Merge git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-linus
* git://git.kernel.org/pub/scm/linux/kernel/git/pkl/squashfs-linus:
Squashfs: simplify CONFIG_SQUASHFS_LZO handling
Squashfs: move squashfs_i() definition from squashfs.h
Squashfs: get rid of default n in Kconfig
Squashfs: add missing check in zlib_wrapper
Squashfs: remove unnecessary variable in zlib_wrapper
Squashfs: Add XZ compression configuration option
Squashfs: add XZ compression support
diff --git a/Documentation/ABI/stable/thermal-notification b/Documentation/ABI/stable/thermal-notification
new file mode 100644
index 0000000..9723e8b
--- /dev/null
+++ b/Documentation/ABI/stable/thermal-notification
@@ -0,0 +1,4 @@
+What: A notification mechanism for thermal related events
+Description:
+ This interface enables notification for thermal related events.
+ The notification is in the form of a netlink event.
diff --git a/Documentation/DocBook/80211.tmpl b/Documentation/DocBook/80211.tmpl
index 03641a0..8906648 100644
--- a/Documentation/DocBook/80211.tmpl
+++ b/Documentation/DocBook/80211.tmpl
@@ -268,10 +268,6 @@
!Finclude/net/mac80211.h ieee80211_ops
!Finclude/net/mac80211.h ieee80211_alloc_hw
!Finclude/net/mac80211.h ieee80211_register_hw
-!Finclude/net/mac80211.h ieee80211_get_tx_led_name
-!Finclude/net/mac80211.h ieee80211_get_rx_led_name
-!Finclude/net/mac80211.h ieee80211_get_assoc_led_name
-!Finclude/net/mac80211.h ieee80211_get_radio_led_name
!Finclude/net/mac80211.h ieee80211_unregister_hw
!Finclude/net/mac80211.h ieee80211_free_hw
</chapter>
@@ -382,6 +378,23 @@
</para>
</partintro>
+ <chapter id="led-support">
+ <title>LED support</title>
+ <para>
+ Mac80211 supports various ways of blinking LEDs. Wherever possible,
+ device LEDs should be exposed as LED class devices and hooked up to
+ the appropriate trigger, which will then be triggered appropriately
+ by mac80211.
+ </para>
+!Finclude/net/mac80211.h ieee80211_get_tx_led_name
+!Finclude/net/mac80211.h ieee80211_get_rx_led_name
+!Finclude/net/mac80211.h ieee80211_get_assoc_led_name
+!Finclude/net/mac80211.h ieee80211_get_radio_led_name
+!Finclude/net/mac80211.h ieee80211_tpt_blink
+!Finclude/net/mac80211.h ieee80211_tpt_led_trigger_flags
+!Finclude/net/mac80211.h ieee80211_create_tpt_led_trigger
+ </chapter>
+
<chapter id="hardware-crypto-offload">
<title>Hardware crypto acceleration</title>
!Pinclude/net/mac80211.h Hardware crypto acceleration
diff --git a/Documentation/IPMI.txt b/Documentation/IPMI.txt
index 69dd29e..b2bea15 100644
--- a/Documentation/IPMI.txt
+++ b/Documentation/IPMI.txt
@@ -533,6 +533,33 @@
Other Pieces
------------
+Get the detailed info related with the IPMI device
+--------------------------------------------------
+
+Some users need more detailed information about a device, like where
+the address came from or the raw base device for the IPMI interface.
+You can use the IPMI smi_watcher to catch the IPMI interfaces as they
+come or go, and to grab the information, you can use the function
+ipmi_get_smi_info(), which returns the following structure:
+
+struct ipmi_smi_info {
+ enum ipmi_addr_src addr_src;
+ struct device *dev;
+ union {
+ struct {
+ void *acpi_handle;
+ } acpi_info;
+ } addr_info;
+};
+
+Currently special info for only for SI_ACPI address sources is
+returned. Others may be added as necessary.
+
+Note that the dev pointer is included in the above structure, and
+assuming ipmi_smi_get_info returns success, you must call put_device
+on the dev pointer.
+
+
Watchdog
--------
diff --git a/Documentation/acpi/apei/output_format.txt b/Documentation/acpi/apei/output_format.txt
new file mode 100644
index 0000000..9146952
--- /dev/null
+++ b/Documentation/acpi/apei/output_format.txt
@@ -0,0 +1,122 @@
+ APEI output format
+ ~~~~~~~~~~~~~~~~~~
+
+APEI uses printk as hardware error reporting interface, the output
+format is as follow.
+
+<error record> :=
+APEI generic hardware error status
+severity: <integer>, <severity string>
+section: <integer>, severity: <integer>, <severity string>
+flags: <integer>
+<section flags strings>
+fru_id: <uuid string>
+fru_text: <string>
+section_type: <section type string>
+<section data>
+
+<severity string>* := recoverable | fatal | corrected | info
+
+<section flags strings># :=
+[primary][, containment warning][, reset][, threshold exceeded]\
+[, resource not accessible][, latent error]
+
+<section type string> := generic processor error | memory error | \
+PCIe error | unknown, <uuid string>
+
+<section data> :=
+<generic processor section data> | <memory section data> | \
+<pcie section data> | <null>
+
+<generic processor section data> :=
+[processor_type: <integer>, <proc type string>]
+[processor_isa: <integer>, <proc isa string>]
+[error_type: <integer>
+<proc error type strings>]
+[operation: <integer>, <proc operation string>]
+[flags: <integer>
+<proc flags strings>]
+[level: <integer>]
+[version_info: <integer>]
+[processor_id: <integer>]
+[target_address: <integer>]
+[requestor_id: <integer>]
+[responder_id: <integer>]
+[IP: <integer>]
+
+<proc type string>* := IA32/X64 | IA64
+
+<proc isa string>* := IA32 | IA64 | X64
+
+<processor error type strings># :=
+[cache error][, TLB error][, bus error][, micro-architectural error]
+
+<proc operation string>* := unknown or generic | data read | data write | \
+instruction execution
+
+<proc flags strings># :=
+[restartable][, precise IP][, overflow][, corrected]
+
+<memory section data> :=
+[error_status: <integer>]
+[physical_address: <integer>]
+[physical_address_mask: <integer>]
+[node: <integer>]
+[card: <integer>]
+[module: <integer>]
+[bank: <integer>]
+[device: <integer>]
+[row: <integer>]
+[column: <integer>]
+[bit_position: <integer>]
+[requestor_id: <integer>]
+[responder_id: <integer>]
+[target_id: <integer>]
+[error_type: <integer>, <mem error type string>]
+
+<mem error type string>* :=
+unknown | no error | single-bit ECC | multi-bit ECC | \
+single-symbol chipkill ECC | multi-symbol chipkill ECC | master abort | \
+target abort | parity error | watchdog timeout | invalid address | \
+mirror Broken | memory sparing | scrub corrected error | \
+scrub uncorrected error
+
+<pcie section data> :=
+[port_type: <integer>, <pcie port type string>]
+[version: <integer>.<integer>]
+[command: <integer>, status: <integer>]
+[device_id: <integer>:<integer>:<integer>.<integer>
+slot: <integer>
+secondary_bus: <integer>
+vendor_id: <integer>, device_id: <integer>
+class_code: <integer>]
+[serial number: <integer>, <integer>]
+[bridge: secondary_status: <integer>, control: <integer>]
+
+<pcie port type string>* := PCIe end point | legacy PCI end point | \
+unknown | unknown | root port | upstream switch port | \
+downstream switch port | PCIe to PCI/PCI-X bridge | \
+PCI/PCI-X to PCIe bridge | root complex integrated endpoint device | \
+root complex event collector
+
+Where, [] designate corresponding content is optional
+
+All <field string> description with * has the following format:
+
+field: <integer>, <field string>
+
+Where value of <integer> should be the position of "string" in <field
+string> description. Otherwise, <field string> will be "unknown".
+
+All <field strings> description with # has the following format:
+
+field: <integer>
+<field strings>
+
+Where each string in <fields strings> corresponding to one set bit of
+<integer>. The bit position is the position of "string" in <field
+strings> description.
+
+For more detailed explanation of every field, please refer to UEFI
+specification version 2.3 or later, section Appendix N: Common
+Platform Error Record.
diff --git a/Documentation/device-mapper/dm-crypt.txt b/Documentation/device-mapper/dm-crypt.txt
index 524de92..59293ac 100644
--- a/Documentation/device-mapper/dm-crypt.txt
+++ b/Documentation/device-mapper/dm-crypt.txt
@@ -8,7 +8,7 @@
<cipher>
Encryption cipher and an optional IV generation mode.
- (In format cipher-chainmode-ivopts:ivmode).
+ (In format cipher[:keycount]-chainmode-ivopts:ivmode).
Examples:
des
aes-cbc-essiv:sha256
@@ -20,6 +20,11 @@
Key used for encryption. It is encoded as a hexadecimal number.
You can only use key sizes that are valid for the selected cipher.
+<keycount>
+ Multi-key compatibility mode. You can define <keycount> keys and
+ then sectors are encrypted according to their offsets (sector 0 uses key0;
+ sector 1 uses key1 etc.). <keycount> must be a power of two.
+
<iv_offset>
The IV offset is a sector count that is added to the sector number
before creating the IV.
diff --git a/Documentation/device-mapper/dm-raid.txt b/Documentation/device-mapper/dm-raid.txt
new file mode 100644
index 0000000..33b6b70
--- /dev/null
+++ b/Documentation/device-mapper/dm-raid.txt
@@ -0,0 +1,70 @@
+Device-mapper RAID (dm-raid) is a bridge from DM to MD. It
+provides a way to use device-mapper interfaces to access the MD RAID
+drivers.
+
+As with all device-mapper targets, the nominal public interfaces are the
+constructor (CTR) tables and the status outputs (both STATUSTYPE_INFO
+and STATUSTYPE_TABLE). The CTR table looks like the following:
+
+1: <s> <l> raid \
+2: <raid_type> <#raid_params> <raid_params> \
+3: <#raid_devs> <meta_dev1> <dev1> .. <meta_devN> <devN>
+
+Line 1 contains the standard first three arguments to any device-mapper
+target - the start, length, and target type fields. The target type in
+this case is "raid".
+
+Line 2 contains the arguments that define the particular raid
+type/personality/level, the required arguments for that raid type, and
+any optional arguments. Possible raid types include: raid4, raid5_la,
+raid5_ls, raid5_rs, raid6_zr, raid6_nr, and raid6_nc. (raid1 is
+planned for the future.) The list of required and optional parameters
+is the same for all the current raid types. The required parameters are
+positional, while the optional parameters are given as key/value pairs.
+The possible parameters are as follows:
+ <chunk_size> Chunk size in sectors.
+ [[no]sync] Force/Prevent RAID initialization
+ [rebuild <idx>] Rebuild the drive indicated by the index
+ [daemon_sleep <ms>] Time between bitmap daemon work to clear bits
+ [min_recovery_rate <kB/sec/disk>] Throttle RAID initialization
+ [max_recovery_rate <kB/sec/disk>] Throttle RAID initialization
+ [max_write_behind <sectors>] See '-write-behind=' (man mdadm)
+ [stripe_cache <sectors>] Stripe cache size for higher RAIDs
+
+Line 3 contains the list of devices that compose the array in
+metadata/data device pairs. If the metadata is stored separately, a '-'
+is given for the metadata device position. If a drive has failed or is
+missing at creation time, a '-' can be given for both the metadata and
+data drives for a given position.
+
+NB. Currently all metadata devices must be specified as '-'.
+
+Examples:
+# RAID4 - 4 data drives, 1 parity
+# No metadata devices specified to hold superblock/bitmap info
+# Chunk size of 1MiB
+# (Lines separated for easy reading)
+0 1960893648 raid \
+ raid4 1 2048 \
+ 5 - 8:17 - 8:33 - 8:49 - 8:65 - 8:81
+
+# RAID4 - 4 data drives, 1 parity (no metadata devices)
+# Chunk size of 1MiB, force RAID initialization,
+# min recovery rate at 20 kiB/sec/disk
+0 1960893648 raid \
+ raid4 4 2048 min_recovery_rate 20 sync\
+ 5 - 8:17 - 8:33 - 8:49 - 8:65 - 8:81
+
+Performing a 'dmsetup table' should display the CTR table used to
+construct the mapping (with possible reordering of optional
+parameters).
+
+Performing a 'dmsetup status' will yield information on the state and
+health of the array. The output is as follows:
+1: <s> <l> raid \
+2: <raid_type> <#devices> <1 health char for each dev> <resync_ratio>
+
+Line 1 is standard DM output. Line 2 is best shown by example:
+ 0 1960893648 raid raid4 5 AAAAA 2/490221568
+Here we can see the RAID type is raid4, there are 5 devices - all of
+which are 'A'live, and the array is 2/490221568 complete with recovery.
diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 6cbbd20..8c594c4 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -248,6 +248,17 @@
---------------------------
+What: CONFIG_ACPI_PROCFS_POWER
+When: 2.6.39
+Why: sysfs I/F for ACPI power devices, including AC and Battery,
+ has been working in upstream kenrel since 2.6.24, Sep 2007.
+ In 2.6.37, we make the sysfs I/F always built in and this option
+ disabled by default.
+ Remove this option and the ACPI power procfs interface in 2.6.39.
+Who: Zhang Rui <rui.zhang@intel.com>
+
+---------------------------
+
What: /proc/acpi/button
When: August 2007
Why: /proc/acpi/button has been replaced by events to the input layer
diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 977d891..651d523 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -19,6 +19,8 @@
void (*d_release)(struct dentry *);
void (*d_iput)(struct dentry *, struct inode *);
char *(*d_dname)((struct dentry *dentry, char *buffer, int buflen);
+ struct vfsmount *(*d_automount)(struct path *path);
+ int (*d_manage)(struct dentry *, bool);
locking rules:
rename_lock ->d_lock may block rcu-walk
@@ -29,6 +31,8 @@
d_release: no no yes no
d_iput: no no yes no
d_dname: no no no no
+d_automount: no no yes no
+d_manage: no no yes (ref-walk) maybe
--------------------------- inode_operations ---------------------------
prototypes:
@@ -343,7 +347,6 @@
int (*fl_grant)(struct file_lock *, struct file_lock *, int);
void (*fl_release_private)(struct file_lock *);
void (*fl_break)(struct file_lock *); /* break_lease callback */
- int (*fl_mylease)(struct file_lock *, struct file_lock *);
int (*fl_change)(struct file_lock **, int);
locking rules:
@@ -353,7 +356,6 @@
fl_grant: no no
fl_release_private: maybe no
fl_break: yes no
-fl_mylease: yes no
fl_change yes no
--------------------------- buffer_head -----------------------------------
diff --git a/Documentation/filesystems/porting b/Documentation/filesystems/porting
index 266d205..dfbcd1b 100644
--- a/Documentation/filesystems/porting
+++ b/Documentation/filesystems/porting
@@ -365,8 +365,8 @@
[recommended]
vfs now tries to do path walking in "rcu-walk mode", which avoids
atomic operations and scalability hazards on dentries and inodes (see
-Documentation/filesystems/path-walk.txt). d_hash and d_compare changes (above)
-are examples of the changes required to support this. For more complex
+Documentation/filesystems/path-lookup.txt). d_hash and d_compare changes
+(above) are examples of the changes required to support this. For more complex
filesystem callbacks, the vfs drops out of rcu-walk mode before the fs call, so
no changes are required to the filesystem. However, this is costly and loses
the benefits of rcu-walk mode. We will begin to add filesystem callbacks that
@@ -383,8 +383,8 @@
permission and check_acl are inode permission checks that are called
on many or all directory inodes on the way down a path walk (to check for
-exec permission). These must now be rcu-walk aware (flags & IPERM_RCU). See
-Documentation/filesystems/vfs.txt for more details.
+exec permission). These must now be rcu-walk aware (flags & IPERM_FLAG_RCU).
+See Documentation/filesystems/vfs.txt for more details.
--
[mandatory]
diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index 9471225..23cae65 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -375,6 +375,7 @@
Swap: 0 kB
KernelPageSize: 4 kB
MMUPageSize: 4 kB
+Locked: 374 kB
The first of these lines shows the same information as is displayed for the
mapping in /proc/PID/maps. The remaining lines show the size of the mapping
@@ -670,6 +671,8 @@
> cat /proc/meminfo
+The "Locked" indicates whether the mapping is locked in memory or not.
+
MemTotal: 16344972 kB
MemFree: 13634064 kB
@@ -1320,6 +1323,10 @@
Writing to /proc/<pid>/oom_score_adj or /proc/<pid>/oom_adj will change the
other with its scaled value.
+The value of /proc/<pid>/oom_score_adj may be reduced no lower than the last
+value set by a CAP_SYS_RESOURCE process. To reduce the value any lower
+requires CAP_SYS_RESOURCE.
+
NOTICE: /proc/<pid>/oom_adj is deprecated and will be removed, please see
Documentation/feature-removal-schedule.txt.
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index fbb324e..94cf97b 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -415,8 +415,8 @@
permission: called by the VFS to check for access rights on a POSIX-like
filesystem.
- May be called in rcu-walk mode (flags & IPERM_RCU). If in rcu-walk
- mode, the filesystem must check the permission without blocking or
+ May be called in rcu-walk mode (flags & IPERM_FLAG_RCU). If in rcu-walk
+ mode, the filesystem must check the permission without blocking or
storing to the inode.
If a situation is encountered that rcu-walk cannot handle, return
@@ -864,6 +864,8 @@
void (*d_release)(struct dentry *);
void (*d_iput)(struct dentry *, struct inode *);
char *(*d_dname)(struct dentry *, char *, int);
+ struct vfsmount *(*d_automount)(struct path *);
+ int (*d_manage)(struct dentry *, bool, bool);
};
d_revalidate: called when the VFS needs to revalidate a dentry. This
@@ -930,6 +932,47 @@
at the end of the buffer, and returns a pointer to the first char.
dynamic_dname() helper function is provided to take care of this.
+ d_automount: called when an automount dentry is to be traversed (optional).
+ This should create a new VFS mount record and return the record to the
+ caller. The caller is supplied with a path parameter giving the
+ automount directory to describe the automount target and the parent
+ VFS mount record to provide inheritable mount parameters. NULL should
+ be returned if someone else managed to make the automount first. If
+ the vfsmount creation failed, then an error code should be returned.
+ If -EISDIR is returned, then the directory will be treated as an
+ ordinary directory and returned to pathwalk to continue walking.
+
+ If a vfsmount is returned, the caller will attempt to mount it on the
+ mountpoint and will remove the vfsmount from its expiration list in
+ the case of failure. The vfsmount should be returned with 2 refs on
+ it to prevent automatic expiration - the caller will clean up the
+ additional ref.
+
+ This function is only used if DCACHE_NEED_AUTOMOUNT is set on the
+ dentry. This is set by __d_instantiate() if S_AUTOMOUNT is set on the
+ inode being added.
+
+ d_manage: called to allow the filesystem to manage the transition from a
+ dentry (optional). This allows autofs, for example, to hold up clients
+ waiting to explore behind a 'mountpoint' whilst letting the daemon go
+ past and construct the subtree there. 0 should be returned to let the
+ calling process continue. -EISDIR can be returned to tell pathwalk to
+ use this directory as an ordinary directory and to ignore anything
+ mounted on it and not to check the automount flag. Any other error
+ code will abort pathwalk completely.
+
+ If the 'mounting_here' parameter is true, then namespace_sem is being
+ held by the caller and the function should not initiate any mounts or
+ unmounts that it will then wait for.
+
+ If the 'rcu_walk' parameter is true, then the caller is doing a
+ pathwalk in RCU-walk mode. Sleeping is not permitted in this mode,
+ and the caller can be asked to leave it and call again by returing
+ -ECHILD.
+
+ This function is only used if DCACHE_MANAGE_TRANSIT is set on the
+ dentry being transited from.
+
Example :
static char *pipefs_dname(struct dentry *dent, char *buffer, int buflen)
diff --git a/Documentation/gpio.txt b/Documentation/gpio.txt
index a492d92..792faa3 100644
--- a/Documentation/gpio.txt
+++ b/Documentation/gpio.txt
@@ -135,7 +135,7 @@
int gpio_direction_input(unsigned gpio);
int gpio_direction_output(unsigned gpio, int value);
-The return value is zero for success, else a negative errno. It must
+The return value is zero for success, else a negative errno. It should
be checked, since the get/set calls don't have error returns and since
misconfiguration is possible. You should normally issue these calls from
a task context. However, for spinlock-safe GPIOs it's OK to use them
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 55fe759..b72e071 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -199,11 +199,6 @@
unusable. The "log_buf_len" parameter may be useful
if you need to capture more output.
- acpi_display_output= [HW,ACPI]
- acpi_display_output=vendor
- acpi_display_output=video
- See above.
-
acpi_irq_balance [HW,ACPI]
ACPI will balance active IRQs
default in APIC mode
diff --git a/Documentation/target/tcm_mod_builder.py b/Documentation/target/tcm_mod_builder.py
new file mode 100755
index 0000000..dbeb8a0
--- /dev/null
+++ b/Documentation/target/tcm_mod_builder.py
@@ -0,0 +1,1094 @@
+#!/usr/bin/python
+# The TCM v4 multi-protocol fabric module generation script for drivers/target/$NEW_MOD
+#
+# Copyright (c) 2010 Rising Tide Systems
+# Copyright (c) 2010 Linux-iSCSI.org
+#
+# Author: nab@kernel.org
+#
+import os, sys
+import subprocess as sub
+import string
+import re
+import optparse
+
+tcm_dir = ""
+
+fabric_ops = []
+fabric_mod_dir = ""
+fabric_mod_port = ""
+fabric_mod_init_port = ""
+
+def tcm_mod_err(msg):
+ print msg
+ sys.exit(1)
+
+def tcm_mod_create_module_subdir(fabric_mod_dir_var):
+
+ if os.path.isdir(fabric_mod_dir_var) == True:
+ return 1
+
+ print "Creating fabric_mod_dir: " + fabric_mod_dir_var
+ ret = os.mkdir(fabric_mod_dir_var)
+ if ret:
+ tcm_mod_err("Unable to mkdir " + fabric_mod_dir_var)
+
+ return
+
+def tcm_mod_build_FC_include(fabric_mod_dir_var, fabric_mod_name):
+ global fabric_mod_port
+ global fabric_mod_init_port
+ buf = ""
+
+ f = fabric_mod_dir_var + "/" + fabric_mod_name + "_base.h"
+ print "Writing file: " + f
+
+ p = open(f, 'w');
+ if not p:
+ tcm_mod_err("Unable to open file: " + f)
+
+ buf = "#define " + fabric_mod_name.upper() + "_VERSION \"v0.1\"\n"
+ buf += "#define " + fabric_mod_name.upper() + "_NAMELEN 32\n"
+ buf += "\n"
+ buf += "struct " + fabric_mod_name + "_nacl {\n"
+ buf += " /* Binary World Wide unique Port Name for FC Initiator Nport */\n"
+ buf += " u64 nport_wwpn;\n"
+ buf += " /* ASCII formatted WWPN for FC Initiator Nport */\n"
+ buf += " char nport_name[" + fabric_mod_name.upper() + "_NAMELEN];\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_nodeacl() */\n"
+ buf += " struct se_node_acl se_node_acl;\n"
+ buf += "};\n"
+ buf += "\n"
+ buf += "struct " + fabric_mod_name + "_tpg {\n"
+ buf += " /* FC lport target portal group tag for TCM */\n"
+ buf += " u16 lport_tpgt;\n"
+ buf += " /* Pointer back to " + fabric_mod_name + "_lport */\n"
+ buf += " struct " + fabric_mod_name + "_lport *lport;\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_tpg() */\n"
+ buf += " struct se_portal_group se_tpg;\n"
+ buf += "};\n"
+ buf += "\n"
+ buf += "struct " + fabric_mod_name + "_lport {\n"
+ buf += " /* SCSI protocol the lport is providing */\n"
+ buf += " u8 lport_proto_id;\n"
+ buf += " /* Binary World Wide unique Port Name for FC Target Lport */\n"
+ buf += " u64 lport_wwpn;\n"
+ buf += " /* ASCII formatted WWPN for FC Target Lport */\n"
+ buf += " char lport_name[" + fabric_mod_name.upper() + "_NAMELEN];\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_lport() */\n"
+ buf += " struct se_wwn lport_wwn;\n"
+ buf += "};\n"
+
+ ret = p.write(buf)
+ if ret:
+ tcm_mod_err("Unable to write f: " + f)
+
+ p.close()
+
+ fabric_mod_port = "lport"
+ fabric_mod_init_port = "nport"
+
+ return
+
+def tcm_mod_build_SAS_include(fabric_mod_dir_var, fabric_mod_name):
+ global fabric_mod_port
+ global fabric_mod_init_port
+ buf = ""
+
+ f = fabric_mod_dir_var + "/" + fabric_mod_name + "_base.h"
+ print "Writing file: " + f
+
+ p = open(f, 'w');
+ if not p:
+ tcm_mod_err("Unable to open file: " + f)
+
+ buf = "#define " + fabric_mod_name.upper() + "_VERSION \"v0.1\"\n"
+ buf += "#define " + fabric_mod_name.upper() + "_NAMELEN 32\n"
+ buf += "\n"
+ buf += "struct " + fabric_mod_name + "_nacl {\n"
+ buf += " /* Binary World Wide unique Port Name for SAS Initiator port */\n"
+ buf += " u64 iport_wwpn;\n"
+ buf += " /* ASCII formatted WWPN for Sas Initiator port */\n"
+ buf += " char iport_name[" + fabric_mod_name.upper() + "_NAMELEN];\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_nodeacl() */\n"
+ buf += " struct se_node_acl se_node_acl;\n"
+ buf += "};\n\n"
+ buf += "struct " + fabric_mod_name + "_tpg {\n"
+ buf += " /* SAS port target portal group tag for TCM */\n"
+ buf += " u16 tport_tpgt;\n"
+ buf += " /* Pointer back to " + fabric_mod_name + "_tport */\n"
+ buf += " struct " + fabric_mod_name + "_tport *tport;\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_tpg() */\n"
+ buf += " struct se_portal_group se_tpg;\n"
+ buf += "};\n\n"
+ buf += "struct " + fabric_mod_name + "_tport {\n"
+ buf += " /* SCSI protocol the tport is providing */\n"
+ buf += " u8 tport_proto_id;\n"
+ buf += " /* Binary World Wide unique Port Name for SAS Target port */\n"
+ buf += " u64 tport_wwpn;\n"
+ buf += " /* ASCII formatted WWPN for SAS Target port */\n"
+ buf += " char tport_name[" + fabric_mod_name.upper() + "_NAMELEN];\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_tport() */\n"
+ buf += " struct se_wwn tport_wwn;\n"
+ buf += "};\n"
+
+ ret = p.write(buf)
+ if ret:
+ tcm_mod_err("Unable to write f: " + f)
+
+ p.close()
+
+ fabric_mod_port = "tport"
+ fabric_mod_init_port = "iport"
+
+ return
+
+def tcm_mod_build_iSCSI_include(fabric_mod_dir_var, fabric_mod_name):
+ global fabric_mod_port
+ global fabric_mod_init_port
+ buf = ""
+
+ f = fabric_mod_dir_var + "/" + fabric_mod_name + "_base.h"
+ print "Writing file: " + f
+
+ p = open(f, 'w');
+ if not p:
+ tcm_mod_err("Unable to open file: " + f)
+
+ buf = "#define " + fabric_mod_name.upper() + "_VERSION \"v0.1\"\n"
+ buf += "#define " + fabric_mod_name.upper() + "_NAMELEN 32\n"
+ buf += "\n"
+ buf += "struct " + fabric_mod_name + "_nacl {\n"
+ buf += " /* ASCII formatted InitiatorName */\n"
+ buf += " char iport_name[" + fabric_mod_name.upper() + "_NAMELEN];\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_nodeacl() */\n"
+ buf += " struct se_node_acl se_node_acl;\n"
+ buf += "};\n\n"
+ buf += "struct " + fabric_mod_name + "_tpg {\n"
+ buf += " /* iSCSI target portal group tag for TCM */\n"
+ buf += " u16 tport_tpgt;\n"
+ buf += " /* Pointer back to " + fabric_mod_name + "_tport */\n"
+ buf += " struct " + fabric_mod_name + "_tport *tport;\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_tpg() */\n"
+ buf += " struct se_portal_group se_tpg;\n"
+ buf += "};\n\n"
+ buf += "struct " + fabric_mod_name + "_tport {\n"
+ buf += " /* SCSI protocol the tport is providing */\n"
+ buf += " u8 tport_proto_id;\n"
+ buf += " /* ASCII formatted TargetName for IQN */\n"
+ buf += " char tport_name[" + fabric_mod_name.upper() + "_NAMELEN];\n"
+ buf += " /* Returned by " + fabric_mod_name + "_make_tport() */\n"
+ buf += " struct se_wwn tport_wwn;\n"
+ buf += "};\n"
+
+ ret = p.write(buf)
+ if ret:
+ tcm_mod_err("Unable to write f: " + f)
+
+ p.close()
+
+ fabric_mod_port = "tport"
+ fabric_mod_init_port = "iport"
+
+ return
+
+def tcm_mod_build_base_includes(proto_ident, fabric_mod_dir_val, fabric_mod_name):
+
+ if proto_ident == "FC":
+ tcm_mod_build_FC_include(fabric_mod_dir_val, fabric_mod_name)
+ elif proto_ident == "SAS":
+ tcm_mod_build_SAS_include(fabric_mod_dir_val, fabric_mod_name)
+ elif proto_ident == "iSCSI":
+ tcm_mod_build_iSCSI_include(fabric_mod_dir_val, fabric_mod_name)
+ else:
+ print "Unsupported proto_ident: " + proto_ident
+ sys.exit(1)
+
+ return
+
+def tcm_mod_build_configfs(proto_ident, fabric_mod_dir_var, fabric_mod_name):
+ buf = ""
+
+ f = fabric_mod_dir_var + "/" + fabric_mod_name + "_configfs.c"
+ print "Writing file: " + f
+
+ p = open(f, 'w');
+ if not p:
+ tcm_mod_err("Unable to open file: " + f)
+
+ buf = "#include <linux/module.h>\n"
+ buf += "#include <linux/moduleparam.h>\n"
+ buf += "#include <linux/version.h>\n"
+ buf += "#include <generated/utsrelease.h>\n"
+ buf += "#include <linux/utsname.h>\n"
+ buf += "#include <linux/init.h>\n"
+ buf += "#include <linux/slab.h>\n"
+ buf += "#include <linux/kthread.h>\n"
+ buf += "#include <linux/types.h>\n"
+ buf += "#include <linux/string.h>\n"
+ buf += "#include <linux/configfs.h>\n"
+ buf += "#include <linux/ctype.h>\n"
+ buf += "#include <asm/unaligned.h>\n\n"
+ buf += "#include <target/target_core_base.h>\n"
+ buf += "#include <target/target_core_transport.h>\n"
+ buf += "#include <target/target_core_fabric_ops.h>\n"
+ buf += "#include <target/target_core_fabric_configfs.h>\n"
+ buf += "#include <target/target_core_fabric_lib.h>\n"
+ buf += "#include <target/target_core_device.h>\n"
+ buf += "#include <target/target_core_tpg.h>\n"
+ buf += "#include <target/target_core_configfs.h>\n"
+ buf += "#include <target/target_core_base.h>\n"
+ buf += "#include <target/configfs_macros.h>\n\n"
+ buf += "#include <" + fabric_mod_name + "_base.h>\n"
+ buf += "#include <" + fabric_mod_name + "_fabric.h>\n\n"
+
+ buf += "/* Local pointer to allocated TCM configfs fabric module */\n"
+ buf += "struct target_fabric_configfs *" + fabric_mod_name + "_fabric_configfs;\n\n"
+
+ buf += "static struct se_node_acl *" + fabric_mod_name + "_make_nodeacl(\n"
+ buf += " struct se_portal_group *se_tpg,\n"
+ buf += " struct config_group *group,\n"
+ buf += " const char *name)\n"
+ buf += "{\n"
+ buf += " struct se_node_acl *se_nacl, *se_nacl_new;\n"
+ buf += " struct " + fabric_mod_name + "_nacl *nacl;\n"
+
+ if proto_ident == "FC" or proto_ident == "SAS":
+ buf += " u64 wwpn = 0;\n"
+
+ buf += " u32 nexus_depth;\n\n"
+ buf += " /* " + fabric_mod_name + "_parse_wwn(name, &wwpn, 1) < 0)\n"
+ buf += " return ERR_PTR(-EINVAL); */\n"
+ buf += " se_nacl_new = " + fabric_mod_name + "_alloc_fabric_acl(se_tpg);\n"
+ buf += " if (!(se_nacl_new))\n"
+ buf += " return ERR_PTR(-ENOMEM);\n"
+ buf += "//#warning FIXME: Hardcoded nexus depth in " + fabric_mod_name + "_make_nodeacl()\n"
+ buf += " nexus_depth = 1;\n"
+ buf += " /*\n"
+ buf += " * se_nacl_new may be released by core_tpg_add_initiator_node_acl()\n"
+ buf += " * when converting a NodeACL from demo mode -> explict\n"
+ buf += " */\n"
+ buf += " se_nacl = core_tpg_add_initiator_node_acl(se_tpg, se_nacl_new,\n"
+ buf += " name, nexus_depth);\n"
+ buf += " if (IS_ERR(se_nacl)) {\n"
+ buf += " " + fabric_mod_name + "_release_fabric_acl(se_tpg, se_nacl_new);\n"
+ buf += " return se_nacl;\n"
+ buf += " }\n"
+ buf += " /*\n"
+ buf += " * Locate our struct " + fabric_mod_name + "_nacl and set the FC Nport WWPN\n"
+ buf += " */\n"
+ buf += " nacl = container_of(se_nacl, struct " + fabric_mod_name + "_nacl, se_node_acl);\n"
+
+ if proto_ident == "FC" or proto_ident == "SAS":
+ buf += " nacl->" + fabric_mod_init_port + "_wwpn = wwpn;\n"
+
+ buf += " /* " + fabric_mod_name + "_format_wwn(&nacl->" + fabric_mod_init_port + "_name[0], " + fabric_mod_name.upper() + "_NAMELEN, wwpn); */\n\n"
+ buf += " return se_nacl;\n"
+ buf += "}\n\n"
+ buf += "static void " + fabric_mod_name + "_drop_nodeacl(struct se_node_acl *se_acl)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_nacl *nacl = container_of(se_acl,\n"
+ buf += " struct " + fabric_mod_name + "_nacl, se_node_acl);\n"
+ buf += " kfree(nacl);\n"
+ buf += "}\n\n"
+
+ buf += "static struct se_portal_group *" + fabric_mod_name + "_make_tpg(\n"
+ buf += " struct se_wwn *wwn,\n"
+ buf += " struct config_group *group,\n"
+ buf += " const char *name)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + "*" + fabric_mod_port + " = container_of(wwn,\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + ", " + fabric_mod_port + "_wwn);\n\n"
+ buf += " struct " + fabric_mod_name + "_tpg *tpg;\n"
+ buf += " unsigned long tpgt;\n"
+ buf += " int ret;\n\n"
+ buf += " if (strstr(name, \"tpgt_\") != name)\n"
+ buf += " return ERR_PTR(-EINVAL);\n"
+ buf += " if (strict_strtoul(name + 5, 10, &tpgt) || tpgt > UINT_MAX)\n"
+ buf += " return ERR_PTR(-EINVAL);\n\n"
+ buf += " tpg = kzalloc(sizeof(struct " + fabric_mod_name + "_tpg), GFP_KERNEL);\n"
+ buf += " if (!(tpg)) {\n"
+ buf += " printk(KERN_ERR \"Unable to allocate struct " + fabric_mod_name + "_tpg\");\n"
+ buf += " return ERR_PTR(-ENOMEM);\n"
+ buf += " }\n"
+ buf += " tpg->" + fabric_mod_port + " = " + fabric_mod_port + ";\n"
+ buf += " tpg->" + fabric_mod_port + "_tpgt = tpgt;\n\n"
+ buf += " ret = core_tpg_register(&" + fabric_mod_name + "_fabric_configfs->tf_ops, wwn,\n"
+ buf += " &tpg->se_tpg, (void *)tpg,\n"
+ buf += " TRANSPORT_TPG_TYPE_NORMAL);\n"
+ buf += " if (ret < 0) {\n"
+ buf += " kfree(tpg);\n"
+ buf += " return NULL;\n"
+ buf += " }\n"
+ buf += " return &tpg->se_tpg;\n"
+ buf += "}\n\n"
+ buf += "static void " + fabric_mod_name + "_drop_tpg(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_tpg *tpg = container_of(se_tpg,\n"
+ buf += " struct " + fabric_mod_name + "_tpg, se_tpg);\n\n"
+ buf += " core_tpg_deregister(se_tpg);\n"
+ buf += " kfree(tpg);\n"
+ buf += "}\n\n"
+
+ buf += "static struct se_wwn *" + fabric_mod_name + "_make_" + fabric_mod_port + "(\n"
+ buf += " struct target_fabric_configfs *tf,\n"
+ buf += " struct config_group *group,\n"
+ buf += " const char *name)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + " *" + fabric_mod_port + ";\n"
+
+ if proto_ident == "FC" or proto_ident == "SAS":
+ buf += " u64 wwpn = 0;\n\n"
+
+ buf += " /* if (" + fabric_mod_name + "_parse_wwn(name, &wwpn, 1) < 0)\n"
+ buf += " return ERR_PTR(-EINVAL); */\n\n"
+ buf += " " + fabric_mod_port + " = kzalloc(sizeof(struct " + fabric_mod_name + "_" + fabric_mod_port + "), GFP_KERNEL);\n"
+ buf += " if (!(" + fabric_mod_port + ")) {\n"
+ buf += " printk(KERN_ERR \"Unable to allocate struct " + fabric_mod_name + "_" + fabric_mod_port + "\");\n"
+ buf += " return ERR_PTR(-ENOMEM);\n"
+ buf += " }\n"
+
+ if proto_ident == "FC" or proto_ident == "SAS":
+ buf += " " + fabric_mod_port + "->" + fabric_mod_port + "_wwpn = wwpn;\n"
+
+ buf += " /* " + fabric_mod_name + "_format_wwn(&" + fabric_mod_port + "->" + fabric_mod_port + "_name[0], " + fabric_mod_name.upper() + "__NAMELEN, wwpn); */\n\n"
+ buf += " return &" + fabric_mod_port + "->" + fabric_mod_port + "_wwn;\n"
+ buf += "}\n\n"
+ buf += "static void " + fabric_mod_name + "_drop_" + fabric_mod_port + "(struct se_wwn *wwn)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + " *" + fabric_mod_port + " = container_of(wwn,\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + ", " + fabric_mod_port + "_wwn);\n"
+ buf += " kfree(" + fabric_mod_port + ");\n"
+ buf += "}\n\n"
+ buf += "static ssize_t " + fabric_mod_name + "_wwn_show_attr_version(\n"
+ buf += " struct target_fabric_configfs *tf,\n"
+ buf += " char *page)\n"
+ buf += "{\n"
+ buf += " return sprintf(page, \"" + fabric_mod_name.upper() + " fabric module %s on %s/%s\"\n"
+ buf += " \"on \"UTS_RELEASE\"\\n\", " + fabric_mod_name.upper() + "_VERSION, utsname()->sysname,\n"
+ buf += " utsname()->machine);\n"
+ buf += "}\n\n"
+ buf += "TF_WWN_ATTR_RO(" + fabric_mod_name + ", version);\n\n"
+ buf += "static struct configfs_attribute *" + fabric_mod_name + "_wwn_attrs[] = {\n"
+ buf += " &" + fabric_mod_name + "_wwn_version.attr,\n"
+ buf += " NULL,\n"
+ buf += "};\n\n"
+
+ buf += "static struct target_core_fabric_ops " + fabric_mod_name + "_ops = {\n"
+ buf += " .get_fabric_name = " + fabric_mod_name + "_get_fabric_name,\n"
+ buf += " .get_fabric_proto_ident = " + fabric_mod_name + "_get_fabric_proto_ident,\n"
+ buf += " .tpg_get_wwn = " + fabric_mod_name + "_get_fabric_wwn,\n"
+ buf += " .tpg_get_tag = " + fabric_mod_name + "_get_tag,\n"
+ buf += " .tpg_get_default_depth = " + fabric_mod_name + "_get_default_depth,\n"
+ buf += " .tpg_get_pr_transport_id = " + fabric_mod_name + "_get_pr_transport_id,\n"
+ buf += " .tpg_get_pr_transport_id_len = " + fabric_mod_name + "_get_pr_transport_id_len,\n"
+ buf += " .tpg_parse_pr_out_transport_id = " + fabric_mod_name + "_parse_pr_out_transport_id,\n"
+ buf += " .tpg_check_demo_mode = " + fabric_mod_name + "_check_false,\n"
+ buf += " .tpg_check_demo_mode_cache = " + fabric_mod_name + "_check_true,\n"
+ buf += " .tpg_check_demo_mode_write_protect = " + fabric_mod_name + "_check_true,\n"
+ buf += " .tpg_check_prod_mode_write_protect = " + fabric_mod_name + "_check_false,\n"
+ buf += " .tpg_alloc_fabric_acl = " + fabric_mod_name + "_alloc_fabric_acl,\n"
+ buf += " .tpg_release_fabric_acl = " + fabric_mod_name + "_release_fabric_acl,\n"
+ buf += " .tpg_get_inst_index = " + fabric_mod_name + "_tpg_get_inst_index,\n"
+ buf += " .release_cmd_to_pool = " + fabric_mod_name + "_release_cmd,\n"
+ buf += " .release_cmd_direct = " + fabric_mod_name + "_release_cmd,\n"
+ buf += " .shutdown_session = " + fabric_mod_name + "_shutdown_session,\n"
+ buf += " .close_session = " + fabric_mod_name + "_close_session,\n"
+ buf += " .stop_session = " + fabric_mod_name + "_stop_session,\n"
+ buf += " .fall_back_to_erl0 = " + fabric_mod_name + "_reset_nexus,\n"
+ buf += " .sess_logged_in = " + fabric_mod_name + "_sess_logged_in,\n"
+ buf += " .sess_get_index = " + fabric_mod_name + "_sess_get_index,\n"
+ buf += " .sess_get_initiator_sid = NULL,\n"
+ buf += " .write_pending = " + fabric_mod_name + "_write_pending,\n"
+ buf += " .write_pending_status = " + fabric_mod_name + "_write_pending_status,\n"
+ buf += " .set_default_node_attributes = " + fabric_mod_name + "_set_default_node_attrs,\n"
+ buf += " .get_task_tag = " + fabric_mod_name + "_get_task_tag,\n"
+ buf += " .get_cmd_state = " + fabric_mod_name + "_get_cmd_state,\n"
+ buf += " .new_cmd_failure = " + fabric_mod_name + "_new_cmd_failure,\n"
+ buf += " .queue_data_in = " + fabric_mod_name + "_queue_data_in,\n"
+ buf += " .queue_status = " + fabric_mod_name + "_queue_status,\n"
+ buf += " .queue_tm_rsp = " + fabric_mod_name + "_queue_tm_rsp,\n"
+ buf += " .get_fabric_sense_len = " + fabric_mod_name + "_get_fabric_sense_len,\n"
+ buf += " .set_fabric_sense_len = " + fabric_mod_name + "_set_fabric_sense_len,\n"
+ buf += " .is_state_remove = " + fabric_mod_name + "_is_state_remove,\n"
+ buf += " .pack_lun = " + fabric_mod_name + "_pack_lun,\n"
+ buf += " /*\n"
+ buf += " * Setup function pointers for generic logic in target_core_fabric_configfs.c\n"
+ buf += " */\n"
+ buf += " .fabric_make_wwn = " + fabric_mod_name + "_make_" + fabric_mod_port + ",\n"
+ buf += " .fabric_drop_wwn = " + fabric_mod_name + "_drop_" + fabric_mod_port + ",\n"
+ buf += " .fabric_make_tpg = " + fabric_mod_name + "_make_tpg,\n"
+ buf += " .fabric_drop_tpg = " + fabric_mod_name + "_drop_tpg,\n"
+ buf += " .fabric_post_link = NULL,\n"
+ buf += " .fabric_pre_unlink = NULL,\n"
+ buf += " .fabric_make_np = NULL,\n"
+ buf += " .fabric_drop_np = NULL,\n"
+ buf += " .fabric_make_nodeacl = " + fabric_mod_name + "_make_nodeacl,\n"
+ buf += " .fabric_drop_nodeacl = " + fabric_mod_name + "_drop_nodeacl,\n"
+ buf += "};\n\n"
+
+ buf += "static int " + fabric_mod_name + "_register_configfs(void)\n"
+ buf += "{\n"
+ buf += " struct target_fabric_configfs *fabric;\n"
+ buf += " int ret;\n\n"
+ buf += " printk(KERN_INFO \"" + fabric_mod_name.upper() + " fabric module %s on %s/%s\"\n"
+ buf += " \" on \"UTS_RELEASE\"\\n\"," + fabric_mod_name.upper() + "_VERSION, utsname()->sysname,\n"
+ buf += " utsname()->machine);\n"
+ buf += " /*\n"
+ buf += " * Register the top level struct config_item_type with TCM core\n"
+ buf += " */\n"
+ buf += " fabric = target_fabric_configfs_init(THIS_MODULE, \"" + fabric_mod_name[4:] + "\");\n"
+ buf += " if (!(fabric)) {\n"
+ buf += " printk(KERN_ERR \"target_fabric_configfs_init() failed\\n\");\n"
+ buf += " return -ENOMEM;\n"
+ buf += " }\n"
+ buf += " /*\n"
+ buf += " * Setup fabric->tf_ops from our local " + fabric_mod_name + "_ops\n"
+ buf += " */\n"
+ buf += " fabric->tf_ops = " + fabric_mod_name + "_ops;\n"
+ buf += " /*\n"
+ buf += " * Setup default attribute lists for various fabric->tf_cit_tmpl\n"
+ buf += " */\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_wwn_cit.ct_attrs = " + fabric_mod_name + "_wwn_attrs;\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_tpg_base_cit.ct_attrs = NULL;\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_tpg_attrib_cit.ct_attrs = NULL;\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_tpg_param_cit.ct_attrs = NULL;\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_tpg_np_base_cit.ct_attrs = NULL;\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_tpg_nacl_base_cit.ct_attrs = NULL;\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_tpg_nacl_attrib_cit.ct_attrs = NULL;\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_tpg_nacl_auth_cit.ct_attrs = NULL;\n"
+ buf += " TF_CIT_TMPL(fabric)->tfc_tpg_nacl_param_cit.ct_attrs = NULL;\n"
+ buf += " /*\n"
+ buf += " * Register the fabric for use within TCM\n"
+ buf += " */\n"
+ buf += " ret = target_fabric_configfs_register(fabric);\n"
+ buf += " if (ret < 0) {\n"
+ buf += " printk(KERN_ERR \"target_fabric_configfs_register() failed\"\n"
+ buf += " \" for " + fabric_mod_name.upper() + "\\n\");\n"
+ buf += " return ret;\n"
+ buf += " }\n"
+ buf += " /*\n"
+ buf += " * Setup our local pointer to *fabric\n"
+ buf += " */\n"
+ buf += " " + fabric_mod_name + "_fabric_configfs = fabric;\n"
+ buf += " printk(KERN_INFO \"" + fabric_mod_name.upper() + "[0] - Set fabric -> " + fabric_mod_name + "_fabric_configfs\\n\");\n"
+ buf += " return 0;\n"
+ buf += "};\n\n"
+ buf += "static void " + fabric_mod_name + "_deregister_configfs(void)\n"
+ buf += "{\n"
+ buf += " if (!(" + fabric_mod_name + "_fabric_configfs))\n"
+ buf += " return;\n\n"
+ buf += " target_fabric_configfs_deregister(" + fabric_mod_name + "_fabric_configfs);\n"
+ buf += " " + fabric_mod_name + "_fabric_configfs = NULL;\n"
+ buf += " printk(KERN_INFO \"" + fabric_mod_name.upper() + "[0] - Cleared " + fabric_mod_name + "_fabric_configfs\\n\");\n"
+ buf += "};\n\n"
+
+ buf += "static int __init " + fabric_mod_name + "_init(void)\n"
+ buf += "{\n"
+ buf += " int ret;\n\n"
+ buf += " ret = " + fabric_mod_name + "_register_configfs();\n"
+ buf += " if (ret < 0)\n"
+ buf += " return ret;\n\n"
+ buf += " return 0;\n"
+ buf += "};\n\n"
+ buf += "static void " + fabric_mod_name + "_exit(void)\n"
+ buf += "{\n"
+ buf += " " + fabric_mod_name + "_deregister_configfs();\n"
+ buf += "};\n\n"
+
+ buf += "#ifdef MODULE\n"
+ buf += "MODULE_DESCRIPTION(\"" + fabric_mod_name.upper() + " series fabric driver\");\n"
+ buf += "MODULE_LICENSE(\"GPL\");\n"
+ buf += "module_init(" + fabric_mod_name + "_init);\n"
+ buf += "module_exit(" + fabric_mod_name + "_exit);\n"
+ buf += "#endif\n"
+
+ ret = p.write(buf)
+ if ret:
+ tcm_mod_err("Unable to write f: " + f)
+
+ p.close()
+
+ return
+
+def tcm_mod_scan_fabric_ops(tcm_dir):
+
+ fabric_ops_api = tcm_dir + "include/target/target_core_fabric_ops.h"
+
+ print "Using tcm_mod_scan_fabric_ops: " + fabric_ops_api
+ process_fo = 0;
+
+ p = open(fabric_ops_api, 'r')
+
+ line = p.readline()
+ while line:
+ if process_fo == 0 and re.search('struct target_core_fabric_ops {', line):
+ line = p.readline()
+ continue
+
+ if process_fo == 0:
+ process_fo = 1;
+ line = p.readline()
+ # Search for function pointer
+ if not re.search('\(\*', line):
+ continue
+
+ fabric_ops.append(line.rstrip())
+ continue
+
+ line = p.readline()
+ # Search for function pointer
+ if not re.search('\(\*', line):
+ continue
+
+ fabric_ops.append(line.rstrip())
+
+ p.close()
+ return
+
+def tcm_mod_dump_fabric_ops(proto_ident, fabric_mod_dir_var, fabric_mod_name):
+ buf = ""
+ bufi = ""
+
+ f = fabric_mod_dir_var + "/" + fabric_mod_name + "_fabric.c"
+ print "Writing file: " + f
+
+ p = open(f, 'w')
+ if not p:
+ tcm_mod_err("Unable to open file: " + f)
+
+ fi = fabric_mod_dir_var + "/" + fabric_mod_name + "_fabric.h"
+ print "Writing file: " + fi
+
+ pi = open(fi, 'w')
+ if not pi:
+ tcm_mod_err("Unable to open file: " + fi)
+
+ buf = "#include <linux/slab.h>\n"
+ buf += "#include <linux/kthread.h>\n"
+ buf += "#include <linux/types.h>\n"
+ buf += "#include <linux/list.h>\n"
+ buf += "#include <linux/types.h>\n"
+ buf += "#include <linux/string.h>\n"
+ buf += "#include <linux/ctype.h>\n"
+ buf += "#include <asm/unaligned.h>\n"
+ buf += "#include <scsi/scsi.h>\n"
+ buf += "#include <scsi/scsi_host.h>\n"
+ buf += "#include <scsi/scsi_device.h>\n"
+ buf += "#include <scsi/scsi_cmnd.h>\n"
+ buf += "#include <scsi/libfc.h>\n\n"
+ buf += "#include <target/target_core_base.h>\n"
+ buf += "#include <target/target_core_transport.h>\n"
+ buf += "#include <target/target_core_fabric_ops.h>\n"
+ buf += "#include <target/target_core_fabric_lib.h>\n"
+ buf += "#include <target/target_core_device.h>\n"
+ buf += "#include <target/target_core_tpg.h>\n"
+ buf += "#include <target/target_core_configfs.h>\n"
+ buf += "#include <" + fabric_mod_name + "_base.h>\n"
+ buf += "#include <" + fabric_mod_name + "_fabric.h>\n\n"
+
+ buf += "int " + fabric_mod_name + "_check_true(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " return 1;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_check_true(struct se_portal_group *);\n"
+
+ buf += "int " + fabric_mod_name + "_check_false(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_check_false(struct se_portal_group *);\n"
+
+ total_fabric_ops = len(fabric_ops)
+ i = 0
+
+ while i < total_fabric_ops:
+ fo = fabric_ops[i]
+ i += 1
+# print "fabric_ops: " + fo
+
+ if re.search('get_fabric_name', fo):
+ buf += "char *" + fabric_mod_name + "_get_fabric_name(void)\n"
+ buf += "{\n"
+ buf += " return \"" + fabric_mod_name[4:] + "\";\n"
+ buf += "}\n\n"
+ bufi += "char *" + fabric_mod_name + "_get_fabric_name(void);\n"
+ continue
+
+ if re.search('get_fabric_proto_ident', fo):
+ buf += "u8 " + fabric_mod_name + "_get_fabric_proto_ident(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_tpg *tpg = container_of(se_tpg,\n"
+ buf += " struct " + fabric_mod_name + "_tpg, se_tpg);\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + " *" + fabric_mod_port + " = tpg->" + fabric_mod_port + ";\n"
+ buf += " u8 proto_id;\n\n"
+ buf += " switch (" + fabric_mod_port + "->" + fabric_mod_port + "_proto_id) {\n"
+ if proto_ident == "FC":
+ buf += " case SCSI_PROTOCOL_FCP:\n"
+ buf += " default:\n"
+ buf += " proto_id = fc_get_fabric_proto_ident(se_tpg);\n"
+ buf += " break;\n"
+ elif proto_ident == "SAS":
+ buf += " case SCSI_PROTOCOL_SAS:\n"
+ buf += " default:\n"
+ buf += " proto_id = sas_get_fabric_proto_ident(se_tpg);\n"
+ buf += " break;\n"
+ elif proto_ident == "iSCSI":
+ buf += " case SCSI_PROTOCOL_ISCSI:\n"
+ buf += " default:\n"
+ buf += " proto_id = iscsi_get_fabric_proto_ident(se_tpg);\n"
+ buf += " break;\n"
+
+ buf += " }\n\n"
+ buf += " return proto_id;\n"
+ buf += "}\n\n"
+ bufi += "u8 " + fabric_mod_name + "_get_fabric_proto_ident(struct se_portal_group *);\n"
+
+ if re.search('get_wwn', fo):
+ buf += "char *" + fabric_mod_name + "_get_fabric_wwn(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_tpg *tpg = container_of(se_tpg,\n"
+ buf += " struct " + fabric_mod_name + "_tpg, se_tpg);\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + " *" + fabric_mod_port + " = tpg->" + fabric_mod_port + ";\n\n"
+ buf += " return &" + fabric_mod_port + "->" + fabric_mod_port + "_name[0];\n"
+ buf += "}\n\n"
+ bufi += "char *" + fabric_mod_name + "_get_fabric_wwn(struct se_portal_group *);\n"
+
+ if re.search('get_tag', fo):
+ buf += "u16 " + fabric_mod_name + "_get_tag(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_tpg *tpg = container_of(se_tpg,\n"
+ buf += " struct " + fabric_mod_name + "_tpg, se_tpg);\n"
+ buf += " return tpg->" + fabric_mod_port + "_tpgt;\n"
+ buf += "}\n\n"
+ bufi += "u16 " + fabric_mod_name + "_get_tag(struct se_portal_group *);\n"
+
+ if re.search('get_default_depth', fo):
+ buf += "u32 " + fabric_mod_name + "_get_default_depth(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " return 1;\n"
+ buf += "}\n\n"
+ bufi += "u32 " + fabric_mod_name + "_get_default_depth(struct se_portal_group *);\n"
+
+ if re.search('get_pr_transport_id\)\(', fo):
+ buf += "u32 " + fabric_mod_name + "_get_pr_transport_id(\n"
+ buf += " struct se_portal_group *se_tpg,\n"
+ buf += " struct se_node_acl *se_nacl,\n"
+ buf += " struct t10_pr_registration *pr_reg,\n"
+ buf += " int *format_code,\n"
+ buf += " unsigned char *buf)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_tpg *tpg = container_of(se_tpg,\n"
+ buf += " struct " + fabric_mod_name + "_tpg, se_tpg);\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + " *" + fabric_mod_port + " = tpg->" + fabric_mod_port + ";\n"
+ buf += " int ret = 0;\n\n"
+ buf += " switch (" + fabric_mod_port + "->" + fabric_mod_port + "_proto_id) {\n"
+ if proto_ident == "FC":
+ buf += " case SCSI_PROTOCOL_FCP:\n"
+ buf += " default:\n"
+ buf += " ret = fc_get_pr_transport_id(se_tpg, se_nacl, pr_reg,\n"
+ buf += " format_code, buf);\n"
+ buf += " break;\n"
+ elif proto_ident == "SAS":
+ buf += " case SCSI_PROTOCOL_SAS:\n"
+ buf += " default:\n"
+ buf += " ret = sas_get_pr_transport_id(se_tpg, se_nacl, pr_reg,\n"
+ buf += " format_code, buf);\n"
+ buf += " break;\n"
+ elif proto_ident == "iSCSI":
+ buf += " case SCSI_PROTOCOL_ISCSI:\n"
+ buf += " default:\n"
+ buf += " ret = iscsi_get_pr_transport_id(se_tpg, se_nacl, pr_reg,\n"
+ buf += " format_code, buf);\n"
+ buf += " break;\n"
+
+ buf += " }\n\n"
+ buf += " return ret;\n"
+ buf += "}\n\n"
+ bufi += "u32 " + fabric_mod_name + "_get_pr_transport_id(struct se_portal_group *,\n"
+ bufi += " struct se_node_acl *, struct t10_pr_registration *,\n"
+ bufi += " int *, unsigned char *);\n"
+
+ if re.search('get_pr_transport_id_len\)\(', fo):
+ buf += "u32 " + fabric_mod_name + "_get_pr_transport_id_len(\n"
+ buf += " struct se_portal_group *se_tpg,\n"
+ buf += " struct se_node_acl *se_nacl,\n"
+ buf += " struct t10_pr_registration *pr_reg,\n"
+ buf += " int *format_code)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_tpg *tpg = container_of(se_tpg,\n"
+ buf += " struct " + fabric_mod_name + "_tpg, se_tpg);\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + " *" + fabric_mod_port + " = tpg->" + fabric_mod_port + ";\n"
+ buf += " int ret = 0;\n\n"
+ buf += " switch (" + fabric_mod_port + "->" + fabric_mod_port + "_proto_id) {\n"
+ if proto_ident == "FC":
+ buf += " case SCSI_PROTOCOL_FCP:\n"
+ buf += " default:\n"
+ buf += " ret = fc_get_pr_transport_id_len(se_tpg, se_nacl, pr_reg,\n"
+ buf += " format_code);\n"
+ buf += " break;\n"
+ elif proto_ident == "SAS":
+ buf += " case SCSI_PROTOCOL_SAS:\n"
+ buf += " default:\n"
+ buf += " ret = sas_get_pr_transport_id_len(se_tpg, se_nacl, pr_reg,\n"
+ buf += " format_code);\n"
+ buf += " break;\n"
+ elif proto_ident == "iSCSI":
+ buf += " case SCSI_PROTOCOL_ISCSI:\n"
+ buf += " default:\n"
+ buf += " ret = iscsi_get_pr_transport_id_len(se_tpg, se_nacl, pr_reg,\n"
+ buf += " format_code);\n"
+ buf += " break;\n"
+
+
+ buf += " }\n\n"
+ buf += " return ret;\n"
+ buf += "}\n\n"
+ bufi += "u32 " + fabric_mod_name + "_get_pr_transport_id_len(struct se_portal_group *,\n"
+ bufi += " struct se_node_acl *, struct t10_pr_registration *,\n"
+ bufi += " int *);\n"
+
+ if re.search('parse_pr_out_transport_id\)\(', fo):
+ buf += "char *" + fabric_mod_name + "_parse_pr_out_transport_id(\n"
+ buf += " struct se_portal_group *se_tpg,\n"
+ buf += " const char *buf,\n"
+ buf += " u32 *out_tid_len,\n"
+ buf += " char **port_nexus_ptr)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_tpg *tpg = container_of(se_tpg,\n"
+ buf += " struct " + fabric_mod_name + "_tpg, se_tpg);\n"
+ buf += " struct " + fabric_mod_name + "_" + fabric_mod_port + " *" + fabric_mod_port + " = tpg->" + fabric_mod_port + ";\n"
+ buf += " char *tid = NULL;\n\n"
+ buf += " switch (" + fabric_mod_port + "->" + fabric_mod_port + "_proto_id) {\n"
+ if proto_ident == "FC":
+ buf += " case SCSI_PROTOCOL_FCP:\n"
+ buf += " default:\n"
+ buf += " tid = fc_parse_pr_out_transport_id(se_tpg, buf, out_tid_len,\n"
+ buf += " port_nexus_ptr);\n"
+ elif proto_ident == "SAS":
+ buf += " case SCSI_PROTOCOL_SAS:\n"
+ buf += " default:\n"
+ buf += " tid = sas_parse_pr_out_transport_id(se_tpg, buf, out_tid_len,\n"
+ buf += " port_nexus_ptr);\n"
+ elif proto_ident == "iSCSI":
+ buf += " case SCSI_PROTOCOL_ISCSI:\n"
+ buf += " default:\n"
+ buf += " tid = iscsi_parse_pr_out_transport_id(se_tpg, buf, out_tid_len,\n"
+ buf += " port_nexus_ptr);\n"
+
+ buf += " }\n\n"
+ buf += " return tid;\n"
+ buf += "}\n\n"
+ bufi += "char *" + fabric_mod_name + "_parse_pr_out_transport_id(struct se_portal_group *,\n"
+ bufi += " const char *, u32 *, char **);\n"
+
+ if re.search('alloc_fabric_acl\)\(', fo):
+ buf += "struct se_node_acl *" + fabric_mod_name + "_alloc_fabric_acl(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_nacl *nacl;\n\n"
+ buf += " nacl = kzalloc(sizeof(struct " + fabric_mod_name + "_nacl), GFP_KERNEL);\n"
+ buf += " if (!(nacl)) {\n"
+ buf += " printk(KERN_ERR \"Unable to alocate struct " + fabric_mod_name + "_nacl\\n\");\n"
+ buf += " return NULL;\n"
+ buf += " }\n\n"
+ buf += " return &nacl->se_node_acl;\n"
+ buf += "}\n\n"
+ bufi += "struct se_node_acl *" + fabric_mod_name + "_alloc_fabric_acl(struct se_portal_group *);\n"
+
+ if re.search('release_fabric_acl\)\(', fo):
+ buf += "void " + fabric_mod_name + "_release_fabric_acl(\n"
+ buf += " struct se_portal_group *se_tpg,\n"
+ buf += " struct se_node_acl *se_nacl)\n"
+ buf += "{\n"
+ buf += " struct " + fabric_mod_name + "_nacl *nacl = container_of(se_nacl,\n"
+ buf += " struct " + fabric_mod_name + "_nacl, se_node_acl);\n"
+ buf += " kfree(nacl);\n"
+ buf += "}\n\n"
+ bufi += "void " + fabric_mod_name + "_release_fabric_acl(struct se_portal_group *,\n"
+ bufi += " struct se_node_acl *);\n"
+
+ if re.search('tpg_get_inst_index\)\(', fo):
+ buf += "u32 " + fabric_mod_name + "_tpg_get_inst_index(struct se_portal_group *se_tpg)\n"
+ buf += "{\n"
+ buf += " return 1;\n"
+ buf += "}\n\n"
+ bufi += "u32 " + fabric_mod_name + "_tpg_get_inst_index(struct se_portal_group *);\n"
+
+ if re.search('release_cmd_to_pool', fo):
+ buf += "void " + fabric_mod_name + "_release_cmd(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return;\n"
+ buf += "}\n\n"
+ bufi += "void " + fabric_mod_name + "_release_cmd(struct se_cmd *);\n"
+
+ if re.search('shutdown_session\)\(', fo):
+ buf += "int " + fabric_mod_name + "_shutdown_session(struct se_session *se_sess)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_shutdown_session(struct se_session *);\n"
+
+ if re.search('close_session\)\(', fo):
+ buf += "void " + fabric_mod_name + "_close_session(struct se_session *se_sess)\n"
+ buf += "{\n"
+ buf += " return;\n"
+ buf += "}\n\n"
+ bufi += "void " + fabric_mod_name + "_close_session(struct se_session *);\n"
+
+ if re.search('stop_session\)\(', fo):
+ buf += "void " + fabric_mod_name + "_stop_session(struct se_session *se_sess, int sess_sleep , int conn_sleep)\n"
+ buf += "{\n"
+ buf += " return;\n"
+ buf += "}\n\n"
+ bufi += "void " + fabric_mod_name + "_stop_session(struct se_session *, int, int);\n"
+
+ if re.search('fall_back_to_erl0\)\(', fo):
+ buf += "void " + fabric_mod_name + "_reset_nexus(struct se_session *se_sess)\n"
+ buf += "{\n"
+ buf += " return;\n"
+ buf += "}\n\n"
+ bufi += "void " + fabric_mod_name + "_reset_nexus(struct se_session *);\n"
+
+ if re.search('sess_logged_in\)\(', fo):
+ buf += "int " + fabric_mod_name + "_sess_logged_in(struct se_session *se_sess)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_sess_logged_in(struct se_session *);\n"
+
+ if re.search('sess_get_index\)\(', fo):
+ buf += "u32 " + fabric_mod_name + "_sess_get_index(struct se_session *se_sess)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "u32 " + fabric_mod_name + "_sess_get_index(struct se_session *);\n"
+
+ if re.search('write_pending\)\(', fo):
+ buf += "int " + fabric_mod_name + "_write_pending(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_write_pending(struct se_cmd *);\n"
+
+ if re.search('write_pending_status\)\(', fo):
+ buf += "int " + fabric_mod_name + "_write_pending_status(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_write_pending_status(struct se_cmd *);\n"
+
+ if re.search('set_default_node_attributes\)\(', fo):
+ buf += "void " + fabric_mod_name + "_set_default_node_attrs(struct se_node_acl *nacl)\n"
+ buf += "{\n"
+ buf += " return;\n"
+ buf += "}\n\n"
+ bufi += "void " + fabric_mod_name + "_set_default_node_attrs(struct se_node_acl *);\n"
+
+ if re.search('get_task_tag\)\(', fo):
+ buf += "u32 " + fabric_mod_name + "_get_task_tag(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "u32 " + fabric_mod_name + "_get_task_tag(struct se_cmd *);\n"
+
+ if re.search('get_cmd_state\)\(', fo):
+ buf += "int " + fabric_mod_name + "_get_cmd_state(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_get_cmd_state(struct se_cmd *);\n"
+
+ if re.search('new_cmd_failure\)\(', fo):
+ buf += "void " + fabric_mod_name + "_new_cmd_failure(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return;\n"
+ buf += "}\n\n"
+ bufi += "void " + fabric_mod_name + "_new_cmd_failure(struct se_cmd *);\n"
+
+ if re.search('queue_data_in\)\(', fo):
+ buf += "int " + fabric_mod_name + "_queue_data_in(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_queue_data_in(struct se_cmd *);\n"
+
+ if re.search('queue_status\)\(', fo):
+ buf += "int " + fabric_mod_name + "_queue_status(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_queue_status(struct se_cmd *);\n"
+
+ if re.search('queue_tm_rsp\)\(', fo):
+ buf += "int " + fabric_mod_name + "_queue_tm_rsp(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_queue_tm_rsp(struct se_cmd *);\n"
+
+ if re.search('get_fabric_sense_len\)\(', fo):
+ buf += "u16 " + fabric_mod_name + "_get_fabric_sense_len(void)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "u16 " + fabric_mod_name + "_get_fabric_sense_len(void);\n"
+
+ if re.search('set_fabric_sense_len\)\(', fo):
+ buf += "u16 " + fabric_mod_name + "_set_fabric_sense_len(struct se_cmd *se_cmd, u32 sense_length)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "u16 " + fabric_mod_name + "_set_fabric_sense_len(struct se_cmd *, u32);\n"
+
+ if re.search('is_state_remove\)\(', fo):
+ buf += "int " + fabric_mod_name + "_is_state_remove(struct se_cmd *se_cmd)\n"
+ buf += "{\n"
+ buf += " return 0;\n"
+ buf += "}\n\n"
+ bufi += "int " + fabric_mod_name + "_is_state_remove(struct se_cmd *);\n"
+
+ if re.search('pack_lun\)\(', fo):
+ buf += "u64 " + fabric_mod_name + "_pack_lun(unsigned int lun)\n"
+ buf += "{\n"
+ buf += " WARN_ON(lun >= 256);\n"
+ buf += " /* Caller wants this byte-swapped */\n"
+ buf += " return cpu_to_le64((lun & 0xff) << 8);\n"
+ buf += "}\n\n"
+ bufi += "u64 " + fabric_mod_name + "_pack_lun(unsigned int);\n"
+
+
+ ret = p.write(buf)
+ if ret:
+ tcm_mod_err("Unable to write f: " + f)
+
+ p.close()
+
+ ret = pi.write(bufi)
+ if ret:
+ tcm_mod_err("Unable to write fi: " + fi)
+
+ pi.close()
+ return
+
+def tcm_mod_build_kbuild(fabric_mod_dir_var, fabric_mod_name):
+
+ buf = ""
+ f = fabric_mod_dir_var + "/Kbuild"
+ print "Writing file: " + f
+
+ p = open(f, 'w')
+ if not p:
+ tcm_mod_err("Unable to open file: " + f)
+
+ buf = "EXTRA_CFLAGS += -I$(srctree)/drivers/target/ -I$(srctree)/include/ -I$(srctree)/drivers/scsi/ -I$(srctree)/include/scsi/ -I$(srctree)/drivers/target/" + fabric_mod_name + "\n\n"
+ buf += fabric_mod_name + "-objs := " + fabric_mod_name + "_fabric.o \\\n"
+ buf += " " + fabric_mod_name + "_configfs.o\n"
+ buf += "obj-$(CONFIG_" + fabric_mod_name.upper() + ") += " + fabric_mod_name + ".o\n"
+
+ ret = p.write(buf)
+ if ret:
+ tcm_mod_err("Unable to write f: " + f)
+
+ p.close()
+ return
+
+def tcm_mod_build_kconfig(fabric_mod_dir_var, fabric_mod_name):
+
+ buf = ""
+ f = fabric_mod_dir_var + "/Kconfig"
+ print "Writing file: " + f
+
+ p = open(f, 'w')
+ if not p:
+ tcm_mod_err("Unable to open file: " + f)
+
+ buf = "config " + fabric_mod_name.upper() + "\n"
+ buf += " tristate \"" + fabric_mod_name.upper() + " fabric module\"\n"
+ buf += " depends on TARGET_CORE && CONFIGFS_FS\n"
+ buf += " default n\n"
+ buf += " ---help---\n"
+ buf += " Say Y here to enable the " + fabric_mod_name.upper() + " fabric module\n"
+
+ ret = p.write(buf)
+ if ret:
+ tcm_mod_err("Unable to write f: " + f)
+
+ p.close()
+ return
+
+def tcm_mod_add_kbuild(tcm_dir, fabric_mod_name):
+ buf = "obj-$(CONFIG_" + fabric_mod_name.upper() + ") += " + fabric_mod_name.lower() + "/\n"
+ kbuild = tcm_dir + "/drivers/target/Kbuild"
+
+ f = open(kbuild, 'a')
+ f.write(buf)
+ f.close()
+ return
+
+def tcm_mod_add_kconfig(tcm_dir, fabric_mod_name):
+ buf = "source \"drivers/target/" + fabric_mod_name.lower() + "/Kconfig\"\n"
+ kconfig = tcm_dir + "/drivers/target/Kconfig"
+
+ f = open(kconfig, 'a')
+ f.write(buf)
+ f.close()
+ return
+
+def main(modname, proto_ident):
+# proto_ident = "FC"
+# proto_ident = "SAS"
+# proto_ident = "iSCSI"
+
+ tcm_dir = os.getcwd();
+ tcm_dir += "/../../"
+ print "tcm_dir: " + tcm_dir
+ fabric_mod_name = modname
+ fabric_mod_dir = tcm_dir + "drivers/target/" + fabric_mod_name
+ print "Set fabric_mod_name: " + fabric_mod_name
+ print "Set fabric_mod_dir: " + fabric_mod_dir
+ print "Using proto_ident: " + proto_ident
+
+ if proto_ident != "FC" and proto_ident != "SAS" and proto_ident != "iSCSI":
+ print "Unsupported proto_ident: " + proto_ident
+ sys.exit(1)
+
+ ret = tcm_mod_create_module_subdir(fabric_mod_dir)
+ if ret:
+ print "tcm_mod_create_module_subdir() failed because module already exists!"
+ sys.exit(1)
+
+ tcm_mod_build_base_includes(proto_ident, fabric_mod_dir, fabric_mod_name)
+ tcm_mod_scan_fabric_ops(tcm_dir)
+ tcm_mod_dump_fabric_ops(proto_ident, fabric_mod_dir, fabric_mod_name)
+ tcm_mod_build_configfs(proto_ident, fabric_mod_dir, fabric_mod_name)
+ tcm_mod_build_kbuild(fabric_mod_dir, fabric_mod_name)
+ tcm_mod_build_kconfig(fabric_mod_dir, fabric_mod_name)
+
+ input = raw_input("Would you like to add " + fabric_mod_name + "to drivers/target/Kbuild..? [yes,no]: ")
+ if input == "yes" or input == "y":
+ tcm_mod_add_kbuild(tcm_dir, fabric_mod_name)
+
+ input = raw_input("Would you like to add " + fabric_mod_name + "to drivers/target/Kconfig..? [yes,no]: ")
+ if input == "yes" or input == "y":
+ tcm_mod_add_kconfig(tcm_dir, fabric_mod_name)
+
+ return
+
+parser = optparse.OptionParser()
+parser.add_option('-m', '--modulename', help='Module name', dest='modname',
+ action='store', nargs=1, type='string')
+parser.add_option('-p', '--protoident', help='Protocol Ident', dest='protoident',
+ action='store', nargs=1, type='string')
+
+(opts, args) = parser.parse_args()
+
+mandatories = ['modname', 'protoident']
+for m in mandatories:
+ if not opts.__dict__[m]:
+ print "mandatory option is missing\n"
+ parser.print_help()
+ exit(-1)
+
+if __name__ == "__main__":
+
+ main(str(opts.modname), opts.protoident)
diff --git a/Documentation/target/tcm_mod_builder.txt b/Documentation/target/tcm_mod_builder.txt
new file mode 100644
index 0000000..84533d8
--- /dev/null
+++ b/Documentation/target/tcm_mod_builder.txt
@@ -0,0 +1,145 @@
+>>>>>>>>>> The TCM v4 fabric module script generator <<<<<<<<<<
+
+Greetings all,
+
+This document is intended to be a mini-HOWTO for using the tcm_mod_builder.py
+script to generate a brand new functional TCM v4 fabric .ko module of your very own,
+that once built can be immediately be loaded to start access the new TCM/ConfigFS
+fabric skeleton, by simply using:
+
+ modprobe $TCM_NEW_MOD
+ mkdir -p /sys/kernel/config/target/$TCM_NEW_MOD
+
+This script will create a new drivers/target/$TCM_NEW_MOD/, and will do the following
+
+ *) Generate new API callers for drivers/target/target_core_fabric_configs.c logic
+ ->make_nodeacl(), ->drop_nodeacl(), ->make_tpg(), ->drop_tpg()
+ ->make_wwn(), ->drop_wwn(). These are created into $TCM_NEW_MOD/$TCM_NEW_MOD_configfs.c
+ *) Generate basic infrastructure for loading/unloading LKMs and TCM/ConfigFS fabric module
+ using a skeleton struct target_core_fabric_ops API template.
+ *) Based on user defined T10 Proto_Ident for the new fabric module being built,
+ the TransportID / Initiator and Target WWPN related handlers for
+ SPC-3 persistent reservation are automatically generated in $TCM_NEW_MOD/$TCM_NEW_MOD_fabric.c
+ using drivers/target/target_core_fabric_lib.c logic.
+ *) NOP API calls for all other Data I/O path and fabric dependent attribute logic
+ in $TCM_NEW_MOD/$TCM_NEW_MOD_fabric.c
+
+tcm_mod_builder.py depends upon the mandatory '-p $PROTO_IDENT' and '-m
+$FABRIC_MOD_name' parameters, and actually running the script looks like:
+
+target:/mnt/sdb/lio-core-2.6.git/Documentation/target# python tcm_mod_builder.py -p iSCSI -m tcm_nab5000
+tcm_dir: /mnt/sdb/lio-core-2.6.git/Documentation/target/../../
+Set fabric_mod_name: tcm_nab5000
+Set fabric_mod_dir:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../drivers/target/tcm_nab5000
+Using proto_ident: iSCSI
+Creating fabric_mod_dir:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../drivers/target/tcm_nab5000
+Writing file:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../drivers/target/tcm_nab5000/tcm_nab5000_base.h
+Using tcm_mod_scan_fabric_ops:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../include/target/target_core_fabric_ops.h
+Writing file:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../drivers/target/tcm_nab5000/tcm_nab5000_fabric.c
+Writing file:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../drivers/target/tcm_nab5000/tcm_nab5000_fabric.h
+Writing file:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../drivers/target/tcm_nab5000/tcm_nab5000_configfs.c
+Writing file:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../drivers/target/tcm_nab5000/Kbuild
+Writing file:
+/mnt/sdb/lio-core-2.6.git/Documentation/target/../../drivers/target/tcm_nab5000/Kconfig
+Would you like to add tcm_nab5000to drivers/target/Kbuild..? [yes,no]: yes
+Would you like to add tcm_nab5000to drivers/target/Kconfig..? [yes,no]: yes
+
+At the end of tcm_mod_builder.py. the script will ask to add the following
+line to drivers/target/Kbuild:
+
+ obj-$(CONFIG_TCM_NAB5000) += tcm_nab5000/
+
+and the same for drivers/target/Kconfig:
+
+ source "drivers/target/tcm_nab5000/Kconfig"
+
+*) Run 'make menuconfig' and select the new CONFIG_TCM_NAB5000 item:
+
+ <M> TCM_NAB5000 fabric module
+
+*) Build using 'make modules', once completed you will have:
+
+target:/mnt/sdb/lio-core-2.6.git# ls -la drivers/target/tcm_nab5000/
+total 1348
+drwxr-xr-x 2 root root 4096 2010-10-05 03:23 .
+drwxr-xr-x 9 root root 4096 2010-10-05 03:22 ..
+-rw-r--r-- 1 root root 282 2010-10-05 03:22 Kbuild
+-rw-r--r-- 1 root root 171 2010-10-05 03:22 Kconfig
+-rw-r--r-- 1 root root 49 2010-10-05 03:23 modules.order
+-rw-r--r-- 1 root root 738 2010-10-05 03:22 tcm_nab5000_base.h
+-rw-r--r-- 1 root root 9096 2010-10-05 03:22 tcm_nab5000_configfs.c
+-rw-r--r-- 1 root root 191200 2010-10-05 03:23 tcm_nab5000_configfs.o
+-rw-r--r-- 1 root root 40504 2010-10-05 03:23 .tcm_nab5000_configfs.o.cmd
+-rw-r--r-- 1 root root 5414 2010-10-05 03:22 tcm_nab5000_fabric.c
+-rw-r--r-- 1 root root 2016 2010-10-05 03:22 tcm_nab5000_fabric.h
+-rw-r--r-- 1 root root 190932 2010-10-05 03:23 tcm_nab5000_fabric.o
+-rw-r--r-- 1 root root 40713 2010-10-05 03:23 .tcm_nab5000_fabric.o.cmd
+-rw-r--r-- 1 root root 401861 2010-10-05 03:23 tcm_nab5000.ko
+-rw-r--r-- 1 root root 265 2010-10-05 03:23 .tcm_nab5000.ko.cmd
+-rw-r--r-- 1 root root 459 2010-10-05 03:23 tcm_nab5000.mod.c
+-rw-r--r-- 1 root root 23896 2010-10-05 03:23 tcm_nab5000.mod.o
+-rw-r--r-- 1 root root 22655 2010-10-05 03:23 .tcm_nab5000.mod.o.cmd
+-rw-r--r-- 1 root root 379022 2010-10-05 03:23 tcm_nab5000.o
+-rw-r--r-- 1 root root 211 2010-10-05 03:23 .tcm_nab5000.o.cmd
+
+*) Load the new module, create a lun_0 configfs group, and add new TCM Core
+ IBLOCK backstore symlink to port:
+
+target:/mnt/sdb/lio-core-2.6.git# insmod drivers/target/tcm_nab5000.ko
+target:/mnt/sdb/lio-core-2.6.git# mkdir -p /sys/kernel/config/target/nab5000/iqn.foo/tpgt_1/lun/lun_0
+target:/mnt/sdb/lio-core-2.6.git# cd /sys/kernel/config/target/nab5000/iqn.foo/tpgt_1/lun/lun_0/
+target:/sys/kernel/config/target/nab5000/iqn.foo/tpgt_1/lun/lun_0# ln -s /sys/kernel/config/target/core/iblock_0/lvm_test0 nab5000_port
+
+target:/sys/kernel/config/target/nab5000/iqn.foo/tpgt_1/lun/lun_0# cd -
+target:/mnt/sdb/lio-core-2.6.git# tree /sys/kernel/config/target/nab5000/
+/sys/kernel/config/target/nab5000/
+|-- discovery_auth
+|-- iqn.foo
+| `-- tpgt_1
+| |-- acls
+| |-- attrib
+| |-- lun
+| | `-- lun_0
+| | |-- alua_tg_pt_gp
+| | |-- alua_tg_pt_offline
+| | |-- alua_tg_pt_status
+| | |-- alua_tg_pt_write_md
+| | `-- nab5000_port -> ../../../../../../target/core/iblock_0/lvm_test0
+| |-- np
+| `-- param
+`-- version
+
+target:/mnt/sdb/lio-core-2.6.git# lsmod
+Module Size Used by
+tcm_nab5000 3935 4
+iscsi_target_mod 193211 0
+target_core_stgt 8090 0
+target_core_pscsi 11122 1
+target_core_file 9172 2
+target_core_iblock 9280 1
+target_core_mod 228575 31
+tcm_nab5000,iscsi_target_mod,target_core_stgt,target_core_pscsi,target_core_file,target_core_iblock
+libfc 73681 0
+scsi_debug 56265 0
+scsi_tgt 8666 1 target_core_stgt
+configfs 20644 2 target_core_mod
+
+----------------------------------------------------------------------
+
+Future TODO items:
+
+ *) Add more T10 proto_idents
+ *) Make tcm_mod_dump_fabric_ops() smarter and generate function pointer
+ defs directly from include/target/target_core_fabric_ops.h:struct target_core_fabric_ops
+ structure members.
+
+October 5th, 2010
+Nicholas A. Bellinger <nab@linux-iscsi.org>
diff --git a/Documentation/thermal/sysfs-api.txt b/Documentation/thermal/sysfs-api.txt
index cb3d15b..b61e46f 100644
--- a/Documentation/thermal/sysfs-api.txt
+++ b/Documentation/thermal/sysfs-api.txt
@@ -278,3 +278,15 @@
|---name: acpitz
|---temp1_input: 37000
|---temp1_crit: 100000
+
+4. Event Notification
+
+The framework includes a simple notification mechanism, in the form of a
+netlink event. Netlink socket initialization is done during the _init_
+of the framework. Drivers which intend to use the notification mechanism
+just need to call generate_netlink_event() with two arguments viz
+(originator, event). Typically the originator will be an integer assigned
+to a thermal_zone_device when it registers itself with the framework. The
+event will be one of:{THERMAL_AUX0, THERMAL_AUX1, THERMAL_CRITICAL,
+THERMAL_DEV_FAULT}. Notification can be sent when the current temperature
+crosses any of the configured thresholds.
diff --git a/Documentation/vm/transhuge.txt b/Documentation/vm/transhuge.txt
new file mode 100644
index 0000000..0924aac
--- /dev/null
+++ b/Documentation/vm/transhuge.txt
@@ -0,0 +1,298 @@
+= Transparent Hugepage Support =
+
+== Objective ==
+
+Performance critical computing applications dealing with large memory
+working sets are already running on top of libhugetlbfs and in turn
+hugetlbfs. Transparent Hugepage Support is an alternative means of
+using huge pages for the backing of virtual memory with huge pages
+that supports the automatic promotion and demotion of page sizes and
+without the shortcomings of hugetlbfs.
+
+Currently it only works for anonymous memory mappings but in the
+future it can expand over the pagecache layer starting with tmpfs.
+
+The reason applications are running faster is because of two
+factors. The first factor is almost completely irrelevant and it's not
+of significant interest because it'll also have the downside of
+requiring larger clear-page copy-page in page faults which is a
+potentially negative effect. The first factor consists in taking a
+single page fault for each 2M virtual region touched by userland (so
+reducing the enter/exit kernel frequency by a 512 times factor). This
+only matters the first time the memory is accessed for the lifetime of
+a memory mapping. The second long lasting and much more important
+factor will affect all subsequent accesses to the memory for the whole
+runtime of the application. The second factor consist of two
+components: 1) the TLB miss will run faster (especially with
+virtualization using nested pagetables but almost always also on bare
+metal without virtualization) and 2) a single TLB entry will be
+mapping a much larger amount of virtual memory in turn reducing the
+number of TLB misses. With virtualization and nested pagetables the
+TLB can be mapped of larger size only if both KVM and the Linux guest
+are using hugepages but a significant speedup already happens if only
+one of the two is using hugepages just because of the fact the TLB
+miss is going to run faster.
+
+== Design ==
+
+- "graceful fallback": mm components which don't have transparent
+ hugepage knowledge fall back to breaking a transparent hugepage and
+ working on the regular pages and their respective regular pmd/pte
+ mappings
+
+- if a hugepage allocation fails because of memory fragmentation,
+ regular pages should be gracefully allocated instead and mixed in
+ the same vma without any failure or significant delay and without
+ userland noticing
+
+- if some task quits and more hugepages become available (either
+ immediately in the buddy or through the VM), guest physical memory
+ backed by regular pages should be relocated on hugepages
+ automatically (with khugepaged)
+
+- it doesn't require memory reservation and in turn it uses hugepages
+ whenever possible (the only possible reservation here is kernelcore=
+ to avoid unmovable pages to fragment all the memory but such a tweak
+ is not specific to transparent hugepage support and it's a generic
+ feature that applies to all dynamic high order allocations in the
+ kernel)
+
+- this initial support only offers the feature in the anonymous memory
+ regions but it'd be ideal to move it to tmpfs and the pagecache
+ later
+
+Transparent Hugepage Support maximizes the usefulness of free memory
+if compared to the reservation approach of hugetlbfs by allowing all
+unused memory to be used as cache or other movable (or even unmovable
+entities). It doesn't require reservation to prevent hugepage
+allocation failures to be noticeable from userland. It allows paging
+and all other advanced VM features to be available on the
+hugepages. It requires no modifications for applications to take
+advantage of it.
+
+Applications however can be further optimized to take advantage of
+this feature, like for example they've been optimized before to avoid
+a flood of mmap system calls for every malloc(4k). Optimizing userland
+is by far not mandatory and khugepaged already can take care of long
+lived page allocations even for hugepage unaware applications that
+deals with large amounts of memory.
+
+In certain cases when hugepages are enabled system wide, application
+may end up allocating more memory resources. An application may mmap a
+large region but only touch 1 byte of it, in that case a 2M page might
+be allocated instead of a 4k page for no good. This is why it's
+possible to disable hugepages system-wide and to only have them inside
+MADV_HUGEPAGE madvise regions.
+
+Embedded systems should enable hugepages only inside madvise regions
+to eliminate any risk of wasting any precious byte of memory and to
+only run faster.
+
+Applications that gets a lot of benefit from hugepages and that don't
+risk to lose memory by using hugepages, should use
+madvise(MADV_HUGEPAGE) on their critical mmapped regions.
+
+== sysfs ==
+
+Transparent Hugepage Support can be entirely disabled (mostly for
+debugging purposes) or only enabled inside MADV_HUGEPAGE regions (to
+avoid the risk of consuming more memory resources) or enabled system
+wide. This can be achieved with one of:
+
+echo always >/sys/kernel/mm/transparent_hugepage/enabled
+echo madvise >/sys/kernel/mm/transparent_hugepage/enabled
+echo never >/sys/kernel/mm/transparent_hugepage/enabled
+
+It's also possible to limit defrag efforts in the VM to generate
+hugepages in case they're not immediately free to madvise regions or
+to never try to defrag memory and simply fallback to regular pages
+unless hugepages are immediately available. Clearly if we spend CPU
+time to defrag memory, we would expect to gain even more by the fact
+we use hugepages later instead of regular pages. This isn't always
+guaranteed, but it may be more likely in case the allocation is for a
+MADV_HUGEPAGE region.
+
+echo always >/sys/kernel/mm/transparent_hugepage/defrag
+echo madvise >/sys/kernel/mm/transparent_hugepage/defrag
+echo never >/sys/kernel/mm/transparent_hugepage/defrag
+
+khugepaged will be automatically started when
+transparent_hugepage/enabled is set to "always" or "madvise, and it'll
+be automatically shutdown if it's set to "never".
+
+khugepaged runs usually at low frequency so while one may not want to
+invoke defrag algorithms synchronously during the page faults, it
+should be worth invoking defrag at least in khugepaged. However it's
+also possible to disable defrag in khugepaged:
+
+echo yes >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag
+echo no >/sys/kernel/mm/transparent_hugepage/khugepaged/defrag
+
+You can also control how many pages khugepaged should scan at each
+pass:
+
+/sys/kernel/mm/transparent_hugepage/khugepaged/pages_to_scan
+
+and how many milliseconds to wait in khugepaged between each pass (you
+can set this to 0 to run khugepaged at 100% utilization of one core):
+
+/sys/kernel/mm/transparent_hugepage/khugepaged/scan_sleep_millisecs
+
+and how many milliseconds to wait in khugepaged if there's an hugepage
+allocation failure to throttle the next allocation attempt.
+
+/sys/kernel/mm/transparent_hugepage/khugepaged/alloc_sleep_millisecs
+
+The khugepaged progress can be seen in the number of pages collapsed:
+
+/sys/kernel/mm/transparent_hugepage/khugepaged/pages_collapsed
+
+for each pass:
+
+/sys/kernel/mm/transparent_hugepage/khugepaged/full_scans
+
+== Boot parameter ==
+
+You can change the sysfs boot time defaults of Transparent Hugepage
+Support by passing the parameter "transparent_hugepage=always" or
+"transparent_hugepage=madvise" or "transparent_hugepage=never"
+(without "") to the kernel command line.
+
+== Need of application restart ==
+
+The transparent_hugepage/enabled values only affect future
+behavior. So to make them effective you need to restart any
+application that could have been using hugepages. This also applies to
+the regions registered in khugepaged.
+
+== get_user_pages and follow_page ==
+
+get_user_pages and follow_page if run on a hugepage, will return the
+head or tail pages as usual (exactly as they would do on
+hugetlbfs). Most gup users will only care about the actual physical
+address of the page and its temporary pinning to release after the I/O
+is complete, so they won't ever notice the fact the page is huge. But
+if any driver is going to mangle over the page structure of the tail
+page (like for checking page->mapping or other bits that are relevant
+for the head page and not the tail page), it should be updated to jump
+to check head page instead (while serializing properly against
+split_huge_page() to avoid the head and tail pages to disappear from
+under it, see the futex code to see an example of that, hugetlbfs also
+needed special handling in futex code for similar reasons).
+
+NOTE: these aren't new constraints to the GUP API, and they match the
+same constrains that applies to hugetlbfs too, so any driver capable
+of handling GUP on hugetlbfs will also work fine on transparent
+hugepage backed mappings.
+
+In case you can't handle compound pages if they're returned by
+follow_page, the FOLL_SPLIT bit can be specified as parameter to
+follow_page, so that it will split the hugepages before returning
+them. Migration for example passes FOLL_SPLIT as parameter to
+follow_page because it's not hugepage aware and in fact it can't work
+at all on hugetlbfs (but it instead works fine on transparent
+hugepages thanks to FOLL_SPLIT). migration simply can't deal with
+hugepages being returned (as it's not only checking the pfn of the
+page and pinning it during the copy but it pretends to migrate the
+memory in regular page sizes and with regular pte/pmd mappings).
+
+== Optimizing the applications ==
+
+To be guaranteed that the kernel will map a 2M page immediately in any
+memory region, the mmap region has to be hugepage naturally
+aligned. posix_memalign() can provide that guarantee.
+
+== Hugetlbfs ==
+
+You can use hugetlbfs on a kernel that has transparent hugepage
+support enabled just fine as always. No difference can be noted in
+hugetlbfs other than there will be less overall fragmentation. All
+usual features belonging to hugetlbfs are preserved and
+unaffected. libhugetlbfs will also work fine as usual.
+
+== Graceful fallback ==
+
+Code walking pagetables but unware about huge pmds can simply call
+split_huge_page_pmd(mm, pmd) where the pmd is the one returned by
+pmd_offset. It's trivial to make the code transparent hugepage aware
+by just grepping for "pmd_offset" and adding split_huge_page_pmd where
+missing after pmd_offset returns the pmd. Thanks to the graceful
+fallback design, with a one liner change, you can avoid to write
+hundred if not thousand of lines of complex code to make your code
+hugepage aware.
+
+If you're not walking pagetables but you run into a physical hugepage
+but you can't handle it natively in your code, you can split it by
+calling split_huge_page(page). This is what the Linux VM does before
+it tries to swapout the hugepage for example.
+
+Example to make mremap.c transparent hugepage aware with a one liner
+change:
+
+diff --git a/mm/mremap.c b/mm/mremap.c
+--- a/mm/mremap.c
++++ b/mm/mremap.c
+@@ -41,6 +41,7 @@ static pmd_t *get_old_pmd(struct mm_stru
+ return NULL;
+
+ pmd = pmd_offset(pud, addr);
++ split_huge_page_pmd(mm, pmd);
+ if (pmd_none_or_clear_bad(pmd))
+ return NULL;
+
+== Locking in hugepage aware code ==
+
+We want as much code as possible hugepage aware, as calling
+split_huge_page() or split_huge_page_pmd() has a cost.
+
+To make pagetable walks huge pmd aware, all you need to do is to call
+pmd_trans_huge() on the pmd returned by pmd_offset. You must hold the
+mmap_sem in read (or write) mode to be sure an huge pmd cannot be
+created from under you by khugepaged (khugepaged collapse_huge_page
+takes the mmap_sem in write mode in addition to the anon_vma lock). If
+pmd_trans_huge returns false, you just fallback in the old code
+paths. If instead pmd_trans_huge returns true, you have to take the
+mm->page_table_lock and re-run pmd_trans_huge. Taking the
+page_table_lock will prevent the huge pmd to be converted into a
+regular pmd from under you (split_huge_page can run in parallel to the
+pagetable walk). If the second pmd_trans_huge returns false, you
+should just drop the page_table_lock and fallback to the old code as
+before. Otherwise you should run pmd_trans_splitting on the pmd. In
+case pmd_trans_splitting returns true, it means split_huge_page is
+already in the middle of splitting the page. So if pmd_trans_splitting
+returns true it's enough to drop the page_table_lock and call
+wait_split_huge_page and then fallback the old code paths. You are
+guaranteed by the time wait_split_huge_page returns, the pmd isn't
+huge anymore. If pmd_trans_splitting returns false, you can proceed to
+process the huge pmd and the hugepage natively. Once finished you can
+drop the page_table_lock.
+
+== compound_lock, get_user_pages and put_page ==
+
+split_huge_page internally has to distribute the refcounts in the head
+page to the tail pages before clearing all PG_head/tail bits from the
+page structures. It can do that easily for refcounts taken by huge pmd
+mappings. But the GUI API as created by hugetlbfs (that returns head
+and tail pages if running get_user_pages on an address backed by any
+hugepage), requires the refcount to be accounted on the tail pages and
+not only in the head pages, if we want to be able to run
+split_huge_page while there are gup pins established on any tail
+page. Failure to be able to run split_huge_page if there's any gup pin
+on any tail page, would mean having to split all hugepages upfront in
+get_user_pages which is unacceptable as too many gup users are
+performance critical and they must work natively on hugepages like
+they work natively on hugetlbfs already (hugetlbfs is simpler because
+hugetlbfs pages cannot be splitted so there wouldn't be requirement of
+accounting the pins on the tail pages for hugetlbfs). If we wouldn't
+account the gup refcounts on the tail pages during gup, we won't know
+anymore which tail page is pinned by gup and which is not while we run
+split_huge_page. But we still have to add the gup pin to the head page
+too, to know when we can free the compound page in case it's never
+splitted during its lifetime. That requires changing not just
+get_page, but put_page as well so that when put_page runs on a tail
+page (and only on a tail page) it will find its respective head page,
+and then it will decrease the head page refcount in addition to the
+tail page refcount. To obtain a head page reliably and to decrease its
+refcount without race conditions, put_page has to serialize against
+__split_huge_page_refcount using a special per-page lock called
+compound_lock.
diff --git a/MAINTAINERS b/MAINTAINERS
index 3dd5c6f..1af022e 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3684,7 +3684,7 @@
KMEMCHECK
M: Vegard Nossum <vegardno@ifi.uio.no>
-M: Pekka Enberg <penberg@cs.helsinki.fi>
+M: Pekka Enberg <penberg@kernel.org>
S: Maintained
F: Documentation/kmemcheck.txt
F: arch/x86/include/asm/kmemcheck.h
@@ -5272,8 +5272,7 @@
F: drivers/s390/net/
S390 ZCRYPT DRIVER
-M: Felix Beck <felix.beck@de.ibm.com>
-M: Ralph Wuerthner <ralph.wuerthner@de.ibm.com>
+M: Holger Dengler <hd@linux.vnet.ibm.com>
M: linux390@de.ibm.com
L: linux-s390@vger.kernel.org
W: http://www.ibm.com/developerworks/linux/linux390/
@@ -5647,7 +5646,7 @@
SLAB ALLOCATOR
M: Christoph Lameter <cl@linux-foundation.org>
-M: Pekka Enberg <penberg@cs.helsinki.fi>
+M: Pekka Enberg <penberg@kernel.org>
M: Matt Mackall <mpm@selenic.com>
L: linux-mm@kvack.org
S: Maintained
@@ -6592,13 +6591,12 @@
F: drivers/i2c/busses/i2c-viapro.c
VIA SD/MMC CARD CONTROLLER DRIVER
-M: Joseph Chan <JosephChan@via.com.tw>
+M: Bruce Chang <brucechang@via.com.tw>
M: Harald Welte <HaraldWelte@viatech.com>
S: Maintained
F: drivers/mmc/host/via-sdmmc.c
VIA UNICHROME(PRO)/CHROME9 FRAMEBUFFER DRIVER
-M: Joseph Chan <JosephChan@via.com.tw>
M: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
L: linux-fbdev@vger.kernel.org
S: Maintained
diff --git a/arch/alpha/include/asm/mman.h b/arch/alpha/include/asm/mman.h
index 99c56d4..72db984 100644
--- a/arch/alpha/include/asm/mman.h
+++ b/arch/alpha/include/asm/mman.h
@@ -53,6 +53,9 @@
#define MADV_MERGEABLE 12 /* KSM may merge identical pages */
#define MADV_UNMERGEABLE 13 /* KSM may not merge identical pages */
+#define MADV_HUGEPAGE 14 /* Worth backing with hugepages */
+#define MADV_NOHUGEPAGE 15 /* Not worth backing with hugepages */
+
/* compatibility flags */
#define MAP_FILE 0
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index e2f8011..5cff165 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -26,6 +26,8 @@
select HAVE_REGS_AND_STACK_ACCESS_API
select HAVE_HW_BREAKPOINT if (PERF_EVENTS && (CPU_V6 || CPU_V7))
select HAVE_C_RECORDMCOUNT
+ select HAVE_GENERIC_HARDIRQS
+ select HAVE_SPARSE_IRQ
help
The ARM series is a line of low-power-consumption RISC chip designs
licensed by ARM Ltd and targeted at embedded applications and
@@ -97,10 +99,6 @@
<file:Documentation/mca.txt> (and especially the web page given
there) before attempting to build an MCA bus kernel.
-config GENERIC_HARDIRQS
- bool
- default y
-
config STACKTRACE_SUPPORT
bool
default y
@@ -180,9 +178,6 @@
config ARCH_MTD_XIP
bool
-config GENERIC_HARDIRQS_NO__DO_IRQ
- def_bool y
-
config ARM_L1_CACHE_SHIFT_6
bool
help
@@ -368,7 +363,7 @@
bool "Freescale MXS-based"
select GENERIC_CLOCKEVENTS
select ARCH_REQUIRE_GPIOLIB
- select COMMON_CLKDEV
+ select CLKDEV_LOOKUP
help
Support for Freescale MXS-based family of processors
@@ -771,6 +766,7 @@
select ARCH_SPARSEMEM_ENABLE
select GENERIC_GPIO
select HAVE_CLK
+ select ARCH_HAS_CPUFREQ
select GENERIC_CLOCKEVENTS
select HAVE_S3C_RTC if RTC_CLASS
select HAVE_S3C2410_I2C if I2C
@@ -1281,7 +1277,7 @@
config SMP_ON_UP
bool "Allow booting SMP kernel on uniprocessor systems (EXPERIMENTAL)"
depends on EXPERIMENTAL
- depends on SMP && !XIP
+ depends on SMP && !XIP_KERNEL
default y
help
SMP kernels contain instructions which fail on non-SMP processors.
@@ -1452,15 +1448,6 @@
Enable hardware performance counter support for perf events. If
disabled, perf events will use software events only.
-config SPARSE_IRQ
- def_bool n
- help
- This enables support for sparse irqs. This is useful in general
- as most CPUs have a fairly sparse array of IRQ vectors, which
- the irq_desc then maps directly on to. Systems with a high
- number of off-chip IRQs will want to treat this as
- experimental until they have been independently verified.
-
source "mm/Kconfig"
config FORCE_MAX_ZONEORDER
diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index 0b89ef0..2243772 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -50,57 +50,56 @@
static struct gic_chip_data gic_data[MAX_GIC_NR] __read_mostly;
-static inline void __iomem *gic_dist_base(unsigned int irq)
+static inline void __iomem *gic_dist_base(struct irq_data *d)
{
- struct gic_chip_data *gic_data = get_irq_chip_data(irq);
+ struct gic_chip_data *gic_data = irq_data_get_irq_chip_data(d);
return gic_data->dist_base;
}
-static inline void __iomem *gic_cpu_base(unsigned int irq)
+static inline void __iomem *gic_cpu_base(struct irq_data *d)
{
- struct gic_chip_data *gic_data = get_irq_chip_data(irq);
+ struct gic_chip_data *gic_data = irq_data_get_irq_chip_data(d);
return gic_data->cpu_base;
}
-static inline unsigned int gic_irq(unsigned int irq)
+static inline unsigned int gic_irq(struct irq_data *d)
{
- struct gic_chip_data *gic_data = get_irq_chip_data(irq);
- return irq - gic_data->irq_offset;
+ struct gic_chip_data *gic_data = irq_data_get_irq_chip_data(d);
+ return d->irq - gic_data->irq_offset;
}
/*
* Routines to acknowledge, disable and enable interrupts
*/
-static void gic_ack_irq(unsigned int irq)
+static void gic_ack_irq(struct irq_data *d)
{
-
spin_lock(&irq_controller_lock);
- writel(gic_irq(irq), gic_cpu_base(irq) + GIC_CPU_EOI);
+ writel(gic_irq(d), gic_cpu_base(d) + GIC_CPU_EOI);
spin_unlock(&irq_controller_lock);
}
-static void gic_mask_irq(unsigned int irq)
+static void gic_mask_irq(struct irq_data *d)
{
- u32 mask = 1 << (irq % 32);
+ u32 mask = 1 << (d->irq % 32);
spin_lock(&irq_controller_lock);
- writel(mask, gic_dist_base(irq) + GIC_DIST_ENABLE_CLEAR + (gic_irq(irq) / 32) * 4);
+ writel(mask, gic_dist_base(d) + GIC_DIST_ENABLE_CLEAR + (gic_irq(d) / 32) * 4);
spin_unlock(&irq_controller_lock);
}
-static void gic_unmask_irq(unsigned int irq)
+static void gic_unmask_irq(struct irq_data *d)
{
- u32 mask = 1 << (irq % 32);
+ u32 mask = 1 << (d->irq % 32);
spin_lock(&irq_controller_lock);
- writel(mask, gic_dist_base(irq) + GIC_DIST_ENABLE_SET + (gic_irq(irq) / 32) * 4);
+ writel(mask, gic_dist_base(d) + GIC_DIST_ENABLE_SET + (gic_irq(d) / 32) * 4);
spin_unlock(&irq_controller_lock);
}
-static int gic_set_type(unsigned int irq, unsigned int type)
+static int gic_set_type(struct irq_data *d, unsigned int type)
{
- void __iomem *base = gic_dist_base(irq);
- unsigned int gicirq = gic_irq(irq);
+ void __iomem *base = gic_dist_base(d);
+ unsigned int gicirq = gic_irq(d);
u32 enablemask = 1 << (gicirq % 32);
u32 enableoff = (gicirq / 32) * 4;
u32 confmask = 0x2 << ((gicirq % 16) * 2);
@@ -143,21 +142,22 @@
}
#ifdef CONFIG_SMP
-static int gic_set_cpu(unsigned int irq, const struct cpumask *mask_val)
+static int
+gic_set_cpu(struct irq_data *d, const struct cpumask *mask_val, bool force)
{
- void __iomem *reg = gic_dist_base(irq) + GIC_DIST_TARGET + (gic_irq(irq) & ~3);
- unsigned int shift = (irq % 4) * 8;
+ void __iomem *reg = gic_dist_base(d) + GIC_DIST_TARGET + (gic_irq(d) & ~3);
+ unsigned int shift = (d->irq % 4) * 8;
unsigned int cpu = cpumask_first(mask_val);
u32 val;
struct irq_desc *desc;
spin_lock(&irq_controller_lock);
- desc = irq_to_desc(irq);
+ desc = irq_to_desc(d->irq);
if (desc == NULL) {
spin_unlock(&irq_controller_lock);
return -EINVAL;
}
- desc->node = cpu;
+ d->node = cpu;
val = readl(reg) & ~(0xff << shift);
val |= 1 << (cpu + shift);
writel(val, reg);
@@ -175,7 +175,7 @@
unsigned long status;
/* primary controller ack'ing */
- chip->ack(irq);
+ chip->irq_ack(&desc->irq_data);
spin_lock(&irq_controller_lock);
status = readl(chip_data->cpu_base + GIC_CPU_INTACK);
@@ -193,17 +193,17 @@
out:
/* primary controller unmasking */
- chip->unmask(irq);
+ chip->irq_unmask(&desc->irq_data);
}
static struct irq_chip gic_chip = {
- .name = "GIC",
- .ack = gic_ack_irq,
- .mask = gic_mask_irq,
- .unmask = gic_unmask_irq,
- .set_type = gic_set_type,
+ .name = "GIC",
+ .irq_ack = gic_ack_irq,
+ .irq_mask = gic_mask_irq,
+ .irq_unmask = gic_unmask_irq,
+ .irq_set_type = gic_set_type,
#ifdef CONFIG_SMP
- .set_affinity = gic_set_cpu,
+ .irq_set_affinity = gic_set_cpu,
#endif
};
@@ -337,7 +337,7 @@
local_irq_save(flags);
irq_to_desc(irq)->status |= IRQ_NOPROBE;
- gic_unmask_irq(irq);
+ gic_unmask_irq(irq_get_irq_data(irq));
local_irq_restore(flags);
}
diff --git a/arch/arm/common/it8152.c b/arch/arm/common/it8152.c
index 665ebf7..fcddd48 100644
--- a/arch/arm/common/it8152.c
+++ b/arch/arm/common/it8152.c
@@ -31,8 +31,10 @@
#define MAX_SLOTS 21
-static void it8152_mask_irq(unsigned int irq)
+static void it8152_mask_irq(struct irq_data *d)
{
+ unsigned int irq = d->irq;
+
if (irq >= IT8152_LD_IRQ(0)) {
__raw_writel((__raw_readl(IT8152_INTC_LDCNIMR) |
(1 << (irq - IT8152_LD_IRQ(0)))),
@@ -48,8 +50,10 @@
}
}
-static void it8152_unmask_irq(unsigned int irq)
+static void it8152_unmask_irq(struct irq_data *d)
{
+ unsigned int irq = d->irq;
+
if (irq >= IT8152_LD_IRQ(0)) {
__raw_writel((__raw_readl(IT8152_INTC_LDCNIMR) &
~(1 << (irq - IT8152_LD_IRQ(0)))),
@@ -67,9 +71,9 @@
static struct irq_chip it8152_irq_chip = {
.name = "it8152",
- .ack = it8152_mask_irq,
- .mask = it8152_mask_irq,
- .unmask = it8152_unmask_irq,
+ .irq_ack = it8152_mask_irq,
+ .irq_mask = it8152_mask_irq,
+ .irq_unmask = it8152_unmask_irq,
};
void it8152_init_irq(void)
diff --git a/arch/arm/common/locomo.c b/arch/arm/common/locomo.c
index 9dff07c..a026a6b 100644
--- a/arch/arm/common/locomo.c
+++ b/arch/arm/common/locomo.c
@@ -144,7 +144,7 @@
int req, i;
/* Acknowledge the parent IRQ */
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
/* check why this interrupt was generated */
req = locomo_readl(lchip->base + LOCOMO_ICR) & 0x0f00;
@@ -161,33 +161,33 @@
}
}
-static void locomo_ack_irq(unsigned int irq)
+static void locomo_ack_irq(struct irq_data *d)
{
}
-static void locomo_mask_irq(unsigned int irq)
+static void locomo_mask_irq(struct irq_data *d)
{
- struct locomo *lchip = get_irq_chip_data(irq);
+ struct locomo *lchip = irq_data_get_irq_chip_data(d);
unsigned int r;
r = locomo_readl(lchip->base + LOCOMO_ICR);
- r &= ~(0x0010 << (irq - lchip->irq_base));
+ r &= ~(0x0010 << (d->irq - lchip->irq_base));
locomo_writel(r, lchip->base + LOCOMO_ICR);
}
-static void locomo_unmask_irq(unsigned int irq)
+static void locomo_unmask_irq(struct irq_data *d)
{
- struct locomo *lchip = get_irq_chip_data(irq);
+ struct locomo *lchip = irq_data_get_irq_chip_data(d);
unsigned int r;
r = locomo_readl(lchip->base + LOCOMO_ICR);
- r |= (0x0010 << (irq - lchip->irq_base));
+ r |= (0x0010 << (d->irq - lchip->irq_base));
locomo_writel(r, lchip->base + LOCOMO_ICR);
}
static struct irq_chip locomo_chip = {
- .name = "LOCOMO",
- .ack = locomo_ack_irq,
- .mask = locomo_mask_irq,
- .unmask = locomo_unmask_irq,
+ .name = "LOCOMO",
+ .irq_ack = locomo_ack_irq,
+ .irq_mask = locomo_mask_irq,
+ .irq_unmask = locomo_unmask_irq,
};
static void locomo_setup_irq(struct locomo *lchip)
diff --git a/arch/arm/common/sa1111.c b/arch/arm/common/sa1111.c
index c0258a8..eb9796b 100644
--- a/arch/arm/common/sa1111.c
+++ b/arch/arm/common/sa1111.c
@@ -210,7 +210,7 @@
sa1111_writel(stat0, mapbase + SA1111_INTSTATCLR0);
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
sa1111_writel(stat1, mapbase + SA1111_INTSTATCLR1);
@@ -228,35 +228,35 @@
generic_handle_irq(i + sachip->irq_base);
/* For level-based interrupts */
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
#define SA1111_IRQMASK_LO(x) (1 << (x - sachip->irq_base))
#define SA1111_IRQMASK_HI(x) (1 << (x - sachip->irq_base - 32))
-static void sa1111_ack_irq(unsigned int irq)
+static void sa1111_ack_irq(struct irq_data *d)
{
}
-static void sa1111_mask_lowirq(unsigned int irq)
+static void sa1111_mask_lowirq(struct irq_data *d)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
unsigned long ie0;
ie0 = sa1111_readl(mapbase + SA1111_INTEN0);
- ie0 &= ~SA1111_IRQMASK_LO(irq);
+ ie0 &= ~SA1111_IRQMASK_LO(d->irq);
writel(ie0, mapbase + SA1111_INTEN0);
}
-static void sa1111_unmask_lowirq(unsigned int irq)
+static void sa1111_unmask_lowirq(struct irq_data *d)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
unsigned long ie0;
ie0 = sa1111_readl(mapbase + SA1111_INTEN0);
- ie0 |= SA1111_IRQMASK_LO(irq);
+ ie0 |= SA1111_IRQMASK_LO(d->irq);
sa1111_writel(ie0, mapbase + SA1111_INTEN0);
}
@@ -267,11 +267,11 @@
* be triggered. In fact, its very difficult, if not impossible to get
* INTSET to re-trigger the interrupt.
*/
-static int sa1111_retrigger_lowirq(unsigned int irq)
+static int sa1111_retrigger_lowirq(struct irq_data *d)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
- unsigned int mask = SA1111_IRQMASK_LO(irq);
+ unsigned int mask = SA1111_IRQMASK_LO(d->irq);
unsigned long ip0;
int i;
@@ -279,21 +279,21 @@
for (i = 0; i < 8; i++) {
sa1111_writel(ip0 ^ mask, mapbase + SA1111_INTPOL0);
sa1111_writel(ip0, mapbase + SA1111_INTPOL0);
- if (sa1111_readl(mapbase + SA1111_INTSTATCLR1) & mask)
+ if (sa1111_readl(mapbase + SA1111_INTSTATCLR0) & mask)
break;
}
if (i == 8)
printk(KERN_ERR "Danger Will Robinson: failed to "
- "re-trigger IRQ%d\n", irq);
+ "re-trigger IRQ%d\n", d->irq);
return i == 8 ? -1 : 0;
}
-static int sa1111_type_lowirq(unsigned int irq, unsigned int flags)
+static int sa1111_type_lowirq(struct irq_data *d, unsigned int flags)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
- unsigned int mask = SA1111_IRQMASK_LO(irq);
+ unsigned int mask = SA1111_IRQMASK_LO(d->irq);
unsigned long ip0;
if (flags == IRQ_TYPE_PROBE)
@@ -313,11 +313,11 @@
return 0;
}
-static int sa1111_wake_lowirq(unsigned int irq, unsigned int on)
+static int sa1111_wake_lowirq(struct irq_data *d, unsigned int on)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
- unsigned int mask = SA1111_IRQMASK_LO(irq);
+ unsigned int mask = SA1111_IRQMASK_LO(d->irq);
unsigned long we0;
we0 = sa1111_readl(mapbase + SA1111_WAKEEN0);
@@ -332,33 +332,33 @@
static struct irq_chip sa1111_low_chip = {
.name = "SA1111-l",
- .ack = sa1111_ack_irq,
- .mask = sa1111_mask_lowirq,
- .unmask = sa1111_unmask_lowirq,
- .retrigger = sa1111_retrigger_lowirq,
- .set_type = sa1111_type_lowirq,
- .set_wake = sa1111_wake_lowirq,
+ .irq_ack = sa1111_ack_irq,
+ .irq_mask = sa1111_mask_lowirq,
+ .irq_unmask = sa1111_unmask_lowirq,
+ .irq_retrigger = sa1111_retrigger_lowirq,
+ .irq_set_type = sa1111_type_lowirq,
+ .irq_set_wake = sa1111_wake_lowirq,
};
-static void sa1111_mask_highirq(unsigned int irq)
+static void sa1111_mask_highirq(struct irq_data *d)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
unsigned long ie1;
ie1 = sa1111_readl(mapbase + SA1111_INTEN1);
- ie1 &= ~SA1111_IRQMASK_HI(irq);
+ ie1 &= ~SA1111_IRQMASK_HI(d->irq);
sa1111_writel(ie1, mapbase + SA1111_INTEN1);
}
-static void sa1111_unmask_highirq(unsigned int irq)
+static void sa1111_unmask_highirq(struct irq_data *d)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
unsigned long ie1;
ie1 = sa1111_readl(mapbase + SA1111_INTEN1);
- ie1 |= SA1111_IRQMASK_HI(irq);
+ ie1 |= SA1111_IRQMASK_HI(d->irq);
sa1111_writel(ie1, mapbase + SA1111_INTEN1);
}
@@ -369,11 +369,11 @@
* be triggered. In fact, its very difficult, if not impossible to get
* INTSET to re-trigger the interrupt.
*/
-static int sa1111_retrigger_highirq(unsigned int irq)
+static int sa1111_retrigger_highirq(struct irq_data *d)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
- unsigned int mask = SA1111_IRQMASK_HI(irq);
+ unsigned int mask = SA1111_IRQMASK_HI(d->irq);
unsigned long ip1;
int i;
@@ -387,15 +387,15 @@
if (i == 8)
printk(KERN_ERR "Danger Will Robinson: failed to "
- "re-trigger IRQ%d\n", irq);
+ "re-trigger IRQ%d\n", d->irq);
return i == 8 ? -1 : 0;
}
-static int sa1111_type_highirq(unsigned int irq, unsigned int flags)
+static int sa1111_type_highirq(struct irq_data *d, unsigned int flags)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
- unsigned int mask = SA1111_IRQMASK_HI(irq);
+ unsigned int mask = SA1111_IRQMASK_HI(d->irq);
unsigned long ip1;
if (flags == IRQ_TYPE_PROBE)
@@ -415,11 +415,11 @@
return 0;
}
-static int sa1111_wake_highirq(unsigned int irq, unsigned int on)
+static int sa1111_wake_highirq(struct irq_data *d, unsigned int on)
{
- struct sa1111 *sachip = get_irq_chip_data(irq);
+ struct sa1111 *sachip = irq_data_get_irq_chip_data(d);
void __iomem *mapbase = sachip->base + SA1111_INTC;
- unsigned int mask = SA1111_IRQMASK_HI(irq);
+ unsigned int mask = SA1111_IRQMASK_HI(d->irq);
unsigned long we1;
we1 = sa1111_readl(mapbase + SA1111_WAKEEN1);
@@ -434,12 +434,12 @@
static struct irq_chip sa1111_high_chip = {
.name = "SA1111-h",
- .ack = sa1111_ack_irq,
- .mask = sa1111_mask_highirq,
- .unmask = sa1111_unmask_highirq,
- .retrigger = sa1111_retrigger_highirq,
- .set_type = sa1111_type_highirq,
- .set_wake = sa1111_wake_highirq,
+ .irq_ack = sa1111_ack_irq,
+ .irq_mask = sa1111_mask_highirq,
+ .irq_unmask = sa1111_unmask_highirq,
+ .irq_retrigger = sa1111_retrigger_highirq,
+ .irq_set_type = sa1111_type_highirq,
+ .irq_set_wake = sa1111_wake_highirq,
};
static void sa1111_setup_irq(struct sa1111 *sachip)
diff --git a/arch/arm/common/vic.c b/arch/arm/common/vic.c
index cb660bc..ae5fe72 100644
--- a/arch/arm/common/vic.c
+++ b/arch/arm/common/vic.c
@@ -204,26 +204,26 @@
static inline void vic_pm_register(void __iomem *base, unsigned int irq, u32 arg1) { }
#endif /* CONFIG_PM */
-static void vic_ack_irq(unsigned int irq)
+static void vic_ack_irq(struct irq_data *d)
{
- void __iomem *base = get_irq_chip_data(irq);
- irq &= 31;
+ void __iomem *base = irq_data_get_irq_chip_data(d);
+ unsigned int irq = d->irq & 31;
writel(1 << irq, base + VIC_INT_ENABLE_CLEAR);
/* moreover, clear the soft-triggered, in case it was the reason */
writel(1 << irq, base + VIC_INT_SOFT_CLEAR);
}
-static void vic_mask_irq(unsigned int irq)
+static void vic_mask_irq(struct irq_data *d)
{
- void __iomem *base = get_irq_chip_data(irq);
- irq &= 31;
+ void __iomem *base = irq_data_get_irq_chip_data(d);
+ unsigned int irq = d->irq & 31;
writel(1 << irq, base + VIC_INT_ENABLE_CLEAR);
}
-static void vic_unmask_irq(unsigned int irq)
+static void vic_unmask_irq(struct irq_data *d)
{
- void __iomem *base = get_irq_chip_data(irq);
- irq &= 31;
+ void __iomem *base = irq_data_get_irq_chip_data(d);
+ unsigned int irq = d->irq & 31;
writel(1 << irq, base + VIC_INT_ENABLE);
}
@@ -242,10 +242,10 @@
return NULL;
}
-static int vic_set_wake(unsigned int irq, unsigned int on)
+static int vic_set_wake(struct irq_data *d, unsigned int on)
{
- struct vic_device *v = vic_from_irq(irq);
- unsigned int off = irq & 31;
+ struct vic_device *v = vic_from_irq(d->irq);
+ unsigned int off = d->irq & 31;
u32 bit = 1 << off;
if (!v)
@@ -267,10 +267,10 @@
static struct irq_chip vic_chip = {
.name = "VIC",
- .ack = vic_ack_irq,
- .mask = vic_mask_irq,
- .unmask = vic_unmask_irq,
- .set_wake = vic_set_wake,
+ .irq_ack = vic_ack_irq,
+ .irq_mask = vic_mask_irq,
+ .irq_unmask = vic_unmask_irq,
+ .irq_set_wake = vic_set_wake,
};
static void __init vic_disable(void __iomem *base)
diff --git a/arch/arm/include/asm/bitops.h b/arch/arm/include/asm/bitops.h
index 338ff19..7b1bb2b 100644
--- a/arch/arm/include/asm/bitops.h
+++ b/arch/arm/include/asm/bitops.h
@@ -285,7 +285,7 @@
if (__builtin_constant_p(x))
return constant_fls(x);
- asm("clz\t%0, %1" : "=r" (ret) : "r" (x) : "cc");
+ asm("clz\t%0, %1" : "=r" (ret) : "r" (x));
ret = 32 - ret;
return ret;
}
diff --git a/arch/arm/include/asm/sched_clock.h b/arch/arm/include/asm/sched_clock.h
index a84628b..c8e6ddf 100644
--- a/arch/arm/include/asm/sched_clock.h
+++ b/arch/arm/include/asm/sched_clock.h
@@ -115,4 +115,6 @@
}
}
+extern void sched_clock_postinit(void);
+
#endif
diff --git a/arch/arm/kernel/ecard.c b/arch/arm/kernel/ecard.c
index eed2f79..2ad62df 100644
--- a/arch/arm/kernel/ecard.c
+++ b/arch/arm/kernel/ecard.c
@@ -443,40 +443,40 @@
*
* They are not meant to be called directly, but via enable/disable_irq.
*/
-static void ecard_irq_unmask(unsigned int irqnr)
+static void ecard_irq_unmask(struct irq_data *d)
{
- ecard_t *ec = slot_to_ecard(irqnr - 32);
+ ecard_t *ec = slot_to_ecard(d->irq - 32);
if (ec) {
if (!ec->ops)
ec->ops = &ecard_default_ops;
if (ec->claimed && ec->ops->irqenable)
- ec->ops->irqenable(ec, irqnr);
+ ec->ops->irqenable(ec, d->irq);
else
printk(KERN_ERR "ecard: rejecting request to "
- "enable IRQs for %d\n", irqnr);
+ "enable IRQs for %d\n", d->irq);
}
}
-static void ecard_irq_mask(unsigned int irqnr)
+static void ecard_irq_mask(struct irq_data *d)
{
- ecard_t *ec = slot_to_ecard(irqnr - 32);
+ ecard_t *ec = slot_to_ecard(d->irq - 32);
if (ec) {
if (!ec->ops)
ec->ops = &ecard_default_ops;
if (ec->ops && ec->ops->irqdisable)
- ec->ops->irqdisable(ec, irqnr);
+ ec->ops->irqdisable(ec, d->irq);
}
}
static struct irq_chip ecard_chip = {
- .name = "ECARD",
- .ack = ecard_irq_mask,
- .mask = ecard_irq_mask,
- .unmask = ecard_irq_unmask,
+ .name = "ECARD",
+ .irq_ack = ecard_irq_mask,
+ .irq_mask = ecard_irq_mask,
+ .irq_unmask = ecard_irq_unmask,
};
void ecard_enablefiq(unsigned int fiqnr)
@@ -551,7 +551,7 @@
printk(KERN_ERR "\nInterrupt lockup detected - "
"disabling all expansion card interrupts\n");
- desc->chip->mask(IRQ_EXPANSIONCARD);
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
ecard_dump_irq_state();
}
} else
@@ -574,7 +574,7 @@
ecard_t *ec;
int called = 0;
- desc->chip->mask(irq);
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
for (ec = cards; ec; ec = ec->next) {
int pending;
@@ -591,7 +591,7 @@
called ++;
}
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
if (called == 0)
ecard_check_lockup(desc);
diff --git a/arch/arm/kernel/head-common.S b/arch/arm/kernel/head-common.S
index bbecaac..8f57515 100644
--- a/arch/arm/kernel/head-common.S
+++ b/arch/arm/kernel/head-common.S
@@ -60,6 +60,8 @@
str_a2: .asciz ").\n\nAvailable machine support:\n\nID (hex)\tNAME\n"
str_a3: .asciz "\nPlease check your kernel config and/or bootloader.\n"
.align
+#else
+ b __error
#endif
/*
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index 8135438..28536e3 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -88,7 +88,7 @@
seq_printf(p, "%*d: ", prec, i);
for_each_present_cpu(cpu)
seq_printf(p, "%10u ", kstat_irqs_cpu(i, cpu));
- seq_printf(p, " %10s", desc->chip->name ? : "-");
+ seq_printf(p, " %10s", desc->irq_data.chip->name ? : "-");
seq_printf(p, " %s", action->name);
for (action = action->next; action; action = action->next)
seq_printf(p, ", %s", action->name);
@@ -181,10 +181,11 @@
static void route_irq(struct irq_desc *desc, unsigned int irq, unsigned int cpu)
{
- pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->node, cpu);
+ pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->irq_data.node, cpu);
raw_spin_lock_irq(&desc->lock);
- desc->chip->set_affinity(irq, cpumask_of(cpu));
+ desc->irq_data.chip->irq_set_affinity(&desc->irq_data,
+ cpumask_of(cpu), false);
raw_spin_unlock_irq(&desc->lock);
}
@@ -199,16 +200,18 @@
struct irq_desc *desc;
for_each_irq_desc(i, desc) {
- if (desc->node == cpu) {
- unsigned int newcpu = cpumask_any_and(desc->affinity,
+ struct irq_data *d = &desc->irq_data;
+
+ if (d->node == cpu) {
+ unsigned int newcpu = cpumask_any_and(d->affinity,
cpu_online_mask);
if (newcpu >= nr_cpu_ids) {
if (printk_ratelimit())
printk(KERN_INFO "IRQ%u no longer affine to CPU%u\n",
i, cpu);
- cpumask_setall(desc->affinity);
- newcpu = cpumask_any_and(desc->affinity,
+ cpumask_setall(d->affinity);
+ newcpu = cpumask_any_and(d->affinity,
cpu_online_mask);
}
diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c
index 0c1bb68..2cfe816 100644
--- a/arch/arm/kernel/module.c
+++ b/arch/arm/kernel/module.c
@@ -38,17 +38,9 @@
#ifdef CONFIG_MMU
void *module_alloc(unsigned long size)
{
- struct vm_struct *area;
-
- size = PAGE_ALIGN(size);
- if (!size)
- return NULL;
-
- area = __get_vm_area(size, VM_ALLOC, MODULES_VADDR, MODULES_END);
- if (!area)
- return NULL;
-
- return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL_EXEC);
+ return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END,
+ GFP_KERNEL, PAGE_KERNEL_EXEC, -1,
+ __builtin_return_address(0));
}
#else /* CONFIG_MMU */
void *module_alloc(unsigned long size)
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index e76fcaa..94bbedb 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -483,6 +483,7 @@
return randomize_range(mm->brk, range_end, 0) ? : mm->brk;
}
+#ifdef CONFIG_MMU
/*
* The vectors page is always readable from user space for the
* atomic helpers and the signal restart code. Let's declare a mapping
@@ -503,3 +504,4 @@
{
return (vma->vm_start == 0xffff0000) ? "[vectors]" : NULL;
}
+#endif
diff --git a/arch/arm/kernel/sched_clock.c b/arch/arm/kernel/sched_clock.c
index 2cdcc92..9a46370 100644
--- a/arch/arm/kernel/sched_clock.c
+++ b/arch/arm/kernel/sched_clock.c
@@ -34,7 +34,7 @@
sched_clock_update_fn = update;
/* calculate the mult/shift to convert counter ticks to ns. */
- clocks_calc_mult_shift(&cd->mult, &cd->shift, rate, NSEC_PER_SEC, 60);
+ clocks_calc_mult_shift(&cd->mult, &cd->shift, rate, NSEC_PER_SEC, 0);
r = rate;
if (r >= 4000000) {
@@ -60,10 +60,15 @@
* sets the initial epoch.
*/
sched_clock_timer.data = msecs_to_jiffies(w - (w / 10));
- sched_clock_poll(sched_clock_timer.data);
+ update();
/*
* Ensure that sched_clock() starts off at 0ns
*/
cd->epoch_ns = 0;
}
+
+void __init sched_clock_postinit(void)
+{
+ sched_clock_poll(sched_clock_timer.data);
+}
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 3455ad3..420b8d6 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -518,25 +518,21 @@
#endif
}
-static void __init
-request_standard_resources(struct meminfo *mi, struct machine_desc *mdesc)
+static void __init request_standard_resources(struct machine_desc *mdesc)
{
+ struct memblock_region *region;
struct resource *res;
- int i;
kernel_code.start = virt_to_phys(_text);
kernel_code.end = virt_to_phys(_etext - 1);
kernel_data.start = virt_to_phys(_sdata);
kernel_data.end = virt_to_phys(_end - 1);
- for (i = 0; i < mi->nr_banks; i++) {
- if (mi->bank[i].size == 0)
- continue;
-
+ for_each_memblock(memory, region) {
res = alloc_bootmem_low(sizeof(*res));
res->name = "System RAM";
- res->start = mi->bank[i].start;
- res->end = mi->bank[i].start + mi->bank[i].size - 1;
+ res->start = __pfn_to_phys(memblock_region_memory_base_pfn(region));
+ res->end = __pfn_to_phys(memblock_region_memory_end_pfn(region)) - 1;
res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
request_resource(&iomem_resource, res);
@@ -650,15 +646,17 @@
__tagtable(ATAG_REVISION, parse_tag_revision);
-#ifndef CONFIG_CMDLINE_FORCE
static int __init parse_tag_cmdline(const struct tag *tag)
{
+#ifndef CONFIG_CMDLINE_FORCE
strlcpy(default_command_line, tag->u.cmdline.cmdline, COMMAND_LINE_SIZE);
+#else
+ pr_warning("Ignoring tag cmdline (using the default kernel command line)\n");
+#endif /* CONFIG_CMDLINE_FORCE */
return 0;
}
__tagtable(ATAG_CMDLINE, parse_tag_cmdline);
-#endif /* CONFIG_CMDLINE_FORCE */
/*
* Scan the tag table for this tag, and call its parse function.
@@ -857,7 +855,7 @@
arm_memblock_init(&meminfo, mdesc);
paging_init(mdesc);
- request_standard_resources(&meminfo, mdesc);
+ request_standard_resources(mdesc);
#ifdef CONFIG_SMP
if (is_smp())
diff --git a/arch/arm/kernel/smp_twd.c b/arch/arm/kernel/smp_twd.c
index dd79074..fd91566 100644
--- a/arch/arm/kernel/smp_twd.c
+++ b/arch/arm/kernel/smp_twd.c
@@ -114,7 +114,7 @@
twd_timer_rate = (0xFFFFFFFFU - count) * (HZ / 5);
printk("%lu.%02luMHz.\n", twd_timer_rate / 1000000,
- (twd_timer_rate / 100000) % 100);
+ (twd_timer_rate / 1000000) % 100);
}
load = twd_timer_rate / HZ;
diff --git a/arch/arm/kernel/stacktrace.c b/arch/arm/kernel/stacktrace.c
index c2e112e..381d23a 100644
--- a/arch/arm/kernel/stacktrace.c
+++ b/arch/arm/kernel/stacktrace.c
@@ -94,10 +94,13 @@
if (tsk != current) {
#ifdef CONFIG_SMP
/*
- * What guarantees do we have here that 'tsk'
- * is not running on another CPU?
+ * What guarantees do we have here that 'tsk' is not
+ * running on another CPU? For now, ignore it as we
+ * can't guarantee we won't explode.
*/
- BUG();
+ if (trace->nr_entries < trace->max_entries)
+ trace->entries[trace->nr_entries++] = ULONG_MAX;
+ return;
#else
data.no_sched_functions = 1;
frame.fp = thread_saved_fp(tsk);
diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c
index f1e2eb1..3d76bf2 100644
--- a/arch/arm/kernel/time.c
+++ b/arch/arm/kernel/time.c
@@ -29,6 +29,7 @@
#include <asm/leds.h>
#include <asm/thread_info.h>
+#include <asm/sched_clock.h>
#include <asm/stacktrace.h>
#include <asm/mach/arch.h>
#include <asm/mach/time.h>
@@ -163,5 +164,8 @@
{
system_timer = machine_desc->timer;
system_timer->init();
+#ifdef CONFIG_HAVE_SCHED_CLOCK
+ sched_clock_postinit();
+#endif
}
diff --git a/arch/arm/lib/delay.S b/arch/arm/lib/delay.S
index 8d6a876..3c9a05c 100644
--- a/arch/arm/lib/delay.S
+++ b/arch/arm/lib/delay.S
@@ -25,11 +25,15 @@
ldr r2, .LC1
mul r0, r2, r0
ENTRY(__const_udelay) @ 0 <= r0 <= 0x7fffff06
+ mov r1, #-1
ldr r2, .LC0
ldr r2, [r2] @ max = 0x01ffffff
+ add r0, r0, r1, lsr #32-14
mov r0, r0, lsr #14 @ max = 0x0001ffff
+ add r2, r2, r1, lsr #32-10
mov r2, r2, lsr #10 @ max = 0x00007fff
mul r0, r2, r0 @ max = 2^32-1
+ add r0, r0, r1, lsr #32-6
movs r0, r0, lsr #6
moveq pc, lr
diff --git a/arch/arm/mach-aaec2000/core.c b/arch/arm/mach-aaec2000/core.c
index 3ef6833..f8465bd 100644
--- a/arch/arm/mach-aaec2000/core.c
+++ b/arch/arm/mach-aaec2000/core.c
@@ -68,25 +68,25 @@
/*
* Interrupt handling routines
*/
-static void aaec2000_int_ack(unsigned int irq)
+static void aaec2000_int_ack(struct irq_data *d)
{
- IRQ_INTSR = 1 << irq;
+ IRQ_INTSR = 1 << d->irq;
}
-static void aaec2000_int_mask(unsigned int irq)
+static void aaec2000_int_mask(struct irq_data *d)
{
- IRQ_INTENC |= (1 << irq);
+ IRQ_INTENC |= (1 << d->irq);
}
-static void aaec2000_int_unmask(unsigned int irq)
+static void aaec2000_int_unmask(struct irq_data *d)
{
- IRQ_INTENS |= (1 << irq);
+ IRQ_INTENS |= (1 << d->irq);
}
static struct irq_chip aaec2000_irq_chip = {
- .ack = aaec2000_int_ack,
- .mask = aaec2000_int_mask,
- .unmask = aaec2000_int_unmask,
+ .irq_ack = aaec2000_int_ack,
+ .irq_mask = aaec2000_int_mask,
+ .irq_unmask = aaec2000_int_unmask,
};
void __init aaec2000_init_irq(void)
diff --git a/arch/arm/mach-at91/Kconfig b/arch/arm/mach-at91/Kconfig
index c015b68..1939023 100644
--- a/arch/arm/mach-at91/Kconfig
+++ b/arch/arm/mach-at91/Kconfig
@@ -362,6 +362,12 @@
Select this if you are using a Eukrea Electromatique's
CPU9G20 Board <http://www.eukrea.com/>
+config MACH_ACMENETUSFOXG20
+ bool "Acme Systems srl FOX Board G20"
+ help
+ Select this if you are using Acme Systems
+ FOX Board G20 <http://www.acmesystems.it>
+
config MACH_PORTUXG20
bool "taskit PortuxG20"
help
@@ -381,6 +387,13 @@
Select this if you are using taskit's Stamp9G20 CPU module on this
carrier board, beeing the decentralized unit of a building automation
system; featuring nvram, eth-switch, iso-rs485, display, io
+
+config MACH_GSIA18S
+ bool "GS_IA18_S board"
+ help
+ This enables support for the GS_IA18_S board
+ produced by GeoSIG Ltd company. This is an internet accelerograph.
+ <http://www.geosig.com>
endif
if (ARCH_AT91SAM9260 || ARCH_AT91SAM9G20)
diff --git a/arch/arm/mach-at91/Makefile b/arch/arm/mach-at91/Makefile
index d13add7..a83835e 100644
--- a/arch/arm/mach-at91/Makefile
+++ b/arch/arm/mach-at91/Makefile
@@ -63,9 +63,11 @@
# AT91SAM9G20 board-specific support
obj-$(CONFIG_MACH_AT91SAM9G20EK) += board-sam9g20ek.o
obj-$(CONFIG_MACH_CPU9G20) += board-cpu9krea.o
+obj-$(CONFIG_MACH_ACMENETUSFOXG20) += board-foxg20.o
obj-$(CONFIG_MACH_STAMP9G20) += board-stamp9g20.o
obj-$(CONFIG_MACH_PORTUXG20) += board-stamp9g20.o
obj-$(CONFIG_MACH_PCONTROL_G20) += board-pcontrol-g20.o board-stamp9g20.o
+obj-$(CONFIG_MACH_GSIA18S) += board-gsia18s.o board-stamp9g20.o
# AT91SAM9260/AT91SAM9G20 board-specific support
obj-$(CONFIG_MACH_SNAPPER_9260) += board-snapper9260.o
diff --git a/arch/arm/mach-at91/board-foxg20.c b/arch/arm/mach-at91/board-foxg20.c
new file mode 100644
index 0000000..dfc7dfe
--- /dev/null
+++ b/arch/arm/mach-at91/board-foxg20.c
@@ -0,0 +1,274 @@
+/*
+ * Copyright (C) 2005 SAN People
+ * Copyright (C) 2008 Atmel
+ * Copyright (C) 2010 Lee McLoughlin - lee@lmmrtech.com
+ * Copyright (C) 2010 Sergio Tanzilli - tanzilli@acmesystems.it
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/at73c213.h>
+#include <linux/gpio.h>
+#include <linux/gpio_keys.h>
+#include <linux/input.h>
+#include <linux/clk.h>
+#include <linux/w1-gpio.h>
+
+#include <mach/hardware.h>
+#include <asm/setup.h>
+#include <asm/mach-types.h>
+#include <asm/irq.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/irq.h>
+
+#include <mach/board.h>
+#include <mach/at91sam9_smc.h>
+
+#include "sam9_smc.h"
+#include "generic.h"
+
+/*
+ * The FOX Board G20 hardware comes as the "Netus G20" board with
+ * just the cpu, ram, dataflash and two header connectors.
+ * This is plugged into the FOX Board which provides the ethernet,
+ * usb, rtc, leds, switch, ...
+ *
+ * For more info visit: http://www.acmesystems.it/foxg20
+ */
+
+
+static void __init foxg20_map_io(void)
+{
+ /* Initialize processor: 18.432 MHz crystal */
+ at91sam9260_initialize(18432000);
+
+ /* DBGU on ttyS0. (Rx & Tx only) */
+ at91_register_uart(0, 0, 0);
+
+ /* USART0 on ttyS1. (Rx, Tx, CTS, RTS, DTR, DSR, DCD, RI) */
+ at91_register_uart(AT91SAM9260_ID_US0, 1,
+ ATMEL_UART_CTS
+ | ATMEL_UART_RTS
+ | ATMEL_UART_DTR
+ | ATMEL_UART_DSR
+ | ATMEL_UART_DCD
+ | ATMEL_UART_RI);
+
+ /* USART1 on ttyS2. (Rx, Tx, RTS, CTS) */
+ at91_register_uart(AT91SAM9260_ID_US1, 2,
+ ATMEL_UART_CTS
+ | ATMEL_UART_RTS);
+
+ /* USART2 on ttyS3. (Rx & Tx only) */
+ at91_register_uart(AT91SAM9260_ID_US2, 3, 0);
+
+ /* USART3 on ttyS4. (Rx, Tx, RTS, CTS) */
+ at91_register_uart(AT91SAM9260_ID_US3, 4,
+ ATMEL_UART_CTS
+ | ATMEL_UART_RTS);
+
+ /* USART4 on ttyS5. (Rx & Tx only) */
+ at91_register_uart(AT91SAM9260_ID_US4, 5, 0);
+
+ /* USART5 on ttyS6. (Rx & Tx only) */
+ at91_register_uart(AT91SAM9260_ID_US5, 6, 0);
+
+ /* set serial console to ttyS0 (ie, DBGU) */
+ at91_set_serial_console(0);
+
+ /* Set the internal pull-up resistor on DRXD */
+ at91_set_A_periph(AT91_PIN_PB14, 1);
+
+}
+
+static void __init foxg20_init_irq(void)
+{
+ at91sam9260_init_interrupts(NULL);
+}
+
+
+/*
+ * USB Host port
+ */
+static struct at91_usbh_data __initdata foxg20_usbh_data = {
+ .ports = 2,
+};
+
+/*
+ * USB Device port
+ */
+static struct at91_udc_data __initdata foxg20_udc_data = {
+ .vbus_pin = AT91_PIN_PC6,
+ .pullup_pin = 0, /* pull-up driven by UDC */
+};
+
+
+/*
+ * SPI devices.
+ */
+static struct spi_board_info foxg20_spi_devices[] = {
+#if !defined(CONFIG_MMC_AT91)
+ {
+ .modalias = "mtd_dataflash",
+ .chip_select = 1,
+ .max_speed_hz = 15 * 1000 * 1000,
+ .bus_num = 0,
+ },
+#endif
+};
+
+
+/*
+ * MACB Ethernet device
+ */
+static struct at91_eth_data __initdata foxg20_macb_data = {
+ .phy_irq_pin = AT91_PIN_PA7,
+ .is_rmii = 1,
+};
+
+/*
+ * MCI (SD/MMC)
+ * det_pin, wp_pin and vcc_pin are not connected
+ */
+static struct at91_mmc_data __initdata foxg20_mmc_data = {
+ .slot_b = 1,
+ .wire4 = 1,
+};
+
+
+/*
+ * LEDs
+ */
+static struct gpio_led foxg20_leds[] = {
+ { /* user led, red */
+ .name = "user_led",
+ .gpio = AT91_PIN_PC7,
+ .active_low = 0,
+ .default_trigger = "heartbeat",
+ },
+};
+
+
+/*
+ * GPIO Buttons
+ */
+#if defined(CONFIG_KEYBOARD_GPIO) || defined(CONFIG_KEYBOARD_GPIO_MODULE)
+static struct gpio_keys_button foxg20_buttons[] = {
+ {
+ .gpio = AT91_PIN_PC4,
+ .code = BTN_1,
+ .desc = "Button 1",
+ .active_low = 1,
+ .wakeup = 1,
+ },
+};
+
+static struct gpio_keys_platform_data foxg20_button_data = {
+ .buttons = foxg20_buttons,
+ .nbuttons = ARRAY_SIZE(foxg20_buttons),
+};
+
+static struct platform_device foxg20_button_device = {
+ .name = "gpio-keys",
+ .id = -1,
+ .num_resources = 0,
+ .dev = {
+ .platform_data = &foxg20_button_data,
+ }
+};
+
+static void __init foxg20_add_device_buttons(void)
+{
+ at91_set_gpio_input(AT91_PIN_PC4, 1); /* btn1 */
+ at91_set_deglitch(AT91_PIN_PC4, 1);
+
+ platform_device_register(&foxg20_button_device);
+}
+#else
+static void __init foxg20_add_device_buttons(void) {}
+#endif
+
+
+#if defined(CONFIG_W1_MASTER_GPIO) || defined(CONFIG_W1_MASTER_GPIO_MODULE)
+static struct w1_gpio_platform_data w1_gpio_pdata = {
+ /* If you choose to use a pin other than PB16 it needs to be 3.3V */
+ .pin = AT91_PIN_PB16,
+ .is_open_drain = 1,
+};
+
+static struct platform_device w1_device = {
+ .name = "w1-gpio",
+ .id = -1,
+ .dev.platform_data = &w1_gpio_pdata,
+};
+
+static void __init at91_add_device_w1(void)
+{
+ at91_set_GPIO_periph(w1_gpio_pdata.pin, 1);
+ at91_set_multi_drive(w1_gpio_pdata.pin, 1);
+ platform_device_register(&w1_device);
+}
+
+#endif
+
+
+static struct i2c_board_info __initdata foxg20_i2c_devices[] = {
+ {
+ I2C_BOARD_INFO("24c512", 0x50),
+ },
+};
+
+
+static void __init foxg20_board_init(void)
+{
+ /* Serial */
+ at91_add_device_serial();
+ /* USB Host */
+ at91_add_device_usbh(&foxg20_usbh_data);
+ /* USB Device */
+ at91_add_device_udc(&foxg20_udc_data);
+ /* SPI */
+ at91_add_device_spi(foxg20_spi_devices, ARRAY_SIZE(foxg20_spi_devices));
+ /* Ethernet */
+ at91_add_device_eth(&foxg20_macb_data);
+ /* MMC */
+ at91_add_device_mmc(0, &foxg20_mmc_data);
+ /* I2C */
+ at91_add_device_i2c(foxg20_i2c_devices, ARRAY_SIZE(foxg20_i2c_devices));
+ /* LEDs */
+ at91_gpio_leds(foxg20_leds, ARRAY_SIZE(foxg20_leds));
+ /* Push Buttons */
+ foxg20_add_device_buttons();
+#if defined(CONFIG_W1_MASTER_GPIO) || defined(CONFIG_W1_MASTER_GPIO_MODULE)
+ at91_add_device_w1();
+#endif
+}
+
+MACHINE_START(ACMENETUSFOXG20, "Acme Systems srl FOX Board G20")
+ /* Maintainer: Sergio Tanzilli */
+ .boot_params = AT91_SDRAM_BASE + 0x100,
+ .timer = &at91sam926x_timer,
+ .map_io = foxg20_map_io,
+ .init_irq = foxg20_init_irq,
+ .init_machine = foxg20_board_init,
+MACHINE_END
diff --git a/arch/arm/mach-at91/board-gsia18s.c b/arch/arm/mach-at91/board-gsia18s.c
new file mode 100644
index 0000000..bc28136
--- /dev/null
+++ b/arch/arm/mach-at91/board-gsia18s.c
@@ -0,0 +1,584 @@
+/*
+ * Copyright (C) 2010 Christian Glindkamp <christian.glindkamp@taskit.de>
+ * taskit GmbH
+ * 2010 Igor Plyatov <plyatov@gmail.com>
+ * GeoSIG 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
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/platform_device.h>
+#include <linux/gpio.h>
+#include <linux/w1-gpio.h>
+#include <linux/i2c.h>
+#include <linux/i2c/pcf857x.h>
+#include <linux/gpio_keys.h>
+#include <linux/input.h>
+
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+
+#include <mach/board.h>
+#include <mach/at91sam9_smc.h>
+#include <mach/gsia18s.h>
+#include <mach/stamp9g20.h>
+
+#include "sam9_smc.h"
+#include "generic.h"
+
+static void __init gsia18s_map_io(void)
+{
+ stamp9g20_map_io();
+
+ /*
+ * USART0 on ttyS1 (Rx, Tx, CTS, RTS, DTR, DSR, DCD, RI).
+ * Used for Internal Analog Modem.
+ */
+ at91_register_uart(AT91SAM9260_ID_US0, 1,
+ ATMEL_UART_CTS | ATMEL_UART_RTS |
+ ATMEL_UART_DTR | ATMEL_UART_DSR |
+ ATMEL_UART_DCD | ATMEL_UART_RI);
+ /*
+ * USART1 on ttyS2 (Rx, Tx, CTS, RTS).
+ * Used for GPS or WiFi or Data stream.
+ */
+ at91_register_uart(AT91SAM9260_ID_US1, 2,
+ ATMEL_UART_CTS | ATMEL_UART_RTS);
+ /*
+ * USART2 on ttyS3 (Rx, Tx, CTS, RTS).
+ * Used for External Modem.
+ */
+ at91_register_uart(AT91SAM9260_ID_US2, 3,
+ ATMEL_UART_CTS | ATMEL_UART_RTS);
+ /*
+ * USART3 on ttyS4 (Rx, Tx, RTS).
+ * Used for RS-485.
+ */
+ at91_register_uart(AT91SAM9260_ID_US3, 4, ATMEL_UART_RTS);
+
+ /*
+ * USART4 on ttyS5 (Rx, Tx).
+ * Used for TRX433 Radio Module.
+ */
+ at91_register_uart(AT91SAM9260_ID_US4, 5, 0);
+}
+
+static void __init init_irq(void)
+{
+ at91sam9260_init_interrupts(NULL);
+}
+
+/*
+ * Two USB Host ports
+ */
+static struct at91_usbh_data __initdata usbh_data = {
+ .ports = 2,
+};
+
+/*
+ * USB Device port
+ */
+static struct at91_udc_data __initdata udc_data = {
+ .vbus_pin = AT91_PIN_PA22,
+ .pullup_pin = 0, /* pull-up driven by UDC */
+};
+
+/*
+ * MACB Ethernet device
+ */
+static struct at91_eth_data __initdata macb_data = {
+ .phy_irq_pin = AT91_PIN_PA28,
+ .is_rmii = 1,
+};
+
+/*
+ * LEDs and GPOs
+ */
+static struct gpio_led gpio_leds[] = {
+ {
+ .name = "gpo:spi1reset",
+ .gpio = AT91_PIN_PC1,
+ .active_low = 0,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ {
+ .name = "gpo:trig_net_out",
+ .gpio = AT91_PIN_PB20,
+ .active_low = 0,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ {
+ .name = "gpo:trig_net_dir",
+ .gpio = AT91_PIN_PB19,
+ .active_low = 0,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ {
+ .name = "gpo:charge_dis",
+ .gpio = AT91_PIN_PC2,
+ .active_low = 0,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ {
+ .name = "led:event",
+ .gpio = AT91_PIN_PB17,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ {
+ .name = "led:lan",
+ .gpio = AT91_PIN_PB18,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ {
+ .name = "led:error",
+ .gpio = AT91_PIN_PB16,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_ON,
+ }
+};
+
+static struct gpio_led_platform_data gpio_led_info = {
+ .leds = gpio_leds,
+ .num_leds = ARRAY_SIZE(gpio_leds),
+};
+
+static struct platform_device leds = {
+ .name = "leds-gpio",
+ .id = 0,
+ .dev = {
+ .platform_data = &gpio_led_info,
+ }
+};
+
+static void __init gsia18s_leds_init(void)
+{
+ platform_device_register(&leds);
+}
+
+/* PCF8574 0x20 GPIO - U1 on the GS_IA18-CB_V3 board */
+static struct gpio_led pcf_gpio_leds1[] = {
+ { /* bit 0 */
+ .name = "gpo:hdc_power",
+ .gpio = PCF_GPIO_HDC_POWER,
+ .active_low = 0,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ { /* bit 1 */
+ .name = "gpo:wifi_setup",
+ .gpio = PCF_GPIO_WIFI_SETUP,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ { /* bit 2 */
+ .name = "gpo:wifi_enable",
+ .gpio = PCF_GPIO_WIFI_ENABLE,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ { /* bit 3 */
+ .name = "gpo:wifi_reset",
+ .gpio = PCF_GPIO_WIFI_RESET,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_ON,
+ },
+ /* bit 4 used as GPI */
+ { /* bit 5 */
+ .name = "gpo:gps_setup",
+ .gpio = PCF_GPIO_GPS_SETUP,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ { /* bit 6 */
+ .name = "gpo:gps_standby",
+ .gpio = PCF_GPIO_GPS_STANDBY,
+ .active_low = 0,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_ON,
+ },
+ { /* bit 7 */
+ .name = "gpo:gps_power",
+ .gpio = PCF_GPIO_GPS_POWER,
+ .active_low = 0,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ }
+};
+
+static struct gpio_led_platform_data pcf_gpio_led_info1 = {
+ .leds = pcf_gpio_leds1,
+ .num_leds = ARRAY_SIZE(pcf_gpio_leds1),
+};
+
+static struct platform_device pcf_leds1 = {
+ .name = "leds-gpio", /* GS_IA18-CB_board */
+ .id = 1,
+ .dev = {
+ .platform_data = &pcf_gpio_led_info1,
+ }
+};
+
+/* PCF8574 0x22 GPIO - U1 on the GS_2G_OPT1-A_V0 board (Alarm) */
+static struct gpio_led pcf_gpio_leds2[] = {
+ { /* bit 0 */
+ .name = "gpo:alarm_1",
+ .gpio = PCF_GPIO_ALARM1,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ { /* bit 1 */
+ .name = "gpo:alarm_2",
+ .gpio = PCF_GPIO_ALARM2,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ { /* bit 2 */
+ .name = "gpo:alarm_3",
+ .gpio = PCF_GPIO_ALARM3,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ { /* bit 3 */
+ .name = "gpo:alarm_4",
+ .gpio = PCF_GPIO_ALARM4,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ /* bits 4, 5, 6 not used */
+ { /* bit 7 */
+ .name = "gpo:alarm_v_relay_on",
+ .gpio = PCF_GPIO_ALARM_V_RELAY_ON,
+ .active_low = 0,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+};
+
+static struct gpio_led_platform_data pcf_gpio_led_info2 = {
+ .leds = pcf_gpio_leds2,
+ .num_leds = ARRAY_SIZE(pcf_gpio_leds2),
+};
+
+static struct platform_device pcf_leds2 = {
+ .name = "leds-gpio",
+ .id = 2,
+ .dev = {
+ .platform_data = &pcf_gpio_led_info2,
+ }
+};
+
+/* PCF8574 0x24 GPIO U1 on the GS_2G-OPT23-A_V0 board (Modem) */
+static struct gpio_led pcf_gpio_leds3[] = {
+ { /* bit 0 */
+ .name = "gpo:modem_power",
+ .gpio = PCF_GPIO_MODEM_POWER,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_OFF,
+ },
+ /* bits 1 and 2 not used */
+ { /* bit 3 */
+ .name = "gpo:modem_reset",
+ .gpio = PCF_GPIO_MODEM_RESET,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_ON,
+ },
+ /* bits 4, 5 and 6 not used */
+ { /* bit 7 */
+ .name = "gpo:trx_reset",
+ .gpio = PCF_GPIO_TRX_RESET,
+ .active_low = 1,
+ .default_trigger = "none",
+ .default_state = LEDS_GPIO_DEFSTATE_ON,
+ }
+};
+
+static struct gpio_led_platform_data pcf_gpio_led_info3 = {
+ .leds = pcf_gpio_leds3,
+ .num_leds = ARRAY_SIZE(pcf_gpio_leds3),
+};
+
+static struct platform_device pcf_leds3 = {
+ .name = "leds-gpio",
+ .id = 3,
+ .dev = {
+ .platform_data = &pcf_gpio_led_info3,
+ }
+};
+
+static void __init gsia18s_pcf_leds_init(void)
+{
+ platform_device_register(&pcf_leds1);
+ platform_device_register(&pcf_leds2);
+ platform_device_register(&pcf_leds3);
+}
+
+/*
+ * SPI busses.
+ */
+static struct spi_board_info gsia18s_spi_devices[] = {
+ { /* User accessible spi0, cs0 used for communication with MSP RTC */
+ .modalias = "spidev",
+ .bus_num = 0,
+ .chip_select = 0,
+ .max_speed_hz = 580000,
+ .mode = SPI_MODE_1,
+ },
+ { /* User accessible spi1, cs0 used for communication with int. DSP */
+ .modalias = "spidev",
+ .bus_num = 1,
+ .chip_select = 0,
+ .max_speed_hz = 5600000,
+ .mode = SPI_MODE_0,
+ },
+ { /* User accessible spi1, cs1 used for communication with ext. DSP */
+ .modalias = "spidev",
+ .bus_num = 1,
+ .chip_select = 1,
+ .max_speed_hz = 5600000,
+ .mode = SPI_MODE_0,
+ },
+ { /* User accessible spi1, cs2 used for communication with ext. DSP */
+ .modalias = "spidev",
+ .bus_num = 1,
+ .chip_select = 2,
+ .max_speed_hz = 5600000,
+ .mode = SPI_MODE_0,
+ },
+ { /* User accessible spi1, cs3 used for communication with ext. DSP */
+ .modalias = "spidev",
+ .bus_num = 1,
+ .chip_select = 3,
+ .max_speed_hz = 5600000,
+ .mode = SPI_MODE_0,
+ }
+};
+
+/*
+ * GPI Buttons
+ */
+static struct gpio_keys_button buttons[] = {
+ {
+ .gpio = GPIO_TRIG_NET_IN,
+ .code = BTN_1,
+ .desc = "TRIG_NET_IN",
+ .type = EV_KEY,
+ .active_low = 0,
+ .wakeup = 1,
+ },
+ { /* SW80 on the GS_IA18_S-MN board*/
+ .gpio = GPIO_CARD_UNMOUNT_0,
+ .code = BTN_2,
+ .desc = "Card umount 0",
+ .type = EV_KEY,
+ .active_low = 1,
+ .wakeup = 1,
+ },
+ { /* SW79 on the GS_IA18_S-MN board*/
+ .gpio = GPIO_CARD_UNMOUNT_1,
+ .code = BTN_3,
+ .desc = "Card umount 1",
+ .type = EV_KEY,
+ .active_low = 1,
+ .wakeup = 1,
+ },
+ { /* SW280 on the GS_IA18-CB board*/
+ .gpio = GPIO_KEY_POWER,
+ .code = KEY_POWER,
+ .desc = "Power Off Button",
+ .type = EV_KEY,
+ .active_low = 0,
+ .wakeup = 1,
+ }
+};
+
+static struct gpio_keys_platform_data button_data = {
+ .buttons = buttons,
+ .nbuttons = ARRAY_SIZE(buttons),
+};
+
+static struct platform_device button_device = {
+ .name = "gpio-keys",
+ .id = -1,
+ .num_resources = 0,
+ .dev = {
+ .platform_data = &button_data,
+ }
+};
+
+static void __init gsia18s_add_device_buttons(void)
+{
+ at91_set_gpio_input(GPIO_TRIG_NET_IN, 1);
+ at91_set_deglitch(GPIO_TRIG_NET_IN, 1);
+ at91_set_gpio_input(GPIO_CARD_UNMOUNT_0, 1);
+ at91_set_deglitch(GPIO_CARD_UNMOUNT_0, 1);
+ at91_set_gpio_input(GPIO_CARD_UNMOUNT_1, 1);
+ at91_set_deglitch(GPIO_CARD_UNMOUNT_1, 1);
+ at91_set_gpio_input(GPIO_KEY_POWER, 0);
+ at91_set_deglitch(GPIO_KEY_POWER, 1);
+
+ platform_device_register(&button_device);
+}
+
+/*
+ * I2C
+ */
+static int pcf8574x_0x20_setup(struct i2c_client *client, int gpio,
+ unsigned int ngpio, void *context)
+{
+ int status;
+
+ status = gpio_request(gpio + PCF_GPIO_ETH_DETECT, "eth_det");
+ if (status < 0) {
+ pr_err("error: can't request GPIO%d\n",
+ gpio + PCF_GPIO_ETH_DETECT);
+ return status;
+ }
+ status = gpio_direction_input(gpio + PCF_GPIO_ETH_DETECT);
+ if (status < 0) {
+ pr_err("error: can't setup GPIO%d as input\n",
+ gpio + PCF_GPIO_ETH_DETECT);
+ return status;
+ }
+ status = gpio_export(gpio + PCF_GPIO_ETH_DETECT, false);
+ if (status < 0) {
+ pr_err("error: can't export GPIO%d\n",
+ gpio + PCF_GPIO_ETH_DETECT);
+ return status;
+ }
+ status = gpio_sysfs_set_active_low(gpio + PCF_GPIO_ETH_DETECT, 1);
+ if (status < 0) {
+ pr_err("error: gpio_sysfs_set active_low(GPIO%d, 1)\n",
+ gpio + PCF_GPIO_ETH_DETECT);
+ return status;
+ }
+
+ return 0;
+}
+
+static int pcf8574x_0x20_teardown(struct i2c_client *client, int gpio,
+ unsigned ngpio, void *context)
+{
+ gpio_free(gpio + PCF_GPIO_ETH_DETECT);
+ return 0;
+}
+
+static struct pcf857x_platform_data pcf20_pdata = {
+ .gpio_base = GS_IA18_S_PCF_GPIO_BASE0,
+ .n_latch = (1 << 4),
+ .setup = pcf8574x_0x20_setup,
+ .teardown = pcf8574x_0x20_teardown,
+};
+
+static struct pcf857x_platform_data pcf22_pdata = {
+ .gpio_base = GS_IA18_S_PCF_GPIO_BASE1,
+};
+
+static struct pcf857x_platform_data pcf24_pdata = {
+ .gpio_base = GS_IA18_S_PCF_GPIO_BASE2,
+};
+
+static struct i2c_board_info __initdata gsia18s_i2c_devices[] = {
+ { /* U1 on the GS_IA18-CB_V3 board */
+ I2C_BOARD_INFO("pcf8574", 0x20),
+ .platform_data = &pcf20_pdata,
+ },
+ { /* U1 on the GS_2G_OPT1-A_V0 board (Alarm) */
+ I2C_BOARD_INFO("pcf8574", 0x22),
+ .platform_data = &pcf22_pdata,
+ },
+ { /* U1 on the GS_2G-OPT23-A_V0 board (Modem) */
+ I2C_BOARD_INFO("pcf8574", 0x24),
+ .platform_data = &pcf24_pdata,
+ },
+ { /* U161 on the GS_IA18_S-MN board */
+ I2C_BOARD_INFO("24c1024", 0x50),
+ },
+ { /* U162 on the GS_IA18_S-MN board */
+ I2C_BOARD_INFO("24c01", 0x53),
+ },
+};
+
+/*
+ * Compact Flash
+ */
+static struct at91_cf_data __initdata gsia18s_cf1_data = {
+ .irq_pin = AT91_PIN_PA27,
+ .det_pin = AT91_PIN_PB30,
+ .rst_pin = AT91_PIN_PB31,
+ .chipselect = 5,
+ .flags = AT91_CF_TRUE_IDE,
+};
+
+/* Power Off by RTC */
+static void gsia18s_power_off(void)
+{
+ pr_notice("Power supply will be switched off automatically now or after 60 seconds without ArmDAS.\n");
+ at91_set_gpio_output(AT91_PIN_PA25, 1);
+ /* Spin to death... */
+ while (1)
+ ;
+}
+
+static int __init gsia18s_power_off_init(void)
+{
+ pm_power_off = gsia18s_power_off;
+ return 0;
+}
+
+/* ---------------------------------------------------------------------------*/
+
+static void __init gsia18s_board_init(void)
+{
+ stamp9g20_board_init();
+ at91_add_device_usbh(&usbh_data);
+ at91_add_device_udc(&udc_data);
+ at91_add_device_eth(&macb_data);
+ gsia18s_leds_init();
+ gsia18s_pcf_leds_init();
+ gsia18s_add_device_buttons();
+ at91_add_device_i2c(gsia18s_i2c_devices,
+ ARRAY_SIZE(gsia18s_i2c_devices));
+ at91_add_device_cf(&gsia18s_cf1_data);
+ at91_add_device_spi(gsia18s_spi_devices,
+ ARRAY_SIZE(gsia18s_spi_devices));
+ gsia18s_power_off_init();
+}
+
+MACHINE_START(GSIA18S, "GS_IA18_S")
+ .boot_params = AT91_SDRAM_BASE + 0x100,
+ .timer = &at91sam926x_timer,
+ .map_io = gsia18s_map_io,
+ .init_irq = init_irq,
+ .init_machine = gsia18s_board_init,
+MACHINE_END
diff --git a/arch/arm/mach-at91/board-sam9m10g45ek.c b/arch/arm/mach-at91/board-sam9m10g45ek.c
index 86ff4b5..6c999db 100644
--- a/arch/arm/mach-at91/board-sam9m10g45ek.c
+++ b/arch/arm/mach-at91/board-sam9m10g45ek.c
@@ -37,7 +37,6 @@
#include <asm/mach/map.h>
#include <asm/mach/irq.h>
-#include <mach/hardware.h>
#include <mach/board.h>
#include <mach/gpio.h>
#include <mach/at91sam9_smc.h>
diff --git a/arch/arm/mach-at91/gpio.c b/arch/arm/mach-at91/gpio.c
index ae4772e..af818a2 100644
--- a/arch/arm/mach-at91/gpio.c
+++ b/arch/arm/mach-at91/gpio.c
@@ -274,10 +274,10 @@
static u32 wakeups[MAX_GPIO_BANKS];
static u32 backups[MAX_GPIO_BANKS];
-static int gpio_irq_set_wake(unsigned pin, unsigned state)
+static int gpio_irq_set_wake(struct irq_data *d, unsigned state)
{
- unsigned mask = pin_to_mask(pin);
- unsigned bank = (pin - PIN_BASE) / 32;
+ unsigned mask = pin_to_mask(d->irq);
+ unsigned bank = (d->irq - PIN_BASE) / 32;
if (unlikely(bank >= MAX_GPIO_BANKS))
return -EINVAL;
@@ -344,25 +344,25 @@
* IRQ0..IRQ6 should be configurable, e.g. level vs edge triggering.
*/
-static void gpio_irq_mask(unsigned pin)
+static void gpio_irq_mask(struct irq_data *d)
{
- void __iomem *pio = pin_to_controller(pin);
- unsigned mask = pin_to_mask(pin);
+ void __iomem *pio = pin_to_controller(d->irq);
+ unsigned mask = pin_to_mask(d->irq);
if (pio)
__raw_writel(mask, pio + PIO_IDR);
}
-static void gpio_irq_unmask(unsigned pin)
+static void gpio_irq_unmask(struct irq_data *d)
{
- void __iomem *pio = pin_to_controller(pin);
- unsigned mask = pin_to_mask(pin);
+ void __iomem *pio = pin_to_controller(d->irq);
+ unsigned mask = pin_to_mask(d->irq);
if (pio)
__raw_writel(mask, pio + PIO_IER);
}
-static int gpio_irq_type(unsigned pin, unsigned type)
+static int gpio_irq_type(struct irq_data *d, unsigned type)
{
switch (type) {
case IRQ_TYPE_NONE:
@@ -375,10 +375,10 @@
static struct irq_chip gpio_irqchip = {
.name = "GPIO",
- .mask = gpio_irq_mask,
- .unmask = gpio_irq_unmask,
- .set_type = gpio_irq_type,
- .set_wake = gpio_irq_set_wake,
+ .irq_mask = gpio_irq_mask,
+ .irq_unmask = gpio_irq_unmask,
+ .irq_set_type = gpio_irq_type,
+ .irq_set_wake = gpio_irq_set_wake,
};
static void gpio_irq_handler(unsigned irq, struct irq_desc *desc)
@@ -393,7 +393,7 @@
pio = at91_gpio->regbase;
/* temporarily mask (level sensitive) parent IRQ */
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
for (;;) {
/* Reading ISR acks pending (edge triggered) GPIO interrupts.
* When there none are pending, we're finished unless we need
@@ -419,7 +419,7 @@
* another IRQ must be generated before it actually gets
* here to be disabled on the GPIO controller.
*/
- gpio_irq_mask(pin);
+ gpio_irq_mask(irq_get_irq_data(pin));
}
else
generic_handle_irq(pin);
@@ -429,7 +429,7 @@
isr >>= 1;
}
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
/* now it may re-trigger */
}
diff --git a/arch/arm/mach-at91/include/mach/gsia18s.h b/arch/arm/mach-at91/include/mach/gsia18s.h
new file mode 100644
index 0000000..307c194
--- /dev/null
+++ b/arch/arm/mach-at91/include/mach/gsia18s.h
@@ -0,0 +1,33 @@
+/* Buttons */
+#define GPIO_TRIG_NET_IN AT91_PIN_PB21
+#define GPIO_CARD_UNMOUNT_0 AT91_PIN_PB13
+#define GPIO_CARD_UNMOUNT_1 AT91_PIN_PB12
+#define GPIO_KEY_POWER AT91_PIN_PA25
+
+/* PCF8574 0x20 GPIO - U1 on the GS_IA18-CB_V3 board */
+#define GS_IA18_S_PCF_GPIO_BASE0 NR_BUILTIN_GPIO
+#define PCF_GPIO_HDC_POWER (GS_IA18_S_PCF_GPIO_BASE0 + 0)
+#define PCF_GPIO_WIFI_SETUP (GS_IA18_S_PCF_GPIO_BASE0 + 1)
+#define PCF_GPIO_WIFI_ENABLE (GS_IA18_S_PCF_GPIO_BASE0 + 2)
+#define PCF_GPIO_WIFI_RESET (GS_IA18_S_PCF_GPIO_BASE0 + 3)
+#define PCF_GPIO_ETH_DETECT 4 /* this is a GPI */
+#define PCF_GPIO_GPS_SETUP (GS_IA18_S_PCF_GPIO_BASE0 + 5)
+#define PCF_GPIO_GPS_STANDBY (GS_IA18_S_PCF_GPIO_BASE0 + 6)
+#define PCF_GPIO_GPS_POWER (GS_IA18_S_PCF_GPIO_BASE0 + 7)
+
+/* PCF8574 0x22 GPIO - U1 on the GS_2G_OPT1-A_V0 board (Alarm) */
+#define GS_IA18_S_PCF_GPIO_BASE1 (GS_IA18_S_PCF_GPIO_BASE0 + 8)
+#define PCF_GPIO_ALARM1 (GS_IA18_S_PCF_GPIO_BASE1 + 0)
+#define PCF_GPIO_ALARM2 (GS_IA18_S_PCF_GPIO_BASE1 + 1)
+#define PCF_GPIO_ALARM3 (GS_IA18_S_PCF_GPIO_BASE1 + 2)
+#define PCF_GPIO_ALARM4 (GS_IA18_S_PCF_GPIO_BASE1 + 3)
+/* bits 4, 5, 6 not used */
+#define PCF_GPIO_ALARM_V_RELAY_ON (GS_IA18_S_PCF_GPIO_BASE1 + 7)
+
+/* PCF8574 0x24 GPIO U1 on the GS_2G-OPT23-A_V0 board (Modem) */
+#define GS_IA18_S_PCF_GPIO_BASE2 (GS_IA18_S_PCF_GPIO_BASE1 + 8)
+#define PCF_GPIO_MODEM_POWER (GS_IA18_S_PCF_GPIO_BASE2 + 0)
+#define PCF_GPIO_MODEM_RESET (GS_IA18_S_PCF_GPIO_BASE2 + 3)
+/* bits 1, 2, 4, 5 not used */
+#define PCF_GPIO_TRX_RESET (GS_IA18_S_PCF_GPIO_BASE2 + 6)
+/* bit 7 not used */
diff --git a/arch/arm/mach-at91/irq.c b/arch/arm/mach-at91/irq.c
index da3494a..b56d6b3a 100644
--- a/arch/arm/mach-at91/irq.c
+++ b/arch/arm/mach-at91/irq.c
@@ -34,23 +34,23 @@
#include <asm/mach/map.h>
-static void at91_aic_mask_irq(unsigned int irq)
+static void at91_aic_mask_irq(struct irq_data *d)
{
/* Disable interrupt on AIC */
- at91_sys_write(AT91_AIC_IDCR, 1 << irq);
+ at91_sys_write(AT91_AIC_IDCR, 1 << d->irq);
}
-static void at91_aic_unmask_irq(unsigned int irq)
+static void at91_aic_unmask_irq(struct irq_data *d)
{
/* Enable interrupt on AIC */
- at91_sys_write(AT91_AIC_IECR, 1 << irq);
+ at91_sys_write(AT91_AIC_IECR, 1 << d->irq);
}
unsigned int at91_extern_irq;
#define is_extern_irq(irq) ((1 << (irq)) & at91_extern_irq)
-static int at91_aic_set_type(unsigned irq, unsigned type)
+static int at91_aic_set_type(struct irq_data *d, unsigned type)
{
unsigned int smr, srctype;
@@ -62,13 +62,13 @@
srctype = AT91_AIC_SRCTYPE_RISING;
break;
case IRQ_TYPE_LEVEL_LOW:
- if ((irq == AT91_ID_FIQ) || is_extern_irq(irq)) /* only supported on external interrupts */
+ if ((d->irq == AT91_ID_FIQ) || is_extern_irq(d->irq)) /* only supported on external interrupts */
srctype = AT91_AIC_SRCTYPE_LOW;
else
return -EINVAL;
break;
case IRQ_TYPE_EDGE_FALLING:
- if ((irq == AT91_ID_FIQ) || is_extern_irq(irq)) /* only supported on external interrupts */
+ if ((d->irq == AT91_ID_FIQ) || is_extern_irq(d->irq)) /* only supported on external interrupts */
srctype = AT91_AIC_SRCTYPE_FALLING;
else
return -EINVAL;
@@ -77,8 +77,8 @@
return -EINVAL;
}
- smr = at91_sys_read(AT91_AIC_SMR(irq)) & ~AT91_AIC_SRCTYPE;
- at91_sys_write(AT91_AIC_SMR(irq), smr | srctype);
+ smr = at91_sys_read(AT91_AIC_SMR(d->irq)) & ~AT91_AIC_SRCTYPE;
+ at91_sys_write(AT91_AIC_SMR(d->irq), smr | srctype);
return 0;
}
@@ -87,15 +87,15 @@
static u32 wakeups;
static u32 backups;
-static int at91_aic_set_wake(unsigned irq, unsigned value)
+static int at91_aic_set_wake(struct irq_data *d, unsigned value)
{
- if (unlikely(irq >= 32))
+ if (unlikely(d->irq >= 32))
return -EINVAL;
if (value)
- wakeups |= (1 << irq);
+ wakeups |= (1 << d->irq);
else
- wakeups &= ~(1 << irq);
+ wakeups &= ~(1 << d->irq);
return 0;
}
@@ -119,11 +119,11 @@
static struct irq_chip at91_aic_chip = {
.name = "AIC",
- .ack = at91_aic_mask_irq,
- .mask = at91_aic_mask_irq,
- .unmask = at91_aic_unmask_irq,
- .set_type = at91_aic_set_type,
- .set_wake = at91_aic_set_wake,
+ .irq_ack = at91_aic_mask_irq,
+ .irq_mask = at91_aic_mask_irq,
+ .irq_unmask = at91_aic_unmask_irq,
+ .irq_set_type = at91_aic_set_type,
+ .irq_set_wake = at91_aic_set_wake,
};
/*
diff --git a/arch/arm/mach-bcmring/irq.c b/arch/arm/mach-bcmring/irq.c
index e315263..84dcda0 100644
--- a/arch/arm/mach-bcmring/irq.c
+++ b/arch/arm/mach-bcmring/irq.c
@@ -30,61 +30,61 @@
#include <mach/csp/intcHw_reg.h>
#include <mach/csp/mm_io.h>
-static void bcmring_mask_irq0(unsigned int irq)
+static void bcmring_mask_irq0(struct irq_data *d)
{
- writel(1 << (irq - IRQ_INTC0_START),
+ writel(1 << (d->irq - IRQ_INTC0_START),
MM_IO_BASE_INTC0 + INTCHW_INTENCLEAR);
}
-static void bcmring_unmask_irq0(unsigned int irq)
+static void bcmring_unmask_irq0(struct irq_data *d)
{
- writel(1 << (irq - IRQ_INTC0_START),
+ writel(1 << (d->irq - IRQ_INTC0_START),
MM_IO_BASE_INTC0 + INTCHW_INTENABLE);
}
-static void bcmring_mask_irq1(unsigned int irq)
+static void bcmring_mask_irq1(struct irq_data *d)
{
- writel(1 << (irq - IRQ_INTC1_START),
+ writel(1 << (d->irq - IRQ_INTC1_START),
MM_IO_BASE_INTC1 + INTCHW_INTENCLEAR);
}
-static void bcmring_unmask_irq1(unsigned int irq)
+static void bcmring_unmask_irq1(struct irq_data *d)
{
- writel(1 << (irq - IRQ_INTC1_START),
+ writel(1 << (d->irq - IRQ_INTC1_START),
MM_IO_BASE_INTC1 + INTCHW_INTENABLE);
}
-static void bcmring_mask_irq2(unsigned int irq)
+static void bcmring_mask_irq2(struct irq_data *d)
{
- writel(1 << (irq - IRQ_SINTC_START),
+ writel(1 << (d->irq - IRQ_SINTC_START),
MM_IO_BASE_SINTC + INTCHW_INTENCLEAR);
}
-static void bcmring_unmask_irq2(unsigned int irq)
+static void bcmring_unmask_irq2(struct irq_data *d)
{
- writel(1 << (irq - IRQ_SINTC_START),
+ writel(1 << (d->irq - IRQ_SINTC_START),
MM_IO_BASE_SINTC + INTCHW_INTENABLE);
}
static struct irq_chip bcmring_irq0_chip = {
.name = "ARM-INTC0",
- .ack = bcmring_mask_irq0,
- .mask = bcmring_mask_irq0, /* mask a specific interrupt, blocking its delivery. */
- .unmask = bcmring_unmask_irq0, /* unmaks an interrupt */
+ .irq_ack = bcmring_mask_irq0,
+ .irq_mask = bcmring_mask_irq0, /* mask a specific interrupt, blocking its delivery. */
+ .irq_unmask = bcmring_unmask_irq0, /* unmaks an interrupt */
};
static struct irq_chip bcmring_irq1_chip = {
.name = "ARM-INTC1",
- .ack = bcmring_mask_irq1,
- .mask = bcmring_mask_irq1,
- .unmask = bcmring_unmask_irq1,
+ .irq_ack = bcmring_mask_irq1,
+ .irq_mask = bcmring_mask_irq1,
+ .irq_unmask = bcmring_unmask_irq1,
};
static struct irq_chip bcmring_irq2_chip = {
.name = "ARM-SINTC",
- .ack = bcmring_mask_irq2,
- .mask = bcmring_mask_irq2,
- .unmask = bcmring_unmask_irq2,
+ .irq_ack = bcmring_mask_irq2,
+ .irq_mask = bcmring_mask_irq2,
+ .irq_unmask = bcmring_unmask_irq2,
};
static void vic_init(void __iomem *base, struct irq_chip *chip,
diff --git a/arch/arm/mach-clps711x/irq.c b/arch/arm/mach-clps711x/irq.c
index 9a12d85..86da7a1 100644
--- a/arch/arm/mach-clps711x/irq.c
+++ b/arch/arm/mach-clps711x/irq.c
@@ -27,24 +27,24 @@
#include <asm/hardware/clps7111.h>
-static void int1_mask(unsigned int irq)
+static void int1_mask(struct irq_data *d)
{
u32 intmr1;
intmr1 = clps_readl(INTMR1);
- intmr1 &= ~(1 << irq);
+ intmr1 &= ~(1 << d->irq);
clps_writel(intmr1, INTMR1);
}
-static void int1_ack(unsigned int irq)
+static void int1_ack(struct irq_data *d)
{
u32 intmr1;
intmr1 = clps_readl(INTMR1);
- intmr1 &= ~(1 << irq);
+ intmr1 &= ~(1 << d->irq);
clps_writel(intmr1, INTMR1);
- switch (irq) {
+ switch (d->irq) {
case IRQ_CSINT: clps_writel(0, COEOI); break;
case IRQ_TC1OI: clps_writel(0, TC1EOI); break;
case IRQ_TC2OI: clps_writel(0, TC2EOI); break;
@@ -54,56 +54,56 @@
}
}
-static void int1_unmask(unsigned int irq)
+static void int1_unmask(struct irq_data *d)
{
u32 intmr1;
intmr1 = clps_readl(INTMR1);
- intmr1 |= 1 << irq;
+ intmr1 |= 1 << d->irq;
clps_writel(intmr1, INTMR1);
}
static struct irq_chip int1_chip = {
- .ack = int1_ack,
- .mask = int1_mask,
- .unmask = int1_unmask,
+ .irq_ack = int1_ack,
+ .irq_mask = int1_mask,
+ .irq_unmask = int1_unmask,
};
-static void int2_mask(unsigned int irq)
+static void int2_mask(struct irq_data *d)
{
u32 intmr2;
intmr2 = clps_readl(INTMR2);
- intmr2 &= ~(1 << (irq - 16));
+ intmr2 &= ~(1 << (d->irq - 16));
clps_writel(intmr2, INTMR2);
}
-static void int2_ack(unsigned int irq)
+static void int2_ack(struct irq_data *d)
{
u32 intmr2;
intmr2 = clps_readl(INTMR2);
- intmr2 &= ~(1 << (irq - 16));
+ intmr2 &= ~(1 << (d->irq - 16));
clps_writel(intmr2, INTMR2);
- switch (irq) {
+ switch (d->irq) {
case IRQ_KBDINT: clps_writel(0, KBDEOI); break;
}
}
-static void int2_unmask(unsigned int irq)
+static void int2_unmask(struct irq_data *d)
{
u32 intmr2;
intmr2 = clps_readl(INTMR2);
- intmr2 |= 1 << (irq - 16);
+ intmr2 |= 1 << (d->irq - 16);
clps_writel(intmr2, INTMR2);
}
static struct irq_chip int2_chip = {
- .ack = int2_ack,
- .mask = int2_mask,
- .unmask = int2_unmask,
+ .irq_ack = int2_ack,
+ .irq_mask = int2_mask,
+ .irq_unmask = int2_unmask,
};
void __init clps711x_init_irq(void)
diff --git a/arch/arm/mach-davinci/cp_intc.c b/arch/arm/mach-davinci/cp_intc.c
index bb4c40e..9abc80a 100644
--- a/arch/arm/mach-davinci/cp_intc.c
+++ b/arch/arm/mach-davinci/cp_intc.c
@@ -26,30 +26,30 @@
__raw_writel(value, davinci_intc_base + offset);
}
-static void cp_intc_ack_irq(unsigned int irq)
+static void cp_intc_ack_irq(struct irq_data *d)
{
- cp_intc_write(irq, CP_INTC_SYS_STAT_IDX_CLR);
+ cp_intc_write(d->irq, CP_INTC_SYS_STAT_IDX_CLR);
}
/* Disable interrupt */
-static void cp_intc_mask_irq(unsigned int irq)
+static void cp_intc_mask_irq(struct irq_data *d)
{
/* XXX don't know why we need to disable nIRQ here... */
cp_intc_write(1, CP_INTC_HOST_ENABLE_IDX_CLR);
- cp_intc_write(irq, CP_INTC_SYS_ENABLE_IDX_CLR);
+ cp_intc_write(d->irq, CP_INTC_SYS_ENABLE_IDX_CLR);
cp_intc_write(1, CP_INTC_HOST_ENABLE_IDX_SET);
}
/* Enable interrupt */
-static void cp_intc_unmask_irq(unsigned int irq)
+static void cp_intc_unmask_irq(struct irq_data *d)
{
- cp_intc_write(irq, CP_INTC_SYS_ENABLE_IDX_SET);
+ cp_intc_write(d->irq, CP_INTC_SYS_ENABLE_IDX_SET);
}
-static int cp_intc_set_irq_type(unsigned int irq, unsigned int flow_type)
+static int cp_intc_set_irq_type(struct irq_data *d, unsigned int flow_type)
{
- unsigned reg = BIT_WORD(irq);
- unsigned mask = BIT_MASK(irq);
+ unsigned reg = BIT_WORD(d->irq);
+ unsigned mask = BIT_MASK(d->irq);
unsigned polarity = cp_intc_read(CP_INTC_SYS_POLARITY(reg));
unsigned type = cp_intc_read(CP_INTC_SYS_TYPE(reg));
@@ -85,18 +85,18 @@
* generic drivers which call {enable|disable}_irq_wake for
* wake up interrupt sources (eg RTC on DA850).
*/
-static int cp_intc_set_wake(unsigned int irq, unsigned int on)
+static int cp_intc_set_wake(struct irq_data *d, unsigned int on)
{
return 0;
}
static struct irq_chip cp_intc_irq_chip = {
.name = "cp_intc",
- .ack = cp_intc_ack_irq,
- .mask = cp_intc_mask_irq,
- .unmask = cp_intc_unmask_irq,
- .set_type = cp_intc_set_irq_type,
- .set_wake = cp_intc_set_wake,
+ .irq_ack = cp_intc_ack_irq,
+ .irq_mask = cp_intc_mask_irq,
+ .irq_unmask = cp_intc_unmask_irq,
+ .irq_set_type = cp_intc_set_irq_type,
+ .irq_set_wake = cp_intc_set_wake,
};
void __init cp_intc_init(void)
diff --git a/arch/arm/mach-davinci/gpio.c b/arch/arm/mach-davinci/gpio.c
index bf0ff58..20d66e5 100644
--- a/arch/arm/mach-davinci/gpio.c
+++ b/arch/arm/mach-davinci/gpio.c
@@ -205,20 +205,20 @@
* serve as EDMA event triggers.
*/
-static void gpio_irq_disable(unsigned irq)
+static void gpio_irq_disable(struct irq_data *d)
{
- struct davinci_gpio_regs __iomem *g = irq2regs(irq);
- u32 mask = (u32) get_irq_data(irq);
+ struct davinci_gpio_regs __iomem *g = irq2regs(d->irq);
+ u32 mask = (u32) irq_data_get_irq_data(d);
__raw_writel(mask, &g->clr_falling);
__raw_writel(mask, &g->clr_rising);
}
-static void gpio_irq_enable(unsigned irq)
+static void gpio_irq_enable(struct irq_data *d)
{
- struct davinci_gpio_regs __iomem *g = irq2regs(irq);
- u32 mask = (u32) get_irq_data(irq);
- unsigned status = irq_desc[irq].status;
+ struct davinci_gpio_regs __iomem *g = irq2regs(d->irq);
+ u32 mask = (u32) irq_data_get_irq_data(d);
+ unsigned status = irq_desc[d->irq].status;
status &= IRQ_TYPE_EDGE_FALLING | IRQ_TYPE_EDGE_RISING;
if (!status)
@@ -230,19 +230,19 @@
__raw_writel(mask, &g->set_rising);
}
-static int gpio_irq_type(unsigned irq, unsigned trigger)
+static int gpio_irq_type(struct irq_data *d, unsigned trigger)
{
- struct davinci_gpio_regs __iomem *g = irq2regs(irq);
- u32 mask = (u32) get_irq_data(irq);
+ struct davinci_gpio_regs __iomem *g = irq2regs(d->irq);
+ u32 mask = (u32) irq_data_get_irq_data(d);
if (trigger & ~(IRQ_TYPE_EDGE_FALLING | IRQ_TYPE_EDGE_RISING))
return -EINVAL;
- irq_desc[irq].status &= ~IRQ_TYPE_SENSE_MASK;
- irq_desc[irq].status |= trigger;
+ irq_desc[d->irq].status &= ~IRQ_TYPE_SENSE_MASK;
+ irq_desc[d->irq].status |= trigger;
/* don't enable the IRQ if it's currently disabled */
- if (irq_desc[irq].depth == 0) {
+ if (irq_desc[d->irq].depth == 0) {
__raw_writel(mask, (trigger & IRQ_TYPE_EDGE_FALLING)
? &g->set_falling : &g->clr_falling);
__raw_writel(mask, (trigger & IRQ_TYPE_EDGE_RISING)
@@ -253,9 +253,9 @@
static struct irq_chip gpio_irqchip = {
.name = "GPIO",
- .enable = gpio_irq_enable,
- .disable = gpio_irq_disable,
- .set_type = gpio_irq_type,
+ .irq_enable = gpio_irq_enable,
+ .irq_disable = gpio_irq_disable,
+ .irq_set_type = gpio_irq_type,
};
static void
@@ -269,8 +269,8 @@
mask <<= 16;
/* temporarily mask (level sensitive) parent IRQ */
- desc->chip->mask(irq);
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
while (1) {
u32 status;
int n;
@@ -293,7 +293,7 @@
status >>= res;
}
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
/* now it may re-trigger */
}
@@ -320,10 +320,10 @@
return -ENODEV;
}
-static int gpio_irq_type_unbanked(unsigned irq, unsigned trigger)
+static int gpio_irq_type_unbanked(struct irq_data *d, unsigned trigger)
{
- struct davinci_gpio_regs __iomem *g = irq2regs(irq);
- u32 mask = (u32) get_irq_data(irq);
+ struct davinci_gpio_regs __iomem *g = irq2regs(d->irq);
+ u32 mask = (u32) irq_data_get_irq_data(d);
if (trigger & ~(IRQ_TYPE_EDGE_FALLING | IRQ_TYPE_EDGE_RISING))
return -EINVAL;
@@ -397,7 +397,7 @@
irq = bank_irq;
gpio_irqchip_unbanked = *get_irq_desc_chip(irq_to_desc(irq));
gpio_irqchip_unbanked.name = "GPIO-AINTC";
- gpio_irqchip_unbanked.set_type = gpio_irq_type_unbanked;
+ gpio_irqchip_unbanked.irq_set_type = gpio_irq_type_unbanked;
/* default trigger: both edges */
g = gpio2regs(0);
diff --git a/arch/arm/mach-davinci/irq.c b/arch/arm/mach-davinci/irq.c
index 784ddf3..5e05c9b 100644
--- a/arch/arm/mach-davinci/irq.c
+++ b/arch/arm/mach-davinci/irq.c
@@ -53,14 +53,14 @@
}
/* Disable interrupt */
-static void davinci_mask_irq(unsigned int irq)
+static void davinci_mask_irq(struct irq_data *d)
{
unsigned int mask;
u32 l;
- mask = 1 << IRQ_BIT(irq);
+ mask = 1 << IRQ_BIT(d->irq);
- if (irq > 31) {
+ if (d->irq > 31) {
l = davinci_irq_readl(IRQ_ENT_REG1_OFFSET);
l &= ~mask;
davinci_irq_writel(l, IRQ_ENT_REG1_OFFSET);
@@ -72,14 +72,14 @@
}
/* Enable interrupt */
-static void davinci_unmask_irq(unsigned int irq)
+static void davinci_unmask_irq(struct irq_data *d)
{
unsigned int mask;
u32 l;
- mask = 1 << IRQ_BIT(irq);
+ mask = 1 << IRQ_BIT(d->irq);
- if (irq > 31) {
+ if (d->irq > 31) {
l = davinci_irq_readl(IRQ_ENT_REG1_OFFSET);
l |= mask;
davinci_irq_writel(l, IRQ_ENT_REG1_OFFSET);
@@ -91,23 +91,23 @@
}
/* EOI interrupt */
-static void davinci_ack_irq(unsigned int irq)
+static void davinci_ack_irq(struct irq_data *d)
{
unsigned int mask;
- mask = 1 << IRQ_BIT(irq);
+ mask = 1 << IRQ_BIT(d->irq);
- if (irq > 31)
+ if (d->irq > 31)
davinci_irq_writel(mask, IRQ_REG1_OFFSET);
else
davinci_irq_writel(mask, IRQ_REG0_OFFSET);
}
static struct irq_chip davinci_irq_chip_0 = {
- .name = "AINTC",
- .ack = davinci_ack_irq,
- .mask = davinci_mask_irq,
- .unmask = davinci_unmask_irq,
+ .name = "AINTC",
+ .irq_ack = davinci_ack_irq,
+ .irq_mask = davinci_mask_irq,
+ .irq_unmask = davinci_unmask_irq,
};
/* ARM Interrupt Controller Initialization */
diff --git a/arch/arm/mach-dove/irq.c b/arch/arm/mach-dove/irq.c
index 61bfcb3..9317f05 100644
--- a/arch/arm/mach-dove/irq.c
+++ b/arch/arm/mach-dove/irq.c
@@ -36,9 +36,9 @@
}
}
-static void pmu_irq_mask(unsigned int irq)
+static void pmu_irq_mask(struct irq_data *d)
{
- int pin = irq_to_pmu(irq);
+ int pin = irq_to_pmu(d->irq);
u32 u;
u = readl(PMU_INTERRUPT_MASK);
@@ -46,9 +46,9 @@
writel(u, PMU_INTERRUPT_MASK);
}
-static void pmu_irq_unmask(unsigned int irq)
+static void pmu_irq_unmask(struct irq_data *d)
{
- int pin = irq_to_pmu(irq);
+ int pin = irq_to_pmu(d->irq);
u32 u;
u = readl(PMU_INTERRUPT_MASK);
@@ -56,9 +56,9 @@
writel(u, PMU_INTERRUPT_MASK);
}
-static void pmu_irq_ack(unsigned int irq)
+static void pmu_irq_ack(struct irq_data *d)
{
- int pin = irq_to_pmu(irq);
+ int pin = irq_to_pmu(d->irq);
u32 u;
u = ~(1 << (pin & 31));
@@ -67,9 +67,9 @@
static struct irq_chip pmu_irq_chip = {
.name = "pmu_irq",
- .mask = pmu_irq_mask,
- .unmask = pmu_irq_unmask,
- .ack = pmu_irq_ack,
+ .irq_mask = pmu_irq_mask,
+ .irq_unmask = pmu_irq_unmask,
+ .irq_ack = pmu_irq_ack,
};
static void pmu_irq_handler(unsigned int irq, struct irq_desc *desc)
diff --git a/arch/arm/mach-ebsa110/core.c b/arch/arm/mach-ebsa110/core.c
index 5df4099..7df083f 100644
--- a/arch/arm/mach-ebsa110/core.c
+++ b/arch/arm/mach-ebsa110/core.c
@@ -35,20 +35,20 @@
#define IRQ_STAT 0xff000000 /* read */
#define IRQ_MCLR 0xff000000 /* write */
-static void ebsa110_mask_irq(unsigned int irq)
+static void ebsa110_mask_irq(struct irq_data *d)
{
- __raw_writeb(1 << irq, IRQ_MCLR);
+ __raw_writeb(1 << d->irq, IRQ_MCLR);
}
-static void ebsa110_unmask_irq(unsigned int irq)
+static void ebsa110_unmask_irq(struct irq_data *d)
{
- __raw_writeb(1 << irq, IRQ_MSET);
+ __raw_writeb(1 << d->irq, IRQ_MSET);
}
static struct irq_chip ebsa110_irq_chip = {
- .ack = ebsa110_mask_irq,
- .mask = ebsa110_mask_irq,
- .unmask = ebsa110_unmask_irq,
+ .irq_ack = ebsa110_mask_irq,
+ .irq_mask = ebsa110_mask_irq,
+ .irq_unmask = ebsa110_unmask_irq,
};
static void __init ebsa110_init_irq(void)
diff --git a/arch/arm/mach-ep93xx/gpio.c b/arch/arm/mach-ep93xx/gpio.c
index cf547ad..f3dc76f 100644
--- a/arch/arm/mach-ep93xx/gpio.c
+++ b/arch/arm/mach-ep93xx/gpio.c
@@ -112,13 +112,13 @@
generic_handle_irq(gpio_irq);
}
-static void ep93xx_gpio_irq_ack(unsigned int irq)
+static void ep93xx_gpio_irq_ack(struct irq_data *d)
{
- int line = irq_to_gpio(irq);
+ int line = irq_to_gpio(d->irq);
int port = line >> 3;
int port_mask = 1 << (line & 7);
- if ((irq_desc[irq].status & IRQ_TYPE_SENSE_MASK) == IRQ_TYPE_EDGE_BOTH) {
+ if ((irq_desc[d->irq].status & IRQ_TYPE_SENSE_MASK) == IRQ_TYPE_EDGE_BOTH) {
gpio_int_type2[port] ^= port_mask; /* switch edge direction */
ep93xx_gpio_update_int_params(port);
}
@@ -126,13 +126,13 @@
__raw_writeb(port_mask, EP93XX_GPIO_REG(eoi_register_offset[port]));
}
-static void ep93xx_gpio_irq_mask_ack(unsigned int irq)
+static void ep93xx_gpio_irq_mask_ack(struct irq_data *d)
{
- int line = irq_to_gpio(irq);
+ int line = irq_to_gpio(d->irq);
int port = line >> 3;
int port_mask = 1 << (line & 7);
- if ((irq_desc[irq].status & IRQ_TYPE_SENSE_MASK) == IRQ_TYPE_EDGE_BOTH)
+ if ((irq_desc[d->irq].status & IRQ_TYPE_SENSE_MASK) == IRQ_TYPE_EDGE_BOTH)
gpio_int_type2[port] ^= port_mask; /* switch edge direction */
gpio_int_unmasked[port] &= ~port_mask;
@@ -141,18 +141,18 @@
__raw_writeb(port_mask, EP93XX_GPIO_REG(eoi_register_offset[port]));
}
-static void ep93xx_gpio_irq_mask(unsigned int irq)
+static void ep93xx_gpio_irq_mask(struct irq_data *d)
{
- int line = irq_to_gpio(irq);
+ int line = irq_to_gpio(d->irq);
int port = line >> 3;
gpio_int_unmasked[port] &= ~(1 << (line & 7));
ep93xx_gpio_update_int_params(port);
}
-static void ep93xx_gpio_irq_unmask(unsigned int irq)
+static void ep93xx_gpio_irq_unmask(struct irq_data *d)
{
- int line = irq_to_gpio(irq);
+ int line = irq_to_gpio(d->irq);
int port = line >> 3;
gpio_int_unmasked[port] |= 1 << (line & 7);
@@ -164,10 +164,10 @@
* edge (1) triggered, while gpio_int_type2 controls whether it
* triggers on low/falling (0) or high/rising (1).
