,
+				  CS4271_MODE2_PDN | CS4271_MODE2_CPEN);
 	if (ret < 0)
 		return ret;
 	ret = snd_soc_update_bits(codec, CS4271_MODE2, CS4271_MODE2_PDN, 0);
diff --git a/sound/soc/codecs/cs42l51.c b/sound/soc/codecs/cs42l51.c
index 8c3c820..1ee66361 100644
--- a/sound/soc/codecs/cs42l51.c
+++ b/sound/soc/codecs/cs42l51.c
@@ -555,7 +555,7 @@
 
 static struct snd_soc_codec_driver soc_codec_device_cs42l51 = {
 	.probe =	cs42l51_probe,
-	.reg_cache_size = CS42L51_NUMREGS,
+	.reg_cache_size = CS42L51_NUMREGS + 1,
 	.reg_word_size = sizeof(u8),
 };
 
diff --git a/sound/soc/codecs/jz4740.c b/sound/soc/codecs/jz4740.c
index e373f8f..3e1f4e1 100644
--- a/sound/soc/codecs/jz4740.c
+++ b/sound/soc/codecs/jz4740.c
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
+#include <linux/io.h>
 
 #include <linux/delay.h>
 
diff --git a/sound/soc/codecs/max9877.c b/sound/soc/codecs/max9877.c
index 9e7e964..dcf6f2a 100644
--- a/sound/soc/codecs/max9877.c
+++ b/sound/soc/codecs/max9877.c
@@ -106,13 +106,13 @@
 	unsigned int mask = mc->max;
 	unsigned int val = (ucontrol->value.integer.value[0] & mask);
 	unsigned int val2 = (ucontrol->value.integer.value[1] & mask);
-	unsigned int change = 1;
+	unsigned int change = 0;
 
-	if (((max9877_regs[reg] >> shift) & mask) == val)
-		change = 0;
+	if (((max9877_regs[reg] >> shift) & mask) != val)
+		change = 1;
 
-	if (((max9877_regs[reg2] >> shift) & mask) == val2)
-		change = 0;
+	if (((max9877_regs[reg2] >> shift) & mask) != val2)
+		change = 1;
 
 	if (change) {
 		max9877_regs[reg] &= ~(mask << shift);
diff --git a/sound/soc/codecs/rt5631.c b/sound/soc/codecs/rt5631.c
index 27a078c..4646e80 100644
--- a/sound/soc/codecs/rt5631.c
+++ b/sound/soc/codecs/rt5631.c
@@ -177,7 +177,7 @@
 static const DECLARE_TLV_DB_SCALE(in_vol_tlv, -3450, 150, 0);
 /* {0, +20, +24, +30, +35, +40, +44, +50, +52}dB */
 static unsigned int mic_bst_tlv[] = {
-	TLV_DB_RANGE_HEAD(6),
+	TLV_DB_RANGE_HEAD(7),
 	0, 0, TLV_DB_SCALE_ITEM(0, 0, 0),
 	1, 1, TLV_DB_SCALE_ITEM(2000, 0, 0),
 	2, 2, TLV_DB_SCALE_ITEM(2400, 0, 0),
diff --git a/sound/soc/codecs/sgtl5000.c b/sound/soc/codecs/sgtl5000.c
index d15695d..bbcf921 100644
--- a/sound/soc/codecs/sgtl5000.c
+++ b/sound/soc/codecs/sgtl5000.c
@@ -365,7 +365,7 @@
 
 /* tlv for mic gain, 0db 20db 30db 40db */
 static const unsigned int mic_gain_tlv[] = {
-	TLV_DB_RANGE_HEAD(4),
+	TLV_DB_RANGE_HEAD(2),
 	0, 0, TLV_DB_SCALE_ITEM(0, 0, 0),
 	1, 3, TLV_DB_SCALE_ITEM(2000, 1000, 0),
 };
diff --git a/sound/soc/codecs/sta32x.c b/sound/soc/codecs/sta32x.c
index bb82408..d2f3715 100644
--- a/sound/soc/codecs/sta32x.c
+++ b/sound/soc/codecs/sta32x.c
@@ -76,6 +76,8 @@
 
 	unsigned int mclk;
 	unsigned int format;
+
+	u32 coef_shadow[STA32X_COEF_COUNT];
 };
 
 static const DECLARE_TLV_DB_SCALE(mvol_tlv, -12700, 50, 1);
@@ -227,6 +229,7 @@
 				  struct snd_ctl_elem_value *ucontrol)
 {
 	struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
+	struct sta32x_priv *sta32x = snd_soc_codec_get_drvdata(codec);
 	int numcoef = kcontrol->private_value >> 16;
 	int index = kcontrol->private_value & 0xffff;
 	unsigned int cfud;
@@ -239,6 +242,11 @@
 	snd_soc_write(codec, STA32X_CFUD, cfud);
 
 	snd_soc_write(codec, STA32X_CFADDR2, index);
+	for (i = 0; i < numcoef && (index + i < STA32X_COEF_COUNT); i++)
+		sta32x->coef_shadow[index + i] =
+			  (ucontrol->value.bytes.data[3 * i] << 16)
+			| (ucontrol->value.bytes.data[3 * i + 1] << 8)
+			| (ucontrol->value.bytes.data[3 * i + 2]);
 	for (i = 0; i < 3 * numcoef; i++)
 		snd_soc_write(codec, STA32X_B1CF1 + i,
 			      ucontrol->value.bytes.data[i]);
@@ -252,6 +260,48 @@
 	return 0;
 }
 
+int sta32x_sync_coef_shadow(struct snd_soc_codec *codec)
+{
+	struct sta32x_priv *sta32x = snd_soc_codec_get_drvdata(codec);
+	unsigned int cfud;
+	int i;
+
+	/* preserve reserved bits in STA32X_CFUD */
+	cfud = snd_soc_read(codec, STA32X_CFUD) & 0xf0;
+
+	for (i = 0; i < STA32X_COEF_COUNT; i++) {
+		snd_soc_write(codec, STA32X_CFADDR2, i);
+		snd_soc_write(codec, STA32X_B1CF1,
+			      (sta32x->coef_shadow[i] >> 16) & 0xff);
+		snd_soc_write(codec, STA32X_B1CF2,
+			      (sta32x->coef_shadow[i] >> 8) & 0xff);
+		snd_soc_write(codec, STA32X_B1CF3,
+			      (sta32x->coef_shadow[i]) & 0xff);
+		/* chip documentation does not say if the bits are
+		 * self-clearing, so do it explicitly */
+		snd_soc_write(codec, STA32X_CFUD, cfud);
+		snd_soc_write(codec, STA32X_CFUD, cfud | 0x01);
+	}
+	return 0;
+}
+
+int sta32x_cache_sync(struct snd_soc_codec *codec)
+{
+	unsigned int mute;
+	int rc;
+
+	if (!codec->cache_sync)
+		return 0;
+
+	/* mute during register sync */
+	mute = snd_soc_read(codec, STA32X_MMUTE);
+	snd_soc_write(codec, STA32X_MMUTE, mute | STA32X_MMUTE_MMUTE);
+	sta32x_sync_coef_shadow(codec);
+	rc = snd_soc_cache_sync(codec);
+	snd_soc_write(codec, STA32X_MMUTE, mute);
+	return rc;
+}
+
 #define SINGLE_COEF(xname, index) \
 {	.iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, \
 	.info = sta32x_coefficient_info, \
@@ -661,7 +711,7 @@
 				return ret;
 			}
 
-			snd_soc_cache_sync(codec);
+			sta32x_cache_sync(codec);
 		}
 
 		/* Power up to mute */
@@ -790,6 +840,17 @@
 			    STA32X_CxCFG_OM_MASK,
 			    2 << STA32X_CxCFG_OM_SHIFT);
 
+	/* initialize coefficient shadow RAM with reset values */
+	for (i = 4; i <= 49; i += 5)
+		sta32x->coef_shadow[i] = 0x400000;
+	for (i = 50; i <= 54; i++)
+		sta32x->coef_shadow[i] = 0x7fffff;
+	sta32x->coef_shadow[55] = 0x5a9df7;
+	sta32x->coef_shadow[56] = 0x7fffff;
+	sta32x->coef_shadow[59] = 0x7fffff;
+	sta32x->coef_shadow[60] = 0x400000;
+	sta32x->coef_shadow[61] = 0x400000;
+
 	sta32x_set_bias_level(codec, SND_SOC_BIAS_STANDBY);
 	/* Bias level configuration will have done an extra enable */
 	regulator_bulk_disable(ARRAY_SIZE(sta32x->supplies), sta32x->supplies);
diff --git a/sound/soc/codecs/sta32x.h b/sound/soc/codecs/sta32x.h
index b97ee5a..d8e32a6 100644
--- a/sound/soc/codecs/sta32x.h
+++ b/sound/soc/codecs/sta32x.h
@@ -19,6 +19,7 @@
 /* STA326 register addresses */
 
 #define STA32X_REGISTER_COUNT	0x2d
+#define STA32X_COEF_COUNT 62
 
 #define STA32X_CONFA	0x00
 #define STA32X_CONFB    0x01
diff --git a/sound/soc/codecs/uda1380.c b/sound/soc/codecs/uda1380.c
index c5ca8cf..0441893 100644
--- a/sound/soc/codecs/uda1380.c
+++ b/sound/soc/codecs/uda1380.c
@@ -863,13 +863,13 @@
 
 static int __init uda1380_modinit(void)
 {
-	int ret;
+	int ret = 0;
 #if defined(CONFIG_I2C) || defined(CONFIG_I2C_MODULE)
 	ret = i2c_add_driver(&uda1380_i2c_driver);
 	if (ret != 0)
 		pr_err("Failed to register UDA1380 I2C driver: %d\n", ret);
 #endif
-	return 0;
+	return ret;
 }
 module_init(uda1380_modinit);
 
diff --git a/sound/soc/codecs/wm8731.c b/sound/soc/codecs/wm8731.c
index 7e5ec03..a7c9ae1 100644
--- a/sound/soc/codecs/wm8731.c
+++ b/sound/soc/codecs/wm8731.c
@@ -453,6 +453,7 @@
 		snd_soc_write(codec, WM8731_PWR, 0xffff);
 		regulator_bulk_disable(ARRAY_SIZE(wm8731->supplies),
 				       wm8731->supplies);
+		codec->cache_sync = 1;
 		break;
 	}
 	codec->dapm.bias_level = level;
diff --git a/sound/soc/codecs/wm8753.c b/sound/soc/codecs/wm8753.c
index a950471..3a629d0 100644
--- a/sound/soc/codecs/wm8753.c
+++ b/sound/soc/codecs/wm8753.c
@@ -190,6 +190,9 @@
 	struct wm8753_priv *wm8753 = snd_soc_codec_get_drvdata(codec);
 	u16 ioctl;
 