*/
-static int ep93xx_gpio_irq_type(unsigned int irq, unsigned int type)
+static int ep93xx_gpio_irq_type(struct irq_data *d, unsigned int type)
{
- struct irq_desc *desc = irq_desc + irq;
- const int gpio = irq_to_gpio(irq);
+ struct irq_desc *desc = irq_desc + d->irq;
+ const int gpio = irq_to_gpio(d->irq);
const int port = gpio >> 3;
const int port_mask = 1 << (gpio & 7);
@@ -220,11 +220,11 @@
static struct irq_chip ep93xx_gpio_irq_chip = {
.name = "GPIO",
- .ack = ep93xx_gpio_irq_ack,
- .mask_ack = ep93xx_gpio_irq_mask_ack,
- .mask = ep93xx_gpio_irq_mask,
- .unmask = ep93xx_gpio_irq_unmask,
- .set_type = ep93xx_gpio_irq_type,
+ .irq_ack = ep93xx_gpio_irq_ack,
+ .irq_mask_ack = ep93xx_gpio_irq_mask_ack,
+ .irq_mask = ep93xx_gpio_irq_mask,
+ .irq_unmask = ep93xx_gpio_irq_unmask,
+ .irq_set_type = ep93xx_gpio_irq_type,
};
void __init ep93xx_gpio_init_irq(void)
diff --git a/arch/arm/mach-footbridge/common.c b/arch/arm/mach-footbridge/common.c
index 88b3dd8..84c5f25 100644
--- a/arch/arm/mach-footbridge/common.c
+++ b/arch/arm/mach-footbridge/common.c
@@ -75,20 +75,20 @@
IRQ_MASK_PCI_PERR, /* 19 */
};
-static void fb_mask_irq(unsigned int irq)
+static void fb_mask_irq(struct irq_data *d)
{
- *CSR_IRQ_DISABLE = fb_irq_mask[_DC21285_INR(irq)];
+ *CSR_IRQ_DISABLE = fb_irq_mask[_DC21285_INR(d->irq)];
}
-static void fb_unmask_irq(unsigned int irq)
+static void fb_unmask_irq(struct irq_data *d)
{
- *CSR_IRQ_ENABLE = fb_irq_mask[_DC21285_INR(irq)];
+ *CSR_IRQ_ENABLE = fb_irq_mask[_DC21285_INR(d->irq)];
}
static struct irq_chip fb_chip = {
- .ack = fb_mask_irq,
- .mask = fb_mask_irq,
- .unmask = fb_unmask_irq,
+ .irq_ack = fb_mask_irq,
+ .irq_mask = fb_mask_irq,
+ .irq_unmask = fb_unmask_irq,
};
static void __init __fb_init_irq(void)
diff --git a/arch/arm/mach-footbridge/isa-irq.c b/arch/arm/mach-footbridge/isa-irq.c
index 8bfd06a..de7a5cb 100644
--- a/arch/arm/mach-footbridge/isa-irq.c
+++ b/arch/arm/mach-footbridge/isa-irq.c
@@ -30,61 +30,61 @@
#include "common.h"
-static void isa_mask_pic_lo_irq(unsigned int irq)
+static void isa_mask_pic_lo_irq(struct irq_data *d)
{
- unsigned int mask = 1 << (irq & 7);
+ unsigned int mask = 1 << (d->irq & 7);
outb(inb(PIC_MASK_LO) | mask, PIC_MASK_LO);
}
-static void isa_ack_pic_lo_irq(unsigned int irq)
+static void isa_ack_pic_lo_irq(struct irq_data *d)
{
- unsigned int mask = 1 << (irq & 7);
+ unsigned int mask = 1 << (d->irq & 7);
outb(inb(PIC_MASK_LO) | mask, PIC_MASK_LO);
outb(0x20, PIC_LO);
}
-static void isa_unmask_pic_lo_irq(unsigned int irq)
+static void isa_unmask_pic_lo_irq(struct irq_data *d)
{
- unsigned int mask = 1 << (irq & 7);
+ unsigned int mask = 1 << (d->irq & 7);
outb(inb(PIC_MASK_LO) & ~mask, PIC_MASK_LO);
}
static struct irq_chip isa_lo_chip = {
- .ack = isa_ack_pic_lo_irq,
- .mask = isa_mask_pic_lo_irq,
- .unmask = isa_unmask_pic_lo_irq,
+ .irq_ack = isa_ack_pic_lo_irq,
+ .irq_mask = isa_mask_pic_lo_irq,
+ .irq_unmask = isa_unmask_pic_lo_irq,
};
-static void isa_mask_pic_hi_irq(unsigned int irq)
+static void isa_mask_pic_hi_irq(struct irq_data *d)
{
- unsigned int mask = 1 << (irq & 7);
+ unsigned int mask = 1 << (d->irq & 7);
outb(inb(PIC_MASK_HI) | mask, PIC_MASK_HI);
}
-static void isa_ack_pic_hi_irq(unsigned int irq)
+static void isa_ack_pic_hi_irq(struct irq_data *d)
{
- unsigned int mask = 1 << (irq & 7);
+ unsigned int mask = 1 << (d->irq & 7);
outb(inb(PIC_MASK_HI) | mask, PIC_MASK_HI);
outb(0x62, PIC_LO);
outb(0x20, PIC_HI);
}
-static void isa_unmask_pic_hi_irq(unsigned int irq)
+static void isa_unmask_pic_hi_irq(struct irq_data *d)
{
- unsigned int mask = 1 << (irq & 7);
+ unsigned int mask = 1 << (d->irq & 7);
outb(inb(PIC_MASK_HI) & ~mask, PIC_MASK_HI);
}
static struct irq_chip isa_hi_chip = {
- .ack = isa_ack_pic_hi_irq,
- .mask = isa_mask_pic_hi_irq,
- .unmask = isa_unmask_pic_hi_irq,
+ .irq_ack = isa_ack_pic_hi_irq,
+ .irq_mask = isa_mask_pic_hi_irq,
+ .irq_unmask = isa_unmask_pic_hi_irq,
};
static void
diff --git a/arch/arm/mach-gemini/gpio.c b/arch/arm/mach-gemini/gpio.c
index fe3bd5a..fa3d333 100644
--- a/arch/arm/mach-gemini/gpio.c
+++ b/arch/arm/mach-gemini/gpio.c
@@ -54,33 +54,33 @@
__raw_writel(reg, base + GPIO_INT_EN);
}
-static void gpio_ack_irq(unsigned int irq)
+static void gpio_ack_irq(struct irq_data *d)
{
- unsigned int gpio = irq_to_gpio(irq);
+ unsigned int gpio = irq_to_gpio(d->irq);
unsigned int base = GPIO_BASE(gpio / 32);
__raw_writel(1 << (gpio % 32), base + GPIO_INT_CLR);
}
-static void gpio_mask_irq(unsigned int irq)
+static void gpio_mask_irq(struct irq_data *d)
{
- unsigned int gpio = irq_to_gpio(irq);
+ unsigned int gpio = irq_to_gpio(d->irq);
unsigned int base = GPIO_BASE(gpio / 32);
_set_gpio_irqenable(base, gpio % 32, 0);
}
-static void gpio_unmask_irq(unsigned int irq)
+static void gpio_unmask_irq(struct irq_data *d)
{
- unsigned int gpio = irq_to_gpio(irq);
+ unsigned int gpio = irq_to_gpio(d->irq);
unsigned int base = GPIO_BASE(gpio / 32);
_set_gpio_irqenable(base, gpio % 32, 1);
}
-static int gpio_set_irq_type(unsigned int irq, unsigned int type)
+static int gpio_set_irq_type(struct irq_data *d, unsigned int type)
{
- unsigned int gpio = irq_to_gpio(irq);
+ unsigned int gpio = irq_to_gpio(d->irq);
unsigned int gpio_mask = 1 << (gpio % 32);
unsigned int base = GPIO_BASE(gpio / 32);
unsigned int reg_both, reg_level, reg_type;
@@ -120,7 +120,7 @@
__raw_writel(reg_level, base + GPIO_INT_LEVEL);
__raw_writel(reg_both, base + GPIO_INT_BOTH_EDGE);
- gpio_ack_irq(irq);
+ gpio_ack_irq(d->irq);
return 0;
}
@@ -146,10 +146,10 @@
static struct irq_chip gpio_irq_chip = {
.name = "GPIO",
- .ack = gpio_ack_irq,
- .mask = gpio_mask_irq,
- .unmask = gpio_unmask_irq,
- .set_type = gpio_set_irq_type,
+ .irq_ack = gpio_ack_irq,
+ .irq_mask = gpio_mask_irq,
+ .irq_unmask = gpio_unmask_irq,
+ .irq_set_type = gpio_set_irq_type,
};
static void _set_gpio_direction(struct gpio_chip *chip, unsigned offset,
diff --git a/arch/arm/mach-gemini/irq.c b/arch/arm/mach-gemini/irq.c
index 9e613ca..96bc227 100644
--- a/arch/arm/mach-gemini/irq.c
+++ b/arch/arm/mach-gemini/irq.c
@@ -32,34 +32,34 @@
#define FIQ_LEVEL(base_addr) (base_addr + 0x30)
#define FIQ_STATUS(base_addr) (base_addr + 0x34)
-static void gemini_ack_irq(unsigned int irq)
+static void gemini_ack_irq(struct irq_data *d)
{
- __raw_writel(1 << irq, IRQ_CLEAR(IO_ADDRESS(GEMINI_INTERRUPT_BASE)));
+ __raw_writel(1 << d->irq, IRQ_CLEAR(IO_ADDRESS(GEMINI_INTERRUPT_BASE)));
}
-static void gemini_mask_irq(unsigned int irq)
+static void gemini_mask_irq(struct irq_data *d)
{
unsigned int mask;
mask = __raw_readl(IRQ_MASK(IO_ADDRESS(GEMINI_INTERRUPT_BASE)));
- mask &= ~(1 << irq);
+ mask &= ~(1 << d->irq);
__raw_writel(mask, IRQ_MASK(IO_ADDRESS(GEMINI_INTERRUPT_BASE)));
}
-static void gemini_unmask_irq(unsigned int irq)
+static void gemini_unmask_irq(struct irq_data *d)
{
unsigned int mask;
mask = __raw_readl(IRQ_MASK(IO_ADDRESS(GEMINI_INTERRUPT_BASE)));
- mask |= (1 << irq);
+ mask |= (1 << d->irq);
__raw_writel(mask, IRQ_MASK(IO_ADDRESS(GEMINI_INTERRUPT_BASE)));
}
static struct irq_chip gemini_irq_chip = {
- .name = "INTC",
- .ack = gemini_ack_irq,
- .mask = gemini_mask_irq,
- .unmask = gemini_unmask_irq,
+ .name = "INTC",
+ .irq_ack = gemini_ack_irq,
+ .irq_mask = gemini_mask_irq,
+ .irq_unmask = gemini_unmask_irq,
};
static struct resource irq_resource = {
diff --git a/arch/arm/mach-h720x/common.c b/arch/arm/mach-h720x/common.c
index bdb3f67..1f28c90 100644
--- a/arch/arm/mach-h720x/common.c
+++ b/arch/arm/mach-h720x/common.c
@@ -52,17 +52,17 @@
/*
* mask Global irq's
*/
-static void mask_global_irq (unsigned int irq )
+static void mask_global_irq(struct irq_data *d)
{
- CPU_REG (IRQC_VIRT, IRQC_IER) &= ~(1 << irq);
+ CPU_REG (IRQC_VIRT, IRQC_IER) &= ~(1 << d->irq);
}
/*
* unmask Global irq's
*/
-static void unmask_global_irq (unsigned int irq )
+static void unmask_global_irq(struct irq_data *d)
{
- CPU_REG (IRQC_VIRT, IRQC_IER) |= (1 << irq);
+ CPU_REG (IRQC_VIRT, IRQC_IER) |= (1 << d->irq);
}
@@ -70,10 +70,10 @@
* ack GPIO irq's
* Ack only for edge triggered int's valid
*/
-static void inline ack_gpio_irq(u32 irq)
+static void inline ack_gpio_irq(struct irq_data *d)
{
- u32 reg_base = GPIO_VIRT(IRQ_TO_REGNO(irq));
- u32 bit = IRQ_TO_BIT(irq);
+ u32 reg_base = GPIO_VIRT(IRQ_TO_REGNO(d->irq));
+ u32 bit = IRQ_TO_BIT(d->irq);
if ( (CPU_REG (reg_base, GPIO_EDGE) & bit))
CPU_REG (reg_base, GPIO_CLR) = bit;
}
@@ -81,20 +81,20 @@
/*
* mask GPIO irq's
*/
-static void inline mask_gpio_irq(u32 irq)
+static void inline mask_gpio_irq(struct irq_data *d)
{
- u32 reg_base = GPIO_VIRT(IRQ_TO_REGNO(irq));
- u32 bit = IRQ_TO_BIT(irq);
+ u32 reg_base = GPIO_VIRT(IRQ_TO_REGNO(d->irq));
+ u32 bit = IRQ_TO_BIT(d->irq);
CPU_REG (reg_base, GPIO_MASK) &= ~bit;
}
/*
* unmask GPIO irq's
*/
-static void inline unmask_gpio_irq(u32 irq)
+static void inline unmask_gpio_irq(struct irq_data *d)
{
- u32 reg_base = GPIO_VIRT(IRQ_TO_REGNO(irq));
- u32 bit = IRQ_TO_BIT(irq);
+ u32 reg_base = GPIO_VIRT(IRQ_TO_REGNO(d->irq));
+ u32 bit = IRQ_TO_BIT(d->irq);
CPU_REG (reg_base, GPIO_MASK) |= bit;
}
@@ -170,15 +170,15 @@
#endif
static struct irq_chip h720x_global_chip = {
- .ack = mask_global_irq,
- .mask = mask_global_irq,
- .unmask = unmask_global_irq,
+ .irq_ack = mask_global_irq,
+ .irq_mask = mask_global_irq,
+ .irq_unmask = unmask_global_irq,
};
static struct irq_chip h720x_gpio_chip = {
- .ack = ack_gpio_irq,
- .mask = mask_gpio_irq,
- .unmask = unmask_gpio_irq,
+ .irq_ack = ack_gpio_irq,
+ .irq_mask = mask_gpio_irq,
+ .irq_unmask = unmask_gpio_irq,
};
/*
diff --git a/arch/arm/mach-h720x/cpu-h7202.c b/arch/arm/mach-h720x/cpu-h7202.c
index fd33a19..ac3f914 100644
--- a/arch/arm/mach-h720x/cpu-h7202.c
+++ b/arch/arm/mach-h720x/cpu-h7202.c
@@ -141,27 +141,27 @@
/*
* mask multiplexed timer IRQs
*/
-static void inline mask_timerx_irq (u32 irq)
+static void inline mask_timerx_irq(struct irq_data *d)
{
unsigned int bit;
- bit = 2 << ((irq == IRQ_TIMER64B) ? 4 : (irq - IRQ_TIMER1));
+ bit = 2 << ((d->irq == IRQ_TIMER64B) ? 4 : (d->irq - IRQ_TIMER1));
CPU_REG (TIMER_VIRT, TIMER_TOPCTRL) &= ~bit;
}
/*
* unmask multiplexed timer IRQs
*/
-static void inline unmask_timerx_irq (u32 irq)
+static void inline unmask_timerx_irq(struct irq_data *d)
{
unsigned int bit;
- bit = 2 << ((irq == IRQ_TIMER64B) ? 4 : (irq - IRQ_TIMER1));
+ bit = 2 << ((d->irq == IRQ_TIMER64B) ? 4 : (d->irq - IRQ_TIMER1));
CPU_REG (TIMER_VIRT, TIMER_TOPCTRL) |= bit;
}
static struct irq_chip h7202_timerx_chip = {
- .ack = mask_timerx_irq,
- .mask = mask_timerx_irq,
- .unmask = unmask_timerx_irq,
+ .irq_ack = mask_timerx_irq,
+ .irq_mask = mask_timerx_irq,
+ .irq_unmask = unmask_timerx_irq,
};
static struct irqaction h7202_timer_irq = {
diff --git a/arch/arm/mach-h720x/h7201-eval.c b/arch/arm/mach-h720x/h7201-eval.c
index 79f0b89..629454d 100644
--- a/arch/arm/mach-h720x/h7201-eval.c
+++ b/arch/arm/mach-h720x/h7201-eval.c
@@ -23,7 +23,6 @@
#include <asm/types.h>
#include <asm/mach-types.h>
#include <asm/page.h>
-#include <asm/pgtable.h>
#include <asm/mach/arch.h>
#include <mach/hardware.h>
#include "common.h"
diff --git a/arch/arm/mach-h720x/h7202-eval.c b/arch/arm/mach-h720x/h7202-eval.c
index cc28b1e..e9f46b6 100644
--- a/arch/arm/mach-h720x/h7202-eval.c
+++ b/arch/arm/mach-h720x/h7202-eval.c
@@ -23,7 +23,6 @@
#include <asm/types.h>
#include <asm/mach-types.h>
#include <asm/page.h>
-#include <asm/pgtable.h>
#include <asm/mach/arch.h>
#include <mach/irqs.h>
#include <mach/hardware.h>
diff --git a/arch/arm/mach-imx/Kconfig b/arch/arm/mach-imx/Kconfig
index 17d2e60..56684b5 100644
--- a/arch/arm/mach-imx/Kconfig
+++ b/arch/arm/mach-imx/Kconfig
@@ -243,6 +243,7 @@
select IMX_HAVE_PLATFORM_MXC_EHCI
select IMX_HAVE_PLATFORM_MXC_MMC
select IMX_HAVE_PLATFORM_SPI_IMX
+ select MXC_DEBUG_BOARD
select MXC_ULPI if USB_ULPI
help
Include support for MX27PDK platform. This includes specific
diff --git a/arch/arm/mach-imx/mach-mx27_3ds.c b/arch/arm/mach-imx/mach-mx27_3ds.c
index 6fd0f8f..1643315 100644
--- a/arch/arm/mach-imx/mach-mx27_3ds.c
+++ b/arch/arm/mach-imx/mach-mx27_3ds.c
@@ -37,12 +37,15 @@
#include <mach/common.h>
#include <mach/iomux-mx27.h>
#include <mach/ulpi.h>
+#include <mach/irqs.h>
+#include <mach/3ds_debugboard.h>
#include "devices-imx27.h"
#define SD1_EN_GPIO (GPIO_PORTB + 25)
#define OTG_PHY_RESET_GPIO (GPIO_PORTB + 23)
#define SPI2_SS0 (GPIO_PORTD + 21)
+#define EXPIO_PARENT_INT (MXC_INTERNAL_IRQS + GPIO_PORTC + 28)
static const int mx27pdk_pins[] __initconst = {
/* UART1 */
@@ -215,10 +218,10 @@
static struct mc13783_regulator_init_data mx27_3ds_regulators[] = {
{
- .id = MC13783_REGU_VMMC1,
+ .id = MC13783_REG_VMMC1,
.init_data = &vmmc1_init,
}, {
- .id = MC13783_REGU_VGEN,
+ .id = MC13783_REG_VGEN,
.init_data = &vgen_init,
},
};
@@ -276,6 +279,9 @@
imx27_add_spi_imx1(&spi2_pdata);
spi_register_board_info(mx27_3ds_spi_devs,
ARRAY_SIZE(mx27_3ds_spi_devs));
+
+ if (mxc_expio_init(MX27_CS5_BASE_ADDR, EXPIO_PARENT_INT))
+ pr_warn("Init of the debugboard failed, all devices on the debugboard are unusable.\n");
}
static void __init mx27pdk_timer_init(void)
diff --git a/arch/arm/mach-integrator/cpu.c b/arch/arm/mach-integrator/cpu.c
index a3fbcb3a..fbb4577 100644
--- a/arch/arm/mach-integrator/cpu.c
+++ b/arch/arm/mach-integrator/cpu.c
@@ -173,7 +173,7 @@
if (machine_is_integrator()) {
vco.s = (cm_osc >> 8) & 7;
- } else if (machine_is_cintegrator()) {
+ } else {
vco.s = 1;
}
vco.v = cm_osc & 255;
diff --git a/arch/arm/mach-integrator/integrator_ap.c b/arch/arm/mach-integrator/integrator_ap.c
index 2774df8..b666443 100644
--- a/arch/arm/mach-integrator/integrator_ap.c
+++ b/arch/arm/mach-integrator/integrator_ap.c
@@ -156,21 +156,21 @@
#define INTEGRATOR_SC_VALID_INT 0x003fffff
-static void sc_mask_irq(unsigned int irq)
+static void sc_mask_irq(struct irq_data *d)
{
- writel(1 << irq, VA_IC_BASE + IRQ_ENABLE_CLEAR);
+ writel(1 << d->irq, VA_IC_BASE + IRQ_ENABLE_CLEAR);
}
-static void sc_unmask_irq(unsigned int irq)
+static void sc_unmask_irq(struct irq_data *d)
{
- writel(1 << irq, VA_IC_BASE + IRQ_ENABLE_SET);
+ writel(1 << d->irq, VA_IC_BASE + IRQ_ENABLE_SET);
}
static struct irq_chip sc_chip = {
- .name = "SC",
- .ack = sc_mask_irq,
- .mask = sc_mask_irq,
- .unmask = sc_unmask_irq,
+ .name = "SC",
+ .irq_ack = sc_mask_irq,
+ .irq_mask = sc_mask_irq,
+ .irq_unmask = sc_unmask_irq,
};
static void __init ap_init_irq(void)
diff --git a/arch/arm/mach-integrator/integrator_cp.c b/arch/arm/mach-integrator/integrator_cp.c
index 85e48a5..e9327da 100644
--- a/arch/arm/mach-integrator/integrator_cp.c
+++ b/arch/arm/mach-integrator/integrator_cp.c
@@ -146,61 +146,61 @@
#define sic_writel __raw_writel
#define sic_readl __raw_readl
-static void cic_mask_irq(unsigned int irq)
+static void cic_mask_irq(struct irq_data *d)
{
- irq -= IRQ_CIC_START;
+ unsigned int irq = d->irq - IRQ_CIC_START;
cic_writel(1 << irq, INTCP_VA_CIC_BASE + IRQ_ENABLE_CLEAR);
}
-static void cic_unmask_irq(unsigned int irq)
+static void cic_unmask_irq(struct irq_data *d)
{
- irq -= IRQ_CIC_START;
+ unsigned int irq = d->irq - IRQ_CIC_START;
cic_writel(1 << irq, INTCP_VA_CIC_BASE + IRQ_ENABLE_SET);
}
static struct irq_chip cic_chip = {
- .name = "CIC",
- .ack = cic_mask_irq,
- .mask = cic_mask_irq,
- .unmask = cic_unmask_irq,
+ .name = "CIC",
+ .irq_ack = cic_mask_irq,
+ .irq_mask = cic_mask_irq,
+ .irq_unmask = cic_unmask_irq,
};
-static void pic_mask_irq(unsigned int irq)
+static void pic_mask_irq(struct irq_data *d)
{
- irq -= IRQ_PIC_START;
+ unsigned int irq = d->irq - IRQ_PIC_START;
pic_writel(1 << irq, INTCP_VA_PIC_BASE + IRQ_ENABLE_CLEAR);
}
-static void pic_unmask_irq(unsigned int irq)
+static void pic_unmask_irq(struct irq_data *d)
{
- irq -= IRQ_PIC_START;
+ unsigned int irq = d->irq - IRQ_PIC_START;
pic_writel(1 << irq, INTCP_VA_PIC_BASE + IRQ_ENABLE_SET);
}
static struct irq_chip pic_chip = {
- .name = "PIC",
- .ack = pic_mask_irq,
- .mask = pic_mask_irq,
- .unmask = pic_unmask_irq,
+ .name = "PIC",
+ .irq_ack = pic_mask_irq,
+ .irq_mask = pic_mask_irq,
+ .irq_unmask = pic_unmask_irq,
};
-static void sic_mask_irq(unsigned int irq)
+static void sic_mask_irq(struct irq_data *d)
{
- irq -= IRQ_SIC_START;
+ unsigned int irq = d->irq - IRQ_SIC_START;
sic_writel(1 << irq, INTCP_VA_SIC_BASE + IRQ_ENABLE_CLEAR);
}
-static void sic_unmask_irq(unsigned int irq)
+static void sic_unmask_irq(struct irq_data *d)
{
- irq -= IRQ_SIC_START;
+ unsigned int irq = d->irq - IRQ_SIC_START;
sic_writel(1 << irq, INTCP_VA_SIC_BASE + IRQ_ENABLE_SET);
}
static struct irq_chip sic_chip = {
- .name = "SIC",
- .ack = sic_mask_irq,
- .mask = sic_mask_irq,
- .unmask = sic_unmask_irq,
+ .name = "SIC",
+ .irq_ack = sic_mask_irq,
+ .irq_mask = sic_mask_irq,
+ .irq_unmask = sic_unmask_irq,
};
static void
diff --git a/arch/arm/mach-iop13xx/irq.c b/arch/arm/mach-iop13xx/irq.c
index 0d099ca..a233470 100644
--- a/arch/arm/mach-iop13xx/irq.c
+++ b/arch/arm/mach-iop13xx/irq.c
@@ -123,79 +123,79 @@
/* 0 = Interrupt Masked and 1 = Interrupt not masked */
static void
-iop13xx_irq_mask0 (unsigned int irq)
+iop13xx_irq_mask0 (struct irq_data *d)
{
- write_intctl_0(read_intctl_0() & ~(1 << (irq - 0)));
+ write_intctl_0(read_intctl_0() & ~(1 << (d->irq - 0)));
}
static void
-iop13xx_irq_mask1 (unsigned int irq)
+iop13xx_irq_mask1 (struct irq_data *d)
{
- write_intctl_1(read_intctl_1() & ~(1 << (irq - 32)));
+ write_intctl_1(read_intctl_1() & ~(1 << (d->irq - 32)));
}
static void
-iop13xx_irq_mask2 (unsigned int irq)
+iop13xx_irq_mask2 (struct irq_data *d)
{
- write_intctl_2(read_intctl_2() & ~(1 << (irq - 64)));
+ write_intctl_2(read_intctl_2() & ~(1 << (d->irq - 64)));
}
static void
-iop13xx_irq_mask3 (unsigned int irq)
+iop13xx_irq_mask3 (struct irq_data *d)
{
- write_intctl_3(read_intctl_3() & ~(1 << (irq - 96)));
+ write_intctl_3(read_intctl_3() & ~(1 << (d->irq - 96)));
}
static void
-iop13xx_irq_unmask0(unsigned int irq)
+iop13xx_irq_unmask0(struct irq_data *d)
{
- write_intctl_0(read_intctl_0() | (1 << (irq - 0)));
+ write_intctl_0(read_intctl_0() | (1 << (d->irq - 0)));
}
static void
-iop13xx_irq_unmask1(unsigned int irq)
+iop13xx_irq_unmask1(struct irq_data *d)
{
- write_intctl_1(read_intctl_1() | (1 << (irq - 32)));
+ write_intctl_1(read_intctl_1() | (1 << (d->irq - 32)));
}
static void
-iop13xx_irq_unmask2(unsigned int irq)
+iop13xx_irq_unmask2(struct irq_data *d)
{
- write_intctl_2(read_intctl_2() | (1 << (irq - 64)));
+ write_intctl_2(read_intctl_2() | (1 << (d->irq - 64)));
}
static void
-iop13xx_irq_unmask3(unsigned int irq)
+iop13xx_irq_unmask3(struct irq_data *d)
{
- write_intctl_3(read_intctl_3() | (1 << (irq - 96)));
+ write_intctl_3(read_intctl_3() | (1 << (d->irq - 96)));
}
static struct irq_chip iop13xx_irqchip1 = {
- .name = "IOP13xx-1",
- .ack = iop13xx_irq_mask0,
- .mask = iop13xx_irq_mask0,
- .unmask = iop13xx_irq_unmask0,
+ .name = "IOP13xx-1",
+ .irq_ack = iop13xx_irq_mask0,
+ .irq_mask = iop13xx_irq_mask0,
+ .irq_unmask = iop13xx_irq_unmask0,
};
static struct irq_chip iop13xx_irqchip2 = {
- .name = "IOP13xx-2",
- .ack = iop13xx_irq_mask1,
- .mask = iop13xx_irq_mask1,
- .unmask = iop13xx_irq_unmask1,
+ .name = "IOP13xx-2",
+ .irq_ack = iop13xx_irq_mask1,
+ .irq_mask = iop13xx_irq_mask1,
+ .irq_unmask = iop13xx_irq_unmask1,
};
static struct irq_chip iop13xx_irqchip3 = {
- .name = "IOP13xx-3",
- .ack = iop13xx_irq_mask2,
- .mask = iop13xx_irq_mask2,
- .unmask = iop13xx_irq_unmask2,
+ .name = "IOP13xx-3",
+ .irq_ack = iop13xx_irq_mask2,
+ .irq_mask = iop13xx_irq_mask2,
+ .irq_unmask = iop13xx_irq_unmask2,
};
static struct irq_chip iop13xx_irqchip4 = {
- .name = "IOP13xx-4",
- .ack = iop13xx_irq_mask3,
- .mask = iop13xx_irq_mask3,
- .unmask = iop13xx_irq_unmask3,
+ .name = "IOP13xx-4",
+ .irq_ack = iop13xx_irq_mask3,
+ .irq_mask = iop13xx_irq_mask3,
+ .irq_unmask = iop13xx_irq_unmask3,
};
extern void iop_init_cp6_handler(void);
diff --git a/arch/arm/mach-iop13xx/msi.c b/arch/arm/mach-iop13xx/msi.c
index 7149fcc..c9c02e3 100644
--- a/arch/arm/mach-iop13xx/msi.c
+++ b/arch/arm/mach-iop13xx/msi.c
@@ -156,14 +156,14 @@
destroy_irq(irq);
}
-static void iop13xx_msi_nop(unsigned int irq)
+static void iop13xx_msi_nop(struct irq_data *d)
{
return;
}
static struct irq_chip iop13xx_msi_chip = {
.name = "PCI-MSI",
- .ack = iop13xx_msi_nop,
+ .irq_ack = iop13xx_msi_nop,
.irq_enable = unmask_msi_irq,
.irq_disable = mask_msi_irq,
.irq_mask = mask_msi_irq,
diff --git a/arch/arm/mach-iop32x/irq.c b/arch/arm/mach-iop32x/irq.c
index ba59b2d..d3426a1 100644
--- a/arch/arm/mach-iop32x/irq.c
+++ b/arch/arm/mach-iop32x/irq.c
@@ -32,24 +32,24 @@
}
static void
-iop32x_irq_mask(unsigned int irq)
+iop32x_irq_mask(struct irq_data *d)
{
- iop32x_mask &= ~(1 << irq);
+ iop32x_mask &= ~(1 << d->irq);
intctl_write(iop32x_mask);
}
static void
-iop32x_irq_unmask(unsigned int irq)
+iop32x_irq_unmask(struct irq_data *d)
{
- iop32x_mask |= 1 << irq;
+ iop32x_mask |= 1 << d->irq;
intctl_write(iop32x_mask);
}
struct irq_chip ext_chip = {
- .name = "IOP32x",
- .ack = iop32x_irq_mask,
- .mask = iop32x_irq_mask,
- .unmask = iop32x_irq_unmask,
+ .name = "IOP32x",
+ .irq_ack = iop32x_irq_mask,
+ .irq_mask = iop32x_irq_mask,
+ .irq_unmask = iop32x_irq_unmask,
};
void __init iop32x_init_irq(void)
diff --git a/arch/arm/mach-iop33x/irq.c b/arch/arm/mach-iop33x/irq.c
index abb4ea2..0ff2f74 100644
--- a/arch/arm/mach-iop33x/irq.c
+++ b/arch/arm/mach-iop33x/irq.c
@@ -53,45 +53,45 @@
}
static void
-iop33x_irq_mask1 (unsigned int irq)
+iop33x_irq_mask1 (struct irq_data *d)
{
- iop33x_mask0 &= ~(1 << irq);
+ iop33x_mask0 &= ~(1 << d->irq);
intctl0_write(iop33x_mask0);
}
static void
-iop33x_irq_mask2 (unsigned int irq)
+iop33x_irq_mask2 (struct irq_data *d)
{
- iop33x_mask1 &= ~(1 << (irq - 32));
+ iop33x_mask1 &= ~(1 << (d->irq - 32));
intctl1_write(iop33x_mask1);
}
static void
-iop33x_irq_unmask1(unsigned int irq)
+iop33x_irq_unmask1(struct irq_data *d)
{
- iop33x_mask0 |= 1 << irq;
+ iop33x_mask0 |= 1 << d->irq;
intctl0_write(iop33x_mask0);
}
static void
-iop33x_irq_unmask2(unsigned int irq)
+iop33x_irq_unmask2(struct irq_data *d)
{
- iop33x_mask1 |= (1 << (irq - 32));
+ iop33x_mask1 |= (1 << (d->irq - 32));
intctl1_write(iop33x_mask1);
}
struct irq_chip iop33x_irqchip1 = {
- .name = "IOP33x-1",
- .ack = iop33x_irq_mask1,
- .mask = iop33x_irq_mask1,
- .unmask = iop33x_irq_unmask1,
+ .name = "IOP33x-1",
+ .irq_ack = iop33x_irq_mask1,
+ .irq_mask = iop33x_irq_mask1,
+ .irq_unmask = iop33x_irq_unmask1,
};
struct irq_chip iop33x_irqchip2 = {
- .name = "IOP33x-2",
- .ack = iop33x_irq_mask2,
- .mask = iop33x_irq_mask2,
- .unmask = iop33x_irq_unmask2,
+ .name = "IOP33x-2",
+ .irq_ack = iop33x_irq_mask2,
+ .irq_mask = iop33x_irq_mask2,
+ .irq_unmask = iop33x_irq_unmask2,
};
void __init iop33x_init_irq(void)
diff --git a/arch/arm/mach-ixp2000/core.c b/arch/arm/mach-ixp2000/core.c
index e24e3d0..5fc4e06 100644
--- a/arch/arm/mach-ixp2000/core.c
+++ b/arch/arm/mach-ixp2000/core.c
@@ -309,9 +309,9 @@
}
}
-static int ixp2000_GPIO_irq_type(unsigned int irq, unsigned int type)
+static int ixp2000_GPIO_irq_type(struct irq_data *d, unsigned int type)
{
- int line = irq - IRQ_IXP2000_GPIO0;
+ int line = d->irq - IRQ_IXP2000_GPIO0;
/*
* First, configure this GPIO line as an input.
@@ -342,8 +342,10 @@
return 0;
}
-static void ixp2000_GPIO_irq_mask_ack(unsigned int irq)
+static void ixp2000_GPIO_irq_mask_ack(struct irq_data *d)
{
+ unsigned int irq = d->irq;
+
ixp2000_reg_write(IXP2000_GPIO_INCR, (1 << (irq - IRQ_IXP2000_GPIO0)));
ixp2000_reg_write(IXP2000_GPIO_EDSR, (1 << (irq - IRQ_IXP2000_GPIO0)));
@@ -351,38 +353,42 @@
ixp2000_reg_wrb(IXP2000_GPIO_INST, (1 << (irq - IRQ_IXP2000_GPIO0)));
}
-static void ixp2000_GPIO_irq_mask(unsigned int irq)
+static void ixp2000_GPIO_irq_mask(struct irq_data *d)
{
+ unsigned int irq = d->irq;
+
ixp2000_reg_wrb(IXP2000_GPIO_INCR, (1 << (irq - IRQ_IXP2000_GPIO0)));
}
-static void ixp2000_GPIO_irq_unmask(unsigned int irq)
+static void ixp2000_GPIO_irq_unmask(struct irq_data *d)
{
+ unsigned int irq = d->irq;
+
ixp2000_reg_write(IXP2000_GPIO_INSR, (1 << (irq - IRQ_IXP2000_GPIO0)));
}
static struct irq_chip ixp2000_GPIO_irq_chip = {
- .ack = ixp2000_GPIO_irq_mask_ack,
- .mask = ixp2000_GPIO_irq_mask,
- .unmask = ixp2000_GPIO_irq_unmask,
- .set_type = ixp2000_GPIO_irq_type,
+ .irq_ack = ixp2000_GPIO_irq_mask_ack,
+ .irq_mask = ixp2000_GPIO_irq_mask,
+ .irq_unmask = ixp2000_GPIO_irq_unmask,
+ .irq_set_type = ixp2000_GPIO_irq_type,
};
-static void ixp2000_pci_irq_mask(unsigned int irq)
+static void ixp2000_pci_irq_mask(struct irq_data *d)
{
unsigned long temp = *IXP2000_PCI_XSCALE_INT_ENABLE;
- if (irq == IRQ_IXP2000_PCIA)
+ if (d->irq == IRQ_IXP2000_PCIA)
ixp2000_reg_wrb(IXP2000_PCI_XSCALE_INT_ENABLE, (temp & ~(1 << 26)));
- else if (irq == IRQ_IXP2000_PCIB)
+ else if (d->irq == IRQ_IXP2000_PCIB)
ixp2000_reg_wrb(IXP2000_PCI_XSCALE_INT_ENABLE, (temp & ~(1 << 27)));
}
-static void ixp2000_pci_irq_unmask(unsigned int irq)
+static void ixp2000_pci_irq_unmask(struct irq_data *d)
{
unsigned long temp = *IXP2000_PCI_XSCALE_INT_ENABLE;
- if (irq == IRQ_IXP2000_PCIA)
+ if (d->irq == IRQ_IXP2000_PCIA)
ixp2000_reg_write(IXP2000_PCI_XSCALE_INT_ENABLE, (temp | (1 << 26)));
- else if (irq == IRQ_IXP2000_PCIB)
+ else if (d->irq == IRQ_IXP2000_PCIB)
ixp2000_reg_write(IXP2000_PCI_XSCALE_INT_ENABLE, (temp | (1 << 27)));
}
@@ -401,44 +407,44 @@
}
}
-static void ixp2000_err_irq_mask(unsigned int irq)
+static void ixp2000_err_irq_mask(struct irq_data *d)
{
ixp2000_reg_write(IXP2000_IRQ_ERR_ENABLE_CLR,
- (1 << (irq - IRQ_IXP2000_DRAM0_MIN_ERR)));
+ (1 << (d->irq - IRQ_IXP2000_DRAM0_MIN_ERR)));
}
-static void ixp2000_err_irq_unmask(unsigned int irq)
+static void ixp2000_err_irq_unmask(struct irq_data *d)
{
ixp2000_reg_write(IXP2000_IRQ_ERR_ENABLE_SET,
- (1 << (irq - IRQ_IXP2000_DRAM0_MIN_ERR)));
+ (1 << (d->irq - IRQ_IXP2000_DRAM0_MIN_ERR)));
}
static struct irq_chip ixp2000_err_irq_chip = {
- .ack = ixp2000_err_irq_mask,
- .mask = ixp2000_err_irq_mask,
- .unmask = ixp2000_err_irq_unmask
+ .irq_ack = ixp2000_err_irq_mask,
+ .irq_mask = ixp2000_err_irq_mask,
+ .irq_unmask = ixp2000_err_irq_unmask
};
static struct irq_chip ixp2000_pci_irq_chip = {
- .ack = ixp2000_pci_irq_mask,
- .mask = ixp2000_pci_irq_mask,
- .unmask = ixp2000_pci_irq_unmask
+ .irq_ack = ixp2000_pci_irq_mask,
+ .irq_mask = ixp2000_pci_irq_mask,
+ .irq_unmask = ixp2000_pci_irq_unmask
};
-static void ixp2000_irq_mask(unsigned int irq)
+static void ixp2000_irq_mask(struct irq_data *d)
{
- ixp2000_reg_wrb(IXP2000_IRQ_ENABLE_CLR, (1 << irq));
+ ixp2000_reg_wrb(IXP2000_IRQ_ENABLE_CLR, (1 << d->irq));
}
-static void ixp2000_irq_unmask(unsigned int irq)
+static void ixp2000_irq_unmask(struct irq_data *d)
{
- ixp2000_reg_write(IXP2000_IRQ_ENABLE_SET, (1 << irq));
+ ixp2000_reg_write(IXP2000_IRQ_ENABLE_SET, (1 << d->irq));
}
static struct irq_chip ixp2000_irq_chip = {
- .ack = ixp2000_irq_mask,
- .mask = ixp2000_irq_mask,
- .unmask = ixp2000_irq_unmask
+ .irq_ack = ixp2000_irq_mask,
+ .irq_mask = ixp2000_irq_mask,
+ .irq_unmask = ixp2000_irq_unmask
};
void __init ixp2000_init_irq(void)
diff --git a/arch/arm/mach-ixp2000/ixdp2x00.c b/arch/arm/mach-ixp2000/ixdp2x00.c
index 91fffb9..7d90d3f 100644
--- a/arch/arm/mach-ixp2000/ixdp2x00.c
+++ b/arch/arm/mach-ixp2000/ixdp2x00.c
@@ -63,7 +63,7 @@
};
#endif
-static void ixdp2x00_irq_mask(unsigned int irq)
+static void ixdp2x00_irq_mask(struct irq_data *d)
{
unsigned long dummy;
static struct slowport_cfg old_cfg;
@@ -78,7 +78,7 @@
#endif
dummy = *board_irq_mask;
- dummy |= IXP2000_BOARD_IRQ_MASK(irq);
+ dummy |= IXP2000_BOARD_IRQ_MASK(d->irq);
ixp2000_reg_wrb(board_irq_mask, dummy);
#ifdef CONFIG_ARCH_IXDP2400
@@ -87,7 +87,7 @@
#endif
}
-static void ixdp2x00_irq_unmask(unsigned int irq)
+static void ixdp2x00_irq_unmask(struct irq_data *d)
{
unsigned long dummy;
static struct slowport_cfg old_cfg;
@@ -98,7 +98,7 @@
#endif
dummy = *board_irq_mask;
- dummy &= ~IXP2000_BOARD_IRQ_MASK(irq);
+ dummy &= ~IXP2000_BOARD_IRQ_MASK(d->irq);
ixp2000_reg_wrb(board_irq_mask, dummy);
if (machine_is_ixdp2400())
@@ -111,7 +111,7 @@
static struct slowport_cfg old_cfg;
int i;
- desc->chip->mask(irq);
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
#ifdef CONFIG_ARCH_IXDP2400
if (machine_is_ixdp2400())
@@ -133,13 +133,13 @@
}
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
static struct irq_chip ixdp2x00_cpld_irq_chip = {
- .ack = ixdp2x00_irq_mask,
- .mask = ixdp2x00_irq_mask,
- .unmask = ixdp2x00_irq_unmask
+ .irq_ack = ixdp2x00_irq_mask,
+ .irq_mask = ixdp2x00_irq_mask,
+ .irq_unmask = ixdp2x00_irq_unmask
};
void __init ixdp2x00_init_irq(volatile unsigned long *stat_reg, volatile unsigned long *mask_reg, unsigned long nr_of_irqs)
diff --git a/arch/arm/mach-ixp2000/ixdp2x01.c b/arch/arm/mach-ixp2000/ixdp2x01.c
index 6c121bd..34b1b2a 100644
--- a/arch/arm/mach-ixp2000/ixdp2x01.c
+++ b/arch/arm/mach-ixp2000/ixdp2x01.c
@@ -48,16 +48,16 @@
/*************************************************************************
* IXDP2x01 IRQ Handling
*************************************************************************/
-static void ixdp2x01_irq_mask(unsigned int irq)
+static void ixdp2x01_irq_mask(struct irq_data *d)
{
ixp2000_reg_wrb(IXDP2X01_INT_MASK_SET_REG,
- IXP2000_BOARD_IRQ_MASK(irq));
+ IXP2000_BOARD_IRQ_MASK(d->irq));
}
-static void ixdp2x01_irq_unmask(unsigned int irq)
+static void ixdp2x01_irq_unmask(struct irq_data *d)
{
ixp2000_reg_write(IXDP2X01_INT_MASK_CLR_REG,
- IXP2000_BOARD_IRQ_MASK(irq));
+ IXP2000_BOARD_IRQ_MASK(d->irq));
}
static u32 valid_irq_mask;
@@ -67,7 +67,7 @@
u32 ex_interrupt;
int i;
- desc->chip->mask(irq);
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
ex_interrupt = *IXDP2X01_INT_STAT_REG & valid_irq_mask;
@@ -83,13 +83,13 @@
}
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
static struct irq_chip ixdp2x01_irq_chip = {
- .mask = ixdp2x01_irq_mask,
- .ack = ixdp2x01_irq_mask,
- .unmask = ixdp2x01_irq_unmask
+ .irq_mask = ixdp2x01_irq_mask,
+ .irq_ack = ixdp2x01_irq_mask,
+ .irq_unmask = ixdp2x01_irq_unmask
};
/*
diff --git a/arch/arm/mach-ixp23xx/core.c b/arch/arm/mach-ixp23xx/core.c
index aa4c442..9c8a339 100644
--- a/arch/arm/mach-ixp23xx/core.c
+++ b/arch/arm/mach-ixp23xx/core.c
@@ -111,9 +111,9 @@
static void ixp23xx_config_irq(unsigned int, enum ixp23xx_irq_type);
-static int ixp23xx_irq_set_type(unsigned int irq, unsigned int type)
+static int ixp23xx_irq_set_type(struct irq_data *d, unsigned int type)
{
- int line = irq - IRQ_IXP23XX_GPIO6 + 6;
+ int line = d->irq - IRQ_IXP23XX_GPIO6 + 6;
u32 int_style;
enum ixp23xx_irq_type irq_type;
volatile u32 *int_reg;
@@ -149,7 +149,7 @@
return -EINVAL;
}
- ixp23xx_config_irq(irq, irq_type);
+ ixp23xx_config_irq(d->irq, irq_type);
if (line >= 8) { /* pins 8-15 */
line -= 8;
@@ -173,9 +173,10 @@
return 0;
}
-static void ixp23xx_irq_mask(unsigned int irq)
+static void ixp23xx_irq_mask(struct irq_data *d)
{
volatile unsigned long *intr_reg;
+ unsigned int irq = d->irq;
if (irq >= 56)
irq += 8;
@@ -184,9 +185,9 @@
*intr_reg &= ~(1 << (irq % 32));
}
-static void ixp23xx_irq_ack(unsigned int irq)
+static void ixp23xx_irq_ack(struct irq_data *d)
{
- int line = irq - IRQ_IXP23XX_GPIO6 + 6;
+ int line = d->irq - IRQ_IXP23XX_GPIO6 + 6;
if ((line < 6) || (line > 15))
return;
@@ -198,11 +199,12 @@
* Level triggered interrupts on GPIO lines can only be cleared when the
* interrupt condition disappears.
*/
-static void ixp23xx_irq_level_unmask(unsigned int irq)
+static void ixp23xx_irq_level_unmask(struct irq_data *d)
{
volatile unsigned long *intr_reg;
+ unsigned int irq = d->irq;
- ixp23xx_irq_ack(irq);
+ ixp23xx_irq_ack(d);
if (irq >= 56)
irq += 8;
@@ -211,9 +213,10 @@
*intr_reg |= (1 << (irq % 32));
}
-static void ixp23xx_irq_edge_unmask(unsigned int irq)
+static void ixp23xx_irq_edge_unmask(struct irq_data *d)
{
volatile unsigned long *intr_reg;
+ unsigned int irq = d->irq;
if (irq >= 56)
irq += 8;
@@ -223,26 +226,30 @@
}
static struct irq_chip ixp23xx_irq_level_chip = {
- .ack = ixp23xx_irq_mask,
- .mask = ixp23xx_irq_mask,
- .unmask = ixp23xx_irq_level_unmask,
- .set_type = ixp23xx_irq_set_type
+ .irq_ack = ixp23xx_irq_mask,
+ .irq_mask = ixp23xx_irq_mask,
+ .irq_unmask = ixp23xx_irq_level_unmask,
+ .irq_set_type = ixp23xx_irq_set_type
};
static struct irq_chip ixp23xx_irq_edge_chip = {
- .ack = ixp23xx_irq_ack,
- .mask = ixp23xx_irq_mask,
- .unmask = ixp23xx_irq_edge_unmask,
- .set_type = ixp23xx_irq_set_type
+ .irq_ack = ixp23xx_irq_ack,
+ .irq_mask = ixp23xx_irq_mask,
+ .irq_unmask = ixp23xx_irq_edge_unmask,
+ .irq_set_type = ixp23xx_irq_set_type
};
-static void ixp23xx_pci_irq_mask(unsigned int irq)
+static void ixp23xx_pci_irq_mask(struct irq_data *d)
{
+ unsigned int irq = d->irq;
+
*IXP23XX_PCI_XSCALE_INT_ENABLE &= ~(1 << (IRQ_IXP23XX_INTA + 27 - irq));
}
-static void ixp23xx_pci_irq_unmask(unsigned int irq)
+static void ixp23xx_pci_irq_unmask(struct irq_data *d)
{
+ unsigned int irq = d->irq;
+
*IXP23XX_PCI_XSCALE_INT_ENABLE |= (1 << (IRQ_IXP23XX_INTA + 27 - irq));
}
@@ -256,7 +263,7 @@
pci_interrupt = *IXP23XX_PCI_XSCALE_INT_STATUS;
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
/* See which PCI_INTA, or PCI_INTB interrupted */
if (pci_interrupt & (1 << 26)) {
@@ -269,13 +276,13 @@
generic_handle_irq(irqno);
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
static struct irq_chip ixp23xx_pci_irq_chip = {
- .ack = ixp23xx_pci_irq_mask,
- .mask = ixp23xx_pci_irq_mask,
- .unmask = ixp23xx_pci_irq_unmask
+ .irq_ack = ixp23xx_pci_irq_mask,
+ .irq_mask = ixp23xx_pci_irq_mask,
+ .irq_unmask = ixp23xx_pci_irq_unmask
};
static void ixp23xx_config_irq(unsigned int irq, enum ixp23xx_irq_type type)
diff --git a/arch/arm/mach-ixp23xx/ixdp2351.c b/arch/arm/mach-ixp23xx/ixdp2351.c
index 664e39c..181116a 100644
--- a/arch/arm/mach-ixp23xx/ixdp2351.c
+++ b/arch/arm/mach-ixp23xx/ixdp2351.c
@@ -48,14 +48,14 @@
/*
* IXDP2351 Interrupt Handling
*/
-static void ixdp2351_inta_mask(unsigned int irq)
+static void ixdp2351_inta_mask(struct irq_data *d)
{
- *IXDP2351_CPLD_INTA_MASK_SET_REG = IXDP2351_INTA_IRQ_MASK(irq);
+ *IXDP2351_CPLD_INTA_MASK_SET_REG = IXDP2351_INTA_IRQ_MASK(d->irq);
}
-static void ixdp2351_inta_unmask(unsigned int irq)
+static void ixdp2351_inta_unmask(struct irq_data *d)
{
- *IXDP2351_CPLD_INTA_MASK_CLR_REG = IXDP2351_INTA_IRQ_MASK(irq);
+ *IXDP2351_CPLD_INTA_MASK_CLR_REG = IXDP2351_INTA_IRQ_MASK(d->irq);
}
static void ixdp2351_inta_handler(unsigned int irq, struct irq_desc *desc)
@@ -64,7 +64,7 @@
*IXDP2351_CPLD_INTA_STAT_REG & IXDP2351_INTA_IRQ_VALID;
int i;
- desc->chip->mask(irq);
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
for (i = 0; i < IXDP2351_INTA_IRQ_NUM; i++) {
if (ex_interrupt & (1 << i)) {
@@ -74,23 +74,23 @@
}
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
static struct irq_chip ixdp2351_inta_chip = {
- .ack = ixdp2351_inta_mask,
- .mask = ixdp2351_inta_mask,
- .unmask = ixdp2351_inta_unmask
+ .irq_ack = ixdp2351_inta_mask,
+ .irq_mask = ixdp2351_inta_mask,
+ .irq_unmask = ixdp2351_inta_unmask
};
-static void ixdp2351_intb_mask(unsigned int irq)
+static void ixdp2351_intb_mask(struct irq_data *d)
{
- *IXDP2351_CPLD_INTB_MASK_SET_REG = IXDP2351_INTB_IRQ_MASK(irq);
+ *IXDP2351_CPLD_INTB_MASK_SET_REG = IXDP2351_INTB_IRQ_MASK(d->irq);
}
-static void ixdp2351_intb_unmask(unsigned int irq)
+static void ixdp2351_intb_unmask(struct irq_data *d)
{
- *IXDP2351_CPLD_INTB_MASK_CLR_REG = IXDP2351_INTB_IRQ_MASK(irq);
+ *IXDP2351_CPLD_INTB_MASK_CLR_REG = IXDP2351_INTB_IRQ_MASK(d->irq);
}
static void ixdp2351_intb_handler(unsigned int irq, struct irq_desc *desc)
@@ -99,7 +99,7 @@
*IXDP2351_CPLD_INTB_STAT_REG & IXDP2351_INTB_IRQ_VALID;
int i;
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
for (i = 0; i < IXDP2351_INTB_IRQ_NUM; i++) {
if (ex_interrupt & (1 << i)) {
@@ -109,13 +109,13 @@
}
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
static struct irq_chip ixdp2351_intb_chip = {
- .ack = ixdp2351_intb_mask,
- .mask = ixdp2351_intb_mask,
- .unmask = ixdp2351_intb_unmask
+ .irq_ack = ixdp2351_intb_mask,
+ .irq_mask = ixdp2351_intb_mask,
+ .irq_unmask = ixdp2351_intb_unmask
};
void __init ixdp2351_init_irq(void)
diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index 4dbfcbb..4dc68d6 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -128,9 +128,9 @@
}
EXPORT_SYMBOL(irq_to_gpio);
-static int ixp4xx_set_irq_type(unsigned int irq, unsigned int type)
+static int ixp4xx_set_irq_type(struct irq_data *d, unsigned int type)
{
- int line = irq2gpio[irq];
+ int line = irq2gpio[d->irq];
u32 int_style;
enum ixp4xx_irq_type irq_type;
volatile u32 *int_reg;
@@ -167,9 +167,9 @@
}
if (irq_type == IXP4XX_IRQ_EDGE)
- ixp4xx_irq_edge |= (1 << irq);
+ ixp4xx_irq_edge |= (1 << d->irq);
else
- ixp4xx_irq_edge &= ~(1 << irq);
+ ixp4xx_irq_edge &= ~(1 << d->irq);
if (line >= 8) { /* pins 8-15 */
line -= 8;
@@ -188,22 +188,22 @@
*int_reg |= (int_style << (line * IXP4XX_GPIO_STYLE_SIZE));
/* Configure the line as an input */
- gpio_line_config(irq2gpio[irq], IXP4XX_GPIO_IN);
+ gpio_line_config(irq2gpio[d->irq], IXP4XX_GPIO_IN);
return 0;
}
-static void ixp4xx_irq_mask(unsigned int irq)
+static void ixp4xx_irq_mask(struct irq_data *d)
{
- if ((cpu_is_ixp46x() || cpu_is_ixp43x()) && irq >= 32)
- *IXP4XX_ICMR2 &= ~(1 << (irq - 32));
+ if ((cpu_is_ixp46x() || cpu_is_ixp43x()) && d->irq >= 32)
+ *IXP4XX_ICMR2 &= ~(1 << (d->irq - 32));
else
- *IXP4XX_ICMR &= ~(1 << irq);
+ *IXP4XX_ICMR &= ~(1 << d->irq);
}
-static void ixp4xx_irq_ack(unsigned int irq)
+static void ixp4xx_irq_ack(struct irq_data *d)
{
- int line = (irq < 32) ? irq2gpio[irq] : -1;
+ int line = (d->irq < 32) ? irq2gpio[d->irq] : -1;
if (line >= 0)
*IXP4XX_GPIO_GPISR = (1 << line);
@@ -213,23 +213,23 @@
* Level triggered interrupts on GPIO lines can only be cleared when the
* interrupt condition disappears.
*/
-static void ixp4xx_irq_unmask(unsigned int irq)
+static void ixp4xx_irq_unmask(struct irq_data *d)
{
- if (!(ixp4xx_irq_edge & (1 << irq)))
- ixp4xx_irq_ack(irq);
+ if (!(ixp4xx_irq_edge & (1 << d->irq)))
+ ixp4xx_irq_ack(d);
- if ((cpu_is_ixp46x() || cpu_is_ixp43x()) && irq >= 32)
- *IXP4XX_ICMR2 |= (1 << (irq - 32));
+ if ((cpu_is_ixp46x() || cpu_is_ixp43x()) && d->irq >= 32)
+ *IXP4XX_ICMR2 |= (1 << (d->irq - 32));
else
- *IXP4XX_ICMR |= (1 << irq);
+ *IXP4XX_ICMR |= (1 << d->irq);
}
static struct irq_chip ixp4xx_irq_chip = {
.name = "IXP4xx",
- .ack = ixp4xx_irq_ack,
- .mask = ixp4xx_irq_mask,
- .unmask = ixp4xx_irq_unmask,
- .set_type = ixp4xx_set_irq_type,
+ .irq_ack = ixp4xx_irq_ack,
+ .irq_mask = ixp4xx_irq_mask,
+ .irq_unmask = ixp4xx_irq_unmask,
+ .irq_set_type = ixp4xx_set_irq_type,
};
void __init ixp4xx_init_irq(void)
diff --git a/arch/arm/mach-ks8695/irq.c b/arch/arm/mach-ks8695/irq.c
index e375c1d..7998cca 100644
--- a/arch/arm/mach-ks8695/irq.c
+++ b/arch/arm/mach-ks8695/irq.c
@@ -34,29 +34,29 @@
#include <mach/regs-irq.h>
#include <mach/regs-gpio.h>
-static void ks8695_irq_mask(unsigned int irqno)
+static void ks8695_irq_mask(struct irq_data *d)
{
unsigned long inten;
inten = __raw_readl(KS8695_IRQ_VA + KS8695_INTEN);
- inten &= ~(1 << irqno);
+ inten &= ~(1 << d->irq);
__raw_writel(inten, KS8695_IRQ_VA + KS8695_INTEN);
}
-static void ks8695_irq_unmask(unsigned int irqno)
+static void ks8695_irq_unmask(struct irq_data *d)
{
unsigned long inten;
inten = __raw_readl(KS8695_IRQ_VA + KS8695_INTEN);
- inten |= (1 << irqno);
+ inten |= (1 << d->irq);
__raw_writel(inten, KS8695_IRQ_VA + KS8695_INTEN);
}
-static void ks8695_irq_ack(unsigned int irqno)
+static void ks8695_irq_ack(struct irq_data *d)
{
- __raw_writel((1 << irqno), KS8695_IRQ_VA + KS8695_INTST);
+ __raw_writel((1 << d->irq), KS8695_IRQ_VA + KS8695_INTST);
}
@@ -64,7 +64,7 @@
static struct irq_chip ks8695_irq_edge_chip;
-static int ks8695_irq_set_type(unsigned int irqno, unsigned int type)
+static int ks8695_irq_set_type(struct irq_data *d, unsigned int type)
{
unsigned long ctrl, mode;
unsigned short level_triggered = 0;
@@ -93,7 +93,7 @@
return -EINVAL;
}
- switch (irqno) {
+ switch (d->irq) {
case KS8695_IRQ_EXTERN0:
ctrl &= ~IOPC_IOEINT0TM;
ctrl |= IOPC_IOEINT0_MODE(mode);
@@ -115,12 +115,12 @@
}
if (level_triggered) {
- set_irq_chip(irqno, &ks8695_irq_level_chip);
- set_irq_handler(irqno, handle_level_irq);
+ set_irq_chip(d->irq, &ks8695_irq_level_chip);
+ set_irq_handler(d->irq, handle_level_irq);
}
else {
- set_irq_chip(irqno, &ks8695_irq_edge_chip);
- set_irq_handler(irqno, handle_edge_irq);
+ set_irq_chip(d->irq, &ks8695_irq_edge_chip);
+ set_irq_handler(d->irq, handle_edge_irq);
}
__raw_writel(ctrl, KS8695_GPIO_VA + KS8695_IOPC);
@@ -128,17 +128,17 @@
}
static struct irq_chip ks8695_irq_level_chip = {
- .ack = ks8695_irq_mask,
- .mask = ks8695_irq_mask,
- .unmask = ks8695_irq_unmask,
- .set_type = ks8695_irq_set_type,
+ .irq_ack = ks8695_irq_mask,
+ .irq_mask = ks8695_irq_mask,
+ .irq_unmask = ks8695_irq_unmask,
+ .irq_set_type = ks8695_irq_set_type,
};
static struct irq_chip ks8695_irq_edge_chip = {
- .ack = ks8695_irq_ack,
- .mask = ks8695_irq_mask,
- .unmask = ks8695_irq_unmask,
- .set_type = ks8695_irq_set_type,
+ .irq_ack = ks8695_irq_ack,
+ .irq_mask = ks8695_irq_mask,
+ .irq_unmask = ks8695_irq_unmask,
+ .irq_set_type = ks8695_irq_set_type,
};
void __init ks8695_init_irq(void)
@@ -164,7 +164,8 @@
/* Edge-triggered interrupts */
default:
- ks8695_irq_ack(irq); /* clear pending bit */
+ /* clear pending bit */
+ ks8695_irq_ack(irq_get_irq_data(irq));
set_irq_chip(irq, &ks8695_irq_edge_chip);
set_irq_handler(irq, handle_edge_irq);
}
diff --git a/arch/arm/mach-lh7a40x/arch-kev7a400.c b/arch/arm/mach-lh7a40x/arch-kev7a400.c
index 9088c166..71129c3 100644
--- a/arch/arm/mach-lh7a40x/arch-kev7a400.c
+++ b/arch/arm/mach-lh7a40x/arch-kev7a400.c
@@ -46,28 +46,28 @@
static u16 CPLD_IRQ_mask; /* Mask for CPLD IRQs, 1 == unmasked */
-static void kev7a400_ack_cpld_irq (u32 irq)
+static void kev7a400_ack_cpld_irq(struct irq_data *d)
{
- CPLD_CL_INT = 1 << (irq - IRQ_KEV7A400_CPLD);
+ CPLD_CL_INT = 1 << (d->irq - IRQ_KEV7A400_CPLD);
}
-static void kev7a400_mask_cpld_irq (u32 irq)
+static void kev7a400_mask_cpld_irq(struct irq_data *d)
{
- CPLD_IRQ_mask &= ~(1 << (irq - IRQ_KEV7A400_CPLD));
+ CPLD_IRQ_mask &= ~(1 << (d->irq - IRQ_KEV7A400_CPLD));
CPLD_WR_PB_INT_MASK = CPLD_IRQ_mask;
}
-static void kev7a400_unmask_cpld_irq (u32 irq)
+static void kev7a400_unmask_cpld_irq(struct irq_data *d)
{
- CPLD_IRQ_mask |= 1 << (irq - IRQ_KEV7A400_CPLD);
+ CPLD_IRQ_mask |= 1 << (d->irq - IRQ_KEV7A400_CPLD);
CPLD_WR_PB_INT_MASK = CPLD_IRQ_mask;
}
static struct irq_chip kev7a400_cpld_chip = {
- .name = "CPLD",
- .ack = kev7a400_ack_cpld_irq,
- .mask = kev7a400_mask_cpld_irq,
- .unmask = kev7a400_unmask_cpld_irq,
+ .name = "CPLD",
+ .irq_ack = kev7a400_ack_cpld_irq,
+ .irq_mask = kev7a400_mask_cpld_irq,
+ .irq_unmask = kev7a400_unmask_cpld_irq,
};
diff --git a/arch/arm/mach-lh7a40x/arch-lpd7a40x.c b/arch/arm/mach-lh7a40x/arch-lpd7a40x.c
index 7315a56..e735546 100644
--- a/arch/arm/mach-lh7a40x/arch-lpd7a40x.c
+++ b/arch/arm/mach-lh7a40x/arch-lpd7a40x.c
@@ -159,7 +159,7 @@
#endif
}
-static void lh7a40x_ack_cpld_irq (u32 irq)
+static void lh7a40x_ack_cpld_irq(struct irq_data *d)
{
/* CPLD doesn't have ack capability, but some devices may */
@@ -167,14 +167,14 @@
/* The touch control *must* mask the interrupt because the
* interrupt bit is read by the driver to determine if the pen
* is still down. */
- if (irq == IRQ_TOUCH)
+ if (d->irq == IRQ_TOUCH)
CPLD_INTERRUPTS |= CPLD_INTMASK_TOUCH;
#endif
}
-static void lh7a40x_mask_cpld_irq (u32 irq)
+static void lh7a40x_mask_cpld_irq(struct irq_data *d)
{
- switch (irq) {
+ switch (d->irq) {
case IRQ_LPD7A40X_ETH_INT:
CPLD_INTERRUPTS |= CPLD_INTMASK_ETHERNET;
break;
@@ -186,9 +186,9 @@
}
}
-static void lh7a40x_unmask_cpld_irq (u32 irq)
+static void lh7a40x_unmask_cpld_irq(struct irq_data *d)
{
- switch (irq) {
+ switch (d->irq) {
case IRQ_LPD7A40X_ETH_INT:
CPLD_INTERRUPTS &= ~CPLD_INTMASK_ETHERNET;
break;
@@ -201,17 +201,17 @@
}
static struct irq_chip lpd7a40x_cpld_chip = {
- .name = "CPLD",
- .ack = lh7a40x_ack_cpld_irq,
- .mask = lh7a40x_mask_cpld_irq,
- .unmask = lh7a40x_unmask_cpld_irq,
+ .name = "CPLD",
+ .irq_ack = lh7a40x_ack_cpld_irq,
+ .irq_mask = lh7a40x_mask_cpld_irq,
+ .irq_unmask = lh7a40x_unmask_cpld_irq,
};
static void lpd7a40x_cpld_handler (unsigned int irq, struct irq_desc *desc)
{
unsigned int mask = CPLD_INTERRUPTS;
- desc->chip->ack (irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
if ((mask & (1<<0)) == 0) /* WLAN */
generic_handle_irq(IRQ_LPD7A40X_ETH_INT);
@@ -221,7 +221,8 @@
generic_handle_irq(IRQ_TOUCH);
#endif
- desc->chip->unmask (irq); /* Level-triggered need this */
+ /* Level-triggered need this */
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
diff --git a/arch/arm/mach-lh7a40x/irq-lh7a400.c b/arch/arm/mach-lh7a40x/irq-lh7a400.c
index 1ad3afc..f2e7e65 100644
--- a/arch/arm/mach-lh7a40x/irq-lh7a400.c
+++ b/arch/arm/mach-lh7a40x/irq-lh7a400.c
@@ -21,34 +21,34 @@
/* CPU IRQ handling */
-static void lh7a400_mask_irq (u32 irq)
+static void lh7a400_mask_irq(struct irq_data *d)
{
- INTC_INTENC = (1 << irq);
+ INTC_INTENC = (1 << d->irq);
}
-static void lh7a400_unmask_irq (u32 irq)
+static void lh7a400_unmask_irq(struct irq_data *d)
{
- INTC_INTENS = (1 << irq);
+ INTC_INTENS = (1 << d->irq);
}
-static void lh7a400_ack_gpio_irq (u32 irq)
+static void lh7a400_ack_gpio_irq(struct irq_data *d)
{
- GPIO_GPIOFEOI = (1 << IRQ_TO_GPIO (irq));
- INTC_INTENC = (1 << irq);
+ GPIO_GPIOFEOI = (1 << IRQ_TO_GPIO (d->irq));
+ INTC_INTENC = (1 << d->irq);
}
static struct irq_chip lh7a400_internal_chip = {
- .name = "MPU",
- .ack = lh7a400_mask_irq, /* Level triggering -> mask is ack */
- .mask = lh7a400_mask_irq,
- .unmask = lh7a400_unmask_irq,
+ .name = "MPU",
+ .irq_ack = lh7a400_mask_irq, /* Level triggering -> mask is ack */
+ .irq_mask = lh7a400_mask_irq,
+ .irq_unmask = lh7a400_unmask_irq,
};
static struct irq_chip lh7a400_gpio_chip = {
- .name = "GPIO",
- .ack = lh7a400_ack_gpio_irq,
- .mask = lh7a400_mask_irq,
- .unmask = lh7a400_unmask_irq,
+ .name = "GPIO",
+ .irq_ack = lh7a400_ack_gpio_irq,
+ .irq_mask = lh7a400_mask_irq,
+ .irq_unmask = lh7a400_unmask_irq,
};
diff --git a/arch/arm/mach-lh7a40x/irq-lh7a404.c b/arch/arm/mach-lh7a40x/irq-lh7a404.c
index 12b045b..14b1733 100644
--- a/arch/arm/mach-lh7a40x/irq-lh7a404.c
+++ b/arch/arm/mach-lh7a40x/irq-lh7a404.c
@@ -43,64 +43,64 @@
/* CPU IRQ handling */
-static void lh7a404_vic1_mask_irq (u32 irq)
+static void lh7a404_vic1_mask_irq(struct irq_data *d)
{
- VIC1_INTENCLR = (1 << irq);
+ VIC1_INTENCLR = (1 << d->irq);
}
-static void lh7a404_vic1_unmask_irq (u32 irq)
+static void lh7a404_vic1_unmask_irq(struct irq_data *d)
{
- VIC1_INTEN = (1 << irq);
+ VIC1_INTEN = (1 << d->irq);
}
-static void lh7a404_vic2_mask_irq (u32 irq)
+static void lh7a404_vic2_mask_irq(struct irq_data *d)
{
- VIC2_INTENCLR = (1 << (irq - 32));
+ VIC2_INTENCLR = (1 << (d->irq - 32));
}
-static void lh7a404_vic2_unmask_irq (u32 irq)
+static void lh7a404_vic2_unmask_irq(struct irq_data *d)
{
- VIC2_INTEN = (1 << (irq - 32));
+ VIC2_INTEN = (1 << (d->irq - 32));
}
-static void lh7a404_vic1_ack_gpio_irq (u32 irq)
+static void lh7a404_vic1_ack_gpio_irq(struct irq_data *d)
{
- GPIO_GPIOFEOI = (1 << IRQ_TO_GPIO (irq));
- VIC1_INTENCLR = (1 << irq);
+ GPIO_GPIOFEOI = (1 << IRQ_TO_GPIO (d->irq));
+ VIC1_INTENCLR = (1 << d->irq);
}
-static void lh7a404_vic2_ack_gpio_irq (u32 irq)
+static void lh7a404_vic2_ack_gpio_irq(struct irq_data *d)
{
- GPIO_GPIOFEOI = (1 << IRQ_TO_GPIO (irq));
- VIC2_INTENCLR = (1 << irq);
+ GPIO_GPIOFEOI = (1 << IRQ_TO_GPIO (d->irq));
+ VIC2_INTENCLR = (1 << d->irq);
}
static struct irq_chip lh7a404_vic1_chip = {
- .name = "VIC1",
- .ack = lh7a404_vic1_mask_irq, /* Because level-triggered */
- .mask = lh7a404_vic1_mask_irq,
- .unmask = lh7a404_vic1_unmask_irq,
+ .name = "VIC1",
+ .irq_ack = lh7a404_vic1_mask_irq, /* Because level-triggered */
+ .irq_mask = lh7a404_vic1_mask_irq,
+ .irq_unmask = lh7a404_vic1_unmask_irq,
};
static struct irq_chip lh7a404_vic2_chip = {
- .name = "VIC2",
- .ack = lh7a404_vic2_mask_irq, /* Because level-triggered */
- .mask = lh7a404_vic2_mask_irq,
- .unmask = lh7a404_vic2_unmask_irq,
+ .name = "VIC2",
+ .irq_ack = lh7a404_vic2_mask_irq, /* Because level-triggered */
+ .irq_mask = lh7a404_vic2_mask_irq,
+ .irq_unmask = lh7a404_vic2_unmask_irq,
};
static struct irq_chip lh7a404_gpio_vic1_chip = {
- .name = "GPIO-VIC1",
- .ack = lh7a404_vic1_ack_gpio_irq,
- .mask = lh7a404_vic1_mask_irq,
- .unmask = lh7a404_vic1_unmask_irq,
+ .name = "GPIO-VIC1",
+ .irq_ack = lh7a404_vic1_ack_gpio_irq,
+ .irq_mask = lh7a404_vic1_mask_irq,
+ .irq_unmask = lh7a404_vic1_unmask_irq,
};
static struct irq_chip lh7a404_gpio_vic2_chip = {
- .name = "GPIO-VIC2",
- .ack = lh7a404_vic2_ack_gpio_irq,
- .mask = lh7a404_vic2_mask_irq,
- .unmask = lh7a404_vic2_unmask_irq,
+ .name = "GPIO-VIC2",
+ .irq_ack = lh7a404_vic2_ack_gpio_irq,
+ .irq_mask = lh7a404_vic2_mask_irq,
+ .irq_unmask = lh7a404_vic2_unmask_irq,
};
/* IRQ initialization */
diff --git a/arch/arm/mach-lh7a40x/irq-lpd7a40x.c b/arch/arm/mach-lh7a40x/irq-lpd7a40x.c
index fd033bb..1bfdcdd 100644
--- a/arch/arm/mach-lh7a40x/irq-lpd7a40x.c
+++ b/arch/arm/mach-lh7a40x/irq-lpd7a40x.c
@@ -20,14 +20,14 @@
#include "common.h"
-static void lh7a40x_ack_cpld_irq (u32 irq)
+static void lh7a40x_ack_cpld_irq(struct irq_data *d)
{
/* CPLD doesn't have ack capability */
}
-static void lh7a40x_mask_cpld_irq (u32 irq)
+static void lh7a40x_mask_cpld_irq(struct irq_data *d)
{
- switch (irq) {
+ switch (d->irq) {
case IRQ_LPD7A40X_ETH_INT:
CPLD_INTERRUPTS = CPLD_INTERRUPTS | 0x4;
break;
@@ -37,9 +37,9 @@
}
}
-static void lh7a40x_unmask_cpld_irq (u32 irq)
+static void lh7a40x_unmask_cpld_irq(struct irq_data *d)
{
- switch (irq) {
+ switch (d->irq) {
case IRQ_LPD7A40X_ETH_INT:
CPLD_INTERRUPTS = CPLD_INTERRUPTS & ~ 0x4;
break;
@@ -50,17 +50,17 @@
}
static struct irq_chip lh7a40x_cpld_chip = {
- .name = "CPLD",
- .ack = lh7a40x_ack_cpld_irq,
- .mask = lh7a40x_mask_cpld_irq,
- .unmask = lh7a40x_unmask_cpld_irq,
+ .name = "CPLD",
+ .irq_ack = lh7a40x_ack_cpld_irq,
+ .irq_mask = lh7a40x_mask_cpld_irq,
+ .irq_unmask = lh7a40x_unmask_cpld_irq,
};
static void lh7a40x_cpld_handler (unsigned int irq, struct irq_desc *desc)
{
unsigned int mask = CPLD_INTERRUPTS;
- desc->chip->ack (irq);
+ desc->irq_data.chip->ack (irq);
if ((mask & 0x1) == 0) /* WLAN */
generic_handle_irq(IRQ_LPD7A40X_ETH_INT);
@@ -68,7 +68,7 @@
if ((mask & 0x2) == 0) /* Touch */
generic_handle_irq(IRQ_LPD7A400_TS);
- desc->chip->unmask (irq); /* Level-triggered need this */
+ desc->irq_data.chip->unmask (irq); /* Level-triggered need this */
}
diff --git a/arch/arm/mach-lpc32xx/irq.c b/arch/arm/mach-lpc32xx/irq.c
index bd0df26c..316ecbf 100644
--- a/arch/arm/mach-lpc32xx/irq.c
+++ b/arch/arm/mach-lpc32xx/irq.c
@@ -191,38 +191,38 @@
}
}
-static void lpc32xx_mask_irq(unsigned int irq)
+static void lpc32xx_mask_irq(struct irq_data *d)
{
unsigned int reg, ctrl, mask;
- get_controller(irq, &ctrl, &mask);
+ get_controller(d->irq, &ctrl, &mask);
reg = __raw_readl(LPC32XX_INTC_MASK(ctrl)) & ~mask;
__raw_writel(reg, LPC32XX_INTC_MASK(ctrl));
}
-static void lpc32xx_unmask_irq(unsigned int irq)
+static void lpc32xx_unmask_irq(struct irq_data *d)
{
unsigned int reg, ctrl, mask;
- get_controller(irq, &ctrl, &mask);
+ get_controller(d->irq, &ctrl, &mask);
reg = __raw_readl(LPC32XX_INTC_MASK(ctrl)) | mask;
__raw_writel(reg, LPC32XX_INTC_MASK(ctrl));
}
-static void lpc32xx_ack_irq(unsigned int irq)
+static void lpc32xx_ack_irq(struct irq_data *d)
{
unsigned int ctrl, mask;
- get_controller(irq, &ctrl, &mask);
+ get_controller(d->irq, &ctrl, &mask);
__raw_writel(mask, LPC32XX_INTC_RAW_STAT(ctrl));
/* Also need to clear pending wake event */
- if (lpc32xx_events[irq].mask != 0)
- __raw_writel(lpc32xx_events[irq].mask,
- lpc32xx_events[irq].event_group->rawstat_reg);
+ if (lpc32xx_events[d->irq].mask != 0)
+ __raw_writel(lpc32xx_events[d->irq].mask,
+ lpc32xx_events[d->irq].event_group->rawstat_reg);
}
static void __lpc32xx_set_irq_type(unsigned int irq, int use_high_level,
@@ -261,27 +261,27 @@
}
}
-static int lpc32xx_set_irq_type(unsigned int irq, unsigned int type)
+static int lpc32xx_set_irq_type(struct irq_data *d, unsigned int type)
{
switch (type) {
case IRQ_TYPE_EDGE_RISING:
/* Rising edge sensitive */
- __lpc32xx_set_irq_type(irq, 1, 1);
+ __lpc32xx_set_irq_type(d->irq, 1, 1);
break;
case IRQ_TYPE_EDGE_FALLING:
/* Falling edge sensitive */
- __lpc32xx_set_irq_type(irq, 0, 1);
+ __lpc32xx_set_irq_type(d->irq, 0, 1);
break;
case IRQ_TYPE_LEVEL_LOW:
/* Low level sensitive */
- __lpc32xx_set_irq_type(irq, 0, 0);
+ __lpc32xx_set_irq_type(d->irq, 0, 0);
break;
case IRQ_TYPE_LEVEL_HIGH:
/* High level sensitive */
- __lpc32xx_set_irq_type(irq, 1, 0);
+ __lpc32xx_set_irq_type(d->irq, 1, 0);
break;
/* Other modes are not supported */
@@ -290,33 +290,33 @@
}
/* Ok to use the level handler for all types */
- set_irq_handler(irq, handle_level_irq);
+ set_irq_handler(d->irq, handle_level_irq);
return 0;
}
-static int lpc32xx_irq_wake(unsigned int irqno, unsigned int state)
+static int lpc32xx_irq_wake(struct irq_data *d, unsigned int state)
{
unsigned long eventreg;
- if (lpc32xx_events[irqno].mask != 0) {
- eventreg = __raw_readl(lpc32xx_events[irqno].
+ if (lpc32xx_events[d->irq].mask != 0) {
+ eventreg = __raw_readl(lpc32xx_events[d->irq].
event_group->enab_reg);
if (state)
- eventreg |= lpc32xx_events[irqno].mask;
+ eventreg |= lpc32xx_events[d->irq].mask;
else
- eventreg &= ~lpc32xx_events[irqno].mask;
+ eventreg &= ~lpc32xx_events[d->irq].mask;
__raw_writel(eventreg,
- lpc32xx_events[irqno].event_group->enab_reg);
+ lpc32xx_events[d->irq].event_group->enab_reg);
return 0;
}
/* Clear event */
- __raw_writel(lpc32xx_events[irqno].mask,
- lpc32xx_events[irqno].event_group->rawstat_reg);
+ __raw_writel(lpc32xx_events[d->irq].mask,
+ lpc32xx_events[d->irq].event_group->rawstat_reg);
return -ENODEV;
}
@@ -336,11 +336,11 @@
}
static struct irq_chip lpc32xx_irq_chip = {
- .ack = lpc32xx_ack_irq,
- .mask = lpc32xx_mask_irq,
- .unmask = lpc32xx_unmask_irq,
- .set_type = lpc32xx_set_irq_type,
- .set_wake = lpc32xx_irq_wake
+ .irq_ack = lpc32xx_ack_irq,
+ .irq_mask = lpc32xx_mask_irq,
+ .irq_unmask = lpc32xx_unmask_irq,
+ .irq_set_type = lpc32xx_set_irq_type,
+ .irq_set_wake = lpc32xx_irq_wake
};
static void lpc32xx_sic1_handler(unsigned int irq, struct irq_desc *desc)
diff --git a/arch/arm/mach-mmp/include/mach/mfp-mmp2.h b/arch/arm/mach-mmp/include/mach/mfp-mmp2.h
index 117e303..4ad3862 100644
--- a/arch/arm/mach-mmp/include/mach/mfp-mmp2.h
+++ b/arch/arm/mach-mmp/include/mach/mfp-mmp2.h
@@ -6,7 +6,7 @@
#define MFP_DRIVE_VERY_SLOW (0x0 << 13)
#define MFP_DRIVE_SLOW (0x2 << 13)
#define MFP_DRIVE_MEDIUM (0x4 << 13)
-#define MFP_DRIVE_FAST (0x8 << 13)
+#define MFP_DRIVE_FAST (0x6 << 13)
/* GPIO */
#define GPIO0_GPIO MFP_CFG(GPIO0, AF0)
diff --git a/arch/arm/mach-mmp/include/mach/mfp-pxa910.h b/arch/arm/mach-mmp/include/mach/mfp-pxa910.h
index 7e8a80f..fbd7ee8 100644
--- a/arch/arm/mach-mmp/include/mach/mfp-pxa910.h
+++ b/arch/arm/mach-mmp/include/mach/mfp-pxa910.h
@@ -6,7 +6,7 @@
#define MFP_DRIVE_VERY_SLOW (0x0 << 13)
#define MFP_DRIVE_SLOW (0x2 << 13)
#define MFP_DRIVE_MEDIUM (0x4 << 13)
-#define MFP_DRIVE_FAST (0x8 << 13)
+#define MFP_DRIVE_FAST (0x6 << 13)
/* UART2 */
#define GPIO47_UART2_RXD MFP_CFG(GPIO47, AF6)
diff --git a/arch/arm/mach-mmp/irq-mmp2.c b/arch/arm/mach-mmp/irq-mmp2.c
index 01342be..fa03703 100644
--- a/arch/arm/mach-mmp/irq-mmp2.c
+++ b/arch/arm/mach-mmp/irq-mmp2.c
@@ -20,48 +20,48 @@
#include "common.h"
-static void icu_mask_irq(unsigned int irq)
+static void icu_mask_irq(struct irq_data *d)
{
- uint32_t r = __raw_readl(ICU_INT_CONF(irq));
+ uint32_t r = __raw_readl(ICU_INT_CONF(d->irq));
r &= ~ICU_INT_ROUTE_PJ4_IRQ;
- __raw_writel(r, ICU_INT_CONF(irq));
+ __raw_writel(r, ICU_INT_CONF(d->irq));
}
-static void icu_unmask_irq(unsigned int irq)
+static void icu_unmask_irq(struct irq_data *d)
{
- uint32_t r = __raw_readl(ICU_INT_CONF(irq));
+ uint32_t r = __raw_readl(ICU_INT_CONF(d->irq));
r |= ICU_INT_ROUTE_PJ4_IRQ;
- __raw_writel(r, ICU_INT_CONF(irq));
+ __raw_writel(r, ICU_INT_CONF(d->irq));
}
static struct irq_chip icu_irq_chip = {
.name = "icu_irq",
- .mask = icu_mask_irq,
- .mask_ack = icu_mask_irq,
- .unmask = icu_unmask_irq,
+ .irq_mask = icu_mask_irq,
+ .irq_mask_ack = icu_mask_irq,
+ .irq_unmask = icu_unmask_irq,
};
-static void pmic_irq_ack(unsigned int irq)
+static void pmic_irq_ack(struct irq_data *d)
{
- if (irq == IRQ_MMP2_PMIC)
+ if (d->irq == IRQ_MMP2_PMIC)
mmp2_clear_pmic_int();
}
#define SECOND_IRQ_MASK(_name_, irq_base, prefix) \
-static void _name_##_mask_irq(unsigned int irq) \
+static void _name_##_mask_irq(struct irq_data *d) \
{ \
uint32_t r; \
- r = __raw_readl(prefix##_MASK) | (1 << (irq - irq_base)); \
+ r = __raw_readl(prefix##_MASK) | (1 << (d->irq - irq_base)); \
__raw_writel(r, prefix##_MASK); \
}
#define SECOND_IRQ_UNMASK(_name_, irq_base, prefix) \
-static void _name_##_unmask_irq(unsigned int irq) \
+static void _name_##_unmask_irq(struct irq_data *d) \
{ \
uint32_t r; \
- r = __raw_readl(prefix##_MASK) & ~(1 << (irq - irq_base)); \
+ r = __raw_readl(prefix##_MASK) & ~(1 << (d->irq - irq_base)); \
__raw_writel(r, prefix##_MASK); \
}
@@ -88,8 +88,8 @@
SECOND_IRQ_DEMUX(_name_, irq_base, prefix) \
static struct irq_chip _name_##_irq_chip = { \
.name = #_name_, \
- .mask = _name_##_mask_irq, \
- .unmask = _name_##_unmask_irq, \
+ .irq_mask = _name_##_mask_irq, \
+ .irq_unmask = _name_##_unmask_irq, \
}
SECOND_IRQ_CHIP(pmic, IRQ_MMP2_PMIC_BASE, MMP2_ICU_INT4);
@@ -103,10 +103,12 @@
int irq;
for (irq = start; num > 0; irq++, num--) {
+ struct irq_data *d = irq_get_irq_data(irq);
+
/* mask and clear the IRQ */
- chip->mask(irq);
- if (chip->ack)
- chip->ack(irq);
+ chip->irq_mask(d);
+ if (chip->irq_ack)
+ chip->irq_ack(d);
set_irq_chip(irq, chip);
set_irq_flags(irq, IRQF_VALID);
@@ -119,7 +121,7 @@
int irq;
for (irq = 0; irq < IRQ_MMP2_MUX_BASE; irq++) {
- icu_mask_irq(irq);
+ icu_mask_irq(irq_get_irq_data(irq));
set_irq_chip(irq, &icu_irq_chip);
set_irq_flags(irq, IRQF_VALID);
@@ -139,7 +141,7 @@
/* NOTE: IRQ_MMP2_PMIC requires the PMIC MFPR register
* to be written to clear the interrupt
*/
- pmic_irq_chip.ack = pmic_irq_ack;
+ pmic_irq_chip.irq_ack = pmic_irq_ack;
init_mux_irq(&pmic_irq_chip, IRQ_MMP2_PMIC_BASE, 2);
init_mux_irq(&rtc_irq_chip, IRQ_MMP2_RTC_BASE, 2);
diff --git a/arch/arm/mach-mmp/irq-pxa168.c b/arch/arm/mach-mmp/irq-pxa168.c
index 52ff2f0..f86b450 100644
--- a/arch/arm/mach-mmp/irq-pxa168.c
+++ b/arch/arm/mach-mmp/irq-pxa168.c
@@ -25,21 +25,21 @@
#define PRIORITY_DEFAULT 0x1
#define PRIORITY_NONE 0x0 /* means IRQ disabled */
-static void icu_mask_irq(unsigned int irq)
+static void icu_mask_irq(struct irq_data *d)
{
- __raw_writel(PRIORITY_NONE, ICU_INT_CONF(irq));
+ __raw_writel(PRIORITY_NONE, ICU_INT_CONF(d->irq));
}
-static void icu_unmask_irq(unsigned int irq)
+static void icu_unmask_irq(struct irq_data *d)
{
- __raw_writel(IRQ_ROUTE_TO_AP | PRIORITY_DEFAULT, ICU_INT_CONF(irq));
+ __raw_writel(IRQ_ROUTE_TO_AP | PRIORITY_DEFAULT, ICU_INT_CONF(d->irq));
}
static struct irq_chip icu_irq_chip = {
- .name = "icu_irq",
- .ack = icu_mask_irq,
- .mask = icu_mask_irq,
- .unmask = icu_unmask_irq,
+ .name = "icu_irq",
+ .irq_ack = icu_mask_irq,
+ .irq_mask = icu_mask_irq,
+ .irq_unmask = icu_unmask_irq,
};
void __init icu_init_irq(void)
@@ -47,7 +47,7 @@
int irq;
for (irq = 0; irq < 64; irq++) {
- icu_mask_irq(irq);
+ icu_mask_irq(irq_get_irq_data(irq));
set_irq_chip(irq, &icu_irq_chip);
set_irq_handler(irq, handle_level_irq);
set_irq_flags(irq, IRQF_VALID);
diff --git a/arch/arm/mach-msm/board-trout-gpio.c b/arch/arm/mach-msm/board-trout-gpio.c
index f8c09ef..a604ec1 100644
--- a/arch/arm/mach-msm/board-trout-gpio.c
+++ b/arch/arm/mach-msm/board-trout-gpio.c
@@ -113,52 +113,52 @@
TROUT_GPIO_BANK("VIRTUAL", 0x12, TROUT_GPIO_VIRTUAL_BASE, 0),
};
-static void trout_gpio_irq_ack(unsigned int irq)
+static void trout_gpio_irq_ack(struct irq_data *d)
{
- int bank = TROUT_INT_TO_BANK(irq);
- uint8_t mask = TROUT_INT_TO_MASK(irq);
+ int bank = TROUT_INT_TO_BANK(d->irq);
+ uint8_t mask = TROUT_INT_TO_MASK(d->irq);
int reg = TROUT_BANK_TO_STAT_REG(bank);
- /*printk(KERN_INFO "trout_gpio_irq_ack irq %d\n", irq);*/
+ /*printk(KERN_INFO "trout_gpio_irq_ack irq %d\n", d->irq);*/
writeb(mask, TROUT_CPLD_BASE + reg);
}
-static void trout_gpio_irq_mask(unsigned int irq)
+static void trout_gpio_irq_mask(struct irq_data *d)
{
unsigned long flags;
uint8_t reg_val;
- int bank = TROUT_INT_TO_BANK(irq);
- uint8_t mask = TROUT_INT_TO_MASK(irq);
+ int bank = TROUT_INT_TO_BANK(d->irq);
+ uint8_t mask = TROUT_INT_TO_MASK(d->irq);
int reg = TROUT_BANK_TO_MASK_REG(bank);
local_irq_save(flags);
reg_val = trout_int_mask[bank] |= mask;
/*printk(KERN_INFO "trout_gpio_irq_mask irq %d => %d:%02x\n",
- irq, bank, reg_val);*/
+ d->irq, bank, reg_val);*/
writeb(reg_val, TROUT_CPLD_BASE + reg);
local_irq_restore(flags);
}
-static void trout_gpio_irq_unmask(unsigned int irq)
+static void trout_gpio_irq_unmask(struct irq_data *d)
{
unsigned long flags;
uint8_t reg_val;
- int bank = TROUT_INT_TO_BANK(irq);
- uint8_t mask = TROUT_INT_TO_MASK(irq);
+ int bank = TROUT_INT_TO_BANK(d->irq);
+ uint8_t mask = TROUT_INT_TO_MASK(d->irq);
int reg = TROUT_BANK_TO_MASK_REG(bank);
local_irq_save(flags);
reg_val = trout_int_mask[bank] &= ~mask;
/*printk(KERN_INFO "trout_gpio_irq_unmask irq %d => %d:%02x\n",
- irq, bank, reg_val);*/
+ d->irq, bank, reg_val);*/
writeb(reg_val, TROUT_CPLD_BASE + reg);
local_irq_restore(flags);
}
-int trout_gpio_irq_set_wake(unsigned int irq, unsigned int on)
+int trout_gpio_irq_set_wake(struct irq_data *d, unsigned int on)
{
unsigned long flags;
- int bank = TROUT_INT_TO_BANK(irq);
- uint8_t mask = TROUT_INT_TO_MASK(irq);
+ int bank = TROUT_INT_TO_BANK(d->irq);
+ uint8_t mask = TROUT_INT_TO_MASK(d->irq);
local_irq_save(flags);
if(on)
@@ -198,15 +198,15 @@
}
int_base += TROUT_INT_BANK0_COUNT;
}
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
}
static struct irq_chip trout_gpio_irq_chip = {
- .name = "troutgpio",
- .ack = trout_gpio_irq_ack,
- .mask = trout_gpio_irq_mask,
- .unmask = trout_gpio_irq_unmask,
- .set_wake = trout_gpio_irq_set_wake,
+ .name = "troutgpio",
+ .irq_ack = trout_gpio_irq_ack,
+ .irq_mask = trout_gpio_irq_mask,
+ .irq_unmask = trout_gpio_irq_unmask,
+ .irq_set_wake = trout_gpio_irq_set_wake,
};
/*
diff --git a/arch/arm/mach-msm/gpio.c b/arch/arm/mach-msm/gpio.c
index 33051b5..176af9d 100644
--- a/arch/arm/mach-msm/gpio.c
+++ b/arch/arm/mach-msm/gpio.c
@@ -225,21 +225,21 @@
#endif
};
-static void msm_gpio_irq_ack(unsigned int irq)
+static void msm_gpio_irq_ack(struct irq_data *d)
{
unsigned long irq_flags;
- struct msm_gpio_chip *msm_chip = get_irq_chip_data(irq);
+ struct msm_gpio_chip *msm_chip = irq_data_get_irq_chip_data(d);
spin_lock_irqsave(&msm_chip->lock, irq_flags);
msm_gpio_clear_detect_status(msm_chip,
- irq - gpio_to_irq(msm_chip->chip.base));
+ d->irq - gpio_to_irq(msm_chip->chip.base));
spin_unlock_irqrestore(&msm_chip->lock, irq_flags);
}
-static void msm_gpio_irq_mask(unsigned int irq)
+static void msm_gpio_irq_mask(struct irq_data *d)
{
unsigned long irq_flags;
- struct msm_gpio_chip *msm_chip = get_irq_chip_data(irq);
- unsigned offset = irq - gpio_to_irq(msm_chip->chip.base);
+ struct msm_gpio_chip *msm_chip = irq_data_get_irq_chip_data(d);
+ unsigned offset = d->irq - gpio_to_irq(msm_chip->chip.base);
spin_lock_irqsave(&msm_chip->lock, irq_flags);
/* level triggered interrupts are also latched */
@@ -250,11 +250,11 @@
spin_unlock_irqrestore(&msm_chip->lock, irq_flags);
}
-static void msm_gpio_irq_unmask(unsigned int irq)
+static void msm_gpio_irq_unmask(struct irq_data *d)
{
unsigned long irq_flags;
- struct msm_gpio_chip *msm_chip = get_irq_chip_data(irq);
- unsigned offset = irq - gpio_to_irq(msm_chip->chip.base);
+ struct msm_gpio_chip *msm_chip = irq_data_get_irq_chip_data(d);
+ unsigned offset = d->irq - gpio_to_irq(msm_chip->chip.base);
spin_lock_irqsave(&msm_chip->lock, irq_flags);
/* level triggered interrupts are also latched */
@@ -265,11 +265,11 @@
spin_unlock_irqrestore(&msm_chip->lock, irq_flags);
}
-static int msm_gpio_irq_set_wake(unsigned int irq, unsigned int on)
+static int msm_gpio_irq_set_wake(struct irq_data *d, unsigned int on)
{
unsigned long irq_flags;
- struct msm_gpio_chip *msm_chip = get_irq_chip_data(irq);
- unsigned offset = irq - gpio_to_irq(msm_chip->chip.base);
+ struct msm_gpio_chip *msm_chip = irq_data_get_irq_chip_data(d);
+ unsigned offset = d->irq - gpio_to_irq(msm_chip->chip.base);
spin_lock_irqsave(&msm_chip->lock, irq_flags);
@@ -282,21 +282,21 @@
return 0;
}
-static int msm_gpio_irq_set_type(unsigned int irq, unsigned int flow_type)
+static int msm_gpio_irq_set_type(struct irq_data *d, unsigned int flow_type)
{
unsigned long irq_flags;
- struct msm_gpio_chip *msm_chip = get_irq_chip_data(irq);
- unsigned offset = irq - gpio_to_irq(msm_chip->chip.base);
+ struct msm_gpio_chip *msm_chip = irq_data_get_irq_chip_data(d);
+ unsigned offset = d->irq - gpio_to_irq(msm_chip->chip.base);
unsigned val, mask = BIT(offset);
spin_lock_irqsave(&msm_chip->lock, irq_flags);
val = readl(msm_chip->regs.int_edge);
if (flow_type & IRQ_TYPE_EDGE_BOTH) {
writel(val | mask, msm_chip->regs.int_edge);
- irq_desc[irq].handle_irq = handle_edge_irq;
+ irq_desc[d->irq].handle_irq = handle_edge_irq;
} else {
writel(val & ~mask, msm_chip->regs.int_edge);
- irq_desc[irq].handle_irq = handle_level_irq;
+ irq_desc[d->irq].handle_irq = handle_level_irq;
}
if ((flow_type & IRQ_TYPE_EDGE_BOTH) == IRQ_TYPE_EDGE_BOTH) {
msm_chip->both_edge_detect |= mask;
@@ -333,16 +333,16 @@
msm_chip->chip.base + j);
}
}
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
}
static struct irq_chip msm_gpio_irq_chip = {
- .name = "msmgpio",
- .ack = msm_gpio_irq_ack,
- .mask = msm_gpio_irq_mask,
- .unmask = msm_gpio_irq_unmask,
- .set_wake = msm_gpio_irq_set_wake,
- .set_type = msm_gpio_irq_set_type,
+ .name = "msmgpio",
+ .irq_ack = msm_gpio_irq_ack,
+ .irq_mask = msm_gpio_irq_mask,
+ .irq_unmask = msm_gpio_irq_unmask,
+ .irq_set_wake = msm_gpio_irq_set_wake,
+ .irq_set_type = msm_gpio_irq_set_type,
};
static int __init msm_init_gpio(void)
diff --git a/arch/arm/mach-msm/irq-vic.c b/arch/arm/mach-msm/irq-vic.c
index 99f2c34..68c28bb 100644
--- a/arch/arm/mach-msm/irq-vic.c
+++ b/arch/arm/mach-msm/irq-vic.c
@@ -226,19 +226,18 @@
writel(val, base + (i * 4));
}
-static void msm_irq_ack(unsigned int irq)
+static void msm_irq_ack(struct irq_data *d)
{
- void __iomem *reg = VIC_INT_TO_REG_ADDR(VIC_INT_CLEAR0, irq);
- irq = 1 << (irq & 31);
- writel(irq, reg);
+ void __iomem *reg = VIC_INT_TO_REG_ADDR(VIC_INT_CLEAR0, d->irq);
+ writel(1 << (d->irq & 31), reg);
}
-static void msm_irq_mask(unsigned int irq)
+static void msm_irq_mask(struct irq_data *d)
{
- void __iomem *reg = VIC_INT_TO_REG_ADDR(VIC_INT_ENCLEAR0, irq);
- unsigned index = VIC_INT_TO_REG_INDEX(irq);
- uint32_t mask = 1UL << (irq & 31);
- int smsm_irq = msm_irq_to_smsm[irq];
+ void __iomem *reg = VIC_INT_TO_REG_ADDR(VIC_INT_ENCLEAR0, d->irq);
+ unsigned index = VIC_INT_TO_REG_INDEX(d->irq);
+ uint32_t mask = 1UL << (d->irq & 31);
+ int smsm_irq = msm_irq_to_smsm[d->irq];
msm_irq_shadow_reg[index].int_en[0] &= ~mask;
writel(mask, reg);
@@ -250,12 +249,12 @@
}
}
-static void msm_irq_unmask(unsigned int irq)
+static void msm_irq_unmask(struct irq_data *d)
{
- void __iomem *reg = VIC_INT_TO_REG_ADDR(VIC_INT_ENSET0, irq);
- unsigned index = VIC_INT_TO_REG_INDEX(irq);
- uint32_t mask = 1UL << (irq & 31);
- int smsm_irq = msm_irq_to_smsm[irq];
+ void __iomem *reg = VIC_INT_TO_REG_ADDR(VIC_INT_ENSET0, d->irq);
+ unsigned index = VIC_INT_TO_REG_INDEX(d->irq);
+ uint32_t mask = 1UL << (d->irq & 31);
+ int smsm_irq = msm_irq_to_smsm[d->irq];
msm_irq_shadow_reg[index].int_en[0] |= mask;
writel(mask, reg);
@@ -268,14 +267,14 @@
}
}
-static int msm_irq_set_wake(unsigned int irq, unsigned int on)
+static int msm_irq_set_wake(struct irq_data *d, unsigned int on)
{
- unsigned index = VIC_INT_TO_REG_INDEX(irq);
- uint32_t mask = 1UL << (irq & 31);
- int smsm_irq = msm_irq_to_smsm[irq];
+ unsigned index = VIC_INT_TO_REG_INDEX(d->irq);
+ uint32_t mask = 1UL << (d->irq & 31);
+ int smsm_irq = msm_irq_to_smsm[d->irq];
if (smsm_irq == 0) {
- printk(KERN_ERR "msm_irq_set_wake: bad wakeup irq %d\n", irq);
+ printk(KERN_ERR "msm_irq_set_wake: bad wakeup irq %d\n", d->irq);
return -EINVAL;
}
if (on)
@@ -294,12 +293,12 @@
return 0;
}
-static int msm_irq_set_type(unsigned int irq, unsigned int flow_type)
+static int msm_irq_set_type(struct irq_data *d, unsigned int flow_type)
{
- void __iomem *treg = VIC_INT_TO_REG_ADDR(VIC_INT_TYPE0, irq);
- void __iomem *preg = VIC_INT_TO_REG_ADDR(VIC_INT_POLARITY0, irq);
- unsigned index = VIC_INT_TO_REG_INDEX(irq);
- int b = 1 << (irq & 31);
+ void __iomem *treg = VIC_INT_TO_REG_ADDR(VIC_INT_TYPE0, d->irq);
+ void __iomem *preg = VIC_INT_TO_REG_ADDR(VIC_INT_POLARITY0, d->irq);
+ unsigned index = VIC_INT_TO_REG_INDEX(d->irq);
+ int b = 1 << (d->irq & 31);
uint32_t polarity;
uint32_t type;
@@ -314,11 +313,11 @@
type = msm_irq_shadow_reg[index].int_type;
if (flow_type & (IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING)) {
type |= b;
- irq_desc[irq].handle_irq = handle_edge_irq;
+ irq_desc[d->irq].handle_irq = handle_edge_irq;
}
if (flow_type & (IRQF_TRIGGER_HIGH | IRQF_TRIGGER_LOW)) {
type &= ~b;
- irq_desc[irq].handle_irq = handle_level_irq;
+ irq_desc[d->irq].handle_irq = handle_level_irq;
}
writel(type, treg);
msm_irq_shadow_reg[index].int_type = type;
@@ -326,13 +325,13 @@
}
static struct irq_chip msm_irq_chip = {
- .name = "msm",
- .disable = msm_irq_mask,
- .ack = msm_irq_ack,
- .mask = msm_irq_mask,
- .unmask = msm_irq_unmask,
- .set_wake = msm_irq_set_wake,
- .set_type = msm_irq_set_type,
+ .name = "msm",
+ .irq_disable = msm_irq_mask,
+ .irq_ack = msm_irq_ack,
+ .irq_mask = msm_irq_mask,
+ .irq_unmask = msm_irq_unmask,
+ .irq_set_wake = msm_irq_set_wake,
+ .irq_set_type = msm_irq_set_type,
};
void __init msm_init_irq(void)
diff --git a/arch/arm/mach-msm/irq.c b/arch/arm/mach-msm/irq.c
index 6c8d5f8..0b27d89 100644
--- a/arch/arm/mach-msm/irq.c
+++ b/arch/arm/mach-msm/irq.c
@@ -64,35 +64,34 @@
#define VIC_VECTPRIORITY(n) VIC_REG(0x0200+((n) * 4))
#define VIC_VECTADDR(n) VIC_REG(0x0400+((n) * 4))
-static void msm_irq_ack(unsigned int irq)
+static void msm_irq_ack(struct irq_data *d)
{
- void __iomem *reg = VIC_INT_CLEAR0 + ((irq & 32) ? 4 : 0);
- irq = 1 << (irq & 31);
- writel(irq, reg);
+ void __iomem *reg = VIC_INT_CLEAR0 + ((d->irq & 32) ? 4 : 0);
+ writel(1 << (d->irq & 31), reg);
}
-static void msm_irq_mask(unsigned int irq)
+static void msm_irq_mask(struct irq_data *d)
{
- void __iomem *reg = VIC_INT_ENCLEAR0 + ((irq & 32) ? 4 : 0);
- writel(1 << (irq & 31), reg);
+ void __iomem *reg = VIC_INT_ENCLEAR0 + ((d->irq & 32) ? 4 : 0);
+ writel(1 << (d->irq & 31), reg);
}
-static void msm_irq_unmask(unsigned int irq)
+static void msm_irq_unmask(struct irq_data *d)
{
- void __iomem *reg = VIC_INT_ENSET0 + ((irq & 32) ? 4 : 0);
- writel(1 << (irq & 31), reg);
+ void __iomem *reg = VIC_INT_ENSET0 + ((d->irq & 32) ? 4 : 0);
+ writel(1 << (d->irq & 31), reg);
}
-static int msm_irq_set_wake(unsigned int irq, unsigned int on)
+static int msm_irq_set_wake(struct irq_data *d, unsigned int on)
{
return -EINVAL;
}
-static int msm_irq_set_type(unsigned int irq, unsigned int flow_type)
+static int msm_irq_set_type(struct irq_data *d, unsigned int flow_type)
{
- void __iomem *treg = VIC_INT_TYPE0 + ((irq & 32) ? 4 : 0);
- void __iomem *preg = VIC_INT_POLARITY0 + ((irq & 32) ? 4 : 0);
- int b = 1 << (irq & 31);
+ void __iomem *treg = VIC_INT_TYPE0 + ((d->irq & 32) ? 4 : 0);
+ void __iomem *preg = VIC_INT_POLARITY0 + ((d->irq & 32) ? 4 : 0);
+ int b = 1 << (d->irq & 31);
if (flow_type & (IRQF_TRIGGER_FALLING | IRQF_TRIGGER_LOW))
writel(readl(preg) | b, preg);
@@ -101,22 +100,22 @@
if (flow_type & (IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING)) {
writel(readl(treg) | b, treg);
- irq_desc[irq].handle_irq = handle_edge_irq;
+ irq_desc[d->irq].handle_irq = handle_edge_irq;
}
if (flow_type & (IRQF_TRIGGER_HIGH | IRQF_TRIGGER_LOW)) {
writel(readl(treg) & (~b), treg);
- irq_desc[irq].handle_irq = handle_level_irq;
+ irq_desc[d->irq].handle_irq = handle_level_irq;
}
return 0;
}
static struct irq_chip msm_irq_chip = {
- .name = "msm",
- .ack = msm_irq_ack,
- .mask = msm_irq_mask,
- .unmask = msm_irq_unmask,
- .set_wake = msm_irq_set_wake,
- .set_type = msm_irq_set_type,
+ .name = "msm",
+ .irq_ack = msm_irq_ack,
+ .irq_mask = msm_irq_mask,
+ .irq_unmask = msm_irq_unmask,
+ .irq_set_wake = msm_irq_set_wake,
+ .irq_set_type = msm_irq_set_type,
};
void __init msm_init_irq(void)
diff --git a/arch/arm/mach-msm/sirc.c b/arch/arm/mach-msm/sirc.c
index 152eefd..11b54c7 100644
--- a/arch/arm/mach-msm/sirc.c
+++ b/arch/arm/mach-msm/sirc.c
@@ -42,12 +42,11 @@
/* Mask off the given interrupt. Keep the int_enable mask in sync with
the enable reg, so it can be restored after power collapse. */
-static void sirc_irq_mask(unsigned int irq)
+static void sirc_irq_mask(struct irq_data *d)
{
unsigned int mask;
-
- mask = 1 << (irq - FIRST_SIRC_IRQ);
+ mask = 1 << (d->irq - FIRST_SIRC_IRQ);
writel(mask, sirc_regs.int_enable_clear);
int_enable &= ~mask;
return;
@@ -55,31 +54,31 @@
/* Unmask the given interrupt. Keep the int_enable mask in sync with
the enable reg, so it can be restored after power collapse. */
-static void sirc_irq_unmask(unsigned int irq)
+static void sirc_irq_unmask(struct irq_data *d)
{
unsigned int mask;
- mask = 1 << (irq - FIRST_SIRC_IRQ);
+ mask = 1 << (d->irq - FIRST_SIRC_IRQ);
writel(mask, sirc_regs.int_enable_set);
int_enable |= mask;
return;
}
-static void sirc_irq_ack(unsigned int irq)
+static void sirc_irq_ack(struct irq_data *d)
{
unsigned int mask;
- mask = 1 << (irq - FIRST_SIRC_IRQ);
+ mask = 1 << (d->irq - FIRST_SIRC_IRQ);
writel(mask, sirc_regs.int_clear);
return;
}
-static int sirc_irq_set_wake(unsigned int irq, unsigned int on)
+static int sirc_irq_set_wake(struct irq_data *d, unsigned int on)
{
unsigned int mask;
/* Used to set the interrupt enable mask during power collapse. */
- mask = 1 << (irq - FIRST_SIRC_IRQ);
+ mask = 1 << (d->irq - FIRST_SIRC_IRQ);
if (on)
wake_enable |= mask;
else
@@ -88,12 +87,12 @@
return 0;
}
-static int sirc_irq_set_type(unsigned int irq, unsigned int flow_type)
+static int sirc_irq_set_type(struct irq_data *d, unsigned int flow_type)
{
unsigned int mask;
unsigned int val;
- mask = 1 << (irq - FIRST_SIRC_IRQ);
+ mask = 1 << (d->irq - FIRST_SIRC_IRQ);
val = readl(sirc_regs.int_polarity);
if (flow_type & (IRQF_TRIGGER_LOW | IRQF_TRIGGER_FALLING))
@@ -106,10 +105,10 @@
val = readl(sirc_regs.int_type);
if (flow_type & (IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING)) {
val |= mask;
- irq_desc[irq].handle_irq = handle_edge_irq;
+ irq_desc[d->irq].handle_irq = handle_edge_irq;
} else {
val &= ~mask;
- irq_desc[irq].handle_irq = handle_level_irq;
+ irq_desc[d->irq].handle_irq = handle_level_irq;
}
writel(val, sirc_regs.int_type);
@@ -139,16 +138,16 @@
;
generic_handle_irq(sirq+FIRST_SIRC_IRQ);
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
}
static struct irq_chip sirc_irq_chip = {
- .name = "sirc",
- .ack = sirc_irq_ack,
- .mask = sirc_irq_mask,
- .unmask = sirc_irq_unmask,
- .set_wake = sirc_irq_set_wake,
- .set_type = sirc_irq_set_type,
+ .name = "sirc",
+ .irq_ack = sirc_irq_ack,
+ .irq_mask = sirc_irq_mask,
+ .irq_unmask = sirc_irq_unmask,
+ .irq_set_wake = sirc_irq_set_wake,
+ .irq_set_type = sirc_irq_set_type,
};
void __init msm_init_sirc(void)
diff --git a/arch/arm/mach-mx3/mach-mx31_3ds.c b/arch/arm/mach-mx3/mach-mx31_3ds.c
index 899a969..0d65db8 100644
--- a/arch/arm/mach-mx3/mach-mx31_3ds.c
+++ b/arch/arm/mach-mx3/mach-mx31_3ds.c
@@ -147,10 +147,10 @@
.init_data = &pwgtx_init,
}, {
- .id = MC13783_REGU_GPO1, /* Turn on 1.8V */
+ .id = MC13783_REG_GPO1, /* Turn on 1.8V */
.init_data = &gpo_init,
}, {
- .id = MC13783_REGU_GPO3, /* Turn on 3.3V */
+ .id = MC13783_REG_GPO3, /* Turn on 3.3V */
.init_data = &gpo_init,
},
};
diff --git a/arch/arm/mach-mx3/mach-mx31ads.c b/arch/arm/mach-mx3/mach-mx31ads.c
index b993b9b..88b97d6 100644
--- a/arch/arm/mach-mx3/mach-mx31ads.c
+++ b/arch/arm/mach-mx3/mach-mx31ads.c
@@ -162,9 +162,9 @@
* Disable an expio pin's interrupt by setting the bit in the imr.
* @param irq an expio virtual irq number
*/
-static void expio_mask_irq(u32 irq)
+static void expio_mask_irq(struct irq_data *d)
{
- u32 expio = MXC_IRQ_TO_EXPIO(irq);
+ u32 expio = MXC_IRQ_TO_EXPIO(d->irq);
/* mask the interrupt */
__raw_writew(1 << expio, PBC_INTMASK_CLEAR_REG);
__raw_readw(PBC_INTMASK_CLEAR_REG);
@@ -174,9 +174,9 @@
* Acknowledge an expanded io pin's interrupt by clearing the bit in the isr.
* @param irq an expanded io virtual irq number
*/
-static void expio_ack_irq(u32 irq)
+static void expio_ack_irq(struct irq_data *d)
{
- u32 expio = MXC_IRQ_TO_EXPIO(irq);
+ u32 expio = MXC_IRQ_TO_EXPIO(d->irq);
/* clear the interrupt status */
__raw_writew(1 << expio, PBC_INTSTATUS_REG);
}
@@ -185,18 +185,18 @@
* Enable a expio pin's interrupt by clearing the bit in the imr.
* @param irq a expio virtual irq number
*/
-static void expio_unmask_irq(u32 irq)
+static void expio_unmask_irq(struct irq_data *d)
{
- u32 expio = MXC_IRQ_TO_EXPIO(irq);
+ u32 expio = MXC_IRQ_TO_EXPIO(d->irq);
/* unmask the interrupt */
__raw_writew(1 << expio, PBC_INTMASK_SET_REG);
}
static struct irq_chip expio_irq_chip = {
.name = "EXPIO(CPLD)",
- .ack = expio_ack_irq,
- .mask = expio_mask_irq,
- .unmask = expio_unmask_irq,
+ .irq_ack = expio_ack_irq,
+ .irq_mask = expio_mask_irq,
+ .irq_unmask = expio_unmask_irq,
};
static void __init mx31ads_init_expio(void)
diff --git a/arch/arm/mach-mx5/Kconfig b/arch/arm/mach-mx5/Kconfig
index 55254b6..de4fa992f 100644
--- a/arch/arm/mach-mx5/Kconfig
+++ b/arch/arm/mach-mx5/Kconfig
@@ -50,6 +50,7 @@
config MACH_MX51_3DS
bool "Support MX51PDK (3DS)"
select SOC_IMX51
+ select IMX_HAVE_PLATFORM_IMX_KEYPAD
select IMX_HAVE_PLATFORM_IMX_UART
select IMX_HAVE_PLATFORM_SDHCI_ESDHC_IMX
select IMX_HAVE_PLATFORM_SPI_IMX
@@ -77,6 +78,7 @@
config MACH_EUKREA_MBIMX51_BASEBOARD
prompt "Eukrea MBIMX51 development board"
bool
+ select IMX_HAVE_PLATFORM_IMX_KEYPAD
select IMX_HAVE_PLATFORM_SDHCI_ESDHC_IMX
help
This adds board specific devices that can be found on Eukrea's
@@ -124,10 +126,28 @@
bool "Support MX53 EVK platforms"
select SOC_IMX53
select IMX_HAVE_PLATFORM_IMX_UART
+ select IMX_HAVE_PLATFORM_IMX_I2C
+ select IMX_HAVE_PLATFORM_SDHCI_ESDHC_IMX
+ select IMX_HAVE_PLATFORM_SPI_IMX
help
Include support for MX53 EVK platform. This includes specific
configurations for the board and its peripherals.
+config MACH_MX53_SMD
+ bool "Support MX53 SMD platforms"
+ select SOC_IMX53
+ select IMX_HAVE_PLATFORM_IMX_UART
+ help
+ Include support for MX53 SMD platform. This includes specific
+ configurations for the board and its peripherals.
+
+config MACH_MX53_LOCO
+ bool "Support MX53 LOCO platforms"
+ select SOC_IMX53
+ select IMX_HAVE_PLATFORM_IMX_UART
+ help
+ Include support for MX53 LOCO platform. This includes specific
+ configurations for the board and its peripherals.
config MACH_MX50_RDP
bool "Support MX50 reference design platform"
diff --git a/arch/arm/mach-mx5/Makefile b/arch/arm/mach-mx5/Makefile
index 0c398ba..0d43be9 100644
--- a/arch/arm/mach-mx5/Makefile
+++ b/arch/arm/mach-mx5/Makefile
@@ -10,6 +10,8 @@
obj-$(CONFIG_MACH_MX51_BABBAGE) += board-mx51_babbage.o
obj-$(CONFIG_MACH_MX51_3DS) += board-mx51_3ds.o
obj-$(CONFIG_MACH_MX53_EVK) += board-mx53_evk.o
+obj-$(CONFIG_MACH_MX53_SMD) += board-mx53_smd.o
+obj-$(CONFIG_MACH_MX53_LOCO) += board-mx53_loco.o
obj-$(CONFIG_MACH_EUKREA_CPUIMX51) += board-cpuimx51.o
obj-$(CONFIG_MACH_EUKREA_MBIMX51_BASEBOARD) += eukrea_mbimx51-baseboard.o
obj-$(CONFIG_MACH_EUKREA_CPUIMX51SD) += board-cpuimx51sd.o
diff --git a/arch/arm/mach-mx5/board-mx51_3ds.c b/arch/arm/mach-mx5/board-mx51_3ds.c
index e42bd2e..49d6448 100644
--- a/arch/arm/mach-mx5/board-mx51_3ds.c
+++ b/arch/arm/mach-mx5/board-mx51_3ds.c
@@ -12,7 +12,6 @@
#include <linux/irq.h>
#include <linux/platform_device.h>
-#include <linux/input/matrix_keypad.h>
#include <linux/spi/spi.h>
#include <asm/mach-types.h>
@@ -120,14 +119,14 @@
KEY(3, 5, KEY_BACK)
};
-static struct matrix_keymap_data mx51_3ds_map_data = {
+static const struct matrix_keymap_data mx51_3ds_map_data __initconst = {
.keymap = mx51_3ds_board_keymap,
.keymap_size = ARRAY_SIZE(mx51_3ds_board_keymap),
};
static void mxc_init_keypad(void)
{
- mxc_register_device(&mxc_keypad_device, &mx51_3ds_map_data);
+ imx51_add_imx_keypad(&mx51_3ds_map_data);
}
#else
static inline void mxc_init_keypad(void)
diff --git a/arch/arm/mach-mx5/board-mx53_evk.c b/arch/arm/mach-mx5/board-mx53_evk.c
index fa97d0d..caee04c 100644
--- a/arch/arm/mach-mx5/board-mx53_evk.c
+++ b/arch/arm/mach-mx5/board-mx53_evk.c
@@ -21,6 +21,11 @@
#include <linux/init.h>
#include <linux/clk.h>
+#include <linux/fec.h>
+#include <linux/delay.h>
+#include <linux/gpio.h>
+#include <linux/spi/flash.h>
+#include <linux/spi/spi.h>
#include <mach/common.h>
#include <mach/hardware.h>
#include <asm/mach-types.h>
@@ -29,6 +34,10 @@
#include <mach/imx-uart.h>
#include <mach/iomux-mx53.h>
+#define SMD_FEC_PHY_RST IMX_GPIO_NR(7, 6)
+#define EVK_ECSPI1_CS0 IMX_GPIO_NR(2, 30)
+#define EVK_ECSPI1_CS1 IMX_GPIO_NR(3, 19)
+
#include "crm_regs.h"
#include "devices-imx53.h"
@@ -47,6 +56,14 @@
MX53_PAD_ATA_CS_1__UART3_RXD,
MX53_PAD_ATA_DA_1__UART3_CTS,
MX53_PAD_ATA_DA_2__UART3_RTS,
+
+ MX53_PAD_EIM_D16__CSPI1_SCLK,
+ MX53_PAD_EIM_D17__CSPI1_MISO,
+ MX53_PAD_EIM_D18__CSPI1_MOSI,
+
+ /* ecspi chip select lines */
+ MX53_PAD_EIM_EB2__GPIO_2_30,
+ MX53_PAD_EIM_D19__GPIO_3_19,
};
static const struct imxuart_platform_data mx53_evk_uart_pdata __initconst = {
@@ -60,11 +77,68 @@
imx53_add_imx_uart(2, &mx53_evk_uart_pdata);
}
+static const struct imxi2c_platform_data mx53_evk_i2c_data __initconst = {
+ .bitrate = 100000,
+};
+
+static inline void mx53_evk_fec_reset(void)
+{
+ int ret;
+
+ /* reset FEC PHY */
+ ret = gpio_request(SMD_FEC_PHY_RST, "fec-phy-reset");
+ if (ret) {
+ printk(KERN_ERR"failed to get GPIO_FEC_PHY_RESET: %d\n", ret);
+ return;
+ }
+ gpio_direction_output(SMD_FEC_PHY_RST, 0);
+ gpio_set_value(SMD_FEC_PHY_RST, 0);
+ msleep(1);
+ gpio_set_value(SMD_FEC_PHY_RST, 1);
+}
+
+static struct fec_platform_data mx53_evk_fec_pdata = {
+ .phy = PHY_INTERFACE_MODE_RMII,
+};
+
+static struct spi_board_info mx53_evk_spi_board_info[] __initdata = {
+ {
+ .modalias = "mtd_dataflash",
+ .max_speed_hz = 25000000,
+ .bus_num = 0,
+ .chip_select = 1,
+ .mode = SPI_MODE_0,
+ .platform_data = NULL,
+ },
+};
+
+static int mx53_evk_spi_cs[] = {
+ EVK_ECSPI1_CS0,
+ EVK_ECSPI1_CS1,
+};
+
+static const struct spi_imx_master mx53_evk_spi_data __initconst = {
+ .chipselect = mx53_evk_spi_cs,
+ .num_chipselect = ARRAY_SIZE(mx53_evk_spi_cs),
+};
+
static void __init mx53_evk_board_init(void)
{
mxc_iomux_v3_setup_multiple_pads(mx53_evk_pads,
ARRAY_SIZE(mx53_evk_pads));
mx53_evk_init_uart();
+ mx53_evk_fec_reset();
+ imx53_add_fec(&mx53_evk_fec_pdata);
+
+ imx53_add_imx_i2c(0, &mx53_evk_i2c_data);
+ imx53_add_imx_i2c(1, &mx53_evk_i2c_data);
+
+ imx53_add_sdhci_esdhc_imx(0, NULL);
+ imx53_add_sdhci_esdhc_imx(1, NULL);
+
+ spi_register_board_info(mx53_evk_spi_board_info,
+ ARRAY_SIZE(mx53_evk_spi_board_info));
+ imx53_add_ecspi(0, &mx53_evk_spi_data);
}
static void __init mx53_evk_timer_init(void)
diff --git a/arch/arm/mach-mx5/board-mx53_loco.c b/arch/arm/mach-mx5/board-mx53_loco.c
new file mode 100644
index 0000000..d1348e0
--- /dev/null
+++ b/arch/arm/mach-mx5/board-mx53_loco.c
@@ -0,0 +1,111 @@
+/*
+ * Copyright (C) 2011 Freescale Semiconductor, Inc. All Rights Reserved.
+ */
+
+/*
+ * 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.
+
+ * 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, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#include <linux/init.h>
+#include <linux/clk.h>
+#include <linux/fec.h>
+#include <linux/delay.h>
+#include <linux/gpio.h>
+
+#include <mach/common.h>
+#include <mach/hardware.h>
+#include <mach/imx-uart.h>
+#include <mach/iomux-mx53.h>
+
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/time.h>
+
+#include "crm_regs.h"
+#include "devices-imx53.h"
+
+#define LOCO_FEC_PHY_RST IMX_GPIO_NR(7, 6)
+
+static iomux_v3_cfg_t mx53_loco_pads[] = {
+ MX53_PAD_CSI0_D10__UART1_TXD,
+ MX53_PAD_CSI0_D11__UART1_RXD,
+ MX53_PAD_ATA_DIOW__UART1_TXD,
+ MX53_PAD_ATA_DMACK__UART1_RXD,
+
+ MX53_PAD_ATA_BUFFER_EN__UART2_RXD,
+ MX53_PAD_ATA_DMARQ__UART2_TXD,
+ MX53_PAD_ATA_DIOR__UART2_RTS,
+ MX53_PAD_ATA_INTRQ__UART2_CTS,
+
+ MX53_PAD_ATA_CS_0__UART3_TXD,
+ MX53_PAD_ATA_CS_1__UART3_RXD,
+ MX53_PAD_ATA_DA_1__UART3_CTS,
+ MX53_PAD_ATA_DA_2__UART3_RTS,
+};
+
+static const struct imxuart_platform_data mx53_loco_uart_data __initconst = {
+ .flags = IMXUART_HAVE_RTSCTS,
+};
+
+static inline void mx53_loco_init_uart(void)
+{
+ imx53_add_imx_uart(0, &mx53_loco_uart_data);
+ imx53_add_imx_uart(1, &mx53_loco_uart_data);
+ imx53_add_imx_uart(2, &mx53_loco_uart_data);
+}
+
+static inline void mx53_loco_fec_reset(void)
+{
+ int ret;
+
+ /* reset FEC PHY */
+ ret = gpio_request(LOCO_FEC_PHY_RST, "fec-phy-reset");
+ if (ret) {
+ printk(KERN_ERR"failed to get GPIO_FEC_PHY_RESET: %d\n", ret);
+ return;
+ }
+ gpio_direction_output(LOCO_FEC_PHY_RST, 0);
+ msleep(1);
+ gpio_set_value(LOCO_FEC_PHY_RST, 1);
+}
+
+static struct fec_platform_data mx53_loco_fec_data = {
+ .phy = PHY_INTERFACE_MODE_RMII,
+};
+
+static void __init mx53_loco_board_init(void)
+{
+ mxc_iomux_v3_setup_multiple_pads(mx53_loco_pads,
+ ARRAY_SIZE(mx53_loco_pads));
+ mx53_loco_init_uart();
+ mx53_loco_fec_reset();
+ imx53_add_fec(&mx53_loco_fec_data);
+}
+
+static void __init mx53_loco_timer_init(void)
+{
+ mx53_clocks_init(32768, 24000000, 0, 0);
+}
+
+static struct sys_timer mx53_loco_timer = {
+ .init = mx53_loco_timer_init,
+};
+
+MACHINE_START(MX53_LOCO, "Freescale MX53 LOCO Board")
+ .map_io = mx53_map_io,
+ .init_irq = mx53_init_irq,
+ .init_machine = mx53_loco_board_init,
+ .timer = &mx53_loco_timer,
+MACHINE_END
diff --git a/arch/arm/mach-mx5/board-mx53_smd.c b/arch/arm/mach-mx5/board-mx53_smd.c
new file mode 100644
index 0000000..7970f7a
--- /dev/null
+++ b/arch/arm/mach-mx5/board-mx53_smd.c
@@ -0,0 +1,111 @@
+/*
+ * Copyright (C) 2011 Freescale Semiconductor, Inc. All Rights Reserved.
+ */
+
+/*
+ * 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.
+
+ * 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, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#include <linux/init.h>
+#include <linux/clk.h>
+#include <linux/fec.h>
+#include <linux/delay.h>
+#include <linux/gpio.h>
+
+#include <mach/common.h>
+#include <mach/hardware.h>
+#include <mach/imx-uart.h>
+#include <mach/iomux-mx53.h>
+
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/time.h>
+
+#include "crm_regs.h"
+#include "devices-imx53.h"
+
+#define SMD_FEC_PHY_RST IMX_GPIO_NR(7, 6)
+
+static iomux_v3_cfg_t mx53_smd_pads[] = {
+ MX53_PAD_CSI0_D10__UART1_TXD,
+ MX53_PAD_CSI0_D11__UART1_RXD,
+ MX53_PAD_ATA_DIOW__UART1_TXD,
+ MX53_PAD_ATA_DMACK__UART1_RXD,
+
+ MX53_PAD_ATA_BUFFER_EN__UART2_RXD,
+ MX53_PAD_ATA_DMARQ__UART2_TXD,
+ MX53_PAD_ATA_DIOR__UART2_RTS,
+ MX53_PAD_ATA_INTRQ__UART2_CTS,
+
+ MX53_PAD_ATA_CS_0__UART3_TXD,
+ MX53_PAD_ATA_CS_1__UART3_RXD,
+ MX53_PAD_ATA_DA_1__UART3_CTS,
+ MX53_PAD_ATA_DA_2__UART3_RTS,
+};
+
+static const struct imxuart_platform_data mx53_smd_uart_data __initconst = {
+ .flags = IMXUART_HAVE_RTSCTS,
+};
+
+static inline void mx53_smd_init_uart(void)
+{
+ imx53_add_imx_uart(0, &mx53_smd_uart_data);
+ imx53_add_imx_uart(1, &mx53_smd_uart_data);
+ imx53_add_imx_uart(2, &mx53_smd_uart_data);
+}
+
+static inline void mx53_smd_fec_reset(void)
+{
+ int ret;
+
+ /* reset FEC PHY */
+ ret = gpio_request(SMD_FEC_PHY_RST, "fec-phy-reset");
+ if (ret) {
+ printk(KERN_ERR"failed to get GPIO_FEC_PHY_RESET: %d\n", ret);
+ return;
+ }
+ gpio_direction_output(SMD_FEC_PHY_RST, 0);
+ msleep(1);
+ gpio_set_value(SMD_FEC_PHY_RST, 1);
+}
+
+static struct fec_platform_data mx53_smd_fec_data = {
+ .phy = PHY_INTERFACE_MODE_RMII,
+};
+
+static void __init mx53_smd_board_init(void)
+{
+ mxc_iomux_v3_setup_multiple_pads(mx53_smd_pads,
+ ARRAY_SIZE(mx53_smd_pads));
+ mx53_smd_init_uart();
+ mx53_smd_fec_reset();
+ imx53_add_fec(&mx53_smd_fec_data);
+}
+
+static void __init mx53_smd_timer_init(void)
+{
+ mx53_clocks_init(32768, 24000000, 22579200, 0);
+}
+
+static struct sys_timer mx53_smd_timer = {
+ .init = mx53_smd_timer_init,
+};
+
+MACHINE_START(MX53_SMD, "Freescale MX53 SMD Board")
+ .map_io = mx53_map_io,
+ .init_irq = mx53_init_irq,
+ .init_machine = mx53_smd_board_init,
+ .timer = &mx53_smd_timer,
+MACHINE_END
diff --git a/arch/arm/mach-mx5/clock-mx51-mx53.c b/arch/arm/mach-mx5/clock-mx51-mx53.c
index 785e1a3..0a19e75 100644
--- a/arch/arm/mach-mx5/clock-mx51-mx53.c
+++ b/arch/arm/mach-mx5/clock-mx51-mx53.c
@@ -1191,6 +1191,11 @@
DEFINE_CLOCK(gpt_clk, 0, MXC_CCM_CCGR2, MXC_CCM_CCGRx_CG9_OFFSET,
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);
+DEFINE_CLOCK(pwm2_clk, 0, MXC_CCM_CCGR2, MXC_CCM_CCGRx_CG8_OFFSET,
+ NULL, NULL, &ipg_clk, NULL);
+
/* I2C */
DEFINE_CLOCK(i2c1_clk, 0, MXC_CCM_CCGR1, MXC_CCM_CCGRx_CG9_OFFSET,
NULL, NULL, &ipg_clk, NULL);
@@ -1283,6 +1288,8 @@
_REGISTER_CLOCK("imx-uart.2", NULL, uart3_clk)
_REGISTER_CLOCK(NULL, "gpt", gpt_clk)
_REGISTER_CLOCK("fec.0", NULL, fec_clk)
+ _REGISTER_CLOCK("mxc_pwm.0", "pwm", pwm1_clk)
+ _REGISTER_CLOCK("mxc_pwm.1", "pwm", pwm2_clk)
_REGISTER_CLOCK("imx-i2c.0", NULL, i2c1_clk)
_REGISTER_CLOCK("imx-i2c.1", NULL, i2c2_clk)
_REGISTER_CLOCK("imx-i2c.2", NULL, hsi2c_clk)
@@ -1295,7 +1302,7 @@
_REGISTER_CLOCK("mxc-ehci.2", "usb_ahb", usb_ahb_clk)
_REGISTER_CLOCK("fsl-usb2-udc", "usb", usboh3_clk)
_REGISTER_CLOCK("fsl-usb2-udc", "usb_ahb", ahb_clk)
- _REGISTER_CLOCK("imx-keypad.0", NULL, kpp_clk)
+ _REGISTER_CLOCK("imx-keypad", NULL, kpp_clk)
_REGISTER_CLOCK("mxc_nand", NULL, nfc_clk)
_REGISTER_CLOCK("imx-ssi.0", NULL, ssi1_clk)
_REGISTER_CLOCK("imx-ssi.1", NULL, ssi2_clk)
@@ -1326,6 +1333,13 @@
_REGISTER_CLOCK(NULL, "gpt", gpt_clk)
_REGISTER_CLOCK("fec.0", NULL, fec_clk)
_REGISTER_CLOCK(NULL, "iim_clk", iim_clk)
+ _REGISTER_CLOCK("imx-i2c.0", NULL, i2c1_clk)
+ _REGISTER_CLOCK("imx-i2c.1", NULL, i2c2_clk)
+ _REGISTER_CLOCK("sdhci-esdhc-imx.0", NULL, esdhc1_clk)
+ _REGISTER_CLOCK("sdhci-esdhc-imx.1", NULL, esdhc2_clk)
+ _REGISTER_CLOCK("imx53-ecspi.0", NULL, ecspi1_clk)
+ _REGISTER_CLOCK("imx53-ecspi.1", NULL, ecspi2_clk)
+ _REGISTER_CLOCK("imx53-cspi.0", NULL, cspi_clk)
};
static void clk_tree_init(void)
@@ -1363,7 +1377,6 @@
clk_tree_init();
- clk_set_parent(&uart_root_clk, &pll3_sw_clk);
clk_enable(&cpu_clk);
clk_enable(&main_bus_clk);
@@ -1406,6 +1419,7 @@
clk_tree_init();
+ clk_set_parent(&uart_root_clk, &pll3_sw_clk);
clk_enable(&cpu_clk);
clk_enable(&main_bus_clk);
diff --git a/arch/arm/mach-mx5/devices-imx51.h b/arch/arm/mach-mx5/devices-imx51.h
index 6302e46..7fff485 100644
--- a/arch/arm/mach-mx5/devices-imx51.h
+++ b/arch/arm/mach-mx5/devices-imx51.h
@@ -47,3 +47,11 @@
extern const struct imx_imx2_wdt_data imx51_imx2_wdt_data[] __initconst;
#define imx51_add_imx2_wdt(id, pdata) \
imx_add_imx2_wdt(&imx51_imx2_wdt_data[id])
+
+extern const struct imx_mxc_pwm_data imx51_mxc_pwm_data[] __initconst;
+#define imx51_add_mxc_pwm(id) \
+ imx_add_mxc_pwm(&imx51_mxc_pwm_data[id])
+
+extern const struct imx_imx_keypad_data imx51_imx_keypad_data __initconst;
+#define imx51_add_imx_keypad(pdata) \
+ imx_add_imx_keypad(&imx51_imx_keypad_data, pdata)
diff --git a/arch/arm/mach-mx5/devices-imx53.h b/arch/arm/mach-mx5/devices-imx53.h
index 9d0ec25..8639735 100644
--- a/arch/arm/mach-mx5/devices-imx53.h
+++ b/arch/arm/mach-mx5/devices-imx53.h
@@ -8,6 +8,24 @@
#include <mach/mx53.h>
#include <mach/devices-common.h>
+extern const struct imx_fec_data imx53_fec_data __initconst;
+#define imx53_add_fec(pdata) \
+ imx_add_fec(&imx53_fec_data, pdata)
+
extern const struct imx_imx_uart_1irq_data imx53_imx_uart_data[] __initconst;
#define imx53_add_imx_uart(id, pdata) \
imx_add_imx_uart_1irq(&imx53_imx_uart_data[id], pdata)
+
+
+extern const struct imx_imx_i2c_data imx53_imx_i2c_data[] __initconst;
+#define imx53_add_imx_i2c(id, pdata) \
+ imx_add_imx_i2c(&imx53_imx_i2c_data[id], pdata)
+
+extern const struct imx_sdhci_esdhc_imx_data
+imx53_sdhci_esdhc_imx_data[] __initconst;
+#define imx53_add_sdhci_esdhc_imx(id, pdata) \
+ imx_add_sdhci_esdhc_imx(&imx53_sdhci_esdhc_imx_data[id], pdata)
+
+extern const struct imx_spi_imx_data imx53_ecspi_data[] __initconst;
+#define imx53_add_ecspi(id, pdata) \
+ imx_add_spi_imx(&imx53_ecspi_data[id], pdata)
diff --git a/arch/arm/mach-mx5/devices.c b/arch/arm/mach-mx5/devices.c
index 1bda5cb..153ada5 100644
--- a/arch/arm/mach-mx5/devices.c
+++ b/arch/arm/mach-mx5/devices.c
@@ -120,25 +120,6 @@
},
};
-static struct resource mxc_kpp_resources[] = {
- {
- .start = MX51_MXC_INT_KPP,
- .end = MX51_MXC_INT_KPP,
- .flags = IORESOURCE_IRQ,
- } , {
- .start = MX51_KPP_BASE_ADDR,
- .end = MX51_KPP_BASE_ADDR + 0x8 - 1,
- .flags = IORESOURCE_MEM,
- },
-};
-
-struct platform_device mxc_keypad_device = {
- .name = "imx-keypad",
- .id = 0,
- .num_resources = ARRAY_SIZE(mxc_kpp_resources),
- .resource = mxc_kpp_resources,
-};
-
static struct mxc_gpio_port mxc_gpio_ports[] = {
{
.chip.label = "gpio-0",
diff --git a/arch/arm/mach-mx5/devices.h b/arch/arm/mach-mx5/devices.h
index 16891aa..55a5129 100644
--- a/arch/arm/mach-mx5/devices.h
+++ b/arch/arm/mach-mx5/devices.h
@@ -3,4 +3,3 @@
extern struct platform_device mxc_usbh2_device;
extern struct platform_device mxc_usbdr_udc_device;
extern struct platform_device mxc_hsi2c_device;
-extern struct platform_device mxc_keypad_device;
diff --git a/arch/arm/mach-mx5/eukrea_mbimx51-baseboard.c b/arch/arm/mach-mx5/eukrea_mbimx51-baseboard.c
index c96d018..e83ffad 100644
--- a/arch/arm/mach-mx5/eukrea_mbimx51-baseboard.c
+++ b/arch/arm/mach-mx5/eukrea_mbimx51-baseboard.c
@@ -21,7 +21,6 @@
#include <linux/fsl_devices.h>
#include <linux/i2c/tsc2007.h>
#include <linux/leds.h>
-#include <linux/input/matrix_keypad.h>
#include <mach/common.h>
#include <mach/hardware.h>
@@ -157,7 +156,7 @@
KEY(3, 3, KEY_ENTER),
};
-static struct matrix_keymap_data mbimx51_map_data = {
+static const struct matrix_keymap_data mbimx51_map_data __initconst = {
.keymap = mbimx51_keymap,
.keymap_size = ARRAY_SIZE(mbimx51_keymap),
};
@@ -209,7 +208,7 @@
platform_add_devices(devices, ARRAY_SIZE(devices));
- mxc_register_device(&mxc_keypad_device, &mbimx51_map_data);
+ imx51_add_imx_keypad(&mbimx51_map_data);
gpio_request(MBIMX51_TSC2007_GPIO, "tsc2007_irq");
gpio_direction_input(MBIMX51_TSC2007_GPIO);
diff --git a/arch/arm/mach-mxs/Kconfig b/arch/arm/mach-mxs/Kconfig
index c4ac7b4..8bfc8df 100644
--- a/arch/arm/mach-mxs/Kconfig
+++ b/arch/arm/mach-mxs/Kconfig
@@ -15,7 +15,7 @@
config MACH_MX23EVK
bool "Support MX23EVK Platform"
select SOC_IMX23
- select MXS_HAVE_PLATFORM_DUART
+ select MXS_HAVE_AMBA_DUART
default y
help
Include support for MX23EVK platform. This includes specific
@@ -24,7 +24,7 @@
config MACH_MX28EVK
bool "Support MX28EVK Platform"
select SOC_IMX28
- select MXS_HAVE_PLATFORM_DUART
+ select MXS_HAVE_AMBA_DUART
select MXS_HAVE_PLATFORM_FEC
default y
help
diff --git a/arch/arm/mach-mxs/clock-mx23.c b/arch/arm/mach-mxs/clock-mx23.c
index 8f5a19a..b1a362e 100644
--- a/arch/arm/mach-mxs/clock-mx23.c
+++ b/arch/arm/mach-mxs/clock-mx23.c
@@ -21,6 +21,7 @@
#include <linux/clk.h>
#include <linux/io.h>
#include <linux/jiffies.h>
+#include <linux/clkdev.h>
#include <asm/clkdev.h>
#include <asm/div64.h>
@@ -437,10 +438,12 @@
},
static struct clk_lookup lookups[] = {
- _REGISTER_CLOCK("mxs-duart.0", NULL, uart_clk)
+ /* for amba bus driver */
+ _REGISTER_CLOCK("duart", "apb_pclk", xbus_clk)
+ /* for amba-pl011 driver */
+ _REGISTER_CLOCK("duart", NULL, uart_clk)
_REGISTER_CLOCK("rtc", NULL, rtc_clk)
_REGISTER_CLOCK(NULL, "hclk", hbus_clk)
- _REGISTER_CLOCK(NULL, "xclk", xbus_clk)
_REGISTER_CLOCK(NULL, "usb", usb_clk)
_REGISTER_CLOCK(NULL, "audio", audio_clk)
_REGISTER_CLOCK(NULL, "pwm", pwm_clk)
@@ -518,6 +521,12 @@
{
clk_misc_init();
+ clk_enable(&cpu_clk);
+ clk_enable(&hbus_clk);
+ clk_enable(&xbus_clk);
+ clk_enable(&emi_clk);
+ clk_enable(&uart_clk);
+
clkdev_add_table(lookups, ARRAY_SIZE(lookups));
mxs_timer_init(&clk32k_clk, MX23_INT_TIMER0);
diff --git a/arch/arm/mach-mxs/clock-mx28.c b/arch/arm/mach-mxs/clock-mx28.c
index 74e2103..56312c0 100644
--- a/arch/arm/mach-mxs/clock-mx28.c
+++ b/arch/arm/mach-mxs/clock-mx28.c
@@ -21,6 +21,7 @@
#include <linux/clk.h>
#include <linux/io.h>
#include <linux/jiffies.h>
+#include <linux/clkdev.h>
#include <asm/clkdev.h>
#include <asm/div64.h>
@@ -602,7 +603,12 @@
},
static struct clk_lookup lookups[] = {
- _REGISTER_CLOCK("mxs-duart.0", NULL, uart_clk)
+ /* for amba bus driver */
+ _REGISTER_CLOCK("duart", "apb_pclk", xbus_clk)
+ /* for amba-pl011 driver */
+ _REGISTER_CLOCK("duart", NULL, uart_clk)
+ _REGISTER_CLOCK("imx28-fec.0", NULL, fec_clk)
+ _REGISTER_CLOCK("imx28-fec.1", NULL, fec_clk)
_REGISTER_CLOCK("fec.0", NULL, fec_clk)
_REGISTER_CLOCK("rtc", NULL, rtc_clk)
_REGISTER_CLOCK("pll2", NULL, pll2_clk)
@@ -726,6 +732,12 @@
{
clk_misc_init();
+ clk_enable(&cpu_clk);
+ clk_enable(&hbus_clk);
+ clk_enable(&xbus_clk);
+ clk_enable(&emi_clk);
+ clk_enable(&uart_clk);
+
clkdev_add_table(lookups, ARRAY_SIZE(lookups));
mxs_timer_init(&clk32k_clk, MX28_INT_TIMER0);
diff --git a/arch/arm/mach-mxs/devices-mx23.h b/arch/arm/mach-mxs/devices-mx23.h
index d0f49fc..1256788 100644
--- a/arch/arm/mach-mxs/devices-mx23.h
+++ b/arch/arm/mach-mxs/devices-mx23.h
@@ -11,6 +11,6 @@
#include <mach/mx23.h>
#include <mach/devices-common.h>
-extern const struct mxs_duart_data mx23_duart_data __initconst;
+extern const struct amba_device mx23_duart_device __initconst;
#define mx23_add_duart() \
- mxs_add_duart(&mx23_duart_data)
+ mxs_add_duart(&mx23_duart_device)
diff --git a/arch/arm/mach-mxs/devices-mx28.h b/arch/arm/mach-mxs/devices-mx28.h
index 00b736c..33773a6 100644
--- a/arch/arm/mach-mxs/devices-mx28.h
+++ b/arch/arm/mach-mxs/devices-mx28.h
@@ -11,9 +11,9 @@
#include <mach/mx28.h>
#include <mach/devices-common.h>
-extern const struct mxs_duart_data mx28_duart_data __initconst;
+extern const struct amba_device mx28_duart_device __initconst;
#define mx28_add_duart() \
- mxs_add_duart(&mx28_duart_data)
+ mxs_add_duart(&mx28_duart_device)
extern const struct mxs_fec_data mx28_fec_data[] __initconst;
#define mx28_add_fec(id, pdata) \
diff --git a/arch/arm/mach-mxs/devices.c b/arch/arm/mach-mxs/devices.c
index 6b60f02..c20d547 100644
--- a/arch/arm/mach-mxs/devices.c
+++ b/arch/arm/mach-mxs/devices.c
@@ -19,9 +19,8 @@
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/init.h>
-#include <linux/err.h>
#include <linux/platform_device.h>
-#include <mach/common.h>
+#include <linux/amba/bus.h>
struct platform_device *__init mxs_add_platform_device_dmamask(
const char *name, int id,
@@ -73,3 +72,17 @@
return pdev;
}
+
+int __init mxs_add_amba_device(const struct amba_device *dev)
+{
+ struct amba_device *adev = kmalloc(sizeof(*adev), GFP_KERNEL);
+
+ if (!adev) {
+ pr_err("%s: failed to allocate memory", __func__);
+ return -ENOMEM;
+ }
+
+ *adev = *dev;
+
+ return amba_device_register(adev, &iomem_resource);
+}
diff --git a/arch/arm/mach-mxs/devices/Kconfig b/arch/arm/mach-mxs/devices/Kconfig
index a35a2dc..cf7dc1a 100644
--- a/arch/arm/mach-mxs/devices/Kconfig
+++ b/arch/arm/mach-mxs/devices/Kconfig
@@ -1,5 +1,6 @@
-config MXS_HAVE_PLATFORM_DUART
+config MXS_HAVE_AMBA_DUART
bool
+ select ARM_AMBA
config MXS_HAVE_PLATFORM_FEC
bool
diff --git a/arch/arm/mach-mxs/devices/Makefile b/arch/arm/mach-mxs/devices/Makefile
index 4b5266a..d0a09f6 100644
--- a/arch/arm/mach-mxs/devices/Makefile
+++ b/arch/arm/mach-mxs/devices/Makefile
@@ -1,2 +1,2 @@
-obj-$(CONFIG_MXS_HAVE_PLATFORM_DUART) += platform-duart.o
+obj-$(CONFIG_MXS_HAVE_AMBA_DUART) += amba-duart.o
obj-$(CONFIG_MXS_HAVE_PLATFORM_FEC) += platform-fec.o
diff --git a/arch/arm/mach-mxs/devices/amba-duart.c b/arch/arm/mach-mxs/devices/amba-duart.c
new file mode 100644
index 0000000..a559db0
--- /dev/null
+++ b/arch/arm/mach-mxs/devices/amba-duart.c
@@ -0,0 +1,40 @@
+/*
+ * Copyright (C) 2009-2010 Pengutronix
+ * Uwe Kleine-Koenig <u.kleine-koenig@pengutronix.de>
+ *
+ * Copyright 2010 Freescale Semiconductor, Inc. All Rights Reserved.
+ *
+ * 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 <asm/irq.h>
+#include <mach/mx23.h>
+#include <mach/mx28.h>
+#include <mach/devices-common.h>
+
+#define MXS_AMBA_DUART_DEVICE(name, soc) \
+const struct amba_device name##_device __initconst = { \
+ .dev = { \
+ .init_name = "duart", \
+ }, \
+ .res = { \
+ .start = soc ## _DUART_BASE_ADDR, \
+ .end = (soc ## _DUART_BASE_ADDR) + SZ_8K - 1, \
+ .flags = IORESOURCE_MEM, \
+ }, \
+ .irq = {soc ## _INT_DUART, NO_IRQ}, \
+}
+
+#ifdef CONFIG_SOC_IMX23
+MXS_AMBA_DUART_DEVICE(mx23_duart, MX23);
+#endif
+
+#ifdef CONFIG_SOC_IMX28
+MXS_AMBA_DUART_DEVICE(mx28_duart, MX28);
+#endif
+
+int __init mxs_add_duart(const struct amba_device *dev)
+{
+ return mxs_add_amba_device(dev);
+}
diff --git a/arch/arm/mach-mxs/devices/platform-duart.c b/arch/arm/mach-mxs/devices/platform-duart.c
deleted file mode 100644
index 2fe0df5..0000000
--- a/arch/arm/mach-mxs/devices/platform-duart.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * Copyright (C) 2009-2010 Pengutronix
- * Uwe Kleine-Koenig <u.kleine-koenig@pengutronix.de>
- *
- * Copyright 2010 Freescale Semiconductor, Inc. All Rights Reserved.
- *
- * 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 <mach/mx23.h>
-#include <mach/mx28.h>
-#include <mach/devices-common.h>
-
-#define mxs_duart_data_entry(soc) \
- { \
- .iobase = soc ## _DUART_BASE_ADDR, \
- .irq = soc ## _INT_DUART, \
- }
-
-#ifdef CONFIG_SOC_IMX23
-const struct mxs_duart_data mx23_duart_data __initconst =
- mxs_duart_data_entry(MX23);
-#endif
-
-#ifdef CONFIG_SOC_IMX28
-const struct mxs_duart_data mx28_duart_data __initconst =
- mxs_duart_data_entry(MX28);
-#endif
-
-struct platform_device *__init mxs_add_duart(
- const struct mxs_duart_data *data)
-{
- struct resource res[] = {
- {
- .start = data->iobase,
- .end = data->iobase + SZ_8K - 1,
- .flags = IORESOURCE_MEM,
- }, {
- .start = data->irq,
- .end = data->irq,
- .flags = IORESOURCE_IRQ,
- },
- };
-
- return mxs_add_platform_device("mxs-duart", 0, res, ARRAY_SIZE(res),
- NULL, 0);
-}
diff --git a/arch/arm/mach-mxs/devices/platform-fec.c b/arch/arm/mach-mxs/devices/platform-fec.c
index c08168c..c42dff7 100644
--- a/arch/arm/mach-mxs/devices/platform-fec.c
+++ b/arch/arm/mach-mxs/devices/platform-fec.c
@@ -45,6 +45,6 @@
},
};
- return mxs_add_platform_device("fec", data->id,
+ return mxs_add_platform_device("imx28-fec", data->id,
res, ARRAY_SIZE(res), pdata, sizeof(*pdata));
}
diff --git a/arch/arm/mach-mxs/include/mach/devices-common.h b/arch/arm/mach-mxs/include/mach/devices-common.h
index 3da48d4..6c3d1a1 100644
--- a/arch/arm/mach-mxs/include/mach/devices-common.h
+++ b/arch/arm/mach-mxs/include/mach/devices-common.h
@@ -9,6 +9,7 @@
#include <linux/kernel.h>
#include <linux/platform_device.h>
#include <linux/init.h>
+#include <linux/amba/bus.h>
struct platform_device *mxs_add_platform_device_dmamask(
const char *name, int id,
@@ -24,14 +25,10 @@
name, id, res, num_resources, data, size_data, 0);
}
+int __init mxs_add_amba_device(const struct amba_device *dev);
+
/* duart */
-struct mxs_duart_data {
- resource_size_t iobase;
- resource_size_t iosize;
- resource_size_t irq;
-};
-struct platform_device *__init mxs_add_duart(
- const struct mxs_duart_data *data);
+int __init mxs_add_duart(const struct amba_device *dev);
/* fec */
#include <linux/fec.h>
diff --git a/arch/arm/mach-mxs/mach-mx28evk.c b/arch/arm/mach-mxs/mach-mx28evk.c
index d162e95..8e2c597 100644
--- a/arch/arm/mach-mxs/mach-mx28evk.c
+++ b/arch/arm/mach-mxs/mach-mx28evk.c
@@ -57,6 +57,19 @@
(MXS_PAD_8MA | MXS_PAD_3V3 | MXS_PAD_PULLUP),
MX28_PAD_ENET_CLK__CLKCTRL_ENET |
(MXS_PAD_8MA | MXS_PAD_3V3 | MXS_PAD_PULLUP),
+ /* fec1 */
+ MX28_PAD_ENET0_CRS__ENET1_RX_EN |
+ (MXS_PAD_8MA | MXS_PAD_3V3 | MXS_PAD_PULLUP),
+ MX28_PAD_ENET0_RXD2__ENET1_RXD0 |
+ (MXS_PAD_8MA | MXS_PAD_3V3 | MXS_PAD_PULLUP),
+ MX28_PAD_ENET0_RXD3__ENET1_RXD1 |
+ (MXS_PAD_8MA | MXS_PAD_3V3 | MXS_PAD_PULLUP),
+ MX28_PAD_ENET0_COL__ENET1_TX_EN |
+ (MXS_PAD_8MA | MXS_PAD_3V3 | MXS_PAD_PULLUP),
+ MX28_PAD_ENET0_TXD2__ENET1_TXD0 |
+ (MXS_PAD_8MA | MXS_PAD_3V3 | MXS_PAD_PULLUP),
+ MX28_PAD_ENET0_TXD3__ENET1_TXD1 |
+ (MXS_PAD_8MA | MXS_PAD_3V3 | MXS_PAD_PULLUP),
/* phy power line */
MX28_PAD_SSP1_DATA3__GPIO_2_15 |
(MXS_PAD_4MA | MXS_PAD_3V3 | MXS_PAD_NOPULL),
@@ -106,8 +119,14 @@
gpio_set_value(MX28EVK_FEC_PHY_RESET, 1);
}
-static const struct fec_platform_data mx28_fec_pdata __initconst = {
- .phy = PHY_INTERFACE_MODE_RMII,
+static struct fec_platform_data mx28_fec_pdata[] = {
+ {
+ /* fec0 */
+ .phy = PHY_INTERFACE_MODE_RMII,
+ }, {
+ /* fec1 */
+ .phy = PHY_INTERFACE_MODE_RMII,
+ },
};
static void __init mx28evk_init(void)
@@ -117,7 +136,8 @@
mx28_add_duart();
mx28evk_fec_reset();
- mx28_add_fec(0, &mx28_fec_pdata);
+ mx28_add_fec(0, &mx28_fec_pdata[0]);
+ mx28_add_fec(1, &mx28_fec_pdata[1]);
}
static void __init mx28evk_timer_init(void)
diff --git a/arch/arm/mach-netx/generic.c b/arch/arm/mach-netx/generic.c
index 43da8bb..29ffa75 100644
--- a/arch/arm/mach-netx/generic.c
+++ b/arch/arm/mach-netx/generic.c
@@ -88,13 +88,13 @@
}
static int
-netx_hif_irq_type(unsigned int _irq, unsigned int type)
+netx_hif_irq_type(struct irq_data *d, unsigned int type)
{
unsigned int val, irq;
val = readl(NETX_DPMAS_IF_CONF1);
- irq = _irq - NETX_IRQ_HIF_CHAINED(0);
+ irq = d->irq - NETX_IRQ_HIF_CHAINED(0);
if (type & IRQ_TYPE_EDGE_RISING) {
DEBUG_IRQ("rising edges\n");
@@ -119,49 +119,49 @@
}
static void
-netx_hif_ack_irq(unsigned int _irq)
+netx_hif_ack_irq(struct irq_data *d)
{
unsigned int val, irq;
- irq = _irq - NETX_IRQ_HIF_CHAINED(0);
+ irq = d->irq - NETX_IRQ_HIF_CHAINED(0);
writel((1 << 24) << irq, NETX_DPMAS_INT_STAT);
val = readl(NETX_DPMAS_INT_EN);
val &= ~((1 << 24) << irq);
writel(val, NETX_DPMAS_INT_EN);
- DEBUG_IRQ("%s: irq %d\n", __func__, _irq);
+ DEBUG_IRQ("%s: irq %d\n", __func__, d->irq);
}
static void
-netx_hif_mask_irq(unsigned int _irq)
+netx_hif_mask_irq(struct irq_data *d)
{
unsigned int val, irq;
- irq = _irq - NETX_IRQ_HIF_CHAINED(0);
+ irq = d->irq - NETX_IRQ_HIF_CHAINED(0);
val = readl(NETX_DPMAS_INT_EN);
val &= ~((1 << 24) << irq);
writel(val, NETX_DPMAS_INT_EN);
- DEBUG_IRQ("%s: irq %d\n", __func__, _irq);
+ DEBUG_IRQ("%s: irq %d\n", __func__, d->irq);
}
static void
-netx_hif_unmask_irq(unsigned int _irq)
+netx_hif_unmask_irq(struct irq_data *d)
{
unsigned int val, irq;
- irq = _irq - NETX_IRQ_HIF_CHAINED(0);
+ irq = d->irq - NETX_IRQ_HIF_CHAINED(0);
val = readl(NETX_DPMAS_INT_EN);
val |= (1 << 24) << irq;
writel(val, NETX_DPMAS_INT_EN);
- DEBUG_IRQ("%s: irq %d\n", __func__, _irq);
+ DEBUG_IRQ("%s: irq %d\n", __func__, d->irq);
}
static struct irq_chip netx_hif_chip = {
- .ack = netx_hif_ack_irq,
- .mask = netx_hif_mask_irq,
- .unmask = netx_hif_unmask_irq,
- .set_type = netx_hif_irq_type,
+ .irq_ack = netx_hif_ack_irq,
+ .irq_mask = netx_hif_mask_irq,
+ .irq_unmask = netx_hif_unmask_irq,
+ .irq_set_type = netx_hif_irq_type,
};
void __init netx_init_irq(void)
diff --git a/arch/arm/mach-ns9xxx/board-a9m9750dev.c b/arch/arm/mach-ns9xxx/board-a9m9750dev.c
index b45bb3b..0c0d524 100644
--- a/arch/arm/mach-ns9xxx/board-a9m9750dev.c
+++ b/arch/arm/mach-ns9xxx/board-a9m9750dev.c
@@ -37,44 +37,44 @@
ARRAY_SIZE(board_a9m9750dev_io_desc));
}
-static void a9m9750dev_fpga_ack_irq(unsigned int irq)
+static void a9m9750dev_fpga_ack_irq(struct irq_data *d)
{
/* nothing */
}
-static void a9m9750dev_fpga_mask_irq(unsigned int irq)
+static void a9m9750dev_fpga_mask_irq(struct irq_data *d)
{
u8 ier;
ier = __raw_readb(FPGA_IER);
- ier &= ~(1 << (irq - FPGA_IRQ(0)));
+ ier &= ~(1 << (d->irq - FPGA_IRQ(0)));
__raw_writeb(ier, FPGA_IER);
}
-static void a9m9750dev_fpga_maskack_irq(unsigned int irq)
+static void a9m9750dev_fpga_maskack_irq(struct irq_data *d)
{
- a9m9750dev_fpga_mask_irq(irq);
- a9m9750dev_fpga_ack_irq(irq);
+ a9m9750dev_fpga_mask_irq(d);
+ a9m9750dev_fpga_ack_irq(d);
}
-static void a9m9750dev_fpga_unmask_irq(unsigned int irq)
+static void a9m9750dev_fpga_unmask_irq(struct irq_data *d)
{
u8 ier;
ier = __raw_readb(FPGA_IER);
- ier |= 1 << (irq - FPGA_IRQ(0));
+ ier |= 1 << (d->irq - FPGA_IRQ(0));
__raw_writeb(ier, FPGA_IER);
}
static struct irq_chip a9m9750dev_fpga_chip = {
- .ack = a9m9750dev_fpga_ack_irq,
- .mask = a9m9750dev_fpga_mask_irq,
- .mask_ack = a9m9750dev_fpga_maskack_irq,
- .unmask = a9m9750dev_fpga_unmask_irq,
+ .irq_ack = a9m9750dev_fpga_ack_irq,
+ .irq_mask = a9m9750dev_fpga_mask_irq,
+ .irq_mask_ack = a9m9750dev_fpga_maskack_irq,
+ .irq_unmask = a9m9750dev_fpga_unmask_irq,
};
static void a9m9750dev_fpga_demux_handler(unsigned int irq,
@@ -82,7 +82,7 @@
{
u8 stat = __raw_readb(FPGA_ISR);
- desc->chip->mask_ack(irq);
+ desc->irq_data.chip->irq_mask_ack(&desc->irq_data);
while (stat != 0) {
int irqno = fls(stat) - 1;
@@ -92,7 +92,7 @@
generic_handle_irq(FPGA_IRQ(irqno));
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
void __init board_a9m9750dev_init_irq(void)
diff --git a/arch/arm/mach-ns9xxx/irq.c b/arch/arm/mach-ns9xxx/irq.c
index 038f24d..389fa5c 100644
--- a/arch/arm/mach-ns9xxx/irq.c
+++ b/arch/arm/mach-ns9xxx/irq.c
@@ -22,40 +22,40 @@
#define irq2prio(i) (i)
#define prio2irq(p) (p)
-static void ns9xxx_mask_irq(unsigned int irq)
+static void ns9xxx_mask_irq(struct irq_data *d)
{
/* XXX: better use cpp symbols */
- int prio = irq2prio(irq);
+ int prio = irq2prio(d->irq);
u32 ic = __raw_readl(SYS_IC(prio / 4));
ic &= ~(1 << (7 + 8 * (3 - (prio & 3))));
__raw_writel(ic, SYS_IC(prio / 4));
}
-static void ns9xxx_ack_irq(unsigned int irq)
+static void ns9xxx_ack_irq(struct irq_data *d)
{
__raw_writel(0, SYS_ISRADDR);
}
-static void ns9xxx_maskack_irq(unsigned int irq)
+static void ns9xxx_maskack_irq(struct irq_data *d)
{
- ns9xxx_mask_irq(irq);
- ns9xxx_ack_irq(irq);
+ ns9xxx_mask_irq(d);
+ ns9xxx_ack_irq(d);
}
-static void ns9xxx_unmask_irq(unsigned int irq)
+static void ns9xxx_unmask_irq(struct irq_data *d)
{
/* XXX: better use cpp symbols */
- int prio = irq2prio(irq);
+ int prio = irq2prio(d->irq);
u32 ic = __raw_readl(SYS_IC(prio / 4));
ic |= 1 << (7 + 8 * (3 - (prio & 3)));
__raw_writel(ic, SYS_IC(prio / 4));
}
static struct irq_chip ns9xxx_chip = {
- .ack = ns9xxx_ack_irq,
- .mask = ns9xxx_mask_irq,
- .mask_ack = ns9xxx_maskack_irq,
- .unmask = ns9xxx_unmask_irq,
+ .irq_ack = ns9xxx_ack_irq,
+ .irq_mask = ns9xxx_mask_irq,
+ .irq_mask_ack = ns9xxx_maskack_irq,
+ .irq_unmask = ns9xxx_unmask_irq,
};
#if 0
@@ -92,10 +92,10 @@
if (desc->status & IRQ_DISABLED)
out_mask:
- desc->chip->mask(irq);
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
/* ack unconditionally to unmask lower prio irqs */
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
raw_spin_unlock(&desc->lock);
}
diff --git a/arch/arm/mach-nuc93x/irq.c b/arch/arm/mach-nuc93x/irq.c
index a7a88ea..1f8a05a 100644
--- a/arch/arm/mach-nuc93x/irq.c
+++ b/arch/arm/mach-nuc93x/irq.c
@@ -25,9 +25,9 @@
#include <mach/hardware.h>
#include <mach/regs-irq.h>
-static void nuc93x_irq_mask(unsigned int irq)
+static void nuc93x_irq_mask(struct irq_data *d)
{
- __raw_writel(1 << irq, REG_AIC_MDCR);
+ __raw_writel(1 << d->irq, REG_AIC_MDCR);
}
/*
@@ -35,21 +35,21 @@
* to REG_AIC_EOSCR for ACK
*/
-static void nuc93x_irq_ack(unsigned int irq)
+static void nuc93x_irq_ack(struct irq_data *d)
{
__raw_writel(0x01, REG_AIC_EOSCR);
}
-static void nuc93x_irq_unmask(unsigned int irq)
+static void nuc93x_irq_unmask(struct irq_data *d)
{
- __raw_writel(1 << irq, REG_AIC_MECR);
+ __raw_writel(1 << d->irq, REG_AIC_MECR);
}
static struct irq_chip nuc93x_irq_chip = {
- .ack = nuc93x_irq_ack,
- .mask = nuc93x_irq_mask,
- .unmask = nuc93x_irq_unmask,
+ .irq_ack = nuc93x_irq_ack,
+ .irq_mask = nuc93x_irq_mask,
+ .irq_unmask = nuc93x_irq_unmask,
};
void __init nuc93x_init_irq(void)
diff --git a/arch/arm/mach-omap1/ams-delta-fiq.c b/arch/arm/mach-omap1/ams-delta-fiq.c
index 6c994e2..152b32c 100644
--- a/arch/arm/mach-omap1/ams-delta-fiq.c
+++ b/arch/arm/mach-omap1/ams-delta-fiq.c
@@ -49,7 +49,7 @@
irq_desc = irq_to_desc(IH_GPIO_BASE);
if (irq_desc)
- irq_chip = irq_desc->chip;
+ irq_chip = irq_desc->irq_data.chip;
/*
* For each handled GPIO interrupt, keep calling its interrupt handler
@@ -62,13 +62,15 @@
while (irq_counter[gpio] < fiq_count) {
if (gpio != AMS_DELTA_GPIO_PIN_KEYBRD_CLK) {
+ struct irq_data *d = irq_get_irq_data(irq_num);
+
/*
* It looks like handle_edge_irq() that
* OMAP GPIO edge interrupts default to,
* expects interrupt already unmasked.
*/
- if (irq_chip && irq_chip->unmask)
- irq_chip->unmask(irq_num);
+ if (irq_chip && irq_chip->irq_unmask)
+ irq_chip->irq_unmask(d);
}
generic_handle_irq(irq_num);
diff --git a/arch/arm/mach-omap1/fpga.c b/arch/arm/mach-omap1/fpga.c
index 8780e75..0ace799 100644
--- a/arch/arm/mach-omap1/fpga.c
+++ b/arch/arm/mach-omap1/fpga.c
@@ -30,9 +30,9 @@
#include <plat/fpga.h>
#include <mach/gpio.h>
-static void fpga_mask_irq(unsigned int irq)
+static void fpga_mask_irq(struct irq_data *d)
{
- irq -= OMAP_FPGA_IRQ_BASE;
+ unsigned int irq = d->irq - OMAP_FPGA_IRQ_BASE;
if (irq < 8)
__raw_writeb((__raw_readb(OMAP1510_FPGA_IMR_LO)
@@ -58,14 +58,14 @@
}
-static void fpga_ack_irq(unsigned int irq)
+static void fpga_ack_irq(struct irq_data *d)
{
/* Don't need to explicitly ACK FPGA interrupts */
}
-static void fpga_unmask_irq(unsigned int irq)
+static void fpga_unmask_irq(struct irq_data *d)
{
- irq -= OMAP_FPGA_IRQ_BASE;
+ unsigned int irq = d->irq - OMAP_FPGA_IRQ_BASE;
if (irq < 8)
__raw_writeb((__raw_readb(OMAP1510_FPGA_IMR_LO) | (1 << irq)),
@@ -78,10 +78,10 @@
| (1 << (irq - 16))), INNOVATOR_FPGA_IMR2);
}
-static void fpga_mask_ack_irq(unsigned int irq)
+static void fpga_mask_ack_irq(struct irq_data *d)
{
- fpga_mask_irq(irq);
- fpga_ack_irq(irq);
+ fpga_mask_irq(d);
+ fpga_ack_irq(d);
}
void innovator_fpga_IRQ_demux(unsigned int irq, struct irq_desc *desc)
@@ -105,17 +105,17 @@
static struct irq_chip omap_fpga_irq_ack = {
.name = "FPGA-ack",
- .ack = fpga_mask_ack_irq,
- .mask = fpga_mask_irq,
- .unmask = fpga_unmask_irq,
+ .irq_ack = fpga_mask_ack_irq,
+ .irq_mask = fpga_mask_irq,
+ .irq_unmask = fpga_unmask_irq,
};
static struct irq_chip omap_fpga_irq = {
.name = "FPGA",
- .ack = fpga_ack_irq,
- .mask = fpga_mask_irq,
- .unmask = fpga_unmask_irq,
+ .irq_ack = fpga_ack_irq,
+ .irq_mask = fpga_mask_irq,
+ .irq_unmask = fpga_unmask_irq,
};
/*
diff --git a/arch/arm/mach-omap1/irq.c b/arch/arm/mach-omap1/irq.c
index 6bddbc8..4770158 100644
--- a/arch/arm/mach-omap1/irq.c
+++ b/arch/arm/mach-omap1/irq.c
@@ -70,48 +70,48 @@
omap_writel(value, irq_banks[bank].base_reg + offset);
}
-static void omap_ack_irq(unsigned int irq)
+static void omap_ack_irq(struct irq_data *d)
{
- if (irq > 31)
+ if (d->irq > 31)
omap_writel(0x1, OMAP_IH2_BASE + IRQ_CONTROL_REG_OFFSET);
omap_writel(0x1, OMAP_IH1_BASE + IRQ_CONTROL_REG_OFFSET);
}
-static void omap_mask_irq(unsigned int irq)
+static void omap_mask_irq(struct irq_data *d)
{
- int bank = IRQ_BANK(irq);
+ int bank = IRQ_BANK(d->irq);
u32 l;
l = omap_readl(irq_banks[bank].base_reg + IRQ_MIR_REG_OFFSET);
- l |= 1 << IRQ_BIT(irq);
+ l |= 1 << IRQ_BIT(d->irq);
omap_writel(l, irq_banks[bank].base_reg + IRQ_MIR_REG_OFFSET);
}
-static void omap_unmask_irq(unsigned int irq)
+static void omap_unmask_irq(struct irq_data *d)
{
- int bank = IRQ_BANK(irq);
+ int bank = IRQ_BANK(d->irq);
u32 l;
l = omap_readl(irq_banks[bank].base_reg + IRQ_MIR_REG_OFFSET);
- l &= ~(1 << IRQ_BIT(irq));
+ l &= ~(1 << IRQ_BIT(d->irq));
omap_writel(l, irq_banks[bank].base_reg + IRQ_MIR_REG_OFFSET);
}
-static void omap_mask_ack_irq(unsigned int irq)
+static void omap_mask_ack_irq(struct irq_data *d)
{
- omap_mask_irq(irq);
- omap_ack_irq(irq);
+ omap_mask_irq(d);
+ omap_ack_irq(d);
}
-static int omap_wake_irq(unsigned int irq, unsigned int enable)
+static int omap_wake_irq(struct irq_data *d, unsigned int enable)
{
- int bank = IRQ_BANK(irq);
+ int bank = IRQ_BANK(d->irq);
if (enable)
- irq_banks[bank].wake_enable |= IRQ_BIT(irq);
+ irq_banks[bank].wake_enable |= IRQ_BIT(d->irq);
else
- irq_banks[bank].wake_enable &= ~IRQ_BIT(irq);
+ irq_banks[bank].wake_enable &= ~IRQ_BIT(d->irq);
return 0;
}
@@ -168,10 +168,10 @@
static struct irq_chip omap_irq_chip = {
.name = "MPU",
- .ack = omap_mask_ack_irq,
- .mask = omap_mask_irq,
- .unmask = omap_unmask_irq,
- .set_wake = omap_wake_irq,
+ .irq_ack = omap_mask_ack_irq,
+ .irq_mask = omap_mask_irq,
+ .irq_unmask = omap_unmask_irq,
+ .irq_set_wake = omap_wake_irq,
};
void __init omap_init_irq(void)
@@ -239,9 +239,9 @@
/* Unmask level 2 handler */
if (cpu_is_omap7xx())
- omap_unmask_irq(INT_7XX_IH2_IRQ);
+ omap_unmask_irq(irq_get_irq_data(INT_7XX_IH2_IRQ));
else if (cpu_is_omap15xx())
- omap_unmask_irq(INT_1510_IH2_IRQ);
+ omap_unmask_irq(irq_get_irq_data(INT_1510_IH2_IRQ));
else if (cpu_is_omap16xx())
- omap_unmask_irq(INT_1610_IH2_IRQ);
+ omap_unmask_irq(irq_get_irq_data(INT_1610_IH2_IRQ));
}
diff --git a/arch/arm/mach-omap2/board-4430sdp.c b/arch/arm/mach-omap2/board-4430sdp.c
index a70bdf2..07d1b20 100644
--- a/arch/arm/mach-omap2/board-4430sdp.c
+++ b/arch/arm/mach-omap2/board-4430sdp.c
@@ -554,6 +554,7 @@
#ifdef CONFIG_OMAP_MUX
static struct omap_board_mux board_mux[] __initdata = {
+ OMAP4_MUX(USBB2_ULPITLL_CLK, OMAP_MUX_MODE4 | OMAP_PIN_OUTPUT),
{ .reg_offset = OMAP_MUX_TERMINATOR },
};
#else
@@ -576,11 +577,12 @@
omap4_twl6030_hsmmc_init(mmc);
/* Power on the ULPI PHY */
- if (gpio_is_valid(OMAP4SDP_MDM_PWR_EN_GPIO)) {
- /* FIXME: Assumes pad is already muxed for GPIO mode */
- gpio_request(OMAP4SDP_MDM_PWR_EN_GPIO, "USBB1 PHY VMDM_3V3");
+ status = gpio_request(OMAP4SDP_MDM_PWR_EN_GPIO, "USBB1 PHY VMDM_3V3");
+ if (status)
+ pr_err("%s: Could not get USBB1 PHY GPIO\n", __func__);
+ else
gpio_direction_output(OMAP4SDP_MDM_PWR_EN_GPIO, 1);
- }
+
usb_ehci_init(&ehci_pdata);
usb_musb_init(&musb_board_data);
diff --git a/arch/arm/mach-omap2/board-igep0020.c b/arch/arm/mach-omap2/board-igep0020.c
index ebaa230..3be85a1 100644
--- a/arch/arm/mach-omap2/board-igep0020.c
+++ b/arch/arm/mach-omap2/board-igep0020.c
@@ -17,6 +17,7 @@
#include <linux/io.h>
#include <linux/gpio.h>
#include <linux/interrupt.h>
+#include <linux/input.h>
#include <linux/regulator/machine.h>
#include <linux/regulator/fixed.h>
@@ -541,6 +542,37 @@
.audio = &igep2_audio_data,
};
+static int igep2_keymap[] = {
+ KEY(0, 0, KEY_LEFT),
+ KEY(0, 1, KEY_RIGHT),
+ KEY(0, 2, KEY_A),
+ KEY(0, 3, KEY_B),
+ KEY(1, 0, KEY_DOWN),
+ KEY(1, 1, KEY_UP),
+ KEY(1, 2, KEY_E),
+ KEY(1, 3, KEY_F),
+ KEY(2, 0, KEY_ENTER),
+ KEY(2, 1, KEY_I),
+ KEY(2, 2, KEY_J),
+ KEY(2, 3, KEY_K),
+ KEY(3, 0, KEY_M),
+ KEY(3, 1, KEY_N),
+ KEY(3, 2, KEY_O),
+ KEY(3, 3, KEY_P)
+};
+
+static struct matrix_keymap_data igep2_keymap_data = {
+ .keymap = igep2_keymap,
+ .keymap_size = ARRAY_SIZE(igep2_keymap),
+};
+
+static struct twl4030_keypad_data igep2_keypad_pdata = {
+ .keymap_data = &igep2_keymap_data,
+ .rows = 4,
+ .cols = 4,
+ .rep = 1,
+};
+
static struct twl4030_platform_data igep2_twldata = {
.irq_base = TWL4030_IRQ_BASE,
.irq_end = TWL4030_IRQ_END,
@@ -549,6 +581,7 @@
.usb = &igep2_usb_data,
.codec = &igep2_codec_data,
.gpio = &igep2_twl4030_gpio_pdata,
+ .keypad = &igep2_keypad_pdata,
.vmmc1 = &igep2_vmmc1,
.vpll2 = &igep2_vpll2,
.vio = &igep2_vio,
diff --git a/arch/arm/mach-omap2/board-igep0030.c b/arch/arm/mach-omap2/board-igep0030.c
index bcccd68..4dc62a9 100644
--- a/arch/arm/mach-omap2/board-igep0030.c
+++ b/arch/arm/mach-omap2/board-igep0030.c
@@ -19,6 +19,7 @@
#include <linux/interrupt.h>
#include <linux/regulator/machine.h>
+#include <linux/regulator/fixed.h>
#include <linux/i2c/twl.h>
#include <linux/mmc/host.h>
@@ -43,7 +44,7 @@
#define IGEP3_GPIO_WIFI_NRESET 139
#define IGEP3_GPIO_BT_NRESET 137
-#define IGEP3_GPIO_USBH_NRESET 115
+#define IGEP3_GPIO_USBH_NRESET 183
#if defined(CONFIG_MTD_ONENAND_OMAP2) || \
@@ -103,7 +104,7 @@
},
};
-void __init igep3_flash_init(void)
+static void __init igep3_flash_init(void)
{
u8 cs = 0;
u8 onenandcs = GPMC_CS_NUM + 1;
@@ -137,12 +138,11 @@
}
#else
-void __init igep3_flash_init(void) {}
+static void __init igep3_flash_init(void) {}
#endif
-static struct regulator_consumer_supply igep3_vmmc1_supply = {
- .supply = "vmmc",
-};
+static struct regulator_consumer_supply igep3_vmmc1_supply =
+ REGULATOR_SUPPLY("vmmc", "mmci-omap-hs.0");
/* VMMC1 for OMAP VDD_MMC1 (i/o) and MMC1 card */
static struct regulator_init_data igep3_vmmc1 = {
@@ -159,6 +159,52 @@
.consumer_supplies = &igep3_vmmc1_supply,
};
+static struct regulator_consumer_supply igep3_vio_supply =
+ REGULATOR_SUPPLY("vmmc_aux", "mmci-omap-hs.1");
+
+static struct regulator_init_data igep3_vio = {
+ .constraints = {
+ .min_uV = 1800000,
+ .max_uV = 1800000,
+ .apply_uV = 1,
+ .valid_modes_mask = REGULATOR_MODE_NORMAL
+ | REGULATOR_MODE_STANDBY,
+ .valid_ops_mask = REGULATOR_CHANGE_VOLTAGE
+ | REGULATOR_CHANGE_MODE
+ | REGULATOR_CHANGE_STATUS,
+ },
+ .num_consumer_supplies = 1,
+ .consumer_supplies = &igep3_vio_supply,
+};
+
+static struct regulator_consumer_supply igep3_vmmc2_supply =
+ REGULATOR_SUPPLY("vmmc", "mmci-omap-hs.1");
+
+static struct regulator_init_data igep3_vmmc2 = {
+ .constraints = {
+ .valid_modes_mask = REGULATOR_MODE_NORMAL,
+ .always_on = 1,
+ },
+ .num_consumer_supplies = 1,
+ .consumer_supplies = &igep3_vmmc2_supply,
+};
+
+static struct fixed_voltage_config igep3_vwlan = {
+ .supply_name = "vwlan",
+ .microvolts = 3300000,
+ .gpio = -EINVAL,
+ .enabled_at_boot = 1,
+ .init_data = &igep3_vmmc2,
+};
+
+static struct platform_device igep3_vwlan_device = {
+ .name = "reg-fixed-voltage",
+ .id = 0,
+ .dev = {
+ .platform_data = &igep3_vwlan,
+ },
+};
+
static struct omap2_hsmmc_info mmc[] = {
[0] = {
.mmc = 1,
@@ -254,12 +300,6 @@
mmc[0].gpio_cd = gpio + 0;
omap2_hsmmc_init(mmc);
- /*
- * link regulators to MMC adapters ... we "know" the
- * regulators will be set up only *after* we return.
- */
- igep3_vmmc1_supply.dev = mmc[0].dev;
-
/* TWL4030_GPIO_MAX + 1 == ledB (out, active low LED) */
#if !defined(CONFIG_LEDS_GPIO) && !defined(CONFIG_LEDS_GPIO_MODULE)
if ((gpio_request(gpio+TWL4030_GPIO_MAX+1, "gpio-led:green:d1") == 0)
@@ -287,6 +327,10 @@
.usb_mode = T2_USB_MODE_ULPI,
};
+static struct platform_device *igep3_devices[] __initdata = {
+ &igep3_vwlan_device,
+};
+
static void __init igep3_init_irq(void)
{
omap2_init_common_infrastructure();
@@ -303,6 +347,7 @@
.usb = &igep3_twl4030_usb_data,
.gpio = &igep3_twl4030_gpio_pdata,
.vmmc1 = &igep3_vmmc1,
+ .vio = &igep3_vio,
};
static struct i2c_board_info __initdata igep3_i2c_boardinfo[] = {
@@ -363,8 +408,20 @@
void __init igep3_wifi_bt_init(void) {}
#endif
+static const struct ehci_hcd_omap_platform_data ehci_pdata __initconst = {
+ .port_mode[0] = EHCI_HCD_OMAP_MODE_UNKNOWN,
+ .port_mode[1] = EHCI_HCD_OMAP_MODE_PHY,
+ .port_mode[2] = EHCI_HCD_OMAP_MODE_UNKNOWN,
+
+ .phy_reset = true,
+ .reset_gpio_port[0] = -EINVAL,
+ .reset_gpio_port[1] = IGEP3_GPIO_USBH_NRESET,
+ .reset_gpio_port[2] = -EINVAL,
+};
+
#ifdef CONFIG_OMAP_MUX
static struct omap_board_mux board_mux[] __initdata = {
+ OMAP3_MUX(I2C2_SDA, OMAP_MUX_MODE4 | OMAP_PIN_OUTPUT),
{ .reg_offset = OMAP_MUX_TERMINATOR },
};
#endif
@@ -375,9 +432,10 @@
/* Register I2C busses and drivers */
igep3_i2c_init();
-
+ platform_add_devices(igep3_devices, ARRAY_SIZE(igep3_devices));
omap_serial_init();
usb_musb_init(&musb_board_data);
+ usb_ehci_init(&ehci_pdata);
igep3_flash_init();
igep3_leds_init();
@@ -392,6 +450,7 @@
MACHINE_START(IGEP0030, "IGEP OMAP3 module")
.boot_params = 0x80000100,
+ .reserve = omap_reserve,
.map_io = omap3_map_io,
.init_irq = igep3_init_irq,
.init_machine = igep3_init,
diff --git a/arch/arm/mach-omap2/board-omap3beagle.c b/arch/arm/mach-omap2/board-omap3beagle.c
index a4fe8e1..46d814a 100644
--- a/arch/arm/mach-omap2/board-omap3beagle.c
+++ b/arch/arm/mach-omap2/board-omap3beagle.c
@@ -207,7 +207,7 @@
.driver_name = "generic_dpi_panel",
.data = &dvi_panel,
.phy.dpi.data_lines = 24,
- .reset_gpio = 170,
+ .reset_gpio = -EINVAL,
};
static struct omap_dss_device beagle_tv_device = {
@@ -279,6 +279,8 @@
static int beagle_twl_gpio_setup(struct device *dev,
unsigned gpio, unsigned ngpio)
{
+ int r;
+
if (omap3_beagle_get_rev() == OMAP3BEAGLE_BOARD_XM) {
mmc[0].gpio_wp = -EINVAL;
} else if ((omap3_beagle_get_rev() == OMAP3BEAGLE_BOARD_C1_3) ||
@@ -299,17 +301,63 @@
/* REVISIT: need ehci-omap hooks for external VBUS
* power switch and overcurrent detect
*/
+ if (omap3_beagle_get_rev() != OMAP3BEAGLE_BOARD_XM) {
+ r = gpio_request(gpio + 1, "EHCI_nOC");
+ if (!r) {
+ r = gpio_direction_input(gpio + 1);
+ if (r)
+ gpio_free(gpio + 1);
+ }
+ if (r)
+ pr_err("%s: unable to configure EHCI_nOC\n", __func__);
+ }
- gpio_request(gpio + 1, "EHCI_nOC");
- gpio_direction_input(gpio + 1);
-
- /* TWL4030_GPIO_MAX + 0 == ledA, EHCI nEN_USB_PWR (out, active low) */
+ /*
+ * TWL4030_GPIO_MAX + 0 == ledA, EHCI nEN_USB_PWR (out, XM active
+ * high / others active low)
+ */
gpio_request(gpio + TWL4030_GPIO_MAX, "nEN_USB_PWR");
- gpio_direction_output(gpio + TWL4030_GPIO_MAX, 0);
+ if (omap3_beagle_get_rev() == OMAP3BEAGLE_BOARD_XM)
+ gpio_direction_output(gpio + TWL4030_GPIO_MAX, 1);
+ else
+ gpio_direction_output(gpio + TWL4030_GPIO_MAX, 0);
+
+ /* DVI reset GPIO is different between beagle revisions */
+ if (omap3_beagle_get_rev() == OMAP3BEAGLE_BOARD_XM)
+ beagle_dvi_device.reset_gpio = 129;
+ else
+ beagle_dvi_device.reset_gpio = 170;
/* TWL4030_GPIO_MAX + 1 == ledB, PMU_STAT (out, active low LED) */
gpio_leds[2].gpio = gpio + TWL4030_GPIO_MAX + 1;
+ /*
+ * gpio + 1 on Xm controls the TFP410's enable line (active low)
+ * gpio + 2 control varies depending on the board rev as follows:
+ * P7/P8 revisions(prototype): Camera EN
+ * A2+ revisions (production): LDO (supplies DVI, serial, led blocks)
+ */
+ if (omap3_beagle_get_rev() == OMAP3BEAGLE_BOARD_XM) {
+ r = gpio_request(gpio + 1, "nDVI_PWR_EN");
+ if (!r) {
+ r = gpio_direction_output(gpio + 1, 0);
+ if (r)
+ gpio_free(gpio + 1);
+ }
+ if (r)
+ pr_err("%s: unable to configure nDVI_PWR_EN\n",
+ __func__);
+ r = gpio_request(gpio + 2, "DVI_LDO_EN");
+ if (!r) {
+ r = gpio_direction_output(gpio + 2, 1);
+ if (r)
+ gpio_free(gpio + 2);
+ }
+ if (r)
+ pr_err("%s: unable to configure DVI_LDO_EN\n",
+ __func__);
+ }
+
return 0;
}
diff --git a/arch/arm/mach-omap2/board-omap4panda.c b/arch/arm/mach-omap2/board-omap4panda.c
index 3094e20..e001a04 100644
--- a/arch/arm/mach-omap2/board-omap4panda.c
+++ b/arch/arm/mach-omap2/board-omap4panda.c
@@ -19,6 +19,7 @@
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/platform_device.h>
+#include <linux/clk.h>
#include <linux/io.h>
#include <linux/leds.h>
#include <linux/gpio.h>
@@ -95,7 +96,16 @@
static void __init omap4_ehci_init(void)
{
int ret;
+ struct clk *phy_ref_clk;
+ /* FREF_CLK3 provides the 19.2 MHz reference clock to the PHY */
+ phy_ref_clk = clk_get(NULL, "auxclk3_ck");
+ if (IS_ERR(phy_ref_clk)) {
+ pr_err("Cannot request auxclk3\n");
+ goto error1;
+ }
+ clk_set_rate(phy_ref_clk, 19200000);
+ clk_enable(phy_ref_clk);
/* disable the power to the usb hub prior to init */
ret = gpio_request(GPIO_HUB_POWER, "hub_power");
diff --git a/arch/arm/mach-omap2/board-zoom-peripherals.c b/arch/arm/mach-omap2/board-zoom-peripherals.c
index 14d95af..e0e040f 100644
--- a/arch/arm/mach-omap2/board-zoom-peripherals.c
+++ b/arch/arm/mach-omap2/board-zoom-peripherals.c
@@ -192,7 +192,7 @@
},
};
-struct wl12xx_platform_data omap_zoom_wlan_data __initdata = {
+static struct wl12xx_platform_data omap_zoom_wlan_data __initdata = {
.irq = OMAP_GPIO_IRQ(OMAP_ZOOM_WLAN_IRQ_GPIO),
/* ZOOM ref clock is 26 MHz */
.board_ref_clock = 1,
@@ -286,7 +286,7 @@
}
/* EXTMUTE callback function */
-void zoom2_set_hs_extmute(int mute)
+static void zoom2_set_hs_extmute(int mute)
{
gpio_set_value(ZOOM2_HEADSET_EXTMUTE_GPIO, mute);
}
diff --git a/arch/arm/mach-omap2/clock3xxx_data.c b/arch/arm/mach-omap2/clock3xxx_data.c
index d3ab1c9e..403a4a1 100644
--- a/arch/arm/mach-omap2/clock3xxx_data.c
+++ b/arch/arm/mach-omap2/clock3xxx_data.c
@@ -3286,7 +3286,7 @@
CLK(NULL, "cpefuse_fck", &cpefuse_fck, CK_3430ES2PLUS | CK_AM35XX | CK_36XX),
CLK(NULL, "ts_fck", &ts_fck, CK_3430ES2PLUS | CK_AM35XX | CK_36XX),
CLK(NULL, "usbtll_fck", &usbtll_fck, CK_3430ES2PLUS | CK_AM35XX | CK_36XX),
- CLK("ehci-omap.0", "usbtll_fck", &usbtll_fck, CK_3430ES2 | CK_AM35XX),
+ CLK("ehci-omap.0", "usbtll_fck", &usbtll_fck, CK_3430ES2PLUS | CK_AM35XX | CK_36XX),
CLK("omap-mcbsp.1", "prcm_fck", &core_96m_fck, CK_3XXX),
CLK("omap-mcbsp.5", "prcm_fck", &core_96m_fck, CK_3XXX),
CLK(NULL, "core_96m_fck", &core_96m_fck, CK_3XXX),
diff --git a/arch/arm/mach-omap2/clockdomain.h b/arch/arm/mach-omap2/clockdomain.h
index de3faa2..9b459c2 100644
--- a/arch/arm/mach-omap2/clockdomain.h
+++ b/arch/arm/mach-omap2/clockdomain.h
@@ -103,9 +103,7 @@
const char *name;
struct powerdomain *ptr;
} pwrdm;
-#if defined(CONFIG_ARCH_OMAP2) || defined(CONFIG_ARCH_OMAP3)
const u16 clktrctrl_mask;
-#endif
const u8 flags;
const u8 dep_bit;
const u8 prcm_partition;
diff --git a/arch/arm/mach-omap2/cpuidle34xx.c b/arch/arm/mach-omap2/cpuidle34xx.c
index 11b89e96..f7b22a1 100644
--- a/arch/arm/mach-omap2/cpuidle34xx.c
+++ b/arch/arm/mach-omap2/cpuidle34xx.c
@@ -47,6 +47,8 @@
#define OMAP3_STATE_MAX OMAP3_STATE_C7
+#define CPUIDLE_FLAG_CHECK_BM 0x10000 /* use omap3_enter_idle_bm() */
+
struct omap3_processor_cx {
u8 valid;
u8 type;
diff --git a/arch/arm/mach-omap2/devices.c b/arch/arm/mach-omap2/devices.c
index 381f4eb..2c9c912 100644
--- a/arch/arm/mach-omap2/devices.c
+++ b/arch/arm/mach-omap2/devices.c
@@ -978,7 +978,7 @@
arch_initcall(omap2_init_devices);
#if defined(CONFIG_OMAP_WATCHDOG) || defined(CONFIG_OMAP_WATCHDOG_MODULE)
-struct omap_device_pm_latency omap_wdt_latency[] = {
+static struct omap_device_pm_latency omap_wdt_latency[] = {
[0] = {
.deactivate_func = omap_device_idle_hwmods,
.activate_func = omap_device_enable_hwmods,
diff --git a/arch/arm/mach-omap2/irq.c b/arch/arm/mach-omap2/irq.c
index 85bf8ca..23049c4 100644
--- a/arch/arm/mach-omap2/irq.c
+++ b/arch/arm/mach-omap2/irq.c
@@ -100,13 +100,14 @@
}
/* XXX: FIQ and additional INTC support (only MPU at the moment) */
-static void omap_ack_irq(unsigned int irq)
+static void omap_ack_irq(struct irq_data *d)
{
intc_bank_write_reg(0x1, &irq_banks[0], INTC_CONTROL);
}
-static void omap_mask_irq(unsigned int irq)
+static void omap_mask_irq(struct irq_data *d)
{
+ unsigned int irq = d->irq;
int offset = irq & (~(IRQ_BITS_PER_REG - 1));
if (cpu_is_omap34xx()) {
@@ -128,8 +129,9 @@
intc_bank_write_reg(1 << irq, &irq_banks[0], INTC_MIR_SET0 + offset);
}
-static void omap_unmask_irq(unsigned int irq)
+static void omap_unmask_irq(struct irq_data *d)
{
+ unsigned int irq = d->irq;
int offset = irq & (~(IRQ_BITS_PER_REG - 1));
irq &= (IRQ_BITS_PER_REG - 1);
@@ -137,17 +139,17 @@
intc_bank_write_reg(1 << irq, &irq_banks[0], INTC_MIR_CLEAR0 + offset);
}
-static void omap_mask_ack_irq(unsigned int irq)
+static void omap_mask_ack_irq(struct irq_data *d)
{
- omap_mask_irq(irq);
- omap_ack_irq(irq);
+ omap_mask_irq(d);
+ omap_ack_irq(d);
}
static struct irq_chip omap_irq_chip = {
- .name = "INTC",
- .ack = omap_mask_ack_irq,
- .mask = omap_mask_irq,
- .unmask = omap_unmask_irq,
+ .name = "INTC",
+ .irq_ack = omap_mask_ack_irq,
+ .irq_mask = omap_mask_irq,
+ .irq_unmask = omap_unmask_irq,
};
static void __init omap_irq_bank_init_one(struct omap_irq_bank *bank)
diff --git a/arch/arm/mach-omap2/mux.c b/arch/arm/mach-omap2/mux.c
index 17bd639..df8d2f2 100644
--- a/arch/arm/mach-omap2/mux.c
+++ b/arch/arm/mach-omap2/mux.c
@@ -893,7 +893,7 @@
return NULL;
m = &entry->mux;
- memcpy(m, src, sizeof(struct omap_mux_entry));
+ entry->mux = *src;
#ifdef CONFIG_OMAP_MUX
if (omap_mux_copy_names(src, m)) {
diff --git a/arch/arm/mach-omap2/mux34xx.c b/arch/arm/mach-omap2/mux34xx.c
index 440c98e..17f80e4 100644
--- a/arch/arm/mach-omap2/mux34xx.c
+++ b/arch/arm/mach-omap2/mux34xx.c
@@ -703,7 +703,7 @@
* Signals different on CBC package compared to the superset
*/
#if defined(CONFIG_OMAP_MUX) && defined(CONFIG_OMAP_PACKAGE_CBC)
-struct omap_mux __initdata omap3_cbc_subset[] = {
+static struct omap_mux __initdata omap3_cbc_subset[] = {
{ .reg_offset = OMAP_MUX_TERMINATOR },
};
#else
@@ -721,7 +721,7 @@
*/
#if defined(CONFIG_OMAP_MUX) && defined(CONFIG_DEBUG_FS) \
&& defined(CONFIG_OMAP_PACKAGE_CBC)
-struct omap_ball __initdata omap3_cbc_ball[] = {
+static struct omap_ball __initdata omap3_cbc_ball[] = {
_OMAP3_BALLENTRY(CAM_D0, "ae16", NULL),
_OMAP3_BALLENTRY(CAM_D1, "ae15", NULL),
_OMAP3_BALLENTRY(CAM_D10, "d25", NULL),
diff --git a/arch/arm/mach-omap2/mux44xx.c b/arch/arm/mach-omap2/mux44xx.c
index 980f11d..c322e7b 100644
--- a/arch/arm/mach-omap2/mux44xx.c
+++ b/arch/arm/mach-omap2/mux44xx.c
@@ -544,7 +544,7 @@
*/
#if defined(CONFIG_OMAP_MUX) && defined(CONFIG_DEBUG_FS) \
&& defined(CONFIG_OMAP_PACKAGE_CBL)
-struct omap_ball __initdata omap4_core_cbl_ball[] = {
+static struct omap_ball __initdata omap4_core_cbl_ball[] = {
_OMAP4_BALLENTRY(GPMC_AD0, "c12", NULL),
_OMAP4_BALLENTRY(GPMC_AD1, "d12", NULL),
_OMAP4_BALLENTRY(GPMC_AD2, "c13", NULL),
@@ -1262,7 +1262,7 @@
*/
#if defined(CONFIG_OMAP_MUX) && defined(CONFIG_DEBUG_FS) \
&& defined(CONFIG_OMAP_PACKAGE_CBS)
-struct omap_ball __initdata omap4_core_cbs_ball[] = {
+static struct omap_ball __initdata omap4_core_cbs_ball[] = {
_OMAP4_BALLENTRY(GPMC_AD0, "c12", NULL),
_OMAP4_BALLENTRY(GPMC_AD1, "d12", NULL),
_OMAP4_BALLENTRY(GPMC_AD2, "c13", NULL),
@@ -1546,7 +1546,7 @@
*/
#if defined(CONFIG_OMAP_MUX) && defined(CONFIG_DEBUG_FS) \
&& defined(CONFIG_OMAP_PACKAGE_CBL)
-struct omap_ball __initdata omap4_wkup_cbl_cbs_ball[] = {
+static struct omap_ball __initdata omap4_wkup_cbl_cbs_ball[] = {
_OMAP4_BALLENTRY(SIM_IO, "h4", NULL),
_OMAP4_BALLENTRY(SIM_CLK, "j2", NULL),
_OMAP4_BALLENTRY(SIM_RESET, "g2", NULL),
diff --git a/arch/arm/mach-omap2/omap_twl.c b/arch/arm/mach-omap2/omap_twl.c
index 15f8c6c..00e1d2b 100644
--- a/arch/arm/mach-omap2/omap_twl.c
+++ b/arch/arm/mach-omap2/omap_twl.c
@@ -20,6 +20,8 @@
#include <plat/voltage.h>
+#include "pm.h"
+
#define OMAP3_SRI2C_SLAVE_ADDR 0x12
#define OMAP3_VDD_MPU_SR_CONTROL_REG 0x00
#define OMAP3_VDD_CORE_SR_CONTROL_REG 0x01
@@ -60,17 +62,17 @@
#define REG_SMPS_OFFSET 0xE0
-unsigned long twl4030_vsel_to_uv(const u8 vsel)
+static unsigned long twl4030_vsel_to_uv(const u8 vsel)
{
return (((vsel * 125) + 6000)) * 100;
}
-u8 twl4030_uv_to_vsel(unsigned long uv)
+static u8 twl4030_uv_to_vsel(unsigned long uv)
{
return DIV_ROUND_UP(uv - 600000, 12500);
}
-unsigned long twl6030_vsel_to_uv(const u8 vsel)
+static unsigned long twl6030_vsel_to_uv(const u8 vsel)
{
/*
* In TWL6030 depending on the value of SMPS_OFFSET
@@ -102,7 +104,7 @@
return ((((vsel - 1) * 125) + 6000)) * 100;
}
-u8 twl6030_uv_to_vsel(unsigned long uv)
+static u8 twl6030_uv_to_vsel(unsigned long uv)
{
/*
* In TWL6030 depending on the value of SMPS_OFFSET
diff --git a/arch/arm/mach-omap2/pm_bus.c b/arch/arm/mach-omap2/pm_bus.c
index 784989f..5acd2ab 100644
--- a/arch/arm/mach-omap2/pm_bus.c
+++ b/arch/arm/mach-omap2/pm_bus.c
@@ -20,7 +20,7 @@
#include <plat/omap-pm.h>
#ifdef CONFIG_PM_RUNTIME
-int omap_pm_runtime_suspend(struct device *dev)
+static int omap_pm_runtime_suspend(struct device *dev)
{
struct platform_device *pdev = to_platform_device(dev);
int r, ret = 0;
@@ -37,7 +37,7 @@
return ret;
};
-int omap_pm_runtime_resume(struct device *dev)
+static int omap_pm_runtime_resume(struct device *dev)
{
struct platform_device *pdev = to_platform_device(dev);
int r;
diff --git a/arch/arm/mach-omap2/prm2xxx_3xxx.h b/arch/arm/mach-omap2/prm2xxx_3xxx.h
index 53d44f6..49654c8 100644
--- a/arch/arm/mach-omap2/prm2xxx_3xxx.h
+++ b/arch/arm/mach-omap2/prm2xxx_3xxx.h
@@ -228,7 +228,67 @@
#ifndef __ASSEMBLER__
-
+/*
+ * Stub omap2xxx/omap3xxx functions so that common files
+ * continue to build when custom builds are used
+ */
+#if defined(CONFIG_ARCH_OMAP4) && !(defined(CONFIG_ARCH_OMAP2) || \
+ defined(CONFIG_ARCH_OMAP3))
+static inline u32 omap2_prm_read_mod_reg(s16 module, u16 idx)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+ return 0;
+}
+static inline void omap2_prm_write_mod_reg(u32 val, s16 module, u16 idx)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+}
+static inline u32 omap2_prm_rmw_mod_reg_bits(u32 mask, u32 bits,
+ s16 module, s16 idx)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+ return 0;
+}
+static inline u32 omap2_prm_set_mod_reg_bits(u32 bits, s16 module, s16 idx)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+ return 0;
+}
+static inline u32 omap2_prm_clear_mod_reg_bits(u32 bits, s16 module, s16 idx)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+ return 0;
+}
+static inline u32 omap2_prm_read_mod_bits_shift(s16 domain, s16 idx, u32 mask)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+ return 0;
+}
+static inline int omap2_prm_is_hardreset_asserted(s16 prm_mod, u8 shift)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+ return 0;
+}
+static inline int omap2_prm_assert_hardreset(s16 prm_mod, u8 shift)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+ return 0;
+}
+static inline int omap2_prm_deassert_hardreset(s16 prm_mod, u8 shift)
+{
+ WARN(1, "prm: omap2xxx/omap3xxx specific function and "
+ "not suppose to be used on omap4\n");
+ return 0;
+}
+#else
/* Power/reset management domain register get/set */
extern u32 omap2_prm_read_mod_reg(s16 module, u16 idx);
extern void omap2_prm_write_mod_reg(u32 val, s16 module, u16 idx);
@@ -242,6 +302,7 @@
extern int omap2_prm_assert_hardreset(s16 prm_mod, u8 shift);
extern int omap2_prm_deassert_hardreset(s16 prm_mod, u8 shift);
+#endif /* CONFIG_ARCH_OMAP4 */
#endif
/*
diff --git a/arch/arm/mach-omap2/sr_device.c b/arch/arm/mach-omap2/sr_device.c
index 786d685..b1e0af1 100644
--- a/arch/arm/mach-omap2/sr_device.c
+++ b/arch/arm/mach-omap2/sr_device.c
@@ -27,6 +27,7 @@
#include <plat/voltage.h>
#include "control.h"
+#include "pm.h"
static bool sr_enable_on_init;
diff --git a/arch/arm/mach-omap2/wd_timer.c b/arch/arm/mach-omap2/wd_timer.c
index b0c4907..4067669 100644
--- a/arch/arm/mach-omap2/wd_timer.c
+++ b/arch/arm/mach-omap2/wd_timer.c
@@ -13,6 +13,8 @@
#include <plat/omap_hwmod.h>
+#include "wd_timer.h"
+
/*
* In order to avoid any assumptions from bootloader regarding WDT
* settings, WDT module is reset during init. This enables the watchdog
diff --git a/arch/arm/mach-pnx4008/irq.c b/arch/arm/mach-pnx4008/irq.c
index a9ce02b..c69c180 100644
--- a/arch/arm/mach-pnx4008/irq.c
+++ b/arch/arm/mach-pnx4008/irq.c
@@ -36,44 +36,44 @@
static u8 pnx4008_irq_type[NR_IRQS] = PNX4008_IRQ_TYPES;
-static void pnx4008_mask_irq(unsigned int irq)
+static void pnx4008_mask_irq(struct irq_data *d)
{
- __raw_writel(__raw_readl(INTC_ER(irq)) & ~INTC_BIT(irq), INTC_ER(irq)); /* mask interrupt */
+ __raw_writel(__raw_readl(INTC_ER(d->irq)) & ~INTC_BIT(d->irq), INTC_ER(d->irq)); /* mask interrupt */
}
-static void pnx4008_unmask_irq(unsigned int irq)
+static void pnx4008_unmask_irq(struct irq_data *d)
{
- __raw_writel(__raw_readl(INTC_ER(irq)) | INTC_BIT(irq), INTC_ER(irq)); /* unmask interrupt */
+ __raw_writel(__raw_readl(INTC_ER(d->irq)) | INTC_BIT(d->irq), INTC_ER(d->irq)); /* unmask interrupt */
}
-static void pnx4008_mask_ack_irq(unsigned int irq)
+static void pnx4008_mask_ack_irq(struct irq_data *d)
{
- __raw_writel(__raw_readl(INTC_ER(irq)) & ~INTC_BIT(irq), INTC_ER(irq)); /* mask interrupt */
- __raw_writel(INTC_BIT(irq), INTC_SR(irq)); /* clear interrupt status */
+ __raw_writel(__raw_readl(INTC_ER(d->irq)) & ~INTC_BIT(d->irq), INTC_ER(d->irq)); /* mask interrupt */
+ __raw_writel(INTC_BIT(d->irq), INTC_SR(d->irq)); /* clear interrupt status */
}
-static int pnx4008_set_irq_type(unsigned int irq, unsigned int type)
+static int pnx4008_set_irq_type(struct irq_data *d, unsigned int type)
{
switch (type) {
case IRQ_TYPE_EDGE_RISING:
- __raw_writel(__raw_readl(INTC_ATR(irq)) | INTC_BIT(irq), INTC_ATR(irq)); /*edge sensitive */
- __raw_writel(__raw_readl(INTC_APR(irq)) | INTC_BIT(irq), INTC_APR(irq)); /*rising edge */
- set_irq_handler(irq, handle_edge_irq);
+ __raw_writel(__raw_readl(INTC_ATR(d->irq)) | INTC_BIT(d->irq), INTC_ATR(d->irq)); /*edge sensitive */
+ __raw_writel(__raw_readl(INTC_APR(d->irq)) | INTC_BIT(d->irq), INTC_APR(d->irq)); /*rising edge */
+ set_irq_handler(d->irq, handle_edge_irq);
break;
case IRQ_TYPE_EDGE_FALLING:
- __raw_writel(__raw_readl(INTC_ATR(irq)) | INTC_BIT(irq), INTC_ATR(irq)); /*edge sensitive */
- __raw_writel(__raw_readl(INTC_APR(irq)) & ~INTC_BIT(irq), INTC_APR(irq)); /*falling edge */
- set_irq_handler(irq, handle_edge_irq);
+ __raw_writel(__raw_readl(INTC_ATR(d->irq)) | INTC_BIT(d->irq), INTC_ATR(d->irq)); /*edge sensitive */
+ __raw_writel(__raw_readl(INTC_APR(d->irq)) & ~INTC_BIT(d->irq), INTC_APR(d->irq)); /*falling edge */
+ set_irq_handler(d->irq, handle_edge_irq);
break;
case IRQ_TYPE_LEVEL_LOW:
- __raw_writel(__raw_readl(INTC_ATR(irq)) & ~INTC_BIT(irq), INTC_ATR(irq)); /*level sensitive */
- __raw_writel(__raw_readl(INTC_APR(irq)) & ~INTC_BIT(irq), INTC_APR(irq)); /*low level */
- set_irq_handler(irq, handle_level_irq);
+ __raw_writel(__raw_readl(INTC_ATR(d->irq)) & ~INTC_BIT(d->irq), INTC_ATR(d->irq)); /*level sensitive */
+ __raw_writel(__raw_readl(INTC_APR(d->irq)) & ~INTC_BIT(d->irq), INTC_APR(d->irq)); /*low level */
+ set_irq_handler(d->irq, handle_level_irq);
break;
case IRQ_TYPE_LEVEL_HIGH:
- __raw_writel(__raw_readl(INTC_ATR(irq)) & ~INTC_BIT(irq), INTC_ATR(irq)); /*level sensitive */
- __raw_writel(__raw_readl(INTC_APR(irq)) | INTC_BIT(irq), INTC_APR(irq)); /* high level */
- set_irq_handler(irq, handle_level_irq);
+ __raw_writel(__raw_readl(INTC_ATR(d->irq)) & ~INTC_BIT(d->irq), INTC_ATR(d->irq)); /*level sensitive */
+ __raw_writel(__raw_readl(INTC_APR(d->irq)) | INTC_BIT(d->irq), INTC_APR(d->irq)); /* high level */
+ set_irq_handler(d->irq, handle_level_irq);
break;
/* IRQ_TYPE_EDGE_BOTH is not supported */
@@ -85,10 +85,10 @@
}
static struct irq_chip pnx4008_irq_chip = {
- .ack = pnx4008_mask_ack_irq,
- .mask = pnx4008_mask_irq,
- .unmask = pnx4008_unmask_irq,
- .set_type = pnx4008_set_irq_type,
+ .irq_ack = pnx4008_mask_ack_irq,
+ .irq_mask = pnx4008_mask_irq,
+ .irq_unmask = pnx4008_unmask_irq,
+ .irq_set_type = pnx4008_set_irq_type,
};
void __init pnx4008_init_irq(void)
@@ -99,14 +99,18 @@
for (i = 0; i < NR_IRQS; i++) {
set_irq_flags(i, IRQF_VALID);
set_irq_chip(i, &pnx4008_irq_chip);
- pnx4008_set_irq_type(i, pnx4008_irq_type[i]);
+ pnx4008_set_irq_type(irq_get_irq_data(i), pnx4008_irq_type[i]);
}
/* configure and enable IRQ 0,1,30,31 (cascade interrupts) */
- pnx4008_set_irq_type(SUB1_IRQ_N, pnx4008_irq_type[SUB1_IRQ_N]);
- pnx4008_set_irq_type(SUB2_IRQ_N, pnx4008_irq_type[SUB2_IRQ_N]);
- pnx4008_set_irq_type(SUB1_FIQ_N, pnx4008_irq_type[SUB1_FIQ_N]);
- pnx4008_set_irq_type(SUB2_FIQ_N, pnx4008_irq_type[SUB2_FIQ_N]);
+ pnx4008_set_irq_type(irq_get_irq_data(SUB1_IRQ_N),
+ pnx4008_irq_type[SUB1_IRQ_N]);
+ pnx4008_set_irq_type(irq_get_irq_data(SUB2_IRQ_N),
+ pnx4008_irq_type[SUB2_IRQ_N]);
+ pnx4008_set_irq_type(irq_get_irq_data(SUB1_FIQ_N),
+ pnx4008_irq_type[SUB1_FIQ_N]);
+ pnx4008_set_irq_type(irq_get_irq_data(SUB2_FIQ_N),
+ pnx4008_irq_type[SUB2_FIQ_N]);
/* mask all others */
__raw_writel((1 << SUB2_FIQ_N) | (1 << SUB1_FIQ_N) |
diff --git a/arch/arm/mach-pxa/balloon3.c b/arch/arm/mach-pxa/balloon3.c
index ccb2d0c..a134a14 100644
--- a/arch/arm/mach-pxa/balloon3.c
+++ b/arch/arm/mach-pxa/balloon3.c
@@ -477,25 +477,25 @@
/******************************************************************************
* FPGA IRQ
******************************************************************************/
-static void balloon3_mask_irq(unsigned int irq)
+static void balloon3_mask_irq(struct irq_data *d)
{
- int balloon3_irq = (irq - BALLOON3_IRQ(0));
+ int balloon3_irq = (d->irq - BALLOON3_IRQ(0));
balloon3_irq_enabled &= ~(1 << balloon3_irq);
__raw_writel(~balloon3_irq_enabled, BALLOON3_INT_CONTROL_REG);
}
-static void balloon3_unmask_irq(unsigned int irq)
+static void balloon3_unmask_irq(struct irq_data *d)
{
- int balloon3_irq = (irq - BALLOON3_IRQ(0));
+ int balloon3_irq = (d->irq - BALLOON3_IRQ(0));
balloon3_irq_enabled |= (1 << balloon3_irq);
__raw_writel(~balloon3_irq_enabled, BALLOON3_INT_CONTROL_REG);
}
static struct irq_chip balloon3_irq_chip = {
.name = "FPGA",
- .ack = balloon3_mask_irq,
- .mask = balloon3_mask_irq,
- .unmask = balloon3_unmask_irq,
+ .irq_ack = balloon3_mask_irq,
+ .irq_mask = balloon3_mask_irq,
+ .irq_unmask = balloon3_unmask_irq,
};
static void balloon3_irq_handler(unsigned int irq, struct irq_desc *desc)
@@ -504,8 +504,13 @@
balloon3_irq_enabled;
do {
/* clear useless edge notification */
- if (desc->chip->ack)
- desc->chip->ack(BALLOON3_AUX_NIRQ);
+ if (desc->irq_data.chip->irq_ack) {
+ struct irq_data *d;
+
+ d = irq_get_irq_data(BALLOON3_AUX_NIRQ);
+ desc->irq_data.chip->irq_ack(d);
+ }
+
while (pending) {
irq = BALLOON3_IRQ(0) + __ffs(pending);
generic_handle_irq(irq);
diff --git a/arch/arm/mach-pxa/clock-pxa3xx.c b/arch/arm/mach-pxa/clock-pxa3xx.c
index 1b08a34..3f864cd 100644
--- a/arch/arm/mach-pxa/clock-pxa3xx.c
+++ b/arch/arm/mach-pxa/clock-pxa3xx.c
@@ -115,7 +115,6 @@
{
unsigned long acsr = ACSR;
unsigned long memclkcfg = __raw_readl(MEMCLKCFG);
- unsigned int smcfs = (acsr >> 23) & 0x7;
return BASE_CLK * smcfs_mult[(acsr >> 23) & 0x7] /
df_clkdiv[(memclkcfg >> 16) & 0x3];
diff --git a/arch/arm/mach-pxa/cm-x2xx-pci.c b/arch/arm/mach-pxa/cm-x2xx-pci.c
index 0f31305..a2380cd 100644
--- a/arch/arm/mach-pxa/cm-x2xx-pci.c
+++ b/arch/arm/mach-pxa/cm-x2xx-pci.c
@@ -59,7 +59,7 @@
static void cmx2xx_it8152_irq_demux(unsigned int irq, struct irq_desc *desc)
{
/* clear our parent irq */
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
it8152_irq_demux(irq, desc);
}
diff --git a/arch/arm/mach-pxa/generic.c b/arch/arm/mach-pxa/generic.c
index d6e15f7..f5d91ef 100644
--- a/arch/arm/mach-pxa/generic.c
+++ b/arch/arm/mach-pxa/generic.c
@@ -22,7 +22,6 @@
#include <mach/hardware.h>
#include <asm/system.h>
-#include <asm/pgtable.h>
#include <asm/mach/map.h>
#include <asm/mach-types.h>
diff --git a/arch/arm/mach-pxa/generic.h b/arch/arm/mach-pxa/generic.h
index 6205dc9..a079d8b 100644
--- a/arch/arm/mach-pxa/generic.h
+++ b/arch/arm/mach-pxa/generic.h
@@ -9,11 +9,13 @@
* published by the Free Software Foundation.
*/
+struct irq_data;
struct sys_timer;
extern struct sys_timer pxa_timer;
extern void __init pxa_init_irq(int irq_nr,
- int (*set_wake)(unsigned int, unsigned int));
+ int (*set_wake)(struct irq_data *,
+ unsigned int));
extern void __init pxa25x_init_irq(void);
#ifdef CONFIG_CPU_PXA26x
extern void __init pxa26x_init_irq(void);
diff --git a/arch/arm/mach-pxa/irq.c b/arch/arm/mach-pxa/irq.c
index 54e91c9..2693e3c 100644
--- a/arch/arm/mach-pxa/irq.c
+++ b/arch/arm/mach-pxa/irq.c
@@ -53,37 +53,48 @@
return !cpu_is_pxa25x();
}
-static void pxa_mask_irq(unsigned int irq)
+static inline void __iomem *irq_base(int i)
{
- void __iomem *base = get_irq_chip_data(irq);
+ static unsigned long phys_base[] = {
+ 0x40d00000,
+ 0x40d0009c,
+ 0x40d00130,
+ };
+
+ return (void __iomem *)io_p2v(phys_base[i]);
+}
+
+static void pxa_mask_irq(struct irq_data *d)
+{
+ void __iomem *base = irq_data_get_irq_chip_data(d);
uint32_t icmr = __raw_readl(base + ICMR);
- icmr &= ~(1 << IRQ_BIT(irq));
+ icmr &= ~(1 << IRQ_BIT(d->irq));
__raw_writel(icmr, base + ICMR);
}
-static void pxa_unmask_irq(unsigned int irq)
+static void pxa_unmask_irq(struct irq_data *d)
{
- void __iomem *base = get_irq_chip_data(irq);
+ void __iomem *base = irq_data_get_irq_chip_data(d);
uint32_t icmr = __raw_readl(base + ICMR);
- icmr |= 1 << IRQ_BIT(irq);
+ icmr |= 1 << IRQ_BIT(d->irq);
__raw_writel(icmr, base + ICMR);
}
static struct irq_chip pxa_internal_irq_chip = {
.name = "SC",
- .ack = pxa_mask_irq,
- .mask = pxa_mask_irq,
- .unmask = pxa_unmask_irq,
+ .irq_ack = pxa_mask_irq,
+ .irq_mask = pxa_mask_irq,
+ .irq_unmask = pxa_unmask_irq,
};
/*
* GPIO IRQs for GPIO 0 and 1
*/
-static int pxa_set_low_gpio_type(unsigned int irq, unsigned int type)
+static int pxa_set_low_gpio_type(struct irq_data *d, unsigned int type)
{
- int gpio = irq - IRQ_GPIO0;
+ int gpio = d->irq - IRQ_GPIO0;
if (__gpio_is_occupied(gpio)) {
pr_err("%s failed: GPIO is configured\n", __func__);
@@ -103,31 +114,17 @@
return 0;
}
-static void pxa_ack_low_gpio(unsigned int irq)
+static void pxa_ack_low_gpio(struct irq_data *d)
{
- GEDR0 = (1 << (irq - IRQ_GPIO0));
-}
-
-static void pxa_mask_low_gpio(unsigned int irq)
-{
- struct irq_desc *desc = irq_to_desc(irq);
-
- desc->chip->mask(irq);
-}
-
-static void pxa_unmask_low_gpio(unsigned int irq)
-{
- struct irq_desc *desc = irq_to_desc(irq);
-
- desc->chip->unmask(irq);
+ GEDR0 = (1 << (d->irq - IRQ_GPIO0));
}
static struct irq_chip pxa_low_gpio_chip = {
.name = "GPIO-l",
- .ack = pxa_ack_low_gpio,
- .mask = pxa_mask_low_gpio,
- .unmask = pxa_unmask_low_gpio,
- .set_type = pxa_set_low_gpio_type,
+ .irq_ack = pxa_ack_low_gpio,
+ .irq_mask = pxa_mask_irq,
+ .irq_unmask = pxa_unmask_irq,
+ .irq_set_type = pxa_set_low_gpio_type,
};
static void __init pxa_init_low_gpio_irq(set_wake_t fn)
@@ -141,22 +138,12 @@
for (irq = IRQ_GPIO0; irq <= IRQ_GPIO1; irq++) {
set_irq_chip(irq, &pxa_low_gpio_chip);
+ set_irq_chip_data(irq, irq_base(0));
set_irq_handler(irq, handle_edge_irq);
set_irq_flags(irq, IRQF_VALID);
}
- pxa_low_gpio_chip.set_wake = fn;
-}
-
-static inline void __iomem *irq_base(int i)
-{
- static unsigned long phys_base[] = {
- 0x40d00000,
- 0x40d0009c,
- 0x40d00130,
- };
-
- return (void __iomem *)io_p2v(phys_base[i >> 5]);
+ pxa_low_gpio_chip.irq_set_wake = fn;
}
void __init pxa_init_irq(int irq_nr, set_wake_t fn)
@@ -168,7 +155,7 @@
pxa_internal_irq_nr = irq_nr;
for (n = 0; n < irq_nr; n += 32) {
- void __iomem *base = irq_base(n);
+ void __iomem *base = irq_base(n >> 5);
__raw_writel(0, base + ICMR); /* disable all IRQs */
__raw_writel(0, base + ICLR); /* all IRQs are IRQ, not FIQ */
@@ -188,7 +175,7 @@
/* only unmasked interrupts kick us out of idle */
__raw_writel(1, irq_base(0) + ICCR);
- pxa_internal_irq_chip.set_wake = fn;
+ pxa_internal_irq_chip.irq_set_wake = fn;
pxa_init_low_gpio_irq(fn);
}
@@ -200,7 +187,7 @@
{
int i;
- for (i = 0; i < pxa_internal_irq_nr; i += 32) {
+ for (i = 0; i < pxa_internal_irq_nr / 32; i++) {
void __iomem *base = irq_base(i);
saved_icmr[i] = __raw_readl(base + ICMR);
@@ -219,14 +206,14 @@
{
int i;
- for (i = 0; i < pxa_internal_irq_nr; i += 32) {
+ for (i = 0; i < pxa_internal_irq_nr / 32; i++) {
void __iomem *base = irq_base(i);
__raw_writel(saved_icmr[i], base + ICMR);
__raw_writel(0, base + ICLR);
}
- if (!cpu_is_pxa25x())
+ if (cpu_has_ipr())
for (i = 0; i < pxa_internal_irq_nr; i++)
__raw_writel(saved_ipr[i], IRQ_BASE + IPR(i));
diff --git a/arch/arm/mach-pxa/lpd270.c b/arch/arm/mach-pxa/lpd270.c
index 8ab62a6..c9a3e77 100644
--- a/arch/arm/mach-pxa/lpd270.c
+++ b/arch/arm/mach-pxa/lpd270.c
@@ -95,9 +95,9 @@
static unsigned int lpd270_irq_enabled;
-static void lpd270_mask_irq(unsigned int irq)
+static void lpd270_mask_irq(struct irq_data *d)
{
- int lpd270_irq = irq - LPD270_IRQ(0);
+ int lpd270_irq = d->irq - LPD270_IRQ(0);
__raw_writew(~(1 << lpd270_irq), LPD270_INT_STATUS);
@@ -105,9 +105,9 @@
__raw_writew(lpd270_irq_enabled, LPD270_INT_MASK);
}
-static void lpd270_unmask_irq(unsigned int irq)
+static void lpd270_unmask_irq(struct irq_data *d)
{
- int lpd270_irq = irq - LPD270_IRQ(0);
+ int lpd270_irq = d->irq - LPD270_IRQ(0);
lpd270_irq_enabled |= 1 << lpd270_irq;
__raw_writew(lpd270_irq_enabled, LPD270_INT_MASK);
@@ -115,9 +115,9 @@
static struct irq_chip lpd270_irq_chip = {
.name = "CPLD",
- .ack = lpd270_mask_irq,
- .mask = lpd270_mask_irq,
- .unmask = lpd270_unmask_irq,
+ .irq_ack = lpd270_mask_irq,
+ .irq_mask = lpd270_mask_irq,
+ .irq_unmask = lpd270_unmask_irq,
};
static void lpd270_irq_handler(unsigned int irq, struct irq_desc *desc)
@@ -126,7 +126,8 @@
pending = __raw_readw(LPD270_INT_STATUS) & lpd270_irq_enabled;
do {
- desc->chip->ack(irq); /* clear useless edge notification */
+ /* clear useless edge notification */
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
if (likely(pending)) {
irq = LPD270_IRQ(0) + __ffs(pending);
generic_handle_irq(irq);
diff --git a/arch/arm/mach-pxa/lubbock.c b/arch/arm/mach-pxa/lubbock.c
index 3072dbe..dca20de 100644
--- a/arch/arm/mach-pxa/lubbock.c
+++ b/arch/arm/mach-pxa/lubbock.c
@@ -122,15 +122,15 @@
static unsigned long lubbock_irq_enabled;
-static void lubbock_mask_irq(unsigned int irq)
+static void lubbock_mask_irq(struct irq_data *d)
{
- int lubbock_irq = (irq - LUBBOCK_IRQ(0));
+ int lubbock_irq = (d->irq - LUBBOCK_IRQ(0));
LUB_IRQ_MASK_EN = (lubbock_irq_enabled &= ~(1 << lubbock_irq));
}
-static void lubbock_unmask_irq(unsigned int irq)
+static void lubbock_unmask_irq(struct irq_data *d)
{
- int lubbock_irq = (irq - LUBBOCK_IRQ(0));
+ int lubbock_irq = (d->irq - LUBBOCK_IRQ(0));
/* the irq can be acknowledged only if deasserted, so it's done here */
LUB_IRQ_SET_CLR &= ~(1 << lubbock_irq);
LUB_IRQ_MASK_EN = (lubbock_irq_enabled |= (1 << lubbock_irq));
@@ -138,16 +138,17 @@
static struct irq_chip lubbock_irq_chip = {
.name = "FPGA",
- .ack = lubbock_mask_irq,
- .mask = lubbock_mask_irq,
- .unmask = lubbock_unmask_irq,
+ .irq_ack = lubbock_mask_irq,
+ .irq_mask = lubbock_mask_irq,
+ .irq_unmask = lubbock_unmask_irq,
};
static void lubbock_irq_handler(unsigned int irq, struct irq_desc *desc)
{
unsigned long pending = LUB_IRQ_SET_CLR & lubbock_irq_enabled;
do {
- desc->chip->ack(irq); /* clear our parent irq */
+ /* clear our parent irq */
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
if (likely(pending)) {
irq = LUBBOCK_IRQ(0) + __ffs(pending);
generic_handle_irq(irq);
diff --git a/arch/arm/mach-pxa/mainstone.c b/arch/arm/mach-pxa/mainstone.c
index 740c035..d4b6f23 100644
--- a/arch/arm/mach-pxa/mainstone.c
+++ b/arch/arm/mach-pxa/mainstone.c
@@ -123,15 +123,15 @@
static unsigned long mainstone_irq_enabled;
-static void mainstone_mask_irq(unsigned int irq)
+static void mainstone_mask_irq(struct irq_data *d)
{
- int mainstone_irq = (irq - MAINSTONE_IRQ(0));
+ int mainstone_irq = (d->irq - MAINSTONE_IRQ(0));
MST_INTMSKENA = (mainstone_irq_enabled &= ~(1 << mainstone_irq));
}
-static void mainstone_unmask_irq(unsigned int irq)
+static void mainstone_unmask_irq(struct irq_data *d)
{
- int mainstone_irq = (irq - MAINSTONE_IRQ(0));
+ int mainstone_irq = (d->irq - MAINSTONE_IRQ(0));
/* the irq can be acknowledged only if deasserted, so it's done here */
MST_INTSETCLR &= ~(1 << mainstone_irq);
MST_INTMSKENA = (mainstone_irq_enabled |= (1 << mainstone_irq));
@@ -139,16 +139,17 @@
static struct irq_chip mainstone_irq_chip = {
.name = "FPGA",
- .ack = mainstone_mask_irq,
- .mask = mainstone_mask_irq,
- .unmask = mainstone_unmask_irq,
+ .irq_ack = mainstone_mask_irq,
+ .irq_mask = mainstone_mask_irq,
+ .irq_unmask = mainstone_unmask_irq,
};
static void mainstone_irq_handler(unsigned int irq, struct irq_desc *desc)
{
unsigned long pending = MST_INTSETCLR & mainstone_irq_enabled;
do {
- desc->chip->ack(irq); /* clear useless edge notification */
+ /* clear useless edge notification */
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
if (likely(pending)) {
irq = MAINSTONE_IRQ(0) + __ffs(pending);
generic_handle_irq(irq);
diff --git a/arch/arm/mach-pxa/pcm990-baseboard.c b/arch/arm/mach-pxa/pcm990-baseboard.c
index f33647a..90820fa 100644
--- a/arch/arm/mach-pxa/pcm990-baseboard.c
+++ b/arch/arm/mach-pxa/pcm990-baseboard.c
@@ -241,23 +241,23 @@
static unsigned long pcm990_irq_enabled;
-static void pcm990_mask_ack_irq(unsigned int irq)
+static void pcm990_mask_ack_irq(struct irq_data *d)
{
- int pcm990_irq = (irq - PCM027_IRQ(0));
+ int pcm990_irq = (d->irq - PCM027_IRQ(0));
PCM990_INTMSKENA = (pcm990_irq_enabled &= ~(1 << pcm990_irq));
}
-static void pcm990_unmask_irq(unsigned int irq)
+static void pcm990_unmask_irq(struct irq_data *d)
{
- int pcm990_irq = (irq - PCM027_IRQ(0));
+ int pcm990_irq = (d->irq - PCM027_IRQ(0));
/* the irq can be acknowledged only if deasserted, so it's done here */
PCM990_INTSETCLR |= 1 << pcm990_irq;
PCM990_INTMSKENA = (pcm990_irq_enabled |= (1 << pcm990_irq));
}
static struct irq_chip pcm990_irq_chip = {
- .mask_ack = pcm990_mask_ack_irq,
- .unmask = pcm990_unmask_irq,
+ .irq_mask_ack = pcm990_mask_ack_irq,
+ .irq_unmask = pcm990_unmask_irq,
};
static void pcm990_irq_handler(unsigned int irq, struct irq_desc *desc)
@@ -265,7 +265,8 @@
unsigned long pending = (~PCM990_INTSETCLR) & pcm990_irq_enabled;
do {
- desc->chip->ack(irq); /* clear our parent IRQ */
+ /* clear our parent IRQ */
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
if (likely(pending)) {
irq = PCM027_IRQ(0) + __ffs(pending);
generic_handle_irq(irq);
diff --git a/arch/arm/mach-pxa/pxa25x.c b/arch/arm/mach-pxa/pxa25x.c
index 3f5241c..fbc5b77 100644
--- a/arch/arm/mach-pxa/pxa25x.c
+++ b/arch/arm/mach-pxa/pxa25x.c
@@ -22,6 +22,7 @@
#include <linux/platform_device.h>
#include <linux/suspend.h>
#include <linux/sysdev.h>
+#include <linux/irq.h>
#include <asm/mach/map.h>
#include <mach/hardware.h>
@@ -282,15 +283,15 @@
/* PXA25x: supports wakeup from GPIO0..GPIO15 and RTC alarm
*/
-static int pxa25x_set_wake(unsigned int irq, unsigned int on)
+static int pxa25x_set_wake(struct irq_data *d, unsigned int on)
{
- int gpio = IRQ_TO_GPIO(irq);
+ int gpio = IRQ_TO_GPIO(d->irq);
uint32_t mask = 0;
if (gpio >= 0 && gpio < 85)
return gpio_set_wake(gpio, on);
- if (irq == IRQ_RTCAlrm) {
+ if (d->irq == IRQ_RTCAlrm) {
mask = PWER_RTC;
goto set_pwer;
}
diff --git a/arch/arm/mach-pxa/pxa27x.c b/arch/arm/mach-pxa/pxa27x.c
index b2130b7..987301f 100644
--- a/arch/arm/mach-pxa/pxa27x.c
+++ b/arch/arm/mach-pxa/pxa27x.c
@@ -18,6 +18,7 @@
#include <linux/platform_device.h>
#include <linux/sysdev.h>
#include <linux/io.h>
+#include <linux/irq.h>
#include <asm/mach/map.h>
#include <mach/hardware.h>
@@ -343,18 +344,18 @@
/* PXA27x: Various gpios can issue wakeup events. This logic only
* handles the simple cases, not the WEMUX2 and WEMUX3 options
*/
-static int pxa27x_set_wake(unsigned int irq, unsigned int on)
+static int pxa27x_set_wake(struct irq_data *d, unsigned int on)
{
- int gpio = IRQ_TO_GPIO(irq);
+ int gpio = IRQ_TO_GPIO(d->irq);
uint32_t mask;
if (gpio >= 0 && gpio < 128)
return gpio_set_wake(gpio, on);
- if (irq == IRQ_KEYPAD)
+ if (d->irq == IRQ_KEYPAD)
return keypad_set_wake(on);
- switch (irq) {
+ switch (d->irq) {
case IRQ_RTCAlrm:
mask = PWER_RTC;
break;
diff --git a/arch/arm/mach-pxa/pxa3xx.c b/arch/arm/mach-pxa/pxa3xx.c
index e14818f..a7a19e1 100644
--- a/arch/arm/mach-pxa/pxa3xx.c
+++ b/arch/arm/mach-pxa/pxa3xx.c
@@ -229,11 +229,11 @@
pxa_cpu_pm_fns = &pxa3xx_cpu_pm_fns;
}
-static int pxa3xx_set_wake(unsigned int irq, unsigned int on)
+static int pxa3xx_set_wake(struct irq_data *d, unsigned int on)
{
unsigned long flags, mask = 0;
- switch (irq) {
+ switch (d->irq) {
case IRQ_SSP3:
mask = ADXER_MFP_WSSP3;
break;
@@ -322,40 +322,40 @@
#define pxa3xx_set_wake NULL
#endif
-static void pxa_ack_ext_wakeup(unsigned int irq)
+static void pxa_ack_ext_wakeup(struct irq_data *d)
{
- PECR |= PECR_IS(irq - IRQ_WAKEUP0);
+ PECR |= PECR_IS(d->irq - IRQ_WAKEUP0);
}
-static void pxa_mask_ext_wakeup(unsigned int irq)
+static void pxa_mask_ext_wakeup(struct irq_data *d)
{
- ICMR2 &= ~(1 << ((irq - PXA_IRQ(0)) & 0x1f));
- PECR &= ~PECR_IE(irq - IRQ_WAKEUP0);
+ ICMR2 &= ~(1 << ((d->irq - PXA_IRQ(0)) & 0x1f));
+ PECR &= ~PECR_IE(d->irq - IRQ_WAKEUP0);
}
-static void pxa_unmask_ext_wakeup(unsigned int irq)
+static void pxa_unmask_ext_wakeup(struct irq_data *d)
{
- ICMR2 |= 1 << ((irq - PXA_IRQ(0)) & 0x1f);
- PECR |= PECR_IE(irq - IRQ_WAKEUP0);
+ ICMR2 |= 1 << ((d->irq - PXA_IRQ(0)) & 0x1f);
+ PECR |= PECR_IE(d->irq - IRQ_WAKEUP0);
}
-static int pxa_set_ext_wakeup_type(unsigned int irq, unsigned int flow_type)
+static int pxa_set_ext_wakeup_type(struct irq_data *d, unsigned int flow_type)
{
if (flow_type & IRQ_TYPE_EDGE_RISING)
- PWER |= 1 << (irq - IRQ_WAKEUP0);
+ PWER |= 1 << (d->irq - IRQ_WAKEUP0);
if (flow_type & IRQ_TYPE_EDGE_FALLING)
- PWER |= 1 << (irq - IRQ_WAKEUP0 + 2);
+ PWER |= 1 << (d->irq - IRQ_WAKEUP0 + 2);
return 0;
}
static struct irq_chip pxa_ext_wakeup_chip = {
.name = "WAKEUP",
- .ack = pxa_ack_ext_wakeup,
- .mask = pxa_mask_ext_wakeup,
- .unmask = pxa_unmask_ext_wakeup,
- .set_type = pxa_set_ext_wakeup_type,
+ .irq_ack = pxa_ack_ext_wakeup,
+ .irq_mask = pxa_mask_ext_wakeup,
+ .irq_unmask = pxa_unmask_ext_wakeup,
+ .irq_set_type = pxa_set_ext_wakeup_type,
};
static void __init pxa_init_ext_wakeup_irq(set_wake_t fn)
@@ -368,7 +368,7 @@
set_irq_flags(irq, IRQF_VALID);
}
- pxa_ext_wakeup_chip.set_wake = fn;
+ pxa_ext_wakeup_chip.irq_set_wake = fn;
}
void __init pxa3xx_init_irq(void)
diff --git a/arch/arm/mach-pxa/spitz.c b/arch/arm/mach-pxa/spitz.c
index 0bc9387..b49a2c2 100644
--- a/arch/arm/mach-pxa/spitz.c
+++ b/arch/arm/mach-pxa/spitz.c
@@ -25,6 +25,7 @@
#include <linux/spi/corgi_lcd.h>
#include <linux/spi/pxa2xx_spi.h>
#include <linux/mtd/sharpsl.h>
+#include <linux/mtd/physmap.h>
#include <linux/input/matrix_keypad.h>
#include <linux/regulator/machine.h>
#include <linux/io.h>
diff --git a/arch/arm/mach-pxa/viper.c b/arch/arm/mach-pxa/viper.c
index de69b20..49eeeab 100644
--- a/arch/arm/mach-pxa/viper.c
+++ b/arch/arm/mach-pxa/viper.c
@@ -249,9 +249,9 @@
return viper_isa_irqs[bit] + PXA_ISA_IRQ(0);
}
-static void viper_ack_irq(unsigned int irq)
+static void viper_ack_irq(struct irq_data *d)
{
- int viper_irq = viper_irq_to_bitmask(irq);
+ int viper_irq = viper_irq_to_bitmask(d->irq);
if (viper_irq & 0xff)
VIPER_LO_IRQ_STATUS = viper_irq;
@@ -259,14 +259,14 @@
VIPER_HI_IRQ_STATUS = (viper_irq >> 8);
}
-static void viper_mask_irq(unsigned int irq)
+static void viper_mask_irq(struct irq_data *d)
{
- viper_irq_enabled_mask &= ~(viper_irq_to_bitmask(irq));
+ viper_irq_enabled_mask &= ~(viper_irq_to_bitmask(d->irq));
}
-static void viper_unmask_irq(unsigned int irq)
+static void viper_unmask_irq(struct irq_data *d)
{
- viper_irq_enabled_mask |= viper_irq_to_bitmask(irq);
+ viper_irq_enabled_mask |= viper_irq_to_bitmask(d->irq);
}
static inline unsigned long viper_irq_pending(void)
@@ -283,7 +283,7 @@
do {
/* we're in a chained irq handler,
* so ack the interrupt by hand */
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
if (likely(pending)) {
irq = viper_bit_to_irq(__ffs(pending));
@@ -294,10 +294,10 @@
}
static struct irq_chip viper_irq_chip = {
- .name = "ISA",
- .ack = viper_ack_irq,
- .mask = viper_mask_irq,
- .unmask = viper_unmask_irq
+ .name = "ISA",
+ .irq_ack = viper_ack_irq,
+ .irq_mask = viper_mask_irq,
+ .irq_unmask = viper_unmask_irq
};
static void __init viper_init_irq(void)
diff --git a/arch/arm/mach-pxa/zeus.c b/arch/arm/mach-pxa/zeus.c
index bf034c7..f4b053b 100644
--- a/arch/arm/mach-pxa/zeus.c
+++ b/arch/arm/mach-pxa/zeus.c
@@ -83,19 +83,19 @@
return zeus_isa_irqs[bit] + PXA_ISA_IRQ(0);
}
-static void zeus_ack_irq(unsigned int irq)
+static void zeus_ack_irq(struct irq_data *d)
{
- __raw_writew(zeus_irq_to_bitmask(irq), ZEUS_CPLD_ISA_IRQ);
+ __raw_writew(zeus_irq_to_bitmask(d->irq), ZEUS_CPLD_ISA_IRQ);
}
-static void zeus_mask_irq(unsigned int irq)
+static void zeus_mask_irq(struct irq_data *d)
{
- zeus_irq_enabled_mask &= ~(zeus_irq_to_bitmask(irq));
+ zeus_irq_enabled_mask &= ~(zeus_irq_to_bitmask(d->irq));
}
-static void zeus_unmask_irq(unsigned int irq)
+static void zeus_unmask_irq(struct irq_data *d)
{
- zeus_irq_enabled_mask |= zeus_irq_to_bitmask(irq);
+ zeus_irq_enabled_mask |= zeus_irq_to_bitmask(d->irq);
}
static inline unsigned long zeus_irq_pending(void)
@@ -111,7 +111,7 @@
do {
/* we're in a chained irq handler,
* so ack the interrupt by hand */
- desc->chip->ack(gpio_to_irq(ZEUS_ISA_GPIO));
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
if (likely(pending)) {
irq = zeus_bit_to_irq(__ffs(pending));
@@ -122,10 +122,10 @@
}
static struct irq_chip zeus_irq_chip = {
- .name = "ISA",
- .ack = zeus_ack_irq,
- .mask = zeus_mask_irq,
- .unmask = zeus_unmask_irq,
+ .name = "ISA",
+ .irq_ack = zeus_ack_irq,
+ .irq_mask = zeus_mask_irq,
+ .irq_unmask = zeus_unmask_irq,
};
static void __init zeus_init_irq(void)
@@ -830,8 +830,8 @@
pr_info("Zeus CPLD V%dI%d\n", (system_rev & 0xf0) >> 4, (system_rev & 0x0f));
/* Fix timings for dm9000s (CS1/CS2)*/
- msc0 = __raw_readl(MSC0) & 0x0000ffff | (dm9000_msc << 16);
- msc1 = __raw_readl(MSC1) & 0xffff0000 | dm9000_msc;
+ msc0 = (__raw_readl(MSC0) & 0x0000ffff) | (dm9000_msc << 16);
+ msc1 = (__raw_readl(MSC1) & 0xffff0000) | dm9000_msc;
__raw_writel(msc0, MSC0);
__raw_writel(msc1, MSC1);
diff --git a/arch/arm/mach-rpc/irq.c b/arch/arm/mach-rpc/irq.c
index 9dd15d6..d29cd9b 100644
--- a/arch/arm/mach-rpc/irq.c
+++ b/arch/arm/mach-rpc/irq.c
@@ -6,110 +6,110 @@
#include <asm/hardware/iomd.h>
#include <asm/irq.h>
-static void iomd_ack_irq_a(unsigned int irq)
+static void iomd_ack_irq_a(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << irq;
+ mask = 1 << d->irq;
val = iomd_readb(IOMD_IRQMASKA);
iomd_writeb(val & ~mask, IOMD_IRQMASKA);
iomd_writeb(mask, IOMD_IRQCLRA);
}
-static void iomd_mask_irq_a(unsigned int irq)
+static void iomd_mask_irq_a(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << irq;
+ mask = 1 << d->irq;
val = iomd_readb(IOMD_IRQMASKA);
iomd_writeb(val & ~mask, IOMD_IRQMASKA);
}
-static void iomd_unmask_irq_a(unsigned int irq)
+static void iomd_unmask_irq_a(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << irq;
+ mask = 1 << d->irq;
val = iomd_readb(IOMD_IRQMASKA);
iomd_writeb(val | mask, IOMD_IRQMASKA);
}
static struct irq_chip iomd_a_chip = {
- .ack = iomd_ack_irq_a,
- .mask = iomd_mask_irq_a,
- .unmask = iomd_unmask_irq_a,
+ .irq_ack = iomd_ack_irq_a,
+ .irq_mask = iomd_mask_irq_a,
+ .irq_unmask = iomd_unmask_irq_a,
};
-static void iomd_mask_irq_b(unsigned int irq)
+static void iomd_mask_irq_b(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << (irq & 7);
+ mask = 1 << (d->irq & 7);
val = iomd_readb(IOMD_IRQMASKB);
iomd_writeb(val & ~mask, IOMD_IRQMASKB);
}
-static void iomd_unmask_irq_b(unsigned int irq)
+static void iomd_unmask_irq_b(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << (irq & 7);
+ mask = 1 << (d->irq & 7);
val = iomd_readb(IOMD_IRQMASKB);
iomd_writeb(val | mask, IOMD_IRQMASKB);
}
static struct irq_chip iomd_b_chip = {
- .ack = iomd_mask_irq_b,
- .mask = iomd_mask_irq_b,
- .unmask = iomd_unmask_irq_b,
+ .irq_ack = iomd_mask_irq_b,
+ .irq_mask = iomd_mask_irq_b,
+ .irq_unmask = iomd_unmask_irq_b,
};
-static void iomd_mask_irq_dma(unsigned int irq)
+static void iomd_mask_irq_dma(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << (irq & 7);
+ mask = 1 << (d->irq & 7);
val = iomd_readb(IOMD_DMAMASK);
iomd_writeb(val & ~mask, IOMD_DMAMASK);
}
-static void iomd_unmask_irq_dma(unsigned int irq)
+static void iomd_unmask_irq_dma(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << (irq & 7);
+ mask = 1 << (d->irq & 7);
val = iomd_readb(IOMD_DMAMASK);
iomd_writeb(val | mask, IOMD_DMAMASK);
}
static struct irq_chip iomd_dma_chip = {
- .ack = iomd_mask_irq_dma,
- .mask = iomd_mask_irq_dma,
- .unmask = iomd_unmask_irq_dma,
+ .irq_ack = iomd_mask_irq_dma,
+ .irq_mask = iomd_mask_irq_dma,
+ .irq_unmask = iomd_unmask_irq_dma,
};
-static void iomd_mask_irq_fiq(unsigned int irq)
+static void iomd_mask_irq_fiq(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << (irq & 7);
+ mask = 1 << (d->irq & 7);
val = iomd_readb(IOMD_FIQMASK);
iomd_writeb(val & ~mask, IOMD_FIQMASK);
}
-static void iomd_unmask_irq_fiq(unsigned int irq)
+static void iomd_unmask_irq_fiq(struct irq_data *d)
{
unsigned int val, mask;
- mask = 1 << (irq & 7);
+ mask = 1 << (d->irq & 7);
val = iomd_readb(IOMD_FIQMASK);
iomd_writeb(val | mask, IOMD_FIQMASK);
}
static struct irq_chip iomd_fiq_chip = {
- .ack = iomd_mask_irq_fiq,
- .mask = iomd_mask_irq_fiq,
- .unmask = iomd_unmask_irq_fiq,
+ .irq_ack = iomd_mask_irq_fiq,
+ .irq_mask = iomd_mask_irq_fiq,
+ .irq_unmask = iomd_unmask_irq_fiq,
};
void __init rpc_init_irq(void)
diff --git a/arch/arm/mach-s3c2410/bast-irq.c b/arch/arm/mach-s3c2410/bast-irq.c
index 217b102..606cb6b 100644
--- a/arch/arm/mach-s3c2410/bast-irq.c
+++ b/arch/arm/mach-s3c2410/bast-irq.c
@@ -75,38 +75,38 @@
static unsigned char bast_pc104_irqs[] = { 3, 5, 7, 10 };
static void
-bast_pc104_mask(unsigned int irqno)
+bast_pc104_mask(struct irq_data *data)
{
unsigned long temp;
temp = __raw_readb(BAST_VA_PC104_IRQMASK);
- temp &= ~bast_pc104_irqmasks[irqno];
+ temp &= ~bast_pc104_irqmasks[data->irq];
__raw_writeb(temp, BAST_VA_PC104_IRQMASK);
}
static void
-bast_pc104_maskack(unsigned int irqno)
+bast_pc104_maskack(struct irq_data *data)
{
struct irq_desc *desc = irq_desc + IRQ_ISA;
- bast_pc104_mask(irqno);
- desc->chip->ack(IRQ_ISA);
+ bast_pc104_mask(data);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
}
static void
-bast_pc104_unmask(unsigned int irqno)
+bast_pc104_unmask(struct irq_data *data)
{
unsigned long temp;
temp = __raw_readb(BAST_VA_PC104_IRQMASK);
- temp |= bast_pc104_irqmasks[irqno];
+ temp |= bast_pc104_irqmasks[data->irq];
__raw_writeb(temp, BAST_VA_PC104_IRQMASK);
}
static struct irq_chip bast_pc104_chip = {
- .mask = bast_pc104_mask,
- .unmask = bast_pc104_unmask,
- .ack = bast_pc104_maskack
+ .irq_mask = bast_pc104_mask,
+ .irq_unmask = bast_pc104_unmask,
+ .irq_ack = bast_pc104_maskack
};
static void
@@ -123,7 +123,7 @@
/* ack if we get an irq with nothing (ie, startup) */
desc = irq_desc + IRQ_ISA;
- desc->chip->ack(IRQ_ISA);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
} else {
/* handle the IRQ */
diff --git a/arch/arm/mach-s3c2410/include/mach/irqs.h b/arch/arm/mach-s3c2410/include/mach/irqs.h
index 11bb0f0..e5a68ea 100644
--- a/arch/arm/mach-s3c2410/include/mach/irqs.h
+++ b/arch/arm/mach-s3c2410/include/mach/irqs.h
@@ -152,8 +152,8 @@
#define IRQ_S3C2416_HSMMC0 S3C2410_IRQ(21) /* S3C2416/S3C2450 */
-#define IRQ_HSMMC0 IRQ_S3C2443_HSMMC
-#define IRQ_HSMMC1 IRQ_S3C2416_HSMMC0
+#define IRQ_HSMMC0 IRQ_S3C2416_HSMMC0
+#define IRQ_HSMMC1 IRQ_S3C2443_HSMMC
#define IRQ_S3C2443_LCD1 S3C2410_IRQSUB(14)
#define IRQ_S3C2443_LCD2 S3C2410_IRQSUB(15)
diff --git a/arch/arm/mach-s3c2410/include/mach/map.h b/arch/arm/mach-s3c2410/include/mach/map.h
index cd3983a..25bbf5a 100644
--- a/arch/arm/mach-s3c2410/include/mach/map.h
+++ b/arch/arm/mach-s3c2410/include/mach/map.h
@@ -112,8 +112,8 @@
#define S3C_PA_IIC S3C2410_PA_IIC
#define S3C_PA_UART S3C24XX_PA_UART
#define S3C_PA_USBHOST S3C2410_PA_USBHOST
-#define S3C_PA_HSMMC0 S3C2443_PA_HSMMC
-#define S3C_PA_HSMMC1 S3C2416_PA_HSMMC0
+#define S3C_PA_HSMMC0 S3C2416_PA_HSMMC0
+#define S3C_PA_HSMMC1 S3C2443_PA_HSMMC
#define S3C_PA_WDT S3C2410_PA_WATCHDOG
#define S3C_PA_NAND S3C24XX_PA_NAND
diff --git a/arch/arm/mach-s3c2410/include/mach/regs-s3c2443-clock.h b/arch/arm/mach-s3c2410/include/mach/regs-s3c2443-clock.h
index 101aeea..44494a5 100644
--- a/arch/arm/mach-s3c2410/include/mach/regs-s3c2443-clock.h
+++ b/arch/arm/mach-s3c2410/include/mach/regs-s3c2443-clock.h
@@ -86,6 +86,7 @@
#define S3C2443_HCLKCON_LCDC (1<<9)
#define S3C2443_HCLKCON_USBH (1<<11)
#define S3C2443_HCLKCON_USBD (1<<12)
+#define S3C2416_HCLKCON_HSMMC0 (1<<15)
#define S3C2443_HCLKCON_HSMMC (1<<16)
#define S3C2443_HCLKCON_CFC (1<<17)
#define S3C2443_HCLKCON_SSMC (1<<18)
diff --git a/arch/arm/mach-s3c2412/irq.c b/arch/arm/mach-s3c2412/irq.c
index 6000ca9..eddb52b 100644
--- a/arch/arm/mach-s3c2412/irq.c
+++ b/arch/arm/mach-s3c2412/irq.c
@@ -49,9 +49,9 @@
*/
static void
-s3c2412_irq_mask(unsigned int irqno)
+s3c2412_irq_mask(struct irq_data *data)
{
- unsigned long bitval = 1UL << (irqno - IRQ_EINT0);
+ unsigned long bitval = 1UL << (data->irq - IRQ_EINT0);
unsigned long mask;
mask = __raw_readl(S3C2410_INTMSK);
@@ -62,9 +62,9 @@
}
static inline void
-s3c2412_irq_ack(unsigned int irqno)
+s3c2412_irq_ack(struct irq_data *data)
{
- unsigned long bitval = 1UL << (irqno - IRQ_EINT0);
+ unsigned long bitval = 1UL << (data->irq - IRQ_EINT0);
__raw_writel(bitval, S3C2412_EINTPEND);
__raw_writel(bitval, S3C2410_SRCPND);
@@ -72,9 +72,9 @@
}
static inline void
-s3c2412_irq_maskack(unsigned int irqno)
+s3c2412_irq_maskack(struct irq_data *data)
{
- unsigned long bitval = 1UL << (irqno - IRQ_EINT0);
+ unsigned long bitval = 1UL << (data->irq - IRQ_EINT0);
unsigned long mask;
mask = __raw_readl(S3C2410_INTMSK);
@@ -89,9 +89,9 @@
}
static void
-s3c2412_irq_unmask(unsigned int irqno)
+s3c2412_irq_unmask(struct irq_data *data)
{
- unsigned long bitval = 1UL << (irqno - IRQ_EINT0);
+ unsigned long bitval = 1UL << (data->irq - IRQ_EINT0);
unsigned long mask;
mask = __raw_readl(S3C2412_EINTMASK);
@@ -102,11 +102,11 @@
}
static struct irq_chip s3c2412_irq_eint0t4 = {
- .ack = s3c2412_irq_ack,
- .mask = s3c2412_irq_mask,
- .unmask = s3c2412_irq_unmask,
- .set_wake = s3c_irq_wake,
- .set_type = s3c_irqext_type,
+ .irq_ack = s3c2412_irq_ack,
+ .irq_mask = s3c2412_irq_mask,
+ .irq_unmask = s3c2412_irq_unmask,
+ .irq_set_wake = s3c_irq_wake,
+ .irq_set_type = s3c_irqext_type,
};
#define INTBIT(x) (1 << ((x) - S3C2410_IRQSUB(0)))
@@ -132,29 +132,29 @@
#define INTMSK_CFSDI (1UL << (IRQ_S3C2412_CFSDI - IRQ_EINT0))
#define SUBMSK_CFSDI INTMSK_SUB(IRQ_S3C2412_SDI, IRQ_S3C2412_CF)
-static void s3c2412_irq_cfsdi_mask(unsigned int irqno)
+static void s3c2412_irq_cfsdi_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_CFSDI, SUBMSK_CFSDI);
+ s3c_irqsub_mask(data->irq, INTMSK_CFSDI, SUBMSK_CFSDI);
}
-static void s3c2412_irq_cfsdi_unmask(unsigned int irqno)
+static void s3c2412_irq_cfsdi_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_CFSDI);
+ s3c_irqsub_unmask(data->irq, INTMSK_CFSDI);
}
-static void s3c2412_irq_cfsdi_ack(unsigned int irqno)
+static void s3c2412_irq_cfsdi_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_CFSDI, SUBMSK_CFSDI);
+ s3c_irqsub_maskack(data->irq, INTMSK_CFSDI, SUBMSK_CFSDI);
}
static struct irq_chip s3c2412_irq_cfsdi = {
.name = "s3c2412-cfsdi",
- .ack = s3c2412_irq_cfsdi_ack,
- .mask = s3c2412_irq_cfsdi_mask,
- .unmask = s3c2412_irq_cfsdi_unmask,
+ .irq_ack = s3c2412_irq_cfsdi_ack,
+ .irq_mask = s3c2412_irq_cfsdi_mask,
+ .irq_unmask = s3c2412_irq_cfsdi_unmask,
};
-static int s3c2412_irq_rtc_wake(unsigned int irqno, unsigned int state)
+static int s3c2412_irq_rtc_wake(struct irq_data *data, unsigned int state)
{
unsigned long pwrcfg;
@@ -165,7 +165,7 @@
pwrcfg |= S3C2412_PWRCFG_RTC_MASKIRQ;
__raw_writel(pwrcfg, S3C2412_PWRCFG);
- return s3c_irq_chip.set_wake(irqno, state);
+ return s3c_irq_chip.irq_set_wake(data, state);
}
static struct irq_chip s3c2412_irq_rtc_chip;
@@ -193,7 +193,7 @@
/* change RTC IRQ's set wake method */
s3c2412_irq_rtc_chip = s3c_irq_chip;
- s3c2412_irq_rtc_chip.set_wake = s3c2412_irq_rtc_wake;
+ s3c2412_irq_rtc_chip.irq_set_wake = s3c2412_irq_rtc_wake;
set_irq_chip(IRQ_RTC, &s3c2412_irq_rtc_chip);
diff --git a/arch/arm/mach-s3c2416/Kconfig b/arch/arm/mach-s3c2416/Kconfig
index df8d149..69b48a7 100644
--- a/arch/arm/mach-s3c2416/Kconfig
+++ b/arch/arm/mach-s3c2416/Kconfig
@@ -31,6 +31,17 @@
help
Internal config node to apply S3C2416 power management
+config S3C2416_SETUP_SDHCI
+ bool
+ select S3C2416_SETUP_SDHCI_GPIO
+ help
+ Internal helper functions for S3C2416 based SDHCI systems
+
+config S3C2416_SETUP_SDHCI_GPIO
+ bool
+ help
+ Common setup code for SDHCI gpio.
+
menu "S3C2416 Machines"
config MACH_SMDK2416
@@ -42,6 +53,7 @@
select S3C_DEV_HSMMC1
select S3C_DEV_NAND
select S3C_DEV_USB_HOST
+ select S3C2416_SETUP_SDHCI
select S3C2416_PM if PM
help
Say Y here if you are using an SMDK2416
diff --git a/arch/arm/mach-s3c2416/Makefile b/arch/arm/mach-s3c2416/Makefile
index ef038d6..7b805b2 100644
--- a/arch/arm/mach-s3c2416/Makefile
+++ b/arch/arm/mach-s3c2416/Makefile
@@ -14,6 +14,10 @@
obj-$(CONFIG_S3C2416_PM) += pm.o
#obj-$(CONFIG_S3C2416_DMA) += dma.o
+# Device setup
+obj-$(CONFIG_S3C2416_SETUP_SDHCI) += setup-sdhci.o
+obj-$(CONFIG_S3C2416_SETUP_SDHCI_GPIO) += setup-sdhci-gpio.o
+
# Machine support
obj-$(CONFIG_MACH_SMDK2416) += mach-smdk2416.o
diff --git a/arch/arm/mach-s3c2416/clock.c b/arch/arm/mach-s3c2416/clock.c
index 7ccf5a2..3b02d85 100644
--- a/arch/arm/mach-s3c2416/clock.c
+++ b/arch/arm/mach-s3c2416/clock.c
@@ -38,12 +38,11 @@
[7] = 8,
};
-/* ID to hardware numbering, 0 is HSMMC1, 1 is HSMMC0 */
static struct clksrc_clk hsmmc_div[] = {
[0] = {
.clk = {
.name = "hsmmc-div",
- .id = 1,
+ .id = 0,
.parent = &clk_esysclk.clk,
},
.reg_div = { .reg = S3C2416_CLKDIV2, .size = 2, .shift = 6 },
@@ -51,7 +50,7 @@
[1] = {
.clk = {
.name = "hsmmc-div",
- .id = 0,
+ .id = 1,
.parent = &clk_esysclk.clk,
},
.reg_div = { .reg = S3C2443_CLKDIV1, .size = 2, .shift = 6 },
@@ -61,7 +60,7 @@
static struct clksrc_clk hsmmc_mux[] = {
[0] = {
.clk = {
- .id = 1,
+ .id = 0,
.name = "hsmmc-if",
.ctrlbit = (1 << 6),
.enable = s3c2443_clkcon_enable_s,
@@ -77,7 +76,7 @@
},
[1] = {
.clk = {
- .id = 0,
+ .id = 1,
.name = "hsmmc-if",
.ctrlbit = (1 << 12),
.enable = s3c2443_clkcon_enable_s,
@@ -93,6 +92,13 @@
},
};
+static struct clk hsmmc0_clk = {
+ .name = "hsmmc",
+ .id = 0,
+ .parent = &clk_h,
+ .enable = s3c2443_clkcon_enable_h,
+ .ctrlbit = S3C2416_HCLKCON_HSMMC0,
+};
static inline unsigned int s3c2416_fclk_div(unsigned long clkcon0)
{
@@ -130,6 +136,8 @@
for (ptr = 0; ptr < ARRAY_SIZE(clksrcs); ptr++)
s3c_register_clksrc(clksrcs[ptr], 1);
+ s3c24xx_register_clock(&hsmmc0_clk);
+
s3c_pwmclk_init();
}
diff --git a/arch/arm/mach-s3c2416/irq.c b/arch/arm/mach-s3c2416/irq.c
index 00174da..680fe38 100644
--- a/arch/arm/mach-s3c2416/irq.c
+++ b/arch/arm/mach-s3c2416/irq.c
@@ -77,28 +77,27 @@
#define INTMSK_WDTAC97 (1UL << (IRQ_WDT - IRQ_EINT0))
#define SUBMSK_WDTAC97 INTMSK(IRQ_S3C2443_WDT, IRQ_S3C2443_AC97)
-static void s3c2416_irq_wdtac97_mask(unsigned int irqno)
+static void s3c2416_irq_wdtac97_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_WDTAC97, SUBMSK_WDTAC97);
+ s3c_irqsub_mask(data->irq, INTMSK_WDTAC97, SUBMSK_WDTAC97);
}
-static void s3c2416_irq_wdtac97_unmask(unsigned int irqno)
+static void s3c2416_irq_wdtac97_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_WDTAC97);
+ s3c_irqsub_unmask(data->irq, INTMSK_WDTAC97);
}
-static void s3c2416_irq_wdtac97_ack(unsigned int irqno)
+static void s3c2416_irq_wdtac97_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_WDTAC97, SUBMSK_WDTAC97);
+ s3c_irqsub_maskack(data->irq, INTMSK_WDTAC97, SUBMSK_WDTAC97);
}
static struct irq_chip s3c2416_irq_wdtac97 = {
- .mask = s3c2416_irq_wdtac97_mask,
- .unmask = s3c2416_irq_wdtac97_unmask,
- .ack = s3c2416_irq_wdtac97_ack,
+ .irq_mask = s3c2416_irq_wdtac97_mask,
+ .irq_unmask = s3c2416_irq_wdtac97_unmask,
+ .irq_ack = s3c2416_irq_wdtac97_ack,
};
-
/* LCD sub interrupts */
static void s3c2416_irq_demux_lcd(unsigned int irq, struct irq_desc *desc)
@@ -109,28 +108,27 @@
#define INTMSK_LCD (1UL << (IRQ_LCD - IRQ_EINT0))
#define SUBMSK_LCD INTMSK(IRQ_S3C2443_LCD1, IRQ_S3C2443_LCD4)
-static void s3c2416_irq_lcd_mask(unsigned int irqno)
+static void s3c2416_irq_lcd_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_LCD, SUBMSK_LCD);
+ s3c_irqsub_mask(data->irq, INTMSK_LCD, SUBMSK_LCD);
}
-static void s3c2416_irq_lcd_unmask(unsigned int irqno)
+static void s3c2416_irq_lcd_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_LCD);
+ s3c_irqsub_unmask(data->irq, INTMSK_LCD);
}
-static void s3c2416_irq_lcd_ack(unsigned int irqno)
+static void s3c2416_irq_lcd_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_LCD, SUBMSK_LCD);
+ s3c_irqsub_maskack(data->irq, INTMSK_LCD, SUBMSK_LCD);
}
static struct irq_chip s3c2416_irq_lcd = {
- .mask = s3c2416_irq_lcd_mask,
- .unmask = s3c2416_irq_lcd_unmask,
- .ack = s3c2416_irq_lcd_ack,
+ .irq_mask = s3c2416_irq_lcd_mask,
+ .irq_unmask = s3c2416_irq_lcd_unmask,
+ .irq_ack = s3c2416_irq_lcd_ack,
};
-
/* DMA sub interrupts */
static void s3c2416_irq_demux_dma(unsigned int irq, struct irq_desc *desc)
@@ -142,28 +140,27 @@
#define SUBMSK_DMA INTMSK(IRQ_S3C2443_DMA0, IRQ_S3C2443_DMA5)
-static void s3c2416_irq_dma_mask(unsigned int irqno)
+static void s3c2416_irq_dma_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_DMA, SUBMSK_DMA);
+ s3c_irqsub_mask(data->irq, INTMSK_DMA, SUBMSK_DMA);
}
-static void s3c2416_irq_dma_unmask(unsigned int irqno)
+static void s3c2416_irq_dma_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_DMA);
+ s3c_irqsub_unmask(data->irq, INTMSK_DMA);
}
-static void s3c2416_irq_dma_ack(unsigned int irqno)
+static void s3c2416_irq_dma_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_DMA, SUBMSK_DMA);
+ s3c_irqsub_maskack(data->irq, INTMSK_DMA, SUBMSK_DMA);
}
static struct irq_chip s3c2416_irq_dma = {
- .mask = s3c2416_irq_dma_mask,
- .unmask = s3c2416_irq_dma_unmask,
- .ack = s3c2416_irq_dma_ack,
+ .irq_mask = s3c2416_irq_dma_mask,
+ .irq_unmask = s3c2416_irq_dma_unmask,
+ .irq_ack = s3c2416_irq_dma_ack,
};
-
/* UART3 sub interrupts */
static void s3c2416_irq_demux_uart3(unsigned int irq, struct irq_desc *desc)
@@ -174,28 +171,27 @@
#define INTMSK_UART3 (1UL << (IRQ_S3C2443_UART3 - IRQ_EINT0))
#define SUBMSK_UART3 (0x7 << (IRQ_S3C2443_RX3 - S3C2410_IRQSUB(0)))
-static void s3c2416_irq_uart3_mask(unsigned int irqno)
+static void s3c2416_irq_uart3_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_UART3, SUBMSK_UART3);
+ s3c_irqsub_mask(data->irq, INTMSK_UART3, SUBMSK_UART3);
}
-static void s3c2416_irq_uart3_unmask(unsigned int irqno)
+static void s3c2416_irq_uart3_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_UART3);
+ s3c_irqsub_unmask(data->irq, INTMSK_UART3);
}
-static void s3c2416_irq_uart3_ack(unsigned int irqno)
+static void s3c2416_irq_uart3_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_UART3, SUBMSK_UART3);
+ s3c_irqsub_maskack(data->irq, INTMSK_UART3, SUBMSK_UART3);
}
static struct irq_chip s3c2416_irq_uart3 = {
- .mask = s3c2416_irq_uart3_mask,
- .unmask = s3c2416_irq_uart3_unmask,
- .ack = s3c2416_irq_uart3_ack,
+ .irq_mask = s3c2416_irq_uart3_mask,
+ .irq_unmask = s3c2416_irq_uart3_unmask,
+ .irq_ack = s3c2416_irq_uart3_ack,
};
-
/* IRQ initialisation code */
static int __init s3c2416_add_sub(unsigned int base,
diff --git a/arch/arm/mach-s3c2416/mach-smdk2416.c b/arch/arm/mach-s3c2416/mach-smdk2416.c
index 7fc3664..3f83177 100644
--- a/arch/arm/mach-s3c2416/mach-smdk2416.c
+++ b/arch/arm/mach-s3c2416/mach-smdk2416.c
@@ -46,6 +46,7 @@
#include <plat/devs.h>
#include <plat/cpu.h>
#include <plat/nand.h>
+#include <plat/sdhci.h>
#include <plat/regs-fb-v4.h>
#include <plat/fb.h>
@@ -110,6 +111,13 @@
.ucon = UCON,
.ulcon = ULCON | 0x50,
.ufcon = UFCON,
+ },
+ [3] = {
+ .hwport = 3,
+ .flags = 0,
+ .ucon = UCON,
+ .ulcon = ULCON,
+ .ufcon = UFCON,
}
};
@@ -159,6 +167,18 @@
.vidcon1 = VIDCON1_INV_HSYNC | VIDCON1_INV_VSYNC,
};
+static struct s3c_sdhci_platdata smdk2416_hsmmc0_pdata __initdata = {
+ .max_width = 4,
+ .cd_type = S3C_SDHCI_CD_GPIO,
+ .ext_cd_gpio = S3C2410_GPF(1),
+ .ext_cd_gpio_invert = 1,
+};
+
+static struct s3c_sdhci_platdata smdk2416_hsmmc1_pdata __initdata = {
+ .max_width = 4,
+ .cd_type = S3C_SDHCI_CD_NONE,
+};
+
static struct platform_device *smdk2416_devices[] __initdata = {
&s3c_device_fb,
&s3c_device_wdt,
@@ -180,6 +200,9 @@
s3c_i2c0_set_platdata(NULL);
s3c_fb_set_platdata(&smdk2416_fb_platdata);
+ s3c_sdhci0_set_platdata(&smdk2416_hsmmc0_pdata);
+ s3c_sdhci1_set_platdata(&smdk2416_hsmmc1_pdata);
+
gpio_request(S3C2410_GPB(4), "USBHost Power");
gpio_direction_output(S3C2410_GPB(4), 1);
diff --git a/arch/arm/mach-s3c2416/s3c2416.c b/arch/arm/mach-s3c2416/s3c2416.c
index 63f39cd..ba7fd87 100644
--- a/arch/arm/mach-s3c2416/s3c2416.c
+++ b/arch/arm/mach-s3c2416/s3c2416.c
@@ -53,6 +53,7 @@
#include <plat/s3c2416.h>
#include <plat/devs.h>
#include <plat/cpu.h>
+#include <plat/sdhci.h>
#include <plat/iic-core.h>
#include <plat/fb-core.h>
@@ -115,6 +116,10 @@
s3c24xx_gpiocfg_default.set_pull = s3c_gpio_setpull_updown;
s3c24xx_gpiocfg_default.get_pull = s3c_gpio_getpull_updown;
+ /* initialize device information early */
+ s3c2416_default_sdhci0();
+ s3c2416_default_sdhci1();
+
iotable_init(s3c2416_iodesc, ARRAY_SIZE(s3c2416_iodesc));
}
diff --git a/arch/arm/mach-s3c2416/setup-sdhci-gpio.c b/arch/arm/mach-s3c2416/setup-sdhci-gpio.c
new file mode 100644
index 0000000..f65cb3e
--- /dev/null
+++ b/arch/arm/mach-s3c2416/setup-sdhci-gpio.c
@@ -0,0 +1,34 @@
+/* linux/arch/arm/plat-s3c2416/setup-sdhci-gpio.c
+ *
+ * Copyright 2010 Promwad Innovation Company
+ * Yauhen Kharuzhy <yauhen.kharuzhy@promwad.com>
+ *
+ * S3C2416 - Helper functions for setting up SDHCI device(s) GPIO (HSMMC)
+ *
+ * Based on mach-s3c64xx/setup-sdhci-gpio.c
+ *
+ * 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/kernel.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/io.h>
+#include <linux/gpio.h>
+
+#include <mach/regs-gpio.h>
+#include <plat/gpio-cfg.h>
+
+void s3c2416_setup_sdhci0_cfg_gpio(struct platform_device *dev, int width)
+{
+ s3c_gpio_cfgrange_nopull(S3C2410_GPE(5), 2 + width, S3C_GPIO_SFN(2));
+}
+
+void s3c2416_setup_sdhci1_cfg_gpio(struct platform_device *dev, int width)
+{
+ s3c_gpio_cfgrange_nopull(S3C2410_GPL(0), width, S3C_GPIO_SFN(2));
+ s3c_gpio_cfgrange_nopull(S3C2410_GPL(8), 2, S3C_GPIO_SFN(2));
+}
diff --git a/arch/arm/mach-s3c2416/setup-sdhci.c b/arch/arm/mach-s3c2416/setup-sdhci.c
new file mode 100644
index 0000000..ed34fad
--- /dev/null
+++ b/arch/arm/mach-s3c2416/setup-sdhci.c
@@ -0,0 +1,61 @@
+/* linux/arch/arm/mach-s3c2416/setup-sdhci.c
+ *
+ * Copyright 2010 Promwad Innovation Company
+ * Yauhen Kharuzhy <yauhen.kharuzhy@promwad.com>
+ *
+ * S3C2416 - Helper functions for settign up SDHCI device(s) (HSMMC)
+ *
+ * Based on mach-s3c64xx/setup-sdhci.c
+ *
+ * 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/kernel.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/io.h>
+
+#include <linux/mmc/card.h>
+#include <linux/mmc/host.h>
+
+#include <plat/regs-sdhci.h>
+#include <plat/sdhci.h>
+
+/* clock sources for the mmc bus clock, order as for the ctrl2[5..4] */
+
+char *s3c2416_hsmmc_clksrcs[4] = {
+ [0] = "hsmmc",
+ [1] = "hsmmc",
+ [2] = "hsmmc-if",
+ /* [3] = "48m", - note not successfully used yet */
+};
+
+void s3c2416_setup_sdhci_cfg_card(struct platform_device *dev,
+ void __iomem *r,
+ struct mmc_ios *ios,
+ struct mmc_card *card)
+{
+ u32 ctrl2, ctrl3;
+
+ ctrl2 = __raw_readl(r + S3C_SDHCI_CONTROL2);
+ ctrl2 &= S3C_SDHCI_CTRL2_SELBASECLK_MASK;
+ ctrl2 |= (S3C64XX_SDHCI_CTRL2_ENSTAASYNCCLR |
+ S3C64XX_SDHCI_CTRL2_ENCMDCNFMSK |
+ S3C_SDHCI_CTRL2_ENFBCLKRX |
+ S3C_SDHCI_CTRL2_DFCNT_NONE |
+ S3C_SDHCI_CTRL2_ENCLKOUTHOLD);
+
+ if (ios->clock < 25 * 1000000)
+ ctrl3 = (S3C_SDHCI_CTRL3_FCSEL3 |
+ S3C_SDHCI_CTRL3_FCSEL2 |
+ S3C_SDHCI_CTRL3_FCSEL1 |
+ S3C_SDHCI_CTRL3_FCSEL0);
+ else
+ ctrl3 = (S3C_SDHCI_CTRL3_FCSEL1 | S3C_SDHCI_CTRL3_FCSEL0);
+
+ __raw_writel(ctrl2, r + S3C_SDHCI_CONTROL2);
+ __raw_writel(ctrl3, r + S3C_SDHCI_CONTROL3);
+}
diff --git a/arch/arm/mach-s3c2440/irq.c b/arch/arm/mach-s3c2440/irq.c
index 0c049b9..acad442 100644
--- a/arch/arm/mach-s3c2440/irq.c
+++ b/arch/arm/mach-s3c2440/irq.c
@@ -69,27 +69,27 @@
#define INTMSK_WDT (1UL << (IRQ_WDT - IRQ_EINT0))
static void
-s3c_irq_wdtac97_mask(unsigned int irqno)
+s3c_irq_wdtac97_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_WDT, 3<<13);
+ s3c_irqsub_mask(data->irq, INTMSK_WDT, 3 << 13);
}
static void
-s3c_irq_wdtac97_unmask(unsigned int irqno)
+s3c_irq_wdtac97_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_WDT);
+ s3c_irqsub_unmask(data->irq, INTMSK_WDT);
}
static void
-s3c_irq_wdtac97_ack(unsigned int irqno)
+s3c_irq_wdtac97_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_WDT, 3<<13);
+ s3c_irqsub_maskack(data->irq, INTMSK_WDT, 3 << 13);
}
static struct irq_chip s3c_irq_wdtac97 = {
- .mask = s3c_irq_wdtac97_mask,
- .unmask = s3c_irq_wdtac97_unmask,
- .ack = s3c_irq_wdtac97_ack,
+ .irq_mask = s3c_irq_wdtac97_mask,
+ .irq_unmask = s3c_irq_wdtac97_unmask,
+ .irq_ack = s3c_irq_wdtac97_ack,
};
static int s3c2440_irq_add(struct sys_device *sysdev)
diff --git a/arch/arm/mach-s3c2440/s3c244x-irq.c b/arch/arm/mach-s3c2440/s3c244x-irq.c
index a75c0c2..83daf4e 100644
--- a/arch/arm/mach-s3c2440/s3c244x-irq.c
+++ b/arch/arm/mach-s3c2440/s3c244x-irq.c
@@ -68,27 +68,27 @@
#define INTMSK_CAM (1UL << (IRQ_CAM - IRQ_EINT0))
static void
-s3c_irq_cam_mask(unsigned int irqno)
+s3c_irq_cam_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_CAM, 3<<11);
+ s3c_irqsub_mask(data->irq, INTMSK_CAM, 3 << 11);
}
static void
-s3c_irq_cam_unmask(unsigned int irqno)
+s3c_irq_cam_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_CAM);
+ s3c_irqsub_unmask(data->irq, INTMSK_CAM);
}
static void
-s3c_irq_cam_ack(unsigned int irqno)
+s3c_irq_cam_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_CAM, 3<<11);
+ s3c_irqsub_maskack(data->irq, INTMSK_CAM, 3 << 11);
}
static struct irq_chip s3c_irq_cam = {
- .mask = s3c_irq_cam_mask,
- .unmask = s3c_irq_cam_unmask,
- .ack = s3c_irq_cam_ack,
+ .irq_mask = s3c_irq_cam_mask,
+ .irq_unmask = s3c_irq_cam_unmask,
+ .irq_ack = s3c_irq_cam_ack,
};
static int s3c244x_irq_add(struct sys_device *sysdev)
diff --git a/arch/arm/mach-s3c2443/Kconfig b/arch/arm/mach-s3c2443/Kconfig
index 31babec..d8eb868 100644
--- a/arch/arm/mach-s3c2443/Kconfig
+++ b/arch/arm/mach-s3c2443/Kconfig
@@ -10,6 +10,7 @@
select CPU_LLSERIAL_S3C2440
select SAMSUNG_CLKSRC
select S3C2443_CLOCK
+ select S3C_GPIO_PULL_S3C2443
help
Support for the S3C2443 SoC from the S3C24XX line
@@ -25,7 +26,7 @@
bool "SMDK2443"
select CPU_S3C2443
select MACH_SMDK
- select S3C_DEV_HSMMC
+ select S3C_DEV_HSMMC1
help
Say Y here if you are using an SMDK2443
diff --git a/arch/arm/mach-s3c2443/clock.c b/arch/arm/mach-s3c2443/clock.c
index 0c3c0c8..f4ec6d5 100644
--- a/arch/arm/mach-s3c2443/clock.c
+++ b/arch/arm/mach-s3c2443/clock.c
@@ -196,7 +196,7 @@
static struct clksrc_clk clk_hsmmc_div = {
.clk = {
.name = "hsmmc-div",
- .id = -1,
+ .id = 1,
.parent = &clk_esysclk.clk,
},
.reg_div = { .reg = S3C2443_CLKDIV1, .size = 2, .shift = 6 },
@@ -231,7 +231,7 @@
static struct clk clk_hsmmc = {
.name = "hsmmc-if",
- .id = -1,
+ .id = 1,
.parent = &clk_hsmmc_div.clk,
.enable = s3c2443_enable_hsmmc,
.ops = &(struct clk_ops) {
diff --git a/arch/arm/mach-s3c2443/irq.c b/arch/arm/mach-s3c2443/irq.c
index 8934247..c7820f9 100644
--- a/arch/arm/mach-s3c2443/irq.c
+++ b/arch/arm/mach-s3c2443/irq.c
@@ -75,28 +75,27 @@
#define INTMSK_WDTAC97 (1UL << (IRQ_WDT - IRQ_EINT0))
#define SUBMSK_WDTAC97 INTMSK(IRQ_S3C2443_WDT, IRQ_S3C2443_AC97)
-static void s3c2443_irq_wdtac97_mask(unsigned int irqno)
+static void s3c2443_irq_wdtac97_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_WDTAC97, SUBMSK_WDTAC97);
+ s3c_irqsub_mask(data->irq, INTMSK_WDTAC97, SUBMSK_WDTAC97);
}
-static void s3c2443_irq_wdtac97_unmask(unsigned int irqno)
+static void s3c2443_irq_wdtac97_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_WDTAC97);
+ s3c_irqsub_unmask(data->irq, INTMSK_WDTAC97);
}
-static void s3c2443_irq_wdtac97_ack(unsigned int irqno)
+static void s3c2443_irq_wdtac97_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_WDTAC97, SUBMSK_WDTAC97);
+ s3c_irqsub_maskack(data->irq, INTMSK_WDTAC97, SUBMSK_WDTAC97);
}
static struct irq_chip s3c2443_irq_wdtac97 = {
- .mask = s3c2443_irq_wdtac97_mask,
- .unmask = s3c2443_irq_wdtac97_unmask,
- .ack = s3c2443_irq_wdtac97_ack,
+ .irq_mask = s3c2443_irq_wdtac97_mask,
+ .irq_unmask = s3c2443_irq_wdtac97_unmask,
+ .irq_ack = s3c2443_irq_wdtac97_ack,
};
-
/* LCD sub interrupts */
static void s3c2443_irq_demux_lcd(unsigned int irq, struct irq_desc *desc)
@@ -107,28 +106,27 @@
#define INTMSK_LCD (1UL << (IRQ_LCD - IRQ_EINT0))
#define SUBMSK_LCD INTMSK(IRQ_S3C2443_LCD1, IRQ_S3C2443_LCD4)
-static void s3c2443_irq_lcd_mask(unsigned int irqno)
+static void s3c2443_irq_lcd_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_LCD, SUBMSK_LCD);
+ s3c_irqsub_mask(data->irq, INTMSK_LCD, SUBMSK_LCD);
}
-static void s3c2443_irq_lcd_unmask(unsigned int irqno)
+static void s3c2443_irq_lcd_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_LCD);
+ s3c_irqsub_unmask(data->irq, INTMSK_LCD);
}
-static void s3c2443_irq_lcd_ack(unsigned int irqno)
+static void s3c2443_irq_lcd_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_LCD, SUBMSK_LCD);
+ s3c_irqsub_maskack(data->irq, INTMSK_LCD, SUBMSK_LCD);
}
static struct irq_chip s3c2443_irq_lcd = {
- .mask = s3c2443_irq_lcd_mask,
- .unmask = s3c2443_irq_lcd_unmask,
- .ack = s3c2443_irq_lcd_ack,
+ .irq_mask = s3c2443_irq_lcd_mask,
+ .irq_unmask = s3c2443_irq_lcd_unmask,
+ .irq_ack = s3c2443_irq_lcd_ack,
};
-
/* DMA sub interrupts */
static void s3c2443_irq_demux_dma(unsigned int irq, struct irq_desc *desc)
@@ -139,29 +137,27 @@
#define INTMSK_DMA (1UL << (IRQ_S3C2443_DMA - IRQ_EINT0))
#define SUBMSK_DMA INTMSK(IRQ_S3C2443_DMA0, IRQ_S3C2443_DMA5)
-
-static void s3c2443_irq_dma_mask(unsigned int irqno)
+static void s3c2443_irq_dma_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_DMA, SUBMSK_DMA);
+ s3c_irqsub_mask(data->irq, INTMSK_DMA, SUBMSK_DMA);
}
-static void s3c2443_irq_dma_unmask(unsigned int irqno)
+static void s3c2443_irq_dma_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_DMA);
+ s3c_irqsub_unmask(data->irq, INTMSK_DMA);
}
-static void s3c2443_irq_dma_ack(unsigned int irqno)
+static void s3c2443_irq_dma_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_DMA, SUBMSK_DMA);
+ s3c_irqsub_maskack(data->irq, INTMSK_DMA, SUBMSK_DMA);
}
static struct irq_chip s3c2443_irq_dma = {
- .mask = s3c2443_irq_dma_mask,
- .unmask = s3c2443_irq_dma_unmask,
- .ack = s3c2443_irq_dma_ack,
+ .irq_mask = s3c2443_irq_dma_mask,
+ .irq_unmask = s3c2443_irq_dma_unmask,
+ .irq_ack = s3c2443_irq_dma_ack,
};
-
/* UART3 sub interrupts */
static void s3c2443_irq_demux_uart3(unsigned int irq, struct irq_desc *desc)
@@ -172,28 +168,27 @@
#define INTMSK_UART3 (1UL << (IRQ_S3C2443_UART3 - IRQ_EINT0))
#define SUBMSK_UART3 (0x7 << (IRQ_S3C2443_RX3 - S3C2410_IRQSUB(0)))
-static void s3c2443_irq_uart3_mask(unsigned int irqno)
+static void s3c2443_irq_uart3_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_UART3, SUBMSK_UART3);
+ s3c_irqsub_mask(data->irq, INTMSK_UART3, SUBMSK_UART3);
}
-static void s3c2443_irq_uart3_unmask(unsigned int irqno)
+static void s3c2443_irq_uart3_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_UART3);
+ s3c_irqsub_unmask(data->irq, INTMSK_UART3);
}
-static void s3c2443_irq_uart3_ack(unsigned int irqno)
+static void s3c2443_irq_uart3_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_UART3, SUBMSK_UART3);
+ s3c_irqsub_maskack(data->irq, INTMSK_UART3, SUBMSK_UART3);
}
static struct irq_chip s3c2443_irq_uart3 = {
- .mask = s3c2443_irq_uart3_mask,
- .unmask = s3c2443_irq_uart3_unmask,
- .ack = s3c2443_irq_uart3_ack,
+ .irq_mask = s3c2443_irq_uart3_mask,
+ .irq_unmask = s3c2443_irq_uart3_unmask,
+ .irq_ack = s3c2443_irq_uart3_ack,
};
-
/* CAM sub interrupts */
static void s3c2443_irq_demux_cam(unsigned int irq, struct irq_desc *desc)
@@ -204,25 +199,25 @@
#define INTMSK_CAM (1UL << (IRQ_CAM - IRQ_EINT0))
#define SUBMSK_CAM INTMSK(IRQ_S3C2440_CAM_C, IRQ_S3C2440_CAM_P)
-static void s3c2443_irq_cam_mask(unsigned int irqno)
+static void s3c2443_irq_cam_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_CAM, SUBMSK_CAM);
+ s3c_irqsub_mask(data->irq, INTMSK_CAM, SUBMSK_CAM);
}
-static void s3c2443_irq_cam_unmask(unsigned int irqno)
+static void s3c2443_irq_cam_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_CAM);
+ s3c_irqsub_unmask(data->irq, INTMSK_CAM);
}
-static void s3c2443_irq_cam_ack(unsigned int irqno)
+static void s3c2443_irq_cam_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_CAM, SUBMSK_CAM);
+ s3c_irqsub_maskack(data->irq, INTMSK_CAM, SUBMSK_CAM);
}
static struct irq_chip s3c2443_irq_cam = {
- .mask = s3c2443_irq_cam_mask,
- .unmask = s3c2443_irq_cam_unmask,
- .ack = s3c2443_irq_cam_ack,
+ .irq_mask = s3c2443_irq_cam_mask,
+ .irq_unmask = s3c2443_irq_cam_unmask,
+ .irq_ack = s3c2443_irq_cam_ack,
};
/* IRQ initialisation code */
diff --git a/arch/arm/mach-s3c2443/mach-smdk2443.c b/arch/arm/mach-s3c2443/mach-smdk2443.c
index 4337f0a..514275e 100644
--- a/arch/arm/mach-s3c2443/mach-smdk2443.c
+++ b/arch/arm/mach-s3c2443/mach-smdk2443.c
@@ -99,13 +99,20 @@
.ucon = 0x3c5,
.ulcon = 0x43,
.ufcon = 0x51,
+ },
+ [3] = {
+ .hwport = 3,
+ .flags = 0,
+ .ucon = 0x3c5,
+ .ulcon = 0x03,
+ .ufcon = 0x51,
}
};
static struct platform_device *smdk2443_devices[] __initdata = {
&s3c_device_wdt,
&s3c_device_i2c0,
- &s3c_device_hsmmc0,
+ &s3c_device_hsmmc1,
#ifdef CONFIG_SND_SOC_SMDK2443_WM9710
&s3c_device_ac97,
#endif
diff --git a/arch/arm/mach-s3c2443/s3c2443.c b/arch/arm/mach-s3c2443/s3c2443.c
index 33d18dd..e6a28ba 100644
--- a/arch/arm/mach-s3c2443/s3c2443.c
+++ b/arch/arm/mach-s3c2443/s3c2443.c
@@ -16,6 +16,7 @@
#include <linux/list.h>
#include <linux/timer.h>
#include <linux/init.h>
+#include <linux/gpio.h>
#include <linux/platform_device.h>
#include <linux/serial_core.h>
#include <linux/sysdev.h>
@@ -32,6 +33,9 @@
#include <mach/regs-s3c2443-clock.h>
#include <mach/reset.h>
+#include <plat/gpio-core.h>
+#include <plat/gpio-cfg.h>
+#include <plat/gpio-cfg-helpers.h>
#include <plat/s3c2443.h>
#include <plat/devs.h>
#include <plat/cpu.h>
@@ -86,6 +90,9 @@
void __init s3c2443_map_io(void)
{
+ s3c24xx_gpiocfg_default.set_pull = s3c_gpio_setpull_s3c2443;
+ s3c24xx_gpiocfg_default.get_pull = s3c_gpio_getpull_s3c2443;
+
iotable_init(s3c2443_iodesc, ARRAY_SIZE(s3c2443_iodesc));
}
diff --git a/arch/arm/mach-s3c64xx/clock.c b/arch/arm/mach-s3c64xx/clock.c
index 1c98d2ff..dd37820 100644
--- a/arch/arm/mach-s3c64xx/clock.c
+++ b/arch/arm/mach-s3c64xx/clock.c
@@ -127,7 +127,7 @@
return s3c64xx_gate(S3C_SCLK_GATE, clk, enable);
}
-static struct clk init_clocks_disable[] = {
+static struct clk init_clocks_off[] = {
{
.name = "nand",
.id = -1,
@@ -834,10 +834,6 @@
void __init s3c64xx_register_clocks(unsigned long xtal,
unsigned armclk_divlimit)
{
- struct clk *clkp;
- int ret;
- int ptr;
-
armclk_mask = armclk_divlimit;
s3c24xx_register_baseclocks(xtal);
@@ -845,17 +841,8 @@
s3c_register_clocks(init_clocks, ARRAY_SIZE(init_clocks));
- clkp = init_clocks_disable;
- for (ptr = 0; ptr < ARRAY_SIZE(init_clocks_disable); ptr++, clkp++) {
-
- ret = s3c24xx_register_clock(clkp);
- if (ret < 0) {
- printk(KERN_ERR "Failed to register clock %s (%d)\n",
- clkp->name, ret);
- }
-
- (clkp->enable)(clkp, 0);
- }
+ s3c_register_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
+ s3c_disable_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
s3c24xx_register_clocks(clks1, ARRAY_SIZE(clks1));
s3c_register_clksrc(clksrcs, ARRAY_SIZE(clksrcs));
diff --git a/arch/arm/mach-s3c64xx/dma.c b/arch/arm/mach-s3c64xx/dma.c
index 372ea68..135db1b 100644
--- a/arch/arm/mach-s3c64xx/dma.c
+++ b/arch/arm/mach-s3c64xx/dma.c
@@ -212,6 +212,7 @@
config = readl(chan->regs + PL080S_CH_CONFIG);
config |= PL080_CONFIG_ENABLE;
+ config &= ~PL080_CONFIG_HALT;
pr_debug("%s: writing config %08x\n", __func__, config);
writel(config, chan->regs + PL080S_CH_CONFIG);
diff --git a/arch/arm/mach-s3c64xx/irq-eint.c b/arch/arm/mach-s3c64xx/irq-eint.c
index 5682d6a..2ead818 100644
--- a/arch/arm/mach-s3c64xx/irq-eint.c
+++ b/arch/arm/mach-s3c64xx/irq-eint.c
@@ -30,41 +30,41 @@
#include <plat/pm.h>
#define eint_offset(irq) ((irq) - IRQ_EINT(0))
-#define eint_irq_to_bit(irq) (1 << eint_offset(irq))
+#define eint_irq_to_bit(irq) ((u32)(1 << eint_offset(irq)))
-static inline void s3c_irq_eint_mask(unsigned int irq)
+static inline void s3c_irq_eint_mask(struct irq_data *data)
{
u32 mask;
mask = __raw_readl(S3C64XX_EINT0MASK);
- mask |= eint_irq_to_bit(irq);
+ mask |= (u32)data->chip_data;
__raw_writel(mask, S3C64XX_EINT0MASK);
}
-static void s3c_irq_eint_unmask(unsigned int irq)
+static void s3c_irq_eint_unmask(struct irq_data *data)
{
u32 mask;
mask = __raw_readl(S3C64XX_EINT0MASK);
- mask &= ~eint_irq_to_bit(irq);
+ mask &= ~((u32)data->chip_data);
__raw_writel(mask, S3C64XX_EINT0MASK);
}
-static inline void s3c_irq_eint_ack(unsigned int irq)
+static inline void s3c_irq_eint_ack(struct irq_data *data)
{
- __raw_writel(eint_irq_to_bit(irq), S3C64XX_EINT0PEND);
+ __raw_writel((u32)data->chip_data, S3C64XX_EINT0PEND);
}
-static void s3c_irq_eint_maskack(unsigned int irq)
+static void s3c_irq_eint_maskack(struct irq_data *data)
{
/* compiler should in-line these */
- s3c_irq_eint_mask(irq);
- s3c_irq_eint_ack(irq);
+ s3c_irq_eint_mask(data);
+ s3c_irq_eint_ack(data);
}
-static int s3c_irq_eint_set_type(unsigned int irq, unsigned int type)
+static int s3c_irq_eint_set_type(struct irq_data *data, unsigned int type)
{
- int offs = eint_offset(irq);
+ int offs = eint_offset(data->irq);
int pin, pin_val;
int shift;
u32 ctrl, mask;
@@ -140,12 +140,12 @@
static struct irq_chip s3c_irq_eint = {
.name = "s3c-eint",
- .mask = s3c_irq_eint_mask,
- .unmask = s3c_irq_eint_unmask,
- .mask_ack = s3c_irq_eint_maskack,
- .ack = s3c_irq_eint_ack,
- .set_type = s3c_irq_eint_set_type,
- .set_wake = s3c_irqext_wake,
+ .irq_mask = s3c_irq_eint_mask,
+ .irq_unmask = s3c_irq_eint_unmask,
+ .irq_mask_ack = s3c_irq_eint_maskack,
+ .irq_ack = s3c_irq_eint_ack,
+ .irq_set_type = s3c_irq_eint_set_type,
+ .irq_set_wake = s3c_irqext_wake,
};
/* s3c_irq_demux_eint
@@ -198,6 +198,7 @@
for (irq = IRQ_EINT(0); irq <= IRQ_EINT(27); irq++) {
set_irq_chip(irq, &s3c_irq_eint);
+ set_irq_chip_data(irq, (void *)eint_irq_to_bit(irq));
set_irq_handler(irq, handle_level_irq);
set_irq_flags(irq, IRQF_VALID);
}
diff --git a/arch/arm/mach-s5p6442/clock.c b/arch/arm/mach-s5p6442/clock.c
index 16d6e7e..fbbc7be 100644
--- a/arch/arm/mach-s5p6442/clock.c
+++ b/arch/arm/mach-s5p6442/clock.c
@@ -340,7 +340,7 @@
clk_pclkd1.rate = pclkd1;
}
-static struct clk init_clocks_disable[] = {
+static struct clk init_clocks_off[] = {
{
.name = "pdma",
.id = -1,
@@ -408,23 +408,13 @@
void __init s5p6442_register_clocks(void)
{
- struct clk *clkptr;
- int i, ret;
-
s3c24xx_register_clocks(clks, ARRAY_SIZE(clks));
s3c_register_clksrc(clksrcs, ARRAY_SIZE(clksrcs));
s3c_register_clocks(init_clocks, ARRAY_SIZE(init_clocks));
- clkptr = init_clocks_disable;
- for (i = 0; i < ARRAY_SIZE(init_clocks_disable); i++, clkptr++) {
- ret = s3c24xx_register_clock(clkptr);
- if (ret < 0) {
- printk(KERN_ERR "Fail to register clock %s (%d)\n",
- clkptr->name, ret);
- } else
- (clkptr->enable)(clkptr, 0);
- }
+ s3c_register_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
+ s3c_disable_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
s3c_pwmclk_init();
}
diff --git a/arch/arm/mach-s5p6442/include/mach/map.h b/arch/arm/mach-s5p6442/include/mach/map.h
index 31fb2e68..203dd5a 100644
--- a/arch/arm/mach-s5p6442/include/mach/map.h
+++ b/arch/arm/mach-s5p6442/include/mach/map.h
@@ -28,6 +28,9 @@
#define S5P6442_PA_VIC1 (0xE4100000)
#define S5P6442_PA_VIC2 (0xE4200000)
+#define S5P6442_PA_SROMC (0xE7000000)
+#define S5P_PA_SROMC S5P6442_PA_SROMC
+
#define S5P6442_PA_MDMA 0xE8000000
#define S5P6442_PA_PDMA 0xE9000000
diff --git a/arch/arm/mach-s5p6442/mach-smdk6442.c b/arch/arm/mach-s5p6442/mach-smdk6442.c
index 819fd80..e69f137 100644
--- a/arch/arm/mach-s5p6442/mach-smdk6442.c
+++ b/arch/arm/mach-s5p6442/mach-smdk6442.c
@@ -12,6 +12,7 @@
#include <linux/types.h>
#include <linux/init.h>
#include <linux/serial_core.h>
+#include <linux/i2c.h>
#include <asm/mach/arch.h>
#include <asm/mach/map.h>
@@ -25,6 +26,7 @@
#include <plat/s5p6442.h>
#include <plat/devs.h>
#include <plat/cpu.h>
+#include <plat/iic.h>
/* Following are default values for UCON, ULCON and UFCON UART registers */
#define SMDK6442_UCON_DEFAULT (S3C2410_UCON_TXILEVEL | \
@@ -65,10 +67,15 @@
};
static struct platform_device *smdk6442_devices[] __initdata = {
+ &s3c_device_i2c0,
&s5p6442_device_iis0,
&s3c_device_wdt,
};
+static struct i2c_board_info smdk6442_i2c_devs0[] __initdata = {
+ { I2C_BOARD_INFO("wm8580", 0x1b), },
+};
+
static void __init smdk6442_map_io(void)
{
s5p_init_io(NULL, 0, S5P_VA_CHIPID);
@@ -78,6 +85,9 @@
static void __init smdk6442_machine_init(void)
{
+ s3c_i2c0_set_platdata(NULL);
+ i2c_register_board_info(0, smdk6442_i2c_devs0,
+ ARRAY_SIZE(smdk6442_i2c_devs0));
platform_add_devices(smdk6442_devices, ARRAY_SIZE(smdk6442_devices));
}
diff --git a/arch/arm/mach-s5p6442/setup-i2c0.c b/arch/arm/mach-s5p6442/setup-i2c0.c
index 662695d..aad8565 100644
--- a/arch/arm/mach-s5p6442/setup-i2c0.c
+++ b/arch/arm/mach-s5p6442/setup-i2c0.c
@@ -14,12 +14,15 @@
#include <linux/kernel.h>
#include <linux/types.h>
+#include <linux/gpio.h>
struct platform_device; /* don't need the contents */
+#include <plat/gpio-cfg.h>
#include <plat/iic.h>
void s3c_i2c0_cfg_gpio(struct platform_device *dev)
{
- /* Will be populated later */
+ s3c_gpio_cfgall_range(S5P6442_GPD1(0), 2,
+ S3C_GPIO_SFN(2), S3C_GPIO_PULL_UP);
}
diff --git a/arch/arm/mach-s5p64x0/Makefile b/arch/arm/mach-s5p64x0/Makefile
index 2655829..ae6bf6f 100644
--- a/arch/arm/mach-s5p64x0/Makefile
+++ b/arch/arm/mach-s5p64x0/Makefile
@@ -12,9 +12,9 @@
# Core support for S5P64X0 system
-obj-$(CONFIG_ARCH_S5P64X0) += cpu.o init.o clock.o dma.o
+obj-$(CONFIG_ARCH_S5P64X0) += cpu.o init.o clock.o dma.o gpiolib.o
obj-$(CONFIG_ARCH_S5P64X0) += setup-i2c0.o
-obj-$(CONFIG_CPU_S5P6440) += clock-s5p6440.o gpio.o
+obj-$(CONFIG_CPU_S5P6440) += clock-s5p6440.o
obj-$(CONFIG_CPU_S5P6450) += clock-s5p6450.o
# machine support
diff --git a/arch/arm/mach-s5p64x0/clock-s5p6440.c b/arch/arm/mach-s5p64x0/clock-s5p6440.c
index 409c5fc..9f12c2e 100644
--- a/arch/arm/mach-s5p64x0/clock-s5p6440.c
+++ b/arch/arm/mach-s5p64x0/clock-s5p6440.c
@@ -133,7 +133,7 @@
* recommended to keep the following clocks disabled until the driver requests
* for enabling the clock.
*/
-static struct clk init_clocks_disable[] = {
+static struct clk init_clocks_off[] = {
{
.name = "nand",
.id = -1,
@@ -419,7 +419,7 @@
static struct clksrc_clk clksrcs[] = {
{
.clk = {
- .name = "mmc_bus",
+ .name = "sclk_mmc",
.id = 0,
.ctrlbit = (1 << 24),
.enable = s5p64x0_sclk_ctrl,
@@ -429,7 +429,7 @@
.reg_div = { .reg = S5P64X0_CLK_DIV1, .shift = 0, .size = 4 },
}, {
.clk = {
- .name = "mmc_bus",
+ .name = "sclk_mmc",
.id = 1,
.ctrlbit = (1 << 25),
.enable = s5p64x0_sclk_ctrl,
@@ -439,7 +439,7 @@
.reg_div = { .reg = S5P64X0_CLK_DIV1, .shift = 4, .size = 4 },
}, {
.clk = {
- .name = "mmc_bus",
+ .name = "sclk_mmc",
.id = 2,
.ctrlbit = (1 << 26),
.enable = s5p64x0_sclk_ctrl,
@@ -602,8 +602,6 @@
void __init s5p6440_register_clocks(void)
{
- struct clk *clkp;
- int ret;
int ptr;
s3c24xx_register_clocks(clks, ARRAY_SIZE(clks));
@@ -614,16 +612,8 @@
s3c_register_clksrc(clksrcs, ARRAY_SIZE(clksrcs));
s3c_register_clocks(init_clocks, ARRAY_SIZE(init_clocks));
- clkp = init_clocks_disable;
- for (ptr = 0; ptr < ARRAY_SIZE(init_clocks_disable); ptr++, clkp++) {
-
- ret = s3c24xx_register_clock(clkp);
- if (ret < 0) {
- printk(KERN_ERR "Failed to register clock %s (%d)\n",
- clkp->name, ret);
- }
- (clkp->enable)(clkp, 0);
- }
+ s3c_register_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
+ s3c_disable_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
s3c_pwmclk_init();
}
diff --git a/arch/arm/mach-s5p64x0/clock-s5p6450.c b/arch/arm/mach-s5p64x0/clock-s5p6450.c
index 7fc6abd..4eec457 100644
--- a/arch/arm/mach-s5p64x0/clock-s5p6450.c
+++ b/arch/arm/mach-s5p64x0/clock-s5p6450.c
@@ -181,7 +181,7 @@
* recommended to keep the following clocks disabled until the driver requests
* for enabling the clock.
*/
-static struct clk init_clocks_disable[] = {
+static struct clk init_clocks_off[] = {
{
.name = "usbhost",
.id = -1,
@@ -231,6 +231,12 @@
.enable = s5p64x0_pclk_ctrl,
.ctrlbit = (1 << 5),
}, {
+ .name = "rtc",
+ .id = -1,
+ .parent = &clk_pclk_low.clk,
+ .enable = s5p64x0_pclk_ctrl,
+ .ctrlbit = (1 << 6),
+ }, {
.name = "adc",
.id = -1,
.parent = &clk_pclk_low.clk,
@@ -261,6 +267,18 @@
.enable = s5p64x0_pclk_ctrl,
.ctrlbit = (1 << 26),
}, {
+ .name = "iis",
+ .id = 1,
+ .parent = &clk_pclk_low.clk,
+ .enable = s5p64x0_pclk_ctrl,
+ .ctrlbit = (1 << 15),
+ }, {
+ .name = "iis",
+ .id = 2,
+ .parent = &clk_pclk_low.clk,
+ .enable = s5p64x0_pclk_ctrl,
+ .ctrlbit = (1 << 16),
+ }, {
.name = "i2c",
.id = 1,
.parent = &clk_pclk_low.clk,
@@ -633,8 +651,6 @@
void __init s5p6450_register_clocks(void)
{
- struct clk *clkp;
- int ret;
int ptr;
for (ptr = 0; ptr < ARRAY_SIZE(sysclks); ptr++)
@@ -643,16 +659,8 @@
s3c_register_clksrc(clksrcs, ARRAY_SIZE(clksrcs));
s3c_register_clocks(init_clocks, ARRAY_SIZE(init_clocks));
- clkp = init_clocks_disable;
- for (ptr = 0; ptr < ARRAY_SIZE(init_clocks_disable); ptr++, clkp++) {
-
- ret = s3c24xx_register_clock(clkp);
- if (ret < 0) {
- printk(KERN_ERR "Failed to register clock %s (%d)\n",
- clkp->name, ret);
- }
- (clkp->enable)(clkp, 0);
- }
+ s3c_register_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
+ s3c_disable_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
s3c_pwmclk_init();
}
diff --git a/arch/arm/mach-s5p64x0/dev-audio.c b/arch/arm/mach-s5p64x0/dev-audio.c
index 14f89e73..35f1f22 100644
--- a/arch/arm/mach-s5p64x0/dev-audio.c
+++ b/arch/arm/mach-s5p64x0/dev-audio.c
@@ -24,13 +24,13 @@
[1] = "sclk_audio2",
};
-static int s5p64x0_cfg_i2s(struct platform_device *pdev)
+static int s5p6440_cfg_i2s(struct platform_device *pdev)
{
- /* configure GPIO for i2s port */
switch (pdev->id) {
case 0:
- s3c_gpio_cfgpin_range(S5P6440_GPR(4), 5, S3C_GPIO_SFN(5));
- s3c_gpio_cfgpin_range(S5P6440_GPR(13), 2, S3C_GPIO_SFN(5));
+ s3c_gpio_cfgpin_range(S5P6440_GPC(4), 2, S3C_GPIO_SFN(5));
+ s3c_gpio_cfgpin(S5P6440_GPC(7), S3C_GPIO_SFN(5));
+ s3c_gpio_cfgpin_range(S5P6440_GPH(6), 4, S3C_GPIO_SFN(5));
break;
default:
printk(KERN_ERR "Invalid Device %d\n", pdev->id);
@@ -40,8 +40,8 @@
return 0;
}
-static struct s3c_audio_pdata s5p64x0_i2s_pdata = {
- .cfg_gpio = s5p64x0_cfg_i2s,
+static struct s3c_audio_pdata s5p6440_i2s_pdata = {
+ .cfg_gpio = s5p6440_cfg_i2s,
.type = {
.i2s = {
.quirks = QUIRK_PRI_6CHAN,
@@ -50,7 +50,7 @@
},
};
-static struct resource s5p64x0_iis0_resource[] = {
+static struct resource s5p64x0_i2s0_resource[] = {
[0] = {
.start = S5P64X0_PA_I2S,
.end = S5P64X0_PA_I2S + 0x100 - 1,
@@ -71,20 +71,117 @@
struct platform_device s5p6440_device_iis = {
.name = "samsung-i2s",
.id = 0,
- .num_resources = ARRAY_SIZE(s5p64x0_iis0_resource),
- .resource = s5p64x0_iis0_resource,
+ .num_resources = ARRAY_SIZE(s5p64x0_i2s0_resource),
+ .resource = s5p64x0_i2s0_resource,
.dev = {
- .platform_data = &s5p64x0_i2s_pdata,
+ .platform_data = &s5p6440_i2s_pdata,
+ },
+};
+
+static int s5p6450_cfg_i2s(struct platform_device *pdev)
+{
+ switch (pdev->id) {
+ case 0:
+ s3c_gpio_cfgpin_range(S5P6450_GPR(4), 5, S3C_GPIO_SFN(5));
+ s3c_gpio_cfgpin_range(S5P6450_GPR(13), 2, S3C_GPIO_SFN(5));
+ break;
+ case 1:
+ s3c_gpio_cfgpin(S5P6440_GPB(4), S3C_GPIO_SFN(5));
+ s3c_gpio_cfgpin_range(S5P6450_GPC(0), 4, S3C_GPIO_SFN(5));
+ break;
+ case 2:
+ s3c_gpio_cfgpin_range(S5P6450_GPK(0), 5, S3C_GPIO_SFN(5));
+ break;
+ default:
+ printk(KERN_ERR "Invalid Device %d\n", pdev->id);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static struct s3c_audio_pdata s5p6450_i2s0_pdata = {
+ .cfg_gpio = s5p6450_cfg_i2s,
+ .type = {
+ .i2s = {
+ .quirks = QUIRK_PRI_6CHAN,
+ .src_clk = rclksrc,
+ },
},
};
struct platform_device s5p6450_device_iis0 = {
.name = "samsung-i2s",
.id = 0,
- .num_resources = ARRAY_SIZE(s5p64x0_iis0_resource),
- .resource = s5p64x0_iis0_resource,
+ .num_resources = ARRAY_SIZE(s5p64x0_i2s0_resource),
+ .resource = s5p64x0_i2s0_resource,
.dev = {
- .platform_data = &s5p64x0_i2s_pdata,
+ .platform_data = &s5p6450_i2s0_pdata,
+ },
+};
+
+static struct s3c_audio_pdata s5p6450_i2s_pdata = {
+ .cfg_gpio = s5p6450_cfg_i2s,
+ .type = {
+ .i2s = {
+ .src_clk = rclksrc,
+ },
+ },
+};
+
+static struct resource s5p6450_i2s1_resource[] = {
+ [0] = {
+ .start = S5P6450_PA_I2S1,
+ .end = S5P6450_PA_I2S1 + 0x100 - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [1] = {
+ .start = DMACH_I2S1_TX,
+ .end = DMACH_I2S1_TX,
+ .flags = IORESOURCE_DMA,
+ },
+ [2] = {
+ .start = DMACH_I2S1_RX,
+ .end = DMACH_I2S1_RX,
+ .flags = IORESOURCE_DMA,
+ },
+};
+
+struct platform_device s5p6450_device_iis1 = {
+ .name = "samsung-i2s",
+ .id = 1,
+ .num_resources = ARRAY_SIZE(s5p6450_i2s1_resource),
+ .resource = s5p6450_i2s1_resource,
+ .dev = {
+ .platform_data = &s5p6450_i2s_pdata,
+ },
+};
+
+static struct resource s5p6450_i2s2_resource[] = {
+ [0] = {
+ .start = S5P6450_PA_I2S2,
+ .end = S5P6450_PA_I2S2 + 0x100 - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [1] = {
+ .start = DMACH_I2S2_TX,
+ .end = DMACH_I2S2_TX,
+ .flags = IORESOURCE_DMA,
+ },
+ [2] = {
+ .start = DMACH_I2S2_RX,
+ .end = DMACH_I2S2_RX,
+ .flags = IORESOURCE_DMA,
+ },
+};
+
+struct platform_device s5p6450_device_iis2 = {
+ .name = "samsung-i2s",
+ .id = 2,
+ .num_resources = ARRAY_SIZE(s5p6450_i2s2_resource),
+ .resource = s5p6450_i2s2_resource,
+ .dev = {
+ .platform_data = &s5p6450_i2s_pdata,
},
};
diff --git a/arch/arm/mach-s5p64x0/gpio.c b/arch/arm/mach-s5p64x0/gpio.c
deleted file mode 100644
index 39159dd..0000000
--- a/arch/arm/mach-s5p64x0/gpio.c
+++ /dev/null
@@ -1,342 +0,0 @@
-/* linux/arch/arm/mach-s5p64x0/gpio.c
- *
- * Copyright (c) 2009-2010 Samsung Electronics Co., Ltd.
- * http://www.samsung.com
- *
- * S5P64X0 - GPIOlib support
- *
- * 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/kernel.h>
-#include <linux/irq.h>
-#include <linux/io.h>
-#include <linux/gpio.h>
-
-#include <mach/map.h>
-#include <mach/regs-gpio.h>
-
-#include <plat/gpio-core.h>
-#include <plat/gpio-cfg.h>
-#include <plat/gpio-cfg-helpers.h>
-
-/* To be implemented S5P6450 GPIO */
-
-/*
- * S5P6440 GPIO bank summary:
- *
- * Bank GPIOs Style SlpCon ExtInt Group
- * A 6 4Bit Yes 1
- * B 7 4Bit Yes 1
- * C 8 4Bit Yes 2
- * F 2 2Bit Yes 4 [1]
- * G 7 4Bit Yes 5
- * H 10 4Bit[2] Yes 6
- * I 16 2Bit Yes None
- * J 12 2Bit Yes None
- * N 16 2Bit No IRQ_EINT
- * P 8 2Bit Yes 8
- * R 15 4Bit[2] Yes 8
- *
- * [1] BANKF pins 14,15 do not form part of the external interrupt sources
- * [2] BANK has two control registers, GPxCON0 and GPxCON1
- */
-
-static int s5p64x0_gpiolib_rbank_4bit2_input(struct gpio_chip *chip,
- unsigned int offset)
-{
- struct s3c_gpio_chip *ourchip = to_s3c_gpio(chip);
- void __iomem *base = ourchip->base;
- void __iomem *regcon = base;
- unsigned long con;
- unsigned long flags;
-
- switch (offset) {
- case 6:
- offset += 1;
- case 0:
- case 1:
- case 2:
- case 3:
- case 4:
- case 5:
- regcon -= 4;
- break;
- default:
- offset -= 7;
- break;
- }
-
- s3c_gpio_lock(ourchip, flags);
-
- con = __raw_readl(regcon);
- con &= ~(0xf << con_4bit_shift(offset));
- __raw_writel(con, regcon);
-
- s3c_gpio_unlock(ourchip, flags);
-
- return 0;
-}
-
-static int s5p64x0_gpiolib_rbank_4bit2_output(struct gpio_chip *chip,
- unsigned int offset, int value)
-{
- struct s3c_gpio_chip *ourchip = to_s3c_gpio(chip);
- void __iomem *base = ourchip->base;
- void __iomem *regcon = base;
- unsigned long con;
- unsigned long dat;
- unsigned long flags;
- unsigned con_offset = offset;
-
- switch (con_offset) {
- case 6:
- con_offset += 1;
- case 0:
- case 1:
- case 2:
- case 3:
- case 4:
- case 5:
- regcon -= 4;
- break;
- default:
- con_offset -= 7;
- break;
- }
-
- s3c_gpio_lock(ourchip, flags);
-
- con = __raw_readl(regcon);
- con &= ~(0xf << con_4bit_shift(con_offset));
- con |= 0x1 << con_4bit_shift(con_offset);
-
- dat = __raw_readl(base + GPIODAT_OFF);
- if (value)
- dat |= 1 << offset;
- else
- dat &= ~(1 << offset);
-
- __raw_writel(con, regcon);
- __raw_writel(dat, base + GPIODAT_OFF);
-
- s3c_gpio_unlock(ourchip, flags);
-
- return 0;
-}
-
-int s5p64x0_gpio_setcfg_4bit_rbank(struct s3c_gpio_chip *chip,
- unsigned int off, unsigned int cfg)
-{
- void __iomem *reg = chip->base;
- unsigned int shift;
- u32 con;
-
- switch (off) {
- case 0:
- case 1:
- case 2:
- case 3:
- case 4:
- case 5:
- shift = (off & 7) * 4;
- reg -= 4;
- break;
- case 6:
- shift = ((off + 1) & 7) * 4;
- reg -= 4;
- default:
- shift = ((off + 1) & 7) * 4;
- break;
- }
-
- if (s3c_gpio_is_cfg_special(cfg)) {
- cfg &= 0xf;
- cfg <<= shift;
- }
-
- con = __raw_readl(reg);
- con &= ~(0xf << shift);
- con |= cfg;
- __raw_writel(con, reg);
-
- return 0;
-}
-
-static struct s3c_gpio_cfg s5p64x0_gpio_cfgs[] = {
- {
- .cfg_eint = 0,
- }, {
- .cfg_eint = 7,
- }, {
- .cfg_eint = 3,
- .set_config = s5p64x0_gpio_setcfg_4bit_rbank,
- }, {
- .cfg_eint = 0,
- .set_config = s3c_gpio_setcfg_s3c24xx,
- .get_config = s3c_gpio_getcfg_s3c24xx,
- }, {
- .cfg_eint = 2,
- .set_config = s3c_gpio_setcfg_s3c24xx,
- .get_config = s3c_gpio_getcfg_s3c24xx,
- }, {
- .cfg_eint = 3,
- .set_config = s3c_gpio_setcfg_s3c24xx,
- .get_config = s3c_gpio_getcfg_s3c24xx,
- },
-};
-
-static struct s3c_gpio_chip s5p6440_gpio_4bit[] = {
- {
- .base = S5P6440_GPA_BASE,
- .config = &s5p64x0_gpio_cfgs[1],
- .chip = {
- .base = S5P6440_GPA(0),
- .ngpio = S5P6440_GPIO_A_NR,
- .label = "GPA",
- },
- }, {
- .base = S5P6440_GPB_BASE,
- .config = &s5p64x0_gpio_cfgs[1],
- .chip = {
- .base = S5P6440_GPB(0),
- .ngpio = S5P6440_GPIO_B_NR,
- .label = "GPB",
- },
- }, {
- .base = S5P6440_GPC_BASE,
- .config = &s5p64x0_gpio_cfgs[1],
- .chip = {
- .base = S5P6440_GPC(0),
- .ngpio = S5P6440_GPIO_C_NR,
- .label = "GPC",
- },
- }, {
- .base = S5P6440_GPG_BASE,
- .config = &s5p64x0_gpio_cfgs[1],
- .chip = {
- .base = S5P6440_GPG(0),
- .ngpio = S5P6440_GPIO_G_NR,
- .label = "GPG",
- },
- },
-};
-
-static struct s3c_gpio_chip s5p6440_gpio_4bit2[] = {
- {
- .base = S5P6440_GPH_BASE + 0x4,
- .config = &s5p64x0_gpio_cfgs[1],
- .chip = {
- .base = S5P6440_GPH(0),
- .ngpio = S5P6440_GPIO_H_NR,
- .label = "GPH",
- },
- },
-};
-
-static struct s3c_gpio_chip s5p6440_gpio_rbank_4bit2[] = {
- {
- .base = S5P6440_GPR_BASE + 0x4,
- .config = &s5p64x0_gpio_cfgs[2],
- .chip = {
- .base = S5P6440_GPR(0),
- .ngpio = S5P6440_GPIO_R_NR,
- .label = "GPR",
- },
- },
-};
-
-static struct s3c_gpio_chip s5p6440_gpio_2bit[] = {
- {
- .base = S5P6440_GPF_BASE,
- .config = &s5p64x0_gpio_cfgs[5],
- .chip = {
- .base = S5P6440_GPF(0),
- .ngpio = S5P6440_GPIO_F_NR,
- .label = "GPF",
- },
- }, {
- .base = S5P6440_GPI_BASE,
- .config = &s5p64x0_gpio_cfgs[3],
- .chip = {
- .base = S5P6440_GPI(0),
- .ngpio = S5P6440_GPIO_I_NR,
- .label = "GPI",
- },
- }, {
- .base = S5P6440_GPJ_BASE,
- .config = &s5p64x0_gpio_cfgs[3],
- .chip = {
- .base = S5P6440_GPJ(0),
- .ngpio = S5P6440_GPIO_J_NR,
- .label = "GPJ",
- },
- }, {
- .base = S5P6440_GPN_BASE,
- .config = &s5p64x0_gpio_cfgs[4],
- .chip = {
- .base = S5P6440_GPN(0),
- .ngpio = S5P6440_GPIO_N_NR,
- .label = "GPN",
- },
- }, {
- .base = S5P6440_GPP_BASE,
- .config = &s5p64x0_gpio_cfgs[5],
- .chip = {
- .base = S5P6440_GPP(0),
- .ngpio = S5P6440_GPIO_P_NR,
- .label = "GPP",
- },
- },
-};
-
-void __init s5p64x0_gpiolib_set_cfg(struct s3c_gpio_cfg *chipcfg, int nr_chips)
-{
- for (; nr_chips > 0; nr_chips--, chipcfg++) {
- if (!chipcfg->set_config)
- chipcfg->set_config = s3c_gpio_setcfg_s3c64xx_4bit;
- if (!chipcfg->get_config)
- chipcfg->get_config = s3c_gpio_getcfg_s3c64xx_4bit;
- if (!chipcfg->set_pull)
- chipcfg->set_pull = s3c_gpio_setpull_updown;
- if (!chipcfg->get_pull)
- chipcfg->get_pull = s3c_gpio_getpull_updown;
- }
-}
-
-static void __init s5p64x0_gpio_add_rbank_4bit2(struct s3c_gpio_chip *chip,
- int nr_chips)
-{
- for (; nr_chips > 0; nr_chips--, chip++) {
- chip->chip.direction_input = s5p64x0_gpiolib_rbank_4bit2_input;
- chip->chip.direction_output =
- s5p64x0_gpiolib_rbank_4bit2_output;
- s3c_gpiolib_add(chip);
- }
-}
-
-static int __init s5p6440_gpiolib_init(void)
-{
- struct s3c_gpio_chip *chips = s5p6440_gpio_2bit;
- int nr_chips = ARRAY_SIZE(s5p6440_gpio_2bit);
-
- s5p64x0_gpiolib_set_cfg(s5p64x0_gpio_cfgs,
- ARRAY_SIZE(s5p64x0_gpio_cfgs));
-
- for (; nr_chips > 0; nr_chips--, chips++)
- s3c_gpiolib_add(chips);
-
- samsung_gpiolib_add_4bit_chips(s5p6440_gpio_4bit,
- ARRAY_SIZE(s5p6440_gpio_4bit));
-
- samsung_gpiolib_add_4bit2_chips(s5p6440_gpio_4bit2,
- ARRAY_SIZE(s5p6440_gpio_4bit2));
-
- s5p64x0_gpio_add_rbank_4bit2(s5p6440_gpio_rbank_4bit2,
- ARRAY_SIZE(s5p6440_gpio_rbank_4bit2));
-
- return 0;
-}
-arch_initcall(s5p6440_gpiolib_init);
diff --git a/arch/arm/mach-s5p64x0/gpiolib.c b/arch/arm/mach-s5p64x0/gpiolib.c
new file mode 100644
index 0000000..e7fb3b0
--- /dev/null
+++ b/arch/arm/mach-s5p64x0/gpiolib.c
@@ -0,0 +1,511 @@
+/* linux/arch/arm/mach-s5p64x0/gpiolib.c
+ *
+ * Copyright (c) 2009-2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * S5P64X0 - GPIOlib support
+ *
+ * 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/kernel.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+#include <linux/gpio.h>
+
+#include <mach/map.h>
+#include <mach/regs-gpio.h>
+#include <mach/regs-clock.h>
+
+#include <plat/gpio-core.h>
+#include <plat/gpio-cfg.h>
+#include <plat/gpio-cfg-helpers.h>
+
+/*
+ * S5P6440 GPIO bank summary:
+ *
+ * Bank GPIOs Style SlpCon ExtInt Group
+ * A 6 4Bit Yes 1
+ * B 7 4Bit Yes 1
+ * C 8 4Bit Yes 2
+ * F 2 2Bit Yes 4 [1]
+ * G 7 4Bit Yes 5
+ * H 10 4Bit[2] Yes 6
+ * I 16 2Bit Yes None
+ * J 12 2Bit Yes None
+ * N 16 2Bit No IRQ_EINT
+ * P 8 2Bit Yes 8
+ * R 15 4Bit[2] Yes 8
+ *
+ * S5P6450 GPIO bank summary:
+ *
+ * Bank GPIOs Style SlpCon ExtInt Group
+ * A 6 4Bit Yes 1
+ * B 7 4Bit Yes 1
+ * C 8 4Bit Yes 2
+ * D 8 4Bit Yes None
+ * F 2 2Bit Yes None
+ * G 14 4Bit[2] Yes 5
+ * H 10 4Bit[2] Yes 6
+ * I 16 2Bit Yes None
+ * J 12 2Bit Yes None
+ * K 5 4Bit Yes None
+ * N 16 2Bit No IRQ_EINT
+ * P 11 2Bit Yes 8
+ * Q 14 2Bit Yes None
+ * R 15 4Bit[2] Yes None
+ * S 8 2Bit Yes None
+ *
+ * [1] BANKF pins 14,15 do not form part of the external interrupt sources
+ * [2] BANK has two control registers, GPxCON0 and GPxCON1
+ */
+
+static int s5p64x0_gpiolib_rbank_4bit2_input(struct gpio_chip *chip,
+ unsigned int offset)
+{
+ struct s3c_gpio_chip *ourchip = to_s3c_gpio(chip);
+ void __iomem *base = ourchip->base;
+ void __iomem *regcon = base;
+ unsigned long con;
+ unsigned long flags;
+
+ switch (offset) {
+ case 6:
+ offset += 1;
+ case 0:
+ case 1:
+ case 2:
+ case 3:
+ case 4:
+ case 5:
+ regcon -= 4;
+ break;
+ default:
+ offset -= 7;
+ break;
+ }
+
+ s3c_gpio_lock(ourchip, flags);
+
+ con = __raw_readl(regcon);
+ con &= ~(0xf << con_4bit_shift(offset));
+ __raw_writel(con, regcon);
+
+ s3c_gpio_unlock(ourchip, flags);
+
+ return 0;
+}
+
+static int s5p64x0_gpiolib_rbank_4bit2_output(struct gpio_chip *chip,
+ unsigned int offset, int value)
+{
+ struct s3c_gpio_chip *ourchip = to_s3c_gpio(chip);
+ void __iomem *base = ourchip->base;
+ void __iomem *regcon = base;
+ unsigned long con;
+ unsigned long dat;
+ unsigned long flags;
+ unsigned con_offset = offset;
+
+ switch (con_offset) {
+ case 6:
+ con_offset += 1;
+ case 0:
+ case 1:
+ case 2:
+ case 3:
+ case 4:
+ case 5:
+ regcon -= 4;
+ break;
+ default:
+ con_offset -= 7;
+ break;
+ }
+
+ s3c_gpio_lock(ourchip, flags);
+
+ con = __raw_readl(regcon);
+ con &= ~(0xf << con_4bit_shift(con_offset));
+ con |= 0x1 << con_4bit_shift(con_offset);
+
+ dat = __raw_readl(base + GPIODAT_OFF);
+ if (value)
+ dat |= 1 << offset;
+ else
+ dat &= ~(1 << offset);
+
+ __raw_writel(con, regcon);
+ __raw_writel(dat, base + GPIODAT_OFF);
+
+ s3c_gpio_unlock(ourchip, flags);
+
+ return 0;
+}
+
+int s5p64x0_gpio_setcfg_4bit_rbank(struct s3c_gpio_chip *chip,
+ unsigned int off, unsigned int cfg)
+{
+ void __iomem *reg = chip->base;
+ unsigned int shift;
+ u32 con;
+
+ switch (off) {
+ case 0:
+ case 1:
+ case 2:
+ case 3:
+ case 4:
+ case 5:
+ shift = (off & 7) * 4;
+ reg -= 4;
+ break;
+ case 6:
+ shift = ((off + 1) & 7) * 4;
+ reg -= 4;
+ default:
+ shift = ((off + 1) & 7) * 4;
+ break;
+ }
+
+ if (s3c_gpio_is_cfg_special(cfg)) {
+ cfg &= 0xf;
+ cfg <<= shift;
+ }
+
+ con = __raw_readl(reg);
+ con &= ~(0xf << shift);
+ con |= cfg;
+ __raw_writel(con, reg);
+
+ return 0;
+}
+
+static struct s3c_gpio_cfg s5p64x0_gpio_cfgs[] = {
+ {
+ .cfg_eint = 0,
+ }, {
+ .cfg_eint = 7,
+ }, {
+ .cfg_eint = 3,
+ .set_config = s5p64x0_gpio_setcfg_4bit_rbank,
+ }, {
+ .cfg_eint = 0,
+ .set_config = s3c_gpio_setcfg_s3c24xx,
+ .get_config = s3c_gpio_getcfg_s3c24xx,
+ }, {
+ .cfg_eint = 2,
+ .set_config = s3c_gpio_setcfg_s3c24xx,
+ .get_config = s3c_gpio_getcfg_s3c24xx,
+ }, {
+ .cfg_eint = 3,
+ .set_config = s3c_gpio_setcfg_s3c24xx,
+ .get_config = s3c_gpio_getcfg_s3c24xx,
+ },
+};
+
+static struct s3c_gpio_chip s5p6440_gpio_4bit[] = {
+ {
+ .base = S5P64X0_GPA_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6440_GPA(0),
+ .ngpio = S5P6440_GPIO_A_NR,
+ .label = "GPA",
+ },
+ }, {
+ .base = S5P64X0_GPB_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6440_GPB(0),
+ .ngpio = S5P6440_GPIO_B_NR,
+ .label = "GPB",
+ },
+ }, {
+ .base = S5P64X0_GPC_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6440_GPC(0),
+ .ngpio = S5P6440_GPIO_C_NR,
+ .label = "GPC",
+ },
+ }, {
+ .base = S5P64X0_GPG_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6440_GPG(0),
+ .ngpio = S5P6440_GPIO_G_NR,
+ .label = "GPG",
+ },
+ },
+};
+
+static struct s3c_gpio_chip s5p6440_gpio_4bit2[] = {
+ {
+ .base = S5P64X0_GPH_BASE + 0x4,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6440_GPH(0),
+ .ngpio = S5P6440_GPIO_H_NR,
+ .label = "GPH",
+ },
+ },
+};
+
+static struct s3c_gpio_chip s5p6440_gpio_rbank_4bit2[] = {
+ {
+ .base = S5P64X0_GPR_BASE + 0x4,
+ .config = &s5p64x0_gpio_cfgs[2],
+ .chip = {
+ .base = S5P6440_GPR(0),
+ .ngpio = S5P6440_GPIO_R_NR,
+ .label = "GPR",
+ },
+ },
+};
+
+static struct s3c_gpio_chip s5p6440_gpio_2bit[] = {
+ {
+ .base = S5P64X0_GPF_BASE,
+ .config = &s5p64x0_gpio_cfgs[5],
+ .chip = {
+ .base = S5P6440_GPF(0),
+ .ngpio = S5P6440_GPIO_F_NR,
+ .label = "GPF",
+ },
+ }, {
+ .base = S5P64X0_GPI_BASE,
+ .config = &s5p64x0_gpio_cfgs[3],
+ .chip = {
+ .base = S5P6440_GPI(0),
+ .ngpio = S5P6440_GPIO_I_NR,
+ .label = "GPI",
+ },
+ }, {
+ .base = S5P64X0_GPJ_BASE,
+ .config = &s5p64x0_gpio_cfgs[3],
+ .chip = {
+ .base = S5P6440_GPJ(0),
+ .ngpio = S5P6440_GPIO_J_NR,
+ .label = "GPJ",
+ },
+ }, {
+ .base = S5P64X0_GPN_BASE,
+ .config = &s5p64x0_gpio_cfgs[4],
+ .chip = {
+ .base = S5P6440_GPN(0),
+ .ngpio = S5P6440_GPIO_N_NR,
+ .label = "GPN",
+ },
+ }, {
+ .base = S5P64X0_GPP_BASE,
+ .config = &s5p64x0_gpio_cfgs[5],
+ .chip = {
+ .base = S5P6440_GPP(0),
+ .ngpio = S5P6440_GPIO_P_NR,
+ .label = "GPP",
+ },
+ },
+};
+
+static struct s3c_gpio_chip s5p6450_gpio_4bit[] = {
+ {
+ .base = S5P64X0_GPA_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6450_GPA(0),
+ .ngpio = S5P6450_GPIO_A_NR,
+ .label = "GPA",
+ },
+ }, {
+ .base = S5P64X0_GPB_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6450_GPB(0),
+ .ngpio = S5P6450_GPIO_B_NR,
+ .label = "GPB",
+ },
+ }, {
+ .base = S5P64X0_GPC_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6450_GPC(0),
+ .ngpio = S5P6450_GPIO_C_NR,
+ .label = "GPC",
+ },
+ }, {
+ .base = S5P6450_GPD_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6450_GPD(0),
+ .ngpio = S5P6450_GPIO_D_NR,
+ .label = "GPD",
+ },
+ }, {
+ .base = S5P6450_GPK_BASE,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6450_GPK(0),
+ .ngpio = S5P6450_GPIO_K_NR,
+ .label = "GPK",
+ },
+ },
+};
+
+static struct s3c_gpio_chip s5p6450_gpio_4bit2[] = {
+ {
+ .base = S5P64X0_GPG_BASE + 0x4,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6450_GPG(0),
+ .ngpio = S5P6450_GPIO_G_NR,
+ .label = "GPG",
+ },
+ }, {
+ .base = S5P64X0_GPH_BASE + 0x4,
+ .config = &s5p64x0_gpio_cfgs[1],
+ .chip = {
+ .base = S5P6450_GPH(0),
+ .ngpio = S5P6450_GPIO_H_NR,
+ .label = "GPH",
+ },
+ },
+};
+
+static struct s3c_gpio_chip s5p6450_gpio_rbank_4bit2[] = {
+ {
+ .base = S5P64X0_GPR_BASE + 0x4,
+ .config = &s5p64x0_gpio_cfgs[2],
+ .chip = {
+ .base = S5P6450_GPR(0),
+ .ngpio = S5P6450_GPIO_R_NR,
+ .label = "GPR",
+ },
+ },
+};
+
+static struct s3c_gpio_chip s5p6450_gpio_2bit[] = {
+ {
+ .base = S5P64X0_GPF_BASE,
+ .config = &s5p64x0_gpio_cfgs[5],
+ .chip = {
+ .base = S5P6450_GPF(0),
+ .ngpio = S5P6450_GPIO_F_NR,
+ .label = "GPF",
+ },
+ }, {
+ .base = S5P64X0_GPI_BASE,
+ .config = &s5p64x0_gpio_cfgs[3],
+ .chip = {
+ .base = S5P6450_GPI(0),
+ .ngpio = S5P6450_GPIO_I_NR,
+ .label = "GPI",
+ },
+ }, {
+ .base = S5P64X0_GPJ_BASE,
+ .config = &s5p64x0_gpio_cfgs[3],
+ .chip = {
+ .base = S5P6450_GPJ(0),
+ .ngpio = S5P6450_GPIO_J_NR,
+ .label = "GPJ",
+ },
+ }, {
+ .base = S5P64X0_GPN_BASE,
+ .config = &s5p64x0_gpio_cfgs[4],
+ .chip = {
+ .base = S5P6450_GPN(0),
+ .ngpio = S5P6450_GPIO_N_NR,
+ .label = "GPN",
+ },
+ }, {
+ .base = S5P64X0_GPP_BASE,
+ .config = &s5p64x0_gpio_cfgs[5],
+ .chip = {
+ .base = S5P6450_GPP(0),
+ .ngpio = S5P6450_GPIO_P_NR,
+ .label = "GPP",
+ },
+ }, {
+ .base = S5P6450_GPQ_BASE,
+ .config = &s5p64x0_gpio_cfgs[4],
+ .chip = {
+ .base = S5P6450_GPQ(0),
+ .ngpio = S5P6450_GPIO_Q_NR,
+ .label = "GPQ",
+ },
+ }, {
+ .base = S5P6450_GPS_BASE,
+ .config = &s5p64x0_gpio_cfgs[5],
+ .chip = {
+ .base = S5P6450_GPS(0),
+ .ngpio = S5P6450_GPIO_S_NR,
+ .label = "GPS",
+ },
+ },
+};
+
+void __init s5p64x0_gpiolib_set_cfg(struct s3c_gpio_cfg *chipcfg, int nr_chips)
+{
+ for (; nr_chips > 0; nr_chips--, chipcfg++) {
+ if (!chipcfg->set_config)
+ chipcfg->set_config = s3c_gpio_setcfg_s3c64xx_4bit;
+ if (!chipcfg->get_config)
+ chipcfg->get_config = s3c_gpio_getcfg_s3c64xx_4bit;
+ if (!chipcfg->set_pull)
+ chipcfg->set_pull = s3c_gpio_setpull_updown;
+ if (!chipcfg->get_pull)
+ chipcfg->get_pull = s3c_gpio_getpull_updown;
+ }
+}
+
+static void __init s5p64x0_gpio_add_rbank_4bit2(struct s3c_gpio_chip *chip,
+ int nr_chips)
+{
+ for (; nr_chips > 0; nr_chips--, chip++) {
+ chip->chip.direction_input = s5p64x0_gpiolib_rbank_4bit2_input;
+ chip->chip.direction_output =
+ s5p64x0_gpiolib_rbank_4bit2_output;
+ s3c_gpiolib_add(chip);
+ }
+}
+
+static int __init s5p64x0_gpiolib_init(void)
+{
+ unsigned int chipid;
+
+ chipid = __raw_readl(S5P64X0_SYS_ID);
+
+ s5p64x0_gpiolib_set_cfg(s5p64x0_gpio_cfgs,
+ ARRAY_SIZE(s5p64x0_gpio_cfgs));
+
+ if ((chipid & 0xff000) == 0x50000) {
+ samsung_gpiolib_add_2bit_chips(s5p6450_gpio_2bit,
+ ARRAY_SIZE(s5p6450_gpio_2bit));
+
+ samsung_gpiolib_add_4bit_chips(s5p6450_gpio_4bit,
+ ARRAY_SIZE(s5p6450_gpio_4bit));
+
+ samsung_gpiolib_add_4bit2_chips(s5p6450_gpio_4bit2,
+ ARRAY_SIZE(s5p6450_gpio_4bit2));
+
+ s5p64x0_gpio_add_rbank_4bit2(s5p6450_gpio_rbank_4bit2,
+ ARRAY_SIZE(s5p6450_gpio_rbank_4bit2));
+ } else {
+ samsung_gpiolib_add_2bit_chips(s5p6440_gpio_2bit,
+ ARRAY_SIZE(s5p6440_gpio_2bit));
+
+ samsung_gpiolib_add_4bit_chips(s5p6440_gpio_4bit,
+ ARRAY_SIZE(s5p6440_gpio_4bit));
+
+ samsung_gpiolib_add_4bit2_chips(s5p6440_gpio_4bit2,
+ ARRAY_SIZE(s5p6440_gpio_4bit2));
+
+ s5p64x0_gpio_add_rbank_4bit2(s5p6440_gpio_rbank_4bit2,
+ ARRAY_SIZE(s5p6440_gpio_rbank_4bit2));
+ }
+
+ return 0;
+}
+core_initcall(s5p64x0_gpiolib_init);
diff --git a/arch/arm/mach-s5p64x0/include/mach/map.h b/arch/arm/mach-s5p64x0/include/mach/map.h
index 31e5341..a9365e5 100644
--- a/arch/arm/mach-s5p64x0/include/mach/map.h
+++ b/arch/arm/mach-s5p64x0/include/mach/map.h
@@ -29,6 +29,9 @@
#define S5P64X0_PA_VIC0 (0xE4000000)
#define S5P64X0_PA_VIC1 (0xE4100000)
+#define S5P64X0_PA_SROMC (0xE7000000)
+#define S5P_PA_SROMC S5P64X0_PA_SROMC
+
#define S5P64X0_PA_PDMA (0xE9000000)
#define S5P64X0_PA_TIMER (0xEA000000)
@@ -63,6 +66,8 @@
#define S5P64X0_PA_HSMMC(x) (0xED800000 + ((x) * 0x100000))
#define S5P64X0_PA_I2S (0xF2000000)
+#define S5P6450_PA_I2S1 0xF2800000
+#define S5P6450_PA_I2S2 0xF2900000
#define S5P64X0_PA_PCM (0xF2100000)
diff --git a/arch/arm/mach-s5p64x0/include/mach/regs-gpio.h b/arch/arm/mach-s5p64x0/include/mach/regs-gpio.h
index 85f448e..0953ef6 100644
--- a/arch/arm/mach-s5p64x0/include/mach/regs-gpio.h
+++ b/arch/arm/mach-s5p64x0/include/mach/regs-gpio.h
@@ -15,48 +15,23 @@
#include <mach/map.h>
-/* Will be implemented S5P6442 GPIOlib */
-
/* Base addresses for each of the banks */
-#define S5P6440_GPA_BASE (S5P_VA_GPIO + 0x0000)
-#define S5P6440_GPB_BASE (S5P_VA_GPIO + 0x0020)
-#define S5P6440_GPC_BASE (S5P_VA_GPIO + 0x0040)
-#define S5P6440_GPF_BASE (S5P_VA_GPIO + 0x00A0)
-#define S5P6440_GPG_BASE (S5P_VA_GPIO + 0x00C0)
-#define S5P6440_GPH_BASE (S5P_VA_GPIO + 0x00E0)
-#define S5P6440_GPI_BASE (S5P_VA_GPIO + 0x0100)
-#define S5P6440_GPJ_BASE (S5P_VA_GPIO + 0x0120)
-#define S5P6440_GPN_BASE (S5P_VA_GPIO + 0x0830)
-#define S5P6440_GPP_BASE (S5P_VA_GPIO + 0x0160)
-#define S5P6440_GPR_BASE (S5P_VA_GPIO + 0x0290)
+#define S5P64X0_GPA_BASE (S5P_VA_GPIO + 0x0000)
+#define S5P64X0_GPB_BASE (S5P_VA_GPIO + 0x0020)
+#define S5P64X0_GPC_BASE (S5P_VA_GPIO + 0x0040)
+#define S5P64X0_GPF_BASE (S5P_VA_GPIO + 0x00A0)
+#define S5P64X0_GPG_BASE (S5P_VA_GPIO + 0x00C0)
+#define S5P64X0_GPH_BASE (S5P_VA_GPIO + 0x00E0)
+#define S5P64X0_GPI_BASE (S5P_VA_GPIO + 0x0100)
+#define S5P64X0_GPJ_BASE (S5P_VA_GPIO + 0x0120)
+#define S5P64X0_GPN_BASE (S5P_VA_GPIO + 0x0830)
+#define S5P64X0_GPP_BASE (S5P_VA_GPIO + 0x0160)
+#define S5P64X0_GPR_BASE (S5P_VA_GPIO + 0x0290)
-#define S5P6440_EINT0CON0 (S5P_VA_GPIO + 0x900)
-#define S5P6440_EINT0FLTCON0 (S5P_VA_GPIO + 0x910)
-#define S5P6440_EINT0FLTCON1 (S5P_VA_GPIO + 0x914)
-#define S5P6440_EINT0MASK (S5P_VA_GPIO + 0x920)
-#define S5P6440_EINT0PEND (S5P_VA_GPIO + 0x924)
-
-/* for LCD */
-
-#define S5P6440_SPCON_LCD_SEL_RGB (1 << 0)
-#define S5P6440_SPCON_LCD_SEL_MASK (3 << 0)
-
-/*
- * These set of macros are not really useful for the
- * GPF/GPI/GPJ/GPN/GPP, useful for others set of GPIO's (4 bit)
- */
-
-#define S5P6440_GPIO_CONMASK(__gpio) (0xf << ((__gpio) * 4))
-#define S5P6440_GPIO_INPUT(__gpio) (0x0 << ((__gpio) * 4))
-#define S5P6440_GPIO_OUTPUT(__gpio) (0x1 << ((__gpio) * 4))
-
-/*
- * Use these macros for GPF/GPI/GPJ/GPN/GPP set of GPIO (2 bit)
- */
-
-#define S5P6440_GPIO2_CONMASK(__gpio) (0x3 << ((__gpio) * 2))
-#define S5P6440_GPIO2_INPUT(__gpio) (0x0 << ((__gpio) * 2))
-#define S5P6440_GPIO2_OUTPUT(__gpio) (0x1 << ((__gpio) * 2))
+#define S5P6450_GPD_BASE (S5P_VA_GPIO + 0x0060)
+#define S5P6450_GPK_BASE (S5P_VA_GPIO + 0x0140)
+#define S5P6450_GPQ_BASE (S5P_VA_GPIO + 0x0180)
+#define S5P6450_GPS_BASE (S5P_VA_GPIO + 0x0300)
#endif /* __ASM_ARCH_REGS_GPIO_H */
diff --git a/arch/arm/mach-s5p64x0/mach-smdk6440.c b/arch/arm/mach-s5p64x0/mach-smdk6440.c
index 87c3f03..e980275 100644
--- a/arch/arm/mach-s5p64x0/mach-smdk6440.c
+++ b/arch/arm/mach-s5p64x0/mach-smdk6440.c
@@ -117,6 +117,7 @@
static struct i2c_board_info smdk6440_i2c_devs0[] __initdata = {
{ I2C_BOARD_INFO("24c08", 0x50), },
+ { I2C_BOARD_INFO("wm8580", 0x1b), },
};
static struct i2c_board_info smdk6440_i2c_devs1[] __initdata = {
diff --git a/arch/arm/mach-s5p64x0/mach-smdk6450.c b/arch/arm/mach-s5p64x0/mach-smdk6450.c
index d609f5a..b78f562 100644
--- a/arch/arm/mach-s5p64x0/mach-smdk6450.c
+++ b/arch/arm/mach-s5p64x0/mach-smdk6450.c
@@ -135,6 +135,7 @@
};
static struct i2c_board_info smdk6450_i2c_devs0[] __initdata = {
+ { I2C_BOARD_INFO("wm8580", 0x1b), },
{ I2C_BOARD_INFO("24c08", 0x50), }, /* Samsung KS24C080C EEPROM */
};
diff --git a/arch/arm/mach-s5pc100/clock.c b/arch/arm/mach-s5pc100/clock.c
index 2d4a761..0305e9b 100644
--- a/arch/arm/mach-s5pc100/clock.c
+++ b/arch/arm/mach-s5pc100/clock.c
@@ -396,7 +396,7 @@
* recommended to keep the following clocks disabled until the driver requests
* for enabling the clock.
*/
-static struct clk init_clocks_disable[] = {
+static struct clk init_clocks_off[] = {
{
.name = "cssys",
.id = -1,
@@ -1381,8 +1381,6 @@
void __init s5pc100_register_clocks(void)
{
- struct clk *clkp;
- int ret;
int ptr;
s3c24xx_register_clocks(clks, ARRAY_SIZE(clks));
@@ -1393,16 +1391,8 @@
s3c_register_clksrc(clksrcs, ARRAY_SIZE(clksrcs));
s3c_register_clocks(init_clocks, ARRAY_SIZE(init_clocks));
- clkp = init_clocks_disable;
- for (ptr = 0; ptr < ARRAY_SIZE(init_clocks_disable); ptr++, clkp++) {
-
- ret = s3c24xx_register_clock(clkp);
- if (ret < 0) {
- printk(KERN_ERR "Failed to register clock %s (%d)\n",
- clkp->name, ret);
- }
- (clkp->enable)(clkp, 0);
- }
+ s3c_register_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
+ s3c_disable_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
s3c_pwmclk_init();
}
diff --git a/arch/arm/mach-s5pc100/include/mach/map.h b/arch/arm/mach-s5pc100/include/mach/map.h
index 32e9cab..328467b 100644
--- a/arch/arm/mach-s5pc100/include/mach/map.h
+++ b/arch/arm/mach-s5pc100/include/mach/map.h
@@ -55,6 +55,8 @@
#define S5PC100_VA_VIC_OFFSET 0x10000
#define S5PC1XX_VA_VIC(x) (S5PC100_VA_VIC + ((x) * S5PC100_VA_VIC_OFFSET))
+#define S5PC100_PA_SROMC (0xE7000000)
+#define S5P_PA_SROMC S5PC100_PA_SROMC
#define S5PC100_PA_ONENAND (0xE7100000)
diff --git a/arch/arm/mach-s5pv210/Kconfig b/arch/arm/mach-s5pv210/Kconfig
index 862f239..53aabef 100644
--- a/arch/arm/mach-s5pv210/Kconfig
+++ b/arch/arm/mach-s5pv210/Kconfig
@@ -118,6 +118,7 @@
config MACH_SMDKV210
bool "SMDKV210"
select CPU_S5PV210
+ select S3C_DEV_FB
select S3C_DEV_HSMMC
select S3C_DEV_HSMMC1
select S3C_DEV_HSMMC2
@@ -130,6 +131,7 @@
select SAMSUNG_DEV_IDE
select SAMSUNG_DEV_KEYPAD
select SAMSUNG_DEV_TS
+ select S5PV210_SETUP_FB_24BPP
select S5PV210_SETUP_I2C1
select S5PV210_SETUP_I2C2
select S5PV210_SETUP_IDE
diff --git a/arch/arm/mach-s5pv210/clock.c b/arch/arm/mach-s5pv210/clock.c
index b774ff1..2d59949 100644
--- a/arch/arm/mach-s5pv210/clock.c
+++ b/arch/arm/mach-s5pv210/clock.c
@@ -309,7 +309,7 @@
.get_rate = s5pv210_clk_fout_apll_get_rate,
};
-static struct clk init_clocks_disable[] = {
+static struct clk init_clocks_off[] = {
{
.name = "pdma",
.id = 0,
@@ -525,6 +525,12 @@
.parent = &clk_pclk_psys.clk,
.enable = s5pv210_clk_ip3_ctrl,
.ctrlbit = (1 << 20),
+ }, {
+ .name = "sromc",
+ .id = -1,
+ .parent = &clk_hclk_psys.clk,
+ .enable = s5pv210_clk_ip1_ctrl,
+ .ctrlbit = (1 << 26),
},
};
@@ -1220,13 +1226,9 @@
void __init s5pv210_register_clocks(void)
{
- struct clk *clkp;
- int ret;
int ptr;
- ret = s3c24xx_register_clocks(clks, ARRAY_SIZE(clks));
- if (ret > 0)
- printk(KERN_ERR "Failed to register %u clocks\n", ret);
+ s3c24xx_register_clocks(clks, ARRAY_SIZE(clks));
for (ptr = 0; ptr < ARRAY_SIZE(sysclks); ptr++)
s3c_register_clksrc(sysclks[ptr], 1);
@@ -1234,15 +1236,8 @@
s3c_register_clksrc(clksrcs, ARRAY_SIZE(clksrcs));
s3c_register_clocks(init_clocks, ARRAY_SIZE(init_clocks));
- clkp = init_clocks_disable;
- for (ptr = 0; ptr < ARRAY_SIZE(init_clocks_disable); ptr++, clkp++) {
- ret = s3c24xx_register_clock(clkp);
- if (ret < 0) {
- printk(KERN_ERR "Failed to register clock %s (%d)\n",
- clkp->name, ret);
- }
- (clkp->enable)(clkp, 0);
- }
+ s3c_register_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
+ s3c_disable_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
s3c_pwmclk_init();
}
diff --git a/arch/arm/mach-s5pv210/cpu.c b/arch/arm/mach-s5pv210/cpu.c
index 8eb480e..61e6c24 100644
--- a/arch/arm/mach-s5pv210/cpu.c
+++ b/arch/arm/mach-s5pv210/cpu.c
@@ -81,11 +81,6 @@
.length = SZ_512K,
.type = MT_DEVICE,
}, {
- .virtual = (unsigned long)S5P_VA_SROMC,
- .pfn = __phys_to_pfn(S5PV210_PA_SROMC),
- .length = SZ_4K,
- .type = MT_DEVICE,
- }, {
.virtual = (unsigned long)S5P_VA_DMC0,
.pfn = __phys_to_pfn(S5PV210_PA_DMC0),
.length = SZ_4K,
diff --git a/arch/arm/mach-s5pv210/include/mach/irqs.h b/arch/arm/mach-s5pv210/include/mach/irqs.h
index 119b95f..26710b3 100644
--- a/arch/arm/mach-s5pv210/include/mach/irqs.h
+++ b/arch/arm/mach-s5pv210/include/mach/irqs.h
@@ -65,7 +65,7 @@
#define IRQ_HSMMC0 S5P_IRQ_VIC1(26)
#define IRQ_HSMMC1 S5P_IRQ_VIC1(27)
#define IRQ_HSMMC2 S5P_IRQ_VIC1(28)
-#define IRQ_MIPICSI S5P_IRQ_VIC1(29)
+#define IRQ_MIPI_CSIS S5P_IRQ_VIC1(29)
#define IRQ_MIPIDSI S5P_IRQ_VIC1(30)
#define IRQ_ONENAND_AUDI S5P_IRQ_VIC1(31)
@@ -132,5 +132,6 @@
#define IRQ_LCD_FIFO IRQ_LCD0
#define IRQ_LCD_VSYNC IRQ_LCD1
#define IRQ_LCD_SYSTEM IRQ_LCD2
+#define IRQ_MIPI_CSIS0 IRQ_MIPI_CSIS
#endif /* ASM_ARCH_IRQS_H */
diff --git a/arch/arm/mach-s5pv210/include/mach/map.h b/arch/arm/mach-s5pv210/include/mach/map.h
index 861d7fe..3611492 100644
--- a/arch/arm/mach-s5pv210/include/mach/map.h
+++ b/arch/arm/mach-s5pv210/include/mach/map.h
@@ -16,6 +16,8 @@
#include <plat/map-base.h>
#include <plat/map-s5p.h>
+#define S5PV210_PA_SROM_BANK5 (0xA8000000)
+
#define S5PC110_PA_ONENAND (0xB0000000)
#define S5P_PA_ONENAND S5PC110_PA_ONENAND
@@ -60,6 +62,7 @@
#define S3C_VA_UARTx(x) (S3C_VA_UART + ((x) * S3C_UART_OFFSET))
#define S5PV210_PA_SROMC (0xE8000000)
+#define S5P_PA_SROMC S5PV210_PA_SROMC
#define S5PV210_PA_CFCON (0xE8200000)
@@ -107,6 +110,8 @@
#define S5PV210_PA_DMC0 (0xF0000000)
#define S5PV210_PA_DMC1 (0xF1400000)
+#define S5PV210_PA_MIPI_CSIS 0xFA600000
+
/* compatibiltiy defines. */
#define S3C_PA_UART S5PV210_PA_UART
#define S3C_PA_HSMMC0 S5PV210_PA_HSMMC(0)
@@ -123,6 +128,7 @@
#define S5P_PA_FIMC0 S5PV210_PA_FIMC0
#define S5P_PA_FIMC1 S5PV210_PA_FIMC1
#define S5P_PA_FIMC2 S5PV210_PA_FIMC2
+#define S5P_PA_MIPI_CSIS0 S5PV210_PA_MIPI_CSIS
#define SAMSUNG_PA_ADC S5PV210_PA_ADC
#define SAMSUNG_PA_CFCON S5PV210_PA_CFCON
diff --git a/arch/arm/mach-s5pv210/include/mach/regs-clock.h b/arch/arm/mach-s5pv210/include/mach/regs-clock.h
index ebaabe0..4c45b74 100644
--- a/arch/arm/mach-s5pv210/include/mach/regs-clock.h
+++ b/arch/arm/mach-s5pv210/include/mach/regs-clock.h
@@ -161,7 +161,7 @@
#define S5P_MDNIE_SEL S5P_CLKREG(0x7008)
#define S5P_MIPI_PHY_CON0 S5P_CLKREG(0x7200)
#define S5P_MIPI_PHY_CON1 S5P_CLKREG(0x7204)
-#define S5P_MIPI_CONTROL S5P_CLKREG(0xE814)
+#define S5P_MIPI_DPHY_CONTROL S5P_CLKREG(0xE814)
#define S5P_IDLE_CFG_TL_MASK (3 << 30)
#define S5P_IDLE_CFG_TM_MASK (3 << 28)
@@ -195,9 +195,6 @@
#define S5P_OTHERS_RET_UART (1 << 28)
#define S5P_OTHERS_USB_SIG_MASK (1 << 16)
-/* MIPI */
-#define S5P_MIPI_DPHY_EN (3)
-
/* S5P_DAC_CONTROL */
#define S5P_DAC_ENABLE (1)
#define S5P_DAC_DISABLE (0)
diff --git a/arch/arm/mach-s5pv210/mach-smdkc110.c b/arch/arm/mach-s5pv210/mach-smdkc110.c
index 5dd1681..bb20a14 100644
--- a/arch/arm/mach-s5pv210/mach-smdkc110.c
+++ b/arch/arm/mach-s5pv210/mach-smdkc110.c
@@ -94,6 +94,7 @@
static struct i2c_board_info smdkc110_i2c_devs0[] __initdata = {
{ I2C_BOARD_INFO("24c08", 0x50), }, /* Samsung S524AD0XD1 */
+ { I2C_BOARD_INFO("wm8580", 0x1b), },
};
static struct i2c_board_info smdkc110_i2c_devs1[] __initdata = {
diff --git a/arch/arm/mach-s5pv210/mach-smdkv210.c b/arch/arm/mach-s5pv210/mach-smdkv210.c
index 1fbc45b..88e4522 100644
--- a/arch/arm/mach-s5pv210/mach-smdkv210.c
+++ b/arch/arm/mach-s5pv210/mach-smdkv210.c
@@ -14,16 +14,25 @@
#include <linux/init.h>
#include <linux/serial_core.h>
#include <linux/sysdev.h>
+#include <linux/dm9000.h>
+#include <linux/fb.h>
+#include <linux/gpio.h>
+#include <linux/delay.h>
#include <asm/mach/arch.h>
#include <asm/mach/map.h>
#include <asm/setup.h>
#include <asm/mach-types.h>
+#include <video/platform_lcd.h>
+
#include <mach/map.h>
#include <mach/regs-clock.h>
+#include <mach/regs-fb.h>
#include <plat/regs-serial.h>
+#include <plat/regs-srom.h>
+#include <plat/gpio-cfg.h>
#include <plat/s5pv210.h>
#include <plat/devs.h>
#include <plat/cpu.h>
@@ -33,6 +42,7 @@
#include <plat/iic.h>
#include <plat/keypad.h>
#include <plat/pm.h>
+#include <plat/fb.h>
/* Following are default values for UCON, ULCON and UFCON UART registers */
#define SMDKV210_UCON_DEFAULT (S3C2410_UCON_TXILEVEL | \
@@ -102,12 +112,106 @@
.cols = 8,
};
+static struct resource smdkv210_dm9000_resources[] = {
+ [0] = {
+ .start = S5PV210_PA_SROM_BANK5,
+ .end = S5PV210_PA_SROM_BANK5,
+ .flags = IORESOURCE_MEM,
+ },
+ [1] = {
+ .start = S5PV210_PA_SROM_BANK5 + 2,
+ .end = S5PV210_PA_SROM_BANK5 + 2,
+ .flags = IORESOURCE_MEM,
+ },
+ [2] = {
+ .start = IRQ_EINT(9),
+ .end = IRQ_EINT(9),
+ .flags = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
+ },
+};
+
+static struct dm9000_plat_data smdkv210_dm9000_platdata = {
+ .flags = DM9000_PLATF_16BITONLY | DM9000_PLATF_NO_EEPROM,
+ .dev_addr = { 0x00, 0x09, 0xc0, 0xff, 0xec, 0x48 },
+};
+
+struct platform_device smdkv210_dm9000 = {
+ .name = "dm9000",
+ .id = -1,
+ .num_resources = ARRAY_SIZE(smdkv210_dm9000_resources),
+ .resource = smdkv210_dm9000_resources,
+ .dev = {
+ .platform_data = &smdkv210_dm9000_platdata,
+ },
+};
+
+static void smdkv210_lte480wv_set_power(struct plat_lcd_data *pd,
+ unsigned int power)
+{
+ if (power) {
+#if !defined(CONFIG_BACKLIGHT_PWM)
+ gpio_request(S5PV210_GPD0(3), "GPD0");
+ gpio_direction_output(S5PV210_GPD0(3), 1);
+ gpio_free(S5PV210_GPD0(3));
+#endif
+
+ /* fire nRESET on power up */
+ gpio_request(S5PV210_GPH0(6), "GPH0");
+
+ gpio_direction_output(S5PV210_GPH0(6), 1);
+
+ gpio_set_value(S5PV210_GPH0(6), 0);
+ mdelay(10);
+
+ gpio_set_value(S5PV210_GPH0(6), 1);
+ mdelay(10);
+
+ gpio_free(S5PV210_GPH0(6));
+ } else {
+#if !defined(CONFIG_BACKLIGHT_PWM)
+ gpio_request(S5PV210_GPD0(3), "GPD0");
+ gpio_direction_output(S5PV210_GPD0(3), 0);
+ gpio_free(S5PV210_GPD0(3));
+#endif
+ }
+}
+
+static struct plat_lcd_data smdkv210_lcd_lte480wv_data = {
+ .set_power = smdkv210_lte480wv_set_power,
+};
+
+static struct platform_device smdkv210_lcd_lte480wv = {
+ .name = "platform-lcd",
+ .dev.parent = &s3c_device_fb.dev,
+ .dev.platform_data = &smdkv210_lcd_lte480wv_data,
+};
+
+static struct s3c_fb_pd_win smdkv210_fb_win0 = {
+ .win_mode = {
+ .left_margin = 13,
+ .right_margin = 8,
+ .upper_margin = 7,
+ .lower_margin = 5,
+ .hsync_len = 3,
+ .vsync_len = 1,
+ .xres = 800,
+ .yres = 480,
+ },
+ .max_bpp = 32,
+ .default_bpp = 24,
+};
+
+static struct s3c_fb_platdata smdkv210_lcd0_pdata __initdata = {
+ .win[0] = &smdkv210_fb_win0,
+ .vidcon0 = VIDCON0_VIDOUT_RGB | VIDCON0_PNRMODE_RGB,
+ .vidcon1 = VIDCON1_INV_HSYNC | VIDCON1_INV_VSYNC,
+ .setup_gpio = s5pv210_fb_gpio_setup_24bpp,
+};
+
static struct platform_device *smdkv210_devices[] __initdata = {
- &s5pv210_device_iis0,
- &s5pv210_device_ac97,
- &s5pv210_device_spdif,
&s3c_device_adc,
&s3c_device_cfcon,
+ &s3c_device_fb,
&s3c_device_hsmmc0,
&s3c_device_hsmmc1,
&s3c_device_hsmmc2,
@@ -115,14 +219,37 @@
&s3c_device_i2c0,
&s3c_device_i2c1,
&s3c_device_i2c2,
- &samsung_device_keypad,
&s3c_device_rtc,
&s3c_device_ts,
&s3c_device_wdt,
+ &s5pv210_device_ac97,
+ &s5pv210_device_iis0,
+ &s5pv210_device_spdif,
+ &samsung_device_keypad,
+ &smdkv210_dm9000,
+ &smdkv210_lcd_lte480wv,
};
+static void __init smdkv210_dm9000_init(void)
+{
+ unsigned int tmp;
+
+ gpio_request(S5PV210_MP01(5), "nCS5");
+ s3c_gpio_cfgpin(S5PV210_MP01(5), S3C_GPIO_SFN(2));
+ gpio_free(S5PV210_MP01(5));
+
+ tmp = (5 << S5P_SROM_BCX__TACC__SHIFT);
+ __raw_writel(tmp, S5P_SROM_BC5);
+
+ tmp = __raw_readl(S5P_SROM_BW);
+ tmp &= (S5P_SROM_BW__CS_MASK << S5P_SROM_BW__NCS5__SHIFT);
+ tmp |= (1 << S5P_SROM_BW__NCS5__SHIFT);
+ __raw_writel(tmp, S5P_SROM_BW);
+}
+
static struct i2c_board_info smdkv210_i2c_devs0[] __initdata = {
{ I2C_BOARD_INFO("24c08", 0x50), }, /* Samsung S524AD0XD1 */
+ { I2C_BOARD_INFO("wm8580", 0x1b), },
};
static struct i2c_board_info smdkv210_i2c_devs1[] __initdata = {
@@ -150,6 +277,8 @@
{
s3c_pm_init();
+ smdkv210_dm9000_init();
+
samsung_keypad_set_platdata(&smdkv210_keypad_data);
s3c24xx_ts_set_platdata(&s3c_ts_platform);
@@ -165,6 +294,8 @@
s3c_ide_set_platdata(&smdkv210_ide_pdata);
+ s3c_fb_set_platdata(&smdkv210_lcd0_pdata);
+
platform_add_devices(smdkv210_devices, ARRAY_SIZE(smdkv210_devices));
}
diff --git a/arch/arm/mach-s5pv310/Kconfig b/arch/arm/mach-s5pv310/Kconfig
index d64efe0..09c4c21 100644
--- a/arch/arm/mach-s5pv310/Kconfig
+++ b/arch/arm/mach-s5pv310/Kconfig
@@ -15,6 +15,11 @@
help
Enable S5PV310 CPU support
+config S5PV310_DEV_PD
+ bool
+ help
+ Compile in platform device definitions for Power Domain
+
config S5PV310_SETUP_I2C1
bool
help
@@ -61,6 +66,11 @@
help
Common setup code for SDHCI gpio.
+config S5PV310_DEV_SYSMMU
+ bool
+ help
+ Common setup code for SYSTEM MMU in S5PV310
+
# machine support
menu "S5PC210 Machines"
@@ -70,11 +80,15 @@
select CPU_S5PV310
select S3C_DEV_RTC
select S3C_DEV_WDT
+ select S3C_DEV_I2C1
select S3C_DEV_HSMMC
select S3C_DEV_HSMMC1
select S3C_DEV_HSMMC2
select S3C_DEV_HSMMC3
+ select S5PV310_DEV_PD
+ select S5PV310_SETUP_I2C1
select S5PV310_SETUP_SDHCI
+ select S5PV310_DEV_SYSMMU
help
Machine support for Samsung SMDKC210
S5PC210(MCP) is one of package option of S5PV310
@@ -83,6 +97,10 @@
bool "Mobile UNIVERSAL_C210 Board"
select CPU_S5PV310
select S5P_DEV_ONENAND
+ select S3C_DEV_HSMMC
+ select S3C_DEV_HSMMC2
+ select S3C_DEV_HSMMC3
+ select S5PV310_SETUP_SDHCI
select S3C_DEV_I2C1
select S5PV310_SETUP_I2C1
help
@@ -98,10 +116,13 @@
select CPU_S5PV310
select S3C_DEV_RTC
select S3C_DEV_WDT
+ select S3C_DEV_I2C1
select S3C_DEV_HSMMC
select S3C_DEV_HSMMC1
select S3C_DEV_HSMMC2
select S3C_DEV_HSMMC3
+ select S5PV310_DEV_PD
+ select S5PV310_SETUP_I2C1
select S5PV310_SETUP_SDHCI
help
Machine support for Samsung SMDKV310
diff --git a/arch/arm/mach-s5pv310/Makefile b/arch/arm/mach-s5pv310/Makefile
index 61e3cb6..036fb38 100644
--- a/arch/arm/mach-s5pv310/Makefile
+++ b/arch/arm/mach-s5pv310/Makefile
@@ -14,6 +14,7 @@
obj-$(CONFIG_CPU_S5PV310) += cpu.o init.o clock.o irq-combiner.o
obj-$(CONFIG_CPU_S5PV310) += setup-i2c0.o time.o gpiolib.o irq-eint.o dma.o
+obj-$(CONFIG_CPU_FREQ) += cpufreq.o
obj-$(CONFIG_SMP) += platsmp.o headsmp.o
obj-$(CONFIG_LOCAL_TIMERS) += localtimer.o
@@ -27,7 +28,10 @@
# device support
-obj-y += dev-audio.o
+obj-y += dev-audio.o
+obj-$(CONFIG_S5PV310_DEV_PD) += dev-pd.o
+obj-$(CONFIG_S5PV310_DEV_SYSMMU) += dev-sysmmu.o
+
obj-$(CONFIG_S5PV310_SETUP_I2C1) += setup-i2c1.o
obj-$(CONFIG_S5PV310_SETUP_I2C2) += setup-i2c2.o
obj-$(CONFIG_S5PV310_SETUP_I2C3) += setup-i2c3.o
diff --git a/arch/arm/mach-s5pv310/clock.c b/arch/arm/mach-s5pv310/clock.c
index 58c9d33..fc7c2f8 100644
--- a/arch/arm/mach-s5pv310/clock.c
+++ b/arch/arm/mach-s5pv310/clock.c
@@ -244,7 +244,7 @@
.id = -1,
},
.sources = &clkset_mout_corebus,
- .reg_src = { .reg = S5P_CLKSRC_CORE, .shift = 4, .size = 1 },
+ .reg_src = { .reg = S5P_CLKSRC_DMC, .shift = 4, .size = 1 },
};
static struct clksrc_clk clk_sclk_dmc = {
@@ -253,7 +253,7 @@
.id = -1,
.parent = &clk_mout_corebus.clk,
},
- .reg_div = { .reg = S5P_CLKDIV_CORE0, .shift = 12, .size = 3 },
+ .reg_div = { .reg = S5P_CLKDIV_DMC0, .shift = 12, .size = 3 },
};
static struct clksrc_clk clk_aclk_cored = {
@@ -262,7 +262,7 @@
.id = -1,
.parent = &clk_sclk_dmc.clk,
},
- .reg_div = { .reg = S5P_CLKDIV_CORE0, .shift = 16, .size = 3 },
+ .reg_div = { .reg = S5P_CLKDIV_DMC0, .shift = 16, .size = 3 },
};
static struct clksrc_clk clk_aclk_corep = {
@@ -271,7 +271,7 @@
.id = -1,
.parent = &clk_aclk_cored.clk,
},
- .reg_div = { .reg = S5P_CLKDIV_CORE0, .shift = 20, .size = 3 },
+ .reg_div = { .reg = S5P_CLKDIV_DMC0, .shift = 20, .size = 3 },
};
static struct clksrc_clk clk_aclk_acp = {
@@ -280,7 +280,7 @@
.id = -1,
.parent = &clk_mout_corebus.clk,
},
- .reg_div = { .reg = S5P_CLKDIV_CORE0, .shift = 0, .size = 3 },
+ .reg_div = { .reg = S5P_CLKDIV_DMC0, .shift = 0, .size = 3 },
};
static struct clksrc_clk clk_pclk_acp = {
@@ -289,7 +289,7 @@
.id = -1,
.parent = &clk_aclk_acp.clk,
},
- .reg_div = { .reg = S5P_CLKDIV_CORE0, .shift = 4, .size = 3 },
+ .reg_div = { .reg = S5P_CLKDIV_DMC0, .shift = 4, .size = 3 },
};
/* Core list of CMU_TOP side */
@@ -384,7 +384,7 @@
.reg_src = { .reg = S5P_CLKSRC_TOP0, .shift = 8, .size = 1 },
};
-static struct clk init_clocks_disable[] = {
+static struct clk init_clocks_off[] = {
{
.name = "timers",
.id = -1,
@@ -467,6 +467,16 @@
.enable = s5pv310_clk_ip_fsys_ctrl,
.ctrlbit = (1 << 10),
}, {
+ .name = "pdma",
+ .id = 0,
+ .enable = s5pv310_clk_ip_fsys_ctrl,
+ .ctrlbit = (1 << 0),
+ }, {
+ .name = "pdma",
+ .id = 1,
+ .enable = s5pv310_clk_ip_fsys_ctrl,
+ .ctrlbit = (1 << 1),
+ }, {
.name = "adc",
.id = -1,
.enable = s5pv310_clk_ip_peril_ctrl,
@@ -507,6 +517,26 @@
.enable = s5pv310_clk_ip_peril_ctrl,
.ctrlbit = (1 << 18),
}, {
+ .name = "iis",
+ .id = 0,
+ .enable = s5pv310_clk_ip_peril_ctrl,
+ .ctrlbit = (1 << 19),
+ }, {
+ .name = "iis",
+ .id = 1,
+ .enable = s5pv310_clk_ip_peril_ctrl,
+ .ctrlbit = (1 << 20),
+ }, {
+ .name = "iis",
+ .id = 2,
+ .enable = s5pv310_clk_ip_peril_ctrl,
+ .ctrlbit = (1 << 21),
+ }, {
+ .name = "ac97",
+ .id = -1,
+ .enable = s5pv310_clk_ip_peril_ctrl,
+ .ctrlbit = (1 << 27),
+ }, {
.name = "fimg2d",
.id = -1,
.enable = s5pv310_clk_ip_image_ctrl,
@@ -990,6 +1020,17 @@
&clk_dout_mmc4,
};
+static int xtal_rate;
+
+static unsigned long s5pv310_fout_apll_get_rate(struct clk *clk)
+{
+ return s5p_get_pll45xx(xtal_rate, __raw_readl(S5P_APLL_CON0), pll_4508);
+}
+
+static struct clk_ops s5pv310_fout_apll_ops = {
+ .get_rate = s5pv310_fout_apll_get_rate,
+};
+
void __init_or_cpufreq s5pv310_setup_clocks(void)
{
struct clk *xtal_clk;
@@ -1013,6 +1054,9 @@
BUG_ON(IS_ERR(xtal_clk));
xtal = clk_get_rate(xtal_clk);
+
+ xtal_rate = xtal;
+
clk_put(xtal_clk);
printk(KERN_DEBUG "%s: xtal is %ld\n", __func__, xtal);
@@ -1026,7 +1070,7 @@
vpll = s5p_get_pll46xx(vpllsrc, __raw_readl(S5P_VPLL_CON0),
__raw_readl(S5P_VPLL_CON1), pll_4650);
- clk_fout_apll.rate = apll;
+ clk_fout_apll.ops = &s5pv310_fout_apll_ops;
clk_fout_mpll.rate = mpll;
clk_fout_epll.rate = epll;
clk_fout_vpll.rate = vpll;
@@ -1061,13 +1105,9 @@
void __init s5pv310_register_clocks(void)
{
- struct clk *clkp;
- int ret;
int ptr;
- ret = s3c24xx_register_clocks(clks, ARRAY_SIZE(clks));
- if (ret > 0)
- printk(KERN_ERR "Failed to register %u clocks\n", ret);
+ s3c24xx_register_clocks(clks, ARRAY_SIZE(clks));
for (ptr = 0; ptr < ARRAY_SIZE(sysclks); ptr++)
s3c_register_clksrc(sysclks[ptr], 1);
@@ -1075,15 +1115,8 @@
s3c_register_clksrc(clksrcs, ARRAY_SIZE(clksrcs));
s3c_register_clocks(init_clocks, ARRAY_SIZE(init_clocks));
- clkp = init_clocks_disable;
- for (ptr = 0; ptr < ARRAY_SIZE(init_clocks_disable); ptr++, clkp++) {
- ret = s3c24xx_register_clock(clkp);
- if (ret < 0) {
- printk(KERN_ERR "Failed to register clock %s (%d)\n",
- clkp->name, ret);
- }
- (clkp->enable)(clkp, 0);
- }
+ s3c_register_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
+ s3c_disable_clocks(init_clocks_off, ARRAY_SIZE(init_clocks_off));
s3c_pwmclk_init();
}
diff --git a/arch/arm/mach-s5pv310/cpu.c b/arch/arm/mach-s5pv310/cpu.c
index 72ab289..0db0fb6 100644
--- a/arch/arm/mach-s5pv310/cpu.c
+++ b/arch/arm/mach-s5pv310/cpu.c
@@ -41,6 +41,11 @@
.length = SZ_128K,
.type = MT_DEVICE,
}, {
+ .virtual = (unsigned long)S5P_VA_PMU,
+ .pfn = __phys_to_pfn(S5PV310_PA_PMU),
+ .length = SZ_64K,
+ .type = MT_DEVICE,
+ }, {
.virtual = (unsigned long)S5P_VA_COMBINER_BASE,
.pfn = __phys_to_pfn(S5PV310_PA_COMBINER),
.length = SZ_4K,
@@ -71,6 +76,11 @@
.length = SZ_256,
.type = MT_DEVICE,
}, {
+ .virtual = (unsigned long)S5P_VA_DMC0,
+ .pfn = __phys_to_pfn(S5PV310_PA_DMC0),
+ .length = SZ_4K,
+ .type = MT_DEVICE,
+ }, {
.virtual = (unsigned long)S3C_VA_UART,
.pfn = __phys_to_pfn(S3C_PA_UART),
.length = SZ_512K,
@@ -123,6 +133,15 @@
gic_init(0, IRQ_LOCALTIMER, S5P_VA_GIC_DIST, S5P_VA_GIC_CPU);
for (irq = 0; irq < MAX_COMBINER_NR; irq++) {
+
+ /*
+ * From SPI(0) to SPI(39) and SPI(51), SPI(53) are
+ * connected to the interrupt combiner. These irqs
+ * should be initialized to support cascade interrupt.
+ */
+ if ((irq >= 40) && !(irq == 51) && !(irq == 53))
+ continue;
+
combiner_init(irq, (void __iomem *)S5P_VA_COMBINER(irq),
COMBINER_IRQ(irq, 0));
combiner_cascade_irq(irq, IRQ_SPI(irq));
@@ -164,7 +183,7 @@
__raw_writel(L2X0_DYNAMIC_CLK_GATING_EN | L2X0_STNDBY_MODE_EN,
S5P_VA_L2CC + L2X0_POWER_CTRL);
- l2x0_init(S5P_VA_L2CC, 0x7C070001, 0xC200ffff);
+ l2x0_init(S5P_VA_L2CC, 0x7C470001, 0xC200ffff);
return 0;
}
diff --git a/arch/arm/mach-s5pv310/cpufreq.c b/arch/arm/mach-s5pv310/cpufreq.c
new file mode 100644
index 0000000..b04cbc7
--- /dev/null
+++ b/arch/arm/mach-s5pv310/cpufreq.c
@@ -0,0 +1,580 @@
+/* linux/arch/arm/mach-s5pv310/cpufreq.c
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * S5PV310 - CPU frequency scaling support
+ *
+ * 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/types.h>
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/slab.h>
+#include <linux/regulator/consumer.h>
+#include <linux/cpufreq.h>
+
+#include <mach/map.h>
+#include <mach/regs-clock.h>
+#include <mach/regs-mem.h>
+
+#include <plat/clock.h>
+#include <plat/pm.h>
+
+static struct clk *cpu_clk;
+static struct clk *moutcore;
+static struct clk *mout_mpll;
+static struct clk *mout_apll;
+
+#ifdef CONFIG_REGULATOR
+static struct regulator *arm_regulator;
+static struct regulator *int_regulator;
+#endif
+
+static struct cpufreq_freqs freqs;
+static unsigned int memtype;
+
+enum s5pv310_memory_type {
+ DDR2 = 4,
+ LPDDR2,
+ DDR3,
+};
+
+enum cpufreq_level_index {
+ L0, L1, L2, L3, CPUFREQ_LEVEL_END,
+};
+
+static struct cpufreq_frequency_table s5pv310_freq_table[] = {
+ {L0, 1000*1000},
+ {L1, 800*1000},
+ {L2, 400*1000},
+ {L3, 100*1000},
+ {0, CPUFREQ_TABLE_END},
+};
+
+static unsigned int clkdiv_cpu0[CPUFREQ_LEVEL_END][7] = {
+ /*
+ * Clock divider value for following
+ * { DIVCORE, DIVCOREM0, DIVCOREM1, DIVPERIPH,
+ * DIVATB, DIVPCLK_DBG, DIVAPLL }
+ */
+
+ /* ARM L0: 1000MHz */
+ { 0, 3, 7, 3, 3, 0, 1 },
+
+ /* ARM L1: 800MHz */
+ { 0, 3, 7, 3, 3, 0, 1 },
+
+ /* ARM L2: 400MHz */
+ { 0, 1, 3, 1, 3, 0, 1 },
+
+ /* ARM L3: 100MHz */
+ { 0, 0, 1, 0, 3, 1, 1 },
+};
+
+static unsigned int clkdiv_cpu1[CPUFREQ_LEVEL_END][2] = {
+ /*
+ * Clock divider value for following
+ * { DIVCOPY, DIVHPM }
+ */
+
+ /* ARM L0: 1000MHz */
+ { 3, 0 },
+
+ /* ARM L1: 800MHz */
+ { 3, 0 },
+
+ /* ARM L2: 400MHz */
+ { 3, 0 },
+
+ /* ARM L3: 100MHz */
+ { 3, 0 },
+};
+
+static unsigned int clkdiv_dmc0[CPUFREQ_LEVEL_END][8] = {
+ /*
+ * Clock divider value for following
+ * { DIVACP, DIVACP_PCLK, DIVDPHY, DIVDMC, DIVDMCD
+ * DIVDMCP, DIVCOPY2, DIVCORE_TIMERS }
+ */
+
+ /* DMC L0: 400MHz */
+ { 3, 1, 1, 1, 1, 1, 3, 1 },
+
+ /* DMC L1: 400MHz */
+ { 3, 1, 1, 1, 1, 1, 3, 1 },
+
+ /* DMC L2: 266.7MHz */
+ { 7, 1, 1, 2, 1, 1, 3, 1 },
+
+ /* DMC L3: 200MHz */
+ { 7, 1, 1, 3, 1, 1, 3, 1 },
+};
+
+static unsigned int clkdiv_top[CPUFREQ_LEVEL_END][5] = {
+ /*
+ * Clock divider value for following
+ * { DIVACLK200, DIVACLK100, DIVACLK160, DIVACLK133, DIVONENAND }
+ */
+
+ /* ACLK200 L0: 200MHz */
+ { 3, 7, 4, 5, 1 },
+
+ /* ACLK200 L1: 200MHz */
+ { 3, 7, 4, 5, 1 },
+
+ /* ACLK200 L2: 160MHz */
+ { 4, 7, 5, 7, 1 },
+
+ /* ACLK200 L3: 133.3MHz */
+ { 5, 7, 7, 7, 1 },
+};
+
+static unsigned int clkdiv_lr_bus[CPUFREQ_LEVEL_END][2] = {
+ /*
+ * Clock divider value for following
+ * { DIVGDL/R, DIVGPL/R }
+ */
+
+ /* ACLK_GDL/R L0: 200MHz */
+ { 3, 1 },
+
+ /* ACLK_GDL/R L1: 200MHz */
+ { 3, 1 },
+
+ /* ACLK_GDL/R L2: 160MHz */
+ { 4, 1 },
+
+ /* ACLK_GDL/R L3: 133.3MHz */
+ { 5, 1 },
+};
+
+struct cpufreq_voltage_table {
+ unsigned int index; /* any */
+ unsigned int arm_volt; /* uV */
+ unsigned int int_volt;
+};
+
+static struct cpufreq_voltage_table s5pv310_volt_table[CPUFREQ_LEVEL_END] = {
+ {
+ .index = L0,
+ .arm_volt = 1200000,
+ .int_volt = 1100000,
+ }, {
+ .index = L1,
+ .arm_volt = 1100000,
+ .int_volt = 1100000,
+ }, {
+ .index = L2,
+ .arm_volt = 1000000,
+ .int_volt = 1000000,
+ }, {
+ .index = L3,
+ .arm_volt = 900000,
+ .int_volt = 1000000,
+ },
+};
+
+static unsigned int s5pv310_apll_pms_table[CPUFREQ_LEVEL_END] = {
+ /* APLL FOUT L0: 1000MHz */
+ ((250 << 16) | (6 << 8) | 1),
+
+ /* APLL FOUT L1: 800MHz */
+ ((200 << 16) | (6 << 8) | 1),
+
+ /* APLL FOUT L2 : 400MHz */
+ ((200 << 16) | (6 << 8) | 2),
+
+ /* APLL FOUT L3: 100MHz */
+ ((200 << 16) | (6 << 8) | 4),
+};
+
+int s5pv310_verify_speed(struct cpufreq_policy *policy)
+{
+ return cpufreq_frequency_table_verify(policy, s5pv310_freq_table);
+}
+
+unsigned int s5pv310_getspeed(unsigned int cpu)
+{
+ return clk_get_rate(cpu_clk) / 1000;
+}
+
+void s5pv310_set_clkdiv(unsigned int div_index)
+{
+ unsigned int tmp;
+
+ /* Change Divider - CPU0 */
+
+ tmp = __raw_readl(S5P_CLKDIV_CPU);
+
+ tmp &= ~(S5P_CLKDIV_CPU0_CORE_MASK | S5P_CLKDIV_CPU0_COREM0_MASK |
+ S5P_CLKDIV_CPU0_COREM1_MASK | S5P_CLKDIV_CPU0_PERIPH_MASK |
+ S5P_CLKDIV_CPU0_ATB_MASK | S5P_CLKDIV_CPU0_PCLKDBG_MASK |
+ S5P_CLKDIV_CPU0_APLL_MASK);
+
+ tmp |= ((clkdiv_cpu0[div_index][0] << S5P_CLKDIV_CPU0_CORE_SHIFT) |
+ (clkdiv_cpu0[div_index][1] << S5P_CLKDIV_CPU0_COREM0_SHIFT) |
+ (clkdiv_cpu0[div_index][2] << S5P_CLKDIV_CPU0_COREM1_SHIFT) |
+ (clkdiv_cpu0[div_index][3] << S5P_CLKDIV_CPU0_PERIPH_SHIFT) |
+ (clkdiv_cpu0[div_index][4] << S5P_CLKDIV_CPU0_ATB_SHIFT) |
+ (clkdiv_cpu0[div_index][5] << S5P_CLKDIV_CPU0_PCLKDBG_SHIFT) |
+ (clkdiv_cpu0[div_index][6] << S5P_CLKDIV_CPU0_APLL_SHIFT));
+
+ __raw_writel(tmp, S5P_CLKDIV_CPU);
+
+ do {
+ tmp = __raw_readl(S5P_CLKDIV_STATCPU);
+ } while (tmp & 0x1111111);
+
+ /* Change Divider - CPU1 */
+
+ tmp = __raw_readl(S5P_CLKDIV_CPU1);
+
+ tmp &= ~((0x7 << 4) | 0x7);
+
+ tmp |= ((clkdiv_cpu1[div_index][0] << 4) |
+ (clkdiv_cpu1[div_index][1] << 0));
+
+ __raw_writel(tmp, S5P_CLKDIV_CPU1);
+
+ do {
+ tmp = __raw_readl(S5P_CLKDIV_STATCPU1);
+ } while (tmp & 0x11);
+
+ /* Change Divider - DMC0 */
+
+ tmp = __raw_readl(S5P_CLKDIV_DMC0);
+
+ tmp &= ~(S5P_CLKDIV_DMC0_ACP_MASK | S5P_CLKDIV_DMC0_ACPPCLK_MASK |
+ S5P_CLKDIV_DMC0_DPHY_MASK | S5P_CLKDIV_DMC0_DMC_MASK |
+ S5P_CLKDIV_DMC0_DMCD_MASK | S5P_CLKDIV_DMC0_DMCP_MASK |
+ S5P_CLKDIV_DMC0_COPY2_MASK | S5P_CLKDIV_DMC0_CORETI_MASK);
+
+ tmp |= ((clkdiv_dmc0[div_index][0] << S5P_CLKDIV_DMC0_ACP_SHIFT) |
+ (clkdiv_dmc0[div_index][1] << S5P_CLKDIV_DMC0_ACPPCLK_SHIFT) |
+ (clkdiv_dmc0[div_index][2] << S5P_CLKDIV_DMC0_DPHY_SHIFT) |
+ (clkdiv_dmc0[div_index][3] << S5P_CLKDIV_DMC0_DMC_SHIFT) |
+ (clkdiv_dmc0[div_index][4] << S5P_CLKDIV_DMC0_DMCD_SHIFT) |
+ (clkdiv_dmc0[div_index][5] << S5P_CLKDIV_DMC0_DMCP_SHIFT) |
+ (clkdiv_dmc0[div_index][6] << S5P_CLKDIV_DMC0_COPY2_SHIFT) |
+ (clkdiv_dmc0[div_index][7] << S5P_CLKDIV_DMC0_CORETI_SHIFT));
+
+ __raw_writel(tmp, S5P_CLKDIV_DMC0);
+
+ do {
+ tmp = __raw_readl(S5P_CLKDIV_STAT_DMC0);
+ } while (tmp & 0x11111111);
+
+ /* Change Divider - TOP */
+
+ tmp = __raw_readl(S5P_CLKDIV_TOP);
+
+ tmp &= ~(S5P_CLKDIV_TOP_ACLK200_MASK | S5P_CLKDIV_TOP_ACLK100_MASK |
+ S5P_CLKDIV_TOP_ACLK160_MASK | S5P_CLKDIV_TOP_ACLK133_MASK |
+ S5P_CLKDIV_TOP_ONENAND_MASK);
+
+ tmp |= ((clkdiv_top[div_index][0] << S5P_CLKDIV_TOP_ACLK200_SHIFT) |
+ (clkdiv_top[div_index][1] << S5P_CLKDIV_TOP_ACLK100_SHIFT) |
+ (clkdiv_top[div_index][2] << S5P_CLKDIV_TOP_ACLK160_SHIFT) |
+ (clkdiv_top[div_index][3] << S5P_CLKDIV_TOP_ACLK133_SHIFT) |
+ (clkdiv_top[div_index][4] << S5P_CLKDIV_TOP_ONENAND_SHIFT));
+
+ __raw_writel(tmp, S5P_CLKDIV_TOP);
+
+ do {
+ tmp = __raw_readl(S5P_CLKDIV_STAT_TOP);
+ } while (tmp & 0x11111);
+
+ /* Change Divider - LEFTBUS */
+
+ tmp = __raw_readl(S5P_CLKDIV_LEFTBUS);
+
+ tmp &= ~(S5P_CLKDIV_BUS_GDLR_MASK | S5P_CLKDIV_BUS_GPLR_MASK);
+
+ tmp |= ((clkdiv_lr_bus[div_index][0] << S5P_CLKDIV_BUS_GDLR_SHIFT) |
+ (clkdiv_lr_bus[div_index][1] << S5P_CLKDIV_BUS_GPLR_SHIFT));
+
+ __raw_writel(tmp, S5P_CLKDIV_LEFTBUS);
+
+ do {
+ tmp = __raw_readl(S5P_CLKDIV_STAT_LEFTBUS);
+ } while (tmp & 0x11);
+
+ /* Change Divider - RIGHTBUS */
+
+ tmp = __raw_readl(S5P_CLKDIV_RIGHTBUS);
+
+ tmp &= ~(S5P_CLKDIV_BUS_GDLR_MASK | S5P_CLKDIV_BUS_GPLR_MASK);
+
+ tmp |= ((clkdiv_lr_bus[div_index][0] << S5P_CLKDIV_BUS_GDLR_SHIFT) |
+ (clkdiv_lr_bus[div_index][1] << S5P_CLKDIV_BUS_GPLR_SHIFT));
+
+ __raw_writel(tmp, S5P_CLKDIV_RIGHTBUS);
+
+ do {
+ tmp = __raw_readl(S5P_CLKDIV_STAT_RIGHTBUS);
+ } while (tmp & 0x11);
+}
+
+static void s5pv310_set_apll(unsigned int index)
+{
+ unsigned int tmp;
+
+ /* 1. MUX_CORE_SEL = MPLL, ARMCLK uses MPLL for lock time */
+ clk_set_parent(moutcore, mout_mpll);
+
+ do {
+ tmp = (__raw_readl(S5P_CLKMUX_STATCPU)
+ >> S5P_CLKSRC_CPU_MUXCORE_SHIFT);
+ tmp &= 0x7;
+ } while (tmp != 0x2);
+
+ /* 2. Set APLL Lock time */
+ __raw_writel(S5P_APLL_LOCKTIME, S5P_APLL_LOCK);
+
+ /* 3. Change PLL PMS values */
+ tmp = __raw_readl(S5P_APLL_CON0);
+ tmp &= ~((0x3ff << 16) | (0x3f << 8) | (0x7 << 0));
+ tmp |= s5pv310_apll_pms_table[index];
+ __raw_writel(tmp, S5P_APLL_CON0);
+
+ /* 4. wait_lock_time */
+ do {
+ tmp = __raw_readl(S5P_APLL_CON0);
+ } while (!(tmp & (0x1 << S5P_APLLCON0_LOCKED_SHIFT)));
+
+ /* 5. MUX_CORE_SEL = APLL */
+ clk_set_parent(moutcore, mout_apll);
+
+ do {
+ tmp = __raw_readl(S5P_CLKMUX_STATCPU);
+ tmp &= S5P_CLKMUX_STATCPU_MUXCORE_MASK;
+ } while (tmp != (0x1 << S5P_CLKSRC_CPU_MUXCORE_SHIFT));
+}
+
+static void s5pv310_set_frequency(unsigned int old_index, unsigned int new_index)
+{
+ unsigned int tmp;
+
+ if (old_index > new_index) {
+ /* The frequency changing to L0 needs to change apll */
+ if (freqs.new == s5pv310_freq_table[L0].frequency) {
+ /* 1. Change the system clock divider values */
+ s5pv310_set_clkdiv(new_index);
+
+ /* 2. Change the apll m,p,s value */
+ s5pv310_set_apll(new_index);
+ } else {
+ /* 1. Change the system clock divider values */
+ s5pv310_set_clkdiv(new_index);
+
+ /* 2. Change just s value in apll m,p,s value */
+ tmp = __raw_readl(S5P_APLL_CON0);
+ tmp &= ~(0x7 << 0);
+ tmp |= (s5pv310_apll_pms_table[new_index] & 0x7);
+ __raw_writel(tmp, S5P_APLL_CON0);
+ }
+ }
+
+ else if (old_index < new_index) {
+ /* The frequency changing from L0 needs to change apll */
+ if (freqs.old == s5pv310_freq_table[L0].frequency) {
+ /* 1. Change the apll m,p,s value */
+ s5pv310_set_apll(new_index);
+
+ /* 2. Change the system clock divider values */
+ s5pv310_set_clkdiv(new_index);
+ } else {
+ /* 1. Change just s value in apll m,p,s value */
+ tmp = __raw_readl(S5P_APLL_CON0);
+ tmp &= ~(0x7 << 0);
+ tmp |= (s5pv310_apll_pms_table[new_index] & 0x7);
+ __raw_writel(tmp, S5P_APLL_CON0);
+
+ /* 2. Change the system clock divider values */
+ s5pv310_set_clkdiv(new_index);
+ }
+ }
+}
+
+static int s5pv310_target(struct cpufreq_policy *policy,
+ unsigned int target_freq,
+ unsigned int relation)
+{
+ unsigned int index, old_index;
+ unsigned int arm_volt, int_volt;
+
+ freqs.old = s5pv310_getspeed(policy->cpu);
+
+ if (cpufreq_frequency_table_target(policy, s5pv310_freq_table,
+ freqs.old, relation, &old_index))
+ return -EINVAL;
+
+ if (cpufreq_frequency_table_target(policy, s5pv310_freq_table,
+ target_freq, relation, &index))
+ return -EINVAL;
+
+ freqs.new = s5pv310_freq_table[index].frequency;
+ freqs.cpu = policy->cpu;
+
+ if (freqs.new == freqs.old)
+ return 0;
+
+ /* get the voltage value */
+ arm_volt = s5pv310_volt_table[index].arm_volt;
+ int_volt = s5pv310_volt_table[index].int_volt;
+
+ cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
+
+ /* control regulator */
+ if (freqs.new > freqs.old) {
+ /* Voltage up */
+#ifdef CONFIG_REGULATOR
+ regulator_set_voltage(arm_regulator, arm_volt, arm_volt);
+ regulator_set_voltage(int_regulator, int_volt, int_volt);
+#endif
+ }
+
+ /* Clock Configuration Procedure */
+ s5pv310_set_frequency(old_index, index);
+
+ /* control regulator */
+ if (freqs.new < freqs.old) {
+ /* Voltage down */
+#ifdef CONFIG_REGULATOR
+ regulator_set_voltage(arm_regulator, arm_volt, arm_volt);
+ regulator_set_voltage(int_regulator, int_volt, int_volt);
+#endif
+ }
+
+ cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
+
+ return 0;
+}
+
+#ifdef CONFIG_PM
+static int s5pv310_cpufreq_suspend(struct cpufreq_policy *policy,
+ pm_message_t pmsg)
+{
+ return 0;
+}
+
+static int s5pv310_cpufreq_resume(struct cpufreq_policy *policy)
+{
+ return 0;
+}
+#endif
+
+static int s5pv310_cpufreq_cpu_init(struct cpufreq_policy *policy)
+{
+ policy->cur = policy->min = policy->max = s5pv310_getspeed(policy->cpu);
+
+ cpufreq_frequency_table_get_attr(s5pv310_freq_table, policy->cpu);
+
+ /* set the transition latency value */
+ policy->cpuinfo.transition_latency = 100000;
+
+ /*
+ * S5PV310 multi-core processors has 2 cores
+ * that the frequency cannot be set independently.
+ * Each cpu is bound to the same speed.
+ * So the affected cpu is all of the cpus.
+ */
+ cpumask_setall(policy->cpus);
+
+ return cpufreq_frequency_table_cpuinfo(policy, s5pv310_freq_table);
+}
+
+static struct cpufreq_driver s5pv310_driver = {
+ .flags = CPUFREQ_STICKY,
+ .verify = s5pv310_verify_speed,
+ .target = s5pv310_target,
+ .get = s5pv310_getspeed,
+ .init = s5pv310_cpufreq_cpu_init,
+ .name = "s5pv310_cpufreq",
+#ifdef CONFIG_PM
+ .suspend = s5pv310_cpufreq_suspend,
+ .resume = s5pv310_cpufreq_resume,
+#endif
+};
+
+static int __init s5pv310_cpufreq_init(void)
+{
+ cpu_clk = clk_get(NULL, "armclk");
+ if (IS_ERR(cpu_clk))
+ return PTR_ERR(cpu_clk);
+
+ moutcore = clk_get(NULL, "moutcore");
+ if (IS_ERR(moutcore))
+ goto out;
+
+ mout_mpll = clk_get(NULL, "mout_mpll");
+ if (IS_ERR(mout_mpll))
+ goto out;
+
+ mout_apll = clk_get(NULL, "mout_apll");
+ if (IS_ERR(mout_apll))
+ goto out;
+
+#ifdef CONFIG_REGULATOR
+ arm_regulator = regulator_get(NULL, "vdd_arm");
+ if (IS_ERR(arm_regulator)) {
+ printk(KERN_ERR "failed to get resource %s\n", "vdd_arm");
+ goto out;
+ }
+
+ int_regulator = regulator_get(NULL, "vdd_int");
+ if (IS_ERR(int_regulator)) {
+ printk(KERN_ERR "failed to get resource %s\n", "vdd_int");
+ goto out;
+ }
+#endif
+
+ /*
+ * Check DRAM type.
+ * Because DVFS level is different according to DRAM type.
+ */
+ memtype = __raw_readl(S5P_VA_DMC0 + S5P_DMC0_MEMCON_OFFSET);
+ memtype = (memtype >> S5P_DMC0_MEMTYPE_SHIFT);
+ memtype &= S5P_DMC0_MEMTYPE_MASK;
+
+ if ((memtype < DDR2) && (memtype > DDR3)) {
+ printk(KERN_ERR "%s: wrong memtype= 0x%x\n", __func__, memtype);
+ goto out;
+ } else {
+ printk(KERN_DEBUG "%s: memtype= 0x%x\n", __func__, memtype);
+ }
+
+ return cpufreq_register_driver(&s5pv310_driver);
+
+out:
+ if (!IS_ERR(cpu_clk))
+ clk_put(cpu_clk);
+
+ if (!IS_ERR(moutcore))
+ clk_put(moutcore);
+
+ if (!IS_ERR(mout_mpll))
+ clk_put(mout_mpll);
+
+ if (!IS_ERR(mout_apll))
+ clk_put(mout_apll);
+
+#ifdef CONFIG_REGULATOR
+ if (!IS_ERR(arm_regulator))
+ regulator_put(arm_regulator);
+
+ if (!IS_ERR(int_regulator))
+ regulator_put(int_regulator);
+#endif
+
+ printk(KERN_ERR "%s: failed initialization\n", __func__);
+
+ return -EINVAL;
+}
+late_initcall(s5pv310_cpufreq_init);
diff --git a/arch/arm/mach-s5pv310/dev-pd.c b/arch/arm/mach-s5pv310/dev-pd.c
new file mode 100644
index 0000000..58a50c2
--- /dev/null
+++ b/arch/arm/mach-s5pv310/dev-pd.c
@@ -0,0 +1,139 @@
+/* linux/arch/arm/mach-s5pv310/dev-pd.c
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * S5PV310 - Power Domain support
+ *
+ * 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/io.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/delay.h>
+
+#include <mach/regs-pmu.h>
+
+#include <plat/pd.h>
+
+static int s5pv310_pd_enable(struct device *dev)
+{
+ struct samsung_pd_info *pdata = dev->platform_data;
+ u32 timeout;
+
+ __raw_writel(S5P_INT_LOCAL_PWR_EN, pdata->base);
+
+ /* Wait max 1ms */
+ timeout = 10;
+ while ((__raw_readl(pdata->base + 0x4) & S5P_INT_LOCAL_PWR_EN)
+ != S5P_INT_LOCAL_PWR_EN) {
+ if (timeout == 0) {
+ printk(KERN_ERR "Power domain %s enable failed.\n",
+ dev_name(dev));
+ return -ETIMEDOUT;
+ }
+ timeout--;
+ udelay(100);
+ }
+
+ return 0;
+}
+
+static int s5pv310_pd_disable(struct device *dev)
+{
+ struct samsung_pd_info *pdata = dev->platform_data;
+ u32 timeout;
+
+ __raw_writel(0, pdata->base);
+
+ /* Wait max 1ms */
+ timeout = 10;
+ while (__raw_readl(pdata->base + 0x4) & S5P_INT_LOCAL_PWR_EN) {
+ if (timeout == 0) {
+ printk(KERN_ERR "Power domain %s disable failed.\n",
+ dev_name(dev));
+ return -ETIMEDOUT;
+ }
+ timeout--;
+ udelay(100);
+ }
+
+ return 0;
+}
+
+struct platform_device s5pv310_device_pd[] = {
+ {
+ .name = "samsung-pd",
+ .id = 0,
+ .dev = {
+ .platform_data = &(struct samsung_pd_info) {
+ .enable = s5pv310_pd_enable,
+ .disable = s5pv310_pd_disable,
+ .base = S5P_PMU_MFC_CONF,
+ },
+ },
+ }, {
+ .name = "samsung-pd",
+ .id = 1,
+ .dev = {
+ .platform_data = &(struct samsung_pd_info) {
+ .enable = s5pv310_pd_enable,
+ .disable = s5pv310_pd_disable,
+ .base = S5P_PMU_G3D_CONF,
+ },
+ },
+ }, {
+ .name = "samsung-pd",
+ .id = 2,
+ .dev = {
+ .platform_data = &(struct samsung_pd_info) {
+ .enable = s5pv310_pd_enable,
+ .disable = s5pv310_pd_disable,
+ .base = S5P_PMU_LCD0_CONF,
+ },
+ },
+ }, {
+ .name = "samsung-pd",
+ .id = 3,
+ .dev = {
+ .platform_data = &(struct samsung_pd_info) {
+ .enable = s5pv310_pd_enable,
+ .disable = s5pv310_pd_disable,
+ .base = S5P_PMU_LCD1_CONF,
+ },
+ },
+ }, {
+ .name = "samsung-pd",
+ .id = 4,
+ .dev = {
+ .platform_data = &(struct samsung_pd_info) {
+ .enable = s5pv310_pd_enable,
+ .disable = s5pv310_pd_disable,
+ .base = S5P_PMU_TV_CONF,
+ },
+ },
+ }, {
+ .name = "samsung-pd",
+ .id = 5,
+ .dev = {
+ .platform_data = &(struct samsung_pd_info) {
+ .enable = s5pv310_pd_enable,
+ .disable = s5pv310_pd_disable,
+ .base = S5P_PMU_CAM_CONF,
+ },
+ },
+ }, {
+ .name = "samsung-pd",
+ .id = 6,
+ .dev = {
+ .platform_data = &(struct samsung_pd_info) {
+ .enable = s5pv310_pd_enable,
+ .disable = s5pv310_pd_disable,
+ .base = S5P_PMU_GPS_CONF,
+ },
+ },
+ },
+};
diff --git a/arch/arm/mach-s5pv310/dev-sysmmu.c b/arch/arm/mach-s5pv310/dev-sysmmu.c
new file mode 100644
index 0000000..e1bb200
--- /dev/null
+++ b/arch/arm/mach-s5pv310/dev-sysmmu.c
@@ -0,0 +1,187 @@
+/* linux/arch/arm/mach-s5pv310/dev-sysmmu.c
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * 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/platform_device.h>
+#include <linux/dma-mapping.h>
+
+#include <mach/map.h>
+#include <mach/irqs.h>
+
+static struct resource s5pv310_sysmmu_resource[] = {
+ [0] = {
+ .start = S5PV310_PA_SYSMMU_MDMA,
+ .end = S5PV310_PA_SYSMMU_MDMA + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [1] = {
+ .start = IRQ_SYSMMU_MDMA0_0,
+ .end = IRQ_SYSMMU_MDMA0_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [2] = {
+ .start = S5PV310_PA_SYSMMU_SSS,
+ .end = S5PV310_PA_SYSMMU_SSS + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [3] = {
+ .start = IRQ_SYSMMU_SSS_0,
+ .end = IRQ_SYSMMU_SSS_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [4] = {
+ .start = S5PV310_PA_SYSMMU_FIMC0,
+ .end = S5PV310_PA_SYSMMU_FIMC0 + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [5] = {
+ .start = IRQ_SYSMMU_FIMC0_0,
+ .end = IRQ_SYSMMU_FIMC0_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [6] = {
+ .start = S5PV310_PA_SYSMMU_FIMC1,
+ .end = S5PV310_PA_SYSMMU_FIMC1 + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [7] = {
+ .start = IRQ_SYSMMU_FIMC1_0,
+ .end = IRQ_SYSMMU_FIMC1_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [8] = {
+ .start = S5PV310_PA_SYSMMU_FIMC2,
+ .end = S5PV310_PA_SYSMMU_FIMC2 + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [9] = {
+ .start = IRQ_SYSMMU_FIMC2_0,
+ .end = IRQ_SYSMMU_FIMC2_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [10] = {
+ .start = S5PV310_PA_SYSMMU_FIMC3,
+ .end = S5PV310_PA_SYSMMU_FIMC3 + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [11] = {
+ .start = IRQ_SYSMMU_FIMC3_0,
+ .end = IRQ_SYSMMU_FIMC3_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [12] = {
+ .start = S5PV310_PA_SYSMMU_JPEG,
+ .end = S5PV310_PA_SYSMMU_JPEG + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [13] = {
+ .start = IRQ_SYSMMU_JPEG_0,
+ .end = IRQ_SYSMMU_JPEG_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [14] = {
+ .start = S5PV310_PA_SYSMMU_FIMD0,
+ .end = S5PV310_PA_SYSMMU_FIMD0 + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [15] = {
+ .start = IRQ_SYSMMU_LCD0_M0_0,
+ .end = IRQ_SYSMMU_LCD0_M0_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [16] = {
+ .start = S5PV310_PA_SYSMMU_FIMD1,
+ .end = S5PV310_PA_SYSMMU_FIMD1 + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [17] = {
+ .start = IRQ_SYSMMU_LCD1_M1_0,
+ .end = IRQ_SYSMMU_LCD1_M1_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [18] = {
+ .start = S5PV310_PA_SYSMMU_PCIe,
+ .end = S5PV310_PA_SYSMMU_PCIe + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [19] = {
+ .start = IRQ_SYSMMU_PCIE_0,
+ .end = IRQ_SYSMMU_PCIE_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [20] = {
+ .start = S5PV310_PA_SYSMMU_G2D,
+ .end = S5PV310_PA_SYSMMU_G2D + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [21] = {
+ .start = IRQ_SYSMMU_2D_0,
+ .end = IRQ_SYSMMU_2D_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [22] = {
+ .start = S5PV310_PA_SYSMMU_ROTATOR,
+ .end = S5PV310_PA_SYSMMU_ROTATOR + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [23] = {
+ .start = IRQ_SYSMMU_ROTATOR_0,
+ .end = IRQ_SYSMMU_ROTATOR_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [24] = {
+ .start = S5PV310_PA_SYSMMU_MDMA2,
+ .end = S5PV310_PA_SYSMMU_MDMA2 + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [25] = {
+ .start = IRQ_SYSMMU_MDMA1_0,
+ .end = IRQ_SYSMMU_MDMA1_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [26] = {
+ .start = S5PV310_PA_SYSMMU_TV,
+ .end = S5PV310_PA_SYSMMU_TV + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [27] = {
+ .start = IRQ_SYSMMU_TV_M0_0,
+ .end = IRQ_SYSMMU_TV_M0_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [28] = {
+ .start = S5PV310_PA_SYSMMU_MFC_L,
+ .end = S5PV310_PA_SYSMMU_MFC_L + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [29] = {
+ .start = IRQ_SYSMMU_MFC_M0_0,
+ .end = IRQ_SYSMMU_MFC_M0_0,
+ .flags = IORESOURCE_IRQ,
+ },
+ [30] = {
+ .start = S5PV310_PA_SYSMMU_MFC_R,
+ .end = S5PV310_PA_SYSMMU_MFC_R + SZ_64K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [31] = {
+ .start = IRQ_SYSMMU_MFC_M1_0,
+ .end = IRQ_SYSMMU_MFC_M1_0,
+ .flags = IORESOURCE_IRQ,
+ },
+};
+
+struct platform_device s5pv310_device_sysmmu = {
+ .name = "s5p-sysmmu",
+ .id = 32,
+ .num_resources = ARRAY_SIZE(s5pv310_sysmmu_resource),
+ .resource = s5pv310_sysmmu_resource,
+};
+
+EXPORT_SYMBOL(s5pv310_device_sysmmu);
diff --git a/arch/arm/mach-s5pv310/hotplug.c b/arch/arm/mach-s5pv310/hotplug.c
index afa5392..c24235c 100644
--- a/arch/arm/mach-s5pv310/hotplug.c
+++ b/arch/arm/mach-s5pv310/hotplug.c
@@ -30,10 +30,10 @@
* Turn off coherency
*/
" mrc p15, 0, %0, c1, c0, 1\n"
- " bic %0, %0, %2\n"
+ " bic %0, %0, #0x20\n"
" mcr p15, 0, %0, c1, c0, 1\n"
" mrc p15, 0, %0, c1, c0, 0\n"
- " bic %0, %0, #0x04\n"
+ " bic %0, %0, %2\n"
" mcr p15, 0, %0, c1, c0, 0\n"
: "=&r" (v)
: "r" (0), "Ir" (CR_C)
diff --git a/arch/arm/mach-s5pv310/include/mach/irqs.h b/arch/arm/mach-s5pv310/include/mach/irqs.h
index 3c05c58..536b0b5 100644
--- a/arch/arm/mach-s5pv310/include/mach/irqs.h
+++ b/arch/arm/mach-s5pv310/include/mach/irqs.h
@@ -25,6 +25,8 @@
#define IRQ_SPI(x) S5P_IRQ(x+32)
+#define IRQ_MCT1 IRQ_SPI(35)
+
#define IRQ_EINT0 IRQ_SPI(40)
#define IRQ_EINT1 IRQ_SPI(41)
#define IRQ_EINT2 IRQ_SPI(42)
@@ -36,9 +38,8 @@
#define IRQ_JPEG IRQ_SPI(48)
#define IRQ_2D IRQ_SPI(49)
#define IRQ_PCIE IRQ_SPI(50)
-#define IRQ_SYSTEM_TIMER IRQ_SPI(51)
+#define IRQ_MCT0 IRQ_SPI(51)
#define IRQ_MFC IRQ_SPI(52)
-#define IRQ_WDT IRQ_SPI(53)
#define IRQ_AUDIO_SS IRQ_SPI(54)
#define IRQ_AC97 IRQ_SPI(55)
#define IRQ_SPDIF IRQ_SPI(56)
@@ -54,6 +55,24 @@
#define COMBINER_GROUP(x) ((x) * MAX_IRQ_IN_COMBINER + IRQ_SPI(64))
#define COMBINER_IRQ(x, y) (COMBINER_GROUP(x) + y)
+#define IRQ_SYSMMU_MDMA0_0 COMBINER_IRQ(4, 0)
+#define IRQ_SYSMMU_SSS_0 COMBINER_IRQ(4, 1)
+#define IRQ_SYSMMU_FIMC0_0 COMBINER_IRQ(4, 2)
+#define IRQ_SYSMMU_FIMC1_0 COMBINER_IRQ(4, 3)
+#define IRQ_SYSMMU_FIMC2_0 COMBINER_IRQ(4, 4)
+#define IRQ_SYSMMU_FIMC3_0 COMBINER_IRQ(4, 5)
+#define IRQ_SYSMMU_JPEG_0 COMBINER_IRQ(4, 6)
+#define IRQ_SYSMMU_2D_0 COMBINER_IRQ(4, 7)
+
+#define IRQ_SYSMMU_ROTATOR_0 COMBINER_IRQ(5, 0)
+#define IRQ_SYSMMU_MDMA1_0 COMBINER_IRQ(5, 1)
+#define IRQ_SYSMMU_LCD0_M0_0 COMBINER_IRQ(5, 2)
+#define IRQ_SYSMMU_LCD1_M1_0 COMBINER_IRQ(5, 3)
+#define IRQ_SYSMMU_TV_M0_0 COMBINER_IRQ(5, 4)
+#define IRQ_SYSMMU_MFC_M0_0 COMBINER_IRQ(5, 5)
+#define IRQ_SYSMMU_MFC_M1_0 COMBINER_IRQ(5, 6)
+#define IRQ_SYSMMU_PCIE_0 COMBINER_IRQ(5, 7)
+
#define IRQ_PDMA0 COMBINER_IRQ(21, 0)
#define IRQ_PDMA1 COMBINER_IRQ(21, 1)
@@ -86,8 +105,13 @@
#define IRQ_HSMMC2 COMBINER_IRQ(29, 2)
#define IRQ_HSMMC3 COMBINER_IRQ(29, 3)
+#define IRQ_MIPI_CSIS0 COMBINER_IRQ(30, 0)
+#define IRQ_MIPI_CSIS1 COMBINER_IRQ(30, 1)
+
#define IRQ_ONENAND_AUDI COMBINER_IRQ(34, 0)
+#define IRQ_MCT_L1 COMBINER_IRQ(35, 3)
+
#define IRQ_EINT4 COMBINER_IRQ(37, 0)
#define IRQ_EINT5 COMBINER_IRQ(37, 1)
#define IRQ_EINT6 COMBINER_IRQ(37, 2)
@@ -104,7 +128,11 @@
#define IRQ_EINT16_31 COMBINER_IRQ(39, 0)
-#define MAX_COMBINER_NR 40
+#define IRQ_MCT_L0 COMBINER_IRQ(51, 0)
+
+#define IRQ_WDT COMBINER_IRQ(53, 0)
+
+#define MAX_COMBINER_NR 54
#define S5P_IRQ_EINT_BASE COMBINER_IRQ(MAX_COMBINER_NR, 0)
diff --git a/arch/arm/mach-s5pv310/include/mach/map.h b/arch/arm/mach-s5pv310/include/mach/map.h
index 5399446..74d4006 100644
--- a/arch/arm/mach-s5pv310/include/mach/map.h
+++ b/arch/arm/mach-s5pv310/include/mach/map.h
@@ -39,11 +39,15 @@
#define S5PV310_PA_SYSCON (0x10010000)
#define S5P_PA_SYSCON S5PV310_PA_SYSCON
+#define S5PV310_PA_PMU (0x10020000)
+
#define S5PV310_PA_CMU (0x10030000)
#define S5PV310_PA_WATCHDOG (0x10060000)
#define S5PV310_PA_RTC (0x10070000)
+#define S5PV310_PA_DMC0 (0x10400000)
+
#define S5PV310_PA_COMBINER (0x10448000)
#define S5PV310_PA_COREPERI (0x10500000)
@@ -61,9 +65,13 @@
#define S5PV310_PA_GPIO2 (0x11000000)
#define S5PV310_PA_GPIO3 (0x03860000)
+#define S5PV310_PA_MIPI_CSIS0 0x11880000
+#define S5PV310_PA_MIPI_CSIS1 0x11890000
+
#define S5PV310_PA_HSMMC(x) (0x12510000 + ((x) * 0x10000))
#define S5PV310_PA_SROMC (0x12570000)
+#define S5P_PA_SROMC S5PV310_PA_SROMC
/* S/PDIF */
#define S5PV310_PA_SPDIF 0xE1100000
@@ -100,6 +108,25 @@
#define S5PV310_PA_SDRAM (0x40000000)
#define S5P_PA_SDRAM S5PV310_PA_SDRAM
+#define S5PV310_PA_SYSMMU_MDMA 0x10A40000
+#define S5PV310_PA_SYSMMU_SSS 0x10A50000
+#define S5PV310_PA_SYSMMU_FIMC0 0x11A20000
+#define S5PV310_PA_SYSMMU_FIMC1 0x11A30000
+#define S5PV310_PA_SYSMMU_FIMC2 0x11A40000
+#define S5PV310_PA_SYSMMU_FIMC3 0x11A50000
+#define S5PV310_PA_SYSMMU_JPEG 0x11A60000
+#define S5PV310_PA_SYSMMU_FIMD0 0x11E20000
+#define S5PV310_PA_SYSMMU_FIMD1 0x12220000
+#define S5PV310_PA_SYSMMU_PCIe 0x12620000
+#define S5PV310_PA_SYSMMU_G2D 0x12A20000
+#define S5PV310_PA_SYSMMU_ROTATOR 0x12A30000
+#define S5PV310_PA_SYSMMU_MDMA2 0x12A40000
+#define S5PV310_PA_SYSMMU_TV 0x12E20000
+#define S5PV310_PA_SYSMMU_MFC_L 0x13620000
+#define S5PV310_PA_SYSMMU_MFC_R 0x13630000
+#define S5PV310_SYSMMU_TOTAL_IPNUM 16
+#define S5P_SYSMMU_TOTAL_IPNUM S5PV310_SYSMMU_TOTAL_IPNUM
+
/* compatibiltiy defines. */
#define S3C_PA_UART S5PV310_PA_UART
#define S3C_PA_HSMMC0 S5PV310_PA_HSMMC(0)
@@ -116,5 +143,7 @@
#define S3C_PA_IIC7 S5PV310_PA_IIC(7)
#define S3C_PA_RTC S5PV310_PA_RTC
#define S3C_PA_WDT S5PV310_PA_WATCHDOG
+#define S5P_PA_MIPI_CSIS0 S5PV310_PA_MIPI_CSIS0
+#define S5P_PA_MIPI_CSIS1 S5PV310_PA_MIPI_CSIS1
#endif /* __ASM_ARCH_MAP_H */
diff --git a/arch/arm/mach-s5pv310/include/mach/regs-clock.h b/arch/arm/mach-s5pv310/include/mach/regs-clock.h
index f1028ca..b5c4ada 100644
--- a/arch/arm/mach-s5pv310/include/mach/regs-clock.h
+++ b/arch/arm/mach-s5pv310/include/mach/regs-clock.h
@@ -19,6 +19,12 @@
#define S5P_INFORM0 S5P_CLKREG(0x800)
+#define S5P_CLKDIV_LEFTBUS S5P_CLKREG(0x04500)
+#define S5P_CLKDIV_STAT_LEFTBUS S5P_CLKREG(0x04600)
+
+#define S5P_CLKDIV_RIGHTBUS S5P_CLKREG(0x08500)
+#define S5P_CLKDIV_STAT_RIGHTBUS S5P_CLKREG(0x08600)
+
#define S5P_EPLL_CON0 S5P_CLKREG(0x0C110)
#define S5P_EPLL_CON1 S5P_CLKREG(0x0C114)
#define S5P_VPLL_CON0 S5P_CLKREG(0x0C120)
@@ -58,6 +64,8 @@
#define S5P_CLKSRC_MASK_PERIL0 S5P_CLKREG(0x0C350)
#define S5P_CLKSRC_MASK_PERIL1 S5P_CLKREG(0x0C354)
+#define S5P_CLKDIV_STAT_TOP S5P_CLKREG(0x0C610)
+
#define S5P_CLKGATE_IP_CAM S5P_CLKREG(0x0C920)
#define S5P_CLKGATE_IP_IMAGE S5P_CLKREG(0x0C930)
#define S5P_CLKGATE_IP_LCD0 S5P_CLKREG(0x0C934)
@@ -66,8 +74,9 @@
#define S5P_CLKGATE_IP_PERIL S5P_CLKREG(0x0C950)
#define S5P_CLKGATE_IP_PERIR S5P_CLKREG(0x0C960)
-#define S5P_CLKSRC_CORE S5P_CLKREG(0x10200)
-#define S5P_CLKDIV_CORE0 S5P_CLKREG(0x10500)
+#define S5P_CLKSRC_DMC S5P_CLKREG(0x10200)
+#define S5P_CLKDIV_DMC0 S5P_CLKREG(0x10500)
+#define S5P_CLKDIV_STAT_DMC0 S5P_CLKREG(0x10600)
#define S5P_APLL_LOCK S5P_CLKREG(0x14000)
#define S5P_MPLL_LOCK S5P_CLKREG(0x14004)
@@ -80,10 +89,77 @@
#define S5P_CLKMUX_STATCPU S5P_CLKREG(0x14400)
#define S5P_CLKDIV_CPU S5P_CLKREG(0x14500)
+#define S5P_CLKDIV_CPU1 S5P_CLKREG(0x14504)
#define S5P_CLKDIV_STATCPU S5P_CLKREG(0x14600)
+#define S5P_CLKDIV_STATCPU1 S5P_CLKREG(0x14604)
#define S5P_CLKGATE_SCLKCPU S5P_CLKREG(0x14800)
+/* APLL_LOCK */
+#define S5P_APLL_LOCKTIME (0x1C20) /* 300us */
+
+/* APLL_CON0 */
+#define S5P_APLLCON0_ENABLE_SHIFT (31)
+#define S5P_APLLCON0_LOCKED_SHIFT (29)
+#define S5P_APLL_VAL_1000 ((250 << 16) | (6 << 8) | 1)
+#define S5P_APLL_VAL_800 ((200 << 16) | (6 << 8) | 1)
+
+/* CLK_SRC_CPU */
+#define S5P_CLKSRC_CPU_MUXCORE_SHIFT (16)
+#define S5P_CLKMUX_STATCPU_MUXCORE_MASK (0x7 << S5P_CLKSRC_CPU_MUXCORE_SHIFT)
+
+/* CLKDIV_CPU0 */
+#define S5P_CLKDIV_CPU0_CORE_SHIFT (0)
+#define S5P_CLKDIV_CPU0_CORE_MASK (0x7 << S5P_CLKDIV_CPU0_CORE_SHIFT)
+#define S5P_CLKDIV_CPU0_COREM0_SHIFT (4)
+#define S5P_CLKDIV_CPU0_COREM0_MASK (0x7 << S5P_CLKDIV_CPU0_COREM0_SHIFT)
+#define S5P_CLKDIV_CPU0_COREM1_SHIFT (8)
+#define S5P_CLKDIV_CPU0_COREM1_MASK (0x7 << S5P_CLKDIV_CPU0_COREM1_SHIFT)
+#define S5P_CLKDIV_CPU0_PERIPH_SHIFT (12)
+#define S5P_CLKDIV_CPU0_PERIPH_MASK (0x7 << S5P_CLKDIV_CPU0_PERIPH_SHIFT)
+#define S5P_CLKDIV_CPU0_ATB_SHIFT (16)
+#define S5P_CLKDIV_CPU0_ATB_MASK (0x7 << S5P_CLKDIV_CPU0_ATB_SHIFT)
+#define S5P_CLKDIV_CPU0_PCLKDBG_SHIFT (20)
+#define S5P_CLKDIV_CPU0_PCLKDBG_MASK (0x7 << S5P_CLKDIV_CPU0_PCLKDBG_SHIFT)
+#define S5P_CLKDIV_CPU0_APLL_SHIFT (24)
+#define S5P_CLKDIV_CPU0_APLL_MASK (0x7 << S5P_CLKDIV_CPU0_APLL_SHIFT)
+
+/* CLKDIV_DMC0 */
+#define S5P_CLKDIV_DMC0_ACP_SHIFT (0)
+#define S5P_CLKDIV_DMC0_ACP_MASK (0x7 << S5P_CLKDIV_DMC0_ACP_SHIFT)
+#define S5P_CLKDIV_DMC0_ACPPCLK_SHIFT (4)
+#define S5P_CLKDIV_DMC0_ACPPCLK_MASK (0x7 << S5P_CLKDIV_DMC0_ACPPCLK_SHIFT)
+#define S5P_CLKDIV_DMC0_DPHY_SHIFT (8)
+#define S5P_CLKDIV_DMC0_DPHY_MASK (0x7 << S5P_CLKDIV_DMC0_DPHY_SHIFT)
+#define S5P_CLKDIV_DMC0_DMC_SHIFT (12)
+#define S5P_CLKDIV_DMC0_DMC_MASK (0x7 << S5P_CLKDIV_DMC0_DMC_SHIFT)
+#define S5P_CLKDIV_DMC0_DMCD_SHIFT (16)
+#define S5P_CLKDIV_DMC0_DMCD_MASK (0x7 << S5P_CLKDIV_DMC0_DMCD_SHIFT)
+#define S5P_CLKDIV_DMC0_DMCP_SHIFT (20)
+#define S5P_CLKDIV_DMC0_DMCP_MASK (0x7 << S5P_CLKDIV_DMC0_DMCP_SHIFT)
+#define S5P_CLKDIV_DMC0_COPY2_SHIFT (24)
+#define S5P_CLKDIV_DMC0_COPY2_MASK (0x7 << S5P_CLKDIV_DMC0_COPY2_SHIFT)
+#define S5P_CLKDIV_DMC0_CORETI_SHIFT (28)
+#define S5P_CLKDIV_DMC0_CORETI_MASK (0x7 << S5P_CLKDIV_DMC0_CORETI_SHIFT)
+
+/* CLKDIV_TOP */
+#define S5P_CLKDIV_TOP_ACLK200_SHIFT (0)
+#define S5P_CLKDIV_TOP_ACLK200_MASK (0x7 << S5P_CLKDIV_TOP_ACLK200_SHIFT)
+#define S5P_CLKDIV_TOP_ACLK100_SHIFT (4)
+#define S5P_CLKDIV_TOP_ACLK100_MASK (0xf << S5P_CLKDIV_TOP_ACLK100_SHIFT)
+#define S5P_CLKDIV_TOP_ACLK160_SHIFT (8)
+#define S5P_CLKDIV_TOP_ACLK160_MASK (0x7 << S5P_CLKDIV_TOP_ACLK160_SHIFT)
+#define S5P_CLKDIV_TOP_ACLK133_SHIFT (12)
+#define S5P_CLKDIV_TOP_ACLK133_MASK (0x7 << S5P_CLKDIV_TOP_ACLK133_SHIFT)
+#define S5P_CLKDIV_TOP_ONENAND_SHIFT (16)
+#define S5P_CLKDIV_TOP_ONENAND_MASK (0x7 << S5P_CLKDIV_TOP_ONENAND_SHIFT)
+
+/* CLKDIV_LEFTBUS / CLKDIV_RIGHTBUS*/
+#define S5P_CLKDIV_BUS_GDLR_SHIFT (0)
+#define S5P_CLKDIV_BUS_GDLR_MASK (0x7 << S5P_CLKDIV_BUS_GDLR_SHIFT)
+#define S5P_CLKDIV_BUS_GPLR_SHIFT (4)
+#define S5P_CLKDIV_BUS_GPLR_MASK (0x7 << S5P_CLKDIV_BUS_GPLR_SHIFT)
+
/* Compatibility defines */
#define S5P_EPLL_CON S5P_EPLL_CON0
diff --git a/arch/arm/mach-s5pv310/include/mach/regs-mem.h b/arch/arm/mach-s5pv310/include/mach/regs-mem.h
new file mode 100644
index 0000000..8342271
--- /dev/null
+++ b/arch/arm/mach-s5pv310/include/mach/regs-mem.h
@@ -0,0 +1,23 @@
+/* linux/arch/arm/mach-s5pv310/include/mach/regs-mem.h
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * S5PV310 - SROMC and DMC register definitions
+ *
+ * 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 __ASM_ARCH_REGS_MEM_H
+#define __ASM_ARCH_REGS_MEM_H __FILE__
+
+#include <mach/map.h>
+
+#define S5P_DMC0_MEMCON_OFFSET 0x04
+
+#define S5P_DMC0_MEMTYPE_SHIFT 8
+#define S5P_DMC0_MEMTYPE_MASK 0xF
+
+#endif /* __ASM_ARCH_REGS_MEM_H */
diff --git a/arch/arm/mach-s5pv310/include/mach/regs-pmu.h b/arch/arm/mach-s5pv310/include/mach/regs-pmu.h
new file mode 100644
index 0000000..fb333d0
--- /dev/null
+++ b/arch/arm/mach-s5pv310/include/mach/regs-pmu.h
@@ -0,0 +1,30 @@
+/* linux/arch/arm/mach-s5pv310/include/mach/regs-pmu.h
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * S5PV310 - Power management unit definition
+ *
+ * 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 __ASM_ARCH_REGS_PMU_H
+#define __ASM_ARCH_REGS_PMU_H __FILE__
+
+#include <mach/map.h>
+
+#define S5P_PMUREG(x) (S5P_VA_PMU + (x))
+
+#define S5P_PMU_CAM_CONF S5P_PMUREG(0x3C00)
+#define S5P_PMU_TV_CONF S5P_PMUREG(0x3C20)
+#define S5P_PMU_MFC_CONF S5P_PMUREG(0x3C40)
+#define S5P_PMU_G3D_CONF S5P_PMUREG(0x3C60)
+#define S5P_PMU_LCD0_CONF S5P_PMUREG(0x3C80)
+#define S5P_PMU_LCD1_CONF S5P_PMUREG(0x3CA0)
+#define S5P_PMU_GPS_CONF S5P_PMUREG(0x3CE0)
+
+#define S5P_INT_LOCAL_PWR_EN 0x7
+
+#endif /* __ASM_ARCH_REGS_PMU_H */
diff --git a/arch/arm/mach-s5pv310/include/mach/regs-srom.h b/arch/arm/mach-s5pv310/include/mach/regs-srom.h
deleted file mode 100644
index 1898b3e..0000000
--- a/arch/arm/mach-s5pv310/include/mach/regs-srom.h
+++ /dev/null
@@ -1,50 +0,0 @@
-/* linux/arch/arm/mach-s5pv310/include/mach/regs-srom.h
- *
- * Copyright (c) 2010 Samsung Electronics Co., Ltd.