+	if (wm8753->dai_func == ucontrol->value.integer.value[0])
+		return 0;
+
 	if (codec->active)
 		return -EBUSY;
 
diff --git a/sound/soc/codecs/wm8958-dsp2.c b/sound/soc/codecs/wm8958-dsp2.c
index 0293763..5a14d5c 100644
--- a/sound/soc/codecs/wm8958-dsp2.c
+++ b/sound/soc/codecs/wm8958-dsp2.c
@@ -60,6 +60,8 @@
 	}
 
 	if (memcmp(fw->data, "WMFW", 4) != 0) {
+		memcpy(&data32, fw->data, sizeof(data32));
+		data32 = be32_to_cpu(data32);
 		dev_err(codec->dev, "%s: firmware has bad file magic %08x\n",
 			name, data32);
 		goto err;
diff --git a/sound/soc/codecs/wm8962.c b/sound/soc/codecs/wm8962.c
index 91d3c6d..53edd9a 100644
--- a/sound/soc/codecs/wm8962.c
+++ b/sound/soc/codecs/wm8962.c
@@ -1973,7 +1973,7 @@
 static const DECLARE_TLV_DB_SCALE(inpga_tlv, -2325, 75, 0);
 static const DECLARE_TLV_DB_SCALE(mixin_tlv, -1500, 300, 0);
 static const unsigned int mixinpga_tlv[] = {
-	TLV_DB_RANGE_HEAD(7),
+	TLV_DB_RANGE_HEAD(5),
 	0, 1, TLV_DB_SCALE_ITEM(0, 600, 0),
 	2, 2, TLV_DB_SCALE_ITEM(1300, 1300, 0),
 	3, 4, TLV_DB_SCALE_ITEM(1800, 200, 0),
@@ -1988,7 +1988,7 @@
 static const DECLARE_TLV_DB_SCALE(out_tlv, -12100, 100, 1);
 static const DECLARE_TLV_DB_SCALE(hp_tlv, -700, 100, 0);
 static const unsigned int classd_tlv[] = {
-	TLV_DB_RANGE_HEAD(7),
+	TLV_DB_RANGE_HEAD(2),
 	0, 6, TLV_DB_SCALE_ITEM(0, 150, 0),
 	7, 7, TLV_DB_SCALE_ITEM(1200, 0, 0),
 };
diff --git a/sound/soc/codecs/wm8993.c b/sound/soc/codecs/wm8993.c
index eec8e14..d1a142f4 100644
--- a/sound/soc/codecs/wm8993.c
+++ b/sound/soc/codecs/wm8993.c
@@ -512,7 +512,7 @@
 static const DECLARE_TLV_DB_SCALE(drc_comp_amp, -2250, 75, 0);
 static const DECLARE_TLV_DB_SCALE(drc_min_tlv, -1800, 600, 0);
 static const unsigned int drc_max_tlv[] = {
-	TLV_DB_RANGE_HEAD(4),
+	TLV_DB_RANGE_HEAD(2),
 	0, 2, TLV_DB_SCALE_ITEM(1200, 600, 0),
 	3, 3, TLV_DB_SCALE_ITEM(3600, 0, 0),
 };
diff --git a/sound/soc/codecs/wm8994.c b/sound/soc/codecs/wm8994.c
index 6b73efd..d0c545b 100644
--- a/sound/soc/codecs/wm8994.c
+++ b/sound/soc/codecs/wm8994.c
@@ -56,7 +56,7 @@
 static int wm8994_readable(struct snd_soc_codec *codec, unsigned int reg)
 {
 	struct wm8994_priv *wm8994 = snd_soc_codec_get_drvdata(codec);
-	struct wm8994 *control = wm8994->control_data;
+	struct wm8994 *control = codec->control_data;
 
 	switch (reg) {
 	case WM8994_GPIO_1:
@@ -1325,15 +1325,15 @@
 };
 
 static const struct snd_soc_dapm_widget wm8994_adc_revd_widgets[] = {
-SND_SOC_DAPM_MUX_E("ADCL Mux", WM8994_POWER_MANAGEMENT_4, 1, 0, &adcl_mux,
-		   adc_mux_ev, SND_SOC_DAPM_PRE_PMU),
-SND_SOC_DAPM_MUX_E("ADCR Mux", WM8994_POWER_MANAGEMENT_4, 0, 0, &adcr_mux,
-		   adc_mux_ev, SND_SOC_DAPM_PRE_PMU),
+SND_SOC_DAPM_VIRT_MUX_E("ADCL Mux", WM8994_POWER_MANAGEMENT_4, 1, 0, &adcl_mux,
+			adc_mux_ev, SND_SOC_DAPM_PRE_PMU),
+SND_SOC_DAPM_VIRT_MUX_E("ADCR Mux", WM8994_POWER_MANAGEMENT_4, 0, 0, &adcr_mux,
+			adc_mux_ev, SND_SOC_DAPM_PRE_PMU),
 };
 
 static const struct snd_soc_dapm_widget wm8994_adc_widgets[] = {
-SND_SOC_DAPM_MUX("ADCL Mux", WM8994_POWER_MANAGEMENT_4, 1, 0, &adcl_mux),
-SND_SOC_DAPM_MUX("ADCR Mux", WM8994_POWER_MANAGEMENT_4, 0, 0, &adcr_mux),
+SND_SOC_DAPM_VIRT_MUX("ADCL Mux", WM8994_POWER_MANAGEMENT_4, 1, 0, &adcl_mux),
+SND_SOC_DAPM_VIRT_MUX("ADCR Mux", WM8994_POWER_MANAGEMENT_4, 0, 0, &adcr_mux),
 };
 
 static const struct snd_soc_dapm_widget wm8994_dapm_widgets[] = {
@@ -2357,6 +2357,11 @@
 	bclk |= best << WM8994_AIF1_BCLK_DIV_SHIFT;
 
 	lrclk = bclk_rate / params_rate(params);
+	if (!lrclk) {
+		dev_err(dai->dev, "Unable to generate LRCLK from %dHz BCLK\n",
+			bclk_rate);
+		return -EINVAL;
+	}
 	dev_dbg(dai->dev, "Using LRCLK rate %d for actual LRCLK %dHz\n",
 		lrclk, bclk_rate / lrclk);
 
@@ -3030,19 +3035,34 @@
 {
 	struct wm8994_priv *wm8994 = data;
 	struct snd_soc_codec *codec = wm8994->codec;
-	int reg;
+	int reg, count;
 
-	reg = snd_soc_read(codec, WM8958_MIC_DETECT_3);
-	if (reg < 0) {
-		dev_err(codec->dev, "Failed to read mic detect status: %d\n",
-			reg);
-		return IRQ_NONE;
-	}
+	/* We may occasionally read a detection without an impedence
+	 * range being provided - if that happens loop again.
+	 */
+	count = 10;
+	do {
+		reg = snd_soc_read(codec, WM8958_MIC_DETECT_3);
+		if (reg < 0) {
+			dev_err(codec->dev,
+				"Failed to read mic detect status: %d\n",
+				reg);
+			return IRQ_NONE;
+		}
 
-	if (!(reg & WM8958_MICD_VALID)) {
-		dev_dbg(codec->dev, "Mic detect data not valid\n");
-		goto out;
-	}
+		if (!(reg & WM8958_MICD_VALID)) {
+			dev_dbg(codec->dev, "Mic detect data not valid\n");
+			goto out;
+		}
+
+		if (!(reg & WM8958_MICD_STS) || (reg & WM8958_MICD_LVL_MASK))
+			break;
+
+		msleep(1);
+	} while (count--);
+
+	if (count == 0)
+		dev_warn(codec->dev, "No impedence range reported for jack\n");
 
 #ifndef CONFIG_SND_SOC_WM8994_MODULE
 	trace_snd_soc_jack_irq(dev_name(codec->dev));
@@ -3163,6 +3183,8 @@
 		switch (wm8994->revision) {
 		case 0:
 		case 1:
+		case 2:
+		case 3:
 			wm8994->hubs.dcs_codes_l = -9;
 			wm8994->hubs.dcs_codes_r = -5;
 			break;
@@ -3180,9 +3202,9 @@
 
 	wm8994_request_irq(codec->control_data, WM8994_IRQ_FIFOS_ERR,
 			   wm8994_fifo_error, "FIFO error", codec);
-	wm8994_request_irq(wm8994->control_data, WM8994_IRQ_TEMP_WARN,
+	wm8994_request_irq(codec->control_data, WM8994_IRQ_TEMP_WARN,
 			   wm8994_temp_warn, "Thermal warning", codec);
-	wm8994_request_irq(wm8994->control_data, WM8994_IRQ_TEMP_SHUT,
+	wm8994_request_irq(codec->control_data, WM8994_IRQ_TEMP_SHUT,
 			   wm8994_temp_shut, "Thermal shutdown", codec);
 
 	ret = wm8994_request_irq(codec->control_data, WM8994_IRQ_DCS_DONE,
diff --git a/sound/soc/codecs/wm8996.c b/sound/soc/codecs/wm8996.c
index 645c980..a33b04d 100644
--- a/sound/soc/codecs/wm8996.c
+++ b/sound/soc/codecs/wm8996.c
@@ -1968,6 +1968,7 @@
 		break;
 	case 24576000:
 		ratediv = WM8996_SYSCLK_DIV;
+		wm8996->sysclk /= 2;
 	case 12288000:
 		snd_soc_update_bits(codec, WM8996_AIF_RATE,
 				    WM8996_SYSCLK_RATE, WM8996_SYSCLK_RATE);
diff --git a/sound/soc/codecs/wm9081.c b/sound/soc/codecs/wm9081.c
index 3cd35a0..4a398c3 100644
--- a/sound/soc/codecs/wm9081.c
+++ b/sound/soc/codecs/wm9081.c
@@ -807,7 +807,6 @@
 			mdelay(100);
 
 			/* Normal bias enable & soft start off */
-			reg |= WM9081_BIAS_ENA;
 			reg &= ~WM9081_VMID_RAMP;
 			snd_soc_write(codec, WM9081_VMID_CONTROL, reg);
 
@@ -818,7 +817,7 @@
 		}
 
 		/* VMID 2*240k */
-		reg = snd_soc_read(codec, WM9081_BIAS_CONTROL_1);
+		reg = snd_soc_read(codec, WM9081_VMID_CONTROL);
 		reg &= ~WM9081_VMID_SEL_MASK;
 		reg |= 0x04;
 		snd_soc_write(codec, WM9081_VMID_CONTROL, reg);
@@ -830,14 +829,15 @@
 		break;
 
 	case SND_SOC_BIAS_OFF:
-		/* Startup bias source */
+		/* Startup bias source and disable bias */
 		reg = snd_soc_read(codec, WM9081_BIAS_CONTROL_1);
 		reg |= WM9081_BIAS_SRC;
+		reg &= ~WM9081_BIAS_ENA;
 		snd_soc_write(codec, WM9081_BIAS_CONTROL_1, reg);
 