- * http://www.samsung.com
- *
- * S5PV310 - SROMC register definitions
- *
- * 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 __ASM_ARCH_REGS_SROM_H
-#define __ASM_ARCH_REGS_SROM_H __FILE__
-
-#include <mach/map.h>
-
-#define S5PV310_SROMREG(x) (S5P_VA_SROMC + (x))
-
-#define S5PV310_SROM_BW S5PV310_SROMREG(0x0)
-#define S5PV310_SROM_BC0 S5PV310_SROMREG(0x4)
-#define S5PV310_SROM_BC1 S5PV310_SROMREG(0x8)
-#define S5PV310_SROM_BC2 S5PV310_SROMREG(0xc)
-#define S5PV310_SROM_BC3 S5PV310_SROMREG(0x10)
-
-/* one register BW holds 4 x 4-bit packed settings for NCS0 - NCS3 */
-
-#define S5PV310_SROM_BW__DATAWIDTH__SHIFT 0
-#define S5PV310_SROM_BW__ADDRMODE__SHIFT 1
-#define S5PV310_SROM_BW__WAITENABLE__SHIFT 2
-#define S5PV310_SROM_BW__BYTEENABLE__SHIFT 3
-
-#define S5PV310_SROM_BW__CS_MASK 0xf
-
-#define S5PV310_SROM_BW__NCS0__SHIFT 0
-#define S5PV310_SROM_BW__NCS1__SHIFT 4
-#define S5PV310_SROM_BW__NCS2__SHIFT 8
-#define S5PV310_SROM_BW__NCS3__SHIFT 12
-
-/* applies to same to BCS0 - BCS3 */
-
-#define S5PV310_SROM_BCX__PMC__SHIFT 0
-#define S5PV310_SROM_BCX__TACP__SHIFT 4
-#define S5PV310_SROM_BCX__TCAH__SHIFT 8
-#define S5PV310_SROM_BCX__TCOH__SHIFT 12
-#define S5PV310_SROM_BCX__TACC__SHIFT 16
-#define S5PV310_SROM_BCX__TCOS__SHIFT 24
-#define S5PV310_SROM_BCX__TACS__SHIFT 28
-
-#endif /* __ASM_ARCH_REGS_SROM_H */
diff --git a/arch/arm/mach-s5pv310/include/mach/regs-sysmmu.h b/arch/arm/mach-s5pv310/include/mach/regs-sysmmu.h
new file mode 100644
index 0000000..0b28e81
--- /dev/null
+++ b/arch/arm/mach-s5pv310/include/mach/regs-sysmmu.h
@@ -0,0 +1,24 @@
+/* linux/arch/arm/mach-s5pv310/include/mach/regs-sysmmu.h
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * S5PV310 - System MMU register
+ *
+ * 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 __ASM_ARCH_REGS_SYSMMU_H
+#define __ASM_ARCH_REGS_SYSMMU_H __FILE__
+
+#define S5P_MMU_CTRL 0x000
+#define S5P_MMU_CFG 0x004
+#define S5P_MMU_STATUS 0x008
+#define S5P_MMU_FLUSH 0x00C
+#define S5P_PT_BASE_ADDR 0x014
+#define S5P_INT_STATUS 0x018
+#define S5P_PAGE_FAULT_ADDR 0x024
+
+#endif /* __ASM_ARCH_REGS_SYSMMU_H */
diff --git a/arch/arm/mach-s5pv310/include/mach/sysmmu.h b/arch/arm/mach-s5pv310/include/mach/sysmmu.h
new file mode 100644
index 0000000..662fe85f
--- /dev/null
+++ b/arch/arm/mach-s5pv310/include/mach/sysmmu.h
@@ -0,0 +1,119 @@
+/* linux/arch/arm/mach-s5pv310/include/mach/sysmmu.h
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com/
+ *
+ * Samsung sysmmu driver for S5PV310
+ *
+ * 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 __ASM_ARM_ARCH_SYSMMU_H
+#define __ASM_ARM_ARCH_SYSMMU_H __FILE__
+
+enum s5pv310_sysmmu_ips {
+ SYSMMU_MDMA,
+ SYSMMU_SSS,
+ SYSMMU_FIMC0,
+ SYSMMU_FIMC1,
+ SYSMMU_FIMC2,
+ SYSMMU_FIMC3,
+ SYSMMU_JPEG,
+ SYSMMU_FIMD0,
+ SYSMMU_FIMD1,
+ SYSMMU_PCIe,
+ SYSMMU_G2D,
+ SYSMMU_ROTATOR,
+ SYSMMU_MDMA2,
+ SYSMMU_TV,
+ SYSMMU_MFC_L,
+ SYSMMU_MFC_R,
+};
+
+static char *sysmmu_ips_name[S5P_SYSMMU_TOTAL_IPNUM] = {
+ "SYSMMU_MDMA" ,
+ "SYSMMU_SSS" ,
+ "SYSMMU_FIMC0" ,
+ "SYSMMU_FIMC1" ,
+ "SYSMMU_FIMC2" ,
+ "SYSMMU_FIMC3" ,
+ "SYSMMU_JPEG" ,
+ "SYSMMU_FIMD0" ,
+ "SYSMMU_FIMD1" ,
+ "SYSMMU_PCIe" ,
+ "SYSMMU_G2D" ,
+ "SYSMMU_ROTATOR",
+ "SYSMMU_MDMA2" ,
+ "SYSMMU_TV" ,
+ "SYSMMU_MFC_L" ,
+ "SYSMMU_MFC_R" ,
+};
+
+typedef enum s5pv310_sysmmu_ips sysmmu_ips;
+
+struct sysmmu_tt_info {
+ unsigned long *pgd;
+ unsigned long pgd_paddr;
+ unsigned long *pte;
+};
+
+struct sysmmu_controller {
+ const char *name;
+
+ /* channels registers */
+ void __iomem *regs;
+
+ /* channel irq */
+ unsigned int irq;
+
+ sysmmu_ips ips;
+
+ /* Translation Table Info. */
+ struct sysmmu_tt_info *tt_info;
+
+ struct resource *mem;
+ struct device *dev;
+
+ /* SysMMU controller enable - true : enable */
+ bool enable;
+};
+
+/**
+ * s5p_sysmmu_enable() - enable system mmu of ip
+ * @ips: The ip connected system mmu.
+ *
+ * This function enable system mmu to transfer address
+ * from virtual address to physical address
+ */
+int s5p_sysmmu_enable(sysmmu_ips ips);
+
+/**
+ * s5p_sysmmu_disable() - disable sysmmu mmu of ip
+ * @ips: The ip connected system mmu.
+ *
+ * This function disable system mmu to transfer address
+ * from virtual address to physical address
+ */
+int s5p_sysmmu_disable(sysmmu_ips ips);
+
+/**
+ * s5p_sysmmu_set_tablebase_pgd() - set page table base address to refer page table
+ * @ips: The ip connected system mmu.
+ * @pgd: The page table base address.
+ *
+ * This function set page table base address
+ * When system mmu transfer address from virtaul address to physical address,
+ * system mmu refer address information from page table
+ */
+int s5p_sysmmu_set_tablebase_pgd(sysmmu_ips ips, unsigned long pgd);
+
+/**
+ * s5p_sysmmu_tlb_invalidate() - flush all TLB entry in system mmu
+ * @ips: The ip connected system mmu.
+ *
+ * This function flush all TLB entry in system mmu
+ */
+int s5p_sysmmu_tlb_invalidate(sysmmu_ips ips);
+#endif /* __ASM_ARM_ARCH_SYSMMU_H */
diff --git a/arch/arm/mach-s5pv310/irq-combiner.c b/arch/arm/mach-s5pv310/irq-combiner.c
index c3f88c3..1ea4a9e 100644
--- a/arch/arm/mach-s5pv310/irq-combiner.c
+++ b/arch/arm/mach-s5pv310/irq-combiner.c
@@ -24,29 +24,32 @@
struct combiner_chip_data {
unsigned int irq_offset;
+ unsigned int irq_mask;
void __iomem *base;
};
static struct combiner_chip_data combiner_data[MAX_COMBINER_NR];
-static inline void __iomem *combiner_base(unsigned int irq)
+static inline void __iomem *combiner_base(struct irq_data *data)
{
- struct combiner_chip_data *combiner_data = get_irq_chip_data(irq);
+ struct combiner_chip_data *combiner_data =
+ irq_data_get_irq_chip_data(data);
+
return combiner_data->base;
}
-static void combiner_mask_irq(unsigned int irq)
+static void combiner_mask_irq(struct irq_data *data)
{
- u32 mask = 1 << (irq % 32);
+ u32 mask = 1 << (data->irq % 32);
- __raw_writel(mask, combiner_base(irq) + COMBINER_ENABLE_CLEAR);
+ __raw_writel(mask, combiner_base(data) + COMBINER_ENABLE_CLEAR);
}
-static void combiner_unmask_irq(unsigned int irq)
+static void combiner_unmask_irq(struct irq_data *data)
{
- u32 mask = 1 << (irq % 32);
+ u32 mask = 1 << (data->irq % 32);
- __raw_writel(mask, combiner_base(irq) + COMBINER_ENABLE_SET);
+ __raw_writel(mask, combiner_base(data) + COMBINER_ENABLE_SET);
}
static void combiner_handle_cascade_irq(unsigned int irq, struct irq_desc *desc)
@@ -57,11 +60,12 @@
unsigned long status;
/* primary controller ack'ing */
- chip->ack(irq);
+ chip->irq_ack(&desc->irq_data);
spin_lock(&irq_controller_lock);
status = __raw_readl(chip_data->base + COMBINER_INT_STATUS);
spin_unlock(&irq_controller_lock);
+ status &= chip_data->irq_mask;
if (status == 0)
goto out;
@@ -76,13 +80,13 @@
out:
/* primary controller unmasking */
- chip->unmask(irq);
+ chip->irq_unmask(&desc->irq_data);
}
static struct irq_chip combiner_chip = {
.name = "COMBINER",
- .mask = combiner_mask_irq,
- .unmask = combiner_unmask_irq,
+ .irq_mask = combiner_mask_irq,
+ .irq_unmask = combiner_unmask_irq,
};
void __init combiner_cascade_irq(unsigned int combiner_nr, unsigned int irq)
@@ -104,10 +108,12 @@
combiner_data[combiner_nr].base = base;
combiner_data[combiner_nr].irq_offset = irq_start;
+ combiner_data[combiner_nr].irq_mask = 0xff << ((combiner_nr % 4) << 3);
/* Disable all interrupts */
- __raw_writel(0xffffffff, base + COMBINER_ENABLE_CLEAR);
+ __raw_writel(combiner_data[combiner_nr].irq_mask,
+ base + COMBINER_ENABLE_CLEAR);
/* Setup the Linux IRQ subsystem */
diff --git a/arch/arm/mach-s5pv310/irq-eint.c b/arch/arm/mach-s5pv310/irq-eint.c
index 5877503..477bd9e 100644
--- a/arch/arm/mach-s5pv310/irq-eint.c
+++ b/arch/arm/mach-s5pv310/irq-eint.c
@@ -48,42 +48,43 @@
return ret;
}
-static inline void s5pv310_irq_eint_mask(unsigned int irq)
+static inline void s5pv310_irq_eint_mask(struct irq_data *data)
{
u32 mask;
spin_lock(&eint_lock);
- mask = __raw_readl(S5P_EINT_MASK(EINT_REG_NR(irq)));
- mask |= eint_irq_to_bit(irq);
- __raw_writel(mask, S5P_EINT_MASK(EINT_REG_NR(irq)));
+ mask = __raw_readl(S5P_EINT_MASK(EINT_REG_NR(data->irq)));
+ mask |= eint_irq_to_bit(data->irq);
+ __raw_writel(mask, S5P_EINT_MASK(EINT_REG_NR(data->irq)));
spin_unlock(&eint_lock);
}
-static void s5pv310_irq_eint_unmask(unsigned int irq)
+static void s5pv310_irq_eint_unmask(struct irq_data *data)
{
u32 mask;
spin_lock(&eint_lock);
- mask = __raw_readl(S5P_EINT_MASK(EINT_REG_NR(irq)));
- mask &= ~(eint_irq_to_bit(irq));
- __raw_writel(mask, S5P_EINT_MASK(EINT_REG_NR(irq)));
+ mask = __raw_readl(S5P_EINT_MASK(EINT_REG_NR(data->irq)));
+ mask &= ~(eint_irq_to_bit(data->irq));
+ __raw_writel(mask, S5P_EINT_MASK(EINT_REG_NR(data->irq)));
spin_unlock(&eint_lock);
}
-static inline void s5pv310_irq_eint_ack(unsigned int irq)
+static inline void s5pv310_irq_eint_ack(struct irq_data *data)
{
- __raw_writel(eint_irq_to_bit(irq), S5P_EINT_PEND(EINT_REG_NR(irq)));
+ __raw_writel(eint_irq_to_bit(data->irq),
+ S5P_EINT_PEND(EINT_REG_NR(data->irq)));
}
-static void s5pv310_irq_eint_maskack(unsigned int irq)
+static void s5pv310_irq_eint_maskack(struct irq_data *data)
{
- s5pv310_irq_eint_mask(irq);
- s5pv310_irq_eint_ack(irq);
+ s5pv310_irq_eint_mask(data);
+ s5pv310_irq_eint_ack(data);
}
-static int s5pv310_irq_eint_set_type(unsigned int irq, unsigned int type)
+static int s5pv310_irq_eint_set_type(struct irq_data *data, unsigned int type)
{
- int offs = EINT_OFFSET(irq);
+ int offs = EINT_OFFSET(data->irq);
int shift;
u32 ctrl, mask;
u32 newvalue = 0;
@@ -118,10 +119,10 @@
mask = 0x7 << shift;
spin_lock(&eint_lock);
- ctrl = __raw_readl(S5P_EINT_CON(EINT_REG_NR(irq)));
+ ctrl = __raw_readl(S5P_EINT_CON(EINT_REG_NR(data->irq)));
ctrl &= ~mask;
ctrl |= newvalue << shift;
- __raw_writel(ctrl, S5P_EINT_CON(EINT_REG_NR(irq)));
+ __raw_writel(ctrl, S5P_EINT_CON(EINT_REG_NR(data->irq)));
spin_unlock(&eint_lock);
switch (offs) {
@@ -146,13 +147,13 @@
static struct irq_chip s5pv310_irq_eint = {
.name = "s5pv310-eint",
- .mask = s5pv310_irq_eint_mask,
- .unmask = s5pv310_irq_eint_unmask,
- .mask_ack = s5pv310_irq_eint_maskack,
- .ack = s5pv310_irq_eint_ack,
- .set_type = s5pv310_irq_eint_set_type,
+ .irq_mask = s5pv310_irq_eint_mask,
+ .irq_unmask = s5pv310_irq_eint_unmask,
+ .irq_mask_ack = s5pv310_irq_eint_maskack,
+ .irq_ack = s5pv310_irq_eint_ack,
+ .irq_set_type = s5pv310_irq_eint_set_type,
#ifdef CONFIG_PM
- .set_wake = s3c_irqext_wake,
+ .irq_set_wake = s3c_irqext_wake,
#endif
};
@@ -192,14 +193,14 @@
u32 *irq_data = get_irq_data(irq);
struct irq_chip *chip = get_irq_chip(irq);
- chip->mask(irq);
+ chip->irq_mask(&desc->irq_data);
- if (chip->ack)
- chip->ack(irq);
+ if (chip->irq_ack)
+ chip->irq_ack(&desc->irq_data);
generic_handle_irq(*irq_data);
- chip->unmask(irq);
+ chip->irq_unmask(&desc->irq_data);
}
int __init s5pv310_init_irq_eint(void)
diff --git a/arch/arm/mach-s5pv310/mach-smdkc210.c b/arch/arm/mach-s5pv310/mach-smdkc210.c
index 2b8d4fc..2d49273 100644
--- a/arch/arm/mach-s5pv310/mach-smdkc210.c
+++ b/arch/arm/mach-s5pv310/mach-smdkc210.c
@@ -14,18 +14,21 @@
#include <linux/platform_device.h>
#include <linux/smsc911x.h>
#include <linux/io.h>
+#include <linux/i2c.h>
#include <asm/mach/arch.h>
#include <asm/mach-types.h>
#include <plat/regs-serial.h>
+#include <plat/regs-srom.h>
#include <plat/s5pv310.h>
#include <plat/cpu.h>
#include <plat/devs.h>
#include <plat/sdhci.h>
+#include <plat/iic.h>
+#include <plat/pd.h>
#include <mach/map.h>
-#include <mach/regs-srom.h>
/* Following are default values for UCON, ULCON and UFCON UART registers */
#define SMDKC210_UCON_DEFAULT (S3C2410_UCON_TXILEVEL | \
@@ -139,14 +142,29 @@
},
};
+static struct i2c_board_info i2c_devs1[] __initdata = {
+ {I2C_BOARD_INFO("wm8994", 0x1a),},
+};
+
static struct platform_device *smdkc210_devices[] __initdata = {
&s3c_device_hsmmc0,
&s3c_device_hsmmc1,
&s3c_device_hsmmc2,
&s3c_device_hsmmc3,
+ &s3c_device_i2c1,
&s3c_device_rtc,
&s3c_device_wdt,
+ &s5pv310_device_ac97,
+ &s5pv310_device_i2s0,
+ &s5pv310_device_pd[PD_MFC],
+ &s5pv310_device_pd[PD_G3D],
+ &s5pv310_device_pd[PD_LCD0],
+ &s5pv310_device_pd[PD_LCD1],
+ &s5pv310_device_pd[PD_CAM],
+ &s5pv310_device_pd[PD_TV],
+ &s5pv310_device_pd[PD_GPS],
&smdkc210_smsc911x,
+ &s5pv310_device_sysmmu,
};
static void __init smdkc210_smsc911x_init(void)
@@ -154,23 +172,22 @@
u32 cs1;
/* configure nCS1 width to 16 bits */
- cs1 = __raw_readl(S5PV310_SROM_BW) &
- ~(S5PV310_SROM_BW__CS_MASK <<
- S5PV310_SROM_BW__NCS1__SHIFT);
- cs1 |= ((1 << S5PV310_SROM_BW__DATAWIDTH__SHIFT) |
- (1 << S5PV310_SROM_BW__WAITENABLE__SHIFT) |
- (1 << S5PV310_SROM_BW__BYTEENABLE__SHIFT)) <<
- S5PV310_SROM_BW__NCS1__SHIFT;
- __raw_writel(cs1, S5PV310_SROM_BW);
+ cs1 = __raw_readl(S5P_SROM_BW) &
+ ~(S5P_SROM_BW__CS_MASK << S5P_SROM_BW__NCS1__SHIFT);
+ cs1 |= ((1 << S5P_SROM_BW__DATAWIDTH__SHIFT) |
+ (1 << S5P_SROM_BW__WAITENABLE__SHIFT) |
+ (1 << S5P_SROM_BW__BYTEENABLE__SHIFT)) <<
+ S5P_SROM_BW__NCS1__SHIFT;
+ __raw_writel(cs1, S5P_SROM_BW);
/* set timing for nCS1 suitable for ethernet chip */
- __raw_writel((0x1 << S5PV310_SROM_BCX__PMC__SHIFT) |
- (0x9 << S5PV310_SROM_BCX__TACP__SHIFT) |
- (0xc << S5PV310_SROM_BCX__TCAH__SHIFT) |
- (0x1 << S5PV310_SROM_BCX__TCOH__SHIFT) |
- (0x6 << S5PV310_SROM_BCX__TACC__SHIFT) |
- (0x1 << S5PV310_SROM_BCX__TCOS__SHIFT) |
- (0x1 << S5PV310_SROM_BCX__TACS__SHIFT), S5PV310_SROM_BC1);
+ __raw_writel((0x1 << S5P_SROM_BCX__PMC__SHIFT) |
+ (0x9 << S5P_SROM_BCX__TACP__SHIFT) |
+ (0xc << S5P_SROM_BCX__TCAH__SHIFT) |
+ (0x1 << S5P_SROM_BCX__TCOH__SHIFT) |
+ (0x6 << S5P_SROM_BCX__TACC__SHIFT) |
+ (0x1 << S5P_SROM_BCX__TCOS__SHIFT) |
+ (0x1 << S5P_SROM_BCX__TACS__SHIFT), S5P_SROM_BC1);
}
static void __init smdkc210_map_io(void)
@@ -182,6 +199,9 @@
static void __init smdkc210_machine_init(void)
{
+ s3c_i2c1_set_platdata(NULL);
+ i2c_register_board_info(1, i2c_devs1, ARRAY_SIZE(i2c_devs1));
+
smdkc210_smsc911x_init();
s3c_sdhci0_set_platdata(&smdkc210_hsmmc0_pdata);
diff --git a/arch/arm/mach-s5pv310/mach-smdkv310.c b/arch/arm/mach-s5pv310/mach-smdkv310.c
index 35826d6..28680cf 100644
--- a/arch/arm/mach-s5pv310/mach-smdkv310.c
+++ b/arch/arm/mach-s5pv310/mach-smdkv310.c
@@ -14,18 +14,21 @@
#include <linux/platform_device.h>
#include <linux/smsc911x.h>
#include <linux/io.h>
+#include <linux/i2c.h>
#include <asm/mach/arch.h>
#include <asm/mach-types.h>
#include <plat/regs-serial.h>
+#include <plat/regs-srom.h>
#include <plat/s5pv310.h>
#include <plat/cpu.h>
#include <plat/devs.h>
#include <plat/sdhci.h>
+#include <plat/iic.h>
+#include <plat/pd.h>
#include <mach/map.h>
-#include <mach/regs-srom.h>
/* Following are default values for UCON, ULCON and UFCON UART registers */
#define SMDKV310_UCON_DEFAULT (S3C2410_UCON_TXILEVEL | \
@@ -139,14 +142,29 @@
},
};
+static struct i2c_board_info i2c_devs1[] __initdata = {
+ {I2C_BOARD_INFO("wm8994", 0x1a),},
+};
+
static struct platform_device *smdkv310_devices[] __initdata = {
&s3c_device_hsmmc0,
&s3c_device_hsmmc1,
&s3c_device_hsmmc2,
&s3c_device_hsmmc3,
+ &s3c_device_i2c1,
&s3c_device_rtc,
&s3c_device_wdt,
+ &s5pv310_device_ac97,
+ &s5pv310_device_i2s0,
+ &s5pv310_device_pd[PD_MFC],
+ &s5pv310_device_pd[PD_G3D],
+ &s5pv310_device_pd[PD_LCD0],
+ &s5pv310_device_pd[PD_LCD1],
+ &s5pv310_device_pd[PD_CAM],
+ &s5pv310_device_pd[PD_TV],
+ &s5pv310_device_pd[PD_GPS],
&smdkv310_smsc911x,
+ &s5pv310_device_sysmmu,
};
static void __init smdkv310_smsc911x_init(void)
@@ -154,23 +172,22 @@
u32 cs1;
/* configure nCS1 width to 16 bits */
- cs1 = __raw_readl(S5PV310_SROM_BW) &
- ~(S5PV310_SROM_BW__CS_MASK <<
- S5PV310_SROM_BW__NCS1__SHIFT);
- cs1 |= ((1 << S5PV310_SROM_BW__DATAWIDTH__SHIFT) |
- (1 << S5PV310_SROM_BW__WAITENABLE__SHIFT) |
- (1 << S5PV310_SROM_BW__BYTEENABLE__SHIFT)) <<
- S5PV310_SROM_BW__NCS1__SHIFT;
- __raw_writel(cs1, S5PV310_SROM_BW);
+ cs1 = __raw_readl(S5P_SROM_BW) &
+ ~(S5P_SROM_BW__CS_MASK << S5P_SROM_BW__NCS1__SHIFT);
+ cs1 |= ((1 << S5P_SROM_BW__DATAWIDTH__SHIFT) |
+ (1 << S5P_SROM_BW__WAITENABLE__SHIFT) |
+ (1 << S5P_SROM_BW__BYTEENABLE__SHIFT)) <<
+ S5P_SROM_BW__NCS1__SHIFT;
+ __raw_writel(cs1, S5P_SROM_BW);
/* set timing for nCS1 suitable for ethernet chip */
- __raw_writel((0x1 << S5PV310_SROM_BCX__PMC__SHIFT) |
- (0x9 << S5PV310_SROM_BCX__TACP__SHIFT) |
- (0xc << S5PV310_SROM_BCX__TCAH__SHIFT) |
- (0x1 << S5PV310_SROM_BCX__TCOH__SHIFT) |
- (0x6 << S5PV310_SROM_BCX__TACC__SHIFT) |
- (0x1 << S5PV310_SROM_BCX__TCOS__SHIFT) |
- (0x1 << S5PV310_SROM_BCX__TACS__SHIFT), S5PV310_SROM_BC1);
+ __raw_writel((0x1 << S5P_SROM_BCX__PMC__SHIFT) |
+ (0x9 << S5P_SROM_BCX__TACP__SHIFT) |
+ (0xc << S5P_SROM_BCX__TCAH__SHIFT) |
+ (0x1 << S5P_SROM_BCX__TCOH__SHIFT) |
+ (0x6 << S5P_SROM_BCX__TACC__SHIFT) |
+ (0x1 << S5P_SROM_BCX__TCOS__SHIFT) |
+ (0x1 << S5P_SROM_BCX__TACS__SHIFT), S5P_SROM_BC1);
}
static void __init smdkv310_map_io(void)
@@ -182,6 +199,9 @@
static void __init smdkv310_machine_init(void)
{
+ s3c_i2c1_set_platdata(NULL);
+ i2c_register_board_info(1, i2c_devs1, ARRAY_SIZE(i2c_devs1));
+
smdkv310_smsc911x_init();
s3c_sdhci0_set_platdata(&smdkv310_hsmmc0_pdata);
diff --git a/arch/arm/mach-s5pv310/mach-universal_c210.c b/arch/arm/mach-s5pv310/mach-universal_c210.c
index 16d8fc0..36bc3cf 100644
--- a/arch/arm/mach-s5pv310/mach-universal_c210.c
+++ b/arch/arm/mach-s5pv310/mach-universal_c210.c
@@ -13,6 +13,9 @@
#include <linux/i2c.h>
#include <linux/gpio_keys.h>
#include <linux/gpio.h>
+#include <linux/regulator/machine.h>
+#include <linux/regulator/fixed.h>
+#include <linux/mmc/host.h>
#include <asm/mach/arch.h>
#include <asm/mach-types.h>
@@ -21,6 +24,7 @@
#include <plat/s5pv310.h>
#include <plat/cpu.h>
#include <plat/devs.h>
+#include <plat/sdhci.h>
#include <mach/map.h>
@@ -116,6 +120,73 @@
},
};
+/* eMMC */
+static struct s3c_sdhci_platdata universal_hsmmc0_data __initdata = {
+ .max_width = 8,
+ .host_caps = (MMC_CAP_8_BIT_DATA | MMC_CAP_4_BIT_DATA |
+ MMC_CAP_MMC_HIGHSPEED | MMC_CAP_SD_HIGHSPEED |
+ MMC_CAP_DISABLE),
+ .cd_type = S3C_SDHCI_CD_PERMANENT,
+ .clk_type = S3C_SDHCI_CLK_DIV_EXTERNAL,
+};
+
+static struct regulator_consumer_supply mmc0_supplies[] = {
+ REGULATOR_SUPPLY("vmmc", "s3c-sdhci.0"),
+};
+
+static struct regulator_init_data mmc0_fixed_voltage_init_data = {
+ .constraints = {
+ .name = "VMEM_VDD_2.8V",
+ .valid_ops_mask = REGULATOR_CHANGE_STATUS,
+ },
+ .num_consumer_supplies = ARRAY_SIZE(mmc0_supplies),
+ .consumer_supplies = mmc0_supplies,
+};
+
+static struct fixed_voltage_config mmc0_fixed_voltage_config = {
+ .supply_name = "MASSMEMORY_EN",
+ .microvolts = 2800000,
+ .gpio = S5PV310_GPE1(3),
+ .enable_high = true,
+ .init_data = &mmc0_fixed_voltage_init_data,
+};
+
+static struct platform_device mmc0_fixed_voltage = {
+ .name = "reg-fixed-voltage",
+ .id = 0,
+ .dev = {
+ .platform_data = &mmc0_fixed_voltage_config,
+ },
+};
+
+/* SD */
+static struct s3c_sdhci_platdata universal_hsmmc2_data __initdata = {
+ .max_width = 4,
+ .host_caps = MMC_CAP_4_BIT_DATA |
+ MMC_CAP_MMC_HIGHSPEED | MMC_CAP_SD_HIGHSPEED |
+ MMC_CAP_DISABLE,
+ .ext_cd_gpio = S5PV310_GPX3(4), /* XEINT_28 */
+ .ext_cd_gpio_invert = 1,
+ .cd_type = S3C_SDHCI_CD_GPIO,
+ .clk_type = S3C_SDHCI_CLK_DIV_EXTERNAL,
+};
+
+/* WiFi */
+static struct s3c_sdhci_platdata universal_hsmmc3_data __initdata = {
+ .max_width = 4,
+ .host_caps = MMC_CAP_4_BIT_DATA |
+ MMC_CAP_MMC_HIGHSPEED | MMC_CAP_SD_HIGHSPEED |
+ MMC_CAP_DISABLE,
+ .cd_type = S3C_SDHCI_CD_EXTERNAL,
+};
+
+static void __init universal_sdhci_init(void)
+{
+ s3c_sdhci0_set_platdata(&universal_hsmmc0_data);
+ s3c_sdhci2_set_platdata(&universal_hsmmc2_data);
+ s3c_sdhci3_set_platdata(&universal_hsmmc3_data);
+}
+
/* I2C0 */
static struct i2c_board_info i2c0_devs[] __initdata = {
/* Camera, To be updated */
@@ -127,6 +198,13 @@
};
static struct platform_device *universal_devices[] __initdata = {
+ /* Samsung Platform Devices */
+ &mmc0_fixed_voltage,
+ &s3c_device_hsmmc0,
+ &s3c_device_hsmmc2,
+ &s3c_device_hsmmc3,
+
+ /* Universal Devices */
&universal_gpio_keys,
&s5p_device_onenand,
};
@@ -140,6 +218,8 @@
static void __init universal_machine_init(void)
{
+ universal_sdhci_init();
+
i2c_register_board_info(0, i2c0_devs, ARRAY_SIZE(i2c0_devs));
i2c_register_board_info(1, i2c1_devs, ARRAY_SIZE(i2c1_devs));
diff --git a/arch/arm/mach-sa1100/generic.c b/arch/arm/mach-sa1100/generic.c
index 59d14f0..e21f347 100644
--- a/arch/arm/mach-sa1100/generic.c
+++ b/arch/arm/mach-sa1100/generic.c
@@ -21,7 +21,6 @@
#include <asm/div64.h>
#include <mach/hardware.h>
#include <asm/system.h>
-#include <asm/pgtable.h>
#include <asm/mach/map.h>
#include <asm/mach/flash.h>
#include <asm/irq.h>
diff --git a/arch/arm/mach-sa1100/irq.c b/arch/arm/mach-sa1100/irq.c
index 3093d46..3d85dfa 100644
--- a/arch/arm/mach-sa1100/irq.c
+++ b/arch/arm/mach-sa1100/irq.c
@@ -37,14 +37,14 @@
#define GPIO_11_27_IRQ(i) ((i) - 21)
#define GPIO11_27_MASK(irq) (1 << GPIO_11_27_IRQ(irq))
-static int sa1100_gpio_type(unsigned int irq, unsigned int type)
+static int sa1100_gpio_type(struct irq_data *d, unsigned int type)
{
unsigned int mask;
- if (irq <= 10)
- mask = 1 << irq;
+ if (d->irq <= 10)
+ mask = 1 << d->irq;
else
- mask = GPIO11_27_MASK(irq);
+ mask = GPIO11_27_MASK(d->irq);
if (type == IRQ_TYPE_PROBE) {
if ((GPIO_IRQ_rising_edge | GPIO_IRQ_falling_edge) & mask)
@@ -70,37 +70,37 @@
/*
* GPIO IRQs must be acknowledged. This is for IRQs from 0 to 10.
*/
-static void sa1100_low_gpio_ack(unsigned int irq)
+static void sa1100_low_gpio_ack(struct irq_data *d)
{
- GEDR = (1 << irq);
+ GEDR = (1 << d->irq);
}
-static void sa1100_low_gpio_mask(unsigned int irq)
+static void sa1100_low_gpio_mask(struct irq_data *d)
{
- ICMR &= ~(1 << irq);
+ ICMR &= ~(1 << d->irq);
}
-static void sa1100_low_gpio_unmask(unsigned int irq)
+static void sa1100_low_gpio_unmask(struct irq_data *d)
{
- ICMR |= 1 << irq;
+ ICMR |= 1 << d->irq;
}
-static int sa1100_low_gpio_wake(unsigned int irq, unsigned int on)
+static int sa1100_low_gpio_wake(struct irq_data *d, unsigned int on)
{
if (on)
- PWER |= 1 << irq;
+ PWER |= 1 << d->irq;
else
- PWER &= ~(1 << irq);
+ PWER &= ~(1 << d->irq);
return 0;
}
static struct irq_chip sa1100_low_gpio_chip = {
.name = "GPIO-l",
- .ack = sa1100_low_gpio_ack,
- .mask = sa1100_low_gpio_mask,
- .unmask = sa1100_low_gpio_unmask,
- .set_type = sa1100_gpio_type,
- .set_wake = sa1100_low_gpio_wake,
+ .irq_ack = sa1100_low_gpio_ack,
+ .irq_mask = sa1100_low_gpio_mask,
+ .irq_unmask = sa1100_low_gpio_unmask,
+ .irq_set_type = sa1100_gpio_type,
+ .irq_set_wake = sa1100_low_gpio_wake,
};
/*
@@ -139,16 +139,16 @@
* In addition, the IRQs are all collected up into one bit in the
* interrupt controller registers.
*/
-static void sa1100_high_gpio_ack(unsigned int irq)
+static void sa1100_high_gpio_ack(struct irq_data *d)
{
- unsigned int mask = GPIO11_27_MASK(irq);
+ unsigned int mask = GPIO11_27_MASK(d->irq);
GEDR = mask;
}
-static void sa1100_high_gpio_mask(unsigned int irq)
+static void sa1100_high_gpio_mask(struct irq_data *d)
{
- unsigned int mask = GPIO11_27_MASK(irq);
+ unsigned int mask = GPIO11_27_MASK(d->irq);
GPIO_IRQ_mask &= ~mask;
@@ -156,9 +156,9 @@
GFER &= ~mask;
}
-static void sa1100_high_gpio_unmask(unsigned int irq)
+static void sa1100_high_gpio_unmask(struct irq_data *d)
{
- unsigned int mask = GPIO11_27_MASK(irq);
+ unsigned int mask = GPIO11_27_MASK(d->irq);
GPIO_IRQ_mask |= mask;
@@ -166,44 +166,44 @@
GFER = GPIO_IRQ_falling_edge & GPIO_IRQ_mask;
}
-static int sa1100_high_gpio_wake(unsigned int irq, unsigned int on)
+static int sa1100_high_gpio_wake(struct irq_data *d, unsigned int on)
{
if (on)
- PWER |= GPIO11_27_MASK(irq);
+ PWER |= GPIO11_27_MASK(d->irq);
else
- PWER &= ~GPIO11_27_MASK(irq);
+ PWER &= ~GPIO11_27_MASK(d->irq);
return 0;
}
static struct irq_chip sa1100_high_gpio_chip = {
.name = "GPIO-h",
- .ack = sa1100_high_gpio_ack,
- .mask = sa1100_high_gpio_mask,
- .unmask = sa1100_high_gpio_unmask,
- .set_type = sa1100_gpio_type,
- .set_wake = sa1100_high_gpio_wake,
+ .irq_ack = sa1100_high_gpio_ack,
+ .irq_mask = sa1100_high_gpio_mask,
+ .irq_unmask = sa1100_high_gpio_unmask,
+ .irq_set_type = sa1100_gpio_type,
+ .irq_set_wake = sa1100_high_gpio_wake,
};
/*
* We don't need to ACK IRQs on the SA1100 unless they're GPIOs
* this is for internal IRQs i.e. from 11 to 31.
*/
-static void sa1100_mask_irq(unsigned int irq)
+static void sa1100_mask_irq(struct irq_data *d)
{
- ICMR &= ~(1 << irq);
+ ICMR &= ~(1 << d->irq);
}
-static void sa1100_unmask_irq(unsigned int irq)
+static void sa1100_unmask_irq(struct irq_data *d)
{
- ICMR |= (1 << irq);
+ ICMR |= (1 << d->irq);
}
/*
* Apart form GPIOs, only the RTC alarm can be a wakeup event.
*/
-static int sa1100_set_wake(unsigned int irq, unsigned int on)
+static int sa1100_set_wake(struct irq_data *d, unsigned int on)
{
- if (irq == IRQ_RTCAlrm) {
+ if (d->irq == IRQ_RTCAlrm) {
if (on)
PWER |= PWER_RTC;
else
@@ -215,10 +215,10 @@
static struct irq_chip sa1100_normal_chip = {
.name = "SC",
- .ack = sa1100_mask_irq,
- .mask = sa1100_mask_irq,
- .unmask = sa1100_unmask_irq,
- .set_wake = sa1100_set_wake,
+ .irq_ack = sa1100_mask_irq,
+ .irq_mask = sa1100_mask_irq,
+ .irq_unmask = sa1100_unmask_irq,
+ .irq_set_wake = sa1100_set_wake,
};
static struct resource irq_resource = {
diff --git a/arch/arm/mach-sa1100/neponset.c b/arch/arm/mach-sa1100/neponset.c
index c601a75..4aad01f 100644
--- a/arch/arm/mach-sa1100/neponset.c
+++ b/arch/arm/mach-sa1100/neponset.c
@@ -35,7 +35,7 @@
/*
* Acknowledge the parent IRQ.
*/
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
/*
* Read the interrupt reason register. Let's have all
@@ -53,7 +53,7 @@
* recheck the register for any pending IRQs.
*/
if (irr & (IRR_ETHERNET | IRR_USAR)) {
- desc->chip->mask(irq);
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
/*
* Ack the interrupt now to prevent re-entering
@@ -61,7 +61,7 @@
* since we'll check the IRR register prior to
* leaving.
*/
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
if (irr & IRR_ETHERNET) {
generic_handle_irq(IRQ_NEPONSET_SMC9196);
@@ -71,7 +71,7 @@
generic_handle_irq(IRQ_NEPONSET_USAR);
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
if (irr & IRR_SA1111) {
diff --git a/arch/arm/mach-shark/irq.c b/arch/arm/mach-shark/irq.c
index c04eb6a..831fc66 100644
--- a/arch/arm/mach-shark/irq.c
+++ b/arch/arm/mach-shark/irq.c
@@ -30,35 +30,35 @@
* These have to be protected by the irq controller spinlock
* before being called.
*/
-static void shark_disable_8259A_irq(unsigned int irq)
+static void shark_disable_8259A_irq(struct irq_data *d)
{
unsigned int mask;
- if (irq<8) {
- mask = 1 << irq;
+ if (d->irq<8) {
+ mask = 1 << d->irq;
cached_irq_mask[0] |= mask;
outb(cached_irq_mask[1],0xA1);
} else {
- mask = 1 << (irq-8);
+ mask = 1 << (d->irq-8);
cached_irq_mask[1] |= mask;
outb(cached_irq_mask[0],0x21);
}
}
-static void shark_enable_8259A_irq(unsigned int irq)
+static void shark_enable_8259A_irq(struct irq_data *d)
{
unsigned int mask;
- if (irq<8) {
- mask = ~(1 << irq);
+ if (d->irq<8) {
+ mask = ~(1 << d->irq);
cached_irq_mask[0] &= mask;
outb(cached_irq_mask[0],0x21);
} else {
- mask = ~(1 << (irq-8));
+ mask = ~(1 << (d->irq-8));
cached_irq_mask[1] &= mask;
outb(cached_irq_mask[1],0xA1);
}
}
-static void shark_ack_8259A_irq(unsigned int irq){}
+static void shark_ack_8259A_irq(struct irq_data *d){}
static irqreturn_t bogus_int(int irq, void *dev_id)
{
@@ -69,10 +69,10 @@
static struct irqaction cascade;
static struct irq_chip fb_chip = {
- .name = "XT-PIC",
- .ack = shark_ack_8259A_irq,
- .mask = shark_disable_8259A_irq,
- .unmask = shark_enable_8259A_irq,
+ .name = "XT-PIC",
+ .irq_ack = shark_ack_8259A_irq,
+ .irq_mask = shark_disable_8259A_irq,
+ .irq_unmask = shark_enable_8259A_irq,
};
void __init shark_init_irq(void)
diff --git a/arch/arm/mach-stmp378x/stmp378x.c b/arch/arm/mach-stmp378x/stmp378x.c
index ddd49a7..c2f9fe04 100644
--- a/arch/arm/mach-stmp378x/stmp378x.c
+++ b/arch/arm/mach-stmp378x/stmp378x.c
@@ -47,7 +47,7 @@
/*
* IRQ handling
*/
-static void stmp378x_ack_irq(unsigned int irq)
+static void stmp378x_ack_irq(struct irq_data *d)
{
/* Tell ICOLL to release IRQ line */
__raw_writel(0, REGS_ICOLL_BASE + HW_ICOLL_VECTOR);
@@ -60,24 +60,24 @@
(void)__raw_readl(REGS_ICOLL_BASE + HW_ICOLL_STAT);
}
-static void stmp378x_mask_irq(unsigned int irq)
+static void stmp378x_mask_irq(struct irq_data *d)
{
/* IRQ disable */
stmp3xxx_clearl(BM_ICOLL_INTERRUPTn_ENABLE,
- REGS_ICOLL_BASE + HW_ICOLL_INTERRUPTn + irq * 0x10);
+ REGS_ICOLL_BASE + HW_ICOLL_INTERRUPTn + d->irq * 0x10);
}
-static void stmp378x_unmask_irq(unsigned int irq)
+static void stmp378x_unmask_irq(struct irq_data *d)
{
/* IRQ enable */
stmp3xxx_setl(BM_ICOLL_INTERRUPTn_ENABLE,
- REGS_ICOLL_BASE + HW_ICOLL_INTERRUPTn + irq * 0x10);
+ REGS_ICOLL_BASE + HW_ICOLL_INTERRUPTn + d->irq * 0x10);
}
static struct irq_chip stmp378x_chip = {
- .ack = stmp378x_ack_irq,
- .mask = stmp378x_mask_irq,
- .unmask = stmp378x_unmask_irq,
+ .irq_ack = stmp378x_ack_irq,
+ .irq_mask = stmp378x_mask_irq,
+ .irq_unmask = stmp378x_unmask_irq,
};
void __init stmp378x_init_irq(void)
diff --git a/arch/arm/mach-stmp37xx/stmp37xx.c b/arch/arm/mach-stmp37xx/stmp37xx.c
index 8c7d6fb..a9aed06 100644
--- a/arch/arm/mach-stmp37xx/stmp37xx.c
+++ b/arch/arm/mach-stmp37xx/stmp37xx.c
@@ -43,11 +43,11 @@
/*
* IRQ handling
*/
-static void stmp37xx_ack_irq(unsigned int irq)
+static void stmp37xx_ack_irq(struct irq_data *d)
{
/* Disable IRQ */
- stmp3xxx_clearl(0x04 << ((irq % 4) * 8),
- REGS_ICOLL_BASE + HW_ICOLL_PRIORITYn + irq / 4 * 0x10);
+ stmp3xxx_clearl(0x04 << ((d->irq % 4) * 8),
+ REGS_ICOLL_BASE + HW_ICOLL_PRIORITYn + d->irq / 4 * 0x10);
/* ACK current interrupt */
__raw_writel(1, REGS_ICOLL_BASE + HW_ICOLL_LEVELACK);
@@ -56,24 +56,24 @@
(void)__raw_readl(REGS_ICOLL_BASE + HW_ICOLL_STAT);
}
-static void stmp37xx_mask_irq(unsigned int irq)
+static void stmp37xx_mask_irq(struct irq_data *d)
{
/* IRQ disable */
- stmp3xxx_clearl(0x04 << ((irq % 4) * 8),
- REGS_ICOLL_BASE + HW_ICOLL_PRIORITYn + irq / 4 * 0x10);
+ stmp3xxx_clearl(0x04 << ((d->irq % 4) * 8),
+ REGS_ICOLL_BASE + HW_ICOLL_PRIORITYn + d->irq / 4 * 0x10);
}
-static void stmp37xx_unmask_irq(unsigned int irq)
+static void stmp37xx_unmask_irq(struct irq_data *d)
{
/* IRQ enable */
- stmp3xxx_setl(0x04 << ((irq % 4) * 8),
- REGS_ICOLL_BASE + HW_ICOLL_PRIORITYn + irq / 4 * 0x10);
+ stmp3xxx_setl(0x04 << ((d->irq % 4) * 8),
+ REGS_ICOLL_BASE + HW_ICOLL_PRIORITYn + d->irq / 4 * 0x10);
}
static struct irq_chip stmp37xx_chip = {
- .ack = stmp37xx_ack_irq,
- .mask = stmp37xx_mask_irq,
- .unmask = stmp37xx_unmask_irq,
+ .irq_ack = stmp37xx_ack_irq,
+ .irq_mask = stmp37xx_mask_irq,
+ .irq_unmask = stmp37xx_unmask_irq,
};
void __init stmp37xx_init_irq(void)
diff --git a/arch/arm/mach-tcc8k/irq.c b/arch/arm/mach-tcc8k/irq.c
index 34575c4..aa9231f 100644
--- a/arch/arm/mach-tcc8k/irq.c
+++ b/arch/arm/mach-tcc8k/irq.c
@@ -18,65 +18,65 @@
#include "common.h"
/* Disable IRQ */
-static void tcc8000_mask_ack_irq0(unsigned int irq)
+static void tcc8000_mask_ack_irq0(struct irq_data *d)
{
- PIC0_IEN &= ~(1 << irq);
- PIC0_CREQ |= (1 << irq);
+ PIC0_IEN &= ~(1 << d->irq);
+ PIC0_CREQ |= (1 << d->irq);
}
-static void tcc8000_mask_ack_irq1(unsigned int irq)
+static void tcc8000_mask_ack_irq1(struct irq_data *d)
{
- PIC1_IEN &= ~(1 << (irq - 32));
- PIC1_CREQ |= (1 << (irq - 32));
+ PIC1_IEN &= ~(1 << (d->irq - 32));
+ PIC1_CREQ |= (1 << (d->irq - 32));
}
-static void tcc8000_mask_irq0(unsigned int irq)
+static void tcc8000_mask_irq0(struct irq_data *d)
{
- PIC0_IEN &= ~(1 << irq);
+ PIC0_IEN &= ~(1 << d->irq);
}
-static void tcc8000_mask_irq1(unsigned int irq)
+static void tcc8000_mask_irq1(struct irq_data *d)
{
- PIC1_IEN &= ~(1 << (irq - 32));
+ PIC1_IEN &= ~(1 << (d->irq - 32));
}
-static void tcc8000_ack_irq0(unsigned int irq)
+static void tcc8000_ack_irq0(struct irq_data *d)
{
- PIC0_CREQ |= (1 << irq);
+ PIC0_CREQ |= (1 << d->irq);
}
-static void tcc8000_ack_irq1(unsigned int irq)
+static void tcc8000_ack_irq1(struct irq_data *d)
{
- PIC1_CREQ |= (1 << (irq - 32));
+ PIC1_CREQ |= (1 << (d->irq - 32));
}
/* Enable IRQ */
-static void tcc8000_unmask_irq0(unsigned int irq)
+static void tcc8000_unmask_irq0(struct irq_data *d)
{
- PIC0_IEN |= (1 << irq);
- PIC0_INTOEN |= (1 << irq);
+ PIC0_IEN |= (1 << d->irq);
+ PIC0_INTOEN |= (1 << d->irq);
}
-static void tcc8000_unmask_irq1(unsigned int irq)
+static void tcc8000_unmask_irq1(struct irq_data *d)
{
- PIC1_IEN |= (1 << (irq - 32));
- PIC1_INTOEN |= (1 << (irq - 32));
+ PIC1_IEN |= (1 << (d->irq - 32));
+ PIC1_INTOEN |= (1 << (d->irq - 32));
}
static struct irq_chip tcc8000_irq_chip0 = {
.name = "tcc_irq0",
- .mask = tcc8000_mask_irq0,
- .ack = tcc8000_ack_irq0,
- .mask_ack = tcc8000_mask_ack_irq0,
- .unmask = tcc8000_unmask_irq0,
+ .irq_mask = tcc8000_mask_irq0,
+ .irq_ack = tcc8000_ack_irq0,
+ .irq_mask_ack = tcc8000_mask_ack_irq0,
+ .irq_unmask = tcc8000_unmask_irq0,
};
static struct irq_chip tcc8000_irq_chip1 = {
.name = "tcc_irq1",
- .mask = tcc8000_mask_irq1,
- .ack = tcc8000_ack_irq1,
- .mask_ack = tcc8000_mask_ack_irq1,
- .unmask = tcc8000_unmask_irq1,
+ .irq_mask = tcc8000_mask_irq1,
+ .irq_ack = tcc8000_ack_irq1,
+ .irq_mask_ack = tcc8000_mask_ack_irq1,
+ .irq_unmask = tcc8000_unmask_irq1,
};
void __init tcc8k_init_irq(void)
diff --git a/arch/arm/mach-tegra/gpio.c b/arch/arm/mach-tegra/gpio.c
index 0775265..bd06620 100644
--- a/arch/arm/mach-tegra/gpio.c
+++ b/arch/arm/mach-tegra/gpio.c
@@ -142,31 +142,31 @@
.ngpio = TEGRA_NR_GPIOS,
};
-static void tegra_gpio_irq_ack(unsigned int irq)
+static void tegra_gpio_irq_ack(struct irq_data *d)
{
- int gpio = irq - INT_GPIO_BASE;
+ int gpio = d->irq - INT_GPIO_BASE;
__raw_writel(1 << GPIO_BIT(gpio), GPIO_INT_CLR(gpio));
}
-static void tegra_gpio_irq_mask(unsigned int irq)
+static void tegra_gpio_irq_mask(struct irq_data *d)
{
- int gpio = irq - INT_GPIO_BASE;
+ int gpio = d->irq - INT_GPIO_BASE;
tegra_gpio_mask_write(GPIO_MSK_INT_ENB(gpio), gpio, 0);
}
-static void tegra_gpio_irq_unmask(unsigned int irq)
+static void tegra_gpio_irq_unmask(struct irq_data *d)
{
- int gpio = irq - INT_GPIO_BASE;
+ int gpio = d->irq - INT_GPIO_BASE;
tegra_gpio_mask_write(GPIO_MSK_INT_ENB(gpio), gpio, 1);
}
-static int tegra_gpio_irq_set_type(unsigned int irq, unsigned int type)
+static int tegra_gpio_irq_set_type(struct irq_data *d, unsigned int type)
{
- int gpio = irq - INT_GPIO_BASE;
- struct tegra_gpio_bank *bank = get_irq_chip_data(irq);
+ int gpio = d->irq - INT_GPIO_BASE;
+ struct tegra_gpio_bank *bank = irq_data_get_irq_chip_data(d);
int port = GPIO_PORT(gpio);
int lvl_type;
int val;
@@ -221,7 +221,7 @@
int pin;
int unmasked = 0;
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
bank = get_irq_data(irq);
@@ -240,7 +240,7 @@
*/
if (lvl & (0x100 << pin)) {
unmasked = 1;
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
generic_handle_irq(gpio_to_irq(gpio + pin));
@@ -248,7 +248,7 @@
}
if (!unmasked)
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
@@ -316,21 +316,21 @@
local_irq_restore(flags);
}
-static int tegra_gpio_wake_enable(unsigned int irq, unsigned int enable)
+static int tegra_gpio_wake_enable(struct irq_data *d, unsigned int enable)
{
- struct tegra_gpio_bank *bank = get_irq_chip_data(irq);
+ struct tegra_gpio_bank *bank = irq_data_get_irq_chip_data(d);
return set_irq_wake(bank->irq, enable);
}
#endif
static struct irq_chip tegra_gpio_irq_chip = {
.name = "GPIO",
- .ack = tegra_gpio_irq_ack,
- .mask = tegra_gpio_irq_mask,
- .unmask = tegra_gpio_irq_unmask,
- .set_type = tegra_gpio_irq_set_type,
+ .irq_ack = tegra_gpio_irq_ack,
+ .irq_mask = tegra_gpio_irq_mask,
+ .irq_unmask = tegra_gpio_irq_unmask,
+ .irq_set_type = tegra_gpio_irq_set_type,
#ifdef CONFIG_PM
- .set_wake = tegra_gpio_wake_enable,
+ .irq_set_wake = tegra_gpio_wake_enable,
#endif
};
diff --git a/arch/arm/mach-tegra/hotplug.c b/arch/arm/mach-tegra/hotplug.c
index a5cb1ce..f329404 100644
--- a/arch/arm/mach-tegra/hotplug.c
+++ b/arch/arm/mach-tegra/hotplug.c
@@ -26,10 +26,10 @@
* Turn off coherency
*/
" mrc p15, 0, %0, c1, c0, 1\n"
- " bic %0, %0, %2\n"
+ " bic %0, %0, #0x20\n"
" mcr p15, 0, %0, c1, c0, 1\n"
" mrc p15, 0, %0, c1, c0, 0\n"
- " bic %0, %0, #0x04\n"
+ " bic %0, %0, %2\n"
" mcr p15, 0, %0, c1, c0, 0\n"
: "=&r" (v)
: "r" (0), "Ir" (CR_C)
diff --git a/arch/arm/mach-tegra/irq.c b/arch/arm/mach-tegra/irq.c
index 5407de0..de7dfad 100644
--- a/arch/arm/mach-tegra/irq.c
+++ b/arch/arm/mach-tegra/irq.c
@@ -46,30 +46,30 @@
#define ICTLR_COP_IER_CLR 0x38
#define ICTLR_COP_IEP_CLASS 0x3c
-static void (*gic_mask_irq)(unsigned int irq);
-static void (*gic_unmask_irq)(unsigned int irq);
+static void (*gic_mask_irq)(struct irq_data *d);
+static void (*gic_unmask_irq)(struct irq_data *d);
#define irq_to_ictlr(irq) (((irq)-32) >> 5)
static void __iomem *tegra_ictlr_base = IO_ADDRESS(TEGRA_PRIMARY_ICTLR_BASE);
#define ictlr_to_virt(ictlr) (tegra_ictlr_base + (ictlr)*0x100)
-static void tegra_mask(unsigned int irq)
+static void tegra_mask(struct irq_data *d)
{
- void __iomem *addr = ictlr_to_virt(irq_to_ictlr(irq));
- gic_mask_irq(irq);
- writel(1<<(irq&31), addr+ICTLR_CPU_IER_CLR);
+ void __iomem *addr = ictlr_to_virt(irq_to_ictlr(d->irq));
+ gic_mask_irq(d);
+ writel(1<<(d->irq&31), addr+ICTLR_CPU_IER_CLR);
}
-static void tegra_unmask(unsigned int irq)
+static void tegra_unmask(struct irq_data *d)
{
- void __iomem *addr = ictlr_to_virt(irq_to_ictlr(irq));
- gic_unmask_irq(irq);
- writel(1<<(irq&31), addr+ICTLR_CPU_IER_SET);
+ void __iomem *addr = ictlr_to_virt(irq_to_ictlr(d->irq));
+ gic_unmask_irq(d);
+ writel(1<<(d->irq&31), addr+ICTLR_CPU_IER_SET);
}
#ifdef CONFIG_PM
-static int tegra_set_wake(unsigned int irq, unsigned int on)
+static int tegra_set_wake(struct irq_data *d, unsigned int on)
{
return 0;
}
@@ -77,10 +77,10 @@
static struct irq_chip tegra_irq = {
.name = "PPI",
- .mask = tegra_mask,
- .unmask = tegra_unmask,
+ .irq_mask = tegra_mask,
+ .irq_unmask = tegra_unmask,
#ifdef CONFIG_PM
- .set_wake = tegra_set_wake,
+ .irq_set_wake = tegra_set_wake,
#endif
};
@@ -98,11 +98,11 @@
IO_ADDRESS(TEGRA_ARM_PERIF_BASE + 0x100));
gic = get_irq_chip(29);
- gic_unmask_irq = gic->unmask;
- gic_mask_irq = gic->mask;
- tegra_irq.ack = gic->ack;
+ gic_unmask_irq = gic->irq_unmask;
+ gic_mask_irq = gic->irq_mask;
+ tegra_irq.irq_ack = gic->irq_ack;
#ifdef CONFIG_SMP
- tegra_irq.set_affinity = gic->set_affinity;
+ tegra_irq.irq_set_affinity = gic->irq_set_affinity;
#endif
for (i = INT_PRI_BASE; i < INT_GPIO_BASE; i++) {
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index 13a83e4..136c32e 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -63,23 +63,25 @@
#define VA_VIC_BASE __io_address(VERSATILE_VIC_BASE)
#define VA_SIC_BASE __io_address(VERSATILE_SIC_BASE)
-static void sic_mask_irq(unsigned int irq)
+static void sic_mask_irq(struct irq_data *d)
{
- irq -= IRQ_SIC_START;
+ unsigned int irq = d->irq - IRQ_SIC_START;
+
writel(1 << irq, VA_SIC_BASE + SIC_IRQ_ENABLE_CLEAR);
}
-static void sic_unmask_irq(unsigned int irq)
+static void sic_unmask_irq(struct irq_data *d)
{
- irq -= IRQ_SIC_START;
+ unsigned int irq = d->irq - IRQ_SIC_START;
+
writel(1 << irq, VA_SIC_BASE + SIC_IRQ_ENABLE_SET);
}
static struct irq_chip sic_chip = {
- .name = "SIC",
- .ack = sic_mask_irq,
- .mask = sic_mask_irq,
- .unmask = sic_unmask_irq,
+ .name = "SIC",
+ .irq_ack = sic_mask_irq,
+ .irq_mask = sic_mask_irq,
+ .irq_unmask = sic_unmask_irq,
};
static void
diff --git a/arch/arm/mach-w90x900/irq.c b/arch/arm/mach-w90x900/irq.c
index 0ce9d8e..9c35010 100644
--- a/arch/arm/mach-w90x900/irq.c
+++ b/arch/arm/mach-w90x900/irq.c
@@ -92,15 +92,15 @@
__raw_writel(regval, REG_AIC_GEN);
}
-static void nuc900_irq_mask(unsigned int irq)
+static void nuc900_irq_mask(struct irq_data *d)
{
struct group_irq *group_irq;
group_irq = NULL;
- __raw_writel(1 << irq, REG_AIC_MDCR);
+ __raw_writel(1 << d->irq, REG_AIC_MDCR);
- switch (irq) {
+ switch (d->irq) {
case IRQ_GROUP0:
group_irq = &group_nirq0;
break;
@@ -143,20 +143,20 @@
* to REG_AIC_EOSCR for ACK
*/
-static void nuc900_irq_ack(unsigned int irq)
+static void nuc900_irq_ack(struct irq_data *d)
{
__raw_writel(0x01, REG_AIC_EOSCR);
}
-static void nuc900_irq_unmask(unsigned int irq)
+static void nuc900_irq_unmask(struct irq_data *d)
{
struct group_irq *group_irq;
group_irq = NULL;
- __raw_writel(1 << irq, REG_AIC_MECR);
+ __raw_writel(1 << d->irq, REG_AIC_MECR);
- switch (irq) {
+ switch (d->irq) {
case IRQ_GROUP0:
group_irq = &group_nirq0;
break;
@@ -195,9 +195,9 @@
}
static struct irq_chip nuc900_irq_chip = {
- .ack = nuc900_irq_ack,
- .mask = nuc900_irq_mask,
- .unmask = nuc900_irq_unmask,
+ .irq_ack = nuc900_irq_ack,
+ .irq_mask = nuc900_irq_mask,
+ .irq_unmask = nuc900_irq_unmask,
};
void __init nuc900_init_irq(void)
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index fcc1e62..9d30c6f 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -644,7 +644,7 @@
config SWP_EMULATE
bool "Emulate SWP/SWPB instructions"
- depends on CPU_V7
+ depends on CPU_V7 && !CPU_V6
select HAVE_PROC_CPU if PROC_FS
default y if SMP
help
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 6b48e0a..4771dba 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -577,7 +577,7 @@
* dma_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg
* @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
* @sg: list of buffers
- * @nents: number of buffers to unmap (returned from dma_map_sg)
+ * @nents: number of buffers to unmap (same as was passed to dma_map_sg)
* @dir: DMA transfer direction (same as was passed to dma_map_sg)
*
* Unmap a set of streaming mode DMA translations. Again, CPU access
diff --git a/arch/arm/mm/pgd.c b/arch/arm/mm/pgd.c
index 93292a1..709244c 100644
--- a/arch/arm/mm/pgd.c
+++ b/arch/arm/mm/pgd.c
@@ -50,7 +50,7 @@
if (!new_pmd)
goto no_pmd;
- new_pte = pte_alloc_map(mm, new_pmd, 0);
+ new_pte = pte_alloc_map(mm, NULL, new_pmd, 0);
if (!new_pte)
goto no_pte;
diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S
index b49fab2..0c1172b 100644
--- a/arch/arm/mm/proc-v7.S
+++ b/arch/arm/mm/proc-v7.S
@@ -159,7 +159,9 @@
tstne r1, #L_PTE_PRESENT
moveq r3, #0
- str r3, [r0, #2048]!
+ ARM( str r3, [r0, #2048]! )
+ THUMB( add r0, r0, #2048 )
+ THUMB( str r3, [r0] )
mcr p15, 0, r0, c7, c10, 1 @ flush_pte
#endif
mov pc, lr
diff --git a/arch/arm/plat-mxc/3ds_debugboard.c b/arch/arm/plat-mxc/3ds_debugboard.c
index 639c54a..c856fa3 100644
--- a/arch/arm/plat-mxc/3ds_debugboard.c
+++ b/arch/arm/plat-mxc/3ds_debugboard.c
@@ -60,7 +60,6 @@
#define EXPIO_INT_BUTTON_B (MXC_BOARD_IRQ_START + 4)
static void __iomem *brd_io;
-static void expio_ack_irq(u32 irq);
static struct resource smsc911x_resources[] = {
{
@@ -93,7 +92,8 @@
u32 int_valid;
u32 expio_irq;
- desc->chip->mask(irq); /* irq = gpio irq number */
+ /* irq = gpio irq number */
+ desc->irq_data.chip->irq_mask(&desc->irq_data);
imr_val = __raw_readw(brd_io + INTR_MASK_REG);
int_valid = __raw_readw(brd_io + INTR_STATUS_REG) & ~imr_val;
@@ -110,37 +110,37 @@
d->handle_irq(expio_irq, d);
}
- desc->chip->ack(irq);
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
/*
* Disable an expio pin's interrupt by setting the bit in the imr.
* Irq is an expio virtual irq number
*/
-static void expio_mask_irq(u32 irq)
+static void expio_mask_irq(struct irq_data *d)
{
u16 reg;
- u32 expio = MXC_IRQ_TO_EXPIO(irq);
+ u32 expio = MXC_IRQ_TO_EXPIO(d->irq);
reg = __raw_readw(brd_io + INTR_MASK_REG);
reg |= (1 << expio);
__raw_writew(reg, brd_io + INTR_MASK_REG);
}
-static void expio_ack_irq(u32 irq)
+static void expio_ack_irq(struct irq_data *d)
{
- u32 expio = MXC_IRQ_TO_EXPIO(irq);
+ u32 expio = MXC_IRQ_TO_EXPIO(d->irq);
__raw_writew(1 << expio, brd_io + INTR_RESET_REG);
__raw_writew(0, brd_io + INTR_RESET_REG);
- expio_mask_irq(irq);
+ expio_mask_irq(d);
}
-static void expio_unmask_irq(u32 irq)
+static void expio_unmask_irq(struct irq_data *d)
{
u16 reg;
- u32 expio = MXC_IRQ_TO_EXPIO(irq);
+ u32 expio = MXC_IRQ_TO_EXPIO(d->irq);
reg = __raw_readw(brd_io + INTR_MASK_REG);
reg &= ~(1 << expio);
@@ -148,9 +148,9 @@
}
static struct irq_chip expio_irq_chip = {
- .ack = expio_ack_irq,
- .mask = expio_mask_irq,
- .unmask = expio_unmask_irq,
+ .irq_ack = expio_ack_irq,
+ .irq_mask = expio_mask_irq,
+ .irq_unmask = expio_unmask_irq,
};
int __init mxc_expio_init(u32 base, u32 p_irq)
diff --git a/arch/arm/plat-mxc/avic.c b/arch/arm/plat-mxc/avic.c
index 9a4e8a2..deb284b 100644
--- a/arch/arm/plat-mxc/avic.c
+++ b/arch/arm/plat-mxc/avic.c
@@ -89,22 +89,22 @@
#endif /* CONFIG_FIQ */
/* Disable interrupt number "irq" in the AVIC */
-static void mxc_mask_irq(unsigned int irq)
+static void mxc_mask_irq(struct irq_data *d)
{
- __raw_writel(irq, avic_base + AVIC_INTDISNUM);
+ __raw_writel(d->irq, avic_base + AVIC_INTDISNUM);
}
/* Enable interrupt number "irq" in the AVIC */
-static void mxc_unmask_irq(unsigned int irq)
+static void mxc_unmask_irq(struct irq_data *d)
{
- __raw_writel(irq, avic_base + AVIC_INTENNUM);
+ __raw_writel(d->irq, avic_base + AVIC_INTENNUM);
}
static struct mxc_irq_chip mxc_avic_chip = {
.base = {
- .ack = mxc_mask_irq,
- .mask = mxc_mask_irq,
- .unmask = mxc_unmask_irq,
+ .irq_ack = mxc_mask_irq,
+ .irq_mask = mxc_mask_irq,
+ .irq_unmask = mxc_unmask_irq,
},
#ifdef CONFIG_MXC_IRQ_PRIOR
.set_priority = avic_irq_set_priority,
diff --git a/arch/arm/plat-mxc/devices/Kconfig b/arch/arm/plat-mxc/devices/Kconfig
index 2537166..b9ab1d5 100644
--- a/arch/arm/plat-mxc/devices/Kconfig
+++ b/arch/arm/plat-mxc/devices/Kconfig
@@ -1,6 +1,6 @@
config IMX_HAVE_PLATFORM_FEC
bool
- default y if ARCH_MX25 || SOC_IMX27 || SOC_IMX35 || SOC_IMX51
+ default y if ARCH_MX25 || SOC_IMX27 || SOC_IMX35 || SOC_IMX51 || SOC_IMX53
config IMX_HAVE_PLATFORM_FLEXCAN
select HAVE_CAN_FLEXCAN if CAN
diff --git a/arch/arm/plat-mxc/devices/platform-fec.c b/arch/arm/plat-mxc/devices/platform-fec.c
index 269ec78..b50c351 100644
--- a/arch/arm/plat-mxc/devices/platform-fec.c
+++ b/arch/arm/plat-mxc/devices/platform-fec.c
@@ -36,6 +36,11 @@
imx_fec_data_entry_single(MX51);
#endif
+#ifdef CONFIG_SOC_IMX53
+const struct imx_fec_data imx53_fec_data __initconst =
+ imx_fec_data_entry_single(MX53);
+#endif
+
struct platform_device *__init imx_add_fec(
const struct imx_fec_data *data,
const struct fec_platform_data *pdata)
diff --git a/arch/arm/plat-mxc/devices/platform-imx-i2c.c b/arch/arm/plat-mxc/devices/platform-imx-i2c.c
index 72ba880..7ba94e1 100644
--- a/arch/arm/plat-mxc/devices/platform-imx-i2c.c
+++ b/arch/arm/plat-mxc/devices/platform-imx-i2c.c
@@ -78,6 +78,15 @@
};
#endif /* ifdef CONFIG_SOC_IMX51 */
+#ifdef CONFIG_SOC_IMX53
+const struct imx_imx_i2c_data imx53_imx_i2c_data[] __initconst = {
+#define imx53_imx_i2c_data_entry(_id, _hwid) \
+ imx_imx_i2c_data_entry(MX53, _id, _hwid, SZ_4K)
+ imx53_imx_i2c_data_entry(0, 1),
+ imx53_imx_i2c_data_entry(1, 2),
+};
+#endif /* ifdef CONFIG_SOC_IMX51 */
+
struct platform_device *__init imx_add_imx_i2c(
const struct imx_imx_i2c_data *data,
const struct imxi2c_platform_data *pdata)
diff --git a/arch/arm/plat-mxc/devices/platform-imx-keypad.c b/arch/arm/plat-mxc/devices/platform-imx-keypad.c
index 40238f0..2636611 100644
--- a/arch/arm/plat-mxc/devices/platform-imx-keypad.c
+++ b/arch/arm/plat-mxc/devices/platform-imx-keypad.c
@@ -41,6 +41,11 @@
imx_imx_keypad_data_entry_single(MX35, SZ_16);
#endif /* ifdef CONFIG_SOC_IMX35 */
+#ifdef CONFIG_SOC_IMX51
+const struct imx_imx_keypad_data imx51_imx_keypad_data __initconst =
+ imx_imx_keypad_data_entry_single(MX51, SZ_16);
+#endif /* ifdef CONFIG_SOC_IMX51 */
+
struct platform_device *__init imx_add_imx_keypad(
const struct imx_imx_keypad_data *data,
const struct matrix_keymap_data *pdata)
diff --git a/arch/arm/plat-mxc/devices/platform-mxc_pwm.c b/arch/arm/plat-mxc/devices/platform-mxc_pwm.c
index 3d8ebdb..b0c4ae2 100644
--- a/arch/arm/plat-mxc/devices/platform-mxc_pwm.c
+++ b/arch/arm/plat-mxc/devices/platform-mxc_pwm.c
@@ -40,6 +40,15 @@
imx_mxc_pwm_data_entry_single(MX27, 0, , SZ_4K);
#endif /* ifdef CONFIG_SOC_IMX27 */
+#ifdef CONFIG_SOC_IMX51
+const struct imx_mxc_pwm_data imx51_mxc_pwm_data[] __initconst = {
+#define imx51_mxc_pwm_data_entry(_id, _hwid) \
+ imx_mxc_pwm_data_entry(MX51, _id, _hwid, SZ_16K)
+ imx51_mxc_pwm_data_entry(0, 1),
+ imx51_mxc_pwm_data_entry(1, 2),
+};
+#endif /* ifdef CONFIG_SOC_IMX51 */
+
struct platform_device *__init imx_add_mxc_pwm(
const struct imx_mxc_pwm_data *data)
{
diff --git a/arch/arm/plat-mxc/devices/platform-sdhci-esdhc-imx.c b/arch/arm/plat-mxc/devices/platform-sdhci-esdhc-imx.c
index b352564..6b2940b 100644
--- a/arch/arm/plat-mxc/devices/platform-sdhci-esdhc-imx.c
+++ b/arch/arm/plat-mxc/devices/platform-sdhci-esdhc-imx.c
@@ -53,6 +53,18 @@
};
#endif /* ifdef CONFIG_SOC_IMX51 */
+#ifdef CONFIG_SOC_IMX53
+const struct imx_sdhci_esdhc_imx_data
+imx53_sdhci_esdhc_imx_data[] __initconst = {
+#define imx53_sdhci_esdhc_imx_data_entry(_id, _hwid) \
+ imx_sdhci_esdhc_imx_data_entry(MX53, _id, _hwid)
+ imx53_sdhci_esdhc_imx_data_entry(0, 1),
+ imx53_sdhci_esdhc_imx_data_entry(1, 2),
+ imx53_sdhci_esdhc_imx_data_entry(2, 3),
+ imx53_sdhci_esdhc_imx_data_entry(3, 4),
+};
+#endif /* ifdef CONFIG_SOC_IMX53 */
+
struct platform_device *__init imx_add_sdhci_esdhc_imx(
const struct imx_sdhci_esdhc_imx_data *data,
const struct esdhc_platform_data *pdata)
diff --git a/arch/arm/plat-mxc/devices/platform-spi_imx.c b/arch/arm/plat-mxc/devices/platform-spi_imx.c
index 8ea49ad..013c85f 100644
--- a/arch/arm/plat-mxc/devices/platform-spi_imx.c
+++ b/arch/arm/plat-mxc/devices/platform-spi_imx.c
@@ -81,6 +81,18 @@
};
#endif /* ifdef CONFIG_SOC_IMX51 */
+#ifdef CONFIG_SOC_IMX53
+const struct imx_spi_imx_data imx53_cspi_data __initconst =
+ imx_spi_imx_data_entry_single(MX53, CSPI, "imx53-cspi", 0, , SZ_4K);
+
+const struct imx_spi_imx_data imx53_ecspi_data[] __initconst = {
+#define imx53_ecspi_data_entry(_id, _hwid) \
+ imx_spi_imx_data_entry(MX53, ECSPI, "imx53-ecspi", _id, _hwid, SZ_4K)
+ imx53_ecspi_data_entry(0, 1),
+ imx53_ecspi_data_entry(1, 2),
+};
+#endif /* ifdef CONFIG_SOC_IMX53 */
+
struct platform_device *__init imx_add_spi_imx(
const struct imx_spi_imx_data *data,
const struct spi_imx_master *pdata)
diff --git a/arch/arm/plat-mxc/gpio.c b/arch/arm/plat-mxc/gpio.c
index bc2c7bc..d17b3c9 100644
--- a/arch/arm/plat-mxc/gpio.c
+++ b/arch/arm/plat-mxc/gpio.c
@@ -63,29 +63,29 @@
__raw_writel(l, port->base + GPIO_IMR);
}
-static void gpio_ack_irq(u32 irq)
+static void gpio_ack_irq(struct irq_data *d)
{
- u32 gpio = irq_to_gpio(irq);
+ u32 gpio = irq_to_gpio(d->irq);
_clear_gpio_irqstatus(&mxc_gpio_ports[gpio / 32], gpio & 0x1f);
}
-static void gpio_mask_irq(u32 irq)
+static void gpio_mask_irq(struct irq_data *d)
{
- u32 gpio = irq_to_gpio(irq);
+ u32 gpio = irq_to_gpio(d->irq);
_set_gpio_irqenable(&mxc_gpio_ports[gpio / 32], gpio & 0x1f, 0);
}
-static void gpio_unmask_irq(u32 irq)
+static void gpio_unmask_irq(struct irq_data *d)
{
- u32 gpio = irq_to_gpio(irq);
+ u32 gpio = irq_to_gpio(d->irq);
_set_gpio_irqenable(&mxc_gpio_ports[gpio / 32], gpio & 0x1f, 1);
}
static int mxc_gpio_get(struct gpio_chip *chip, unsigned offset);
-static int gpio_set_irq_type(u32 irq, u32 type)
+static int gpio_set_irq_type(struct irq_data *d, u32 type)
{
- u32 gpio = irq_to_gpio(irq);
+ u32 gpio = irq_to_gpio(d->irq);
struct mxc_gpio_port *port = &mxc_gpio_ports[gpio / 32];
u32 bit, val;
int edge;
@@ -211,9 +211,9 @@
* @param enable enable as wake-up if equal to non-zero
* @return This function returns 0 on success.
*/
-static int gpio_set_wake_irq(u32 irq, u32 enable)
+static int gpio_set_wake_irq(struct irq_data *d, u32 enable)
{
- u32 gpio = irq_to_gpio(irq);
+ u32 gpio = irq_to_gpio(d->irq);
u32 gpio_idx = gpio & 0x1F;
struct mxc_gpio_port *port = &mxc_gpio_ports[gpio / 32];
@@ -233,11 +233,11 @@
}
static struct irq_chip gpio_irq_chip = {
- .ack = gpio_ack_irq,
- .mask = gpio_mask_irq,
- .unmask = gpio_unmask_irq,
- .set_type = gpio_set_irq_type,
- .set_wake = gpio_set_wake_irq,
+ .irq_ack = gpio_ack_irq,
+ .irq_mask = gpio_mask_irq,
+ .irq_unmask = gpio_unmask_irq,
+ .irq_set_type = gpio_set_irq_type,
+ .irq_set_wake = gpio_set_wake_irq,
};
static void _set_gpio_direction(struct gpio_chip *chip, unsigned offset,
diff --git a/arch/arm/plat-mxc/include/mach/iomux-mx53.h b/arch/arm/plat-mxc/include/mach/iomux-mx53.h
index 5deee01..68e11d7 100644
--- a/arch/arm/plat-mxc/include/mach/iomux-mx53.h
+++ b/arch/arm/plat-mxc/include/mach/iomux-mx53.h
@@ -34,7 +34,6 @@
IOMUX_CONFIG_ALT6,
IOMUX_CONFIG_ALT7,
IOMUX_CONFIG_GPIO, /* added to help user use GPIO mode */
- IOMUX_CONFIG_SION = 0x1 << 4, /* LOOPBACK:MUX SION bit */
} iomux_pin_cfg_t;
/* These 2 defines are for pins that may not have a mux register, but could
@@ -135,6 +134,9 @@
#define MX53_PAD_EIM_D16__GPIO_3_16 IOMUX_PAD(0x460, 0x118,IOMUX_CONFIG_ALT1, 0x0, 0, NO_PAD_CTRL)
#define MX53_PAD_EIM_D17__GPIO_3_17 IOMUX_PAD(0x464, 0x11C,IOMUX_CONFIG_ALT1, 0x0, 0, NO_PAD_CTRL)
#define MX53_PAD_EIM_D18__GPIO_3_18 IOMUX_PAD(0x468, 0x120,IOMUX_CONFIG_ALT1, 0x0, 0, NO_PAD_CTRL)
+#define MX53_PAD_EIM_D16__CSPI1_SCLK IOMUX_PAD(0x460, 0x118,IOMUX_CONFIG_ALT4, 0x79c, 3, NO_PAD_CTRL)
+#define MX53_PAD_EIM_D17__CSPI1_MISO IOMUX_PAD(0x464, 0x11C,IOMUX_CONFIG_ALT4, 0x7a0, 3, NO_PAD_CTRL)
+#define MX53_PAD_EIM_D18__CSPI1_MOSI IOMUX_PAD(0x468, 0x120,IOMUX_CONFIG_ALT4, 0x7a4, 3, NO_PAD_CTRL)
#define MX53_PAD_EIM_D19__GPIO_3_19 IOMUX_PAD(0x46C, 0x124,IOMUX_CONFIG_ALT1, 0x0, 0, NO_PAD_CTRL)
#define MX53_PAD_EIM_D20__GPIO_3_20 IOMUX_PAD(0x470, 0x128,IOMUX_CONFIG_ALT1, 0x0, 0, NO_PAD_CTRL)
#define MX53_PAD_EIM_D21__GPIO_3_21 IOMUX_PAD(0x474, 0x12C,IOMUX_CONFIG_ALT1, 0x0, 0, NO_PAD_CTRL)
diff --git a/arch/arm/plat-mxc/include/mach/iomux-v3.h b/arch/arm/plat-mxc/include/mach/iomux-v3.h
index 2277b01..82620af 100644
--- a/arch/arm/plat-mxc/include/mach/iomux-v3.h
+++ b/arch/arm/plat-mxc/include/mach/iomux-v3.h
@@ -105,6 +105,7 @@
#define PAD_CTL_SRE_FAST (1 << 0)
#define PAD_CTL_SRE_SLOW (0 << 0)
+#define IOMUX_CONFIG_SION (0x1 << 4)
#define MX51_NUM_GPIO_PORT 4
diff --git a/arch/arm/plat-mxc/include/mach/mx51.h b/arch/arm/plat-mxc/include/mach/mx51.h
index 873807f..1eb339e 100644
--- a/arch/arm/plat-mxc/include/mach/mx51.h
+++ b/arch/arm/plat-mxc/include/mach/mx51.h
@@ -301,8 +301,8 @@
#define MX51_MXC_INT_GPIO4_HIGH 57
#define MX51_MXC_INT_WDOG1 58
#define MX51_MXC_INT_WDOG2 59
-#define MX51_MXC_INT_KPP 60
-#define MX51_MXC_INT_PWM1 61
+#define MX51_INT_KPP 60
+#define MX51_INT_PWM1 61
#define MX51_INT_I2C1 62
#define MX51_INT_I2C2 63
#define MX51_MXC_INT_HS_I2C 64
@@ -335,7 +335,7 @@
#define MX51_MXC_INT_SPDIF 91
#define MX51_MXC_INT_TVE 92
#define MX51_MXC_INT_FIRI 93
-#define MX51_MXC_INT_PWM2 94
+#define MX51_INT_PWM2 94
#define MX51_MXC_INT_SLIM_EXP 95
#define MX51_INT_SSI3 96
#define MX51_MXC_INT_EMI_BOOT 97
diff --git a/arch/arm/plat-mxc/include/mach/mx53.h b/arch/arm/plat-mxc/include/mach/mx53.h
index 9577cdb..d7a8e52 100644
--- a/arch/arm/plat-mxc/include/mach/mx53.h
+++ b/arch/arm/plat-mxc/include/mach/mx53.h
@@ -53,13 +53,13 @@
#define MX53_SPBA0_BASE_ADDR 0x50000000
#define MX53_SPBA0_SIZE SZ_1M
-#define MX53_MMC_SDHC1_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00004000)
-#define MX53_MMC_SDHC2_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00008000)
+#define MX53_ESDHC1_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00004000)
+#define MX53_ESDHC2_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00008000)
#define MX53_UART3_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x0000C000)
-#define MX53_CSPI1_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00010000)
+#define MX53_ECSPI1_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00010000)
#define MX53_SSI2_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00014000)
-#define MX53_MMC_SDHC3_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00020000)
-#define MX53_MMC_SDHC4_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00024000)
+#define MX53_ESDHC3_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00020000)
+#define MX53_ESDHC4_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00024000)
#define MX53_SPDIF_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00028000)
#define MX53_ASRC_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x0002C000)
#define MX53_ATA_DMA_BASE_ADDR (MX53_SPBA0_BASE_ADDR + 0x00030000)
@@ -117,12 +117,12 @@
#define MX53_ARM_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000A0000)
#define MX53_OWIRE_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000A4000)
#define MX53_FIRI_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000A8000)
-#define MX53_CSPI2_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000AC000)
+#define MX53_ECSPI2_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000AC000)
#define MX53_SDMA_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000B0000)
#define MX53_SCC_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000B4000)
#define MX53_ROMCP_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000B8000)
#define MX53_RTIC_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000BC000)
-#define MX53_CSPI3_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000C0000)
+#define MX53_CSPI_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000C0000)
#define MX53_I2C2_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000C4000)
#define MX53_I2C1_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000C8000)
#define MX53_SSI1_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000CC000)
@@ -136,7 +136,7 @@
#define MX53_MIPI_HSC_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000DC000)
#define MX53_MLB_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000E4000)
#define MX53_SSI3_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000E8000)
-#define MX53_MXC_FEC_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000EC000)
+#define MX53_FEC_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000EC000)
#define MX53_TVE_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000F0000)
#define MX53_VPU_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000F4000)
#define MX53_SAHARA_BASE_ADDR (MX53_AIPS2_BASE_ADDR + 0x000F8000)
@@ -229,10 +229,10 @@
* Interrupt numbers
*/
#define MX53_INT_RESV0 0
-#define MX53_INT_MMC_SDHC1 1
-#define MX53_INT_MMC_SDHC2 2
-#define MX53_INT_MMC_SDHC3 3
-#define MX53_INT_MMC_SDHC4 4
+#define MX53_INT_ESDHC1 1
+#define MX53_INT_ESDHC2 2
+#define MX53_INT_ESDHC3 3
+#define MX53_INT_ESDHC4 4
#define MX53_INT_RESV5 5
#define MX53_INT_SDMA 6
#define MX53_INT_IOMUX 7
@@ -264,8 +264,8 @@
#define MX53_INT_UART3 33
#define MX53_INT_RESV34 34
#define MX53_INT_RESV35 35
-#define MX53_INT_CSPI1 36
-#define MX53_INT_CSPI2 37
+#define MX53_INT_ECSPI1 36
+#define MX53_INT_ECSPI2 37
#define MX53_INT_CSPI 38
#define MX53_INT_GPT 39
#define MX53_INT_EPIT1 40
diff --git a/arch/arm/plat-mxc/pwm.c b/arch/arm/plat-mxc/pwm.c
index c36f263..7a61ef8 100644
--- a/arch/arm/plat-mxc/pwm.c
+++ b/arch/arm/plat-mxc/pwm.c
@@ -57,7 +57,7 @@
if (pwm == NULL || period_ns == 0 || duty_ns > period_ns)
return -EINVAL;
- if (cpu_is_mx27() || cpu_is_mx3() || cpu_is_mx25()) {
+ if (cpu_is_mx27() || cpu_is_mx3() || cpu_is_mx25() || cpu_is_mx51()) {
unsigned long long c;
unsigned long period_cycles, duty_cycles, prescale;
u32 cr;
diff --git a/arch/arm/plat-mxc/tzic.c b/arch/arm/plat-mxc/tzic.c
index e69ed8a..bc3a6be 100644
--- a/arch/arm/plat-mxc/tzic.c
+++ b/arch/arm/plat-mxc/tzic.c
@@ -69,50 +69,50 @@
#endif
/**
- * tzic_mask_irq() - Disable interrupt number "irq" in the TZIC
+ * tzic_mask_irq() - Disable interrupt source "d" in the TZIC
*
- * @param irq interrupt source number
+ * @param d interrupt source
*/
-static void tzic_mask_irq(unsigned int irq)
+static void tzic_mask_irq(struct irq_data *d)
{
int index, off;
- index = irq >> 5;
- off = irq & 0x1F;
+ index = d->irq >> 5;
+ off = d->irq & 0x1F;
__raw_writel(1 << off, tzic_base + TZIC_ENCLEAR0(index));
}
/**
- * tzic_unmask_irq() - Enable interrupt number "irq" in the TZIC
+ * tzic_unmask_irq() - Enable interrupt source "d" in the TZIC
*
- * @param irq interrupt source number
+ * @param d interrupt source
*/
-static void tzic_unmask_irq(unsigned int irq)
+static void tzic_unmask_irq(struct irq_data *d)
{
int index, off;
- index = irq >> 5;
- off = irq & 0x1F;
+ index = d->irq >> 5;
+ off = d->irq & 0x1F;
__raw_writel(1 << off, tzic_base + TZIC_ENSET0(index));
}
static unsigned int wakeup_intr[4];
/**
- * tzic_set_wake_irq() - Set interrupt number "irq" in the TZIC as a wake-up source.
+ * tzic_set_wake_irq() - Set interrupt source "d" in the TZIC as a wake-up source.
*
- * @param irq interrupt source number
+ * @param d interrupt source
* @param enable enable as wake-up if equal to non-zero
* disble as wake-up if equal to zero
*
* @return This function returns 0 on success.
*/
-static int tzic_set_wake_irq(unsigned int irq, unsigned int enable)
+static int tzic_set_wake_irq(struct irq_data *d, unsigned int enable)
{
unsigned int index, off;
- index = irq >> 5;
- off = irq & 0x1F;
+ index = d->irq >> 5;
+ off = d->irq & 0x1F;
if (index > 3)
return -EINVAL;
@@ -128,10 +128,10 @@
static struct mxc_irq_chip mxc_tzic_chip = {
.base = {
.name = "MXC_TZIC",
- .ack = tzic_mask_irq,
- .mask = tzic_mask_irq,
- .unmask = tzic_unmask_irq,
- .set_wake = tzic_set_wake_irq,
+ .irq_ack = tzic_mask_irq,
+ .irq_mask = tzic_mask_irq,
+ .irq_unmask = tzic_unmask_irq,
+ .irq_set_wake = tzic_set_wake_irq,
},
#ifdef CONFIG_FIQ
.set_irq_fiq = tzic_set_irq_fiq,
diff --git a/arch/arm/plat-nomadik/gpio.c b/arch/arm/plat-nomadik/gpio.c
index eda4e3a..1e88ecb 100644
--- a/arch/arm/plat-nomadik/gpio.c
+++ b/arch/arm/plat-nomadik/gpio.c
@@ -356,13 +356,13 @@
return 1 << (gpio % 32);
}
-static void nmk_gpio_irq_ack(unsigned int irq)
+static void nmk_gpio_irq_ack(struct irq_data *d)
{
int gpio;
struct nmk_gpio_chip *nmk_chip;
- gpio = NOMADIK_IRQ_TO_GPIO(irq);
- nmk_chip = get_irq_chip_data(irq);
+ gpio = NOMADIK_IRQ_TO_GPIO(d->irq);
+ nmk_chip = irq_data_get_irq_chip_data(d);
if (!nmk_chip)
return;
writel(nmk_gpio_get_bitmask(gpio), nmk_chip->addr + NMK_GPIO_IC);
@@ -401,7 +401,7 @@
}
}
-static int nmk_gpio_irq_modify(unsigned int irq, enum nmk_gpio_irq_type which,
+static int nmk_gpio_irq_modify(struct irq_data *d, enum nmk_gpio_irq_type which,
bool enable)
{
int gpio;
@@ -409,8 +409,8 @@
unsigned long flags;
u32 bitmask;
- gpio = NOMADIK_IRQ_TO_GPIO(irq);
- nmk_chip = get_irq_chip_data(irq);
+ gpio = NOMADIK_IRQ_TO_GPIO(d->irq);
+ nmk_chip = irq_data_get_irq_chip_data(d);
bitmask = nmk_gpio_get_bitmask(gpio);
if (!nmk_chip)
return -EINVAL;
@@ -422,24 +422,24 @@
return 0;
}
-static void nmk_gpio_irq_mask(unsigned int irq)
+static void nmk_gpio_irq_mask(struct irq_data *d)
{
- nmk_gpio_irq_modify(irq, NORMAL, false);
+ nmk_gpio_irq_modify(d, NORMAL, false);
}
-static void nmk_gpio_irq_unmask(unsigned int irq)
+static void nmk_gpio_irq_unmask(struct irq_data *d)
{
- nmk_gpio_irq_modify(irq, NORMAL, true);
+ nmk_gpio_irq_modify(d, NORMAL, true);
}
-static int nmk_gpio_irq_set_wake(unsigned int irq, unsigned int on)
+static int nmk_gpio_irq_set_wake(struct irq_data *d, unsigned int on)
{
struct nmk_gpio_chip *nmk_chip;
unsigned long flags;
int gpio;
- gpio = NOMADIK_IRQ_TO_GPIO(irq);
- nmk_chip = get_irq_chip_data(irq);
+ gpio = NOMADIK_IRQ_TO_GPIO(d->irq);
+ nmk_chip = irq_data_get_irq_chip_data(d);
if (!nmk_chip)
return -EINVAL;
@@ -457,9 +457,9 @@
return 0;
}
-static int nmk_gpio_irq_set_type(unsigned int irq, unsigned int type)
+static int nmk_gpio_irq_set_type(struct irq_data *d, unsigned int type)
{
- struct irq_desc *desc = irq_to_desc(irq);
+ struct irq_desc *desc = irq_to_desc(d->irq);
bool enabled = !(desc->status & IRQ_DISABLED);
bool wake = desc->wake_depth;
int gpio;
@@ -467,8 +467,8 @@
unsigned long flags;
u32 bitmask;
- gpio = NOMADIK_IRQ_TO_GPIO(irq);
- nmk_chip = get_irq_chip_data(irq);
+ gpio = NOMADIK_IRQ_TO_GPIO(d->irq);
+ nmk_chip = irq_data_get_irq_chip_data(d);
bitmask = nmk_gpio_get_bitmask(gpio);
if (!nmk_chip)
return -EINVAL;
@@ -507,11 +507,11 @@
static struct irq_chip nmk_gpio_irq_chip = {
.name = "Nomadik-GPIO",
- .ack = nmk_gpio_irq_ack,
- .mask = nmk_gpio_irq_mask,
- .unmask = nmk_gpio_irq_unmask,
- .set_type = nmk_gpio_irq_set_type,
- .set_wake = nmk_gpio_irq_set_wake,
+ .irq_ack = nmk_gpio_irq_ack,
+ .irq_mask = nmk_gpio_irq_mask,
+ .irq_unmask = nmk_gpio_irq_unmask,
+ .irq_set_type = nmk_gpio_irq_set_type,
+ .irq_set_wake = nmk_gpio_irq_set_wake,
};
static void nmk_gpio_irq_handler(unsigned int irq, struct irq_desc *desc)
@@ -522,12 +522,12 @@
u32 pending;
unsigned int first_irq;
- if (host_chip->mask_ack)
- host_chip->mask_ack(irq);
+ if (host_chip->irq_mask_ack)
+ host_chip->irq_mask_ack(&desc->irq_data);
else {
- host_chip->mask(irq);
- if (host_chip->ack)
- host_chip->ack(irq);
+ host_chip->irq_mask(&desc->irq_data);
+ if (host_chip->irq_ack)
+ host_chip->irq_ack(&desc->irq_data);
}
nmk_chip = get_irq_data(irq);
@@ -537,7 +537,7 @@
generic_handle_irq(gpio_irq);
}
- host_chip->unmask(irq);
+ host_chip->irq_unmask(&desc->irq_data);
}
static int nmk_gpio_init_irq(struct nmk_gpio_chip *nmk_chip)
diff --git a/arch/arm/plat-omap/gpio.c b/arch/arm/plat-omap/gpio.c
index 1f98e0b..971d186 100644
--- a/arch/arm/plat-omap/gpio.c
+++ b/arch/arm/plat-omap/gpio.c
@@ -718,7 +718,7 @@
case METHOD_GPIO_24XX:
case METHOD_GPIO_44XX:
set_24xx_gpio_triggering(bank, gpio, trigger);
- break;
+ return 0;
#endif
default:
goto bad;
@@ -729,17 +729,17 @@
return -EINVAL;
}
-static int gpio_irq_type(unsigned irq, unsigned type)
+static int gpio_irq_type(struct irq_data *d, unsigned type)
{
struct gpio_bank *bank;
unsigned gpio;
int retval;
unsigned long flags;
- if (!cpu_class_is_omap2() && irq > IH_MPUIO_BASE)
- gpio = OMAP_MPUIO(irq - IH_MPUIO_BASE);
+ if (!cpu_class_is_omap2() && d->irq > IH_MPUIO_BASE)
+ gpio = OMAP_MPUIO(d->irq - IH_MPUIO_BASE);
else
- gpio = irq - IH_GPIO_BASE;
+ gpio = d->irq - IH_GPIO_BASE;
if (check_gpio(gpio) < 0)
return -EINVAL;
@@ -752,19 +752,21 @@
&& (type & (IRQ_TYPE_LEVEL_LOW|IRQ_TYPE_LEVEL_HIGH)))
return -EINVAL;
- bank = get_irq_chip_data(irq);
+ bank = irq_data_get_irq_chip_data(d);
spin_lock_irqsave(&bank->lock, flags);
retval = _set_gpio_triggering(bank, get_gpio_index(gpio), type);
if (retval == 0) {
- irq_desc[irq].status &= ~IRQ_TYPE_SENSE_MASK;
- irq_desc[irq].status |= type;
+ struct irq_desc *desc = irq_to_desc(d->irq);
+
+ desc->status &= ~IRQ_TYPE_SENSE_MASK;
+ desc->status |= type;
}
spin_unlock_irqrestore(&bank->lock, flags);
if (type & (IRQ_TYPE_LEVEL_LOW | IRQ_TYPE_LEVEL_HIGH))
- __set_irq_handler_unlocked(irq, handle_level_irq);
+ __set_irq_handler_unlocked(d->irq, handle_level_irq);
else if (type & (IRQ_TYPE_EDGE_FALLING | IRQ_TYPE_EDGE_RISING))
- __set_irq_handler_unlocked(irq, handle_edge_irq);
+ __set_irq_handler_unlocked(d->irq, handle_edge_irq);
return retval;
}
@@ -1021,15 +1023,15 @@
}
/* Use disable_irq_wake() and enable_irq_wake() functions from drivers */
-static int gpio_wake_enable(unsigned int irq, unsigned int enable)
+static int gpio_wake_enable(struct irq_data *d, unsigned int enable)
{
- unsigned int gpio = irq - IH_GPIO_BASE;
+ unsigned int gpio = d->irq - IH_GPIO_BASE;
struct gpio_bank *bank;
int retval;
if (check_gpio(gpio) < 0)
return -ENODEV;
- bank = get_irq_chip_data(irq);
+ bank = irq_data_get_irq_chip_data(d);
retval = _set_gpio_wakeup(bank, get_gpio_index(gpio), enable);
return retval;
@@ -1142,7 +1144,7 @@
u32 retrigger = 0;
int unmasked = 0;
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
bank = get_irq_data(irq);
#ifdef CONFIG_ARCH_OMAP1
@@ -1199,7 +1201,7 @@
configured, we could unmask GPIO bank interrupt immediately */
if (!level_mask && !unmasked) {
unmasked = 1;
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
isr |= retrigger;
@@ -1235,41 +1237,40 @@
interrupt */
exit:
if (!unmasked)
- desc->chip->unmask(irq);
-
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
-static void gpio_irq_shutdown(unsigned int irq)
+static void gpio_irq_shutdown(struct irq_data *d)
{
- unsigned int gpio = irq - IH_GPIO_BASE;
- struct gpio_bank *bank = get_irq_chip_data(irq);
+ unsigned int gpio = d->irq - IH_GPIO_BASE;
+ struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
_reset_gpio(bank, gpio);
}
-static void gpio_ack_irq(unsigned int irq)
+static void gpio_ack_irq(struct irq_data *d)
{
- unsigned int gpio = irq - IH_GPIO_BASE;
- struct gpio_bank *bank = get_irq_chip_data(irq);
+ unsigned int gpio = d->irq - IH_GPIO_BASE;
+ struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
_clear_gpio_irqstatus(bank, gpio);
}
-static void gpio_mask_irq(unsigned int irq)
+static void gpio_mask_irq(struct irq_data *d)
{
- unsigned int gpio = irq - IH_GPIO_BASE;
- struct gpio_bank *bank = get_irq_chip_data(irq);
+ unsigned int gpio = d->irq - IH_GPIO_BASE;
+ struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
_set_gpio_irqenable(bank, gpio, 0);
_set_gpio_triggering(bank, get_gpio_index(gpio), IRQ_TYPE_NONE);
}
-static void gpio_unmask_irq(unsigned int irq)
+static void gpio_unmask_irq(struct irq_data *d)
{
- unsigned int gpio = irq - IH_GPIO_BASE;
- struct gpio_bank *bank = get_irq_chip_data(irq);
+ unsigned int gpio = d->irq - IH_GPIO_BASE;
+ struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
unsigned int irq_mask = 1 << get_gpio_index(gpio);
- struct irq_desc *desc = irq_to_desc(irq);
+ struct irq_desc *desc = irq_to_desc(d->irq);
u32 trigger = desc->status & IRQ_TYPE_SENSE_MASK;
if (trigger)
@@ -1287,12 +1288,12 @@
static struct irq_chip gpio_irq_chip = {
.name = "GPIO",
- .shutdown = gpio_irq_shutdown,
- .ack = gpio_ack_irq,
- .mask = gpio_mask_irq,
- .unmask = gpio_unmask_irq,
- .set_type = gpio_irq_type,
- .set_wake = gpio_wake_enable,
+ .irq_shutdown = gpio_irq_shutdown,
+ .irq_ack = gpio_ack_irq,
+ .irq_mask = gpio_mask_irq,
+ .irq_unmask = gpio_unmask_irq,
+ .irq_set_type = gpio_irq_type,
+ .irq_set_wake = gpio_wake_enable,
};
/*---------------------------------------------------------------------*/
@@ -1301,36 +1302,36 @@
/* MPUIO uses the always-on 32k clock */
-static void mpuio_ack_irq(unsigned int irq)
+static void mpuio_ack_irq(struct irq_data *d)
{
/* The ISR is reset automatically, so do nothing here. */
}
-static void mpuio_mask_irq(unsigned int irq)
+static void mpuio_mask_irq(struct irq_data *d)
{
- unsigned int gpio = OMAP_MPUIO(irq - IH_MPUIO_BASE);
- struct gpio_bank *bank = get_irq_chip_data(irq);
+ unsigned int gpio = OMAP_MPUIO(d->irq - IH_MPUIO_BASE);
+ struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
_set_gpio_irqenable(bank, gpio, 0);
}
-static void mpuio_unmask_irq(unsigned int irq)
+static void mpuio_unmask_irq(struct irq_data *d)
{
- unsigned int gpio = OMAP_MPUIO(irq - IH_MPUIO_BASE);
- struct gpio_bank *bank = get_irq_chip_data(irq);
+ unsigned int gpio = OMAP_MPUIO(d->irq - IH_MPUIO_BASE);
+ struct gpio_bank *bank = irq_data_get_irq_chip_data(d);
_set_gpio_irqenable(bank, gpio, 1);
}
static struct irq_chip mpuio_irq_chip = {
.name = "MPUIO",
- .ack = mpuio_ack_irq,
- .mask = mpuio_mask_irq,
- .unmask = mpuio_unmask_irq,
- .set_type = gpio_irq_type,
+ .irq_ack = mpuio_ack_irq,
+ .irq_mask = mpuio_mask_irq,
+ .irq_unmask = mpuio_unmask_irq,
+ .irq_set_type = gpio_irq_type,
#ifdef CONFIG_ARCH_OMAP16XX
/* REVISIT: assuming only 16xx supports MPUIO wake events */
- .set_wake = gpio_wake_enable,
+ .irq_set_wake = gpio_wake_enable,
#endif
};
@@ -1671,7 +1672,9 @@
for (j = bank->virtual_irq_start;
j < bank->virtual_irq_start + bank_width; j++) {
- lockdep_set_class(&irq_desc[j].lock, &gpio_lock_class);
+ struct irq_desc *d = irq_to_desc(j);
+
+ lockdep_set_class(&d->lock, &gpio_lock_class);
set_irq_chip_data(j, bank);
if (bank_is_mpuio(bank))
set_irq_chip(j, &mpuio_irq_chip);
diff --git a/arch/arm/plat-omap/include/plat/voltage.h b/arch/arm/plat-omap/include/plat/voltage.h
index 0ff1233..5bd204e 100644
--- a/arch/arm/plat-omap/include/plat/voltage.h
+++ b/arch/arm/plat-omap/include/plat/voltage.h
@@ -14,6 +14,8 @@
#ifndef __ARCH_ARM_MACH_OMAP2_VOLTAGE_H
#define __ARCH_ARM_MACH_OMAP2_VOLTAGE_H
+#include <linux/err.h>
+
#define VOLTSCALE_VPFORCEUPDATE 1
#define VOLTSCALE_VCBYPASS 2
@@ -65,9 +67,6 @@
char *name;
};
-/* API to get the voltagedomain pointer */
-struct voltagedomain *omap_voltage_domain_lookup(char *name);
-
/**
* struct omap_volt_data - Omap voltage specific data.
* @voltage_nominal: The possible voltage value in uV
@@ -131,16 +130,26 @@
struct omap_volt_pmic_info *pmic_info);
void omap_change_voltscale_method(struct voltagedomain *voltdm,
int voltscale_method);
+/* API to get the voltagedomain pointer */
+struct voltagedomain *omap_voltage_domain_lookup(char *name);
+
int omap_voltage_late_init(void);
#else
static inline int omap_voltage_register_pmic(struct voltagedomain *voltdm,
- struct omap_volt_pmic_info *pmic_info) {}
+ struct omap_volt_pmic_info *pmic_info)
+{
+ return -EINVAL;
+}
static inline void omap_change_voltscale_method(struct voltagedomain *voltdm,
int voltscale_method) {}
static inline int omap_voltage_late_init(void)
{
return -EINVAL;
}
+static inline struct voltagedomain *omap_voltage_domain_lookup(char *name)
+{
+ return ERR_PTR(-EINVAL);
+}
#endif
#endif
diff --git a/arch/arm/plat-orion/gpio.c b/arch/arm/plat-orion/gpio.c
index e814803..5f35223 100644
--- a/arch/arm/plat-orion/gpio.c
+++ b/arch/arm/plat-orion/gpio.c
@@ -232,20 +232,19 @@
* polarity LEVEL mask
*
****************************************************************************/
-
-static void gpio_irq_ack(u32 irq)
+static void gpio_irq_ack(struct irq_data *d)
{
- int type = irq_desc[irq].status & IRQ_TYPE_SENSE_MASK;
+ int type = irq_desc[d->irq].status & IRQ_TYPE_SENSE_MASK;
if (type & (IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING)) {
- int pin = irq_to_gpio(irq);
+ int pin = irq_to_gpio(d->irq);
writel(~(1 << (pin & 31)), GPIO_EDGE_CAUSE(pin));
}
}
-static void gpio_irq_mask(u32 irq)
+static void gpio_irq_mask(struct irq_data *d)
{
- int pin = irq_to_gpio(irq);
- int type = irq_desc[irq].status & IRQ_TYPE_SENSE_MASK;
+ int pin = irq_to_gpio(d->irq);
+ int type = irq_desc[d->irq].status & IRQ_TYPE_SENSE_MASK;
u32 reg = (type & (IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING)) ?
GPIO_EDGE_MASK(pin) : GPIO_LEVEL_MASK(pin);
u32 u = readl(reg);
@@ -253,10 +252,10 @@
writel(u, reg);
}
-static void gpio_irq_unmask(u32 irq)
+static void gpio_irq_unmask(struct irq_data *d)
{
- int pin = irq_to_gpio(irq);
- int type = irq_desc[irq].status & IRQ_TYPE_SENSE_MASK;
+ int pin = irq_to_gpio(d->irq);
+ int type = irq_desc[d->irq].status & IRQ_TYPE_SENSE_MASK;
u32 reg = (type & (IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING)) ?
GPIO_EDGE_MASK(pin) : GPIO_LEVEL_MASK(pin);
u32 u = readl(reg);
@@ -264,20 +263,20 @@
writel(u, reg);
}
-static int gpio_irq_set_type(u32 irq, u32 type)
+static int gpio_irq_set_type(struct irq_data *d, u32 type)
{
- int pin = irq_to_gpio(irq);
+ int pin = irq_to_gpio(d->irq);
struct irq_desc *desc;
u32 u;
u = readl(GPIO_IO_CONF(pin)) & (1 << (pin & 31));
if (!u) {
printk(KERN_ERR "orion gpio_irq_set_type failed "
- "(irq %d, pin %d).\n", irq, pin);
+ "(irq %d, pin %d).\n", d->irq, pin);
return -EINVAL;
}
- desc = irq_desc + irq;
+ desc = irq_desc + d->irq;
/*
* Set edge/level type.
@@ -287,7 +286,7 @@
} else if (type & (IRQ_TYPE_LEVEL_HIGH | IRQ_TYPE_LEVEL_LOW)) {
desc->handle_irq = handle_level_irq;
} else {
- printk(KERN_ERR "failed to set irq=%d (type=%d)\n", irq, type);
+ printk(KERN_ERR "failed to set irq=%d (type=%d)\n", d->irq, type);
return -EINVAL;
}
@@ -325,10 +324,10 @@
struct irq_chip orion_gpio_irq_chip = {
.name = "orion_gpio_irq",
- .ack = gpio_irq_ack,
- .mask = gpio_irq_mask,
- .unmask = gpio_irq_unmask,
- .set_type = gpio_irq_set_type,
+ .irq_ack = gpio_irq_ack,
+ .irq_mask = gpio_irq_mask,
+ .irq_unmask = gpio_irq_unmask,
+ .irq_set_type = gpio_irq_set_type,
};
void orion_gpio_irq_handler(int pinoff)
diff --git a/arch/arm/plat-orion/irq.c b/arch/arm/plat-orion/irq.c
index 3f9d34f..7d0c7eb 100644
--- a/arch/arm/plat-orion/irq.c
+++ b/arch/arm/plat-orion/irq.c
@@ -14,31 +14,31 @@
#include <linux/io.h>
#include <plat/irq.h>
-static void orion_irq_mask(u32 irq)
+static void orion_irq_mask(struct irq_data *d)
{
- void __iomem *maskaddr = get_irq_chip_data(irq);
+ void __iomem *maskaddr = irq_data_get_irq_chip_data(d);
u32 mask;
mask = readl(maskaddr);
- mask &= ~(1 << (irq & 31));
+ mask &= ~(1 << (d->irq & 31));
writel(mask, maskaddr);
}
-static void orion_irq_unmask(u32 irq)
+static void orion_irq_unmask(struct irq_data *d)
{
- void __iomem *maskaddr = get_irq_chip_data(irq);
+ void __iomem *maskaddr = irq_data_get_irq_chip_data(d);
u32 mask;
mask = readl(maskaddr);
- mask |= 1 << (irq & 31);
+ mask |= 1 << (d->irq & 31);
writel(mask, maskaddr);
}
static struct irq_chip orion_irq_chip = {
.name = "orion_irq",
- .mask = orion_irq_mask,
- .mask_ack = orion_irq_mask,
- .unmask = orion_irq_unmask,
+ .irq_mask = orion_irq_mask,
+ .irq_mask_ack = orion_irq_mask,
+ .irq_unmask = orion_irq_unmask,
};
void __init orion_irq_init(unsigned int irq_start, void __iomem *maskaddr)
diff --git a/arch/arm/plat-pxa/gpio.c b/arch/arm/plat-pxa/gpio.c
index 98548c6..e7de6ae 100644
--- a/arch/arm/plat-pxa/gpio.c
+++ b/arch/arm/plat-pxa/gpio.c
@@ -155,10 +155,10 @@
__raw_writel(gfer, c->regbase + GFER_OFFSET);
}
-static int pxa_gpio_irq_type(unsigned int irq, unsigned int type)
+static int pxa_gpio_irq_type(struct irq_data *d, unsigned int type)
{
struct pxa_gpio_chip *c;
- int gpio = irq_to_gpio(irq);
+ int gpio = irq_to_gpio(d->irq);
unsigned long gpdr, mask = GPIO_bit(gpio);
c = gpio_to_chip(gpio);
@@ -195,7 +195,7 @@
update_edge_detect(c);
- pr_debug("%s: IRQ%d (GPIO%d) - edge%s%s\n", __func__, irq, gpio,
+ pr_debug("%s: IRQ%d (GPIO%d) - edge%s%s\n", __func__, d->irq, gpio,
((type & IRQ_TYPE_EDGE_RISING) ? " rising" : ""),
((type & IRQ_TYPE_EDGE_FALLING) ? " falling" : ""));
return 0;
@@ -227,17 +227,17 @@
} while (loop);
}
-static void pxa_ack_muxed_gpio(unsigned int irq)
+static void pxa_ack_muxed_gpio(struct irq_data *d)
{
- int gpio = irq_to_gpio(irq);
+ int gpio = irq_to_gpio(d->irq);
struct pxa_gpio_chip *c = gpio_to_chip(gpio);
__raw_writel(GPIO_bit(gpio), c->regbase + GEDR_OFFSET);
}
-static void pxa_mask_muxed_gpio(unsigned int irq)
+static void pxa_mask_muxed_gpio(struct irq_data *d)
{
- int gpio = irq_to_gpio(irq);
+ int gpio = irq_to_gpio(d->irq);
struct pxa_gpio_chip *c = gpio_to_chip(gpio);
uint32_t grer, gfer;
@@ -249,9 +249,9 @@
__raw_writel(gfer, c->regbase + GFER_OFFSET);
}
-static void pxa_unmask_muxed_gpio(unsigned int irq)
+static void pxa_unmask_muxed_gpio(struct irq_data *d)
{
- int gpio = irq_to_gpio(irq);
+ int gpio = irq_to_gpio(d->irq);
struct pxa_gpio_chip *c = gpio_to_chip(gpio);
c->irq_mask |= GPIO_bit(gpio);
@@ -260,10 +260,10 @@
static struct irq_chip pxa_muxed_gpio_chip = {
.name = "GPIO",
- .ack = pxa_ack_muxed_gpio,
- .mask = pxa_mask_muxed_gpio,
- .unmask = pxa_unmask_muxed_gpio,
- .set_type = pxa_gpio_irq_type,
+ .irq_ack = pxa_ack_muxed_gpio,
+ .irq_mask = pxa_mask_muxed_gpio,
+ .irq_unmask = pxa_unmask_muxed_gpio,
+ .irq_set_type = pxa_gpio_irq_type,
};
void __init pxa_init_gpio(int mux_irq, int start, int end, set_wake_t fn)
@@ -291,7 +291,7 @@
/* Install handler for GPIO>=2 edge detect interrupts */
set_irq_chained_handler(mux_irq, pxa_gpio_demux_handler);
- pxa_muxed_gpio_chip.set_wake = fn;
+ pxa_muxed_gpio_chip.irq_set_wake = fn;
}
#ifdef CONFIG_PM
diff --git a/arch/arm/plat-pxa/include/plat/gpio.h b/arch/arm/plat-pxa/include/plat/gpio.h
index 44248cb..1ddd2b9 100644
--- a/arch/arm/plat-pxa/include/plat/gpio.h
+++ b/arch/arm/plat-pxa/include/plat/gpio.h
@@ -1,6 +1,8 @@
#ifndef __PLAT_GPIO_H
#define __PLAT_GPIO_H
+struct irq_data;
+
/*
* We handle the GPIOs by banks, each bank covers up to 32 GPIOs with
* one set of registers. The register offsets are organized below:
@@ -56,7 +58,7 @@
*/
extern int pxa_last_gpio;
-typedef int (*set_wake_t)(unsigned int irq, unsigned int on);
+typedef int (*set_wake_t)(struct irq_data *d, unsigned int on);
extern void pxa_init_gpio(int mux_irq, int start, int end, set_wake_t fn);
#endif /* __PLAT_GPIO_H */
diff --git a/arch/arm/plat-s3c24xx/devs.c b/arch/arm/plat-s3c24xx/devs.c
index 8a42bc4..268f3ed 100644
--- a/arch/arm/plat-s3c24xx/devs.c
+++ b/arch/arm/plat-s3c24xx/devs.c
@@ -194,7 +194,6 @@
memcpy(&s3c2410ts_info, hard_s3c2410ts_info, sizeof(struct s3c2410_ts_mach_info));
s3c_device_ts.dev.platform_data = &s3c2410ts_info;
}
-EXPORT_SYMBOL(s3c24xx_ts_set_platdata);
/* USB Device (Gadget)*/
diff --git a/arch/arm/plat-s3c24xx/include/plat/irq.h b/arch/arm/plat-s3c24xx/include/plat/irq.h
index 69e1be8..ec087d6 100644
--- a/arch/arm/plat-s3c24xx/include/plat/irq.h
+++ b/arch/arm/plat-s3c24xx/include/plat/irq.h
@@ -107,9 +107,9 @@
/* exported for use in arch/arm/mach-s3c2410 */
#ifdef CONFIG_PM
-extern int s3c_irq_wake(unsigned int irqno, unsigned int state);
+extern int s3c_irq_wake(struct irq_data *data, unsigned int state);
#else
#define s3c_irq_wake NULL
#endif
-extern int s3c_irqext_type(unsigned int irq, unsigned int type);
+extern int s3c_irqext_type(struct irq_data *d, unsigned int type);
diff --git a/arch/arm/plat-s3c24xx/irq-pm.c b/arch/arm/plat-s3c24xx/irq-pm.c
index ea8dea3..c3624d8 100644
--- a/arch/arm/plat-s3c24xx/irq-pm.c
+++ b/arch/arm/plat-s3c24xx/irq-pm.c
@@ -15,11 +15,14 @@
#include <linux/module.h>
#include <linux/interrupt.h>
#include <linux/sysdev.h>
+#include <linux/irq.h>
#include <plat/cpu.h>
#include <plat/pm.h>
#include <plat/irq.h>
+#include <asm/irq.h>
+
/* state for IRQs over sleep */
/* default is to allow for EINT0..EINT15, and IRQ_RTC as wakeup sources
@@ -30,15 +33,15 @@
unsigned long s3c_irqwake_intallow = 1L << (IRQ_RTC - IRQ_EINT0) | 0xfL;
unsigned long s3c_irqwake_eintallow = 0x0000fff0L;
-int s3c_irq_wake(unsigned int irqno, unsigned int state)
+int s3c_irq_wake(struct irq_data *data, unsigned int state)
{
- unsigned long irqbit = 1 << (irqno - IRQ_EINT0);
+ unsigned long irqbit = 1 << (data->irq - IRQ_EINT0);
if (!(s3c_irqwake_intallow & irqbit))
return -ENOENT;
printk(KERN_INFO "wake %s for irq %d\n",
- state ? "enabled" : "disabled", irqno);
+ state ? "enabled" : "disabled", data->irq);
if (!state)
s3c_irqwake_intmask |= irqbit;
diff --git a/arch/arm/plat-s3c24xx/irq.c b/arch/arm/plat-s3c24xx/irq.c
index ad0d44e..4434cb5 100644
--- a/arch/arm/plat-s3c24xx/irq.c
+++ b/arch/arm/plat-s3c24xx/irq.c
@@ -34,30 +34,29 @@
#include <plat/irq.h>
static void
-s3c_irq_mask(unsigned int irqno)
+s3c_irq_mask(struct irq_data *data)
{
+ unsigned int irqno = data->irq - IRQ_EINT0;
unsigned long mask;
- irqno -= IRQ_EINT0;
-
mask = __raw_readl(S3C2410_INTMSK);
mask |= 1UL << irqno;
__raw_writel(mask, S3C2410_INTMSK);
}
static inline void
-s3c_irq_ack(unsigned int irqno)
+s3c_irq_ack(struct irq_data *data)
{
- unsigned long bitval = 1UL << (irqno - IRQ_EINT0);
+ unsigned long bitval = 1UL << (data->irq - IRQ_EINT0);
__raw_writel(bitval, S3C2410_SRCPND);
__raw_writel(bitval, S3C2410_INTPND);
}
static inline void
-s3c_irq_maskack(unsigned int irqno)
+s3c_irq_maskack(struct irq_data *data)
{
- unsigned long bitval = 1UL << (irqno - IRQ_EINT0);
+ unsigned long bitval = 1UL << (data->irq - IRQ_EINT0);
unsigned long mask;
mask = __raw_readl(S3C2410_INTMSK);
@@ -69,8 +68,9 @@
static void
-s3c_irq_unmask(unsigned int irqno)
+s3c_irq_unmask(struct irq_data *data)
{
+ unsigned int irqno = data->irq;
unsigned long mask;
if (irqno != IRQ_TIMER4 && irqno != IRQ_EINT8t23)
@@ -85,40 +85,39 @@
struct irq_chip s3c_irq_level_chip = {
.name = "s3c-level",
- .ack = s3c_irq_maskack,
- .mask = s3c_irq_mask,
- .unmask = s3c_irq_unmask,
- .set_wake = s3c_irq_wake
+ .irq_ack = s3c_irq_maskack,
+ .irq_mask = s3c_irq_mask,
+ .irq_unmask = s3c_irq_unmask,
+ .irq_set_wake = s3c_irq_wake
};
struct irq_chip s3c_irq_chip = {
.name = "s3c",
- .ack = s3c_irq_ack,
- .mask = s3c_irq_mask,
- .unmask = s3c_irq_unmask,
- .set_wake = s3c_irq_wake
+ .irq_ack = s3c_irq_ack,
+ .irq_mask = s3c_irq_mask,
+ .irq_unmask = s3c_irq_unmask,
+ .irq_set_wake = s3c_irq_wake
};
static void
-s3c_irqext_mask(unsigned int irqno)
+s3c_irqext_mask(struct irq_data *data)
{
+ unsigned int irqno = data->irq - EXTINT_OFF;
unsigned long mask;
- irqno -= EXTINT_OFF;
-
mask = __raw_readl(S3C24XX_EINTMASK);
mask |= ( 1UL << irqno);
__raw_writel(mask, S3C24XX_EINTMASK);
}
static void
-s3c_irqext_ack(unsigned int irqno)
+s3c_irqext_ack(struct irq_data *data)
{
unsigned long req;
unsigned long bit;
unsigned long mask;
- bit = 1UL << (irqno - EXTINT_OFF);
+ bit = 1UL << (data->irq - EXTINT_OFF);
mask = __raw_readl(S3C24XX_EINTMASK);
@@ -129,64 +128,57 @@
/* not sure if we should be acking the parent irq... */
- if (irqno <= IRQ_EINT7 ) {
+ if (data->irq <= IRQ_EINT7) {
if ((req & 0xf0) == 0)
- s3c_irq_ack(IRQ_EINT4t7);
+ s3c_irq_ack(irq_get_irq_data(IRQ_EINT4t7));
} else {
if ((req >> 8) == 0)
- s3c_irq_ack(IRQ_EINT8t23);
+ s3c_irq_ack(irq_get_irq_data(IRQ_EINT8t23));
}
}
static void
-s3c_irqext_unmask(unsigned int irqno)
+s3c_irqext_unmask(struct irq_data *data)
{
+ unsigned int irqno = data->irq - EXTINT_OFF;
unsigned long mask;
- irqno -= EXTINT_OFF;
-
mask = __raw_readl(S3C24XX_EINTMASK);
- mask &= ~( 1UL << irqno);
+ mask &= ~(1UL << irqno);
__raw_writel(mask, S3C24XX_EINTMASK);
}
int
-s3c_irqext_type(unsigned int irq, unsigned int type)
+s3c_irqext_type(struct irq_data *data, unsigned int type)
{
void __iomem *extint_reg;
void __iomem *gpcon_reg;
unsigned long gpcon_offset, extint_offset;
unsigned long newvalue = 0, value;
- if ((irq >= IRQ_EINT0) && (irq <= IRQ_EINT3))
- {
+ if ((data->irq >= IRQ_EINT0) && (data->irq <= IRQ_EINT3)) {
gpcon_reg = S3C2410_GPFCON;
extint_reg = S3C24XX_EXTINT0;
- gpcon_offset = (irq - IRQ_EINT0) * 2;
- extint_offset = (irq - IRQ_EINT0) * 4;
- }
- else if ((irq >= IRQ_EINT4) && (irq <= IRQ_EINT7))
- {
+ gpcon_offset = (data->irq - IRQ_EINT0) * 2;
+ extint_offset = (data->irq - IRQ_EINT0) * 4;
+ } else if ((data->irq >= IRQ_EINT4) && (data->irq <= IRQ_EINT7)) {
gpcon_reg = S3C2410_GPFCON;
extint_reg = S3C24XX_EXTINT0;
- gpcon_offset = (irq - (EXTINT_OFF)) * 2;
- extint_offset = (irq - (EXTINT_OFF)) * 4;
- }
- else if ((irq >= IRQ_EINT8) && (irq <= IRQ_EINT15))
- {
+ gpcon_offset = (data->irq - (EXTINT_OFF)) * 2;
+ extint_offset = (data->irq - (EXTINT_OFF)) * 4;
+ } else if ((data->irq >= IRQ_EINT8) && (data->irq <= IRQ_EINT15)) {
gpcon_reg = S3C2410_GPGCON;
extint_reg = S3C24XX_EXTINT1;
- gpcon_offset = (irq - IRQ_EINT8) * 2;
- extint_offset = (irq - IRQ_EINT8) * 4;
- }
- else if ((irq >= IRQ_EINT16) && (irq <= IRQ_EINT23))
- {
+ gpcon_offset = (data->irq - IRQ_EINT8) * 2;
+ extint_offset = (data->irq - IRQ_EINT8) * 4;
+ } else if ((data->irq >= IRQ_EINT16) && (data->irq <= IRQ_EINT23)) {
gpcon_reg = S3C2410_GPGCON;
extint_reg = S3C24XX_EXTINT2;
- gpcon_offset = (irq - IRQ_EINT8) * 2;
- extint_offset = (irq - IRQ_EINT16) * 4;
- } else
+ gpcon_offset = (data->irq - IRQ_EINT8) * 2;
+ extint_offset = (data->irq - IRQ_EINT16) * 4;
+ } else {
return -1;
+ }
/* Set the GPIO to external interrupt mode */
value = __raw_readl(gpcon_reg);
@@ -234,20 +226,20 @@
static struct irq_chip s3c_irqext_chip = {
.name = "s3c-ext",
- .mask = s3c_irqext_mask,
- .unmask = s3c_irqext_unmask,
- .ack = s3c_irqext_ack,
- .set_type = s3c_irqext_type,
- .set_wake = s3c_irqext_wake
+ .irq_mask = s3c_irqext_mask,
+ .irq_unmask = s3c_irqext_unmask,
+ .irq_ack = s3c_irqext_ack,
+ .irq_set_type = s3c_irqext_type,
+ .irq_set_wake = s3c_irqext_wake
};
static struct irq_chip s3c_irq_eint0t4 = {
.name = "s3c-ext0",
- .ack = s3c_irq_ack,
- .mask = s3c_irq_mask,
- .unmask = s3c_irq_unmask,
- .set_wake = s3c_irq_wake,
- .set_type = s3c_irqext_type,
+ .irq_ack = s3c_irq_ack,
+ .irq_mask = s3c_irq_mask,
+ .irq_unmask = s3c_irq_unmask,
+ .irq_set_wake = s3c_irq_wake,
+ .irq_set_type = s3c_irqext_type,
};
/* mask values for the parent registers for each of the interrupt types */
@@ -261,109 +253,109 @@
/* UART0 */
static void
-s3c_irq_uart0_mask(unsigned int irqno)
+s3c_irq_uart0_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_UART0, 7);
+ s3c_irqsub_mask(data->irq, INTMSK_UART0, 7);
}
static void
-s3c_irq_uart0_unmask(unsigned int irqno)
+s3c_irq_uart0_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_UART0);
+ s3c_irqsub_unmask(data->irq, INTMSK_UART0);
}
static void
-s3c_irq_uart0_ack(unsigned int irqno)
+s3c_irq_uart0_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_UART0, 7);
+ s3c_irqsub_maskack(data->irq, INTMSK_UART0, 7);
}
static struct irq_chip s3c_irq_uart0 = {
.name = "s3c-uart0",
- .mask = s3c_irq_uart0_mask,
- .unmask = s3c_irq_uart0_unmask,
- .ack = s3c_irq_uart0_ack,
+ .irq_mask = s3c_irq_uart0_mask,
+ .irq_unmask = s3c_irq_uart0_unmask,
+ .irq_ack = s3c_irq_uart0_ack,
};
/* UART1 */
static void
-s3c_irq_uart1_mask(unsigned int irqno)
+s3c_irq_uart1_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_UART1, 7 << 3);
+ s3c_irqsub_mask(data->irq, INTMSK_UART1, 7 << 3);
}
static void
-s3c_irq_uart1_unmask(unsigned int irqno)
+s3c_irq_uart1_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_UART1);
+ s3c_irqsub_unmask(data->irq, INTMSK_UART1);
}
static void
-s3c_irq_uart1_ack(unsigned int irqno)
+s3c_irq_uart1_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_UART1, 7 << 3);
+ s3c_irqsub_maskack(data->irq, INTMSK_UART1, 7 << 3);
}
static struct irq_chip s3c_irq_uart1 = {
.name = "s3c-uart1",
- .mask = s3c_irq_uart1_mask,
- .unmask = s3c_irq_uart1_unmask,
- .ack = s3c_irq_uart1_ack,
+ .irq_mask = s3c_irq_uart1_mask,
+ .irq_unmask = s3c_irq_uart1_unmask,
+ .irq_ack = s3c_irq_uart1_ack,
};
/* UART2 */
static void
-s3c_irq_uart2_mask(unsigned int irqno)
+s3c_irq_uart2_mask(struct irq_data *data)
{
- s3c_irqsub_mask(irqno, INTMSK_UART2, 7 << 6);
+ s3c_irqsub_mask(data->irq, INTMSK_UART2, 7 << 6);
}
static void
-s3c_irq_uart2_unmask(unsigned int irqno)
+s3c_irq_uart2_unmask(struct irq_data *data)
{
- s3c_irqsub_unmask(irqno, INTMSK_UART2);
+ s3c_irqsub_unmask(data->irq, INTMSK_UART2);
}
static void
-s3c_irq_uart2_ack(unsigned int irqno)
+s3c_irq_uart2_ack(struct irq_data *data)
{
- s3c_irqsub_maskack(irqno, INTMSK_UART2, 7 << 6);
+ s3c_irqsub_maskack(data->irq, INTMSK_UART2, 7 << 6);
}
static struct irq_chip s3c_irq_uart2 = {
.name = "s3c-uart2",
- .mask = s3c_irq_uart2_mask,
- .unmask = s3c_irq_uart2_unmask,
- .ack = s3c_irq_uart2_ack,
+ .irq_mask = s3c_irq_uart2_mask,
+ .irq_unmask = s3c_irq_uart2_unmask,
+ .irq_ack = s3c_irq_uart2_ack,
};
/* ADC and Touchscreen */
static void
-s3c_irq_adc_mask(unsigned int irqno)
+s3c_irq_adc_mask(struct irq_data *d)
{
- s3c_irqsub_mask(irqno, INTMSK_ADCPARENT, 3 << 9);
+ s3c_irqsub_mask(d->irq, INTMSK_ADCPARENT, 3 << 9);
}
static void
-s3c_irq_adc_unmask(unsigned int irqno)
+s3c_irq_adc_unmask(struct irq_data *d)
{
- s3c_irqsub_unmask(irqno, INTMSK_ADCPARENT);
+ s3c_irqsub_unmask(d->irq, INTMSK_ADCPARENT);
}
static void
-s3c_irq_adc_ack(unsigned int irqno)
+s3c_irq_adc_ack(struct irq_data *d)
{
- s3c_irqsub_ack(irqno, INTMSK_ADCPARENT, 3 << 9);
+ s3c_irqsub_ack(d->irq, INTMSK_ADCPARENT, 3 << 9);
}
static struct irq_chip s3c_irq_adc = {
.name = "s3c-adc",
- .mask = s3c_irq_adc_mask,
- .unmask = s3c_irq_adc_unmask,
- .ack = s3c_irq_adc_ack,
+ .irq_mask = s3c_irq_adc_mask,
+ .irq_unmask = s3c_irq_adc_unmask,
+ .irq_ack = s3c_irq_adc_ack,
};
/* irq demux for adc */
diff --git a/arch/arm/plat-s3c24xx/s3c2443-clock.c b/arch/arm/plat-s3c24xx/s3c2443-clock.c
index 461f070..82f2d4a 100644
--- a/arch/arm/plat-s3c24xx/s3c2443-clock.c
+++ b/arch/arm/plat-s3c24xx/s3c2443-clock.c
@@ -271,7 +271,7 @@
.ctrlbit = S3C2443_HCLKCON_DMA5,
}, {
.name = "hsmmc",
- .id = 0,
+ .id = 1,
.parent = &clk_h,
.enable = s3c2443_clkcon_enable_h,
.ctrlbit = S3C2443_HCLKCON_HSMMC,
diff --git a/arch/arm/plat-s5p/Kconfig b/arch/arm/plat-s5p/Kconfig
index 65dbfa8..deb3995 100644
--- a/arch/arm/plat-s5p/Kconfig
+++ b/arch/arm/plat-s5p/Kconfig
@@ -56,3 +56,29 @@
bool
help
Compile in platform device definition for OneNAND controller
+
+config S5P_DEV_CSIS0
+ bool
+ help
+ Compile in platform device definitions for MIPI-CSIS channel 0
+
+config S5P_DEV_CSIS1
+ bool
+ help
+ Compile in platform device definitions for MIPI-CSIS channel 1
+
+menuconfig S5P_SYSMMU
+ bool "SYSMMU support"
+ depends on ARCH_S5PV310
+ help
+ This is a System MMU driver for Samsung ARM based Soc.
+
+if S5P_SYSMMU
+
+config S5P_SYSMMU_DEBUG
+ bool "Enables debug messages"
+ depends on S5P_SYSMMU
+ help
+ This enables SYSMMU driver debug massages.
+
+endif
diff --git a/arch/arm/plat-s5p/Makefile b/arch/arm/plat-s5p/Makefile
index de65238..92efe1a 100644
--- a/arch/arm/plat-s5p/Makefile
+++ b/arch/arm/plat-s5p/Makefile
@@ -28,3 +28,6 @@
obj-$(CONFIG_S5P_DEV_FIMC1) += dev-fimc1.o
obj-$(CONFIG_S5P_DEV_FIMC2) += dev-fimc2.o
obj-$(CONFIG_S5P_DEV_ONENAND) += dev-onenand.o
+obj-$(CONFIG_S5P_DEV_CSIS0) += dev-csis0.o
+obj-$(CONFIG_S5P_DEV_CSIS1) += dev-csis1.o
+obj-$(CONFIG_S5P_SYSMMU) += sysmmu.o
diff --git a/arch/arm/plat-s5p/cpu.c b/arch/arm/plat-s5p/cpu.c
index 74f7f5a..047d31c 100644
--- a/arch/arm/plat-s5p/cpu.c
+++ b/arch/arm/plat-s5p/cpu.c
@@ -108,6 +108,11 @@
.pfn = __phys_to_pfn(S3C_PA_WDT),
.length = SZ_4K,
.type = MT_DEVICE,
+ }, {
+ .virtual = (unsigned long)S5P_VA_SROMC,
+ .pfn = __phys_to_pfn(S5P_PA_SROMC),
+ .length = SZ_4K,
+ .type = MT_DEVICE,
},
};
diff --git a/arch/arm/plat-s5p/dev-csis0.c b/arch/arm/plat-s5p/dev-csis0.c
new file mode 100644
index 0000000..dfab1c8
--- /dev/null
+++ b/arch/arm/plat-s5p/dev-csis0.c
@@ -0,0 +1,34 @@
+/*
+ * Copyright (C) 2010 Samsung Electronics
+ *
+ * S5P series device definition for MIPI-CSIS channel 0
+ *
+ * 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/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <mach/map.h>
+
+static struct resource s5p_mipi_csis0_resource[] = {
+ [0] = {
+ .start = S5P_PA_MIPI_CSIS0,
+ .end = S5P_PA_MIPI_CSIS0 + SZ_4K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [1] = {
+ .start = IRQ_MIPI_CSIS0,
+ .end = IRQ_MIPI_CSIS0,
+ .flags = IORESOURCE_IRQ,
+ }
+};
+
+struct platform_device s5p_device_mipi_csis0 = {
+ .name = "s5p-mipi-csis",
+ .id = 0,
+ .num_resources = ARRAY_SIZE(s5p_mipi_csis0_resource),
+ .resource = s5p_mipi_csis0_resource,
+};
diff --git a/arch/arm/plat-s5p/dev-csis1.c b/arch/arm/plat-s5p/dev-csis1.c
new file mode 100644
index 0000000..e3053f2
--- /dev/null
+++ b/arch/arm/plat-s5p/dev-csis1.c
@@ -0,0 +1,34 @@
+/*
+ * Copyright (C) 2010 Samsung Electronics
+ *
+ * S5P series device definition for MIPI-CSIS channel 1
+ *
+ * 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/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <mach/map.h>
+
+static struct resource s5p_mipi_csis1_resource[] = {
+ [0] = {
+ .start = S5P_PA_MIPI_CSIS1,
+ .end = S5P_PA_MIPI_CSIS1 + SZ_4K - 1,
+ .flags = IORESOURCE_MEM,
+ },
+ [1] = {
+ .start = IRQ_MIPI_CSIS1,
+ .end = IRQ_MIPI_CSIS1,
+ .flags = IORESOURCE_IRQ,
+ },
+};
+
+struct platform_device s5p_device_mipi_csis1 = {
+ .name = "s5p-mipi-csis",
+ .id = 1,
+ .num_resources = ARRAY_SIZE(s5p_mipi_csis1_resource),
+ .resource = s5p_mipi_csis1_resource,
+};
diff --git a/arch/arm/plat-s5p/include/plat/csis.h b/arch/arm/plat-s5p/include/plat/csis.h
new file mode 100644
index 0000000..51e308c
--- /dev/null
+++ b/arch/arm/plat-s5p/include/plat/csis.h
@@ -0,0 +1,28 @@
+/*
+ * Copyright (C) 2010 Samsung Electronics
+ *
+ * S5P series MIPI CSI slave device support
+ *
+ * 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 PLAT_S5P_CSIS_H_
+#define PLAT_S5P_CSIS_H_ __FILE__
+
+/**
+ * struct s5p_platform_mipi_csis - platform data for MIPI-CSIS
+ * @clk_rate: bus clock frequency
+ * @lanes: number of data lanes used
+ * @alignment: data alignment in bits
+ * @hs_settle: HS-RX settle time
+ */
+struct s5p_platform_mipi_csis {
+ unsigned long clk_rate;
+ u8 lanes;
+ u8 alignment;
+ u8 hs_settle;
+};
+
+#endif /* PLAT_S5P_CSIS_H_ */
diff --git a/arch/arm/plat-s5p/include/plat/map-s5p.h b/arch/arm/plat-s5p/include/plat/map-s5p.h
index fef353d..d973d39 100644
--- a/arch/arm/plat-s5p/include/plat/map-s5p.h
+++ b/arch/arm/plat-s5p/include/plat/map-s5p.h
@@ -15,6 +15,7 @@
#define S5P_VA_CHIPID S3C_ADDR(0x02000000)
#define S5P_VA_CMU S3C_ADDR(0x02100000)
+#define S5P_VA_PMU S3C_ADDR(0x02180000)
#define S5P_VA_GPIO S3C_ADDR(0x02200000)
#define S5P_VA_GPIO1 S5P_VA_GPIO
#define S5P_VA_GPIO2 S3C_ADDR(0x02240000)
diff --git a/arch/arm/plat-s5p/include/plat/regs-srom.h b/arch/arm/plat-s5p/include/plat/regs-srom.h
new file mode 100644
index 0000000..f121ab5
--- /dev/null
+++ b/arch/arm/plat-s5p/include/plat/regs-srom.h
@@ -0,0 +1,54 @@
+/* linux/arch/arm/plat-s5p/include/plat/regs-srom.h
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * S5P SROMC register definitions
+ *
+ * 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 __ASM_PLAT_S5P_REGS_SROM_H
+#define __ASM_PLAT_S5P_REGS_SROM_H __FILE__
+
+#include <mach/map.h>
+
+#define S5P_SROMREG(x) (S5P_VA_SROMC + (x))
+
+#define S5P_SROM_BW S5P_SROMREG(0x0)
+#define S5P_SROM_BC0 S5P_SROMREG(0x4)
+#define S5P_SROM_BC1 S5P_SROMREG(0x8)
+#define S5P_SROM_BC2 S5P_SROMREG(0xc)
+#define S5P_SROM_BC3 S5P_SROMREG(0x10)
+#define S5P_SROM_BC4 S5P_SROMREG(0x14)
+#define S5P_SROM_BC5 S5P_SROMREG(0x18)
+
+/* one register BW holds 4 x 4-bit packed settings for NCS0 - NCS3 */
+
+#define S5P_SROM_BW__DATAWIDTH__SHIFT 0
+#define S5P_SROM_BW__ADDRMODE__SHIFT 1
+#define S5P_SROM_BW__WAITENABLE__SHIFT 2
+#define S5P_SROM_BW__BYTEENABLE__SHIFT 3
+
+#define S5P_SROM_BW__CS_MASK 0xf
+
+#define S5P_SROM_BW__NCS0__SHIFT 0
+#define S5P_SROM_BW__NCS1__SHIFT 4
+#define S5P_SROM_BW__NCS2__SHIFT 8
+#define S5P_SROM_BW__NCS3__SHIFT 12
+#define S5P_SROM_BW__NCS4__SHIFT 16
+#define S5P_SROM_BW__NCS5__SHIFT 20
+
+/* applies to same to BCS0 - BCS3 */
+
+#define S5P_SROM_BCX__PMC__SHIFT 0
+#define S5P_SROM_BCX__TACP__SHIFT 4
+#define S5P_SROM_BCX__TCAH__SHIFT 8
+#define S5P_SROM_BCX__TCOH__SHIFT 12
+#define S5P_SROM_BCX__TACC__SHIFT 16
+#define S5P_SROM_BCX__TCOS__SHIFT 24
+#define S5P_SROM_BCX__TACS__SHIFT 28
+
+#endif /* __ASM_PLAT_S5P_REGS_SROM_H */
diff --git a/arch/arm/plat-s5p/include/plat/sysmmu.h b/arch/arm/plat-s5p/include/plat/sysmmu.h
new file mode 100644
index 0000000..db298fc
--- /dev/null
+++ b/arch/arm/plat-s5p/include/plat/sysmmu.h
@@ -0,0 +1,23 @@
+/* linux/arch/arm/plat-s5p/include/plat/sysmmu.h
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com/
+ *
+ * Samsung sysmmu 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 __ASM_PLAT_S5P_SYSMMU_H
+#define __ASM_PLAT_S5P_SYSMMU_H __FILE__
+
+/* debug macro */
+#ifdef CONFIG_S5P_SYSMMU_DEBUG
+#define sysmmu_debug(fmt, arg...) printk(KERN_INFO "[%s] " fmt, __func__, ## arg)
+#else
+#define sysmmu_debug(fmt, arg...) do { } while (0)
+#endif
+
+#endif /* __ASM_PLAT_S5P_SYSMMU_H */
diff --git a/arch/arm/plat-s5p/irq-eint.c b/arch/arm/plat-s5p/irq-eint.c
index 752f1a6..225aa25 100644
--- a/arch/arm/plat-s5p/irq-eint.c
+++ b/arch/arm/plat-s5p/irq-eint.c
@@ -28,39 +28,40 @@
#include <plat/gpio-cfg.h>
#include <mach/regs-gpio.h>
-static inline void s5p_irq_eint_mask(unsigned int irq)
+static inline void s5p_irq_eint_mask(struct irq_data *data)
{
u32 mask;
- mask = __raw_readl(S5P_EINT_MASK(EINT_REG_NR(irq)));
- mask |= eint_irq_to_bit(irq);
- __raw_writel(mask, S5P_EINT_MASK(EINT_REG_NR(irq)));
+ mask = __raw_readl(S5P_EINT_MASK(EINT_REG_NR(data->irq)));
+ mask |= eint_irq_to_bit(data->irq);
+ __raw_writel(mask, S5P_EINT_MASK(EINT_REG_NR(data->irq)));
}
-static void s5p_irq_eint_unmask(unsigned int irq)
+static void s5p_irq_eint_unmask(struct irq_data *data)
{
u32 mask;
- mask = __raw_readl(S5P_EINT_MASK(EINT_REG_NR(irq)));
- mask &= ~(eint_irq_to_bit(irq));
- __raw_writel(mask, S5P_EINT_MASK(EINT_REG_NR(irq)));
+ mask = __raw_readl(S5P_EINT_MASK(EINT_REG_NR(data->irq)));
+ mask &= ~(eint_irq_to_bit(data->irq));
+ __raw_writel(mask, S5P_EINT_MASK(EINT_REG_NR(data->irq)));
}
-static inline void s5p_irq_eint_ack(unsigned int irq)
+static inline void s5p_irq_eint_ack(struct irq_data *data)
{
- __raw_writel(eint_irq_to_bit(irq), S5P_EINT_PEND(EINT_REG_NR(irq)));
+ __raw_writel(eint_irq_to_bit(data->irq),
+ S5P_EINT_PEND(EINT_REG_NR(data->irq)));
}
-static void s5p_irq_eint_maskack(unsigned int irq)
+static void s5p_irq_eint_maskack(struct irq_data *data)
{
/* compiler should in-line these */
- s5p_irq_eint_mask(irq);
- s5p_irq_eint_ack(irq);
+ s5p_irq_eint_mask(data);
+ s5p_irq_eint_ack(data);
}
-static int s5p_irq_eint_set_type(unsigned int irq, unsigned int type)
+static int s5p_irq_eint_set_type(struct irq_data *data, unsigned int type)
{
- int offs = EINT_OFFSET(irq);
+ int offs = EINT_OFFSET(data->irq);
int shift;
u32 ctrl, mask;
u32 newvalue = 0;
@@ -94,10 +95,10 @@
shift = (offs & 0x7) * 4;
mask = 0x7 << shift;
- ctrl = __raw_readl(S5P_EINT_CON(EINT_REG_NR(irq)));
+ ctrl = __raw_readl(S5P_EINT_CON(EINT_REG_NR(data->irq)));
ctrl &= ~mask;
ctrl |= newvalue << shift;
- __raw_writel(ctrl, S5P_EINT_CON(EINT_REG_NR(irq)));
+ __raw_writel(ctrl, S5P_EINT_CON(EINT_REG_NR(data->irq)));
if ((0 <= offs) && (offs < 8))
s3c_gpio_cfgpin(EINT_GPIO_0(offs & 0x7), EINT_MODE);
@@ -119,13 +120,13 @@
static struct irq_chip s5p_irq_eint = {
.name = "s5p-eint",
- .mask = s5p_irq_eint_mask,
- .unmask = s5p_irq_eint_unmask,
- .mask_ack = s5p_irq_eint_maskack,
- .ack = s5p_irq_eint_ack,
- .set_type = s5p_irq_eint_set_type,
+ .irq_mask = s5p_irq_eint_mask,
+ .irq_unmask = s5p_irq_eint_unmask,
+ .irq_mask_ack = s5p_irq_eint_maskack,
+ .irq_ack = s5p_irq_eint_ack,
+ .irq_set_type = s5p_irq_eint_set_type,
#ifdef CONFIG_PM
- .set_wake = s3c_irqext_wake,
+ .irq_set_wake = s3c_irqext_wake,
#endif
};
@@ -159,42 +160,43 @@
s5p_irq_demux_eint(IRQ_EINT(24));
}
-static inline void s5p_irq_vic_eint_mask(unsigned int irq)
+static inline void s5p_irq_vic_eint_mask(struct irq_data *data)
{
- void __iomem *base = get_irq_chip_data(irq);
+ void __iomem *base = irq_data_get_irq_chip_data(data);
- s5p_irq_eint_mask(irq);
- writel(1 << EINT_OFFSET(irq), base + VIC_INT_ENABLE_CLEAR);
+ s5p_irq_eint_mask(data);
+ writel(1 << EINT_OFFSET(data->irq), base + VIC_INT_ENABLE_CLEAR);
}
-static void s5p_irq_vic_eint_unmask(unsigned int irq)
+static void s5p_irq_vic_eint_unmask(struct irq_data *data)
{
- void __iomem *base = get_irq_chip_data(irq);
+ void __iomem *base = irq_data_get_irq_chip_data(data);
- s5p_irq_eint_unmask(irq);
- writel(1 << EINT_OFFSET(irq), base + VIC_INT_ENABLE);
+ s5p_irq_eint_unmask(data);
+ writel(1 << EINT_OFFSET(data->irq), base + VIC_INT_ENABLE);
}
-static inline void s5p_irq_vic_eint_ack(unsigned int irq)
+static inline void s5p_irq_vic_eint_ack(struct irq_data *data)
{
- __raw_writel(eint_irq_to_bit(irq), S5P_EINT_PEND(EINT_REG_NR(irq)));
+ __raw_writel(eint_irq_to_bit(data->irq),
+ S5P_EINT_PEND(EINT_REG_NR(data->irq)));
}
-static void s5p_irq_vic_eint_maskack(unsigned int irq)
+static void s5p_irq_vic_eint_maskack(struct irq_data *data)
{
- s5p_irq_vic_eint_mask(irq);
- s5p_irq_vic_eint_ack(irq);
+ s5p_irq_vic_eint_mask(data);
+ s5p_irq_vic_eint_ack(data);
}
static struct irq_chip s5p_irq_vic_eint = {
.name = "s5p_vic_eint",
- .mask = s5p_irq_vic_eint_mask,
- .unmask = s5p_irq_vic_eint_unmask,
- .mask_ack = s5p_irq_vic_eint_maskack,
- .ack = s5p_irq_vic_eint_ack,
- .set_type = s5p_irq_eint_set_type,
+ .irq_mask = s5p_irq_vic_eint_mask,
+ .irq_unmask = s5p_irq_vic_eint_unmask,
+ .irq_mask_ack = s5p_irq_vic_eint_maskack,
+ .irq_ack = s5p_irq_vic_eint_ack,
+ .irq_set_type = s5p_irq_eint_set_type,
#ifdef CONFIG_PM
- .set_wake = s3c_irqext_wake,
+ .irq_set_wake = s3c_irqext_wake,
#endif
};
diff --git a/arch/arm/plat-s5p/irq-gpioint.c b/arch/arm/plat-s5p/irq-gpioint.c
index 0e5dc8c..3b6bf89 100644
--- a/arch/arm/plat-s5p/irq-gpioint.c
+++ b/arch/arm/plat-s5p/irq-gpioint.c
@@ -30,9 +30,9 @@
static struct s3c_gpio_chip *irq_chips[S5P_GPIOINT_GROUP_MAXNR];
-static int s5p_gpioint_get_group(unsigned int irq)
+static int s5p_gpioint_get_group(struct irq_data *data)
{
- struct gpio_chip *chip = get_irq_data(irq);
+ struct gpio_chip *chip = irq_data_get_irq_data(data);
struct s3c_gpio_chip *s3c_chip = container_of(chip,
struct s3c_gpio_chip, chip);
int group;
@@ -44,22 +44,22 @@
return group;
}
-static int s5p_gpioint_get_offset(unsigned int irq)
+static int s5p_gpioint_get_offset(struct irq_data *data)
{
- struct gpio_chip *chip = get_irq_data(irq);
+ struct gpio_chip *chip = irq_data_get_irq_data(data);
struct s3c_gpio_chip *s3c_chip = container_of(chip,
struct s3c_gpio_chip, chip);
- return irq - s3c_chip->irq_base;
+ return data->irq - s3c_chip->irq_base;
}
-static void s5p_gpioint_ack(unsigned int irq)
+static void s5p_gpioint_ack(struct irq_data *data)
{
int group, offset, pend_offset;
unsigned int value;
- group = s5p_gpioint_get_group(irq);
- offset = s5p_gpioint_get_offset(irq);
+ group = s5p_gpioint_get_group(data);
+ offset = s5p_gpioint_get_offset(data);
pend_offset = group << 2;
value = __raw_readl(S5P_GPIOREG(GPIOINT_PEND_OFFSET) + pend_offset);
@@ -67,13 +67,13 @@
__raw_writel(value, S5P_GPIOREG(GPIOINT_PEND_OFFSET) + pend_offset);
}
-static void s5p_gpioint_mask(unsigned int irq)
+static void s5p_gpioint_mask(struct irq_data *data)
{
int group, offset, mask_offset;
unsigned int value;
- group = s5p_gpioint_get_group(irq);
- offset = s5p_gpioint_get_offset(irq);
+ group = s5p_gpioint_get_group(data);
+ offset = s5p_gpioint_get_offset(data);
mask_offset = group << 2;
value = __raw_readl(S5P_GPIOREG(GPIOINT_MASK_OFFSET) + mask_offset);
@@ -81,13 +81,13 @@
__raw_writel(value, S5P_GPIOREG(GPIOINT_MASK_OFFSET) + mask_offset);
}
-static void s5p_gpioint_unmask(unsigned int irq)
+static void s5p_gpioint_unmask(struct irq_data *data)
{
int group, offset, mask_offset;
unsigned int value;
- group = s5p_gpioint_get_group(irq);
- offset = s5p_gpioint_get_offset(irq);
+ group = s5p_gpioint_get_group(data);
+ offset = s5p_gpioint_get_offset(data);
mask_offset = group << 2;
value = __raw_readl(S5P_GPIOREG(GPIOINT_MASK_OFFSET) + mask_offset);
@@ -95,19 +95,19 @@
__raw_writel(value, S5P_GPIOREG(GPIOINT_MASK_OFFSET) + mask_offset);
}
-static void s5p_gpioint_mask_ack(unsigned int irq)
+static void s5p_gpioint_mask_ack(struct irq_data *data)
{
- s5p_gpioint_mask(irq);
- s5p_gpioint_ack(irq);
+ s5p_gpioint_mask(data);
+ s5p_gpioint_ack(data);
}
-static int s5p_gpioint_set_type(unsigned int irq, unsigned int type)
+static int s5p_gpioint_set_type(struct irq_data *data, unsigned int type)
{
int group, offset, con_offset;
unsigned int value;
- group = s5p_gpioint_get_group(irq);
- offset = s5p_gpioint_get_offset(irq);
+ group = s5p_gpioint_get_group(data);
+ offset = s5p_gpioint_get_offset(data);
con_offset = group << 2;
switch (type) {
@@ -142,11 +142,11 @@
struct irq_chip s5p_gpioint = {
.name = "s5p_gpioint",
- .ack = s5p_gpioint_ack,
- .mask = s5p_gpioint_mask,
- .mask_ack = s5p_gpioint_mask_ack,
- .unmask = s5p_gpioint_unmask,
- .set_type = s5p_gpioint_set_type,
+ .irq_ack = s5p_gpioint_ack,
+ .irq_mask = s5p_gpioint_mask,
+ .irq_mask_ack = s5p_gpioint_mask_ack,
+ .irq_unmask = s5p_gpioint_unmask,
+ .irq_set_type = s5p_gpioint_set_type,
};
static void s5p_gpioint_handler(unsigned int irq, struct irq_desc *desc)
diff --git a/arch/arm/plat-s5p/irq-pm.c b/arch/arm/plat-s5p/irq-pm.c
index dc33b9e..5259ad4 100644
--- a/arch/arm/plat-s5p/irq-pm.c
+++ b/arch/arm/plat-s5p/irq-pm.c
@@ -37,14 +37,14 @@
unsigned long s3c_irqwake_intallow = 0x00000006L;
unsigned long s3c_irqwake_eintallow = 0xffffffffL;
-int s3c_irq_wake(unsigned int irqno, unsigned int state)
+int s3c_irq_wake(struct irq_data *data, unsigned int state)
{
unsigned long irqbit;
- switch (irqno) {
+ switch (data->irq) {
case IRQ_RTC_TIC:
case IRQ_RTC_ALARM:
- irqbit = 1 << (irqno + 1 - IRQ_RTC_ALARM);
+ irqbit = 1 << (data->irq + 1 - IRQ_RTC_ALARM);
if (!state)
s3c_irqwake_intmask |= irqbit;
else
diff --git a/arch/arm/plat-s5p/sysmmu.c b/arch/arm/plat-s5p/sysmmu.c
new file mode 100644
index 0000000..d804914
--- /dev/null
+++ b/arch/arm/plat-s5p/sysmmu.c
@@ -0,0 +1,328 @@
+/* linux/arch/arm/plat-s5p/sysmmu.c
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * 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/io.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+
+#include <mach/map.h>
+#include <mach/regs-sysmmu.h>
+#include <mach/sysmmu.h>
+
+#include <plat/sysmmu.h>
+
+struct sysmmu_controller s5p_sysmmu_cntlrs[S5P_SYSMMU_TOTAL_IPNUM];
+
+void s5p_sysmmu_register(struct sysmmu_controller *sysmmuconp)
+{
+ unsigned int reg_mmu_ctrl;
+ unsigned int reg_mmu_status;
+ unsigned int reg_pt_base_addr;
+ unsigned int reg_int_status;
+ unsigned int reg_page_ft_addr;
+
+ reg_int_status = __raw_readl(sysmmuconp->regs + S5P_INT_STATUS);
+ reg_mmu_ctrl = __raw_readl(sysmmuconp->regs + S5P_MMU_CTRL);
+ reg_mmu_status = __raw_readl(sysmmuconp->regs + S5P_MMU_STATUS);
+ reg_pt_base_addr = __raw_readl(sysmmuconp->regs + S5P_PT_BASE_ADDR);
+ reg_page_ft_addr = __raw_readl(sysmmuconp->regs + S5P_PAGE_FAULT_ADDR);
+
+ printk(KERN_INFO "%s: ips:%s\n", __func__, sysmmuconp->name);
+ printk(KERN_INFO "%s: MMU_CTRL:0x%X, ", __func__, reg_mmu_ctrl);
+ printk(KERN_INFO "MMU_STATUS:0x%X, PT_BASE_ADDR:0x%X\n", reg_mmu_status, reg_pt_base_addr);
+ printk(KERN_INFO "%s: INT_STATUS:0x%X, PAGE_FAULT_ADDR:0x%X\n", __func__, reg_int_status, reg_page_ft_addr);
+
+ switch (reg_int_status & 0xFF) {
+ case 0x1:
+ printk(KERN_INFO "%s: Page fault\n", __func__);
+ printk(KERN_INFO "%s: Virtual address causing last page fault or bus error : 0x%x\n", __func__ , reg_page_ft_addr);
+ break;
+ case 0x2:
+ printk(KERN_INFO "%s: AR multi-hit fault\n", __func__);
+ break;
+ case 0x4:
+ printk(KERN_INFO "%s: AW multi-hit fault\n", __func__);
+ break;
+ case 0x8:
+ printk(KERN_INFO "%s: Bus error\n", __func__);
+ break;
+ case 0x10:
+ printk(KERN_INFO "%s: AR Security protection fault\n", __func__);
+ break;
+ case 0x20:
+ printk(KERN_INFO "%s: AR Access protection fault\n", __func__);
+ break;
+ case 0x40:
+ printk(KERN_INFO "%s: AW Security protection fault\n", __func__);
+ break;
+ case 0x80:
+ printk(KERN_INFO "%s: AW Access protection fault\n", __func__);
+ break;
+ }
+}
+
+static irqreturn_t s5p_sysmmu_irq(int irq, void *dev_id)
+{
+ unsigned int i;
+ unsigned int reg_int_status;
+ struct sysmmu_controller *sysmmuconp;
+
+ for (i = 0; i < S5P_SYSMMU_TOTAL_IPNUM; i++) {
+ sysmmuconp = &s5p_sysmmu_cntlrs[i];
+
+ if (sysmmuconp->enable == true) {
+ reg_int_status = __raw_readl(sysmmuconp->regs + S5P_INT_STATUS);
+
+ if (reg_int_status & 0xFF)
+ s5p_sysmmu_register(sysmmuconp);
+ }
+ }
+ return IRQ_HANDLED;
+}
+
+int s5p_sysmmu_set_tablebase_pgd(sysmmu_ips ips, unsigned long pgd)
+{
+ struct sysmmu_controller *sysmmuconp = NULL;
+
+ sysmmuconp = &s5p_sysmmu_cntlrs[ips];
+
+ if (sysmmuconp == NULL) {
+ printk(KERN_ERR "failed to get ip's sysmmu info\n");
+ return 1;
+ }
+
+ /* Set sysmmu page table base address */
+ __raw_writel(pgd, sysmmuconp->regs + S5P_PT_BASE_ADDR);
+
+ if (s5p_sysmmu_tlb_invalidate(ips) != 0)
+ printk(KERN_ERR "failed s5p_sysmmu_tlb_invalidate\n");
+
+ return 0;
+}
+
+static int s5p_sysmmu_set_tablebase(sysmmu_ips ips)
+{
+ unsigned int pg;
+ struct sysmmu_controller *sysmmuconp;
+
+ sysmmuconp = &s5p_sysmmu_cntlrs[ips];
+
+ if (sysmmuconp == NULL) {
+ printk(KERN_ERR "failed to get ip's sysmmu info\n");
+ return 1;
+ }
+
+ __asm__("mrc p15, 0, %0, c2, c0, 0" \
+ : "=r" (pg) : : "cc"); \
+ pg &= ~0x3fff;
+
+ sysmmu_debug("CP15 TTBR0 : 0x%x\n", pg);
+
+ /* Set sysmmu page table base address */
+ __raw_writel(pg, sysmmuconp->regs + S5P_PT_BASE_ADDR);
+
+ return 0;
+}
+
+int s5p_sysmmu_enable(sysmmu_ips ips)
+{
+ unsigned int reg;
+
+ struct sysmmu_controller *sysmmuconp;
+
+ sysmmuconp = &s5p_sysmmu_cntlrs[ips];
+
+ if (sysmmuconp == NULL) {
+ printk(KERN_ERR "failed to get ip's sysmmu info\n");
+ return 1;
+ }
+
+ s5p_sysmmu_set_tablebase(ips);
+
+ /* replacement policy : LRU */
+ reg = __raw_readl(sysmmuconp->regs + S5P_MMU_CFG);
+ reg |= 0x1;
+ __raw_writel(reg, sysmmuconp->regs + S5P_MMU_CFG);
+
+ /* Enable interrupt, Enable MMU */
+ reg = __raw_readl(sysmmuconp->regs + S5P_MMU_CTRL);
+ reg |= (0x1 << 2) | (0x1 << 0);
+
+ __raw_writel(reg, sysmmuconp->regs + S5P_MMU_CTRL);
+
+ sysmmuconp->enable = true;
+
+ return 0;
+}
+
+int s5p_sysmmu_disable(sysmmu_ips ips)
+{
+ unsigned int reg;
+
+ struct sysmmu_controller *sysmmuconp = NULL;
+
+ if (ips > S5P_SYSMMU_TOTAL_IPNUM)
+ printk(KERN_ERR "failed to get ips parameter\n");
+
+ sysmmuconp = &s5p_sysmmu_cntlrs[ips];
+
+ if (sysmmuconp == NULL) {
+ printk(KERN_ERR "failed to get ip's sysmmu info\n");
+ return 1;
+ }
+
+ reg = __raw_readl(sysmmuconp->regs + S5P_MMU_CFG);
+
+ /* replacement policy : LRU */
+ reg |= 0x1;
+ __raw_writel(reg, sysmmuconp->regs + S5P_MMU_CFG);
+
+ reg = __raw_readl(sysmmuconp->regs + S5P_MMU_CTRL);
+
+ /* Disable MMU */
+ reg &= ~0x1;
+ __raw_writel(reg, sysmmuconp->regs + S5P_MMU_CTRL);
+
+ sysmmuconp->enable = false;
+
+ return 0;
+}
+
+int s5p_sysmmu_tlb_invalidate(sysmmu_ips ips)
+{
+ unsigned int reg;
+ struct sysmmu_controller *sysmmuconp = NULL;
+
+ sysmmuconp = &s5p_sysmmu_cntlrs[ips];
+
+ if (sysmmuconp == NULL) {
+ printk(KERN_ERR "failed to get ip's sysmmu info\n");
+ return 1;
+ }
+
+ /* set Block MMU for flush TLB */
+ reg = __raw_readl(sysmmuconp->regs + S5P_MMU_CTRL);
+ reg |= 0x1 << 1;
+ __raw_writel(reg, sysmmuconp->regs + S5P_MMU_CTRL);
+
+ /* flush all TLB entry */
+ __raw_writel(0x1, sysmmuconp->regs + S5P_MMU_FLUSH);
+
+ /* set Un-block MMU after flush TLB */
+ reg = __raw_readl(sysmmuconp->regs + S5P_MMU_CTRL);
+ reg &= ~(0x1 << 1);
+ __raw_writel(reg, sysmmuconp->regs + S5P_MMU_CTRL);
+
+ return 0;
+}
+
+static int s5p_sysmmu_probe(struct platform_device *pdev)
+{
+ int i;
+ int ret;
+ struct resource *res;
+ struct sysmmu_controller *sysmmuconp;
+ sysmmu_ips ips;
+
+ for (i = 0; i < S5P_SYSMMU_TOTAL_IPNUM; i++) {
+ sysmmuconp = &s5p_sysmmu_cntlrs[i];
+ if (sysmmuconp == NULL) {
+ printk(KERN_ERR "failed to get ip's sysmmu info\n");
+ ret = -ENOENT;
+ goto err_res;
+ }
+
+ sysmmuconp->name = sysmmu_ips_name[i];
+
+ res = platform_get_resource(pdev, IORESOURCE_MEM, i);
+ if (!res) {
+ printk(KERN_ERR "failed to get sysmmu resource\n");
+ ret = -ENODEV;
+ goto err_res;
+ }
+
+ sysmmuconp->mem = request_mem_region(res->start,
+ ((res->end) - (res->start)) + 1, pdev->name);
+ if (!sysmmuconp->mem) {
+ pr_err("failed to request sysmmu memory region\n");
+ ret = -EBUSY;
+ goto err_res;
+ }
+
+ sysmmuconp->regs = ioremap(res->start, res->end - res->start + 1);
+ if (!sysmmuconp->regs) {
+ pr_err("failed to sysmmu ioremap\n");
+ ret = -ENXIO;
+ goto err_reg;
+ }
+
+ sysmmuconp->irq = platform_get_irq(pdev, i);
+ if (sysmmuconp->irq <= 0) {
+ pr_err("failed to get sysmmu irq resource\n");
+ ret = -ENOENT;
+ goto err_map;
+ }
+
+ ret = request_irq(sysmmuconp->irq, s5p_sysmmu_irq, IRQF_DISABLED, pdev->name, sysmmuconp);
+ if (ret) {
+ pr_err("failed to request irq\n");
+ ret = -ENOENT;
+ goto err_map;
+ }
+
+ ips = (sysmmu_ips)i;
+
+ sysmmuconp->ips = ips;
+ }
+
+ return 0;
+
+err_reg:
+ release_mem_region((resource_size_t)sysmmuconp->mem, (resource_size_t)((res->end) - (res->start) + 1));
+err_map:
+ iounmap(sysmmuconp->regs);
+err_res:
+ return ret;
+}
+
+static int s5p_sysmmu_remove(struct platform_device *pdev)
+{
+ return 0;
+}
+int s5p_sysmmu_runtime_suspend(struct device *dev)
+{
+ return 0;
+}
+
+int s5p_sysmmu_runtime_resume(struct device *dev)
+{
+ return 0;
+}
+
+const struct dev_pm_ops s5p_sysmmu_pm_ops = {
+ .runtime_suspend = s5p_sysmmu_runtime_suspend,
+ .runtime_resume = s5p_sysmmu_runtime_resume,
+};
+
+static struct platform_driver s5p_sysmmu_driver = {
+ .probe = s5p_sysmmu_probe,
+ .remove = s5p_sysmmu_remove,
+ .driver = {
+ .owner = THIS_MODULE,
+ .name = "s5p-sysmmu",
+ .pm = &s5p_sysmmu_pm_ops,
+ }
+};
+
+static int __init s5p_sysmmu_init(void)
+{
+ return platform_driver_register(&s5p_sysmmu_driver);
+}
+arch_initcall(s5p_sysmmu_init);
diff --git a/arch/arm/plat-samsung/Kconfig b/arch/arm/plat-samsung/Kconfig
index dcd6eff4e..32be05c 100644
--- a/arch/arm/plat-samsung/Kconfig
+++ b/arch/arm/plat-samsung/Kconfig
@@ -95,6 +95,12 @@
help
Internal configuration to enable the correct GPIO pull helper
+config S3C_GPIO_PULL_S3C2443
+ bool
+ select S3C_GPIO_PULL_UPDOWN
+ help
+ Internal configuration to enable the correct GPIO pull helper for S3C2443-style GPIO
+
config S3C_GPIO_PULL_DOWN
bool
help
@@ -333,4 +339,12 @@
and above. This code allows a set of interrupt to wakeup-mask
mappings. See <plat/wakeup-mask.h>
+comment "Power Domain"
+
+config SAMSUNG_PD
+ bool "Samsung Power Domain"
+ depends on PM_RUNTIME
+ help
+ Say Y here if you want to control Power Domain by Runtime PM.
+
endif
diff --git a/arch/arm/plat-samsung/Makefile b/arch/arm/plat-samsung/Makefile
index 19d8a16..29932f8 100644
--- a/arch/arm/plat-samsung/Makefile
+++ b/arch/arm/plat-samsung/Makefile
@@ -74,6 +74,10 @@
obj-$(CONFIG_SAMSUNG_WAKEMASK) += wakeup-mask.o
+# PD support
+
+obj-$(CONFIG_SAMSUNG_PD) += pd.o
+
# PWM support
obj-$(CONFIG_HAVE_PWM) += pwm.o
diff --git a/arch/arm/plat-samsung/clock.c b/arch/arm/plat-samsung/clock.c
index e8d20b0..7728928 100644
--- a/arch/arm/plat-samsung/clock.c
+++ b/arch/arm/plat-samsung/clock.c
@@ -39,6 +39,9 @@
#include <linux/clk.h>
#include <linux/spinlock.h>
#include <linux/io.h>
+#if defined(CONFIG_DEBUG_FS)
+#include <linux/debugfs.h>
+#endif
#include <mach/hardware.h>
#include <asm/irq.h>
@@ -447,3 +450,92 @@
return 0;
}
+#if defined(CONFIG_PM_DEBUG) && defined(CONFIG_DEBUG_FS)
+/* 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, *child, *child_tmp;
+ struct clk *pa = c->parent;
+ char s[255];
+ char *p = s;
+
+ p += sprintf(p, "%s", c->name);
+
+ if (c->id >= 0)
+ sprintf(p, ":%d", c->id);
+
+ d = debugfs_create_dir(s, pa ? pa->dent : clk_debugfs_root);
+ if (!d)
+ return -ENOMEM;
+
+ c->dent = d;
+
+ d = debugfs_create_u8("usecount", S_IRUGO, c->dent, (u8 *)&c->usage);
+ if (!d) {
+ err = -ENOMEM;
+ goto err_out;
+ }
+
+ d = debugfs_create_u32("rate", S_IRUGO, c->dent, (u32 *)&c->rate);
+ if (!d) {
+ err = -ENOMEM;
+ goto err_out;
+ }
+ return 0;
+
+err_out:
+ d = c->dent;
+ list_for_each_entry_safe(child, child_tmp, &d->d_subdirs, d_u.d_child)
+ debugfs_remove(child);
+ debugfs_remove(c->dent);
+ return err;
+}
+
+static int clk_debugfs_register(struct clk *c)
+{
+ int err;
+ struct clk *pa = c->parent;
+
+ if (pa && !pa->dent) {
+ err = clk_debugfs_register(pa);
+ if (err)
+ return err;
+ }
+
+ if (!c->dent) {
+ 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, &clocks, list) {
+ 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);
+
+#endif /* defined(CONFIG_PM_DEBUG) && defined(CONFIG_DEBUG_FS) */
diff --git a/arch/arm/plat-samsung/dev-nand.c b/arch/arm/plat-samsung/dev-nand.c
index 3a7b889..6927ae8 100644
--- a/arch/arm/plat-samsung/dev-nand.c
+++ b/arch/arm/plat-samsung/dev-nand.c
@@ -126,5 +126,3 @@
s3c_device_nand.dev.platform_data = npd;
}
-
-EXPORT_SYMBOL_GPL(s3c_nand_set_platdata);
diff --git a/arch/arm/plat-samsung/gpio-config.c b/arch/arm/plat-samsung/gpio-config.c
index 0aa32f2..1c0b040 100644
--- a/arch/arm/plat-samsung/gpio-config.c
+++ b/arch/arm/plat-samsung/gpio-config.c
@@ -278,6 +278,48 @@
pup &= 0x3;
return (__force s3c_gpio_pull_t)pup;
}
+
+#ifdef CONFIG_S3C_GPIO_PULL_S3C2443
+int s3c_gpio_setpull_s3c2443(struct s3c_gpio_chip *chip,
+ unsigned int off, s3c_gpio_pull_t pull)
+{
+ switch (pull) {
+ case S3C_GPIO_PULL_NONE:
+ pull = 0x01;
+ break;
+ case S3C_GPIO_PULL_UP:
+ pull = 0x00;
+ break;
+ case S3C_GPIO_PULL_DOWN:
+ pull = 0x02;
+ break;
+ }
+ return s3c_gpio_setpull_updown(chip, off, pull);
+}
+
+s3c_gpio_pull_t s3c_gpio_getpull_s3c2443(struct s3c_gpio_chip *chip,
+ unsigned int off)
+{
+ s3c_gpio_pull_t pull;
+
+ pull = s3c_gpio_getpull_updown(chip, off);
+
+ switch (pull) {
+ case 0x00:
+ pull = S3C_GPIO_PULL_UP;
+ break;
+ case 0x01:
+ case 0x03:
+ pull = S3C_GPIO_PULL_NONE;
+ break;
+ case 0x02:
+ pull = S3C_GPIO_PULL_DOWN;
+ break;
+ }
+
+ return pull;
+}
+#endif
#endif
#if defined(CONFIG_S3C_GPIO_PULL_UP) || defined(CONFIG_S3C_GPIO_PULL_DOWN)
diff --git a/arch/arm/plat-samsung/gpiolib.c b/arch/arm/plat-samsung/gpiolib.c
index c354089..ea37c04 100644
--- a/arch/arm/plat-samsung/gpiolib.c
+++ b/arch/arm/plat-samsung/gpiolib.c
@@ -197,3 +197,10 @@
s3c_gpiolib_add(chip);
}
}
+
+void __init samsung_gpiolib_add_2bit_chips(struct s3c_gpio_chip *chip,
+ int nr_chips)
+{
+ for (; nr_chips > 0; nr_chips--, chip++)
+ s3c_gpiolib_add(chip);
+}
diff --git a/arch/arm/plat-samsung/include/plat/clock.h b/arch/arm/plat-samsung/include/plat/clock.h
index 0fbcd0e..9a82b88 100644
--- a/arch/arm/plat-samsung/include/plat/clock.h
+++ b/arch/arm/plat-samsung/include/plat/clock.h
@@ -47,6 +47,9 @@
struct clk_ops *ops;
int (*enable)(struct clk *, int enable);
+#if defined(CONFIG_PM_DEBUG) && defined(CONFIG_DEBUG_FS)
+ struct dentry *dent; /* For visible tree hierarchy */
+#endif
};
/* other clocks which may be registered by board support */
diff --git a/arch/arm/plat-samsung/include/plat/devs.h b/arch/arm/plat-samsung/include/plat/devs.h
index e9e3b6e..b4d208b 100644
--- a/arch/arm/plat-samsung/include/plat/devs.h
+++ b/arch/arm/plat-samsung/include/plat/devs.h
@@ -104,6 +104,7 @@
extern struct platform_device s5pv310_device_i2s1;
extern struct platform_device s5pv310_device_i2s2;
extern struct platform_device s5pv310_device_spdif;
+extern struct platform_device s5pv310_device_pd[];
extern struct platform_device s5p6442_device_pcm0;
extern struct platform_device s5p6442_device_pcm1;
@@ -115,6 +116,8 @@
extern struct platform_device s5p6440_device_iis;
extern struct platform_device s5p6450_device_iis0;
+extern struct platform_device s5p6450_device_iis1;
+extern struct platform_device s5p6450_device_iis2;
extern struct platform_device s5p6450_device_pcm0;
extern struct platform_device s5pc100_device_ac97;
@@ -131,6 +134,11 @@
extern struct platform_device s5p_device_fimc1;
extern struct platform_device s5p_device_fimc2;
+extern struct platform_device s5p_device_mipi_csis0;
+extern struct platform_device s5p_device_mipi_csis1;
+
+extern struct platform_device s5pv310_device_sysmmu;
+
/* s3c2440 specific devices */
#ifdef CONFIG_CPU_S3C2440
diff --git a/arch/arm/plat-samsung/include/plat/gpio-cfg-helpers.h b/arch/arm/plat-samsung/include/plat/gpio-cfg-helpers.h
index 0d2c570..5603db0 100644
--- a/arch/arm/plat-samsung/include/plat/gpio-cfg-helpers.h
+++ b/arch/arm/plat-samsung/include/plat/gpio-cfg-helpers.h
@@ -244,7 +244,7 @@
* This helper function reads the state of the pull-{up,down} resistor for the
* given GPIO in the same case as s3c_gpio_setpull_upown.
*/
-extern s3c_gpio_pull_t s3c_gpio_getpull_s3c24xx(struct s3c_gpio_chip *chip,
+extern s3c_gpio_pull_t s3c_gpio_getpull_s3c2443(struct s3c_gpio_chip *chip,
unsigned int off);
#endif /* __PLAT_GPIO_CFG_HELPERS_H */
diff --git a/arch/arm/plat-samsung/include/plat/gpio-core.h b/arch/arm/plat-samsung/include/plat/gpio-core.h
index 13a22b8..dac35d0 100644
--- a/arch/arm/plat-samsung/include/plat/gpio-core.h
+++ b/arch/arm/plat-samsung/include/plat/gpio-core.h
@@ -118,6 +118,8 @@
int nr_chips);
extern void samsung_gpiolib_add_4bit2_chips(struct s3c_gpio_chip *chip,
int nr_chips);
+extern void samsung_gpiolib_add_2bit_chips(struct s3c_gpio_chip *chip,
+ int nr_chips);
extern void samsung_gpiolib_add_4bit(struct s3c_gpio_chip *chip);
extern void samsung_gpiolib_add_4bit2(struct s3c_gpio_chip *chip);
diff --git a/arch/arm/plat-samsung/include/plat/pd.h b/arch/arm/plat-samsung/include/plat/pd.h
new file mode 100644
index 0000000..5f0ad85
--- /dev/null
+++ b/arch/arm/plat-samsung/include/plat/pd.h
@@ -0,0 +1,30 @@
+/* linux/arch/arm/plat-samsung/include/plat/pd.h
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * 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 __ASM_PLAT_SAMSUNG_PD_H
+#define __ASM_PLAT_SAMSUNG_PD_H __FILE__
+
+struct samsung_pd_info {
+ int (*enable)(struct device *dev);
+ int (*disable)(struct device *dev);
+ void __iomem *base;
+};
+
+enum s5pv310_pd_block {
+ PD_MFC,
+ PD_G3D,
+ PD_LCD0,
+ PD_LCD1,
+ PD_TV,
+ PD_CAM,
+ PD_GPS
+};
+
+#endif /* __ASM_PLAT_SAMSUNG_PD_H */
diff --git a/arch/arm/plat-samsung/include/plat/pm.h b/arch/arm/plat-samsung/include/plat/pm.h
index 245836d..d9025e3 100644
--- a/arch/arm/plat-samsung/include/plat/pm.h
+++ b/arch/arm/plat-samsung/include/plat/pm.h
@@ -15,6 +15,8 @@
* management
*/
+#include <linux/irq.h>
+
#ifdef CONFIG_PM
extern __init int s3c_pm_init(void);
@@ -100,7 +102,7 @@
extern void s3c_pm_do_restore_core(struct sleep_save *ptr, int count);
#ifdef CONFIG_PM
-extern int s3c_irqext_wake(unsigned int irqno, unsigned int state);
+extern int s3c_irqext_wake(struct irq_data *data, unsigned int state);
extern int s3c24xx_irq_suspend(struct sys_device *dev, pm_message_t state);
extern int s3c24xx_irq_resume(struct sys_device *dev);
#else
diff --git a/arch/arm/plat-samsung/include/plat/sdhci.h b/arch/arm/plat-samsung/include/plat/sdhci.h
index 85853f8..5a41a0b 100644
--- a/arch/arm/plat-samsung/include/plat/sdhci.h
+++ b/arch/arm/plat-samsung/include/plat/sdhci.h
@@ -107,6 +107,8 @@
/* Helper function availablity */
+extern void s3c2416_setup_sdhci0_cfg_gpio(struct platform_device *, int w);
+extern void s3c2416_setup_sdhci1_cfg_gpio(struct platform_device *, int w);
extern void s3c64xx_setup_sdhci0_cfg_gpio(struct platform_device *, int w);
extern void s3c64xx_setup_sdhci1_cfg_gpio(struct platform_device *, int w);
extern void s5pc100_setup_sdhci0_cfg_gpio(struct platform_device *, int w);
@@ -122,6 +124,39 @@
extern void s5pv310_setup_sdhci2_cfg_gpio(struct platform_device *, int w);
extern void s5pv310_setup_sdhci3_cfg_gpio(struct platform_device *, int w);
+/* S3C2416 SDHCI setup */
+
+#ifdef CONFIG_S3C2416_SETUP_SDHCI
+extern char *s3c2416_hsmmc_clksrcs[4];
+
+extern void s3c2416_setup_sdhci_cfg_card(struct platform_device *dev,
+ void __iomem *r,
+ struct mmc_ios *ios,
+ struct mmc_card *card);
+
+static inline void s3c2416_default_sdhci0(void)
+{
+#ifdef CONFIG_S3C_DEV_HSMMC
+ s3c_hsmmc0_def_platdata.clocks = s3c2416_hsmmc_clksrcs;
+ s3c_hsmmc0_def_platdata.cfg_gpio = s3c2416_setup_sdhci0_cfg_gpio;
+ s3c_hsmmc0_def_platdata.cfg_card = s3c2416_setup_sdhci_cfg_card;
+#endif /* CONFIG_S3C_DEV_HSMMC */
+}
+
+static inline void s3c2416_default_sdhci1(void)
+{
+#ifdef CONFIG_S3C_DEV_HSMMC1
+ s3c_hsmmc1_def_platdata.clocks = s3c2416_hsmmc_clksrcs;
+ s3c_hsmmc1_def_platdata.cfg_gpio = s3c2416_setup_sdhci1_cfg_gpio;
+ s3c_hsmmc1_def_platdata.cfg_card = s3c2416_setup_sdhci_cfg_card;
+#endif /* CONFIG_S3C_DEV_HSMMC1 */
+}
+
+#else
+static inline void s3c2416_default_sdhci0(void) { }
+static inline void s3c2416_default_sdhci1(void) { }
+
+#endif /* CONFIG_S3C2416_SETUP_SDHCI */
/* S3C64XX SDHCI setup */
#ifdef CONFIG_S3C64XX_SETUP_SDHCI
diff --git a/arch/arm/plat-samsung/irq-uart.c b/arch/arm/plat-samsung/irq-uart.c
index 4f8c102..4e77035 100644
--- a/arch/arm/plat-samsung/irq-uart.c
+++ b/arch/arm/plat-samsung/irq-uart.c
@@ -28,9 +28,9 @@
* are consecutive when looking up the interrupt in the demux routines.