-		/* Disable VMID and biases with soft ramping */
+		/* Disable VMID with soft ramping */
 		reg = snd_soc_read(codec, WM9081_VMID_CONTROL);
-		reg &= ~(WM9081_VMID_SEL_MASK | WM9081_BIAS_ENA);
+		reg &= ~WM9081_VMID_SEL_MASK;
 		reg |= WM9081_VMID_RAMP;
 		snd_soc_write(codec, WM9081_VMID_CONTROL, reg);
 
diff --git a/sound/soc/codecs/wm9090.c b/sound/soc/codecs/wm9090.c
index 2b5252c..f94c060 100644
--- a/sound/soc/codecs/wm9090.c
+++ b/sound/soc/codecs/wm9090.c
@@ -177,19 +177,19 @@
 }
 
 static const unsigned int in_tlv[] = {
-	TLV_DB_RANGE_HEAD(6),
+	TLV_DB_RANGE_HEAD(3),
 	0, 0, TLV_DB_SCALE_ITEM(-600, 0, 0),
 	1, 3, TLV_DB_SCALE_ITEM(-350, 350, 0),
 	4, 6, TLV_DB_SCALE_ITEM(600, 600, 0),
 };
 static const unsigned int mix_tlv[] = {
-	TLV_DB_RANGE_HEAD(4),
+	TLV_DB_RANGE_HEAD(2),
 	0, 2, TLV_DB_SCALE_ITEM(-1200, 300, 0),
 	3, 3, TLV_DB_SCALE_ITEM(0, 0, 0),
 };
 static const DECLARE_TLV_DB_SCALE(out_tlv, -5700, 100, 0);
 static const unsigned int spkboost_tlv[] = {
-	TLV_DB_RANGE_HEAD(7),
+	TLV_DB_RANGE_HEAD(2),
 	0, 6, TLV_DB_SCALE_ITEM(0, 150, 0),
 	7, 7, TLV_DB_SCALE_ITEM(1200, 0, 0),
 };
diff --git a/sound/soc/codecs/wm_hubs.c b/sound/soc/codecs/wm_hubs.c
index 84f33d4..48e61e9 100644
--- a/sound/soc/codecs/wm_hubs.c
+++ b/sound/soc/codecs/wm_hubs.c
@@ -40,7 +40,7 @@
 static const DECLARE_TLV_DB_SCALE(spkmixout_tlv, -1800, 600, 1);
 static const DECLARE_TLV_DB_SCALE(outpga_tlv, -5700, 100, 0);
 static const unsigned int spkboost_tlv[] = {
-	TLV_DB_RANGE_HEAD(7),
+	TLV_DB_RANGE_HEAD(2),
 	0, 6, TLV_DB_SCALE_ITEM(0, 150, 0),
 	7, 7, TLV_DB_SCALE_ITEM(1200, 0, 0),
 };
diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c
index 0268cf9..83c4bd5 100644
--- a/sound/soc/fsl/fsl_ssi.c
+++ b/sound/soc/fsl/fsl_ssi.c
@@ -694,6 +694,7 @@
 
 	/* Initialize the the device_attribute structure */
 	dev_attr = &ssi_private->dev_attr;
+	sysfs_attr_init(&dev_attr->attr);
 	dev_attr->attr.name = "statistics";
 	dev_attr->attr.mode = S_IRUGO;
 	dev_attr->show = fsl_sysfs_ssi_show;
diff --git a/sound/soc/fsl/mpc8610_hpcd.c b/sound/soc/fsl/mpc8610_hpcd.c
index 31af405..ae49f1c 100644
--- a/sound/soc/fsl/mpc8610_hpcd.c
+++ b/sound/soc/fsl/mpc8610_hpcd.c
@@ -392,7 +392,8 @@
 	}
 
 	if (strcasecmp(sprop, "i2s-slave") == 0) {
-		machine_data->dai_format = SND_SOC_DAIFMT_I2S;
+		machine_data->dai_format =
+			SND_SOC_DAIFMT_I2S | SND_SOC_DAIFMT_CBM_CFM;
 		machine_data->codec_clk_direction = SND_SOC_CLOCK_OUT;
 		machine_data->cpu_clk_direction = SND_SOC_CLOCK_IN;
 
@@ -409,31 +410,38 @@
 		}
 		machine_data->clk_frequency = be32_to_cpup(iprop);
 	} else if (strcasecmp(sprop, "i2s-master") == 0) {
-		machine_data->dai_format = SND_SOC_DAIFMT_I2S;
+		machine_data->dai_format =
+			SND_SOC_DAIFMT_I2S | SND_SOC_DAIFMT_CBS_CFS;
 		machine_data->codec_clk_direction = SND_SOC_CLOCK_IN;
 		machine_data->cpu_clk_direction = SND_SOC_CLOCK_OUT;
 	} else if (strcasecmp(sprop, "lj-slave") == 0) {
-		machine_data->dai_format = SND_SOC_DAIFMT_LEFT_J;
+		machine_data->dai_format =
+			SND_SOC_DAIFMT_LEFT_J | SND_SOC_DAIFMT_CBM_CFM;
 		machine_data->codec_clk_direction = SND_SOC_CLOCK_OUT;
 		machine_data->cpu_clk_direction = SND_SOC_CLOCK_IN;
 	} else if (strcasecmp(sprop, "lj-master") == 0) {
-		machine_data->dai_format = SND_SOC_DAIFMT_LEFT_J;
+		machine_data->dai_format =
+			SND_SOC_DAIFMT_LEFT_J | SND_SOC_DAIFMT_CBS_CFS;
 		machine_data->codec_clk_direction = SND_SOC_CLOCK_IN;
 		machine_data->cpu_clk_direction = SND_SOC_CLOCK_OUT;
 	} else if (strcasecmp(sprop, "rj-slave") == 0) {
-		machine_data->dai_format = SND_SOC_DAIFMT_RIGHT_J;
+		machine_data->dai_format =
+			SND_SOC_DAIFMT_RIGHT_J | SND_SOC_DAIFMT_CBM_CFM;
 		machine_data->codec_clk_direction = SND_SOC_CLOCK_OUT;
 		machine_data->cpu_clk_direction = SND_SOC_CLOCK_IN;
 	} else if (strcasecmp(sprop, "rj-master") == 0) {
-		machine_data->dai_format = SND_SOC_DAIFMT_RIGHT_J;
+		machine_data->dai_format =
+			SND_SOC_DAIFMT_RIGHT_J | SND_SOC_DAIFMT_CBS_CFS;
 		machine_data->codec_clk_direction = SND_SOC_CLOCK_IN;
 		machine_data->cpu_clk_direction = SND_SOC_CLOCK_OUT;
 	} else if (strcasecmp(sprop, "ac97-slave") == 0) {
-		machine_data->dai_format = SND_SOC_DAIFMT_AC97;
+		machine_data->dai_format =
+			SND_SOC_DAIFMT_AC97 | SND_SOC_DAIFMT_CBM_CFM;
 		machine_data->codec_clk_direction = SND_SOC_CLOCK_OUT;
 		machine_data->cpu_clk_direction = SND_SOC_CLOCK_IN;
 	} else if (strcasecmp(sprop, "ac97-master") == 0) {
-		machine_data->dai_format = SND_SOC_DAIFMT_AC97;
+		machine_data->dai_format =
+			SND_SOC_DAIFMT_AC97 | SND_SOC_DAIFMT_CBS_CFS;
 		machine_data->codec_clk_direction = SND_SOC_CLOCK_IN;
 		machine_data->cpu_clk_direction = SND_SOC_CLOCK_OUT;
 	} else {
diff --git a/sound/soc/imx/Kconfig b/sound/soc/imx/Kconfig
index b133bfc..7383917 100644
--- a/sound/soc/imx/Kconfig
+++ b/sound/soc/imx/Kconfig
@@ -28,7 +28,7 @@
 
 config SND_SOC_MX27VIS_AIC32X4
 	tristate "SoC audio support for Visstrim M10 boards"
-	depends on MACH_IMX27_VISSTRIM_M10
+	depends on MACH_IMX27_VISSTRIM_M10 && I2C
 	select SND_SOC_TLV320AIC32X4
 	select SND_MXC_SOC_MX2
 	help
diff --git a/sound/soc/kirkwood/Kconfig b/sound/soc/kirkwood/Kconfig
index 8f49e16..c62d715 100644
--- a/sound/soc/kirkwood/Kconfig
+++ b/sound/soc/kirkwood/Kconfig
@@ -12,6 +12,7 @@
 config SND_KIRKWOOD_SOC_OPENRD
 	tristate "SoC Audio support for Kirkwood Openrd Client"
 	depends on SND_KIRKWOOD_SOC && (MACH_OPENRD_CLIENT || MACH_OPENRD_ULTIMATE)
+	depends on I2C
 	select SND_KIRKWOOD_SOC_I2S
 	select SND_SOC_CS42L51
 	help
@@ -20,7 +21,7 @@
 
 config SND_KIRKWOOD_SOC_T5325
 	tristate "SoC Audio support for HP t5325"
-	depends on SND_KIRKWOOD_SOC && MACH_T5325
+	depends on SND_KIRKWOOD_SOC && MACH_T5325 && I2C
 	select SND_KIRKWOOD_SOC_I2S
 	select SND_SOC_ALC5623
 	help
diff --git a/sound/soc/mxs/mxs-pcm.c b/sound/soc/mxs/mxs-pcm.c
index dea5aa4..f39d7dd 100644
--- a/sound/soc/mxs/mxs-pcm.c
+++ b/sound/soc/mxs/mxs-pcm.c
@@ -357,3 +357,6 @@
 	platform_driver_unregister(&mxs_pcm_driver);
 }
 module_exit(snd_mxs_pcm_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:mxs-pcm-audio");
diff --git a/sound/soc/mxs/mxs-sgtl5000.c b/sound/soc/mxs/mxs-sgtl5000.c
index 7fbeaec..1c57f66 100644
--- a/sound/soc/mxs/mxs-sgtl5000.c
+++ b/sound/soc/mxs/mxs-sgtl5000.c
@@ -171,3 +171,4 @@
 MODULE_AUTHOR("Freescale Semiconductor, Inc.");
 MODULE_DESCRIPTION("MXS ALSA SoC Machine driver");
 MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:mxs-sgtl5000");
diff --git a/sound/soc/nuc900/nuc900-ac97.c b/sound/soc/nuc900/nuc900-ac97.c
index 9c0edad..a4e3237 100644
--- a/sound/soc/nuc900/nuc900-ac97.c
+++ b/sound/soc/nuc900/nuc900-ac97.c
@@ -365,7 +365,8 @@
 	if (ret)
 		goto out3;
 
-	mfp_set_groupg(nuc900_audio->dev); /* enbale ac97 multifunction pin*/
+	/* enbale ac97 multifunction pin */
+	mfp_set_groupg(nuc900_audio->dev, "nuc900-audio");
 
 	return 0;
 
diff --git a/sound/soc/pxa/Kconfig b/sound/soc/pxa/Kconfig
index ffd2242..a0f7d3c 100644
--- a/sound/soc/pxa/Kconfig
+++ b/sound/soc/pxa/Kconfig
@@ -151,6 +151,7 @@
 config SND_SOC_RAUMFELD
 	tristate "SoC Audio support Raumfeld audio adapter"
 	depends on SND_PXA2XX_SOC && (MACH_RAUMFELD_SPEAKER || MACH_RAUMFELD_CONNECTOR)
+	depends on I2C && SPI_MASTER
 	select SND_PXA_SOC_SSP
 	select SND_SOC_CS4270
 	select SND_SOC_AK4104
@@ -159,7 +160,7 @@
 
 config SND_PXA2XX_SOC_HX4700
 	tristate "SoC Audio support for HP iPAQ hx4700"
-	depends on SND_PXA2XX_SOC && MACH_H4700
+	depends on SND_PXA2XX_SOC && MACH_H4700 && I2C
 	select SND_PXA2XX_SOC_I2S
 	select SND_SOC_AK4641
 	help
diff --git a/sound/soc/pxa/hx4700.c b/sound/soc/pxa/hx4700.c
index 65c1248..c664e33 100644
--- a/sound/soc/pxa/hx4700.c
+++ b/sound/soc/pxa/hx4700.c
@@ -209,9 +209,10 @@
 	snd_soc_card_hx4700.dev = &pdev->dev;
 	ret = snd_soc_register_card(&snd_soc_card_hx4700);
 	if (ret)
-		return ret;
+		gpio_free_array(hx4700_audio_gpios,
+				ARRAY_SIZE(hx4700_audio_gpios));
 
-	return 0;
+	return ret;
 }
 
 static int __devexit hx4700_audio_remove(struct platform_device *pdev)
diff --git a/sound/soc/samsung/jive_wm8750.c b/sound/soc/samsung/jive_wm8750.c
index 1826acf..8e523fd 100644
--- a/sound/soc/samsung/jive_wm8750.c
+++ b/sound/soc/samsung/jive_wm8750.c
@@ -101,7 +101,6 @@
 {
 	struct snd_soc_codec *codec = rtd->codec;
 	struct snd_soc_dapm_context *dapm = &codec->dapm;
-	int err;
 
 	/* These endpoints are not being used. */
 	snd_soc_dapm_nc_pin(dapm, "LINPUT2");
@@ -131,7 +130,7 @@
 	.dai_link	= &jive_dai,
 	.num_links	= 1,
 
-	.dapm_widgtets	= wm8750_dapm_widgets,
+	.dapm_widgets	= wm8750_dapm_widgets,
 	.num_dapm_widgets = ARRAY_SIZE(wm8750_dapm_widgets),
 	.dapm_routes	= audio_map,
 	.num_dapm_routes = ARRAY_SIZE(audio_map),
diff --git a/sound/soc/samsung/smdk2443_wm9710.c b/sound/soc/samsung/smdk2443_wm9710.c
index 3a0dbfc..8bd1dc5 100644
--- a/sound/soc/samsung/smdk2443_wm9710.c
+++ b/sound/soc/samsung/smdk2443_wm9710.c
@@ -12,6 +12,7 @@
  *
  */
 
+#include <linux/module.h>
 #include <sound/soc.h>
 
 static struct snd_soc_card smdk2443;
diff --git a/sound/soc/samsung/smdk_wm8994.c b/sound/soc/samsung/smdk_wm8994.c
index f75e439..ad9ac42 100644
--- a/sound/soc/samsung/smdk_wm8994.c
+++ b/sound/soc/samsung/smdk_wm8994.c
@@ -9,6 +9,7 @@
 
 #include "../codecs/wm8994.h"
 #include <sound/pcm_params.h>
+#include <linux/module.h>
 
  /*
   * Default CFG switch settings to use this driver:
diff --git a/sound/soc/samsung/speyside.c b/sound/soc/samsung/speyside.c
index 85bf541..4b8e354 100644
--- a/sound/soc/samsung/speyside.c
+++ b/sound/soc/samsung/speyside.c
@@ -191,7 +191,7 @@
 	snd_soc_dapm_ignore_suspend(&card->dapm, "Headset Mic");
 	snd_soc_dapm_ignore_suspend(&card->dapm, "Main AMIC");
 	snd_soc_dapm_ignore_suspend(&card->dapm, "Main DMIC");
-	snd_soc_dapm_ignore_suspend(&card->dapm, "Speaker");
+	snd_soc_dapm_ignore_suspend(&card->dapm, "Main Speaker");
 	snd_soc_dapm_ignore_suspend(&card->dapm, "WM1250 Output");
 	snd_soc_dapm_ignore_suspend(&card->dapm, "WM1250 Input");
 
diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c
index a5d3685..a25fa63 100644
--- a/sound/soc/soc-core.c
+++ b/sound/soc/soc-core.c
@@ -709,6 +709,12 @@
 	struct snd_soc_card *card = dev_get_drvdata(dev);
 	int i, ac97_control = 0;
 
+	/* If the initialization of this soc device failed, there is no codec
+	 * associated with it. Just bail out in this case.
+	 */
+	if (list_empty(&card->codec_dev_list))
+		return 0;
+
 	/* AC97 devices might have other drivers hanging off them so
 	 * need to resume immediately.  Other drivers don't have that
 	 * problem and may take a substantial amount of time to resume
diff --git a/sound/soc/soc-utils.c b/sound/soc/soc-utils.c
index 0c12b98..4220bb0 100644
--- a/sound/soc/soc-utils.c
+++ b/sound/soc/soc-utils.c
@@ -58,7 +58,36 @@
 }
 EXPORT_SYMBOL_GPL(snd_soc_params_to_bclk);
 
-static struct snd_soc_platform_driver dummy_platform;
+static const struct snd_pcm_hardware dummy_dma_hardware = {
+	.formats		= 0xffffffff,
+	.channels_min		= 1,
+	.channels_max		= UINT_MAX,
+
+	/* Random values to keep userspace happy when checking constraints */
+	.info			= SNDRV_PCM_INFO_INTERLEAVED |
+				  SNDRV_PCM_INFO_BLOCK_TRANSFER,
+	.buffer_bytes_max	= 128*1024,
+	.period_bytes_min	= PAGE_SIZE,
+	.period_bytes_max	= PAGE_SIZE*2,
+	.periods_min		= 2,
+	.periods_max		= 128,
+};
+
+static int dummy_dma_open(struct snd_pcm_substream *substream)
+{
+	snd_soc_set_runtime_hwparams(substream, &dummy_dma_hardware);
+
+	return 0;
+}
+
+static struct snd_pcm_ops dummy_dma_ops = {
+	.open		= dummy_dma_open,
+	.ioctl		= snd_pcm_lib_ioctl,
+};
+
+static struct snd_soc_platform_driver dummy_platform = {
+	.ops = &dummy_dma_ops,
+};
 
 static __devinit int snd_soc_dummy_probe(struct platform_device *pdev)
 {
diff --git a/sound/usb/mixer.c b/sound/usb/mixer.c
index 60f65ac..ab23869 100644
--- a/sound/usb/mixer.c
+++ b/sound/usb/mixer.c
@@ -765,10 +765,61 @@
  * interface to ALSA control for feature/mixer units
  */
 
+/* volume control quirks */
+static void volume_control_quirks(struct usb_mixer_elem_info *cval,
+				  struct snd_kcontrol *kctl)
+{
+	switch (cval->mixer->chip->usb_id) {
+	case USB_ID(0x0471, 0x0101):
+	case USB_ID(0x0471, 0x0104):
+	case USB_ID(0x0471, 0x0105):
+	case USB_ID(0x0672, 0x1041):
+	/* quirk for UDA1321/N101.
+	 * note that detection between firmware 2.1.1.7 (N101)
+	 * and later 2.1.1.21 is not very clear from datasheets.
+	 * I hope that the min value is -15360 for newer firmware --jk
+	 */
+		if (!strcmp(kctl->id.name, "PCM Playback Volume") &&
+		    cval->min == -15616) {
+			snd_printk(KERN_INFO
+				 "set volume quirk for UDA1321/N101 chip\n");
+			cval->max = -256;
+		}
+		break;
+
+	case USB_ID(0x046d, 0x09a4):
+		if (!strcmp(kctl->id.name, "Mic Capture Volume")) {
+			snd_printk(KERN_INFO
+				"set volume quirk for QuickCam E3500\n");
+			cval->min = 6080;
+			cval->max = 8768;
+			cval->res = 192;
+		}
+		break;
+
+	case USB_ID(0x046d, 0x0808):
+	case USB_ID(0x046d, 0x0809):
+	case USB_ID(0x046d, 0x081d): /* HD Webcam c510 */
+	case USB_ID(0x046d, 0x0991):
+	/* Most audio usb devices lie about volume resolution.
+	 * Most Logitech webcams have res = 384.
+	 * Proboly there is some logitech magic behind this number --fishor
+	 */
+		if (!strcmp(kctl->id.name, "Mic Capture Volume")) {
+			snd_printk(KERN_INFO
+				"set resolution quirk: cval->res = 384\n");
+			cval->res = 384;
+		}
+		break;
+
+	}
+}
+
 /*
  * retrieve the minimum and maximum values for the specified control
  */
-static int get_min_max(struct usb_mixer_elem_info *cval, int default_min)
+static int get_min_max_with_quirks(struct usb_mixer_elem_info *cval,
+				   int default_min, struct snd_kcontrol *kctl)
 {
 	/* for failsafe */
 	cval->min = default_min;
@@ -844,6 +895,9 @@
 		cval->initialized = 1;
 	}
 
+	if (kctl)
+		volume_control_quirks(cval, kctl);
+
 	/* USB descriptions contain the dB scale in 1/256 dB unit
 	 * while ALSA TLV contains in 1/100 dB unit
 	 */
@@ -864,6 +918,7 @@
 	return 0;
 }
 
+#define get_min_max(cval, def)	get_min_max_with_quirks(cval, def, NULL)
 
 /* get a feature/mixer unit info */
 static int mixer_ctl_feature_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo)
@@ -882,7 +937,7 @@
 		uinfo->value.integer.max = 1;
 	} else {
 		if (!cval->initialized) {
-			get_min_max(cval, 0);
+			get_min_max_with_quirks(cval, 0, kcontrol);
 			if (cval->initialized && cval->dBmin >= cval->dBmax) {
 				kcontrol->vd[0].access &= 
 					~(SNDRV_CTL_ELEM_ACCESS_TLV_READ |
@@ -1045,9 +1100,6 @@
 		cval->ch_readonly = readonly_mask;
 	}
 