*/
-static inline void __iomem *s3c_irq_uart_base(unsigned int irq)
+static inline void __iomem *s3c_irq_uart_base(struct irq_data *data)
{
- struct s3c_uart_irq *uirq = get_irq_chip_data(irq);
+ struct s3c_uart_irq *uirq = irq_data_get_irq_chip_data(data);
return uirq->regs;
}
@@ -39,10 +39,10 @@
return irq & 3;
}
-static void s3c_irq_uart_mask(unsigned int irq)
+static void s3c_irq_uart_mask(struct irq_data *data)
{
- void __iomem *regs = s3c_irq_uart_base(irq);
- unsigned int bit = s3c_irq_uart_bit(irq);
+ void __iomem *regs = s3c_irq_uart_base(data);
+ unsigned int bit = s3c_irq_uart_bit(data->irq);
u32 reg;
reg = __raw_readl(regs + S3C64XX_UINTM);
@@ -50,10 +50,10 @@
__raw_writel(reg, regs + S3C64XX_UINTM);
}
-static void s3c_irq_uart_maskack(unsigned int irq)
+static void s3c_irq_uart_maskack(struct irq_data *data)
{
- void __iomem *regs = s3c_irq_uart_base(irq);
- unsigned int bit = s3c_irq_uart_bit(irq);
+ void __iomem *regs = s3c_irq_uart_base(data);
+ unsigned int bit = s3c_irq_uart_bit(data->irq);
u32 reg;
reg = __raw_readl(regs + S3C64XX_UINTM);
@@ -62,10 +62,10 @@
__raw_writel(1 << bit, regs + S3C64XX_UINTP);
}
-static void s3c_irq_uart_unmask(unsigned int irq)
+static void s3c_irq_uart_unmask(struct irq_data *data)
{
- void __iomem *regs = s3c_irq_uart_base(irq);
- unsigned int bit = s3c_irq_uart_bit(irq);
+ void __iomem *regs = s3c_irq_uart_base(data);
+ unsigned int bit = s3c_irq_uart_bit(data->irq);
u32 reg;
reg = __raw_readl(regs + S3C64XX_UINTM);
@@ -73,17 +73,17 @@
__raw_writel(reg, regs + S3C64XX_UINTM);
}
-static void s3c_irq_uart_ack(unsigned int irq)
+static void s3c_irq_uart_ack(struct irq_data *data)
{
- void __iomem *regs = s3c_irq_uart_base(irq);
- unsigned int bit = s3c_irq_uart_bit(irq);
+ void __iomem *regs = s3c_irq_uart_base(data);
+ unsigned int bit = s3c_irq_uart_bit(data->irq);
__raw_writel(1 << bit, regs + S3C64XX_UINTP);
}
static void s3c_irq_demux_uart(unsigned int irq, struct irq_desc *desc)
{
- struct s3c_uart_irq *uirq = desc->handler_data;
+ struct s3c_uart_irq *uirq = desc->irq_data.handler_data;
u32 pend = __raw_readl(uirq->regs + S3C64XX_UINTP);
int base = uirq->base_irq;
@@ -99,10 +99,10 @@
static struct irq_chip s3c_irq_uart = {
.name = "s3c-uart",
- .mask = s3c_irq_uart_mask,
- .unmask = s3c_irq_uart_unmask,
- .mask_ack = s3c_irq_uart_maskack,
- .ack = s3c_irq_uart_ack,
+ .irq_mask = s3c_irq_uart_mask,
+ .irq_unmask = s3c_irq_uart_unmask,
+ .irq_mask_ack = s3c_irq_uart_maskack,
+ .irq_ack = s3c_irq_uart_ack,
};
static void __init s3c_init_uart_irq(struct s3c_uart_irq *uirq)
@@ -124,7 +124,7 @@
set_irq_flags(irq, IRQF_VALID);
}
- desc->handler_data = uirq;
+ desc->irq_data.handler_data = uirq;
set_irq_chained_handler(uirq->parent_irq, s3c_irq_demux_uart);
}
diff --git a/arch/arm/plat-samsung/irq-vic-timer.c b/arch/arm/plat-samsung/irq-vic-timer.c
index 0270519..dd8692a 100644
--- a/arch/arm/plat-samsung/irq-vic-timer.c
+++ b/arch/arm/plat-samsung/irq-vic-timer.c
@@ -24,43 +24,46 @@
static void s3c_irq_demux_vic_timer(unsigned int irq, struct irq_desc *desc)
{
- generic_handle_irq((int)desc->handler_data);
+ generic_handle_irq((int)desc->irq_data.handler_data);
}
/* We assume the IRQ_TIMER0..IRQ_TIMER4 range is continuous. */
-static void s3c_irq_timer_mask(unsigned int irq)
+static void s3c_irq_timer_mask(struct irq_data *data)
{
u32 reg = __raw_readl(S3C64XX_TINT_CSTAT);
+ u32 mask = (u32)data->chip_data;
reg &= 0x1f; /* mask out pending interrupts */
- reg &= ~(1 << (irq - IRQ_TIMER0));
+ reg &= ~mask;
__raw_writel(reg, S3C64XX_TINT_CSTAT);
}
-static void s3c_irq_timer_unmask(unsigned int irq)
+static void s3c_irq_timer_unmask(struct irq_data *data)
{
u32 reg = __raw_readl(S3C64XX_TINT_CSTAT);
+ u32 mask = (u32)data->chip_data;
reg &= 0x1f; /* mask out pending interrupts */
- reg |= 1 << (irq - IRQ_TIMER0);
+ reg |= mask;
__raw_writel(reg, S3C64XX_TINT_CSTAT);
}
-static void s3c_irq_timer_ack(unsigned int irq)
+static void s3c_irq_timer_ack(struct irq_data *data)
{
u32 reg = __raw_readl(S3C64XX_TINT_CSTAT);
+ u32 mask = (u32)data->chip_data;
reg &= 0x1f;
- reg |= (1 << 5) << (irq - IRQ_TIMER0);
+ reg |= mask << 5;
__raw_writel(reg, S3C64XX_TINT_CSTAT);
}
static struct irq_chip s3c_irq_timer = {
.name = "s3c-timer",
- .mask = s3c_irq_timer_mask,
- .unmask = s3c_irq_timer_unmask,
- .ack = s3c_irq_timer_ack,
+ .irq_mask = s3c_irq_timer_mask,
+ .irq_unmask = s3c_irq_timer_unmask,
+ .irq_ack = s3c_irq_timer_ack,
};
/**
@@ -79,8 +82,9 @@
set_irq_chained_handler(parent_irq, s3c_irq_demux_vic_timer);
set_irq_chip(timer_irq, &s3c_irq_timer);
+ set_irq_chip_data(timer_irq, (void *)(1 << (timer_irq - IRQ_TIMER0)));
set_irq_handler(timer_irq, handle_level_irq);
set_irq_flags(timer_irq, IRQF_VALID);
- desc->handler_data = (void *)timer_irq;
+ desc->irq_data.handler_data = (void *)timer_irq;
}
diff --git a/arch/arm/plat-samsung/pd.c b/arch/arm/plat-samsung/pd.c
new file mode 100644
index 0000000..efe1d56
--- /dev/null
+++ b/arch/arm/plat-samsung/pd.c
@@ -0,0 +1,95 @@
+/* linux/arch/arm/plat-samsung/pd.c
+ *
+ * Copyright (c) 2010 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com
+ *
+ * Samsung Power domain support
+ *
+ * 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/init.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/err.h>
+#include <linux/pm_runtime.h>
+
+#include <plat/pd.h>
+
+static int samsung_pd_probe(struct platform_device *pdev)
+{
+ struct samsung_pd_info *pdata = pdev->dev.platform_data;
+ struct device *dev = &pdev->dev;
+
+ if (!pdata) {
+ dev_err(dev, "no device data specified\n");
+ return -ENOENT;
+ }
+
+ pm_runtime_set_active(dev);
+ pm_runtime_enable(dev);
+
+ dev_info(dev, "power domain registered\n");
+ return 0;
+}
+
+static int __devexit samsung_pd_remove(struct platform_device *pdev)
+{
+ struct device *dev = &pdev->dev;
+
+ pm_runtime_disable(dev);
+ return 0;
+}
+
+static int samsung_pd_runtime_suspend(struct device *dev)
+{
+ struct samsung_pd_info *pdata = dev->platform_data;
+ int ret = 0;
+
+ if (pdata->disable)
+ ret = pdata->disable(dev);
+
+ dev_dbg(dev, "suspended\n");
+ return ret;
+}
+
+static int samsung_pd_runtime_resume(struct device *dev)
+{
+ struct samsung_pd_info *pdata = dev->platform_data;
+ int ret = 0;
+
+ if (pdata->enable)
+ ret = pdata->enable(dev);
+
+ dev_dbg(dev, "resumed\n");
+ return ret;
+}
+
+static const struct dev_pm_ops samsung_pd_pm_ops = {
+ .runtime_suspend = samsung_pd_runtime_suspend,
+ .runtime_resume = samsung_pd_runtime_resume,
+};
+
+static struct platform_driver samsung_pd_driver = {
+ .driver = {
+ .name = "samsung-pd",
+ .owner = THIS_MODULE,
+ .pm = &samsung_pd_pm_ops,
+ },
+ .probe = samsung_pd_probe,
+ .remove = __devexit_p(samsung_pd_remove),
+};
+
+static int __init samsung_pd_init(void)
+{
+ int ret;
+
+ ret = platform_driver_register(&samsung_pd_driver);
+ if (ret)
+ printk(KERN_ERR "%s: failed to add PD driver\n", __func__);
+
+ return ret;
+}
+arch_initcall(samsung_pd_init);
diff --git a/arch/arm/plat-samsung/pm.c b/arch/arm/plat-samsung/pm.c
index 5bf3f2f..02d531f 100644
--- a/arch/arm/plat-samsung/pm.c
+++ b/arch/arm/plat-samsung/pm.c
@@ -136,15 +136,15 @@
unsigned long s3c_irqwake_intmask = 0xffffffffL;
unsigned long s3c_irqwake_eintmask = 0xffffffffL;
-int s3c_irqext_wake(unsigned int irqno, unsigned int state)
+int s3c_irqext_wake(struct irq_data *data, unsigned int state)
{
- unsigned long bit = 1L << IRQ_EINT_BIT(irqno);
+ unsigned long bit = 1L << IRQ_EINT_BIT(data->irq);
if (!(s3c_irqwake_eintallow & bit))
return -ENOENT;
printk(KERN_INFO "wake %s for irq %d\n",
- state ? "enabled" : "disabled", irqno);
+ state ? "enabled" : "disabled", data->irq);
if (!state)
s3c_irqwake_eintmask |= bit;
diff --git a/arch/arm/plat-spear/shirq.c b/arch/arm/plat-spear/shirq.c
index 2172d69..7818903 100644
--- a/arch/arm/plat-spear/shirq.c
+++ b/arch/arm/plat-spear/shirq.c
@@ -20,10 +20,10 @@
struct spear_shirq *shirq;
static DEFINE_SPINLOCK(lock);
-static void shirq_irq_mask(unsigned irq)
+static void shirq_irq_mask(struct irq_data *d)
{
- struct spear_shirq *shirq = get_irq_chip_data(irq);
- u32 val, id = irq - shirq->dev_config[0].virq;
+ struct spear_shirq *shirq = irq_data_get_irq_chip_data(d);
+ u32 val, id = d->irq - shirq->dev_config[0].virq;
unsigned long flags;
if ((shirq->regs.enb_reg == -1) || shirq->dev_config[id].enb_mask == -1)
@@ -39,10 +39,10 @@
spin_unlock_irqrestore(&lock, flags);
}
-static void shirq_irq_unmask(unsigned irq)
+static void shirq_irq_unmask(struct irq_data *d)
{
- struct spear_shirq *shirq = get_irq_chip_data(irq);
- u32 val, id = irq - shirq->dev_config[0].virq;
+ struct spear_shirq *shirq = irq_data_get_irq_chip_data(d);
+ u32 val, id = d->irq - shirq->dev_config[0].virq;
unsigned long flags;
if ((shirq->regs.enb_reg == -1) || shirq->dev_config[id].enb_mask == -1)
@@ -60,9 +60,9 @@
static struct irq_chip shirq_chip = {
.name = "spear_shirq",
- .ack = shirq_irq_mask,
- .mask = shirq_irq_mask,
- .unmask = shirq_irq_unmask,
+ .irq_ack = shirq_irq_mask,
+ .irq_mask = shirq_irq_mask,
+ .irq_unmask = shirq_irq_unmask,
};
static void shirq_handler(unsigned irq, struct irq_desc *desc)
@@ -70,7 +70,7 @@
u32 i, val, mask;
struct spear_shirq *shirq = get_irq_data(irq);
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
while ((val = readl(shirq->regs.base + shirq->regs.status_reg) &
shirq->regs.status_reg_mask)) {
for (i = 0; (i < shirq->dev_count) && val; i++) {
@@ -92,7 +92,7 @@
writel(mask, shirq->regs.base + shirq->regs.clear_reg);
}
}
- desc->chip->unmask(irq);
+ desc->irq_data.chip->irq_unmask(&desc->irq_data);
}
int spear_shirq_register(struct spear_shirq *shirq)
diff --git a/arch/arm/plat-stmp3xxx/irq.c b/arch/arm/plat-stmp3xxx/irq.c
index 20de4e0..aaa1686 100644
--- a/arch/arm/plat-stmp3xxx/irq.c
+++ b/arch/arm/plat-stmp3xxx/irq.c
@@ -34,7 +34,7 @@
/* Disable all interrupts initially */
for (i = 0; i < NR_REAL_IRQS; i++) {
- chip->mask(i);
+ chip->irq_mask(irq_get_irq_data(i));
set_irq_chip(i, chip);
set_irq_handler(i, handle_level_irq);
set_irq_flags(i, IRQF_VALID | IRQF_PROBE);
diff --git a/arch/arm/plat-stmp3xxx/pinmux.c b/arch/arm/plat-stmp3xxx/pinmux.c
index 6d6b1a4..66d5bac 100644
--- a/arch/arm/plat-stmp3xxx/pinmux.c
+++ b/arch/arm/plat-stmp3xxx/pinmux.c
@@ -351,27 +351,27 @@
}
EXPORT_SYMBOL(stmp3xxx_release_pin_group);
-static int stmp3xxx_irq_to_gpio(int irq,
+static int stmp3xxx_irq_data_to_gpio(struct irq_data *d,
struct stmp3xxx_pinmux_bank **bank, unsigned *gpio)
{
struct stmp3xxx_pinmux_bank *pm;
for (pm = pinmux_banks; pm < pinmux_banks + NR_BANKS; pm++)
- if (pm->virq <= irq && irq < pm->virq + 32) {
+ if (pm->virq <= d->irq && d->irq < pm->virq + 32) {
*bank = pm;
- *gpio = irq - pm->virq;
+ *gpio = d->irq - pm->virq;
return 0;
}
return -ENOENT;
}
-static int stmp3xxx_set_irqtype(unsigned irq, unsigned type)
+static int stmp3xxx_set_irqtype(struct irq_data *d, unsigned type)
{
struct stmp3xxx_pinmux_bank *pm;
unsigned gpio;
int l, p;
- stmp3xxx_irq_to_gpio(irq, &pm, &gpio);
+ stmp3xxx_irq_data_to_gpio(d, &pm, &gpio);
switch (type) {
case IRQ_TYPE_EDGE_RISING:
l = 0; p = 1; break;
@@ -398,33 +398,33 @@
return 0;
}
-static void stmp3xxx_pin_ack_irq(unsigned irq)
+static void stmp3xxx_pin_ack_irq(struct irq_data *d)
{
u32 stat;
struct stmp3xxx_pinmux_bank *pm;
unsigned gpio;
- stmp3xxx_irq_to_gpio(irq, &pm, &gpio);
+ stmp3xxx_irq_data_to_gpio(d, &pm, &gpio);
stat = __raw_readl(pm->irqstat) & (1 << gpio);
stmp3xxx_clearl(stat, pm->irqstat);
}
-static void stmp3xxx_pin_mask_irq(unsigned irq)
+static void stmp3xxx_pin_mask_irq(struct irq_data *d)
{
struct stmp3xxx_pinmux_bank *pm;
unsigned gpio;
- stmp3xxx_irq_to_gpio(irq, &pm, &gpio);
+ stmp3xxx_irq_data_to_gpio(d, &pm, &gpio);
stmp3xxx_clearl(1 << gpio, pm->irqen);
stmp3xxx_clearl(1 << gpio, pm->pin2irq);
}
-static void stmp3xxx_pin_unmask_irq(unsigned irq)
+static void stmp3xxx_pin_unmask_irq(struct irq_data *d)
{
struct stmp3xxx_pinmux_bank *pm;
unsigned gpio;
- stmp3xxx_irq_to_gpio(irq, &pm, &gpio);
+ stmp3xxx_irq_data_to_gpio(d, &pm, &gpio);
stmp3xxx_setl(1 << gpio, pm->irqen);
stmp3xxx_setl(1 << gpio, pm->pin2irq);
}
@@ -503,10 +503,10 @@
}
static struct irq_chip gpio_irq_chip = {
- .ack = stmp3xxx_pin_ack_irq,
- .mask = stmp3xxx_pin_mask_irq,
- .unmask = stmp3xxx_pin_unmask_irq,
- .set_type = stmp3xxx_set_irqtype,
+ .irq_ack = stmp3xxx_pin_ack_irq,
+ .irq_mask = stmp3xxx_pin_mask_irq,
+ .irq_unmask = stmp3xxx_pin_unmask_irq,
+ .irq_set_type = stmp3xxx_set_irqtype,
};
int __init stmp3xxx_pinmux_init(int virtual_irq_start)
@@ -533,7 +533,7 @@
pm->virq = virtual_irq_start + b * 32;
for (virq = pm->virq; virq < pm->virq; virq++) {
- gpio_irq_chip.mask(virq);
+ gpio_irq_chip.irq_mask(irq_get_irq_data(virq));
set_irq_chip(virq, &gpio_irq_chip);
set_irq_handler(virq, handle_level_irq);
set_irq_flags(virq, IRQF_VALID);
diff --git a/arch/avr32/boards/atngw100/setup.c b/arch/avr32/boards/atngw100/setup.c
index 8c6a244..659d119 100644
--- a/arch/avr32/boards/atngw100/setup.c
+++ b/arch/avr32/boards/atngw100/setup.c
@@ -188,7 +188,7 @@
*/
regs = (void __iomem __force *)res->start;
pclk = clk_get(&pdev->dev, "pclk");
- if (!pclk)
+ if (IS_ERR(pclk))
return;
clk_enable(pclk);
diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c
index 2adc261..6ce30fb 100644
--- a/arch/avr32/boards/atstk1000/atstk1002.c
+++ b/arch/avr32/boards/atstk1000/atstk1002.c
@@ -203,7 +203,7 @@
*/
regs = (void __iomem __force *)res->start;
pclk = clk_get(&pdev->dev, "pclk");
- if (!pclk)
+ if (IS_ERR(pclk))
return;
clk_enable(pclk);
diff --git a/arch/avr32/boards/favr-32/setup.c b/arch/avr32/boards/favr-32/setup.c
index 75f19f4..86fab77 100644
--- a/arch/avr32/boards/favr-32/setup.c
+++ b/arch/avr32/boards/favr-32/setup.c
@@ -206,7 +206,7 @@
*/
regs = (void __iomem __force *)res->start;
pclk = clk_get(&pdev->dev, "pclk");
- if (!pclk)
+ if (IS_ERR(pclk))
return;
clk_enable(pclk);
diff --git a/arch/avr32/boards/hammerhead/setup.c b/arch/avr32/boards/hammerhead/setup.c
index dd009875..da14fbd 100644
--- a/arch/avr32/boards/hammerhead/setup.c
+++ b/arch/avr32/boards/hammerhead/setup.c
@@ -150,7 +150,7 @@
regs = (void __iomem __force *)res->start;
pclk = clk_get(&pdev->dev, "pclk");
- if (!pclk)
+ if (IS_ERR(pclk))
return;
clk_enable(pclk);
diff --git a/arch/avr32/boards/merisc/setup.c b/arch/avr32/boards/merisc/setup.c
index 623b077..e61bc94 100644
--- a/arch/avr32/boards/merisc/setup.c
+++ b/arch/avr32/boards/merisc/setup.c
@@ -134,7 +134,7 @@
regs = (void __iomem __force *)res->start;
pclk = clk_get(&pdev->dev, "pclk");
- if (!pclk)
+ if (IS_ERR(pclk))
return;
clk_enable(pclk);
diff --git a/arch/avr32/boards/mimc200/setup.c b/arch/avr32/boards/mimc200/setup.c
index 523d8e1..c4da5cb 100644
--- a/arch/avr32/boards/mimc200/setup.c
+++ b/arch/avr32/boards/mimc200/setup.c
@@ -162,7 +162,7 @@
*/
regs = (void __iomem __force *)res->start;
pclk = clk_get(&pdev->dev, "pclk");
- if (!pclk)
+ if (IS_ERR(pclk))
return;
clk_enable(pclk);
diff --git a/arch/avr32/configs/atngw100_defconfig b/arch/avr32/configs/atngw100_defconfig
index 9854013..6f9ca56 100644
--- a/arch/avr32/configs/atngw100_defconfig
+++ b/arch/avr32/configs/atngw100_defconfig
@@ -2,20 +2,17 @@
# CONFIG_LOCALVERSION_AUTO is not set
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-CONFIG_BSD_PROCESS_ACCT_V3=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
# CONFIG_BASE_FULL is not set
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
-CONFIG_MODULE_FORCE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
# CONFIG_IOSCHED_DEADLINE is not set
CONFIG_NO_HZ=y
@@ -29,6 +26,7 @@
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -72,8 +70,8 @@
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_TCLIB=y
-CONFIG_EEPROM_AT24=m
CONFIG_NETDEVICES=y
CONFIG_TUN=m
CONFIG_NET_ETHERNET=y
@@ -106,6 +104,7 @@
CONFIG_WATCHDOG=y
CONFIG_AT32AP700X_WDT=y
CONFIG_USB_GADGET=y
+CONFIG_USB_GADGET_VBUS_DRAW=350
CONFIG_USB_ZERO=m
CONFIG_USB_ETH=m
CONFIG_USB_GADGETFS=m
@@ -115,14 +114,12 @@
CONFIG_MMC=y
CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
-CONFIG_MMC_SPI=m
CONFIG_NEW_LEDS=y
CONFIG_LEDS_CLASS=y
CONFIG_LEDS_GPIO=y
CONFIG_LEDS_TRIGGERS=y
CONFIG_LEDS_TRIGGER_TIMER=y
CONFIG_LEDS_TRIGGER_HEARTBEAT=y
-CONFIG_LEDS_TRIGGER_DEFAULT_ON=y
CONFIG_RTC_CLASS=y
CONFIG_RTC_DRV_AT32AP700X=y
CONFIG_DMADEVICES=y
@@ -130,21 +127,23 @@
CONFIG_EXT3_FS=y
# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
CONFIG_FAT_DEFAULT_CODEPAGE=850
+CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
-CONFIG_CONFIGFS_FS=m
+CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
-CONFIG_UFS_FS=y
+CONFIG_UBIFS_FS=y
CONFIG_NFS_FS=y
CONFIG_NFS_V3=y
CONFIG_ROOT_NFS=y
CONFIG_NFSD=m
CONFIG_NFSD_V3=y
-CONFIG_SMB_FS=m
CONFIG_CIFS=m
CONFIG_NLS_CODEPAGE_437=m
CONFIG_NLS_CODEPAGE_850=m
@@ -155,5 +154,3 @@
CONFIG_DEBUG_KERNEL=y
CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
-CONFIG_CRYPTO_PCBC=m
diff --git a/arch/avr32/configs/atngw100_evklcd100_defconfig b/arch/avr32/configs/atngw100_evklcd100_defconfig
index 7ceda35..7eece0a 100644
--- a/arch/avr32/configs/atngw100_evklcd100_defconfig
+++ b/arch/avr32/configs/atngw100_evklcd100_defconfig
@@ -2,20 +2,17 @@
# CONFIG_LOCALVERSION_AUTO is not set
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-CONFIG_BSD_PROCESS_ACCT_V3=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
# CONFIG_BASE_FULL is not set
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
-CONFIG_MODULE_FORCE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
# CONFIG_IOSCHED_DEADLINE is not set
CONFIG_NO_HZ=y
@@ -31,6 +28,7 @@
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -74,8 +72,10 @@
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_TCLIB=y
CONFIG_NETDEVICES=y
+CONFIG_TUN=m
CONFIG_NET_ETHERNET=y
CONFIG_MACB=y
# CONFIG_NETDEV_1000 is not set
@@ -104,6 +104,7 @@
CONFIG_SPI=y
CONFIG_SPI_ATMEL=y
CONFIG_SPI_SPIDEV=m
+CONFIG_GPIO_SYSFS=y
# CONFIG_HWMON is not set
CONFIG_WATCHDOG=y
CONFIG_AT32AP700X_WDT=y
@@ -127,6 +128,7 @@
CONFIG_USB_G_SERIAL=m
CONFIG_USB_CDC_COMPOSITE=m
CONFIG_MMC=y
+CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
CONFIG_NEW_LEDS=y
CONFIG_LEDS_CLASS=y
@@ -141,11 +143,14 @@
CONFIG_EXT3_FS=y
# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
CONFIG_FAT_DEFAULT_CODEPAGE=850
+CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
@@ -155,7 +160,6 @@
CONFIG_ROOT_NFS=y
CONFIG_NFSD=m
CONFIG_NFSD_V3=y
-CONFIG_SMB_FS=m
CONFIG_CIFS=m
CONFIG_NLS_CODEPAGE_437=m
CONFIG_NLS_CODEPAGE_850=m
@@ -166,4 +170,3 @@
CONFIG_DEBUG_KERNEL=y
CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
diff --git a/arch/avr32/configs/atngw100_evklcd101_defconfig b/arch/avr32/configs/atngw100_evklcd101_defconfig
index 7bc5b2c..387eb9d 100644
--- a/arch/avr32/configs/atngw100_evklcd101_defconfig
+++ b/arch/avr32/configs/atngw100_evklcd101_defconfig
@@ -2,20 +2,17 @@
# CONFIG_LOCALVERSION_AUTO is not set
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-CONFIG_BSD_PROCESS_ACCT_V3=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
# CONFIG_BASE_FULL is not set
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
-CONFIG_MODULE_FORCE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
# CONFIG_IOSCHED_DEADLINE is not set
CONFIG_NO_HZ=y
@@ -30,6 +27,7 @@
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -73,8 +71,10 @@
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_TCLIB=y
CONFIG_NETDEVICES=y
+CONFIG_TUN=m
CONFIG_NET_ETHERNET=y
CONFIG_MACB=y
# CONFIG_NETDEV_1000 is not set
@@ -103,6 +103,7 @@
CONFIG_SPI=y
CONFIG_SPI_ATMEL=y
CONFIG_SPI_SPIDEV=m
+CONFIG_GPIO_SYSFS=y
# CONFIG_HWMON is not set
CONFIG_WATCHDOG=y
CONFIG_AT32AP700X_WDT=y
@@ -126,6 +127,7 @@
CONFIG_USB_G_SERIAL=m
CONFIG_USB_CDC_COMPOSITE=m
CONFIG_MMC=y
+CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
CONFIG_NEW_LEDS=y
CONFIG_LEDS_CLASS=y
@@ -140,11 +142,14 @@
CONFIG_EXT3_FS=y
# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
CONFIG_FAT_DEFAULT_CODEPAGE=850
+CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
@@ -154,7 +159,6 @@
CONFIG_ROOT_NFS=y
CONFIG_NFSD=m
CONFIG_NFSD_V3=y
-CONFIG_SMB_FS=m
CONFIG_CIFS=m
CONFIG_NLS_CODEPAGE_437=m
CONFIG_NLS_CODEPAGE_850=m
@@ -165,4 +169,3 @@
CONFIG_DEBUG_KERNEL=y
CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
diff --git a/arch/avr32/configs/atngw100mkii_defconfig b/arch/avr32/configs/atngw100mkii_defconfig
index 4bd3682..f0fe237 100644
--- a/arch/avr32/configs/atngw100mkii_defconfig
+++ b/arch/avr32/configs/atngw100mkii_defconfig
@@ -2,20 +2,17 @@
# CONFIG_LOCALVERSION_AUTO is not set
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-CONFIG_BSD_PROCESS_ACCT_V3=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
# CONFIG_BASE_FULL is not set
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
-CONFIG_MODULE_FORCE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
# CONFIG_IOSCHED_DEADLINE is not set
CONFIG_NO_HZ=y
@@ -29,6 +26,7 @@
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -74,6 +72,7 @@
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_TCLIB=y
CONFIG_NETDEVICES=y
CONFIG_TUN=m
@@ -107,6 +106,7 @@
CONFIG_WATCHDOG=y
CONFIG_AT32AP700X_WDT=y
CONFIG_USB_GADGET=y
+CONFIG_USB_GADGET_VBUS_DRAW=350
CONFIG_USB_ZERO=m
CONFIG_USB_ETH=m
CONFIG_USB_GADGETFS=m
@@ -116,14 +116,12 @@
CONFIG_MMC=y
CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
-CONFIG_MMC_SPI=m
CONFIG_NEW_LEDS=y
CONFIG_LEDS_CLASS=y
CONFIG_LEDS_GPIO=y
CONFIG_LEDS_TRIGGERS=y
CONFIG_LEDS_TRIGGER_TIMER=y
CONFIG_LEDS_TRIGGER_HEARTBEAT=y
-CONFIG_LEDS_TRIGGER_DEFAULT_ON=y
CONFIG_RTC_CLASS=y
CONFIG_RTC_DRV_AT32AP700X=y
CONFIG_DMADEVICES=y
@@ -131,21 +129,23 @@
CONFIG_EXT3_FS=y
# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
CONFIG_FAT_DEFAULT_CODEPAGE=850
+CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
-CONFIG_CONFIGFS_FS=m
+CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
-CONFIG_UFS_FS=y
+CONFIG_UBIFS_FS=y
CONFIG_NFS_FS=y
CONFIG_NFS_V3=y
CONFIG_ROOT_NFS=y
CONFIG_NFSD=m
CONFIG_NFSD_V3=y
-CONFIG_SMB_FS=m
CONFIG_CIFS=m
CONFIG_NLS_CODEPAGE_437=m
CONFIG_NLS_CODEPAGE_850=m
@@ -156,5 +156,3 @@
CONFIG_DEBUG_KERNEL=y
CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
-CONFIG_CRYPTO_PCBC=m
diff --git a/arch/avr32/configs/atngw100mkii_evklcd100_defconfig b/arch/avr32/configs/atngw100mkii_evklcd100_defconfig
index f8437ef..e4a7c1d 100644
--- a/arch/avr32/configs/atngw100mkii_evklcd100_defconfig
+++ b/arch/avr32/configs/atngw100mkii_evklcd100_defconfig
@@ -2,20 +2,17 @@
# CONFIG_LOCALVERSION_AUTO is not set
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-CONFIG_BSD_PROCESS_ACCT_V3=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
# CONFIG_BASE_FULL is not set
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
-CONFIG_MODULE_FORCE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
# CONFIG_IOSCHED_DEADLINE is not set
CONFIG_NO_HZ=y
@@ -32,6 +29,7 @@
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -77,8 +75,10 @@
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_TCLIB=y
CONFIG_NETDEVICES=y
+CONFIG_TUN=m
CONFIG_NET_ETHERNET=y
CONFIG_MACB=y
# CONFIG_NETDEV_1000 is not set
@@ -107,6 +107,7 @@
CONFIG_SPI=y
CONFIG_SPI_ATMEL=y
CONFIG_SPI_SPIDEV=m
+CONFIG_GPIO_SYSFS=y
# CONFIG_HWMON is not set
CONFIG_WATCHDOG=y
CONFIG_AT32AP700X_WDT=y
@@ -130,6 +131,7 @@
CONFIG_USB_G_SERIAL=m
CONFIG_USB_CDC_COMPOSITE=m
CONFIG_MMC=y
+CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
CONFIG_NEW_LEDS=y
CONFIG_LEDS_CLASS=y
@@ -144,11 +146,14 @@
CONFIG_EXT3_FS=y
# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
CONFIG_FAT_DEFAULT_CODEPAGE=850
+CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
@@ -158,7 +163,6 @@
CONFIG_ROOT_NFS=y
CONFIG_NFSD=m
CONFIG_NFSD_V3=y
-CONFIG_SMB_FS=m
CONFIG_CIFS=m
CONFIG_NLS_CODEPAGE_437=m
CONFIG_NLS_CODEPAGE_850=m
@@ -169,4 +173,3 @@
CONFIG_DEBUG_KERNEL=y
CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
diff --git a/arch/avr32/configs/atngw100mkii_evklcd101_defconfig b/arch/avr32/configs/atngw100mkii_evklcd101_defconfig
index 7f58f99..6f37f70 100644
--- a/arch/avr32/configs/atngw100mkii_evklcd101_defconfig
+++ b/arch/avr32/configs/atngw100mkii_evklcd101_defconfig
@@ -2,20 +2,17 @@
# CONFIG_LOCALVERSION_AUTO is not set
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-CONFIG_BSD_PROCESS_ACCT_V3=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
# CONFIG_BASE_FULL is not set
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
-CONFIG_MODULE_FORCE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
# CONFIG_IOSCHED_DEADLINE is not set
CONFIG_NO_HZ=y
@@ -31,6 +28,7 @@
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -76,8 +74,10 @@
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_TCLIB=y
CONFIG_NETDEVICES=y
+CONFIG_TUN=m
CONFIG_NET_ETHERNET=y
CONFIG_MACB=y
# CONFIG_NETDEV_1000 is not set
@@ -106,6 +106,7 @@
CONFIG_SPI=y
CONFIG_SPI_ATMEL=y
CONFIG_SPI_SPIDEV=m
+CONFIG_GPIO_SYSFS=y
# CONFIG_HWMON is not set
CONFIG_WATCHDOG=y
CONFIG_AT32AP700X_WDT=y
@@ -129,6 +130,7 @@
CONFIG_USB_G_SERIAL=m
CONFIG_USB_CDC_COMPOSITE=m
CONFIG_MMC=y
+CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
CONFIG_NEW_LEDS=y
CONFIG_LEDS_CLASS=y
@@ -143,11 +145,14 @@
CONFIG_EXT3_FS=y
# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
CONFIG_FAT_DEFAULT_CODEPAGE=850
+CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
@@ -157,7 +162,6 @@
CONFIG_ROOT_NFS=y
CONFIG_NFSD=m
CONFIG_NFSD_V3=y
-CONFIG_SMB_FS=m
CONFIG_CIFS=m
CONFIG_NLS_CODEPAGE_437=m
CONFIG_NLS_CODEPAGE_850=m
@@ -168,4 +172,3 @@
CONFIG_DEBUG_KERNEL=y
CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
diff --git a/arch/avr32/configs/atstk1002_defconfig b/arch/avr32/configs/atstk1002_defconfig
index aec4c43..4fb01f5 100644
--- a/arch/avr32/configs/atstk1002_defconfig
+++ b/arch/avr32/configs/atstk1002_defconfig
@@ -3,7 +3,6 @@
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
@@ -11,7 +10,7 @@
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
@@ -26,6 +25,7 @@
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -35,6 +35,7 @@
CONFIG_IP_PNP=y
CONFIG_IP_PNP_DHCP=y
CONFIG_NET_IPIP=m
+CONFIG_NET_IPGRE_DEMUX=m
CONFIG_NET_IPGRE=m
CONFIG_INET_AH=m
CONFIG_INET_ESP=m
@@ -58,16 +59,14 @@
CONFIG_MTD_CFI=y
CONFIG_MTD_CFI_AMDSTD=y
CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_DATAFLASH=m
-CONFIG_MTD_M25P80=m
CONFIG_MTD_UBI=y
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_PWM=m
CONFIG_ATMEL_TCLIB=y
CONFIG_ATMEL_SSC=m
-CONFIG_EEPROM_AT24=m
# CONFIG_SCSI_PROC_FS is not set
CONFIG_BLK_DEV_SD=m
CONFIG_BLK_DEV_SR=m
@@ -120,7 +119,6 @@
CONFIG_SND_PCM_OSS=m
# CONFIG_SND_SUPPORT_OLD_API is not set
# CONFIG_SND_VERBOSE_PROCFS is not set
-# CONFIG_SND_DRIVERS is not set
CONFIG_SND_AT73C213=m
# CONFIG_HID_SUPPORT is not set
CONFIG_USB_GADGET=y
@@ -131,16 +129,15 @@
CONFIG_USB_G_SERIAL=m
CONFIG_USB_CDC_COMPOSITE=m
CONFIG_MMC=y
+CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
-CONFIG_MMC_SPI=m
CONFIG_NEW_LEDS=y
-CONFIG_LEDS_CLASS=m
+CONFIG_LEDS_CLASS=y
CONFIG_LEDS_ATMEL_PWM=m
CONFIG_LEDS_GPIO=m
CONFIG_LEDS_TRIGGERS=y
CONFIG_LEDS_TRIGGER_TIMER=m
CONFIG_LEDS_TRIGGER_HEARTBEAT=m
-CONFIG_LEDS_TRIGGER_DEFAULT_ON=m
CONFIG_RTC_CLASS=y
CONFIG_RTC_DRV_AT32AP700X=y
CONFIG_DMADEVICES=y
@@ -149,20 +146,23 @@
# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
# CONFIG_EXT3_FS_XATTR is not set
CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=850
CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
+CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
-# CONFIG_JFFS2_FS_WRITEBUFFER is not set
CONFIG_UBIFS_FS=y
-CONFIG_MINIX_FS=m
CONFIG_NFS_FS=y
CONFIG_NFS_V3=y
CONFIG_ROOT_NFS=y
+CONFIG_CIFS=m
CONFIG_NLS_CODEPAGE_437=m
+CONFIG_NLS_CODEPAGE_850=m
CONFIG_NLS_ISO8859_1=m
CONFIG_NLS_UTF8=m
CONFIG_MAGIC_SYSRQ=y
@@ -170,6 +170,3 @@
CONFIG_DEBUG_KERNEL=y
CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
-# CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_T10DIF=m
diff --git a/arch/avr32/configs/atstk1003_defconfig b/arch/avr32/configs/atstk1003_defconfig
index 50ba3db..9faaf9b 100644
--- a/arch/avr32/configs/atstk1003_defconfig
+++ b/arch/avr32/configs/atstk1003_defconfig
@@ -2,22 +2,15 @@
# CONFIG_LOCALVERSION_AUTO is not set
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
-CONFIG_BSD_PROCESS_ACCT=y
-CONFIG_BSD_PROCESS_ACCT_V3=y
-CONFIG_TASKSTATS=y
-CONFIG_TASK_DELAY_ACCT=y
-CONFIG_AUDIT=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
# CONFIG_BASE_FULL is not set
-# CONFIG_SLUB_DEBUG is not set
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
@@ -33,6 +26,7 @@
CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
CONFIG_CPU_FREQ_GOV_USERSPACE=y
CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -54,18 +48,18 @@
CONFIG_MTD_CFI=y
CONFIG_MTD_CFI_AMDSTD=y
CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_DATAFLASH=m
-CONFIG_MTD_M25P80=m
+CONFIG_MTD_UBI=y
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_PWM=m
CONFIG_ATMEL_TCLIB=y
CONFIG_ATMEL_SSC=m
-CONFIG_EEPROM_AT24=m
# CONFIG_SCSI_PROC_FS is not set
CONFIG_BLK_DEV_SD=m
CONFIG_BLK_DEV_SR=m
+# CONFIG_SCSI_LOWLEVEL is not set
CONFIG_ATA=m
# CONFIG_SATA_PMP is not set
CONFIG_PATA_AT32=m
@@ -77,6 +71,7 @@
CONFIG_PPP_DEFLATE=m
CONFIG_PPP_BSDCOMP=m
CONFIG_INPUT=m
+CONFIG_INPUT_EVDEV=m
# CONFIG_KEYBOARD_ATKBD is not set
CONFIG_KEYBOARD_GPIO=m
# CONFIG_MOUSE_PS2 is not set
@@ -106,7 +101,6 @@
CONFIG_SND_AT73C213=m
# CONFIG_HID_SUPPORT is not set
CONFIG_USB_GADGET=y
-CONFIG_USB_GADGET_DEBUG_FS=y
CONFIG_USB_ZERO=m
CONFIG_USB_ETH=m
CONFIG_USB_GADGETFS=m
@@ -116,36 +110,39 @@
CONFIG_MMC=y
CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
-CONFIG_MMC_SPI=m
CONFIG_NEW_LEDS=y
CONFIG_LEDS_CLASS=y
CONFIG_LEDS_ATMEL_PWM=m
-CONFIG_LEDS_GPIO=y
+CONFIG_LEDS_GPIO=m
CONFIG_LEDS_TRIGGERS=y
-CONFIG_LEDS_TRIGGER_TIMER=y
-CONFIG_LEDS_TRIGGER_HEARTBEAT=y
-CONFIG_LEDS_TRIGGER_DEFAULT_ON=y
+CONFIG_LEDS_TRIGGER_TIMER=m
+CONFIG_LEDS_TRIGGER_HEARTBEAT=m
CONFIG_RTC_CLASS=y
CONFIG_RTC_DRV_AT32AP700X=y
CONFIG_DMADEVICES=y
-CONFIG_DW_DMAC=y
-CONFIG_EXT2_FS=m
-CONFIG_EXT3_FS=m
+CONFIG_EXT2_FS=y
+CONFIG_EXT3_FS=y
+# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=850
CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
-CONFIG_CONFIGFS_FS=m
+CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
+CONFIG_UBIFS_FS=y
# CONFIG_NETWORK_FILESYSTEMS is not set
CONFIG_NLS_CODEPAGE_437=m
+CONFIG_NLS_CODEPAGE_850=m
CONFIG_NLS_ISO8859_1=m
CONFIG_NLS_UTF8=m
CONFIG_MAGIC_SYSRQ=y
CONFIG_DEBUG_FS=y
CONFIG_DEBUG_KERNEL=y
+CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-CONFIG_CRC_T10DIF=m
diff --git a/arch/avr32/configs/atstk1004_defconfig b/arch/avr32/configs/atstk1004_defconfig
index 329e10b..3d2a5d8 100644
--- a/arch/avr32/configs/atstk1004_defconfig
+++ b/arch/avr32/configs/atstk1004_defconfig
@@ -1,19 +1,32 @@
CONFIG_EXPERIMENTAL=y
# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
+CONFIG_RELAY=y
+CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
# CONFIG_BASE_FULL is not set
-# CONFIG_FUTEX is not set
-# CONFIG_EPOLL is not set
-# CONFIG_SIGNALFD is not set
-# CONFIG_TIMERFD is not set
-# CONFIG_EVENTFD is not set
# CONFIG_COMPAT_BRK is not set
-CONFIG_SLOB=y
-# CONFIG_BLOCK is not set
+CONFIG_PROFILING=y
+CONFIG_OPROFILE=m
+# CONFIG_KPROBES is not set
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_BLK_DEV_BSG is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
CONFIG_BOARD_ATSTK1004=y
# CONFIG_OWNERSHIP_TRACE is not set
+CONFIG_NMI_DEBUGGING=y
+CONFIG_PM=y
+CONFIG_CPU_FREQ=y
+# CONFIG_CPU_FREQ_STAT is not set
+CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
+CONFIG_CPU_FREQ_GOV_USERSPACE=y
+CONFIG_CPU_FREQ_AT32AP=y
+CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -31,40 +44,104 @@
CONFIG_MTD_PARTITIONS=y
CONFIG_MTD_CMDLINE_PARTS=y
CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLOCK=y
CONFIG_MTD_CFI=y
CONFIG_MTD_CFI_AMDSTD=y
CONFIG_MTD_PHYSMAP=y
-# CONFIG_MISC_DEVICES is not set
-# CONFIG_INPUT is not set
+CONFIG_MTD_UBI=y
+CONFIG_BLK_DEV_LOOP=m
+CONFIG_BLK_DEV_NBD=m
+CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
+CONFIG_ATMEL_PWM=m
+CONFIG_ATMEL_TCLIB=y
+CONFIG_ATMEL_SSC=m
+# CONFIG_SCSI_PROC_FS is not set
+CONFIG_BLK_DEV_SD=m
+CONFIG_BLK_DEV_SR=m
+# CONFIG_SCSI_LOWLEVEL is not set
+CONFIG_ATA=m
+# CONFIG_SATA_PMP is not set
+CONFIG_PATA_AT32=m
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+CONFIG_PPP=m
+CONFIG_PPP_ASYNC=m
+CONFIG_PPP_DEFLATE=m
+CONFIG_PPP_BSDCOMP=m
+CONFIG_INPUT=m
+CONFIG_INPUT_EVDEV=m
+# CONFIG_KEYBOARD_ATKBD is not set
+CONFIG_KEYBOARD_GPIO=m
+# CONFIG_MOUSE_PS2 is not set
+CONFIG_MOUSE_GPIO=m
# CONFIG_SERIO is not set
# CONFIG_VT is not set
# CONFIG_DEVKMEM is not set
CONFIG_SERIAL_ATMEL=y
CONFIG_SERIAL_ATMEL_CONSOLE=y
-# CONFIG_SERIAL_ATMEL_PDC is not set
# CONFIG_LEGACY_PTYS is not set
# CONFIG_HW_RANDOM is not set
+CONFIG_I2C=m
+CONFIG_I2C_CHARDEV=m
+CONFIG_I2C_GPIO=m
CONFIG_SPI=y
CONFIG_SPI_ATMEL=y
+CONFIG_SPI_SPIDEV=m
+CONFIG_GPIO_SYSFS=y
# CONFIG_HWMON is not set
CONFIG_WATCHDOG=y
CONFIG_AT32AP700X_WDT=y
CONFIG_FB=y
CONFIG_FB_ATMEL=y
CONFIG_BACKLIGHT_LCD_SUPPORT=y
+CONFIG_LCD_CLASS_DEVICE=y
CONFIG_LCD_LTV350QV=y
# CONFIG_BACKLIGHT_CLASS_DEVICE is not set
CONFIG_USB_GADGET=y
-CONFIG_USB_ETH=y
-# CONFIG_USB_ETH_RNDIS is not set
+CONFIG_USB_ZERO=m
+CONFIG_USB_ETH=m
+CONFIG_USB_GADGETFS=m
+CONFIG_USB_FILE_STORAGE=m
+CONFIG_USB_G_SERIAL=m
+CONFIG_USB_CDC_COMPOSITE=m
+CONFIG_MMC=y
+CONFIG_MMC_TEST=m
+CONFIG_MMC_ATMELMCI=y
+CONFIG_NEW_LEDS=y
+CONFIG_LEDS_CLASS=y
+CONFIG_LEDS_ATMEL_PWM=m
+CONFIG_LEDS_GPIO=m
+CONFIG_LEDS_TRIGGERS=y
+CONFIG_LEDS_TRIGGER_TIMER=m
+CONFIG_LEDS_TRIGGER_HEARTBEAT=m
CONFIG_RTC_CLASS=y
-# CONFIG_RTC_INTF_PROC is not set
CONFIG_RTC_DRV_AT32AP700X=y
+CONFIG_DMADEVICES=y
+CONFIG_EXT2_FS=y
+CONFIG_EXT3_FS=y
+# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
+# CONFIG_EXT3_FS_XATTR is not set
+CONFIG_EXT4_FS=y
+# CONFIG_EXT4_FS_XATTR is not set
# CONFIG_DNOTIFY is not set
+CONFIG_FUSE_FS=m
+CONFIG_MSDOS_FS=m
+CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=850
CONFIG_PROC_KCORE=y
-# CONFIG_PROC_PAGE_MONITOR is not set
CONFIG_TMPFS=y
+CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
-# CONFIG_JFFS2_FS_WRITEBUFFER is not set
+CONFIG_UBIFS_FS=y
# CONFIG_NETWORK_FILESYSTEMS is not set
+CONFIG_NLS_CODEPAGE_437=m
+CONFIG_NLS_CODEPAGE_850=m
+CONFIG_NLS_ISO8859_1=m
+CONFIG_NLS_UTF8=m
CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_FS=y
+CONFIG_DEBUG_KERNEL=y
+CONFIG_DETECT_HUNG_TASK=y
+CONFIG_FRAME_POINTER=y
diff --git a/arch/avr32/configs/atstk1006_defconfig b/arch/avr32/configs/atstk1006_defconfig
index dbcc1b5..1ed8f22 100644
--- a/arch/avr32/configs/atstk1006_defconfig
+++ b/arch/avr32/configs/atstk1006_defconfig
@@ -3,7 +3,6 @@
CONFIG_SYSVIPC=y
CONFIG_POSIX_MQUEUE=y
CONFIG_LOG_BUF_SHIFT=14
-CONFIG_SYSFS_DEPRECATED_V2=y
CONFIG_RELAY=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_SYSCTL_SYSCALL is not set
@@ -11,7 +10,7 @@
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
@@ -37,6 +36,7 @@
CONFIG_IP_PNP=y
CONFIG_IP_PNP_DHCP=y
CONFIG_NET_IPIP=m
+CONFIG_NET_IPGRE_DEMUX=m
CONFIG_NET_IPGRE=m
CONFIG_INET_AH=m
CONFIG_INET_ESP=m
@@ -60,15 +60,13 @@
CONFIG_MTD_CFI=y
CONFIG_MTD_CFI_AMDSTD=y
CONFIG_MTD_PHYSMAP=y
-CONFIG_MTD_DATAFLASH=m
-CONFIG_MTD_DATAFLASH_OTP=y
-CONFIG_MTD_M25P80=m
CONFIG_MTD_NAND=y
CONFIG_MTD_NAND_ATMEL=y
CONFIG_MTD_UBI=y
CONFIG_BLK_DEV_LOOP=m
CONFIG_BLK_DEV_NBD=m
CONFIG_BLK_DEV_RAM=m
+CONFIG_MISC_DEVICES=y
CONFIG_ATMEL_PWM=m
CONFIG_ATMEL_TCLIB=y
CONFIG_ATMEL_SSC=m
@@ -132,17 +130,17 @@
CONFIG_USB_GADGETFS=m
CONFIG_USB_FILE_STORAGE=m
CONFIG_USB_G_SERIAL=m
+CONFIG_USB_CDC_COMPOSITE=m
CONFIG_MMC=y
+CONFIG_MMC_TEST=m
CONFIG_MMC_ATMELMCI=y
-CONFIG_MMC_SPI=m
CONFIG_NEW_LEDS=y
-CONFIG_LEDS_CLASS=m
+CONFIG_LEDS_CLASS=y
CONFIG_LEDS_ATMEL_PWM=m
CONFIG_LEDS_GPIO=m
CONFIG_LEDS_TRIGGERS=y
CONFIG_LEDS_TRIGGER_TIMER=m
CONFIG_LEDS_TRIGGER_HEARTBEAT=m
-CONFIG_LEDS_TRIGGER_DEFAULT_ON=m
CONFIG_RTC_CLASS=y
CONFIG_RTC_DRV_AT32AP700X=y
CONFIG_DMADEVICES=y
@@ -156,15 +154,18 @@
CONFIG_FUSE_FS=m
CONFIG_MSDOS_FS=m
CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=850
CONFIG_PROC_KCORE=y
CONFIG_TMPFS=y
+CONFIG_CONFIGFS_FS=y
CONFIG_JFFS2_FS=y
CONFIG_UBIFS_FS=y
-CONFIG_MINIX_FS=m
CONFIG_NFS_FS=y
CONFIG_NFS_V3=y
CONFIG_ROOT_NFS=y
+CONFIG_CIFS=m
CONFIG_NLS_CODEPAGE_437=m
+CONFIG_NLS_CODEPAGE_850=m
CONFIG_NLS_ISO8859_1=m
CONFIG_NLS_UTF8=m
CONFIG_MAGIC_SYSRQ=y
@@ -172,7 +173,3 @@
CONFIG_DEBUG_KERNEL=y
CONFIG_DETECT_HUNG_TASK=y
CONFIG_FRAME_POINTER=y
-# CONFIG_RCU_CPU_STALL_DETECTOR is not set
-CONFIG_CRYPTO_FIPS=y
-# CONFIG_CRYPTO_HW is not set
-CONFIG_CRC_T10DIF=m
diff --git a/arch/avr32/configs/favr-32_defconfig b/arch/avr32/configs/favr-32_defconfig
index 0c813b6..aeadc95 100644
--- a/arch/avr32/configs/favr-32_defconfig
+++ b/arch/avr32/configs/favr-32_defconfig
@@ -11,7 +11,7 @@
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
diff --git a/arch/avr32/configs/hammerhead_defconfig b/arch/avr32/configs/hammerhead_defconfig
index dcc01f0..1692bee 100644
--- a/arch/avr32/configs/hammerhead_defconfig
+++ b/arch/avr32/configs/hammerhead_defconfig
@@ -12,7 +12,7 @@
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
CONFIG_OPROFILE=m
-CONFIG_KPROBES=y
+# CONFIG_KPROBES is not set
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
CONFIG_MODULE_FORCE_UNLOAD=y
diff --git a/arch/avr32/include/asm/syscalls.h b/arch/avr32/include/asm/syscalls.h
index ab608b7..244f2ac 100644
--- a/arch/avr32/include/asm/syscalls.h
+++ b/arch/avr32/include/asm/syscalls.h
@@ -15,20 +15,6 @@
#include <linux/types.h>
#include <linux/signal.h>
-/* kernel/process.c */
-asmlinkage int sys_fork(struct pt_regs *);
-asmlinkage int sys_clone(unsigned long, unsigned long,
- unsigned long, unsigned long,
- struct pt_regs *);
-asmlinkage int sys_vfork(struct pt_regs *);
-asmlinkage int sys_execve(const char __user *, char __user *__user *,
- char __user *__user *, struct pt_regs *);
-
-/* kernel/signal.c */
-asmlinkage int sys_sigaltstack(const stack_t __user *, stack_t __user *,
- struct pt_regs *);
-asmlinkage int sys_rt_sigreturn(struct pt_regs *);
-
/* mm/cache.c */
asmlinkage int sys_cacheflush(int, void __user *, size_t);
diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c
index 9c46aaa..ef5a2a0 100644
--- a/arch/avr32/kernel/process.c
+++ b/arch/avr32/kernel/process.c
@@ -367,14 +367,13 @@
}
asmlinkage int sys_clone(unsigned long clone_flags, unsigned long newsp,
- unsigned long parent_tidptr,
- unsigned long child_tidptr, struct pt_regs *regs)
+ void __user *parent_tidptr, void __user *child_tidptr,
+ struct pt_regs *regs)
{
if (!newsp)
newsp = regs->sp;
- return do_fork(clone_flags, newsp, regs, 0,
- (int __user *)parent_tidptr,
- (int __user *)child_tidptr);
+ return do_fork(clone_flags, newsp, regs, 0, parent_tidptr,
+ child_tidptr);
}
asmlinkage int sys_vfork(struct pt_regs *regs)
diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c
index 668ed28..05ad291 100644
--- a/arch/avr32/kernel/time.c
+++ b/arch/avr32/kernel/time.c
@@ -35,7 +35,6 @@
.rating = 50,
.read = read_cycle_count,
.mask = CLOCKSOURCE_MASK(32),
- .shift = 16,
.flags = CLOCK_SOURCE_IS_CONTINUOUS,
};
@@ -123,9 +122,7 @@
/* figure rate for counter */
counter_hz = clk_get_rate(boot_cpu_data.clk);
- counter.mult = clocksource_hz2mult(counter_hz, counter.shift);
-
- ret = clocksource_register(&counter);
+ ret = clocksource_register_hz(&counter, counter_hz);
if (ret)
pr_debug("timer: could not register clocksource: %d\n", ret);
diff --git a/arch/ia64/include/asm/io.h b/arch/ia64/include/asm/io.h
index cc8335e..e5a6c35 100644
--- a/arch/ia64/include/asm/io.h
+++ b/arch/ia64/include/asm/io.h
@@ -426,6 +426,11 @@
extern void iounmap (volatile void __iomem *addr);
extern void __iomem * early_ioremap (unsigned long phys_addr, unsigned long size);
extern void early_iounmap (volatile void __iomem *addr, unsigned long size);
+static inline void __iomem * ioremap_cache (unsigned long phys_addr, unsigned long size)
+{
+ return ioremap(phys_addr, size);
+}
+
/*
* String version of IO memory access ops:
diff --git a/arch/ia64/include/asm/page.h b/arch/ia64/include/asm/page.h
index 41b6d31..961a16f 100644
--- a/arch/ia64/include/asm/page.h
+++ b/arch/ia64/include/asm/page.h
@@ -189,6 +189,7 @@
# define pgprot_val(x) ((x).pgprot)
# define __pte(x) ((pte_t) { (x) } )
+# define __pmd(x) ((pmd_t) { (x) } )
# define __pgprot(x) ((pgprot_t) { (x) } )
#else /* !STRICT_MM_TYPECHECKS */
diff --git a/arch/ia64/include/asm/processor.h b/arch/ia64/include/asm/processor.h
index 348e44d..03afe79 100644
--- a/arch/ia64/include/asm/processor.h
+++ b/arch/ia64/include/asm/processor.h
@@ -717,8 +717,9 @@
#define spin_lock_prefetch(x) prefetchw(x)
extern unsigned long boot_option_idle_override;
-extern unsigned long idle_halt;
-extern unsigned long idle_nomwait;
+
+enum idle_boot_override {IDLE_NO_OVERRIDE=0, IDLE_HALT, IDLE_FORCE_MWAIT,
+ IDLE_NOMWAIT, IDLE_POLL};
#endif /* !__ASSEMBLY__ */
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index ac76da0..89accc6 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -618,7 +618,7 @@
}
/* forward declaration */
-static static const struct dentry_operations pfmfs_dentry_operations;
+static const struct dentry_operations pfmfs_dentry_operations;
static struct dentry *
pfmfs_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data)
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index 16f1c7b..6d33c5c 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -53,12 +53,8 @@
void (*ia64_mark_idle)(int);
-unsigned long boot_option_idle_override = 0;
+unsigned long boot_option_idle_override = IDLE_NO_OVERRIDE;
EXPORT_SYMBOL(boot_option_idle_override);
-unsigned long idle_halt;
-EXPORT_SYMBOL(idle_halt);
-unsigned long idle_nomwait;
-EXPORT_SYMBOL(idle_nomwait);
void (*pm_idle) (void);
EXPORT_SYMBOL(pm_idle);
void (*pm_power_off) (void);
diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
index 1841ee7..5ca674b 100644
--- a/arch/ia64/mm/hugetlbpage.c
+++ b/arch/ia64/mm/hugetlbpage.c
@@ -38,7 +38,7 @@
if (pud) {
pmd = pmd_alloc(mm, pud, taddr);
if (pmd)
- pte = pte_alloc_map(mm, pmd, taddr);
+ pte = pte_alloc_map(mm, NULL, pmd, taddr);
}
return pte;
}
diff --git a/arch/mips/include/asm/mman.h b/arch/mips/include/asm/mman.h
index c892bfb..785b4ea 100644
--- a/arch/mips/include/asm/mman.h
+++ b/arch/mips/include/asm/mman.h
@@ -77,6 +77,9 @@
#define MADV_UNMERGEABLE 13 /* KSM may not merge identical pages */
#define MADV_HWPOISON 100 /* poison a page for testing */
+#define MADV_HUGEPAGE 14 /* Worth backing with hugepages */
+#define MADV_NOHUGEPAGE 15 /* Not worth backing with hugepages */
+
/* compatibility flags */
#define MAP_FILE 0
diff --git a/arch/mips/kernel/module.c b/arch/mips/kernel/module.c
index 6f51dda..d87a72e 100644
--- a/arch/mips/kernel/module.c
+++ b/arch/mips/kernel/module.c
@@ -46,17 +46,9 @@
void *module_alloc(unsigned long size)
{
#ifdef MODULE_START
- struct vm_struct *area;
-
- size = PAGE_ALIGN(size);
- if (!size)
- return NULL;
-
- area = __get_vm_area(size, VM_ALLOC, MODULE_START, MODULE_END);
- if (!area)
- return NULL;
-
- return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL);
+ return __vmalloc_node_range(size, 1, MODULE_START, MODULE_END,
+ GFP_KERNEL, PAGE_KERNEL, -1,
+ __builtin_return_address(0));
#else
if (size == 0)
return NULL;
diff --git a/arch/parisc/include/asm/mman.h b/arch/parisc/include/asm/mman.h
index 9749c8a..f5b7bf5 100644
--- a/arch/parisc/include/asm/mman.h
+++ b/arch/parisc/include/asm/mman.h
@@ -59,6 +59,9 @@
#define MADV_MERGEABLE 65 /* KSM may merge identical pages */
#define MADV_UNMERGEABLE 66 /* KSM may not merge identical pages */
+#define MADV_HUGEPAGE 67 /* Worth backing with hugepages */
+#define MADV_NOHUGEPAGE 68 /* Not worth backing with hugepages */
+
/* compatibility flags */
#define MAP_FILE 0
#define MAP_VARIABLE 0
diff --git a/arch/powerpc/mm/gup.c b/arch/powerpc/mm/gup.c
index d7efdbf..fec1320 100644
--- a/arch/powerpc/mm/gup.c
+++ b/arch/powerpc/mm/gup.c
@@ -16,6 +16,16 @@
#ifdef __HAVE_ARCH_PTE_SPECIAL
+static inline void get_huge_page_tail(struct page *page)
+{
+ /*
+ * __split_huge_page_refcount() cannot run
+ * from under us.
+ */
+ VM_BUG_ON(atomic_read(&page->_count) < 0);
+ atomic_inc(&page->_count);
+}
+
/*
* The performance critical leaf functions are made noinline otherwise gcc
* inlines everything into a single function which results in too much
@@ -47,6 +57,8 @@
put_page(page);
return 0;
}
+ if (PageTail(page))
+ get_huge_page_tail(page);
pages[*nr] = page;
(*nr)++;
diff --git a/arch/s390/defconfig b/arch/s390/defconfig
index d796971..29c82c6 100644
--- a/arch/s390/defconfig
+++ b/arch/s390/defconfig
@@ -5,10 +5,21 @@
CONFIG_RCU_TRACE=y
CONFIG_IKCONFIG=y
CONFIG_IKCONFIG_PROC=y
+CONFIG_CGROUPS=y
+CONFIG_CPUSETS=y
+CONFIG_CGROUP_CPUACCT=y
+CONFIG_RESOURCE_COUNTERS=y
+CONFIG_CGROUP_MEM_RES_CTLR=y
+CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y
+CONFIG_CGROUP_SCHED=y
+CONFIG_RT_GROUP_SCHED=y
+CONFIG_BLK_CGROUP=y
CONFIG_BLK_DEV_INITRD=y
# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
-CONFIG_PERF_EVENTS=y
+# CONFIG_COMPAT_BRK is not set
CONFIG_SLAB=y
+CONFIG_PROFILING=y
+CONFIG_OPROFILE=y
CONFIG_KPROBES=y
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
@@ -19,7 +30,9 @@
CONFIG_PREEMPT=y
CONFIG_MEMORY_HOTPLUG=y
CONFIG_MEMORY_HOTREMOVE=y
+CONFIG_KSM=y
CONFIG_BINFMT_MISC=m
+CONFIG_CMM=m
CONFIG_HZ_100=y
CONFIG_KEXEC=y
CONFIG_PM=y
@@ -105,6 +118,7 @@
CONFIG_DEBUG_NOTIFIERS=y
# CONFIG_RCU_CPU_STALL_DETECTOR is not set
CONFIG_KPROBES_SANITY_TEST=y
+CONFIG_DEBUG_FORCE_WEAK_PER_CPU=y
CONFIG_CPU_NOTIFIER_ERROR_INJECT=m
CONFIG_LATENCYTOP=y
CONFIG_SYSCTL_SYSCALL_CHECK=y
diff --git a/arch/s390/include/asm/compat.h b/arch/s390/include/asm/compat.h
index a875c2f..da359ca 100644
--- a/arch/s390/include/asm/compat.h
+++ b/arch/s390/include/asm/compat.h
@@ -169,7 +169,7 @@
static inline int is_compat_task(void)
{
- return test_thread_flag(TIF_31BIT);
+ return is_32bit_task();
}
#else
diff --git a/arch/s390/include/asm/elf.h b/arch/s390/include/asm/elf.h
index 354d426..10c029c 100644
--- a/arch/s390/include/asm/elf.h
+++ b/arch/s390/include/asm/elf.h
@@ -161,7 +161,9 @@
use of this is to invoke "./ld.so someprog" to test out a new version of
the loader. We need to make sure that it is out of the way of the program
that it will "exec", and that there is sufficient room for the brk. */
-#define ELF_ET_DYN_BASE (STACK_TOP / 3 * 2)
+
+extern unsigned long randomize_et_dyn(unsigned long base);
+#define ELF_ET_DYN_BASE (randomize_et_dyn(STACK_TOP / 3 * 2))
/* This yields a mask that user programs can use to figure out what
instruction set this CPU supports. */
@@ -206,6 +208,8 @@
current->mm->context.noexec == 0; \
})
+#define STACK_RND_MASK 0x7ffUL
+
#define ARCH_DLINFO \
do { \
if (vdso_enabled) \
@@ -218,4 +222,7 @@
#define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1
int arch_setup_additional_pages(struct linux_binprm *, int);
+extern unsigned long arch_randomize_brk(struct mm_struct *mm);
+#define arch_randomize_brk arch_randomize_brk
+
#endif
diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h
index 6710b0e..8f8d759 100644
--- a/arch/s390/include/asm/system.h
+++ b/arch/s390/include/asm/system.h
@@ -449,7 +449,7 @@
extern void (*_machine_halt)(void);
extern void (*_machine_power_off)(void);
-#define arch_align_stack(x) (x)
+extern unsigned long arch_align_stack(unsigned long sp);
static inline int tprot(unsigned long addr)
{
diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h
index ebc7709..ad1382f 100644
--- a/arch/s390/include/asm/thread_info.h
+++ b/arch/s390/include/asm/thread_info.h
@@ -118,6 +118,12 @@
#define _TIF_SINGLE_STEP (1<<TIF_FREEZE)
#define _TIF_FREEZE (1<<TIF_FREEZE)
+#ifdef CONFIG_64BIT
+#define is_32bit_task() (test_thread_flag(TIF_31BIT))
+#else
+#define is_32bit_task() (1)
+#endif
+
#endif /* __KERNEL__ */
#define PREEMPT_ACTIVE 0x4000000
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 6ba4222..a895e69 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -30,9 +30,11 @@
#include <linux/tick.h>
#include <linux/elfcore.h>
#include <linux/kernel_stat.h>
+#include <linux/personality.h>
#include <linux/syscalls.h>
#include <linux/compat.h>
#include <linux/kprobes.h>
+#include <linux/random.h>
#include <asm/compat.h>
#include <asm/uaccess.h>
#include <asm/pgtable.h>
@@ -332,3 +334,39 @@
}
return 0;
}
+
+unsigned long arch_align_stack(unsigned long sp)
+{
+ if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+ sp -= get_random_int() & ~PAGE_MASK;
+ return sp & ~0xf;
+}
+
+static inline unsigned long brk_rnd(void)
+{
+ /* 8MB for 32bit, 1GB for 64bit */
+ if (is_32bit_task())
+ return (get_random_int() & 0x7ffUL) << PAGE_SHIFT;
+ else
+ return (get_random_int() & 0x3ffffUL) << PAGE_SHIFT;
+}
+
+unsigned long arch_randomize_brk(struct mm_struct *mm)
+{
+ unsigned long ret = PAGE_ALIGN(mm->brk + brk_rnd());
+
+ if (ret < mm->brk)
+ return mm->brk;
+ return ret;
+}
+
+unsigned long randomize_et_dyn(unsigned long base)
+{
+ unsigned long ret = PAGE_ALIGN(base + brk_rnd());
+
+ if (!(current->flags & PF_RANDOMIZE))
+ return base;
+ if (ret < base)
+ return base;
+ return ret;
+}
diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c
index e3150dd..f438d74 100644
--- a/arch/s390/kernel/vdso.c
+++ b/arch/s390/kernel/vdso.c
@@ -203,7 +203,6 @@
if (!uses_interp)
return 0;
- vdso_base = mm->mmap_base;
#ifdef CONFIG_64BIT
vdso_pagelist = vdso64_pagelist;
vdso_pages = vdso64_pages;
@@ -233,8 +232,7 @@
* fail and end up putting it elsewhere.
*/
down_write(&mm->mmap_sem);
- vdso_base = get_unmapped_area(NULL, vdso_base,
- vdso_pages << PAGE_SHIFT, 0, 0);
+ vdso_base = get_unmapped_area(NULL, 0, vdso_pages << PAGE_SHIFT, 0, 0);
if (IS_ERR_VALUE(vdso_base)) {
rc = vdso_base;
goto out_up;
diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c
index 869efba..c9a9f7f 100644
--- a/arch/s390/mm/mmap.c
+++ b/arch/s390/mm/mmap.c
@@ -27,17 +27,44 @@
#include <linux/personality.h>
#include <linux/mm.h>
#include <linux/module.h>
+#include <linux/random.h>
#include <asm/pgalloc.h>
#include <asm/compat.h>
+static unsigned long stack_maxrandom_size(void)
+{
+ if (!(current->flags & PF_RANDOMIZE))
+ return 0;
+ if (current->personality & ADDR_NO_RANDOMIZE)
+ return 0;
+ return STACK_RND_MASK << PAGE_SHIFT;
+}
+
/*
* Top of mmap area (just below the process stack).
*
- * Leave an at least ~128 MB hole.
+ * Leave at least a ~32 MB hole.
*/
-#define MIN_GAP (128*1024*1024)
+#define MIN_GAP (32*1024*1024)
#define MAX_GAP (STACK_TOP/6*5)
+static inline int mmap_is_legacy(void)
+{
+ if (current->personality & ADDR_COMPAT_LAYOUT)
+ return 1;
+ if (rlimit(RLIMIT_STACK) == RLIM_INFINITY)
+ return 1;
+ return sysctl_legacy_va_layout;
+}
+
+static unsigned long mmap_rnd(void)
+{
+ if (!(current->flags & PF_RANDOMIZE))
+ return 0;
+ /* 8MB randomization for mmap_base */
+ return (get_random_int() & 0x7ffUL) << PAGE_SHIFT;
+}
+
static inline unsigned long mmap_base(void)
{
unsigned long gap = rlimit(RLIMIT_STACK);
@@ -46,22 +73,8 @@
gap = MIN_GAP;
else if (gap > MAX_GAP)
gap = MAX_GAP;
-
- return STACK_TOP - (gap & PAGE_MASK);
-}
-
-static inline int mmap_is_legacy(void)
-{
-#ifdef CONFIG_64BIT
- /*
- * Force standard allocation for 64 bit programs.
- */
- if (!is_compat_task())
- return 1;
-#endif
- return sysctl_legacy_va_layout ||
- (current->personality & ADDR_COMPAT_LAYOUT) ||
- rlimit(RLIMIT_STACK) == RLIM_INFINITY;
+ gap &= PAGE_MASK;
+ return STACK_TOP - stack_maxrandom_size() - mmap_rnd() - gap;
}
#ifndef CONFIG_64BIT
diff --git a/arch/sh/kernel/cpu/shmobile/cpuidle.c b/arch/sh/kernel/cpu/shmobile/cpuidle.c
index 83972aa..c19e2a9 100644
--- a/arch/sh/kernel/cpu/shmobile/cpuidle.c
+++ b/arch/sh/kernel/cpu/shmobile/cpuidle.c
@@ -81,7 +81,6 @@
state->target_residency = 1 * 2;
state->power_usage = 3;
state->flags = 0;
- state->flags |= CPUIDLE_FLAG_SHALLOW;
state->flags |= CPUIDLE_FLAG_TIME_VALID;
state->enter = cpuidle_sleep_enter;
diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c
index 9163db3..d776234 100644
--- a/arch/sh/mm/hugetlbpage.c
+++ b/arch/sh/mm/hugetlbpage.c
@@ -35,7 +35,7 @@
if (pud) {
pmd = pmd_alloc(mm, pud, addr);
if (pmd)
- pte = pte_alloc_map(mm, pmd, addr);
+ pte = pte_alloc_map(mm, NULL, pmd, addr);
}
}
diff --git a/arch/sparc/kernel/module.c b/arch/sparc/kernel/module.c
index ee3c7dd..8d348c4 100644
--- a/arch/sparc/kernel/module.c
+++ b/arch/sparc/kernel/module.c
@@ -23,17 +23,11 @@
static void *module_map(unsigned long size)
{
- struct vm_struct *area;
-
- size = PAGE_ALIGN(size);
- if (!size || size > MODULES_LEN)
+ if (PAGE_ALIGN(size) > MODULES_LEN)
return NULL;
-
- area = __get_vm_area(size, VM_ALLOC, MODULES_VADDR, MODULES_END);
- if (!area)
- return NULL;
-
- return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL);
+ return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END,
+ GFP_KERNEL, PAGE_KERNEL, -1,
+ __builtin_return_address(0));
}
static char *dot2underscore(char *name)
diff --git a/arch/sparc/mm/generic_32.c b/arch/sparc/mm/generic_32.c
index 5edcac1..e6067b7 100644
--- a/arch/sparc/mm/generic_32.c
+++ b/arch/sparc/mm/generic_32.c
@@ -50,7 +50,7 @@
end = PGDIR_SIZE;
offset -= address;
do {
- pte_t * pte = pte_alloc_map(mm, pmd, address);
+ 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);
diff --git a/arch/sparc/mm/generic_64.c b/arch/sparc/mm/generic_64.c
index 04f2bf4..3cb00df 100644
--- a/arch/sparc/mm/generic_64.c
+++ b/arch/sparc/mm/generic_64.c
@@ -92,7 +92,7 @@
end = PGDIR_SIZE;
offset -= address;
do {
- pte_t * pte = pte_alloc_map(mm, pmd, address);
+ 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);
diff --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c
index 5fdddf1..f4e9764 100644
--- a/arch/sparc/mm/hugetlbpage.c
+++ b/arch/sparc/mm/hugetlbpage.c
@@ -214,7 +214,7 @@
if (pud) {
pmd = pmd_alloc(mm, pud, addr);
if (pmd)
- pte = pte_alloc_map(mm, pmd, addr);
+ pte = pte_alloc_map(mm, NULL, pmd, addr);
}
return pte;
}
diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c
index 3d099f9..1aee587 100644
--- a/arch/um/kernel/skas/mmu.c
+++ b/arch/um/kernel/skas/mmu.c
@@ -31,7 +31,7 @@
if (!pmd)
goto out_pmd;
- pte = pte_alloc_map(mm, pmd, proc);
+ pte = pte_alloc_map(mm, NULL, pmd, proc);
if (!pte)
goto out_pte;
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 36ed2e2..3ed5ad9 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1934,13 +1934,19 @@
depends on X86_64 && PCI && ACPI
config PCI_CNB20LE_QUIRK
- bool "Read CNB20LE Host Bridge Windows"
- depends on PCI
+ bool "Read CNB20LE Host Bridge Windows" if EMBEDDED
+ default n
+ depends on PCI && EXPERIMENTAL
help
Read the PCI windows out of the CNB20LE host bridge. This allows
PCI hotplug to work on systems with the CNB20LE chipset which do
not have ACPI.
+ There's no public spec for this chipset, and this functionality
+ is known to be incomplete.
+
+ You should say N unless you know you need this.
+
config DMAR
bool "Support for DMA Remapping Devices (EXPERIMENTAL)"
depends on PCI_MSI && ACPI && EXPERIMENTAL
@@ -2062,13 +2068,14 @@
bool "One Laptop Per Child support"
select GPIOLIB
select OLPC_OPENFIRMWARE
+ depends on !X86_64 && !X86_PAE
---help---
Add support for detecting the unique features of the OLPC
XO hardware.
config OLPC_XO1
tristate "OLPC XO-1 support"
- depends on OLPC && PCI
+ depends on OLPC && MFD_CS5535
---help---
Add support for non-essential features of the OLPC XO-1 laptop.
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index aa75f21..ffd7f8d 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -822,6 +822,7 @@
#define KVM_ARCH_WANT_MMU_NOTIFIER
int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
int kvm_age_hva(struct kvm *kvm, unsigned long hva);
+int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);
int cpuid_maxphyaddr(struct kvm_vcpu *vcpu);
int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index 7709c12..2071a8b 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -435,6 +435,11 @@
{
PVOP_VCALL3(pv_mmu_ops.pte_update, mm, addr, ptep);
}
+static inline void pmd_update(struct mm_struct *mm, unsigned long addr,
+ pmd_t *pmdp)
+{
+ PVOP_VCALL3(pv_mmu_ops.pmd_update, mm, addr, pmdp);
+}
static inline void pte_update_defer(struct mm_struct *mm, unsigned long addr,
pte_t *ptep)
@@ -442,6 +447,12 @@
PVOP_VCALL3(pv_mmu_ops.pte_update_defer, mm, addr, ptep);
}
+static inline void pmd_update_defer(struct mm_struct *mm, unsigned long addr,
+ pmd_t *pmdp)
+{
+ PVOP_VCALL3(pv_mmu_ops.pmd_update_defer, mm, addr, pmdp);
+}
+
static inline pte_t __pte(pteval_t val)
{
pteval_t ret;
@@ -543,6 +554,20 @@
PVOP_VCALL4(pv_mmu_ops.set_pte_at, mm, addr, ptep, pte.pte);
}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr,
+ pmd_t *pmdp, pmd_t pmd)
+{
+#if PAGETABLE_LEVELS >= 3
+ if (sizeof(pmdval_t) > sizeof(long))
+ /* 5 arg words */
+ pv_mmu_ops.set_pmd_at(mm, addr, pmdp, pmd);
+ else
+ PVOP_VCALL4(pv_mmu_ops.set_pmd_at, mm, addr, pmdp, pmd.pmd);
+#endif
+}
+#endif
+
static inline void set_pmd(pmd_t *pmdp, pmd_t pmd)
{
pmdval_t val = native_pmd_val(pmd);
diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h
index b82bac9..8288509 100644
--- a/arch/x86/include/asm/paravirt_types.h
+++ b/arch/x86/include/asm/paravirt_types.h
@@ -265,10 +265,16 @@
void (*set_pte_at)(struct mm_struct *mm, unsigned long addr,
pte_t *ptep, pte_t pteval);
void (*set_pmd)(pmd_t *pmdp, pmd_t pmdval);
+ void (*set_pmd_at)(struct mm_struct *mm, unsigned long addr,
+ pmd_t *pmdp, pmd_t pmdval);
void (*pte_update)(struct mm_struct *mm, unsigned long addr,
pte_t *ptep);
void (*pte_update_defer)(struct mm_struct *mm,
unsigned long addr, pte_t *ptep);
+ void (*pmd_update)(struct mm_struct *mm, unsigned long addr,
+ pmd_t *pmdp);
+ void (*pmd_update_defer)(struct mm_struct *mm,
+ unsigned long addr, pmd_t *pmdp);
pte_t (*ptep_modify_prot_start)(struct mm_struct *mm, unsigned long addr,
pte_t *ptep);
diff --git a/arch/x86/include/asm/pgtable-2level.h b/arch/x86/include/asm/pgtable-2level.h
index 2334982..98391db 100644
--- a/arch/x86/include/asm/pgtable-2level.h
+++ b/arch/x86/include/asm/pgtable-2level.h
@@ -46,6 +46,15 @@
#define native_ptep_get_and_clear(xp) native_local_ptep_get_and_clear(xp)
#endif
+#ifdef CONFIG_SMP
+static inline pmd_t native_pmdp_get_and_clear(pmd_t *xp)
+{
+ return __pmd(xchg((pmdval_t *)xp, 0));
+}
+#else
+#define native_pmdp_get_and_clear(xp) native_local_pmdp_get_and_clear(xp)
+#endif
+
/*
* Bits _PAGE_BIT_PRESENT, _PAGE_BIT_FILE and _PAGE_BIT_PROTNONE are taken,
* split up the 29 bits of offset into this range:
diff --git a/arch/x86/include/asm/pgtable-3level.h b/arch/x86/include/asm/pgtable-3level.h
index 177b016..94b979d 100644
--- a/arch/x86/include/asm/pgtable-3level.h
+++ b/arch/x86/include/asm/pgtable-3level.h
@@ -104,6 +104,29 @@
#define native_ptep_get_and_clear(xp) native_local_ptep_get_and_clear(xp)
#endif
+#ifdef CONFIG_SMP
+union split_pmd {
+ struct {
+ u32 pmd_low;
+ u32 pmd_high;
+ };
+ pmd_t pmd;
+};
+static inline pmd_t native_pmdp_get_and_clear(pmd_t *pmdp)
+{
+ union split_pmd res, *orig = (union split_pmd *)pmdp;
+
+ /* xchg acts as a barrier before setting of the high bits */
+ res.pmd_low = xchg(&orig->pmd_low, 0);
+ res.pmd_high = orig->pmd_high;
+ orig->pmd_high = 0;
+
+ return res.pmd;
+}
+#else
+#define native_pmdp_get_and_clear(xp) native_local_pmdp_get_and_clear(xp)
+#endif
+
/*
* Bits 0, 6 and 7 are taken in the low part of the pte,
* put the 32 bits of offset into the high part.
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index ada823a..18601c8 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -35,6 +35,7 @@
#else /* !CONFIG_PARAVIRT */
#define set_pte(ptep, pte) native_set_pte(ptep, pte)
#define set_pte_at(mm, addr, ptep, pte) native_set_pte_at(mm, addr, ptep, pte)
+#define set_pmd_at(mm, addr, pmdp, pmd) native_set_pmd_at(mm, addr, pmdp, pmd)
#define set_pte_atomic(ptep, pte) \
native_set_pte_atomic(ptep, pte)
@@ -59,6 +60,8 @@
#define pte_update(mm, addr, ptep) do { } while (0)
#define pte_update_defer(mm, addr, ptep) do { } while (0)
+#define pmd_update(mm, addr, ptep) do { } while (0)
+#define pmd_update_defer(mm, addr, ptep) do { } while (0)
#define pgd_val(x) native_pgd_val(x)
#define __pgd(x) native_make_pgd(x)
@@ -94,6 +97,11 @@
return pte_flags(pte) & _PAGE_ACCESSED;
}
+static inline int pmd_young(pmd_t pmd)
+{
+ return pmd_flags(pmd) & _PAGE_ACCESSED;
+}
+
static inline int pte_write(pte_t pte)
{
return pte_flags(pte) & _PAGE_RW;
@@ -142,6 +150,23 @@
(_PAGE_PSE | _PAGE_PRESENT);
}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline int pmd_trans_splitting(pmd_t pmd)
+{
+ return pmd_val(pmd) & _PAGE_SPLITTING;
+}
+
+static inline int pmd_trans_huge(pmd_t pmd)
+{
+ return pmd_val(pmd) & _PAGE_PSE;
+}
+
+static inline int has_transparent_hugepage(void)
+{
+ return cpu_has_pse;
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+
static inline pte_t pte_set_flags(pte_t pte, pteval_t set)
{
pteval_t v = native_pte_val(pte);
@@ -216,6 +241,55 @@
return pte_set_flags(pte, _PAGE_SPECIAL);
}
+static inline pmd_t pmd_set_flags(pmd_t pmd, pmdval_t set)
+{
+ pmdval_t v = native_pmd_val(pmd);
+
+ return __pmd(v | set);
+}
+
+static inline pmd_t pmd_clear_flags(pmd_t pmd, pmdval_t clear)
+{
+ pmdval_t v = native_pmd_val(pmd);
+
+ return __pmd(v & ~clear);
+}
+
+static inline pmd_t pmd_mkold(pmd_t pmd)
+{
+ return pmd_clear_flags(pmd, _PAGE_ACCESSED);
+}
+
+static inline pmd_t pmd_wrprotect(pmd_t pmd)
+{
+ return pmd_clear_flags(pmd, _PAGE_RW);
+}
+
+static inline pmd_t pmd_mkdirty(pmd_t pmd)
+{
+ return pmd_set_flags(pmd, _PAGE_DIRTY);
+}
+
+static inline pmd_t pmd_mkhuge(pmd_t pmd)
+{
+ return pmd_set_flags(pmd, _PAGE_PSE);
+}
+
+static inline pmd_t pmd_mkyoung(pmd_t pmd)
+{
+ return pmd_set_flags(pmd, _PAGE_ACCESSED);
+}
+
+static inline pmd_t pmd_mkwrite(pmd_t pmd)
+{
+ return pmd_set_flags(pmd, _PAGE_RW);
+}
+
+static inline pmd_t pmd_mknotpresent(pmd_t pmd)
+{
+ return pmd_clear_flags(pmd, _PAGE_PRESENT);
+}
+
/*
* Mask out unsupported bits in a present pgprot. Non-present pgprots
* can use those bits for other purposes, so leave them be.
@@ -256,6 +330,16 @@
return __pte(val);
}
+static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)
+{
+ pmdval_t val = pmd_val(pmd);
+
+ val &= _HPAGE_CHG_MASK;
+ val |= massage_pgprot(newprot) & ~_HPAGE_CHG_MASK;
+
+ return __pmd(val);
+}
+
/* mprotect needs to preserve PAT bits when updating vm_page_prot */
#define pgprot_modify pgprot_modify
static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot)
@@ -350,7 +434,7 @@
* Currently stuck as a macro due to indirect forward reference to
* linux/mmzone.h's __section_mem_map_addr() definition:
*/
-#define pmd_page(pmd) pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT)
+#define pmd_page(pmd) pfn_to_page((pmd_val(pmd) & PTE_PFN_MASK) >> PAGE_SHIFT)
/*
* the pmd page can be thought of an array like this: pmd_t[PTRS_PER_PMD]
@@ -524,12 +608,26 @@
return res;
}
+static inline pmd_t native_local_pmdp_get_and_clear(pmd_t *pmdp)
+{
+ pmd_t res = *pmdp;
+
+ native_pmd_clear(pmdp);
+ return res;
+}
+
static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr,
pte_t *ptep , pte_t pte)
{
native_set_pte(ptep, pte);
}
+static inline void native_set_pmd_at(struct mm_struct *mm, unsigned long addr,
+ pmd_t *pmdp , pmd_t pmd)
+{
+ native_set_pmd(pmdp, pmd);
+}
+
#ifndef CONFIG_PARAVIRT
/*
* Rules for using pte_update - it must be called after any PTE update which
@@ -607,6 +705,49 @@
#define flush_tlb_fix_spurious_fault(vma, address)
+#define mk_pmd(page, pgprot) pfn_pmd(page_to_pfn(page), (pgprot))
+
+#define __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS
+extern int pmdp_set_access_flags(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp,
+ pmd_t entry, int dirty);
+
+#define __HAVE_ARCH_PMDP_TEST_AND_CLEAR_YOUNG
+extern int pmdp_test_and_clear_young(struct vm_area_struct *vma,
+ unsigned long addr, pmd_t *pmdp);
+
+#define __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH
+extern int pmdp_clear_flush_young(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp);
+
+
+#define __HAVE_ARCH_PMDP_SPLITTING_FLUSH
+extern void pmdp_splitting_flush(struct vm_area_struct *vma,
+ unsigned long addr, pmd_t *pmdp);
+
+#define __HAVE_ARCH_PMD_WRITE
+static inline int pmd_write(pmd_t pmd)
+{
+ return pmd_flags(pmd) & _PAGE_RW;
+}
+
+#define __HAVE_ARCH_PMDP_GET_AND_CLEAR
+static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm, unsigned long addr,
+ pmd_t *pmdp)
+{
+ pmd_t pmd = native_pmdp_get_and_clear(pmdp);
+ pmd_update(mm, addr, pmdp);
+ return pmd;
+}
+
+#define __HAVE_ARCH_PMDP_SET_WRPROTECT
+static inline void pmdp_set_wrprotect(struct mm_struct *mm,
+ unsigned long addr, pmd_t *pmdp)
+{
+ clear_bit(_PAGE_BIT_RW, (unsigned long *)pmdp);
+ pmd_update(mm, addr, pmdp);
+}
+
/*
* clone_pgd_range(pgd_t *dst, pgd_t *src, int count);
*
diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
index f86da20..975f709 100644
--- a/arch/x86/include/asm/pgtable_64.h
+++ b/arch/x86/include/asm/pgtable_64.h
@@ -59,6 +59,16 @@
native_set_pte(ptep, pte);
}
+static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
+{
+ *pmdp = pmd;
+}
+
+static inline void native_pmd_clear(pmd_t *pmd)
+{
+ native_set_pmd(pmd, native_make_pmd(0));
+}
+
static inline pte_t native_ptep_get_and_clear(pte_t *xp)
{
#ifdef CONFIG_SMP
@@ -72,14 +82,17 @@
#endif
}
-static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd)
+static inline pmd_t native_pmdp_get_and_clear(pmd_t *xp)
{
- *pmdp = pmd;
-}
-
-static inline void native_pmd_clear(pmd_t *pmd)
-{
- native_set_pmd(pmd, native_make_pmd(0));
+#ifdef CONFIG_SMP
+ return native_make_pmd(xchg(&xp->pmd, 0));
+#else
+ /* native_local_pmdp_get_and_clear,
+ but duplicated because of cyclic dependency */
+ pmd_t ret = *xp;
+ native_pmd_clear(xp);
+ return ret;
+#endif
}
static inline void native_set_pud(pud_t *pudp, pud_t pud)
@@ -168,6 +181,7 @@
#define kc_offset_to_vaddr(o) ((o) | ~__VIRTUAL_MASK)
#define __HAVE_ARCH_PTE_SAME
+
#endif /* !__ASSEMBLY__ */
#endif /* _ASM_X86_PGTABLE_64_H */
diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h
index d1f4a76..7db7723 100644
--- a/arch/x86/include/asm/pgtable_types.h
+++ b/arch/x86/include/asm/pgtable_types.h
@@ -22,6 +22,7 @@
#define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */
#define _PAGE_BIT_SPECIAL _PAGE_BIT_UNUSED1
#define _PAGE_BIT_CPA_TEST _PAGE_BIT_UNUSED1
+#define _PAGE_BIT_SPLITTING _PAGE_BIT_UNUSED1 /* only valid on a PSE pmd */
#define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */
/* If _PAGE_BIT_PRESENT is clear, we use these: */
@@ -45,6 +46,7 @@
#define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE)
#define _PAGE_SPECIAL (_AT(pteval_t, 1) << _PAGE_BIT_SPECIAL)
#define _PAGE_CPA_TEST (_AT(pteval_t, 1) << _PAGE_BIT_CPA_TEST)
+#define _PAGE_SPLITTING (_AT(pteval_t, 1) << _PAGE_BIT_SPLITTING)
#define __HAVE_ARCH_PTE_SPECIAL
#ifdef CONFIG_KMEMCHECK
@@ -70,6 +72,7 @@
/* Set of bits not changed in pte_modify */
#define _PAGE_CHG_MASK (PTE_PFN_MASK | _PAGE_PCD | _PAGE_PWT | \
_PAGE_SPECIAL | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE)
#define _PAGE_CACHE_MASK (_PAGE_PCD | _PAGE_PWT)
#define _PAGE_CACHE_WB (0)
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index 53fd1d5..45636ce 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -761,10 +761,11 @@
extern void init_c1e_mask(void);
extern unsigned long boot_option_idle_override;
-extern unsigned long idle_halt;
-extern unsigned long idle_nomwait;
extern bool c1e_detected;
+enum idle_boot_override {IDLE_NO_OVERRIDE=0, IDLE_HALT, IDLE_NOMWAIT,
+ IDLE_POLL, IDLE_FORCE_MWAIT};
+
extern void enable_sep_cpu(void);
extern int sysenter_setup(void);
diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h
index 8760cc6..f25bdf2 100644
--- a/arch/x86/include/asm/xen/page.h
+++ b/arch/x86/include/asm/xen/page.h
@@ -42,6 +42,11 @@
extern unsigned long get_phys_to_machine(unsigned long pfn);
extern bool set_phys_to_machine(unsigned long pfn, unsigned long mfn);
+extern int m2p_add_override(unsigned long mfn, struct page *page);
+extern int m2p_remove_override(struct page *page);
+extern struct page *m2p_find_override(unsigned long mfn);
+extern unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn);
+
static inline unsigned long pfn_to_mfn(unsigned long pfn)
{
unsigned long mfn;
@@ -72,9 +77,6 @@
if (xen_feature(XENFEAT_auto_translated_physmap))
return mfn;
- if (unlikely((mfn >> machine_to_phys_order) != 0))
- return ~0;
-
pfn = 0;
/*
* The array access can fail (e.g., device space beyond end of RAM).
@@ -83,6 +85,14 @@
*/
__get_user(pfn, &machine_to_phys_mapping[mfn]);
+ /*
+ * If this appears to be a foreign mfn (because the pfn
+ * doesn't map back to the mfn), then check the local override
+ * table to see if there's a better pfn to use.
+ */
+ if (get_phys_to_machine(pfn) != mfn)
+ pfn = m2p_find_override_pfn(mfn, pfn);
+
return pfn;
}
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index ec881c6..b3a7113 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -509,6 +509,7 @@
return 0;
}
+EXPORT_SYMBOL_GPL(acpi_gsi_to_irq);
int acpi_isa_irq_to_gsi(unsigned isa_irq, u32 *gsi)
{
diff --git a/arch/x86/kernel/apb_timer.c b/arch/x86/kernel/apb_timer.c
index 7c9ab59..51ef31a 100644
--- a/arch/x86/kernel/apb_timer.c
+++ b/arch/x86/kernel/apb_timer.c
@@ -313,14 +313,16 @@
if (adev->irq == 0)
return;
+ irq_modify_status(adev->irq, 0, IRQ_MOVE_PCNTXT);
+ irq_set_affinity(adev->irq, cpumask_of(adev->cpu));
+ /* APB timer irqs are set up as mp_irqs, timer is edge type */
+ __set_irq_handler(adev->irq, handle_edge_irq, 0, "edge");
+
if (system_state == SYSTEM_BOOTING) {
- irq_modify_status(adev->irq, 0, IRQ_MOVE_PCNTXT);
- irq_set_affinity(adev->irq, cpumask_of(adev->cpu));
- /* APB timer irqs are set up as mp_irqs, timer is edge type */
- __set_irq_handler(adev->irq, handle_edge_irq, 0, "edge");
if (request_irq(adev->irq, apbt_interrupt_handler,
- IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING,
- adev->name, adev)) {
+ IRQF_TIMER | IRQF_DISABLED |
+ IRQF_NOBALANCING,
+ adev->name, adev)) {
printk(KERN_ERR "Failed request IRQ for APBT%d\n",
adev->num);
}
diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
index d6fb146..df20723 100644
--- a/arch/x86/kernel/dumpstack.c
+++ b/arch/x86/kernel/dumpstack.c
@@ -234,6 +234,7 @@
bust_spinlocks(1);
return flags;
}
+EXPORT_SYMBOL_GPL(oops_begin);
void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
{
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 0c2b7ef..294f26d 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -14,6 +14,7 @@
#include <linux/bootmem.h>
#include <linux/pfn.h>
#include <linux/suspend.h>
+#include <linux/acpi.h>
#include <linux/firmware-map.h>
#include <linux/memblock.h>
diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c
index 8f29560..ab23f1a 100644
--- a/arch/x86/kernel/module.c
+++ b/arch/x86/kernel/module.c
@@ -37,20 +37,11 @@
void *module_alloc(unsigned long size)
{
- struct vm_struct *area;
-
- if (!size)
+ if (PAGE_ALIGN(size) > MODULES_LEN)
return NULL;
- size = PAGE_ALIGN(size);
- if (size > MODULES_LEN)
- return NULL;
-
- area = __get_vm_area(size, VM_ALLOC, MODULES_VADDR, MODULES_END);
- if (!area)
- return NULL;
-
- return __vmalloc_area(area, GFP_KERNEL | __GFP_HIGHMEM,
- PAGE_KERNEL_EXEC);
+ return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END,
+ GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC,
+ -1, __builtin_return_address(0));
}
/* Free memory returned from module_alloc */
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index c5b2500..869e1ae 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -421,8 +421,11 @@
.set_pte = native_set_pte,
.set_pte_at = native_set_pte_at,
.set_pmd = native_set_pmd,
+ .set_pmd_at = native_set_pmd_at,
.pte_update = paravirt_nop,
.pte_update_defer = paravirt_nop,
+ .pmd_update = paravirt_nop,
+ .pmd_update_defer = paravirt_nop,
.ptep_modify_prot_start = __ptep_modify_prot_start,
.ptep_modify_prot_commit = __ptep_modify_prot_commit,
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 09c08a1..d8286ed 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -22,11 +22,6 @@
#include <asm/i387.h>
#include <asm/debugreg.h>
-unsigned long idle_halt;
-EXPORT_SYMBOL(idle_halt);
-unsigned long idle_nomwait;
-EXPORT_SYMBOL(idle_nomwait);
-
struct kmem_cache *task_xstate_cachep;
EXPORT_SYMBOL_GPL(task_xstate_cachep);
@@ -327,7 +322,7 @@
/*
* Idle related variables and functions
*/
-unsigned long boot_option_idle_override = 0;
+unsigned long boot_option_idle_override = IDLE_NO_OVERRIDE;
EXPORT_SYMBOL(boot_option_idle_override);
/*
@@ -386,6 +381,8 @@
else
local_irq_enable();
current_thread_info()->status |= TS_POLLING;
+ trace_power_end(smp_processor_id());
+ trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
} else {
local_irq_enable();
/* loop is done by the caller */
@@ -443,8 +440,6 @@
*/
void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
{
- trace_power_start(POWER_CSTATE, (ax>>4)+1, smp_processor_id());
- trace_cpu_idle((ax>>4)+1, smp_processor_id());
if (!need_resched()) {
if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLUSH_MONITOR))
clflush((void *)¤t_thread_info()->flags);
@@ -471,6 +466,8 @@
__sti_mwait(0, 0);
else
local_irq_enable();
+ trace_power_end(smp_processor_id());
+ trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
} else
local_irq_enable();
}
@@ -503,7 +500,6 @@
*
* idle=mwait overrides this decision and forces the usage of mwait.
*/
-static int __cpuinitdata force_mwait;
#define MWAIT_INFO 0x05
#define MWAIT_ECX_EXTENDED_INFO 0x01
@@ -513,7 +509,7 @@
{
u32 eax, ebx, ecx, edx;
- if (force_mwait)
+ if (boot_option_idle_override == IDLE_FORCE_MWAIT)
return 1;
if (c->cpuid_level < MWAIT_INFO)
@@ -633,9 +629,10 @@
if (!strcmp(str, "poll")) {
printk("using polling idle threads.\n");
pm_idle = poll_idle;
- } else if (!strcmp(str, "mwait"))
- force_mwait = 1;
- else if (!strcmp(str, "halt")) {
+ boot_option_idle_override = IDLE_POLL;
+ } else if (!strcmp(str, "mwait")) {
+ boot_option_idle_override = IDLE_FORCE_MWAIT;
+ } else if (!strcmp(str, "halt")) {
/*
* When the boot option of idle=halt is added, halt is
* forced to be used for CPU idle. In such case CPU C2/C3
@@ -644,8 +641,7 @@
* the boot_option_idle_override.
*/
pm_idle = default_idle;
- idle_halt = 1;
- return 0;
+ boot_option_idle_override = IDLE_HALT;
} else if (!strcmp(str, "nomwait")) {
/*
* If the boot option of "idle=nomwait" is added,
@@ -653,12 +649,10 @@
* states. In such case it won't touch the variable
* of boot_option_idle_override.
*/
- idle_nomwait = 1;
- return 0;
+ boot_option_idle_override = IDLE_NOMWAIT;
} else
return -1;
- boot_option_idle_override = 1;
return 0;
}
early_param("idle", idle_setup);
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 4b9befa..8d12878 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -57,8 +57,6 @@
#include <asm/syscalls.h>
#include <asm/debugreg.h>
-#include <trace/events/power.h>
-
asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
/*
@@ -113,8 +111,6 @@
stop_critical_timings();
pm_idle();
start_critical_timings();
- trace_power_end(smp_processor_id());
- trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
}
tick_nohz_restart_sched_tick();
preempt_enable_no_resched();
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 4c818a7..bd387e8 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -51,8 +51,6 @@
#include <asm/syscalls.h>
#include <asm/debugreg.h>
-#include <trace/events/power.h>
-
asmlinkage extern void ret_from_fork(void);
DEFINE_PER_CPU(unsigned long, old_rsp);
@@ -141,10 +139,6 @@
pm_idle();
start_critical_timings();
- trace_power_end(smp_processor_id());
- trace_cpu_idle(PWR_EVENT_EXIT,
- smp_processor_id());
-
/* In many cases the interrupt that ended idle
has already called exit_idle. But some idle
loops can be woken up without interrupt. */
diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c
index c2f1b26..998e972 100644
--- a/arch/x86/kernel/tboot.c
+++ b/arch/x86/kernel/tboot.c
@@ -133,7 +133,7 @@
pmd = pmd_alloc(&tboot_mm, pud, vaddr);
if (!pmd)
return -1;
- pte = pte_alloc_map(&tboot_mm, pmd, vaddr);
+ pte = pte_alloc_map(&tboot_mm, NULL, pmd, vaddr);
if (!pte)
return -1;
set_pte_at(&tboot_mm, vaddr, pte, pfn_pte(pfn, prot));
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 823f79a..ffe5755 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -464,7 +464,7 @@
tsc_pit_min = min(tsc_pit_min, tsc_pit_khz);
/* hpet or pmtimer available ? */
- if (!hpet && !ref1 && !ref2)
+ if (ref1 == ref2)
continue;
/* Check, whether the sampling was disturbed by an SMI */
@@ -935,7 +935,7 @@
tsc_stop = tsc_read_refs(&ref_stop, hpet);
/* hpet or pmtimer available ? */
- if (!hpet && !ref_start && !ref_stop)
+ if (ref_start == ref_stop)
goto out;
/* Check, whether the sampling was disturbed by an SMI */
diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c
index 61fb985..863f875 100644
--- a/arch/x86/kernel/vm86_32.c
+++ b/arch/x86/kernel/vm86_32.c
@@ -179,6 +179,7 @@
if (pud_none_or_clear_bad(pud))
goto out;
pmd = pmd_offset(pud, 0xA0000);
+ split_huge_page_pmd(mm, pmd);
if (pmd_none_or_clear_bad(pmd))
goto out;
pte = pte_offset_map_lock(mm, pmd, 0xA0000, &ptl);
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 9cafbb4..f02b8ed 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -554,14 +554,18 @@
return ret;
}
-static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn)
+static bool mapping_level_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t large_gfn)
{
struct kvm_memory_slot *slot;
- int host_level, level, max_level;
-
slot = gfn_to_memslot(vcpu->kvm, large_gfn);
if (slot && slot->dirty_bitmap)
- return PT_PAGE_TABLE_LEVEL;
+ return true;
+ return false;
+}
+
+static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn)
+{
+ int host_level, level, max_level;
host_level = host_mapping_level(vcpu->kvm, large_gfn);
@@ -941,6 +945,35 @@
return young;
}
+static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
+ unsigned long data)
+{
+ u64 *spte;
+ int young = 0;
+
+ /*
+ * If there's no access bit in the secondary pte set by the
+ * hardware it's up to gup-fast/gup to set the access bit in
+ * the primary pte or in the page structure.
+ */
+ if (!shadow_accessed_mask)
+ goto out;
+
+ spte = rmap_next(kvm, rmapp, NULL);
+ while (spte) {
+ u64 _spte = *spte;
+ BUG_ON(!(_spte & PT_PRESENT_MASK));
+ young = _spte & PT_ACCESSED_MASK;
+ if (young) {
+ young = 1;
+ break;
+ }
+ spte = rmap_next(kvm, rmapp, spte);
+ }
+out:
+ return young;
+}
+
#define RMAP_RECYCLE_THRESHOLD 1000
static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
@@ -961,6 +994,11 @@
return kvm_handle_hva(kvm, hva, 0, kvm_age_rmapp);
}
+int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
+{
+ return kvm_handle_hva(kvm, hva, 0, kvm_test_age_rmapp);
+}
+
#ifdef MMU_DEBUG
static int is_empty_shadow_page(u64 *spt)
{
@@ -2281,6 +2319,48 @@
return 1;
}
+static void transparent_hugepage_adjust(struct kvm_vcpu *vcpu,
+ gfn_t *gfnp, pfn_t *pfnp, int *levelp)
+{
+ pfn_t pfn = *pfnp;
+ gfn_t gfn = *gfnp;
+ int level = *levelp;
+
+ /*
+ * Check if it's a transparent hugepage. If this would be an
+ * hugetlbfs page, level wouldn't be set to
+ * PT_PAGE_TABLE_LEVEL and there would be no adjustment done
+ * here.
+ */
+ if (!is_error_pfn(pfn) && !kvm_is_mmio_pfn(pfn) &&
+ level == PT_PAGE_TABLE_LEVEL &&
+ PageTransCompound(pfn_to_page(pfn)) &&
+ !has_wrprotected_page(vcpu->kvm, gfn, PT_DIRECTORY_LEVEL)) {
+ unsigned long mask;
+ /*
+ * mmu_notifier_retry was successful and we hold the
+ * mmu_lock here, so the pmd can't become splitting
+ * from under us, and in turn
+ * __split_huge_page_refcount() can't run from under
+ * us and we can safely transfer the refcount from
+ * PG_tail to PG_head as we switch the pfn to tail to
+ * head.
+ */
+ *levelp = level = PT_DIRECTORY_LEVEL;
+ mask = KVM_PAGES_PER_HPAGE(level) - 1;
+ VM_BUG_ON((gfn & mask) != (pfn & mask));
+ if (pfn & mask) {
+ gfn &= ~mask;
+ *gfnp = gfn;
+ kvm_release_pfn_clean(pfn);
+ pfn &= ~mask;
+ if (!get_page_unless_zero(pfn_to_page(pfn)))
+ BUG();
+ *pfnp = pfn;
+ }
+ }
+}
+
static bool try_async_pf(struct kvm_vcpu *vcpu, bool prefault, gfn_t gfn,
gva_t gva, pfn_t *pfn, bool write, bool *writable);
@@ -2289,20 +2369,25 @@
{
int r;
int level;
+ int force_pt_level;
pfn_t pfn;
unsigned long mmu_seq;
bool map_writable;
- level = mapping_level(vcpu, gfn);
+ force_pt_level = mapping_level_dirty_bitmap(vcpu, gfn);
+ if (likely(!force_pt_level)) {
+ level = mapping_level(vcpu, gfn);
+ /*
+ * This path builds a PAE pagetable - so we can map
+ * 2mb pages at maximum. Therefore check if the level
+ * is larger than that.
+ */
+ if (level > PT_DIRECTORY_LEVEL)
+ level = PT_DIRECTORY_LEVEL;
- /*
- * This path builds a PAE pagetable - so we can map 2mb pages at
- * maximum. Therefore check if the level is larger than that.
- */
- if (level > PT_DIRECTORY_LEVEL)
- level = PT_DIRECTORY_LEVEL;
-
- gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1);
+ gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1);
+ } else
+ level = PT_PAGE_TABLE_LEVEL;
mmu_seq = vcpu->kvm->mmu_notifier_seq;
smp_rmb();
@@ -2318,6 +2403,8 @@
if (mmu_notifier_retry(vcpu, mmu_seq))
goto out_unlock;
kvm_mmu_free_some_pages(vcpu);
+ if (likely(!force_pt_level))
+ transparent_hugepage_adjust(vcpu, &gfn, &pfn, &level);
r = __direct_map(vcpu, v, write, map_writable, level, gfn, pfn,
prefault);
spin_unlock(&vcpu->kvm->mmu_lock);
@@ -2655,6 +2742,7 @@
pfn_t pfn;
int r;
int level;
+ int force_pt_level;
gfn_t gfn = gpa >> PAGE_SHIFT;
unsigned long mmu_seq;
int write = error_code & PFERR_WRITE_MASK;
@@ -2667,9 +2755,12 @@
if (r)
return r;
- level = mapping_level(vcpu, gfn);
-
- gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1);
+ force_pt_level = mapping_level_dirty_bitmap(vcpu, gfn);
+ if (likely(!force_pt_level)) {
+ level = mapping_level(vcpu, gfn);
+ gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1);
+ } else
+ level = PT_PAGE_TABLE_LEVEL;
mmu_seq = vcpu->kvm->mmu_notifier_seq;
smp_rmb();
@@ -2684,6 +2775,8 @@
if (mmu_notifier_retry(vcpu, mmu_seq))
goto out_unlock;
kvm_mmu_free_some_pages(vcpu);
+ if (likely(!force_pt_level))
+ transparent_hugepage_adjust(vcpu, &gfn, &pfn, &level);
r = __direct_map(vcpu, gpa, write, map_writable,
level, gfn, pfn, prefault);
spin_unlock(&vcpu->kvm->mmu_lock);
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 53210f1..6bccc24 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -550,6 +550,7 @@
int r;
pfn_t pfn;
int level = PT_PAGE_TABLE_LEVEL;
+ int force_pt_level;
unsigned long mmu_seq;
bool map_writable;
@@ -577,7 +578,11 @@
return 0;
}
- if (walker.level >= PT_DIRECTORY_LEVEL) {
+ if (walker.level >= PT_DIRECTORY_LEVEL)
+ force_pt_level = mapping_level_dirty_bitmap(vcpu, walker.gfn);
+ else
+ force_pt_level = 1;
+ if (!force_pt_level) {
level = min(walker.level, mapping_level(vcpu, walker.gfn));
walker.gfn = walker.gfn & ~(KVM_PAGES_PER_HPAGE(level) - 1);
}
@@ -599,6 +604,8 @@
trace_kvm_mmu_audit(vcpu, AUDIT_PRE_PAGE_FAULT);
kvm_mmu_free_some_pages(vcpu);
+ if (!force_pt_level)
+ transparent_hugepage_adjust(vcpu, &walker.gfn, &pfn, &level);
sptep = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
level, &write_pt, pfn, map_writable, prefault);
(void)sptep;
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
index 738e659..dbe34b9 100644
--- a/arch/x86/mm/gup.c
+++ b/arch/x86/mm/gup.c
@@ -8,6 +8,7 @@
#include <linux/mm.h>
#include <linux/vmstat.h>
#include <linux/highmem.h>
+#include <linux/swap.h>
#include <asm/pgtable.h>
@@ -89,6 +90,7 @@
VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
page = pte_page(pte);
get_page(page);
+ SetPageReferenced(page);
pages[*nr] = page;
(*nr)++;
@@ -103,6 +105,17 @@
VM_BUG_ON(page != compound_head(page));
VM_BUG_ON(page_count(page) == 0);
atomic_add(nr, &page->_count);
+ SetPageReferenced(page);
+}
+
+static inline void get_huge_page_tail(struct page *page)
+{
+ /*
+ * __split_huge_page_refcount() cannot run
+ * from under us.
+ */
+ VM_BUG_ON(atomic_read(&page->_count) < 0);
+ atomic_inc(&page->_count);
}
static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr,
@@ -128,6 +141,8 @@
do {
VM_BUG_ON(compound_head(page) != head);
pages[*nr] = page;
+ if (PageTail(page))
+ get_huge_page_tail(page);
(*nr)++;
page++;
refs++;
@@ -148,7 +163,18 @@
pmd_t pmd = *pmdp;
next = pmd_addr_end(addr, end);
- if (pmd_none(pmd))
+ /*
+ * The pmd_trans_splitting() check below explains why
+ * pmdp_splitting_flush has to flush the tlb, to stop
+ * this gup-fast code from running while we set the
+ * splitting bit in the pmd. Returning zero will take
+ * the slow path that will call wait_split_huge_page()
+ * if the pmd is still in splitting state. gup-fast
+ * can't because it has irq disabled and
+ * wait_split_huge_page() would never return as the
+ * tlb flush IPI wouldn't run.
+ */
+ if (pmd_none(pmd) || pmd_trans_splitting(pmd))
return 0;
if (unlikely(pmd_large(pmd))) {
if (!gup_huge_pmd(pmd, addr, next, write, pages, nr))
diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index 8be8c7d..500242d 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -320,6 +320,25 @@
return changed;
}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+int pmdp_set_access_flags(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp,
+ pmd_t entry, int dirty)
+{
+ int changed = !pmd_same(*pmdp, entry);
+
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+
+ if (changed && dirty) {
+ *pmdp = entry;
+ pmd_update_defer(vma->vm_mm, address, pmdp);
+ flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+ }
+
+ return changed;
+}
+#endif
+
int ptep_test_and_clear_young(struct vm_area_struct *vma,
unsigned long addr, pte_t *ptep)
{
@@ -335,6 +354,23 @@
return ret;
}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+int pmdp_test_and_clear_young(struct vm_area_struct *vma,
+ unsigned long addr, pmd_t *pmdp)
+{
+ int ret = 0;
+
+ if (pmd_young(*pmdp))
+ ret = test_and_clear_bit(_PAGE_BIT_ACCESSED,
+ (unsigned long *)pmdp);
+
+ if (ret)
+ pmd_update(vma->vm_mm, addr, pmdp);
+
+ return ret;
+}
+#endif
+
int ptep_clear_flush_young(struct vm_area_struct *vma,
unsigned long address, pte_t *ptep)
{
@@ -347,6 +383,36 @@
return young;
}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+int pmdp_clear_flush_young(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp)
+{
+ int young;
+
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+
+ young = pmdp_test_and_clear_young(vma, address, pmdp);
+ if (young)
+ flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+
+ return young;
+}
+
+void pmdp_splitting_flush(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp)
+{
+ int set;
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+ set = !test_and_set_bit(_PAGE_BIT_SPLITTING,
+ (unsigned long *)pmdp);
+ if (set) {
+ pmd_update(vma->vm_mm, address, pmdp);
+ /* need tlb flush only to serialize against gup-fast */
+ flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+ }
+}
+#endif
+
/**
* reserve_top_address - reserves a hole in the top of kernel address space
* @reserve - size of hole to reserve
diff --git a/arch/x86/pci/broadcom_bus.c b/arch/x86/pci/broadcom_bus.c
index 0846a5b..ab8269b 100644
--- a/arch/x86/pci/broadcom_bus.c
+++ b/arch/x86/pci/broadcom_bus.c
@@ -9,6 +9,7 @@
* option) any later version.
*/
+#include <linux/acpi.h>
#include <linux/delay.h>
#include <linux/dmi.h>
#include <linux/pci.h>
@@ -25,12 +26,14 @@
u8 fbus, lbus;
int i;
+#ifdef CONFIG_ACPI
/*
- * The x86_pci_root_bus_res_quirks() function already refuses to use
- * this information if ACPI _CRS was used. Therefore, we don't bother
- * checking if ACPI is enabled, and just generate the information
- * for both the ACPI _CRS and no ACPI cases.
+ * We should get host bridge information from ACPI unless the BIOS
+ * doesn't support it.
*/
+ if (acpi_os_get_root_pointer())
+ return;
+#endif
info = &pci_root_info[pci_root_num];
pci_root_num++;
diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index f7c8a39..5fe7502 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -22,6 +22,7 @@
unsigned int pci_early_dump_regs;
static int pci_bf_sort;
+static int smbios_type_b1_flag;
int pci_routeirq;
int noioapicquirk;
#ifdef CONFIG_X86_REROUTE_FOR_BROKEN_BOOT_IRQS
@@ -185,6 +186,39 @@
return 0;
}
+static void __devinit read_dmi_type_b1(const struct dmi_header *dm,
+ void *private_data)
+{
+ u8 *d = (u8 *)dm + 4;
+
+ if (dm->type != 0xB1)
+ return;
+ switch (((*(u32 *)d) >> 9) & 0x03) {
+ case 0x00:
+ printk(KERN_INFO "dmi type 0xB1 record - unknown flag\n");
+ break;
+ case 0x01: /* set pci=bfsort */
+ smbios_type_b1_flag = 1;
+ break;
+ case 0x02: /* do not set pci=bfsort */
+ smbios_type_b1_flag = 2;
+ break;
+ default:
+ break;
+ }
+}
+
+static int __devinit find_sort_method(const struct dmi_system_id *d)
+{
+ dmi_walk(read_dmi_type_b1, NULL);
+
+ if (smbios_type_b1_flag == 1) {
+ set_bf_sort(d);
+ return 0;
+ }
+ return -1;
+}
+
/*
* Enable renumbering of PCI bus# ranges to reach all PCI busses (Cardbus)
*/
@@ -213,6 +247,13 @@
},
#endif /* __i386__ */
{
+ .callback = find_sort_method,
+ .ident = "Dell System",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc"),
+ },
+ },
+ {
.callback = set_bf_sort,
.ident = "Dell PowerEdge 1950",
.matches = {
diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c
index 9f9bfb7..87e6c83 100644
--- a/arch/x86/pci/irq.c
+++ b/arch/x86/pci/irq.c
@@ -589,7 +589,8 @@
case PCI_DEVICE_ID_INTEL_ICH10_1:
case PCI_DEVICE_ID_INTEL_ICH10_2:
case PCI_DEVICE_ID_INTEL_ICH10_3:
- case PCI_DEVICE_ID_INTEL_PATSBURG_LPC:
+ case PCI_DEVICE_ID_INTEL_PATSBURG_LPC_0:
+ case PCI_DEVICE_ID_INTEL_PATSBURG_LPC_1:
r->name = "PIIX/ICH";
r->get = pirq_piix_get;
r->set = pirq_piix_set;
diff --git a/arch/x86/platform/olpc/olpc-xo1.c b/arch/x86/platform/olpc/olpc-xo1.c
index f5442c0..1277756 100644
--- a/arch/x86/platform/olpc/olpc-xo1.c
+++ b/arch/x86/platform/olpc/olpc-xo1.c
@@ -1,6 +1,7 @@
/*
* Support for features of the OLPC XO-1 laptop
*
+ * Copyright (C) 2010 Andres Salomon <dilinger@queued.net>
* Copyright (C) 2010 One Laptop per Child
* Copyright (C) 2006 Red Hat, Inc.
* Copyright (C) 2006 Advanced Micro Devices, Inc.
@@ -12,8 +13,6 @@
*/
#include <linux/module.h>
-#include <linux/pci.h>
-#include <linux/pci_ids.h>
#include <linux/platform_device.h>
#include <linux/pm.h>
@@ -22,9 +21,6 @@
#define DRV_NAME "olpc-xo1"
-#define PMS_BAR 4
-#define ACPI_BAR 5
-
/* PMC registers (PMS block) */
#define PM_SCLK 0x10
#define PM_IN_SLPCTL 0x20
@@ -57,65 +53,67 @@
outl(0x00002000, acpi_base + PM1_CNT);
}
-/* Read the base addresses from the PCI BAR info */
-static int __devinit setup_bases(struct pci_dev *pdev)
-{
- int r;
-
- r = pci_enable_device_io(pdev);
- if (r) {
- dev_err(&pdev->dev, "can't enable device IO\n");
- return r;
- }
-
- r = pci_request_region(pdev, ACPI_BAR, DRV_NAME);
- if (r) {
- dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", ACPI_BAR);
- return r;
- }
-
- r = pci_request_region(pdev, PMS_BAR, DRV_NAME);
- if (r) {
- dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", PMS_BAR);
- pci_release_region(pdev, ACPI_BAR);
- return r;
- }
-
- acpi_base = pci_resource_start(pdev, ACPI_BAR);
- pms_base = pci_resource_start(pdev, PMS_BAR);
-
- return 0;
-}
-
static int __devinit olpc_xo1_probe(struct platform_device *pdev)
{
- struct pci_dev *pcidev;
- int r;
+ struct resource *res;
- pcidev = pci_get_device(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA,
- NULL);
- if (!pdev)
+ /* don't run on non-XOs */
+ if (!machine_is_olpc())
return -ENODEV;
- r = setup_bases(pcidev);
- if (r)
- return r;
+ res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+ if (!res) {
+ dev_err(&pdev->dev, "can't fetch device resource info\n");
+ return -EIO;
+ }
- pm_power_off = xo1_power_off;
+ if (!request_region(res->start, resource_size(res), DRV_NAME)) {
+ dev_err(&pdev->dev, "can't request region\n");
+ return -EIO;
+ }
- printk(KERN_INFO "OLPC XO-1 support registered\n");
+ if (strcmp(pdev->name, "cs5535-pms") == 0)
+ pms_base = res->start;
+ else if (strcmp(pdev->name, "cs5535-acpi") == 0)
+ acpi_base = res->start;
+
+ /* If we have both addresses, we can override the poweroff hook */
+ if (pms_base && acpi_base) {
+ pm_power_off = xo1_power_off;
+ printk(KERN_INFO "OLPC XO-1 support registered\n");
+ }
+
return 0;
}
static int __devexit olpc_xo1_remove(struct platform_device *pdev)
{
+ struct resource *r;
+
+ r = platform_get_resource(pdev, IORESOURCE_IO, 0);
+ release_region(r->start, resource_size(r));
+
+ if (strcmp(pdev->name, "cs5535-pms") == 0)
+ pms_base = 0;
+ else if (strcmp(pdev->name, "cs5535-acpi") == 0)
+ acpi_base = 0;
+
pm_power_off = NULL;
return 0;
}
-static struct platform_driver olpc_xo1_driver = {
+static struct platform_driver cs5535_pms_drv = {
.driver = {
- .name = DRV_NAME,
+ .name = "cs5535-pms",
+ .owner = THIS_MODULE,
+ },
+ .probe = olpc_xo1_probe,
+ .remove = __devexit_p(olpc_xo1_remove),
+};
+
+static struct platform_driver cs5535_acpi_drv = {
+ .driver = {
+ .name = "cs5535-acpi",
.owner = THIS_MODULE,
},
.probe = olpc_xo1_probe,
@@ -124,12 +122,23 @@
static int __init olpc_xo1_init(void)
{
- return platform_driver_register(&olpc_xo1_driver);
+ int r;
+
+ r = platform_driver_register(&cs5535_pms_drv);
+ if (r)
+ return r;
+
+ r = platform_driver_register(&cs5535_acpi_drv);
+ if (r)
+ platform_driver_unregister(&cs5535_pms_drv);
+
+ return r;
}
static void __exit olpc_xo1_exit(void)
{
- platform_driver_unregister(&olpc_xo1_driver);
+ platform_driver_unregister(&cs5535_acpi_drv);
+ platform_driver_unregister(&cs5535_pms_drv);
}
MODULE_AUTHOR("Daniel Drake <dsd@laptop.org>");
diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile
index 7793851..17c565d 100644
--- a/arch/x86/xen/Makefile
+++ b/arch/x86/xen/Makefile
@@ -12,7 +12,8 @@
obj-y := enlighten.o setup.o multicalls.o mmu.o irq.o \
time.o xen-asm.o xen-asm_$(BITS).o \
- grant-table.o suspend.o platform-pci-unplug.o
+ grant-table.o suspend.o platform-pci-unplug.o \
+ p2m.o
obj-$(CONFIG_SMP) += smp.o
obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= spinlock.o
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 44924e5..5e92b61 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -173,371 +173,6 @@
*/
#define USER_LIMIT ((STACK_TOP_MAX + PGDIR_SIZE - 1) & PGDIR_MASK)
-/*
- * Xen leaves the responsibility for maintaining p2m mappings to the
- * guests themselves, but it must also access and update the p2m array
- * during suspend/resume when all the pages are reallocated.
- *
- * The p2m table is logically a flat array, but we implement it as a
- * three-level tree to allow the address space to be sparse.
- *
- * Xen
- * |
- * p2m_top p2m_top_mfn
- * / \ / \
- * p2m_mid p2m_mid p2m_mid_mfn p2m_mid_mfn
- * / \ / \ / /
- * p2m p2m p2m p2m p2m p2m p2m ...
- *
- * The p2m_mid_mfn pages are mapped by p2m_top_mfn_p.
- *
- * The p2m_top and p2m_top_mfn levels are limited to 1 page, so the
- * maximum representable pseudo-physical address space is:
- * P2M_TOP_PER_PAGE * P2M_MID_PER_PAGE * P2M_PER_PAGE pages
- *
- * P2M_PER_PAGE depends on the architecture, as a mfn is always
- * unsigned long (8 bytes on 64-bit, 4 bytes on 32), leading to
- * 512 and 1024 entries respectively.
- */
-
-unsigned long xen_max_p2m_pfn __read_mostly;
-
-#define P2M_PER_PAGE (PAGE_SIZE / sizeof(unsigned long))
-#define P2M_MID_PER_PAGE (PAGE_SIZE / sizeof(unsigned long *))
-#define P2M_TOP_PER_PAGE (PAGE_SIZE / sizeof(unsigned long **))
-
-#define MAX_P2M_PFN (P2M_TOP_PER_PAGE * P2M_MID_PER_PAGE * P2M_PER_PAGE)
-
-/* Placeholders for holes in the address space */
-static RESERVE_BRK_ARRAY(unsigned long, p2m_missing, P2M_PER_PAGE);
-static RESERVE_BRK_ARRAY(unsigned long *, p2m_mid_missing, P2M_MID_PER_PAGE);
-static RESERVE_BRK_ARRAY(unsigned long, p2m_mid_missing_mfn, P2M_MID_PER_PAGE);
-
-static RESERVE_BRK_ARRAY(unsigned long **, p2m_top, P2M_TOP_PER_PAGE);
-static RESERVE_BRK_ARRAY(unsigned long, p2m_top_mfn, P2M_TOP_PER_PAGE);
-static RESERVE_BRK_ARRAY(unsigned long *, p2m_top_mfn_p, P2M_TOP_PER_PAGE);
-
-RESERVE_BRK(p2m_mid, PAGE_SIZE * (MAX_DOMAIN_PAGES / (P2M_PER_PAGE * P2M_MID_PER_PAGE)));
-RESERVE_BRK(p2m_mid_mfn, PAGE_SIZE * (MAX_DOMAIN_PAGES / (P2M_PER_PAGE * P2M_MID_PER_PAGE)));
-
-static inline unsigned p2m_top_index(unsigned long pfn)
-{
- BUG_ON(pfn >= MAX_P2M_PFN);
- return pfn / (P2M_MID_PER_PAGE * P2M_PER_PAGE);
-}
-
-static inline unsigned p2m_mid_index(unsigned long pfn)
-{
- return (pfn / P2M_PER_PAGE) % P2M_MID_PER_PAGE;
-}
-
-static inline unsigned p2m_index(unsigned long pfn)
-{
- return pfn % P2M_PER_PAGE;
-}
-
-static void p2m_top_init(unsigned long ***top)
-{
- unsigned i;
-
- for (i = 0; i < P2M_TOP_PER_PAGE; i++)
- top[i] = p2m_mid_missing;
-}
-
-static void p2m_top_mfn_init(unsigned long *top)
-{
- unsigned i;
-
- for (i = 0; i < P2M_TOP_PER_PAGE; i++)
- top[i] = virt_to_mfn(p2m_mid_missing_mfn);
-}
-
-static void p2m_top_mfn_p_init(unsigned long **top)
-{
- unsigned i;
-
- for (i = 0; i < P2M_TOP_PER_PAGE; i++)
- top[i] = p2m_mid_missing_mfn;
-}
-
-static void p2m_mid_init(unsigned long **mid)
-{
- unsigned i;
-
- for (i = 0; i < P2M_MID_PER_PAGE; i++)
- mid[i] = p2m_missing;
-}
-
-static void p2m_mid_mfn_init(unsigned long *mid)
-{
- unsigned i;
-
- for (i = 0; i < P2M_MID_PER_PAGE; i++)
- mid[i] = virt_to_mfn(p2m_missing);
-}
-
-static void p2m_init(unsigned long *p2m)
-{
- unsigned i;
-
- for (i = 0; i < P2M_MID_PER_PAGE; i++)
- p2m[i] = INVALID_P2M_ENTRY;
-}
-
-/*
- * Build the parallel p2m_top_mfn and p2m_mid_mfn structures
- *
- * This is called both at boot time, and after resuming from suspend:
- * - At boot time we're called very early, and must use extend_brk()
- * to allocate memory.
- *
- * - After resume we're called from within stop_machine, but the mfn
- * tree should alreay be completely allocated.
- */
-void xen_build_mfn_list_list(void)
-{
- unsigned long pfn;
-
- /* Pre-initialize p2m_top_mfn to be completely missing */
- if (p2m_top_mfn == NULL) {
- p2m_mid_missing_mfn = extend_brk(PAGE_SIZE, PAGE_SIZE);
- p2m_mid_mfn_init(p2m_mid_missing_mfn);
-
- p2m_top_mfn_p = extend_brk(PAGE_SIZE, PAGE_SIZE);
- p2m_top_mfn_p_init(p2m_top_mfn_p);
-
- p2m_top_mfn = extend_brk(PAGE_SIZE, PAGE_SIZE);
- p2m_top_mfn_init(p2m_top_mfn);
- } else {
- /* Reinitialise, mfn's all change after migration */
- p2m_mid_mfn_init(p2m_mid_missing_mfn);
- }
-
- for (pfn = 0; pfn < xen_max_p2m_pfn; pfn += P2M_PER_PAGE) {
- unsigned topidx = p2m_top_index(pfn);
- unsigned mididx = p2m_mid_index(pfn);
- unsigned long **mid;
- unsigned long *mid_mfn_p;
-
- mid = p2m_top[topidx];
- mid_mfn_p = p2m_top_mfn_p[topidx];
-
- /* Don't bother allocating any mfn mid levels if
- * they're just missing, just update the stored mfn,
- * since all could have changed over a migrate.
- */
- if (mid == p2m_mid_missing) {
- BUG_ON(mididx);
- BUG_ON(mid_mfn_p != p2m_mid_missing_mfn);
- p2m_top_mfn[topidx] = virt_to_mfn(p2m_mid_missing_mfn);
- pfn += (P2M_MID_PER_PAGE - 1) * P2M_PER_PAGE;
- continue;
- }
-
- if (mid_mfn_p == p2m_mid_missing_mfn) {
- /*
- * XXX boot-time only! We should never find
- * missing parts of the mfn tree after
- * runtime. extend_brk() will BUG if we call
- * it too late.
- */
- mid_mfn_p = extend_brk(PAGE_SIZE, PAGE_SIZE);
- p2m_mid_mfn_init(mid_mfn_p);
-
- p2m_top_mfn_p[topidx] = mid_mfn_p;
- }
-
- p2m_top_mfn[topidx] = virt_to_mfn(mid_mfn_p);
- mid_mfn_p[mididx] = virt_to_mfn(mid[mididx]);
- }
-}
-
-void xen_setup_mfn_list_list(void)
-{
- BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info);
-
- HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
- virt_to_mfn(p2m_top_mfn);
- HYPERVISOR_shared_info->arch.max_pfn = xen_max_p2m_pfn;
-}
-
-/* Set up p2m_top to point to the domain-builder provided p2m pages */
-void __init xen_build_dynamic_phys_to_machine(void)
-{
- unsigned long *mfn_list = (unsigned long *)xen_start_info->mfn_list;
- unsigned long max_pfn = min(MAX_DOMAIN_PAGES, xen_start_info->nr_pages);
- unsigned long pfn;
-
- xen_max_p2m_pfn = max_pfn;
-
- p2m_missing = extend_brk(PAGE_SIZE, PAGE_SIZE);
- p2m_init(p2m_missing);
-
- p2m_mid_missing = extend_brk(PAGE_SIZE, PAGE_SIZE);
- p2m_mid_init(p2m_mid_missing);
-
- p2m_top = extend_brk(PAGE_SIZE, PAGE_SIZE);
- p2m_top_init(p2m_top);
-
- /*
- * The domain builder gives us a pre-constructed p2m array in
- * mfn_list for all the pages initially given to us, so we just
- * need to graft that into our tree structure.
- */
- for (pfn = 0; pfn < max_pfn; pfn += P2M_PER_PAGE) {
- unsigned topidx = p2m_top_index(pfn);
- unsigned mididx = p2m_mid_index(pfn);
-
- if (p2m_top[topidx] == p2m_mid_missing) {
- unsigned long **mid = extend_brk(PAGE_SIZE, PAGE_SIZE);
- p2m_mid_init(mid);
-
- p2m_top[topidx] = mid;
- }
-
- p2m_top[topidx][mididx] = &mfn_list[pfn];
- }
-}
-
-unsigned long get_phys_to_machine(unsigned long pfn)
-{
- unsigned topidx, mididx, idx;
-
- if (unlikely(pfn >= MAX_P2M_PFN))
- return INVALID_P2M_ENTRY;
-
- topidx = p2m_top_index(pfn);
- mididx = p2m_mid_index(pfn);
- idx = p2m_index(pfn);
-
- return p2m_top[topidx][mididx][idx];
-}
-EXPORT_SYMBOL_GPL(get_phys_to_machine);
-
-static void *alloc_p2m_page(void)
-{
- return (void *)__get_free_page(GFP_KERNEL | __GFP_REPEAT);
-}
-
-static void free_p2m_page(void *p)
-{
- free_page((unsigned long)p);
-}
-
-/*
- * Fully allocate the p2m structure for a given pfn. We need to check
- * that both the top and mid levels are allocated, and make sure the
- * parallel mfn tree is kept in sync. We may race with other cpus, so
- * the new pages are installed with cmpxchg; if we lose the race then
- * simply free the page we allocated and use the one that's there.
- */
-static bool alloc_p2m(unsigned long pfn)
-{
- unsigned topidx, mididx;
- unsigned long ***top_p, **mid;
- unsigned long *top_mfn_p, *mid_mfn;
-
- topidx = p2m_top_index(pfn);
- mididx = p2m_mid_index(pfn);
-
- top_p = &p2m_top[topidx];
- mid = *top_p;
-
- if (mid == p2m_mid_missing) {
- /* Mid level is missing, allocate a new one */
- mid = alloc_p2m_page();
- if (!mid)
- return false;
-
- p2m_mid_init(mid);
-
- if (cmpxchg(top_p, p2m_mid_missing, mid) != p2m_mid_missing)
- free_p2m_page(mid);
- }
-
- top_mfn_p = &p2m_top_mfn[topidx];
- mid_mfn = p2m_top_mfn_p[topidx];
-
- BUG_ON(virt_to_mfn(mid_mfn) != *top_mfn_p);
-
- if (mid_mfn == p2m_mid_missing_mfn) {
- /* Separately check the mid mfn level */
- unsigned long missing_mfn;
- unsigned long mid_mfn_mfn;
-
- mid_mfn = alloc_p2m_page();
- if (!mid_mfn)
- return false;
-
- p2m_mid_mfn_init(mid_mfn);
-
- missing_mfn = virt_to_mfn(p2m_mid_missing_mfn);
- mid_mfn_mfn = virt_to_mfn(mid_mfn);
- if (cmpxchg(top_mfn_p, missing_mfn, mid_mfn_mfn) != missing_mfn)
- free_p2m_page(mid_mfn);
- else
- p2m_top_mfn_p[topidx] = mid_mfn;
- }
-
- if (p2m_top[topidx][mididx] == p2m_missing) {
- /* p2m leaf page is missing */
- unsigned long *p2m;
-
- p2m = alloc_p2m_page();
- if (!p2m)
- return false;
-
- p2m_init(p2m);
-
- if (cmpxchg(&mid[mididx], p2m_missing, p2m) != p2m_missing)
- free_p2m_page(p2m);
- else
- mid_mfn[mididx] = virt_to_mfn(p2m);
- }
-
- return true;
-}
-
-/* Try to install p2m mapping; fail if intermediate bits missing */
-bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn)
-{
- unsigned topidx, mididx, idx;
-
- if (unlikely(pfn >= MAX_P2M_PFN)) {
- BUG_ON(mfn != INVALID_P2M_ENTRY);
- return true;
- }
-
- topidx = p2m_top_index(pfn);
- mididx = p2m_mid_index(pfn);
- idx = p2m_index(pfn);
-
- if (p2m_top[topidx][mididx] == p2m_missing)
- return mfn == INVALID_P2M_ENTRY;
-
- p2m_top[topidx][mididx][idx] = mfn;
-
- return true;
-}
-
-bool set_phys_to_machine(unsigned long pfn, unsigned long mfn)
-{
- if (unlikely(xen_feature(XENFEAT_auto_translated_physmap))) {
- BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
- return true;
- }
-
- if (unlikely(!__set_phys_to_machine(pfn, mfn))) {
- if (!alloc_p2m(pfn))
- return false;
-
- if (!__set_phys_to_machine(pfn, mfn))
- return false;
- }
-
- return true;
-}
-
unsigned long arbitrary_virt_to_mfn(void *vaddr)
{
xmaddr_t maddr = arbitrary_virt_to_machine(vaddr);
@@ -566,6 +201,7 @@
offset = address & ~PAGE_MASK;
return XMADDR(((phys_addr_t)pte_mfn(*pte) << PAGE_SHIFT) + offset);
}
+EXPORT_SYMBOL_GPL(arbitrary_virt_to_machine);
void make_lowmem_page_readonly(void *vaddr)
{
diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c
new file mode 100644
index 0000000..8f2251d
--- /dev/null
+++ b/arch/x86/xen/p2m.c
@@ -0,0 +1,510 @@
+/*
+ * Xen leaves the responsibility for maintaining p2m mappings to the
+ * guests themselves, but it must also access and update the p2m array
+ * during suspend/resume when all the pages are reallocated.
+ *
+ * The p2m table is logically a flat array, but we implement it as a
+ * three-level tree to allow the address space to be sparse.
+ *
+ * Xen
+ * |
+ * p2m_top p2m_top_mfn
+ * / \ / \
+ * p2m_mid p2m_mid p2m_mid_mfn p2m_mid_mfn
+ * / \ / \ / /
+ * p2m p2m p2m p2m p2m p2m p2m ...
+ *
+ * The p2m_mid_mfn pages are mapped by p2m_top_mfn_p.
+ *
+ * The p2m_top and p2m_top_mfn levels are limited to 1 page, so the
+ * maximum representable pseudo-physical address space is:
+ * P2M_TOP_PER_PAGE * P2M_MID_PER_PAGE * P2M_PER_PAGE pages
+ *
+ * P2M_PER_PAGE depends on the architecture, as a mfn is always
+ * unsigned long (8 bytes on 64-bit, 4 bytes on 32), leading to
+ * 512 and 1024 entries respectively.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/hash.h>
+#include <linux/sched.h>
+
+#include <asm/cache.h>
+#include <asm/setup.h>
+
+#include <asm/xen/page.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/hypervisor.h>
+
+#include "xen-ops.h"
+
+static void __init m2p_override_init(void);
+
+unsigned long xen_max_p2m_pfn __read_mostly;
+
+#define P2M_PER_PAGE (PAGE_SIZE / sizeof(unsigned long))
+#define P2M_MID_PER_PAGE (PAGE_SIZE / sizeof(unsigned long *))
+#define P2M_TOP_PER_PAGE (PAGE_SIZE / sizeof(unsigned long **))
+
+#define MAX_P2M_PFN (P2M_TOP_PER_PAGE * P2M_MID_PER_PAGE * P2M_PER_PAGE)
+
+/* Placeholders for holes in the address space */
+static RESERVE_BRK_ARRAY(unsigned long, p2m_missing, P2M_PER_PAGE);
+static RESERVE_BRK_ARRAY(unsigned long *, p2m_mid_missing, P2M_MID_PER_PAGE);
+static RESERVE_BRK_ARRAY(unsigned long, p2m_mid_missing_mfn, P2M_MID_PER_PAGE);
+
+static RESERVE_BRK_ARRAY(unsigned long **, p2m_top, P2M_TOP_PER_PAGE);
+static RESERVE_BRK_ARRAY(unsigned long, p2m_top_mfn, P2M_TOP_PER_PAGE);
+static RESERVE_BRK_ARRAY(unsigned long *, p2m_top_mfn_p, P2M_TOP_PER_PAGE);
+
+RESERVE_BRK(p2m_mid, PAGE_SIZE * (MAX_DOMAIN_PAGES / (P2M_PER_PAGE * P2M_MID_PER_PAGE)));
+RESERVE_BRK(p2m_mid_mfn, PAGE_SIZE * (MAX_DOMAIN_PAGES / (P2M_PER_PAGE * P2M_MID_PER_PAGE)));
+
+static inline unsigned p2m_top_index(unsigned long pfn)
+{
+ BUG_ON(pfn >= MAX_P2M_PFN);
+ return pfn / (P2M_MID_PER_PAGE * P2M_PER_PAGE);
+}
+
+static inline unsigned p2m_mid_index(unsigned long pfn)
+{
+ return (pfn / P2M_PER_PAGE) % P2M_MID_PER_PAGE;
+}
+
+static inline unsigned p2m_index(unsigned long pfn)
+{
+ return pfn % P2M_PER_PAGE;
+}
+
+static void p2m_top_init(unsigned long ***top)
+{
+ unsigned i;
+
+ for (i = 0; i < P2M_TOP_PER_PAGE; i++)
+ top[i] = p2m_mid_missing;
+}
+
+static void p2m_top_mfn_init(unsigned long *top)
+{
+ unsigned i;
+
+ for (i = 0; i < P2M_TOP_PER_PAGE; i++)
+ top[i] = virt_to_mfn(p2m_mid_missing_mfn);
+}
+
+static void p2m_top_mfn_p_init(unsigned long **top)
+{
+ unsigned i;
+
+ for (i = 0; i < P2M_TOP_PER_PAGE; i++)
+ top[i] = p2m_mid_missing_mfn;
+}
+
+static void p2m_mid_init(unsigned long **mid)
+{
+ unsigned i;
+
+ for (i = 0; i < P2M_MID_PER_PAGE; i++)
+ mid[i] = p2m_missing;
+}
+
+static void p2m_mid_mfn_init(unsigned long *mid)
+{
+ unsigned i;
+
+ for (i = 0; i < P2M_MID_PER_PAGE; i++)
+ mid[i] = virt_to_mfn(p2m_missing);
+}
+
+static void p2m_init(unsigned long *p2m)
+{
+ unsigned i;
+
+ for (i = 0; i < P2M_MID_PER_PAGE; i++)
+ p2m[i] = INVALID_P2M_ENTRY;
+}
+
+/*
+ * Build the parallel p2m_top_mfn and p2m_mid_mfn structures
+ *
+ * This is called both at boot time, and after resuming from suspend:
+ * - At boot time we're called very early, and must use extend_brk()
+ * to allocate memory.
+ *
+ * - After resume we're called from within stop_machine, but the mfn
+ * tree should alreay be completely allocated.
+ */
+void xen_build_mfn_list_list(void)
+{
+ unsigned long pfn;
+
+ /* Pre-initialize p2m_top_mfn to be completely missing */
+ if (p2m_top_mfn == NULL) {
+ p2m_mid_missing_mfn = extend_brk(PAGE_SIZE, PAGE_SIZE);
+ p2m_mid_mfn_init(p2m_mid_missing_mfn);
+
+ p2m_top_mfn_p = extend_brk(PAGE_SIZE, PAGE_SIZE);
+ p2m_top_mfn_p_init(p2m_top_mfn_p);
+
+ p2m_top_mfn = extend_brk(PAGE_SIZE, PAGE_SIZE);
+ p2m_top_mfn_init(p2m_top_mfn);
+ } else {
+ /* Reinitialise, mfn's all change after migration */
+ p2m_mid_mfn_init(p2m_mid_missing_mfn);
+ }
+
+ for (pfn = 0; pfn < xen_max_p2m_pfn; pfn += P2M_PER_PAGE) {
+ unsigned topidx = p2m_top_index(pfn);
+ unsigned mididx = p2m_mid_index(pfn);
+ unsigned long **mid;
+ unsigned long *mid_mfn_p;
+
+ mid = p2m_top[topidx];
+ mid_mfn_p = p2m_top_mfn_p[topidx];
+
+ /* Don't bother allocating any mfn mid levels if
+ * they're just missing, just update the stored mfn,
+ * since all could have changed over a migrate.
+ */
+ if (mid == p2m_mid_missing) {
+ BUG_ON(mididx);
+ BUG_ON(mid_mfn_p != p2m_mid_missing_mfn);
+ p2m_top_mfn[topidx] = virt_to_mfn(p2m_mid_missing_mfn);
+ pfn += (P2M_MID_PER_PAGE - 1) * P2M_PER_PAGE;
+ continue;
+ }
+
+ if (mid_mfn_p == p2m_mid_missing_mfn) {
+ /*
+ * XXX boot-time only! We should never find
+ * missing parts of the mfn tree after
+ * runtime. extend_brk() will BUG if we call
+ * it too late.
+ */
+ mid_mfn_p = extend_brk(PAGE_SIZE, PAGE_SIZE);
+ p2m_mid_mfn_init(mid_mfn_p);
+
+ p2m_top_mfn_p[topidx] = mid_mfn_p;
+ }
+
+ p2m_top_mfn[topidx] = virt_to_mfn(mid_mfn_p);
+ mid_mfn_p[mididx] = virt_to_mfn(mid[mididx]);
+ }
+}
+
+void xen_setup_mfn_list_list(void)
+{
+ BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info);
+
+ HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
+ virt_to_mfn(p2m_top_mfn);
+ HYPERVISOR_shared_info->arch.max_pfn = xen_max_p2m_pfn;
+}
+
+/* Set up p2m_top to point to the domain-builder provided p2m pages */
+void __init xen_build_dynamic_phys_to_machine(void)
+{
+ unsigned long *mfn_list = (unsigned long *)xen_start_info->mfn_list;
+ unsigned long max_pfn = min(MAX_DOMAIN_PAGES, xen_start_info->nr_pages);
+ unsigned long pfn;
+
+ xen_max_p2m_pfn = max_pfn;
+
+ p2m_missing = extend_brk(PAGE_SIZE, PAGE_SIZE);
+ p2m_init(p2m_missing);
+
+ p2m_mid_missing = extend_brk(PAGE_SIZE, PAGE_SIZE);
+ p2m_mid_init(p2m_mid_missing);
+
+ p2m_top = extend_brk(PAGE_SIZE, PAGE_SIZE);
+ p2m_top_init(p2m_top);
+
+ /*
+ * The domain builder gives us a pre-constructed p2m array in
+ * mfn_list for all the pages initially given to us, so we just
+ * need to graft that into our tree structure.
+ */
+ for (pfn = 0; pfn < max_pfn; pfn += P2M_PER_PAGE) {
+ unsigned topidx = p2m_top_index(pfn);
+ unsigned mididx = p2m_mid_index(pfn);
+
+ if (p2m_top[topidx] == p2m_mid_missing) {
+ unsigned long **mid = extend_brk(PAGE_SIZE, PAGE_SIZE);
+ p2m_mid_init(mid);
+
+ p2m_top[topidx] = mid;
+ }
+
+ p2m_top[topidx][mididx] = &mfn_list[pfn];
+ }
+
+ m2p_override_init();
+}
+
+unsigned long get_phys_to_machine(unsigned long pfn)
+{
+ unsigned topidx, mididx, idx;
+
+ if (unlikely(pfn >= MAX_P2M_PFN))
+ return INVALID_P2M_ENTRY;
+
+ topidx = p2m_top_index(pfn);
+ mididx = p2m_mid_index(pfn);
+ idx = p2m_index(pfn);
+
+ return p2m_top[topidx][mididx][idx];
+}
+EXPORT_SYMBOL_GPL(get_phys_to_machine);
+
+static void *alloc_p2m_page(void)
+{
+ return (void *)__get_free_page(GFP_KERNEL | __GFP_REPEAT);
+}
+
+static void free_p2m_page(void *p)
+{
+ free_page((unsigned long)p);
+}
+
+/*
+ * Fully allocate the p2m structure for a given pfn. We need to check
+ * that both the top and mid levels are allocated, and make sure the
+ * parallel mfn tree is kept in sync. We may race with other cpus, so
+ * the new pages are installed with cmpxchg; if we lose the race then
+ * simply free the page we allocated and use the one that's there.
+ */
+static bool alloc_p2m(unsigned long pfn)
+{
+ unsigned topidx, mididx;
+ unsigned long ***top_p, **mid;
+ unsigned long *top_mfn_p, *mid_mfn;
+
+ topidx = p2m_top_index(pfn);
+ mididx = p2m_mid_index(pfn);
+
+ top_p = &p2m_top[topidx];
+ mid = *top_p;
+
+ if (mid == p2m_mid_missing) {
+ /* Mid level is missing, allocate a new one */
+ mid = alloc_p2m_page();
+ if (!mid)
+ return false;
+
+ p2m_mid_init(mid);
+
+ if (cmpxchg(top_p, p2m_mid_missing, mid) != p2m_mid_missing)
+ free_p2m_page(mid);
+ }
+
+ top_mfn_p = &p2m_top_mfn[topidx];
+ mid_mfn = p2m_top_mfn_p[topidx];
+
+ BUG_ON(virt_to_mfn(mid_mfn) != *top_mfn_p);
+
+ if (mid_mfn == p2m_mid_missing_mfn) {
+ /* Separately check the mid mfn level */
+ unsigned long missing_mfn;
+ unsigned long mid_mfn_mfn;
+
+ mid_mfn = alloc_p2m_page();
+ if (!mid_mfn)
+ return false;
+
+ p2m_mid_mfn_init(mid_mfn);
+
+ missing_mfn = virt_to_mfn(p2m_mid_missing_mfn);
+ mid_mfn_mfn = virt_to_mfn(mid_mfn);
+ if (cmpxchg(top_mfn_p, missing_mfn, mid_mfn_mfn) != missing_mfn)
+ free_p2m_page(mid_mfn);
+ else
+ p2m_top_mfn_p[topidx] = mid_mfn;
+ }
+
+ if (p2m_top[topidx][mididx] == p2m_missing) {
+ /* p2m leaf page is missing */
+ unsigned long *p2m;
+
+ p2m = alloc_p2m_page();
+ if (!p2m)
+ return false;
+
+ p2m_init(p2m);
+
+ if (cmpxchg(&mid[mididx], p2m_missing, p2m) != p2m_missing)
+ free_p2m_page(p2m);
+ else
+ mid_mfn[mididx] = virt_to_mfn(p2m);
+ }
+
+ return true;
+}
+
+/* Try to install p2m mapping; fail if intermediate bits missing */
+bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn)
+{
+ unsigned topidx, mididx, idx;
+
+ if (unlikely(pfn >= MAX_P2M_PFN)) {
+ BUG_ON(mfn != INVALID_P2M_ENTRY);
+ return true;
+ }
+
+ topidx = p2m_top_index(pfn);
+ mididx = p2m_mid_index(pfn);
+ idx = p2m_index(pfn);
+
+ if (p2m_top[topidx][mididx] == p2m_missing)
+ return mfn == INVALID_P2M_ENTRY;
+
+ p2m_top[topidx][mididx][idx] = mfn;
+
+ return true;
+}
+
+bool set_phys_to_machine(unsigned long pfn, unsigned long mfn)
+{
+ if (unlikely(xen_feature(XENFEAT_auto_translated_physmap))) {
+ BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
+ return true;
+ }
+
+ if (unlikely(!__set_phys_to_machine(pfn, mfn))) {
+ if (!alloc_p2m(pfn))
+ return false;
+
+ if (!__set_phys_to_machine(pfn, mfn))
+ return false;
+ }
+
+ return true;
+}
+
+#define M2P_OVERRIDE_HASH_SHIFT 10
+#define M2P_OVERRIDE_HASH (1 << M2P_OVERRIDE_HASH_SHIFT)
+
+static RESERVE_BRK_ARRAY(struct list_head, m2p_overrides, M2P_OVERRIDE_HASH);
+static DEFINE_SPINLOCK(m2p_override_lock);
+
+static void __init m2p_override_init(void)
+{
+ unsigned i;
+
+ m2p_overrides = extend_brk(sizeof(*m2p_overrides) * M2P_OVERRIDE_HASH,
+ sizeof(unsigned long));
+
+ for (i = 0; i < M2P_OVERRIDE_HASH; i++)
+ INIT_LIST_HEAD(&m2p_overrides[i]);
+}
+
+static unsigned long mfn_hash(unsigned long mfn)
+{
+ return hash_long(mfn, M2P_OVERRIDE_HASH_SHIFT);
+}
+
+/* Add an MFN override for a particular page */
+int m2p_add_override(unsigned long mfn, struct page *page)
+{
+ unsigned long flags;
+ unsigned long pfn;
+ unsigned long address;
+ unsigned level;
+ pte_t *ptep = NULL;
+
+ pfn = page_to_pfn(page);
+ if (!PageHighMem(page)) {
+ address = (unsigned long)__va(pfn << PAGE_SHIFT);
+ ptep = lookup_address(address, &level);
+
+ if (WARN(ptep == NULL || level != PG_LEVEL_4K,
+ "m2p_add_override: pfn %lx not mapped", pfn))
+ return -EINVAL;
+ }
+
+ page->private = mfn;
+ page->index = pfn_to_mfn(pfn);
+
+ __set_phys_to_machine(pfn, FOREIGN_FRAME(mfn));
+ if (!PageHighMem(page))
+ /* Just zap old mapping for now */
+ pte_clear(&init_mm, address, ptep);
+
+ spin_lock_irqsave(&m2p_override_lock, flags);
+ list_add(&page->lru, &m2p_overrides[mfn_hash(mfn)]);
+ spin_unlock_irqrestore(&m2p_override_lock, flags);
+
+ return 0;
+}
+
+int m2p_remove_override(struct page *page)
+{
+ unsigned long flags;
+ unsigned long mfn;
+ unsigned long pfn;
+ unsigned long address;
+ unsigned level;
+ pte_t *ptep = NULL;
+
+ pfn = page_to_pfn(page);
+ mfn = get_phys_to_machine(pfn);
+ if (mfn == INVALID_P2M_ENTRY || !(mfn & FOREIGN_FRAME_BIT))
+ return -EINVAL;
+
+ if (!PageHighMem(page)) {
+ address = (unsigned long)__va(pfn << PAGE_SHIFT);
+ ptep = lookup_address(address, &level);
+
+ if (WARN(ptep == NULL || level != PG_LEVEL_4K,
+ "m2p_remove_override: pfn %lx not mapped", pfn))
+ return -EINVAL;
+ }
+
+ spin_lock_irqsave(&m2p_override_lock, flags);
+ list_del(&page->lru);
+ spin_unlock_irqrestore(&m2p_override_lock, flags);
+ __set_phys_to_machine(pfn, page->index);
+
+ if (!PageHighMem(page))
+ set_pte_at(&init_mm, address, ptep,
+ pfn_pte(pfn, PAGE_KERNEL));
+ /* No tlb flush necessary because the caller already
+ * left the pte unmapped. */
+
+ return 0;
+}
+
+struct page *m2p_find_override(unsigned long mfn)
+{
+ unsigned long flags;
+ struct list_head *bucket = &m2p_overrides[mfn_hash(mfn)];
+ struct page *p, *ret;
+
+ ret = NULL;
+
+ spin_lock_irqsave(&m2p_override_lock, flags);
+
+ list_for_each_entry(p, bucket, lru) {
+ if (p->private == mfn) {
+ ret = p;
+ break;
+ }
+ }
+
+ spin_unlock_irqrestore(&m2p_override_lock, flags);
+
+ return ret;
+}
+
+unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn)
+{
+ struct page *p = m2p_find_override(mfn);
+ unsigned long ret = pfn;
+
+ if (p)
+ ret = page_to_pfn(p);
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(m2p_find_override_pfn);
diff --git a/arch/xtensa/include/asm/mman.h b/arch/xtensa/include/asm/mman.h
index fca4db4..3078901 100644
--- a/arch/xtensa/include/asm/mman.h
+++ b/arch/xtensa/include/asm/mman.h
@@ -83,6 +83,9 @@
#define MADV_MERGEABLE 12 /* KSM may merge identical pages */
#define MADV_UNMERGEABLE 13 /* KSM may not merge identical pages */
+#define MADV_HUGEPAGE 14 /* Worth backing with hugepages */
+#define MADV_NOHUGEPAGE 15 /* Not worth backing with hugepages */
+
/* compatibility flags */
#define MAP_FILE 0
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 8427697..501ffdf 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -598,8 +598,8 @@
return cfq_target_latency * cfqg->weight / st->total_weight;
}
-static inline void
-cfq_set_prio_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+static inline unsigned
+cfq_scaled_group_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
{
unsigned slice = cfq_prio_to_slice(cfqd, cfqq);
if (cfqd->cfq_latency) {
@@ -625,6 +625,14 @@
low_slice);
}
}
+ return slice;
+}
+
+static inline void
+cfq_set_prio_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+ unsigned slice = cfq_scaled_group_slice(cfqd, cfqq);
+
cfqq->slice_start = jiffies;
cfqq->slice_end = jiffies + slice;
cfqq->allocated_slice = slice;
@@ -1661,8 +1669,11 @@
/*
* store what was left of this slice, if the queue idled/timed out
*/
- if (timed_out && !cfq_cfqq_slice_new(cfqq)) {
- cfqq->slice_resid = cfqq->slice_end - jiffies;
+ if (timed_out) {
+ if (cfq_cfqq_slice_new(cfqq))
+ cfqq->slice_resid = cfq_scaled_group_slice(cfqd, cfqq);
+ else
+ cfqq->slice_resid = cfqq->slice_end - jiffies;
cfq_log_cfqq(cfqd, cfqq, "resid=%ld", cfqq->slice_resid);
}
@@ -3284,10 +3295,19 @@
*/
static void cfq_preempt_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
{
+ struct cfq_queue *old_cfqq = cfqd->active_queue;
+
cfq_log_cfqq(cfqd, cfqq, "preempt");
cfq_slice_expired(cfqd, 1);
/*
+ * workload type is changed, don't save slice, otherwise preempt
+ * doesn't happen
+ */
+ if (cfqq_type(old_cfqq) != cfqq_type(cfqq))
+ cfqq->cfqg->saved_workload_slice = 0;
+
+ /*
* Put the new queue at the front of the of the current list,
* so we know that it will be selected next.
*/
diff --git a/drivers/Kconfig b/drivers/Kconfig
index dd0a5b5..9bfb71f 100644
--- a/drivers/Kconfig
+++ b/drivers/Kconfig
@@ -26,6 +26,8 @@
source "drivers/md/Kconfig"
+source "drivers/target/Kconfig"
+
source "drivers/message/fusion/Kconfig"
source "drivers/firewire/Kconfig"
diff --git a/drivers/Makefile b/drivers/Makefile
index ef51324..7eb35f4 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -46,6 +46,7 @@
obj-$(CONFIG_IDE) += ide/
obj-$(CONFIG_SCSI) += scsi/
obj-$(CONFIG_ATA) += ata/
+obj-$(CONFIG_TARGET_CORE) += target/
obj-$(CONFIG_MTD) += mtd/
obj-$(CONFIG_SPI) += spi/
obj-y += net/
diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
index 3f3489c..10c7ad5 100644
--- a/drivers/acpi/Kconfig
+++ b/drivers/acpi/Kconfig
@@ -51,12 +51,7 @@
For backwards compatibility, this option allows
deprecated /proc/acpi/ files to exist, even when
they have been replaced by functions in /sys.
- The deprecated files (and their replacements) include:
- /proc/acpi/processor/*/throttling (/sys/class/thermal/
- cooling_device*/*)
- /proc/acpi/video/*/brightness (/sys/class/backlight/)
- /proc/acpi/thermal_zone/*/* (/sys/class/thermal/)
This option has no effect on /proc/acpi/ files
and functions which do not yet exist in /sys.
@@ -74,6 +69,8 @@
/proc/acpi/ac_adapter/* (sys/class/power_supply/*)
This option has no effect on /proc/acpi/ directories
and functions, which do not yet exist in /sys
+ This option, together with the proc directories, will be
+ deleted in 2.6.39.
Say N to delete power /proc/acpi/ directories that have moved to /sys/
@@ -209,6 +206,17 @@
To compile this driver as a module, choose M here:
the module will be called processor.
+config ACPI_IPMI
+ tristate "IPMI"
+ depends on EXPERIMENTAL && IPMI_SI && IPMI_HANDLER
+ default n
+ help
+ This driver enables the ACPI to access the BMC controller. And it
+ uses the IPMI request/response message to communicate with BMC
+ controller, which can be found on on the server.
+
+ To compile this driver as a module, choose M here:
+ the module will be called as acpi_ipmi.
config ACPI_HOTPLUG_CPU
bool
diff --git a/drivers/acpi/Makefile b/drivers/acpi/Makefile
index 3d031d0..d113fa5 100644
--- a/drivers/acpi/Makefile
+++ b/drivers/acpi/Makefile
@@ -24,7 +24,7 @@
# sleep related files
acpi-y += wakeup.o
acpi-y += sleep.o
-acpi-$(CONFIG_ACPI_SLEEP) += proc.o
+acpi-$(CONFIG_ACPI_SLEEP) += proc.o nvs.o
#
@@ -69,5 +69,6 @@
processor-$(CONFIG_CPU_FREQ) += processor_perflib.o
obj-$(CONFIG_ACPI_PROCESSOR_AGGREGATOR) += acpi_pad.o
+obj-$(CONFIG_ACPI_IPMI) += acpi_ipmi.o
obj-$(CONFIG_ACPI_APEI) += apei/
diff --git a/drivers/acpi/ac.c b/drivers/acpi/ac.c
index 25d3aae..58c3f74 100644
--- a/drivers/acpi/ac.c
+++ b/drivers/acpi/ac.c
@@ -197,7 +197,8 @@
{
struct proc_dir_entry *entry = NULL;
-
+ printk(KERN_WARNING PREFIX "Deprecated procfs I/F for AC is loaded,"
+ " please retry with CONFIG_ACPI_PROCFS_POWER cleared\n");
if (!acpi_device_dir(device)) {
acpi_device_dir(device) = proc_mkdir(acpi_device_bid(device),
acpi_ac_dir);
diff --git a/drivers/acpi/acpi_ipmi.c b/drivers/acpi/acpi_ipmi.c
new file mode 100644
index 0000000..f40acef
--- /dev/null
+++ b/drivers/acpi/acpi_ipmi.c
@@ -0,0 +1,525 @@
+/*
+ * acpi_ipmi.c - ACPI IPMI opregion
+ *
+ * Copyright (C) 2010 Intel Corporation
+ * Copyright (C) 2010 Zhao Yakui <yakui.zhao@intel.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.
+ *
+ * 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, write to the Free Software Foundation, Inc.,
+ * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/delay.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/io.h>
+#include <acpi/acpi_bus.h>
+#include <acpi/acpi_drivers.h>
+#include <linux/ipmi.h>
+#include <linux/device.h>
+#include <linux/pnp.h>
+
+MODULE_AUTHOR("Zhao Yakui");
+MODULE_DESCRIPTION("ACPI IPMI Opregion driver");
+MODULE_LICENSE("GPL");
+
+#define IPMI_FLAGS_HANDLER_INSTALL 0
+
+#define ACPI_IPMI_OK 0
+#define ACPI_IPMI_TIMEOUT 0x10
+#define ACPI_IPMI_UNKNOWN 0x07
+/* the IPMI timeout is 5s */
+#define IPMI_TIMEOUT (5 * HZ)
+
+struct acpi_ipmi_device {
+ /* the device list attached to driver_data.ipmi_devices */
+ struct list_head head;
+ /* the IPMI request message list */
+ struct list_head tx_msg_list;
+ struct mutex tx_msg_lock;
+ acpi_handle handle;
+ struct pnp_dev *pnp_dev;
+ ipmi_user_t user_interface;
+ int ipmi_ifnum; /* IPMI interface number */
+ long curr_msgid;
+ unsigned long flags;
+ struct ipmi_smi_info smi_data;
+};
+
+struct ipmi_driver_data {
+ struct list_head ipmi_devices;
+ struct ipmi_smi_watcher bmc_events;
+ struct ipmi_user_hndl ipmi_hndlrs;
+ struct mutex ipmi_lock;
+};
+
+struct acpi_ipmi_msg {
+ struct list_head head;
+ /*
+ * General speaking the addr type should be SI_ADDR_TYPE. And
+ * the addr channel should be BMC.
+ * In fact it can also be IPMB type. But we will have to
+ * parse it from the Netfn command buffer. It is so complex
+ * that it is skipped.
+ */
+ struct ipmi_addr addr;
+ long tx_msgid;
+ /* it is used to track whether the IPMI message is finished */
+ struct completion tx_complete;
+ struct kernel_ipmi_msg tx_message;
+ int msg_done;
+ /* tx data . And copy it from ACPI object buffer */
+ u8 tx_data[64];
+ int tx_len;
+ u8 rx_data[64];
+ int rx_len;
+ struct acpi_ipmi_device *device;
+};
+
+/* IPMI request/response buffer per ACPI 4.0, sec 5.5.2.4.3.2 */
+struct acpi_ipmi_buffer {
+ u8 status;
+ u8 length;
+ u8 data[64];
+};
+
+static void ipmi_register_bmc(int iface, struct device *dev);
+static void ipmi_bmc_gone(int iface);
+static void ipmi_msg_handler(struct ipmi_recv_msg *msg, void *user_msg_data);
+static void acpi_add_ipmi_device(struct acpi_ipmi_device *ipmi_device);
+static void acpi_remove_ipmi_device(struct acpi_ipmi_device *ipmi_device);
+
+static struct ipmi_driver_data driver_data = {
+ .ipmi_devices = LIST_HEAD_INIT(driver_data.ipmi_devices),
+ .bmc_events = {
+ .owner = THIS_MODULE,
+ .new_smi = ipmi_register_bmc,
+ .smi_gone = ipmi_bmc_gone,
+ },
+ .ipmi_hndlrs = {
+ .ipmi_recv_hndl = ipmi_msg_handler,
+ },
+};
+
+static struct acpi_ipmi_msg *acpi_alloc_ipmi_msg(struct acpi_ipmi_device *ipmi)
+{
+ struct acpi_ipmi_msg *ipmi_msg;
+ struct pnp_dev *pnp_dev = ipmi->pnp_dev;
+
+ ipmi_msg = kzalloc(sizeof(struct acpi_ipmi_msg), GFP_KERNEL);
+ if (!ipmi_msg) {
+ dev_warn(&pnp_dev->dev, "Can't allocate memory for ipmi_msg\n");
+ return NULL;
+ }
+ init_completion(&ipmi_msg->tx_complete);
+ INIT_LIST_HEAD(&ipmi_msg->head);
+ ipmi_msg->device = ipmi;
+ return ipmi_msg;
+}
+
+#define IPMI_OP_RGN_NETFN(offset) ((offset >> 8) & 0xff)
+#define IPMI_OP_RGN_CMD(offset) (offset & 0xff)
+static void acpi_format_ipmi_msg(struct acpi_ipmi_msg *tx_msg,
+ acpi_physical_address address,
+ acpi_integer *value)
+{
+ struct kernel_ipmi_msg *msg;
+ struct acpi_ipmi_buffer *buffer;
+ struct acpi_ipmi_device *device;
+
+ msg = &tx_msg->tx_message;
+ /*
+ * IPMI network function and command are encoded in the address
+ * within the IPMI OpRegion; see ACPI 4.0, sec 5.5.2.4.3.
+ */
+ msg->netfn = IPMI_OP_RGN_NETFN(address);
+ msg->cmd = IPMI_OP_RGN_CMD(address);
+ msg->data = tx_msg->tx_data;
+ /*
+ * value is the parameter passed by the IPMI opregion space handler.
+ * It points to the IPMI request message buffer
+ */
+ buffer = (struct acpi_ipmi_buffer *)value;
+ /* copy the tx message data */
+ msg->data_len = buffer->length;
+ memcpy(tx_msg->tx_data, buffer->data, msg->data_len);
+ /*
+ * now the default type is SYSTEM_INTERFACE and channel type is BMC.
+ * If the netfn is APP_REQUEST and the cmd is SEND_MESSAGE,
+ * the addr type should be changed to IPMB. Then we will have to parse
+ * the IPMI request message buffer to get the IPMB address.
+ * If so, please fix me.
+ */
+ tx_msg->addr.addr_type = IPMI_SYSTEM_INTERFACE_ADDR_TYPE;
+ tx_msg->addr.channel = IPMI_BMC_CHANNEL;
+ tx_msg->addr.data[0] = 0;
+
+ /* Get the msgid */
+ device = tx_msg->device;
+ mutex_lock(&device->tx_msg_lock);
+ device->curr_msgid++;
+ tx_msg->tx_msgid = device->curr_msgid;
+ mutex_unlock(&device->tx_msg_lock);
+}
+
+static void acpi_format_ipmi_response(struct acpi_ipmi_msg *msg,
+ acpi_integer *value, int rem_time)
+{
+ struct acpi_ipmi_buffer *buffer;
+
+ /*
+ * value is also used as output parameter. It represents the response
+ * IPMI message returned by IPMI command.
+ */
+ buffer = (struct acpi_ipmi_buffer *)value;
+ if (!rem_time && !msg->msg_done) {
+ buffer->status = ACPI_IPMI_TIMEOUT;
+ return;
+ }
+ /*
+ * If the flag of msg_done is not set or the recv length is zero, it
+ * means that the IPMI command is not executed correctly.
+ * The status code will be ACPI_IPMI_UNKNOWN.
+ */
+ if (!msg->msg_done || !msg->rx_len) {
+ buffer->status = ACPI_IPMI_UNKNOWN;
+ return;
+ }
+ /*
+ * If the IPMI response message is obtained correctly, the status code
+ * will be ACPI_IPMI_OK
+ */
+ buffer->status = ACPI_IPMI_OK;
+ buffer->length = msg->rx_len;
+ memcpy(buffer->data, msg->rx_data, msg->rx_len);
+}
+
+static void ipmi_flush_tx_msg(struct acpi_ipmi_device *ipmi)
+{
+ struct acpi_ipmi_msg *tx_msg, *temp;
+ int count = HZ / 10;
+ struct pnp_dev *pnp_dev = ipmi->pnp_dev;
+
+ list_for_each_entry_safe(tx_msg, temp, &ipmi->tx_msg_list, head) {
+ /* wake up the sleep thread on the Tx msg */
+ complete(&tx_msg->tx_complete);
+ }
+
+ /* wait for about 100ms to flush the tx message list */
+ while (count--) {
+ if (list_empty(&ipmi->tx_msg_list))
+ break;
+ schedule_timeout(1);
+ }
+ if (!list_empty(&ipmi->tx_msg_list))
+ dev_warn(&pnp_dev->dev, "tx msg list is not NULL\n");
+}
+
+static void ipmi_msg_handler(struct ipmi_recv_msg *msg, void *user_msg_data)
+{
+ struct acpi_ipmi_device *ipmi_device = user_msg_data;
+ int msg_found = 0;
+ struct acpi_ipmi_msg *tx_msg;
+ struct pnp_dev *pnp_dev = ipmi_device->pnp_dev;
+
+ if (msg->user != ipmi_device->user_interface) {
+ dev_warn(&pnp_dev->dev, "Unexpected response is returned. "
+ "returned user %p, expected user %p\n",
+ msg->user, ipmi_device->user_interface);
+ ipmi_free_recv_msg(msg);
+ return;
+ }
+ mutex_lock(&ipmi_device->tx_msg_lock);
+ list_for_each_entry(tx_msg, &ipmi_device->tx_msg_list, head) {
+ if (msg->msgid == tx_msg->tx_msgid) {
+ msg_found = 1;
+ break;
+ }
+ }
+
+ mutex_unlock(&ipmi_device->tx_msg_lock);
+ if (!msg_found) {
+ dev_warn(&pnp_dev->dev, "Unexpected response (msg id %ld) is "
+ "returned.\n", msg->msgid);
+ ipmi_free_recv_msg(msg);
+ return;
+ }
+
+ if (msg->msg.data_len) {
+ /* copy the response data to Rx_data buffer */
+ memcpy(tx_msg->rx_data, msg->msg_data, msg->msg.data_len);
+ tx_msg->rx_len = msg->msg.data_len;
+ tx_msg->msg_done = 1;
+ }
+ complete(&tx_msg->tx_complete);
+ ipmi_free_recv_msg(msg);
+};
+
+static void ipmi_register_bmc(int iface, struct device *dev)
+{
+ struct acpi_ipmi_device *ipmi_device, *temp;
+ struct pnp_dev *pnp_dev;
+ ipmi_user_t user;
+ int err;
+ struct ipmi_smi_info smi_data;
+ acpi_handle handle;
+
+ err = ipmi_get_smi_info(iface, &smi_data);
+
+ if (err)
+ return;
+
+ if (smi_data.addr_src != SI_ACPI) {
+ put_device(smi_data.dev);
+ return;
+ }
+
+ handle = smi_data.addr_info.acpi_info.acpi_handle;
+
+ mutex_lock(&driver_data.ipmi_lock);
+ list_for_each_entry(temp, &driver_data.ipmi_devices, head) {
+ /*
+ * if the corresponding ACPI handle is already added
+ * to the device list, don't add it again.
+ */
+ if (temp->handle == handle)
+ goto out;
+ }
+
+ ipmi_device = kzalloc(sizeof(*ipmi_device), GFP_KERNEL);
+
+ if (!ipmi_device)
+ goto out;
+
+ pnp_dev = to_pnp_dev(smi_data.dev);
+ ipmi_device->handle = handle;
+ ipmi_device->pnp_dev = pnp_dev;
+
+ err = ipmi_create_user(iface, &driver_data.ipmi_hndlrs,
+ ipmi_device, &user);
+ if (err) {
+ dev_warn(&pnp_dev->dev, "Can't create IPMI user interface\n");
+ kfree(ipmi_device);
+ goto out;
+ }
+ acpi_add_ipmi_device(ipmi_device);
+ ipmi_device->user_interface = user;
+ ipmi_device->ipmi_ifnum = iface;
+ mutex_unlock(&driver_data.ipmi_lock);
+ memcpy(&ipmi_device->smi_data, &smi_data, sizeof(struct ipmi_smi_info));
+ return;
+
+out:
+ mutex_unlock(&driver_data.ipmi_lock);
+ put_device(smi_data.dev);
+ return;
+}
+
+static void ipmi_bmc_gone(int iface)
+{
+ struct acpi_ipmi_device *ipmi_device, *temp;
+
+ mutex_lock(&driver_data.ipmi_lock);
+ list_for_each_entry_safe(ipmi_device, temp,
+ &driver_data.ipmi_devices, head) {
+ if (ipmi_device->ipmi_ifnum != iface)
+ continue;
+
+ acpi_remove_ipmi_device(ipmi_device);
+ put_device(ipmi_device->smi_data.dev);
+ kfree(ipmi_device);
+ break;
+ }
+ mutex_unlock(&driver_data.ipmi_lock);
+}
+/* --------------------------------------------------------------------------
+ * Address Space Management
+ * -------------------------------------------------------------------------- */
+/*
+ * This is the IPMI opregion space handler.
+ * @function: indicates the read/write. In fact as the IPMI message is driven
+ * by command, only write is meaningful.
+ * @address: This contains the netfn/command of IPMI request message.
+ * @bits : not used.
+ * @value : it is an in/out parameter. It points to the IPMI message buffer.
+ * Before the IPMI message is sent, it represents the actual request
+ * IPMI message. After the IPMI message is finished, it represents
+ * the response IPMI message returned by IPMI command.
+ * @handler_context: IPMI device context.
+ */
+
+static acpi_status
+acpi_ipmi_space_handler(u32 function, acpi_physical_address address,
+ u32 bits, acpi_integer *value,
+ void *handler_context, void *region_context)
+{
+ struct acpi_ipmi_msg *tx_msg;
+ struct acpi_ipmi_device *ipmi_device = handler_context;
+ int err, rem_time;
+ acpi_status status;
+ /*
+ * IPMI opregion message.
+ * IPMI message is firstly written to the BMC and system software
+ * can get the respsonse. So it is unmeaningful for the read access
+ * of IPMI opregion.
+ */
+ if ((function & ACPI_IO_MASK) == ACPI_READ)
+ return AE_TYPE;
+
+ if (!ipmi_device->user_interface)
+ return AE_NOT_EXIST;
+
+ tx_msg = acpi_alloc_ipmi_msg(ipmi_device);
+ if (!tx_msg)
+ return AE_NO_MEMORY;
+
+ acpi_format_ipmi_msg(tx_msg, address, value);
+ mutex_lock(&ipmi_device->tx_msg_lock);
+ list_add_tail(&tx_msg->head, &ipmi_device->tx_msg_list);
+ mutex_unlock(&ipmi_device->tx_msg_lock);
+ err = ipmi_request_settime(ipmi_device->user_interface,
+ &tx_msg->addr,
+ tx_msg->tx_msgid,
+ &tx_msg->tx_message,
+ NULL, 0, 0, 0);
+ if (err) {
+ status = AE_ERROR;
+ goto end_label;
+ }
+ rem_time = wait_for_completion_timeout(&tx_msg->tx_complete,
+ IPMI_TIMEOUT);
+ acpi_format_ipmi_response(tx_msg, value, rem_time);
+ status = AE_OK;
+
+end_label:
+ mutex_lock(&ipmi_device->tx_msg_lock);
+ list_del(&tx_msg->head);
+ mutex_unlock(&ipmi_device->tx_msg_lock);
+ kfree(tx_msg);
+ return status;
+}
+
+static void ipmi_remove_space_handler(struct acpi_ipmi_device *ipmi)
+{
+ if (!test_bit(IPMI_FLAGS_HANDLER_INSTALL, &ipmi->flags))
+ return;
+
+ acpi_remove_address_space_handler(ipmi->handle,
+ ACPI_ADR_SPACE_IPMI, &acpi_ipmi_space_handler);
+
+ clear_bit(IPMI_FLAGS_HANDLER_INSTALL, &ipmi->flags);
+}
+
+static int ipmi_install_space_handler(struct acpi_ipmi_device *ipmi)
+{
+ acpi_status status;
+
+ if (test_bit(IPMI_FLAGS_HANDLER_INSTALL, &ipmi->flags))
+ return 0;
+
+ status = acpi_install_address_space_handler(ipmi->handle,
+ ACPI_ADR_SPACE_IPMI,
+ &acpi_ipmi_space_handler,
+ NULL, ipmi);
+ if (ACPI_FAILURE(status)) {
+ struct pnp_dev *pnp_dev = ipmi->pnp_dev;
+ dev_warn(&pnp_dev->dev, "Can't register IPMI opregion space "
+ "handle\n");
+ return -EINVAL;
+ }
+ set_bit(IPMI_FLAGS_HANDLER_INSTALL, &ipmi->flags);
+ return 0;
+}
+
+static void acpi_add_ipmi_device(struct acpi_ipmi_device *ipmi_device)
+{
+
+ INIT_LIST_HEAD(&ipmi_device->head);
+
+ mutex_init(&ipmi_device->tx_msg_lock);
+ INIT_LIST_HEAD(&ipmi_device->tx_msg_list);
+ ipmi_install_space_handler(ipmi_device);
+
+ list_add_tail(&ipmi_device->head, &driver_data.ipmi_devices);
+}
+
+static void acpi_remove_ipmi_device(struct acpi_ipmi_device *ipmi_device)
+{
+ /*
+ * If the IPMI user interface is created, it should be
+ * destroyed.
+ */
+ if (ipmi_device->user_interface) {
+ ipmi_destroy_user(ipmi_device->user_interface);
+ ipmi_device->user_interface = NULL;
+ }
+ /* flush the Tx_msg list */
+ if (!list_empty(&ipmi_device->tx_msg_list))
+ ipmi_flush_tx_msg(ipmi_device);
+
+ list_del(&ipmi_device->head);
+ ipmi_remove_space_handler(ipmi_device);
+}
+
+static int __init acpi_ipmi_init(void)
+{
+ int result = 0;
+
+ if (acpi_disabled)
+ return result;
+
+ mutex_init(&driver_data.ipmi_lock);
+
+ result = ipmi_smi_watcher_register(&driver_data.bmc_events);
+
+ return result;
+}
+
+static void __exit acpi_ipmi_exit(void)
+{
+ struct acpi_ipmi_device *ipmi_device, *temp;
+
+ if (acpi_disabled)
+ return;
+
+ ipmi_smi_watcher_unregister(&driver_data.bmc_events);
+
+ /*
+ * When one smi_watcher is unregistered, it is only deleted
+ * from the smi_watcher list. But the smi_gone callback function
+ * is not called. So explicitly uninstall the ACPI IPMI oregion
+ * handler and free it.
+ */
+ mutex_lock(&driver_data.ipmi_lock);
+ list_for_each_entry_safe(ipmi_device, temp,
+ &driver_data.ipmi_devices, head) {
+ acpi_remove_ipmi_device(ipmi_device);
+ put_device(ipmi_device->smi_data.dev);
+ kfree(ipmi_device);
+ }
+ mutex_unlock(&driver_data.ipmi_lock);
+}
+
+module_init(acpi_ipmi_init);
+module_exit(acpi_ipmi_exit);
diff --git a/drivers/acpi/acpica/Makefile b/drivers/acpi/acpica/Makefile
index a7e1d1a..eec2ead 100644
--- a/drivers/acpi/acpica/Makefile
+++ b/drivers/acpi/acpica/Makefile
@@ -14,7 +14,7 @@
acpi-y += evevent.o evregion.o evsci.o evxfevnt.o \
evmisc.o evrgnini.o evxface.o evxfregn.o \
- evgpe.o evgpeblk.o evgpeinit.o evgpeutil.o
+ evgpe.o evgpeblk.o evgpeinit.o evgpeutil.o evxfgpe.o
acpi-y += exconfig.o exfield.o exnames.o exoparg6.o exresolv.o exstorob.o\
exconvrt.o exfldio.o exoparg1.o exprep.o exresop.o exsystem.o\
diff --git a/drivers/acpi/acpica/acevents.h b/drivers/acpi/acpica/acevents.h
index a6f99cc..70e0b28 100644
--- a/drivers/acpi/acpica/acevents.h
+++ b/drivers/acpi/acpica/acevents.h
@@ -51,8 +51,6 @@
acpi_status acpi_ev_install_xrupt_handlers(void);
-acpi_status acpi_ev_install_fadt_gpes(void);
-
u32 acpi_ev_fixed_event_detect(void);
/*
@@ -82,9 +80,9 @@
acpi_status acpi_ev_enable_gpe(struct acpi_gpe_event_info *gpe_event_info);
-acpi_status acpi_raw_enable_gpe(struct acpi_gpe_event_info *gpe_event_info);
+acpi_status acpi_ev_add_gpe_reference(struct acpi_gpe_event_info *gpe_event_info);
-acpi_status acpi_raw_disable_gpe(struct acpi_gpe_event_info *gpe_event_info);
+acpi_status acpi_ev_remove_gpe_reference(struct acpi_gpe_event_info *gpe_event_info);
struct acpi_gpe_event_info *acpi_ev_get_gpe_event_info(acpi_handle gpe_device,
u32 gpe_number);
@@ -93,6 +91,8 @@
struct acpi_gpe_block_info
*gpe_block);
+acpi_status acpi_ev_finish_gpe(struct acpi_gpe_event_info *gpe_event_info);
+
/*
* evgpeblk - Upper-level GPE block support
*/
@@ -107,12 +107,13 @@
acpi_status
acpi_ev_initialize_gpe_block(struct acpi_gpe_xrupt_info *gpe_xrupt_info,
struct acpi_gpe_block_info *gpe_block,
- void *ignored);
+ void *context);
acpi_status acpi_ev_delete_gpe_block(struct acpi_gpe_block_info *gpe_block);
u32
-acpi_ev_gpe_dispatch(struct acpi_gpe_event_info *gpe_event_info,
+acpi_ev_gpe_dispatch(struct acpi_namespace_node *gpe_device,
+ struct acpi_gpe_event_info *gpe_event_info,
u32 gpe_number);
/*
@@ -126,10 +127,6 @@
acpi_ev_match_gpe_method(acpi_handle obj_handle,
u32 level, void *context, void **return_value);
-acpi_status
-acpi_ev_match_prw_and_gpe(acpi_handle obj_handle,
- u32 level, void *context, void **return_value);
-
/*
* evgpeutil - GPE utilities
*/
@@ -138,6 +135,10 @@
u8 acpi_ev_valid_gpe_event(struct acpi_gpe_event_info *gpe_event_info);
+acpi_status
+acpi_ev_get_gpe_device(struct acpi_gpe_xrupt_info *gpe_xrupt_info,
+ struct acpi_gpe_block_info *gpe_block, void *context);
+
struct acpi_gpe_xrupt_info *acpi_ev_get_gpe_xrupt_block(u32 interrupt_number);
acpi_status acpi_ev_delete_gpe_xrupt(struct acpi_gpe_xrupt_info *gpe_xrupt);
diff --git a/drivers/acpi/acpica/acglobal.h b/drivers/acpi/acpica/acglobal.h
index ad88fca..0e4dba0 100644
--- a/drivers/acpi/acpica/acglobal.h
+++ b/drivers/acpi/acpica/acglobal.h
@@ -146,6 +146,9 @@
extern u32 acpi_gbl_nesting_level;
+ACPI_EXTERN u32 acpi_gpe_count;
+ACPI_EXTERN u32 acpi_fixed_event_count[ACPI_NUM_FIXED_EVENTS];
+
/* Support for dynamic control method tracing mechanism */
ACPI_EXTERN u32 acpi_gbl_original_dbg_level;
@@ -225,8 +228,10 @@
*/
ACPI_EXTERN spinlock_t _acpi_gbl_gpe_lock; /* For GPE data structs and registers */
ACPI_EXTERN spinlock_t _acpi_gbl_hardware_lock; /* For ACPI H/W except GPE registers */
+ACPI_EXTERN spinlock_t _acpi_ev_global_lock_pending_lock; /* For global lock */
#define acpi_gbl_gpe_lock &_acpi_gbl_gpe_lock
#define acpi_gbl_hardware_lock &_acpi_gbl_hardware_lock
+#define acpi_ev_global_lock_pending_lock &_acpi_ev_global_lock_pending_lock
/*****************************************************************************
*
@@ -370,7 +375,9 @@
ACPI_EXTERN struct acpi_gpe_xrupt_info *acpi_gbl_gpe_xrupt_list_head;
ACPI_EXTERN struct acpi_gpe_block_info
*acpi_gbl_gpe_fadt_blocks[ACPI_MAX_GPE_BLOCKS];
-ACPI_EXTERN u8 acpi_all_gpes_initialized;
+ACPI_EXTERN u8 acpi_gbl_all_gpes_initialized;
+ACPI_EXTERN ACPI_GBL_EVENT_HANDLER acpi_gbl_global_event_handler;
+ACPI_EXTERN void *acpi_gbl_global_event_handler_context;
/*****************************************************************************
*
diff --git a/drivers/acpi/acpica/achware.h b/drivers/acpi/acpica/achware.h
index 167470a..258d628 100644
--- a/drivers/acpi/acpica/achware.h
+++ b/drivers/acpi/acpica/achware.h
@@ -94,7 +94,7 @@
struct acpi_gpe_register_info *gpe_register_info);
acpi_status
-acpi_hw_low_set_gpe(struct acpi_gpe_event_info *gpe_event_info, u8 action);
+acpi_hw_low_set_gpe(struct acpi_gpe_event_info *gpe_event_info, u32 action);
acpi_status
acpi_hw_disable_gpe_block(struct acpi_gpe_xrupt_info *gpe_xrupt_info,
diff --git a/drivers/acpi/acpica/aclocal.h b/drivers/acpi/acpica/aclocal.h
index 2ceb0c0..74000f5 100644
--- a/drivers/acpi/acpica/aclocal.h
+++ b/drivers/acpi/acpica/aclocal.h
@@ -408,17 +408,18 @@
/* Dispatch info for each GPE -- either a method or handler, cannot be both */
-struct acpi_handler_info {
- acpi_event_handler address; /* Address of handler, if any */
+struct acpi_gpe_handler_info {
+ acpi_gpe_handler address; /* Address of handler, if any */
void *context; /* Context to be passed to handler */
struct acpi_namespace_node *method_node; /* Method node for this GPE level (saved) */
- u8 orig_flags; /* Original misc info about this GPE */
- u8 orig_enabled; /* Set if the GPE was originally enabled */
+ u8 original_flags; /* Original (pre-handler) GPE info */
+ u8 originally_enabled; /* True if GPE was originally enabled */
};
union acpi_gpe_dispatch_info {
struct acpi_namespace_node *method_node; /* Method node for this GPE level */
- struct acpi_handler_info *handler;
+ struct acpi_gpe_handler_info *handler; /* Installed GPE handler */
+ struct acpi_namespace_node *device_node; /* Parent _PRW device for implicit notify */
};
/*
@@ -458,7 +459,7 @@
u32 register_count; /* Number of register pairs in block */
u16 gpe_count; /* Number of individual GPEs in block */
u8 block_base_number; /* Base GPE number for this block */
- u8 initialized; /* If set, the GPE block has been initialized */
+ u8 initialized; /* TRUE if this block is initialized */
};
/* Information about GPE interrupt handlers, one per each interrupt level used for GPEs */
diff --git a/drivers/acpi/acpica/evevent.c b/drivers/acpi/acpica/evevent.c
index c61c3039..e5e313c 100644
--- a/drivers/acpi/acpica/evevent.c
+++ b/drivers/acpi/acpica/evevent.c
@@ -217,9 +217,17 @@
status_bit_mask)
&& (fixed_enable & acpi_gbl_fixed_event_info[i].
enable_bit_mask)) {
+ /*
+ * Found an active (signalled) event. Invoke global event
+ * handler if present.
+ */
+ acpi_fixed_event_count[i]++;
+ if (acpi_gbl_global_event_handler) {
+ acpi_gbl_global_event_handler
+ (ACPI_EVENT_TYPE_FIXED, NULL, i,
+ acpi_gbl_global_event_handler_context);
+ }
- /* Found an active (signalled) event */
- acpi_os_fixed_event_count(i);
int_status |= acpi_ev_fixed_event_dispatch(i);
}
}
diff --git a/drivers/acpi/acpica/evgpe.c b/drivers/acpi/acpica/evgpe.c
index f226eac..7c339d3 100644
--- a/drivers/acpi/acpica/evgpe.c
+++ b/drivers/acpi/acpica/evgpe.c
@@ -52,6 +52,8 @@
/* Local prototypes */
static void ACPI_SYSTEM_XFACE acpi_ev_asynch_execute_gpe_method(void *context);
+static void ACPI_SYSTEM_XFACE acpi_ev_asynch_enable_gpe(void *context);
+
/*******************************************************************************
*
* FUNCTION: acpi_ev_update_gpe_enable_mask
@@ -102,7 +104,7 @@
*
* RETURN: Status
*
- * DESCRIPTION: Clear the given GPE from stale events and enable it.
+ * DESCRIPTION: Clear a GPE of stale events and enable it.
*
******************************************************************************/
acpi_status
@@ -113,12 +115,13 @@
ACPI_FUNCTION_TRACE(ev_enable_gpe);
/*
- * We will only allow a GPE to be enabled if it has either an
- * associated method (_Lxx/_Exx) or a handler. Otherwise, the
- * GPE will be immediately disabled by acpi_ev_gpe_dispatch the
- * first time it fires.
+ * We will only allow a GPE to be enabled if it has either an associated
+ * method (_Lxx/_Exx) or a handler, or is using the implicit notify
+ * feature. Otherwise, the GPE will be immediately disabled by
+ * acpi_ev_gpe_dispatch the first time it fires.
*/
- if (!(gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK)) {
+ if ((gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK) ==
+ ACPI_GPE_DISPATCH_NONE) {
return_ACPI_STATUS(AE_NO_HANDLER);
}
@@ -137,9 +140,9 @@
/*******************************************************************************
*
- * FUNCTION: acpi_raw_enable_gpe
+ * FUNCTION: acpi_ev_add_gpe_reference
*
- * PARAMETERS: gpe_event_info - GPE to enable
+ * PARAMETERS: gpe_event_info - Add a reference to this GPE
*
* RETURN: Status
*
@@ -148,16 +151,21 @@
*
******************************************************************************/
-acpi_status acpi_raw_enable_gpe(struct acpi_gpe_event_info *gpe_event_info)
+acpi_status acpi_ev_add_gpe_reference(struct acpi_gpe_event_info *gpe_event_info)
{
acpi_status status = AE_OK;
+ ACPI_FUNCTION_TRACE(ev_add_gpe_reference);
+
if (gpe_event_info->runtime_count == ACPI_UINT8_MAX) {
return_ACPI_STATUS(AE_LIMIT);
}
gpe_event_info->runtime_count++;
if (gpe_event_info->runtime_count == 1) {
+
+ /* Enable on first reference */
+
status = acpi_ev_update_gpe_enable_mask(gpe_event_info);
if (ACPI_SUCCESS(status)) {
status = acpi_ev_enable_gpe(gpe_event_info);
@@ -173,9 +181,9 @@
/*******************************************************************************
*
- * FUNCTION: acpi_raw_disable_gpe
+ * FUNCTION: acpi_ev_remove_gpe_reference
*
- * PARAMETERS: gpe_event_info - GPE to disable
+ * PARAMETERS: gpe_event_info - Remove a reference to this GPE
*
* RETURN: Status
*
@@ -184,16 +192,21 @@
*
******************************************************************************/
-acpi_status acpi_raw_disable_gpe(struct acpi_gpe_event_info *gpe_event_info)
+acpi_status acpi_ev_remove_gpe_reference(struct acpi_gpe_event_info *gpe_event_info)
{
acpi_status status = AE_OK;
+ ACPI_FUNCTION_TRACE(ev_remove_gpe_reference);
+
if (!gpe_event_info->runtime_count) {
return_ACPI_STATUS(AE_LIMIT);
}
gpe_event_info->runtime_count--;
if (!gpe_event_info->runtime_count) {
+
+ /* Disable on last reference */
+
status = acpi_ev_update_gpe_enable_mask(gpe_event_info);
if (ACPI_SUCCESS(status)) {
status = acpi_hw_low_set_gpe(gpe_event_info,
@@ -379,7 +392,7 @@
}
ACPI_DEBUG_PRINT((ACPI_DB_INTERRUPTS,
- "Read GPE Register at GPE%X: Status=%02X, Enable=%02X\n",
+ "Read GPE Register at GPE%02X: Status=%02X, Enable=%02X\n",
gpe_register_info->base_gpe_number,
status_reg, enable_reg));
@@ -405,7 +418,9 @@
* or method.
*/
int_status |=
- acpi_ev_gpe_dispatch(&gpe_block->
+ acpi_ev_gpe_dispatch(gpe_block->
+ node,
+ &gpe_block->
event_info[((acpi_size) i * ACPI_GPE_REGISTER_WIDTH) + j], j + gpe_register_info->base_gpe_number);
}
}
@@ -435,17 +450,25 @@
* an interrupt handler.
*
******************************************************************************/
-static void acpi_ev_asynch_enable_gpe(void *context);
static void ACPI_SYSTEM_XFACE acpi_ev_asynch_execute_gpe_method(void *context)
{
- struct acpi_gpe_event_info *gpe_event_info = (void *)context;
+ struct acpi_gpe_event_info *gpe_event_info = context;
acpi_status status;
- struct acpi_gpe_event_info local_gpe_event_info;
+ struct acpi_gpe_event_info *local_gpe_event_info;
struct acpi_evaluate_info *info;
ACPI_FUNCTION_TRACE(ev_asynch_execute_gpe_method);
+ /* Allocate a local GPE block */
+
+ local_gpe_event_info =
+ ACPI_ALLOCATE_ZEROED(sizeof(struct acpi_gpe_event_info));
+ if (!local_gpe_event_info) {
+ ACPI_EXCEPTION((AE_INFO, AE_NO_MEMORY, "while handling a GPE"));
+ return_VOID;
+ }
+
status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
if (ACPI_FAILURE(status)) {
return_VOID;
@@ -462,7 +485,7 @@
* Take a snapshot of the GPE info for this level - we copy the info to
* prevent a race condition with remove_handler/remove_block.
*/
- ACPI_MEMCPY(&local_gpe_event_info, gpe_event_info,
+ ACPI_MEMCPY(local_gpe_event_info, gpe_event_info,
sizeof(struct acpi_gpe_event_info));
status = acpi_ut_release_mutex(ACPI_MTX_EVENTS);
@@ -470,12 +493,26 @@
return_VOID;
}
- /*
- * Must check for control method type dispatch one more time to avoid a
- * race with ev_gpe_install_handler
- */
- if ((local_gpe_event_info.flags & ACPI_GPE_DISPATCH_MASK) ==
- ACPI_GPE_DISPATCH_METHOD) {
+ /* Do the correct dispatch - normal method or implicit notify */
+
+ switch (local_gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK) {
+ case ACPI_GPE_DISPATCH_NOTIFY:
+
+ /*
+ * Implicit notify.
+ * Dispatch a DEVICE_WAKE notify to the appropriate handler.
+ * NOTE: the request is queued for execution after this method
+ * completes. The notify handlers are NOT invoked synchronously
+ * from this thread -- because handlers may in turn run other
+ * control methods.
+ */
+ status =
+ acpi_ev_queue_notify_request(local_gpe_event_info->dispatch.
+ device_node,
+ ACPI_NOTIFY_DEVICE_WAKE);
+ break;
+
+ case ACPI_GPE_DISPATCH_METHOD:
/* Allocate the evaluation information block */
@@ -488,7 +525,7 @@
* control method that corresponds to this GPE
*/
info->prefix_node =
- local_gpe_event_info.dispatch.method_node;
+ local_gpe_event_info->dispatch.method_node;
info->flags = ACPI_IGNORE_RETURN_VALUE;
status = acpi_ns_evaluate(info);
@@ -499,46 +536,98 @@
ACPI_EXCEPTION((AE_INFO, status,
"while evaluating GPE method [%4.4s]",
acpi_ut_get_node_name
- (local_gpe_event_info.dispatch.
+ (local_gpe_event_info->dispatch.
method_node)));
}
+
+ break;
+
+ default:
+ return_VOID; /* Should never happen */
}
+
/* Defer enabling of GPE until all notify handlers are done */
- acpi_os_execute(OSL_NOTIFY_HANDLER, acpi_ev_asynch_enable_gpe,
- gpe_event_info);
+
+ status = acpi_os_execute(OSL_NOTIFY_HANDLER,
+ acpi_ev_asynch_enable_gpe,
+ local_gpe_event_info);
+ if (ACPI_FAILURE(status)) {
+ ACPI_FREE(local_gpe_event_info);
+ }
return_VOID;
}
-static void acpi_ev_asynch_enable_gpe(void *context)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_ev_asynch_enable_gpe
+ *
+ * PARAMETERS: Context (gpe_event_info) - Info for this GPE
+ * Callback from acpi_os_execute
+ *
+ * RETURN: None
+ *
+ * DESCRIPTION: Asynchronous clear/enable for GPE. This allows the GPE to
+ * complete (i.e., finish execution of Notify)
+ *
+ ******************************************************************************/
+
+static void ACPI_SYSTEM_XFACE acpi_ev_asynch_enable_gpe(void *context)
{
struct acpi_gpe_event_info *gpe_event_info = context;
+
+ (void)acpi_ev_finish_gpe(gpe_event_info);
+
+ ACPI_FREE(gpe_event_info);
+ return;
+}
+
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_ev_finish_gpe
+ *
+ * PARAMETERS: gpe_event_info - Info for this GPE
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Clear/Enable a GPE. Common code that is used after execution
+ * of a GPE method or a synchronous or asynchronous GPE handler.
+ *
+ ******************************************************************************/
+
+acpi_status acpi_ev_finish_gpe(struct acpi_gpe_event_info *gpe_event_info)
+{
acpi_status status;
+
if ((gpe_event_info->flags & ACPI_GPE_XRUPT_TYPE_MASK) ==
ACPI_GPE_LEVEL_TRIGGERED) {
/*
- * GPE is level-triggered, we clear the GPE status bit after handling
- * the event.
+ * GPE is level-triggered, we clear the GPE status bit after
+ * handling the event.
*/
status = acpi_hw_clear_gpe(gpe_event_info);
if (ACPI_FAILURE(status)) {
- return_VOID;
+ return (status);
}
}
/*
- * Enable this GPE, conditionally. This means that the GPE will only be
- * physically enabled if the enable_for_run bit is set in the event_info
+ * Enable this GPE, conditionally. This means that the GPE will
+ * only be physically enabled if the enable_for_run bit is set
+ * in the event_info.
*/
- (void)acpi_hw_low_set_gpe(gpe_event_info, ACPI_GPE_COND_ENABLE);
-
- return_VOID;
+ (void)acpi_hw_low_set_gpe(gpe_event_info, ACPI_GPE_CONDITIONAL_ENABLE);
+ return (AE_OK);
}
+
/*******************************************************************************
*
* FUNCTION: acpi_ev_gpe_dispatch
*
- * PARAMETERS: gpe_event_info - Info for this GPE
+ * PARAMETERS: gpe_device - Device node. NULL for GPE0/GPE1
+ * gpe_event_info - Info for this GPE
* gpe_number - Number relative to the parent GPE block
*
* RETURN: INTERRUPT_HANDLED or INTERRUPT_NOT_HANDLED
@@ -551,13 +640,22 @@
******************************************************************************/
u32
-acpi_ev_gpe_dispatch(struct acpi_gpe_event_info *gpe_event_info, u32 gpe_number)
+acpi_ev_gpe_dispatch(struct acpi_namespace_node *gpe_device,
+ struct acpi_gpe_event_info *gpe_event_info, u32 gpe_number)
{
acpi_status status;
+ u32 return_value;
ACPI_FUNCTION_TRACE(ev_gpe_dispatch);
- acpi_os_gpe_count(gpe_number);
+ /* Invoke global event handler if present */
+
+ acpi_gpe_count++;
+ if (acpi_gbl_global_event_handler) {
+ acpi_gbl_global_event_handler(ACPI_EVENT_TYPE_GPE, gpe_device,
+ gpe_number,
+ acpi_gbl_global_event_handler_context);
+ }
/*
* If edge-triggered, clear the GPE status bit now. Note that
@@ -568,59 +666,55 @@
status = acpi_hw_clear_gpe(gpe_event_info);
if (ACPI_FAILURE(status)) {
ACPI_EXCEPTION((AE_INFO, status,
- "Unable to clear GPE[0x%2X]",
- gpe_number));
+ "Unable to clear GPE%02X", gpe_number));
return_UINT32(ACPI_INTERRUPT_NOT_HANDLED);
}
}
/*
- * Dispatch the GPE to either an installed handler, or the control method
- * associated with this GPE (_Lxx or _Exx). If a handler exists, we invoke
- * it and do not attempt to run the method. If there is neither a handler
- * nor a method, we disable this GPE to prevent further such pointless
- * events from firing.
+ * Always disable the GPE so that it does not keep firing before
+ * any asynchronous activity completes (either from the execution
+ * of a GPE method or an asynchronous GPE handler.)
+ *
+ * If there is no handler or method to run, just disable the
+ * GPE and leave it disabled permanently to prevent further such
+ * pointless events from firing.
+ */
+ status = acpi_hw_low_set_gpe(gpe_event_info, ACPI_GPE_DISABLE);
+ if (ACPI_FAILURE(status)) {
+ ACPI_EXCEPTION((AE_INFO, status,
+ "Unable to disable GPE%02X", gpe_number));
+ return_UINT32(ACPI_INTERRUPT_NOT_HANDLED);
+ }
+
+ /*
+ * Dispatch the GPE to either an installed handler or the control
+ * method associated with this GPE (_Lxx or _Exx). If a handler
+ * exists, we invoke it and do not attempt to run the method.
+ * If there is neither a handler nor a method, leave the GPE
+ * disabled.
*/
switch (gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK) {
case ACPI_GPE_DISPATCH_HANDLER:
- /*
- * Invoke the installed handler (at interrupt level)
- * Ignore return status for now.
- * TBD: leave GPE disabled on error?
- */
- (void)gpe_event_info->dispatch.handler->address(gpe_event_info->
- dispatch.
- handler->
- context);
+ /* Invoke the installed handler (at interrupt level) */
- /* It is now safe to clear level-triggered events. */
+ return_value =
+ gpe_event_info->dispatch.handler->address(gpe_device,
+ gpe_number,
+ gpe_event_info->
+ dispatch.handler->
+ context);
- if ((gpe_event_info->flags & ACPI_GPE_XRUPT_TYPE_MASK) ==
- ACPI_GPE_LEVEL_TRIGGERED) {
- status = acpi_hw_clear_gpe(gpe_event_info);
- if (ACPI_FAILURE(status)) {
- ACPI_EXCEPTION((AE_INFO, status,
- "Unable to clear GPE[0x%2X]",
- gpe_number));
- return_UINT32(ACPI_INTERRUPT_NOT_HANDLED);
- }
+ /* If requested, clear (if level-triggered) and reenable the GPE */
+
+ if (return_value & ACPI_REENABLE_GPE) {
+ (void)acpi_ev_finish_gpe(gpe_event_info);
}
break;
case ACPI_GPE_DISPATCH_METHOD:
-
- /*
- * Disable the GPE, so it doesn't keep firing before the method has a
- * chance to run (it runs asynchronously with interrupts enabled).
- */
- status = acpi_hw_low_set_gpe(gpe_event_info, ACPI_GPE_DISABLE);
- if (ACPI_FAILURE(status)) {
- ACPI_EXCEPTION((AE_INFO, status,
- "Unable to disable GPE[0x%2X]",
- gpe_number));
- return_UINT32(ACPI_INTERRUPT_NOT_HANDLED);
- }
+ case ACPI_GPE_DISPATCH_NOTIFY:
/*
* Execute the method associated with the GPE
@@ -631,7 +725,7 @@
gpe_event_info);
if (ACPI_FAILURE(status)) {
ACPI_EXCEPTION((AE_INFO, status,
- "Unable to queue handler for GPE[0x%2X] - event disabled",
+ "Unable to queue handler for GPE%2X - event disabled",
gpe_number));
}
break;
@@ -644,20 +738,9 @@
* a GPE to be enabled if it has no handler or method.
*/
ACPI_ERROR((AE_INFO,
- "No handler or method for GPE[0x%2X], disabling event",
+ "No handler or method for GPE%02X, disabling event",
gpe_number));
- /*
- * Disable the GPE. The GPE will remain disabled a handler
- * is installed or ACPICA is restarted.
- */
- status = acpi_hw_low_set_gpe(gpe_event_info, ACPI_GPE_DISABLE);
- if (ACPI_FAILURE(status)) {
- ACPI_EXCEPTION((AE_INFO, status,
- "Unable to disable GPE[0x%2X]",
- gpe_number));
- return_UINT32(ACPI_INTERRUPT_NOT_HANDLED);
- }
break;
}
diff --git a/drivers/acpi/acpica/evgpeblk.c b/drivers/acpi/acpica/evgpeblk.c
index 020add3..9acb869 100644
--- a/drivers/acpi/acpica/evgpeblk.c
+++ b/drivers/acpi/acpica/evgpeblk.c
@@ -361,9 +361,9 @@
gpe_block->node = gpe_device;
gpe_block->gpe_count = (u16)(register_count * ACPI_GPE_REGISTER_WIDTH);
+ gpe_block->initialized = FALSE;
gpe_block->register_count = register_count;
gpe_block->block_base_number = gpe_block_base_number;
- gpe_block->initialized = FALSE;
ACPI_MEMCPY(&gpe_block->block_address, gpe_block_address,
sizeof(struct acpi_generic_address));
@@ -386,7 +386,7 @@
return_ACPI_STATUS(status);
}
- acpi_all_gpes_initialized = FALSE;
+ acpi_gbl_all_gpes_initialized = FALSE;
/* Find all GPE methods (_Lxx or_Exx) for this block */
@@ -423,14 +423,12 @@
*
* FUNCTION: acpi_ev_initialize_gpe_block
*
- * PARAMETERS: gpe_device - Handle to the parent GPE block
- * gpe_block - Gpe Block info
+ * PARAMETERS: acpi_gpe_callback
*
* RETURN: Status
*
- * DESCRIPTION: Initialize and enable a GPE block. First find and run any
- * _PRT methods associated with the block, then enable the
- * appropriate GPEs.
+ * DESCRIPTION: Initialize and enable a GPE block. Enable GPEs that have
+ * associated methods.
* Note: Assumes namespace is locked.
*
******************************************************************************/
@@ -450,8 +448,8 @@
ACPI_FUNCTION_TRACE(ev_initialize_gpe_block);
/*
- * Ignore a null GPE block (e.g., if no GPE block 1 exists) and
- * GPE blocks that have been initialized already.
+ * Ignore a null GPE block (e.g., if no GPE block 1 exists), and
+ * any GPE blocks that have been initialized already.
*/
if (!gpe_block || gpe_block->initialized) {
return_ACPI_STATUS(AE_OK);
@@ -459,8 +457,8 @@
/*
* Enable all GPEs that have a corresponding method and have the
- * ACPI_GPE_CAN_WAKE flag unset. Any other GPEs within this block must
- * be enabled via the acpi_enable_gpe() interface.
+ * ACPI_GPE_CAN_WAKE flag unset. Any other GPEs within this block
+ * must be enabled via the acpi_enable_gpe() interface.
*/
gpe_enabled_count = 0;
@@ -472,14 +470,19 @@
gpe_index = (i * ACPI_GPE_REGISTER_WIDTH) + j;
gpe_event_info = &gpe_block->event_info[gpe_index];
- /* Ignore GPEs that have no corresponding _Lxx/_Exx method */
-
- if (!(gpe_event_info->flags & ACPI_GPE_DISPATCH_METHOD)
+ /*
+ * Ignore GPEs that have no corresponding _Lxx/_Exx method
+ * and GPEs that are used to wake the system
+ */
+ if (((gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK) ==
+ ACPI_GPE_DISPATCH_NONE)
+ || ((gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK)
+ == ACPI_GPE_DISPATCH_HANDLER)
|| (gpe_event_info->flags & ACPI_GPE_CAN_WAKE)) {
continue;
}
- status = acpi_raw_enable_gpe(gpe_event_info);
+ status = acpi_ev_add_gpe_reference(gpe_event_info);
if (ACPI_FAILURE(status)) {
ACPI_EXCEPTION((AE_INFO, status,
"Could not enable GPE 0x%02X",
diff --git a/drivers/acpi/acpica/evgpeinit.c b/drivers/acpi/acpica/evgpeinit.c
index 4c8dea5..c59dc23 100644
--- a/drivers/acpi/acpica/evgpeinit.c
+++ b/drivers/acpi/acpica/evgpeinit.c
@@ -45,11 +45,27 @@
#include "accommon.h"
#include "acevents.h"
#include "acnamesp.h"
-#include "acinterp.h"
#define _COMPONENT ACPI_EVENTS
ACPI_MODULE_NAME("evgpeinit")
+/*
+ * Note: History of _PRW support in ACPICA
+ *
+ * Originally (2000 - 2010), the GPE initialization code performed a walk of
+ * the entire namespace to execute the _PRW methods and detect all GPEs
+ * capable of waking the system.
+ *
+ * As of 10/2010, the _PRW method execution has been removed since it is
+ * actually unnecessary. The host OS must in fact execute all _PRW methods
+ * in order to identify the device/power-resource dependencies. We now put
+ * the onus on the host OS to identify the wake GPEs as part of this process
+ * and to inform ACPICA of these GPEs via the acpi_setup_gpe_for_wake interface. This
+ * not only reduces the complexity of the ACPICA initialization code, but in
+ * some cases (on systems with very large namespaces) it should reduce the
+ * kernel boot time as well.
+ */
+
/*******************************************************************************
*
* FUNCTION: acpi_ev_gpe_initialize
@@ -222,7 +238,7 @@
acpi_status status = AE_OK;
/*
- * 2) Find any _Lxx/_Exx GPE methods that have just been loaded.
+ * Find any _Lxx/_Exx GPE methods that have just been loaded.
*
* Any GPEs that correspond to new _Lxx/_Exx methods are immediately
* enabled.
@@ -235,9 +251,9 @@
return;
}
+ walk_info.count = 0;
walk_info.owner_id = table_owner_id;
walk_info.execute_by_owner_id = TRUE;
- walk_info.count = 0;
/* Walk the interrupt level descriptor list */
@@ -298,7 +314,7 @@
* xx - is the GPE number [in HEX]
*
* If walk_info->execute_by_owner_id is TRUE, we only execute examine GPE methods
- * with that owner.
+ * with that owner.
*
******************************************************************************/
@@ -415,6 +431,7 @@
* Add the GPE information from above to the gpe_event_info block for
* use during dispatch of this GPE.
*/
+ gpe_event_info->flags &= ~(ACPI_GPE_DISPATCH_MASK);
gpe_event_info->flags |= (u8)(type | ACPI_GPE_DISPATCH_METHOD);
gpe_event_info->dispatch.method_node = method_node;
diff --git a/drivers/acpi/acpica/evgpeutil.c b/drivers/acpi/acpica/evgpeutil.c
index 19a0e51..10e4774 100644
--- a/drivers/acpi/acpica/evgpeutil.c
+++ b/drivers/acpi/acpica/evgpeutil.c
@@ -154,6 +154,45 @@
/*******************************************************************************
*
+ * FUNCTION: acpi_ev_get_gpe_device
+ *
+ * PARAMETERS: GPE_WALK_CALLBACK
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Matches the input GPE index (0-current_gpe_count) with a GPE
+ * block device. NULL if the GPE is one of the FADT-defined GPEs.
+ *
+ ******************************************************************************/
+
+acpi_status
+acpi_ev_get_gpe_device(struct acpi_gpe_xrupt_info *gpe_xrupt_info,
+ struct acpi_gpe_block_info *gpe_block, void *context)
+{
+ struct acpi_gpe_device_info *info = context;
+
+ /* Increment Index by the number of GPEs in this block */
+
+ info->next_block_base_index += gpe_block->gpe_count;
+
+ if (info->index < info->next_block_base_index) {
+ /*
+ * The GPE index is within this block, get the node. Leave the node
+ * NULL for the FADT-defined GPEs
+ */
+ if ((gpe_block->node)->type == ACPI_TYPE_DEVICE) {
+ info->gpe_device = gpe_block->node;
+ }
+
+ info->status = AE_OK;
+ return (AE_CTRL_END);
+ }
+
+ return (AE_OK);
+}
+
+/*******************************************************************************
+ *
* FUNCTION: acpi_ev_get_gpe_xrupt_block
*
* PARAMETERS: interrupt_number - Interrupt for a GPE block
diff --git a/drivers/acpi/acpica/evmisc.c b/drivers/acpi/acpica/evmisc.c
index fcaed9f..38bba66 100644
--- a/drivers/acpi/acpica/evmisc.c
+++ b/drivers/acpi/acpica/evmisc.c
@@ -284,41 +284,39 @@
* RETURN: ACPI_INTERRUPT_HANDLED
*
* DESCRIPTION: Invoked directly from the SCI handler when a global lock
- * release interrupt occurs. Attempt to acquire the global lock,
- * if successful, signal the thread waiting for the lock.
+ * release interrupt occurs. If there's a thread waiting for
+ * the global lock, signal it.
*
* NOTE: Assumes that the semaphore can be signaled from interrupt level. If
* this is not possible for some reason, a separate thread will have to be
* scheduled to do this.
*
******************************************************************************/
+static u8 acpi_ev_global_lock_pending;
static u32 acpi_ev_global_lock_handler(void *context)
{
- u8 acquired = FALSE;
+ acpi_status status;
+ acpi_cpu_flags flags;
- /*
- * Attempt to get the lock.
- *
- * If we don't get it now, it will be marked pending and we will
- * take another interrupt when it becomes free.
- */
- ACPI_ACQUIRE_GLOBAL_LOCK(acpi_gbl_FACS, acquired);
- if (acquired) {
+ flags = acpi_os_acquire_lock(acpi_ev_global_lock_pending_lock);
- /* Got the lock, now wake all threads waiting for it */
-
- acpi_gbl_global_lock_acquired = TRUE;
- /* Send a unit to the semaphore */
-
- if (ACPI_FAILURE
- (acpi_os_signal_semaphore
- (acpi_gbl_global_lock_semaphore, 1))) {
- ACPI_ERROR((AE_INFO,
- "Could not signal Global Lock semaphore"));
- }
+ if (!acpi_ev_global_lock_pending) {
+ goto out;
}
+ /* Send a unit to the semaphore */
+
+ status = acpi_os_signal_semaphore(acpi_gbl_global_lock_semaphore, 1);
+ if (ACPI_FAILURE(status)) {
+ ACPI_ERROR((AE_INFO, "Could not signal Global Lock semaphore"));
+ }
+
+ acpi_ev_global_lock_pending = FALSE;
+
+ out:
+ acpi_os_release_lock(acpi_ev_global_lock_pending_lock, flags);
+
return (ACPI_INTERRUPT_HANDLED);
}
@@ -415,6 +413,7 @@
acpi_status acpi_ev_acquire_global_lock(u16 timeout)
{
+ acpi_cpu_flags flags;
acpi_status status = AE_OK;
u8 acquired = FALSE;
@@ -467,32 +466,47 @@
return_ACPI_STATUS(AE_OK);
}
- /* Attempt to acquire the actual hardware lock */
+ flags = acpi_os_acquire_lock(acpi_ev_global_lock_pending_lock);
- ACPI_ACQUIRE_GLOBAL_LOCK(acpi_gbl_FACS, acquired);
- if (acquired) {
+ do {
- /* We got the lock */
+ /* Attempt to acquire the actual hardware lock */
+ ACPI_ACQUIRE_GLOBAL_LOCK(acpi_gbl_FACS, acquired);
+ if (acquired) {
+ acpi_gbl_global_lock_acquired = TRUE;
+
+ ACPI_DEBUG_PRINT((ACPI_DB_EXEC,
+ "Acquired hardware Global Lock\n"));
+ break;
+ }
+
+ acpi_ev_global_lock_pending = TRUE;
+
+ acpi_os_release_lock(acpi_ev_global_lock_pending_lock, flags);
+
+ /*
+ * Did not get the lock. The pending bit was set above, and we
+ * must wait until we get the global lock released interrupt.
+ */
ACPI_DEBUG_PRINT((ACPI_DB_EXEC,
- "Acquired hardware Global Lock\n"));
+ "Waiting for hardware Global Lock\n"));
- acpi_gbl_global_lock_acquired = TRUE;
- return_ACPI_STATUS(AE_OK);
- }
+ /*
+ * Wait for handshake with the global lock interrupt handler.
+ * This interface releases the interpreter if we must wait.
+ */
+ status = acpi_ex_system_wait_semaphore(
+ acpi_gbl_global_lock_semaphore,
+ ACPI_WAIT_FOREVER);
- /*
- * Did not get the lock. The pending bit was set above, and we must now
- * wait until we get the global lock released interrupt.
- */
- ACPI_DEBUG_PRINT((ACPI_DB_EXEC, "Waiting for hardware Global Lock\n"));
+ flags = acpi_os_acquire_lock(acpi_ev_global_lock_pending_lock);
- /*
- * Wait for handshake with the global lock interrupt handler.
- * This interface releases the interpreter if we must wait.
- */
- status = acpi_ex_system_wait_semaphore(acpi_gbl_global_lock_semaphore,
- ACPI_WAIT_FOREVER);
+ } while (ACPI_SUCCESS(status));
+
+ acpi_ev_global_lock_pending = FALSE;
+
+ acpi_os_release_lock(acpi_ev_global_lock_pending_lock, flags);
return_ACPI_STATUS(status);
}
diff --git a/drivers/acpi/acpica/evxface.c b/drivers/acpi/acpica/evxface.c
index 36af222..1226689 100644
--- a/drivers/acpi/acpica/evxface.c
+++ b/drivers/acpi/acpica/evxface.c
@@ -92,6 +92,57 @@
ACPI_EXPORT_SYMBOL(acpi_install_exception_handler)
#endif /* ACPI_FUTURE_USAGE */
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_install_global_event_handler
+ *
+ * PARAMETERS: Handler - Pointer to the global event handler function
+ * Context - Value passed to the handler on each event
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Saves the pointer to the handler function. The global handler
+ * is invoked upon each incoming GPE and Fixed Event. It is
+ * invoked at interrupt level at the time of the event dispatch.
+ * Can be used to update event counters, etc.
+ *
+ ******************************************************************************/
+acpi_status
+acpi_install_global_event_handler(ACPI_GBL_EVENT_HANDLER handler, void *context)
+{
+ acpi_status status;
+
+ ACPI_FUNCTION_TRACE(acpi_install_global_event_handler);
+
+ /* Parameter validation */
+
+ if (!handler) {
+ return_ACPI_STATUS(AE_BAD_PARAMETER);
+ }
+
+ status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
+ if (ACPI_FAILURE(status)) {
+ return_ACPI_STATUS(status);
+ }
+
+ /* Don't allow two handlers. */
+
+ if (acpi_gbl_global_event_handler) {
+ status = AE_ALREADY_EXISTS;
+ goto cleanup;
+ }
+
+ acpi_gbl_global_event_handler = handler;
+ acpi_gbl_global_event_handler_context = context;
+
+ cleanup:
+ (void)acpi_ut_release_mutex(ACPI_MTX_EVENTS);
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_install_global_event_handler)
+
/*******************************************************************************
*
* FUNCTION: acpi_install_fixed_event_handler
@@ -671,10 +722,10 @@
acpi_status
acpi_install_gpe_handler(acpi_handle gpe_device,
u32 gpe_number,
- u32 type, acpi_event_handler address, void *context)
+ u32 type, acpi_gpe_handler address, void *context)
{
struct acpi_gpe_event_info *gpe_event_info;
- struct acpi_handler_info *handler;
+ struct acpi_gpe_handler_info *handler;
acpi_status status;
acpi_cpu_flags flags;
@@ -693,7 +744,7 @@
/* Allocate memory for the handler object */
- handler = ACPI_ALLOCATE_ZEROED(sizeof(struct acpi_handler_info));
+ handler = ACPI_ALLOCATE_ZEROED(sizeof(struct acpi_gpe_handler_info));
if (!handler) {
status = AE_NO_MEMORY;
goto unlock_and_exit;
@@ -722,7 +773,7 @@
handler->address = address;
handler->context = context;
handler->method_node = gpe_event_info->dispatch.method_node;
- handler->orig_flags = gpe_event_info->flags &
+ handler->original_flags = gpe_event_info->flags &
(ACPI_GPE_XRUPT_TYPE_MASK | ACPI_GPE_DISPATCH_MASK);
/*
@@ -731,10 +782,10 @@
* disabled now to avoid spurious execution of the handler.
*/
- if ((handler->orig_flags & ACPI_GPE_DISPATCH_METHOD)
+ if ((handler->original_flags & ACPI_GPE_DISPATCH_METHOD)
&& gpe_event_info->runtime_count) {
- handler->orig_enabled = 1;
- (void)acpi_raw_disable_gpe(gpe_event_info);
+ handler->originally_enabled = 1;
+ (void)acpi_ev_remove_gpe_reference(gpe_event_info);
}
/* Install the handler */
@@ -777,10 +828,10 @@
******************************************************************************/
acpi_status
acpi_remove_gpe_handler(acpi_handle gpe_device,
- u32 gpe_number, acpi_event_handler address)
+ u32 gpe_number, acpi_gpe_handler address)
{
struct acpi_gpe_event_info *gpe_event_info;
- struct acpi_handler_info *handler;
+ struct acpi_gpe_handler_info *handler;
acpi_status status;
acpi_cpu_flags flags;
@@ -835,7 +886,7 @@
gpe_event_info->dispatch.method_node = handler->method_node;
gpe_event_info->flags &=
~(ACPI_GPE_XRUPT_TYPE_MASK | ACPI_GPE_DISPATCH_MASK);
- gpe_event_info->flags |= handler->orig_flags;
+ gpe_event_info->flags |= handler->original_flags;
/*
* If the GPE was previously associated with a method and it was
@@ -843,9 +894,9 @@
* post-initialization configuration.
*/
- if ((handler->orig_flags & ACPI_GPE_DISPATCH_METHOD)
- && handler->orig_enabled)
- (void)acpi_raw_enable_gpe(gpe_event_info);
+ if ((handler->original_flags & ACPI_GPE_DISPATCH_METHOD)
+ && handler->originally_enabled)
+ (void)acpi_ev_add_gpe_reference(gpe_event_info);
/* Now we can free the handler object */
diff --git a/drivers/acpi/acpica/evxfevnt.c b/drivers/acpi/acpica/evxfevnt.c
index a1dabe3..90488c1 100644
--- a/drivers/acpi/acpica/evxfevnt.c
+++ b/drivers/acpi/acpica/evxfevnt.c
@@ -43,18 +43,11 @@
#include <acpi/acpi.h>
#include "accommon.h"
-#include "acevents.h"
-#include "acnamesp.h"
#include "actables.h"
#define _COMPONENT ACPI_EVENTS
ACPI_MODULE_NAME("evxfevnt")
-/* Local prototypes */
-static acpi_status
-acpi_ev_get_gpe_device(struct acpi_gpe_xrupt_info *gpe_xrupt_info,
- struct acpi_gpe_block_info *gpe_block, void *context);
-
/*******************************************************************************
*
* FUNCTION: acpi_enable
@@ -213,185 +206,6 @@
/*******************************************************************************
*
- * FUNCTION: acpi_gpe_wakeup
- *
- * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
- * gpe_number - GPE level within the GPE block
- * Action - Enable or Disable
- *
- * RETURN: Status
- *
- * DESCRIPTION: Set or clear the GPE's wakeup enable mask bit.
- *
- ******************************************************************************/
-acpi_status acpi_gpe_wakeup(acpi_handle gpe_device, u32 gpe_number, u8 action)
-{
- acpi_status status = AE_OK;
- struct acpi_gpe_event_info *gpe_event_info;
- struct acpi_gpe_register_info *gpe_register_info;
- acpi_cpu_flags flags;
- u32 register_bit;
-
- ACPI_FUNCTION_TRACE(acpi_gpe_wakeup);
-
- flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
-
- /* Ensure that we have a valid GPE number */
-
- gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
- if (!gpe_event_info || !(gpe_event_info->flags & ACPI_GPE_CAN_WAKE)) {
- status = AE_BAD_PARAMETER;
- goto unlock_and_exit;
- }
-
- gpe_register_info = gpe_event_info->register_info;
- if (!gpe_register_info) {
- status = AE_NOT_EXIST;
- goto unlock_and_exit;
- }
-
- register_bit =
- acpi_hw_get_gpe_register_bit(gpe_event_info, gpe_register_info);
-
- /* Perform the action */
-
- switch (action) {
- case ACPI_GPE_ENABLE:
- ACPI_SET_BIT(gpe_register_info->enable_for_wake,
- (u8)register_bit);
- break;
-
- case ACPI_GPE_DISABLE:
- ACPI_CLEAR_BIT(gpe_register_info->enable_for_wake,
- (u8)register_bit);
- break;
-
- default:
- ACPI_ERROR((AE_INFO, "%u, Invalid action", action));
- status = AE_BAD_PARAMETER;
- break;
- }
-
-unlock_and_exit:
- acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
- return_ACPI_STATUS(status);
-}
-
-ACPI_EXPORT_SYMBOL(acpi_gpe_wakeup)
-
-/*******************************************************************************
- *
- * FUNCTION: acpi_enable_gpe
- *
- * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
- * gpe_number - GPE level within the GPE block
- *
- * RETURN: Status
- *
- * DESCRIPTION: Add a reference to a GPE. On the first reference, the GPE is
- * hardware-enabled.
- *
- ******************************************************************************/
-acpi_status acpi_enable_gpe(acpi_handle gpe_device, u32 gpe_number)
-{
- acpi_status status = AE_BAD_PARAMETER;
- struct acpi_gpe_event_info *gpe_event_info;
- acpi_cpu_flags flags;
-
- ACPI_FUNCTION_TRACE(acpi_enable_gpe);
-
- flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
-
- /* Ensure that we have a valid GPE number */
-
- gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
- if (gpe_event_info) {
- status = acpi_raw_enable_gpe(gpe_event_info);
- }
-
- acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
- return_ACPI_STATUS(status);
-}
-ACPI_EXPORT_SYMBOL(acpi_enable_gpe)
-
-/*******************************************************************************
- *
- * FUNCTION: acpi_disable_gpe
- *
- * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
- * gpe_number - GPE level within the GPE block
- *
- * RETURN: Status
- *
- * DESCRIPTION: Remove a reference to a GPE. When the last reference is
- * removed, only then is the GPE disabled (for runtime GPEs), or
- * the GPE mask bit disabled (for wake GPEs)
- *
- ******************************************************************************/
-acpi_status acpi_disable_gpe(acpi_handle gpe_device, u32 gpe_number)
-{
- acpi_status status = AE_BAD_PARAMETER;
- struct acpi_gpe_event_info *gpe_event_info;
- acpi_cpu_flags flags;
-
- ACPI_FUNCTION_TRACE(acpi_disable_gpe);
-
- flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
-
- /* Ensure that we have a valid GPE number */
-
- gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
- if (gpe_event_info) {
- status = acpi_raw_disable_gpe(gpe_event_info) ;
- }
-
- acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
- return_ACPI_STATUS(status);
-}
-ACPI_EXPORT_SYMBOL(acpi_disable_gpe)
-
-/*******************************************************************************
- *
- * FUNCTION: acpi_gpe_can_wake
- *
- * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
- * gpe_number - GPE level within the GPE block
- *
- * RETURN: Status
- *
- * DESCRIPTION: Set the ACPI_GPE_CAN_WAKE flag for the given GPE. If the GPE
- * has a corresponding method and is currently enabled, disable it
- * (GPEs with corresponding methods are enabled unconditionally
- * during initialization, but GPEs that can wake up are expected
- * to be initially disabled).
- *
- ******************************************************************************/
-acpi_status acpi_gpe_can_wake(acpi_handle gpe_device, u32 gpe_number)
-{
- acpi_status status = AE_OK;
- struct acpi_gpe_event_info *gpe_event_info;
- acpi_cpu_flags flags;
-
- ACPI_FUNCTION_TRACE(acpi_gpe_can_wake);
-
- flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
-
- /* Ensure that we have a valid GPE number */
-
- gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
- if (gpe_event_info) {
- gpe_event_info->flags |= ACPI_GPE_CAN_WAKE;
- } else {
- status = AE_BAD_PARAMETER;
- }
-
- acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
- return_ACPI_STATUS(status);
-}
-ACPI_EXPORT_SYMBOL(acpi_gpe_can_wake)
-
-/*******************************************************************************
- *
* FUNCTION: acpi_disable_event
*
* PARAMETERS: Event - The fixed eventto be enabled
@@ -483,44 +297,6 @@
/*******************************************************************************
*
- * FUNCTION: acpi_clear_gpe
- *
- * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
- * gpe_number - GPE level within the GPE block
- *
- * RETURN: Status
- *
- * DESCRIPTION: Clear an ACPI event (general purpose)
- *
- ******************************************************************************/
-acpi_status acpi_clear_gpe(acpi_handle gpe_device, u32 gpe_number)
-{
- acpi_status status = AE_OK;
- struct acpi_gpe_event_info *gpe_event_info;
- acpi_cpu_flags flags;
-
- ACPI_FUNCTION_TRACE(acpi_clear_gpe);
-
- flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
-
- /* Ensure that we have a valid GPE number */
-
- gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
- if (!gpe_event_info) {
- status = AE_BAD_PARAMETER;
- goto unlock_and_exit;
- }
-
- status = acpi_hw_clear_gpe(gpe_event_info);
-
- unlock_and_exit:
- acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
- return_ACPI_STATUS(status);
-}
-
-ACPI_EXPORT_SYMBOL(acpi_clear_gpe)
-/*******************************************************************************
- *
* FUNCTION: acpi_get_event_status
*
* PARAMETERS: Event - The fixed event
@@ -575,379 +351,3 @@
}
ACPI_EXPORT_SYMBOL(acpi_get_event_status)
-
-/*******************************************************************************
- *
- * FUNCTION: acpi_get_gpe_status
- *
- * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
- * gpe_number - GPE level within the GPE block
- * event_status - Where the current status of the event will
- * be returned
- *
- * RETURN: Status
- *
- * DESCRIPTION: Get status of an event (general purpose)
- *
- ******************************************************************************/
-acpi_status
-acpi_get_gpe_status(acpi_handle gpe_device,
- u32 gpe_number, acpi_event_status *event_status)
-{
- acpi_status status = AE_OK;
- struct acpi_gpe_event_info *gpe_event_info;
- acpi_cpu_flags flags;
-
- ACPI_FUNCTION_TRACE(acpi_get_gpe_status);
-
- flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
-
- /* Ensure that we have a valid GPE number */
-
- gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
- if (!gpe_event_info) {
- status = AE_BAD_PARAMETER;
- goto unlock_and_exit;
- }
-
- /* Obtain status on the requested GPE number */
-
- status = acpi_hw_get_gpe_status(gpe_event_info, event_status);
-
- if (gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK)
- *event_status |= ACPI_EVENT_FLAG_HANDLE;
-
- unlock_and_exit:
- acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
- return_ACPI_STATUS(status);
-}
-
-ACPI_EXPORT_SYMBOL(acpi_get_gpe_status)
-/*******************************************************************************
- *
- * FUNCTION: acpi_install_gpe_block
- *
- * PARAMETERS: gpe_device - Handle to the parent GPE Block Device
- * gpe_block_address - Address and space_iD
- * register_count - Number of GPE register pairs in the block
- * interrupt_number - H/W interrupt for the block
- *
- * RETURN: Status
- *
- * DESCRIPTION: Create and Install a block of GPE registers
- *
- ******************************************************************************/
-acpi_status
-acpi_install_gpe_block(acpi_handle gpe_device,
- struct acpi_generic_address *gpe_block_address,
- u32 register_count, u32 interrupt_number)
-{
- acpi_status status = AE_OK;
- union acpi_operand_object *obj_desc;
- struct acpi_namespace_node *node;
- struct acpi_gpe_block_info *gpe_block;
-
- ACPI_FUNCTION_TRACE(acpi_install_gpe_block);
-
- if ((!gpe_device) || (!gpe_block_address) || (!register_count)) {
- return_ACPI_STATUS(AE_BAD_PARAMETER);
- }
-
- status = acpi_ut_acquire_mutex(ACPI_MTX_NAMESPACE);
- if (ACPI_FAILURE(status)) {
- return (status);
- }
-
- node = acpi_ns_validate_handle(gpe_device);
- if (!node) {
- status = AE_BAD_PARAMETER;
- goto unlock_and_exit;
- }
-
- /*
- * For user-installed GPE Block Devices, the gpe_block_base_number
- * is always zero
- */
- status =
- acpi_ev_create_gpe_block(node, gpe_block_address, register_count, 0,
- interrupt_number, &gpe_block);
- if (ACPI_FAILURE(status)) {
- goto unlock_and_exit;
- }
-
- /* Install block in the device_object attached to the node */
-
- obj_desc = acpi_ns_get_attached_object(node);
- if (!obj_desc) {
-
- /*
- * No object, create a new one (Device nodes do not always have
- * an attached object)
- */
- obj_desc = acpi_ut_create_internal_object(ACPI_TYPE_DEVICE);
- if (!obj_desc) {
- status = AE_NO_MEMORY;
- goto unlock_and_exit;
- }
-
- status =
- acpi_ns_attach_object(node, obj_desc, ACPI_TYPE_DEVICE);
-
- /* Remove local reference to the object */
-
- acpi_ut_remove_reference(obj_desc);
-
- if (ACPI_FAILURE(status)) {
- goto unlock_and_exit;
- }
- }
-
- /* Now install the GPE block in the device_object */
-
- obj_desc->device.gpe_block = gpe_block;
-
- unlock_and_exit:
- (void)acpi_ut_release_mutex(ACPI_MTX_NAMESPACE);
- return_ACPI_STATUS(status);
-}
-
-ACPI_EXPORT_SYMBOL(acpi_install_gpe_block)
-
-/*******************************************************************************
- *
- * FUNCTION: acpi_remove_gpe_block
- *
- * PARAMETERS: gpe_device - Handle to the parent GPE Block Device
- *
- * RETURN: Status
- *
- * DESCRIPTION: Remove a previously installed block of GPE registers
- *
- ******************************************************************************/
-acpi_status acpi_remove_gpe_block(acpi_handle gpe_device)
-{
- union acpi_operand_object *obj_desc;
- acpi_status status;
- struct acpi_namespace_node *node;
-
- ACPI_FUNCTION_TRACE(acpi_remove_gpe_block);
-
- if (!gpe_device) {
- return_ACPI_STATUS(AE_BAD_PARAMETER);
- }
-
- status = acpi_ut_acquire_mutex(ACPI_MTX_NAMESPACE);
- if (ACPI_FAILURE(status)) {
- return (status);
- }
-
- node = acpi_ns_validate_handle(gpe_device);
- if (!node) {
- status = AE_BAD_PARAMETER;
- goto unlock_and_exit;
- }
-
- /* Get the device_object attached to the node */
-
- obj_desc = acpi_ns_get_attached_object(node);
- if (!obj_desc || !obj_desc->device.gpe_block) {
- return_ACPI_STATUS(AE_NULL_OBJECT);
- }
-
- /* Delete the GPE block (but not the device_object) */
-
- status = acpi_ev_delete_gpe_block(obj_desc->device.gpe_block);
- if (ACPI_SUCCESS(status)) {
- obj_desc->device.gpe_block = NULL;
- }
-
- unlock_and_exit:
- (void)acpi_ut_release_mutex(ACPI_MTX_NAMESPACE);
- return_ACPI_STATUS(status);
-}
-
-ACPI_EXPORT_SYMBOL(acpi_remove_gpe_block)
-
-/*******************************************************************************
- *
- * FUNCTION: acpi_get_gpe_device
- *
- * PARAMETERS: Index - System GPE index (0-current_gpe_count)
- * gpe_device - Where the parent GPE Device is returned
- *
- * RETURN: Status
- *
- * DESCRIPTION: Obtain the GPE device associated with the input index. A NULL
- * gpe device indicates that the gpe number is contained in one of
- * the FADT-defined gpe blocks. Otherwise, the GPE block device.
- *
- ******************************************************************************/
-acpi_status
-acpi_get_gpe_device(u32 index, acpi_handle *gpe_device)
-{
- struct acpi_gpe_device_info info;
- acpi_status status;
-
- ACPI_FUNCTION_TRACE(acpi_get_gpe_device);
-
- if (!gpe_device) {
- return_ACPI_STATUS(AE_BAD_PARAMETER);
- }
-
- if (index >= acpi_current_gpe_count) {
- return_ACPI_STATUS(AE_NOT_EXIST);
- }
-
- /* Setup and walk the GPE list */
-
- info.index = index;
- info.status = AE_NOT_EXIST;
- info.gpe_device = NULL;
- info.next_block_base_index = 0;
-
- status = acpi_ev_walk_gpe_list(acpi_ev_get_gpe_device, &info);
- if (ACPI_FAILURE(status)) {
- return_ACPI_STATUS(status);
- }
-
- *gpe_device = info.gpe_device;
- return_ACPI_STATUS(info.status);
-}
-
-ACPI_EXPORT_SYMBOL(acpi_get_gpe_device)
-
-/*******************************************************************************
- *
- * FUNCTION: acpi_ev_get_gpe_device
- *
- * PARAMETERS: GPE_WALK_CALLBACK
- *
- * RETURN: Status
- *
- * DESCRIPTION: Matches the input GPE index (0-current_gpe_count) with a GPE
- * block device. NULL if the GPE is one of the FADT-defined GPEs.
- *
- ******************************************************************************/
-static acpi_status
-acpi_ev_get_gpe_device(struct acpi_gpe_xrupt_info *gpe_xrupt_info,
- struct acpi_gpe_block_info *gpe_block, void *context)
-{
- struct acpi_gpe_device_info *info = context;
-
- /* Increment Index by the number of GPEs in this block */
-
- info->next_block_base_index += gpe_block->gpe_count;
-
- if (info->index < info->next_block_base_index) {
- /*
- * The GPE index is within this block, get the node. Leave the node
- * NULL for the FADT-defined GPEs
- */
- if ((gpe_block->node)->type == ACPI_TYPE_DEVICE) {
- info->gpe_device = gpe_block->node;
- }
-
- info->status = AE_OK;
- return (AE_CTRL_END);
- }
-
- return (AE_OK);
-}
-
-/******************************************************************************
- *
- * FUNCTION: acpi_disable_all_gpes
- *
- * PARAMETERS: None
- *
- * RETURN: Status
- *
- * DESCRIPTION: Disable and clear all GPEs in all GPE blocks
- *
- ******************************************************************************/
-
-acpi_status acpi_disable_all_gpes(void)
-{
- acpi_status status;
-
- ACPI_FUNCTION_TRACE(acpi_disable_all_gpes);
-
- status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
- if (ACPI_FAILURE(status)) {
- return_ACPI_STATUS(status);
- }
-
- status = acpi_hw_disable_all_gpes();
- (void)acpi_ut_release_mutex(ACPI_MTX_EVENTS);
-
- return_ACPI_STATUS(status);
-}
-
-/******************************************************************************
- *
- * FUNCTION: acpi_enable_all_runtime_gpes
- *
- * PARAMETERS: None
- *
- * RETURN: Status
- *
- * DESCRIPTION: Enable all "runtime" GPEs, in all GPE blocks
- *
- ******************************************************************************/
-
-acpi_status acpi_enable_all_runtime_gpes(void)
-{
- acpi_status status;
-
- ACPI_FUNCTION_TRACE(acpi_enable_all_runtime_gpes);
-
- status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
- if (ACPI_FAILURE(status)) {
- return_ACPI_STATUS(status);
- }
-
- status = acpi_hw_enable_all_runtime_gpes();
- (void)acpi_ut_release_mutex(ACPI_MTX_EVENTS);
-
- return_ACPI_STATUS(status);
-}
-
-/******************************************************************************
- *
- * FUNCTION: acpi_update_gpes
- *
- * PARAMETERS: None
- *
- * RETURN: None
- *
- * DESCRIPTION: Enable all GPEs that have associated _Lxx or _Exx methods and
- * are not pointed to by any device _PRW methods indicating that
- * these GPEs are generally intended for system or device wakeup
- * (such GPEs have to be enabled directly when the devices whose
- * _PRW methods point to them are set up for wakeup signaling).
- *
- ******************************************************************************/
-
-acpi_status acpi_update_gpes(void)
-{
- acpi_status status;
-
- ACPI_FUNCTION_TRACE(acpi_update_gpes);
-
- status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
- if (ACPI_FAILURE(status)) {
- return_ACPI_STATUS(status);
- } else if (acpi_all_gpes_initialized) {
- goto unlock;
- }
-
- status = acpi_ev_walk_gpe_list(acpi_ev_initialize_gpe_block, NULL);
- if (ACPI_SUCCESS(status)) {
- acpi_all_gpes_initialized = TRUE;
- }
-
-unlock:
- (void)acpi_ut_release_mutex(ACPI_MTX_EVENTS);
-
- return_ACPI_STATUS(status);
-}
diff --git a/drivers/acpi/acpica/evxfgpe.c b/drivers/acpi/acpica/evxfgpe.c
new file mode 100644
index 0000000..416845b
--- /dev/null
+++ b/drivers/acpi/acpica/evxfgpe.c
@@ -0,0 +1,669 @@
+/******************************************************************************
+ *
+ * Module Name: evxfgpe - External Interfaces for General Purpose Events (GPEs)
+ *
+ *****************************************************************************/
+
+/*
+ * Copyright (C) 2000 - 2010, Intel Corp.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions, and the following disclaimer,
+ * without modification.
+ * 2. Redistributions in binary form must reproduce at minimum a disclaimer
+ * substantially similar to the "NO WARRANTY" disclaimer below
+ * ("Disclaimer") and any redistribution must be conditioned upon
+ * including a substantially similar Disclaimer requirement for further
+ * binary redistribution.
+ * 3. Neither the names of the above-listed copyright holders nor the names
+ * of any contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * Alternatively, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") version 2 as published by the Free
+ * Software Foundation.
+ *
+ * NO WARRANTY
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * HOLDERS OR CONTRIBUTORS BE LIABLE FOR SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
+ * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
+ * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGES.
+ */
+
+#include <acpi/acpi.h>
+#include "accommon.h"
+#include "acevents.h"
+#include "acnamesp.h"
+
+#define _COMPONENT ACPI_EVENTS
+ACPI_MODULE_NAME("evxfgpe")
+
+/******************************************************************************
+ *
+ * FUNCTION: acpi_update_all_gpes
+ *
+ * PARAMETERS: None
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Complete GPE initialization and enable all GPEs that have
+ * associated _Lxx or _Exx methods and are not pointed to by any
+ * device _PRW methods (this indicates that these GPEs are
+ * generally intended for system or device wakeup. Such GPEs
+ * have to be enabled directly when the devices whose _PRW
+ * methods point to them are set up for wakeup signaling.)
+ *
+ * NOTE: Should be called after any GPEs are added to the system. Primarily,
+ * after the system _PRW methods have been run, but also after a GPE Block
+ * Device has been added or if any new GPE methods have been added via a
+ * dynamic table load.
+ *
+ ******************************************************************************/
+
+acpi_status acpi_update_all_gpes(void)
+{
+ acpi_status status;
+
+ ACPI_FUNCTION_TRACE(acpi_update_all_gpes);
+
+ status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
+ if (ACPI_FAILURE(status)) {
+ return_ACPI_STATUS(status);
+ }
+
+ if (acpi_gbl_all_gpes_initialized) {
+ goto unlock_and_exit;
+ }
+
+ status = acpi_ev_walk_gpe_list(acpi_ev_initialize_gpe_block, NULL);
+ if (ACPI_SUCCESS(status)) {
+ acpi_gbl_all_gpes_initialized = TRUE;
+ }
+
+unlock_and_exit:
+ (void)acpi_ut_release_mutex(ACPI_MTX_EVENTS);
+
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_update_all_gpes)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_enable_gpe
+ *
+ * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
+ * gpe_number - GPE level within the GPE block
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Add a reference to a GPE. On the first reference, the GPE is
+ * hardware-enabled.
+ *
+ ******************************************************************************/
+
+acpi_status acpi_enable_gpe(acpi_handle gpe_device, u32 gpe_number)
+{
+ acpi_status status = AE_BAD_PARAMETER;
+ struct acpi_gpe_event_info *gpe_event_info;
+ acpi_cpu_flags flags;
+
+ ACPI_FUNCTION_TRACE(acpi_enable_gpe);
+
+ flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
+
+ /* Ensure that we have a valid GPE number */
+
+ gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
+ if (gpe_event_info) {
+ status = acpi_ev_add_gpe_reference(gpe_event_info);
+ }
+
+ acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
+ return_ACPI_STATUS(status);
+}
+ACPI_EXPORT_SYMBOL(acpi_enable_gpe)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_disable_gpe
+ *
+ * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
+ * gpe_number - GPE level within the GPE block
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Remove a reference to a GPE. When the last reference is
+ * removed, only then is the GPE disabled (for runtime GPEs), or
+ * the GPE mask bit disabled (for wake GPEs)
+ *
+ ******************************************************************************/
+
+acpi_status acpi_disable_gpe(acpi_handle gpe_device, u32 gpe_number)
+{
+ acpi_status status = AE_BAD_PARAMETER;
+ struct acpi_gpe_event_info *gpe_event_info;
+ acpi_cpu_flags flags;
+
+ ACPI_FUNCTION_TRACE(acpi_disable_gpe);
+
+ flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
+
+ /* Ensure that we have a valid GPE number */
+
+ gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
+ if (gpe_event_info) {
+ status = acpi_ev_remove_gpe_reference(gpe_event_info) ;
+ }
+
+ acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
+ return_ACPI_STATUS(status);
+}
+ACPI_EXPORT_SYMBOL(acpi_disable_gpe)
+
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_setup_gpe_for_wake
+ *
+ * PARAMETERS: wake_device - Device associated with the GPE (via _PRW)
+ * gpe_device - Parent GPE Device. NULL for GPE0/GPE1
+ * gpe_number - GPE level within the GPE block
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Mark a GPE as having the ability to wake the system. This
+ * interface is intended to be used as the host executes the
+ * _PRW methods (Power Resources for Wake) in the system tables.
+ * Each _PRW appears under a Device Object (The wake_device), and
+ * contains the info for the wake GPE associated with the
+ * wake_device.
+ *
+ ******************************************************************************/
+acpi_status
+acpi_setup_gpe_for_wake(acpi_handle wake_device,
+ acpi_handle gpe_device, u32 gpe_number)
+{
+ acpi_status status = AE_BAD_PARAMETER;
+ struct acpi_gpe_event_info *gpe_event_info;
+ struct acpi_namespace_node *device_node;
+ acpi_cpu_flags flags;
+
+ ACPI_FUNCTION_TRACE(acpi_setup_gpe_for_wake);
+
+ /* Parameter Validation */
+
+ if (!wake_device) {
+ /*
+ * By forcing wake_device to be valid, we automatically enable the
+ * implicit notify feature on all hosts.
+ */
+ return_ACPI_STATUS(AE_BAD_PARAMETER);
+ }
+
+ /* Validate wake_device is of type Device */
+
+ device_node = ACPI_CAST_PTR(struct acpi_namespace_node, wake_device);
+ if (device_node->type != ACPI_TYPE_DEVICE) {
+ return_ACPI_STATUS(AE_BAD_PARAMETER);
+ }
+
+ flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
+
+ /* Ensure that we have a valid GPE number */
+
+ gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
+ if (gpe_event_info) {
+ /*
+ * If there is no method or handler for this GPE, then the
+ * wake_device will be notified whenever this GPE fires (aka
+ * "implicit notify") Note: The GPE is assumed to be
+ * level-triggered (for windows compatibility).
+ */
+ if ((gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK) ==
+ ACPI_GPE_DISPATCH_NONE) {
+ gpe_event_info->flags =
+ (ACPI_GPE_DISPATCH_NOTIFY |
+ ACPI_GPE_LEVEL_TRIGGERED);
+ gpe_event_info->dispatch.device_node = device_node;
+ }
+
+ gpe_event_info->flags |= ACPI_GPE_CAN_WAKE;
+ status = AE_OK;
+ }
+
+ acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
+ return_ACPI_STATUS(status);
+}
+ACPI_EXPORT_SYMBOL(acpi_setup_gpe_for_wake)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_set_gpe_wake_mask
+ *
+ * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
+ * gpe_number - GPE level within the GPE block
+ * Action - Enable or Disable
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Set or clear the GPE's wakeup enable mask bit. The GPE must
+ * already be marked as a WAKE GPE.
+ *
+ ******************************************************************************/
+
+acpi_status acpi_set_gpe_wake_mask(acpi_handle gpe_device, u32 gpe_number, u8 action)
+{
+ acpi_status status = AE_OK;
+ struct acpi_gpe_event_info *gpe_event_info;
+ struct acpi_gpe_register_info *gpe_register_info;
+ acpi_cpu_flags flags;
+ u32 register_bit;
+
+ ACPI_FUNCTION_TRACE(acpi_set_gpe_wake_mask);
+
+ flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
+
+ /*
+ * Ensure that we have a valid GPE number and that this GPE is in
+ * fact a wake GPE
+ */
+ gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
+ if (!gpe_event_info) {
+ status = AE_BAD_PARAMETER;
+ goto unlock_and_exit;
+ }
+
+ if (!(gpe_event_info->flags & ACPI_GPE_CAN_WAKE)) {
+ status = AE_TYPE;
+ goto unlock_and_exit;
+ }
+
+ gpe_register_info = gpe_event_info->register_info;
+ if (!gpe_register_info) {
+ status = AE_NOT_EXIST;
+ goto unlock_and_exit;
+ }
+
+ register_bit =
+ acpi_hw_get_gpe_register_bit(gpe_event_info, gpe_register_info);
+
+ /* Perform the action */
+
+ switch (action) {
+ case ACPI_GPE_ENABLE:
+ ACPI_SET_BIT(gpe_register_info->enable_for_wake,
+ (u8)register_bit);
+ break;
+
+ case ACPI_GPE_DISABLE:
+ ACPI_CLEAR_BIT(gpe_register_info->enable_for_wake,
+ (u8)register_bit);
+ break;
+
+ default:
+ ACPI_ERROR((AE_INFO, "%u, Invalid action", action));
+ status = AE_BAD_PARAMETER;
+ break;
+ }
+
+unlock_and_exit:
+ acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_set_gpe_wake_mask)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_clear_gpe
+ *
+ * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
+ * gpe_number - GPE level within the GPE block
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Clear an ACPI event (general purpose)
+ *
+ ******************************************************************************/
+acpi_status acpi_clear_gpe(acpi_handle gpe_device, u32 gpe_number)
+{
+ acpi_status status = AE_OK;
+ struct acpi_gpe_event_info *gpe_event_info;
+ acpi_cpu_flags flags;
+
+ ACPI_FUNCTION_TRACE(acpi_clear_gpe);
+
+ flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
+
+ /* Ensure that we have a valid GPE number */
+
+ gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
+ if (!gpe_event_info) {
+ status = AE_BAD_PARAMETER;
+ goto unlock_and_exit;
+ }
+
+ status = acpi_hw_clear_gpe(gpe_event_info);
+
+ unlock_and_exit:
+ acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_clear_gpe)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_get_gpe_status
+ *
+ * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1
+ * gpe_number - GPE level within the GPE block
+ * event_status - Where the current status of the event will
+ * be returned
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Get the current status of a GPE (signalled/not_signalled)
+ *
+ ******************************************************************************/
+acpi_status
+acpi_get_gpe_status(acpi_handle gpe_device,
+ u32 gpe_number, acpi_event_status *event_status)
+{
+ acpi_status status = AE_OK;
+ struct acpi_gpe_event_info *gpe_event_info;
+ acpi_cpu_flags flags;
+
+ ACPI_FUNCTION_TRACE(acpi_get_gpe_status);
+
+ flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock);
+
+ /* Ensure that we have a valid GPE number */
+
+ gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number);
+ if (!gpe_event_info) {
+ status = AE_BAD_PARAMETER;
+ goto unlock_and_exit;
+ }
+
+ /* Obtain status on the requested GPE number */
+
+ status = acpi_hw_get_gpe_status(gpe_event_info, event_status);
+
+ if (gpe_event_info->flags & ACPI_GPE_DISPATCH_MASK)
+ *event_status |= ACPI_EVENT_FLAG_HANDLE;
+
+ unlock_and_exit:
+ acpi_os_release_lock(acpi_gbl_gpe_lock, flags);
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_get_gpe_status)
+
+/******************************************************************************
+ *
+ * FUNCTION: acpi_disable_all_gpes
+ *
+ * PARAMETERS: None
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Disable and clear all GPEs in all GPE blocks
+ *
+ ******************************************************************************/
+
+acpi_status acpi_disable_all_gpes(void)
+{
+ acpi_status status;
+
+ ACPI_FUNCTION_TRACE(acpi_disable_all_gpes);
+
+ status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
+ if (ACPI_FAILURE(status)) {
+ return_ACPI_STATUS(status);
+ }
+
+ status = acpi_hw_disable_all_gpes();
+ (void)acpi_ut_release_mutex(ACPI_MTX_EVENTS);
+
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_disable_all_gpes)
+
+/******************************************************************************
+ *
+ * FUNCTION: acpi_enable_all_runtime_gpes
+ *
+ * PARAMETERS: None
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Enable all "runtime" GPEs, in all GPE blocks
+ *
+ ******************************************************************************/
+
+acpi_status acpi_enable_all_runtime_gpes(void)
+{
+ acpi_status status;
+
+ ACPI_FUNCTION_TRACE(acpi_enable_all_runtime_gpes);
+
+ status = acpi_ut_acquire_mutex(ACPI_MTX_EVENTS);
+ if (ACPI_FAILURE(status)) {
+ return_ACPI_STATUS(status);
+ }
+
+ status = acpi_hw_enable_all_runtime_gpes();
+ (void)acpi_ut_release_mutex(ACPI_MTX_EVENTS);
+
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_enable_all_runtime_gpes)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_install_gpe_block
+ *
+ * PARAMETERS: gpe_device - Handle to the parent GPE Block Device
+ * gpe_block_address - Address and space_iD
+ * register_count - Number of GPE register pairs in the block
+ * interrupt_number - H/W interrupt for the block
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Create and Install a block of GPE registers. The GPEs are not
+ * enabled here.
+ *
+ ******************************************************************************/
+acpi_status
+acpi_install_gpe_block(acpi_handle gpe_device,
+ struct acpi_generic_address *gpe_block_address,
+ u32 register_count, u32 interrupt_number)
+{
+ acpi_status status;
+ union acpi_operand_object *obj_desc;
+ struct acpi_namespace_node *node;
+ struct acpi_gpe_block_info *gpe_block;
+
+ ACPI_FUNCTION_TRACE(acpi_install_gpe_block);
+
+ if ((!gpe_device) || (!gpe_block_address) || (!register_count)) {
+ return_ACPI_STATUS(AE_BAD_PARAMETER);
+ }
+
+ status = acpi_ut_acquire_mutex(ACPI_MTX_NAMESPACE);
+ if (ACPI_FAILURE(status)) {
+ return (status);
+ }
+
+ node = acpi_ns_validate_handle(gpe_device);
+ if (!node) {
+ status = AE_BAD_PARAMETER;
+ goto unlock_and_exit;
+ }
+
+ /*
+ * For user-installed GPE Block Devices, the gpe_block_base_number
+ * is always zero
+ */
+ status =
+ acpi_ev_create_gpe_block(node, gpe_block_address, register_count, 0,
+ interrupt_number, &gpe_block);
+ if (ACPI_FAILURE(status)) {
+ goto unlock_and_exit;
+ }
+
+ /* Install block in the device_object attached to the node */
+
+ obj_desc = acpi_ns_get_attached_object(node);
+ if (!obj_desc) {
+
+ /*
+ * No object, create a new one (Device nodes do not always have
+ * an attached object)
+ */
+ obj_desc = acpi_ut_create_internal_object(ACPI_TYPE_DEVICE);
+ if (!obj_desc) {
+ status = AE_NO_MEMORY;
+ goto unlock_and_exit;
+ }
+
+ status =
+ acpi_ns_attach_object(node, obj_desc, ACPI_TYPE_DEVICE);
+
+ /* Remove local reference to the object */
+
+ acpi_ut_remove_reference(obj_desc);
+
+ if (ACPI_FAILURE(status)) {
+ goto unlock_and_exit;
+ }
+ }
+
+ /* Now install the GPE block in the device_object */
+
+ obj_desc->device.gpe_block = gpe_block;
+
+ unlock_and_exit:
+ (void)acpi_ut_release_mutex(ACPI_MTX_NAMESPACE);
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_install_gpe_block)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_remove_gpe_block
+ *
+ * PARAMETERS: gpe_device - Handle to the parent GPE Block Device
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Remove a previously installed block of GPE registers
+ *
+ ******************************************************************************/
+acpi_status acpi_remove_gpe_block(acpi_handle gpe_device)
+{
+ union acpi_operand_object *obj_desc;
+ acpi_status status;
+ struct acpi_namespace_node *node;
+
+ ACPI_FUNCTION_TRACE(acpi_remove_gpe_block);
+
+ if (!gpe_device) {
+ return_ACPI_STATUS(AE_BAD_PARAMETER);
+ }
+
+ status = acpi_ut_acquire_mutex(ACPI_MTX_NAMESPACE);
+ if (ACPI_FAILURE(status)) {
+ return (status);
+ }
+
+ node = acpi_ns_validate_handle(gpe_device);
+ if (!node) {
+ status = AE_BAD_PARAMETER;
+ goto unlock_and_exit;
+ }
+
+ /* Get the device_object attached to the node */
+
+ obj_desc = acpi_ns_get_attached_object(node);
+ if (!obj_desc || !obj_desc->device.gpe_block) {
+ return_ACPI_STATUS(AE_NULL_OBJECT);
+ }
+
+ /* Delete the GPE block (but not the device_object) */
+
+ status = acpi_ev_delete_gpe_block(obj_desc->device.gpe_block);
+ if (ACPI_SUCCESS(status)) {
+ obj_desc->device.gpe_block = NULL;
+ }
+
+ unlock_and_exit:
+ (void)acpi_ut_release_mutex(ACPI_MTX_NAMESPACE);
+ return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_remove_gpe_block)
+
+/*******************************************************************************
+ *
+ * FUNCTION: acpi_get_gpe_device
+ *
+ * PARAMETERS: Index - System GPE index (0-current_gpe_count)
+ * gpe_device - Where the parent GPE Device is returned
+ *
+ * RETURN: Status
+ *
+ * DESCRIPTION: Obtain the GPE device associated with the input index. A NULL
+ * gpe device indicates that the gpe number is contained in one of
+ * the FADT-defined gpe blocks. Otherwise, the GPE block device.
+ *
+ ******************************************************************************/
+acpi_status
+acpi_get_gpe_device(u32 index, acpi_handle *gpe_device)
+{
+ struct acpi_gpe_device_info info;
+ acpi_status status;
+
+ ACPI_FUNCTION_TRACE(acpi_get_gpe_device);
+
+ if (!gpe_device) {
+ return_ACPI_STATUS(AE_BAD_PARAMETER);
+ }
+
+ if (index >= acpi_current_gpe_count) {
+ return_ACPI_STATUS(AE_NOT_EXIST);
+ }
+
+ /* Setup and walk the GPE list */
+
+ info.index = index;
+ info.status = AE_NOT_EXIST;
+ info.gpe_device = NULL;
+ info.next_block_base_index = 0;
+
+ status = acpi_ev_walk_gpe_list(acpi_ev_get_gpe_device, &info);
+ if (ACPI_FAILURE(status)) {
+ return_ACPI_STATUS(status);
+ }
+
+ *gpe_device = ACPI_CAST_PTR(acpi_handle, info.gpe_device);
+ return_ACPI_STATUS(info.status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_get_gpe_device)
diff --git a/drivers/acpi/acpica/hwgpe.c b/drivers/acpi/acpica/hwgpe.c
index 14750db..85c3cbd 100644
--- a/drivers/acpi/acpica/hwgpe.c
+++ b/drivers/acpi/acpica/hwgpe.c
@@ -62,10 +62,10 @@
* PARAMETERS: gpe_event_info - Info block for the GPE
* gpe_register_info - Info block for the GPE register
*
- * RETURN: Status
+ * RETURN: Register mask with a one in the GPE bit position
*
- * DESCRIPTION: Compute GPE enable mask with one bit corresponding to the given
- * GPE set.
+ * DESCRIPTION: Compute the register mask for this GPE. One bit is set in the
+ * correct position for the input GPE.
*
******************************************************************************/
@@ -85,12 +85,12 @@
*
* RETURN: Status
*
- * DESCRIPTION: Enable or disable a single GPE in its enable register.
+ * DESCRIPTION: Enable or disable a single GPE in the parent enable register.
*
******************************************************************************/
acpi_status
-acpi_hw_low_set_gpe(struct acpi_gpe_event_info *gpe_event_info, u8 action)
+acpi_hw_low_set_gpe(struct acpi_gpe_event_info *gpe_event_info, u32 action)
{
struct acpi_gpe_register_info *gpe_register_info;
acpi_status status;
@@ -113,14 +113,20 @@
return (status);
}
- /* Set ot clear just the bit that corresponds to this GPE */
+ /* Set or clear just the bit that corresponds to this GPE */
register_bit = acpi_hw_get_gpe_register_bit(gpe_event_info,
gpe_register_info);
switch (action) {
- case ACPI_GPE_COND_ENABLE:
- if (!(register_bit & gpe_register_info->enable_for_run))
+ case ACPI_GPE_CONDITIONAL_ENABLE:
+
+ /* Only enable if the enable_for_run bit is set */
+
+ if (!(register_bit & gpe_register_info->enable_for_run)) {
return (AE_BAD_PARAMETER);
+ }
+
+ /*lint -fallthrough */
case ACPI_GPE_ENABLE:
ACPI_SET_BIT(enable_mask, register_bit);
@@ -131,7 +137,7 @@
break;
default:
- ACPI_ERROR((AE_INFO, "Invalid action\n"));
+ ACPI_ERROR((AE_INFO, "Invalid GPE Action, %u\n", action));
return (AE_BAD_PARAMETER);
}
@@ -168,13 +174,13 @@
return (AE_NOT_EXIST);
}
- register_bit = acpi_hw_get_gpe_register_bit(gpe_event_info,
- gpe_register_info);
-
/*
* Write a one to the appropriate bit in the status register to
* clear this GPE.
*/
+ register_bit =
+ acpi_hw_get_gpe_register_bit(gpe_event_info, gpe_register_info);
+
status = acpi_hw_write(register_bit,
&gpe_register_info->status_address);
@@ -201,8 +207,8 @@
u32 in_byte;
u32 register_bit;
struct acpi_gpe_register_info *gpe_register_info;
- acpi_status status;
acpi_event_status local_event_status = 0;
+ acpi_status status;
ACPI_FUNCTION_ENTRY();
diff --git a/drivers/acpi/acpica/utglobal.c b/drivers/acpi/acpica/utglobal.c
index e87bc67..508537f 100644
--- a/drivers/acpi/acpica/utglobal.c
+++ b/drivers/acpi/acpica/utglobal.c
@@ -768,7 +768,7 @@
acpi_gbl_gpe_fadt_blocks[0] = NULL;
acpi_gbl_gpe_fadt_blocks[1] = NULL;
acpi_current_gpe_count = 0;
- acpi_all_gpes_initialized = FALSE;
+ acpi_gbl_all_gpes_initialized = FALSE;
/* Global handlers */
@@ -778,6 +778,7 @@
acpi_gbl_init_handler = NULL;
acpi_gbl_table_handler = NULL;
acpi_gbl_interface_handler = NULL;
+ acpi_gbl_global_event_handler = NULL;
/* Global Lock support */
diff --git a/drivers/acpi/acpica/utmutex.c b/drivers/acpi/acpica/utmutex.c
index d9efa49..199528f 100644
--- a/drivers/acpi/acpica/utmutex.c
+++ b/drivers/acpi/acpica/utmutex.c
@@ -85,6 +85,7 @@
spin_lock_init(acpi_gbl_gpe_lock);
spin_lock_init(acpi_gbl_hardware_lock);
+ spin_lock_init(acpi_ev_global_lock_pending_lock);
/* Mutex for _OSI support */
status = acpi_os_create_mutex(&acpi_gbl_osi_mutex);
diff --git a/drivers/acpi/apei/apei-internal.h b/drivers/acpi/apei/apei-internal.h
index 18df1e9..ef0581f 100644
--- a/drivers/acpi/apei/apei-internal.h
+++ b/drivers/acpi/apei/apei-internal.h
@@ -109,6 +109,8 @@
return sizeof(*estatus) + estatus->data_length;
}
+void apei_estatus_print(const char *pfx,
+ const struct acpi_hest_generic_status *estatus);
int apei_estatus_check_header(const struct acpi_hest_generic_status *estatus);
int apei_estatus_check(const struct acpi_hest_generic_status *estatus);
#endif
diff --git a/drivers/acpi/apei/cper.c b/drivers/acpi/apei/cper.c
index f4cf2fc..31464a0 100644
--- a/drivers/acpi/apei/cper.c
+++ b/drivers/acpi/apei/cper.c
@@ -46,6 +46,317 @@
}
EXPORT_SYMBOL_GPL(cper_next_record_id);
+static const char *cper_severity_strs[] = {
+ "recoverable",
+ "fatal",
+ "corrected",
+ "info",
+};
+
+static const char *cper_severity_str(unsigned int severity)
+{
+ return severity < ARRAY_SIZE(cper_severity_strs) ?
+ cper_severity_strs[severity] : "unknown";
+}
+
+/*
+ * cper_print_bits - print strings for set bits
+ * @pfx: prefix for each line, including log level and prefix string
+ * @bits: bit mask
+ * @strs: string array, indexed by bit position
+ * @strs_size: size of the string array: @strs
+ *
+ * For each set bit in @bits, print the corresponding string in @strs.
+ * If the output length is longer than 80, multiple line will be
+ * printed, with @pfx is printed at the beginning of each line.
+ */
+static void cper_print_bits(const char *pfx, unsigned int bits,
+ const char *strs[], unsigned int strs_size)
+{
+ int i, len = 0;
+ const char *str;
+ char buf[84];
+
+ for (i = 0; i < strs_size; i++) {
+ if (!(bits & (1U << i)))
+ continue;
+ str = strs[i];
+ if (len && len + strlen(str) + 2 > 80) {
+ printk("%s\n", buf);
+ len = 0;
+ }
+ if (!len)
+ len = snprintf(buf, sizeof(buf), "%s%s", pfx, str);
+ else
+ len += snprintf(buf+len, sizeof(buf)-len, ", %s", str);
+ }
+ if (len)
+ printk("%s\n", buf);
+}
+
+static const char *cper_proc_type_strs[] = {
+ "IA32/X64",
+ "IA64",
+};
+
+static const char *cper_proc_isa_strs[] = {
+ "IA32",
+ "IA64",
+ "X64",
+};
+
+static const char *cper_proc_error_type_strs[] = {
+ "cache error",
+ "TLB error",
+ "bus error",
+ "micro-architectural error",
+};
+
+static const char *cper_proc_op_strs[] = {
+ "unknown or generic",
+ "data read",
+ "data write",
+ "instruction execution",
+};
+
+static const char *cper_proc_flag_strs[] = {
+ "restartable",
+ "precise IP",
+ "overflow",
+ "corrected",
+};
+
+static void cper_print_proc_generic(const char *pfx,
+ const struct cper_sec_proc_generic *proc)
+{
+ if (proc->validation_bits & CPER_PROC_VALID_TYPE)
+ printk("%s""processor_type: %d, %s\n", pfx, proc->proc_type,
+ proc->proc_type < ARRAY_SIZE(cper_proc_type_strs) ?
+ cper_proc_type_strs[proc->proc_type] : "unknown");
+ if (proc->validation_bits & CPER_PROC_VALID_ISA)
+ printk("%s""processor_isa: %d, %s\n", pfx, proc->proc_isa,
+ proc->proc_isa < ARRAY_SIZE(cper_proc_isa_strs) ?
+ cper_proc_isa_strs[proc->proc_isa] : "unknown");
+ if (proc->validation_bits & CPER_PROC_VALID_ERROR_TYPE) {
+ printk("%s""error_type: 0x%02x\n", pfx, proc->proc_error_type);
+ cper_print_bits(pfx, proc->proc_error_type,
+ cper_proc_error_type_strs,
+ ARRAY_SIZE(cper_proc_error_type_strs));
+ }
+ if (proc->validation_bits & CPER_PROC_VALID_OPERATION)
+ printk("%s""operation: %d, %s\n", pfx, proc->operation,
+ proc->operation < ARRAY_SIZE(cper_proc_op_strs) ?
+ cper_proc_op_strs[proc->operation] : "unknown");
+ if (proc->validation_bits & CPER_PROC_VALID_FLAGS) {
+ printk("%s""flags: 0x%02x\n", pfx, proc->flags);
+ cper_print_bits(pfx, proc->flags, cper_proc_flag_strs,
+ ARRAY_SIZE(cper_proc_flag_strs));
+ }
+ if (proc->validation_bits & CPER_PROC_VALID_LEVEL)
+ printk("%s""level: %d\n", pfx, proc->level);
+ if (proc->validation_bits & CPER_PROC_VALID_VERSION)
+ printk("%s""version_info: 0x%016llx\n", pfx, proc->cpu_version);
+ if (proc->validation_bits & CPER_PROC_VALID_ID)
+ printk("%s""processor_id: 0x%016llx\n", pfx, proc->proc_id);
+ if (proc->validation_bits & CPER_PROC_VALID_TARGET_ADDRESS)
+ printk("%s""target_address: 0x%016llx\n",
+ pfx, proc->target_addr);
+ if (proc->validation_bits & CPER_PROC_VALID_REQUESTOR_ID)
+ printk("%s""requestor_id: 0x%016llx\n",
+ pfx, proc->requestor_id);
+ if (proc->validation_bits & CPER_PROC_VALID_RESPONDER_ID)
+ printk("%s""responder_id: 0x%016llx\n",
+ pfx, proc->responder_id);
+ if (proc->validation_bits & CPER_PROC_VALID_IP)
+ printk("%s""IP: 0x%016llx\n", pfx, proc->ip);
+}
+
+static const char *cper_mem_err_type_strs[] = {
+ "unknown",
+ "no error",
+ "single-bit ECC",
+ "multi-bit ECC",
+ "single-symbol chipkill ECC",
+ "multi-symbol chipkill ECC",
+ "master abort",
+ "target abort",
+ "parity error",
+ "watchdog timeout",
+ "invalid address",
+ "mirror Broken",
+ "memory sparing",
+ "scrub corrected error",
+ "scrub uncorrected error",
+};
+
+static void cper_print_mem(const char *pfx, const struct cper_sec_mem_err *mem)
+{
+ if (mem->validation_bits & CPER_MEM_VALID_ERROR_STATUS)
+ printk("%s""error_status: 0x%016llx\n", pfx, mem->error_status);
+ if (mem->validation_bits & CPER_MEM_VALID_PHYSICAL_ADDRESS)
+ printk("%s""physical_address: 0x%016llx\n",
+ pfx, mem->physical_addr);
+ if (mem->validation_bits & CPER_MEM_VALID_PHYSICAL_ADDRESS_MASK)
+ printk("%s""physical_address_mask: 0x%016llx\n",
+ pfx, mem->physical_addr_mask);
+ if (mem->validation_bits & CPER_MEM_VALID_NODE)
+ printk("%s""node: %d\n", pfx, mem->node);
+ if (mem->validation_bits & CPER_MEM_VALID_CARD)
+ printk("%s""card: %d\n", pfx, mem->card);
+ if (mem->validation_bits & CPER_MEM_VALID_MODULE)
+ printk("%s""module: %d\n", pfx, mem->module);
+ if (mem->validation_bits & CPER_MEM_VALID_BANK)
+ printk("%s""bank: %d\n", pfx, mem->bank);
+ if (mem->validation_bits & CPER_MEM_VALID_DEVICE)
+ printk("%s""device: %d\n", pfx, mem->device);
+ if (mem->validation_bits & CPER_MEM_VALID_ROW)
+ printk("%s""row: %d\n", pfx, mem->row);
+ if (mem->validation_bits & CPER_MEM_VALID_COLUMN)
+ printk("%s""column: %d\n", pfx, mem->column);
+ if (mem->validation_bits & CPER_MEM_VALID_BIT_POSITION)
+ printk("%s""bit_position: %d\n", pfx, mem->bit_pos);
+ if (mem->validation_bits & CPER_MEM_VALID_REQUESTOR_ID)
+ printk("%s""requestor_id: 0x%016llx\n", pfx, mem->requestor_id);
+ if (mem->validation_bits & CPER_MEM_VALID_RESPONDER_ID)
+ printk("%s""responder_id: 0x%016llx\n", pfx, mem->responder_id);
+ if (mem->validation_bits & CPER_MEM_VALID_TARGET_ID)
+ printk("%s""target_id: 0x%016llx\n", pfx, mem->target_id);
+ if (mem->validation_bits & CPER_MEM_VALID_ERROR_TYPE) {
+ u8 etype = mem->error_type;
+ printk("%s""error_type: %d, %s\n", pfx, etype,
+ etype < ARRAY_SIZE(cper_mem_err_type_strs) ?
+ cper_mem_err_type_strs[etype] : "unknown");
+ }
+}
+
+static const char *cper_pcie_port_type_strs[] = {
+ "PCIe end point",
+ "legacy PCI end point",
+ "unknown",
+ "unknown",
+ "root port",
+ "upstream switch port",
+ "downstream switch port",
+ "PCIe to PCI/PCI-X bridge",
+ "PCI/PCI-X to PCIe bridge",
+ "root complex integrated endpoint device",
+ "root complex event collector",
+};
+
+static void cper_print_pcie(const char *pfx, const struct cper_sec_pcie *pcie)
+{
+ if (pcie->validation_bits & CPER_PCIE_VALID_PORT_TYPE)
+ printk("%s""port_type: %d, %s\n", pfx, pcie->port_type,
+ pcie->port_type < ARRAY_SIZE(cper_pcie_port_type_strs) ?
+ cper_pcie_port_type_strs[pcie->port_type] : "unknown");
+ if (pcie->validation_bits & CPER_PCIE_VALID_VERSION)
+ printk("%s""version: %d.%d\n", pfx,
+ pcie->version.major, pcie->version.minor);
+ if (pcie->validation_bits & CPER_PCIE_VALID_COMMAND_STATUS)
+ printk("%s""command: 0x%04x, status: 0x%04x\n", pfx,
+ pcie->command, pcie->status);
+ if (pcie->validation_bits & CPER_PCIE_VALID_DEVICE_ID) {
+ const __u8 *p;
+ printk("%s""device_id: %04x:%02x:%02x.%x\n", pfx,
+ pcie->device_id.segment, pcie->device_id.bus,
+ pcie->device_id.device, pcie->device_id.function);
+ printk("%s""slot: %d\n", pfx,
+ pcie->device_id.slot >> CPER_PCIE_SLOT_SHIFT);
+ printk("%s""secondary_bus: 0x%02x\n", pfx,
+ pcie->device_id.secondary_bus);
+ printk("%s""vendor_id: 0x%04x, device_id: 0x%04x\n", pfx,
+ pcie->device_id.vendor_id, pcie->device_id.device_id);
+ p = pcie->device_id.class_code;
+ printk("%s""class_code: %02x%02x%02x\n", pfx, p[0], p[1], p[2]);
+ }
+ if (pcie->validation_bits & CPER_PCIE_VALID_SERIAL_NUMBER)
+ printk("%s""serial number: 0x%04x, 0x%04x\n", pfx,
+ pcie->serial_number.lower, pcie->serial_number.upper);
+ if (pcie->validation_bits & CPER_PCIE_VALID_BRIDGE_CONTROL_STATUS)
+ printk(
+ "%s""bridge: secondary_status: 0x%04x, control: 0x%04x\n",
+ pfx, pcie->bridge.secondary_status, pcie->bridge.control);
+}
+
+static const char *apei_estatus_section_flag_strs[] = {
+ "primary",
+ "containment warning",
+ "reset",
+ "threshold exceeded",
+ "resource not accessible",
+ "latent error",
+};
+
+static void apei_estatus_print_section(
+ const char *pfx, const struct acpi_hest_generic_data *gdata, int sec_no)
+{
+ uuid_le *sec_type = (uuid_le *)gdata->section_type;
+ __u16 severity;
+
+ severity = gdata->error_severity;
+ printk("%s""section: %d, severity: %d, %s\n", pfx, sec_no, severity,
+ cper_severity_str(severity));
+ printk("%s""flags: 0x%02x\n", pfx, gdata->flags);
+ cper_print_bits(pfx, gdata->flags, apei_estatus_section_flag_strs,
+ ARRAY_SIZE(apei_estatus_section_flag_strs));
+ if (gdata->validation_bits & CPER_SEC_VALID_FRU_ID)
+ printk("%s""fru_id: %pUl\n", pfx, (uuid_le *)gdata->fru_id);
+ if (gdata->validation_bits & CPER_SEC_VALID_FRU_TEXT)
+ printk("%s""fru_text: %.20s\n", pfx, gdata->fru_text);
+
+ if (!uuid_le_cmp(*sec_type, CPER_SEC_PROC_GENERIC)) {
+ struct cper_sec_proc_generic *proc_err = (void *)(gdata + 1);
+ printk("%s""section_type: general processor error\n", pfx);
+ if (gdata->error_data_length >= sizeof(*proc_err))
+ cper_print_proc_generic(pfx, proc_err);
+ else
+ goto err_section_too_small;
+ } else if (!uuid_le_cmp(*sec_type, CPER_SEC_PLATFORM_MEM)) {
+ struct cper_sec_mem_err *mem_err = (void *)(gdata + 1);
+ printk("%s""section_type: memory error\n", pfx);
+ if (gdata->error_data_length >= sizeof(*mem_err))
+ cper_print_mem(pfx, mem_err);
+ else
+ goto err_section_too_small;
+ } else if (!uuid_le_cmp(*sec_type, CPER_SEC_PCIE)) {
+ struct cper_sec_pcie *pcie = (void *)(gdata + 1);
+ printk("%s""section_type: PCIe error\n", pfx);
+ if (gdata->error_data_length >= sizeof(*pcie))
+ cper_print_pcie(pfx, pcie);
+ else
+ goto err_section_too_small;
+ } else
+ printk("%s""section type: unknown, %pUl\n", pfx, sec_type);
+
+ return;
+
+err_section_too_small:
+ pr_err(FW_WARN "error section length is too small\n");
+}
+
+void apei_estatus_print(const char *pfx,
+ const struct acpi_hest_generic_status *estatus)
+{
+ struct acpi_hest_generic_data *gdata;
+ unsigned int data_len, gedata_len;
+ int sec_no = 0;
+ __u16 severity;
+
+ printk("%s""APEI generic hardware error status\n", pfx);
+ severity = estatus->error_severity;
+ printk("%s""severity: %d, %s\n", pfx, severity,
+ cper_severity_str(severity));
+ data_len = estatus->data_length;
+ gdata = (struct acpi_hest_generic_data *)(estatus + 1);
+ while (data_len > sizeof(*gdata)) {
+ gedata_len = gdata->error_data_length;
+ apei_estatus_print_section(pfx, gdata, sec_no);
+ data_len -= gedata_len + sizeof(*gdata);
+ sec_no++;
+ }
+}
+EXPORT_SYMBOL_GPL(apei_estatus_print);
+
int apei_estatus_check_header(const struct acpi_hest_generic_status *estatus)
{
if (estatus->data_length &&
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 0d505e5..d1d484d 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -12,10 +12,6 @@
* For more information about Generic Hardware Error Source, please
* refer to ACPI Specification version 4.0, section 17.3.2.6
*
- * Now, only SCI notification type and memory errors are
- * supported. More notification type and hardware error type will be
- * added later.
- *
* Copyright 2010 Intel Corp.
* Author: Huang Ying <ying.huang@intel.com>
*
@@ -39,14 +35,18 @@
#include <linux/acpi.h>
#include <linux/io.h>
#include <linux/interrupt.h>
+#include <linux/timer.h>
#include <linux/cper.h>
#include <linux/kdebug.h>
#include <linux/platform_device.h>
#include <linux/mutex.h>
+#include <linux/ratelimit.h>
+#include <linux/vmalloc.h>
#include <acpi/apei.h>
#include <acpi/atomicio.h>
#include <acpi/hed.h>
#include <asm/mce.h>
+#include <asm/tlbflush.h>
#include "apei-internal.h"
@@ -55,42 +55,131 @@
#define GHES_ESTATUS_MAX_SIZE 65536
/*
- * One struct ghes is created for each generic hardware error
- * source.
- *
+ * One struct ghes is created for each generic hardware error source.
* It provides the context for APEI hardware error timer/IRQ/SCI/NMI
- * handler. Handler for one generic hardware error source is only
- * triggered after the previous one is done. So handler can uses
- * struct ghes without locking.
+ * handler.
*
* estatus: memory buffer for error status block, allocated during
* HEST parsing.
*/
#define GHES_TO_CLEAR 0x0001
+#define GHES_EXITING 0x0002
struct ghes {
struct acpi_hest_generic *generic;
struct acpi_hest_generic_status *estatus;
- struct list_head list;
u64 buffer_paddr;
unsigned long flags;
+ union {
+ struct list_head list;
+ struct timer_list timer;
+ unsigned int irq;
+ };
};
+static int ghes_panic_timeout __read_mostly = 30;
+
/*
- * Error source lists, one list for each notification method. The
- * members in lists are struct ghes.
+ * All error sources notified with SCI shares one notifier function,
+ * so they need to be linked and checked one by one. This is applied
+ * to NMI too.
*
- * The list members are only added in HEST parsing and deleted during
- * module_exit, that is, single-threaded. So no lock is needed for
- * that.
- *
- * But the mutual exclusion is needed between members adding/deleting
- * and timer/IRQ/SCI/NMI handler, which may traverse the list. RCU is
- * used for that.
+ * RCU is used for these lists, so ghes_list_mutex is only used for
+ * list changing, not for traversing.
*/
static LIST_HEAD(ghes_sci);
+static LIST_HEAD(ghes_nmi);
static DEFINE_MUTEX(ghes_list_mutex);
+/*
+ * NMI may be triggered on any CPU, so ghes_nmi_lock is used for
+ * mutual exclusion.
+ */
+static DEFINE_RAW_SPINLOCK(ghes_nmi_lock);
+
+/*
+ * Because the memory area used to transfer hardware error information
+ * from BIOS to Linux can be determined only in NMI, IRQ or timer
+ * handler, but general ioremap can not be used in atomic context, so
+ * a special version of atomic ioremap is implemented for that.
+ */
+
+/*
+ * Two virtual pages are used, one for NMI context, the other for
+ * IRQ/PROCESS context
+ */
+#define GHES_IOREMAP_PAGES 2
+#define GHES_IOREMAP_NMI_PAGE(base) (base)
+#define GHES_IOREMAP_IRQ_PAGE(base) ((base) + PAGE_SIZE)
+
+/* virtual memory area for atomic ioremap */
+static struct vm_struct *ghes_ioremap_area;
+/*
+ * These 2 spinlock is used to prevent atomic ioremap virtual memory
+ * area from being mapped simultaneously.
+ */
+static DEFINE_RAW_SPINLOCK(ghes_ioremap_lock_nmi);
+static DEFINE_SPINLOCK(ghes_ioremap_lock_irq);
+
+static int ghes_ioremap_init(void)
+{
+ ghes_ioremap_area = __get_vm_area(PAGE_SIZE * GHES_IOREMAP_PAGES,
+ VM_IOREMAP, VMALLOC_START, VMALLOC_END);
+ if (!ghes_ioremap_area) {
+ pr_err(GHES_PFX "Failed to allocate virtual memory area for atomic ioremap.\n");
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
+static void ghes_ioremap_exit(void)
+{
+ free_vm_area(ghes_ioremap_area);
+}
+
+static void __iomem *ghes_ioremap_pfn_nmi(u64 pfn)
+{
+ unsigned long vaddr;
+
+ vaddr = (unsigned long)GHES_IOREMAP_NMI_PAGE(ghes_ioremap_area->addr);
+ ioremap_page_range(vaddr, vaddr + PAGE_SIZE,
+ pfn << PAGE_SHIFT, PAGE_KERNEL);
+
+ return (void __iomem *)vaddr;
+}
+
+static void __iomem *ghes_ioremap_pfn_irq(u64 pfn)
+{
+ unsigned long vaddr;
+
+ vaddr = (unsigned long)GHES_IOREMAP_IRQ_PAGE(ghes_ioremap_area->addr);
+ ioremap_page_range(vaddr, vaddr + PAGE_SIZE,
+ pfn << PAGE_SHIFT, PAGE_KERNEL);
+
+ return (void __iomem *)vaddr;
+}
+
+static void ghes_iounmap_nmi(void __iomem *vaddr_ptr)
+{
+ unsigned long vaddr = (unsigned long __force)vaddr_ptr;
+ void *base = ghes_ioremap_area->addr;
+
+ BUG_ON(vaddr != (unsigned long)GHES_IOREMAP_NMI_PAGE(base));
+ unmap_kernel_range_noflush(vaddr, PAGE_SIZE);
+ __flush_tlb_one(vaddr);
+}
+
+static void ghes_iounmap_irq(void __iomem *vaddr_ptr)
+{
+ unsigned long vaddr = (unsigned long __force)vaddr_ptr;
+ void *base = ghes_ioremap_area->addr;
+
+ BUG_ON(vaddr != (unsigned long)GHES_IOREMAP_IRQ_PAGE(base));
+ unmap_kernel_range_noflush(vaddr, PAGE_SIZE);
+ __flush_tlb_one(vaddr);
+}
+
static struct ghes *ghes_new(struct acpi_hest_generic *generic)
{
struct ghes *ghes;
@@ -101,7 +190,6 @@
if (!ghes)
return ERR_PTR(-ENOMEM);
ghes->generic = generic;
- INIT_LIST_HEAD(&ghes->list);
rc = acpi_pre_map_gar(&generic->error_status_address);
if (rc)
goto err_free;
@@ -158,22 +246,41 @@
}
}
-/* SCI handler run in work queue, so ioremap can be used here */
-static int ghes_copy_tofrom_phys(void *buffer, u64 paddr, u32 len,
- int from_phys)
+static void ghes_copy_tofrom_phys(void *buffer, u64 paddr, u32 len,
+ int from_phys)
{
- void *vaddr;
+ void __iomem *vaddr;
+ unsigned long flags = 0;
+ int in_nmi = in_nmi();
+ u64 offset;
+ u32 trunk;
- vaddr = ioremap_cache(paddr, len);
- if (!vaddr)
- return -ENOMEM;
- if (from_phys)
- memcpy(buffer, vaddr, len);
- else
- memcpy(vaddr, buffer, len);
- iounmap(vaddr);
-
- return 0;
+ while (len > 0) {
+ offset = paddr - (paddr & PAGE_MASK);
+ if (in_nmi) {
+ raw_spin_lock(&ghes_ioremap_lock_nmi);
+ vaddr = ghes_ioremap_pfn_nmi(paddr >> PAGE_SHIFT);
+ } else {
+ spin_lock_irqsave(&ghes_ioremap_lock_irq, flags);
+ vaddr = ghes_ioremap_pfn_irq(paddr >> PAGE_SHIFT);
+ }
+ trunk = PAGE_SIZE - offset;
+ trunk = min(trunk, len);
+ if (from_phys)
+ memcpy_fromio(buffer, vaddr + offset, trunk);
+ else
+ memcpy_toio(vaddr + offset, buffer, trunk);
+ len -= trunk;
+ paddr += trunk;
+ buffer += trunk;
+ if (in_nmi) {
+ ghes_iounmap_nmi(vaddr);
+ raw_spin_unlock(&ghes_ioremap_lock_nmi);
+ } else {
+ ghes_iounmap_irq(vaddr);
+ spin_unlock_irqrestore(&ghes_ioremap_lock_irq, flags);
+ }
+ }
}
static int ghes_read_estatus(struct ghes *ghes, int silent)
@@ -194,10 +301,8 @@
if (!buf_paddr)
return -ENOENT;
- rc = ghes_copy_tofrom_phys(ghes->estatus, buf_paddr,
- sizeof(*ghes->estatus), 1);
- if (rc)
- return rc;
+ ghes_copy_tofrom_phys(ghes->estatus, buf_paddr,
+ sizeof(*ghes->estatus), 1);
if (!ghes->estatus->block_status)
return -ENOENT;
@@ -212,17 +317,15 @@
goto err_read_block;
if (apei_estatus_check_header(ghes->estatus))
goto err_read_block;
- rc = ghes_copy_tofrom_phys(ghes->estatus + 1,
- buf_paddr + sizeof(*ghes->estatus),
- len - sizeof(*ghes->estatus), 1);
- if (rc)
- return rc;
+ ghes_copy_tofrom_phys(ghes->estatus + 1,
+ buf_paddr + sizeof(*ghes->estatus),
+ len - sizeof(*ghes->estatus), 1);
if (apei_estatus_check(ghes->estatus))
goto err_read_block;
rc = 0;
err_read_block:
- if (rc && !silent)
+ if (rc && !silent && printk_ratelimit())
pr_warning(FW_WARN GHES_PFX
"Failed to read error status block!\n");
return rc;
@@ -255,11 +358,26 @@
}
#endif
}
+}
- if (!processed && printk_ratelimit())
- pr_warning(GHES_PFX
- "Unknown error record from generic hardware error source: %d\n",
- ghes->generic->header.source_id);
+static void ghes_print_estatus(const char *pfx, struct ghes *ghes)
+{
+ /* Not more than 2 messages every 5 seconds */
+ static DEFINE_RATELIMIT_STATE(ratelimit, 5*HZ, 2);
+
+ if (pfx == NULL) {
+ if (ghes_severity(ghes->estatus->error_severity) <=
+ GHES_SEV_CORRECTED)
+ pfx = KERN_WARNING HW_ERR;
+ else
+ pfx = KERN_ERR HW_ERR;
+ }
+ if (__ratelimit(&ratelimit)) {
+ printk(
+ "%s""Hardware error from APEI Generic Hardware Error Source: %d\n",
+ pfx, ghes->generic->header.source_id);
+ apei_estatus_print(pfx, ghes->estatus);
+ }
}
static int ghes_proc(struct ghes *ghes)
@@ -269,6 +387,7 @@
rc = ghes_read_estatus(ghes, 0);
if (rc)
goto out;
+ ghes_print_estatus(NULL, ghes);
ghes_do_proc(ghes);
out:
@@ -276,6 +395,42 @@
return 0;
}
+static void ghes_add_timer(struct ghes *ghes)
+{
+ struct acpi_hest_generic *g = ghes->generic;
+ unsigned long expire;
+
+ if (!g->notify.poll_interval) {
+ pr_warning(FW_WARN GHES_PFX "Poll interval is 0 for generic hardware error source: %d, disabled.\n",
+ g->header.source_id);
+ return;
+ }
+ expire = jiffies + msecs_to_jiffies(g->notify.poll_interval);
+ ghes->timer.expires = round_jiffies_relative(expire);
+ add_timer(&ghes->timer);
+}
+
+static void ghes_poll_func(unsigned long data)
+{
+ struct ghes *ghes = (void *)data;
+
+ ghes_proc(ghes);
+ if (!(ghes->flags & GHES_EXITING))
+ ghes_add_timer(ghes);
+}
+
+static irqreturn_t ghes_irq_func(int irq, void *data)
+{
+ struct ghes *ghes = data;
+ int rc;
+
+ rc = ghes_proc(ghes);
+ if (rc)
+ return IRQ_NONE;
+
+ return IRQ_HANDLED;
+}
+
static int ghes_notify_sci(struct notifier_block *this,
unsigned long event, void *data)
{
@@ -292,10 +447,63 @@
return ret;
}
+static int ghes_notify_nmi(struct notifier_block *this,
+ unsigned long cmd, void *data)
+{
+ struct ghes *ghes, *ghes_global = NULL;
+ int sev, sev_global = -1;
+ int ret = NOTIFY_DONE;
+
+ if (cmd != DIE_NMI)
+ return ret;
+
+ raw_spin_lock(&ghes_nmi_lock);
+ list_for_each_entry_rcu(ghes, &ghes_nmi, list) {
+ if (ghes_read_estatus(ghes, 1)) {
+ ghes_clear_estatus(ghes);
+ continue;
+ }
+ sev = ghes_severity(ghes->estatus->error_severity);
+ if (sev > sev_global) {
+ sev_global = sev;
+ ghes_global = ghes;
+ }
+ ret = NOTIFY_STOP;
+ }
+
+ if (ret == NOTIFY_DONE)
+ goto out;
+
+ if (sev_global >= GHES_SEV_PANIC) {
+ oops_begin();
+ ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global);
+ /* reboot to log the error! */
+ if (panic_timeout == 0)
+ panic_timeout = ghes_panic_timeout;
+ panic("Fatal hardware error!");
+ }
+
+ list_for_each_entry_rcu(ghes, &ghes_nmi, list) {
+ if (!(ghes->flags & GHES_TO_CLEAR))
+ continue;
+ /* Do not print estatus because printk is not NMI safe */
+ ghes_do_proc(ghes);
+ ghes_clear_estatus(ghes);
+ }
+
+out:
+ raw_spin_unlock(&ghes_nmi_lock);
+ return ret;
+}
+
static struct notifier_block ghes_notifier_sci = {
.notifier_call = ghes_notify_sci,
};
+static struct notifier_block ghes_notifier_nmi = {
+ .notifier_call = ghes_notify_nmi,
+};
+
static int __devinit ghes_probe(struct platform_device *ghes_dev)
{
struct acpi_hest_generic *generic;
@@ -306,18 +514,27 @@
if (!generic->enabled)
return -ENODEV;
- if (generic->error_block_length <
- sizeof(struct acpi_hest_generic_status)) {
- pr_warning(FW_BUG GHES_PFX
-"Invalid error block length: %u for generic hardware error source: %d\n",
- generic->error_block_length,
+ switch (generic->notify.type) {
+ case ACPI_HEST_NOTIFY_POLLED:
+ case ACPI_HEST_NOTIFY_EXTERNAL:
+ case ACPI_HEST_NOTIFY_SCI:
+ case ACPI_HEST_NOTIFY_NMI:
+ break;
+ case ACPI_HEST_NOTIFY_LOCAL:
+ pr_warning(GHES_PFX "Generic hardware error source: %d notified via local interrupt is not supported!\n",
generic->header.source_id);
goto err;
+ default:
+ pr_warning(FW_WARN GHES_PFX "Unknown notification type: %u for generic hardware error source: %d\n",
+ generic->notify.type, generic->header.source_id);
+ goto err;
}
- if (generic->records_to_preallocate == 0) {
- pr_warning(FW_BUG GHES_PFX
-"Invalid records to preallocate: %u for generic hardware error source: %d\n",
- generic->records_to_preallocate,
+
+ rc = -EIO;
+ if (generic->error_block_length <
+ sizeof(struct acpi_hest_generic_status)) {
+ pr_warning(FW_BUG GHES_PFX "Invalid error block length: %u for generic hardware error source: %d\n",
+ generic->error_block_length,
generic->header.source_id);
goto err;
}
@@ -327,38 +544,43 @@
ghes = NULL;
goto err;
}
- if (generic->notify.type == ACPI_HEST_NOTIFY_SCI) {
+ switch (generic->notify.type) {
+ case ACPI_HEST_NOTIFY_POLLED:
+ ghes->timer.function = ghes_poll_func;
+ ghes->timer.data = (unsigned long)ghes;
+ init_timer_deferrable(&ghes->timer);
+ ghes_add_timer(ghes);
+ break;
+ case ACPI_HEST_NOTIFY_EXTERNAL:
+ /* External interrupt vector is GSI */
+ if (acpi_gsi_to_irq(generic->notify.vector, &ghes->irq)) {
+ pr_err(GHES_PFX "Failed to map GSI to IRQ for generic hardware error source: %d\n",
+ generic->header.source_id);
+ goto err;
+ }
+ if (request_irq(ghes->irq, ghes_irq_func,
+ 0, "GHES IRQ", ghes)) {
+ pr_err(GHES_PFX "Failed to register IRQ for generic hardware error source: %d\n",
+ generic->header.source_id);
+ goto err;
+ }
+ break;
+ case ACPI_HEST_NOTIFY_SCI:
mutex_lock(&ghes_list_mutex);
if (list_empty(&ghes_sci))
register_acpi_hed_notifier(&ghes_notifier_sci);
list_add_rcu(&ghes->list, &ghes_sci);
mutex_unlock(&ghes_list_mutex);
- } else {
- unsigned char *notify = NULL;
-
- switch (generic->notify.type) {
- case ACPI_HEST_NOTIFY_POLLED:
- notify = "POLL";
- break;
- case ACPI_HEST_NOTIFY_EXTERNAL:
- case ACPI_HEST_NOTIFY_LOCAL:
- notify = "IRQ";
- break;
- case ACPI_HEST_NOTIFY_NMI:
- notify = "NMI";
- break;
- }
- if (notify) {
- pr_warning(GHES_PFX
-"Generic hardware error source: %d notified via %s is not supported!\n",
- generic->header.source_id, notify);
- } else {
- pr_warning(FW_WARN GHES_PFX
-"Unknown notification type: %u for generic hardware error source: %d\n",
- generic->notify.type, generic->header.source_id);
- }
- rc = -ENODEV;
- goto err;
+ break;
+ case ACPI_HEST_NOTIFY_NMI:
+ mutex_lock(&ghes_list_mutex);
+ if (list_empty(&ghes_nmi))
+ register_die_notifier(&ghes_notifier_nmi);
+ list_add_rcu(&ghes->list, &ghes_nmi);
+ mutex_unlock(&ghes_list_mutex);
+ break;
+ default:
+ BUG();
}
platform_set_drvdata(ghes_dev, ghes);
@@ -379,7 +601,14 @@
ghes = platform_get_drvdata(ghes_dev);
generic = ghes->generic;
+ ghes->flags |= GHES_EXITING;
switch (generic->notify.type) {
+ case ACPI_HEST_NOTIFY_POLLED:
+ del_timer_sync(&ghes->timer);
+ break;
+ case ACPI_HEST_NOTIFY_EXTERNAL:
+ free_irq(ghes->irq, ghes);
+ break;
case ACPI_HEST_NOTIFY_SCI:
mutex_lock(&ghes_list_mutex);
list_del_rcu(&ghes->list);
@@ -387,12 +616,23 @@
unregister_acpi_hed_notifier(&ghes_notifier_sci);
mutex_unlock(&ghes_list_mutex);
break;
+ case ACPI_HEST_NOTIFY_NMI:
+ mutex_lock(&ghes_list_mutex);
+ list_del_rcu(&ghes->list);
+ if (list_empty(&ghes_nmi))
+ unregister_die_notifier(&ghes_notifier_nmi);
+ mutex_unlock(&ghes_list_mutex);
+ /*
+ * To synchronize with NMI handler, ghes can only be
+ * freed after NMI handler finishes.
+ */
+ synchronize_rcu();
+ break;
default:
BUG();
break;
}
- synchronize_rcu();
ghes_fini(ghes);
kfree(ghes);
@@ -412,6 +652,8 @@
static int __init ghes_init(void)
{
+ int rc;
+
if (acpi_disabled)
return -ENODEV;
@@ -420,12 +662,25 @@
return -EINVAL;
}
- return platform_driver_register(&ghes_platform_driver);
+ rc = ghes_ioremap_init();
+ if (rc)
+ goto err;
+
+ rc = platform_driver_register(&ghes_platform_driver);
+ if (rc)
+ goto err_ioremap_exit;
+
+ return 0;
+err_ioremap_exit:
+ ghes_ioremap_exit();
+err:
+ return rc;
}
static void __exit ghes_exit(void)
{
platform_driver_unregister(&ghes_platform_driver);
+ ghes_ioremap_exit();
}
module_init(ghes_init);
diff --git a/drivers/acpi/apei/hest.c b/drivers/acpi/apei/hest.c
index daa7bc6..abda378 100644
--- a/drivers/acpi/apei/hest.c
+++ b/drivers/acpi/apei/hest.c
@@ -195,24 +195,24 @@
__setup("hest_disable", setup_hest_disable);
-static int __init hest_init(void)
+void __init acpi_hest_init(void)
{
acpi_status status;
int rc = -ENODEV;
unsigned int ghes_count = 0;
+ if (hest_disable) {
+ pr_info(HEST_PFX "Table parsing disabled.\n");
+ return;
+ }
+
if (acpi_disabled)
goto err;
- if (hest_disable) {
- pr_info(HEST_PFX "HEST tabling parsing is disabled.\n");
- goto err;
- }
-
status = acpi_get_table(ACPI_SIG_HEST, 0,
(struct acpi_table_header **)&hest_tab);
if (status == AE_NOT_FOUND) {
- pr_info(HEST_PFX "Table is not found!\n");
+ pr_info(HEST_PFX "Table not found.\n");
goto err;
} else if (ACPI_FAILURE(status)) {
const char *msg = acpi_format_exception(status);
@@ -226,15 +226,11 @@
goto err;
rc = hest_ghes_dev_register(ghes_count);
- if (rc)
- goto err;
+ if (!rc) {
+ pr_info(HEST_PFX "Table parsing has been initialized.\n");
+ return;
+ }
- pr_info(HEST_PFX "HEST table parsing is initialized.\n");
-
- return 0;
err:
hest_disable = 1;
- return rc;
}
-
-subsys_initcall(hest_init);
diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c
index 95649d3..68bc227 100644
--- a/drivers/acpi/battery.c
+++ b/drivers/acpi/battery.c
@@ -631,6 +631,17 @@
return result;
}
+static void acpi_battery_refresh(struct acpi_battery *battery)
+{
+ if (!battery->bat.dev)
+ return;
+
+ acpi_battery_get_info(battery);
+ /* The battery may have changed its reporting units. */
+ sysfs_remove_battery(battery);
+ sysfs_add_battery(battery);
+}
+
/* --------------------------------------------------------------------------
FS Interface (/proc)
-------------------------------------------------------------------------- */
@@ -868,6 +879,8 @@
struct proc_dir_entry *entry = NULL;
int i;
+ printk(KERN_WARNING PREFIX "Deprecated procfs I/F for battery is loaded,"
+ " please retry with CONFIG_ACPI_PROCFS_POWER cleared\n");
if (!acpi_device_dir(device)) {
acpi_device_dir(device) = proc_mkdir(acpi_device_bid(device),
acpi_battery_dir);
@@ -914,6 +927,8 @@
if (!battery)
return;
old = battery->bat.dev;
+ if (event == ACPI_BATTERY_NOTIFY_INFO)
+ acpi_battery_refresh(battery);
acpi_battery_update(battery);
acpi_bus_generate_proc_event(device, event,
acpi_battery_present(battery));
@@ -983,6 +998,7 @@
if (!device)
return -EINVAL;
battery = acpi_driver_data(device);
+ acpi_battery_refresh(battery);
battery->update_time = 0;
acpi_battery_update(battery);
return 0;
diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index d68bd61..7ced61f 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -52,22 +52,6 @@
#define STRUCT_TO_INT(s) (*((int*)&s))
-static int set_power_nocheck(const struct dmi_system_id *id)
-{
- printk(KERN_NOTICE PREFIX "%s detected - "
- "disable power check in power transition\n", id->ident);
- acpi_power_nocheck = 1;
- return 0;
-}
-static struct dmi_system_id __cpuinitdata power_nocheck_dmi_table[] = {
- {
- set_power_nocheck, "HP Pavilion 05", {
- DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies LTD"),
- DMI_MATCH(DMI_SYS_VENDOR, "HP Pavilion 05"),
- DMI_MATCH(DMI_PRODUCT_VERSION, "2001211RE101GLEND") }, NULL},
- {},
-};
-
#ifdef CONFIG_X86
static int set_copy_dsdt(const struct dmi_system_id *id)
@@ -196,33 +180,24 @@
Power Management
-------------------------------------------------------------------------- */
-int acpi_bus_get_power(acpi_handle handle, int *state)
+static int __acpi_bus_get_power(struct acpi_device *device, int *state)
{
int result = 0;
acpi_status status = 0;
- struct acpi_device *device = NULL;
unsigned long long psc = 0;
-
- result = acpi_bus_get_device(handle, &device);
- if (result)
- return result;
+ if (!device || !state)
+ return -EINVAL;
*state = ACPI_STATE_UNKNOWN;
- if (!device->flags.power_manageable) {
- /* TBD: Non-recursive algorithm for walking up hierarchy */
- if (device->parent)
- *state = device->parent->power.state;
- else
- *state = ACPI_STATE_D0;
- } else {
+ if (device->flags.power_manageable) {
/*
* Get the device's power state either directly (via _PSC) or
* indirectly (via power resources).
*/
if (device->power.flags.power_resources) {
- result = acpi_power_get_inferred_state(device);
+ result = acpi_power_get_inferred_state(device, state);
if (result)
return result;
} else if (device->power.flags.explicit_get) {
@@ -230,59 +205,33 @@
NULL, &psc);
if (ACPI_FAILURE(status))
return -ENODEV;
- device->power.state = (int)psc;
+ *state = (int)psc;
}
-
- *state = device->power.state;
+ } else {
+ /* TBD: Non-recursive algorithm for walking up hierarchy. */
+ *state = device->parent ?
+ device->parent->power.state : ACPI_STATE_D0;
}
ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Device [%s] power state is D%d\n",
- device->pnp.bus_id, device->power.state));
+ device->pnp.bus_id, *state));
return 0;
}
-EXPORT_SYMBOL(acpi_bus_get_power);
-int acpi_bus_set_power(acpi_handle handle, int state)
+static int __acpi_bus_set_power(struct acpi_device *device, int state)
{
int result = 0;
acpi_status status = AE_OK;
- struct acpi_device *device = NULL;
char object_name[5] = { '_', 'P', 'S', '0' + state, '\0' };
-
- result = acpi_bus_get_device(handle, &device);
- if (result)
- return result;
-
- if ((state < ACPI_STATE_D0) || (state > ACPI_STATE_D3))
+ if (!device || (state < ACPI_STATE_D0) || (state > ACPI_STATE_D3))
return -EINVAL;
/* Make sure this is a valid target state */
- if (!device->flags.power_manageable) {
- ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Device `[%s]' is not power manageable\n",
- kobject_name(&device->dev.kobj)));
- return -ENODEV;
- }
- /*
- * Get device's current power state
- */
- if (!acpi_power_nocheck) {
- /*
- * Maybe the incorrect power state is returned on the bogus
- * bios, which is different with the real power state.
- * For example: the bios returns D0 state and the real power
- * state is D3. OS expects to set the device to D0 state. In
- * such case if OS uses the power state returned by the BIOS,
- * the device can't be transisted to the correct power state.
- * So if the acpi_power_nocheck is set, it is unnecessary to
- * get the power state by calling acpi_bus_get_power.
- */
- acpi_bus_get_power(device->handle, &device->power.state);
- }
- if ((state == device->power.state) && !device->flags.force_power_state) {
+ if (state == device->power.state) {
ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Device is already at D%d\n",
state));
return 0;
@@ -351,8 +300,75 @@
return result;
}
+
+int acpi_bus_set_power(acpi_handle handle, int state)
+{
+ struct acpi_device *device;
+ int result;
+
+ result = acpi_bus_get_device(handle, &device);
+ if (result)
+ return result;
+
+ if (!device->flags.power_manageable) {
+ ACPI_DEBUG_PRINT((ACPI_DB_INFO,
+ "Device [%s] is not power manageable\n",
+ dev_name(&device->dev)));
+ return -ENODEV;
+ }
+
+ return __acpi_bus_set_power(device, state);
+}
EXPORT_SYMBOL(acpi_bus_set_power);
+
+int acpi_bus_init_power(struct acpi_device *device)
+{
+ int state;
+ int result;
+
+ if (!device)
+ return -EINVAL;
+
+ device->power.state = ACPI_STATE_UNKNOWN;
+
+ result = __acpi_bus_get_power(device, &state);
+ if (result)
+ return result;
+
+ if (device->power.flags.power_resources)
+ result = acpi_power_on_resources(device, state);
+
+ if (!result)
+ device->power.state = state;
+
+ return result;
+}
+
+
+int acpi_bus_update_power(acpi_handle handle, int *state_p)
+{
+ struct acpi_device *device;
+ int state;
+ int result;
+
+ result = acpi_bus_get_device(handle, &device);
+ if (result)
+ return result;
+
+ result = __acpi_bus_get_power(device, &state);
+ if (result)
+ return result;
+
+ result = __acpi_bus_set_power(device, state);
+ if (!result && state_p)
+ *state_p = state;
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(acpi_bus_update_power);
+
+
bool acpi_bus_power_manageable(acpi_handle handle)
{
struct acpi_device *device;
@@ -1023,15 +1039,8 @@
if (acpi_disabled)
return result;
- /*
- * If the laptop falls into the DMI check table, the power state check
- * will be disabled in the course of device power transition.
- */
- dmi_check_system(power_nocheck_dmi_table);
-
acpi_scan_init();
acpi_ec_init();
- acpi_power_init();
acpi_debugfs_init();
acpi_sleep_proc_init();
acpi_wakeup_device_init();
diff --git a/drivers/acpi/button.c b/drivers/acpi/button.c
index 71ef9cd..76bbb78 100644
--- a/drivers/acpi/button.c
+++ b/drivers/acpi/button.c
@@ -279,6 +279,9 @@
input_report_switch(button->input, SW_LID, !state);
input_sync(button->input);
+ if (state)
+ pm_wakeup_event(&device->dev, 0);
+
ret = blocking_notifier_call_chain(&acpi_lid_notifier, state, device);
if (ret == NOTIFY_DONE)
ret = blocking_notifier_call_chain(&acpi_lid_notifier, state,
@@ -314,6 +317,8 @@
input_sync(input);
input_report_key(input, keycode, 0);
input_sync(input);
+
+ pm_wakeup_event(&device->dev, 0);
}
acpi_bus_generate_proc_event(device, event, ++button->pushed);
@@ -426,7 +431,7 @@
acpi_enable_gpe(device->wakeup.gpe_device,
device->wakeup.gpe_number);
device->wakeup.run_wake_count++;
- device->wakeup.state.enabled = 1;
+ device_set_wakeup_enable(&device->dev, true);
}
printk(KERN_INFO PREFIX "%s [%s]\n", name, acpi_device_bid(device));
@@ -449,7 +454,7 @@
acpi_disable_gpe(device->wakeup.gpe_device,
device->wakeup.gpe_number);
device->wakeup.run_wake_count--;
- device->wakeup.state.enabled = 0;
+ device_set_wakeup_enable(&device->dev, false);
}
acpi_button_remove_fs(device);
diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c
index 81514a4..1864ad3 100644
--- a/drivers/acpi/dock.c
+++ b/drivers/acpi/dock.c
@@ -725,7 +725,7 @@
complete_dock(ds);
dock_event(ds, event, DOCK_EVENT);
dock_lock(ds, 1);
- acpi_update_gpes();
+ acpi_update_all_gpes();
break;
}
if (dock_present(ds) || dock_in_progress(ds))
diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c
index 302b31e..fa848c4 100644
--- a/drivers/acpi/ec.c
+++ b/drivers/acpi/ec.c
@@ -606,7 +606,8 @@
return 0;
}
-static u32 acpi_ec_gpe_handler(void *data)
+static u32 acpi_ec_gpe_handler(acpi_handle gpe_device,
+ u32 gpe_number, void *data)
{
struct acpi_ec *ec = data;
@@ -618,7 +619,7 @@
wake_up(&ec->wait);
ec_check_sci(ec, acpi_ec_read_status(ec));
}
- return ACPI_INTERRUPT_HANDLED;
+ return ACPI_INTERRUPT_HANDLED | ACPI_REENABLE_GPE;
}
/* --------------------------------------------------------------------------
diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
index 6004908..467479f 100644
--- a/drivers/acpi/fan.c
+++ b/drivers/acpi/fan.c
@@ -86,7 +86,7 @@
if (!device)
return -EINVAL;
- result = acpi_bus_get_power(device->handle, &acpi_state);
+ result = acpi_bus_update_power(device->handle, &acpi_state);
if (result)
return result;
@@ -123,7 +123,6 @@
static int acpi_fan_add(struct acpi_device *device)
{
int result = 0;
- int state = 0;
struct thermal_cooling_device *cdev;
if (!device)
@@ -132,16 +131,12 @@
strcpy(acpi_device_name(device), "Fan");
strcpy(acpi_device_class(device), ACPI_FAN_CLASS);
- result = acpi_bus_get_power(device->handle, &state);
+ result = acpi_bus_update_power(device->handle, NULL);
if (result) {
- printk(KERN_ERR PREFIX "Reading power state\n");
+ printk(KERN_ERR PREFIX "Setting initial power state\n");
goto end;
}
- device->flags.force_power_state = 1;
- acpi_bus_set_power(device->handle, state);
- device->flags.force_power_state = 0;
-
cdev = thermal_cooling_device_register("Fan", device,
&fan_cooling_ops);
if (IS_ERR(cdev)) {
@@ -200,22 +195,14 @@
static int acpi_fan_resume(struct acpi_device *device)
{
- int result = 0;
- int power_state = 0;
+ int result;
if (!device)
return -EINVAL;
- result = acpi_bus_get_power(device->handle, &power_state);
- if (result) {
- printk(KERN_ERR PREFIX
- "Error reading fan power state\n");
- return result;
- }
-
- device->flags.force_power_state = 1;
- acpi_bus_set_power(device->handle, power_state);
- device->flags.force_power_state = 0;
+ result = acpi_bus_update_power(device->handle, NULL);
+ if (result)
+ printk(KERN_ERR PREFIX "Error updating fan power state\n");
return result;
}
diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
index 78b0164..7c47ed5 100644
--- a/drivers/acpi/glue.c
+++ b/drivers/acpi/glue.c
@@ -167,11 +167,8 @@
"firmware_node");
ret = sysfs_create_link(&acpi_dev->dev.kobj, &dev->kobj,
"physical_node");
- if (acpi_dev->wakeup.flags.valid) {
+ if (acpi_dev->wakeup.flags.valid)
device_set_wakeup_capable(dev, true);
- device_set_wakeup_enable(dev,
- acpi_dev->wakeup.state.enabled);
- }
}
return 0;
diff --git a/drivers/acpi/internal.h b/drivers/acpi/internal.h
index a212bfe..b1cc81a 100644
--- a/drivers/acpi/internal.h
+++ b/drivers/acpi/internal.h
@@ -41,9 +41,10 @@
int acpi_power_init(void);
int acpi_device_sleep_wake(struct acpi_device *dev,
int enable, int sleep_state, int dev_state);
-int acpi_power_get_inferred_state(struct acpi_device *device);
+int acpi_power_get_inferred_state(struct acpi_device *device, int *state);
+int acpi_power_on_resources(struct acpi_device *device, int state);
int acpi_power_transition(struct acpi_device *device, int state);
-extern int acpi_power_nocheck;
+int acpi_bus_init_power(struct acpi_device *device);
int acpi_wakeup_device_init(void);
void acpi_early_processor_set_pdc(void);
@@ -82,8 +83,16 @@
#ifdef CONFIG_ACPI_SLEEP
int acpi_sleep_proc_init(void);
+int suspend_nvs_alloc(void);
+void suspend_nvs_free(void);
+int suspend_nvs_save(void);
+void suspend_nvs_restore(void);
#else
static inline int acpi_sleep_proc_init(void) { return 0; }
+static inline int suspend_nvs_alloc(void) { return 0; }
+static inline void suspend_nvs_free(void) {}
+static inline int suspend_nvs_save(void) { return 0; }
+static inline void suspend_nvs_restore(void) {}
#endif
#endif /* _ACPI_INTERNAL_H_ */
diff --git a/drivers/acpi/numa.c b/drivers/acpi/numa.c
index d9926af..5eb25eb 100644
--- a/drivers/acpi/numa.c
+++ b/drivers/acpi/numa.c
@@ -275,23 +275,19 @@
int __init acpi_numa_init(void)
{
int ret = 0;
- int nr_cpu_entries = nr_cpu_ids;
-#ifdef CONFIG_X86
/*
* Should not limit number with cpu num that is from NR_CPUS or nr_cpus=
* SRAT cpu entries could have different order with that in MADT.
* So go over all cpu entries in SRAT to get apicid to node mapping.
*/
- nr_cpu_entries = MAX_LOCAL_APIC;
-#endif
/* SRAT: Static Resource Affinity Table */
if (!acpi_table_parse(ACPI_SIG_SRAT, acpi_parse_srat)) {
acpi_table_parse_srat(ACPI_SRAT_TYPE_X2APIC_CPU_AFFINITY,
- acpi_parse_x2apic_affinity, nr_cpu_entries);
+ acpi_parse_x2apic_affinity, 0);
acpi_table_parse_srat(ACPI_SRAT_TYPE_CPU_AFFINITY,
- acpi_parse_processor_affinity, nr_cpu_entries);
+ acpi_parse_processor_affinity, 0);
ret = acpi_table_parse_srat(ACPI_SRAT_TYPE_MEMORY_AFFINITY,
acpi_parse_memory_affinity,
NR_NODE_MEMBLKS);
diff --git a/kernel/power/nvs.c b/drivers/acpi/nvs.c
similarity index 86%
rename from kernel/power/nvs.c
rename to drivers/acpi/nvs.c
index 1836db6..54b6ab8 100644
--- a/kernel/power/nvs.c
+++ b/drivers/acpi/nvs.c
@@ -1,7 +1,7 @@
/*
- * linux/kernel/power/hibernate_nvs.c - Routines for handling NVS memory
+ * nvs.c - Routines for saving and restoring ACPI NVS memory region
*
- * Copyright (C) 2008,2009 Rafael J. Wysocki <rjw@sisk.pl>, Novell Inc.
+ * Copyright (C) 2008-2011 Rafael J. Wysocki <rjw@sisk.pl>, Novell Inc.
*
* This file is released under the GPLv2.
*/
@@ -11,7 +11,8 @@
#include <linux/list.h>
#include <linux/mm.h>
#include <linux/slab.h>
-#include <linux/suspend.h>
+#include <linux/acpi.h>
+#include <acpi/acpiosxf.h>
/*
* Platforms, like ACPI, may want us to save some memory used by them during
@@ -79,7 +80,7 @@
free_page((unsigned long)entry->data);
entry->data = NULL;
if (entry->kaddr) {
- iounmap(entry->kaddr);
+ acpi_os_unmap_memory(entry->kaddr, entry->size);
entry->kaddr = NULL;
}
}
@@ -105,7 +106,7 @@
/**
* suspend_nvs_save - save NVS memory regions
*/
-void suspend_nvs_save(void)
+int suspend_nvs_save(void)
{
struct nvs_page *entry;
@@ -113,9 +114,16 @@
list_for_each_entry(entry, &nvs_list, node)
if (entry->data) {
- entry->kaddr = ioremap(entry->phys_start, entry->size);
+ entry->kaddr = acpi_os_map_memory(entry->phys_start,
+ entry->size);
+ if (!entry->kaddr) {
+ suspend_nvs_free();
+ return -ENOMEM;
+ }
memcpy(entry->data, entry->kaddr, entry->size);
}
+
+ return 0;
}
/**
diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index 055d7b7..e2dd6de 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -320,7 +320,7 @@
pg_off = round_down(phys, PAGE_SIZE);
pg_sz = round_up(phys + size, PAGE_SIZE) - pg_off;
- virt = ioremap(pg_off, pg_sz);
+ virt = ioremap_cache(pg_off, pg_sz);
if (!virt) {
kfree(map);
return NULL;
@@ -642,7 +642,7 @@
virt_addr = acpi_map_vaddr_lookup(phys_addr, size);
rcu_read_unlock();
if (!virt_addr) {
- virt_addr = ioremap(phys_addr, size);
+ virt_addr = ioremap_cache(phys_addr, size);
unmap = 1;
}
if (!value)
@@ -678,7 +678,7 @@
virt_addr = acpi_map_vaddr_lookup(phys_addr, size);
rcu_read_unlock();
if (!virt_addr) {
- virt_addr = ioremap(phys_addr, size);
+ virt_addr = ioremap_cache(phys_addr, size);
unmap = 1;
}
@@ -1233,8 +1233,7 @@
int acpi_check_resource_conflict(const struct resource *res)
{
struct acpi_res_list *res_list_elem;
- int ioport;
- int clash = 0;
+ int ioport = 0, clash = 0;
if (acpi_enforce_resources == ENFORCE_RESOURCES_NO)
return 0;
@@ -1264,9 +1263,13 @@
if (clash) {
if (acpi_enforce_resources != ENFORCE_RESOURCES_NO) {
printk(KERN_WARNING "ACPI: resource %s %pR"
- " conflicts with ACPI region %s %pR\n",
+ " conflicts with ACPI region %s "
+ "[%s 0x%zx-0x%zx]\n",
res->name, res, res_list_elem->name,
- res_list_elem);
+ (res_list_elem->resource_type ==
+ ACPI_ADR_SPACE_SYSTEM_IO) ? "io" : "mem",
+ (size_t) res_list_elem->start,
+ (size_t) res_list_elem->end);
if (acpi_enforce_resources == ENFORCE_RESOURCES_LAX)
printk(KERN_NOTICE "ACPI: This conflict may"
" cause random problems and system"
diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 96668ad..8524939 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -36,6 +36,7 @@
#include <linux/slab.h>
#include <acpi/acpi_bus.h>
#include <acpi/acpi_drivers.h>
+#include <acpi/apei.h>
#define PREFIX "ACPI: "
@@ -47,6 +48,11 @@
static int acpi_pci_root_remove(struct acpi_device *device, int type);
static int acpi_pci_root_start(struct acpi_device *device);
+#define ACPI_PCIE_REQ_SUPPORT (OSC_EXT_PCI_CONFIG_SUPPORT \
+ | OSC_ACTIVE_STATE_PWR_SUPPORT \
+ | OSC_CLOCK_PWR_CAPABILITY_SUPPORT \
+ | OSC_MSI_SUPPORT)
+
static const struct acpi_device_id root_device_ids[] = {
{"PNP0A03", 0},
{"", 0},
@@ -566,6 +572,33 @@
if (flags != base_flags)
acpi_pci_osc_support(root, flags);
+ if (!pcie_ports_disabled
+ && (flags & ACPI_PCIE_REQ_SUPPORT) == ACPI_PCIE_REQ_SUPPORT) {
+ flags = OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL
+ | OSC_PCI_EXPRESS_NATIVE_HP_CONTROL
+ | OSC_PCI_EXPRESS_PME_CONTROL;
+
+ if (pci_aer_available()) {
+ if (aer_acpi_firmware_first())
+ dev_dbg(root->bus->bridge,
+ "PCIe errors handled by BIOS.\n");
+ else
+ flags |= OSC_PCI_EXPRESS_AER_CONTROL;
+ }
+
+ dev_info(root->bus->bridge,
+ "Requesting ACPI _OSC control (0x%02x)\n", flags);
+
+ status = acpi_pci_osc_control_set(device->handle, &flags,
+ OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL);
+ if (ACPI_SUCCESS(status))
+ dev_info(root->bus->bridge,
+ "ACPI _OSC control (0x%02x) granted\n", flags);
+ else
+ dev_dbg(root->bus->bridge,
+ "ACPI _OSC request failed (code %d)\n", status);
+ }
+
pci_acpi_add_bus_pm_notifier(device, root->bus);
if (device->wakeup.flags.run_wake)
device_set_run_wake(root->bus->bridge, true);
@@ -600,6 +633,8 @@
static int __init acpi_pci_root_init(void)
{
+ acpi_hest_init();
+
if (acpi_pci_disabled)
return 0;
diff --git a/drivers/acpi/power.c b/drivers/acpi/power.c
index 4c9c2fb..9ac2a9f 100644
--- a/drivers/acpi/power.c
+++ b/drivers/acpi/power.c
@@ -56,9 +56,6 @@
#define ACPI_POWER_RESOURCE_STATE_ON 0x01
#define ACPI_POWER_RESOURCE_STATE_UNKNOWN 0xFF
-int acpi_power_nocheck;
-module_param_named(power_nocheck, acpi_power_nocheck, bool, 000);
-
static int acpi_power_add(struct acpi_device *device);
static int acpi_power_remove(struct acpi_device *device, int type);
static int acpi_power_resume(struct acpi_device *device);
@@ -148,9 +145,8 @@
static int acpi_power_get_list_state(struct acpi_handle_list *list, int *state)
{
- int result = 0, state1;
- u32 i = 0;
-
+ int cur_state;
+ int i = 0;
if (!list || !state)
return -EINVAL;
@@ -158,25 +154,33 @@
/* The state of the list is 'on' IFF all resources are 'on'. */
for (i = 0; i < list->count; i++) {
- /*
- * The state of the power resource can be obtained by
- * using the ACPI handle. In such case it is unnecessary to
- * get the Power resource first and then get its state again.
- */
- result = acpi_power_get_state(list->handles[i], &state1);
+ struct acpi_power_resource *resource;
+ acpi_handle handle = list->handles[i];
+ int result;
+
+ result = acpi_power_get_context(handle, &resource);
if (result)
return result;
- *state = state1;
+ mutex_lock(&resource->resource_lock);
- if (*state != ACPI_POWER_RESOURCE_STATE_ON)
+ result = acpi_power_get_state(handle, &cur_state);
+
+ mutex_unlock(&resource->resource_lock);
+
+ if (result)
+ return result;
+
+ if (cur_state != ACPI_POWER_RESOURCE_STATE_ON)
break;
}
ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Resource list is %s\n",
- *state ? "on" : "off"));
+ cur_state ? "on" : "off"));
- return result;
+ *state = cur_state;
+
+ return 0;
}
static int __acpi_power_on(struct acpi_power_resource *resource)
@@ -222,7 +226,7 @@
return result;
}
-static int acpi_power_off_device(acpi_handle handle)
+static int acpi_power_off(acpi_handle handle)
{
int result = 0;
acpi_status status = AE_OK;
@@ -266,6 +270,35 @@
return result;
}
+static void __acpi_power_off_list(struct acpi_handle_list *list, int num_res)
+{
+ int i;
+
+ for (i = num_res - 1; i >= 0 ; i--)
+ acpi_power_off(list->handles[i]);
+}
+
+static void acpi_power_off_list(struct acpi_handle_list *list)
+{
+ __acpi_power_off_list(list, list->count);
+}
+
+static int acpi_power_on_list(struct acpi_handle_list *list)
+{
+ int result = 0;
+ int i;
+
+ for (i = 0; i < list->count; i++) {
+ result = acpi_power_on(list->handles[i]);
+ if (result) {
+ __acpi_power_off_list(list, i);
+ break;
+ }
+ }
+
+ return result;
+}
+
/**
* acpi_device_sleep_wake - execute _DSW (Device Sleep Wake) or (deprecated in
* ACPI 3.0) _PSW (Power State Wake)
@@ -404,8 +437,7 @@
/* Close power resource */
for (i = 0; i < dev->wakeup.resources.count; i++) {
- int ret = acpi_power_off_device(
- dev->wakeup.resources.handles[i]);
+ int ret = acpi_power_off(dev->wakeup.resources.handles[i]);
if (ret) {
printk(KERN_ERR PREFIX "Transition power state\n");
dev->wakeup.flags.valid = 0;
@@ -423,19 +455,16 @@
Device Power Management
-------------------------------------------------------------------------- */
-int acpi_power_get_inferred_state(struct acpi_device *device)
+int acpi_power_get_inferred_state(struct acpi_device *device, int *state)
{
int result = 0;
struct acpi_handle_list *list = NULL;
int list_state = 0;
int i = 0;
-
- if (!device)
+ if (!device || !state)
return -EINVAL;
- device->power.state = ACPI_STATE_UNKNOWN;
-
/*
* We know a device's inferred power state when all the resources
* required for a given D-state are 'on'.
@@ -450,22 +479,26 @@
return result;
if (list_state == ACPI_POWER_RESOURCE_STATE_ON) {
- device->power.state = i;
+ *state = i;
return 0;
}
}
- device->power.state = ACPI_STATE_D3;
-
+ *state = ACPI_STATE_D3;
return 0;
}
+int acpi_power_on_resources(struct acpi_device *device, int state)
+{
+ if (!device || state < ACPI_STATE_D0 || state > ACPI_STATE_D3)
+ return -EINVAL;
+
+ return acpi_power_on_list(&device->power.states[state].resources);
+}
+
int acpi_power_transition(struct acpi_device *device, int state)
{
- int result = 0;
- struct acpi_handle_list *cl = NULL; /* Current Resources */
- struct acpi_handle_list *tl = NULL; /* Target Resources */
- int i = 0;
+ int result;
if (!device || (state < ACPI_STATE_D0) || (state > ACPI_STATE_D3))
return -EINVAL;
@@ -477,37 +510,20 @@
|| (device->power.state > ACPI_STATE_D3))
return -ENODEV;
- cl = &device->power.states[device->power.state].resources;
- tl = &device->power.states[state].resources;
-
/* TBD: Resources must be ordered. */
/*
* First we reference all power resources required in the target list
- * (e.g. so the device doesn't lose power while transitioning).
+ * (e.g. so the device doesn't lose power while transitioning). Then,
+ * we dereference all power resources used in the current list.
*/
- for (i = 0; i < tl->count; i++) {
- result = acpi_power_on(tl->handles[i]);
- if (result)
- goto end;
- }
+ result = acpi_power_on_list(&device->power.states[state].resources);
+ if (!result)
+ acpi_power_off_list(
+ &device->power.states[device->power.state].resources);
- /*
- * Then we dereference all power resources used in the current list.
- */
- for (i = 0; i < cl->count; i++) {
- result = acpi_power_off_device(cl->handles[i]);
- if (result)
- goto end;
- }
-
- end:
- if (result)
- device->power.state = ACPI_STATE_UNKNOWN;
- else {
- /* We shouldn't change the state till all above operations succeed */
- device->power.state = state;
- }
+ /* We shouldn't change the state unless the above operations succeed. */
+ device->power.state = result ? ACPI_STATE_UNKNOWN : state;
return result;
}
diff --git a/drivers/acpi/proc.c b/drivers/acpi/proc.c
index afad677..f5f9869 100644
--- a/drivers/acpi/proc.c
+++ b/drivers/acpi/proc.c
@@ -311,7 +311,9 @@
dev->pnp.bus_id,
(u32) dev->wakeup.sleep_state,
dev->wakeup.flags.run_wake ? '*' : ' ',
- dev->wakeup.state.enabled ? "enabled" : "disabled");
+ (device_may_wakeup(&dev->dev)
+ || (ldev && device_may_wakeup(ldev))) ?
+ "enabled" : "disabled");
if (ldev)
seq_printf(seq, "%s:%s",
ldev->bus ? ldev->bus->name : "no-bus",
@@ -328,8 +330,10 @@
{
struct device *dev = acpi_get_physical_device(adev->handle);
- if (dev && device_can_wakeup(dev))
- device_set_wakeup_enable(dev, adev->wakeup.state.enabled);
+ if (dev && device_can_wakeup(dev)) {
+ bool enable = !device_may_wakeup(dev);
+ device_set_wakeup_enable(dev, enable);
+ }
}
static ssize_t
@@ -341,7 +345,6 @@
char strbuf[5];
char str[5] = "";
unsigned int len = count;
- struct acpi_device *found_dev = NULL;
if (len > 4)
len = 4;
@@ -361,33 +364,13 @@
continue;
if (!strncmp(dev->pnp.bus_id, str, 4)) {
- dev->wakeup.state.enabled =
- dev->wakeup.state.enabled ? 0 : 1;
- found_dev = dev;
- break;
- }
- }
- if (found_dev) {
- physical_device_enable_wakeup(found_dev);
- list_for_each_safe(node, next, &acpi_wakeup_device_list) {
- struct acpi_device *dev = container_of(node,
- struct
- acpi_device,
- wakeup_list);
-
- if ((dev != found_dev) &&
- (dev->wakeup.gpe_number ==
- found_dev->wakeup.gpe_number)
- && (dev->wakeup.gpe_device ==
- found_dev->wakeup.gpe_device)) {
- printk(KERN_WARNING
- "ACPI: '%s' and '%s' have the same GPE, "
- "can't disable/enable one separately\n",
- dev->pnp.bus_id, found_dev->pnp.bus_id);
- dev->wakeup.state.enabled =
- found_dev->wakeup.state.enabled;
+ if (device_can_wakeup(&dev->dev)) {
+ bool enable = !device_may_wakeup(&dev->dev);
+ device_set_wakeup_enable(&dev->dev, enable);
+ } else {
physical_device_enable_wakeup(dev);
}
+ break;
}
}
mutex_unlock(&acpi_device_lock);
diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index bec561c..3c1a2fe 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -23,7 +23,7 @@
{
printk(KERN_NOTICE PREFIX "%s detected - "
"disabling mwait for CPU C-states\n", id->ident);
- idle_nomwait = 1;
+ boot_option_idle_override = IDLE_NOMWAIT;
return 0;
}
@@ -283,7 +283,7 @@
{
acpi_status status = AE_OK;
- if (idle_nomwait) {
+ if (boot_option_idle_override == IDLE_NOMWAIT) {
/*
* If mwait is disabled for CPU C-states, the C2C3_FFH access
* mode will be disabled in the parameter of _PDC object.
diff --git a/drivers/acpi/processor_driver.c b/drivers/acpi/processor_driver.c
index 85e4804..360a74e 100644
--- a/drivers/acpi/processor_driver.c
+++ b/drivers/acpi/processor_driver.c
@@ -40,10 +40,6 @@
#include <linux/pm.h>
#include <linux/cpufreq.h>
#include <linux/cpu.h>
-#ifdef CONFIG_ACPI_PROCFS
-#include <linux/proc_fs.h>
-#include <linux/seq_file.h>
-#endif
#include <linux/dmi.h>
#include <linux/moduleparam.h>
#include <linux/cpuidle.h>
@@ -246,53 +242,6 @@
return result;
}
-#ifdef CONFIG_ACPI_PROCFS
-static struct proc_dir_entry *acpi_processor_dir = NULL;
-
-static int __cpuinit acpi_processor_add_fs(struct acpi_device *device)
-{
- struct proc_dir_entry *entry = NULL;
-
-
- if (!acpi_device_dir(device)) {
- acpi_device_dir(device) = proc_mkdir(acpi_device_bid(device),
- acpi_processor_dir);
- if (!acpi_device_dir(device))
- return -ENODEV;
- }
-
- /* 'throttling' [R/W] */
- entry = proc_create_data(ACPI_PROCESSOR_FILE_THROTTLING,
- S_IFREG | S_IRUGO | S_IWUSR,
- acpi_device_dir(device),
- &acpi_processor_throttling_fops,
- acpi_driver_data(device));
- if (!entry)
- return -EIO;
- return 0;
-}
-static int acpi_processor_remove_fs(struct acpi_device *device)
-{
-
- if (acpi_device_dir(device)) {
- remove_proc_entry(ACPI_PROCESSOR_FILE_THROTTLING,
- acpi_device_dir(device));
- remove_proc_entry(acpi_device_bid(device), acpi_processor_dir);
- acpi_device_dir(device) = NULL;
- }
-
- return 0;
-}
-#else
-static inline int acpi_processor_add_fs(struct acpi_device *device)
-{
- return 0;
-}
-static inline int acpi_processor_remove_fs(struct acpi_device *device)
-{
- return 0;
-}
-#endif
/* --------------------------------------------------------------------------
Driver Interface
-------------------------------------------------------------------------- */
@@ -478,8 +427,13 @@
if (action == CPU_ONLINE && pr) {
acpi_processor_ppc_has_changed(pr, 0);
acpi_processor_cst_has_changed(pr);
+ acpi_processor_reevaluate_tstate(pr, action);
acpi_processor_tstate_has_changed(pr);
}
+ if (action == CPU_DEAD && pr) {
+ /* invalidate the flag.throttling after one CPU is offline */
+ acpi_processor_reevaluate_tstate(pr, action);
+ }
return NOTIFY_OK;
}
@@ -537,14 +491,10 @@
per_cpu(processors, pr->id) = pr;
- result = acpi_processor_add_fs(device);
- if (result)
- goto err_free_cpumask;
-
sysdev = get_cpu_sysdev(pr->id);
if (sysfs_create_link(&device->dev.kobj, &sysdev->kobj, "sysdev")) {
result = -EFAULT;
- goto err_remove_fs;
+ goto err_free_cpumask;
}
#ifdef CONFIG_CPU_FREQ
@@ -590,8 +540,6 @@
thermal_cooling_device_unregister(pr->cdev);
err_power_exit:
acpi_processor_power_exit(pr, device);
-err_remove_fs:
- acpi_processor_remove_fs(device);
err_free_cpumask:
free_cpumask_var(pr->throttling.shared_cpu_map);
@@ -620,8 +568,6 @@
sysfs_remove_link(&device->dev.kobj, "sysdev");
- acpi_processor_remove_fs(device);
-
if (pr->cdev) {
sysfs_remove_link(&device->dev.kobj, "thermal_cooling");
sysfs_remove_link(&pr->cdev->device.kobj, "device");
@@ -854,12 +800,6 @@
memset(&errata, 0, sizeof(errata));
-#ifdef CONFIG_ACPI_PROCFS
- acpi_processor_dir = proc_mkdir(ACPI_PROCESSOR_CLASS, acpi_root_dir);
- if (!acpi_processor_dir)
- return -ENOMEM;
-#endif
-
if (!cpuidle_register_driver(&acpi_idle_driver)) {
printk(KERN_DEBUG "ACPI: %s registered with cpuidle\n",
acpi_idle_driver.name);
@@ -885,10 +825,6 @@
out_cpuidle:
cpuidle_unregister_driver(&acpi_idle_driver);
-#ifdef CONFIG_ACPI_PROCFS
- remove_proc_entry(ACPI_PROCESSOR_CLASS, acpi_root_dir);
-#endif
-
return result;
}
@@ -907,10 +843,6 @@
cpuidle_unregister_driver(&acpi_idle_driver);
-#ifdef CONFIG_ACPI_PROCFS
- remove_proc_entry(ACPI_PROCESSOR_CLASS, acpi_root_dir);
-#endif
-
return;
}
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index a765b82..d615b7d 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -79,6 +79,13 @@
static unsigned int latency_factor __read_mostly = 2;
module_param(latency_factor, uint, 0644);
+static int disabled_by_idle_boot_param(void)
+{
+ return boot_option_idle_override == IDLE_POLL ||
+ boot_option_idle_override == IDLE_FORCE_MWAIT ||
+ boot_option_idle_override == IDLE_HALT;
+}
+
/*
* IBM ThinkPad R40e crashes mysteriously when going into C2 or C3.
* For now disable this. Probably a bug somewhere else.
@@ -455,7 +462,7 @@
continue;
}
if (cx.type == ACPI_STATE_C1 &&
- (idle_halt || idle_nomwait)) {
+ (boot_option_idle_override == IDLE_NOMWAIT)) {
/*
* In most cases the C1 space_id obtained from
* _CST object is FIXED_HARDWARE access mode.
@@ -1016,7 +1023,6 @@
state->flags = 0;
switch (cx->type) {
case ACPI_STATE_C1:
- state->flags |= CPUIDLE_FLAG_SHALLOW;
if (cx->entry_method == ACPI_CSTATE_FFH)
state->flags |= CPUIDLE_FLAG_TIME_VALID;
@@ -1025,16 +1031,13 @@
break;
case ACPI_STATE_C2:
- state->flags |= CPUIDLE_FLAG_BALANCED;
state->flags |= CPUIDLE_FLAG_TIME_VALID;
state->enter = acpi_idle_enter_simple;
dev->safe_state = state;
break;
case ACPI_STATE_C3:
- state->flags |= CPUIDLE_FLAG_DEEP;
state->flags |= CPUIDLE_FLAG_TIME_VALID;
- state->flags |= CPUIDLE_FLAG_CHECK_BM;
state->enter = pr->flags.bm_check ?
acpi_idle_enter_bm :
acpi_idle_enter_simple;
@@ -1058,7 +1061,7 @@
{
int ret = 0;
- if (boot_option_idle_override)
+ if (disabled_by_idle_boot_param())
return 0;
if (!pr)
@@ -1089,19 +1092,10 @@
acpi_status status = 0;
static int first_run;
- if (boot_option_idle_override)
+ if (disabled_by_idle_boot_param())
return 0;
if (!first_run) {
- if (idle_halt) {
- /*
- * When the boot option of "idle=halt" is added, halt
- * is used for CPU IDLE.
- * In such case C2/C3 is meaningless. So the max_cstate
- * is set to one.
- */
- max_cstate = 1;
- }
dmi_check_system(processor_power_dmi_table);
max_cstate = acpi_processor_cstate_check(max_cstate);
if (max_cstate < ACPI_C_STATES_MAX)
@@ -1142,7 +1136,7 @@
int acpi_processor_power_exit(struct acpi_processor *pr,
struct acpi_device *device)
{
- if (boot_option_idle_override)
+ if (disabled_by_idle_boot_param())
return 0;
cpuidle_unregister_device(&pr->power.dev);
diff --git a/drivers/acpi/processor_throttling.c b/drivers/acpi/processor_throttling.c
index ff36327..fa84e97 100644
--- a/drivers/acpi/processor_throttling.c
+++ b/drivers/acpi/processor_throttling.c
@@ -32,10 +32,6 @@
#include <linux/init.h>
#include <linux/sched.h>
#include <linux/cpufreq.h>
-#ifdef CONFIG_ACPI_PROCFS
-#include <linux/proc_fs.h>
-#include <linux/seq_file.h>
-#endif
#include <asm/io.h>
#include <asm/uaccess.h>
@@ -370,6 +366,58 @@
}
/*
+ * This function is used to reevaluate whether the T-state is valid
+ * after one CPU is onlined/offlined.
+ * It is noted that it won't reevaluate the following properties for
+ * the T-state.
+ * 1. Control method.
+ * 2. the number of supported T-state
+ * 3. TSD domain
+ */
+void acpi_processor_reevaluate_tstate(struct acpi_processor *pr,
+ unsigned long action)
+{
+ int result = 0;
+
+ if (action == CPU_DEAD) {
+ /* When one CPU is offline, the T-state throttling
+ * will be invalidated.
+ */
+ pr->flags.throttling = 0;
+ return;
+ }
+ /* the following is to recheck whether the T-state is valid for
+ * the online CPU
+ */
+ if (!pr->throttling.state_count) {
+ /* If the number of T-state is invalid, it is
+ * invalidated.
+ */
+ pr->flags.throttling = 0;
+ return;
+ }
+ pr->flags.throttling = 1;
+
+ /* Disable throttling (if enabled). We'll let subsequent
+ * policy (e.g.thermal) decide to lower performance if it
+ * so chooses, but for now we'll crank up the speed.
+ */
+
+ result = acpi_processor_get_throttling(pr);
+ if (result)
+ goto end;
+
+ if (pr->throttling.state) {
+ result = acpi_processor_set_throttling(pr, 0, false);
+ if (result)
+ goto end;
+ }
+
+end:
+ if (result)
+ pr->flags.throttling = 0;
+}
+/*
* _PTC - Processor Throttling Control (and status) register location
*/
static int acpi_processor_get_throttling_control(struct acpi_processor *pr)
@@ -876,7 +924,11 @@
*/
cpumask_copy(saved_mask, ¤t->cpus_allowed);
/* FIXME: use work_on_cpu() */
- set_cpus_allowed_ptr(current, cpumask_of(pr->id));
+ if (set_cpus_allowed_ptr(current, cpumask_of(pr->id))) {
+ /* Can't migrate to the target pr->id CPU. Exit */
+ free_cpumask_var(saved_mask);
+ return -ENODEV;
+ }
ret = pr->throttling.acpi_processor_get_throttling(pr);
/* restore the previous state */
set_cpus_allowed_ptr(current, saved_mask);
@@ -1051,6 +1103,14 @@
return -ENOMEM;
}
+ if (cpu_is_offline(pr->id)) {
+ /*
+ * the cpu pointed by pr->id is offline. Unnecessary to change
+ * the throttling state any more.
+ */
+ return -ENODEV;
+ }
+
cpumask_copy(saved_mask, ¤t->cpus_allowed);
t_state.target_state = state;
p_throttling = &(pr->throttling);
@@ -1074,7 +1134,11 @@
*/
if (p_throttling->shared_type == DOMAIN_COORD_TYPE_SW_ANY) {
/* FIXME: use work_on_cpu() */
- set_cpus_allowed_ptr(current, cpumask_of(pr->id));
+ if (set_cpus_allowed_ptr(current, cpumask_of(pr->id))) {
+ /* Can't migrate to the pr->id CPU. Exit */
+ ret = -ENODEV;
+ goto exit;
+ }
ret = p_throttling->acpi_processor_set_throttling(pr,
t_state.target_state, force);
} else {
@@ -1106,7 +1170,8 @@
}
t_state.cpu = i;
/* FIXME: use work_on_cpu() */
- set_cpus_allowed_ptr(current, cpumask_of(i));
+ if (set_cpus_allowed_ptr(current, cpumask_of(i)))
+ continue;
ret = match_pr->throttling.
acpi_processor_set_throttling(
match_pr, t_state.target_state, force);
@@ -1126,6 +1191,7 @@
/* restore the previous state */
/* FIXME: use work_on_cpu() */
set_cpus_allowed_ptr(current, saved_mask);
+exit:
free_cpumask_var(online_throttling_cpus);
free_cpumask_var(saved_mask);
return ret;
@@ -1216,113 +1282,3 @@
return result;
}
-#ifdef CONFIG_ACPI_PROCFS
-/* proc interface */
-static int acpi_processor_throttling_seq_show(struct seq_file *seq,
- void *offset)
-{
- struct acpi_processor *pr = seq->private;
- int i = 0;
- int result = 0;
-
- if (!pr)
- goto end;
-
- if (!(pr->throttling.state_count > 0)) {
- seq_puts(seq, "<not supported>\n");
- goto end;
- }
-
- result = acpi_processor_get_throttling(pr);
-
- if (result) {
- seq_puts(seq,
- "Could not determine current throttling state.\n");
- goto end;
- }
-
- seq_printf(seq, "state count: %d\n"
- "active state: T%d\n"
- "state available: T%d to T%d\n",
- pr->throttling.state_count, pr->throttling.state,
- pr->throttling_platform_limit,
- pr->throttling.state_count - 1);
-
- seq_puts(seq, "states:\n");
- if (pr->throttling.acpi_processor_get_throttling ==
- acpi_processor_get_throttling_fadt) {
- for (i = 0; i < pr->throttling.state_count; i++)
- seq_printf(seq, " %cT%d: %02d%%\n",
- (i == pr->throttling.state ? '*' : ' '), i,
- (pr->throttling.states[i].performance ? pr->
- throttling.states[i].performance / 10 : 0));
- } else {
- for (i = 0; i < pr->throttling.state_count; i++)
- seq_printf(seq, " %cT%d: %02d%%\n",
- (i == pr->throttling.state ? '*' : ' '), i,
- (int)pr->throttling.states_tss[i].
- freqpercentage);
- }
-
- end:
- return 0;
-}
-
-static int acpi_processor_throttling_open_fs(struct inode *inode,
- struct file *file)
-{
- return single_open(file, acpi_processor_throttling_seq_show,
- PDE(inode)->data);
-}
-
-static ssize_t acpi_processor_write_throttling(struct file *file,
- const char __user * buffer,
- size_t count, loff_t * data)
-{
- int result = 0;
- struct seq_file *m = file->private_data;
- struct acpi_processor *pr = m->private;
- char state_string[5] = "";
- char *charp = NULL;
- size_t state_val = 0;
- char tmpbuf[5] = "";
-
- if (!pr || (count > sizeof(state_string) - 1))
- return -EINVAL;
-
- if (copy_from_user(state_string, buffer, count))
- return -EFAULT;
-
- state_string[count] = '\0';
- if ((count > 0) && (state_string[count-1] == '\n'))
- state_string[count-1] = '\0';
-
- charp = state_string;
- if ((state_string[0] == 't') || (state_string[0] == 'T'))
- charp++;
-
- state_val = simple_strtoul(charp, NULL, 0);
- if (state_val >= pr->throttling.state_count)
- return -EINVAL;
-
- snprintf(tmpbuf, 5, "%zu", state_val);
-
- if (strcmp(tmpbuf, charp) != 0)
- return -EINVAL;
-
- result = acpi_processor_set_throttling(pr, state_val, false);
- if (result)
- return result;
-
- return count;
-}
-
-const struct file_operations acpi_processor_throttling_fops = {
- .owner = THIS_MODULE,
- .open = acpi_processor_throttling_open_fs,
- .read = seq_read,
- .write = acpi_processor_write_throttling,
- .llseek = seq_lseek,
- .release = single_release,
-};
-#endif
diff --git a/drivers/acpi/sbs.c b/drivers/acpi/sbs.c
index e5dbedb..51ae379 100644
--- a/drivers/acpi/sbs.c
+++ b/drivers/acpi/sbs.c
@@ -484,6 +484,8 @@
const struct file_operations *state_fops,
const struct file_operations *alarm_fops, void *data)
{
+ printk(KERN_WARNING PREFIX "Deprecated procfs I/F for SBS is loaded,"
+ " please retry with CONFIG_ACPI_PROCFS_POWER cleared\n");
if (!*dir) {
*dir = proc_mkdir(dir_name, parent_dir);
if (!*dir) {
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 29ef505..b99e624 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -778,7 +778,7 @@
wakeup->resources.handles[i] = element->reference.handle;
}
- acpi_gpe_can_wake(wakeup->gpe_device, wakeup->gpe_number);
+ acpi_setup_gpe_for_wake(handle, wakeup->gpe_device, wakeup->gpe_number);
out:
kfree(buffer.pointer);
@@ -803,7 +803,7 @@
/* Power button, Lid switch always enable wakeup */
if (!acpi_match_device_ids(device, button_device_ids)) {
device->wakeup.flags.run_wake = 1;
- device->wakeup.flags.always_enabled = 1;
+ device_set_wakeup_capable(&device->dev, true);
return;
}
@@ -815,16 +815,22 @@
!!(event_status & ACPI_EVENT_FLAG_HANDLE);
}
-static int acpi_bus_get_wakeup_device_flags(struct acpi_device *device)
+static void acpi_bus_get_wakeup_device_flags(struct acpi_device *device)
{
+ acpi_handle temp;
acpi_status status = 0;
int psw_error;
+ /* Presence of _PRW indicates wake capable */
+ status = acpi_get_handle(device->handle, "_PRW", &temp);
+ if (ACPI_FAILURE(status))
+ return;
+
status = acpi_bus_extract_wakeup_device_power_package(device->handle,
&device->wakeup);
if (ACPI_FAILURE(status)) {
ACPI_EXCEPTION((AE_INFO, status, "Extracting _PRW package"));
- goto end;
+ return;
}
device->wakeup.flags.valid = 1;
@@ -840,13 +846,10 @@
if (psw_error)
ACPI_DEBUG_PRINT((ACPI_DB_INFO,
"error in _DSW or _PSW evaluation\n"));
-
-end:
- if (ACPI_FAILURE(status))
- device->flags.wake_capable = 0;
- return 0;
}
+static void acpi_bus_add_power_resource(acpi_handle handle);
+
static int acpi_bus_get_power_flags(struct acpi_device *device)
{
acpi_status status = 0;
@@ -875,8 +878,12 @@
acpi_evaluate_reference(device->handle, object_name, NULL,
&ps->resources);
if (ps->resources.count) {
+ int j;
+
device->power.flags.power_resources = 1;
ps->flags.valid = 1;
+ for (j = 0; j < ps->resources.count; j++)
+ acpi_bus_add_power_resource(ps->resources.handles[j]);
}
/* Evaluate "_PSx" to see if we can do explicit sets */
@@ -901,10 +908,7 @@
device->power.states[ACPI_STATE_D3].flags.valid = 1;
device->power.states[ACPI_STATE_D3].power = 0;
- /* TBD: System wake support and resource requirements. */
-
- device->power.state = ACPI_STATE_UNKNOWN;
- acpi_bus_get_power(device->handle, &(device->power.state));
+ acpi_bus_init_power(device);
return 0;
}
@@ -947,11 +951,6 @@
if (ACPI_SUCCESS(status))
device->flags.power_manageable = 1;
- /* Presence of _PRW indicates wake capable */
- status = acpi_get_handle(device->handle, "_PRW", &temp);
- if (ACPI_SUCCESS(status))
- device->flags.wake_capable = 1;
-
/* TBD: Performance management */
return 0;
@@ -1278,11 +1277,7 @@
* Wakeup device management
*-----------------------
*/
- if (device->flags.wake_capable) {
- result = acpi_bus_get_wakeup_device_flags(device);
- if (result)
- goto end;
- }
+ acpi_bus_get_wakeup_device_flags(device);
/*
* Performance Management
@@ -1326,6 +1321,20 @@
#define ACPI_STA_DEFAULT (ACPI_STA_DEVICE_PRESENT | ACPI_STA_DEVICE_ENABLED | \
ACPI_STA_DEVICE_UI | ACPI_STA_DEVICE_FUNCTIONING)
+static void acpi_bus_add_power_resource(acpi_handle handle)
+{
+ struct acpi_bus_ops ops = {
+ .acpi_op_add = 1,
+ .acpi_op_start = 1,
+ };
+ struct acpi_device *device = NULL;
+
+ acpi_bus_get_device(handle, &device);
+ if (!device)
+ acpi_add_single_object(&device, handle, ACPI_BUS_TYPE_POWER,
+ ACPI_STA_DEFAULT, &ops);
+}
+
static int acpi_bus_type_and_status(acpi_handle handle, int *type,
unsigned long long *sta)
{
@@ -1371,7 +1380,6 @@
struct acpi_bus_ops *ops = context;
int type;
unsigned long long sta;
- struct acpi_device_wakeup wakeup;
struct acpi_device *device;
acpi_status status;
int result;
@@ -1382,7 +1390,13 @@
if (!(sta & ACPI_STA_DEVICE_PRESENT) &&
!(sta & ACPI_STA_DEVICE_FUNCTIONING)) {
- acpi_bus_extract_wakeup_device_power_package(handle, &wakeup);
+ struct acpi_device_wakeup wakeup;
+ acpi_handle temp;
+
+ status = acpi_get_handle(handle, "_PRW", &temp);
+ if (ACPI_SUCCESS(status))
+ acpi_bus_extract_wakeup_device_power_package(handle,
+ &wakeup);
return AE_CTRL_DEPTH;
}
@@ -1467,7 +1481,7 @@
result = acpi_bus_scan(device->handle, &ops, NULL);
- acpi_update_gpes();
+ acpi_update_all_gpes();
return result;
}
@@ -1573,6 +1587,8 @@
printk(KERN_ERR PREFIX "Could not register bus type\n");
}
+ acpi_power_init();
+
/*
* Enumerate devices in the ACPI namespace.
*/
@@ -1584,7 +1600,7 @@
if (result)
acpi_device_unregister(acpi_root, ACPI_BUS_REMOVAL_NORMAL);
else
- acpi_update_gpes();
+ acpi_update_all_gpes();
return result;
}
diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c
index c423231..fdd3aee 100644
--- a/drivers/acpi/sleep.c
+++ b/drivers/acpi/sleep.c
@@ -124,8 +124,7 @@
static int acpi_pm_pre_suspend(void)
{
acpi_pm_freeze();
- suspend_nvs_save();
- return 0;
+ return suspend_nvs_save();
}
/**
@@ -151,7 +150,7 @@
{
int error = __acpi_pm_prepare();
if (!error)
- acpi_pm_pre_suspend();
+ error = acpi_pm_pre_suspend();
return error;
}
@@ -435,6 +434,14 @@
DMI_MATCH(DMI_PRODUCT_NAME, "VGN-NW130D"),
},
},
+ {
+ .callback = init_nvs_nosave,
+ .ident = "Averatec AV1020-ED2",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "AVERATEC"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "1000 Series"),
+ },
+ },
{},
};
#endif /* CONFIG_SUSPEND */
diff --git a/drivers/acpi/sysfs.c b/drivers/acpi/sysfs.c
index f8588f8..61891e7 100644
--- a/drivers/acpi/sysfs.c
+++ b/drivers/acpi/sysfs.c
@@ -438,7 +438,7 @@
return;
}
-void acpi_os_gpe_count(u32 gpe_number)
+static void gpe_count(u32 gpe_number)
{
acpi_gpe_count++;
@@ -454,7 +454,7 @@
return;
}
-void acpi_os_fixed_event_count(u32 event_number)
+static void fixed_event_count(u32 event_number)
{
if (!all_counters)
return;
@@ -468,6 +468,16 @@
return;
}
+static void acpi_gbl_event_handler(u32 event_type, acpi_handle device,
+ u32 event_number, void *context)
+{
+ if (event_type == ACPI_EVENT_TYPE_GPE)
+ gpe_count(event_number);
+
+ if (event_type == ACPI_EVENT_TYPE_FIXED)
+ fixed_event_count(event_number);
+}
+
static int get_status(u32 index, acpi_event_status *status,
acpi_handle *handle)
{
@@ -601,6 +611,7 @@
void acpi_irq_stats_init(void)
{
+ acpi_status status;
int i;
if (all_counters)
@@ -619,6 +630,10 @@
if (all_counters == NULL)
goto fail;
+ status = acpi_install_global_event_handler(acpi_gbl_event_handler, NULL);
+ if (ACPI_FAILURE(status))
+ goto fail;
+
counter_attrs = kzalloc(sizeof(struct kobj_attribute) * (num_counters),
GFP_KERNEL);
if (counter_attrs == NULL)
diff --git a/drivers/acpi/thermal.c b/drivers/acpi/thermal.c
index 5a27b0a..2607e17 100644
--- a/drivers/acpi/thermal.c
+++ b/drivers/acpi/thermal.c
@@ -1059,8 +1059,9 @@
break;
tz->trips.active[i].flags.enabled = 1;
for (j = 0; j < tz->trips.active[i].devices.count; j++) {
- result = acpi_bus_get_power(tz->trips.active[i].devices.
- handles[j], &power_state);
+ result = acpi_bus_update_power(
+ tz->trips.active[i].devices.handles[j],
+ &power_state);
if (result || (power_state != ACPI_STATE_D0)) {
tz->trips.active[i].flags.enabled = 0;
break;
diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c
index 15a0fde..90f8f76 100644
--- a/drivers/acpi/video.c
+++ b/drivers/acpi/video.c
@@ -33,7 +33,6 @@
#include <linux/input.h>
#include <linux/backlight.h>
#include <linux/thermal.h>
-#include <linux/video_output.h>
#include <linux/sort.h>
#include <linux/pci.h>
#include <linux/pci_ids.h>
@@ -81,6 +80,13 @@
static int allow_duplicates;
module_param(allow_duplicates, bool, 0644);
+/*
+ * Some BIOSes claim they use minimum backlight at boot,
+ * and this may bring dimming screen after boot
+ */
+static int use_bios_initial_backlight = 1;
+module_param(use_bios_initial_backlight, bool, 0644);
+
static int register_count = 0;
static int acpi_video_bus_add(struct acpi_device *device);
static int acpi_video_bus_remove(struct acpi_device *device, int type);
@@ -172,9 +178,6 @@
u8 _BQC:1; /* Get current brightness level */
u8 _BCQ:1; /* Some buggy BIOS uses _BCQ instead of _BQC */
u8 _DDC:1; /*Return the EDID for this device */
- u8 _DCS:1; /*Return status of output device */
- u8 _DGS:1; /*Query graphics state */
- u8 _DSS:1; /*Device state set */
};
struct acpi_video_brightness_flags {
@@ -202,7 +205,6 @@
struct acpi_video_device_brightness *brightness;
struct backlight_device *backlight;
struct thermal_cooling_device *cooling_dev;
- struct output_device *output_dev;
};
static const char device_decode[][30] = {
@@ -226,10 +228,6 @@
u32 level_current, u32 event);
static int acpi_video_switch_brightness(struct acpi_video_device *device,
int event);
-static int acpi_video_device_get_state(struct acpi_video_device *device,
- unsigned long long *state);
-static int acpi_video_output_get(struct output_device *od);
-static int acpi_video_device_set_state(struct acpi_video_device *device, int state);
/*backlight device sysfs support*/
static int acpi_video_get_brightness(struct backlight_device *bd)
@@ -265,30 +263,6 @@
.update_status = acpi_video_set_brightness,
};
-/*video output device sysfs support*/
-static int acpi_video_output_get(struct output_device *od)
-{
- unsigned long long state;
- struct acpi_video_device *vd =
- (struct acpi_video_device *)dev_get_drvdata(&od->dev);
- acpi_video_device_get_state(vd, &state);
- return (int)state;
-}
-
-static int acpi_video_output_set(struct output_device *od)
-{
- unsigned long state = od->request_state;
- struct acpi_video_device *vd=
- (struct acpi_video_device *)dev_get_drvdata(&od->dev);
- return acpi_video_device_set_state(vd, state);
-}
-
-static struct output_properties acpi_output_properties = {
- .set_state = acpi_video_output_set,
- .get_status = acpi_video_output_get,
-};
-
-
/* thermal cooling device callbacks */
static int video_get_max_state(struct thermal_cooling_device *cooling_dev, unsigned
long *state)
@@ -344,34 +318,6 @@
Video Management
-------------------------------------------------------------------------- */
-/* device */
-
-static int
-acpi_video_device_get_state(struct acpi_video_device *device,
- unsigned long long *state)
-{
- int status;
-
- status = acpi_evaluate_integer(device->dev->handle, "_DCS", NULL, state);
-
- return status;
-}
-
-static int
-acpi_video_device_set_state(struct acpi_video_device *device, int state)
-{
- int status;
- union acpi_object arg0 = { ACPI_TYPE_INTEGER };
- struct acpi_object_list args = { 1, &arg0 };
- unsigned long long ret;
-
-
- arg0.integer.value = state;
- status = acpi_evaluate_integer(device->dev->handle, "_DSS", &args, &ret);
-
- return status;
-}
-
static int
acpi_video_device_lcd_query_levels(struct acpi_video_device *device,
union acpi_object **levels)
@@ -766,9 +712,11 @@
* when invoked for the first time, i.e. level_old is invalid.
* set the backlight to max_level in this case
*/
- for (i = 2; i < br->count; i++)
- if (level_old == br->levels[i])
- level = level_old;
+ if (use_bios_initial_backlight) {
+ for (i = 2; i < br->count; i++)
+ if (level_old == br->levels[i])
+ level = level_old;
+ }
goto set_level;
}
@@ -831,15 +779,6 @@
if (ACPI_SUCCESS(acpi_get_handle(device->dev->handle, "_DDC", &h_dummy1))) {
device->cap._DDC = 1;
}
- if (ACPI_SUCCESS(acpi_get_handle(device->dev->handle, "_DCS", &h_dummy1))) {
- device->cap._DCS = 1;
- }
- if (ACPI_SUCCESS(acpi_get_handle(device->dev->handle, "_DGS", &h_dummy1))) {
- device->cap._DGS = 1;
- }
- if (ACPI_SUCCESS(acpi_get_handle(device->dev->handle, "_DSS", &h_dummy1))) {
- device->cap._DSS = 1;
- }
if (acpi_video_backlight_support()) {
struct backlight_properties props;
@@ -904,21 +843,6 @@
printk(KERN_ERR PREFIX "Create sysfs link\n");
}
-
- if (acpi_video_display_switch_support()) {
-
- if (device->cap._DCS && device->cap._DSS) {
- static int count;
- char *name;
- name = kasprintf(GFP_KERNEL, "acpi_video%d", count);
- if (!name)
- return;
- count++;
- device->output_dev = video_output_register(name,
- NULL, device, &acpi_output_properties);
- kfree(name);
- }
- }
}
/*
@@ -1360,6 +1284,9 @@
if (!video_device)
continue;
+ if (!video_device->cap._DDC)
+ continue;
+
if (type) {
switch (type) {
case ACPI_VIDEO_DISPLAY_CRT:
@@ -1452,7 +1379,6 @@
thermal_cooling_device_unregister(device->cooling_dev);
device->cooling_dev = NULL;
}
- video_output_unregister(device->output_dev);
return 0;
}
diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c
index b836761..42d3d72 100644
--- a/drivers/acpi/video_detect.c
+++ b/drivers/acpi/video_detect.c
@@ -17,15 +17,14 @@
* capabilities the graphics cards plugged in support. The check for general
* video capabilities will be triggered by the first caller of
* acpi_video_get_capabilities(NULL); which will happen when the first
- * backlight (or display output) switching supporting driver calls:
+ * backlight switching supporting driver calls:
* acpi_video_backlight_support();
*
* Depending on whether ACPI graphics extensions (cmp. ACPI spec Appendix B)
* are available, video.ko should be used to handle the device.
*
* Otherwise vendor specific drivers like thinkpad_acpi, asus_acpi,
- * sony_acpi,... can take care about backlight brightness and display output
- * switching.
+ * sony_acpi,... can take care about backlight brightness.
*
* If CONFIG_ACPI_VIDEO is neither set as "compiled in" (y) nor as a module (m)
* this file will not be compiled, acpi_video_get_capabilities() and
@@ -83,11 +82,6 @@
if (!device)
return 0;
- /* Is this device able to support video switching ? */
- if (ACPI_SUCCESS(acpi_get_handle(device->handle, "_DOD", &h_dummy)) ||
- ACPI_SUCCESS(acpi_get_handle(device->handle, "_DOS", &h_dummy)))
- video_caps |= ACPI_VIDEO_OUTPUT_SWITCHING;
-
/* Is this device able to retrieve a video ROM ? */
if (ACPI_SUCCESS(acpi_get_handle(device->handle, "_ROM", &h_dummy)))
video_caps |= ACPI_VIDEO_ROM_AVAILABLE;
@@ -161,8 +155,6 @@
*
* if (dmi_name_in_vendors("XY")) {
* acpi_video_support |=
- * ACPI_VIDEO_OUTPUT_SWITCHING_DMI_VENDOR;
- * acpi_video_support |=
* ACPI_VIDEO_BACKLIGHT_DMI_VENDOR;
*}
*/
@@ -212,33 +204,8 @@
EXPORT_SYMBOL(acpi_video_backlight_support);
/*
- * Returns true if video.ko can do display output switching.
- * This does not work well/at all with binary graphics drivers
- * which disable system io ranges and do it on their own.
- */
-int acpi_video_display_switch_support(void)
-{
- if (!acpi_video_caps_checked)
- acpi_video_get_capabilities(NULL);
-
- if (acpi_video_support & ACPI_VIDEO_OUTPUT_SWITCHING_FORCE_VENDOR)
- return 0;
- else if (acpi_video_support & ACPI_VIDEO_OUTPUT_SWITCHING_FORCE_VIDEO)
- return 1;
-
- if (acpi_video_support & ACPI_VIDEO_OUTPUT_SWITCHING_DMI_VENDOR)
- return 0;
- else if (acpi_video_support & ACPI_VIDEO_OUTPUT_SWITCHING_DMI_VIDEO)
- return 1;
-
- return acpi_video_support & ACPI_VIDEO_OUTPUT_SWITCHING;
-}
-EXPORT_SYMBOL(acpi_video_display_switch_support);
-
-/*
- * Use acpi_display_output=vendor/video or acpi_backlight=vendor/video
- * To force that backlight or display output switching is processed by vendor
- * specific acpi drivers or video.ko driver.
+ * Use acpi_backlight=vendor/video to force that backlight switching
+ * is processed by vendor specific acpi drivers or video.ko driver.
*/
static int __init acpi_backlight(char *str)
{
@@ -255,19 +222,3 @@
return 1;
}
__setup("acpi_backlight=", acpi_backlight);
-
-static int __init acpi_display_output(char *str)
-{
- if (str == NULL || *str == '\0')
- return 1;
- else {
- if (!strcmp("vendor", str))
- acpi_video_support |=
- ACPI_VIDEO_OUTPUT_SWITCHING_FORCE_VENDOR;
- if (!strcmp("video", str))
- acpi_video_support |=
- ACPI_VIDEO_OUTPUT_SWITCHING_FORCE_VIDEO;
- }
- return 1;
-}
-__setup("acpi_display_output=", acpi_display_output);
diff --git a/drivers/acpi/wakeup.c b/drivers/acpi/wakeup.c
index f62a50c..ed65014 100644
--- a/drivers/acpi/wakeup.c
+++ b/drivers/acpi/wakeup.c
@@ -37,15 +37,16 @@
container_of(node, struct acpi_device, wakeup_list);
if (!dev->wakeup.flags.valid
- || !(dev->wakeup.state.enabled || dev->wakeup.prepare_count)
- || sleep_state > (u32) dev->wakeup.sleep_state)
+ || sleep_state > (u32) dev->wakeup.sleep_state
+ || !(device_may_wakeup(&dev->dev)
+ || dev->wakeup.prepare_count))
continue;
- if (dev->wakeup.state.enabled)
+ if (device_may_wakeup(&dev->dev))
acpi_enable_wakeup_device_power(dev, sleep_state);
/* The wake-up power should have been enabled already. */
- acpi_gpe_wakeup(dev->wakeup.gpe_device, dev->wakeup.gpe_number,
+ acpi_set_gpe_wake_mask(dev->wakeup.gpe_device, dev->wakeup.gpe_number,
ACPI_GPE_ENABLE);
}
}
@@ -63,14 +64,15 @@
container_of(node, struct acpi_device, wakeup_list);
if (!dev->wakeup.flags.valid
- || !(dev->wakeup.state.enabled || dev->wakeup.prepare_count)
- || (sleep_state > (u32) dev->wakeup.sleep_state))
+ || sleep_state > (u32) dev->wakeup.sleep_state
+ || !(device_may_wakeup(&dev->dev)
+ || dev->wakeup.prepare_count))
continue;
- acpi_gpe_wakeup(dev->wakeup.gpe_device, dev->wakeup.gpe_number,
+ acpi_set_gpe_wake_mask(dev->wakeup.gpe_device, dev->wakeup.gpe_number,
ACPI_GPE_DISABLE);
- if (dev->wakeup.state.enabled)
+ if (device_may_wakeup(&dev->dev))
acpi_disable_wakeup_device_power(dev);
}
}
@@ -84,8 +86,8 @@
struct acpi_device *dev = container_of(node,
struct acpi_device,
wakeup_list);
- if (dev->wakeup.flags.always_enabled)
- dev->wakeup.state.enabled = 1;
+ if (device_can_wakeup(&dev->dev))
+ device_set_wakeup_enable(&dev->dev, true);
}
mutex_unlock(&acpi_device_lock);
return 0;
diff --git a/drivers/base/node.c b/drivers/base/node.c
index ce012a9..36b4305 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -117,12 +117,21 @@
"Node %d WritebackTmp: %8lu kB\n"
"Node %d Slab: %8lu kB\n"
"Node %d SReclaimable: %8lu kB\n"
- "Node %d SUnreclaim: %8lu kB\n",
+ "Node %d SUnreclaim: %8lu kB\n"
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ "Node %d AnonHugePages: %8lu kB\n"
+#endif
+ ,
nid, K(node_page_state(nid, NR_FILE_DIRTY)),
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)),
+ nid, K(node_page_state(nid, NR_ANON_PAGES)
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ + node_page_state(nid, NR_ANON_TRANSPARENT_HUGEPAGES) *
+ HPAGE_PMD_NR
+#endif
+ ),
nid, K(node_page_state(nid, NR_SHMEM)),
nid, node_page_state(nid, NR_KERNEL_STACK) *
THREAD_SIZE / 1024,
@@ -133,7 +142,13 @@
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)));
+ nid, K(node_page_state(nid, NR_SLAB_UNRECLAIMABLE))
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ , nid,
+ K(node_page_state(nid, NR_ANON_TRANSPARENT_HUGEPAGES) *
+ HPAGE_PMD_NR)
+#endif
+ );
n += hugetlb_report_node_meminfo(nid, buf + n);
return n;
}
diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c
index 07e9796..857df10 100644
--- a/drivers/char/agp/intel-agp.c
+++ b/drivers/char/agp/intel-agp.c
@@ -717,8 +717,8 @@
{ PCI_DEVICE_ID_INTEL_82820_UP_HB, "i820", &intel_820_driver },
{ PCI_DEVICE_ID_INTEL_82830_HB, "830M", &intel_830mp_driver },
{ PCI_DEVICE_ID_INTEL_82840_HB, "i840", &intel_840_driver },
- { PCI_DEVICE_ID_INTEL_82845_HB, "845G", &intel_845_driver },
- { PCI_DEVICE_ID_INTEL_82845G_HB, "830M", &intel_845_driver },
+ { PCI_DEVICE_ID_INTEL_82845_HB, "i845", &intel_845_driver },
+ { PCI_DEVICE_ID_INTEL_82845G_HB, "845G", &intel_845_driver },
{ PCI_DEVICE_ID_INTEL_82850_HB, "i850", &intel_850_driver },
{ PCI_DEVICE_ID_INTEL_82854_HB, "854", &intel_845_driver },
{ PCI_DEVICE_ID_INTEL_82855PM_HB, "855PM", &intel_845_driver },
diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index e921b69..826ab09 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -1361,7 +1361,7 @@
&i81x_gtt_driver},
{ PCI_DEVICE_ID_INTEL_82830_CGC, "830M",
&i8xx_gtt_driver},
- { PCI_DEVICE_ID_INTEL_82845G_IG, "830M",
+ { PCI_DEVICE_ID_INTEL_82845G_IG, "845G",
&i8xx_gtt_driver},
{ PCI_DEVICE_ID_INTEL_82854_IG, "854",
&i8xx_gtt_driver},
diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c
index 2fe72f8..38223e9 100644
--- a/drivers/char/ipmi/ipmi_msghandler.c
+++ b/drivers/char/ipmi/ipmi_msghandler.c
@@ -970,6 +970,33 @@
}
EXPORT_SYMBOL(ipmi_create_user);
+int ipmi_get_smi_info(int if_num, struct ipmi_smi_info *data)
+{
+ int rv = 0;
+ ipmi_smi_t intf;
+ struct ipmi_smi_handlers *handlers;
+
+ mutex_lock(&ipmi_interfaces_mutex);
+ list_for_each_entry_rcu(intf, &ipmi_interfaces, link) {
+ if (intf->intf_num == if_num)
+ goto found;
+ }
+ /* Not found, return an error */
+ rv = -EINVAL;
+ mutex_unlock(&ipmi_interfaces_mutex);
+ return rv;
+
+found:
+ handlers = intf->handlers;
+ rv = -ENOSYS;
+ if (handlers->get_smi_info)
+ rv = handlers->get_smi_info(intf->send_info, data);
+ mutex_unlock(&ipmi_interfaces_mutex);
+
+ return rv;
+}
+EXPORT_SYMBOL(ipmi_get_smi_info);
+
static void free_user(struct kref *ref)
{
ipmi_user_t user = container_of(ref, struct ipmi_user, refcount);
diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c
index f27c04e..b6ae6e9 100644
--- a/drivers/char/ipmi/ipmi_si_intf.c
+++ b/drivers/char/ipmi/ipmi_si_intf.c
@@ -57,6 +57,7 @@
#include <asm/irq.h>
#include <linux/interrupt.h>
#include <linux/rcupdate.h>
+#include <linux/ipmi.h>
#include <linux/ipmi_smi.h>
#include <asm/io.h>
#include "ipmi_si_sm.h"
@@ -109,10 +110,6 @@
};
static char *si_to_str[] = { "kcs", "smic", "bt" };
-enum ipmi_addr_src {
- SI_INVALID = 0, SI_HOTMOD, SI_HARDCODED, SI_SPMI, SI_ACPI, SI_SMBIOS,
- SI_PCI, SI_DEVICETREE, SI_DEFAULT
-};
static char *ipmi_addr_src_to_str[] = { NULL, "hotmod", "hardcoded", "SPMI",
"ACPI", "SMBIOS", "PCI",
"device-tree", "default" };
@@ -293,6 +290,7 @@
struct task_struct *thread;
struct list_head link;
+ union ipmi_smi_info_union addr_info;
};
#define smi_inc_stat(smi, stat) \
@@ -1188,6 +1186,18 @@
return 0;
}
+static int get_smi_info(void *send_info, struct ipmi_smi_info *data)
+{
+ struct smi_info *smi = send_info;
+
+ data->addr_src = smi->addr_source;
+ data->dev = smi->dev;
+ data->addr_info = smi->addr_info;
+ get_device(smi->dev);
+
+ return 0;
+}
+
static void set_maintenance_mode(void *send_info, int enable)
{
struct smi_info *smi_info = send_info;
@@ -1199,6 +1209,7 @@
static struct ipmi_smi_handlers handlers = {
.owner = THIS_MODULE,
.start_processing = smi_start_processing,
+ .get_smi_info = get_smi_info,
.sender = sender,
.request_events = request_events,
.set_maintenance_mode = set_maintenance_mode,
@@ -1930,7 +1941,8 @@
static int acpi_failure;
/* For GPE-type interrupts. */
-static u32 ipmi_acpi_gpe(void *context)
+static u32 ipmi_acpi_gpe(acpi_handle gpe_device,
+ u32 gpe_number, void *context)
{
struct smi_info *smi_info = context;
unsigned long flags;
@@ -2158,6 +2170,7 @@
printk(KERN_INFO PFX "probing via ACPI\n");
handle = acpi_dev->handle;
+ info->addr_info.acpi_info.acpi_handle = handle;
/* _IFT tells us the interface type: KCS, BT, etc */
status = acpi_evaluate_integer(handle, "_IFT", NULL, &tmp);
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 386888f..bf50924 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -96,7 +96,15 @@
/* enter the state and update stats */
dev->last_state = target_state;
+
+ trace_power_start(POWER_CSTATE, next_state, dev->cpu);
+ trace_cpu_idle(next_state, dev->cpu);
+
dev->last_residency = target_state->enter(dev, target_state);
+
+ trace_power_end(dev->cpu);
+ trace_cpu_idle(PWR_EVENT_EXIT, dev->cpu);
+
if (dev->last_state)
target_state = dev->last_state;
@@ -106,8 +114,6 @@
/* give the governor an opportunity to reflect on the outcome */
if (cpuidle_curr_governor->reflect)
cpuidle_curr_governor->reflect(dev);
- trace_power_end(smp_processor_id());
- trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
}
/**
@@ -155,6 +161,45 @@
EXPORT_SYMBOL_GPL(cpuidle_resume_and_unlock);
+#ifdef CONFIG_ARCH_HAS_CPU_RELAX
+static int poll_idle(struct cpuidle_device *dev, struct cpuidle_state *st)
+{
+ ktime_t t1, t2;
+ s64 diff;
+ int ret;
+
+ t1 = ktime_get();
+ local_irq_enable();
+ while (!need_resched())
+ cpu_relax();
+
+ t2 = ktime_get();
+ diff = ktime_to_us(ktime_sub(t2, t1));
+ if (diff > INT_MAX)
+ diff = INT_MAX;
+
+ ret = (int) diff;
+ return ret;
+}
+
+static void poll_idle_init(struct cpuidle_device *dev)
+{
+ struct cpuidle_state *state = &dev->states[0];
+
+ cpuidle_set_statedata(state, NULL);
+
+ snprintf(state->name, CPUIDLE_NAME_LEN, "POLL");
+ snprintf(state->desc, CPUIDLE_DESC_LEN, "CPUIDLE CORE POLL IDLE");
+ state->exit_latency = 0;
+ state->target_residency = 0;
+ state->power_usage = -1;
+ state->flags = 0;
+ state->enter = poll_idle;
+}
+#else
+static void poll_idle_init(struct cpuidle_device *dev) {}
+#endif /* CONFIG_ARCH_HAS_CPU_RELAX */
+
/**
* cpuidle_enable_device - enables idle PM for a CPU
* @dev: the CPU
@@ -179,6 +224,8 @@
return ret;
}
+ poll_idle_init(dev);
+
if ((ret = cpuidle_add_state_sysfs(dev)))
return ret;
@@ -233,45 +280,6 @@
EXPORT_SYMBOL_GPL(cpuidle_disable_device);
-#ifdef CONFIG_ARCH_HAS_CPU_RELAX
-static int poll_idle(struct cpuidle_device *dev, struct cpuidle_state *st)
-{
- ktime_t t1, t2;
- s64 diff;
- int ret;
-
- t1 = ktime_get();
- local_irq_enable();
- while (!need_resched())
- cpu_relax();
-
- t2 = ktime_get();
- diff = ktime_to_us(ktime_sub(t2, t1));
- if (diff > INT_MAX)
- diff = INT_MAX;
-
- ret = (int) diff;
- return ret;
-}
-
-static void poll_idle_init(struct cpuidle_device *dev)
-{
- struct cpuidle_state *state = &dev->states[0];
-
- cpuidle_set_statedata(state, NULL);
-
- snprintf(state->name, CPUIDLE_NAME_LEN, "C0");
- snprintf(state->desc, CPUIDLE_DESC_LEN, "CPUIDLE CORE POLL IDLE");
- state->exit_latency = 0;
- state->target_residency = 0;
- state->power_usage = -1;
- state->flags = CPUIDLE_FLAG_POLL;
- state->enter = poll_idle;
-}
-#else
-static void poll_idle_init(struct cpuidle_device *dev) {}
-#endif /* CONFIG_ARCH_HAS_CPU_RELAX */
-
/**
* __cpuidle_register_device - internal register function called before register
* and enable routines
@@ -292,8 +300,6 @@
init_completion(&dev->kobj_unregister);
- poll_idle_init(dev);
-
/*
* cpuidle driver should set the dev->power_specified bit
* before registering the device if the driver provides
diff --git a/drivers/gpio/cs5535-gpio.c b/drivers/gpio/cs5535-gpio.c
index 815d98b2..0d05ea7 100644
--- a/drivers/gpio/cs5535-gpio.c
+++ b/drivers/gpio/cs5535-gpio.c
@@ -11,14 +11,13 @@
#include <linux/kernel.h>
#include <linux/spinlock.h>
#include <linux/module.h>
-#include <linux/pci.h>
+#include <linux/platform_device.h>
#include <linux/gpio.h>
#include <linux/io.h>
#include <linux/cs5535.h>
#include <asm/msr.h>
#define DRV_NAME "cs5535-gpio"
-#define GPIO_BAR 1
/*
* Some GPIO pins
@@ -47,7 +46,7 @@
struct gpio_chip chip;
resource_size_t base;
- struct pci_dev *pdev;
+ struct platform_device *pdev;
spinlock_t lock;
} cs5535_gpio_chip;
@@ -301,10 +300,10 @@
},
};
-static int __init cs5535_gpio_probe(struct pci_dev *pdev,
- const struct pci_device_id *pci_id)
+static int __devinit cs5535_gpio_probe(struct platform_device *pdev)
{
- int err;
+ struct resource *res;
+ int err = -EIO;
ulong mask_orig = mask;
/* There are two ways to get the GPIO base address; one is by
@@ -314,25 +313,23 @@
* it turns out to be unreliable in the face of crappy BIOSes, we
* can always go back to using MSRs.. */
- err = pci_enable_device_io(pdev);
- if (err) {
- dev_err(&pdev->dev, "can't enable device IO\n");
+ res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+ if (!res) {
+ dev_err(&pdev->dev, "can't fetch device resource info\n");
goto done;
}
- err = pci_request_region(pdev, GPIO_BAR, DRV_NAME);
- if (err) {
- dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", GPIO_BAR);
+ if (!request_region(res->start, resource_size(res), pdev->name)) {
+ dev_err(&pdev->dev, "can't request region\n");
goto done;
}
/* set up the driver-specific struct */
- cs5535_gpio_chip.base = pci_resource_start(pdev, GPIO_BAR);
+ cs5535_gpio_chip.base = res->start;
cs5535_gpio_chip.pdev = pdev;
spin_lock_init(&cs5535_gpio_chip.lock);
- dev_info(&pdev->dev, "allocated PCI BAR #%d: base 0x%llx\n", GPIO_BAR,
- (unsigned long long) cs5535_gpio_chip.base);
+ dev_info(&pdev->dev, "reserved resource region %pR\n", res);
/* mask out reserved pins */
mask &= 0x1F7FFFFF;
@@ -350,78 +347,49 @@
if (err)
goto release_region;
- dev_info(&pdev->dev, DRV_NAME ": GPIO support successfully loaded.\n");
+ dev_info(&pdev->dev, "GPIO support successfully loaded.\n");
return 0;
release_region:
- pci_release_region(pdev, GPIO_BAR);
+ release_region(res->start, resource_size(res));
done:
return err;
}
-static void __exit cs5535_gpio_remove(struct pci_dev *pdev)
+static int __devexit cs5535_gpio_remove(struct platform_device *pdev)
{
+ struct resource *r;
int err;
err = gpiochip_remove(&cs5535_gpio_chip.chip);
if (err) {
/* uhh? */
dev_err(&pdev->dev, "unable to remove gpio_chip?\n");
+ return err;
}
- pci_release_region(pdev, GPIO_BAR);
+
+ r = platform_get_resource(pdev, IORESOURCE_IO, 0);
+ release_region(r->start, resource_size(r));
+ return 0;
}
-static struct pci_device_id cs5535_gpio_pci_tbl[] = {
- { PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_CS5535_ISA) },
- { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA) },
- { 0, },
+static struct platform_driver cs5535_gpio_drv = {
+ .driver = {
+ .name = DRV_NAME,
+ .owner = THIS_MODULE,
+ },
+ .probe = cs5535_gpio_probe,
+ .remove = __devexit_p(cs5535_gpio_remove),
};
-MODULE_DEVICE_TABLE(pci, cs5535_gpio_pci_tbl);
-
-/*
- * We can't use the standard PCI driver registration stuff here, since
- * that allows only one driver to bind to each PCI device (and we want
- * multiple drivers to be able to bind to the device). Instead, manually
- * scan for the PCI device, request a single region, and keep track of the
- * devices that we're using.
- */
-
-static int __init cs5535_gpio_scan_pci(void)
-{
- struct pci_dev *pdev;
- int err = -ENODEV;
- int i;
-
- for (i = 0; i < ARRAY_SIZE(cs5535_gpio_pci_tbl); i++) {
- pdev = pci_get_device(cs5535_gpio_pci_tbl[i].vendor,
- cs5535_gpio_pci_tbl[i].device, NULL);
- if (pdev) {
- err = cs5535_gpio_probe(pdev, &cs5535_gpio_pci_tbl[i]);
- if (err)
- pci_dev_put(pdev);
-
- /* we only support a single CS5535/6 southbridge */
- break;
- }
- }
-
- return err;
-}
-
-static void __exit cs5535_gpio_free_pci(void)
-{
- cs5535_gpio_remove(cs5535_gpio_chip.pdev);
- pci_dev_put(cs5535_gpio_chip.pdev);
-}
static int __init cs5535_gpio_init(void)
{
- return cs5535_gpio_scan_pci();
+ return platform_driver_register(&cs5535_gpio_drv);
}
static void __exit cs5535_gpio_exit(void)
{
- cs5535_gpio_free_pci();
+ platform_driver_unregister(&cs5535_gpio_drv);
}
module_init(cs5535_gpio_init);
@@ -430,3 +398,4 @@
MODULE_AUTHOR("Andres Salomon <dilinger@queued.net>");
MODULE_DESCRIPTION("AMD CS5535/CS5536 GPIO driver");
MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:" DRV_NAME);
diff --git a/drivers/gpio/timbgpio.c b/drivers/gpio/timbgpio.c
index 349131e..58c8f30 100644
--- a/drivers/gpio/timbgpio.c
+++ b/drivers/gpio/timbgpio.c
@@ -193,13 +193,13 @@
return ret;
}
-static void timbgpio_irq(struct irq_data *d, struct irq_desc *desc)
+static void timbgpio_irq(unsigned int irq, struct irq_desc *desc)
{
- struct timbgpio *tgpio = irq_data_get_irq_data(d);
+ struct timbgpio *tgpio = get_irq_data(irq);
unsigned long ipr;
int offset;
- desc->irq_data.chip->ack(irq_get_irq_data(d));
+ desc->irq_data.chip->irq_ack(irq_get_irq_data(irq));
ipr = ioread32(tgpio->membase + TGPIO_IPR);
iowrite32(ipr, tgpio->membase + TGPIO_ICR);
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 7af4436..64828a7 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -107,7 +107,6 @@
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 VIDEO_OUTPUT_CONTROL if ACPI
select BACKLIGHT_CLASS_DEVICE if ACPI
select INPUT if ACPI
select ACPI_VIDEO if ACPI
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 0307d60..5c4f9b9 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -607,25 +607,6 @@
}
EXPORT_SYMBOL(drm_fb_helper_fini);
-void drm_fb_helper_fill_fix(struct fb_info *info, struct drm_framebuffer *fb)
-{
- info->fix.type = FB_TYPE_PACKED_PIXELS;
- info->fix.visual = fb->depth == 8 ? FB_VISUAL_PSEUDOCOLOR :
- FB_VISUAL_TRUECOLOR;
- info->fix.mmio_start = 0;
- info->fix.mmio_len = 0;
- info->fix.type_aux = 0;
- info->fix.xpanstep = 1; /* doing it in hw */
- info->fix.ypanstep = 1; /* doing it in hw */
- info->fix.ywrapstep = 0;
- info->fix.accel = FB_ACCEL_NONE;
- info->fix.type_aux = 0;
-
- info->fix.line_length = fb->pitch;
- return;
-}
-EXPORT_SYMBOL(drm_fb_helper_fill_fix);
-
static int setcolreg(struct drm_crtc *crtc, u16 red, u16 green,
u16 blue, u16 regno, struct fb_info *info)
{
@@ -835,7 +816,6 @@
mutex_unlock(&dev->mode_config.mutex);
return ret;
}
- drm_fb_helper_fill_fix(info, fb_helper->fb);
}
mutex_unlock(&dev->mode_config.mutex);
@@ -973,7 +953,6 @@
if (new_fb) {
info->var.pixclock = 0;
- drm_fb_helper_fill_fix(info, fb_helper->fb);
if (register_framebuffer(info) < 0) {
return -EINVAL;
}
@@ -1000,6 +979,26 @@
}
EXPORT_SYMBOL(drm_fb_helper_single_fb_probe);
+void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
+ uint32_t depth)
+{
+ info->fix.type = FB_TYPE_PACKED_PIXELS;
+ info->fix.visual = depth == 8 ? FB_VISUAL_PSEUDOCOLOR :
+ FB_VISUAL_TRUECOLOR;
+ info->fix.mmio_start = 0;
+ info->fix.mmio_len = 0;
+ info->fix.type_aux = 0;
+ info->fix.xpanstep = 1; /* doing it in hw */
+ info->fix.ypanstep = 1; /* doing it in hw */
+ info->fix.ywrapstep = 0;
+ info->fix.accel = FB_ACCEL_NONE;
+ info->fix.type_aux = 0;
+
+ info->fix.line_length = pitch;
+ return;
+}
+EXPORT_SYMBOL(drm_fb_helper_fill_fix);
+
void drm_fb_helper_fill_var(struct fb_info *info, struct drm_fb_helper *fb_helper,
uint32_t fb_width, uint32_t fb_height)
{
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 19a3d58..3601466 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -703,7 +703,7 @@
seq_printf(m, "%s [%d]:\n", name, count);
while (count--) {
- seq_printf(m, " %08x %8zd %04x %04x %08x%s%s%s%s%s%s",
+ seq_printf(m, " %08x %8u %04x %04x %08x%s%s%s%s%s%s",
err->gtt_offset,
err->size,
err->read_domains,
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 0de75a2..72fea2b 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -49,6 +49,9 @@
unsigned int i915_lvds_downclock = 0;
module_param_named(lvds_downclock, i915_lvds_downclock, int, 0400);
+unsigned int i915_panel_use_ssc = 1;
+module_param_named(lvds_use_ssc, i915_panel_use_ssc, int, 0600);
+
bool i915_try_reset = true;
module_param_named(reset, i915_try_reset, bool, 0600);
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 385fc7e..5969f46 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -954,6 +954,7 @@
extern unsigned int i915_fbpercrtc;
extern unsigned int i915_powersave;
extern unsigned int i915_lvds_downclock;
+extern unsigned int i915_panel_use_ssc;
extern int i915_suspend(struct drm_device *dev, pm_message_t state);
extern int i915_resume(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index e698343..dcfdf41 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -464,8 +464,6 @@
int ret;
list_for_each_entry(obj, objects, exec_list) {
- obj->base.pending_read_domains = 0;
- obj->base.pending_write_domain = 0;
ret = i915_gem_execbuffer_relocate_object(obj, eb);
if (ret)
return ret;
@@ -505,6 +503,9 @@
list_move(&obj->exec_list, &ordered_objects);
else
list_move_tail(&obj->exec_list, &ordered_objects);
+
+ obj->base.pending_read_domains = 0;
+ obj->base.pending_write_domain = 0;
}
list_splice(&ordered_objects, objects);
@@ -636,6 +637,7 @@
{
struct drm_i915_gem_relocation_entry *reloc;
struct drm_i915_gem_object *obj;
+ int *reloc_offset;
int i, total, ret;
/* We may process another execbuffer during the unlock... */
@@ -653,8 +655,11 @@
for (i = 0; i < count; i++)
total += exec[i].relocation_count;
+ reloc_offset = drm_malloc_ab(count, sizeof(*reloc_offset));
reloc = drm_malloc_ab(total, sizeof(*reloc));
- if (reloc == NULL) {
+ if (reloc == NULL || reloc_offset == NULL) {
+ drm_free_large(reloc);
+ drm_free_large(reloc_offset);
mutex_lock(&dev->struct_mutex);
return -ENOMEM;
}
@@ -672,6 +677,7 @@
goto err;
}
+ reloc_offset[i] = total;
total += exec[i].relocation_count;
}
@@ -705,17 +711,12 @@
if (ret)
goto err;
- total = 0;
list_for_each_entry(obj, objects, exec_list) {
- obj->base.pending_read_domains = 0;
- obj->base.pending_write_domain = 0;
+ int offset = obj->exec_entry - exec;
ret = i915_gem_execbuffer_relocate_object_slow(obj, eb,
- reloc + total);
+ reloc + reloc_offset[offset]);
if (ret)
goto err;
-
- total += exec->relocation_count;
- exec++;
}
/* Leave the user relocations as are, this is the painfully slow path,
@@ -726,6 +727,7 @@
err:
drm_free_large(reloc);
+ drm_free_large(reloc_offset);
return ret;
}
@@ -770,7 +772,8 @@
if (from == NULL || to == from)
return 0;
- if (INTEL_INFO(obj->base.dev)->gen < 6)
+ /* XXX gpu semaphores are currently causing hard hangs on SNB mobile */
+ if (INTEL_INFO(obj->base.dev)->gen < 6 || IS_MOBILE(obj->base.dev))
return i915_gem_object_wait_rendering(obj, true);
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 e418e8b..b8e509a 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -720,7 +720,7 @@
if (obj->ring != ring)
continue;
- if (!i915_seqno_passed(obj->last_rendering_seqno, seqno))
+ if (i915_seqno_passed(seqno, obj->last_rendering_seqno))
continue;
if ((obj->base.read_domains & I915_GEM_DOMAIN_COMMAND) == 0)
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index b0b1200..0b44956 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -264,17 +264,12 @@
dev_priv->int_crt_support = general->int_crt_support;
dev_priv->lvds_use_ssc = general->enable_ssc;
- if (dev_priv->lvds_use_ssc) {
- if (IS_I85X(dev))
- dev_priv->lvds_ssc_freq =
- general->ssc_freq ? 66 : 48;
- else if (IS_GEN5(dev) || IS_GEN6(dev))
- dev_priv->lvds_ssc_freq =
- general->ssc_freq ? 100 : 120;
- else
- dev_priv->lvds_ssc_freq =
- general->ssc_freq ? 100 : 96;
- }
+ if (IS_I85X(dev))
+ dev_priv->lvds_ssc_freq = general->ssc_freq ? 66 : 48;
+ else if (IS_GEN5(dev) || IS_GEN6(dev))
+ dev_priv->lvds_ssc_freq = general->ssc_freq ? 100 : 120;
+ else
+ dev_priv->lvds_ssc_freq = general->ssc_freq ? 100 : 96;
}
}
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 25d9688..98967f3 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -3822,6 +3822,11 @@
sr_hdisplay, sr_htotal, pixel_size);
}
+static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
+{
+ return dev_priv->lvds_use_ssc && i915_panel_use_ssc;
+}
+
static int intel_crtc_mode_set(struct drm_crtc *crtc,
struct drm_display_mode *mode,
struct drm_display_mode *adjusted_mode,
@@ -3884,7 +3889,7 @@
num_connectors++;
}
- if (is_lvds && dev_priv->lvds_use_ssc && num_connectors < 2) {
+ if (is_lvds && intel_panel_use_ssc(dev_priv) && num_connectors < 2) {
refclk = dev_priv->lvds_ssc_freq * 1000;
DRM_DEBUG_KMS("using SSC reference clock of %d MHz\n",
refclk / 1000);
@@ -4059,7 +4064,7 @@
udelay(200);
if (has_edp_encoder) {
- if (dev_priv->lvds_use_ssc) {
+ if (intel_panel_use_ssc(dev_priv)) {
temp |= DREF_SSC1_ENABLE;
I915_WRITE(PCH_DREF_CONTROL, temp);
@@ -4070,13 +4075,13 @@
/* Enable CPU source on CPU attached eDP */
if (!intel_encoder_is_pch_edp(&has_edp_encoder->base)) {
- if (dev_priv->lvds_use_ssc)
+ if (intel_panel_use_ssc(dev_priv))
temp |= DREF_CPU_SOURCE_OUTPUT_DOWNSPREAD;
else
temp |= DREF_CPU_SOURCE_OUTPUT_NONSPREAD;
} else {
/* Enable SSC on PCH eDP if needed */
- if (dev_priv->lvds_use_ssc) {
+ if (intel_panel_use_ssc(dev_priv)) {
DRM_ERROR("enabling SSC on PCH\n");
temp |= DREF_SUPERSPREAD_SOURCE_ENABLE;
}
@@ -4104,7 +4109,7 @@
int factor = 21;
if (is_lvds) {
- if ((dev_priv->lvds_use_ssc &&
+ if ((intel_panel_use_ssc(dev_priv) &&
dev_priv->lvds_ssc_freq == 100) ||
(I915_READ(PCH_LVDS) & LVDS_CLKB_POWER_MASK) == LVDS_CLKB_POWER_UP)
factor = 25;
@@ -4183,7 +4188,7 @@
/* XXX: just matching BIOS for now */
/* dpll |= PLL_REF_INPUT_TVCLKINBC; */
dpll |= 3;
- else if (is_lvds && dev_priv->lvds_use_ssc && num_connectors < 2)
+ else if (is_lvds && intel_panel_use_ssc(dev_priv) && num_connectors < 2)
dpll |= PLLB_REF_INPUT_SPREADSPECTRUMIN;
else
dpll |= PLL_REF_INPUT_DREFCLK;
diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c
index ee145a2..5127827 100644
--- a/drivers/gpu/drm/i915/intel_fb.c
+++ b/drivers/gpu/drm/i915/intel_fb.c
@@ -148,6 +148,7 @@
// memset(info->screen_base, 0, size);
+ drm_fb_helper_fill_fix(info, fb->pitch, 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_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 8f4f6bd..ace8d5d 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -704,6 +704,14 @@
},
{
.callback = intel_no_lvds_dmi_callback,
+ .ident = "AOpen i915GMm-HFS",
+ .matches = {
+ DMI_MATCH(DMI_BOARD_VENDOR, "AOpen"),
+ DMI_MATCH(DMI_BOARD_NAME, "i915GMm-HFS"),
+ },
+ },
+ {
+ .callback = intel_no_lvds_dmi_callback,
.ident = "Aopen i945GTt-VFA",
.matches = {
DMI_MATCH(DMI_PRODUCT_VERSION, "AO00001JW"),
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index e00d200..c65992d 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -278,6 +278,6 @@
{
struct drm_i915_private *dev_priv = dev->dev_private;
- dev_priv->backlight_level = intel_panel_get_max_backlight(dev);
+ dev_priv->backlight_level = intel_panel_get_backlight(dev);
dev_priv->backlight_enabled = dev_priv->backlight_level != 0;
}
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index a26d047..6d56a54 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -359,6 +359,7 @@
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_var(info, &nfbdev->helper, sizes->fb_width, sizes->fb_height);
/* Set aperture base/size for vesafb takeover */
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index ca32e9c..66324b5 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -225,6 +225,8 @@
strcpy(info->fix.id, "radeondrmfb");
+ drm_fb_helper_fill_fix(info, fb->pitch, fb->depth);
+
info->flags = FBINFO_DEFAULT | FBINFO_CAN_FORCE_OUTPUT;
info->fbops = &radeonfb_ops;
diff --git a/drivers/gpu/stub/Kconfig b/drivers/gpu/stub/Kconfig
index 0e1edd7..09aea5f 100644
--- a/drivers/gpu/stub/Kconfig
+++ b/drivers/gpu/stub/Kconfig
@@ -3,7 +3,6 @@
depends on PCI
# Poulsbo stub depends on ACPI_VIDEO when ACPI is enabled
# but for select to work, need to select ACPI_VIDEO's dependencies, ick
- select VIDEO_OUTPUT_CONTROL if ACPI
select BACKLIGHT_CLASS_DEVICE if ACPI
select INPUT if ACPI
select ACPI_VIDEO if ACPI
diff --git a/drivers/i2c/busses/scx200_acb.c b/drivers/i2c/busses/scx200_acb.c
index 53fab51..986e5f6 100644
--- a/drivers/i2c/busses/scx200_acb.c
+++ b/drivers/i2c/busses/scx200_acb.c
@@ -29,6 +29,7 @@
#include <linux/init.h>
#include <linux/i2c.h>
#include <linux/pci.h>
+#include <linux/platform_device.h>
#include <linux/delay.h>
#include <linux/mutex.h>
#include <linux/slab.h>
@@ -40,6 +41,7 @@
MODULE_AUTHOR("Christer Weinigel <wingel@nano-system.com>");
MODULE_DESCRIPTION("NatSemi SCx200 ACCESS.bus Driver");
+MODULE_ALIAS("platform:cs5535-smb");
MODULE_LICENSE("GPL");
#define MAX_DEVICES 4
@@ -84,10 +86,6 @@
u8 *ptr;
char needs_reset;
unsigned len;
-
- /* PCI device info */
- struct pci_dev *pdev;
- int bar;
};
/* Register Definitions */
@@ -391,7 +389,7 @@
static struct scx200_acb_iface *scx200_acb_list;
static DEFINE_MUTEX(scx200_acb_list_mutex);
-static __init int scx200_acb_probe(struct scx200_acb_iface *iface)
+static __devinit int scx200_acb_probe(struct scx200_acb_iface *iface)
{
u8 val;
@@ -427,7 +425,7 @@
return 0;
}
-static __init struct scx200_acb_iface *scx200_create_iface(const char *text,
+static __devinit struct scx200_acb_iface *scx200_create_iface(const char *text,
struct device *dev, int index)
{
struct scx200_acb_iface *iface;
@@ -452,7 +450,7 @@
return iface;
}
-static int __init scx200_acb_create(struct scx200_acb_iface *iface)
+static int __devinit scx200_acb_create(struct scx200_acb_iface *iface)
{
struct i2c_adapter *adapter;
int rc;
@@ -472,67 +470,31 @@
return -ENODEV;
}
- mutex_lock(&scx200_acb_list_mutex);
- iface->next = scx200_acb_list;
- scx200_acb_list = iface;
- mutex_unlock(&scx200_acb_list_mutex);
+ if (!adapter->dev.parent) {
+ /* If there's no dev, we're tracking (ISA) ifaces manually */
+ mutex_lock(&scx200_acb_list_mutex);
+ iface->next = scx200_acb_list;
+ scx200_acb_list = iface;
+ mutex_unlock(&scx200_acb_list_mutex);
+ }
return 0;
}
-static __init int scx200_create_pci(const char *text, struct pci_dev *pdev,
- int bar)
+static struct scx200_acb_iface * __devinit scx200_create_dev(const char *text,
+ unsigned long base, int index, struct device *dev)
{
struct scx200_acb_iface *iface;
int rc;
- iface = scx200_create_iface(text, &pdev->dev, 0);
+ iface = scx200_create_iface(text, dev, index);
if (iface == NULL)
- return -ENOMEM;
-
- iface->pdev = pdev;
- iface->bar = bar;
-
- rc = pci_enable_device_io(iface->pdev);
- if (rc)
- goto errout_free;
-
- rc = pci_request_region(iface->pdev, iface->bar, iface->adapter.name);
- if (rc) {
- printk(KERN_ERR NAME ": can't allocate PCI BAR %d\n",
- iface->bar);
- goto errout_free;
- }
-
- iface->base = pci_resource_start(iface->pdev, iface->bar);
- rc = scx200_acb_create(iface);
-
- if (rc == 0)
- return 0;
-
- pci_release_region(iface->pdev, iface->bar);
- pci_dev_put(iface->pdev);
- errout_free:
- kfree(iface);
- return rc;
-}
-
-static int __init scx200_create_isa(const char *text, unsigned long base,
- int index)
-{
- struct scx200_acb_iface *iface;
- int rc;
-
- iface = scx200_create_iface(text, NULL, index);
-
- if (iface == NULL)
- return -ENOMEM;
+ return NULL;
if (!request_region(base, 8, iface->adapter.name)) {
printk(KERN_ERR NAME ": can't allocate io 0x%lx-0x%lx\n",
base, base + 8 - 1);
- rc = -EBUSY;
goto errout_free;
}
@@ -540,115 +502,113 @@
rc = scx200_acb_create(iface);
if (rc == 0)
- return 0;
+ return iface;
release_region(base, 8);
errout_free:
kfree(iface);
- return rc;
+ return NULL;
}
-/* Driver data is an index into the scx200_data array that indicates
- * the name and the BAR where the I/O address resource is located. ISA
- * devices are flagged with a bar value of -1 */
+static int __devinit scx200_probe(struct platform_device *pdev)
+{
+ struct scx200_acb_iface *iface;
+ struct resource *res;
-static const struct pci_device_id scx200_pci[] __initconst = {
- { PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_SCx200_BRIDGE),
- .driver_data = 0 },
- { PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_SC1100_BRIDGE),
- .driver_data = 0 },
- { PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_CS5535_ISA),
- .driver_data = 1 },
- { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA),
- .driver_data = 2 },
+ res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+ if (!res) {
+ dev_err(&pdev->dev, "can't fetch device resource info\n");
+ return -ENODEV;
+ }
+
+ iface = scx200_create_dev("CS5535", res->start, 0, &pdev->dev);
+ if (!iface)
+ return -EIO;
+
+ dev_info(&pdev->dev, "SCx200 device '%s' registered\n",
+ iface->adapter.name);
+ platform_set_drvdata(pdev, iface);
+
+ return 0;
+}
+
+static void __devexit scx200_cleanup_iface(struct scx200_acb_iface *iface)
+{
+ i2c_del_adapter(&iface->adapter);
+ release_region(iface->base, 8);
+ kfree(iface);
+}
+
+static int __devexit scx200_remove(struct platform_device *pdev)
+{
+ struct scx200_acb_iface *iface;
+
+ iface = platform_get_drvdata(pdev);
+ platform_set_drvdata(pdev, NULL);
+ scx200_cleanup_iface(iface);
+
+ return 0;
+}
+
+static struct platform_driver scx200_pci_drv = {
+ .driver = {
+ .name = "cs5535-smb",
+ .owner = THIS_MODULE,
+ },
+ .probe = scx200_probe,
+ .remove = __devexit_p(scx200_remove),
+};
+
+static const struct pci_device_id scx200_isa[] __initconst = {
+ { PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_SCx200_BRIDGE) },
+ { PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_SC1100_BRIDGE) },
{ 0, }
};
-static struct {
- const char *name;
- int bar;
-} scx200_data[] = {
- { "SCx200", -1 },
- { "CS5535", 0 },
- { "CS5536", 0 }
-};
-
-static __init int scx200_scan_pci(void)
+static __init void scx200_scan_isa(void)
{
- int data, dev;
- int rc = -ENODEV;
- struct pci_dev *pdev;
+ int i;
- for(dev = 0; dev < ARRAY_SIZE(scx200_pci); dev++) {
- pdev = pci_get_device(scx200_pci[dev].vendor,
- scx200_pci[dev].device, NULL);
+ if (!pci_dev_present(scx200_isa))
+ return;
- if (pdev == NULL)
+ for (i = 0; i < MAX_DEVICES; ++i) {
+ if (base[i] == 0)
continue;
- data = scx200_pci[dev].driver_data;
-
- /* if .bar is greater or equal to zero, this is a
- * PCI device - otherwise, we assume
- that the ports are ISA based
- */
-
- if (scx200_data[data].bar >= 0)
- rc = scx200_create_pci(scx200_data[data].name, pdev,
- scx200_data[data].bar);
- else {
- int i;
-
- pci_dev_put(pdev);
- for (i = 0; i < MAX_DEVICES; ++i) {
- if (base[i] == 0)
- continue;
-
- rc = scx200_create_isa(scx200_data[data].name,
- base[i],
- i);
- }
- }
-
- break;
+ /* XXX: should we care about failures? */
+ scx200_create_dev("SCx200", base[i], i, NULL);
}
-
- return rc;
}
static int __init scx200_acb_init(void)
{
- int rc;
-
pr_debug(NAME ": NatSemi SCx200 ACCESS.bus Driver\n");
- rc = scx200_scan_pci();
+ /* First scan for ISA-based devices */
+ scx200_scan_isa(); /* XXX: should we care about errors? */
/* If at least one bus was created, init must succeed */
if (scx200_acb_list)
return 0;
- return rc;
+
+ /* No ISA devices; register the platform driver for PCI-based devices */
+ return platform_driver_register(&scx200_pci_drv);
}
static void __exit scx200_acb_cleanup(void)
{
struct scx200_acb_iface *iface;
+ platform_driver_unregister(&scx200_pci_drv);
+
mutex_lock(&scx200_acb_list_mutex);
while ((iface = scx200_acb_list) != NULL) {
scx200_acb_list = iface->next;
mutex_unlock(&scx200_acb_list_mutex);
- i2c_del_adapter(&iface->adapter);
+ scx200_cleanup_iface(iface);
- if (iface->pdev) {
- pci_release_region(iface->pdev, iface->bar);
- pci_dev_put(iface->pdev);
- }
- else
- release_region(iface->base, 8);
-
- kfree(iface);
mutex_lock(&scx200_acb_list_mutex);
}
mutex_unlock(&scx200_acb_list_mutex);
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index c7db698..f0bd5bc 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -196,88 +196,60 @@
{
const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
- if (pm) {
- if (pm_runtime_suspended(dev))
- return 0;
- else
- return pm->suspend ? pm->suspend(dev) : 0;
- }
-
- return i2c_legacy_suspend(dev, PMSG_SUSPEND);
+ if (pm)
+ return pm_generic_suspend(dev);
+ else
+ return i2c_legacy_suspend(dev, PMSG_SUSPEND);
}
static int i2c_device_pm_resume(struct device *dev)
{
const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
- int ret;
if (pm)
- ret = pm->resume ? pm->resume(dev) : 0;
+ return pm_generic_resume(dev);
else
- ret = i2c_legacy_resume(dev);
-
- return ret;
+ return i2c_legacy_resume(dev);
}
static int i2c_device_pm_freeze(struct device *dev)
{
const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
- if (pm) {
- if (pm_runtime_suspended(dev))
- return 0;
- else
- return pm->freeze ? pm->freeze(dev) : 0;
- }
-
- return i2c_legacy_suspend(dev, PMSG_FREEZE);
+ if (pm)
+ return pm_generic_freeze(dev);
+ else
+ return i2c_legacy_suspend(dev, PMSG_FREEZE);
}
static int i2c_device_pm_thaw(struct device *dev)
{
const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
- if (pm) {
- if (pm_runtime_suspended(dev))
- return 0;
- else
- return pm->thaw ? pm->thaw(dev) : 0;
- }
-
- return i2c_legacy_resume(dev);
+ if (pm)
+ return pm_generic_thaw(dev);
+ else
+ return i2c_legacy_resume(dev);
}
static int i2c_device_pm_poweroff(struct device *dev)
{
const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
- if (pm) {
- if (pm_runtime_suspended(dev))
- return 0;
- else
- return pm->poweroff ? pm->poweroff(dev) : 0;
- }
-
- return i2c_legacy_suspend(dev, PMSG_HIBERNATE);
+ if (pm)
+ return pm_generic_poweroff(dev);
+ else
+ return i2c_legacy_suspend(dev, PMSG_HIBERNATE);
}
static int i2c_device_pm_restore(struct device *dev)
{
const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
- int ret;
if (pm)
- ret = pm->restore ? pm->restore(dev) : 0;
+ return pm_generic_restore(dev);
else
- ret = i2c_legacy_resume(dev);
-
- if (!ret) {
- pm_runtime_disable(dev);
- pm_runtime_set_active(dev);
- pm_runtime_enable(dev);
- }
-
- return ret;
+ return i2c_legacy_resume(dev);
}
#else /* !CONFIG_PM_SLEEP */
#define i2c_device_pm_suspend NULL
@@ -1021,6 +993,14 @@
static int __unregister_client(struct device *dev, void *dummy)
{
struct i2c_client *client = i2c_verify_client(dev);
+ if (client && strcmp(client->name, "dummy"))
+ i2c_unregister_device(client);
+ return 0;
+}
+
+static int __unregister_dummy(struct device *dev, void *dummy)
+{
+ struct i2c_client *client = i2c_verify_client(dev);
if (client)
i2c_unregister_device(client);
return 0;
@@ -1075,8 +1055,12 @@
mutex_unlock(&adap->userspace_clients_lock);
/* Detach any active clients. This can't fail, thus we do not
- checking the returned value. */
+ * check the returned value. This is a two-pass process, because
+ * we can't remove the dummy devices during the first pass: they
+ * could have been instantiated by real devices wishing to clean
+ * them up properly, so we give them a chance to do that first. */
res = device_for_each_child(&adap->dev, NULL, __unregister_client);
+ res = device_for_each_child(&adap->dev, NULL, __unregister_dummy);
#ifdef CONFIG_I2C_COMPAT
class_compat_remove_link(i2c_adapter_compat_class, &adap->dev,
@@ -1140,6 +1124,14 @@
if (res)
return res;
+ /* Drivers should switch to dev_pm_ops instead. */
+ if (driver->suspend)
+ pr_warn("i2c-core: driver [%s] using legacy suspend method\n",
+ driver->driver.name);
+ if (driver->resume)
+ pr_warn("i2c-core: driver [%s] using legacy resume method\n",
+ driver->driver.name);
+
pr_debug("i2c-core: driver [%s] registered\n", driver->driver.name);
INIT_LIST_HEAD(&driver->clients);
diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 56ac09d..7acb32e 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -59,6 +59,8 @@
#include <linux/hrtimer.h> /* ktime_get_real() */
#include <trace/events/power.h>
#include <linux/sched.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
#include <asm/mwait.h>
#define INTEL_IDLE_VERSION "0.4"
@@ -73,6 +75,7 @@
static unsigned int mwait_substates;
+#define LAPIC_TIMER_ALWAYS_RELIABLE 0xFFFFFFFF
/* Reliable LAPIC Timer States, bit 1 for C1 etc. */
static unsigned int lapic_timer_reliable_states = (1 << 1); /* Default to only C1 */
@@ -82,6 +85,14 @@
static struct cpuidle_state *cpuidle_state_table;
/*
+ * Set this flag for states where the HW flushes the TLB for us
+ * and so we don't need cross-calls to keep it consistent.
+ * If this flag is set, SW flushes the TLB, so even if the
+ * HW doesn't do the flushing, this flag is safe to use.
+ */
+#define CPUIDLE_FLAG_TLB_FLUSHED 0x10000
+
+/*
* States are indexed by the cstate number,
* which is also the index into the MWAIT hint array.
* Thus C0 is a dummy.
@@ -122,7 +133,7 @@
.driver_data = (void *) 0x00,
.flags = CPUIDLE_FLAG_TIME_VALID,
.exit_latency = 1,
- .target_residency = 4,
+ .target_residency = 1,
.enter = &intel_idle },
{ /* MWAIT C2 */
.name = "SNB-C3",
@@ -130,7 +141,7 @@
.driver_data = (void *) 0x10,
.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 80,
- .target_residency = 160,
+ .target_residency = 211,
.enter = &intel_idle },
{ /* MWAIT C3 */
.name = "SNB-C6",
@@ -138,7 +149,7 @@
.driver_data = (void *) 0x20,
.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 104,
- .target_residency = 208,
+ .target_residency = 345,
.enter = &intel_idle },
{ /* MWAIT C4 */
.name = "SNB-C7",
@@ -146,7 +157,7 @@
.driver_data = (void *) 0x30,
.flags = CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
.exit_latency = 109,
- .target_residency = 300,
+ .target_residency = 345,
.enter = &intel_idle },
};
@@ -220,8 +231,6 @@
kt_before = ktime_get_real();
stop_critical_timings();
- trace_power_start(POWER_CSTATE, (eax >> 4) + 1, cpu);
- trace_cpu_idle((eax >> 4) + 1, cpu);
if (!need_resched()) {
__monitor((void *)¤t_thread_info()->flags, 0, 0);
@@ -243,6 +252,39 @@
return usec_delta;
}
+static void __setup_broadcast_timer(void *arg)
+{
+ unsigned long reason = (unsigned long)arg;
+ int cpu = smp_processor_id();
+
+ reason = reason ?
+ CLOCK_EVT_NOTIFY_BROADCAST_ON : CLOCK_EVT_NOTIFY_BROADCAST_OFF;
+
+ clockevents_notify(reason, &cpu);
+}
+
+static int __cpuinit setup_broadcast_cpuhp_notify(struct notifier_block *n,
+ unsigned long action, void *hcpu)
+{
+ int hotcpu = (unsigned long)hcpu;
+
+ switch (action & 0xf) {
+ case CPU_ONLINE:
+ smp_call_function_single(hotcpu, __setup_broadcast_timer,
+ (void *)true, 1);
+ break;
+ case CPU_DOWN_PREPARE:
+ smp_call_function_single(hotcpu, __setup_broadcast_timer,
+ (void *)false, 1);
+ break;
+ }
+ return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata setup_broadcast_notifier = {
+ .notifier_call = setup_broadcast_cpuhp_notify,
+};
+
/*
* intel_idle_probe()
*/
@@ -305,7 +347,11 @@
}
if (boot_cpu_has(X86_FEATURE_ARAT)) /* Always Reliable APIC Timer */
- lapic_timer_reliable_states = 0xFFFFFFFF;
+ lapic_timer_reliable_states = LAPIC_TIMER_ALWAYS_RELIABLE;
+ else {
+ smp_call_function(__setup_broadcast_timer, (void *)true, 1);
+ register_cpu_notifier(&setup_broadcast_notifier);
+ }
pr_debug(PREFIX "v" INTEL_IDLE_VERSION
" model 0x%X\n", boot_cpu_data.x86_model);
@@ -403,6 +449,10 @@
{
int retval;
+ /* Do not load intel_idle at all for now if idle= is passed */
+ if (boot_option_idle_override != IDLE_NO_OVERRIDE)
+ return -ENODEV;
+
retval = intel_idle_probe();
if (retval)
return retval;
@@ -428,6 +478,11 @@
intel_idle_cpuidle_devices_uninit();
cpuidle_unregister_driver(&intel_idle_driver);
+ if (lapic_timer_reliable_states != LAPIC_TIMER_ALWAYS_RELIABLE) {
+ smp_call_function(__setup_broadcast_timer, (void *)false, 1);
+ unregister_cpu_notifier(&setup_broadcast_notifier);
+ }
+
return;
}
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index bf1a95e3..98d9ec8 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -240,6 +240,30 @@
Allow volume managers to mirror logical volumes, also
needed for live data migration tools such as 'pvmove'.
+config DM_RAID
+ tristate "RAID 4/5/6 target (EXPERIMENTAL)"
+ depends on BLK_DEV_DM && EXPERIMENTAL
+ select MD_RAID456
+ select BLK_DEV_MD
+ ---help---
+ A dm target that supports RAID4, RAID5 and RAID6 mappings
+
+ A RAID-5 set of N drives with a capacity of C MB per drive provides
+ the capacity of C * (N - 1) MB, and protects against a failure
+ of a single drive. For a given sector (row) number, (N - 1) drives
+ contain data sectors, and one drive contains the parity protection.
+ For a RAID-4 set, the parity blocks are present on a single drive,
+ while a RAID-5 set distributes the parity across the drives in one
+ of the available parity distribution methods.
+
+ A RAID-6 set of N drives with a capacity of C MB per drive
+ provides the capacity of C * (N - 2) MB, and protects
+ against a failure of any two drives. For a given sector
+ (row) number, (N - 2) drives contain data sectors, and two
+ drives contains two independent redundancy syndromes. Like
+ RAID-5, RAID-6 distributes the syndromes across the drives
+ in one of the available parity distribution methods.
+
config DM_LOG_USERSPACE
tristate "Mirror userspace logging (EXPERIMENTAL)"
depends on DM_MIRROR && EXPERIMENTAL && NET
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index 5e3aac4..d013860 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -36,6 +36,7 @@
obj-$(CONFIG_DM_MIRROR) += dm-mirror.o dm-log.o dm-region-hash.o
obj-$(CONFIG_DM_LOG_USERSPACE) += dm-log-userspace.o
obj-$(CONFIG_DM_ZERO) += dm-zero.o
+obj-$(CONFIG_DM_RAID) += dm-raid.o
ifeq ($(CONFIG_DM_UEVENT),y)
dm-mod-objs += dm-uevent.o
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 5a1ffe3..9a35320 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -210,11 +210,11 @@
|| test_bit(Faulty, &rdev->flags))
continue;
- target = rdev->sb_start + offset + index * (PAGE_SIZE/512);
+ target = offset + index * (PAGE_SIZE/512);
if (sync_page_io(rdev, target,
roundup(size, bdev_logical_block_size(rdev->bdev)),
- page, READ)) {
+ page, READ, true)) {
page->index = index;
attach_page_buffers(page, NULL); /* so that free_buffer will
* quietly no-op */
@@ -264,14 +264,18 @@
static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
{
mdk_rdev_t *rdev = NULL;
+ struct block_device *bdev;
mddev_t *mddev = bitmap->mddev;
while ((rdev = next_active_rdev(rdev, mddev)) != NULL) {
int size = PAGE_SIZE;
loff_t offset = mddev->bitmap_info.offset;
+
+ bdev = (rdev->meta_bdev) ? rdev->meta_bdev : rdev->bdev;
+
if (page->index == bitmap->file_pages-1)
size = roundup(bitmap->last_page_size,
- bdev_logical_block_size(rdev->bdev));
+ bdev_logical_block_size(bdev));
/* Just make sure we aren't corrupting data or
* metadata
*/
@@ -1542,7 +1546,7 @@
wait_event(bitmap->mddev->recovery_wait,
atomic_read(&bitmap->mddev->recovery_active) == 0);
- bitmap->mddev->curr_resync_completed = bitmap->mddev->curr_resync;
+ bitmap->mddev->curr_resync_completed = sector;
set_bit(MD_CHANGE_CLEAN, &bitmap->mddev->flags);
sector &= ~((1ULL << CHUNK_BLOCK_SHIFT(bitmap)) - 1);
s = 0;
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index d5b0e4c..4e054bd 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -18,10 +18,14 @@
#include <linux/crypto.h>
#include <linux/workqueue.h>
#include <linux/backing-dev.h>
+#include <linux/percpu.h>
#include <asm/atomic.h>
#include <linux/scatterlist.h>
#include <asm/page.h>
#include <asm/unaligned.h>
+#include <crypto/hash.h>
+#include <crypto/md5.h>
+#include <crypto/algapi.h>
#include <linux/device-mapper.h>
@@ -63,6 +67,7 @@
struct convert_context *ctx;
struct scatterlist sg_in;
struct scatterlist sg_out;
+ sector_t iv_sector;
};
struct crypt_config;
@@ -73,11 +78,13 @@
void (*dtr)(struct crypt_config *cc);
int (*init)(struct crypt_config *cc);
int (*wipe)(struct crypt_config *cc);
- int (*generator)(struct crypt_config *cc, u8 *iv, sector_t sector);
+ int (*generator)(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq);
+ int (*post)(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq);
};
struct iv_essiv_private {
- struct crypto_cipher *tfm;
struct crypto_hash *hash_tfm;
u8 *salt;
};
@@ -86,11 +93,32 @@
int shift;
};
+#define LMK_SEED_SIZE 64 /* hash + 0 */
+struct iv_lmk_private {
+ struct crypto_shash *hash_tfm;
+ u8 *seed;
+};
+
/*
* Crypt: maps a linear range of a block device
* and encrypts / decrypts at the same time.
*/
enum flags { DM_CRYPT_SUSPENDED, DM_CRYPT_KEY_VALID };
+
+/*
+ * Duplicated per-CPU state for cipher.
+ */
+struct crypt_cpu {
+ struct ablkcipher_request *req;
+ /* ESSIV: struct crypto_cipher *essiv_tfm */
+ void *iv_private;
+ struct crypto_ablkcipher *tfms[0];
+};
+
+/*
+ * The fields in here must be read only after initialization,
+ * changing state should be in crypt_cpu.
+ */
struct crypt_config {
struct dm_dev *dev;
sector_t start;
@@ -108,17 +136,25 @@
struct workqueue_struct *crypt_queue;
char *cipher;
- char *cipher_mode;
+ char *cipher_string;
struct crypt_iv_operations *iv_gen_ops;
union {
struct iv_essiv_private essiv;
struct iv_benbi_private benbi;
+ struct iv_lmk_private lmk;
} iv_gen_private;
sector_t iv_offset;
unsigned int iv_size;
/*
+ * Duplicated per cpu state. Access through
+ * per_cpu_ptr() only.
+ */
+ struct crypt_cpu __percpu *cpu;
+ unsigned tfms_count;
+
+ /*
* Layout of each crypto request:
*
* struct ablkcipher_request
@@ -132,11 +168,10 @@
* correctly aligned.
*/
unsigned int dmreq_start;
- struct ablkcipher_request *req;
- struct crypto_ablkcipher *tfm;
unsigned long flags;
unsigned int key_size;
+ unsigned int key_parts;
u8 key[0];
};
@@ -148,6 +183,20 @@
static void clone_init(struct dm_crypt_io *, struct bio *);
static void kcryptd_queue_crypt(struct dm_crypt_io *io);
+static u8 *iv_of_dmreq(struct crypt_config *cc, struct dm_crypt_request *dmreq);
+
+static struct crypt_cpu *this_crypt_config(struct crypt_config *cc)
+{
+ return this_cpu_ptr(cc->cpu);
+}
+
+/*
+ * Use this to access cipher attributes that are the same for each CPU.
+ */
+static struct crypto_ablkcipher *any_tfm(struct crypt_config *cc)
+{
+ return __this_cpu_ptr(cc->cpu)->tfms[0];
+}
/*
* Different IV generation algorithms:
@@ -168,23 +217,38 @@
* null: the initial vector is always zero. Provides compatibility with
* obsolete loop_fish2 devices. Do not use for new devices.
*
+ * lmk: Compatible implementation of the block chaining mode used
+ * by the Loop-AES block device encryption system
+ * designed by Jari Ruusu. See http://loop-aes.sourceforge.net/
+ * It operates on full 512 byte sectors and uses CBC
+ * with an IV derived from the sector number, the data and
+ * optionally extra IV seed.
+ * This means that after decryption the first block
+ * of sector must be tweaked according to decrypted data.