-	/* get min/max values */
-	get_min_max(cval, 0);
-
 	/* if all channels in the mask are marked read-only, make the control
 	 * read-only. set_cur_mix_value() will check the mask again and won't
 	 * issue write commands to read-only channels. */
@@ -1069,6 +1121,9 @@
 		len = snd_usb_copy_string_desc(state, nameid,
 				kctl->id.name, sizeof(kctl->id.name));
 
+	/* get min/max values */
+	get_min_max_with_quirks(cval, 0, kctl);
+
 	switch (control) {
 	case UAC_FU_MUTE:
 	case UAC_FU_VOLUME:
@@ -1118,51 +1173,6 @@
 		break;
 	}
 
-	/* volume control quirks */
-	switch (state->chip->usb_id) {
-	case USB_ID(0x0471, 0x0101):
-	case USB_ID(0x0471, 0x0104):
-	case USB_ID(0x0471, 0x0105):
-	case USB_ID(0x0672, 0x1041):
-	/* quirk for UDA1321/N101.
-	 * note that detection between firmware 2.1.1.7 (N101)
-	 * and later 2.1.1.21 is not very clear from datasheets.
-	 * I hope that the min value is -15360 for newer firmware --jk
-	 */
-		if (!strcmp(kctl->id.name, "PCM Playback Volume") &&
-		    cval->min == -15616) {
-			snd_printk(KERN_INFO
-				 "set volume quirk for UDA1321/N101 chip\n");
-			cval->max = -256;
-		}
-		break;
-
-	case USB_ID(0x046d, 0x09a4):
-		if (!strcmp(kctl->id.name, "Mic Capture Volume")) {
-			snd_printk(KERN_INFO
-				"set volume quirk for QuickCam E3500\n");
-			cval->min = 6080;
-			cval->max = 8768;
-			cval->res = 192;
-		}
-		break;
-
-	case USB_ID(0x046d, 0x0808):
-	case USB_ID(0x046d, 0x0809):
-	case USB_ID(0x046d, 0x0991):
-	/* Most audio usb devices lie about volume resolution.
-	 * Most Logitech webcams have res = 384.
-	 * Proboly there is some logitech magic behind this number --fishor
-	 */
-		if (!strcmp(kctl->id.name, "Mic Capture Volume")) {
-			snd_printk(KERN_INFO
-				"set resolution quirk: cval->res = 384\n");
-			cval->res = 384;
-		}
-		break;
-
-	}
-
 	range = (cval->max - cval->min) / cval->res;
 	/* Are there devices with volume range more than 255? I use a bit more
 	 * to be sure. 384 is a resolution magic number found on Logitech
diff --git a/sound/usb/quirks-table.h b/sound/usb/quirks-table.h
index b61945f..32d2a21 100644
--- a/sound/usb/quirks-table.h
+++ b/sound/usb/quirks-table.h
@@ -1633,6 +1633,37 @@
 	}
 },
 {
+	/* Roland GAIA SH-01 */
+	USB_DEVICE(0x0582, 0x0111),
+	.driver_info = (unsigned long) &(const struct snd_usb_audio_quirk) {
+		.vendor_name = "Roland",
+		.product_name = "GAIA",
+		.ifnum = QUIRK_ANY_INTERFACE,
+		.type = QUIRK_COMPOSITE,
+		.data = (const struct snd_usb_audio_quirk[]) {
+			{
+				.ifnum = 0,
+				.type = QUIRK_AUDIO_STANDARD_INTERFACE
+			},
+			{
+				.ifnum = 1,
+				.type = QUIRK_AUDIO_STANDARD_INTERFACE
+			},
+			{
+				.ifnum = 2,
+				.type = QUIRK_MIDI_FIXED_ENDPOINT,
+				.data = &(const struct snd_usb_midi_endpoint_info) {
+				.out_cables = 0x0003,
+				.in_cables  = 0x0003
+				}
+			},
+			{
+				.ifnum = -1
+			}
+		}
+	}
+},
+{
 	USB_DEVICE(0x0582, 0x0113),
 	.driver_info = (unsigned long) & (const struct snd_usb_audio_quirk) {
 		/* .vendor_name = "BOSS", */
diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c
index 2e5bc73..a3ddac0 100644
--- a/sound/usb/quirks.c
+++ b/sound/usb/quirks.c
@@ -137,12 +137,12 @@
 		return -ENOMEM;
 	}
 	if (fp->nr_rates > 0) {
-		rate_table = kmalloc(sizeof(int) * fp->nr_rates, GFP_KERNEL);
+		rate_table = kmemdup(fp->rate_table,
+				     sizeof(int) * fp->nr_rates, GFP_KERNEL);
 		if (!rate_table) {
 			kfree(fp);
 			return -ENOMEM;
 		}
-		memcpy(rate_table, fp->rate_table, sizeof(int) * fp->nr_rates);
 		fp->rate_table = rate_table;
 	}
 
@@ -224,10 +224,9 @@
 	if (altsd->bNumEndpoints != 1)
 		return -ENXIO;
 
-	fp = kmalloc(sizeof(*fp), GFP_KERNEL);
+	fp = kmemdup(&ua_format, sizeof(*fp), GFP_KERNEL);
 	if (!fp)
 		return -ENOMEM;
-	memcpy(fp, &ua_format, sizeof(*fp));
 
 	fp->iface = altsd->bInterfaceNumber;
 	fp->endpoint = get_endpoint(alts, 0)->bEndpointAddress;
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 7d98676..955930e 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -463,7 +463,8 @@
 
 	list_for_each_entry(counter, &evsel_list->entries, node) {
 		if (create_perf_stat_counter(counter, first) < 0) {
-			if (errno == EINVAL || errno == ENOSYS || errno == ENOENT) {
+			if (errno == EINVAL || errno == ENOSYS ||
+			    errno == ENOENT || errno == EOPNOTSUPP) {
 				if (verbose)
 					ui__warning("%s event is not supported by the kernel.\n",
 						    event_name(counter));
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index e426264..d7915d4 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -34,6 +34,16 @@
 	return size;
 }
 
+static void hists__init(struct hists *hists)
+{
+	memset(hists, 0, sizeof(*hists));
+	hists->entries_in_array[0] = hists->entries_in_array[1] = RB_ROOT;
+	hists->entries_in = &hists->entries_in_array[0];
+	hists->entries_collapsed = RB_ROOT;
+	hists->entries = RB_ROOT;
+	pthread_mutex_init(&hists->lock, NULL);
+}
+
 void perf_evsel__init(struct perf_evsel *evsel,
 		      struct perf_event_attr *attr, int idx)
 {
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index bcd05d0..33c17a2 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -388,7 +388,7 @@
 		/*
 		 * write event string as passed on cmdline
 		 */
-		ret = do_write_string(fd, attr->name);
+		ret = do_write_string(fd, event_name(attr));
 		if (ret < 0)
 			return ret;
 		/*
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index a36a3fa..abef270 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -1211,13 +1211,3 @@
 
 	return ret;
 }
-
-void hists__init(struct hists *hists)
-{
-	memset(hists, 0, sizeof(*hists));
-	hists->entries_in_array[0] = hists->entries_in_array[1] = RB_ROOT;
-	hists->entries_in = &hists->entries_in_array[0];
-	hists->entries_collapsed = RB_ROOT;
-	hists->entries = RB_ROOT;
-	pthread_mutex_init(&hists->lock, NULL);
-}
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
index c86c1d2..89289c8 100644
--- a/tools/perf/util/hist.h
+++ b/tools/perf/util/hist.h
@@ -63,8 +63,6 @@
 	struct callchain_cursor	callchain_cursor;
 };
 
-void hists__init(struct hists *hists);
-
 struct hist_entry *__hists__add_entry(struct hists *self,
 				      struct addr_location *al,
 				      struct symbol *parent, u64 period);
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index 85c1e6b7..0f4555c 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -1333,6 +1333,10 @@
 	}
 
 	map = cpu_map__new(cpu_list);
+	if (map == NULL) {
+		pr_err("Invalid cpu_list\n");
+		return -1;
+	}
 
 	for (i = 0; i < map->nr; i++) {
 		int cpu = map->map[i];
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
index 0a7ed5b..6c164dc 100644
--- a/tools/perf/util/trace-event-parse.c
+++ b/tools/perf/util/trace-event-parse.c
@@ -1537,6 +1537,8 @@
 	field = malloc_or_die(sizeof(*field));
 
 	type = process_arg(event, field, &token);
+	while (type == EVENT_OP)
+		type = process_op(event, field, &token);
 	if (test_type_token(type, token, EVENT_DELIM, ","))
 		goto out_free;
 
diff --git a/tools/testing/ktest/ktest.pl b/tools/testing/ktest/ktest.pl
index 8d02ccb..8b4c253 100755
--- a/tools/testing/ktest/ktest.pl
+++ b/tools/testing/ktest/ktest.pl
@@ -42,6 +42,7 @@
 $default{"BISECT_SKIP"}		= 1;
 $default{"SUCCESS_LINE"}	= "login:";
 $default{"DETECT_TRIPLE_FAULT"} = 1;
+$default{"NO_INSTALL"}		= 0;
 $default{"BOOTED_TIMEOUT"}	= 1;
 $default{"DIE_ON_FAILURE"}	= 1;
 $default{"SSH_EXEC"}		= "ssh \$SSH_USER\@\$MACHINE \$SSH_COMMAND";
@@ -84,6 +85,7 @@
 my $target;
 my $make;
 my $post_install;
+my $no_install;
 my $noclean;
 my $minconfig;
 my $start_minconfig;
@@ -115,6 +117,7 @@
 my $booted_timeout;
 my $detect_triplefault;
 my $console;
+my $reboot_success_line;
 my $success_line;
 my $stop_after_success;
 my $stop_after_failure;
@@ -130,6 +133,12 @@
 my %variable;
 my %force_config;
 
+# do not force reboots on config problems
+my $no_reboot = 1;
+
+# default variables that can be used
+chomp ($variable{"PWD"} = `pwd`);
+
 $config_help{"MACHINE"} = << "EOF"
  The machine hostname that you will test.
 EOF
@@ -241,6 +250,7 @@
 
 sub get_ktest_config {
     my ($config) = @_;
+    my $ans;
 
     return if (defined($opt{$config}));
 
@@ -254,16 +264,17 @@
 	if (defined($default{$config})) {
 	    print "\[$default{$config}\] ";
 	}
-	$entered_configs{$config} = <STDIN>;
-	$entered_configs{$config} =~ s/^\s*(.*\S)\s*$/$1/;
-	if ($entered_configs{$config} =~ /^\s*$/) {
+	$ans = <STDIN>;
+	$ans =~ s/^\s*(.*\S)\s*$/$1/;
+	if ($ans =~ /^\s*$/) {
 	    if ($default{$config}) {
-		$entered_configs{$config} = $default{$config};
+		$ans = $default{$config};
 	    } else {
 		print "Your answer can not be blank\n";
 		next;
 	    }
 	}
+	$entered_configs{$config} = process_variables($ans);
 	last;
     }
 }
@@ -298,7 +309,7 @@
 }
 
 sub process_variables {
-    my ($value) = @_;
+    my ($value, $remove_undef) = @_;
     my $retval = "";
 
     # We want to check for '\', and it is just easier
@@ -316,6 +327,10 @@
 	$retval = "$retval$begin";
 	if (defined($variable{$var})) {
 	    $retval = "$retval$variable{$var}";
+	} elsif (defined($remove_undef) && $remove_undef) {
+	    # for if statements, any variable that is not defined,
+	    # we simple convert to 0
+	    $retval = "${retval}0";
 	} else {
 	    # put back the origin piece.
 	    $retval = "$retval\$\{$var\}";
@@ -331,10 +346,17 @@
 }
 
 sub set_value {
-    my ($lvalue, $rvalue) = @_;
+    my ($lvalue, $rvalue, $override, $overrides, $name) = @_;
 
     if (defined($opt{$lvalue})) {
-	die "Error: Option $lvalue defined more than once!\n";
+	if (!$override || defined(${$overrides}{$lvalue})) {
+	    my $extra = "";
+	    if ($override) {
+		$extra = "In the same override section!\n";
+	    }
+	    die "$name: $.: Option $lvalue defined more than once!\n$extra";
+	}
+	${$overrides}{$lvalue} = $rvalue;
     }
     if ($rvalue =~ /^\s*$/) {
 	delete $opt{$lvalue};
@@ -355,86 +377,274 @@
     }
 }
 
-sub read_config {
-    my ($config) = @_;
+sub process_compare {
+    my ($lval, $cmp, $rval) = @_;
 
-    open(IN, $config) || die "can't read file $config";
+    # remove whitespace
+
+    $lval =~ s/^\s*//;
+    $lval =~ s/\s*$//;
+
+    $rval =~ s/^\s*//;
+    $rval =~ s/\s*$//;
+
+    if ($cmp eq "==") {
+	return $lval eq $rval;
+    } elsif ($cmp eq "!=") {
+	return $lval ne $rval;
+    }
+
+    my $statement = "$lval $cmp $rval";
+    my $ret = eval $statement;
+
+    # $@ stores error of eval
+    if ($@) {
+	return -1;
+    }
+
+    return $ret;
+}
+
+sub value_defined {
+    my ($val) = @_;
+
+    return defined($variable{$2}) ||
+	defined($opt{$2});
+}
+
+my $d = 0;
+sub process_expression {
+    my ($name, $val) = @_;
+
+    my $c = $d++;
+
+    while ($val =~ s/\(([^\(]*?)\)/\&\&\&\&VAL\&\&\&\&/) {
+	my $express = $1;
+
+	if (process_expression($name, $express)) {
+	    $val =~ s/\&\&\&\&VAL\&\&\&\&/ 1 /;
+	} else {
+	    $val =~ s/\&\&\&\&VAL\&\&\&\&/ 0 /;
+	}
+    }
+
+    $d--;
+    my $OR = "\\|\\|";
+    my $AND = "\\&\\&";
+
+    while ($val =~ s/^(.*?)($OR|$AND)//) {
+	my $express = $1;
+	my $op = $2;
+
+	if (process_expression($name, $express)) {
+	    if ($op eq "||") {
+		return 1;
+	    }
+	} else {
+	    if ($op eq "&&") {
+		return 0;
+	    }
+	}
+    }
+
+    if ($val =~ /(.*)(==|\!=|>=|<=|>|<)(.*)/) {
+	my $ret = process_compare($1, $2, $3);
+	if ($ret < 0) {
+	    die "$name: $.: Unable to process comparison\n";
+	}
+	return $ret;
+    }
+
+    if ($val =~ /^\s*(NOT\s*)?DEFINED\s+(\S+)\s*$/) {
+	if (defined $1) {
+	    return !value_defined($2);
+	} else {
+	    return value_defined($2);
+	}
+    }
+
+    if ($val =~ /^\s*0\s*$/) {
+	return 0;
+    } elsif ($val =~ /^\s*\d+\s*$/) {
+	return 1;
+    }
+
+    die ("$name: $.: Undefined content $val in if statement\n");
+}
+
+sub process_if {
+    my ($name, $value) = @_;
+
+    # Convert variables and replace undefined ones with 0
+    my $val = process_variables($value, 1);
+    my $ret = process_expression $name, $val;
+
+    return $ret;
+}
+
+sub __read_config {
+    my ($config, $current_test_num) = @_;
+
+    my $in;
+    open($in, $config) || die "can't read file $config";
 
     my $name = $config;
     $name =~ s,.*/(.*),$1,;
 
-    my $test_num = 0;
+    my $test_num = $$current_test_num;
     my $default = 1;
     my $repeat = 1;
     my $num_tests_set = 0;
     my $skip = 0;
     my $rest;
+    my $line;
     my $test_case = 0;
+    my $if = 0;
+    my $if_set = 0;
+    my $override = 0;
 
-    while (<IN>) {
+    my %overrides;
+
+    while (<$in>) {
 
 	# ignore blank lines and comments
 	next if (/^\s*$/ || /\s*\#/);
 
-	if (/^\s*TEST_START(.*)/) {
+	if (/^\s*(TEST_START|DEFAULTS)\b(.*)/) {
 
-	    $rest = $1;
+	    my $type = $1;
+	    $rest = $2;
+	    $line = $2;
 
-	    if ($num_tests_set) {
-		die "$name: $.: Can not specify both NUM_TESTS and TEST_START\n";
+	    my $old_test_num;
+	    my $old_repeat;
+	    $override = 0;
+
+	    if ($type eq "TEST_START") {
+
+		if ($num_tests_set) {
+		    die "$name: $.: Can not specify both NUM_TESTS and TEST_START\n";
+		}
+
+		$old_test_num = $test_num;
+		$old_repeat = $repeat;
+
+		$test_num += $repeat;
+		$default = 0;
+		$repeat = 1;
+	    } else {
+		$default = 1;
 	    }
 
-	    my $old_test_num = $test_num;
-	    my $old_repeat = $repeat;
-
-	    $test_num += $repeat;
-	    $default = 0;
-	    $repeat = 1;
-
-	    if ($rest =~ /\s+SKIP(.*)/) {
-		$rest = $1;
+	    # If SKIP is anywhere in the line, the command will be skipped
+	    if ($rest =~ s/\s+SKIP\b//) {
 		$skip = 1;
 	    } else {
 		$test_case = 1;
 		$skip = 0;
 	    }
 
-	    if ($rest =~ /\s+ITERATE\s+(\d+)(.*)$/) {
-		$repeat = $1;
-		$rest = $2;
-		$repeat_tests{"$test_num"} = $repeat;
+	    if ($rest =~ s/\sELSE\b//) {
+		if (!$if) {
+		    die "$name: $.: ELSE found with out matching IF section\n$_";
+		}
+		$if = 0;
+
+		if ($if_set) {
+		    $skip = 1;
+		} else {
+		    $skip = 0;
+		}
 	    }
 
-	    if ($rest =~ /\s+SKIP(.*)/) {
-		$rest = $1;
-		$skip = 1;
+	    if ($rest =~ s/\sIF\s+(.*)//) {
+		if (process_if($name, $1)) {
+		    $if_set = 1;
+		} else {
+		    $skip = 1;
+		}
+		$if = 1;
+	    } else {
+		$if = 0;
+		$if_set = 0;
 	    }
 
-	    if ($rest !~ /^\s*$/) {
-		die "$name: $.: Gargbage found after TEST_START\n$_";
+	    if (!$skip) {
+		if ($type eq "TEST_START") {
+		    if ($rest =~ s/\s+ITERATE\s+(\d+)//) {
+			$repeat = $1;
+			$repeat_tests{"$test_num"} = $repeat;
+		    }
+		} elsif ($rest =~ s/\sOVERRIDE\b//) {
+		    # DEFAULT only
+		    $override = 1;
+		    # Clear previous overrides
+		    %overrides = ();
+		}
 	    }
 
-	    if ($skip) {
+	    if (!$skip && $rest !~ /^\s*$/) {
+		die "$name: $.: Gargbage found after $type\n$_";
+	    }
+
+	    if ($skip && $type eq "TEST_START") {
 		$test_num = $old_test_num;
 		$repeat = $old_repeat;
 	    }
 
-	} elsif (/^\s*DEFAULTS(.*)$/) {
-	    $default = 1;
-
+	} elsif (/^\s*ELSE\b(.*)$/) {
+	    if (!$if) {
+		die "$name: $.: ELSE found with out matching IF section\n$_";
+	    }
 	    $rest = $1;
-
-	    if ($rest =~ /\s+SKIP(.*)/) {
-		$rest = $1;
+	    if ($if_set) {
 		$skip = 1;
+		$rest = "";
 	    } else {
 		$skip = 0;
+
+		if ($rest =~ /\sIF\s+(.*)/) {
+		    # May be a ELSE IF section.
+		    if (!process_if($name, $1)) {
+			$skip = 1;
+		    }
+		    $rest = "";
+		} else {
+		    $if = 0;
+		}
 	    }
 
 	    if ($rest !~ /^\s*$/) {
 		die "$name: $.: Gargbage found after DEFAULTS\n$_";
 	    }
 
+	} elsif (/^\s*INCLUDE\s+(\S+)/) {
+
+	    next if ($skip);
+
+	    if (!$default) {
+		die "$name: $.: INCLUDE can only be done in default sections\n$_";
+	    }
+
+	    my $file = process_variables($1);
+
+	    if ($file !~ m,^/,) {
+		# check the path of the config file first
+		if ($config =~ m,(.*)/,) {
+		    if (-f "$1/$file") {
+			$file = "$1/$file";
+		    }
+		}
+	    }
+		
+	    if ( ! -r $file ) {
+		die "$name: $.: Can't read file $file\n$_";
+	    }
+
+	    if (__read_config($file, \$test_num)) {
+		$test_case = 1;
+	    }
+
 	} elsif (/^\s*([A-Z_\[\]\d]+)\s*=\s*(.*?)\s*$/) {
 
 	    next if ($skip);
@@ -460,10 +670,10 @@
 	    }
 
 	    if ($default || $lvalue =~ /\[\d+\]$/) {
-		set_value($lvalue, $rvalue);
+		set_value($lvalue, $rvalue, $override, \%overrides, $name);
 	    } else {
 		my $val = "$lvalue\[$test_num\]";
-		set_value($val, $rvalue);
+		set_value($val, $rvalue, $override, \%overrides, $name);
 
 		if ($repeat > 1) {
 		    $repeats{$val} = $repeat;
@@ -490,13 +700,26 @@
 	}
     }
 
-    close(IN);
-
     if ($test_num) {
 	$test_num += $repeat - 1;
 	$opt{"NUM_TESTS"} = $test_num;
     }
 