+ * Loop-AES can use three encryption schemes:
+ * version 1: is plain aes-cbc mode
+ * version 2: uses 64 multikey scheme with lmk IV generator
+ * version 3: the same as version 2 with additional IV seed
+ * (it uses 65 keys, last key is used as IV seed)
+ *
* plumb: unimplemented, see:
* http://article.gmane.org/gmane.linux.kernel.device-mapper.dm-crypt/454
*/
-static int crypt_iv_plain_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
+static int crypt_iv_plain_gen(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq)
{
memset(iv, 0, cc->iv_size);
- *(u32 *)iv = cpu_to_le32(sector & 0xffffffff);
+ *(u32 *)iv = cpu_to_le32(dmreq->iv_sector & 0xffffffff);
return 0;
}
static int crypt_iv_plain64_gen(struct crypt_config *cc, u8 *iv,
- sector_t sector)
+ struct dm_crypt_request *dmreq)
{
memset(iv, 0, cc->iv_size);
- *(u64 *)iv = cpu_to_le64(sector);
+ *(u64 *)iv = cpu_to_le64(dmreq->iv_sector);
return 0;
}
@@ -195,7 +259,8 @@
struct iv_essiv_private *essiv = &cc->iv_gen_private.essiv;
struct hash_desc desc;
struct scatterlist sg;
- int err;
+ struct crypto_cipher *essiv_tfm;
+ int err, cpu;
sg_init_one(&sg, cc->key, cc->key_size);
desc.tfm = essiv->hash_tfm;
@@ -205,8 +270,16 @@
if (err)
return err;
- return crypto_cipher_setkey(essiv->tfm, essiv->salt,
+ for_each_possible_cpu(cpu) {
+ essiv_tfm = per_cpu_ptr(cc->cpu, cpu)->iv_private,
+
+ err = crypto_cipher_setkey(essiv_tfm, essiv->salt,
crypto_hash_digestsize(essiv->hash_tfm));
+ if (err)
+ return err;
+ }
+
+ return 0;
}
/* Wipe salt and reset key derived from volume key */
@@ -214,24 +287,76 @@
{
struct iv_essiv_private *essiv = &cc->iv_gen_private.essiv;
unsigned salt_size = crypto_hash_digestsize(essiv->hash_tfm);
+ struct crypto_cipher *essiv_tfm;
+ int cpu, r, err = 0;
memset(essiv->salt, 0, salt_size);
- return crypto_cipher_setkey(essiv->tfm, essiv->salt, salt_size);
+ for_each_possible_cpu(cpu) {
+ essiv_tfm = per_cpu_ptr(cc->cpu, cpu)->iv_private;
+ r = crypto_cipher_setkey(essiv_tfm, essiv->salt, salt_size);
+ if (r)
+ err = r;
+ }
+
+ return err;
+}
+
+/* Set up per cpu cipher state */
+static struct crypto_cipher *setup_essiv_cpu(struct crypt_config *cc,
+ struct dm_target *ti,
+ u8 *salt, unsigned saltsize)
+{
+ struct crypto_cipher *essiv_tfm;
+ int err;
+
+ /* Setup the essiv_tfm with the given salt */
+ essiv_tfm = crypto_alloc_cipher(cc->cipher, 0, CRYPTO_ALG_ASYNC);
+ if (IS_ERR(essiv_tfm)) {
+ ti->error = "Error allocating crypto tfm for ESSIV";
+ return essiv_tfm;
+ }
+
+ if (crypto_cipher_blocksize(essiv_tfm) !=
+ crypto_ablkcipher_ivsize(any_tfm(cc))) {
+ ti->error = "Block size of ESSIV cipher does "
+ "not match IV size of block cipher";
+ crypto_free_cipher(essiv_tfm);
+ return ERR_PTR(-EINVAL);
+ }
+
+ err = crypto_cipher_setkey(essiv_tfm, salt, saltsize);
+ if (err) {
+ ti->error = "Failed to set key for ESSIV cipher";
+ crypto_free_cipher(essiv_tfm);
+ return ERR_PTR(err);
+ }
+
+ return essiv_tfm;
}
static void crypt_iv_essiv_dtr(struct crypt_config *cc)
{
+ int cpu;
+ struct crypt_cpu *cpu_cc;
+ struct crypto_cipher *essiv_tfm;
struct iv_essiv_private *essiv = &cc->iv_gen_private.essiv;
- crypto_free_cipher(essiv->tfm);
- essiv->tfm = NULL;
-
crypto_free_hash(essiv->hash_tfm);
essiv->hash_tfm = NULL;
kzfree(essiv->salt);
essiv->salt = NULL;
+
+ for_each_possible_cpu(cpu) {
+ cpu_cc = per_cpu_ptr(cc->cpu, cpu);
+ essiv_tfm = cpu_cc->iv_private;
+
+ if (essiv_tfm)
+ crypto_free_cipher(essiv_tfm);
+
+ cpu_cc->iv_private = NULL;
+ }
}
static int crypt_iv_essiv_ctr(struct crypt_config *cc, struct dm_target *ti,
@@ -240,7 +365,7 @@
struct crypto_cipher *essiv_tfm = NULL;
struct crypto_hash *hash_tfm = NULL;
u8 *salt = NULL;
- int err;
+ int err, cpu;
if (!opts) {
ti->error = "Digest algorithm missing for ESSIV mode";
@@ -262,48 +387,44 @@
goto bad;
}
- /* Allocate essiv_tfm */
- essiv_tfm = crypto_alloc_cipher(cc->cipher, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(essiv_tfm)) {
- ti->error = "Error allocating crypto tfm for ESSIV";
- err = PTR_ERR(essiv_tfm);
- goto bad;
- }
- if (crypto_cipher_blocksize(essiv_tfm) !=
- crypto_ablkcipher_ivsize(cc->tfm)) {
- ti->error = "Block size of ESSIV cipher does "
- "not match IV size of block cipher";
- err = -EINVAL;
- goto bad;
- }
-
cc->iv_gen_private.essiv.salt = salt;
- cc->iv_gen_private.essiv.tfm = essiv_tfm;
cc->iv_gen_private.essiv.hash_tfm = hash_tfm;
+ for_each_possible_cpu(cpu) {
+ essiv_tfm = setup_essiv_cpu(cc, ti, salt,
+ crypto_hash_digestsize(hash_tfm));
+ if (IS_ERR(essiv_tfm)) {
+ crypt_iv_essiv_dtr(cc);
+ return PTR_ERR(essiv_tfm);
+ }
+ per_cpu_ptr(cc->cpu, cpu)->iv_private = essiv_tfm;
+ }
+
return 0;
bad:
- if (essiv_tfm && !IS_ERR(essiv_tfm))
- crypto_free_cipher(essiv_tfm);
if (hash_tfm && !IS_ERR(hash_tfm))
crypto_free_hash(hash_tfm);
kfree(salt);
return err;
}
-static int crypt_iv_essiv_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
+static int crypt_iv_essiv_gen(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq)
{
+ struct crypto_cipher *essiv_tfm = this_crypt_config(cc)->iv_private;
+
memset(iv, 0, cc->iv_size);
- *(u64 *)iv = cpu_to_le64(sector);
- crypto_cipher_encrypt_one(cc->iv_gen_private.essiv.tfm, iv, iv);
+ *(u64 *)iv = cpu_to_le64(dmreq->iv_sector);
+ crypto_cipher_encrypt_one(essiv_tfm, iv, iv);
+
return 0;
}
static int crypt_iv_benbi_ctr(struct crypt_config *cc, struct dm_target *ti,
const char *opts)
{
- unsigned bs = crypto_ablkcipher_blocksize(cc->tfm);
+ unsigned bs = crypto_ablkcipher_blocksize(any_tfm(cc));
int log = ilog2(bs);
/* we need to calculate how far we must shift the sector count
@@ -328,25 +449,177 @@
{
}
-static int crypt_iv_benbi_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
+static int crypt_iv_benbi_gen(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq)
{
__be64 val;
memset(iv, 0, cc->iv_size - sizeof(u64)); /* rest is cleared below */
- val = cpu_to_be64(((u64)sector << cc->iv_gen_private.benbi.shift) + 1);
+ val = cpu_to_be64(((u64)dmreq->iv_sector << cc->iv_gen_private.benbi.shift) + 1);
put_unaligned(val, (__be64 *)(iv + cc->iv_size - sizeof(u64)));
return 0;
}
-static int crypt_iv_null_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
+static int crypt_iv_null_gen(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq)
{
memset(iv, 0, cc->iv_size);
return 0;
}
+static void crypt_iv_lmk_dtr(struct crypt_config *cc)
+{
+ struct iv_lmk_private *lmk = &cc->iv_gen_private.lmk;
+
+ if (lmk->hash_tfm && !IS_ERR(lmk->hash_tfm))
+ crypto_free_shash(lmk->hash_tfm);
+ lmk->hash_tfm = NULL;
+
+ kzfree(lmk->seed);
+ lmk->seed = NULL;
+}
+
+static int crypt_iv_lmk_ctr(struct crypt_config *cc, struct dm_target *ti,
+ const char *opts)
+{
+ struct iv_lmk_private *lmk = &cc->iv_gen_private.lmk;
+
+ lmk->hash_tfm = crypto_alloc_shash("md5", 0, 0);
+ if (IS_ERR(lmk->hash_tfm)) {
+ ti->error = "Error initializing LMK hash";
+ return PTR_ERR(lmk->hash_tfm);
+ }
+
+ /* No seed in LMK version 2 */
+ if (cc->key_parts == cc->tfms_count) {
+ lmk->seed = NULL;
+ return 0;
+ }
+
+ lmk->seed = kzalloc(LMK_SEED_SIZE, GFP_KERNEL);
+ if (!lmk->seed) {
+ crypt_iv_lmk_dtr(cc);
+ ti->error = "Error kmallocing seed storage in LMK";
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
+static int crypt_iv_lmk_init(struct crypt_config *cc)
+{
+ struct iv_lmk_private *lmk = &cc->iv_gen_private.lmk;
+ int subkey_size = cc->key_size / cc->key_parts;
+
+ /* LMK seed is on the position of LMK_KEYS + 1 key */
+ if (lmk->seed)
+ memcpy(lmk->seed, cc->key + (cc->tfms_count * subkey_size),
+ crypto_shash_digestsize(lmk->hash_tfm));
+
+ return 0;
+}
+
+static int crypt_iv_lmk_wipe(struct crypt_config *cc)
+{
+ struct iv_lmk_private *lmk = &cc->iv_gen_private.lmk;
+
+ if (lmk->seed)
+ memset(lmk->seed, 0, LMK_SEED_SIZE);
+
+ return 0;
+}
+
+static int crypt_iv_lmk_one(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq,
+ u8 *data)
+{
+ struct iv_lmk_private *lmk = &cc->iv_gen_private.lmk;
+ struct {
+ struct shash_desc desc;
+ char ctx[crypto_shash_descsize(lmk->hash_tfm)];
+ } sdesc;
+ struct md5_state md5state;
+ u32 buf[4];
+ int i, r;
+
+ sdesc.desc.tfm = lmk->hash_tfm;
+ sdesc.desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP;
+
+ r = crypto_shash_init(&sdesc.desc);
+ if (r)
+ return r;
+
+ if (lmk->seed) {
+ r = crypto_shash_update(&sdesc.desc, lmk->seed, LMK_SEED_SIZE);
+ if (r)
+ return r;
+ }
+
+ /* Sector is always 512B, block size 16, add data of blocks 1-31 */
+ r = crypto_shash_update(&sdesc.desc, data + 16, 16 * 31);
+ if (r)
+ return r;
+
+ /* Sector is cropped to 56 bits here */
+ buf[0] = cpu_to_le32(dmreq->iv_sector & 0xFFFFFFFF);
+ buf[1] = cpu_to_le32((((u64)dmreq->iv_sector >> 32) & 0x00FFFFFF) | 0x80000000);
+ buf[2] = cpu_to_le32(4024);
+ buf[3] = 0;
+ r = crypto_shash_update(&sdesc.desc, (u8 *)buf, sizeof(buf));
+ if (r)
+ return r;
+
+ /* No MD5 padding here */
+ r = crypto_shash_export(&sdesc.desc, &md5state);
+ if (r)
+ return r;
+
+ for (i = 0; i < MD5_HASH_WORDS; i++)
+ __cpu_to_le32s(&md5state.hash[i]);
+ memcpy(iv, &md5state.hash, cc->iv_size);
+
+ return 0;
+}
+
+static int crypt_iv_lmk_gen(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq)
+{
+ u8 *src;
+ int r = 0;
+
+ if (bio_data_dir(dmreq->ctx->bio_in) == WRITE) {
+ src = kmap_atomic(sg_page(&dmreq->sg_in), KM_USER0);
+ r = crypt_iv_lmk_one(cc, iv, dmreq, src + dmreq->sg_in.offset);
+ kunmap_atomic(src, KM_USER0);
+ } else
+ memset(iv, 0, cc->iv_size);
+
+ return r;
+}
+
+static int crypt_iv_lmk_post(struct crypt_config *cc, u8 *iv,
+ struct dm_crypt_request *dmreq)
+{
+ u8 *dst;
+ int r;
+
+ if (bio_data_dir(dmreq->ctx->bio_in) == WRITE)
+ return 0;
+
+ dst = kmap_atomic(sg_page(&dmreq->sg_out), KM_USER0);
+ r = crypt_iv_lmk_one(cc, iv, dmreq, dst + dmreq->sg_out.offset);
+
+ /* Tweak the first block of plaintext sector */
+ if (!r)
+ crypto_xor(dst + dmreq->sg_out.offset, iv, cc->iv_size);
+
+ kunmap_atomic(dst, KM_USER0);
+ return r;
+}
+
static struct crypt_iv_operations crypt_iv_plain_ops = {
.generator = crypt_iv_plain_gen
};
@@ -373,6 +646,15 @@
.generator = crypt_iv_null_gen
};
+static struct crypt_iv_operations crypt_iv_lmk_ops = {
+ .ctr = crypt_iv_lmk_ctr,
+ .dtr = crypt_iv_lmk_dtr,
+ .init = crypt_iv_lmk_init,
+ .wipe = crypt_iv_lmk_wipe,
+ .generator = crypt_iv_lmk_gen,
+ .post = crypt_iv_lmk_post
+};
+
static void crypt_convert_init(struct crypt_config *cc,
struct convert_context *ctx,
struct bio *bio_out, struct bio *bio_in,
@@ -400,6 +682,13 @@
return (struct ablkcipher_request *)((char *)dmreq - cc->dmreq_start);
}
+static u8 *iv_of_dmreq(struct crypt_config *cc,
+ struct dm_crypt_request *dmreq)
+{
+ return (u8 *)ALIGN((unsigned long)(dmreq + 1),
+ crypto_ablkcipher_alignmask(any_tfm(cc)) + 1);
+}
+
static int crypt_convert_block(struct crypt_config *cc,
struct convert_context *ctx,
struct ablkcipher_request *req)
@@ -411,9 +700,9 @@
int r = 0;
dmreq = dmreq_of_req(cc, req);
- iv = (u8 *)ALIGN((unsigned long)(dmreq + 1),
- crypto_ablkcipher_alignmask(cc->tfm) + 1);
+ iv = iv_of_dmreq(cc, dmreq);
+ dmreq->iv_sector = ctx->sector;
dmreq->ctx = ctx;
sg_init_table(&dmreq->sg_in, 1);
sg_set_page(&dmreq->sg_in, bv_in->bv_page, 1 << SECTOR_SHIFT,
@@ -436,7 +725,7 @@
}
if (cc->iv_gen_ops) {
- r = cc->iv_gen_ops->generator(cc, iv, ctx->sector);
+ r = cc->iv_gen_ops->generator(cc, iv, dmreq);
if (r < 0)
return r;
}
@@ -449,21 +738,28 @@
else
r = crypto_ablkcipher_decrypt(req);
+ if (!r && cc->iv_gen_ops && cc->iv_gen_ops->post)
+ r = cc->iv_gen_ops->post(cc, iv, dmreq);
+
return r;
}
static void kcryptd_async_done(struct crypto_async_request *async_req,
int error);
+
static void crypt_alloc_req(struct crypt_config *cc,
struct convert_context *ctx)
{
- if (!cc->req)
- cc->req = mempool_alloc(cc->req_pool, GFP_NOIO);
- ablkcipher_request_set_tfm(cc->req, cc->tfm);
- ablkcipher_request_set_callback(cc->req, CRYPTO_TFM_REQ_MAY_BACKLOG |
- CRYPTO_TFM_REQ_MAY_SLEEP,
- kcryptd_async_done,
- dmreq_of_req(cc, cc->req));
+ struct crypt_cpu *this_cc = this_crypt_config(cc);
+ unsigned key_index = ctx->sector & (cc->tfms_count - 1);
+
+ if (!this_cc->req)
+ this_cc->req = mempool_alloc(cc->req_pool, GFP_NOIO);
+
+ ablkcipher_request_set_tfm(this_cc->req, this_cc->tfms[key_index]);
+ ablkcipher_request_set_callback(this_cc->req,
+ CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
+ kcryptd_async_done, dmreq_of_req(cc, this_cc->req));
}
/*
@@ -472,6 +768,7 @@
static int crypt_convert(struct crypt_config *cc,
struct convert_context *ctx)
{
+ struct crypt_cpu *this_cc = this_crypt_config(cc);
int r;
atomic_set(&ctx->pending, 1);
@@ -483,7 +780,7 @@
atomic_inc(&ctx->pending);
- r = crypt_convert_block(cc, ctx, cc->req);
+ r = crypt_convert_block(cc, ctx, this_cc->req);
switch (r) {
/* async */
@@ -492,7 +789,7 @@
INIT_COMPLETION(ctx->restart);
/* fall through*/
case -EINPROGRESS:
- cc->req = NULL;
+ this_cc->req = NULL;
ctx->sector++;
continue;
@@ -651,6 +948,9 @@
* They must be separated as otherwise the final stages could be
* starved by new requests which can block in the first stages due
* to memory allocation.
+ *
+ * The work is done per CPU global for all dm-crypt instances.
+ * They should not depend on each other and do not block.
*/
static void crypt_endio(struct bio *clone, int error)
{
@@ -691,26 +991,30 @@
clone->bi_destructor = dm_crypt_bio_destructor;
}
-static void kcryptd_io_read(struct dm_crypt_io *io)
+static void kcryptd_unplug(struct crypt_config *cc)
+{
+ blk_unplug(bdev_get_queue(cc->dev->bdev));
+}
+
+static int kcryptd_io_read(struct dm_crypt_io *io, gfp_t gfp)
{
struct crypt_config *cc = io->target->private;
struct bio *base_bio = io->base_bio;
struct bio *clone;
- crypt_inc_pending(io);
-
/*
* The block layer might modify the bvec array, so always
* copy the required bvecs because we need the original
* one in order to decrypt the whole bio data *afterwards*.
*/
- clone = bio_alloc_bioset(GFP_NOIO, bio_segments(base_bio), cc->bs);
- if (unlikely(!clone)) {
- io->error = -ENOMEM;
- crypt_dec_pending(io);
- return;
+ clone = bio_alloc_bioset(gfp, bio_segments(base_bio), cc->bs);
+ if (!clone) {
+ kcryptd_unplug(cc);
+ return 1;
}
+ crypt_inc_pending(io);
+
clone_init(io, clone);
clone->bi_idx = 0;
clone->bi_vcnt = bio_segments(base_bio);
@@ -720,6 +1024,7 @@
sizeof(struct bio_vec) * clone->bi_vcnt);
generic_make_request(clone);
+ return 0;
}
static void kcryptd_io_write(struct dm_crypt_io *io)
@@ -732,9 +1037,12 @@
{
struct dm_crypt_io *io = container_of(work, struct dm_crypt_io, work);
- if (bio_data_dir(io->base_bio) == READ)
- kcryptd_io_read(io);
- else
+ if (bio_data_dir(io->base_bio) == READ) {
+ crypt_inc_pending(io);
+ if (kcryptd_io_read(io, GFP_NOIO))
+ io->error = -ENOMEM;
+ crypt_dec_pending(io);
+ } else
kcryptd_io_write(io);
}
@@ -901,6 +1209,9 @@
return;
}
+ if (!error && cc->iv_gen_ops && cc->iv_gen_ops->post)
+ error = cc->iv_gen_ops->post(cc, iv_of_dmreq(cc, dmreq), dmreq);
+
mempool_free(req_of_dmreq(cc, dmreq), cc->req_pool);
if (!atomic_dec_and_test(&ctx->pending))
@@ -971,34 +1282,84 @@
}
}
+static void crypt_free_tfms(struct crypt_config *cc, int cpu)
+{
+ struct crypt_cpu *cpu_cc = per_cpu_ptr(cc->cpu, cpu);
+ unsigned i;
+
+ for (i = 0; i < cc->tfms_count; i++)
+ if (cpu_cc->tfms[i] && !IS_ERR(cpu_cc->tfms[i])) {
+ crypto_free_ablkcipher(cpu_cc->tfms[i]);
+ cpu_cc->tfms[i] = NULL;
+ }
+}
+
+static int crypt_alloc_tfms(struct crypt_config *cc, int cpu, char *ciphermode)
+{
+ struct crypt_cpu *cpu_cc = per_cpu_ptr(cc->cpu, cpu);
+ unsigned i;
+ int err;
+
+ for (i = 0; i < cc->tfms_count; i++) {
+ cpu_cc->tfms[i] = crypto_alloc_ablkcipher(ciphermode, 0, 0);
+ if (IS_ERR(cpu_cc->tfms[i])) {
+ err = PTR_ERR(cpu_cc->tfms[i]);
+ crypt_free_tfms(cc, cpu);
+ return err;
+ }
+ }
+
+ return 0;
+}
+
+static int crypt_setkey_allcpus(struct crypt_config *cc)
+{
+ unsigned subkey_size = cc->key_size >> ilog2(cc->tfms_count);
+ int cpu, err = 0, i, r;
+
+ for_each_possible_cpu(cpu) {
+ for (i = 0; i < cc->tfms_count; i++) {
+ r = crypto_ablkcipher_setkey(per_cpu_ptr(cc->cpu, cpu)->tfms[i],
+ cc->key + (i * subkey_size), subkey_size);
+ if (r)
+ err = r;
+ }
+ }
+
+ return err;
+}
+
static int crypt_set_key(struct crypt_config *cc, char *key)
{
- unsigned key_size = strlen(key) >> 1;
-
- if (cc->key_size && cc->key_size != key_size)
+ /* The key size may not be changed. */
+ if (cc->key_size != (strlen(key) >> 1))
return -EINVAL;
- cc->key_size = key_size; /* initial settings */
+ /* Hyphen (which gives a key_size of zero) means there is no key. */
+ if (!cc->key_size && strcmp(key, "-"))
+ return -EINVAL;
- if ((!key_size && strcmp(key, "-")) ||
- (key_size && crypt_decode_key(cc->key, key, key_size) < 0))
+ if (cc->key_size && crypt_decode_key(cc->key, key, cc->key_size) < 0)
return -EINVAL;
set_bit(DM_CRYPT_KEY_VALID, &cc->flags);
- return crypto_ablkcipher_setkey(cc->tfm, cc->key, cc->key_size);
+ return crypt_setkey_allcpus(cc);
}
static int crypt_wipe_key(struct crypt_config *cc)
{
clear_bit(DM_CRYPT_KEY_VALID, &cc->flags);
memset(&cc->key, 0, cc->key_size * sizeof(u8));
- return crypto_ablkcipher_setkey(cc->tfm, cc->key, cc->key_size);
+
+ return crypt_setkey_allcpus(cc);
}
static void crypt_dtr(struct dm_target *ti)
{
struct crypt_config *cc = ti->private;
+ struct crypt_cpu *cpu_cc;
+ int cpu;
ti->private = NULL;
@@ -1010,6 +1371,14 @@
if (cc->crypt_queue)
destroy_workqueue(cc->crypt_queue);
+ if (cc->cpu)
+ for_each_possible_cpu(cpu) {
+ cpu_cc = per_cpu_ptr(cc->cpu, cpu);
+ if (cpu_cc->req)
+ mempool_free(cpu_cc->req, cc->req_pool);
+ crypt_free_tfms(cc, cpu);
+ }
+
if (cc->bs)
bioset_free(cc->bs);
@@ -1023,14 +1392,14 @@
if (cc->iv_gen_ops && cc->iv_gen_ops->dtr)
cc->iv_gen_ops->dtr(cc);
- if (cc->tfm && !IS_ERR(cc->tfm))
- crypto_free_ablkcipher(cc->tfm);
-
if (cc->dev)
dm_put_device(ti, cc->dev);
+ if (cc->cpu)
+ free_percpu(cc->cpu);
+
kzfree(cc->cipher);
- kzfree(cc->cipher_mode);
+ kzfree(cc->cipher_string);
/* Must zero key material before freeing */
kzfree(cc);
@@ -1040,9 +1409,9 @@
char *cipher_in, char *key)
{
struct crypt_config *cc = ti->private;
- char *tmp, *cipher, *chainmode, *ivmode, *ivopts;
+ char *tmp, *cipher, *chainmode, *ivmode, *ivopts, *keycount;
char *cipher_api = NULL;
- int ret = -EINVAL;
+ int cpu, ret = -EINVAL;
/* Convert to crypto api definition? */
if (strchr(cipher_in, '(')) {
@@ -1050,23 +1419,31 @@
return -EINVAL;
}
+ cc->cipher_string = kstrdup(cipher_in, GFP_KERNEL);
+ if (!cc->cipher_string)
+ goto bad_mem;
+
/*
* Legacy dm-crypt cipher specification
- * cipher-mode-iv:ivopts
+ * cipher[:keycount]-mode-iv:ivopts
*/
tmp = cipher_in;
- cipher = strsep(&tmp, "-");
+ keycount = strsep(&tmp, "-");
+ cipher = strsep(&keycount, ":");
+
+ if (!keycount)
+ cc->tfms_count = 1;
+ else if (sscanf(keycount, "%u", &cc->tfms_count) != 1 ||
+ !is_power_of_2(cc->tfms_count)) {
+ ti->error = "Bad cipher key count specification";
+ return -EINVAL;
+ }
+ cc->key_parts = cc->tfms_count;
cc->cipher = kstrdup(cipher, GFP_KERNEL);
if (!cc->cipher)
goto bad_mem;
- if (tmp) {
- cc->cipher_mode = kstrdup(tmp, GFP_KERNEL);
- if (!cc->cipher_mode)
- goto bad_mem;
- }
-
chainmode = strsep(&tmp, "-");
ivopts = strsep(&tmp, "-");
ivmode = strsep(&ivopts, ":");
@@ -1074,10 +1451,19 @@
if (tmp)
DMWARN("Ignoring unexpected additional cipher options");
- /* Compatibility mode for old dm-crypt mappings */
+ cc->cpu = __alloc_percpu(sizeof(*(cc->cpu)) +
+ cc->tfms_count * sizeof(*(cc->cpu->tfms)),
+ __alignof__(struct crypt_cpu));
+ if (!cc->cpu) {
+ ti->error = "Cannot allocate per cpu state";
+ goto bad_mem;
+ }
+
+ /*
+ * For compatibility with the original dm-crypt mapping format, if
+ * only the cipher name is supplied, use cbc-plain.
+ */
if (!chainmode || (!strcmp(chainmode, "plain") && !ivmode)) {
- kfree(cc->cipher_mode);
- cc->cipher_mode = kstrdup("cbc-plain", GFP_KERNEL);
chainmode = "cbc";
ivmode = "plain";
}
@@ -1099,11 +1485,12 @@
}
/* Allocate cipher */
- cc->tfm = crypto_alloc_ablkcipher(cipher_api, 0, 0);
- if (IS_ERR(cc->tfm)) {
- ret = PTR_ERR(cc->tfm);
- ti->error = "Error allocating crypto tfm";
- goto bad;
+ for_each_possible_cpu(cpu) {
+ ret = crypt_alloc_tfms(cc, cpu, cipher_api);
+ if (ret < 0) {
+ ti->error = "Error allocating crypto tfm";
+ goto bad;
+ }
}
/* Initialize and set key */
@@ -1114,7 +1501,7 @@
}
/* Initialize IV */
- cc->iv_size = crypto_ablkcipher_ivsize(cc->tfm);
+ cc->iv_size = crypto_ablkcipher_ivsize(any_tfm(cc));
if (cc->iv_size)
/* at least a 64 bit sector number should fit in our buffer */
cc->iv_size = max(cc->iv_size,
@@ -1137,7 +1524,15 @@
cc->iv_gen_ops = &crypt_iv_benbi_ops;
else if (strcmp(ivmode, "null") == 0)
cc->iv_gen_ops = &crypt_iv_null_ops;
- else {
+ else if (strcmp(ivmode, "lmk") == 0) {
+ cc->iv_gen_ops = &crypt_iv_lmk_ops;
+ /* Version 2 and 3 is recognised according
+ * to length of provided multi-key string.
+ * If present (version 3), last key is used as IV seed.
+ */
+ if (cc->key_size % cc->key_parts)
+ cc->key_parts++;
+ } else {
ret = -EINVAL;
ti->error = "Invalid IV mode";
goto bad;
@@ -1194,6 +1589,7 @@
ti->error = "Cannot allocate encryption context";
return -ENOMEM;
}
+ cc->key_size = key_size;
ti->private = cc;
ret = crypt_ctr_cipher(ti, argv[0], argv[1]);
@@ -1208,9 +1604,9 @@
}
cc->dmreq_start = sizeof(struct ablkcipher_request);
- cc->dmreq_start += crypto_ablkcipher_reqsize(cc->tfm);
+ cc->dmreq_start += crypto_ablkcipher_reqsize(any_tfm(cc));
cc->dmreq_start = ALIGN(cc->dmreq_start, crypto_tfm_ctx_alignment());
- cc->dmreq_start += crypto_ablkcipher_alignmask(cc->tfm) &
+ cc->dmreq_start += crypto_ablkcipher_alignmask(any_tfm(cc)) &
~(crypto_tfm_ctx_alignment() - 1);
cc->req_pool = mempool_create_kmalloc_pool(MIN_IOS, cc->dmreq_start +
@@ -1219,7 +1615,6 @@
ti->error = "Cannot allocate crypt request mempool";
goto bad;
}
- cc->req = NULL;
cc->page_pool = mempool_create_page_pool(MIN_POOL_PAGES, 0);
if (!cc->page_pool) {
@@ -1252,13 +1647,20 @@
cc->start = tmpll;
ret = -ENOMEM;
- cc->io_queue = create_singlethread_workqueue("kcryptd_io");
+ cc->io_queue = alloc_workqueue("kcryptd_io",
+ WQ_NON_REENTRANT|
+ WQ_MEM_RECLAIM,
+ 1);
if (!cc->io_queue) {
ti->error = "Couldn't create kcryptd io queue";
goto bad;
}
- cc->crypt_queue = create_singlethread_workqueue("kcryptd");
+ cc->crypt_queue = alloc_workqueue("kcryptd",
+ WQ_NON_REENTRANT|
+ WQ_CPU_INTENSIVE|
+ WQ_MEM_RECLAIM,
+ 1);
if (!cc->crypt_queue) {
ti->error = "Couldn't create kcryptd queue";
goto bad;
@@ -1286,9 +1688,10 @@
io = crypt_io_alloc(ti, bio, dm_target_offset(ti, bio->bi_sector));
- if (bio_data_dir(io->base_bio) == READ)
- kcryptd_queue_io(io);
- else
+ if (bio_data_dir(io->base_bio) == READ) {
+ if (kcryptd_io_read(io, GFP_NOWAIT))
+ kcryptd_queue_io(io);
+ } else
kcryptd_queue_crypt(io);
return DM_MAPIO_SUBMITTED;
@@ -1306,10 +1709,7 @@
break;
case STATUSTYPE_TABLE:
- if (cc->cipher_mode)
- DMEMIT("%s-%s ", cc->cipher, cc->cipher_mode);
- else
- DMEMIT("%s ", cc->cipher);
+ DMEMIT("%s ", cc->cipher_string);
if (cc->key_size > 0) {
if ((maxlen - sz) < ((cc->key_size << 1) + 1))
@@ -1421,7 +1821,7 @@
static struct target_type crypt_target = {
.name = "crypt",
- .version = {1, 7, 0},
+ .version = {1, 10, 0},
.module = THIS_MODULE,
.ctr = crypt_ctr,
.dtr = crypt_dtr,
diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c
index baa1191..f18375d 100644
--- a/drivers/md/dm-delay.c
+++ b/drivers/md/dm-delay.c
@@ -352,7 +352,7 @@
{
int r = -ENOMEM;
- kdelayd_wq = create_workqueue("kdelayd");
+ kdelayd_wq = alloc_workqueue("kdelayd", WQ_MEM_RECLAIM, 0);
if (!kdelayd_wq) {
DMERR("Couldn't start kdelayd");
goto bad_queue;
diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index 4b54618..6d12775 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -295,19 +295,55 @@
DMWARN("remove_all left %d open device(s)", dev_skipped);
}
+/*
+ * Set the uuid of a hash_cell that isn't already set.
+ */
+static void __set_cell_uuid(struct hash_cell *hc, char *new_uuid)
+{
+ mutex_lock(&dm_hash_cells_mutex);
+ hc->uuid = new_uuid;
+ mutex_unlock(&dm_hash_cells_mutex);
+
+ list_add(&hc->uuid_list, _uuid_buckets + hash_str(new_uuid));
+}
+
+/*
+ * Changes the name of a hash_cell and returns the old name for
+ * the caller to free.
+ */
+static char *__change_cell_name(struct hash_cell *hc, char *new_name)
+{
+ char *old_name;
+
+ /*
+ * Rename and move the name cell.
+ */
+ list_del(&hc->name_list);
+ old_name = hc->name;
+
+ mutex_lock(&dm_hash_cells_mutex);
+ hc->name = new_name;
+ mutex_unlock(&dm_hash_cells_mutex);
+
+ list_add(&hc->name_list, _name_buckets + hash_str(new_name));
+
+ return old_name;
+}
+
static struct mapped_device *dm_hash_rename(struct dm_ioctl *param,
const char *new)
{
- char *new_name, *old_name;
+ char *new_data, *old_name = NULL;
struct hash_cell *hc;
struct dm_table *table;
struct mapped_device *md;
+ unsigned change_uuid = (param->flags & DM_UUID_FLAG) ? 1 : 0;
/*
* duplicate new.
*/
- new_name = kstrdup(new, GFP_KERNEL);
- if (!new_name)
+ new_data = kstrdup(new, GFP_KERNEL);
+ if (!new_data)
return ERR_PTR(-ENOMEM);
down_write(&_hash_lock);
@@ -315,13 +351,19 @@
/*
* Is new free ?
*/
- hc = __get_name_cell(new);
+ if (change_uuid)
+ hc = __get_uuid_cell(new);
+ else
+ hc = __get_name_cell(new);
+
if (hc) {
- DMWARN("asked to rename to an already-existing name %s -> %s",
+ DMWARN("Unable to change %s on mapped device %s to one that "
+ "already exists: %s",
+ change_uuid ? "uuid" : "name",
param->name, new);
dm_put(hc->md);
up_write(&_hash_lock);
- kfree(new_name);
+ kfree(new_data);
return ERR_PTR(-EBUSY);
}
@@ -330,22 +372,30 @@
*/
hc = __get_name_cell(param->name);
if (!hc) {
- DMWARN("asked to rename a non-existent device %s -> %s",
- param->name, new);
+ DMWARN("Unable to rename non-existent device, %s to %s%s",
+ param->name, change_uuid ? "uuid " : "", new);
up_write(&_hash_lock);
- kfree(new_name);
+ kfree(new_data);
return ERR_PTR(-ENXIO);
}
/*
- * rename and move the name cell.
+ * Does this device already have a uuid?
*/
- list_del(&hc->name_list);
- old_name = hc->name;
- mutex_lock(&dm_hash_cells_mutex);
- hc->name = new_name;
- mutex_unlock(&dm_hash_cells_mutex);
- list_add(&hc->name_list, _name_buckets + hash_str(new_name));
+ if (change_uuid && hc->uuid) {
+ DMWARN("Unable to change uuid of mapped device %s to %s "
+ "because uuid is already set to %s",
+ param->name, new, hc->uuid);
+ dm_put(hc->md);
+ up_write(&_hash_lock);
+ kfree(new_data);
+ return ERR_PTR(-EINVAL);
+ }
+
+ if (change_uuid)
+ __set_cell_uuid(hc, new_data);
+ else
+ old_name = __change_cell_name(hc, new_data);
/*
* Wake up any dm event waiters.
@@ -729,7 +779,7 @@
hc = __find_device_hash_cell(param);
if (!hc) {
- DMWARN("device doesn't appear to be in the dev hash table.");
+ DMDEBUG_LIMIT("device doesn't appear to be in the dev hash table.");
up_write(&_hash_lock);
return -ENXIO;
}
@@ -741,7 +791,7 @@
*/
r = dm_lock_for_deletion(md);
if (r) {
- DMWARN("unable to remove open device %s", hc->name);
+ DMDEBUG_LIMIT("unable to remove open device %s", hc->name);
up_write(&_hash_lock);
dm_put(md);
return r;
@@ -774,21 +824,24 @@
static int dev_rename(struct dm_ioctl *param, size_t param_size)
{
int r;
- char *new_name = (char *) param + param->data_start;
+ char *new_data = (char *) param + param->data_start;
struct mapped_device *md;
+ unsigned change_uuid = (param->flags & DM_UUID_FLAG) ? 1 : 0;
- if (new_name < param->data ||
- invalid_str(new_name, (void *) param + param_size) ||
- strlen(new_name) > DM_NAME_LEN - 1) {
- DMWARN("Invalid new logical volume name supplied.");
+ if (new_data < param->data ||
+ invalid_str(new_data, (void *) param + param_size) ||
+ strlen(new_data) > (change_uuid ? DM_UUID_LEN - 1 : DM_NAME_LEN - 1)) {
+ DMWARN("Invalid new mapped device name or uuid string supplied.");
return -EINVAL;
}
- r = check_name(new_name);
- if (r)
- return r;
+ if (!change_uuid) {
+ r = check_name(new_data);
+ if (r)
+ return r;
+ }
- md = dm_hash_rename(param, new_name);
+ md = dm_hash_rename(param, new_data);
if (IS_ERR(md))
return PTR_ERR(md);
@@ -885,7 +938,7 @@
hc = __find_device_hash_cell(param);
if (!hc) {
- DMWARN("device doesn't appear to be in the dev hash table.");
+ DMDEBUG_LIMIT("device doesn't appear to be in the dev hash table.");
up_write(&_hash_lock);
return -ENXIO;
}
@@ -1212,7 +1265,7 @@
hc = __find_device_hash_cell(param);
if (!hc) {
- DMWARN("device doesn't appear to be in the dev hash table.");
+ DMDEBUG_LIMIT("device doesn't appear to be in the dev hash table.");
up_write(&_hash_lock);
return -ENXIO;
}
diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c
index d8587ba..924f5f0 100644
--- a/drivers/md/dm-kcopyd.c
+++ b/drivers/md/dm-kcopyd.c
@@ -37,6 +37,13 @@
unsigned int nr_pages;
unsigned int nr_free_pages;
+ /*
+ * Block devices to unplug.
+ * Non-NULL pointer means that a block device has some pending requests
+ * and needs to be unplugged.
+ */
+ struct block_device *unplug[2];
+
struct dm_io_client *io_client;
wait_queue_head_t destroyq;
@@ -308,6 +315,31 @@
return 0;
}
+/*
+ * Unplug the block device at the specified index.
+ */
+static void unplug(struct dm_kcopyd_client *kc, int rw)
+{
+ if (kc->unplug[rw] != NULL) {
+ blk_unplug(bdev_get_queue(kc->unplug[rw]));
+ kc->unplug[rw] = NULL;
+ }
+}
+
+/*
+ * Prepare block device unplug. If there's another device
+ * to be unplugged at the same array index, we unplug that
+ * device first.
+ */
+static void prepare_unplug(struct dm_kcopyd_client *kc, int rw,
+ struct block_device *bdev)
+{
+ if (likely(kc->unplug[rw] == bdev))
+ return;
+ unplug(kc, rw);
+ kc->unplug[rw] = bdev;
+}
+
static void complete_io(unsigned long error, void *context)
{
struct kcopyd_job *job = (struct kcopyd_job *) context;
@@ -345,7 +377,7 @@
{
int r;
struct dm_io_request io_req = {
- .bi_rw = job->rw | REQ_SYNC | REQ_UNPLUG,
+ .bi_rw = job->rw,
.mem.type = DM_IO_PAGE_LIST,
.mem.ptr.pl = job->pages,
.mem.offset = job->offset,
@@ -354,10 +386,16 @@
.client = job->kc->io_client,
};
- if (job->rw == READ)
+ if (job->rw == READ) {
r = dm_io(&io_req, 1, &job->source, NULL);
- else
+ prepare_unplug(job->kc, READ, job->source.bdev);
+ } else {
+ if (job->num_dests > 1)
+ io_req.bi_rw |= REQ_UNPLUG;
r = dm_io(&io_req, job->num_dests, job->dests, NULL);
+ if (!(io_req.bi_rw & REQ_UNPLUG))
+ prepare_unplug(job->kc, WRITE, job->dests[0].bdev);
+ }
return r;
}
@@ -435,10 +473,18 @@
* Pages jobs when successful will jump onto the io jobs
* list. io jobs call wake when they complete and it all
* starts again.
+ *
+ * Note that io_jobs add block devices to the unplug array,
+ * this array is cleared with "unplug" calls. It is thus
+ * forbidden to run complete_jobs after io_jobs and before
+ * unplug because the block device could be destroyed in
+ * job completion callback.
*/
process_jobs(&kc->complete_jobs, kc, run_complete_job);
process_jobs(&kc->pages_jobs, kc, run_pages_job);
process_jobs(&kc->io_jobs, kc, run_io_job);
+ unplug(kc, READ);
+ unplug(kc, WRITE);
}
/*
@@ -619,12 +665,15 @@
INIT_LIST_HEAD(&kc->io_jobs);
INIT_LIST_HEAD(&kc->pages_jobs);
+ memset(kc->unplug, 0, sizeof(kc->unplug));
+
kc->job_pool = mempool_create_slab_pool(MIN_JOBS, _job_cache);
if (!kc->job_pool)
goto bad_slab;
INIT_WORK(&kc->kcopyd_work, do_work);
- kc->kcopyd_wq = create_singlethread_workqueue("kcopyd");
+ kc->kcopyd_wq = alloc_workqueue("kcopyd",
+ WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0);
if (!kc->kcopyd_wq)
goto bad_workqueue;
diff --git a/drivers/md/dm-log-userspace-base.c b/drivers/md/dm-log-userspace-base.c
index 1ed0094..aa2e0c3 100644
--- a/drivers/md/dm-log-userspace-base.c
+++ b/drivers/md/dm-log-userspace-base.c
@@ -12,12 +12,22 @@
#include "dm-log-userspace-transfer.h"
+#define DM_LOG_USERSPACE_VSN "1.1.0"
+
struct flush_entry {
int type;
region_t region;
struct list_head list;
};
+/*
+ * This limit on the number of mark and clear request is, to a degree,
+ * arbitrary. However, there is some basis for the choice in the limits
+ * imposed on the size of data payload by dm-log-userspace-transfer.c:
+ * dm_consult_userspace().
+ */
+#define MAX_FLUSH_GROUP_COUNT 32
+
struct log_c {
struct dm_target *ti;
uint32_t region_size;
@@ -37,8 +47,15 @@
*/
uint64_t in_sync_hint;
+ /*
+ * Mark and clear requests are held until a flush is issued
+ * so that we can group, and thereby limit, the amount of
+ * network traffic between kernel and userspace. The 'flush_lock'
+ * is used to protect these lists.
+ */
spinlock_t flush_lock;
- struct list_head flush_list; /* only for clear and mark requests */
+ struct list_head mark_list;
+ struct list_head clear_list;
};
static mempool_t *flush_entry_pool;
@@ -169,7 +186,8 @@
strncpy(lc->uuid, argv[0], DM_UUID_LEN);
spin_lock_init(&lc->flush_lock);
- INIT_LIST_HEAD(&lc->flush_list);
+ INIT_LIST_HEAD(&lc->mark_list);
+ INIT_LIST_HEAD(&lc->clear_list);
str_size = build_constructor_string(ti, argc - 1, argv + 1, &ctr_str);
if (str_size < 0) {
@@ -181,8 +199,11 @@
r = dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_CTR,
ctr_str, str_size, NULL, NULL);
- if (r == -ESRCH) {
- DMERR("Userspace log server not found");
+ if (r < 0) {
+ if (r == -ESRCH)
+ DMERR("Userspace log server not found");
+ else
+ DMERR("Userspace log server failed to create log");
goto out;
}
@@ -214,10 +235,9 @@
static void userspace_dtr(struct dm_dirty_log *log)
{
- int r;
struct log_c *lc = log->context;
- r = dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_DTR,
+ (void) dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_DTR,
NULL, 0,
NULL, NULL);
@@ -338,6 +358,71 @@
return (r) ? 0 : (int)in_sync;
}
+static int flush_one_by_one(struct log_c *lc, struct list_head *flush_list)
+{
+ int r = 0;
+ struct flush_entry *fe;
+
+ list_for_each_entry(fe, flush_list, list) {
+ r = userspace_do_request(lc, lc->uuid, fe->type,
+ (char *)&fe->region,
+ sizeof(fe->region),
+ NULL, NULL);
+ if (r)
+ break;
+ }
+
+ return r;
+}
+
+static int flush_by_group(struct log_c *lc, struct list_head *flush_list)
+{
+ int r = 0;
+ int count;
+ uint32_t type = 0;
+ struct flush_entry *fe, *tmp_fe;
+ LIST_HEAD(tmp_list);
+ uint64_t group[MAX_FLUSH_GROUP_COUNT];
+
+ /*
+ * Group process the requests
+ */
+ while (!list_empty(flush_list)) {
+ count = 0;
+
+ list_for_each_entry_safe(fe, tmp_fe, flush_list, list) {
+ group[count] = fe->region;
+ count++;
+
+ list_del(&fe->list);
+ list_add(&fe->list, &tmp_list);
+
+ type = fe->type;
+ if (count >= MAX_FLUSH_GROUP_COUNT)
+ break;
+ }
+
+ r = userspace_do_request(lc, lc->uuid, type,
+ (char *)(group),
+ count * sizeof(uint64_t),
+ NULL, NULL);
+ if (r) {
+ /* Group send failed. Attempt one-by-one. */
+ list_splice_init(&tmp_list, flush_list);
+ r = flush_one_by_one(lc, flush_list);
+ break;
+ }
+ }
+
+ /*
+ * Must collect flush_entrys that were successfully processed
+ * as a group so that they will be free'd by the caller.
+ */
+ list_splice_init(&tmp_list, flush_list);
+
+ return r;
+}
+
/*
* userspace_flush
*
@@ -360,31 +445,25 @@
int r = 0;
unsigned long flags;
struct log_c *lc = log->context;
- LIST_HEAD(flush_list);
+ LIST_HEAD(mark_list);
+ LIST_HEAD(clear_list);
struct flush_entry *fe, *tmp_fe;
spin_lock_irqsave(&lc->flush_lock, flags);
- list_splice_init(&lc->flush_list, &flush_list);
+ list_splice_init(&lc->mark_list, &mark_list);
+ list_splice_init(&lc->clear_list, &clear_list);
spin_unlock_irqrestore(&lc->flush_lock, flags);
- if (list_empty(&flush_list))
+ if (list_empty(&mark_list) && list_empty(&clear_list))
return 0;
- /*
- * FIXME: Count up requests, group request types,
- * allocate memory to stick all requests in and
- * send to server in one go. Failing the allocation,
- * do it one by one.
- */
+ r = flush_by_group(lc, &mark_list);
+ if (r)
+ goto fail;
- list_for_each_entry(fe, &flush_list, list) {
- r = userspace_do_request(lc, lc->uuid, fe->type,
- (char *)&fe->region,
- sizeof(fe->region),
- NULL, NULL);
- if (r)
- goto fail;
- }
+ r = flush_by_group(lc, &clear_list);
+ if (r)
+ goto fail;
r = userspace_do_request(lc, lc->uuid, DM_ULOG_FLUSH,
NULL, 0, NULL, NULL);
@@ -395,7 +474,11 @@
* Calling code will receive an error and will know that
* the log facility has failed.
*/
- list_for_each_entry_safe(fe, tmp_fe, &flush_list, list) {
+ list_for_each_entry_safe(fe, tmp_fe, &mark_list, list) {
+ list_del(&fe->list);
+ mempool_free(fe, flush_entry_pool);
+ }
+ list_for_each_entry_safe(fe, tmp_fe, &clear_list, list) {
list_del(&fe->list);
mempool_free(fe, flush_entry_pool);
}
@@ -425,7 +508,7 @@
spin_lock_irqsave(&lc->flush_lock, flags);
fe->type = DM_ULOG_MARK_REGION;
fe->region = region;
- list_add(&fe->list, &lc->flush_list);
+ list_add(&fe->list, &lc->mark_list);
spin_unlock_irqrestore(&lc->flush_lock, flags);
return;
@@ -462,7 +545,7 @@
spin_lock_irqsave(&lc->flush_lock, flags);
fe->type = DM_ULOG_CLEAR_REGION;
fe->region = region;
- list_add(&fe->list, &lc->flush_list);
+ list_add(&fe->list, &lc->clear_list);
spin_unlock_irqrestore(&lc->flush_lock, flags);
return;
@@ -684,7 +767,7 @@
return r;
}
- DMINFO("version 1.0.0 loaded");
+ DMINFO("version " DM_LOG_USERSPACE_VSN " loaded");
return 0;
}
@@ -694,7 +777,7 @@
dm_ulog_tfr_exit();
mempool_destroy(flush_entry_pool);
- DMINFO("version 1.0.0 unloaded");
+ DMINFO("version " DM_LOG_USERSPACE_VSN " unloaded");
return;
}
diff --git a/drivers/md/dm-log-userspace-transfer.c b/drivers/md/dm-log-userspace-transfer.c
index 075cbcf..049eaf1 100644
--- a/drivers/md/dm-log-userspace-transfer.c
+++ b/drivers/md/dm-log-userspace-transfer.c
@@ -198,6 +198,7 @@
memset(tfr, 0, DM_ULOG_PREALLOCED_SIZE - sizeof(struct cn_msg));
memcpy(tfr->uuid, uuid, DM_UUID_LEN);
+ tfr->version = DM_ULOG_REQUEST_VERSION;
tfr->luid = luid;
tfr->seq = dm_ulog_seq++;
diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c
index 33420e6..6951536 100644
--- a/drivers/md/dm-log.c
+++ b/drivers/md/dm-log.c
@@ -455,7 +455,7 @@
r = PTR_ERR(lc->io_req.client);
DMWARN("couldn't allocate disk io client");
kfree(lc);
- return -ENOMEM;
+ return r;
}
lc->disk_header = vmalloc(buf_size);
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 487ecda..b82d288 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -23,6 +23,8 @@
#define DM_MSG_PREFIX "multipath"
#define MESG_STR(x) x, sizeof(x)
+#define DM_PG_INIT_DELAY_MSECS 2000
+#define DM_PG_INIT_DELAY_DEFAULT ((unsigned) -1)
/* Path properties */
struct pgpath {
@@ -33,8 +35,7 @@
unsigned fail_count; /* Cumulative failure count */
struct dm_path path;
- struct work_struct deactivate_path;
- struct work_struct activate_path;
+ struct delayed_work activate_path;
};
#define path_to_pgpath(__pgp) container_of((__pgp), struct pgpath, path)
@@ -65,11 +66,15 @@
const char *hw_handler_name;
char *hw_handler_params;
+
unsigned nr_priority_groups;
struct list_head priority_groups;
+
+ wait_queue_head_t pg_init_wait; /* Wait for pg_init completion */
+
unsigned pg_init_required; /* pg_init needs calling? */
unsigned pg_init_in_progress; /* Only one pg_init allowed at once */
- wait_queue_head_t pg_init_wait; /* Wait for pg_init completion */
+ unsigned pg_init_delay_retry; /* Delay pg_init retry? */
unsigned nr_valid_paths; /* Total number of usable paths */
struct pgpath *current_pgpath;
@@ -82,6 +87,7 @@
unsigned saved_queue_if_no_path;/* Saved state during suspension */
unsigned pg_init_retries; /* Number of times to retry pg_init */
unsigned pg_init_count; /* Number of times pg_init called */
+ unsigned pg_init_delay_msecs; /* Number of msecs before pg_init retry */
struct work_struct process_queued_ios;
struct list_head queued_ios;
@@ -116,7 +122,6 @@
static void process_queued_ios(struct work_struct *work);
static void trigger_event(struct work_struct *work);
static void activate_path(struct work_struct *work);
-static void deactivate_path(struct work_struct *work);
/*-----------------------------------------------
@@ -129,8 +134,7 @@
if (pgpath) {
pgpath->is_active = 1;
- INIT_WORK(&pgpath->deactivate_path, deactivate_path);
- INIT_WORK(&pgpath->activate_path, activate_path);
+ INIT_DELAYED_WORK(&pgpath->activate_path, activate_path);
}
return pgpath;
@@ -141,14 +145,6 @@
kfree(pgpath);
}
-static void deactivate_path(struct work_struct *work)
-{
- struct pgpath *pgpath =
- container_of(work, struct pgpath, deactivate_path);
-
- blk_abort_queue(pgpath->path.dev->bdev->bd_disk->queue);
-}
-
static struct priority_group *alloc_priority_group(void)
{
struct priority_group *pg;
@@ -199,6 +195,7 @@
INIT_LIST_HEAD(&m->queued_ios);
spin_lock_init(&m->lock);
m->queue_io = 1;
+ m->pg_init_delay_msecs = DM_PG_INIT_DELAY_DEFAULT;
INIT_WORK(&m->process_queued_ios, process_queued_ios);
INIT_WORK(&m->trigger_event, trigger_event);
init_waitqueue_head(&m->pg_init_wait);
@@ -238,14 +235,19 @@
static void __pg_init_all_paths(struct multipath *m)
{
struct pgpath *pgpath;
+ unsigned long pg_init_delay = 0;
m->pg_init_count++;
m->pg_init_required = 0;
+ if (m->pg_init_delay_retry)
+ pg_init_delay = msecs_to_jiffies(m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT ?
+ m->pg_init_delay_msecs : DM_PG_INIT_DELAY_MSECS);
list_for_each_entry(pgpath, &m->current_pg->pgpaths, list) {
/* Skip failed paths */
if (!pgpath->is_active)
continue;
- if (queue_work(kmpath_handlerd, &pgpath->activate_path))
+ if (queue_delayed_work(kmpath_handlerd, &pgpath->activate_path,
+ pg_init_delay))
m->pg_init_in_progress++;
}
}
@@ -793,8 +795,9 @@
const char *param_name;
static struct param _params[] = {
- {0, 3, "invalid number of feature args"},
+ {0, 5, "invalid number of feature args"},
{1, 50, "pg_init_retries must be between 1 and 50"},
+ {0, 60000, "pg_init_delay_msecs must be between 0 and 60000"},
};
r = read_param(_params, shift(as), &argc, &ti->error);
@@ -821,6 +824,14 @@
continue;
}
+ if (!strnicmp(param_name, MESG_STR("pg_init_delay_msecs")) &&
+ (argc >= 1)) {
+ r = read_param(_params + 2, shift(as),
+ &m->pg_init_delay_msecs, &ti->error);
+ argc--;
+ continue;
+ }
+
ti->error = "Unrecognised multipath feature request";
r = -EINVAL;
} while (argc && !r);
@@ -931,7 +942,7 @@
flush_workqueue(kmpath_handlerd);
multipath_wait_for_pg_init_completion(m);
flush_workqueue(kmultipathd);
- flush_scheduled_work();
+ flush_work_sync(&m->trigger_event);
}
static void multipath_dtr(struct dm_target *ti)
@@ -995,7 +1006,6 @@
pgpath->path.dev->name, m->nr_valid_paths);
schedule_work(&m->trigger_event);
- queue_work(kmultipathd, &pgpath->deactivate_path);
out:
spin_unlock_irqrestore(&m->lock, flags);
@@ -1034,7 +1044,7 @@
m->current_pgpath = NULL;
queue_work(kmultipathd, &m->process_queued_ios);
} else if (m->hw_handler_name && (m->current_pg == pgpath->pg)) {
- if (queue_work(kmpath_handlerd, &pgpath->activate_path))
+ if (queue_work(kmpath_handlerd, &pgpath->activate_path.work))
m->pg_init_in_progress++;
}
@@ -1169,6 +1179,7 @@
struct priority_group *pg = pgpath->pg;
struct multipath *m = pg->m;
unsigned long flags;
+ unsigned delay_retry = 0;
/* device or driver problems */
switch (errors) {
@@ -1193,8 +1204,9 @@
*/
bypass_pg(m, pg, 1);
break;
- /* TODO: For SCSI_DH_RETRY we should wait a couple seconds */
case SCSI_DH_RETRY:
+ /* Wait before retrying. */
+ delay_retry = 1;
case SCSI_DH_IMM_RETRY:
case SCSI_DH_RES_TEMP_UNAVAIL:
if (pg_init_limit_reached(m, pgpath))
@@ -1227,6 +1239,7 @@
if (!m->pg_init_required)
m->queue_io = 0;
+ m->pg_init_delay_retry = delay_retry;
queue_work(kmultipathd, &m->process_queued_ios);
/*
@@ -1241,7 +1254,7 @@
static void activate_path(struct work_struct *work)
{
struct pgpath *pgpath =
- container_of(work, struct pgpath, activate_path);
+ container_of(work, struct pgpath, activate_path.work);
scsi_dh_activate(bdev_get_queue(pgpath->path.dev->bdev),
pg_init_done, pgpath);
@@ -1382,11 +1395,14 @@
DMEMIT("2 %u %u ", m->queue_size, m->pg_init_count);
else {
DMEMIT("%u ", m->queue_if_no_path +
- (m->pg_init_retries > 0) * 2);
+ (m->pg_init_retries > 0) * 2 +
+ (m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT) * 2);
if (m->queue_if_no_path)
DMEMIT("queue_if_no_path ");
if (m->pg_init_retries)
DMEMIT("pg_init_retries %u ", m->pg_init_retries);
+ if (m->pg_init_delay_msecs != DM_PG_INIT_DELAY_DEFAULT)
+ DMEMIT("pg_init_delay_msecs %u ", m->pg_init_delay_msecs);
}
if (!m->hw_handler_name || type == STATUSTYPE_INFO)
@@ -1655,7 +1671,7 @@
*---------------------------------------------------------------*/
static struct target_type multipath_target = {
.name = "multipath",
- .version = {1, 1, 1},
+ .version = {1, 2, 0},
.module = THIS_MODULE,
.ctr = multipath_ctr,
.dtr = multipath_dtr,
@@ -1687,7 +1703,7 @@
return -EINVAL;
}
- kmultipathd = create_workqueue("kmpathd");
+ kmultipathd = alloc_workqueue("kmpathd", WQ_MEM_RECLAIM, 0);
if (!kmultipathd) {
DMERR("failed to create workqueue kmpathd");
dm_unregister_target(&multipath_target);
@@ -1701,7 +1717,8 @@
* old workqueue would also create a bottleneck in the
* path of the storage hardware device activation.
*/
- kmpath_handlerd = create_singlethread_workqueue("kmpath_handlerd");
+ kmpath_handlerd = alloc_ordered_workqueue("kmpath_handlerd",
+ WQ_MEM_RECLAIM);
if (!kmpath_handlerd) {
DMERR("failed to create workqueue kmpath_handlerd");
destroy_workqueue(kmultipathd);
diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c
new file mode 100644
index 0000000..b9e1e15
--- /dev/null
+++ b/drivers/md/dm-raid.c
@@ -0,0 +1,697 @@
+/*
+ * Copyright (C) 2010-2011 Neil Brown
+ * Copyright (C) 2010-2011 Red Hat, Inc. All rights reserved.
+ *
+ * This file is released under the GPL.
+ */
+
+#include <linux/slab.h>
+
+#include "md.h"
+#include "raid5.h"
+#include "dm.h"
+#include "bitmap.h"
+
+#define DM_MSG_PREFIX "raid"
+
+/*
+ * If the MD doesn't support MD_SYNC_STATE_FORCED yet, then
+ * make it so the flag doesn't set anything.
+ */
+#ifndef MD_SYNC_STATE_FORCED
+#define MD_SYNC_STATE_FORCED 0
+#endif
+
+struct raid_dev {
+ /*
+ * Two DM devices, one to hold metadata and one to hold the
+ * actual data/parity. The reason for this is to not confuse
+ * ti->len and give more flexibility in altering size and
+ * characteristics.
+ *
+ * While it is possible for this device to be associated
+ * with a different physical device than the data_dev, it
+ * is intended for it to be the same.
+ * |--------- Physical Device ---------|
+ * |- meta_dev -|------ data_dev ------|
+ */
+ struct dm_dev *meta_dev;
+ struct dm_dev *data_dev;
+ struct mdk_rdev_s rdev;
+};
+
+/*
+ * Flags for rs->print_flags field.
+ */
+#define DMPF_DAEMON_SLEEP 0x1
+#define DMPF_MAX_WRITE_BEHIND 0x2
+#define DMPF_SYNC 0x4
+#define DMPF_NOSYNC 0x8
+#define DMPF_STRIPE_CACHE 0x10
+#define DMPF_MIN_RECOVERY_RATE 0x20
+#define DMPF_MAX_RECOVERY_RATE 0x40
+
+struct raid_set {
+ struct dm_target *ti;
+
+ uint64_t print_flags;
+
+ struct mddev_s md;
+ struct raid_type *raid_type;
+ struct dm_target_callbacks callbacks;
+
+ struct raid_dev dev[0];
+};
+
+/* Supported raid types and properties. */
+static struct raid_type {
+ const char *name; /* RAID algorithm. */
+ const char *descr; /* Descriptor text for logging. */
+ const unsigned parity_devs; /* # of parity devices. */
+ const unsigned minimal_devs; /* minimal # of devices in set. */
+ const unsigned level; /* RAID level. */
+ const unsigned algorithm; /* RAID algorithm. */
+} raid_types[] = {
+ {"raid4", "RAID4 (dedicated parity disk)", 1, 2, 5, ALGORITHM_PARITY_0},
+ {"raid5_la", "RAID5 (left asymmetric)", 1, 2, 5, ALGORITHM_LEFT_ASYMMETRIC},
+ {"raid5_ra", "RAID5 (right asymmetric)", 1, 2, 5, ALGORITHM_RIGHT_ASYMMETRIC},
+ {"raid5_ls", "RAID5 (left symmetric)", 1, 2, 5, ALGORITHM_LEFT_SYMMETRIC},
+ {"raid5_rs", "RAID5 (right symmetric)", 1, 2, 5, ALGORITHM_RIGHT_SYMMETRIC},
+ {"raid6_zr", "RAID6 (zero restart)", 2, 4, 6, ALGORITHM_ROTATING_ZERO_RESTART},
+ {"raid6_nr", "RAID6 (N restart)", 2, 4, 6, ALGORITHM_ROTATING_N_RESTART},
+ {"raid6_nc", "RAID6 (N continue)", 2, 4, 6, ALGORITHM_ROTATING_N_CONTINUE}
+};
+
+static struct raid_type *get_raid_type(char *name)
+{
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(raid_types); i++)
+ if (!strcmp(raid_types[i].name, name))
+ return &raid_types[i];
+
+ return NULL;
+}
+
+static struct raid_set *context_alloc(struct dm_target *ti, struct raid_type *raid_type, unsigned raid_devs)
+{
+ unsigned i;
+ struct raid_set *rs;
+ sector_t sectors_per_dev;
+
+ if (raid_devs <= raid_type->parity_devs) {
+ ti->error = "Insufficient number of devices";
+ return ERR_PTR(-EINVAL);
+ }
+
+ sectors_per_dev = ti->len;
+ if (sector_div(sectors_per_dev, (raid_devs - raid_type->parity_devs))) {
+ ti->error = "Target length not divisible by number of data devices";
+ return ERR_PTR(-EINVAL);
+ }
+
+ rs = kzalloc(sizeof(*rs) + raid_devs * sizeof(rs->dev[0]), GFP_KERNEL);
+ if (!rs) {
+ ti->error = "Cannot allocate raid context";
+ return ERR_PTR(-ENOMEM);
+ }
+
+ mddev_init(&rs->md);
+
+ rs->ti = ti;
+ rs->raid_type = raid_type;
+ rs->md.raid_disks = raid_devs;
+ rs->md.level = raid_type->level;
+ rs->md.new_level = rs->md.level;
+ rs->md.dev_sectors = sectors_per_dev;
+ rs->md.layout = raid_type->algorithm;
+ rs->md.new_layout = rs->md.layout;
+ rs->md.delta_disks = 0;
+ rs->md.recovery_cp = 0;
+
+ for (i = 0; i < raid_devs; i++)
+ md_rdev_init(&rs->dev[i].rdev);
+
+ /*
+ * Remaining items to be initialized by further RAID params:
+ * rs->md.persistent
+ * rs->md.external
+ * rs->md.chunk_sectors
+ * rs->md.new_chunk_sectors
+ */
+
+ return rs;
+}
+
+static void context_free(struct raid_set *rs)
+{
+ int i;
+
+ for (i = 0; i < rs->md.raid_disks; i++)
+ if (rs->dev[i].data_dev)
+ dm_put_device(rs->ti, rs->dev[i].data_dev);
+
+ kfree(rs);
+}
+
+/*
+ * For every device we have two words
+ * <meta_dev>: meta device name or '-' if missing
+ * <data_dev>: data device name or '-' if missing
+ *
+ * This code parses those words.
+ */
+static int dev_parms(struct raid_set *rs, char **argv)
+{
+ int i;
+ int rebuild = 0;
+ int metadata_available = 0;
+ int ret = 0;
+
+ for (i = 0; i < rs->md.raid_disks; i++, argv += 2) {
+ rs->dev[i].rdev.raid_disk = i;
+
+ rs->dev[i].meta_dev = NULL;
+ rs->dev[i].data_dev = NULL;
+
+ /*
+ * There are no offsets, since there is a separate device
+ * for data and metadata.
+ */
+ rs->dev[i].rdev.data_offset = 0;
+ rs->dev[i].rdev.mddev = &rs->md;
+
+ if (strcmp(argv[0], "-")) {
+ rs->ti->error = "Metadata devices not supported";
+ return -EINVAL;
+ }
+
+ if (!strcmp(argv[1], "-")) {
+ if (!test_bit(In_sync, &rs->dev[i].rdev.flags) &&
+ (!rs->dev[i].rdev.recovery_offset)) {
+ rs->ti->error = "Drive designated for rebuild not specified";
+ return -EINVAL;
+ }
+
+ continue;
+ }
+
+ ret = dm_get_device(rs->ti, argv[1],
+ dm_table_get_mode(rs->ti->table),
+ &rs->dev[i].data_dev);
+ if (ret) {
+ rs->ti->error = "RAID device lookup failure";
+ return ret;
+ }
+
+ rs->dev[i].rdev.bdev = rs->dev[i].data_dev->bdev;
+ list_add(&rs->dev[i].rdev.same_set, &rs->md.disks);
+ if (!test_bit(In_sync, &rs->dev[i].rdev.flags))
+ rebuild++;
+ }
+
+ if (metadata_available) {
+ rs->md.external = 0;
+ rs->md.persistent = 1;
+ rs->md.major_version = 2;
+ } else if (rebuild && !rs->md.recovery_cp) {
+ /*
+ * Without metadata, we will not be able to tell if the array
+ * is in-sync or not - we must assume it is not. Therefore,
+ * it is impossible to rebuild a drive.
+ *
+ * Even if there is metadata, the on-disk information may
+ * indicate that the array is not in-sync and it will then
+ * fail at that time.
+ *
+ * User could specify 'nosync' option if desperate.
+ */
+ DMERR("Unable to rebuild drive while array is not in-sync");
+ rs->ti->error = "RAID device lookup failure";
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/*
+ * Possible arguments are...
+ * RAID456:
+ * <chunk_size> [optional_args]
+ *
+ * Optional args:
+ * [[no]sync] Force or prevent recovery of the entire array
+ * [rebuild <idx>] Rebuild the drive indicated by the index
+ * [daemon_sleep <ms>] Time between bitmap daemon work to clear bits
+ * [min_recovery_rate <kB/sec/disk>] Throttle RAID initialization
+ * [max_recovery_rate <kB/sec/disk>] Throttle RAID initialization
+ * [max_write_behind <sectors>] See '-write-behind=' (man mdadm)
+ * [stripe_cache <sectors>] Stripe cache size for higher RAIDs
+ */
+static int parse_raid_params(struct raid_set *rs, char **argv,
+ unsigned num_raid_params)
+{
+ unsigned i, rebuild_cnt = 0;
+ unsigned long value;
+ char *key;
+
+ /*
+ * First, parse the in-order required arguments
+ */
+ if ((strict_strtoul(argv[0], 10, &value) < 0) ||
+ !is_power_of_2(value) || (value < 8)) {
+ rs->ti->error = "Bad chunk size";
+ return -EINVAL;
+ }
+
+ rs->md.new_chunk_sectors = rs->md.chunk_sectors = value;
+ argv++;
+ num_raid_params--;
+
+ /*
+ * Second, parse the unordered optional arguments
+ */
+ for (i = 0; i < rs->md.raid_disks; i++)
+ set_bit(In_sync, &rs->dev[i].rdev.flags);
+
+ for (i = 0; i < num_raid_params; i++) {
+ if (!strcmp(argv[i], "nosync")) {
+ rs->md.recovery_cp = MaxSector;
+ rs->print_flags |= DMPF_NOSYNC;
+ rs->md.flags |= MD_SYNC_STATE_FORCED;
+ continue;
+ }
+ if (!strcmp(argv[i], "sync")) {
+ rs->md.recovery_cp = 0;
+ rs->print_flags |= DMPF_SYNC;
+ rs->md.flags |= MD_SYNC_STATE_FORCED;
+ continue;
+ }
+
+ /* The rest of the optional arguments come in key/value pairs */
+ if ((i + 1) >= num_raid_params) {
+ rs->ti->error = "Wrong number of raid parameters given";
+ return -EINVAL;
+ }
+
+ key = argv[i++];
+ if (strict_strtoul(argv[i], 10, &value) < 0) {
+ rs->ti->error = "Bad numerical argument given in raid params";
+ return -EINVAL;
+ }
+
+ if (!strcmp(key, "rebuild")) {
+ if (++rebuild_cnt > rs->raid_type->parity_devs) {
+ rs->ti->error = "Too many rebuild drives given";
+ return -EINVAL;
+ }
+ if (value > rs->md.raid_disks) {
+ rs->ti->error = "Invalid rebuild index given";
+ return -EINVAL;
+ }
+ clear_bit(In_sync, &rs->dev[value].rdev.flags);
+ rs->dev[value].rdev.recovery_offset = 0;
+ } else if (!strcmp(key, "max_write_behind")) {
+ rs->print_flags |= DMPF_MAX_WRITE_BEHIND;
+
+ /*
+ * In device-mapper, we specify things in sectors, but
+ * MD records this value in kB
+ */
+ value /= 2;
+ if (value > COUNTER_MAX) {
+ rs->ti->error = "Max write-behind limit out of range";
+ return -EINVAL;
+ }
+ rs->md.bitmap_info.max_write_behind = value;
+ } else if (!strcmp(key, "daemon_sleep")) {
+ rs->print_flags |= DMPF_DAEMON_SLEEP;
+ if (!value || (value > MAX_SCHEDULE_TIMEOUT)) {
+ rs->ti->error = "daemon sleep period out of range";
+ return -EINVAL;
+ }
+ rs->md.bitmap_info.daemon_sleep = value;
+ } else if (!strcmp(key, "stripe_cache")) {
+ rs->print_flags |= DMPF_STRIPE_CACHE;
+
+ /*
+ * In device-mapper, we specify things in sectors, but
+ * MD records this value in kB
+ */
+ value /= 2;
+
+ if (rs->raid_type->level < 5) {
+ rs->ti->error = "Inappropriate argument: stripe_cache";
+ return -EINVAL;
+ }
+ if (raid5_set_cache_size(&rs->md, (int)value)) {
+ rs->ti->error = "Bad stripe_cache size";
+ return -EINVAL;
+ }
+ } else if (!strcmp(key, "min_recovery_rate")) {
+ rs->print_flags |= DMPF_MIN_RECOVERY_RATE;
+ if (value > INT_MAX) {
+ rs->ti->error = "min_recovery_rate out of range";
+ return -EINVAL;
+ }
+ rs->md.sync_speed_min = (int)value;
+ } else if (!strcmp(key, "max_recovery_rate")) {
+ rs->print_flags |= DMPF_MAX_RECOVERY_RATE;
+ if (value > INT_MAX) {
+ rs->ti->error = "max_recovery_rate out of range";
+ return -EINVAL;
+ }
+ rs->md.sync_speed_max = (int)value;
+ } else {
+ DMERR("Unable to parse RAID parameter: %s", key);
+ rs->ti->error = "Unable to parse RAID parameters";
+ return -EINVAL;
+ }
+ }
+
+ /* Assume there are no metadata devices until the drives are parsed */
+ rs->md.persistent = 0;
+ rs->md.external = 1;
+
+ return 0;
+}
+
+static void do_table_event(struct work_struct *ws)
+{
+ struct raid_set *rs = container_of(ws, struct raid_set, md.event_work);
+
+ dm_table_event(rs->ti->table);
+}
+
+static int raid_is_congested(struct dm_target_callbacks *cb, int bits)
+{
+ struct raid_set *rs = container_of(cb, struct raid_set, callbacks);
+
+ return md_raid5_congested(&rs->md, bits);
+}
+
+static void raid_unplug(struct dm_target_callbacks *cb)
+{
+ struct raid_set *rs = container_of(cb, struct raid_set, callbacks);
+
+ md_raid5_unplug_device(rs->md.private);
+}
+
+/*
+ * Construct a RAID4/5/6 mapping:
+ * Args:
+ * <raid_type> <#raid_params> <raid_params> \
+ * <#raid_devs> { <meta_dev1> <dev1> .. <meta_devN> <devN> }
+ *
+ * ** metadata devices are not supported yet, use '-' instead **
+ *
+ * <raid_params> varies by <raid_type>. See 'parse_raid_params' for
+ * details on possible <raid_params>.
+ */
+static int raid_ctr(struct dm_target *ti, unsigned argc, char **argv)
+{
+ int ret;
+ struct raid_type *rt;
+ unsigned long num_raid_params, num_raid_devs;
+ struct raid_set *rs = NULL;
+
+ /* Must have at least <raid_type> <#raid_params> */
+ if (argc < 2) {
+ ti->error = "Too few arguments";
+ return -EINVAL;
+ }
+
+ /* raid type */
+ rt = get_raid_type(argv[0]);
+ if (!rt) {
+ ti->error = "Unrecognised raid_type";
+ return -EINVAL;
+ }
+ argc--;
+ argv++;
+
+ /* number of RAID parameters */
+ if (strict_strtoul(argv[0], 10, &num_raid_params) < 0) {
+ ti->error = "Cannot understand number of RAID parameters";
+ return -EINVAL;
+ }
+ argc--;
+ argv++;
+
+ /* Skip over RAID params for now and find out # of devices */
+ if (num_raid_params + 1 > argc) {
+ ti->error = "Arguments do not agree with counts given";
+ return -EINVAL;
+ }
+
+ if ((strict_strtoul(argv[num_raid_params], 10, &num_raid_devs) < 0) ||
+ (num_raid_devs >= INT_MAX)) {
+ ti->error = "Cannot understand number of raid devices";
+ return -EINVAL;
+ }
+
+ rs = context_alloc(ti, rt, (unsigned)num_raid_devs);
+ if (IS_ERR(rs))
+ return PTR_ERR(rs);
+
+ ret = parse_raid_params(rs, argv, (unsigned)num_raid_params);
+ if (ret)
+ goto bad;
+
+ ret = -EINVAL;
+
+ argc -= num_raid_params + 1; /* +1: we already have num_raid_devs */
+ argv += num_raid_params + 1;
+
+ if (argc != (num_raid_devs * 2)) {
+ ti->error = "Supplied RAID devices does not match the count given";
+ goto bad;
+ }
+
+ ret = dev_parms(rs, argv);
+ if (ret)
+ goto bad;
+
+ INIT_WORK(&rs->md.event_work, do_table_event);
+ ti->split_io = rs->md.chunk_sectors;
+ ti->private = rs;
+
+ mutex_lock(&rs->md.reconfig_mutex);
+ ret = md_run(&rs->md);
+ rs->md.in_sync = 0; /* Assume already marked dirty */
+ mutex_unlock(&rs->md.reconfig_mutex);
+
+ if (ret) {
+ ti->error = "Fail to run raid array";
+ goto bad;
+ }
+
+ rs->callbacks.congested_fn = raid_is_congested;
+ rs->callbacks.unplug_fn = raid_unplug;
+ dm_table_add_target_callbacks(ti->table, &rs->callbacks);
+
+ return 0;
+
+bad:
+ context_free(rs);
+
+ return ret;
+}
+
+static void raid_dtr(struct dm_target *ti)
+{
+ struct raid_set *rs = ti->private;
+
+ list_del_init(&rs->callbacks.list);
+ md_stop(&rs->md);
+ context_free(rs);
+}
+
+static int raid_map(struct dm_target *ti, struct bio *bio, union map_info *map_context)
+{
+ struct raid_set *rs = ti->private;
+ mddev_t *mddev = &rs->md;
+
+ mddev->pers->make_request(mddev, bio);
+
+ return DM_MAPIO_SUBMITTED;
+}
+
+static int raid_status(struct dm_target *ti, status_type_t type,
+ char *result, unsigned maxlen)
+{
+ struct raid_set *rs = ti->private;
+ unsigned raid_param_cnt = 1; /* at least 1 for chunksize */
+ unsigned sz = 0;
+ int i;
+ sector_t sync;
+
+ switch (type) {
+ case STATUSTYPE_INFO:
+ DMEMIT("%s %d ", rs->raid_type->name, rs->md.raid_disks);
+
+ for (i = 0; i < rs->md.raid_disks; i++) {
+ if (test_bit(Faulty, &rs->dev[i].rdev.flags))
+ DMEMIT("D");
+ else if (test_bit(In_sync, &rs->dev[i].rdev.flags))
+ DMEMIT("A");
+ else
+ DMEMIT("a");
+ }
+
+ if (test_bit(MD_RECOVERY_RUNNING, &rs->md.recovery))
+ sync = rs->md.curr_resync_completed;
+ else
+ sync = rs->md.recovery_cp;
+
+ if (sync > rs->md.resync_max_sectors)
+ sync = rs->md.resync_max_sectors;
+
+ DMEMIT(" %llu/%llu",
+ (unsigned long long) sync,
+ (unsigned long long) rs->md.resync_max_sectors);
+
+ break;
+ case STATUSTYPE_TABLE:
+ /* The string you would use to construct this array */
+ for (i = 0; i < rs->md.raid_disks; i++)
+ if (rs->dev[i].data_dev &&
+ !test_bit(In_sync, &rs->dev[i].rdev.flags))
+ raid_param_cnt++; /* for rebuilds */
+
+ raid_param_cnt += (hweight64(rs->print_flags) * 2);
+ if (rs->print_flags & (DMPF_SYNC | DMPF_NOSYNC))
+ raid_param_cnt--;
+
+ DMEMIT("%s %u %u", rs->raid_type->name,
+ raid_param_cnt, rs->md.chunk_sectors);
+
+ if ((rs->print_flags & DMPF_SYNC) &&
+ (rs->md.recovery_cp == MaxSector))
+ DMEMIT(" sync");
+ if (rs->print_flags & DMPF_NOSYNC)
+ DMEMIT(" nosync");
+
+ for (i = 0; i < rs->md.raid_disks; i++)
+ if (rs->dev[i].data_dev &&
+ !test_bit(In_sync, &rs->dev[i].rdev.flags))
+ DMEMIT(" rebuild %u", i);
+
+ if (rs->print_flags & DMPF_DAEMON_SLEEP)
+ DMEMIT(" daemon_sleep %lu",
+ rs->md.bitmap_info.daemon_sleep);
+
+ if (rs->print_flags & DMPF_MIN_RECOVERY_RATE)
+ DMEMIT(" min_recovery_rate %d", rs->md.sync_speed_min);
+
+ if (rs->print_flags & DMPF_MAX_RECOVERY_RATE)
+ DMEMIT(" max_recovery_rate %d", rs->md.sync_speed_max);
+
+ if (rs->print_flags & DMPF_MAX_WRITE_BEHIND)
+ DMEMIT(" max_write_behind %lu",
+ rs->md.bitmap_info.max_write_behind);
+
+ if (rs->print_flags & DMPF_STRIPE_CACHE) {
+ raid5_conf_t *conf = rs->md.private;
+
+ /* convert from kiB to sectors */
+ DMEMIT(" stripe_cache %d",
+ conf ? conf->max_nr_stripes * 2 : 0);
+ }
+
+ DMEMIT(" %d", rs->md.raid_disks);
+ for (i = 0; i < rs->md.raid_disks; i++) {
+ DMEMIT(" -"); /* metadata device */
+
+ if (rs->dev[i].data_dev)
+ DMEMIT(" %s", rs->dev[i].data_dev->name);
+ else
+ DMEMIT(" -");
+ }
+ }
+
+ return 0;
+}
+
+static int raid_iterate_devices(struct dm_target *ti, iterate_devices_callout_fn fn, void *data)
+{
+ struct raid_set *rs = ti->private;
+ unsigned i;
+ int ret = 0;
+
+ for (i = 0; !ret && i < rs->md.raid_disks; i++)
+ if (rs->dev[i].data_dev)
+ ret = fn(ti,
+ rs->dev[i].data_dev,
+ 0, /* No offset on data devs */
+ rs->md.dev_sectors,
+ data);
+
+ return ret;
+}
+
+static void raid_io_hints(struct dm_target *ti, struct queue_limits *limits)
+{
+ struct raid_set *rs = ti->private;
+ unsigned chunk_size = rs->md.chunk_sectors << 9;
+ raid5_conf_t *conf = rs->md.private;
+
+ blk_limits_io_min(limits, chunk_size);
+ blk_limits_io_opt(limits, chunk_size * (conf->raid_disks - conf->max_degraded));
+}
+
+static void raid_presuspend(struct dm_target *ti)
+{
+ struct raid_set *rs = ti->private;
+
+ md_stop_writes(&rs->md);
+}
+
+static void raid_postsuspend(struct dm_target *ti)
+{
+ struct raid_set *rs = ti->private;
+
+ mddev_suspend(&rs->md);
+}
+
+static void raid_resume(struct dm_target *ti)
+{
+ struct raid_set *rs = ti->private;
+
+ mddev_resume(&rs->md);
+}
+
+static struct target_type raid_target = {
+ .name = "raid",
+ .version = {1, 0, 0},
+ .module = THIS_MODULE,
+ .ctr = raid_ctr,
+ .dtr = raid_dtr,
+ .map = raid_map,
+ .status = raid_status,
+ .iterate_devices = raid_iterate_devices,
+ .io_hints = raid_io_hints,
+ .presuspend = raid_presuspend,
+ .postsuspend = raid_postsuspend,
+ .resume = raid_resume,
+};
+
+static int __init dm_raid_init(void)
+{
+ return dm_register_target(&raid_target);
+}
+
+static void __exit dm_raid_exit(void)
+{
+ dm_unregister_target(&raid_target);
+}
+
+module_init(dm_raid_init);
+module_exit(dm_raid_exit);
+
+MODULE_DESCRIPTION(DM_NAME " raid4/5/6 target");
+MODULE_ALIAS("dm-raid4");
+MODULE_ALIAS("dm-raid5");
+MODULE_ALIAS("dm-raid6");
+MODULE_AUTHOR("Neil Brown <dm-devel@redhat.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c
index 19a59b0..dee3267 100644
--- a/drivers/md/dm-raid1.c
+++ b/drivers/md/dm-raid1.c
@@ -261,7 +261,7 @@
struct dm_io_request io_req = {
.bi_rw = WRITE_FLUSH,
.mem.type = DM_IO_KMEM,
- .mem.ptr.bvec = NULL,
+ .mem.ptr.addr = NULL,
.client = ms->io_client,
};
@@ -637,6 +637,12 @@
.client = ms->io_client,
};
+ if (bio->bi_rw & REQ_DISCARD) {
+ io_req.bi_rw |= REQ_DISCARD;
+ io_req.mem.type = DM_IO_KMEM;
+ io_req.mem.ptr.addr = NULL;
+ }
+
for (i = 0, m = ms->mirror; i < ms->nr_mirrors; i++, m++)
map_region(dest++, m, bio);
@@ -670,7 +676,8 @@
bio_list_init(&requeue);
while ((bio = bio_list_pop(writes))) {
- if (bio->bi_rw & REQ_FLUSH) {
+ if ((bio->bi_rw & REQ_FLUSH) ||
+ (bio->bi_rw & REQ_DISCARD)) {
bio_list_add(&sync, bio);
continue;
}
@@ -1076,8 +1083,10 @@
ti->private = ms;
ti->split_io = dm_rh_get_region_size(ms->rh);
ti->num_flush_requests = 1;
+ ti->num_discard_requests = 1;
- ms->kmirrord_wq = create_singlethread_workqueue("kmirrord");
+ ms->kmirrord_wq = alloc_workqueue("kmirrord",
+ WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0);
if (!ms->kmirrord_wq) {
DMERR("couldn't start kmirrord");
r = -ENOMEM;
@@ -1130,7 +1139,7 @@
del_timer_sync(&ms->timer);
flush_workqueue(ms->kmirrord_wq);
- flush_scheduled_work();
+ flush_work_sync(&ms->trigger_event);
dm_kcopyd_client_destroy(ms->kcopyd_client);
destroy_workqueue(ms->kmirrord_wq);
free_context(ms, ti, ms->nr_mirrors);
@@ -1406,7 +1415,7 @@
static struct target_type mirror_target = {
.name = "mirror",
- .version = {1, 12, 0},
+ .version = {1, 12, 1},
.module = THIS_MODULE,
.ctr = mirror_ctr,
.dtr = mirror_dtr,
diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c
index 2129cdb..95891df 100644
--- a/drivers/md/dm-snap-persistent.c
+++ b/drivers/md/dm-snap-persistent.c
@@ -256,7 +256,7 @@
*/
INIT_WORK_ONSTACK(&req.work, do_metadata);
queue_work(ps->metadata_wq, &req.work);
- flush_workqueue(ps->metadata_wq);
+ flush_work(&req.work);
return req.result;
}
@@ -818,7 +818,7 @@
atomic_set(&ps->pending_count, 0);
ps->callbacks = NULL;
- ps->metadata_wq = create_singlethread_workqueue("ksnaphd");
+ ps->metadata_wq = alloc_workqueue("ksnaphd", WQ_MEM_RECLAIM, 0);
if (!ps->metadata_wq) {
kfree(ps);
DMERR("couldn't start header metadata update thread");
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 53cf79d..fdde53c 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -19,7 +19,6 @@
#include <linux/vmalloc.h>
#include <linux/log2.h>
#include <linux/dm-kcopyd.h>
-#include <linux/workqueue.h>
#include "dm-exception-store.h"
@@ -80,9 +79,6 @@
/* Origin writes don't trigger exceptions until this is set */
int active;
- /* Whether or not owning mapped_device is suspended */
- int suspended;
-
atomic_t pending_exceptions_count;
mempool_t *pending_pool;
@@ -106,10 +102,6 @@
struct dm_kcopyd_client *kcopyd_client;
- /* Queue of snapshot writes for ksnapd to flush */
- struct bio_list queued_bios;
- struct work_struct queued_bios_work;
-
/* Wait for events based on state_bits */
unsigned long state_bits;
@@ -160,9 +152,6 @@
}
EXPORT_SYMBOL(dm_snap_cow);
-static struct workqueue_struct *ksnapd;
-static void flush_queued_bios(struct work_struct *work);
-
static sector_t chunk_to_sector(struct dm_exception_store *store,
chunk_t chunk)
{
@@ -1110,7 +1099,6 @@
s->ti = ti;
s->valid = 1;
s->active = 0;
- s->suspended = 0;
atomic_set(&s->pending_exceptions_count, 0);
init_rwsem(&s->lock);
INIT_LIST_HEAD(&s->list);
@@ -1153,9 +1141,6 @@
spin_lock_init(&s->tracked_chunk_lock);
- bio_list_init(&s->queued_bios);
- INIT_WORK(&s->queued_bios_work, flush_queued_bios);
-
ti->private = s;
ti->num_flush_requests = num_flush_requests;
@@ -1279,8 +1264,6 @@
struct dm_snapshot *s = ti->private;
struct dm_snapshot *snap_src = NULL, *snap_dest = NULL;
- flush_workqueue(ksnapd);
-
down_read(&_origins_lock);
/* Check whether exception handover must be cancelled */
(void) __find_snapshots_sharing_cow(s, &snap_src, &snap_dest, NULL);
@@ -1342,20 +1325,6 @@
}
}
-static void flush_queued_bios(struct work_struct *work)
-{
- struct dm_snapshot *s =
- container_of(work, struct dm_snapshot, queued_bios_work);
- struct bio *queued_bios;
- unsigned long flags;
-
- spin_lock_irqsave(&s->pe_lock, flags);
- queued_bios = bio_list_get(&s->queued_bios);
- spin_unlock_irqrestore(&s->pe_lock, flags);
-
- flush_bios(queued_bios);
-}
-
static int do_origin(struct dm_dev *origin, struct bio *bio);
/*
@@ -1760,15 +1729,6 @@
stop_merge(s);
}
-static void snapshot_postsuspend(struct dm_target *ti)
-{
- struct dm_snapshot *s = ti->private;
-
- down_write(&s->lock);
- s->suspended = 1;
- up_write(&s->lock);
-}
-
static int snapshot_preresume(struct dm_target *ti)
{
int r = 0;
@@ -1783,7 +1743,7 @@
DMERR("Unable to resume snapshot source until "
"handover completes.");
r = -EINVAL;
- } else if (!snap_src->suspended) {
+ } else if (!dm_suspended(snap_src->ti)) {
DMERR("Unable to perform snapshot handover until "
"source is suspended.");
r = -EINVAL;
@@ -1816,7 +1776,6 @@
down_write(&s->lock);
s->active = 1;
- s->suspended = 0;
up_write(&s->lock);
}
@@ -2194,7 +2153,7 @@
static struct target_type origin_target = {
.name = "snapshot-origin",
- .version = {1, 7, 0},
+ .version = {1, 7, 1},
.module = THIS_MODULE,
.ctr = origin_ctr,
.dtr = origin_dtr,
@@ -2207,13 +2166,12 @@
static struct target_type snapshot_target = {
.name = "snapshot",
- .version = {1, 9, 0},
+ .version = {1, 10, 0},
.module = THIS_MODULE,
.ctr = snapshot_ctr,
.dtr = snapshot_dtr,
.map = snapshot_map,
.end_io = snapshot_end_io,
- .postsuspend = snapshot_postsuspend,
.preresume = snapshot_preresume,
.resume = snapshot_resume,
.status = snapshot_status,
@@ -2222,14 +2180,13 @@
static struct target_type merge_target = {
.name = dm_snapshot_merge_target_name,
- .version = {1, 0, 0},
+ .version = {1, 1, 0},
.module = THIS_MODULE,
.ctr = snapshot_ctr,
.dtr = snapshot_dtr,
.map = snapshot_merge_map,
.end_io = snapshot_end_io,
.presuspend = snapshot_merge_presuspend,
- .postsuspend = snapshot_postsuspend,
.preresume = snapshot_preresume,
.resume = snapshot_merge_resume,
.status = snapshot_status,
@@ -2291,17 +2248,8 @@
goto bad_tracked_chunk_cache;
}
- ksnapd = create_singlethread_workqueue("ksnapd");
- if (!ksnapd) {
- DMERR("Failed to create ksnapd workqueue.");
- r = -ENOMEM;
- goto bad_pending_pool;
- }
-
return 0;
-bad_pending_pool:
- kmem_cache_destroy(tracked_chunk_cache);
bad_tracked_chunk_cache:
kmem_cache_destroy(pending_cache);
bad_pending_cache:
@@ -2322,8 +2270,6 @@
static void __exit dm_snapshot_exit(void)
{
- destroy_workqueue(ksnapd);
-
dm_unregister_target(&snapshot_target);
dm_unregister_target(&origin_target);
dm_unregister_target(&merge_target);
diff --git a/drivers/md/dm-stripe.c b/drivers/md/dm-stripe.c
index f0371b4..dddfa14 100644
--- a/drivers/md/dm-stripe.c
+++ b/drivers/md/dm-stripe.c
@@ -39,23 +39,20 @@
struct dm_target *ti;
/* Work struct used for triggering events*/
- struct work_struct kstriped_ws;
+ struct work_struct trigger_event;
struct stripe stripe[0];
};
-static struct workqueue_struct *kstriped;
-
/*
* An event is triggered whenever a drive
* drops out of a stripe volume.
*/
static void trigger_event(struct work_struct *work)
{
- struct stripe_c *sc = container_of(work, struct stripe_c, kstriped_ws);
-
+ struct stripe_c *sc = container_of(work, struct stripe_c,
+ trigger_event);
dm_table_event(sc->ti->table);
-
}
static inline struct stripe_c *alloc_context(unsigned int stripes)
@@ -160,7 +157,7 @@
return -ENOMEM;
}
- INIT_WORK(&sc->kstriped_ws, trigger_event);
+ INIT_WORK(&sc->trigger_event, trigger_event);
/* Set pointer to dm target; used in trigger_event */
sc->ti = ti;
@@ -211,7 +208,7 @@
for (i = 0; i < sc->stripes; i++)
dm_put_device(ti, sc->stripe[i].dev);
- flush_workqueue(kstriped);
+ flush_work_sync(&sc->trigger_event);
kfree(sc);
}
@@ -367,7 +364,7 @@
atomic_inc(&(sc->stripe[i].error_count));
if (atomic_read(&(sc->stripe[i].error_count)) <
DM_IO_ERROR_THRESHOLD)
- queue_work(kstriped, &sc->kstriped_ws);
+ schedule_work(&sc->trigger_event);
}
return error;
@@ -401,7 +398,7 @@
static struct target_type stripe_target = {
.name = "striped",
- .version = {1, 3, 0},
+ .version = {1, 3, 1},
.module = THIS_MODULE,
.ctr = stripe_ctr,
.dtr = stripe_dtr,
@@ -422,20 +419,10 @@
return r;
}
- kstriped = create_singlethread_workqueue("kstriped");
- if (!kstriped) {
- DMERR("failed to create workqueue kstriped");
- dm_unregister_target(&stripe_target);
- return -ENOMEM;
- }
-
return r;
}
void dm_stripe_exit(void)
{
dm_unregister_target(&stripe_target);
- destroy_workqueue(kstriped);
-
- return;
}
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 985c20a..38e4eb1 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -71,6 +71,8 @@
void *event_context;
struct dm_md_mempools *mempools;
+
+ struct list_head target_callbacks;
};
/*
@@ -204,6 +206,7 @@
return -ENOMEM;
INIT_LIST_HEAD(&t->devices);
+ INIT_LIST_HEAD(&t->target_callbacks);
atomic_set(&t->holders, 0);
t->discards_supported = 1;
@@ -347,6 +350,7 @@
if (!d->dm_dev.bdev)
return;
+ bd_unlink_disk_holder(d->dm_dev.bdev, dm_disk(md));
blkdev_put(d->dm_dev.bdev, d->dm_dev.mode | FMODE_EXCL);
d->dm_dev.bdev = NULL;
}
@@ -1225,10 +1229,17 @@
return 0;
}
+void dm_table_add_target_callbacks(struct dm_table *t, struct dm_target_callbacks *cb)
+{
+ list_add(&cb->list, &t->target_callbacks);
+}
+EXPORT_SYMBOL_GPL(dm_table_add_target_callbacks);
+
int dm_table_any_congested(struct dm_table *t, int bdi_bits)
{
struct dm_dev_internal *dd;
struct list_head *devices = dm_table_get_devices(t);
+ struct dm_target_callbacks *cb;
int r = 0;
list_for_each_entry(dd, devices, list) {
@@ -1243,6 +1254,10 @@
bdevname(dd->dm_dev.bdev, b));
}
+ list_for_each_entry(cb, &t->target_callbacks, list)
+ if (cb->congested_fn)
+ r |= cb->congested_fn(cb, bdi_bits);
+
return r;
}
@@ -1264,6 +1279,7 @@
{
struct dm_dev_internal *dd;
struct list_head *devices = dm_table_get_devices(t);
+ struct dm_target_callbacks *cb;
list_for_each_entry(dd, devices, list) {
struct request_queue *q = bdev_get_queue(dd->dm_dev.bdev);
@@ -1276,6 +1292,10 @@
dm_device_name(t->md),
bdevname(dd->dm_dev.bdev, b));
}
+
+ list_for_each_entry(cb, &t->target_callbacks, list)
+ if (cb->unplug_fn)
+ cb->unplug_fn(cb);
}
struct mapped_device *dm_table_get_md(struct dm_table *t)
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index f48a2f3..eaa3af0 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -32,7 +32,6 @@
#define DM_COOKIE_ENV_VAR_NAME "DM_COOKIE"
#define DM_COOKIE_LENGTH 24
-static DEFINE_MUTEX(dm_mutex);
static const char *_name = DM_NAME;
static unsigned int major = 0;
@@ -328,7 +327,6 @@
{
struct mapped_device *md;
- mutex_lock(&dm_mutex);
spin_lock(&_minor_lock);
md = bdev->bd_disk->private_data;
@@ -346,7 +344,6 @@
out:
spin_unlock(&_minor_lock);
- mutex_unlock(&dm_mutex);
return md ? 0 : -ENXIO;
}
@@ -355,10 +352,12 @@
{
struct mapped_device *md = disk->private_data;
- mutex_lock(&dm_mutex);
+ spin_lock(&_minor_lock);
+
atomic_dec(&md->open_count);
dm_put(md);
- mutex_unlock(&dm_mutex);
+
+ spin_unlock(&_minor_lock);
return 0;
}
@@ -1638,13 +1637,15 @@
if (map_request(ti, clone, md))
goto requeued;
- spin_lock_irq(q->queue_lock);
+ BUG_ON(!irqs_disabled());
+ spin_lock(q->queue_lock);
}
goto out;
requeued:
- spin_lock_irq(q->queue_lock);
+ BUG_ON(!irqs_disabled());
+ spin_lock(q->queue_lock);
plug_and_out:
if (!elv_queue_empty(q))
@@ -1884,7 +1885,8 @@
add_disk(md->disk);
format_dev_t(md->name, MKDEV(_major, minor));
- md->wq = create_singlethread_workqueue("kdmflush");
+ md->wq = alloc_workqueue("kdmflush",
+ WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0);
if (!md->wq)
goto bad_thread;
@@ -1992,13 +1994,14 @@
wake_up(&md->eventq);
}
+/*
+ * Protected by md->suspend_lock obtained by dm_swap_table().
+ */
static void __set_size(struct mapped_device *md, sector_t size)
{
set_capacity(md->disk, size);
- mutex_lock(&md->bdev->bd_inode->i_mutex);
i_size_write(md->bdev->bd_inode, (loff_t)size << SECTOR_SHIFT);
- mutex_unlock(&md->bdev->bd_inode->i_mutex);
}
/*
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 7fc090a..b76cfc8 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -288,10 +288,12 @@
int rv;
int cpu;
- if (mddev == NULL || mddev->pers == NULL) {
+ if (mddev == NULL || mddev->pers == NULL
+ || !mddev->ready) {
bio_io_error(bio);
return 0;
}
+ smp_rmb(); /* Ensure implications of 'active' are visible */
rcu_read_lock();
if (mddev->suspended) {
DEFINE_WAIT(__wait);
@@ -703,9 +705,9 @@
}
/* return the offset of the super block in 512byte sectors */
-static inline sector_t calc_dev_sboffset(struct block_device *bdev)
+static inline sector_t calc_dev_sboffset(mdk_rdev_t *rdev)
{
- sector_t num_sectors = i_size_read(bdev->bd_inode) / 512;
+ sector_t num_sectors = i_size_read(rdev->bdev->bd_inode) / 512;
return MD_NEW_SIZE_SECTORS(num_sectors);
}
@@ -763,7 +765,7 @@
*/
struct bio *bio = bio_alloc_mddev(GFP_NOIO, 1, mddev);
- bio->bi_bdev = rdev->bdev;
+ bio->bi_bdev = rdev->meta_bdev ? rdev->meta_bdev : rdev->bdev;
bio->bi_sector = sector;
bio_add_page(bio, page, size, 0);
bio->bi_private = rdev;
@@ -793,7 +795,7 @@
}
int sync_page_io(mdk_rdev_t *rdev, sector_t sector, int size,
- struct page *page, int rw)
+ struct page *page, int rw, bool metadata_op)
{
struct bio *bio = bio_alloc_mddev(GFP_NOIO, 1, rdev->mddev);
struct completion event;
@@ -801,8 +803,12 @@
rw |= REQ_SYNC | REQ_UNPLUG;
- bio->bi_bdev = rdev->bdev;
- bio->bi_sector = sector;
+ bio->bi_bdev = (metadata_op && rdev->meta_bdev) ?
+ rdev->meta_bdev : rdev->bdev;
+ if (metadata_op)
+ bio->bi_sector = sector + rdev->sb_start;
+ else
+ bio->bi_sector = sector + rdev->data_offset;
bio_add_page(bio, page, size, 0);
init_completion(&event);
bio->bi_private = &event;
@@ -827,7 +833,7 @@
return 0;
- if (!sync_page_io(rdev, rdev->sb_start, size, rdev->sb_page, READ))
+ if (!sync_page_io(rdev, 0, size, rdev->sb_page, READ, true))
goto fail;
rdev->sb_loaded = 1;
return 0;
@@ -989,7 +995,7 @@
*
* It also happens to be a multiple of 4Kb.
*/
- rdev->sb_start = calc_dev_sboffset(rdev->bdev);
+ rdev->sb_start = calc_dev_sboffset(rdev);
ret = read_disk_sb(rdev, MD_SB_BYTES);
if (ret) return ret;
@@ -1330,7 +1336,7 @@
return 0; /* component must fit device */
if (rdev->mddev->bitmap_info.offset)
return 0; /* can't move bitmap */
- rdev->sb_start = calc_dev_sboffset(rdev->bdev);
+ rdev->sb_start = calc_dev_sboffset(rdev);
if (!num_sectors || num_sectors > rdev->sb_start)
num_sectors = rdev->sb_start;
md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size,
@@ -1906,6 +1912,7 @@
MD_BUG();
return;
}
+ bd_unlink_disk_holder(rdev->bdev, rdev->mddev->gendisk);
list_del_rcu(&rdev->same_set);
printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b));
rdev->mddev = NULL;
@@ -2465,6 +2472,10 @@
if (rdev2->raid_disk == slot)
return -EEXIST;
+ if (slot >= rdev->mddev->raid_disks &&
+ slot >= rdev->mddev->raid_disks + rdev->mddev->delta_disks)
+ return -ENOSPC;
+
rdev->raid_disk = slot;
if (test_bit(In_sync, &rdev->flags))
rdev->saved_raid_disk = slot;
@@ -2482,7 +2493,8 @@
/* failure here is OK */;
/* don't wakeup anyone, leave that to userspace. */
} else {
- if (slot >= rdev->mddev->raid_disks)
+ if (slot >= rdev->mddev->raid_disks &&
+ slot >= rdev->mddev->raid_disks + rdev->mddev->delta_disks)
return -ENOSPC;
rdev->raid_disk = slot;
/* assume it is working */
@@ -3107,7 +3119,7 @@
char nm[20];
if (rdev->raid_disk < 0)
continue;
- if (rdev->new_raid_disk > mddev->raid_disks)
+ if (rdev->new_raid_disk >= mddev->raid_disks)
rdev->new_raid_disk = -1;
if (rdev->new_raid_disk == rdev->raid_disk)
continue;
@@ -3736,6 +3748,8 @@
return sprintf(page, "%s\n", type);
}
+static void reap_sync_thread(mddev_t *mddev);
+
static ssize_t
action_store(mddev_t *mddev, const char *page, size_t len)
{
@@ -3750,9 +3764,7 @@
if (cmd_match(page, "idle") || cmd_match(page, "frozen")) {
if (mddev->sync_thread) {
set_bit(MD_RECOVERY_INTR, &mddev->recovery);
- md_unregister_thread(mddev->sync_thread);
- mddev->sync_thread = NULL;
- mddev->recovery = 0;
+ reap_sync_thread(mddev);
}
} else if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) ||
test_bit(MD_RECOVERY_NEEDED, &mddev->recovery))
@@ -3904,7 +3916,7 @@
static ssize_t
sync_completed_show(mddev_t *mddev, char *page)
{
- unsigned long max_sectors, resync;
+ unsigned long long max_sectors, resync;
if (!test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
return sprintf(page, "none\n");
@@ -3915,7 +3927,7 @@
max_sectors = mddev->dev_sectors;
resync = mddev->curr_resync_completed;
- return sprintf(page, "%lu / %lu\n", resync, max_sectors);
+ return sprintf(page, "%llu / %llu\n", resync, max_sectors);
}
static struct md_sysfs_entry md_sync_completed = __ATTR_RO(sync_completed);
@@ -4002,19 +4014,24 @@
{
char *e;
unsigned long long new = simple_strtoull(buf, &e, 10);
+ unsigned long long old = mddev->suspend_lo;
if (mddev->pers == NULL ||
mddev->pers->quiesce == NULL)
return -EINVAL;
if (buf == e || (*e && *e != '\n'))
return -EINVAL;
- if (new >= mddev->suspend_hi ||
- (new > mddev->suspend_lo && new < mddev->suspend_hi)) {
- mddev->suspend_lo = new;
+
+ mddev->suspend_lo = new;
+ if (new >= old)
+ /* Shrinking suspended region */
mddev->pers->quiesce(mddev, 2);
- return len;
- } else
- return -EINVAL;
+ else {
+ /* Expanding suspended region - need to wait */
+ mddev->pers->quiesce(mddev, 1);
+ mddev->pers->quiesce(mddev, 0);
+ }
+ return len;
}
static struct md_sysfs_entry md_suspend_lo =
__ATTR(suspend_lo, S_IRUGO|S_IWUSR, suspend_lo_show, suspend_lo_store);
@@ -4031,20 +4048,24 @@
{
char *e;
unsigned long long new = simple_strtoull(buf, &e, 10);
+ unsigned long long old = mddev->suspend_hi;
if (mddev->pers == NULL ||
mddev->pers->quiesce == NULL)
return -EINVAL;
if (buf == e || (*e && *e != '\n'))
return -EINVAL;
- if ((new <= mddev->suspend_lo && mddev->suspend_lo >= mddev->suspend_hi) ||
- (new > mddev->suspend_lo && new > mddev->suspend_hi)) {
- mddev->suspend_hi = new;
+
+ mddev->suspend_hi = new;
+ if (new <= old)
+ /* Shrinking suspended region */
+ mddev->pers->quiesce(mddev, 2);
+ else {
+ /* Expanding suspended region - need to wait */
mddev->pers->quiesce(mddev, 1);
mddev->pers->quiesce(mddev, 0);
- return len;
- } else
- return -EINVAL;
+ }
+ return len;
}
static struct md_sysfs_entry md_suspend_hi =
__ATTR(suspend_hi, S_IRUGO|S_IWUSR, suspend_hi_show, suspend_hi_store);
@@ -4422,7 +4443,9 @@
* We don't want the data to overlap the metadata,
* Internal Bitmap issues have been handled elsewhere.
*/
- if (rdev->data_offset < rdev->sb_start) {
+ if (rdev->meta_bdev) {
+ /* Nothing to check */;
+ } else if (rdev->data_offset < rdev->sb_start) {
if (mddev->dev_sectors &&
rdev->data_offset + mddev->dev_sectors
> rdev->sb_start) {
@@ -4556,7 +4579,8 @@
mddev->safemode_timer.data = (unsigned long) mddev;
mddev->safemode_delay = (200 * HZ)/1000 +1; /* 200 msec delay */
mddev->in_sync = 1;
-
+ smp_wmb();
+ mddev->ready = 1;
list_for_each_entry(rdev, &mddev->disks, same_set)
if (rdev->raid_disk >= 0) {
char nm[20];
@@ -4693,13 +4717,12 @@
mddev->plug = NULL;
}
-void md_stop_writes(mddev_t *mddev)
+static void __md_stop_writes(mddev_t *mddev)
{
if (mddev->sync_thread) {
set_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
set_bit(MD_RECOVERY_INTR, &mddev->recovery);
- md_unregister_thread(mddev->sync_thread);
- mddev->sync_thread = NULL;
+ reap_sync_thread(mddev);
}
del_timer_sync(&mddev->safemode_timer);
@@ -4713,10 +4736,18 @@
md_update_sb(mddev, 1);
}
}
+
+void md_stop_writes(mddev_t *mddev)
+{
+ mddev_lock(mddev);
+ __md_stop_writes(mddev);
+ mddev_unlock(mddev);
+}
EXPORT_SYMBOL_GPL(md_stop_writes);
void md_stop(mddev_t *mddev)
{
+ mddev->ready = 0;
mddev->pers->stop(mddev);
if (mddev->pers->sync_request && mddev->to_remove == NULL)
mddev->to_remove = &md_redundancy_group;
@@ -4736,7 +4767,7 @@
goto out;
}
if (mddev->pers) {
- md_stop_writes(mddev);
+ __md_stop_writes(mddev);
err = -ENXIO;
if (mddev->ro==1)
@@ -4773,7 +4804,7 @@
if (mddev->ro)
set_disk_ro(disk, 0);
- md_stop_writes(mddev);
+ __md_stop_writes(mddev);
md_stop(mddev);
mddev->queue->merge_bvec_fn = NULL;
mddev->queue->unplug_fn = NULL;
@@ -5151,9 +5182,10 @@
/* set saved_raid_disk if appropriate */
if (!mddev->persistent) {
if (info->state & (1<<MD_DISK_SYNC) &&
- info->raid_disk < mddev->raid_disks)
+ info->raid_disk < mddev->raid_disks) {
rdev->raid_disk = info->raid_disk;
- else
+ set_bit(In_sync, &rdev->flags);
+ } else
rdev->raid_disk = -1;
} else
super_types[mddev->major_version].
@@ -5230,7 +5262,7 @@
printk(KERN_INFO "md: nonpersistent superblock ...\n");
rdev->sb_start = i_size_read(rdev->bdev->bd_inode) / 512;
} else
- rdev->sb_start = calc_dev_sboffset(rdev->bdev);
+ rdev->sb_start = calc_dev_sboffset(rdev);
rdev->sectors = rdev->sb_start;
err = bind_rdev_to_array(rdev, mddev);
@@ -5297,7 +5329,7 @@
}
if (mddev->persistent)
- rdev->sb_start = calc_dev_sboffset(rdev->bdev);
+ rdev->sb_start = calc_dev_sboffset(rdev);
else
rdev->sb_start = i_size_read(rdev->bdev->bd_inode) / 512;
@@ -5510,7 +5542,6 @@
* sb_start or, if that is <data_offset, it must fit before the size
* of each device. If num_sectors is zero, we find the largest size
* that fits.
-
*/
if (mddev->sync_thread)
return -EBUSY;
@@ -6033,7 +6064,8 @@
|| kthread_should_stop(),
thread->timeout);
- if (test_and_clear_bit(THREAD_WAKEUP, &thread->flags))
+ clear_bit(THREAD_WAKEUP, &thread->flags);
+ if (!kthread_should_stop())
thread->run(thread->mddev);
}
@@ -6799,7 +6831,7 @@
desc, mdname(mddev));
mddev->curr_resync = j;
}
- mddev->curr_resync_completed = mddev->curr_resync;
+ mddev->curr_resync_completed = j;
while (j < max_sectors) {
sector_t sectors;
@@ -6817,8 +6849,7 @@
md_unplug(mddev);
wait_event(mddev->recovery_wait,
atomic_read(&mddev->recovery_active) == 0);
- mddev->curr_resync_completed =
- mddev->curr_resync;
+ mddev->curr_resync_completed = j;
set_bit(MD_CHANGE_CLEAN, &mddev->flags);
sysfs_notify(&mddev->kobj, NULL, "sync_completed");
}
@@ -7023,6 +7054,45 @@
}
return spares;
}
+
+static void reap_sync_thread(mddev_t *mddev)
+{
+ mdk_rdev_t *rdev;
+
+ /* resync has finished, collect result */
+ md_unregister_thread(mddev->sync_thread);
+ mddev->sync_thread = NULL;
+ if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery) &&
+ !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
+ /* success...*/
+ /* activate any spares */
+ if (mddev->pers->spare_active(mddev))
+ sysfs_notify(&mddev->kobj, NULL,
+ "degraded");
+ }
+ if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&
+ mddev->pers->finish_reshape)
+ mddev->pers->finish_reshape(mddev);
+ md_update_sb(mddev, 1);
+
+ /* if array is no-longer degraded, then any saved_raid_disk
+ * information must be scrapped
+ */
+ if (!mddev->degraded)
+ list_for_each_entry(rdev, &mddev->disks, same_set)
+ rdev->saved_raid_disk = -1;
+
+ clear_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
+ clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
+ clear_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
+ clear_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
+ clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
+ /* flag recovery needed just to double check */
+ set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
+ sysfs_notify_dirent_safe(mddev->sysfs_action);
+ md_new_event(mddev);
+}
+
/*
* This routine is regularly called by all per-raid-array threads to
* deal with generic issues like resync and super-block update.
@@ -7047,9 +7117,6 @@
*/
void md_check_recovery(mddev_t *mddev)
{
- mdk_rdev_t *rdev;
-
-
if (mddev->bitmap)
bitmap_daemon_work(mddev);
@@ -7117,34 +7184,7 @@
goto unlock;
}
if (mddev->sync_thread) {
- /* resync has finished, collect result */
- md_unregister_thread(mddev->sync_thread);
- mddev->sync_thread = NULL;
- if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery) &&
- !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
- /* success...*/
- /* activate any spares */
- if (mddev->pers->spare_active(mddev))
- sysfs_notify(&mddev->kobj, NULL,
- "degraded");
- }
- if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&
- mddev->pers->finish_reshape)
- mddev->pers->finish_reshape(mddev);
- md_update_sb(mddev, 1);
-
- /* if array is no-longer degraded, then any saved_raid_disk
- * information must be scrapped
- */
- if (!mddev->degraded)
- list_for_each_entry(rdev, &mddev->disks, same_set)
- rdev->saved_raid_disk = -1;
-
- mddev->recovery = 0;
- /* flag recovery needed just to double check */
- set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
- sysfs_notify_dirent_safe(mddev->sysfs_action);
- md_new_event(mddev);
+ reap_sync_thread(mddev);
goto unlock;
}
/* Set RUNNING before clearing NEEDED to avoid
@@ -7202,7 +7242,11 @@
" thread...\n",
mdname(mddev));
/* leave the spares where they are, it shouldn't hurt */
- mddev->recovery = 0;
+ clear_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
+ clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
+ clear_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
+ clear_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
+ clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
} else
md_wakeup_thread(mddev->sync_thread);
sysfs_notify_dirent_safe(mddev->sysfs_action);
diff --git a/drivers/md/md.h b/drivers/md/md.h
index d05bab5..eec517c 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -60,6 +60,12 @@
mddev_t *mddev; /* RAID array if running */
int last_events; /* IO event timestamp */
+ /*
+ * If meta_bdev is non-NULL, it means that a separate device is
+ * being used to store the metadata (superblock/bitmap) which
+ * would otherwise be contained on the same device as the data (bdev).
+ */
+ struct block_device *meta_bdev;
struct block_device *bdev; /* block device handle */
struct page *sb_page;
@@ -148,7 +154,8 @@
* are happening, so run/
* takeover/stop are not safe
*/
-
+ int ready; /* See when safe to pass
+ * IO requests down */
struct gendisk *gendisk;
struct kobject kobj;
@@ -497,8 +504,8 @@
extern void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev,
sector_t sector, int size, struct page *page);
extern void md_super_wait(mddev_t *mddev);
-extern int sync_page_io(mdk_rdev_t *rdev, sector_t sector, int size,
- struct page *page, int rw);
+extern int sync_page_io(mdk_rdev_t *rdev, sector_t sector, int size,
+ struct page *page, int rw, bool metadata_op);
extern void md_do_sync(mddev_t *mddev);
extern void md_new_event(mddev_t *mddev);
extern int md_allow_write(mddev_t *mddev);
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 845cf95..a23ffa3 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1027,8 +1027,9 @@
} else
set_bit(Faulty, &rdev->flags);
set_bit(MD_CHANGE_DEVS, &mddev->flags);
- printk(KERN_ALERT "md/raid1:%s: Disk failure on %s, disabling device.\n"
- KERN_ALERT "md/raid1:%s: Operation continuing on %d devices.\n",
+ printk(KERN_ALERT
+ "md/raid1:%s: Disk failure on %s, disabling device.\n"
+ "md/raid1:%s: Operation continuing on %d devices.\n",
mdname(mddev), bdevname(rdev->bdev, b),
mdname(mddev), conf->raid_disks - mddev->degraded);
}
@@ -1364,10 +1365,10 @@
*/
rdev = conf->mirrors[d].rdev;
if (sync_page_io(rdev,
- sect + rdev->data_offset,
+ sect,
s<<9,
bio->bi_io_vec[idx].bv_page,
- READ)) {
+ READ, false)) {
success = 1;
break;
}
@@ -1390,10 +1391,10 @@
rdev = conf->mirrors[d].rdev;
atomic_add(s, &rdev->corrected_errors);
if (sync_page_io(rdev,
- sect + rdev->data_offset,
+ sect,
s<<9,
bio->bi_io_vec[idx].bv_page,
- WRITE) == 0)
+ WRITE, false) == 0)
md_error(mddev, rdev);
}
d = start;
@@ -1405,10 +1406,10 @@
continue;
rdev = conf->mirrors[d].rdev;
if (sync_page_io(rdev,
- sect + rdev->data_offset,
+ sect,
s<<9,
bio->bi_io_vec[idx].bv_page,
- READ) == 0)
+ READ, false) == 0)
md_error(mddev, rdev);
}
} else {
@@ -1488,10 +1489,8 @@
rdev = conf->mirrors[d].rdev;
if (rdev &&
test_bit(In_sync, &rdev->flags) &&
- sync_page_io(rdev,
- sect + rdev->data_offset,
- s<<9,
- conf->tmppage, READ))
+ sync_page_io(rdev, sect, s<<9,
+ conf->tmppage, READ, false))
success = 1;
else {
d++;
@@ -1514,9 +1513,8 @@
rdev = conf->mirrors[d].rdev;
if (rdev &&
test_bit(In_sync, &rdev->flags)) {
- if (sync_page_io(rdev,
- sect + rdev->data_offset,
- s<<9, conf->tmppage, WRITE)
+ if (sync_page_io(rdev, sect, s<<9,
+ conf->tmppage, WRITE, false)
== 0)
/* Well, this device is dead */
md_error(mddev, rdev);
@@ -1531,9 +1529,8 @@
rdev = conf->mirrors[d].rdev;
if (rdev &&
test_bit(In_sync, &rdev->flags)) {
- if (sync_page_io(rdev,
- sect + rdev->data_offset,
- s<<9, conf->tmppage, READ)
+ if (sync_page_io(rdev, sect, s<<9,
+ conf->tmppage, READ, false)
== 0)
/* Well, this device is dead */
md_error(mddev, rdev);
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 0641674..69b6595 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1051,8 +1051,9 @@
}
set_bit(Faulty, &rdev->flags);
set_bit(MD_CHANGE_DEVS, &mddev->flags);
- printk(KERN_ALERT "md/raid10:%s: Disk failure on %s, disabling device.\n"
- KERN_ALERT "md/raid10:%s: Operation continuing on %d devices.\n",
+ printk(KERN_ALERT
+ "md/raid10:%s: Disk failure on %s, disabling device.\n"
+ "md/raid10:%s: Operation continuing on %d devices.\n",
mdname(mddev), bdevname(rdev->bdev, b),
mdname(mddev), conf->raid_disks - mddev->degraded);
}
@@ -1559,9 +1560,9 @@
rcu_read_unlock();
success = sync_page_io(rdev,
r10_bio->devs[sl].addr +
- sect + rdev->data_offset,
+ sect,
s<<9,
- conf->tmppage, READ);
+ conf->tmppage, READ, false);
rdev_dec_pending(rdev, mddev);
rcu_read_lock();
if (success)
@@ -1598,8 +1599,8 @@
atomic_add(s, &rdev->corrected_errors);
if (sync_page_io(rdev,
r10_bio->devs[sl].addr +
- sect + rdev->data_offset,
- s<<9, conf->tmppage, WRITE)
+ sect,
+ s<<9, conf->tmppage, WRITE, false)
== 0) {
/* Well, this device is dead */
printk(KERN_NOTICE
@@ -1635,9 +1636,9 @@
rcu_read_unlock();
if (sync_page_io(rdev,
r10_bio->devs[sl].addr +
- sect + rdev->data_offset,
+ sect,
s<<9, conf->tmppage,
- READ) == 0) {
+ READ, false) == 0) {
/* Well, this device is dead */
printk(KERN_NOTICE
"md/raid10:%s: unable to read back "
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index dc574f3..5044bab 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -1721,7 +1721,6 @@
set_bit(Faulty, &rdev->flags);
printk(KERN_ALERT
"md/raid:%s: Disk failure on %s, disabling device.\n"
- KERN_ALERT
"md/raid:%s: Operation continuing on %d devices.\n",
mdname(mddev),
bdevname(rdev->bdev, b),
@@ -4237,7 +4236,7 @@
wait_event(conf->wait_for_overlap,
atomic_read(&conf->reshape_stripes)==0);
mddev->reshape_position = conf->reshape_progress;
- mddev->curr_resync_completed = mddev->curr_resync;
+ mddev->curr_resync_completed = sector_nr;
conf->reshape_checkpoint = jiffies;
set_bit(MD_CHANGE_DEVS, &mddev->flags);
md_wakeup_thread(mddev->thread);
@@ -4338,7 +4337,7 @@
wait_event(conf->wait_for_overlap,
atomic_read(&conf->reshape_stripes) == 0);
mddev->reshape_position = conf->reshape_progress;
- mddev->curr_resync_completed = mddev->curr_resync + reshape_sectors;
+ mddev->curr_resync_completed = sector_nr;
conf->reshape_checkpoint = jiffies;
set_bit(MD_CHANGE_DEVS, &mddev->flags);
md_wakeup_thread(mddev->thread);
@@ -5339,7 +5338,7 @@
&& !test_bit(Faulty, &tmp->rdev->flags)
&& !test_and_set_bit(In_sync, &tmp->rdev->flags)) {
count++;
- sysfs_notify_dirent(tmp->rdev->sysfs_state);
+ sysfs_notify_dirent_safe(tmp->rdev->sysfs_state);
}
}
spin_lock_irqsave(&conf->device_lock, flags);
@@ -5528,8 +5527,8 @@
return -ENOSPC;
list_for_each_entry(rdev, &mddev->disks, same_set)
- if (rdev->raid_disk < 0 &&
- !test_bit(Faulty, &rdev->flags))
+ if ((rdev->raid_disk < 0 || rdev->raid_disk >= conf->raid_disks)
+ && !test_bit(Faulty, &rdev->flags))
spares++;
if (spares - mddev->degraded < mddev->delta_disks - conf->max_degraded)
@@ -5589,6 +5588,11 @@
/* Failure here is OK */;
} else
break;
+ } else if (rdev->raid_disk >= conf->previous_raid_disks
+ && !test_bit(Faulty, &rdev->flags)) {
+ /* This is a spare that was manually added */
+ set_bit(In_sync, &rdev->flags);
+ added_devices++;
}
/* When a reshape changes the number of devices, ->degraded
diff --git a/drivers/media/video/cafe_ccic.c b/drivers/media/video/cafe_ccic.c
index 789087c..49f1b8f 100644
--- a/drivers/media/video/cafe_ccic.c
+++ b/drivers/media/video/cafe_ccic.c
@@ -2184,9 +2184,7 @@
struct cafe_camera *cam = to_cam(v4l2_dev);
int ret = 0;
- ret = pci_restore_state(pdev);
- if (ret)
- return ret;
+ pci_restore_state(pdev);
ret = pci_enable_device(pdev);
if (ret) {
diff --git a/drivers/mfd/88pm860x-core.c b/drivers/mfd/88pm860x-core.c
index 20895e7..793300c 100644
--- a/drivers/mfd/88pm860x-core.c
+++ b/drivers/mfd/88pm860x-core.c
@@ -361,12 +361,6 @@
},
};
-static inline struct pm860x_irq_data *irq_to_pm860x(struct pm860x_chip *chip,
- int irq)
-{
- return &pm860x_irqs[irq - chip->irq_base];
-}
-
static irqreturn_t pm860x_irq(int irq, void *data)
{
struct pm860x_chip *chip = data;
@@ -388,16 +382,16 @@
return IRQ_HANDLED;
}
-static void pm860x_irq_lock(unsigned int irq)
+static void pm860x_irq_lock(struct irq_data *data)
{
- struct pm860x_chip *chip = get_irq_chip_data(irq);
+ struct pm860x_chip *chip = irq_data_get_irq_chip_data(data);
mutex_lock(&chip->irq_lock);
}
-static void pm860x_irq_sync_unlock(unsigned int irq)
+static void pm860x_irq_sync_unlock(struct irq_data *data)
{
- struct pm860x_chip *chip = get_irq_chip_data(irq);
+ struct pm860x_chip *chip = irq_data_get_irq_chip_data(data);
struct pm860x_irq_data *irq_data;
struct i2c_client *i2c;
static unsigned char cached[3] = {0x0, 0x0, 0x0};
@@ -439,25 +433,25 @@
mutex_unlock(&chip->irq_lock);
}
-static void pm860x_irq_enable(unsigned int irq)
+static void pm860x_irq_enable(struct irq_data *data)
{
- struct pm860x_chip *chip = get_irq_chip_data(irq);
- pm860x_irqs[irq - chip->irq_base].enable
- = pm860x_irqs[irq - chip->irq_base].offs;
+ struct pm860x_chip *chip = irq_data_get_irq_chip_data(data);
+ pm860x_irqs[data->irq - chip->irq_base].enable
+ = pm860x_irqs[data->irq - chip->irq_base].offs;
}
-static void pm860x_irq_disable(unsigned int irq)
+static void pm860x_irq_disable(struct irq_data *data)
{
- struct pm860x_chip *chip = get_irq_chip_data(irq);
- pm860x_irqs[irq - chip->irq_base].enable = 0;
+ struct pm860x_chip *chip = irq_data_get_irq_chip_data(data);
+ pm860x_irqs[data->irq - chip->irq_base].enable = 0;
}
static struct irq_chip pm860x_irq_chip = {
.name = "88pm860x",
- .bus_lock = pm860x_irq_lock,
- .bus_sync_unlock = pm860x_irq_sync_unlock,
- .enable = pm860x_irq_enable,
- .disable = pm860x_irq_disable,
+ .irq_bus_lock = pm860x_irq_lock,
+ .irq_bus_sync_unlock = pm860x_irq_sync_unlock,
+ .irq_enable = pm860x_irq_enable,
+ .irq_disable = pm860x_irq_disable,
};
static int __devinit device_gpadc_init(struct pm860x_chip *chip,
diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index da9d297..fd01836 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -496,13 +496,13 @@
config AB8500_CORE
bool "ST-Ericsson AB8500 Mixed Signal Power Management chip"
- depends on GENERIC_HARDIRQS && ABX500_CORE && SPI_MASTER && ARCH_U8500
+ depends on GENERIC_HARDIRQS && ABX500_CORE
select MFD_CORE
help
Select this option to enable access to AB8500 power management
- chip. This connects to U8500 either on the SSP/SPI bus
- or the I2C bus via PRCMU. It also adds the irq_chip
- parts for handling the Mixed Signal chip events.
+ chip. This connects to U8500 either on the SSP/SPI bus (deprecated
+ since hardware version v1.0) or the I2C bus via PRCMU. It also adds
+ the irq_chip parts for handling the Mixed Signal chip events.
This chip embeds various other multimedia funtionalities as well.
config AB8500_I2C_CORE
@@ -537,6 +537,14 @@
LEDs, vibrator, system power and temperature, power management
and ALSA sound.
+config MFD_CS5535
+ tristate "Support for CS5535 and CS5536 southbridge core functions"
+ select MFD_CORE
+ depends on PCI
+ ---help---
+ This is the core driver for CS5535/CS5536 MFD functions. This is
+ necessary for using the board's GPIO and MFGPT functionality.
+
config MFD_TIMBERDALE
tristate "Support for the Timberdale FPGA"
select MFD_CORE
diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
index 848e7ea..a54e2c7 100644
--- a/drivers/mfd/Makefile
+++ b/drivers/mfd/Makefile
@@ -70,7 +70,7 @@
obj-$(CONFIG_AB3100_CORE) += ab3100-core.o
obj-$(CONFIG_AB3100_OTP) += ab3100-otp.o
obj-$(CONFIG_AB3550_CORE) += ab3550-core.o
-obj-$(CONFIG_AB8500_CORE) += ab8500-core.o ab8500-spi.o
+obj-$(CONFIG_AB8500_CORE) += ab8500-core.o
obj-$(CONFIG_AB8500_I2C_CORE) += ab8500-i2c.o
obj-$(CONFIG_AB8500_DEBUG) += ab8500-debugfs.o
obj-$(CONFIG_MFD_TIMBERDALE) += timberdale.o
@@ -82,3 +82,4 @@
obj-$(CONFIG_MFD_TPS6586X) += tps6586x.o
obj-$(CONFIG_MFD_VX855) += vx855.o
obj-$(CONFIG_MFD_WL1273_CORE) += wl1273-core.o
+obj-$(CONFIG_MFD_CS5535) += cs5535-mfd.o
diff --git a/drivers/mfd/ab3550-core.c b/drivers/mfd/ab3550-core.c
index 8a98739..5fbca34 100644
--- a/drivers/mfd/ab3550-core.c
+++ b/drivers/mfd/ab3550-core.c
@@ -1159,15 +1159,16 @@
}
}
-static void ab3550_mask(unsigned int irq)
+static void ab3550_mask(struct irq_data *data)
{
unsigned long flags;
struct ab3550 *ab;
struct ab3550_platform_data *plf_data;
+ int irq;
- ab = get_irq_chip_data(irq);
+ ab = irq_data_get_irq_chip_data(data);
plf_data = ab->i2c_client[0]->dev.platform_data;
- irq -= plf_data->irq.base;
+ irq = data->irq - plf_data->irq.base;
spin_lock_irqsave(&ab->event_lock, flags);
ab->event_mask[irq / 8] |= BIT(irq % 8);
@@ -1176,15 +1177,16 @@
schedule_work(&ab->mask_work);
}
-static void ab3550_unmask(unsigned int irq)
+static void ab3550_unmask(struct irq_data *data)
{
unsigned long flags;
struct ab3550 *ab;
struct ab3550_platform_data *plf_data;
+ int irq;
- ab = get_irq_chip_data(irq);
+ ab = irq_data_get_irq_chip_data(data);
plf_data = ab->i2c_client[0]->dev.platform_data;
- irq -= plf_data->irq.base;
+ irq = data->irq - plf_data->irq.base;
spin_lock_irqsave(&ab->event_lock, flags);
ab->event_mask[irq / 8] &= ~BIT(irq % 8);
@@ -1193,20 +1195,16 @@
schedule_work(&ab->mask_work);
}
-static void noop(unsigned int irq)
+static void noop(struct irq_data *data)
{
}
static struct irq_chip ab3550_irq_chip = {
.name = "ab3550-core", /* Keep the same name as the request */
- .startup = NULL, /* defaults to enable */
- .shutdown = NULL, /* defaults to disable */
- .enable = NULL, /* defaults to unmask */
- .disable = ab3550_mask, /* No default to mask in chip.c */
- .ack = noop,
- .mask = ab3550_mask,
- .unmask = ab3550_unmask,
- .end = NULL,
+ .irq_disable = ab3550_mask, /* No default to mask in chip.c */
+ .irq_ack = noop,
+ .irq_mask = ab3550_mask,
+ .irq_unmask = ab3550_unmask,
};
struct ab_family_id {
diff --git a/drivers/mfd/ab8500-core.c b/drivers/mfd/ab8500-core.c
index d9640a6..b688701 100644
--- a/drivers/mfd/ab8500-core.c
+++ b/drivers/mfd/ab8500-core.c
@@ -52,6 +52,7 @@
#define AB8500_IT_LATCH8_REG 0x27
#define AB8500_IT_LATCH9_REG 0x28
#define AB8500_IT_LATCH10_REG 0x29
+#define AB8500_IT_LATCH12_REG 0x2B
#define AB8500_IT_LATCH19_REG 0x32
#define AB8500_IT_LATCH20_REG 0x33
#define AB8500_IT_LATCH21_REG 0x34
@@ -98,13 +99,17 @@
* offset 0.
*/
static const int ab8500_irq_regoffset[AB8500_NUM_IRQ_REGS] = {
- 0, 1, 2, 3, 4, 6, 7, 8, 9, 18, 19, 20, 21,
+ 0, 1, 2, 3, 4, 6, 7, 8, 9, 11, 18, 19, 20, 21,
};
static int ab8500_get_chip_id(struct device *dev)
{
- struct ab8500 *ab8500 = dev_get_drvdata(dev->parent);
- return (int)ab8500->chip_id;
+ struct ab8500 *ab8500;
+
+ if (!dev)
+ return -EINVAL;
+ ab8500 = dev_get_drvdata(dev->parent);
+ return ab8500 ? (int)ab8500->chip_id : -EINVAL;
}
static int set_register_interruptible(struct ab8500 *ab8500, u8 bank,
@@ -228,16 +233,16 @@
.startup_irq_enabled = NULL,
};
-static void ab8500_irq_lock(unsigned int irq)
+static void ab8500_irq_lock(struct irq_data *data)
{
- struct ab8500 *ab8500 = get_irq_chip_data(irq);
+ struct ab8500 *ab8500 = irq_data_get_irq_chip_data(data);
mutex_lock(&ab8500->irq_lock);
}
-static void ab8500_irq_sync_unlock(unsigned int irq)
+static void ab8500_irq_sync_unlock(struct irq_data *data)
{
- struct ab8500 *ab8500 = get_irq_chip_data(irq);
+ struct ab8500 *ab8500 = irq_data_get_irq_chip_data(data);
int i;
for (i = 0; i < AB8500_NUM_IRQ_REGS; i++) {
@@ -248,6 +253,10 @@
if (new == old)
continue;
+ /* Interrupt register 12 does'nt exist prior to version 0x20 */
+ if (ab8500_irq_regoffset[i] == 11 && ab8500->chip_id < 0x20)
+ continue;
+
ab8500->oldmask[i] = new;
reg = AB8500_IT_MASK1_REG + ab8500_irq_regoffset[i];
@@ -257,20 +266,20 @@
mutex_unlock(&ab8500->irq_lock);
}
-static void ab8500_irq_mask(unsigned int irq)
+static void ab8500_irq_mask(struct irq_data *data)
{
- struct ab8500 *ab8500 = get_irq_chip_data(irq);
- int offset = irq - ab8500->irq_base;
+ struct ab8500 *ab8500 = irq_data_get_irq_chip_data(data);
+ int offset = data->irq - ab8500->irq_base;
int index = offset / 8;
int mask = 1 << (offset % 8);
ab8500->mask[index] |= mask;
}
-static void ab8500_irq_unmask(unsigned int irq)
+static void ab8500_irq_unmask(struct irq_data *data)
{
- struct ab8500 *ab8500 = get_irq_chip_data(irq);
- int offset = irq - ab8500->irq_base;
+ struct ab8500 *ab8500 = irq_data_get_irq_chip_data(data);
+ int offset = data->irq - ab8500->irq_base;
int index = offset / 8;
int mask = 1 << (offset % 8);
@@ -279,10 +288,10 @@
static struct irq_chip ab8500_irq_chip = {
.name = "ab8500",
- .bus_lock = ab8500_irq_lock,
- .bus_sync_unlock = ab8500_irq_sync_unlock,
- .mask = ab8500_irq_mask,
- .unmask = ab8500_irq_unmask,
+ .irq_bus_lock = ab8500_irq_lock,
+ .irq_bus_sync_unlock = ab8500_irq_sync_unlock,
+ .irq_mask = ab8500_irq_mask,
+ .irq_unmask = ab8500_irq_unmask,
};
static irqreturn_t ab8500_irq(int irq, void *dev)
@@ -297,6 +306,10 @@
int status;
u8 value;
+ /* Interrupt register 12 does'nt exist prior to version 0x20 */
+ if (regoffset == 11 && ab8500->chip_id < 0x20)
+ continue;
+
status = get_register_interruptible(ab8500, AB8500_INTERRUPT,
AB8500_IT_LATCH1_REG + regoffset, &value);
if (status < 0 || value == 0)
@@ -393,13 +406,195 @@
},
};
+static struct resource ab8500_bm_resources[] = {
+ {
+ .name = "MAIN_EXT_CH_NOT_OK",
+ .start = AB8500_INT_MAIN_EXT_CH_NOT_OK,
+ .end = AB8500_INT_MAIN_EXT_CH_NOT_OK,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "BATT_OVV",
+ .start = AB8500_INT_BATT_OVV,
+ .end = AB8500_INT_BATT_OVV,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "MAIN_CH_UNPLUG_DET",
+ .start = AB8500_INT_MAIN_CH_UNPLUG_DET,
+ .end = AB8500_INT_MAIN_CH_UNPLUG_DET,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "MAIN_CHARGE_PLUG_DET",
+ .start = AB8500_INT_MAIN_CH_PLUG_DET,
+ .end = AB8500_INT_MAIN_CH_PLUG_DET,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "VBUS_DET_F",
+ .start = AB8500_INT_VBUS_DET_F,
+ .end = AB8500_INT_VBUS_DET_F,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "VBUS_DET_R",
+ .start = AB8500_INT_VBUS_DET_R,
+ .end = AB8500_INT_VBUS_DET_R,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "BAT_CTRL_INDB",
+ .start = AB8500_INT_BAT_CTRL_INDB,
+ .end = AB8500_INT_BAT_CTRL_INDB,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "CH_WD_EXP",
+ .start = AB8500_INT_CH_WD_EXP,
+ .end = AB8500_INT_CH_WD_EXP,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "VBUS_OVV",
+ .start = AB8500_INT_VBUS_OVV,
+ .end = AB8500_INT_VBUS_OVV,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "NCONV_ACCU",
+ .start = AB8500_INT_CCN_CONV_ACC,
+ .end = AB8500_INT_CCN_CONV_ACC,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "LOW_BAT_F",
+ .start = AB8500_INT_LOW_BAT_F,
+ .end = AB8500_INT_LOW_BAT_F,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "LOW_BAT_R",
+ .start = AB8500_INT_LOW_BAT_R,
+ .end = AB8500_INT_LOW_BAT_R,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "BTEMP_LOW",
+ .start = AB8500_INT_BTEMP_LOW,
+ .end = AB8500_INT_BTEMP_LOW,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "BTEMP_HIGH",
+ .start = AB8500_INT_BTEMP_HIGH,
+ .end = AB8500_INT_BTEMP_HIGH,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "USB_CHARGER_NOT_OKR",
+ .start = AB8500_INT_USB_CHARGER_NOT_OK,
+ .end = AB8500_INT_USB_CHARGER_NOT_OK,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "USB_CHARGE_DET_DONE",
+ .start = AB8500_INT_USB_CHG_DET_DONE,
+ .end = AB8500_INT_USB_CHG_DET_DONE,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "USB_CH_TH_PROT_R",
+ .start = AB8500_INT_USB_CH_TH_PROT_R,
+ .end = AB8500_INT_USB_CH_TH_PROT_R,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "MAIN_CH_TH_PROT_R",
+ .start = AB8500_INT_MAIN_CH_TH_PROT_R,
+ .end = AB8500_INT_MAIN_CH_TH_PROT_R,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "USB_CHARGER_NOT_OKF",
+ .start = AB8500_INT_USB_CHARGER_NOT_OKF,
+ .end = AB8500_INT_USB_CHARGER_NOT_OKF,
+ .flags = IORESOURCE_IRQ,
+ },
+};
+
+static struct resource ab8500_debug_resources[] = {
+ {
+ .name = "IRQ_FIRST",
+ .start = AB8500_INT_MAIN_EXT_CH_NOT_OK,
+ .end = AB8500_INT_MAIN_EXT_CH_NOT_OK,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "IRQ_LAST",
+ .start = AB8500_INT_USB_CHARGER_NOT_OKF,
+ .end = AB8500_INT_USB_CHARGER_NOT_OKF,
+ .flags = IORESOURCE_IRQ,
+ },
+};
+
+static struct resource ab8500_usb_resources[] = {
+ {
+ .name = "ID_WAKEUP_R",
+ .start = AB8500_INT_ID_WAKEUP_R,
+ .end = AB8500_INT_ID_WAKEUP_R,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "ID_WAKEUP_F",
+ .start = AB8500_INT_ID_WAKEUP_F,
+ .end = AB8500_INT_ID_WAKEUP_F,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "VBUS_DET_F",
+ .start = AB8500_INT_VBUS_DET_F,
+ .end = AB8500_INT_VBUS_DET_F,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "VBUS_DET_R",
+ .start = AB8500_INT_VBUS_DET_R,
+ .end = AB8500_INT_VBUS_DET_R,
+ .flags = IORESOURCE_IRQ,
+ },
+ {
+ .name = "USB_LINK_STATUS",
+ .start = AB8500_INT_USB_LINK_STATUS,
+ .end = AB8500_INT_USB_LINK_STATUS,
+ .flags = IORESOURCE_IRQ,
+ },
+};
+
+static struct resource ab8500_temp_resources[] = {
+ {
+ .name = "AB8500_TEMP_WARM",
+ .start = AB8500_INT_TEMP_WARM,
+ .end = AB8500_INT_TEMP_WARM,
+ .flags = IORESOURCE_IRQ,
+ },
+};
+
static struct mfd_cell ab8500_devs[] = {
#ifdef CONFIG_DEBUG_FS
{
.name = "ab8500-debug",
+ .num_resources = ARRAY_SIZE(ab8500_debug_resources),
+ .resources = ab8500_debug_resources,
},
#endif
{
+ .name = "ab8500-sysctrl",
+ },
+ {
+ .name = "ab8500-regulator",
+ },
+ {
.name = "ab8500-gpadc",
.num_resources = ARRAY_SIZE(ab8500_gpadc_resources),
.resources = ab8500_gpadc_resources,
@@ -410,6 +605,22 @@
.resources = ab8500_rtc_resources,
},
{
+ .name = "ab8500-bm",
+ .num_resources = ARRAY_SIZE(ab8500_bm_resources),
+ .resources = ab8500_bm_resources,
+ },
+ { .name = "ab8500-codec", },
+ {
+ .name = "ab8500-usb",
+ .num_resources = ARRAY_SIZE(ab8500_usb_resources),
+ .resources = ab8500_usb_resources,
+ },
+ {
+ .name = "ab8500-poweron-key",
+ .num_resources = ARRAY_SIZE(ab8500_poweronkey_db_resources),
+ .resources = ab8500_poweronkey_db_resources,
+ },
+ {
.name = "ab8500-pwm",
.id = 1,
},
@@ -421,15 +632,35 @@
.name = "ab8500-pwm",
.id = 3,
},
- { .name = "ab8500-charger", },
- { .name = "ab8500-audio", },
- { .name = "ab8500-usb", },
- { .name = "ab8500-regulator", },
+ { .name = "ab8500-leds", },
{
- .name = "ab8500-poweron-key",
- .num_resources = ARRAY_SIZE(ab8500_poweronkey_db_resources),
- .resources = ab8500_poweronkey_db_resources,
+ .name = "ab8500-denc",
},
+ {
+ .name = "ab8500-temp",
+ .num_resources = ARRAY_SIZE(ab8500_temp_resources),
+ .resources = ab8500_temp_resources,
+ },
+};
+
+static ssize_t show_chip_id(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct ab8500 *ab8500;
+
+ ab8500 = dev_get_drvdata(dev);
+ return sprintf(buf, "%#x\n", ab8500 ? ab8500->chip_id : -EINVAL);
+}
+
+static DEVICE_ATTR(chip_id, S_IRUGO, show_chip_id, NULL);
+
+static struct attribute *ab8500_sysfs_entries[] = {
+ &dev_attr_chip_id.attr,
+ NULL,
+};
+
+static struct attribute_group ab8500_attr_group = {
+ .attrs = ab8500_sysfs_entries,
};
int __devinit ab8500_init(struct ab8500 *ab8500)
@@ -454,8 +685,9 @@
* 0x0 - Early Drop
* 0x10 - Cut 1.0
* 0x11 - Cut 1.1
+ * 0x20 - Cut 2.0
*/
- if (value == 0x0 || value == 0x10 || value == 0x11) {
+ if (value == 0x0 || value == 0x10 || value == 0x11 || value == 0x20) {
ab8500->revision = value;
dev_info(ab8500->dev, "detected chip, revision: %#x\n", value);
} else {
@@ -468,18 +700,16 @@
plat->init(ab8500);
/* Clear and mask all interrupts */
- for (i = 0; i < 10; i++) {
- get_register_interruptible(ab8500, AB8500_INTERRUPT,
- AB8500_IT_LATCH1_REG + i, &value);
- set_register_interruptible(ab8500, AB8500_INTERRUPT,
- AB8500_IT_MASK1_REG + i, 0xff);
- }
+ for (i = 0; i < AB8500_NUM_IRQ_REGS; i++) {
+ /* Interrupt register 12 does'nt exist prior to version 0x20 */
+ if (ab8500_irq_regoffset[i] == 11 && ab8500->chip_id < 0x20)
+ continue;
- for (i = 18; i < 24; i++) {
get_register_interruptible(ab8500, AB8500_INTERRUPT,
- AB8500_IT_LATCH1_REG + i, &value);
+ AB8500_IT_LATCH1_REG + ab8500_irq_regoffset[i],
+ &value);
set_register_interruptible(ab8500, AB8500_INTERRUPT,
- AB8500_IT_MASK1_REG + i, 0xff);
+ AB8500_IT_MASK1_REG + ab8500_irq_regoffset[i], 0xff);
}
ret = abx500_register_ops(ab8500->dev, &ab8500_ops);
@@ -495,7 +725,8 @@
return ret;
ret = request_threaded_irq(ab8500->irq, NULL, ab8500_irq,
- IRQF_ONESHOT, "ab8500", ab8500);
+ IRQF_ONESHOT | IRQF_NO_SUSPEND,
+ "ab8500", ab8500);
if (ret)
goto out_removeirq;
}
@@ -506,6 +737,10 @@
if (ret)
goto out_freeirq;
+ ret = sysfs_create_group(&ab8500->dev->kobj, &ab8500_attr_group);
+ if (ret)
+ dev_err(ab8500->dev, "error creating sysfs entries\n");
+
return ret;
out_freeirq:
@@ -519,6 +754,7 @@
int __devexit ab8500_exit(struct ab8500 *ab8500)
{
+ sysfs_remove_group(&ab8500->dev->kobj, &ab8500_attr_group);
mfd_remove_devices(ab8500->dev);
if (ab8500->irq_base) {
free_irq(ab8500->irq, ab8500);
diff --git a/drivers/mfd/ab8500-debugfs.c b/drivers/mfd/ab8500-debugfs.c
index 8d1e05a..3c1541a 100644
--- a/drivers/mfd/ab8500-debugfs.c
+++ b/drivers/mfd/ab8500-debugfs.c
@@ -24,9 +24,9 @@
* @perm: access permissions for the range
*/
struct ab8500_reg_range {
- u8 first;
- u8 last;
- u8 perm;
+ u8 first;
+ u8 last;
+ u8 perm;
};
/**
@@ -36,9 +36,9 @@
* @range: the list of register ranges
*/
struct ab8500_i2c_ranges {
- u8 num_ranges;
- u8 bankid;
- const struct ab8500_reg_range *range;
+ u8 num_ranges;
+ u8 bankid;
+ const struct ab8500_reg_range *range;
};
#define AB8500_NAME_STRING "ab8500"
@@ -47,521 +47,521 @@
#define AB8500_REV_REG 0x80
static struct ab8500_i2c_ranges debug_ranges[AB8500_NUM_BANKS] = {
- [0x0] = {
- .num_ranges = 0,
- .range = 0,
- },
- [AB8500_SYS_CTRL1_BLOCK] = {
- .num_ranges = 3,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x02,
- },
- {
- .first = 0x42,
- .last = 0x42,
- },
- {
- .first = 0x80,
- .last = 0x81,
- },
- },
- },
- [AB8500_SYS_CTRL2_BLOCK] = {
- .num_ranges = 4,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x0D,
- },
- {
- .first = 0x0F,
- .last = 0x17,
- },
- {
- .first = 0x30,
- .last = 0x30,
- },
- {
- .first = 0x32,
- .last = 0x33,
- },
- },
- },
- [AB8500_REGU_CTRL1] = {
- .num_ranges = 3,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x00,
- },
- {
- .first = 0x03,
- .last = 0x10,
- },
- {
- .first = 0x80,
- .last = 0x84,
- },
- },
- },
- [AB8500_REGU_CTRL2] = {
- .num_ranges = 5,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x15,
- },
- {
- .first = 0x17,
- .last = 0x19,
- },
- {
- .first = 0x1B,
- .last = 0x1D,
- },
- {
- .first = 0x1F,
- .last = 0x22,
- },
- {
- .first = 0x40,
- .last = 0x44,
- },
- /* 0x80-0x8B is SIM registers and should
- * not be accessed from here */
- },
- },
- [AB8500_USB] = {
- .num_ranges = 2,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x80,
- .last = 0x83,
- },
- {
- .first = 0x87,
- .last = 0x8A,
- },
- },
- },
- [AB8500_TVOUT] = {
- .num_ranges = 9,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x12,
- },
- {
- .first = 0x15,
- .last = 0x17,
- },
- {
- .first = 0x19,
- .last = 0x21,
- },
- {
- .first = 0x27,
- .last = 0x2C,
- },
- {
- .first = 0x41,
- .last = 0x41,
- },
- {
- .first = 0x45,
- .last = 0x5B,
- },
- {
- .first = 0x5D,
- .last = 0x5D,
- },
- {
- .first = 0x69,
- .last = 0x69,
- },
- {
- .first = 0x80,
- .last = 0x81,
- },
- },
- },
- [AB8500_DBI] = {
- .num_ranges = 0,
- .range = 0,
- },
- [AB8500_ECI_AV_ACC] = {
- .num_ranges = 1,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x80,
- .last = 0x82,
- },
- },
- },
- [0x9] = {
- .num_ranges = 0,
- .range = 0,
- },
- [AB8500_GPADC] = {
- .num_ranges = 1,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x08,
- },
- },
- },
- [AB8500_CHARGER] = {
- .num_ranges = 8,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x03,
- },
- {
- .first = 0x05,
- .last = 0x05,
- },
- {
- .first = 0x40,
- .last = 0x40,
- },
- {
- .first = 0x42,
- .last = 0x42,
- },
- {
- .first = 0x44,
- .last = 0x44,
- },
- {
- .first = 0x50,
- .last = 0x55,
- },
- {
- .first = 0x80,
- .last = 0x82,
- },
- {
- .first = 0xC0,
- .last = 0xC2,
- },
- },
- },
- [AB8500_GAS_GAUGE] = {
- .num_ranges = 3,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x00,
- },
- {
- .first = 0x07,
- .last = 0x0A,
- },
- {
- .first = 0x10,
- .last = 0x14,
- },
- },
- },
- [AB8500_AUDIO] = {
- .num_ranges = 1,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x6F,
- },
- },
- },
- [AB8500_INTERRUPT] = {
- .num_ranges = 0,
- .range = 0,
- },
- [AB8500_RTC] = {
- .num_ranges = 1,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x0F,
- },
- },
- },
- [AB8500_MISC] = {
- .num_ranges = 8,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x00,
- .last = 0x05,
- },
- {
- .first = 0x10,
- .last = 0x15,
- },
- {
- .first = 0x20,
- .last = 0x25,
- },
- {
- .first = 0x30,
- .last = 0x35,
- },
- {
- .first = 0x40,
- .last = 0x45,
- },
- {
- .first = 0x50,
- .last = 0x50,
- },
- {
- .first = 0x60,
- .last = 0x67,
- },
- {
- .first = 0x80,
- .last = 0x80,
- },
- },
- },
- [0x11] = {
- .num_ranges = 0,
- .range = 0,
- },
- [0x12] = {
- .num_ranges = 0,
- .range = 0,
- },
- [0x13] = {
- .num_ranges = 0,
- .range = 0,
- },
- [0x14] = {
- .num_ranges = 0,
- .range = 0,
- },
- [AB8500_OTP_EMUL] = {
- .num_ranges = 1,
- .range = (struct ab8500_reg_range[]) {
- {
- .first = 0x01,
- .last = 0x0F,
- },
- },
- },
+ [0x0] = {
+ .num_ranges = 0,
+ .range = 0,
+ },
+ [AB8500_SYS_CTRL1_BLOCK] = {
+ .num_ranges = 3,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x02,
+ },
+ {
+ .first = 0x42,
+ .last = 0x42,
+ },
+ {
+ .first = 0x80,
+ .last = 0x81,
+ },
+ },
+ },
+ [AB8500_SYS_CTRL2_BLOCK] = {
+ .num_ranges = 4,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x0D,
+ },
+ {
+ .first = 0x0F,
+ .last = 0x17,
+ },
+ {
+ .first = 0x30,
+ .last = 0x30,
+ },
+ {
+ .first = 0x32,
+ .last = 0x33,
+ },
+ },
+ },
+ [AB8500_REGU_CTRL1] = {
+ .num_ranges = 3,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x00,
+ },
+ {
+ .first = 0x03,
+ .last = 0x10,
+ },
+ {
+ .first = 0x80,
+ .last = 0x84,
+ },
+ },
+ },
+ [AB8500_REGU_CTRL2] = {
+ .num_ranges = 5,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x15,
+ },
+ {
+ .first = 0x17,
+ .last = 0x19,
+ },
+ {
+ .first = 0x1B,
+ .last = 0x1D,
+ },
+ {
+ .first = 0x1F,
+ .last = 0x22,
+ },
+ {
+ .first = 0x40,
+ .last = 0x44,
+ },
+ /* 0x80-0x8B is SIM registers and should
+ * not be accessed from here */
+ },
+ },
+ [AB8500_USB] = {
+ .num_ranges = 2,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x80,
+ .last = 0x83,
+ },
+ {
+ .first = 0x87,
+ .last = 0x8A,
+ },
+ },
+ },
+ [AB8500_TVOUT] = {
+ .num_ranges = 9,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x12,
+ },
+ {
+ .first = 0x15,
+ .last = 0x17,
+ },
+ {
+ .first = 0x19,
+ .last = 0x21,
+ },
+ {
+ .first = 0x27,
+ .last = 0x2C,
+ },
+ {
+ .first = 0x41,
+ .last = 0x41,
+ },
+ {
+ .first = 0x45,
+ .last = 0x5B,
+ },
+ {
+ .first = 0x5D,
+ .last = 0x5D,
+ },
+ {
+ .first = 0x69,
+ .last = 0x69,
+ },
+ {
+ .first = 0x80,
+ .last = 0x81,
+ },
+ },
+ },
+ [AB8500_DBI] = {
+ .num_ranges = 0,
+ .range = NULL,
+ },
+ [AB8500_ECI_AV_ACC] = {
+ .num_ranges = 1,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x80,
+ .last = 0x82,
+ },
+ },
+ },
+ [0x9] = {
+ .num_ranges = 0,
+ .range = NULL,
+ },
+ [AB8500_GPADC] = {
+ .num_ranges = 1,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x08,
+ },
+ },
+ },
+ [AB8500_CHARGER] = {
+ .num_ranges = 8,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x03,
+ },
+ {
+ .first = 0x05,
+ .last = 0x05,
+ },
+ {
+ .first = 0x40,
+ .last = 0x40,
+ },
+ {
+ .first = 0x42,
+ .last = 0x42,
+ },
+ {
+ .first = 0x44,
+ .last = 0x44,
+ },
+ {
+ .first = 0x50,
+ .last = 0x55,
+ },
+ {
+ .first = 0x80,
+ .last = 0x82,
+ },
+ {
+ .first = 0xC0,
+ .last = 0xC2,
+ },
+ },
+ },
+ [AB8500_GAS_GAUGE] = {
+ .num_ranges = 3,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x00,
+ },
+ {
+ .first = 0x07,
+ .last = 0x0A,
+ },
+ {
+ .first = 0x10,
+ .last = 0x14,
+ },
+ },
+ },
+ [AB8500_AUDIO] = {
+ .num_ranges = 1,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x6F,
+ },
+ },
+ },
+ [AB8500_INTERRUPT] = {
+ .num_ranges = 0,
+ .range = NULL,
+ },
+ [AB8500_RTC] = {
+ .num_ranges = 1,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x0F,
+ },
+ },
+ },
+ [AB8500_MISC] = {
+ .num_ranges = 8,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x00,
+ .last = 0x05,
+ },
+ {
+ .first = 0x10,
+ .last = 0x15,
+ },
+ {
+ .first = 0x20,
+ .last = 0x25,
+ },
+ {
+ .first = 0x30,
+ .last = 0x35,
+ },
+ {
+ .first = 0x40,
+ .last = 0x45,
+ },
+ {
+ .first = 0x50,
+ .last = 0x50,
+ },
+ {
+ .first = 0x60,
+ .last = 0x67,
+ },
+ {
+ .first = 0x80,
+ .last = 0x80,
+ },
+ },
+ },
+ [0x11] = {
+ .num_ranges = 0,
+ .range = NULL,
+ },
+ [0x12] = {
+ .num_ranges = 0,
+ .range = NULL,
+ },
+ [0x13] = {
+ .num_ranges = 0,
+ .range = NULL,
+ },
+ [0x14] = {
+ .num_ranges = 0,
+ .range = NULL,
+ },
+ [AB8500_OTP_EMUL] = {
+ .num_ranges = 1,
+ .range = (struct ab8500_reg_range[]) {
+ {
+ .first = 0x01,
+ .last = 0x0F,
+ },
+ },
+ },
};
static int ab8500_registers_print(struct seq_file *s, void *p)
{
- struct device *dev = s->private;
- unsigned int i;
- u32 bank = debug_bank;
+ struct device *dev = s->private;
+ unsigned int i;
+ u32 bank = debug_bank;
- seq_printf(s, AB8500_NAME_STRING " register values:\n");
+ seq_printf(s, AB8500_NAME_STRING " register values:\n");
- seq_printf(s, " bank %u:\n", bank);
- for (i = 0; i < debug_ranges[bank].num_ranges; i++) {
- u32 reg;
+ seq_printf(s, " bank %u:\n", bank);
+ for (i = 0; i < debug_ranges[bank].num_ranges; i++) {
+ u32 reg;
- for (reg = debug_ranges[bank].range[i].first;
- reg <= debug_ranges[bank].range[i].last;
- reg++) {
- u8 value;
- int err;
+ for (reg = debug_ranges[bank].range[i].first;
+ reg <= debug_ranges[bank].range[i].last;
+ reg++) {
+ u8 value;
+ int err;
- err = abx500_get_register_interruptible(dev,
- (u8)bank, (u8)reg, &value);
- if (err < 0) {
- dev_err(dev, "ab->read fail %d\n", err);
- return err;
- }
+ err = abx500_get_register_interruptible(dev,
+ (u8)bank, (u8)reg, &value);
+ if (err < 0) {
+ dev_err(dev, "ab->read fail %d\n", err);
+ return err;
+ }
- err = seq_printf(s, " [%u/0x%02X]: 0x%02X\n", bank,
- reg, value);
- if (err < 0) {
- dev_err(dev, "seq_printf overflow\n");
- /* Error is not returned here since
- * the output is wanted in any case */
- return 0;
- }
- }
- }
- return 0;
+ err = seq_printf(s, " [%u/0x%02X]: 0x%02X\n", bank,
+ reg, value);
+ if (err < 0) {
+ dev_err(dev, "seq_printf overflow\n");
+ /* Error is not returned here since
+ * the output is wanted in any case */
+ return 0;
+ }
+ }
+ }
+ return 0;
}
static int ab8500_registers_open(struct inode *inode, struct file *file)
{
- return single_open(file, ab8500_registers_print, inode->i_private);
+ return single_open(file, ab8500_registers_print, inode->i_private);
}
static const struct file_operations ab8500_registers_fops = {
- .open = ab8500_registers_open,
- .read = seq_read,
- .llseek = seq_lseek,
- .release = single_release,
- .owner = THIS_MODULE,
+ .open = ab8500_registers_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+ .owner = THIS_MODULE,
};
static int ab8500_bank_print(struct seq_file *s, void *p)
{
- return seq_printf(s, "%d\n", debug_bank);
+ return seq_printf(s, "%d\n", debug_bank);
}
static int ab8500_bank_open(struct inode *inode, struct file *file)
{
- return single_open(file, ab8500_bank_print, inode->i_private);
+ return single_open(file, ab8500_bank_print, inode->i_private);
}
static ssize_t ab8500_bank_write(struct file *file,
- const char __user *user_buf,
- size_t count, loff_t *ppos)
+ const char __user *user_buf,
+ size_t count, loff_t *ppos)
{
- struct device *dev = ((struct seq_file *)(file->private_data))->private;
- char buf[32];
- int buf_size;
- unsigned long user_bank;
- int err;
+ struct device *dev = ((struct seq_file *)(file->private_data))->private;
+ char buf[32];
+ int buf_size;
+ unsigned long user_bank;
+ int err;
- /* Get userspace string and assure termination */
- buf_size = min(count, (sizeof(buf) - 1));
- if (copy_from_user(buf, user_buf, buf_size))
- return -EFAULT;
- buf[buf_size] = 0;
+ /* Get userspace string and assure termination */
+ buf_size = min(count, (sizeof(buf) - 1));
+ if (copy_from_user(buf, user_buf, buf_size))
+ return -EFAULT;
+ buf[buf_size] = 0;
- err = strict_strtoul(buf, 0, &user_bank);
- if (err)
- return -EINVAL;
+ err = strict_strtoul(buf, 0, &user_bank);
+ if (err)
+ return -EINVAL;
- if (user_bank >= AB8500_NUM_BANKS) {
- dev_err(dev, "debugfs error input > number of banks\n");
- return -EINVAL;
- }
+ if (user_bank >= AB8500_NUM_BANKS) {
+ dev_err(dev, "debugfs error input > number of banks\n");
+ return -EINVAL;
+ }
- debug_bank = user_bank;
+ debug_bank = user_bank;
- return buf_size;
+ return buf_size;
}
static int ab8500_address_print(struct seq_file *s, void *p)
{
- return seq_printf(s, "0x%02X\n", debug_address);
+ return seq_printf(s, "0x%02X\n", debug_address);
}
static int ab8500_address_open(struct inode *inode, struct file *file)
{
- return single_open(file, ab8500_address_print, inode->i_private);
+ return single_open(file, ab8500_address_print, inode->i_private);
}
static ssize_t ab8500_address_write(struct file *file,
- const char __user *user_buf,
- size_t count, loff_t *ppos)
+ const char __user *user_buf,
+ size_t count, loff_t *ppos)
{
- struct device *dev = ((struct seq_file *)(file->private_data))->private;
- char buf[32];
- int buf_size;
- unsigned long user_address;
- int err;
+ struct device *dev = ((struct seq_file *)(file->private_data))->private;
+ char buf[32];
+ int buf_size;
+ unsigned long user_address;
+ int err;
- /* Get userspace string and assure termination */
- buf_size = min(count, (sizeof(buf) - 1));
- if (copy_from_user(buf, user_buf, buf_size))
- return -EFAULT;
- buf[buf_size] = 0;
+ /* Get userspace string and assure termination */
+ buf_size = min(count, (sizeof(buf) - 1));
+ if (copy_from_user(buf, user_buf, buf_size))
+ return -EFAULT;
+ buf[buf_size] = 0;
- err = strict_strtoul(buf, 0, &user_address);
- if (err)
- return -EINVAL;
- if (user_address > 0xff) {
- dev_err(dev, "debugfs error input > 0xff\n");
- return -EINVAL;
- }
- debug_address = user_address;
- return buf_size;
+ err = strict_strtoul(buf, 0, &user_address);
+ if (err)
+ return -EINVAL;
+ if (user_address > 0xff) {
+ dev_err(dev, "debugfs error input > 0xff\n");
+ return -EINVAL;
+ }
+ debug_address = user_address;
+ return buf_size;
}
static int ab8500_val_print(struct seq_file *s, void *p)
{
- struct device *dev = s->private;
- int ret;
- u8 regvalue;
+ struct device *dev = s->private;
+ int ret;
+ u8 regvalue;
- ret = abx500_get_register_interruptible(dev,
- (u8)debug_bank, (u8)debug_address, ®value);
- if (ret < 0) {
- dev_err(dev, "abx500_get_reg fail %d, %d\n",
- ret, __LINE__);
- return -EINVAL;
- }
- seq_printf(s, "0x%02X\n", regvalue);
+ ret = abx500_get_register_interruptible(dev,
+ (u8)debug_bank, (u8)debug_address, ®value);
+ if (ret < 0) {
+ dev_err(dev, "abx500_get_reg fail %d, %d\n",
+ ret, __LINE__);
+ return -EINVAL;
+ }
+ seq_printf(s, "0x%02X\n", regvalue);
- return 0;
+ return 0;
}
static int ab8500_val_open(struct inode *inode, struct file *file)
{
- return single_open(file, ab8500_val_print, inode->i_private);
+ return single_open(file, ab8500_val_print, inode->i_private);
}
static ssize_t ab8500_val_write(struct file *file,
- const char __user *user_buf,
- size_t count, loff_t *ppos)
+ const char __user *user_buf,
+ size_t count, loff_t *ppos)
{
- struct device *dev = ((struct seq_file *)(file->private_data))->private;
- char buf[32];
- int buf_size;
- unsigned long user_val;
- int err;
+ struct device *dev = ((struct seq_file *)(file->private_data))->private;
+ char buf[32];
+ int buf_size;
+ unsigned long user_val;
+ int err;
- /* Get userspace string and assure termination */
- buf_size = min(count, (sizeof(buf)-1));
- if (copy_from_user(buf, user_buf, buf_size))
- return -EFAULT;
- buf[buf_size] = 0;
+ /* Get userspace string and assure termination */
+ buf_size = min(count, (sizeof(buf)-1));
+ if (copy_from_user(buf, user_buf, buf_size))
+ return -EFAULT;
+ buf[buf_size] = 0;
- err = strict_strtoul(buf, 0, &user_val);
- if (err)
- return -EINVAL;
- if (user_val > 0xff) {
- dev_err(dev, "debugfs error input > 0xff\n");
- return -EINVAL;
- }
- err = abx500_set_register_interruptible(dev,
- (u8)debug_bank, debug_address, (u8)user_val);
- if (err < 0) {
- printk(KERN_ERR "abx500_set_reg failed %d, %d", err, __LINE__);
- return -EINVAL;
- }
+ err = strict_strtoul(buf, 0, &user_val);
+ if (err)
+ return -EINVAL;
+ if (user_val > 0xff) {
+ dev_err(dev, "debugfs error input > 0xff\n");
+ return -EINVAL;
+ }
+ err = abx500_set_register_interruptible(dev,
+ (u8)debug_bank, debug_address, (u8)user_val);
+ if (err < 0) {
+ printk(KERN_ERR "abx500_set_reg failed %d, %d", err, __LINE__);
+ return -EINVAL;
+ }
- return buf_size;
+ return buf_size;
}
static const struct file_operations ab8500_bank_fops = {
- .open = ab8500_bank_open,
- .write = ab8500_bank_write,
- .read = seq_read,
- .llseek = seq_lseek,
- .release = single_release,
- .owner = THIS_MODULE,
+ .open = ab8500_bank_open,
+ .write = ab8500_bank_write,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+ .owner = THIS_MODULE,
};
static const struct file_operations ab8500_address_fops = {
- .open = ab8500_address_open,
- .write = ab8500_address_write,
- .read = seq_read,
- .llseek = seq_lseek,
- .release = single_release,
- .owner = THIS_MODULE,
+ .open = ab8500_address_open,
+ .write = ab8500_address_write,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+ .owner = THIS_MODULE,
};
static const struct file_operations ab8500_val_fops = {
- .open = ab8500_val_open,
- .write = ab8500_val_write,
- .read = seq_read,
- .llseek = seq_lseek,
- .release = single_release,
- .owner = THIS_MODULE,
+ .open = ab8500_val_open,
+ .write = ab8500_val_write,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+ .owner = THIS_MODULE,
};
static struct dentry *ab8500_dir;
@@ -572,77 +572,77 @@
static int __devinit ab8500_debug_probe(struct platform_device *plf)
{
- debug_bank = AB8500_MISC;
- debug_address = AB8500_REV_REG & 0x00FF;
+ debug_bank = AB8500_MISC;
+ debug_address = AB8500_REV_REG & 0x00FF;
- ab8500_dir = debugfs_create_dir(AB8500_NAME_STRING, NULL);
- if (!ab8500_dir)
- goto exit_no_debugfs;
+ ab8500_dir = debugfs_create_dir(AB8500_NAME_STRING, NULL);
+ if (!ab8500_dir)
+ goto exit_no_debugfs;
- ab8500_reg_file = debugfs_create_file("all-bank-registers",
- S_IRUGO, ab8500_dir, &plf->dev, &ab8500_registers_fops);
- if (!ab8500_reg_file)
- goto exit_destroy_dir;
+ ab8500_reg_file = debugfs_create_file("all-bank-registers",
+ S_IRUGO, ab8500_dir, &plf->dev, &ab8500_registers_fops);
+ if (!ab8500_reg_file)
+ goto exit_destroy_dir;
- ab8500_bank_file = debugfs_create_file("register-bank",
- (S_IRUGO | S_IWUGO), ab8500_dir, &plf->dev, &ab8500_bank_fops);
- if (!ab8500_bank_file)
- goto exit_destroy_reg;
+ ab8500_bank_file = debugfs_create_file("register-bank",
+ (S_IRUGO | S_IWUGO), ab8500_dir, &plf->dev, &ab8500_bank_fops);
+ if (!ab8500_bank_file)
+ goto exit_destroy_reg;
- ab8500_address_file = debugfs_create_file("register-address",
- (S_IRUGO | S_IWUGO), ab8500_dir, &plf->dev,
- &ab8500_address_fops);
- if (!ab8500_address_file)
- goto exit_destroy_bank;
+ ab8500_address_file = debugfs_create_file("register-address",
+ (S_IRUGO | S_IWUGO), ab8500_dir, &plf->dev,
+ &ab8500_address_fops);
+ if (!ab8500_address_file)
+ goto exit_destroy_bank;
- ab8500_val_file = debugfs_create_file("register-value",
- (S_IRUGO | S_IWUGO), ab8500_dir, &plf->dev, &ab8500_val_fops);
- if (!ab8500_val_file)
- goto exit_destroy_address;
+ ab8500_val_file = debugfs_create_file("register-value",
+ (S_IRUGO | S_IWUGO), ab8500_dir, &plf->dev, &ab8500_val_fops);
+ if (!ab8500_val_file)
+ goto exit_destroy_address;
- return 0;
+ return 0;
exit_destroy_address:
- debugfs_remove(ab8500_address_file);
+ debugfs_remove(ab8500_address_file);
exit_destroy_bank:
- debugfs_remove(ab8500_bank_file);
+ debugfs_remove(ab8500_bank_file);
exit_destroy_reg:
- debugfs_remove(ab8500_reg_file);
+ debugfs_remove(ab8500_reg_file);
exit_destroy_dir:
- debugfs_remove(ab8500_dir);
+ debugfs_remove(ab8500_dir);
exit_no_debugfs:
- dev_err(&plf->dev, "failed to create debugfs entries.\n");
- return -ENOMEM;
+ dev_err(&plf->dev, "failed to create debugfs entries.\n");
+ return -ENOMEM;
}
static int __devexit ab8500_debug_remove(struct platform_device *plf)
{
- debugfs_remove(ab8500_val_file);
- debugfs_remove(ab8500_address_file);
- debugfs_remove(ab8500_bank_file);
- debugfs_remove(ab8500_reg_file);
- debugfs_remove(ab8500_dir);
+ debugfs_remove(ab8500_val_file);
+ debugfs_remove(ab8500_address_file);
+ debugfs_remove(ab8500_bank_file);
+ debugfs_remove(ab8500_reg_file);
+ debugfs_remove(ab8500_dir);
- return 0;
+ return 0;
}
static struct platform_driver ab8500_debug_driver = {
- .driver = {
- .name = "ab8500-debug",
- .owner = THIS_MODULE,
- },
- .probe = ab8500_debug_probe,
- .remove = __devexit_p(ab8500_debug_remove)
+ .driver = {
+ .name = "ab8500-debug",
+ .owner = THIS_MODULE,
+ },
+ .probe = ab8500_debug_probe,
+ .remove = __devexit_p(ab8500_debug_remove)
};
static int __init ab8500_debug_init(void)
{
- return platform_driver_register(&ab8500_debug_driver);
+ return platform_driver_register(&ab8500_debug_driver);
}
static void __exit ab8500_debug_exit(void)
{
- platform_driver_unregister(&ab8500_debug_driver);
+ platform_driver_unregister(&ab8500_debug_driver);
}
subsys_initcall(ab8500_debug_init);
module_exit(ab8500_debug_exit);
diff --git a/drivers/mfd/ab8500-spi.c b/drivers/mfd/ab8500-spi.c
deleted file mode 100644
index b165342..0000000
--- a/drivers/mfd/ab8500-spi.c
+++ /dev/null
@@ -1,143 +0,0 @@
-/*
- * Copyright (C) ST-Ericsson SA 2010
- *
- * License Terms: GNU General Public License v2
- * Author: Srinidhi Kasagar <srinidhi.kasagar@stericsson.com>
- */
-
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/init.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/spi/spi.h>
-#include <linux/mfd/ab8500.h>
-
-/*
- * This funtion writes to any AB8500 registers using
- * SPI protocol & before it writes it packs the data
- * in the below 24 bit frame format
- *
- * *|------------------------------------|
- * *| 23|22...18|17.......10|9|8|7......0|
- * *| r/w bank adr data |
- * * ------------------------------------
- *
- * This function shouldn't be called from interrupt
- * context
- */
-static int ab8500_spi_write(struct ab8500 *ab8500, u16 addr, u8 data)
-{
- struct spi_device *spi = container_of(ab8500->dev, struct spi_device,
- dev);
- unsigned long spi_data = addr << 10 | data;
- struct spi_transfer xfer;
- struct spi_message msg;
-
- ab8500->tx_buf[0] = spi_data;
- ab8500->rx_buf[0] = 0;
-
- xfer.tx_buf = ab8500->tx_buf;
- xfer.rx_buf = NULL;
- xfer.len = sizeof(unsigned long);
-
- spi_message_init(&msg);
- spi_message_add_tail(&xfer, &msg);
-
- return spi_sync(spi, &msg);
-}
-
-static int ab8500_spi_read(struct ab8500 *ab8500, u16 addr)
-{
- struct spi_device *spi = container_of(ab8500->dev, struct spi_device,
- dev);
- unsigned long spi_data = 1 << 23 | addr << 10;
- struct spi_transfer xfer;
- struct spi_message msg;
- int ret;
-
- ab8500->tx_buf[0] = spi_data;
- ab8500->rx_buf[0] = 0;
-
- xfer.tx_buf = ab8500->tx_buf;
- xfer.rx_buf = ab8500->rx_buf;
- xfer.len = sizeof(unsigned long);
-
- spi_message_init(&msg);
- spi_message_add_tail(&xfer, &msg);
-
- ret = spi_sync(spi, &msg);
- if (!ret)
- /*
- * Only the 8 lowermost bytes are
- * defined with value, the rest may
- * vary depending on chip/board noise.
- */
- ret = ab8500->rx_buf[0] & 0xFFU;
-
- return ret;
-}
-
-static int __devinit ab8500_spi_probe(struct spi_device *spi)
-{
- struct ab8500 *ab8500;
- int ret;
-
- spi->bits_per_word = 24;
- ret = spi_setup(spi);
- if (ret < 0)
- return ret;
-
- ab8500 = kzalloc(sizeof *ab8500, GFP_KERNEL);
- if (!ab8500)
- return -ENOMEM;
-
- ab8500->dev = &spi->dev;
- ab8500->irq = spi->irq;
-
- ab8500->read = ab8500_spi_read;
- ab8500->write = ab8500_spi_write;
-
- spi_set_drvdata(spi, ab8500);
-
- ret = ab8500_init(ab8500);
- if (ret)
- kfree(ab8500);
-
- return ret;
-}
-
-static int __devexit ab8500_spi_remove(struct spi_device *spi)
-{
- struct ab8500 *ab8500 = spi_get_drvdata(spi);
-
- ab8500_exit(ab8500);
- kfree(ab8500);
-
- return 0;
-}
-
-static struct spi_driver ab8500_spi_driver = {
- .driver = {
- .name = "ab8500-spi",
- .owner = THIS_MODULE,
- },
- .probe = ab8500_spi_probe,
- .remove = __devexit_p(ab8500_spi_remove)
-};
-
-static int __init ab8500_spi_init(void)
-{
- return spi_register_driver(&ab8500_spi_driver);
-}
-subsys_initcall(ab8500_spi_init);
-
-static void __exit ab8500_spi_exit(void)
-{
- spi_unregister_driver(&ab8500_spi_driver);
-}
-module_exit(ab8500_spi_exit);
-
-MODULE_AUTHOR("Srinidhi KASAGAR <srinidhi.kasagar@stericsson.com");
-MODULE_DESCRIPTION("AB8500 SPI");
-MODULE_LICENSE("GPL v2");
diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
index 7de708d..6a1f940 100644
--- a/drivers/mfd/asic3.c
+++ b/drivers/mfd/asic3.c
@@ -57,7 +57,7 @@
.rate = _rate, \
}
-struct asic3_clk asic3_clk_init[] __initdata = {
+static struct asic3_clk asic3_clk_init[] __initdata = {
INIT_CDEX(SPI, 0),
INIT_CDEX(OWM, 5000000),
INIT_CDEX(PWM0, 0),
@@ -102,7 +102,7 @@
(reg >> asic->bus_shift));
}
-void asic3_set_register(struct asic3 *asic, u32 reg, u32 bits, bool set)
+static void asic3_set_register(struct asic3 *asic, u32 reg, u32 bits, bool set)
{
unsigned long flags;
u32 val;
@@ -226,14 +226,14 @@
return (irq - asic->irq_base) & 0xf;
}
-static void asic3_mask_gpio_irq(unsigned int irq)
+static void asic3_mask_gpio_irq(struct irq_data *data)
{
- struct asic3 *asic = get_irq_chip_data(irq);
+ struct asic3 *asic = irq_data_get_irq_chip_data(data);
u32 val, bank, index;
unsigned long flags;
- bank = asic3_irq_to_bank(asic, irq);
- index = asic3_irq_to_index(asic, irq);
+ bank = asic3_irq_to_bank(asic, data->irq);
+ index = asic3_irq_to_index(asic, data->irq);
spin_lock_irqsave(&asic->lock, flags);
val = asic3_read_register(asic, bank + ASIC3_GPIO_MASK);
@@ -242,9 +242,9 @@
spin_unlock_irqrestore(&asic->lock, flags);
}
-static void asic3_mask_irq(unsigned int irq)
+static void asic3_mask_irq(struct irq_data *data)
{
- struct asic3 *asic = get_irq_chip_data(irq);
+ struct asic3 *asic = irq_data_get_irq_chip_data(data);
int regval;
unsigned long flags;
@@ -254,7 +254,7 @@
ASIC3_INTR_INT_MASK);
regval &= ~(ASIC3_INTMASK_MASK0 <<
- (irq - (asic->irq_base + ASIC3_NUM_GPIOS)));
+ (data->irq - (asic->irq_base + ASIC3_NUM_GPIOS)));
asic3_write_register(asic,
ASIC3_INTR_BASE +
@@ -263,14 +263,14 @@
spin_unlock_irqrestore(&asic->lock, flags);
}
-static void asic3_unmask_gpio_irq(unsigned int irq)
+static void asic3_unmask_gpio_irq(struct irq_data *data)
{
- struct asic3 *asic = get_irq_chip_data(irq);
+ struct asic3 *asic = irq_data_get_irq_chip_data(data);
u32 val, bank, index;
unsigned long flags;
- bank = asic3_irq_to_bank(asic, irq);
- index = asic3_irq_to_index(asic, irq);
+ bank = asic3_irq_to_bank(asic, data->irq);
+ index = asic3_irq_to_index(asic, data->irq);
spin_lock_irqsave(&asic->lock, flags);
val = asic3_read_register(asic, bank + ASIC3_GPIO_MASK);
@@ -279,9 +279,9 @@
spin_unlock_irqrestore(&asic->lock, flags);
}
-static void asic3_unmask_irq(unsigned int irq)
+static void asic3_unmask_irq(struct irq_data *data)
{
- struct asic3 *asic = get_irq_chip_data(irq);
+ struct asic3 *asic = irq_data_get_irq_chip_data(data);
int regval;
unsigned long flags;
@@ -291,7 +291,7 @@
ASIC3_INTR_INT_MASK);
regval |= (ASIC3_INTMASK_MASK0 <<
- (irq - (asic->irq_base + ASIC3_NUM_GPIOS)));
+ (data->irq - (asic->irq_base + ASIC3_NUM_GPIOS)));
asic3_write_register(asic,
ASIC3_INTR_BASE +
@@ -300,15 +300,15 @@
spin_unlock_irqrestore(&asic->lock, flags);
}
-static int asic3_gpio_irq_type(unsigned int irq, unsigned int type)
+static int asic3_gpio_irq_type(struct irq_data *data, unsigned int type)
{
- struct asic3 *asic = get_irq_chip_data(irq);
+ struct asic3 *asic = irq_data_get_irq_chip_data(data);
u32 bank, index;
u16 trigger, level, edge, bit;
unsigned long flags;
- bank = asic3_irq_to_bank(asic, irq);
- index = asic3_irq_to_index(asic, irq);
+ bank = asic3_irq_to_bank(asic, data->irq);
+ index = asic3_irq_to_index(asic, data->irq);
bit = 1<<index;
spin_lock_irqsave(&asic->lock, flags);
@@ -318,7 +318,7 @@
bank + ASIC3_GPIO_EDGE_TRIGGER);
trigger = asic3_read_register(asic,
bank + ASIC3_GPIO_TRIGGER_TYPE);
- asic->irq_bothedge[(irq - asic->irq_base) >> 4] &= ~bit;
+ asic->irq_bothedge[(data->irq - asic->irq_base) >> 4] &= ~bit;
if (type == IRQ_TYPE_EDGE_RISING) {
trigger |= bit;
@@ -328,11 +328,11 @@
edge &= ~bit;
} else if (type == IRQ_TYPE_EDGE_BOTH) {
trigger |= bit;
- if (asic3_gpio_get(&asic->gpio, irq - asic->irq_base))
+ if (asic3_gpio_get(&asic->gpio, data->irq - asic->irq_base))
edge &= ~bit;
else
edge |= bit;
- asic->irq_bothedge[(irq - asic->irq_base) >> 4] |= bit;
+ asic->irq_bothedge[(data->irq - asic->irq_base) >> 4] |= bit;
} else if (type == IRQ_TYPE_LEVEL_LOW) {
trigger &= ~bit;
level &= ~bit;
@@ -359,17 +359,17 @@
static struct irq_chip asic3_gpio_irq_chip = {
.name = "ASIC3-GPIO",
- .ack = asic3_mask_gpio_irq,
- .mask = asic3_mask_gpio_irq,
- .unmask = asic3_unmask_gpio_irq,
- .set_type = asic3_gpio_irq_type,
+ .irq_ack = asic3_mask_gpio_irq,
+ .irq_mask = asic3_mask_gpio_irq,
+ .irq_unmask = asic3_unmask_gpio_irq,
+ .irq_set_type = asic3_gpio_irq_type,
};
static struct irq_chip asic3_irq_chip = {
.name = "ASIC3",
- .ack = asic3_mask_irq,
- .mask = asic3_mask_irq,
- .unmask = asic3_unmask_irq,
+ .irq_ack = asic3_mask_irq,
+ .irq_mask = asic3_mask_irq,
+ .irq_unmask = asic3_unmask_irq,
};
static int __init asic3_irq_probe(struct platform_device *pdev)
@@ -635,7 +635,7 @@
},
{
.start = ASIC3_IRQ_OWM,
- .start = ASIC3_IRQ_OWM,
+ .end = ASIC3_IRQ_OWM,
.flags = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHEDGE,
},
};
diff --git a/drivers/mfd/cs5535-mfd.c b/drivers/mfd/cs5535-mfd.c
new file mode 100644
index 0000000..59ca6f1
--- /dev/null
+++ b/drivers/mfd/cs5535-mfd.c
@@ -0,0 +1,151 @@
+/*
+ * cs5535-mfd.c - core MFD driver for CS5535/CS5536 southbridges
+ *
+ * The CS5535 and CS5536 has an ISA bridge on the PCI bus that is
+ * used for accessing GPIOs, MFGPTs, ACPI, etc. Each subdevice has
+ * an IO range that's specified in a single BAR. The BAR order is
+ * hardcoded in the CS553x specifications.
+ *
+ * Copyright (c) 2010 Andres Salomon <dilinger@queued.net>
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/mfd/core.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+
+#define DRV_NAME "cs5535-mfd"
+
+enum cs5535_mfd_bars {
+ SMB_BAR = 0,
+ GPIO_BAR = 1,
+ MFGPT_BAR = 2,
+ PMS_BAR = 4,
+ ACPI_BAR = 5,
+ NR_BARS,
+};
+
+static __devinitdata struct resource cs5535_mfd_resources[NR_BARS];
+
+static __devinitdata struct mfd_cell cs5535_mfd_cells[] = {
+ {
+ .id = SMB_BAR,
+ .name = "cs5535-smb",
+ .num_resources = 1,
+ .resources = &cs5535_mfd_resources[SMB_BAR],
+ },
+ {
+ .id = GPIO_BAR,
+ .name = "cs5535-gpio",
+ .num_resources = 1,
+ .resources = &cs5535_mfd_resources[GPIO_BAR],
+ },
+ {
+ .id = MFGPT_BAR,
+ .name = "cs5535-mfgpt",
+ .num_resources = 1,
+ .resources = &cs5535_mfd_resources[MFGPT_BAR],
+ },
+ {
+ .id = PMS_BAR,
+ .name = "cs5535-pms",
+ .num_resources = 1,
+ .resources = &cs5535_mfd_resources[PMS_BAR],
+ },
+ {
+ .id = ACPI_BAR,
+ .name = "cs5535-acpi",
+ .num_resources = 1,
+ .resources = &cs5535_mfd_resources[ACPI_BAR],
+ },
+};
+
+static int __devinit cs5535_mfd_probe(struct pci_dev *pdev,
+ const struct pci_device_id *id)
+{
+ int err, i;
+
+ err = pci_enable_device(pdev);
+ if (err)
+ return err;
+
+ /* fill in IO range for each cell; subdrivers handle the region */
+ for (i = 0; i < ARRAY_SIZE(cs5535_mfd_cells); i++) {
+ int bar = cs5535_mfd_cells[i].id;
+ struct resource *r = &cs5535_mfd_resources[bar];
+
+ r->flags = IORESOURCE_IO;
+ r->start = pci_resource_start(pdev, bar);
+ r->end = pci_resource_end(pdev, bar);
+
+ /* id is used for temporarily storing BAR; unset it now */
+ cs5535_mfd_cells[i].id = 0;
+ }
+
+ err = mfd_add_devices(&pdev->dev, -1, cs5535_mfd_cells,
+ ARRAY_SIZE(cs5535_mfd_cells), NULL, 0);
+ if (err) {
+ dev_err(&pdev->dev, "MFD add devices failed: %d\n", err);
+ goto err_disable;
+ }
+
+ dev_info(&pdev->dev, "%zu devices registered.\n",
+ ARRAY_SIZE(cs5535_mfd_cells));
+
+ return 0;
+
+err_disable:
+ pci_disable_device(pdev);
+ return err;
+}
+
+static void __devexit cs5535_mfd_remove(struct pci_dev *pdev)
+{
+ mfd_remove_devices(&pdev->dev);
+ pci_disable_device(pdev);
+}
+
+static struct pci_device_id cs5535_mfd_pci_tbl[] = {
+ { PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_CS5535_ISA) },
+ { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA) },
+ { 0, }
+};
+MODULE_DEVICE_TABLE(pci, cs5535_mfd_pci_tbl);
+
+static struct pci_driver cs5535_mfd_drv = {
+ .name = DRV_NAME,
+ .id_table = cs5535_mfd_pci_tbl,
+ .probe = cs5535_mfd_probe,
+ .remove = __devexit_p(cs5535_mfd_remove),
+};
+
+static int __init cs5535_mfd_init(void)
+{
+ return pci_register_driver(&cs5535_mfd_drv);
+}
+
+static void __exit cs5535_mfd_exit(void)
+{
+ pci_unregister_driver(&cs5535_mfd_drv);
+}
+
+module_init(cs5535_mfd_init);
+module_exit(cs5535_mfd_exit);
+
+MODULE_AUTHOR("Andres Salomon <dilinger@queued.net>");
+MODULE_DESCRIPTION("MFD driver for CS5535/CS5536 southbridge's ISA PCI device");
+MODULE_LICENSE("GPL");
diff --git a/drivers/mfd/ezx-pcap.c b/drivers/mfd/ezx-pcap.c
index c2b698d..9e2d8dd 100644
--- a/drivers/mfd/ezx-pcap.c
+++ b/drivers/mfd/ezx-pcap.c
@@ -144,26 +144,26 @@
}
EXPORT_SYMBOL_GPL(pcap_to_irq);
-static void pcap_mask_irq(unsigned int irq)
+static void pcap_mask_irq(struct irq_data *d)
{
- struct pcap_chip *pcap = get_irq_chip_data(irq);
+ struct pcap_chip *pcap = irq_data_get_irq_chip_data(d);
- pcap->msr |= 1 << irq_to_pcap(pcap, irq);
+ pcap->msr |= 1 << irq_to_pcap(pcap, d->irq);
queue_work(pcap->workqueue, &pcap->msr_work);
}
-static void pcap_unmask_irq(unsigned int irq)
+static void pcap_unmask_irq(struct irq_data *d)
{
- struct pcap_chip *pcap = get_irq_chip_data(irq);
+ struct pcap_chip *pcap = irq_data_get_irq_chip_data(d);
- pcap->msr &= ~(1 << irq_to_pcap(pcap, irq));
+ pcap->msr &= ~(1 << irq_to_pcap(pcap, d->irq));
queue_work(pcap->workqueue, &pcap->msr_work);
}
static struct irq_chip pcap_irq_chip = {
- .name = "pcap",
- .mask = pcap_mask_irq,
- .unmask = pcap_unmask_irq,
+ .name = "pcap",
+ .irq_mask = pcap_mask_irq,
+ .irq_unmask = pcap_unmask_irq,
};
static void pcap_msr_work(struct work_struct *work)
@@ -199,8 +199,7 @@
if (service & 1) {
struct irq_desc *desc = irq_to_desc(irq);
- if (WARN(!desc, KERN_WARNING
- "Invalid PCAP IRQ %d\n", irq))
+ if (WARN(!desc, "Invalid PCAP IRQ %d\n", irq))
break;
if (desc->status & IRQ_DISABLED)
@@ -218,7 +217,7 @@
{
struct pcap_chip *pcap = get_irq_data(irq);
- desc->chip->ack(irq);
+ desc->irq_data.chip->irq_ack(&desc->irq_data);
queue_work(pcap->workqueue, &pcap->isr_work);
return;
}
@@ -282,7 +281,7 @@
mutex_lock(&pcap->adc_mutex);
req = pcap->adc_queue[pcap->adc_head];
- if (WARN(!req, KERN_WARNING "adc irq without pending request\n")) {
+ if (WARN(!req, "adc irq without pending request\n")) {
mutex_unlock(&pcap->adc_mutex);
return IRQ_HANDLED;
}
diff --git a/drivers/mfd/htc-egpio.c b/drivers/mfd/htc-egpio.c
index d3e74f8..d00b6d1 100644
--- a/drivers/mfd/htc-egpio.c
+++ b/drivers/mfd/htc-egpio.c
@@ -70,31 +70,32 @@
ei->ack_write, ei->ack_register << ei->bus_shift);
}
-static void egpio_ack(unsigned int irq)
+static void egpio_ack(struct irq_data *data)
{
}
/* There does not appear to be a way to proactively mask interrupts
* on the egpio chip itself. So, we simply ignore interrupts that
* aren't desired. */
-static void egpio_mask(unsigned int irq)
+static void egpio_mask(struct irq_data *data)
{
- struct egpio_info *ei = get_irq_chip_data(irq);
- ei->irqs_enabled &= ~(1 << (irq - ei->irq_start));
- pr_debug("EGPIO mask %d %04x\n", irq, ei->irqs_enabled);
+ struct egpio_info *ei = irq_data_get_irq_chip_data(data);
+ ei->irqs_enabled &= ~(1 << (data->irq - ei->irq_start));
+ pr_debug("EGPIO mask %d %04x\n", data->irq, ei->irqs_enabled);
}
-static void egpio_unmask(unsigned int irq)
+
+static void egpio_unmask(struct irq_data *data)
{
- struct egpio_info *ei = get_irq_chip_data(irq);
- ei->irqs_enabled |= 1 << (irq - ei->irq_start);
- pr_debug("EGPIO unmask %d %04x\n", irq, ei->irqs_enabled);
+ struct egpio_info *ei = irq_data_get_irq_chip_data(data);
+ ei->irqs_enabled |= 1 << (data->irq - ei->irq_start);
+ pr_debug("EGPIO unmask %d %04x\n", data->irq, ei->irqs_enabled);
}
static struct irq_chip egpio_muxed_chip = {
- .name = "htc-egpio",
- .ack = egpio_ack,
- .mask = egpio_mask,
- .unmask = egpio_unmask,
+ .name = "htc-egpio",
+ .irq_ack = egpio_ack,
+ .irq_mask = egpio_mask,
+ .irq_unmask = egpio_unmask,
};
static void egpio_handler(unsigned int irq, struct irq_desc *desc)
diff --git a/drivers/mfd/htc-i2cpld.c b/drivers/mfd/htc-i2cpld.c
index 594c9a8..296ad15 100644
--- a/drivers/mfd/htc-i2cpld.c
+++ b/drivers/mfd/htc-i2cpld.c
@@ -82,25 +82,25 @@
/* There does not appear to be a way to proactively mask interrupts
* on the htcpld chip itself. So, we simply ignore interrupts that
* aren't desired. */
-static void htcpld_mask(unsigned int irq)
+static void htcpld_mask(struct irq_data *data)
{
- struct htcpld_chip *chip = get_irq_chip_data(irq);
- chip->irqs_enabled &= ~(1 << (irq - chip->irq_start));
- pr_debug("HTCPLD mask %d %04x\n", irq, chip->irqs_enabled);
+ struct htcpld_chip *chip = irq_data_get_irq_chip_data(data);
+ chip->irqs_enabled &= ~(1 << (data->irq - chip->irq_start));
+ pr_debug("HTCPLD mask %d %04x\n", data->irq, chip->irqs_enabled);
}
-static void htcpld_unmask(unsigned int irq)
+static void htcpld_unmask(struct irq_data *data)
{
- struct htcpld_chip *chip = get_irq_chip_data(irq);
- chip->irqs_enabled |= 1 << (irq - chip->irq_start);
- pr_debug("HTCPLD unmask %d %04x\n", irq, chip->irqs_enabled);
+ struct htcpld_chip *chip = irq_data_get_irq_chip_data(data);
+ chip->irqs_enabled |= 1 << (data->irq - chip->irq_start);
+ pr_debug("HTCPLD unmask %d %04x\n", data->irq, chip->irqs_enabled);
}
-static int htcpld_set_type(unsigned int irq, unsigned int flags)
+static int htcpld_set_type(struct irq_data *data, unsigned int flags)
{
- struct irq_desc *d = irq_to_desc(irq);
+ struct irq_desc *d = irq_to_desc(data->irq);
if (!d) {
- pr_err("HTCPLD invalid IRQ: %d\n", irq);
+ pr_err("HTCPLD invalid IRQ: %d\n", data->irq);
return -EINVAL;
}
@@ -118,10 +118,10 @@
}
static struct irq_chip htcpld_muxed_chip = {
- .name = "htcpld",
- .mask = htcpld_mask,
- .unmask = htcpld_unmask,
- .set_type = htcpld_set_type,
+ .name = "htcpld",
+ .irq_mask = htcpld_mask,
+ .irq_unmask = htcpld_unmask,
+ .irq_set_type = htcpld_set_type,
};
/* To properly dispatch IRQ events, we need to read from the
@@ -235,7 +235,7 @@
* and that work is scheduled in the set routine. The kernel can then run
* the I2C functions, which will sleep, in process context.
*/
-void htcpld_chip_set(struct gpio_chip *chip, unsigned offset, int val)
+static void htcpld_chip_set(struct gpio_chip *chip, unsigned offset, int val)
{
struct i2c_client *client;
struct htcpld_chip *chip_data;
@@ -259,7 +259,7 @@
schedule_work(&(chip_data->set_val_work));
}
-void htcpld_chip_set_ni(struct work_struct *work)
+static void htcpld_chip_set_ni(struct work_struct *work)
{
struct htcpld_chip *chip_data;
struct i2c_client *client;
@@ -269,7 +269,7 @@
i2c_smbus_read_byte_data(client, chip_data->cache_out);
}
-int htcpld_chip_get(struct gpio_chip *chip, unsigned offset)
+static int htcpld_chip_get(struct gpio_chip *chip, unsigned offset)
{
struct htcpld_chip *chip_data;
int val = 0;
@@ -316,7 +316,7 @@
return (offset < chip->ngpio) ? 0 : -EINVAL;
}
-int htcpld_chip_to_irq(struct gpio_chip *chip, unsigned offset)
+static int htcpld_chip_to_irq(struct gpio_chip *chip, unsigned offset)
{
struct htcpld_chip *chip_data;
@@ -328,7 +328,7 @@
return -EINVAL;
}
-void htcpld_chip_reset(struct i2c_client *client)
+static void htcpld_chip_reset(struct i2c_client *client)
{
struct htcpld_chip *chip_data = i2c_get_clientdata(client);
if (!chip_data)
diff --git a/drivers/mfd/jz4740-adc.c b/drivers/mfd/jz4740-adc.c
index 9dd1b33..0cc5979 100644
--- a/drivers/mfd/jz4740-adc.c
+++ b/drivers/mfd/jz4740-adc.c
@@ -84,31 +84,30 @@
spin_unlock_irqrestore(&adc->lock, flags);
}
-static void jz4740_adc_irq_mask(unsigned int irq)
+static void jz4740_adc_irq_mask(struct irq_data *data)
{
- struct jz4740_adc *adc = get_irq_chip_data(irq);
- jz4740_adc_irq_set_masked(adc, irq, true);
+ struct jz4740_adc *adc = irq_data_get_irq_chip_data(data);
+ jz4740_adc_irq_set_masked(adc, data->irq, true);
}
-static void jz4740_adc_irq_unmask(unsigned int irq)
+static void jz4740_adc_irq_unmask(struct irq_data *data)
{
- struct jz4740_adc *adc = get_irq_chip_data(irq);
- jz4740_adc_irq_set_masked(adc, irq, false);
+ struct jz4740_adc *adc = irq_data_get_irq_chip_data(data);
+ jz4740_adc_irq_set_masked(adc, data->irq, false);
}
-static void jz4740_adc_irq_ack(unsigned int irq)
+static void jz4740_adc_irq_ack(struct irq_data *data)
{
- struct jz4740_adc *adc = get_irq_chip_data(irq);
-
- irq -= adc->irq_base;
+ struct jz4740_adc *adc = irq_data_get_irq_chip_data(data);
+ unsigned int irq = data->irq - adc->irq_base;
writeb(BIT(irq), adc->base + JZ_REG_ADC_STATUS);
}
static struct irq_chip jz4740_adc_irq_chip = {
.name = "jz4740-adc",
- .mask = jz4740_adc_irq_mask,
- .unmask = jz4740_adc_irq_unmask,
- .ack = jz4740_adc_irq_ack,
+ .irq_mask = jz4740_adc_irq_mask,
+ .irq_unmask = jz4740_adc_irq_unmask,
+ .irq_ack = jz4740_adc_irq_ack,
};
static void jz4740_adc_irq_demux(unsigned int irq, struct irq_desc *desc)
diff --git a/drivers/mfd/max8925-core.c b/drivers/mfd/max8925-core.c
index 44695f5..0e998dc 100644
--- a/drivers/mfd/max8925-core.c
+++ b/drivers/mfd/max8925-core.c
@@ -407,16 +407,16 @@
return IRQ_HANDLED;
}
-static void max8925_irq_lock(unsigned int irq)
+static void max8925_irq_lock(struct irq_data *data)
{
- struct max8925_chip *chip = get_irq_chip_data(irq);
+ struct max8925_chip *chip = irq_data_get_irq_chip_data(data);
mutex_lock(&chip->irq_lock);
}
-static void max8925_irq_sync_unlock(unsigned int irq)
+static void max8925_irq_sync_unlock(struct irq_data *data)
{
- struct max8925_chip *chip = get_irq_chip_data(irq);
+ struct max8925_chip *chip = irq_data_get_irq_chip_data(data);
struct max8925_irq_data *irq_data;
static unsigned char cache_chg[2] = {0xff, 0xff};
static unsigned char cache_on[2] = {0xff, 0xff};
@@ -492,25 +492,25 @@
mutex_unlock(&chip->irq_lock);
}
-static void max8925_irq_enable(unsigned int irq)
+static void max8925_irq_enable(struct irq_data *data)
{
- struct max8925_chip *chip = get_irq_chip_data(irq);
- max8925_irqs[irq - chip->irq_base].enable
- = max8925_irqs[irq - chip->irq_base].offs;
+ struct max8925_chip *chip = irq_data_get_irq_chip_data(data);
+ max8925_irqs[data->irq - chip->irq_base].enable
+ = max8925_irqs[data->irq - chip->irq_base].offs;
}
-static void max8925_irq_disable(unsigned int irq)
+static void max8925_irq_disable(struct irq_data *data)
{
- struct max8925_chip *chip = get_irq_chip_data(irq);
- max8925_irqs[irq - chip->irq_base].enable = 0;
+ struct max8925_chip *chip = irq_data_get_irq_chip_data(data);
+ max8925_irqs[data->irq - chip->irq_base].enable = 0;
}
static struct irq_chip max8925_irq_chip = {
.name = "max8925",
- .bus_lock = max8925_irq_lock,
- .bus_sync_unlock = max8925_irq_sync_unlock,
- .enable = max8925_irq_enable,
- .disable = max8925_irq_disable,
+ .irq_bus_lock = max8925_irq_lock,
+ .irq_bus_sync_unlock = max8925_irq_sync_unlock,
+ .irq_enable = max8925_irq_enable,
+ .irq_disable = max8925_irq_disable,
};
static int max8925_irq_init(struct max8925_chip *chip, int irq,
diff --git a/drivers/mfd/max8998-irq.c b/drivers/mfd/max8998-irq.c
index 45bfe77..3903e1f 100644
--- a/drivers/mfd/max8998-irq.c
+++ b/drivers/mfd/max8998-irq.c
@@ -102,16 +102,16 @@
return &max8998_irqs[irq - max8998->irq_base];
}
-static void max8998_irq_lock(unsigned int irq)
+static void max8998_irq_lock(struct irq_data *data)
{
- struct max8998_dev *max8998 = get_irq_chip_data(irq);
+ struct max8998_dev *max8998 = irq_data_get_irq_chip_data(data);
mutex_lock(&max8998->irqlock);
}
-static void max8998_irq_sync_unlock(unsigned int irq)
+static void max8998_irq_sync_unlock(struct irq_data *data)
{
- struct max8998_dev *max8998 = get_irq_chip_data(irq);
+ struct max8998_dev *max8998 = irq_data_get_irq_chip_data(data);
int i;
for (i = 0; i < ARRAY_SIZE(max8998->irq_masks_cur); i++) {
@@ -129,28 +129,30 @@
mutex_unlock(&max8998->irqlock);
}
-static void max8998_irq_unmask(unsigned int irq)
+static void max8998_irq_unmask(struct irq_data *data)
{
- struct max8998_dev *max8998 = get_irq_chip_data(irq);
- struct max8998_irq_data *irq_data = irq_to_max8998_irq(max8998, irq);
+ struct max8998_dev *max8998 = irq_data_get_irq_chip_data(data);
+ struct max8998_irq_data *irq_data = irq_to_max8998_irq(max8998,
+ data->irq);
max8998->irq_masks_cur[irq_data->reg - 1] &= ~irq_data->mask;
}
-static void max8998_irq_mask(unsigned int irq)
+static void max8998_irq_mask(struct irq_data *data)
{
- struct max8998_dev *max8998 = get_irq_chip_data(irq);
- struct max8998_irq_data *irq_data = irq_to_max8998_irq(max8998, irq);
+ struct max8998_dev *max8998 = irq_data_get_irq_chip_data(data);
+ struct max8998_irq_data *irq_data = irq_to_max8998_irq(max8998,
+ data->irq);
max8998->irq_masks_cur[irq_data->reg - 1] |= irq_data->mask;
}
static struct irq_chip max8998_irq_chip = {
.name = "max8998",
- .bus_lock = max8998_irq_lock,
- .bus_sync_unlock = max8998_irq_sync_unlock,
- .mask = max8998_irq_mask,
- .unmask = max8998_irq_unmask,
+ .irq_bus_lock = max8998_irq_lock,
+ .irq_bus_sync_unlock = max8998_irq_sync_unlock,
+ .irq_mask = max8998_irq_mask,
+ .irq_unmask = max8998_irq_unmask,
};
static irqreturn_t max8998_irq_thread(int irq, void *data)
@@ -181,6 +183,13 @@
return IRQ_HANDLED;
}
+int max8998_irq_resume(struct max8998_dev *max8998)
+{
+ if (max8998->irq && max8998->irq_base)
+ max8998_irq_thread(max8998->irq_base, max8998);
+ return 0;
+}
+
int max8998_irq_init(struct max8998_dev *max8998)
{
int i;
diff --git a/drivers/mfd/max8998.c b/drivers/mfd/max8998.c
index bb9977b..bbfe867 100644
--- a/drivers/mfd/max8998.c
+++ b/drivers/mfd/max8998.c
@@ -25,6 +25,8 @@
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/i2c.h>
+#include <linux/interrupt.h>
+#include <linux/pm_runtime.h>
#include <linux/mutex.h>
#include <linux/mfd/core.h>
#include <linux/mfd/max8998.h>
@@ -40,6 +42,14 @@
},
};
+static struct mfd_cell lp3974_devs[] = {
+ {
+ .name = "lp3974-pmic",
+ }, {
+ .name = "lp3974-rtc",
+ },
+};
+
int max8998_read_reg(struct i2c_client *i2c, u8 reg, u8 *dest)
{
struct max8998_dev *max8998 = i2c_get_clientdata(i2c);
@@ -135,6 +145,7 @@
if (pdata) {
max8998->ono = pdata->ono;
max8998->irq_base = pdata->irq_base;
+ max8998->wakeup = pdata->wakeup;
}
mutex_init(&max8998->iolock);
@@ -143,9 +154,23 @@
max8998_irq_init(max8998);
- ret = mfd_add_devices(max8998->dev, -1,
- max8998_devs, ARRAY_SIZE(max8998_devs),
- NULL, 0);
+ pm_runtime_set_active(max8998->dev);
+
+ switch (id->driver_data) {
+ case TYPE_LP3974:
+ ret = mfd_add_devices(max8998->dev, -1,
+ lp3974_devs, ARRAY_SIZE(lp3974_devs),
+ NULL, 0);
+ break;
+ case TYPE_MAX8998:
+ ret = mfd_add_devices(max8998->dev, -1,
+ max8998_devs, ARRAY_SIZE(max8998_devs),
+ NULL, 0);
+ break;
+ default:
+ ret = -EINVAL;
+ }
+
if (ret < 0)
goto err;
@@ -178,10 +203,113 @@
};
MODULE_DEVICE_TABLE(i2c, max8998_i2c_id);
+static int max8998_suspend(struct device *dev)
+{
+ struct i2c_client *i2c = container_of(dev, struct i2c_client, dev);
+ struct max8998_dev *max8998 = i2c_get_clientdata(i2c);
+
+ if (max8998->wakeup)
+ set_irq_wake(max8998->irq, 1);
+ return 0;
+}
+
+static int max8998_resume(struct device *dev)
+{
+ struct i2c_client *i2c = container_of(dev, struct i2c_client, dev);
+ struct max8998_dev *max8998 = i2c_get_clientdata(i2c);
+
+ if (max8998->wakeup)
+ set_irq_wake(max8998->irq, 0);
+ /*
+ * In LP3974, if IRQ registers are not "read & clear"
+ * when it's set during sleep, the interrupt becomes
+ * disabled.