+    close($in);
+
+    $$current_test_num = $test_num;
+
+    return $test_case;
+}
+
+sub read_config {
+    my ($config) = @_;
+
+    my $test_case;
+    my $test_num = 0;
+
+    $test_case = __read_config $config, \$test_num;
+
     # make sure we have all mandatory configs
     get_ktest_configs;
 
@@ -524,6 +747,18 @@
     # Add space to evaluate the character before $
     $option = " $option";
     my $retval = "";
+    my $repeated = 0;
+    my $parent = 0;
+
+    foreach my $test (keys %repeat_tests) {
+	if ($i >= $test &&
+	    $i < $test + $repeat_tests{$test}) {
+
+	    $repeated = 1;
+	    $parent = $test;
+	    last;
+	}
+    }
 
     while ($option =~ /(.*?[^\\])\$\{(.*?)\}(.*)/) {
 	my $start = $1;
@@ -537,10 +772,14 @@
 	# otherwise see if the default OPT (without [$i]) exists.
 
 	my $o = "$var\[$i\]";
+	my $parento = "$var\[$parent\]";
 
 	if (defined($opt{$o})) {
 	    $o = $opt{$o};
 	    $retval = "$retval$o";
+	} elsif ($repeated && defined($opt{$parento})) {
+	    $o = $opt{$parento};
+	    $retval = "$retval$o";
 	} elsif (defined($opt{$var})) {
 	    $o = $opt{$var};
 	    $retval = "$retval$o";
@@ -603,8 +842,20 @@
 }
 
 sub run_command;
+sub start_monitor;
+sub end_monitor;
+sub wait_for_monitor;
 
 sub reboot {
+    my ($time) = @_;
+
+    if (defined($time)) {
+	start_monitor;
+	# flush out current monitor
+	# May contain the reboot success line
+	wait_for_monitor 1;
+    }
+
     # try to reboot normally
     if (run_command $reboot) {
 	if (defined($powercycle_after_reboot)) {
@@ -615,12 +866,17 @@
 	# nope? power cycle it.
 	run_command "$power_cycle";
     }
+
+    if (defined($time)) {
+	wait_for_monitor($time, $reboot_success_line);
+	end_monitor;
+    }
 }
 
 sub do_not_reboot {
     my $i = $iteration;
 
-    return $test_type eq "build" ||
+    return $test_type eq "build" || $no_reboot ||
 	($test_type eq "patchcheck" && $opt{"PATCHCHECK_TYPE[$i]"} eq "build") ||
 	($test_type eq "bisect" && $opt{"BISECT_TYPE[$i]"} eq "build");
 }
@@ -693,16 +949,29 @@
 }
 
 sub wait_for_monitor {
-    my ($time) = @_;
+    my ($time, $stop) = @_;
+    my $full_line = "";
     my $line;
+    my $booted = 0;
 
     doprint "** Wait for monitor to settle down **\n";
 
     # read the monitor and wait for the system to calm down
-    do {
+    while (!$booted) {
 	$line = wait_for_input($monitor_fp, $time);
-	print "$line" if (defined($line));
-    } while (defined($line));
+	last if (!defined($line));
+	print "$line";
+	$full_line .= $line;
+
+	if (defined($stop) && $full_line =~ /$stop/) {
+	    doprint "wait for monitor detected $stop\n";
+	    $booted = 1;
+	}
+
+	if ($line =~ /\n/) {
+	    $full_line = "";
+	}
+    }
     print "** Monitor flushed **\n";
 }
 
@@ -719,10 +988,7 @@
 	# no need to reboot for just building.
 	if (!do_not_reboot) {
 	    doprint "REBOOTING\n";
-	    reboot;
-	    start_monitor;
-	    wait_for_monitor $sleep_time;
-	    end_monitor;
+	    reboot $sleep_time;
 	}
 
 	my $name = "";
@@ -854,9 +1120,12 @@
     open(IN, "$ssh_grub |")
 	or die "unable to get menu.lst";
 
+    my $found = 0;
+
     while (<IN>) {
 	if (/^\s*title\s+$grub_menu\s*$/) {
 	    $grub_number++;
+	    $found = 1;
 	    last;
 	} elsif (/^\s*title\s/) {
 	    $grub_number++;
@@ -865,7 +1134,7 @@
     close(IN);
 
     die "Could not find '$grub_menu' in /boot/grub/menu on $machine"
-	if ($grub_number < 0);
+	if (!$found);
     doprint "$grub_number\n";
 }
 
@@ -902,7 +1171,8 @@
 
 sub reboot_to {
     if ($reboot_type eq "grub") {
-	run_ssh "'(echo \"savedefault --default=$grub_number --once\" | grub --batch && reboot)'";
+	run_ssh "'(echo \"savedefault --default=$grub_number --once\" | grub --batch)'";
+	reboot;
 	return;
     }
 
@@ -1083,6 +1353,8 @@
 
 sub install {
 
+    return if ($no_install);
+
     run_scp "$outputdir/$build_target", "$target_image" or
 	dodie "failed to copy image";
 
@@ -1140,6 +1412,11 @@
 }
 
 sub start_monitor_and_boot {
+    # Make sure the stable kernel has finished booting
+    start_monitor;
+    wait_for_monitor 5;
+    end_monitor;
+
     get_grub_index;
     get_version;
     install;
@@ -1250,6 +1527,10 @@
 
     unlink $buildlog;
 
+    # Failed builds should not reboot the target
+    my $save_no_reboot = $no_reboot;
+    $no_reboot = 1;
+
     if (defined($pre_build)) {
 	my $ret = run_command $pre_build;
 	if (!$ret && defined($pre_build_die) &&
@@ -1272,16 +1553,16 @@
 	# allow for empty configs
 	run_command "touch $output_config";
 
-	run_command "mv $output_config $outputdir/config_temp" or
-	    dodie "moving .config";
+	if (!$noclean) {
+	    run_command "mv $output_config $outputdir/config_temp" or
+		dodie "moving .config";
 
-	if (!$noclean && !run_command "$make mrproper") {
-	    dodie "make mrproper";
+	    run_command "$make mrproper" or dodie "make mrproper";
+
+	    run_command "mv $outputdir/config_temp $output_config" or
+		dodie "moving config_temp";
 	}
 
-	run_command "mv $outputdir/config_temp $output_config" or
-	    dodie "moving config_temp";
-
     } elsif (!$noclean) {
 	unlink "$output_config";
 	run_command "$make mrproper" or
@@ -1318,10 +1599,15 @@
 
     if (!$build_ret) {
 	# bisect may need this to pass
-	return 0 if ($in_bisect);
+	if ($in_bisect) {
+	    $no_reboot = $save_no_reboot;
+	    return 0;
+	}
 	fail "failed build" and return 0;
     }
 
+    $no_reboot = $save_no_reboot;
+
     return 1;
 }
 
@@ -1356,10 +1642,7 @@
 
     if ($i != $opt{"NUM_TESTS"} && !do_not_reboot) {
 	doprint "Reboot and wait $sleep_time seconds\n";
-	reboot;
-	start_monitor;
-	wait_for_monitor $sleep_time;
-	end_monitor;
+	reboot $sleep_time;
     }
 }
 
@@ -1500,10 +1783,7 @@
 
 sub bisect_reboot {
     doprint "Reboot and sleep $bisect_sleep_time seconds\n";
-    reboot;
-    start_monitor;
-    wait_for_monitor $bisect_sleep_time;
-    end_monitor;
+    reboot $bisect_sleep_time;
 }
 
 # returns 1 on success, 0 on failure, -1 on skip
@@ -2066,10 +2346,7 @@
 
 sub patchcheck_reboot {
     doprint "Reboot and sleep $patchcheck_sleep_time seconds\n";
-    reboot;
-    start_monitor;
-    wait_for_monitor $patchcheck_sleep_time;
-    end_monitor;
+    reboot $patchcheck_sleep_time;
 }
 
 sub patchcheck {
@@ -2178,12 +2455,31 @@
 }
 
 my %depends;
+my %depcount;
 my $iflevel = 0;
 my @ifdeps;
 
 # prevent recursion
 my %read_kconfigs;
 
+sub add_dep {
+    # $config depends on $dep
+    my ($config, $dep) = @_;
+
+    if (defined($depends{$config})) {
+	$depends{$config} .= " " . $dep;
+    } else {
+	$depends{$config} = $dep;
+    }
+
+    # record the number of configs depending on $dep
+    if (defined $depcount{$dep}) {
+	$depcount{$dep}++;
+    } else {
+	$depcount{$dep} = 1;
+    } 
+}
+
 # taken from streamline_config.pl
 sub read_kconfig {
     my ($kconfig) = @_;
@@ -2230,30 +2526,19 @@
 	    $config = $2;
 
 	    for (my $i = 0; $i < $iflevel; $i++) {
-		if ($i) {
-		    $depends{$config} .= " " . $ifdeps[$i];
-		} else {
-		    $depends{$config} = $ifdeps[$i];
-		}
-		$state = "DEP";
+		add_dep $config, $ifdeps[$i];
 	    }
 
 	# collect the depends for the config
 	} elsif ($state eq "NEW" && /^\s*depends\s+on\s+(.*)$/) {
 
-	    if (defined($depends{$1})) {
-		$depends{$config} .= " " . $1;
-	    } else {
-		$depends{$config} = $1;
-	    }
+	    add_dep $config, $1;
 
 	# Get the configs that select this config
-	} elsif ($state ne "NONE" && /^\s*select\s+(\S+)/) {
-	    if (defined($depends{$1})) {
-		$depends{$1} .= " " . $config;
-	    } else {
-		$depends{$1} = $config;
-	    }
+	} elsif ($state eq "NEW" && /^\s*select\s+(\S+)/) {
+
+	    # selected by depends on config
+	    add_dep $1, $config;
 
 	# Check for if statements
 	} elsif (/^if\s+(.*\S)\s*$/) {
@@ -2365,11 +2650,18 @@
     close OUT;
 }
 
+sub chomp_config {
+    my ($config) = @_;
+
+    $config =~ s/CONFIG_//;
+
+    return $config;
+}
+
 sub get_depends {
     my ($dep) = @_;
 
-    my $kconfig = $dep;
-    $kconfig =~ s/CONFIG_//;
+    my $kconfig = chomp_config $dep;
 
     $dep = $depends{"$kconfig"};
 
@@ -2419,8 +2711,7 @@
 	return undef;
     }
 
-    my $kconfig = $config;
-    $kconfig =~ s/CONFIG_//;
+    my $kconfig = chomp_config $config;
 
     # Test dependencies first
     if (defined($depends{"$kconfig"})) {
@@ -2510,6 +2801,14 @@
 
     my @config_keys = keys %min_configs;
 
+    # All configs need a depcount
+    foreach my $config (@config_keys) {
+	my $kconfig = chomp_config $config;
+	if (!defined $depcount{$kconfig}) {
+		$depcount{$kconfig} = 0;
+	}
+    }
+
     # Remove anything that was set by the make allnoconfig
     # we shouldn't need them as they get set for us anyway.
     foreach my $config (@config_keys) {
@@ -2548,8 +2847,13 @@
 	# Now disable each config one by one and do a make oldconfig
 	# till we find a config that changes our list.
 
-	# Put configs that did not modify the config at the end.
 	my @test_configs = keys %min_configs;
+
+	# Sort keys by who is most dependent on
+	@test_configs = sort  { $depcount{chomp_config($b)} <=> $depcount{chomp_config($a)} }
+			  @test_configs ;
+
+	# Put configs that did not modify the config at the end.
 	my $reset = 1;
 	for (my $i = 0; $i < $#test_configs; $i++) {
 	    if (!defined($nochange_config{$test_configs[0]})) {
@@ -2659,10 +2963,7 @@
 	}
 
 	doprint "Reboot and wait $sleep_time seconds\n";
-	reboot;
-	start_monitor;
-	wait_for_monitor $sleep_time;
-	end_monitor;
+	reboot $sleep_time;
     }
 
     success $i;
@@ -2783,6 +3084,9 @@
 # First we need to do is the builds
 for (my $i = 1; $i <= $opt{"NUM_TESTS"}; $i++) {
 
+    # Do not reboot on failing test options
+    $no_reboot = 1;
+
     $iteration = $i;
 
     my $makecmd = set_test_option("MAKE_CMD", $i);
@@ -2811,6 +3115,7 @@
     $reboot_type = set_test_option("REBOOT_TYPE", $i);
     $grub_menu = set_test_option("GRUB_MENU", $i);
     $post_install = set_test_option("POST_INSTALL", $i);
+    $no_install = set_test_option("NO_INSTALL", $i);
     $reboot_script = set_test_option("REBOOT_SCRIPT", $i);
     $reboot_on_error = set_test_option("REBOOT_ON_ERROR", $i);
     $poweroff_on_error = set_test_option("POWEROFF_ON_ERROR", $i);
@@ -2832,6 +3137,7 @@
     $console = set_test_option("CONSOLE", $i);
     $detect_triplefault = set_test_option("DETECT_TRIPLE_FAULT", $i);
     $success_line = set_test_option("SUCCESS_LINE", $i);
+    $reboot_success_line = set_test_option("REBOOT_SUCCESS_LINE", $i);
     $stop_after_success = set_test_option("STOP_AFTER_SUCCESS", $i);
     $stop_after_failure = set_test_option("STOP_AFTER_FAILURE", $i);
     $stop_test_after = set_test_option("STOP_TEST_AFTER", $i);
@@ -2850,9 +3156,11 @@
 
     chdir $builddir || die "can't change directory to $builddir";
 
-    if (!-d $tmpdir) {
-	mkpath($tmpdir) or
-	    die "can't create $tmpdir";
+    foreach my $dir ($tmpdir, $outputdir) {
+	if (!-d $dir) {
+	    mkpath($dir) or
+		die "can't create $dir";
+	}
     }
 
     $ENV{"SSH_USER"} = $ssh_user;
@@ -2889,8 +3197,11 @@
 	$run_type = "ERROR";
     }
 
+    my $installme = "";
+    $installme = " no_install" if ($no_install);
+
     doprint "\n\n";
-    doprint "RUNNING TEST $i of $opt{NUM_TESTS} with option $test_type $run_type\n\n";
+    doprint "RUNNING TEST $i of $opt{NUM_TESTS} with option $test_type $run_type$installme\n\n";
 
     unlink $dmesg;
     unlink $buildlog;
@@ -2911,6 +3222,9 @@
 	    die "failed to checkout $checkout";
     }
 
+    $no_reboot = 0;
+
+
     if ($test_type eq "bisect") {
 	bisect $i;
 	next;
@@ -2929,6 +3243,13 @@
 	build $build_type or next;
     }
 
+    if ($test_type eq "install") {
+	get_version;
+	install;
+	success $i;
+	next;
+    }
+
     if ($test_type ne "build") {
 	my $failed = 0;
 	start_monitor_and_boot or $failed = 1;
diff --git a/tools/testing/ktest/sample.conf b/tools/testing/ktest/sample.conf
index b8bcd14..dbedfa1 100644
--- a/tools/testing/ktest/sample.conf
+++ b/tools/testing/ktest/sample.conf
@@ -72,6 +72,128 @@
 # the same option name under the same test or as default
 # ktest will fail to execute, and no tests will run.
 #
+# DEFAULTS OVERRIDE
+#
+# Options defined in the DEFAULTS section can not be duplicated
+# even if they are defined in two different DEFAULT sections.
+# This is done to catch mistakes where an option is added but
+# the previous option was forgotten about and not commented.
+#
+# The OVERRIDE keyword can be added to a section to allow this
+# section to override other DEFAULT sections values that have
+# been defined previously. It will only override options that
+# have been defined before its use. Options defined later
+# in a non override section will still error. The same option
+# can not be defined in the same section even if that section
+# is marked OVERRIDE.
+#
+#
+#
+# Both TEST_START and DEFAULTS sections can also have the IF keyword
+# The value after the IF must evaluate into a 0 or non 0 positive
+# integer, and can use the config variables (explained below).
+#
+# DEFAULTS IF ${IS_X86_32}
+#
+# The above will process the DEFAULTS section if the config
+# variable IS_X86_32 evaluates to a non zero positive integer
+# otherwise if it evaluates to zero, it will act the same
+# as if the SKIP keyword was used.
+#
+# The ELSE keyword can be used directly after a section with
+# a IF statement.
+#
+# TEST_START IF ${RUN_NET_TESTS}
+# BUILD_TYPE = useconfig:${CONFIG_DIR}/config-network
+#
+# ELSE
+#
+# BUILD_TYPE = useconfig:${CONFIG_DIR}/config-normal
+#
+#
+# The ELSE keyword can also contain an IF statement to allow multiple
+# if then else sections. But all the sections must be either
+# DEFAULT or TEST_START, they can not be a mixture.
+#
+# TEST_START IF ${RUN_NET_TESTS}
+# BUILD_TYPE = useconfig:${CONFIG_DIR}/config-network
+#
+# ELSE IF ${RUN_DISK_TESTS}
+# BUILD_TYPE = useconfig:${CONFIG_DIR}/config-tests
+#
+# ELSE IF ${RUN_CPU_TESTS}
+# BUILD_TYPE = useconfig:${CONFIG_DIR}/config-cpu
+#
+# ELSE
+# BUILD_TYPE = useconfig:${CONFIG_DIR}/config-network
+#
+# The if statement may also have comparisons that will and for
+# == and !=, strings may be used for both sides.
+#
+# BOX_TYPE := x86_32
+#
+# DEFAULTS IF ${BOX_TYPE} == x86_32
+# BUILD_TYPE = useconfig:${CONFIG_DIR}/config-32
+# ELSE
+# BUILD_TYPE = useconfig:${CONFIG_DIR}/config-64
+#
+# The DEFINED keyword can be used by the IF statements too.
+# It returns true if the given config variable or option has been defined
+# or false otherwise.
+#
+# 
+# DEFAULTS IF DEFINED USE_CC
+# CC := ${USE_CC}
+# ELSE
+# CC := gcc
+#
+#
+# As well as NOT DEFINED.
+#
+# DEFAULTS IF NOT DEFINED MAKE_CMD
+# MAKE_CMD := make ARCH=x86
+#
+#
+# And/or ops (&&,||) may also be used to make complex conditionals.
+#
+# TEST_START IF (DEFINED ALL_TESTS || ${MYTEST} == boottest) && ${MACHINE} == gandalf
+#
+# Notice the use of paranthesis. Without any paranthesis the above would be
+# processed the same as:
+#
+# TEST_START IF DEFINED ALL_TESTS || (${MYTEST} == boottest && ${MACHINE} == gandalf)
+#
+#
+#
+# INCLUDE file
+#
+# The INCLUDE keyword may be used in DEFAULT sections. This will
+# read another config file and process that file as well. The included
+# file can include other files, add new test cases or default
+# statements. Config variables will be passed to these files and changes
+# to config variables will be seen by top level config files. Including
+# a file is processed just like the contents of the file was cut and pasted
+# into the top level file, except, that include files that end with
+# TEST_START sections will have that section ended at the end of
+# the include file. That is, an included file is included followed
+# by another DEFAULT keyword.
+#
+# Unlike other files referenced in this config, the file path does not need
+# to be absolute. If the file does not start with '/', then the directory
+# that the current config file was located in is used. If no config by the
+# given name is found there, then the current directory is searched.
+#
+# INCLUDE myfile
+# DEFAULT
+#
+# is the same as:
+#
+# INCLUDE myfile
+#
+# Note, if the include file does not contain a full path, the file is
+# searched first by the location of the original include file, and then
+# by the location that ktest.pl was executed in.
+#
 
 #### Config variables ####
 #
@@ -253,9 +375,10 @@
 
 # The default test type (default test)
 # The test types may be:
-#   build - only build the kernel, do nothing else
-#   boot - build and boot the kernel
-#   test - build, boot and if TEST is set, run the test script
+#   build   - only build the kernel, do nothing else
+#   install - build and install, but do nothing else (does not reboot)
+#   boot    - build, install, and boot the kernel
+#   test    - build, boot and if TEST is set, run the test script
 #          (If TEST is not set, it defaults back to boot)
 #   bisect - Perform a bisect on the kernel (see BISECT_TYPE below)
 #   patchcheck - Do a test on a series of commits in git (see PATCHCHECK below)
@@ -293,6 +416,13 @@
 # or on some systems:
 #POST_INSTALL = ssh user@target /sbin/dracut -f /boot/initramfs-test.img $KERNEL_VERSION
 
+# If for some reason you just want to boot the kernel and you do not
+# want the test to install anything new. For example, you may just want
+# to boot test the same kernel over and over and do not want to go through
+# the hassle of installing anything, you can set this option to 1
+# (default 0)
+#NO_INSTALL = 1
+
 # If there is a script that you require to run before the build is done
 # you can specify it with PRE_BUILD.
 #
@@ -415,6 +545,14 @@
 # (default "login:")
 #SUCCESS_LINE = login:
 
+# To speed up between reboots, defining a line that the
+# default kernel produces that represents that the default
+# kernel has successfully booted and can be used to pass
+# a new test kernel to it. Otherwise ktest.pl will wait till
+# SLEEP_TIME to continue.
+# (default undefined)
+#REBOOT_SUCCESS_LINE = login:
+
 # In case the console constantly fills the screen, having
 # a specified time to stop the test after success is recommended.
 # (in seconds)
@@ -480,6 +618,8 @@
 # another test. If a reboot to the reliable kernel happens,
 # we wait SLEEP_TIME for the console to stop producing output
 # before starting the next test.
+#
+# You can speed up reboot times even more by setting REBOOT_SUCCESS_LINE.
 # (default 60)
 #SLEEP_TIME = 60