+ */
+ return max8998_irq_resume(i2c_get_clientdata(i2c));
+}
+
+struct max8998_reg_dump {
+ u8 addr;
+ u8 val;
+};
+#define SAVE_ITEM(x) { .addr = (x), .val = 0x0, }
+struct max8998_reg_dump max8998_dump[] = {
+ SAVE_ITEM(MAX8998_REG_IRQM1),
+ SAVE_ITEM(MAX8998_REG_IRQM2),
+ SAVE_ITEM(MAX8998_REG_IRQM3),
+ SAVE_ITEM(MAX8998_REG_IRQM4),
+ SAVE_ITEM(MAX8998_REG_STATUSM1),
+ SAVE_ITEM(MAX8998_REG_STATUSM2),
+ SAVE_ITEM(MAX8998_REG_CHGR1),
+ SAVE_ITEM(MAX8998_REG_CHGR2),
+ SAVE_ITEM(MAX8998_REG_LDO_ACTIVE_DISCHARGE1),
+ SAVE_ITEM(MAX8998_REG_LDO_ACTIVE_DISCHARGE1),
+ SAVE_ITEM(MAX8998_REG_BUCK_ACTIVE_DISCHARGE3),
+ SAVE_ITEM(MAX8998_REG_ONOFF1),
+ SAVE_ITEM(MAX8998_REG_ONOFF2),
+ SAVE_ITEM(MAX8998_REG_ONOFF3),
+ SAVE_ITEM(MAX8998_REG_ONOFF4),
+ SAVE_ITEM(MAX8998_REG_BUCK1_VOLTAGE1),
+ SAVE_ITEM(MAX8998_REG_BUCK1_VOLTAGE2),
+ SAVE_ITEM(MAX8998_REG_BUCK1_VOLTAGE3),
+ SAVE_ITEM(MAX8998_REG_BUCK1_VOLTAGE4),
+ SAVE_ITEM(MAX8998_REG_BUCK2_VOLTAGE1),
+ SAVE_ITEM(MAX8998_REG_BUCK2_VOLTAGE2),
+ SAVE_ITEM(MAX8998_REG_LDO2_LDO3),
+ SAVE_ITEM(MAX8998_REG_LDO4),
+ SAVE_ITEM(MAX8998_REG_LDO5),
+ SAVE_ITEM(MAX8998_REG_LDO6),
+ SAVE_ITEM(MAX8998_REG_LDO7),
+ SAVE_ITEM(MAX8998_REG_LDO8_LDO9),
+ SAVE_ITEM(MAX8998_REG_LDO10_LDO11),
+ SAVE_ITEM(MAX8998_REG_LDO12),
+ SAVE_ITEM(MAX8998_REG_LDO13),
+ SAVE_ITEM(MAX8998_REG_LDO14),
+ SAVE_ITEM(MAX8998_REG_LDO15),
+ SAVE_ITEM(MAX8998_REG_LDO16),
+ SAVE_ITEM(MAX8998_REG_LDO17),
+ SAVE_ITEM(MAX8998_REG_BKCHR),
+ SAVE_ITEM(MAX8998_REG_LBCNFG1),
+ SAVE_ITEM(MAX8998_REG_LBCNFG2),
+};
+/* Save registers before hibernation */
+static int max8998_freeze(struct device *dev)
+{
+ struct i2c_client *i2c = container_of(dev, struct i2c_client, dev);
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(max8998_dump); i++)
+ max8998_read_reg(i2c, max8998_dump[i].addr,
+ &max8998_dump[i].val);
+
+ return 0;
+}
+
+/* Restore registers after hibernation */
+static int max8998_restore(struct device *dev)
+{
+ struct i2c_client *i2c = container_of(dev, struct i2c_client, dev);
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(max8998_dump); i++)
+ max8998_write_reg(i2c, max8998_dump[i].addr,
+ max8998_dump[i].val);
+
+ return 0;
+}
+
+const struct dev_pm_ops max8998_pm = {
+ .suspend = max8998_suspend,
+ .resume = max8998_resume,
+ .freeze = max8998_freeze,
+ .restore = max8998_restore,
+};
+
static struct i2c_driver max8998_i2c_driver = {
.driver = {
.name = "max8998",
.owner = THIS_MODULE,
+ .pm = &max8998_pm,
},
.probe = max8998_i2c_probe,
.remove = max8998_i2c_remove,
diff --git a/drivers/mfd/mc13xxx-core.c b/drivers/mfd/mc13xxx-core.c
index a2ac2ed..b9fcaf0 100644
--- a/drivers/mfd/mc13xxx-core.c
+++ b/drivers/mfd/mc13xxx-core.c
@@ -749,7 +749,7 @@
if (ret) {
err_mask:
err_revision:
- mutex_unlock(&mc13xxx->lock);
+ mc13xxx_unlock(mc13xxx);
dev_set_drvdata(&spi->dev, NULL);
kfree(mc13xxx);
return ret;
diff --git a/drivers/mfd/mfd-core.c b/drivers/mfd/mfd-core.c
index ec99f68..d83ad0f 100644
--- a/drivers/mfd/mfd-core.c
+++ b/drivers/mfd/mfd-core.c
@@ -15,6 +15,7 @@
#include <linux/platform_device.h>
#include <linux/acpi.h>
#include <linux/mfd/core.h>
+#include <linux/pm_runtime.h>
#include <linux/slab.h>
static int mfd_add_device(struct device *parent, int id,
@@ -82,6 +83,9 @@
if (ret)
goto fail_res;
+ if (cell->pm_runtime_no_callbacks)
+ pm_runtime_no_callbacks(&pdev->dev);
+
kfree(res);
return 0;
diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index bc9275c..5de3a76 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -26,7 +26,7 @@
#include <linux/sm501-regs.h>
#include <linux/serial_8250.h>
-#include <asm/io.h>
+#include <linux/io.h>
struct sm501_device {
struct list_head list;
@@ -745,11 +745,8 @@
int ret;
for (ptr = 0; ptr < pdev->num_resources; ptr++) {
- printk(KERN_DEBUG "%s[%d] flags %08lx: %08llx..%08llx\n",
- pdev->name, ptr,
- pdev->resource[ptr].flags,
- (unsigned long long)pdev->resource[ptr].start,
- (unsigned long long)pdev->resource[ptr].end);
+ printk(KERN_DEBUG "%s[%d] %pR\n",
+ pdev->name, ptr, &pdev->resource[ptr]);
}
ret = platform_device_register(pdev);
diff --git a/drivers/mfd/stmpe.c b/drivers/mfd/stmpe.c
index b11487f..3e5732b 100644
--- a/drivers/mfd/stmpe.c
+++ b/drivers/mfd/stmpe.c
@@ -699,16 +699,16 @@
return IRQ_HANDLED;
}
-static void stmpe_irq_lock(unsigned int irq)
+static void stmpe_irq_lock(struct irq_data *data)
{
- struct stmpe *stmpe = get_irq_chip_data(irq);
+ struct stmpe *stmpe = irq_data_get_irq_chip_data(data);
mutex_lock(&stmpe->irq_lock);
}
-static void stmpe_irq_sync_unlock(unsigned int irq)
+static void stmpe_irq_sync_unlock(struct irq_data *data)
{
- struct stmpe *stmpe = get_irq_chip_data(irq);
+ struct stmpe *stmpe = irq_data_get_irq_chip_data(data);
struct stmpe_variant_info *variant = stmpe->variant;
int num = DIV_ROUND_UP(variant->num_irqs, 8);
int i;
@@ -727,20 +727,20 @@
mutex_unlock(&stmpe->irq_lock);
}
-static void stmpe_irq_mask(unsigned int irq)
+static void stmpe_irq_mask(struct irq_data *data)
{
- struct stmpe *stmpe = get_irq_chip_data(irq);
- int offset = irq - stmpe->irq_base;
+ struct stmpe *stmpe = irq_data_get_irq_chip_data(data);
+ int offset = data->irq - stmpe->irq_base;
int regoffset = offset / 8;
int mask = 1 << (offset % 8);
stmpe->ier[regoffset] &= ~mask;
}
-static void stmpe_irq_unmask(unsigned int irq)
+static void stmpe_irq_unmask(struct irq_data *data)
{
- struct stmpe *stmpe = get_irq_chip_data(irq);
- int offset = irq - stmpe->irq_base;
+ struct stmpe *stmpe = irq_data_get_irq_chip_data(data);
+ int offset = data->irq - stmpe->irq_base;
int regoffset = offset / 8;
int mask = 1 << (offset % 8);
@@ -749,10 +749,10 @@
static struct irq_chip stmpe_irq_chip = {
.name = "stmpe",
- .bus_lock = stmpe_irq_lock,
- .bus_sync_unlock = stmpe_irq_sync_unlock,
- .mask = stmpe_irq_mask,
- .unmask = stmpe_irq_unmask,
+ .irq_bus_lock = stmpe_irq_lock,
+ .irq_bus_sync_unlock = stmpe_irq_sync_unlock,
+ .irq_mask = stmpe_irq_mask,
+ .irq_unmask = stmpe_irq_unmask,
};
static int __devinit stmpe_irq_init(struct stmpe *stmpe)
diff --git a/drivers/mfd/t7l66xb.c b/drivers/mfd/t7l66xb.c
index 006c121..9caeb4a 100644
--- a/drivers/mfd/t7l66xb.c
+++ b/drivers/mfd/t7l66xb.c
@@ -199,37 +199,37 @@
generic_handle_irq(irq_base + i);
}
-static void t7l66xb_irq_mask(unsigned int irq)
+static void t7l66xb_irq_mask(struct irq_data *data)
{
- struct t7l66xb *t7l66xb = get_irq_chip_data(irq);
+ struct t7l66xb *t7l66xb = irq_data_get_irq_chip_data(data);
unsigned long flags;
u8 imr;
spin_lock_irqsave(&t7l66xb->lock, flags);
imr = tmio_ioread8(t7l66xb->scr + SCR_IMR);
- imr |= 1 << (irq - t7l66xb->irq_base);
+ imr |= 1 << (data->irq - t7l66xb->irq_base);
tmio_iowrite8(imr, t7l66xb->scr + SCR_IMR);
spin_unlock_irqrestore(&t7l66xb->lock, flags);
}
-static void t7l66xb_irq_unmask(unsigned int irq)
+static void t7l66xb_irq_unmask(struct irq_data *data)
{
- struct t7l66xb *t7l66xb = get_irq_chip_data(irq);
+ struct t7l66xb *t7l66xb = irq_data_get_irq_chip_data(data);
unsigned long flags;
u8 imr;
spin_lock_irqsave(&t7l66xb->lock, flags);
imr = tmio_ioread8(t7l66xb->scr + SCR_IMR);
- imr &= ~(1 << (irq - t7l66xb->irq_base));
+ imr &= ~(1 << (data->irq - t7l66xb->irq_base));
tmio_iowrite8(imr, t7l66xb->scr + SCR_IMR);
spin_unlock_irqrestore(&t7l66xb->lock, flags);
}
static struct irq_chip t7l66xb_chip = {
- .name = "t7l66xb",
- .ack = t7l66xb_irq_mask,
- .mask = t7l66xb_irq_mask,
- .unmask = t7l66xb_irq_unmask,
+ .name = "t7l66xb",
+ .irq_ack = t7l66xb_irq_mask,
+ .irq_mask = t7l66xb_irq_mask,
+ .irq_unmask = t7l66xb_irq_unmask,
};
/*--------------------------------------------------------------------------*/
diff --git a/drivers/mfd/tc6393xb.c b/drivers/mfd/tc6393xb.c
index 1ea80d8a..9a23863 100644
--- a/drivers/mfd/tc6393xb.c
+++ b/drivers/mfd/tc6393xb.c
@@ -527,41 +527,41 @@
}
}
-static void tc6393xb_irq_ack(unsigned int irq)
+static void tc6393xb_irq_ack(struct irq_data *data)
{
}
-static void tc6393xb_irq_mask(unsigned int irq)
+static void tc6393xb_irq_mask(struct irq_data *data)
{
- struct tc6393xb *tc6393xb = get_irq_chip_data(irq);
+ struct tc6393xb *tc6393xb = irq_data_get_irq_chip_data(data);
unsigned long flags;
u8 imr;
spin_lock_irqsave(&tc6393xb->lock, flags);
imr = tmio_ioread8(tc6393xb->scr + SCR_IMR);
- imr |= 1 << (irq - tc6393xb->irq_base);
+ imr |= 1 << (data->irq - tc6393xb->irq_base);
tmio_iowrite8(imr, tc6393xb->scr + SCR_IMR);
spin_unlock_irqrestore(&tc6393xb->lock, flags);
}
-static void tc6393xb_irq_unmask(unsigned int irq)
+static void tc6393xb_irq_unmask(struct irq_data *data)
{
- struct tc6393xb *tc6393xb = get_irq_chip_data(irq);
+ struct tc6393xb *tc6393xb = irq_data_get_irq_chip_data(data);
unsigned long flags;
u8 imr;
spin_lock_irqsave(&tc6393xb->lock, flags);
imr = tmio_ioread8(tc6393xb->scr + SCR_IMR);
- imr &= ~(1 << (irq - tc6393xb->irq_base));
+ imr &= ~(1 << (data->irq - tc6393xb->irq_base));
tmio_iowrite8(imr, tc6393xb->scr + SCR_IMR);
spin_unlock_irqrestore(&tc6393xb->lock, flags);
}
static struct irq_chip tc6393xb_chip = {
- .name = "tc6393xb",
- .ack = tc6393xb_irq_ack,
- .mask = tc6393xb_irq_mask,
- .unmask = tc6393xb_irq_unmask,
+ .name = "tc6393xb",
+ .irq_ack = tc6393xb_irq_ack,
+ .irq_mask = tc6393xb_irq_mask,
+ .irq_unmask = tc6393xb_irq_unmask,
};
static void tc6393xb_attach_irq(struct platform_device *dev)
diff --git a/drivers/mfd/tps65010.c b/drivers/mfd/tps65010.c
index 90187fe..93d5fdf 100644
--- a/drivers/mfd/tps65010.c
+++ b/drivers/mfd/tps65010.c
@@ -34,7 +34,7 @@
#include <linux/i2c/tps65010.h>
-#include <asm/gpio.h>
+#include <linux/gpio.h>
/*-------------------------------------------------------------------------*/
diff --git a/drivers/mfd/tps6586x.c b/drivers/mfd/tps6586x.c
index b4931ab..627cf57 100644
--- a/drivers/mfd/tps6586x.c
+++ b/drivers/mfd/tps6586x.c
@@ -46,8 +46,6 @@
/* device id */
#define TPS6586X_VERSIONCRC 0xcd
-#define TPS658621A_VERSIONCRC 0x15
-#define TPS658621C_VERSIONCRC 0x2c
struct tps6586x_irq_data {
u8 mask_reg;
@@ -325,37 +323,37 @@
return device_for_each_child(tps6586x->dev, NULL, __remove_subdev);
}
-static void tps6586x_irq_lock(unsigned int irq)
+static void tps6586x_irq_lock(struct irq_data *data)
{
- struct tps6586x *tps6586x = get_irq_chip_data(irq);
+ struct tps6586x *tps6586x = irq_data_get_irq_chip_data(data);
mutex_lock(&tps6586x->irq_lock);
}
-static void tps6586x_irq_enable(unsigned int irq)
+static void tps6586x_irq_enable(struct irq_data *irq_data)
{
- struct tps6586x *tps6586x = get_irq_chip_data(irq);
- unsigned int __irq = irq - tps6586x->irq_base;
+ struct tps6586x *tps6586x = irq_data_get_irq_chip_data(irq_data);
+ unsigned int __irq = irq_data->irq - tps6586x->irq_base;
const struct tps6586x_irq_data *data = &tps6586x_irqs[__irq];
tps6586x->mask_reg[data->mask_reg] &= ~data->mask_mask;
tps6586x->irq_en |= (1 << __irq);
}
-static void tps6586x_irq_disable(unsigned int irq)
+static void tps6586x_irq_disable(struct irq_data *irq_data)
{
- struct tps6586x *tps6586x = get_irq_chip_data(irq);
+ struct tps6586x *tps6586x = irq_data_get_irq_chip_data(irq_data);
- unsigned int __irq = irq - tps6586x->irq_base;
+ unsigned int __irq = irq_data->irq - tps6586x->irq_base;
const struct tps6586x_irq_data *data = &tps6586x_irqs[__irq];
tps6586x->mask_reg[data->mask_reg] |= data->mask_mask;
tps6586x->irq_en &= ~(1 << __irq);
}
-static void tps6586x_irq_sync_unlock(unsigned int irq)
+static void tps6586x_irq_sync_unlock(struct irq_data *data)
{
- struct tps6586x *tps6586x = get_irq_chip_data(irq);
+ struct tps6586x *tps6586x = irq_data_get_irq_chip_data(data);
int i;
for (i = 0; i < ARRAY_SIZE(tps6586x->mask_reg); i++) {
@@ -421,10 +419,10 @@
tps6586x->irq_base = irq_base;
tps6586x->irq_chip.name = "tps6586x";
- tps6586x->irq_chip.enable = tps6586x_irq_enable;
- tps6586x->irq_chip.disable = tps6586x_irq_disable;
- tps6586x->irq_chip.bus_lock = tps6586x_irq_lock;
- tps6586x->irq_chip.bus_sync_unlock = tps6586x_irq_sync_unlock;
+ tps6586x->irq_chip.irq_enable = tps6586x_irq_enable;
+ tps6586x->irq_chip.irq_disable = tps6586x_irq_disable;
+ tps6586x->irq_chip.irq_bus_lock = tps6586x_irq_lock;
+ tps6586x->irq_chip.irq_bus_sync_unlock = tps6586x_irq_sync_unlock;
for (i = 0; i < ARRAY_SIZE(tps6586x_irqs); i++) {
int __irq = i + tps6586x->irq_base;
@@ -498,11 +496,7 @@
return -EIO;
}
- if ((ret != TPS658621A_VERSIONCRC) &&
- (ret != TPS658621C_VERSIONCRC)) {
- dev_err(&client->dev, "Unsupported chip ID: %x\n", ret);
- return -ENODEV;
- }
+ dev_info(&client->dev, "VERSIONCRC is %02x\n", ret);
tps6586x = kzalloc(sizeof(struct tps6586x), GFP_KERNEL);
if (tps6586x == NULL)
diff --git a/drivers/mfd/twl-core.c b/drivers/mfd/twl-core.c
index 12abd5b..a35fa7dc 100644
--- a/drivers/mfd/twl-core.c
+++ b/drivers/mfd/twl-core.c
@@ -1003,7 +1003,7 @@
}
/* NOTE: this driver only handles a single twl4030/tps659x0 chip */
-static int __init
+static int __devinit
twl_probe(struct i2c_client *client, const struct i2c_device_id *id)
{
int status;
diff --git a/drivers/mfd/twl4030-irq.c b/drivers/mfd/twl4030-irq.c
index 5d3a147..63a30e8 100644
--- a/drivers/mfd/twl4030-irq.c
+++ b/drivers/mfd/twl4030-irq.c
@@ -599,38 +599,38 @@
* completion, potentially including some re-ordering, of these requests.
*/
-static void twl4030_sih_mask(unsigned irq)
+static void twl4030_sih_mask(struct irq_data *data)
{
- struct sih_agent *sih = get_irq_chip_data(irq);
+ struct sih_agent *sih = irq_data_get_irq_chip_data(data);
unsigned long flags;
spin_lock_irqsave(&sih_agent_lock, flags);
- sih->imr |= BIT(irq - sih->irq_base);
+ sih->imr |= BIT(data->irq - sih->irq_base);
sih->imr_change_pending = true;
queue_work(wq, &sih->mask_work);
spin_unlock_irqrestore(&sih_agent_lock, flags);
}
-static void twl4030_sih_unmask(unsigned irq)
+static void twl4030_sih_unmask(struct irq_data *data)
{
- struct sih_agent *sih = get_irq_chip_data(irq);
+ struct sih_agent *sih = irq_data_get_irq_chip_data(data);
unsigned long flags;
spin_lock_irqsave(&sih_agent_lock, flags);
- sih->imr &= ~BIT(irq - sih->irq_base);
+ sih->imr &= ~BIT(data->irq - sih->irq_base);
sih->imr_change_pending = true;
queue_work(wq, &sih->mask_work);
spin_unlock_irqrestore(&sih_agent_lock, flags);
}
-static int twl4030_sih_set_type(unsigned irq, unsigned trigger)
+static int twl4030_sih_set_type(struct irq_data *data, unsigned trigger)
{
- struct sih_agent *sih = get_irq_chip_data(irq);
- struct irq_desc *desc = irq_to_desc(irq);
+ struct sih_agent *sih = irq_data_get_irq_chip_data(data);
+ struct irq_desc *desc = irq_to_desc(data->irq);
unsigned long flags;
if (!desc) {
- pr_err("twl4030: Invalid IRQ: %d\n", irq);
+ pr_err("twl4030: Invalid IRQ: %d\n", data->irq);
return -EINVAL;
}
@@ -641,7 +641,7 @@
if ((desc->status & IRQ_TYPE_SENSE_MASK) != trigger) {
desc->status &= ~IRQ_TYPE_SENSE_MASK;
desc->status |= trigger;
- sih->edge_change |= BIT(irq - sih->irq_base);
+ sih->edge_change |= BIT(data->irq - sih->irq_base);
queue_work(wq, &sih->edge_work);
}
spin_unlock_irqrestore(&sih_agent_lock, flags);
@@ -650,9 +650,9 @@
static struct irq_chip twl4030_sih_irq_chip = {
.name = "twl4030",
- .mask = twl4030_sih_mask,
- .unmask = twl4030_sih_unmask,
- .set_type = twl4030_sih_set_type,
+ .irq_mask = twl4030_sih_mask,
+ .irq_unmask = twl4030_sih_unmask,
+ .irq_set_type = twl4030_sih_set_type,
};
/*----------------------------------------------------------------------*/
diff --git a/drivers/mfd/twl6030-irq.c b/drivers/mfd/twl6030-irq.c
index 06c8955..4082ed7 100644
--- a/drivers/mfd/twl6030-irq.c
+++ b/drivers/mfd/twl6030-irq.c
@@ -332,7 +332,7 @@
*/
twl6030_irq_chip = dummy_irq_chip;
twl6030_irq_chip.name = "twl6030";
- twl6030_irq_chip.set_type = NULL;
+ twl6030_irq_chip.irq_set_type = NULL;
for (i = irq_base; i < irq_end; i++) {
set_irq_chip_and_handler(i, &twl6030_irq_chip,
diff --git a/drivers/mfd/vx855.c b/drivers/mfd/vx855.c
index ebb0597..348052a 100644
--- a/drivers/mfd/vx855.c
+++ b/drivers/mfd/vx855.c
@@ -112,7 +112,7 @@
return ret;
}
-static void vx855_remove(struct pci_dev *pdev)
+static void __devexit vx855_remove(struct pci_dev *pdev)
{
mfd_remove_devices(&pdev->dev);
pci_disable_device(pdev);
diff --git a/drivers/mfd/wm831x-core.c b/drivers/mfd/wm831x-core.c
index 76cadcf..3fe9a58 100644
--- a/drivers/mfd/wm831x-core.c
+++ b/drivers/mfd/wm831x-core.c
@@ -1541,6 +1541,12 @@
dev_info(wm831x->dev, "WM8325 revision %c\n", 'A' + rev);
break;
+ case WM8326:
+ parent = WM8326;
+ wm831x->num_gpio = 12;
+ dev_info(wm831x->dev, "WM8326 revision %c\n", 'A' + rev);
+ break;
+
default:
dev_err(wm831x->dev, "Unknown WM831x device %04x\n", ret);
ret = -EINVAL;
@@ -1610,18 +1616,9 @@
break;
case WM8320:
- ret = mfd_add_devices(wm831x->dev, -1,
- wm8320_devs, ARRAY_SIZE(wm8320_devs),
- NULL, 0);
- break;
-
case WM8321:
- ret = mfd_add_devices(wm831x->dev, -1,
- wm8320_devs, ARRAY_SIZE(wm8320_devs),
- NULL, 0);
- break;
-
case WM8325:
+ case WM8326:
ret = mfd_add_devices(wm831x->dev, -1,
wm8320_devs, ARRAY_SIZE(wm8320_devs),
NULL, wm831x->irq_base);
diff --git a/drivers/mfd/wm831x-i2c.c b/drivers/mfd/wm831x-i2c.c
index 156b198..3853fa8 100644
--- a/drivers/mfd/wm831x-i2c.c
+++ b/drivers/mfd/wm831x-i2c.c
@@ -94,9 +94,9 @@
return 0;
}
-static int wm831x_i2c_suspend(struct i2c_client *i2c, pm_message_t mesg)
+static int wm831x_i2c_suspend(struct device *dev)
{
- struct wm831x *wm831x = i2c_get_clientdata(i2c);
+ struct wm831x *wm831x = dev_get_drvdata(dev);
return wm831x_device_suspend(wm831x);
}
@@ -108,19 +108,23 @@
{ "wm8320", WM8320 },
{ "wm8321", WM8321 },
{ "wm8325", WM8325 },
+ { "wm8326", WM8326 },
{ }
};
MODULE_DEVICE_TABLE(i2c, wm831x_i2c_id);
+static const struct dev_pm_ops wm831x_pm_ops = {
+ .suspend = wm831x_i2c_suspend,
+};
static struct i2c_driver wm831x_i2c_driver = {
.driver = {
- .name = "wm831x",
- .owner = THIS_MODULE,
+ .name = "wm831x",
+ .owner = THIS_MODULE,
+ .pm = &wm831x_pm_ops,
},
.probe = wm831x_i2c_probe,
.remove = wm831x_i2c_remove,
- .suspend = wm831x_i2c_suspend,
.id_table = wm831x_i2c_id,
};
diff --git a/drivers/mfd/wm831x-irq.c b/drivers/mfd/wm831x-irq.c
index 294183b..f7192d4 100644
--- a/drivers/mfd/wm831x-irq.c
+++ b/drivers/mfd/wm831x-irq.c
@@ -345,16 +345,16 @@
return &wm831x_irqs[irq - wm831x->irq_base];
}
-static void wm831x_irq_lock(unsigned int irq)
+static void wm831x_irq_lock(struct irq_data *data)
{
- struct wm831x *wm831x = get_irq_chip_data(irq);
+ struct wm831x *wm831x = irq_data_get_irq_chip_data(data);
mutex_lock(&wm831x->irq_lock);
}
-static void wm831x_irq_sync_unlock(unsigned int irq)
+static void wm831x_irq_sync_unlock(struct irq_data *data)
{
- struct wm831x *wm831x = get_irq_chip_data(irq);
+ struct wm831x *wm831x = irq_data_get_irq_chip_data(data);
int i;
for (i = 0; i < ARRAY_SIZE(wm831x->irq_masks_cur); i++) {
@@ -371,28 +371,30 @@
mutex_unlock(&wm831x->irq_lock);
}
-static void wm831x_irq_unmask(unsigned int irq)
+static void wm831x_irq_unmask(struct irq_data *data)
{
- struct wm831x *wm831x = get_irq_chip_data(irq);
- struct wm831x_irq_data *irq_data = irq_to_wm831x_irq(wm831x, irq);
+ struct wm831x *wm831x = irq_data_get_irq_chip_data(data);
+ struct wm831x_irq_data *irq_data = irq_to_wm831x_irq(wm831x,
+ data->irq);
wm831x->irq_masks_cur[irq_data->reg - 1] &= ~irq_data->mask;
}
-static void wm831x_irq_mask(unsigned int irq)
+static void wm831x_irq_mask(struct irq_data *data)
{
- struct wm831x *wm831x = get_irq_chip_data(irq);
- struct wm831x_irq_data *irq_data = irq_to_wm831x_irq(wm831x, irq);
+ struct wm831x *wm831x = irq_data_get_irq_chip_data(data);
+ struct wm831x_irq_data *irq_data = irq_to_wm831x_irq(wm831x,
+ data->irq);
wm831x->irq_masks_cur[irq_data->reg - 1] |= irq_data->mask;
}
-static int wm831x_irq_set_type(unsigned int irq, unsigned int type)
+static int wm831x_irq_set_type(struct irq_data *data, unsigned int type)
{
- struct wm831x *wm831x = get_irq_chip_data(irq);
- int val;
+ struct wm831x *wm831x = irq_data_get_irq_chip_data(data);
+ int val, irq;
- irq = irq - wm831x->irq_base;
+ irq = data->irq - wm831x->irq_base;
if (irq < WM831X_IRQ_GPIO_1 || irq > WM831X_IRQ_GPIO_11) {
/* Ignore internal-only IRQs */
@@ -421,12 +423,12 @@
}
static struct irq_chip wm831x_irq_chip = {
- .name = "wm831x",
- .bus_lock = wm831x_irq_lock,
- .bus_sync_unlock = wm831x_irq_sync_unlock,
- .mask = wm831x_irq_mask,
- .unmask = wm831x_irq_unmask,
- .set_type = wm831x_irq_set_type,
+ .name = "wm831x",
+ .irq_bus_lock = wm831x_irq_lock,
+ .irq_bus_sync_unlock = wm831x_irq_sync_unlock,
+ .irq_mask = wm831x_irq_mask,
+ .irq_unmask = wm831x_irq_unmask,
+ .irq_set_type = wm831x_irq_set_type,
};
/* The processing of the primary interrupt occurs in a thread so that
@@ -515,6 +517,17 @@
return 0;
}
+ /* Try to flag /IRQ as a wake source; there are a number of
+ * unconditional wake sources in the PMIC so this isn't
+ * conditional but we don't actually care *too* much if it
+ * fails.
+ */
+ ret = enable_irq_wake(irq);
+ if (ret != 0) {
+ dev_warn(wm831x->dev, "Can't enable IRQ as wake source: %d\n",
+ ret);
+ }
+
wm831x->irq = irq;
wm831x->irq_base = pdata->irq_base;
diff --git a/drivers/mfd/wm831x-spi.c b/drivers/mfd/wm831x-spi.c
index 2789b15..0a8f772 100644
--- a/drivers/mfd/wm831x-spi.c
+++ b/drivers/mfd/wm831x-spi.c
@@ -81,6 +81,8 @@
type = WM8321;
else if (strcmp(spi->modalias, "wm8325") == 0)
type = WM8325;
+ else if (strcmp(spi->modalias, "wm8326") == 0)
+ type = WM8326;
else {
dev_err(&spi->dev, "Unknown device type\n");
return -EINVAL;
@@ -184,6 +186,17 @@
.suspend = wm831x_spi_suspend,
};
+static struct spi_driver wm8326_spi_driver = {
+ .driver = {
+ .name = "wm8326",
+ .bus = &spi_bus_type,
+ .owner = THIS_MODULE,
+ },
+ .probe = wm831x_spi_probe,
+ .remove = __devexit_p(wm831x_spi_remove),
+ .suspend = wm831x_spi_suspend,
+};
+
static int __init wm831x_spi_init(void)
{
int ret;
@@ -212,12 +225,17 @@
if (ret != 0)
pr_err("Failed to register WM8325 SPI driver: %d\n", ret);
+ ret = spi_register_driver(&wm8326_spi_driver);
+ if (ret != 0)
+ pr_err("Failed to register WM8326 SPI driver: %d\n", ret);
+
return 0;
}
subsys_initcall(wm831x_spi_init);
static void __exit wm831x_spi_exit(void)
{
+ spi_unregister_driver(&wm8326_spi_driver);
spi_unregister_driver(&wm8325_spi_driver);
spi_unregister_driver(&wm8321_spi_driver);
spi_unregister_driver(&wm8320_spi_driver);
diff --git a/drivers/mfd/wm8350-irq.c b/drivers/mfd/wm8350-irq.c
index f56c9ad..5839966 100644
--- a/drivers/mfd/wm8350-irq.c
+++ b/drivers/mfd/wm8350-irq.c
@@ -417,16 +417,16 @@
return IRQ_HANDLED;
}
-static void wm8350_irq_lock(unsigned int irq)
+static void wm8350_irq_lock(struct irq_data *data)
{
- struct wm8350 *wm8350 = get_irq_chip_data(irq);
+ struct wm8350 *wm8350 = irq_data_get_irq_chip_data(data);
mutex_lock(&wm8350->irq_lock);
}
-static void wm8350_irq_sync_unlock(unsigned int irq)
+static void wm8350_irq_sync_unlock(struct irq_data *data)
{
- struct wm8350 *wm8350 = get_irq_chip_data(irq);
+ struct wm8350 *wm8350 = irq_data_get_irq_chip_data(data);
int i;
for (i = 0; i < ARRAY_SIZE(wm8350->irq_masks); i++) {
@@ -442,28 +442,30 @@
mutex_unlock(&wm8350->irq_lock);
}
-static void wm8350_irq_enable(unsigned int irq)
+static void wm8350_irq_enable(struct irq_data *data)
{
- struct wm8350 *wm8350 = get_irq_chip_data(irq);
- struct wm8350_irq_data *irq_data = irq_to_wm8350_irq(wm8350, irq);
+ struct wm8350 *wm8350 = irq_data_get_irq_chip_data(data);
+ struct wm8350_irq_data *irq_data = irq_to_wm8350_irq(wm8350,
+ data->irq);
wm8350->irq_masks[irq_data->reg] &= ~irq_data->mask;
}
-static void wm8350_irq_disable(unsigned int irq)
+static void wm8350_irq_disable(struct irq_data *data)
{
- struct wm8350 *wm8350 = get_irq_chip_data(irq);
- struct wm8350_irq_data *irq_data = irq_to_wm8350_irq(wm8350, irq);
+ struct wm8350 *wm8350 = irq_data_get_irq_chip_data(data);
+ struct wm8350_irq_data *irq_data = irq_to_wm8350_irq(wm8350,
+ data->irq);
wm8350->irq_masks[irq_data->reg] |= irq_data->mask;
}
static struct irq_chip wm8350_irq_chip = {
- .name = "wm8350",
- .bus_lock = wm8350_irq_lock,
- .bus_sync_unlock = wm8350_irq_sync_unlock,
- .disable = wm8350_irq_disable,
- .enable = wm8350_irq_enable,
+ .name = "wm8350",
+ .irq_bus_lock = wm8350_irq_lock,
+ .irq_bus_sync_unlock = wm8350_irq_sync_unlock,
+ .irq_disable = wm8350_irq_disable,
+ .irq_enable = wm8350_irq_enable,
};
int wm8350_irq_init(struct wm8350 *wm8350, int irq,
diff --git a/drivers/mfd/wm8994-core.c b/drivers/mfd/wm8994-core.c
index 8d221ba..41233c7 100644
--- a/drivers/mfd/wm8994-core.c
+++ b/drivers/mfd/wm8994-core.c
@@ -18,6 +18,7 @@
#include <linux/i2c.h>
#include <linux/delay.h>
#include <linux/mfd/core.h>
+#include <linux/pm_runtime.h>
#include <linux/regulator/consumer.h>
#include <linux/regulator/machine.h>
@@ -169,8 +170,16 @@
EXPORT_SYMBOL_GPL(wm8994_set_bits);
static struct mfd_cell wm8994_regulator_devs[] = {
- { .name = "wm8994-ldo", .id = 1 },
- { .name = "wm8994-ldo", .id = 2 },
+ {
+ .name = "wm8994-ldo",
+ .id = 1,
+ .pm_runtime_no_callbacks = true,
+ },
+ {
+ .name = "wm8994-ldo",
+ .id = 2,
+ .pm_runtime_no_callbacks = true,
+ },
};
static struct resource wm8994_codec_resources[] = {
@@ -200,6 +209,7 @@
.name = "wm8994-gpio",
.num_resources = ARRAY_SIZE(wm8994_gpio_resources),
.resources = wm8994_gpio_resources,
+ .pm_runtime_no_callbacks = true,
},
};
@@ -231,7 +241,7 @@
};
#ifdef CONFIG_PM
-static int wm8994_device_suspend(struct device *dev)
+static int wm8994_suspend(struct device *dev)
{
struct wm8994 *wm8994 = dev_get_drvdata(dev);
int ret;
@@ -261,7 +271,7 @@
return 0;
}
-static int wm8994_device_resume(struct device *dev)
+static int wm8994_resume(struct device *dev)
{
struct wm8994 *wm8994 = dev_get_drvdata(dev);
int ret;
@@ -471,6 +481,9 @@
goto err_irq;
}
+ pm_runtime_enable(wm8994->dev);
+ pm_runtime_resume(wm8994->dev);
+
return 0;
err_irq:
@@ -490,6 +503,7 @@
static void wm8994_device_exit(struct wm8994 *wm8994)
{
+ pm_runtime_disable(wm8994->dev);
mfd_remove_devices(wm8994->dev);
wm8994_irq_exit(wm8994);
regulator_bulk_disable(wm8994->num_supplies,
@@ -573,21 +587,6 @@
return 0;
}
-#ifdef CONFIG_PM
-static int wm8994_i2c_suspend(struct i2c_client *i2c, pm_message_t state)
-{
- return wm8994_device_suspend(&i2c->dev);
-}
-
-static int wm8994_i2c_resume(struct i2c_client *i2c)
-{
- return wm8994_device_resume(&i2c->dev);
-}
-#else
-#define wm8994_i2c_suspend NULL
-#define wm8994_i2c_resume NULL
-#endif
-
static const struct i2c_device_id wm8994_i2c_id[] = {
{ "wm8994", WM8994 },
{ "wm8958", WM8958 },
@@ -595,15 +594,16 @@
};
MODULE_DEVICE_TABLE(i2c, wm8994_i2c_id);
+UNIVERSAL_DEV_PM_OPS(wm8994_pm_ops, wm8994_suspend, wm8994_resume, NULL);
+
static struct i2c_driver wm8994_i2c_driver = {
.driver = {
- .name = "wm8994",
- .owner = THIS_MODULE,
+ .name = "wm8994",
+ .owner = THIS_MODULE,
+ .pm = &wm8994_pm_ops,
},
.probe = wm8994_i2c_probe,
.remove = wm8994_i2c_remove,
- .suspend = wm8994_i2c_suspend,
- .resume = wm8994_i2c_resume,
.id_table = wm8994_i2c_id,
};
diff --git a/drivers/mfd/wm8994-irq.c b/drivers/mfd/wm8994-irq.c
index 8400eb1..29e8faf 100644
--- a/drivers/mfd/wm8994-irq.c
+++ b/drivers/mfd/wm8994-irq.c
@@ -156,16 +156,16 @@
return &wm8994_irqs[irq - wm8994->irq_base];
}
-static void wm8994_irq_lock(unsigned int irq)
+static void wm8994_irq_lock(struct irq_data *data)
{
- struct wm8994 *wm8994 = get_irq_chip_data(irq);
+ struct wm8994 *wm8994 = irq_data_get_irq_chip_data(data);
mutex_lock(&wm8994->irq_lock);
}
-static void wm8994_irq_sync_unlock(unsigned int irq)
+static void wm8994_irq_sync_unlock(struct irq_data *data)
{
- struct wm8994 *wm8994 = get_irq_chip_data(irq);
+ struct wm8994 *wm8994 = irq_data_get_irq_chip_data(data);
int i;
for (i = 0; i < ARRAY_SIZE(wm8994->irq_masks_cur); i++) {
@@ -182,28 +182,30 @@
mutex_unlock(&wm8994->irq_lock);
}
-static void wm8994_irq_unmask(unsigned int irq)
+static void wm8994_irq_unmask(struct irq_data *data)
{
- struct wm8994 *wm8994 = get_irq_chip_data(irq);
- struct wm8994_irq_data *irq_data = irq_to_wm8994_irq(wm8994, irq);
+ struct wm8994 *wm8994 = irq_data_get_irq_chip_data(data);
+ struct wm8994_irq_data *irq_data = irq_to_wm8994_irq(wm8994,
+ data->irq);
wm8994->irq_masks_cur[irq_data->reg - 1] &= ~irq_data->mask;
}
-static void wm8994_irq_mask(unsigned int irq)
+static void wm8994_irq_mask(struct irq_data *data)
{
- struct wm8994 *wm8994 = get_irq_chip_data(irq);
- struct wm8994_irq_data *irq_data = irq_to_wm8994_irq(wm8994, irq);
+ struct wm8994 *wm8994 = irq_data_get_irq_chip_data(data);
+ struct wm8994_irq_data *irq_data = irq_to_wm8994_irq(wm8994,
+ data->irq);
wm8994->irq_masks_cur[irq_data->reg - 1] |= irq_data->mask;
}
static struct irq_chip wm8994_irq_chip = {
- .name = "wm8994",
- .bus_lock = wm8994_irq_lock,
- .bus_sync_unlock = wm8994_irq_sync_unlock,
- .mask = wm8994_irq_mask,
- .unmask = wm8994_irq_unmask,
+ .name = "wm8994",
+ .irq_bus_lock = wm8994_irq_lock,
+ .irq_bus_sync_unlock = wm8994_irq_sync_unlock,
+ .irq_mask = wm8994_irq_mask,
+ .irq_unmask = wm8994_irq_unmask,
};
/* The processing of the primary interrupt occurs in a thread so that
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 1e1a4be..cc8e49d 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -64,7 +64,7 @@
config AB8500_PWM
bool "AB8500 PWM support"
- depends on AB8500_CORE
+ depends on AB8500_CORE && ARCH_U8500
select HAVE_PWM
help
This driver exports functions to enable/disble/config/free Pulse
diff --git a/drivers/misc/cs5535-mfgpt.c b/drivers/misc/cs5535-mfgpt.c
index 6f62180..d02d302 100644
--- a/drivers/misc/cs5535-mfgpt.c
+++ b/drivers/misc/cs5535-mfgpt.c
@@ -16,12 +16,11 @@
#include <linux/spinlock.h>
#include <linux/interrupt.h>
#include <linux/module.h>
-#include <linux/pci.h>
+#include <linux/platform_device.h>
#include <linux/cs5535.h>
#include <linux/slab.h>
#define DRV_NAME "cs5535-mfgpt"
-#define MFGPT_BAR 2
static int mfgpt_reset_timers;
module_param_named(mfgptfix, mfgpt_reset_timers, int, 0644);
@@ -37,7 +36,7 @@
DECLARE_BITMAP(avail, MFGPT_MAX_TIMERS);
resource_size_t base;
- struct pci_dev *pdev;
+ struct platform_device *pdev;
spinlock_t lock;
int initialized;
} cs5535_mfgpt_chip;
@@ -290,10 +289,10 @@
return timers;
}
-static int __init cs5535_mfgpt_probe(struct pci_dev *pdev,
- const struct pci_device_id *pci_id)
+static int __devinit cs5535_mfgpt_probe(struct platform_device *pdev)
{
- int err, t;
+ struct resource *res;
+ int err = -EIO, t;
/* There are two ways to get the MFGPT base address; one is by
* fetching it from MSR_LBAR_MFGPT, the other is by reading the
@@ -302,29 +301,27 @@
* it turns out to be unreliable in the face of crappy BIOSes, we
* can always go back to using MSRs.. */
- err = pci_enable_device_io(pdev);
- if (err) {
- dev_err(&pdev->dev, "can't enable device IO\n");
+ res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+ if (!res) {
+ dev_err(&pdev->dev, "can't fetch device resource info\n");
goto done;
}
- err = pci_request_region(pdev, MFGPT_BAR, DRV_NAME);
- if (err) {
- dev_err(&pdev->dev, "can't alloc PCI BAR #%d\n", MFGPT_BAR);
+ if (!request_region(res->start, resource_size(res), pdev->name)) {
+ dev_err(&pdev->dev, "can't request region\n");
goto done;
}
/* set up the driver-specific struct */
- cs5535_mfgpt_chip.base = pci_resource_start(pdev, MFGPT_BAR);
+ cs5535_mfgpt_chip.base = res->start;
cs5535_mfgpt_chip.pdev = pdev;
spin_lock_init(&cs5535_mfgpt_chip.lock);
- dev_info(&pdev->dev, "allocated PCI BAR #%d: base 0x%llx\n", MFGPT_BAR,
- (unsigned long long) cs5535_mfgpt_chip.base);
+ dev_info(&pdev->dev, "reserved resource region %pR\n", res);
/* detect the available timers */
t = scan_timers(&cs5535_mfgpt_chip);
- dev_info(&pdev->dev, DRV_NAME ": %d MFGPT timers available\n", t);
+ dev_info(&pdev->dev, "%d MFGPT timers available\n", t);
cs5535_mfgpt_chip.initialized = 1;
return 0;
@@ -332,47 +329,18 @@
return err;
}
-static struct pci_device_id cs5535_mfgpt_pci_tbl[] = {
- { PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_CS5535_ISA) },
- { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_ISA) },
- { 0, },
+static struct platform_driver cs5535_mfgpt_drv = {
+ .driver = {
+ .name = DRV_NAME,
+ .owner = THIS_MODULE,
+ },
+ .probe = cs5535_mfgpt_probe,
};
-MODULE_DEVICE_TABLE(pci, cs5535_mfgpt_pci_tbl);
-/*
- * Just like with the cs5535-gpio driver, we can't use the standard PCI driver
- * registration stuff. It only allows only one driver to bind to each PCI
- * device, and we want the GPIO and MFGPT drivers to be able to share a PCI
- * device. Instead, we manually scan for the PCI device, request a single
- * region, and keep track of the devices that we're using.
- */
-
-static int __init cs5535_mfgpt_scan_pci(void)
-{
- struct pci_dev *pdev;
- int err = -ENODEV;
- int i;
-
- for (i = 0; i < ARRAY_SIZE(cs5535_mfgpt_pci_tbl); i++) {
- pdev = pci_get_device(cs5535_mfgpt_pci_tbl[i].vendor,
- cs5535_mfgpt_pci_tbl[i].device, NULL);
- if (pdev) {
- err = cs5535_mfgpt_probe(pdev,
- &cs5535_mfgpt_pci_tbl[i]);
- if (err)
- pci_dev_put(pdev);
-
- /* we only support a single CS5535/6 southbridge */
- break;
- }
- }
-
- return err;
-}
static int __init cs5535_mfgpt_init(void)
{
- return cs5535_mfgpt_scan_pci();
+ return platform_driver_register(&cs5535_mfgpt_drv);
}
module_init(cs5535_mfgpt_init);
@@ -380,3 +348,4 @@
MODULE_AUTHOR("Andres Salomon <dilinger@queued.net>");
MODULE_DESCRIPTION("CS5535/CS5536 MFGPT timer driver");
MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:" DRV_NAME);
diff --git a/drivers/mmc/host/sdhci-of-core.c b/drivers/mmc/host/sdhci-of-core.c
index fa19d84..dd84124f 100644
--- a/drivers/mmc/host/sdhci-of-core.c
+++ b/drivers/mmc/host/sdhci-of-core.c
@@ -13,6 +13,7 @@
* your option) any later version.
*/
+#include <linux/err.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/io.h>
@@ -20,8 +21,12 @@
#include <linux/delay.h>
#include <linux/of.h>
#include <linux/of_platform.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
#include <linux/mmc/host.h>
+#ifdef CONFIG_PPC
#include <asm/machdep.h>
+#endif
#include "sdhci-of.h"
#include "sdhci.h"
@@ -112,7 +117,11 @@
return true;
/* Old device trees don't have the wp-inverted property. */
+#ifdef CONFIG_PPC
return machine_is(mpc837x_rdb) || machine_is(mpc837x_mds);
+#else
+ return false;
+#endif
}
static int __devinit sdhci_of_probe(struct platform_device *ofdev,
diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index ee4bb33..9824057 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -1201,7 +1201,7 @@
static void __exit cleanup_mtdchar(void)
{
unregister_mtd_user(&mtdchar_notifier);
- mntput_long(mtd_inode_mnt);
+ mntput(mtd_inode_mnt);
unregister_filesystem(&mtd_inodefs_type);
__unregister_chrdev(MTD_CHAR_MAJOR, 0, 1 << MINORBITS, "mtd");
}
diff --git a/drivers/net/arm/ks8695net.c b/drivers/net/arm/ks8695net.c
index 54c6d84..62d6f88 100644
--- a/drivers/net/arm/ks8695net.c
+++ b/drivers/net/arm/ks8695net.c
@@ -854,12 +854,12 @@
}
/**
- * ks8695_get_settings - Get device-specific settings.
+ * ks8695_wan_get_settings - Get device-specific settings.
* @ndev: The network device to read settings from
* @cmd: The ethtool structure to read into
*/
static int
-ks8695_get_settings(struct net_device *ndev, struct ethtool_cmd *cmd)
+ks8695_wan_get_settings(struct net_device *ndev, struct ethtool_cmd *cmd)
{
struct ks8695_priv *ksp = netdev_priv(ndev);
u32 ctrl;
@@ -870,69 +870,50 @@
SUPPORTED_TP | SUPPORTED_MII);
cmd->transceiver = XCVR_INTERNAL;
- /* Port specific extras */
- switch (ksp->dtype) {
- case KS8695_DTYPE_HPNA:
- cmd->phy_address = 0;
- /* not supported for HPNA */
+ cmd->advertising = ADVERTISED_TP | ADVERTISED_MII;
+ cmd->port = PORT_MII;
+ cmd->supported |= (SUPPORTED_Autoneg | SUPPORTED_Pause);
+ cmd->phy_address = 0;
+
+ ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
+ if ((ctrl & WMC_WAND) == 0) {
+ /* auto-negotiation is enabled */
+ cmd->advertising |= ADVERTISED_Autoneg;
+ if (ctrl & WMC_WANA100F)
+ cmd->advertising |= ADVERTISED_100baseT_Full;
+ if (ctrl & WMC_WANA100H)
+ cmd->advertising |= ADVERTISED_100baseT_Half;
+ if (ctrl & WMC_WANA10F)
+ cmd->advertising |= ADVERTISED_10baseT_Full;
+ if (ctrl & WMC_WANA10H)
+ cmd->advertising |= ADVERTISED_10baseT_Half;
+ if (ctrl & WMC_WANAP)
+ cmd->advertising |= ADVERTISED_Pause;
+ cmd->autoneg = AUTONEG_ENABLE;
+
+ cmd->speed = (ctrl & WMC_WSS) ? SPEED_100 : SPEED_10;
+ cmd->duplex = (ctrl & WMC_WDS) ?
+ DUPLEX_FULL : DUPLEX_HALF;
+ } else {
+ /* auto-negotiation is disabled */
cmd->autoneg = AUTONEG_DISABLE;
- /* BUG: Erm, dtype hpna implies no phy regs */
- /*
- ctrl = readl(KS8695_MISC_VA + KS8695_HMC);
- cmd->speed = (ctrl & HMC_HSS) ? SPEED_100 : SPEED_10;
- cmd->duplex = (ctrl & HMC_HDS) ? DUPLEX_FULL : DUPLEX_HALF;
- */
- return -EOPNOTSUPP;
- case KS8695_DTYPE_WAN:
- cmd->advertising = ADVERTISED_TP | ADVERTISED_MII;
- cmd->port = PORT_MII;
- cmd->supported |= (SUPPORTED_Autoneg | SUPPORTED_Pause);
- cmd->phy_address = 0;
-
- ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
- if ((ctrl & WMC_WAND) == 0) {
- /* auto-negotiation is enabled */
- cmd->advertising |= ADVERTISED_Autoneg;
- if (ctrl & WMC_WANA100F)
- cmd->advertising |= ADVERTISED_100baseT_Full;
- if (ctrl & WMC_WANA100H)
- cmd->advertising |= ADVERTISED_100baseT_Half;
- if (ctrl & WMC_WANA10F)
- cmd->advertising |= ADVERTISED_10baseT_Full;
- if (ctrl & WMC_WANA10H)
- cmd->advertising |= ADVERTISED_10baseT_Half;
- if (ctrl & WMC_WANAP)
- cmd->advertising |= ADVERTISED_Pause;
- cmd->autoneg = AUTONEG_ENABLE;
-
- cmd->speed = (ctrl & WMC_WSS) ? SPEED_100 : SPEED_10;
- cmd->duplex = (ctrl & WMC_WDS) ?
- DUPLEX_FULL : DUPLEX_HALF;
- } else {
- /* auto-negotiation is disabled */
- cmd->autoneg = AUTONEG_DISABLE;
-
- cmd->speed = (ctrl & WMC_WANF100) ?
- SPEED_100 : SPEED_10;
- cmd->duplex = (ctrl & WMC_WANFF) ?
- DUPLEX_FULL : DUPLEX_HALF;
- }
- break;
- case KS8695_DTYPE_LAN:
- return -EOPNOTSUPP;
+ cmd->speed = (ctrl & WMC_WANF100) ?
+ SPEED_100 : SPEED_10;
+ cmd->duplex = (ctrl & WMC_WANFF) ?
+ DUPLEX_FULL : DUPLEX_HALF;
}
return 0;
}
/**
- * ks8695_set_settings - Set device-specific settings.
+ * ks8695_wan_set_settings - Set device-specific settings.
* @ndev: The network device to configure
* @cmd: The settings to configure
*/
static int
-ks8695_set_settings(struct net_device *ndev, struct ethtool_cmd *cmd)
+ks8695_wan_set_settings(struct net_device *ndev, struct ethtool_cmd *cmd)
{
struct ks8695_priv *ksp = netdev_priv(ndev);
u32 ctrl;
@@ -956,171 +937,85 @@
ADVERTISED_100baseT_Full)) == 0)
return -EINVAL;
- switch (ksp->dtype) {
- case KS8695_DTYPE_HPNA:
- /* HPNA does not support auto-negotiation. */
- return -EINVAL;
- case KS8695_DTYPE_WAN:
- ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
+ ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
- ctrl &= ~(WMC_WAND | WMC_WANA100F | WMC_WANA100H |
- WMC_WANA10F | WMC_WANA10H);
- if (cmd->advertising & ADVERTISED_100baseT_Full)
- ctrl |= WMC_WANA100F;
- if (cmd->advertising & ADVERTISED_100baseT_Half)
- ctrl |= WMC_WANA100H;
- if (cmd->advertising & ADVERTISED_10baseT_Full)
- ctrl |= WMC_WANA10F;
- if (cmd->advertising & ADVERTISED_10baseT_Half)
- ctrl |= WMC_WANA10H;
+ ctrl &= ~(WMC_WAND | WMC_WANA100F | WMC_WANA100H |
+ WMC_WANA10F | WMC_WANA10H);
+ if (cmd->advertising & ADVERTISED_100baseT_Full)
+ ctrl |= WMC_WANA100F;
+ if (cmd->advertising & ADVERTISED_100baseT_Half)
+ ctrl |= WMC_WANA100H;
+ if (cmd->advertising & ADVERTISED_10baseT_Full)
+ ctrl |= WMC_WANA10F;
+ if (cmd->advertising & ADVERTISED_10baseT_Half)
+ ctrl |= WMC_WANA10H;
- /* force a re-negotiation */
- ctrl |= WMC_WANR;
- writel(ctrl, ksp->phyiface_regs + KS8695_WMC);
- break;
- case KS8695_DTYPE_LAN:
- return -EOPNOTSUPP;
- }
-
+ /* force a re-negotiation */
+ ctrl |= WMC_WANR;
+ writel(ctrl, ksp->phyiface_regs + KS8695_WMC);
} else {
- switch (ksp->dtype) {
- case KS8695_DTYPE_HPNA:
- /* BUG: dtype_hpna implies no phy registers */
- /*
- ctrl = __raw_readl(KS8695_MISC_VA + KS8695_HMC);
+ ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
- ctrl &= ~(HMC_HSS | HMC_HDS);
- if (cmd->speed == SPEED_100)
- ctrl |= HMC_HSS;
- if (cmd->duplex == DUPLEX_FULL)
- ctrl |= HMC_HDS;
+ /* disable auto-negotiation */
+ ctrl |= WMC_WAND;
+ ctrl &= ~(WMC_WANF100 | WMC_WANFF);
- __raw_writel(ctrl, KS8695_MISC_VA + KS8695_HMC);
- */
- return -EOPNOTSUPP;
- case KS8695_DTYPE_WAN:
- ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
+ if (cmd->speed == SPEED_100)
+ ctrl |= WMC_WANF100;
+ if (cmd->duplex == DUPLEX_FULL)
+ ctrl |= WMC_WANFF;
- /* disable auto-negotiation */
- ctrl |= WMC_WAND;
- ctrl &= ~(WMC_WANF100 | WMC_WANFF);
-
- if (cmd->speed == SPEED_100)
- ctrl |= WMC_WANF100;
- if (cmd->duplex == DUPLEX_FULL)
- ctrl |= WMC_WANFF;
-
- writel(ctrl, ksp->phyiface_regs + KS8695_WMC);
- break;
- case KS8695_DTYPE_LAN:
- return -EOPNOTSUPP;
- }
+ writel(ctrl, ksp->phyiface_regs + KS8695_WMC);
}
return 0;
}
/**
- * ks8695_nwayreset - Restart the autonegotiation on the port.
+ * ks8695_wan_nwayreset - Restart the autonegotiation on the port.
* @ndev: The network device to restart autoneotiation on
*/
static int
-ks8695_nwayreset(struct net_device *ndev)
+ks8695_wan_nwayreset(struct net_device *ndev)
{
struct ks8695_priv *ksp = netdev_priv(ndev);
u32 ctrl;
- switch (ksp->dtype) {
- case KS8695_DTYPE_HPNA:
- /* No phy means no autonegotiation on hpna */
+ ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
+
+ if ((ctrl & WMC_WAND) == 0)
+ writel(ctrl | WMC_WANR,
+ ksp->phyiface_regs + KS8695_WMC);
+ else
+ /* auto-negotiation not enabled */
return -EINVAL;
- case KS8695_DTYPE_WAN:
- ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
-
- if ((ctrl & WMC_WAND) == 0)
- writel(ctrl | WMC_WANR,
- ksp->phyiface_regs + KS8695_WMC);
- else
- /* auto-negotiation not enabled */
- return -EINVAL;
- break;
- case KS8695_DTYPE_LAN:
- return -EOPNOTSUPP;
- }
return 0;
}
/**
- * ks8695_get_link - Retrieve link status of network interface
- * @ndev: The network interface to retrive the link status of.
- */
-static u32
-ks8695_get_link(struct net_device *ndev)
-{
- struct ks8695_priv *ksp = netdev_priv(ndev);
- u32 ctrl;
-
- switch (ksp->dtype) {
- case KS8695_DTYPE_HPNA:
- /* HPNA always has link */
- return 1;
- case KS8695_DTYPE_WAN:
- /* WAN we can read the PHY for */
- ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
- return ctrl & WMC_WLS;
- case KS8695_DTYPE_LAN:
- return -EOPNOTSUPP;
- }
- return 0;
-}
-
-/**
- * ks8695_get_pause - Retrieve network pause/flow-control advertising
+ * ks8695_wan_get_pause - Retrieve network pause/flow-control advertising
* @ndev: The device to retrieve settings from
* @param: The structure to fill out with the information
*/
static void
-ks8695_get_pause(struct net_device *ndev, struct ethtool_pauseparam *param)
+ks8695_wan_get_pause(struct net_device *ndev, struct ethtool_pauseparam *param)
{
struct ks8695_priv *ksp = netdev_priv(ndev);
u32 ctrl;
- switch (ksp->dtype) {
- case KS8695_DTYPE_HPNA:
- /* No phy link on hpna to configure */
- return;
- case KS8695_DTYPE_WAN:
- ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
+ ctrl = readl(ksp->phyiface_regs + KS8695_WMC);
- /* advertise Pause */
- param->autoneg = (ctrl & WMC_WANAP);
+ /* advertise Pause */
+ param->autoneg = (ctrl & WMC_WANAP);
- /* current Rx Flow-control */
- ctrl = ks8695_readreg(ksp, KS8695_DRXC);
- param->rx_pause = (ctrl & DRXC_RFCE);
+ /* current Rx Flow-control */
+ ctrl = ks8695_readreg(ksp, KS8695_DRXC);
+ param->rx_pause = (ctrl & DRXC_RFCE);
- /* current Tx Flow-control */
- ctrl = ks8695_readreg(ksp, KS8695_DTXC);
- param->tx_pause = (ctrl & DTXC_TFCE);
- break;
- case KS8695_DTYPE_LAN:
- /* The LAN's "phy" is a direct-attached switch */
- return;
- }
-}
-
-/**
- * ks8695_set_pause - Configure pause/flow-control
- * @ndev: The device to configure
- * @param: The pause parameters to set
- *
- * TODO: Implement this
- */
-static int
-ks8695_set_pause(struct net_device *ndev, struct ethtool_pauseparam *param)
-{
- return -EOPNOTSUPP;
+ /* current Tx Flow-control */
+ ctrl = ks8695_readreg(ksp, KS8695_DTXC);
+ param->tx_pause = (ctrl & DTXC_TFCE);
}
/**
@@ -1140,12 +1035,17 @@
static const struct ethtool_ops ks8695_ethtool_ops = {
.get_msglevel = ks8695_get_msglevel,
.set_msglevel = ks8695_set_msglevel,
- .get_settings = ks8695_get_settings,
- .set_settings = ks8695_set_settings,
- .nway_reset = ks8695_nwayreset,
- .get_link = ks8695_get_link,
- .get_pauseparam = ks8695_get_pause,
- .set_pauseparam = ks8695_set_pause,
+ .get_drvinfo = ks8695_get_drvinfo,
+};
+
+static const struct ethtool_ops ks8695_wan_ethtool_ops = {
+ .get_msglevel = ks8695_get_msglevel,
+ .set_msglevel = ks8695_set_msglevel,
+ .get_settings = ks8695_wan_get_settings,
+ .set_settings = ks8695_wan_set_settings,
+ .nway_reset = ks8695_wan_nwayreset,
+ .get_link = ethtool_op_get_link,
+ .get_pauseparam = ks8695_wan_get_pause,
.get_drvinfo = ks8695_get_drvinfo,
};
@@ -1541,7 +1441,6 @@
/* driver system setup */
ndev->netdev_ops = &ks8695_netdev_ops;
- SET_ETHTOOL_OPS(ndev, &ks8695_ethtool_ops);
ndev->watchdog_timeo = msecs_to_jiffies(watchdog);
netif_napi_add(ndev, &ksp->napi, ks8695_poll, NAPI_WEIGHT);
@@ -1608,12 +1507,15 @@
if (ksp->phyiface_regs && ksp->link_irq == -1) {
ks8695_init_switch(ksp);
ksp->dtype = KS8695_DTYPE_LAN;
+ SET_ETHTOOL_OPS(ndev, &ks8695_ethtool_ops);
} else if (ksp->phyiface_regs && ksp->link_irq != -1) {
ks8695_init_wan_phy(ksp);
ksp->dtype = KS8695_DTYPE_WAN;
+ SET_ETHTOOL_OPS(ndev, &ks8695_wan_ethtool_ops);
} else {
/* No initialisation since HPNA does not have a PHY */
ksp->dtype = KS8695_DTYPE_HPNA;
+ SET_ETHTOOL_OPS(ndev, &ks8695_ethtool_ops);
}
/* And bring up the net_device with the net core */
diff --git a/drivers/net/bfin_mac.c b/drivers/net/bfin_mac.c
index 0b9fc51..22abfb3 100644
--- a/drivers/net/bfin_mac.c
+++ b/drivers/net/bfin_mac.c
@@ -1284,19 +1284,12 @@
{
u32 emac_hashhi, emac_hashlo;
struct netdev_hw_addr *ha;
- char *addrs;
u32 crc;
emac_hashhi = emac_hashlo = 0;
netdev_for_each_mc_addr(ha, dev) {
- addrs = ha->addr;
-
- /* skip non-multicast addresses */
- if (!(*addrs & 1))
- continue;
-
- crc = ether_crc(ETH_ALEN, addrs);
+ crc = ether_crc(ETH_ALEN, ha->addr);
crc >>= 26;
if (crc & 0x20)
diff --git a/drivers/net/bna/bnad_ethtool.c b/drivers/net/bna/bnad_ethtool.c
index 99be5ae..142d604 100644
--- a/drivers/net/bna/bnad_ethtool.c
+++ b/drivers/net/bna/bnad_ethtool.c
@@ -275,7 +275,6 @@
ioc_attr = kzalloc(sizeof(*ioc_attr), GFP_KERNEL);
if (ioc_attr) {
- memset(ioc_attr, 0, sizeof(*ioc_attr));
spin_lock_irqsave(&bnad->bna_lock, flags);
bfa_nw_ioc_get_attr(&bnad->bna.device.ioc, ioc_attr);
spin_unlock_irqrestore(&bnad->bna_lock, flags);
diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c
index 7206ab2..3437613 100644
--- a/drivers/net/cassini.c
+++ b/drivers/net/cassini.c
@@ -3203,7 +3203,7 @@
int phy_type = CAS_PHY_MII_MDIO0; /* default phy type */
int mac_off = 0;
-#if defined(CONFIG_OF)
+#if defined(CONFIG_SPARC)
const unsigned char *addr;
#endif
@@ -3354,7 +3354,7 @@
if (found & VPD_FOUND_MAC)
goto done;
-#if defined(CONFIG_OF)
+#if defined(CONFIG_SPARC)
addr = of_get_property(cp->of_node, "local-mac-address", NULL);
if (addr != NULL) {
memcpy(dev_addr, addr, 6);
@@ -5031,7 +5031,7 @@
cp->msg_enable = (cassini_debug < 0) ? CAS_DEF_MSG_ENABLE :
cassini_debug;
-#if defined(CONFIG_OF)
+#if defined(CONFIG_SPARC)
cp->of_node = pci_device_to_OF_node(pdev);
#endif
diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index de69c54..bfab140 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -3478,9 +3478,17 @@
struct e1000_hw *hw = &adapter->hw;
u32 icr = er32(ICR);
- if (unlikely((!icr) || test_bit(__E1000_DOWN, &adapter->flags)))
+ if (unlikely((!icr)))
return IRQ_NONE; /* Not our interrupt */
+ /*
+ * we might have caused the interrupt, but the above
+ * read cleared it, and just in case the driver is
+ * down there is nothing to do so return handled
+ */
+ if (unlikely(test_bit(__E1000_DOWN, &adapter->flags)))
+ return IRQ_HANDLED;
+
if (unlikely(icr & (E1000_ICR_RXSEQ | E1000_ICR_LSC))) {
hw->get_link_status = 1;
/* guard against interrupt when we're going down */
diff --git a/drivers/net/e1000e/82571.c b/drivers/net/e1000e/82571.c
index 1397da1..89a6903 100644
--- a/drivers/net/e1000e/82571.c
+++ b/drivers/net/e1000e/82571.c
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
@@ -1310,7 +1310,7 @@
* apply workaround for hardware errata documented in errata
* docs Fixes issue where some error prone or unreliable PCIe
* completions are occurring, particularly with ASPM enabled.
- * Without fix, issue can cause tx timeouts.
+ * Without fix, issue can cause Tx timeouts.
*/
reg = er32(GCR2);
reg |= 1;
diff --git a/drivers/net/e1000e/Makefile b/drivers/net/e1000e/Makefile
index 360c913..28519ac 100644
--- a/drivers/net/e1000e/Makefile
+++ b/drivers/net/e1000e/Makefile
@@ -1,7 +1,7 @@
################################################################################
#
# Intel PRO/1000 Linux driver
-# Copyright(c) 1999 - 2008 Intel Corporation.
+# Copyright(c) 1999 - 2011 Intel Corporation.
#
# This program is free software; you can redistribute it and/or modify it
# under the terms and conditions of the GNU General Public License,
diff --git a/drivers/net/e1000e/defines.h b/drivers/net/e1000e/defines.h
index 7245dc2..1314998 100644
--- a/drivers/net/e1000e/defines.h
+++ b/drivers/net/e1000e/defines.h
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
diff --git a/drivers/net/e1000e/e1000.h b/drivers/net/e1000e/e1000.h
index 5255be7..e610e136 100644
--- a/drivers/net/e1000e/e1000.h
+++ b/drivers/net/e1000e/e1000.h
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
diff --git a/drivers/net/e1000e/es2lan.c b/drivers/net/e1000e/es2lan.c
index e45a61c..2fefa82 100644
--- a/drivers/net/e1000e/es2lan.c
+++ b/drivers/net/e1000e/es2lan.c
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
diff --git a/drivers/net/e1000e/ethtool.c b/drivers/net/e1000e/ethtool.c
index f8ed03d..fa08b63 100644
--- a/drivers/net/e1000e/ethtool.c
+++ b/drivers/net/e1000e/ethtool.c
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
diff --git a/drivers/net/e1000e/hw.h b/drivers/net/e1000e/hw.h
index e774380..bc0860a 100644
--- a/drivers/net/e1000e/hw.h
+++ b/drivers/net/e1000e/hw.h
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
@@ -102,7 +102,7 @@
E1000_RDTR = 0x02820, /* Rx Delay Timer - RW */
E1000_RXDCTL_BASE = 0x02828, /* Rx Descriptor Control - RW */
#define E1000_RXDCTL(_n) (E1000_RXDCTL_BASE + (_n << 8))
- E1000_RADV = 0x0282C, /* RX Interrupt Absolute Delay Timer - RW */
+ E1000_RADV = 0x0282C, /* Rx Interrupt Absolute Delay Timer - RW */
/* Convenience macros
*
diff --git a/drivers/net/e1000e/ich8lan.c b/drivers/net/e1000e/ich8lan.c
index 5bb65b7..fb46974 100644
--- a/drivers/net/e1000e/ich8lan.c
+++ b/drivers/net/e1000e/ich8lan.c
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
diff --git a/drivers/net/e1000e/lib.c b/drivers/net/e1000e/lib.c
index ff28721..68aa174 100644
--- a/drivers/net/e1000e/lib.c
+++ b/drivers/net/e1000e/lib.c
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
@@ -533,7 +533,7 @@
mac->autoneg_failed = 1;
return 0;
}
- e_dbg("NOT RXing /C/, disable AutoNeg and force link.\n");
+ e_dbg("NOT Rx'ing /C/, disable AutoNeg and force link.\n");
/* Disable auto-negotiation in the TXCW register */
ew32(TXCW, (mac->txcw & ~E1000_TXCW_ANE));
@@ -556,7 +556,7 @@
* and disable forced link in the Device Control register
* in an attempt to auto-negotiate with our link partner.
*/
- e_dbg("RXing /C/, enable AutoNeg and stop forcing link.\n");
+ e_dbg("Rx'ing /C/, enable AutoNeg and stop forcing link.\n");
ew32(TXCW, mac->txcw);
ew32(CTRL, (ctrl & ~E1000_CTRL_SLU));
@@ -598,7 +598,7 @@
mac->autoneg_failed = 1;
return 0;
}
- e_dbg("NOT RXing /C/, disable AutoNeg and force link.\n");
+ e_dbg("NOT Rx'ing /C/, disable AutoNeg and force link.\n");
/* Disable auto-negotiation in the TXCW register */
ew32(TXCW, (mac->txcw & ~E1000_TXCW_ANE));
@@ -621,7 +621,7 @@
* and disable forced link in the Device Control register
* in an attempt to auto-negotiate with our link partner.
*/
- e_dbg("RXing /C/, enable AutoNeg and stop forcing link.\n");
+ e_dbg("Rx'ing /C/, enable AutoNeg and stop forcing link.\n");
ew32(TXCW, mac->txcw);
ew32(CTRL, (ctrl & ~E1000_CTRL_SLU));
@@ -800,9 +800,9 @@
* The possible values of the "fc" parameter are:
* 0: Flow control is completely disabled
* 1: Rx flow control is enabled (we can receive pause frames,
- * but not send pause frames).
+ * but not send pause frames).
* 2: Tx flow control is enabled (we can send pause frames but we
- * do not support receiving pause frames).
+ * do not support receiving pause frames).
* 3: Both Rx and Tx flow control (symmetric) are enabled.
*/
switch (hw->fc.current_mode) {
@@ -1031,9 +1031,9 @@
* The possible values of the "fc" parameter are:
* 0: Flow control is completely disabled
* 1: Rx flow control is enabled (we can receive pause
- * frames but not send pause frames).
+ * frames but not send pause frames).
* 2: Tx flow control is enabled (we can send pause frames
- * frames but we do not receive pause frames).
+ * frames but we do not receive pause frames).
* 3: Both Rx and Tx flow control (symmetric) is enabled.
* other: No other values should be possible at this point.
*/
@@ -1189,7 +1189,7 @@
} else {
hw->fc.current_mode = e1000_fc_rx_pause;
e_dbg("Flow Control = "
- "RX PAUSE frames only.\r\n");
+ "Rx PAUSE frames only.\r\n");
}
}
/*
diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index fa5b604..1c18f26 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
@@ -77,17 +77,17 @@
char *name;
};
-#define E1000_RDFH 0x02410 /* Rx Data FIFO Head - RW */
-#define E1000_RDFT 0x02418 /* Rx Data FIFO Tail - RW */
-#define E1000_RDFHS 0x02420 /* Rx Data FIFO Head Saved - RW */
-#define E1000_RDFTS 0x02428 /* Rx Data FIFO Tail Saved - RW */
-#define E1000_RDFPC 0x02430 /* Rx Data FIFO Packet Count - RW */
+#define E1000_RDFH 0x02410 /* Rx Data FIFO Head - RW */
+#define E1000_RDFT 0x02418 /* Rx Data FIFO Tail - RW */
+#define E1000_RDFHS 0x02420 /* Rx Data FIFO Head Saved - RW */
+#define E1000_RDFTS 0x02428 /* Rx Data FIFO Tail Saved - RW */
+#define E1000_RDFPC 0x02430 /* Rx Data FIFO Packet Count - RW */
-#define E1000_TDFH 0x03410 /* Tx Data FIFO Head - RW */
-#define E1000_TDFT 0x03418 /* Tx Data FIFO Tail - RW */
-#define E1000_TDFHS 0x03420 /* Tx Data FIFO Head Saved - RW */
-#define E1000_TDFTS 0x03428 /* Tx Data FIFO Tail Saved - RW */
-#define E1000_TDFPC 0x03430 /* Tx Data FIFO Packet Count - RW */
+#define E1000_TDFH 0x03410 /* Tx Data FIFO Head - RW */
+#define E1000_TDFT 0x03418 /* Tx Data FIFO Tail - RW */
+#define E1000_TDFHS 0x03420 /* Tx Data FIFO Head Saved - RW */
+#define E1000_TDFTS 0x03428 /* Tx Data FIFO Tail Saved - RW */
+#define E1000_TDFPC 0x03430 /* Tx Data FIFO Packet Count - RW */
static const struct e1000_reg_info e1000_reg_info_tbl[] = {
@@ -99,7 +99,7 @@
/* Interrupt Registers */
{E1000_ICR, "ICR"},
- /* RX Registers */
+ /* Rx Registers */
{E1000_RCTL, "RCTL"},
{E1000_RDLEN, "RDLEN"},
{E1000_RDH, "RDH"},
@@ -115,7 +115,7 @@
{E1000_RDFTS, "RDFTS"},
{E1000_RDFPC, "RDFPC"},
- /* TX Registers */
+ /* Tx Registers */
{E1000_TCTL, "TCTL"},
{E1000_TDBAL, "TDBAL"},
{E1000_TDBAH, "TDBAH"},
@@ -160,7 +160,7 @@
break;
default:
printk(KERN_INFO "%-15s %08x\n",
- reginfo->name, __er32(hw, reginfo->ofs));
+ reginfo->name, __er32(hw, reginfo->ofs));
return;
}
@@ -171,9 +171,8 @@
printk(KERN_CONT "\n");
}
-
/*
- * e1000e_dump - Print registers, tx-ring and rx-ring
+ * e1000e_dump - Print registers, Tx-ring and Rx-ring
*/
static void e1000e_dump(struct e1000_adapter *adapter)
{
@@ -182,12 +181,20 @@
struct e1000_reg_info *reginfo;
struct e1000_ring *tx_ring = adapter->tx_ring;
struct e1000_tx_desc *tx_desc;
- struct my_u0 { u64 a; u64 b; } *u0;
+ struct my_u0 {
+ u64 a;
+ u64 b;
+ } *u0;
struct e1000_buffer *buffer_info;
struct e1000_ring *rx_ring = adapter->rx_ring;
union e1000_rx_desc_packet_split *rx_desc_ps;
struct e1000_rx_desc *rx_desc;
- struct my_u1 { u64 a; u64 b; u64 c; u64 d; } *u1;
+ struct my_u1 {
+ u64 a;
+ u64 b;
+ u64 c;
+ u64 d;
+ } *u1;
u32 staterr;
int i = 0;
@@ -198,12 +205,10 @@
if (netdev) {
dev_info(&adapter->pdev->dev, "Net device Info\n");
printk(KERN_INFO "Device Name state "
- "trans_start last_rx\n");
+ "trans_start last_rx\n");
printk(KERN_INFO "%-15s %016lX %016lX %016lX\n",
- netdev->name,
- netdev->state,
- netdev->trans_start,
- netdev->last_rx);
+ netdev->name, netdev->state, netdev->trans_start,
+ netdev->last_rx);
}
/* Print Registers */
@@ -214,26 +219,26 @@
e1000_regdump(hw, reginfo);
}
- /* Print TX Ring Summary */
+ /* Print Tx Ring Summary */
if (!netdev || !netif_running(netdev))
goto exit;
- dev_info(&adapter->pdev->dev, "TX Rings Summary\n");
+ dev_info(&adapter->pdev->dev, "Tx Ring Summary\n");
printk(KERN_INFO "Queue [NTU] [NTC] [bi(ntc)->dma ]"
- " leng ntw timestamp\n");
+ " leng ntw timestamp\n");
buffer_info = &tx_ring->buffer_info[tx_ring->next_to_clean];
printk(KERN_INFO " %5d %5X %5X %016llX %04X %3X %016llX\n",
- 0, tx_ring->next_to_use, tx_ring->next_to_clean,
- (unsigned long long)buffer_info->dma,
- buffer_info->length,
- buffer_info->next_to_watch,
- (unsigned long long)buffer_info->time_stamp);
+ 0, tx_ring->next_to_use, tx_ring->next_to_clean,
+ (unsigned long long)buffer_info->dma,
+ buffer_info->length,
+ buffer_info->next_to_watch,
+ (unsigned long long)buffer_info->time_stamp);
- /* Print TX Rings */
+ /* Print Tx Ring */
if (!netif_msg_tx_done(adapter))
goto rx_ring_summary;
- dev_info(&adapter->pdev->dev, "TX Rings Dump\n");
+ dev_info(&adapter->pdev->dev, "Tx Ring Dump\n");
/* Transmit Descriptor Formats - DEXT[29] is 0 (Legacy) or 1 (Extended)
*
@@ -263,22 +268,22 @@
* 63 48 47 40 39 36 35 32 31 24 23 20 19 0
*/
printk(KERN_INFO "Tl[desc] [address 63:0 ] [SpeCssSCmCsLen]"
- " [bi->dma ] leng ntw timestamp bi->skb "
- "<-- Legacy format\n");
+ " [bi->dma ] leng ntw timestamp bi->skb "
+ "<-- Legacy format\n");
printk(KERN_INFO "Tc[desc] [Ce CoCsIpceCoS] [MssHlRSCm0Plen]"
- " [bi->dma ] leng ntw timestamp bi->skb "
- "<-- Ext Context format\n");
+ " [bi->dma ] leng ntw timestamp bi->skb "
+ "<-- Ext Context format\n");
printk(KERN_INFO "Td[desc] [address 63:0 ] [VlaPoRSCm1Dlen]"
- " [bi->dma ] leng ntw timestamp bi->skb "
- "<-- Ext Data format\n");
+ " [bi->dma ] leng ntw timestamp bi->skb "
+ "<-- Ext Data format\n");
for (i = 0; tx_ring->desc && (i < tx_ring->count); i++) {
tx_desc = E1000_TX_DESC(*tx_ring, i);
buffer_info = &tx_ring->buffer_info[i];
u0 = (struct my_u0 *)tx_desc;
printk(KERN_INFO "T%c[0x%03X] %016llX %016llX %016llX "
- "%04X %3X %016llX %p",
- (!(le64_to_cpu(u0->b) & (1<<29)) ? 'l' :
- ((le64_to_cpu(u0->b) & (1<<20)) ? 'd' : 'c')), i,
+ "%04X %3X %016llX %p",
+ (!(le64_to_cpu(u0->b) & (1 << 29)) ? 'l' :
+ ((le64_to_cpu(u0->b) & (1 << 20)) ? 'd' : 'c')), i,
(unsigned long long)le64_to_cpu(u0->a),
(unsigned long long)le64_to_cpu(u0->b),
(unsigned long long)buffer_info->dma,
@@ -296,22 +301,22 @@
if (netif_msg_pktdata(adapter) && buffer_info->dma != 0)
print_hex_dump(KERN_INFO, "", DUMP_PREFIX_ADDRESS,
- 16, 1, phys_to_virt(buffer_info->dma),
- buffer_info->length, true);
+ 16, 1, phys_to_virt(buffer_info->dma),
+ buffer_info->length, true);
}
- /* Print RX Rings Summary */
+ /* Print Rx Ring Summary */
rx_ring_summary:
- dev_info(&adapter->pdev->dev, "RX Rings Summary\n");
+ dev_info(&adapter->pdev->dev, "Rx Ring Summary\n");
printk(KERN_INFO "Queue [NTU] [NTC]\n");
printk(KERN_INFO " %5d %5X %5X\n", 0,
- rx_ring->next_to_use, rx_ring->next_to_clean);
+ rx_ring->next_to_use, rx_ring->next_to_clean);
- /* Print RX Rings */
+ /* Print Rx Ring */
if (!netif_msg_rx_status(adapter))
goto exit;
- dev_info(&adapter->pdev->dev, "RX Rings Dump\n");
+ dev_info(&adapter->pdev->dev, "Rx Ring Dump\n");
switch (adapter->rx_ps_pages) {
case 1:
case 2:
@@ -329,7 +334,7 @@
* +-----------------------------------------------------+
*/
printk(KERN_INFO "R [desc] [buffer 0 63:0 ] "
- "[buffer 1 63:0 ] "
+ "[buffer 1 63:0 ] "
"[buffer 2 63:0 ] [buffer 3 63:0 ] [bi->dma ] "
"[bi->skb] <-- Ext Pkt Split format\n");
/* [Extended] Receive Descriptor (Write-Back) Format
@@ -344,7 +349,7 @@
* 63 48 47 32 31 20 19 0
*/
printk(KERN_INFO "RWB[desc] [ck ipid mrqhsh] "
- "[vl l0 ee es] "
+ "[vl l0 ee es] "
"[ l3 l2 l1 hs] [reserved ] ---------------- "
"[bi->skb] <-- Ext Rx Write-Back format\n");
for (i = 0; i < rx_ring->count; i++) {
@@ -352,26 +357,26 @@
rx_desc_ps = E1000_RX_DESC_PS(*rx_ring, i);
u1 = (struct my_u1 *)rx_desc_ps;
staterr =
- le32_to_cpu(rx_desc_ps->wb.middle.status_error);
+ le32_to_cpu(rx_desc_ps->wb.middle.status_error);
if (staterr & E1000_RXD_STAT_DD) {
/* Descriptor Done */
printk(KERN_INFO "RWB[0x%03X] %016llX "
- "%016llX %016llX %016llX "
- "---------------- %p", i,
- (unsigned long long)le64_to_cpu(u1->a),
- (unsigned long long)le64_to_cpu(u1->b),
- (unsigned long long)le64_to_cpu(u1->c),
- (unsigned long long)le64_to_cpu(u1->d),
- buffer_info->skb);
+ "%016llX %016llX %016llX "
+ "---------------- %p", i,
+ (unsigned long long)le64_to_cpu(u1->a),
+ (unsigned long long)le64_to_cpu(u1->b),
+ (unsigned long long)le64_to_cpu(u1->c),
+ (unsigned long long)le64_to_cpu(u1->d),
+ buffer_info->skb);
} else {
printk(KERN_INFO "R [0x%03X] %016llX "
- "%016llX %016llX %016llX %016llX %p", i,
- (unsigned long long)le64_to_cpu(u1->a),
- (unsigned long long)le64_to_cpu(u1->b),
- (unsigned long long)le64_to_cpu(u1->c),
- (unsigned long long)le64_to_cpu(u1->d),
- (unsigned long long)buffer_info->dma,
- buffer_info->skb);
+ "%016llX %016llX %016llX %016llX %p", i,
+ (unsigned long long)le64_to_cpu(u1->a),
+ (unsigned long long)le64_to_cpu(u1->b),
+ (unsigned long long)le64_to_cpu(u1->c),
+ (unsigned long long)le64_to_cpu(u1->d),
+ (unsigned long long)buffer_info->dma,
+ buffer_info->skb);
if (netif_msg_pktdata(adapter))
print_hex_dump(KERN_INFO, "",
@@ -400,18 +405,18 @@
* 63 48 47 40 39 32 31 16 15 0
*/
printk(KERN_INFO "Rl[desc] [address 63:0 ] "
- "[vl er S cks ln] [bi->dma ] [bi->skb] "
- "<-- Legacy format\n");
+ "[vl er S cks ln] [bi->dma ] [bi->skb] "
+ "<-- Legacy format\n");
for (i = 0; rx_ring->desc && (i < rx_ring->count); i++) {
rx_desc = E1000_RX_DESC(*rx_ring, i);
buffer_info = &rx_ring->buffer_info[i];
u0 = (struct my_u0 *)rx_desc;
printk(KERN_INFO "Rl[0x%03X] %016llX %016llX "
- "%016llX %p", i,
- (unsigned long long)le64_to_cpu(u0->a),
- (unsigned long long)le64_to_cpu(u0->b),
- (unsigned long long)buffer_info->dma,
- buffer_info->skb);
+ "%016llX %p", i,
+ (unsigned long long)le64_to_cpu(u0->a),
+ (unsigned long long)le64_to_cpu(u0->b),
+ (unsigned long long)buffer_info->dma,
+ buffer_info->skb);
if (i == rx_ring->next_to_use)
printk(KERN_CONT " NTU\n");
else if (i == rx_ring->next_to_clean)
@@ -421,9 +426,10 @@
if (netif_msg_pktdata(adapter))
print_hex_dump(KERN_INFO, "",
- DUMP_PREFIX_ADDRESS,
- 16, 1, phys_to_virt(buffer_info->dma),
- adapter->rx_buffer_len, true);
+ DUMP_PREFIX_ADDRESS,
+ 16, 1,
+ phys_to_virt(buffer_info->dma),
+ adapter->rx_buffer_len, true);
}
}
@@ -450,8 +456,7 @@
* @skb: pointer to sk_buff to be indicated to stack
**/
static void e1000_receive_skb(struct e1000_adapter *adapter,
- struct net_device *netdev,
- struct sk_buff *skb,
+ struct net_device *netdev, struct sk_buff *skb,
u8 status, __le16 vlan)
{
skb->protocol = eth_type_trans(skb, netdev);
@@ -464,7 +469,7 @@
}
/**
- * e1000_rx_checksum - Receive Checksum Offload for 82543
+ * e1000_rx_checksum - Receive Checksum Offload
* @adapter: board private structure
* @status_err: receive descriptor status and error fields
* @csum: receive descriptor csum field
@@ -548,7 +553,7 @@
adapter->rx_buffer_len,
DMA_FROM_DEVICE);
if (dma_mapping_error(&pdev->dev, buffer_info->dma)) {
- dev_err(&pdev->dev, "RX DMA map failed\n");
+ dev_err(&pdev->dev, "Rx DMA map failed\n");
adapter->rx_dma_failed++;
break;
}
@@ -601,7 +606,8 @@
ps_page = &buffer_info->ps_pages[j];
if (j >= adapter->rx_ps_pages) {
/* all unused desc entries get hw null ptr */
- rx_desc->read.buffer_addr[j+1] = ~cpu_to_le64(0);
+ rx_desc->read.buffer_addr[j + 1] =
+ ~cpu_to_le64(0);
continue;
}
if (!ps_page->page) {
@@ -617,7 +623,7 @@
if (dma_mapping_error(&pdev->dev,
ps_page->dma)) {
dev_err(&adapter->pdev->dev,
- "RX DMA page map failed\n");
+ "Rx DMA page map failed\n");
adapter->rx_dma_failed++;
goto no_buffers;
}
@@ -627,8 +633,8 @@
* didn't change because each write-back
* erases this info.
*/
- rx_desc->read.buffer_addr[j+1] =
- cpu_to_le64(ps_page->dma);
+ rx_desc->read.buffer_addr[j + 1] =
+ cpu_to_le64(ps_page->dma);
}
skb = netdev_alloc_skb_ip_align(netdev,
@@ -644,7 +650,7 @@
adapter->rx_ps_bsize0,
DMA_FROM_DEVICE);
if (dma_mapping_error(&pdev->dev, buffer_info->dma)) {
- dev_err(&pdev->dev, "RX DMA map failed\n");
+ dev_err(&pdev->dev, "Rx DMA map failed\n");
adapter->rx_dma_failed++;
/* cleanup skb */
dev_kfree_skb_any(skb);
@@ -662,7 +668,7 @@
* such as IA-64).
*/
wmb();
- writel(i<<1, adapter->hw.hw_addr + rx_ring->tail);
+ writel(i << 1, adapter->hw.hw_addr + rx_ring->tail);
}
i++;
@@ -1106,11 +1112,10 @@
cleaned = 1;
cleaned_count++;
dma_unmap_single(&pdev->dev, buffer_info->dma,
- adapter->rx_ps_bsize0,
- DMA_FROM_DEVICE);
+ adapter->rx_ps_bsize0, DMA_FROM_DEVICE);
buffer_info->dma = 0;
- /* see !EOP comment in other rx routine */
+ /* see !EOP comment in other Rx routine */
if (!(staterr & E1000_RXD_STAT_EOP))
adapter->flags2 |= FLAG2_IS_DISCARDING;
@@ -2610,7 +2615,7 @@
}
/**
- * e1000_configure_tx - Configure 8254x Transmit Unit after Reset
+ * e1000_configure_tx - Configure Transmit Unit after Reset
* @adapter: board private structure
*
* Configure the Tx unit of the MAC after a reset.
@@ -2663,7 +2668,7 @@
* hthresh = 1 ==> prefetch when one or more available
* pthresh = 0x1f ==> prefetch if internal cache 31 or less
* BEWARE: this seems to work but should be considered first if
- * there are tx hangs or other tx related bugs
+ * there are Tx hangs or other Tx related bugs
*/
txdctl |= E1000_TXDCTL_DMA_BURST_ENABLE;
ew32(TXDCTL(0), txdctl);
@@ -2877,7 +2882,7 @@
if (adapter->rx_ps_pages) {
/* this is a 32 byte descriptor */
rdlen = rx_ring->count *
- sizeof(union e1000_rx_desc_packet_split);
+ sizeof(union e1000_rx_desc_packet_split);
adapter->clean_rx = e1000_clean_rx_irq_ps;
adapter->alloc_rx_buf = e1000_alloc_rx_buffers_ps;
} else if (adapter->netdev->mtu > ETH_FRAME_LEN + ETH_FCS_LEN) {
@@ -2900,7 +2905,7 @@
/*
* set the writeback threshold (only takes effect if the RDTR
* is set). set GRAN=1 and write back up to 0x4 worth, and
- * enable prefetching of 0x20 rx descriptors
+ * enable prefetching of 0x20 Rx descriptors
* granularity = 01
* wthresh = 04,
* hthresh = 04,
@@ -2981,12 +2986,10 @@
* excessive C-state transition latencies result in
* dropped transactions.
*/
- pm_qos_update_request(
- &adapter->netdev->pm_qos_req, 55);
+ pm_qos_update_request(&adapter->netdev->pm_qos_req, 55);
} else {
- pm_qos_update_request(
- &adapter->netdev->pm_qos_req,
- PM_QOS_DEFAULT_VALUE);
+ pm_qos_update_request(&adapter->netdev->pm_qos_req,
+ PM_QOS_DEFAULT_VALUE);
}
}
@@ -3152,7 +3155,7 @@
/* lower 16 bits has Rx packet buffer allocation size in KB */
pba &= 0xffff;
/*
- * the Tx fifo also stores 16 bytes of information about the tx
+ * the Tx fifo also stores 16 bytes of information about the Tx
* but don't include ethernet FCS because hardware appends it
*/
min_tx_space = (adapter->max_frame_size +
@@ -3175,7 +3178,7 @@
pba -= min_tx_space - tx_space;
/*
- * if short on Rx space, Rx wins and must trump tx
+ * if short on Rx space, Rx wins and must trump Tx
* adjustment or use Early Receive if available
*/
if ((pba < min_rx_space) &&
@@ -4039,11 +4042,11 @@
adapter->netdev->name,
adapter->link_speed,
(adapter->link_duplex == FULL_DUPLEX) ?
- "Full Duplex" : "Half Duplex",
+ "Full Duplex" : "Half Duplex",
((ctrl & E1000_CTRL_TFCE) && (ctrl & E1000_CTRL_RFCE)) ?
- "RX/TX" :
- ((ctrl & E1000_CTRL_RFCE) ? "RX" :
- ((ctrl & E1000_CTRL_TFCE) ? "TX" : "None" )));
+ "Rx/Tx" :
+ ((ctrl & E1000_CTRL_RFCE) ? "Rx" :
+ ((ctrl & E1000_CTRL_TFCE) ? "Tx" : "None")));
}
static bool e1000e_has_link(struct e1000_adapter *adapter)
@@ -4338,7 +4341,7 @@
/* Force detection of hung controller every watchdog period */
adapter->detect_tx_hung = 1;
- /* flush partial descriptors to memory before detecting tx hang */
+ /* flush partial descriptors to memory before detecting Tx hang */
if (adapter->flags2 & FLAG2_DMA_BURST) {
ew32(TIDV, adapter->tx_int_delay | E1000_TIDV_FPD);
ew32(RDTR, adapter->rx_int_delay | E1000_RDTR_FPD);
@@ -4529,7 +4532,7 @@
buffer_info->next_to_watch = i;
buffer_info->dma = dma_map_single(&pdev->dev,
skb->data + offset,
- size, DMA_TO_DEVICE);
+ size, DMA_TO_DEVICE);
buffer_info->mapped_as_page = false;
if (dma_mapping_error(&pdev->dev, buffer_info->dma))
goto dma_error;
@@ -4576,7 +4579,7 @@
}
}
- segs = skb_shinfo(skb)->gso_segs ?: 1;
+ segs = skb_shinfo(skb)->gso_segs ? : 1;
/* multiply data chunks by size of headers */
bytecount = ((segs - 1) * skb_headlen(skb)) + skb->len;
@@ -4588,13 +4591,13 @@
return count;
dma_error:
- dev_err(&pdev->dev, "TX DMA map failed\n");
+ dev_err(&pdev->dev, "Tx DMA map failed\n");
buffer_info->dma = 0;
if (count)
count--;
while (count--) {
- if (i==0)
+ if (i == 0)
i += tx_ring->count;
i--;
buffer_info = &tx_ring->buffer_info[i];
@@ -6193,7 +6196,7 @@
int ret;
pr_info("Intel(R) PRO/1000 Network Driver - %s\n",
e1000e_driver_version);
- pr_info("Copyright (c) 1999 - 2010 Intel Corporation.\n");
+ pr_info("Copyright(c) 1999 - 2011 Intel Corporation.\n");
ret = pci_register_driver(&e1000_driver);
return ret;
diff --git a/drivers/net/e1000e/param.c b/drivers/net/e1000e/param.c
index a9612b0..4dd9b63 100644
--- a/drivers/net/e1000e/param.c
+++ b/drivers/net/e1000e/param.c
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
@@ -62,10 +62,9 @@
module_param_array_named(X, X, int, &num_##X, 0); \
MODULE_PARM_DESC(X, desc);
-
/*
* Transmit Interrupt Delay in units of 1.024 microseconds
- * Tx interrupt delay needs to typically be set to something non zero
+ * Tx interrupt delay needs to typically be set to something non-zero
*
* Valid Range: 0-65535
*/
@@ -112,6 +111,7 @@
#define DEFAULT_ITR 3
#define MAX_ITR 100000
#define MIN_ITR 100
+
/* IntMode (Interrupt Mode)
*
* Valid Range: 0 - 2
diff --git a/drivers/net/e1000e/phy.c b/drivers/net/e1000e/phy.c
index 00f89e8..6bea051 100644
--- a/drivers/net/e1000e/phy.c
+++ b/drivers/net/e1000e/phy.c
@@ -1,7 +1,7 @@
/*******************************************************************************
Intel PRO/1000 Linux driver
- Copyright(c) 1999 - 2010 Intel Corporation.
+ Copyright(c) 1999 - 2011 Intel Corporation.
This program is free software; you can redistribute it and/or modify it
under the terms and conditions of the GNU General Public License,
@@ -640,7 +640,7 @@
s32 ret_val;
u16 phy_data;
- /* Enable CRS on TX. This must be set for half-duplex operation. */
+ /* Enable CRS on Tx. This must be set for half-duplex operation. */
ret_val = e1e_rphy(hw, I82577_CFG_REG, &phy_data);
if (ret_val)
goto out;
diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c
index 6de4675..119aa20 100644
--- a/drivers/net/gianfar.c
+++ b/drivers/net/gianfar.c
@@ -434,7 +434,6 @@
static struct net_device_stats *gfar_get_stats(struct net_device *dev)
{
struct gfar_private *priv = netdev_priv(dev);
- struct netdev_queue *txq;
unsigned long rx_packets = 0, rx_bytes = 0, rx_dropped = 0;
unsigned long tx_packets = 0, tx_bytes = 0;
int i = 0;
@@ -450,9 +449,8 @@
dev->stats.rx_dropped = rx_dropped;
for (i = 0; i < priv->num_tx_queues; i++) {
- txq = netdev_get_tx_queue(dev, i);
- tx_bytes += txq->tx_bytes;
- tx_packets += txq->tx_packets;
+ tx_bytes += priv->tx_queue[i]->stats.tx_bytes;
+ tx_packets += priv->tx_queue[i]->stats.tx_packets;
}
dev->stats.tx_bytes = tx_bytes;
@@ -2109,8 +2107,8 @@
}
/* Update transmit stats */
- txq->tx_bytes += skb->len;
- txq->tx_packets ++;
+ tx_queue->stats.tx_bytes += skb->len;
+ tx_queue->stats.tx_packets++;
txbdp = txbdp_start = tx_queue->cur_tx;
lstatus = txbdp->lstatus;
diff --git a/drivers/net/gianfar.h b/drivers/net/gianfar.h
index 68984eb..54de413 100644
--- a/drivers/net/gianfar.h
+++ b/drivers/net/gianfar.h
@@ -907,12 +907,21 @@
MQ_MG_MODE
};
+/*
+ * Per TX queue stats
+ */
+struct tx_q_stats {
+ unsigned long tx_packets;
+ unsigned long tx_bytes;
+};
+
/**
* struct gfar_priv_tx_q - per tx queue structure
* @txlock: per queue tx spin lock
* @tx_skbuff:skb pointers
* @skb_curtx: to be used skb pointer
* @skb_dirtytx:the last used skb pointer
+ * @stats: bytes/packets stats
* @qindex: index of this queue
* @dev: back pointer to the dev structure
* @grp: back pointer to the group to which this queue belongs
@@ -934,6 +943,7 @@
struct txbd8 *tx_bd_base;
struct txbd8 *cur_tx;
struct txbd8 *dirty_tx;
+ struct tx_q_stats stats;
struct net_device *dev;
struct gfar_priv_grp *grp;
u16 skb_curtx;
diff --git a/drivers/net/greth.c b/drivers/net/greth.c
index 27d6960..fdb0333 100644
--- a/drivers/net/greth.c
+++ b/drivers/net/greth.c
@@ -1,7 +1,7 @@
/*
* Aeroflex Gaisler GRETH 10/100/1G Ethernet MAC.
*
- * 2005-2009 (c) Aeroflex Gaisler AB
+ * 2005-2010 (c) Aeroflex Gaisler AB
*
* This driver supports GRETH 10/100 and GRETH 10/100/1G Ethernet MACs
* available in the GRLIB VHDL IP core library.
@@ -356,6 +356,8 @@
dev_dbg(&dev->dev, " starting queue\n");
netif_start_queue(dev);
+ GRETH_REGSAVE(greth->regs->status, 0xFF);
+
napi_enable(&greth->napi);
greth_enable_irqs(greth);
@@ -371,7 +373,9 @@
napi_disable(&greth->napi);
+ greth_disable_irqs(greth);
greth_disable_tx(greth);
+ greth_disable_rx(greth);
netif_stop_queue(dev);
@@ -388,12 +392,20 @@
struct greth_private *greth = netdev_priv(dev);
struct greth_bd *bdp;
int err = NETDEV_TX_OK;
- u32 status, dma_addr;
+ u32 status, dma_addr, ctrl;
+ unsigned long flags;
- bdp = greth->tx_bd_base + greth->tx_next;
+ /* Clean TX Ring */
+ greth_clean_tx(greth->netdev);
if (unlikely(greth->tx_free <= 0)) {
+ spin_lock_irqsave(&greth->devlock, flags);/*save from poll/irq*/
+ ctrl = GRETH_REGLOAD(greth->regs->control);
+ /* Enable TX IRQ only if not already in poll() routine */
+ if (ctrl & GRETH_RXI)
+ GRETH_REGSAVE(greth->regs->control, ctrl | GRETH_TXI);
netif_stop_queue(dev);
+ spin_unlock_irqrestore(&greth->devlock, flags);
return NETDEV_TX_BUSY;
}
@@ -406,13 +418,14 @@
goto out;
}
+ bdp = greth->tx_bd_base + greth->tx_next;
dma_addr = greth_read_bd(&bdp->addr);
memcpy((unsigned char *) phys_to_virt(dma_addr), skb->data, skb->len);
dma_sync_single_for_device(greth->dev, dma_addr, skb->len, DMA_TO_DEVICE);
- status = GRETH_BD_EN | (skb->len & GRETH_BD_LEN);
+ status = GRETH_BD_EN | GRETH_BD_IE | (skb->len & GRETH_BD_LEN);
/* Wrap around descriptor ring */
if (greth->tx_next == GRETH_TXBD_NUM_MASK) {
@@ -422,22 +435,11 @@
greth->tx_next = NEXT_TX(greth->tx_next);
greth->tx_free--;
- /* No more descriptors */
- if (unlikely(greth->tx_free == 0)) {
-
- /* Free transmitted descriptors */
- greth_clean_tx(dev);
-
- /* If nothing was cleaned, stop queue & wait for irq */
- if (unlikely(greth->tx_free == 0)) {
- status |= GRETH_BD_IE;
- netif_stop_queue(dev);
- }
- }
-
/* Write descriptor control word and enable transmission */
greth_write_bd(&bdp->stat, status);
+ spin_lock_irqsave(&greth->devlock, flags); /*save from poll/irq*/
greth_enable_tx(greth);
+ spin_unlock_irqrestore(&greth->devlock, flags);
out:
dev_kfree_skb(skb);
@@ -450,13 +452,23 @@
{
struct greth_private *greth = netdev_priv(dev);
struct greth_bd *bdp;
- u32 status = 0, dma_addr;
+ u32 status = 0, dma_addr, ctrl;
int curr_tx, nr_frags, i, err = NETDEV_TX_OK;
+ unsigned long flags;
nr_frags = skb_shinfo(skb)->nr_frags;
+ /* Clean TX Ring */
+ greth_clean_tx_gbit(dev);
+
if (greth->tx_free < nr_frags + 1) {
+ spin_lock_irqsave(&greth->devlock, flags);/*save from poll/irq*/
+ ctrl = GRETH_REGLOAD(greth->regs->control);
+ /* Enable TX IRQ only if not already in poll() routine */
+ if (ctrl & GRETH_RXI)
+ GRETH_REGSAVE(greth->regs->control, ctrl | GRETH_TXI);
netif_stop_queue(dev);
+ spin_unlock_irqrestore(&greth->devlock, flags);
err = NETDEV_TX_BUSY;
goto out;
}
@@ -499,7 +511,7 @@
greth->tx_skbuff[curr_tx] = NULL;
bdp = greth->tx_bd_base + curr_tx;
- status = GRETH_TXBD_CSALL;
+ status = GRETH_TXBD_CSALL | GRETH_BD_EN;
status |= frag->size & GRETH_BD_LEN;
/* Wrap around descriptor ring */
@@ -509,14 +521,8 @@
/* More fragments left */
if (i < nr_frags - 1)
status |= GRETH_TXBD_MORE;
-
- /* ... last fragment, check if out of descriptors */
- else if (greth->tx_free - nr_frags - 1 < (MAX_SKB_FRAGS + 1)) {
-
- /* Enable interrupts and stop queue */
- status |= GRETH_BD_IE;
- netif_stop_queue(dev);
- }
+ else
+ status |= GRETH_BD_IE; /* enable IRQ on last fragment */
greth_write_bd(&bdp->stat, status);
@@ -536,26 +542,29 @@
wmb();
- /* Enable the descriptors that we configured ... */
- for (i = 0; i < nr_frags + 1; i++) {
- bdp = greth->tx_bd_base + greth->tx_next;
- greth_write_bd(&bdp->stat, greth_read_bd(&bdp->stat) | GRETH_BD_EN);
- greth->tx_next = NEXT_TX(greth->tx_next);
- greth->tx_free--;
- }
+ /* Enable the descriptor chain by enabling the first descriptor */
+ bdp = greth->tx_bd_base + greth->tx_next;
+ greth_write_bd(&bdp->stat, greth_read_bd(&bdp->stat) | GRETH_BD_EN);
+ greth->tx_next = curr_tx;
+ greth->tx_free -= nr_frags + 1;
+ wmb();
+
+ spin_lock_irqsave(&greth->devlock, flags); /*save from poll/irq*/
greth_enable_tx(greth);
+ spin_unlock_irqrestore(&greth->devlock, flags);
return NETDEV_TX_OK;
frag_map_error:
- /* Unmap SKB mappings that succeeded */
+ /* Unmap SKB mappings that succeeded and disable descriptor */
for (i = 0; greth->tx_next + i != curr_tx; i++) {
bdp = greth->tx_bd_base + greth->tx_next + i;
dma_unmap_single(greth->dev,
greth_read_bd(&bdp->addr),
greth_read_bd(&bdp->stat) & GRETH_BD_LEN,
DMA_TO_DEVICE);
+ greth_write_bd(&bdp->stat, 0);
}
map_error:
if (net_ratelimit())
@@ -565,12 +574,11 @@
return err;
}
-
static irqreturn_t greth_interrupt(int irq, void *dev_id)
{
struct net_device *dev = dev_id;
struct greth_private *greth;
- u32 status;
+ u32 status, ctrl;
irqreturn_t retval = IRQ_NONE;
greth = netdev_priv(dev);
@@ -580,13 +588,15 @@
/* Get the interrupt events that caused us to be here. */
status = GRETH_REGLOAD(greth->regs->status);
+ /* Must see if interrupts are enabled also, INT_TX|INT_RX flags may be
+ * set regardless of whether IRQ is enabled or not. Especially
+ * important when shared IRQ.
+ */
+ ctrl = GRETH_REGLOAD(greth->regs->control);
+
/* Handle rx and tx interrupts through poll */
- if (status & (GRETH_INT_RX | GRETH_INT_TX)) {
-
- /* Clear interrupt status */
- GRETH_REGORIN(greth->regs->status,
- status & (GRETH_INT_RX | GRETH_INT_TX));
-
+ if (((status & (GRETH_INT_RE | GRETH_INT_RX)) && (ctrl & GRETH_RXI)) ||
+ ((status & (GRETH_INT_TE | GRETH_INT_TX)) && (ctrl & GRETH_TXI))) {
retval = IRQ_HANDLED;
/* Disable interrupts and schedule poll() */
@@ -610,6 +620,8 @@
while (1) {
bdp = greth->tx_bd_base + greth->tx_last;
+ GRETH_REGSAVE(greth->regs->status, GRETH_INT_TE | GRETH_INT_TX);
+ mb();
stat = greth_read_bd(&bdp->stat);
if (unlikely(stat & GRETH_BD_EN))
@@ -670,7 +682,10 @@
/* We only clean fully completed SKBs */
bdp_last_frag = greth->tx_bd_base + SKIP_TX(greth->tx_last, nr_frags);
- stat = bdp_last_frag->stat;
+
+ GRETH_REGSAVE(greth->regs->status, GRETH_INT_TE | GRETH_INT_TX);
+ mb();
+ stat = greth_read_bd(&bdp_last_frag->stat);
if (stat & GRETH_BD_EN)
break;
@@ -702,21 +717,9 @@
greth->tx_free += nr_frags+1;
dev_kfree_skb(skb);
}
- if (greth->tx_free > (MAX_SKB_FRAGS + 1)) {
- netif_wake_queue(dev);
- }
-}
-static int greth_pending_packets(struct greth_private *greth)
-{
- struct greth_bd *bdp;
- u32 status;
- bdp = greth->rx_bd_base + greth->rx_cur;
- status = greth_read_bd(&bdp->stat);
- if (status & GRETH_BD_EN)
- return 0;
- else
- return 1;
+ if (netif_queue_stopped(dev) && (greth->tx_free > (MAX_SKB_FRAGS+1)))
+ netif_wake_queue(dev);
}
static int greth_rx(struct net_device *dev, int limit)
@@ -727,20 +730,24 @@
int pkt_len;
int bad, count;
u32 status, dma_addr;
+ unsigned long flags;
greth = netdev_priv(dev);
for (count = 0; count < limit; ++count) {
bdp = greth->rx_bd_base + greth->rx_cur;
+ GRETH_REGSAVE(greth->regs->status, GRETH_INT_RE | GRETH_INT_RX);
+ mb();
status = greth_read_bd(&bdp->stat);
- dma_addr = greth_read_bd(&bdp->addr);
- bad = 0;
if (unlikely(status & GRETH_BD_EN)) {
break;
}
+ dma_addr = greth_read_bd(&bdp->addr);
+ bad = 0;
+
/* Check status for errors. */
if (unlikely(status & GRETH_RXBD_STATUS)) {
if (status & GRETH_RXBD_ERR_FT) {
@@ -802,7 +809,9 @@
dma_sync_single_for_device(greth->dev, dma_addr, MAX_FRAME_SIZE, DMA_FROM_DEVICE);
+ spin_lock_irqsave(&greth->devlock, flags); /* save from XMIT */
greth_enable_rx(greth);
+ spin_unlock_irqrestore(&greth->devlock, flags);
greth->rx_cur = NEXT_RX(greth->rx_cur);
}
@@ -836,6 +845,7 @@
int pkt_len;
int bad, count = 0;
u32 status, dma_addr;
+ unsigned long flags;
greth = netdev_priv(dev);
@@ -843,6 +853,8 @@
bdp = greth->rx_bd_base + greth->rx_cur;
skb = greth->rx_skbuff[greth->rx_cur];
+ GRETH_REGSAVE(greth->regs->status, GRETH_INT_RE | GRETH_INT_RX);
+ mb();
status = greth_read_bd(&bdp->stat);
bad = 0;
@@ -865,10 +877,9 @@
}
}
- /* Allocate new skb to replace current */
- newskb = netdev_alloc_skb(dev, MAX_FRAME_SIZE + NET_IP_ALIGN);
-
- if (!bad && newskb) {
+ /* Allocate new skb to replace current, not needed if the
+ * current skb can be reused */
+ if (!bad && (newskb=netdev_alloc_skb(dev, MAX_FRAME_SIZE + NET_IP_ALIGN))) {
skb_reserve(newskb, NET_IP_ALIGN);
dma_addr = dma_map_single(greth->dev,
@@ -905,11 +916,22 @@
if (net_ratelimit())
dev_warn(greth->dev, "Could not create DMA mapping, dropping packet\n");
dev_kfree_skb(newskb);
+ /* reusing current skb, so it is a drop */
dev->stats.rx_dropped++;
}
+ } else if (bad) {
+ /* Bad Frame transfer, the skb is reused */
+ dev->stats.rx_dropped++;
} else {
+ /* Failed Allocating a new skb. This is rather stupid
+ * but the current "filled" skb is reused, as if
+ * transfer failure. One could argue that RX descriptor
+ * table handling should be divided into cleaning and
+ * filling as the TX part of the driver
+ */
if (net_ratelimit())
dev_warn(greth->dev, "Could not allocate SKB, dropping packet\n");
+ /* reusing current skb, so it is a drop */
dev->stats.rx_dropped++;
}
@@ -920,7 +942,9 @@
wmb();
greth_write_bd(&bdp->stat, status);
+ spin_lock_irqsave(&greth->devlock, flags);
greth_enable_rx(greth);
+ spin_unlock_irqrestore(&greth->devlock, flags);
greth->rx_cur = NEXT_RX(greth->rx_cur);
}
@@ -932,15 +956,18 @@
{
struct greth_private *greth;
int work_done = 0;
+ unsigned long flags;
+ u32 mask, ctrl;
greth = container_of(napi, struct greth_private, napi);
- if (greth->gbit_mac) {
- greth_clean_tx_gbit(greth->netdev);
- } else {
- greth_clean_tx(greth->netdev);
+restart_txrx_poll:
+ if (netif_queue_stopped(greth->netdev)) {
+ if (greth->gbit_mac)
+ greth_clean_tx_gbit(greth->netdev);
+ else
+ greth_clean_tx(greth->netdev);
}
-restart_poll:
if (greth->gbit_mac) {
work_done += greth_rx_gbit(greth->netdev, budget - work_done);
} else {
@@ -949,15 +976,29 @@
if (work_done < budget) {
- napi_complete(napi);
+ spin_lock_irqsave(&greth->devlock, flags);
- if (greth_pending_packets(greth)) {
- napi_reschedule(napi);
- goto restart_poll;
+ ctrl = GRETH_REGLOAD(greth->regs->control);
+ if (netif_queue_stopped(greth->netdev)) {
+ GRETH_REGSAVE(greth->regs->control,
+ ctrl | GRETH_TXI | GRETH_RXI);
+ mask = GRETH_INT_RX | GRETH_INT_RE |
+ GRETH_INT_TX | GRETH_INT_TE;
+ } else {
+ GRETH_REGSAVE(greth->regs->control, ctrl | GRETH_RXI);
+ mask = GRETH_INT_RX | GRETH_INT_RE;
+ }
+
+ if (GRETH_REGLOAD(greth->regs->status) & mask) {
+ GRETH_REGSAVE(greth->regs->control, ctrl);
+ spin_unlock_irqrestore(&greth->devlock, flags);
+ goto restart_txrx_poll;
+ } else {
+ __napi_complete(napi);
+ spin_unlock_irqrestore(&greth->devlock, flags);
}
}
- greth_enable_irqs(greth);
return work_done;
}
@@ -1152,11 +1193,11 @@
};
static struct net_device_ops greth_netdev_ops = {
- .ndo_open = greth_open,
- .ndo_stop = greth_close,
- .ndo_start_xmit = greth_start_xmit,
- .ndo_set_mac_address = greth_set_mac_add,
- .ndo_validate_addr = eth_validate_addr,
+ .ndo_open = greth_open,
+ .ndo_stop = greth_close,
+ .ndo_start_xmit = greth_start_xmit,
+ .ndo_set_mac_address = greth_set_mac_add,
+ .ndo_validate_addr = eth_validate_addr,
};
static inline int wait_for_mdio(struct greth_private *greth)
@@ -1217,29 +1258,26 @@
struct greth_private *greth = netdev_priv(dev);
struct phy_device *phydev = greth->phy;
unsigned long flags;
-
int status_change = 0;
+ u32 ctrl;
spin_lock_irqsave(&greth->devlock, flags);
if (phydev->link) {
if ((greth->speed != phydev->speed) || (greth->duplex != phydev->duplex)) {
-
- GRETH_REGANDIN(greth->regs->control,
- ~(GRETH_CTRL_FD | GRETH_CTRL_SP | GRETH_CTRL_GB));
+ ctrl = GRETH_REGLOAD(greth->regs->control) &
+ ~(GRETH_CTRL_FD | GRETH_CTRL_SP | GRETH_CTRL_GB);
if (phydev->duplex)
- GRETH_REGORIN(greth->regs->control, GRETH_CTRL_FD);
+ ctrl |= GRETH_CTRL_FD;
- if (phydev->speed == SPEED_100) {
-
- GRETH_REGORIN(greth->regs->control, GRETH_CTRL_SP);
- }
-
+ if (phydev->speed == SPEED_100)
+ ctrl |= GRETH_CTRL_SP;
else if (phydev->speed == SPEED_1000)
- GRETH_REGORIN(greth->regs->control, GRETH_CTRL_GB);
+ ctrl |= GRETH_CTRL_GB;
+ GRETH_REGSAVE(greth->regs->control, ctrl);
greth->speed = phydev->speed;
greth->duplex = phydev->duplex;
status_change = 1;
@@ -1600,6 +1638,9 @@
{
.name = "GAISLER_ETHMAC",
},
+ {
+ .name = "01_01d",
+ },
{},
};
diff --git a/drivers/net/greth.h b/drivers/net/greth.h
index 03ad903..be0f206 100644
--- a/drivers/net/greth.h
+++ b/drivers/net/greth.h
@@ -23,6 +23,7 @@
#define GRETH_BD_LEN 0x7FF
#define GRETH_TXEN 0x1
+#define GRETH_INT_TE 0x2
#define GRETH_INT_TX 0x8
#define GRETH_TXI 0x4
#define GRETH_TXBD_STATUS 0x0001C000
@@ -35,6 +36,7 @@
#define GRETH_TXBD_ERR_UE 0x4000
#define GRETH_TXBD_ERR_AL 0x8000
+#define GRETH_INT_RE 0x1
#define GRETH_INT_RX 0x4
#define GRETH_RXEN 0x2
#define GRETH_RXI 0x8
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index a060610..602078b 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -6667,8 +6667,6 @@
struct ixgbe_adapter *adapter,
struct ixgbe_ring *tx_ring)
{
- struct net_device *netdev = tx_ring->netdev;
- struct netdev_queue *txq;
unsigned int first;
unsigned int tx_flags = 0;
u8 hdr_len = 0;
@@ -6765,9 +6763,6 @@
/* add the ATR filter if ATR is on */
if (test_bit(__IXGBE_TX_FDIR_INIT_DONE, &tx_ring->state))
ixgbe_atr(tx_ring, skb, tx_flags, protocol);
- txq = netdev_get_tx_queue(netdev, tx_ring->queue_index);
- txq->tx_bytes += skb->len;
- txq->tx_packets++;
ixgbe_tx_queue(tx_ring, tx_flags, count, skb->len, hdr_len);
ixgbe_maybe_stop_tx(tx_ring, DESC_NEEDED);
@@ -6925,8 +6920,6 @@
struct ixgbe_adapter *adapter = netdev_priv(netdev);
int i;
- /* accurate rx/tx bytes/packets stats */
- dev_txq_stats_fold(netdev, stats);
rcu_read_lock();
for (i = 0; i < adapter->num_rx_queues; i++) {
struct ixgbe_ring *ring = ACCESS_ONCE(adapter->rx_ring[i]);
@@ -6943,6 +6936,22 @@
stats->rx_bytes += bytes;
}
}
+
+ for (i = 0; i < adapter->num_tx_queues; i++) {
+ struct ixgbe_ring *ring = ACCESS_ONCE(adapter->tx_ring[i]);
+ u64 bytes, packets;
+ unsigned int start;
+
+ if (ring) {
+ do {
+ start = u64_stats_fetch_begin_bh(&ring->syncp);
+ packets = ring->stats.packets;
+ bytes = ring->stats.bytes;
+ } while (u64_stats_fetch_retry_bh(&ring->syncp, start));
+ stats->tx_packets += packets;
+ stats->tx_bytes += bytes;
+ }
+ }
rcu_read_unlock();
/* following stats updated by ixgbe_watchdog_task() */
stats->multicast = netdev->stats.multicast;
diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 21845af..5933621 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -585,7 +585,7 @@
rcu_read_lock_bh();
vlan = rcu_dereference(q->vlan);
if (vlan)
- netdev_get_tx_queue(vlan->dev, 0)->tx_dropped++;
+ vlan->dev->stats.tx_dropped++;
rcu_read_unlock_bh();
return err;
diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c
index a37fcf1..ea5cfe2 100644
--- a/drivers/net/myri10ge/myri10ge.c
+++ b/drivers/net/myri10ge/myri10ge.c
@@ -3403,9 +3403,7 @@
return -EIO;
}
- status = pci_restore_state(pdev);
- if (status)
- return status;
+ pci_restore_state(pdev);
status = pci_enable_device(pdev);
if (status) {
diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index bb8645a..bde7d61 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -554,6 +554,8 @@
struct mii_if_info mii;
struct rtl8169_counters counters;
u32 saved_wolopts;
+
+ const struct firmware *fw;
};
MODULE_AUTHOR("Realtek and the Linux r8169 crew <netdev@vger.kernel.org>");
@@ -1766,6 +1768,29 @@
}
}
+static void rtl_release_firmware(struct rtl8169_private *tp)
+{
+ release_firmware(tp->fw);
+ tp->fw = NULL;
+}
+
+static int rtl_apply_firmware(struct rtl8169_private *tp, const char *fw_name)
+{
+ const struct firmware **fw = &tp->fw;
+ int rc = !*fw;
+
+ if (rc) {
+ rc = request_firmware(fw, fw_name, &tp->pci_dev->dev);
+ if (rc < 0)
+ goto out;
+ }
+
+ /* TODO: release firmware once rtl_phy_write_fw signals failures. */
+ rtl_phy_write_fw(tp, *fw);
+out:
+ return rc;
+}
+
static void rtl8169s_hw_phy_config(struct rtl8169_private *tp)
{
static const struct phy_reg phy_reg_init[] = {
@@ -2139,7 +2164,6 @@
{ 0x0d, 0xf880 }
};
void __iomem *ioaddr = tp->mmio_addr;
- const struct firmware *fw;
rtl_writephy_batch(tp, phy_reg_init_0, ARRAY_SIZE(phy_reg_init_0));
@@ -2203,11 +2227,8 @@
rtl_writephy(tp, 0x1f, 0x0005);
rtl_writephy(tp, 0x05, 0x001b);
- if (rtl_readphy(tp, 0x06) == 0xbf00 &&
- request_firmware(&fw, FIRMWARE_8168D_1, &tp->pci_dev->dev) == 0) {
- rtl_phy_write_fw(tp, fw);
- release_firmware(fw);
- } else {
+ if ((rtl_readphy(tp, 0x06) != 0xbf00) ||
+ (rtl_apply_firmware(tp, FIRMWARE_8168D_1) < 0)) {
netif_warn(tp, probe, tp->dev, "unable to apply firmware patch\n");
}
@@ -2257,7 +2278,6 @@
{ 0x0d, 0xf880 }
};
void __iomem *ioaddr = tp->mmio_addr;
- const struct firmware *fw;
rtl_writephy_batch(tp, phy_reg_init_0, ARRAY_SIZE(phy_reg_init_0));
@@ -2312,11 +2332,8 @@
rtl_writephy(tp, 0x1f, 0x0005);
rtl_writephy(tp, 0x05, 0x001b);
- if (rtl_readphy(tp, 0x06) == 0xb300 &&
- request_firmware(&fw, FIRMWARE_8168D_2, &tp->pci_dev->dev) == 0) {
- rtl_phy_write_fw(tp, fw);
- release_firmware(fw);
- } else {
+ if ((rtl_readphy(tp, 0x06) != 0xb300) ||
+ (rtl_apply_firmware(tp, FIRMWARE_8168D_2) < 0)) {
netif_warn(tp, probe, tp->dev, "unable to apply firmware patch\n");
}
@@ -3200,6 +3217,8 @@
cancel_delayed_work_sync(&tp->task);
+ rtl_release_firmware(tp);
+
unregister_netdev(dev);
if (pci_dev_run_wake(pdev))
diff --git a/drivers/net/sfc/efx.c b/drivers/net/sfc/efx.c
index 711449c..002bac7 100644
--- a/drivers/net/sfc/efx.c
+++ b/drivers/net/sfc/efx.c
@@ -1153,6 +1153,9 @@
int count;
int cpu;
+ if (rss_cpus)
+ return rss_cpus;
+
if (unlikely(!zalloc_cpumask_var(&core_mask, GFP_KERNEL))) {
printk(KERN_WARNING
"sfc: RSS disabled due to allocation failure\n");
@@ -1266,27 +1269,18 @@
efx->legacy_irq = 0;
}
-struct efx_tx_queue *
-efx_get_tx_queue(struct efx_nic *efx, unsigned index, unsigned type)
-{
- unsigned tx_channel_offset =
- separate_tx_channels ? efx->n_channels - efx->n_tx_channels : 0;
- EFX_BUG_ON_PARANOID(index >= efx->n_tx_channels ||
- type >= EFX_TXQ_TYPES);
- return &efx->channel[tx_channel_offset + index]->tx_queue[type];
-}
-
static void efx_set_channels(struct efx_nic *efx)
{
struct efx_channel *channel;
struct efx_tx_queue *tx_queue;
- unsigned tx_channel_offset =
+
+ efx->tx_channel_offset =
separate_tx_channels ? efx->n_channels - efx->n_tx_channels : 0;
/* Channel pointers were set in efx_init_struct() but we now
* need to clear them for TX queues in any RX-only channels. */
efx_for_each_channel(channel, efx) {
- if (channel->channel - tx_channel_offset >=
+ if (channel->channel - efx->tx_channel_offset >=
efx->n_tx_channels) {
efx_for_each_channel_tx_queue(tx_queue, channel)
tx_queue->channel = NULL;
diff --git a/drivers/net/sfc/falcon.c b/drivers/net/sfc/falcon.c
index 70e4f7d..61ddd2c 100644
--- a/drivers/net/sfc/falcon.c
+++ b/drivers/net/sfc/falcon.c
@@ -1107,22 +1107,9 @@
/* Restore PCI configuration if needed */
if (method == RESET_TYPE_WORLD) {
- if (efx_nic_is_dual_func(efx)) {
- rc = pci_restore_state(nic_data->pci_dev2);
- if (rc) {
- netif_err(efx, drv, efx->net_dev,
- "failed to restore PCI config for "
- "the secondary function\n");
- goto fail3;
- }
- }
- rc = pci_restore_state(efx->pci_dev);
- if (rc) {
- netif_err(efx, drv, efx->net_dev,
- "failed to restore PCI config for the "
- "primary function\n");
- goto fail4;
- }
+ if (efx_nic_is_dual_func(efx))
+ pci_restore_state(nic_data->pci_dev2);
+ pci_restore_state(efx->pci_dev);
netif_dbg(efx, drv, efx->net_dev,
"successfully restored PCI config\n");
}
@@ -1133,7 +1120,7 @@
rc = -ETIMEDOUT;
netif_err(efx, hw, efx->net_dev,
"timed out waiting for hardware reset\n");
- goto fail5;
+ goto fail3;
}
netif_dbg(efx, hw, efx->net_dev, "hardware reset complete\n");
@@ -1141,11 +1128,9 @@
/* pci_save_state() and pci_restore_state() MUST be called in pairs */
fail2:
-fail3:
pci_restore_state(efx->pci_dev);
fail1:
-fail4:
-fail5:
+fail3:
return rc;
}
diff --git a/drivers/net/sfc/net_driver.h b/drivers/net/sfc/net_driver.h
index bdce66d..28df866 100644
--- a/drivers/net/sfc/net_driver.h
+++ b/drivers/net/sfc/net_driver.h
@@ -735,6 +735,7 @@
unsigned next_buffer_table;
unsigned n_channels;
unsigned n_rx_channels;
+ unsigned tx_channel_offset;
unsigned n_tx_channels;
unsigned int rx_buffer_len;
unsigned int rx_buffer_order;
@@ -929,8 +930,13 @@
_channel = (_channel->channel + 1 < (_efx)->n_channels) ? \
(_efx)->channel[_channel->channel + 1] : NULL)
-extern struct efx_tx_queue *
-efx_get_tx_queue(struct efx_nic *efx, unsigned index, unsigned type);
+static inline struct efx_tx_queue *
+efx_get_tx_queue(struct efx_nic *efx, unsigned index, unsigned type)
+{
+ EFX_BUG_ON_PARANOID(index >= efx->n_tx_channels ||
+ type >= EFX_TXQ_TYPES);
+ return &efx->channel[efx->tx_channel_offset + index]->tx_queue[type];
+}
static inline struct efx_tx_queue *
efx_channel_get_tx_queue(struct efx_channel *channel, unsigned type)
diff --git a/drivers/net/tile/tilepro.c b/drivers/net/tile/tilepro.c
index 0e6bac5..7cb301d 100644
--- a/drivers/net/tile/tilepro.c
+++ b/drivers/net/tile/tilepro.c
@@ -142,14 +142,6 @@
MODULE_AUTHOR("Tilera");
MODULE_LICENSE("GPL");
-
-#define IS_MULTICAST(mac_addr) \
- (((u8 *)(mac_addr))[0] & 0x01)
-
-#define IS_BROADCAST(mac_addr) \
- (((u16 *)(mac_addr))[0] == 0xffff)
-
-
/*
* Queue of incoming packets for a specific cpu and device.
*
@@ -795,7 +787,7 @@
/*
* FIXME: Implement HW multicast filter.
*/
- if (!IS_MULTICAST(buf) && !IS_BROADCAST(buf)) {
+ if (is_unicast_ether_addr(buf)) {
/* Filter packets not for our address. */
const u8 *mine = dev->dev_addr;
filter = compare_ether_addr(mine, buf);
diff --git a/drivers/net/ucc_geth.c b/drivers/net/ucc_geth.c
index 73a3e0d..715e7b4 100644
--- a/drivers/net/ucc_geth.c
+++ b/drivers/net/ucc_geth.c
@@ -2032,7 +2032,7 @@
netdev_for_each_mc_addr(ha, dev) {
/* Only support group multicast for now.
*/
- if (!(ha->addr[0] & 1))
+ if (!is_multicast_ether_addr(ha->addr))
continue;
/* Ask CPM to run CRC and set bit in
diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c
index 593c104..d776c4a 100644
--- a/drivers/net/usb/cdc_ncm.c
+++ b/drivers/net/usb/cdc_ncm.c
@@ -1021,13 +1021,15 @@
(temp > CDC_NCM_MAX_DATAGRAM_SIZE) || (temp < ETH_HLEN)) {
pr_debug("invalid frame detected (ignored)"
"offset[%u]=%u, length=%u, skb=%p\n",
- x, offset, temp, skb);
+ x, offset, temp, skb_in);
if (!x)
goto error;
break;
} else {
skb = skb_clone(skb_in, GFP_ATOMIC);
+ if (!skb)
+ goto error;
skb->len = temp;
skb->data = ((u8 *)skb_in->data) + offset;
skb_set_tail_pointer(skb, temp);
diff --git a/drivers/net/vxge/vxge-main.c b/drivers/net/vxge/vxge-main.c
index 1ac9b56..c81a651 100644
--- a/drivers/net/vxge/vxge-main.c
+++ b/drivers/net/vxge/vxge-main.c
@@ -4120,6 +4120,7 @@
"hotplug event.\n");
out:
+ release_firmware(fw);
return ret;
}
diff --git a/drivers/net/wireless/ath/ath9k/ar9002_calib.c b/drivers/net/wireless/ath/ath9k/ar9002_calib.c
index 01880aa..ea2e7d7 100644
--- a/drivers/net/wireless/ath/ath9k/ar9002_calib.c
+++ b/drivers/net/wireless/ath/ath9k/ar9002_calib.c
@@ -954,6 +954,9 @@
&adc_dc_cal_multi_sample;
}
ah->supp_cals = ADC_GAIN_CAL | ADC_DC_CAL | IQ_MISMATCH_CAL;
+
+ if (AR_SREV_9287(ah))
+ ah->supp_cals &= ~ADC_GAIN_CAL;
}
}
diff --git a/drivers/net/wireless/ath/ath9k/eeprom_def.c b/drivers/net/wireless/ath/ath9k/eeprom_def.c
index 088f141..749a936 100644
--- a/drivers/net/wireless/ath/ath9k/eeprom_def.c
+++ b/drivers/net/wireless/ath/ath9k/eeprom_def.c
@@ -226,6 +226,10 @@
eep->baseEepHeader.pwdclkind == 0)
ah->need_an_top2_fixup = 1;
+ if ((common->bus_ops->ath_bus_type == ATH_USB) &&
+ (AR_SREV_9280(ah)))
+ eep->modalHeader[0].xpaBiasLvl = 0;
+
return 0;
}
diff --git a/drivers/net/wireless/ath/ath9k/htc.h b/drivers/net/wireless/ath/ath9k/htc.h
index a099b3e..1ce506f 100644
--- a/drivers/net/wireless/ath/ath9k/htc.h
+++ b/drivers/net/wireless/ath/ath9k/htc.h
@@ -433,6 +433,7 @@
void ath9k_htc_beaconep(void *drv_priv, struct sk_buff *skb,
enum htc_endpoint_id ep_id, bool txok);
+int ath9k_htc_update_cap_target(struct ath9k_htc_priv *priv);
void ath9k_htc_station_work(struct work_struct *work);
void ath9k_htc_aggr_work(struct work_struct *work);
void ath9k_ani_work(struct work_struct *work);;
diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_main.c b/drivers/net/wireless/ath/ath9k/htc_drv_main.c
index 845b4c9..f4d576b 100644
--- a/drivers/net/wireless/ath/ath9k/htc_drv_main.c
+++ b/drivers/net/wireless/ath/ath9k/htc_drv_main.c
@@ -301,6 +301,16 @@
priv->nstations++;
+ /*
+ * Set chainmask etc. on the target.
+ */
+ ret = ath9k_htc_update_cap_target(priv);
+ if (ret)
+ ath_dbg(common, ATH_DBG_CONFIG,
+ "Failed to update capability in target\n");
+
+ priv->ah->is_monitoring = true;
+
return 0;
err_vif:
@@ -328,6 +338,7 @@
}
priv->nstations--;
+ priv->ah->is_monitoring = false;
return 0;
}
@@ -419,7 +430,7 @@
return 0;
}
-static int ath9k_htc_update_cap_target(struct ath9k_htc_priv *priv)
+int ath9k_htc_update_cap_target(struct ath9k_htc_priv *priv)
{
struct ath9k_htc_cap_target tcap;
int ret;
@@ -1186,6 +1197,20 @@
}
}
+ /*
+ * Monitor interface should be added before
+ * IEEE80211_CONF_CHANGE_CHANNEL is handled.
+ */
+ if (changed & IEEE80211_CONF_CHANGE_MONITOR) {
+ if (conf->flags & IEEE80211_CONF_MONITOR) {
+ if (ath9k_htc_add_monitor_interface(priv))
+ ath_err(common, "Failed to set monitor mode\n");
+ else
+ ath_dbg(common, ATH_DBG_CONFIG,
+ "HW opmode set to Monitor mode\n");
+ }
+ }
+
if (changed & IEEE80211_CONF_CHANGE_CHANNEL) {
struct ieee80211_channel *curchan = hw->conf.channel;
int pos = curchan->hw_value;
@@ -1221,16 +1246,6 @@
ath_update_txpow(priv);
}
- if (changed & IEEE80211_CONF_CHANGE_MONITOR) {
- if (conf->flags & IEEE80211_CONF_MONITOR) {
- if (ath9k_htc_add_monitor_interface(priv))
- ath_err(common, "Failed to set monitor mode\n");
- else
- ath_dbg(common, ATH_DBG_CONFIG,
- "HW opmode set to Monitor mode\n");
- }
- }
-
if (changed & IEEE80211_CONF_CHANGE_IDLE) {
mutex_lock(&priv->htc_pm_lock);
if (!priv->ps_idle) {
diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c
index fde9786..1afb8bb 100644
--- a/drivers/net/wireless/ath/ath9k/hw.c
+++ b/drivers/net/wireless/ath/ath9k/hw.c
@@ -436,9 +436,10 @@
static int ath9k_hw_post_init(struct ath_hw *ah)
{
+ struct ath_common *common = ath9k_hw_common(ah);
int ecode;
- if (!AR_SREV_9271(ah)) {
+ if (common->bus_ops->ath_bus_type != ATH_USB) {
if (!ath9k_hw_chip_test(ah))
return -ENODEV;
}
@@ -1213,7 +1214,7 @@
ah->txchainmask = common->tx_chainmask;
ah->rxchainmask = common->rx_chainmask;
- if (!ah->chip_fullsleep) {
+ if ((common->bus_ops->ath_bus_type != ATH_USB) && !ah->chip_fullsleep) {
ath9k_hw_abortpcurecv(ah);
if (!ath9k_hw_stopdmarecv(ah)) {
ath_dbg(common, ATH_DBG_XMIT,
diff --git a/drivers/net/wireless/hostap/hostap_cs.c b/drivers/net/wireless/hostap/hostap_cs.c
index bd8a413..2176ede 100644
--- a/drivers/net/wireless/hostap/hostap_cs.c
+++ b/drivers/net/wireless/hostap/hostap_cs.c
@@ -518,22 +518,21 @@
hw_priv->link = link;
/*
- * Make sure the IRQ handler cannot proceed until at least
- * dev->base_addr is initialized.
+ * We enable IRQ here, but IRQ handler will not proceed
+ * until dev->base_addr is set below. This protect us from
+ * receive interrupts when driver is not initialized.
*/
- spin_lock_irqsave(&local->irq_init_lock, flags);
-
ret = pcmcia_request_irq(link, prism2_interrupt);
if (ret)
- goto failed_unlock;
+ goto failed;
ret = pcmcia_enable_device(link);
if (ret)
- goto failed_unlock;
+ goto failed;
+ spin_lock_irqsave(&local->irq_init_lock, flags);
dev->irq = link->irq;
dev->base_addr = link->resource[0]->start;
-
spin_unlock_irqrestore(&local->irq_init_lock, flags);
local->shutdown = 0;
@@ -546,8 +545,6 @@
return ret;
- failed_unlock:
- spin_unlock_irqrestore(&local->irq_init_lock, flags);
failed:
kfree(hw_priv);
prism2_release((u_long)link);
diff --git a/drivers/net/wireless/ipw2x00/ipw2200.c b/drivers/net/wireless/ipw2x00/ipw2200.c
index 8d6ed5f..ae438ed 100644
--- a/drivers/net/wireless/ipw2x00/ipw2200.c
+++ b/drivers/net/wireless/ipw2x00/ipw2200.c
@@ -1973,6 +1973,13 @@
inta = ipw_read32(priv, IPW_INTA_RW);
inta_mask = ipw_read32(priv, IPW_INTA_MASK_R);
+
+ if (inta == 0xFFFFFFFF) {
+ /* Hardware disappeared */
+ IPW_WARNING("TASKLET INTA == 0xFFFFFFFF\n");
+ /* Only handle the cached INTA values */
+ inta = 0;
+ }
inta &= (IPW_INTA_MASK_ALL & inta_mask);
/* Add any cached INTA values that need to be handled */
diff --git a/drivers/net/wireless/p54/txrx.c b/drivers/net/wireless/p54/txrx.c
index 76b2318a..f618b96 100644
--- a/drivers/net/wireless/p54/txrx.c
+++ b/drivers/net/wireless/p54/txrx.c
@@ -618,7 +618,7 @@
else
*burst_possible = false;
- if (info->flags & IEEE80211_TX_CTL_ASSIGN_SEQ)
+ if (!(info->flags & IEEE80211_TX_CTL_ASSIGN_SEQ))
*flags |= P54_HDR_FLAG_DATA_OUT_SEQNR;
if (info->flags & IEEE80211_TX_CTL_PSPOLL_RESPONSE)
diff --git a/drivers/net/wireless/rt2x00/rt2x00pci.c b/drivers/net/wireless/rt2x00/rt2x00pci.c
index 73631c6..ace0b66 100644
--- a/drivers/net/wireless/rt2x00/rt2x00pci.c
+++ b/drivers/net/wireless/rt2x00/rt2x00pci.c
@@ -363,12 +363,12 @@
struct rt2x00_dev *rt2x00dev = hw->priv;
if (pci_set_power_state(pci_dev, PCI_D0) ||
- pci_enable_device(pci_dev) ||
- pci_restore_state(pci_dev)) {
+ pci_enable_device(pci_dev)) {
ERROR(rt2x00dev, "Failed to resume device.\n");
return -EIO;
}
+ pci_restore_state(pci_dev);
return rt2x00lib_resume(rt2x00dev);
}
EXPORT_SYMBOL_GPL(rt2x00pci_resume);
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 7c24dce..44b0aee 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -168,8 +168,9 @@
u32 mask_bits = desc->masked;
unsigned offset = desc->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE +
PCI_MSIX_ENTRY_VECTOR_CTRL;
- mask_bits &= ~1;
- mask_bits |= flag;
+ mask_bits &= ~PCI_MSIX_ENTRY_CTRL_MASKBIT;
+ if (flag)
+ mask_bits |= PCI_MSIX_ENTRY_CTRL_MASKBIT;
writel(mask_bits, desc->mask_base + offset);
return mask_bits;
diff --git a/drivers/pci/msi.h b/drivers/pci/msi.h
index feff3be..65c42f8 100644
--- a/drivers/pci/msi.h
+++ b/drivers/pci/msi.h
@@ -6,12 +6,6 @@
#ifndef MSI_H
#define MSI_H
-#define PCI_MSIX_ENTRY_SIZE 16
-#define PCI_MSIX_ENTRY_LOWER_ADDR 0
-#define PCI_MSIX_ENTRY_UPPER_ADDR 4
-#define PCI_MSIX_ENTRY_DATA 8
-#define PCI_MSIX_ENTRY_VECTOR_CTRL 12
-
#define msi_control_reg(base) (base + PCI_MSI_FLAGS)
#define msi_lower_address_reg(base) (base + PCI_MSI_ADDRESS_LO)
#define msi_upper_address_reg(base) (base + PCI_MSI_ADDRESS_HI)
diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index 24e19c5..6fe0772 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -46,9 +46,9 @@
struct pci_dev *pci_dev = context;
if (event == ACPI_NOTIFY_DEVICE_WAKE && pci_dev) {
+ pci_wakeup_event(pci_dev);
pci_check_pme_status(pci_dev);
pm_runtime_resume(&pci_dev->dev);
- pci_wakeup_event(pci_dev);
if (pci_dev->subordinate)
pci_pme_wakeup_bus(pci_dev->subordinate);
}
@@ -399,6 +399,7 @@
if (acpi_gbl_FADT.boot_flags & ACPI_FADT_NO_ASPM) {
printk(KERN_INFO"ACPI FADT declares the system doesn't support PCIe ASPM, so disable it\n");
+ pcie_clear_aspm();
pcie_no_aspm();
}
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 8a6f797..88246dd 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -338,7 +338,7 @@
}
/**
- * __pci_device_probe()
+ * __pci_device_probe - check if a driver wants to claim a specific PCI device
* @drv: driver to call to check if it wants the PCI device
* @pci_dev: PCI device being probed
*
@@ -449,7 +449,8 @@
return error;
}
- return pci_restore_state(pci_dev);
+ pci_restore_state(pci_dev);
+ return 0;
}
static void pci_pm_default_resume_early(struct pci_dev *pci_dev)
diff --git a/drivers/pci/pci-stub.c b/drivers/pci/pci-stub.c
index f7b68ca..775e933 100644
--- a/drivers/pci/pci-stub.c
+++ b/drivers/pci/pci-stub.c
@@ -47,6 +47,10 @@
if (rc)
return rc;
+ /* no ids passed actually */
+ if (ids[0] == '\0')
+ return 0;
+
/* add ids specified in the module parameter */
p = ids;
while ((id = strsep(&p, ","))) {
@@ -54,6 +58,9 @@
subdevice = PCI_ANY_ID, class=0, class_mask=0;
int fields;
+ if (!strlen(id))
+ continue;
+
fields = sscanf(id, "%x:%x:%x:%x:%x:%x",
&vendor, &device, &subvendor, &subdevice,
&class, &class_mask);
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 63d5042..8ecaac9 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -1149,7 +1149,7 @@
sysfs_bin_attr_init(attr);
attr->size = rom_size;
attr->attr.name = "rom";
- attr->attr.mode = S_IRUSR;
+ attr->attr.mode = S_IRUSR | S_IWUSR;
attr->read = pci_read_rom;
attr->write = pci_write_rom;
retval = sysfs_create_bin_file(&pdev->dev.kobj, attr);
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 710c8a2..b714d78 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -937,14 +937,13 @@
* pci_restore_state - Restore the saved state of a PCI device
* @dev: - PCI device that we're dealing with
*/
-int
-pci_restore_state(struct pci_dev *dev)
+void pci_restore_state(struct pci_dev *dev)
{
int i;
u32 val;
if (!dev->state_saved)
- return 0;
+ return;
/* PCI Express register must be restored first */
pci_restore_pcie_state(dev);
@@ -968,8 +967,6 @@
pci_restore_iov_state(dev);
dev->state_saved = false;
-
- return 0;
}
static int do_pci_enable_device(struct pci_dev *dev, int bars)
@@ -1300,22 +1297,6 @@
return ret;
}
-/*
- * Time to wait before the system can be put into a sleep state after reporting
- * a wakeup event signaled by a PCI device.
- */
-#define PCI_WAKEUP_COOLDOWN 100
-
-/**
- * pci_wakeup_event - Report a wakeup event related to a given PCI device.
- * @dev: Device to report the wakeup event for.
- */
-void pci_wakeup_event(struct pci_dev *dev)
-{
- if (device_may_wakeup(&dev->dev))
- pm_wakeup_event(&dev->dev, PCI_WAKEUP_COOLDOWN);
-}
-
/**
* pci_pme_wakeup - Wake up a PCI device if its PME Status bit is set.
* @dev: Device to handle.
@@ -1327,8 +1308,8 @@
static int pci_pme_wakeup(struct pci_dev *dev, void *ign)
{
if (pci_check_pme_status(dev)) {
- pm_request_resume(&dev->dev);
pci_wakeup_event(dev);
+ pm_request_resume(&dev->dev);
}
return 0;
}
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 7d33f66..f69d6e0 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -74,6 +74,12 @@
extern void platform_pci_wakeup_init(struct pci_dev *dev);
extern void pci_allocate_cap_save_buffers(struct pci_dev *dev);
+static inline void pci_wakeup_event(struct pci_dev *dev)
+{
+ /* Wait 100 ms before the system can be put into a sleep state. */
+ pm_wakeup_event(&dev->dev, 100);
+}
+
static inline bool pci_is_bridge(struct pci_dev *pci_dev)
{
return !!(pci_dev->subordinate);
@@ -140,14 +146,6 @@
static inline void pci_msi_init_pci_dev(struct pci_dev *dev) { }
#endif
-#ifdef CONFIG_PCIEAER
-void pci_no_aer(void);
-bool pci_aer_available(void);
-#else
-static inline void pci_no_aer(void) { }
-static inline bool pci_aer_available(void) { return false; }
-#endif
-
static inline int pci_no_d1d2(struct pci_dev *dev)
{
unsigned int parent_dstates = 0;
diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c
index 2b2b650..58ad791 100644
--- a/drivers/pci/pcie/aer/aerdrv.c
+++ b/drivers/pci/pcie/aer/aerdrv.c
@@ -17,6 +17,7 @@
#include <linux/module.h>
#include <linux/pci.h>
+#include <linux/pci-acpi.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/errno.h>
diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h
index 9656e30..80c11d1 100644
--- a/drivers/pci/pcie/aer/aerdrv.h
+++ b/drivers/pci/pcie/aer/aerdrv.h
@@ -132,7 +132,6 @@
#ifdef CONFIG_ACPI_APEI
extern int pcie_aer_get_firmware_first(struct pci_dev *pci_dev);
-extern bool aer_acpi_firmware_first(void);
#else
static inline int pcie_aer_get_firmware_first(struct pci_dev *pci_dev)
{
@@ -140,8 +139,6 @@
return pci_dev->__aer_firmware_first;
return 0;
}
-
-static inline bool aer_acpi_firmware_first(void) { return false; }
#endif
static inline void pcie_aer_force_firmware_first(struct pci_dev *pci_dev,
diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 7122281..3188cd9 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -68,7 +68,7 @@
struct aspm_latency acceptable[8];
};
-static int aspm_disabled, aspm_force;
+static int aspm_disabled, aspm_force, aspm_clear_state;
static DEFINE_MUTEX(aspm_lock);
static LIST_HEAD(link_list);
@@ -139,7 +139,7 @@
{
/* Don't enable Clock PM if the link is not Clock PM capable */
if (!link->clkpm_capable && enable)
- return;
+ enable = 0;
/* Need nothing if the specified equals to current state */
if (link->clkpm_enabled == enable)
return;
@@ -498,6 +498,10 @@
struct pci_dev *child;
int pos;
u32 reg32;
+
+ if (aspm_clear_state)
+ return -EINVAL;
+
/*
* Some functions in a slot might not all be PCIe functions,
* very strange. Disable ASPM for the whole slot
@@ -563,12 +567,15 @@
struct pcie_link_state *link;
int blacklist = !!pcie_aspm_sanity_check(pdev);
- if (aspm_disabled || !pci_is_pcie(pdev) || pdev->link_state)
+ if (!pci_is_pcie(pdev) || pdev->link_state)
return;
if (pdev->pcie_type != PCI_EXP_TYPE_ROOT_PORT &&
pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM)
return;
+ if (aspm_disabled && !aspm_clear_state)
+ return;
+
/* VIA has a strange chipset, root port is under a bridge */
if (pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT &&
pdev->bus->self)
@@ -641,7 +648,7 @@
struct pci_dev *parent = pdev->bus->self;
struct pcie_link_state *link, *root, *parent_link;
- if (aspm_disabled || !pci_is_pcie(pdev) ||
+ if ((aspm_disabled && !aspm_clear_state) || !pci_is_pcie(pdev) ||
!parent || !parent->link_state)
return;
if ((parent->pcie_type != PCI_EXP_TYPE_ROOT_PORT) &&
@@ -899,6 +906,12 @@
__setup("pcie_aspm=", pcie_aspm_disable);
+void pcie_clear_aspm(void)
+{
+ if (!aspm_force)
+ aspm_clear_state = 1;
+}
+
void pcie_no_aspm(void)
{
if (!aspm_force)
diff --git a/drivers/pci/pcie/pme.c b/drivers/pci/pcie/pme.c
index 2f3c904..0057344 100644
--- a/drivers/pci/pcie/pme.c
+++ b/drivers/pci/pcie/pme.c
@@ -26,9 +26,6 @@
#include "../pci.h"
#include "portdrv.h"
-#define PCI_EXP_RTSTA_PME 0x10000 /* PME status */
-#define PCI_EXP_RTSTA_PENDING 0x20000 /* PME pending */
-
/*
* If this switch is set, MSI will not be used for PCIe PME signaling. This
* causes the PCIe port driver to use INTx interrupts only, but it turns out
@@ -74,22 +71,6 @@
}
/**
- * pcie_pme_clear_status - Clear root port PME interrupt status.
- * @dev: PCIe root port or event collector.
- */
-static void pcie_pme_clear_status(struct pci_dev *dev)
-{
- int rtsta_pos;
- u32 rtsta;
-
- rtsta_pos = pci_pcie_cap(dev) + PCI_EXP_RTSTA;
-
- pci_read_config_dword(dev, rtsta_pos, &rtsta);
- rtsta |= PCI_EXP_RTSTA_PME;
- pci_write_config_dword(dev, rtsta_pos, rtsta);
-}
-
-/**
* pcie_pme_walk_bus - Scan a PCI bus for devices asserting PME#.
* @bus: PCI bus to scan.
*
@@ -103,8 +84,8 @@
list_for_each_entry(dev, &bus->devices, bus_list) {
/* Skip PCIe devices in case we started from a root port. */
if (!pci_is_pcie(dev) && pci_check_pme_status(dev)) {
- pm_request_resume(&dev->dev);
pci_wakeup_event(dev);
+ pm_request_resume(&dev->dev);
ret = true;
}
@@ -206,8 +187,8 @@
/* The device is there, but we have to check its PME status. */
found = pci_check_pme_status(dev);
if (found) {
- pm_request_resume(&dev->dev);
pci_wakeup_event(dev);
+ pm_request_resume(&dev->dev);
}
pci_dev_put(dev);
} else if (devfn) {
@@ -253,7 +234,7 @@
* Clear PME status of the port. If there are other
* pending PMEs, the status will be set again.
*/
- pcie_pme_clear_status(port);
+ pcie_clear_root_pme_status(port);
spin_unlock_irq(&data->lock);
pcie_pme_handle_request(port, rtsta & 0xffff);
@@ -378,7 +359,7 @@
port = srv->port;
pcie_pme_interrupt_enable(port, false);
- pcie_pme_clear_status(port);
+ pcie_clear_root_pme_status(port);
ret = request_irq(srv->irq, pcie_pme_irq, IRQF_SHARED, "PCIe PME", srv);
if (ret) {
@@ -402,7 +383,7 @@
spin_lock_irq(&data->lock);
pcie_pme_interrupt_enable(port, false);
- pcie_pme_clear_status(port);
+ pcie_clear_root_pme_status(port);
data->noirq = true;
spin_unlock_irq(&data->lock);
@@ -422,7 +403,7 @@
spin_lock_irq(&data->lock);
data->noirq = false;
- pcie_pme_clear_status(port);
+ pcie_clear_root_pme_status(port);
pcie_pme_interrupt_enable(port, true);
spin_unlock_irq(&data->lock);
diff --git a/drivers/pci/pcie/portdrv.h b/drivers/pci/pcie/portdrv.h
index 7b5aba0..bd00a01 100644
--- a/drivers/pci/pcie/portdrv.h
+++ b/drivers/pci/pcie/portdrv.h
@@ -20,9 +20,6 @@
#define get_descriptor_id(type, service) (((type - 4) << 4) | service)
-extern bool pcie_ports_disabled;
-extern bool pcie_ports_auto;
-
extern struct bus_type pcie_port_bus_type;
extern int pcie_port_device_register(struct pci_dev *dev);
#ifdef CONFIG_PM
@@ -35,6 +32,8 @@
struct pci_dev;
+extern void pcie_clear_root_pme_status(struct pci_dev *dev);
+
#ifdef CONFIG_PCIE_PME
extern bool pcie_pme_msi_disabled;
diff --git a/drivers/pci/pcie/portdrv_acpi.c b/drivers/pci/pcie/portdrv_acpi.c
index 5982b6a..a86b56e 100644
--- a/drivers/pci/pcie/portdrv_acpi.c
+++ b/drivers/pci/pcie/portdrv_acpi.c
@@ -33,7 +33,7 @@
*/
int pcie_port_acpi_setup(struct pci_dev *port, int *srv_mask)
{
- acpi_status status;
+ struct acpi_pci_root *root;
acpi_handle handle;
u32 flags;
@@ -44,26 +44,11 @@
if (!handle)
return -EINVAL;
- flags = OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL
- | OSC_PCI_EXPRESS_NATIVE_HP_CONTROL
- | OSC_PCI_EXPRESS_PME_CONTROL;
-
- if (pci_aer_available()) {
- if (aer_acpi_firmware_first())
- dev_dbg(&port->dev, "PCIe errors handled by BIOS.\n");
- else
- flags |= OSC_PCI_EXPRESS_AER_CONTROL;
- }
-
- status = acpi_pci_osc_control_set(handle, &flags,
- OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL);
- if (ACPI_FAILURE(status)) {
- dev_dbg(&port->dev, "ACPI _OSC request failed (code %d)\n",
- status);
+ root = acpi_pci_find_root(handle);
+ if (!root)
return -ENODEV;
- }
- dev_info(&port->dev, "ACPI _OSC control granted for 0x%02x\n", flags);
+ flags = root->osc_control_set;
*srv_mask = PCIE_PORT_SERVICE_VC;
if (flags & OSC_PCI_EXPRESS_NATIVE_HP_CONTROL)
diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c
index a9c222d..5130d0d 100644
--- a/drivers/pci/pcie/portdrv_core.c
+++ b/drivers/pci/pcie/portdrv_core.c
@@ -241,17 +241,17 @@
int cap_mask;
int err;
+ if (pcie_ports_disabled)
+ return 0;
+
err = pcie_port_platform_notify(dev, &cap_mask);
- if (pcie_ports_auto) {
- if (err) {
- pcie_no_aspm();
- return 0;
- }
- } else {
+ if (!pcie_ports_auto) {
cap_mask = PCIE_PORT_SERVICE_PME | PCIE_PORT_SERVICE_HP
| PCIE_PORT_SERVICE_VC;
if (pci_aer_available())
cap_mask |= PCIE_PORT_SERVICE_AER;
+ } else if (err) {
+ return 0;
}
pos = pci_pcie_cap(dev);
@@ -349,15 +349,18 @@
int status, capabilities, i, nr_service;
int irqs[PCIE_PORT_DEVICE_MAXSERVICES];
- /* Get and check PCI Express port services */
- capabilities = get_port_device_capability(dev);
- if (!capabilities)
- return -ENODEV;
-
/* Enable PCI Express port device */
status = pci_enable_device(dev);
if (status)
return status;
+
+ /* Get and check PCI Express port services */
+ capabilities = get_port_device_capability(dev);
+ if (!capabilities) {
+ pcie_no_aspm();
+ return 0;
+ }
+
pci_set_master(dev);
/*
* Initialize service irqs. Don't use service devices that
diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c
index f9033e1..e0610bda 100644
--- a/drivers/pci/pcie/portdrv_pci.c
+++ b/drivers/pci/pcie/portdrv_pci.c
@@ -57,6 +57,22 @@
/* global data */
+/**
+ * pcie_clear_root_pme_status - Clear root port PME interrupt status.
+ * @dev: PCIe root port or event collector.
+ */
+void pcie_clear_root_pme_status(struct pci_dev *dev)
+{
+ int rtsta_pos;
+ u32 rtsta;
+
+ rtsta_pos = pci_pcie_cap(dev) + PCI_EXP_RTSTA;
+
+ pci_read_config_dword(dev, rtsta_pos, &rtsta);
+ rtsta |= PCI_EXP_RTSTA_PME;
+ pci_write_config_dword(dev, rtsta_pos, rtsta);
+}
+
static int pcie_portdrv_restore_config(struct pci_dev *dev)
{
int retval;
@@ -69,6 +85,20 @@
}
#ifdef CONFIG_PM
+static int pcie_port_resume_noirq(struct device *dev)
+{
+ struct pci_dev *pdev = to_pci_dev(dev);
+
+ /*
+ * Some BIOSes forget to clear Root PME Status bits after system wakeup
+ * which breaks ACPI-based runtime wakeup on PCI Express, so clear those
+ * bits now just in case (shouldn't hurt).
+ */
+ if(pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT)
+ pcie_clear_root_pme_status(pdev);
+ return 0;
+}
+
static const struct dev_pm_ops pcie_portdrv_pm_ops = {
.suspend = pcie_port_device_suspend,
.resume = pcie_port_device_resume,
@@ -76,6 +106,7 @@
.thaw = pcie_port_device_resume,
.poweroff = pcie_port_device_suspend,
.restore = pcie_port_device_resume,
+ .resume_noirq = pcie_port_resume_noirq,
};
#define PCIE_PORTDRV_PM_OPS (&pcie_portdrv_pm_ops)
@@ -327,10 +358,8 @@
{
int retval;
- if (pcie_ports_disabled) {
- pcie_no_aspm();
- return -EACCES;
- }
+ if (pcie_ports_disabled)
+ return pci_register_driver(&pcie_portdriver);
dmi_check_system(pcie_portdrv_dmi_table);
diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
index 19e92b2..95e3b09 100644
--- a/drivers/platform/x86/fujitsu-laptop.c
+++ b/drivers/platform/x86/fujitsu-laptop.c
@@ -689,7 +689,7 @@
if (error)
goto err_free_input_dev;
- result = acpi_bus_get_power(fujitsu->acpi_handle, &state);
+ result = acpi_bus_update_power(fujitsu->acpi_handle, &state);
if (result) {
printk(KERN_ERR "Error reading power state\n");
goto err_unregister_input_dev;
@@ -857,7 +857,7 @@
if (error)
goto err_free_input_dev;
- result = acpi_bus_get_power(fujitsu_hotkey->acpi_handle, &state);
+ result = acpi_bus_update_power(fujitsu_hotkey->acpi_handle, &state);
if (result) {
printk(KERN_ERR "Error reading power state\n");
goto err_unregister_input_dev;
diff --git a/drivers/pnp/Makefile b/drivers/pnp/Makefile
index 8de3775..bfba893 100644
--- a/drivers/pnp/Makefile
+++ b/drivers/pnp/Makefile
@@ -2,11 +2,13 @@
# Makefile for the Linux Plug-and-Play Support.
#
-obj-y := core.o card.o driver.o resource.o manager.o support.o interface.o quirks.o
+obj-y := pnp.o
+
+pnp-y := core.o card.o driver.o resource.o manager.o support.o interface.o quirks.o
obj-$(CONFIG_PNPACPI) += pnpacpi/
obj-$(CONFIG_PNPBIOS) += pnpbios/
obj-$(CONFIG_ISAPNP) += isapnp/
# pnp_system_init goes after pnpacpi/pnpbios init
-obj-y += system.o
+pnp-y += system.o
diff --git a/drivers/pnp/core.c b/drivers/pnp/core.c
index 0f34d96..cb6ce42 100644
--- a/drivers/pnp/core.c
+++ b/drivers/pnp/core.c
@@ -220,10 +220,5 @@
int pnp_debug;
#if defined(CONFIG_PNP_DEBUG_MESSAGES)
-static int __init pnp_debug_setup(char *__unused)
-{
- pnp_debug = 1;
- return 1;
-}
-__setup("pnp.debug", pnp_debug_setup);
+module_param_named(debug, pnp_debug, int, 0644);
#endif
diff --git a/drivers/pnp/driver.c b/drivers/pnp/driver.c
index d1dbb9d..00e94032 100644
--- a/drivers/pnp/driver.c
+++ b/drivers/pnp/driver.c
@@ -189,8 +189,11 @@
if (!pnp_drv)
return 0;
- if (pnp_dev->protocol->resume)
- pnp_dev->protocol->resume(pnp_dev);
+ if (pnp_dev->protocol->resume) {
+ error = pnp_dev->protocol->resume(pnp_dev);
+ if (error)
+ return error;
+ }
if (pnp_can_write(pnp_dev)) {
error = pnp_start_dev(pnp_dev);
diff --git a/drivers/pnp/isapnp/Makefile b/drivers/pnp/isapnp/Makefile
index cac18bb..6e607aa 100644
--- a/drivers/pnp/isapnp/Makefile
+++ b/drivers/pnp/isapnp/Makefile
@@ -1,7 +1,7 @@
#
# Makefile for the kernel ISAPNP driver.
#
+obj-y += pnp.o
+pnp-y := core.o compat.o
-isapnp-proc-$(CONFIG_PROC_FS) = proc.o
-
-obj-y := core.o compat.o $(isapnp-proc-y)
+pnp-$(CONFIG_PROC_FS) += proc.o
diff --git a/drivers/pnp/pnpacpi/Makefile b/drivers/pnp/pnpacpi/Makefile
index 905326f..40c93da 100644
--- a/drivers/pnp/pnpacpi/Makefile
+++ b/drivers/pnp/pnpacpi/Makefile
@@ -1,5 +1,6 @@
#
# Makefile for the kernel PNPACPI driver.
#
+obj-y += pnp.o
-obj-y := core.o rsparser.o
+pnp-y := core.o rsparser.o
diff --git a/drivers/pnp/pnpacpi/core.c b/drivers/pnp/pnpacpi/core.c
index 57313f4..ca84d50 100644
--- a/drivers/pnp/pnpacpi/core.c
+++ b/drivers/pnp/pnpacpi/core.c
@@ -81,12 +81,19 @@
static int pnpacpi_set_resources(struct pnp_dev *dev)
{
- struct acpi_device *acpi_dev = dev->data;
- acpi_handle handle = acpi_dev->handle;
+ struct acpi_device *acpi_dev;
+ acpi_handle handle;
struct acpi_buffer buffer;
int ret;
pnp_dbg(&dev->dev, "set resources\n");
+
+ handle = DEVICE_ACPI_HANDLE(&dev->dev);
+ if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &acpi_dev))) {
+ dev_dbg(&dev->dev, "ACPI device not found in %s!\n", __func__);
+ return -ENODEV;
+ }
+
ret = pnpacpi_build_resource_template(dev, &buffer);
if (ret)
return ret;
@@ -105,12 +112,18 @@
static int pnpacpi_disable_resources(struct pnp_dev *dev)
{
- struct acpi_device *acpi_dev = dev->data;
- acpi_handle handle = acpi_dev->handle;
+ struct acpi_device *acpi_dev;
+ acpi_handle handle;
int ret;
dev_dbg(&dev->dev, "disable resources\n");
+ handle = DEVICE_ACPI_HANDLE(&dev->dev);
+ if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &acpi_dev))) {
+ dev_dbg(&dev->dev, "ACPI device not found in %s!\n", __func__);
+ return 0;
+ }
+
/* acpi_unregister_gsi(pnp_irq(dev, 0)); */
ret = 0;
if (acpi_bus_power_manageable(handle))
@@ -124,46 +137,74 @@
#ifdef CONFIG_ACPI_SLEEP
static bool pnpacpi_can_wakeup(struct pnp_dev *dev)
{
- struct acpi_device *acpi_dev = dev->data;
- acpi_handle handle = acpi_dev->handle;
+ struct acpi_device *acpi_dev;
+ acpi_handle handle;
+
+ handle = DEVICE_ACPI_HANDLE(&dev->dev);
+ if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &acpi_dev))) {
+ dev_dbg(&dev->dev, "ACPI device not found in %s!\n", __func__);
+ return false;
+ }
return acpi_bus_can_wakeup(handle);
}
static int pnpacpi_suspend(struct pnp_dev *dev, pm_message_t state)
{
- struct acpi_device *acpi_dev = dev->data;
- acpi_handle handle = acpi_dev->handle;
- int power_state;
+ struct acpi_device *acpi_dev;
+ acpi_handle handle;
+ int error = 0;
+
+ handle = DEVICE_ACPI_HANDLE(&dev->dev);
+ if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &acpi_dev))) {
+ dev_dbg(&dev->dev, "ACPI device not found in %s!\n", __func__);
+ return 0;
+ }
if (device_can_wakeup(&dev->dev)) {
- int rc = acpi_pm_device_sleep_wake(&dev->dev,
+ error = acpi_pm_device_sleep_wake(&dev->dev,
device_may_wakeup(&dev->dev));
-
- if (rc)
- return rc;
+ if (error)
+ return error;
}
- power_state = acpi_pm_device_sleep_state(&dev->dev, NULL);
- if (power_state < 0)
- power_state = (state.event == PM_EVENT_ON) ?
- ACPI_STATE_D0 : ACPI_STATE_D3;
- /* acpi_bus_set_power() often fails (keyboard port can't be
- * powered-down?), and in any case, our return value is ignored
- * by pnp_bus_suspend(). Hence we don't revert the wakeup
- * setting if the set_power fails.
- */
- return acpi_bus_set_power(handle, power_state);
+ if (acpi_bus_power_manageable(handle)) {
+ int power_state = acpi_pm_device_sleep_state(&dev->dev, NULL);
+
+ if (power_state < 0)
+ power_state = (state.event == PM_EVENT_ON) ?
+ ACPI_STATE_D0 : ACPI_STATE_D3;
+
+ /*
+ * acpi_bus_set_power() often fails (keyboard port can't be
+ * powered-down?), and in any case, our return value is ignored
+ * by pnp_bus_suspend(). Hence we don't revert the wakeup
+ * setting if the set_power fails.
+ */
+ error = acpi_bus_set_power(handle, power_state);
+ }
+
+ return error;
}
static int pnpacpi_resume(struct pnp_dev *dev)
{
- struct acpi_device *acpi_dev = dev->data;
- acpi_handle handle = acpi_dev->handle;
+ struct acpi_device *acpi_dev;
+ acpi_handle handle = DEVICE_ACPI_HANDLE(&dev->dev);
+ int error = 0;
+
+ if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &acpi_dev))) {
+ dev_dbg(&dev->dev, "ACPI device not found in %s!\n", __func__);
+ return -ENODEV;
+ }
if (device_may_wakeup(&dev->dev))
acpi_pm_device_sleep_wake(&dev->dev, false);
- return acpi_bus_set_power(handle, ACPI_STATE_D0);
+
+ if (acpi_bus_power_manageable(handle))
+ error = acpi_bus_set_power(handle, ACPI_STATE_D0);
+
+ return error;
}
#endif
diff --git a/drivers/pnp/pnpbios/Makefile b/drivers/pnp/pnpbios/Makefile
index 3cd3ed7..240b0ff 100644
--- a/drivers/pnp/pnpbios/Makefile
+++ b/drivers/pnp/pnpbios/Makefile
@@ -1,7 +1,8 @@
#
# Makefile for the kernel PNPBIOS driver.
#
+obj-y := pnp.o
-pnpbios-proc-$(CONFIG_PNPBIOS_PROC_FS) = proc.o
+pnp-y := core.o bioscalls.o rsparser.o
-obj-y := core.o bioscalls.o rsparser.o $(pnpbios-proc-y)
+pnp-$(CONFIG_PNPBIOS_PROC_FS) += proc.o
diff --git a/drivers/power/Kconfig b/drivers/power/Kconfig
index 60d83d9..61bf5d7 100644
--- a/drivers/power/Kconfig
+++ b/drivers/power/Kconfig
@@ -136,6 +136,16 @@
in handheld and portable equipment. The MAX17040 is configured
to operate with a single lithium cell
+config BATTERY_MAX17042
+ tristate "Maxim MAX17042/8997/8966 Fuel Gauge"
+ depends on I2C
+ help
+ MAX17042 is fuel-gauge systems for lithium-ion (Li+) batteries
+ in handheld and portable equipment. The MAX17042 is configured
+ to operate with a single lithium cell. MAX8997 and MAX8966 are
+ multi-function devices that include fuel gauages that are compatible
+ with MAX17042.
+
config BATTERY_Z2
tristate "Z2 battery driver"
depends on I2C && MACH_ZIPIT2
@@ -185,4 +195,14 @@
help
Say Y here to enable support for TWL4030 Battery Charge Interface.
+config CHARGER_GPIO
+ tristate "GPIO charger"
+ depends on GPIOLIB
+ help
+ Say Y to include support for chargers which report their online status
+ through a GPIO pin.
+
+ This driver can be build as a module. If so, the module will be
+ called gpio-charger.
+
endif # POWER_SUPPLY
diff --git a/drivers/power/Makefile b/drivers/power/Makefile
index c75772e..8385bfa 100644
--- a/drivers/power/Makefile
+++ b/drivers/power/Makefile
@@ -25,6 +25,7 @@
obj-$(CONFIG_BATTERY_BQ27x00) += bq27x00_battery.o
obj-$(CONFIG_BATTERY_DA9030) += da9030_battery.o
obj-$(CONFIG_BATTERY_MAX17040) += max17040_battery.o
+obj-$(CONFIG_BATTERY_MAX17042) += max17042_battery.o
obj-$(CONFIG_BATTERY_Z2) += z2_battery.o
obj-$(CONFIG_BATTERY_S3C_ADC) += s3c_adc_battery.o
obj-$(CONFIG_CHARGER_PCF50633) += pcf50633-charger.o
@@ -32,3 +33,4 @@
obj-$(CONFIG_BATTERY_INTEL_MID) += intel_mid_battery.o
obj-$(CONFIG_CHARGER_ISP1704) += isp1704_charger.o
obj-$(CONFIG_CHARGER_TWL4030) += twl4030_charger.o
+obj-$(CONFIG_CHARGER_GPIO) += gpio-charger.o
diff --git a/drivers/power/collie_battery.c b/drivers/power/collie_battery.c
index 039f41a..548d263 100644
--- a/drivers/power/collie_battery.c
+++ b/drivers/power/collie_battery.c
@@ -295,7 +295,7 @@
static int collie_bat_suspend(struct ucb1x00_dev *dev, pm_message_t state)
{
/* flush all pending status updates */
- flush_scheduled_work();
+ flush_work_sync(&bat_work);
return 0;
}
@@ -362,7 +362,7 @@
err_psy_reg_main:
/* see comment in collie_bat_remove */
- flush_scheduled_work();
+ cancel_work_sync(&bat_work);
i--;
err_gpio:
@@ -382,12 +382,11 @@
power_supply_unregister(&collie_bat_main.psy);
/*
- * now flush all pending work.
- * we won't get any more schedules, since all
- * sources (isr and external_power_changed)
- * are unregistered now.
+ * Now cancel the bat_work. We won't get any more schedules,
+ * since all sources (isr and external_power_changed) are
+ * unregistered now.
*/
- flush_scheduled_work();
+ cancel_work_sync(&bat_work);
for (i = ARRAY_SIZE(gpios) - 1; i >= 0; i--)
gpio_free(gpios[i].gpio);
diff --git a/drivers/power/ds2760_battery.c b/drivers/power/ds2760_battery.c
index e7f8978..e534290 100644
--- a/drivers/power/ds2760_battery.c
+++ b/drivers/power/ds2760_battery.c
@@ -212,7 +212,7 @@
if (di->rem_capacity > 100)
di->rem_capacity = 100;
- if (di->current_uA >= 100L)
+ if (di->current_uA < -100L)
di->life_sec = -((di->accum_current_uAh - di->empty_uAh) * 36L)
/ (di->current_uA / 100L);
else
diff --git a/drivers/power/gpio-charger.c b/drivers/power/gpio-charger.c
new file mode 100644
index 0000000..25b88ac
--- /dev/null
+++ b/drivers/power/gpio-charger.c
@@ -0,0 +1,188 @@
+/*
+ * Copyright (C) 2010, Lars-Peter Clausen <lars@metafoo.de>
+ * Driver for chargers which report their online status through a GPIO pin
+ *
+ * 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.
+ *
+ * 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.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+#include <linux/device.h>
+#include <linux/gpio.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/power_supply.h>
+#include <linux/slab.h>
+
+#include <linux/power/gpio-charger.h>
+
+struct gpio_charger {
+ const struct gpio_charger_platform_data *pdata;
+ unsigned int irq;
+
+ struct power_supply charger;
+};
+
+static irqreturn_t gpio_charger_irq(int irq, void *devid)
+{
+ struct power_supply *charger = devid;
+
+ power_supply_changed(charger);
+
+ return IRQ_HANDLED;
+}
+
+static inline struct gpio_charger *psy_to_gpio_charger(struct power_supply *psy)
+{
+ return container_of(psy, struct gpio_charger, charger);
+}
+
+static int gpio_charger_get_property(struct power_supply *psy,
+ enum power_supply_property psp, union power_supply_propval *val)
+{
+ struct gpio_charger *gpio_charger = psy_to_gpio_charger(psy);
+ const struct gpio_charger_platform_data *pdata = gpio_charger->pdata;
+
+ switch (psp) {
+ case POWER_SUPPLY_PROP_ONLINE:
+ val->intval = gpio_get_value(pdata->gpio);
+ val->intval ^= pdata->gpio_active_low;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static enum power_supply_property gpio_charger_properties[] = {
+ POWER_SUPPLY_PROP_ONLINE,
+};
+
+static int __devinit gpio_charger_probe(struct platform_device *pdev)
+{
+ const struct gpio_charger_platform_data *pdata = pdev->dev.platform_data;
+ struct gpio_charger *gpio_charger;
+ struct power_supply *charger;
+ int ret;
+ int irq;
+
+ if (!pdata) {
+ dev_err(&pdev->dev, "No platform data\n");
+ return -EINVAL;
+ }
+
+ if (!gpio_is_valid(pdata->gpio)) {
+ dev_err(&pdev->dev, "Invalid gpio pin\n");
+ return -EINVAL;
+ }
+
+ gpio_charger = kzalloc(sizeof(*gpio_charger), GFP_KERNEL);
+ if (!gpio_charger) {
+ dev_err(&pdev->dev, "Failed to alloc driver structure\n");
+ return -ENOMEM;
+ }
+
+ charger = &gpio_charger->charger;
+
+ charger->name = pdata->name ? pdata->name : "gpio-charger";
+ charger->type = pdata->type;
+ charger->properties = gpio_charger_properties;
+ charger->num_properties = ARRAY_SIZE(gpio_charger_properties);
+ charger->get_property = gpio_charger_get_property;
+ charger->supplied_to = pdata->supplied_to;
+ charger->num_supplicants = pdata->num_supplicants;
+
+ ret = gpio_request(pdata->gpio, dev_name(&pdev->dev));
+ if (ret) {
+ dev_err(&pdev->dev, "Failed to request gpio pin: %d\n", ret);
+ goto err_free;
+ }
+ ret = gpio_direction_input(pdata->gpio);
+ if (ret) {
+ dev_err(&pdev->dev, "Failed to set gpio to input: %d\n", ret);
+ goto err_gpio_free;
+ }
+
+ gpio_charger->pdata = pdata;
+
+ ret = power_supply_register(&pdev->dev, charger);
+ if (ret < 0) {
+ dev_err(&pdev->dev, "Failed to register power supply: %d\n",
+ ret);
+ goto err_gpio_free;
+ }
+
+ irq = gpio_to_irq(pdata->gpio);
+ if (irq > 0) {
+ ret = request_any_context_irq(irq, gpio_charger_irq,
+ IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING,
+ dev_name(&pdev->dev), charger);
+ if (ret)
+ dev_warn(&pdev->dev, "Failed to request irq: %d\n", ret);
+ else
+ gpio_charger->irq = irq;
+ }
+
+ platform_set_drvdata(pdev, gpio_charger);
+
+ return 0;
+
+err_gpio_free:
+ gpio_free(pdata->gpio);
+err_free:
+ kfree(gpio_charger);
+ return ret;
+}
+
+static int __devexit gpio_charger_remove(struct platform_device *pdev)
+{
+ struct gpio_charger *gpio_charger = platform_get_drvdata(pdev);
+
+ if (gpio_charger->irq)
+ free_irq(gpio_charger->irq, &gpio_charger->charger);
+
+ power_supply_unregister(&gpio_charger->charger);
+
+ gpio_free(gpio_charger->pdata->gpio);
+
+ platform_set_drvdata(pdev, NULL);
+ kfree(gpio_charger);
+
+ return 0;
+}
+
+static struct platform_driver gpio_charger_driver = {
+ .probe = gpio_charger_probe,
+ .remove = __devexit_p(gpio_charger_remove),
+ .driver = {
+ .name = "gpio-charger",
+ .owner = THIS_MODULE,
+ },
+};
+
+static int __init gpio_charger_init(void)
+{
+ return platform_driver_register(&gpio_charger_driver);
+}
+module_init(gpio_charger_init);
+
+static void __exit gpio_charger_exit(void)
+{
+ platform_driver_unregister(&gpio_charger_driver);
+}
+module_exit(gpio_charger_exit);
+
+MODULE_AUTHOR("Lars-Peter Clausen <lars@metafoo.de>");
+MODULE_DESCRIPTION("Driver for chargers which report their online status through a GPIO");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:gpio-charger");
diff --git a/drivers/power/intel_mid_battery.c b/drivers/power/intel_mid_battery.c
index 36cf402..bce3a01 100644
--- a/drivers/power/intel_mid_battery.c
+++ b/drivers/power/intel_mid_battery.c
@@ -765,7 +765,7 @@
power_supply_unregister(&pbi->usb);
power_supply_unregister(&pbi->batt);
- flush_scheduled_work();
+ cancel_work_sync(&pbi->handler);
kfree(pbi);
return 0;
}
diff --git a/drivers/power/isp1704_charger.c b/drivers/power/isp1704_charger.c
index 7251218..2ad9b14 100644
--- a/drivers/power/isp1704_charger.c
+++ b/drivers/power/isp1704_charger.c
@@ -59,11 +59,61 @@
struct notifier_block nb;
struct work_struct work;
- char model[7];
+ /* properties */
+ char model[8];
unsigned present:1;
+ unsigned online:1;
+ unsigned current_max;
+
+ /* temp storage variables */
+ unsigned long event;
+ unsigned max_power;
};
/*
+ * Determine is the charging port DCP (dedicated charger) or CDP (Host/HUB
+ * chargers).
+ *
+ * REVISIT: The method is defined in Battery Charging Specification and is
+ * applicable to any ULPI transceiver. Nothing isp170x specific here.
+ */
+static inline int isp1704_charger_type(struct isp1704_charger *isp)
+{
+ u8 reg;
+ u8 func_ctrl;
+ u8 otg_ctrl;
+ int type = POWER_SUPPLY_TYPE_USB_DCP;
+
+ func_ctrl = otg_io_read(isp->otg, ULPI_FUNC_CTRL);
+ otg_ctrl = otg_io_read(isp->otg, ULPI_OTG_CTRL);
+
+ /* disable pulldowns */
+ reg = ULPI_OTG_CTRL_DM_PULLDOWN | ULPI_OTG_CTRL_DP_PULLDOWN;
+ otg_io_write(isp->otg, ULPI_CLR(ULPI_OTG_CTRL), reg);
+
+ /* full speed */
+ otg_io_write(isp->otg, ULPI_CLR(ULPI_FUNC_CTRL),
+ ULPI_FUNC_CTRL_XCVRSEL_MASK);
+ otg_io_write(isp->otg, ULPI_SET(ULPI_FUNC_CTRL),
+ ULPI_FUNC_CTRL_FULL_SPEED);
+
+ /* Enable strong pull-up on DP (1.5K) and reset */
+ reg = ULPI_FUNC_CTRL_TERMSELECT | ULPI_FUNC_CTRL_RESET;
+ otg_io_write(isp->otg, ULPI_SET(ULPI_FUNC_CTRL), reg);
+ usleep_range(1000, 2000);
+
+ reg = otg_io_read(isp->otg, ULPI_DEBUG);
+ if ((reg & 3) != 3)
+ type = POWER_SUPPLY_TYPE_USB_CDP;
+
+ /* recover original state */
+ otg_io_write(isp->otg, ULPI_FUNC_CTRL, func_ctrl);
+ otg_io_write(isp->otg, ULPI_OTG_CTRL, otg_ctrl);
+
+ return type;
+}
+
+/*
* ISP1704 detects PS/2 adapters as charger. To make sure the detected charger
* is actually a dedicated charger, the following steps need to be taken.
*/
@@ -127,16 +177,19 @@
static inline int isp1704_charger_detect(struct isp1704_charger *isp)
{
unsigned long timeout;
- u8 r;
+ u8 pwr_ctrl;
int ret = 0;
+ pwr_ctrl = otg_io_read(isp->otg, ISP1704_PWR_CTRL);
+
/* set SW control bit in PWR_CTRL register */
otg_io_write(isp->otg, ISP1704_PWR_CTRL,
ISP1704_PWR_CTRL_SWCTRL);
/* enable manual charger detection */
- r = (ISP1704_PWR_CTRL_SWCTRL | ISP1704_PWR_CTRL_DPVSRC_EN);
- otg_io_write(isp->otg, ULPI_SET(ISP1704_PWR_CTRL), r);
+ otg_io_write(isp->otg, ULPI_SET(ISP1704_PWR_CTRL),
+ ISP1704_PWR_CTRL_SWCTRL
+ | ISP1704_PWR_CTRL_DPVSRC_EN);
usleep_range(1000, 2000);
timeout = jiffies + msecs_to_jiffies(300);
@@ -147,7 +200,10 @@
ret = isp1704_charger_verify(isp);
break;
}
- } while (!time_after(jiffies, timeout));
+ } while (!time_after(jiffies, timeout) && isp->online);
+
+ /* recover original state */
+ otg_io_write(isp->otg, ISP1704_PWR_CTRL, pwr_ctrl);
return ret;
}
@@ -155,52 +211,92 @@
static void isp1704_charger_work(struct work_struct *data)
{
int detect;
+ unsigned long event;
+ unsigned power;
struct isp1704_charger *isp =
container_of(data, struct isp1704_charger, work);
+ static DEFINE_MUTEX(lock);
- /*
- * FIXME Only supporting dedicated chargers even though isp1704 can
- * detect HUB and HOST chargers. If the device has already been
- * enumerated, the detection will break the connection.
- */
- if (isp->otg->state != OTG_STATE_B_IDLE)
- return;
+ event = isp->event;
+ power = isp->max_power;
- /* disable data pullups */
- if (isp->otg->gadget)
- usb_gadget_disconnect(isp->otg->gadget);
+ mutex_lock(&lock);
- /* detect charger */
- detect = isp1704_charger_detect(isp);
- if (detect) {
- isp->present = detect;
- power_supply_changed(&isp->psy);
+ switch (event) {
+ case USB_EVENT_VBUS:
+ isp->online = true;
+
+ /* detect charger */
+ detect = isp1704_charger_detect(isp);
+
+ if (detect) {
+ isp->present = detect;
+ isp->psy.type = isp1704_charger_type(isp);
+ }
+
+ switch (isp->psy.type) {
+ case POWER_SUPPLY_TYPE_USB_DCP:
+ isp->current_max = 1800;
+ break;
+ case POWER_SUPPLY_TYPE_USB_CDP:
+ /*
+ * Only 500mA here or high speed chirp
+ * handshaking may break
+ */
+ isp->current_max = 500;
+ /* FALLTHROUGH */
+ case POWER_SUPPLY_TYPE_USB:
+ default:
+ /* enable data pullups */
+ if (isp->otg->gadget)
+ usb_gadget_connect(isp->otg->gadget);
+ }
+ break;
+ case USB_EVENT_NONE:
+ isp->online = false;
+ isp->current_max = 0;
+ isp->present = 0;
+ isp->current_max = 0;
+ isp->psy.type = POWER_SUPPLY_TYPE_USB;
+
+ /*
+ * Disable data pullups. We need to prevent the controller from
+ * enumerating.
+ *
+ * FIXME: This is here to allow charger detection with Host/HUB
+ * chargers. The pullups may be enabled elsewhere, so this can
+ * not be the final solution.
+ */
+ if (isp->otg->gadget)
+ usb_gadget_disconnect(isp->otg->gadget);
+ break;
+ case USB_EVENT_ENUMERATED:
+ if (isp->present)
+ isp->current_max = 1800;
+ else
+ isp->current_max = power;
+ break;
+ default:
+ goto out;
}
- /* enable data pullups */
- if (isp->otg->gadget)
- usb_gadget_connect(isp->otg->gadget);
+ power_supply_changed(&isp->psy);
+out:
+ mutex_unlock(&lock);
}
static int isp1704_notifier_call(struct notifier_block *nb,
- unsigned long event, void *unused)
+ unsigned long event, void *power)
{
struct isp1704_charger *isp =
container_of(nb, struct isp1704_charger, nb);
- switch (event) {
- case USB_EVENT_VBUS:
- schedule_work(&isp->work);
- break;
- case USB_EVENT_NONE:
- if (isp->present) {
- isp->present = 0;
- power_supply_changed(&isp->psy);
- }
- break;
- default:
- return NOTIFY_DONE;
- }
+ isp->event = event;
+
+ if (power)
+ isp->max_power = *((unsigned *)power);
+
+ schedule_work(&isp->work);
return NOTIFY_OK;
}
@@ -216,6 +312,12 @@
case POWER_SUPPLY_PROP_PRESENT:
val->intval = isp->present;
break;
+ case POWER_SUPPLY_PROP_ONLINE:
+ val->intval = isp->online;
+ break;
+ case POWER_SUPPLY_PROP_CURRENT_MAX:
+ val->intval = isp->current_max;
+ break;
case POWER_SUPPLY_PROP_MODEL_NAME:
val->strval = isp->model;
break;
@@ -230,6 +332,8 @@
static enum power_supply_property power_props[] = {
POWER_SUPPLY_PROP_PRESENT,
+ POWER_SUPPLY_PROP_ONLINE,
+ POWER_SUPPLY_PROP_CURRENT_MAX,
POWER_SUPPLY_PROP_MODEL_NAME,
POWER_SUPPLY_PROP_MANUFACTURER,
};
@@ -287,13 +391,13 @@
if (!isp->otg)
goto fail0;
+ isp->dev = &pdev->dev;
+ platform_set_drvdata(pdev, isp);
+
ret = isp1704_test_ulpi(isp);
if (ret < 0)
goto fail1;
- isp->dev = &pdev->dev;
- platform_set_drvdata(pdev, isp);
-
isp->psy.name = "isp1704";
isp->psy.type = POWER_SUPPLY_TYPE_USB;
isp->psy.properties = power_props;
@@ -318,6 +422,23 @@
dev_info(isp->dev, "registered with product id %s\n", isp->model);
+ /*
+ * Taking over the D+ pullup.
+ *
+ * FIXME: The device will be disconnected if it was already
+ * enumerated. The charger driver should be always loaded before any
+ * gadget is loaded.
+ */
+ if (isp->otg->gadget)
+ usb_gadget_disconnect(isp->otg->gadget);
+
+ /* Detect charger if VBUS is valid (the cable was already plugged). */
+ ret = otg_io_read(isp->otg, ULPI_USB_INT_STS);
+ if ((ret & ULPI_INT_VBUS_VALID) && !isp->otg->default_a) {
+ isp->event = USB_EVENT_VBUS;
+ schedule_work(&isp->work);
+ }
+
return 0;
fail2:
power_supply_unregister(&isp->psy);
diff --git a/drivers/power/jz4740-battery.c b/drivers/power/jz4740-battery.c
index a8108a7..02414db 100644
--- a/drivers/power/jz4740-battery.c
+++ b/drivers/power/jz4740-battery.c
@@ -19,6 +19,7 @@
#include <linux/module.h>
#include <linux/platform_device.h>
#include <linux/slab.h>
+#include <linux/io.h>
#include <linux/delay.h>
#include <linux/gpio.h>
@@ -47,6 +48,8 @@
struct power_supply battery;
struct delayed_work work;
+
+ struct mutex lock;
};
static inline struct jz_battery *psy_to_jz_battery(struct power_supply *psy)
@@ -68,6 +71,8 @@
unsigned long val;
long voltage;
+ mutex_lock(&battery->lock);
+
INIT_COMPLETION(battery->read_completion);
enable_irq(battery->irq);
@@ -91,6 +96,8 @@
battery->cell->disable(battery->pdev);
disable_irq(battery->irq);
+ mutex_unlock(&battery->lock);
+
return voltage;
}
@@ -240,6 +247,11 @@
struct jz_battery *jz_battery;
struct power_supply *battery;
+ if (!pdata) {
+ dev_err(&pdev->dev, "No platform_data supplied\n");
+ return -ENXIO;
+ }
+
jz_battery = kzalloc(sizeof(*jz_battery), GFP_KERNEL);
if (!jz_battery) {
dev_err(&pdev->dev, "Failed to allocate driver structure\n");
@@ -291,6 +303,7 @@
jz_battery->pdev = pdev;
init_completion(&jz_battery->read_completion);
+ mutex_init(&jz_battery->lock);
INIT_DELAYED_WORK(&jz_battery->work, jz_battery_work);
diff --git a/drivers/power/max17042_battery.c b/drivers/power/max17042_battery.c
new file mode 100644
index 0000000..c5c8805
--- /dev/null
+++ b/drivers/power/max17042_battery.c
@@ -0,0 +1,239 @@
+/*
+ * Fuel gauge driver for Maxim 17042 / 8966 / 8997
+ * Note that Maxim 8966 and 8997 are mfd and this is its subdevice.
+ *
+ * Copyright (C) 2011 Samsung Electronics
+ * MyungJoo Ham <myungjoo.ham@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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * This driver is based on max17040_battery.c
+ */
+
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/i2c.h>
+#include <linux/mod_devicetable.h>
+#include <linux/power_supply.h>
+#include <linux/power/max17042_battery.h>
+
+enum max17042_register {
+ MAX17042_STATUS = 0x00,
+ MAX17042_VALRT_Th = 0x01,
+ MAX17042_TALRT_Th = 0x02,
+ MAX17042_SALRT_Th = 0x03,
+ MAX17042_AtRate = 0x04,
+ MAX17042_RepCap = 0x05,
+ MAX17042_RepSOC = 0x06,
+ MAX17042_Age = 0x07,
+ MAX17042_TEMP = 0x08,
+ MAX17042_VCELL = 0x09,
+ MAX17042_Current = 0x0A,
+ MAX17042_AvgCurrent = 0x0B,
+ MAX17042_Qresidual = 0x0C,
+ MAX17042_SOC = 0x0D,
+ MAX17042_AvSOC = 0x0E,
+ MAX17042_RemCap = 0x0F,
+ MAX17402_FullCAP = 0x10,
+ MAX17042_TTE = 0x11,
+ MAX17042_V_empty = 0x12,
+
+ MAX17042_RSLOW = 0x14,
+
+ MAX17042_AvgTA = 0x16,
+ MAX17042_Cycles = 0x17,
+ MAX17042_DesignCap = 0x18,
+ MAX17042_AvgVCELL = 0x19,
+ MAX17042_MinMaxTemp = 0x1A,
+ MAX17042_MinMaxVolt = 0x1B,
+ MAX17042_MinMaxCurr = 0x1C,
+ MAX17042_CONFIG = 0x1D,
+ MAX17042_ICHGTerm = 0x1E,
+ MAX17042_AvCap = 0x1F,
+ MAX17042_ManName = 0x20,
+ MAX17042_DevName = 0x21,
+ MAX17042_DevChem = 0x22,
+
+ MAX17042_TempNom = 0x24,
+ MAX17042_TempCold = 0x25,
+ MAX17042_TempHot = 0x26,
+ MAX17042_AIN = 0x27,
+ MAX17042_LearnCFG = 0x28,
+ MAX17042_SHFTCFG = 0x29,
+ MAX17042_RelaxCFG = 0x2A,
+ MAX17042_MiscCFG = 0x2B,
+ MAX17042_TGAIN = 0x2C,
+ MAx17042_TOFF = 0x2D,
+ MAX17042_CGAIN = 0x2E,
+ MAX17042_COFF = 0x2F,
+
+ MAX17042_Q_empty = 0x33,
+ MAX17042_T_empty = 0x34,
+
+ MAX17042_RCOMP0 = 0x38,
+ MAX17042_TempCo = 0x39,
+ MAX17042_Rx = 0x3A,
+ MAX17042_T_empty0 = 0x3B,
+ MAX17042_TaskPeriod = 0x3C,
+ MAX17042_FSTAT = 0x3D,
+
+ MAX17042_SHDNTIMER = 0x3F,
+
+ MAX17042_VFRemCap = 0x4A,
+
+ MAX17042_QH = 0x4D,
+ MAX17042_QL = 0x4E,
+};
+
+struct max17042_chip {
+ struct i2c_client *client;
+ struct power_supply battery;
+ struct max17042_platform_data *pdata;
+};
+
+static int max17042_write_reg(struct i2c_client *client, u8 reg, u16 value)
+{
+ int ret = i2c_smbus_write_word_data(client, reg, value);
+
+ if (ret < 0)
+ dev_err(&client->dev, "%s: err %d\n", __func__, ret);
+
+ return ret;
+}
+
+static int max17042_read_reg(struct i2c_client *client, u8 reg)
+{
+ int ret = i2c_smbus_read_word_data(client, reg);
+
+ if (ret < 0)
+ dev_err(&client->dev, "%s: err %d\n", __func__, ret);
+
+ return ret;
+}
+
+static enum power_supply_property max17042_battery_props[] = {
+ POWER_SUPPLY_PROP_VOLTAGE_NOW,
+ POWER_SUPPLY_PROP_VOLTAGE_AVG,
+ POWER_SUPPLY_PROP_CAPACITY,
+};
+
+static int max17042_get_property(struct power_supply *psy,
+ enum power_supply_property psp,
+ union power_supply_propval *val)
+{
+ struct max17042_chip *chip = container_of(psy,
+ struct max17042_chip, battery);
+
+ switch (psp) {
+ case POWER_SUPPLY_PROP_VOLTAGE_NOW:
+ val->intval = max17042_read_reg(chip->client,
+ MAX17042_VCELL) * 83; /* 1000 / 12 = 83 */
+ break;
+ case POWER_SUPPLY_PROP_VOLTAGE_AVG:
+ val->intval = max17042_read_reg(chip->client,
+ MAX17042_AvgVCELL) * 83;
+ break;
+ case POWER_SUPPLY_PROP_CAPACITY:
+ val->intval = max17042_read_reg(chip->client,
+ MAX17042_SOC) / 256;
+ break;
+ default:
+ return -EINVAL;
+ }
+ return 0;
+}
+
+static int __devinit max17042_probe(struct i2c_client *client,
+ const struct i2c_device_id *id)
+{
+ struct i2c_adapter *adapter = to_i2c_adapter(client->dev.parent);
+ struct max17042_chip *chip;
+ int ret;
+
+ if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_WORD_DATA))
+ return -EIO;
+
+ chip = kzalloc(sizeof(*chip), GFP_KERNEL);
+ if (!chip)
+ return -ENOMEM;
+
+ chip->client = client;
+ chip->pdata = client->dev.platform_data;
+
+ i2c_set_clientdata(client, chip);
+
+ chip->battery.name = "max17042_battery";
+ chip->battery.type = POWER_SUPPLY_TYPE_BATTERY;
+ chip->battery.get_property = max17042_get_property;
+ chip->battery.properties = max17042_battery_props;
+ chip->battery.num_properties = ARRAY_SIZE(max17042_battery_props);
+
+ ret = power_supply_register(&client->dev, &chip->battery);
+ if (ret) {
+ dev_err(&client->dev, "failed: power supply register\n");
+ i2c_set_clientdata(client, NULL);
+ kfree(chip);
+ return ret;
+ }
+
+ if (!chip->pdata->enable_current_sense) {
+ max17042_write_reg(client, MAX17042_CGAIN, 0x0000);
+ max17042_write_reg(client, MAX17042_MiscCFG, 0x0003);
+ max17042_write_reg(client, MAX17042_LearnCFG, 0x0007);
+ }
+
+ return 0;
+}
+
+static int __devexit max17042_remove(struct i2c_client *client)
+{
+ struct max17042_chip *chip = i2c_get_clientdata(client);
+
+ power_supply_unregister(&chip->battery);
+ i2c_set_clientdata(client, NULL);
+ kfree(chip);
+ return 0;
+}
+
+static const struct i2c_device_id max17042_id[] = {
+ { "max17042", 0 },
+ { }
+};
+MODULE_DEVICE_TABLE(i2c, max17042_id);
+
+static struct i2c_driver max17042_i2c_driver = {
+ .driver = {
+ .name = "max17042",
+ },
+ .probe = max17042_probe,
+ .remove = __devexit_p(max17042_remove),
+ .id_table = max17042_id,
+};
+
+static int __init max17042_init(void)
+{
+ return i2c_add_driver(&max17042_i2c_driver);
+}
+module_init(max17042_init);
+
+static void __exit max17042_exit(void)
+{
+ i2c_del_driver(&max17042_i2c_driver);
+}
+module_exit(max17042_exit);
+
+MODULE_AUTHOR("MyungJoo Ham <myungjoo.ham@samsung.com>");
+MODULE_DESCRIPTION("MAX17042 Fuel Gauge");
+MODULE_LICENSE("GPL");
diff --git a/drivers/power/olpc_battery.c b/drivers/power/olpc_battery.c
index 5bc1dcf..0b0ff3a 100644
--- a/drivers/power/olpc_battery.c
+++ b/drivers/power/olpc_battery.c
@@ -201,6 +201,72 @@
return ret;
}
+static int olpc_bat_get_charge_full_design(union power_supply_propval *val)
+{
+ uint8_t ec_byte;
+ union power_supply_propval tech;
+ int ret, mfr;
+
+ ret = olpc_bat_get_tech(&tech);
+ if (ret)
+ return ret;
+
+ ec_byte = BAT_ADDR_MFR_TYPE;
+ ret = olpc_ec_cmd(EC_BAT_EEPROM, &ec_byte, 1, &ec_byte, 1);
+ if (ret)
+ return ret;
+
+ mfr = ec_byte >> 4;
+
+ switch (tech.intval) {
+ case POWER_SUPPLY_TECHNOLOGY_NiMH:
+ switch (mfr) {
+ case 1: /* Gold Peak */
+ val->intval = 3000000*.8;
+ break;
+ default:
+ return -EIO;
+ }
+ break;
+
+ case POWER_SUPPLY_TECHNOLOGY_LiFe:
+ switch (mfr) {
+ case 1: /* Gold Peak */
+ val->intval = 2800000;
+ break;
+ case 2: /* BYD */
+ val->intval = 3100000;
+ break;
+ default:
+ return -EIO;
+ }
+ break;
+
+ default:
+ return -EIO;
+ }
+
+ return ret;
+}
+
+static int olpc_bat_get_charge_now(union power_supply_propval *val)
+{
+ uint8_t soc;
+ union power_supply_propval full;
+ int ret;
+
+ ret = olpc_ec_cmd(EC_BAT_SOC, NULL, 0, &soc, 1);
+ if (ret)
+ return ret;
+
+ ret = olpc_bat_get_charge_full_design(&full);
+ if (ret)
+ return ret;
+
+ val->intval = soc * (full.intval / 100);
+ return 0;
+}
+
/*********************************************************************
* Battery properties
*********************************************************************/
@@ -267,6 +333,7 @@
return ret;
break;
case POWER_SUPPLY_PROP_VOLTAGE_AVG:
+ case POWER_SUPPLY_PROP_VOLTAGE_NOW:
ret = olpc_ec_cmd(EC_BAT_VOLTAGE, NULL, 0, (void *)&ec_word, 2);
if (ret)
return ret;
@@ -274,6 +341,7 @@
val->intval = (s16)be16_to_cpu(ec_word) * 9760L / 32;
break;
case POWER_SUPPLY_PROP_CURRENT_AVG:
+ case POWER_SUPPLY_PROP_CURRENT_NOW:
ret = olpc_ec_cmd(EC_BAT_CURRENT, NULL, 0, (void *)&ec_word, 2);
if (ret)
return ret;
@@ -294,6 +362,16 @@
else
val->intval = POWER_SUPPLY_CAPACITY_LEVEL_NORMAL;
break;
+ case POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN:
+ ret = olpc_bat_get_charge_full_design(val);
+ if (ret)
+ return ret;
+ break;
+ case POWER_SUPPLY_PROP_CHARGE_NOW:
+ ret = olpc_bat_get_charge_now(val);
+ if (ret)
+ return ret;
+ break;
case POWER_SUPPLY_PROP_TEMP:
ret = olpc_ec_cmd(EC_BAT_TEMP, NULL, 0, (void *)&ec_word, 2);
if (ret)
@@ -331,16 +409,20 @@
return ret;
}
-static enum power_supply_property olpc_bat_props[] = {
+static enum power_supply_property olpc_xo1_bat_props[] = {
POWER_SUPPLY_PROP_STATUS,
POWER_SUPPLY_PROP_CHARGE_TYPE,
POWER_SUPPLY_PROP_PRESENT,
POWER_SUPPLY_PROP_HEALTH,
POWER_SUPPLY_PROP_TECHNOLOGY,
POWER_SUPPLY_PROP_VOLTAGE_AVG,
+ POWER_SUPPLY_PROP_VOLTAGE_NOW,
POWER_SUPPLY_PROP_CURRENT_AVG,
+ POWER_SUPPLY_PROP_CURRENT_NOW,
POWER_SUPPLY_PROP_CAPACITY,
POWER_SUPPLY_PROP_CAPACITY_LEVEL,
+ POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN,
+ POWER_SUPPLY_PROP_CHARGE_NOW,
POWER_SUPPLY_PROP_TEMP,
POWER_SUPPLY_PROP_TEMP_AMBIENT,
POWER_SUPPLY_PROP_MANUFACTURER,
@@ -348,6 +430,27 @@
POWER_SUPPLY_PROP_CHARGE_COUNTER,
};
+/* XO-1.5 does not have ambient temperature property */
+static enum power_supply_property olpc_xo15_bat_props[] = {
+ POWER_SUPPLY_PROP_STATUS,
+ POWER_SUPPLY_PROP_CHARGE_TYPE,
+ POWER_SUPPLY_PROP_PRESENT,
+ POWER_SUPPLY_PROP_HEALTH,
+ POWER_SUPPLY_PROP_TECHNOLOGY,
+ POWER_SUPPLY_PROP_VOLTAGE_AVG,
+ POWER_SUPPLY_PROP_VOLTAGE_NOW,
+ POWER_SUPPLY_PROP_CURRENT_AVG,
+ POWER_SUPPLY_PROP_CURRENT_NOW,
+ POWER_SUPPLY_PROP_CAPACITY,
+ POWER_SUPPLY_PROP_CAPACITY_LEVEL,
+ POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN,
+ POWER_SUPPLY_PROP_CHARGE_NOW,
+ POWER_SUPPLY_PROP_TEMP,
+ POWER_SUPPLY_PROP_MANUFACTURER,
+ POWER_SUPPLY_PROP_SERIAL_NUMBER,
+ POWER_SUPPLY_PROP_CHARGE_COUNTER,
+};
+
/* EEPROM reading goes completely around the power_supply API, sadly */
#define EEPROM_START 0x20
@@ -419,8 +522,6 @@
static struct platform_device *bat_pdev;
static struct power_supply olpc_bat = {
- .properties = olpc_bat_props,
- .num_properties = ARRAY_SIZE(olpc_bat_props),
.get_property = olpc_bat_get_property,
.use_for_apm = 1,
};
@@ -466,6 +567,13 @@
goto ac_failed;
olpc_bat.name = bat_pdev->name;
+ if (olpc_board_at_least(olpc_board_pre(0xd0))) { /* XO-1.5 */
+ olpc_bat.properties = olpc_xo15_bat_props;
+ olpc_bat.num_properties = ARRAY_SIZE(olpc_xo15_bat_props);
+ } else { /* XO-1 */
+ olpc_bat.properties = olpc_xo1_bat_props;
+ olpc_bat.num_properties = ARRAY_SIZE(olpc_xo1_bat_props);
+ }
ret = power_supply_register(&bat_pdev->dev, &olpc_bat);
if (ret)
diff --git a/drivers/power/power_supply_core.c b/drivers/power/power_supply_core.c
index 91606bb..970f733 100644
--- a/drivers/power/power_supply_core.c
+++ b/drivers/power/power_supply_core.c
@@ -190,10 +190,10 @@
goto success;
create_triggers_failed:
- device_unregister(psy->dev);
+ device_del(dev);
kobject_set_name_failed:
device_add_failed:
- kfree(dev);
+ put_device(dev);
success:
return rc;
}
@@ -201,7 +201,7 @@
void power_supply_unregister(struct power_supply *psy)
{
- flush_scheduled_work();
+ cancel_work_sync(&psy->changed_work);
power_supply_remove_triggers(psy);
device_unregister(psy->dev);
}
diff --git a/drivers/power/s3c_adc_battery.c b/drivers/power/s3c_adc_battery.c
index 4a8ae39..4255f23 100644
--- a/drivers/power/s3c_adc_battery.c
+++ b/drivers/power/s3c_adc_battery.c
@@ -112,6 +112,13 @@
return volt_val + cur_val * impedance / 1000;
}
+static int charge_finished(struct s3c_adc_bat *bat)
+{
+ return bat->pdata->gpio_inverted ?
+ !gpio_get_value(bat->pdata->gpio_charge_finished) :
+ gpio_get_value(bat->pdata->gpio_charge_finished);
+}
+
static int s3c_adc_bat_get_property(struct power_supply *psy,
enum power_supply_property psp,
union power_supply_propval *val)
@@ -140,7 +147,7 @@
if (bat->cable_plugged &&
((bat->pdata->gpio_charge_finished < 0) ||
- !gpio_get_value(bat->pdata->gpio_charge_finished))) {
+ !charge_finished(bat))) {
lut = bat->pdata->lut_acin;
lut_size = bat->pdata->lut_acin_cnt;
}
@@ -236,8 +243,7 @@
}
} else {
if ((bat->pdata->gpio_charge_finished >= 0) && is_plugged) {
- is_charged = gpio_get_value(
- main_bat.pdata->gpio_charge_finished);
+ is_charged = charge_finished(&main_bat);
if (is_charged) {
if (bat->pdata->disable_charger)
bat->pdata->disable_charger();
diff --git a/drivers/power/tosa_battery.c b/drivers/power/tosa_battery.c
index ee04936..53f0d35 100644
--- a/drivers/power/tosa_battery.c
+++ b/drivers/power/tosa_battery.c
@@ -332,7 +332,7 @@
static int tosa_bat_suspend(struct platform_device *dev, pm_message_t state)
{
/* flush all pending status updates */
- flush_scheduled_work();
+ flush_work_sync(&bat_work);
return 0;
}
@@ -422,7 +422,7 @@
err_psy_reg_main:
/* see comment in tosa_bat_remove */
- flush_scheduled_work();
+ cancel_work_sync(&bat_work);
i--;
err_gpio:
@@ -445,12 +445,11 @@
power_supply_unregister(&tosa_bat_main.psy);
/*
- * now flush all pending work.
- * we won't get any more schedules, since all
- * sources (isr and external_power_changed)
- * are unregistered now.
+ * Now cancel the bat_work. We won't get any more schedules,
+ * since all sources (isr and external_power_changed) are
+ * unregistered now.
*/
- flush_scheduled_work();
+ cancel_work_sync(&bat_work);
for (i = ARRAY_SIZE(gpios) - 1; i >= 0; i--)
gpio_free(gpios[i].gpio);
diff --git a/drivers/power/wm97xx_battery.c b/drivers/power/wm97xx_battery.c
index 5071d85..156559e 100644
--- a/drivers/power/wm97xx_battery.c
+++ b/drivers/power/wm97xx_battery.c
@@ -147,7 +147,7 @@
#ifdef CONFIG_PM
static int wm97xx_bat_suspend(struct device *dev)
{
- flush_scheduled_work();
+ flush_work_sync(&bat_work);
return 0;
}
@@ -273,7 +273,7 @@
free_irq(gpio_to_irq(pdata->charge_gpio), dev);
gpio_free(pdata->charge_gpio);
}
- flush_scheduled_work();
+ cancel_work_sync(&bat_work);
power_supply_unregister(&bat_ps);
kfree(prop);
return 0;
diff --git a/drivers/power/z2_battery.c b/drivers/power/z2_battery.c
index 85064a9..e5ed52d 100644
--- a/drivers/power/z2_battery.c
+++ b/drivers/power/z2_battery.c
@@ -254,7 +254,7 @@
struct z2_charger *charger = i2c_get_clientdata(client);
struct z2_battery_info *info = charger->info;
- flush_scheduled_work();
+ cancel_work_sync(&charger->bat_work);
power_supply_unregister(&charger->batt_ps);
kfree(charger->batt_ps.properties);
@@ -271,7 +271,9 @@
#ifdef CONFIG_PM
static int z2_batt_suspend(struct i2c_client *client, pm_message_t state)
{
- flush_scheduled_work();
+ struct z2_charger *charger = i2c_get_clientdata(client);
+
+ flush_work_sync(&charger->bat_work);
return 0;
}
diff --git a/drivers/regulator/max8998.c b/drivers/regulator/max8998.c
index 7568df6..0ec49ca 100644
--- a/drivers/regulator/max8998.c
+++ b/drivers/regulator/max8998.c
@@ -424,6 +424,9 @@
}
}
+ if (pdata->buck_voltage_lock)
+ return -EINVAL;
+
/* no predefine regulator found */
max8998->buck1_idx = (buck1_last_val % 2) + 2;
dev_dbg(max8998->dev, "max8998->buck1_idx:%d\n",
@@ -451,18 +454,26 @@
"BUCK2, i:%d buck2_vol1:%d, buck2_vol2:%d\n"
, i, max8998->buck2_vol[0], max8998->buck2_vol[1]);
if (gpio_is_valid(pdata->buck2_set3)) {
- if (max8998->buck2_vol[0] == i) {
- max8998->buck1_idx = 0;
- buck2_gpio_set(pdata->buck2_set3, 0);
- } else {
- max8998->buck1_idx = 1;
- ret = max8998_get_voltage_register(rdev, ®,
- &shift,
- &mask);
- ret = max8998_write_reg(i2c, reg, i);
- max8998->buck2_vol[1] = i;
- buck2_gpio_set(pdata->buck2_set3, 1);
+
+ /* check if requested voltage */
+ /* value is already defined */
+ for (j = 0; j < ARRAY_SIZE(max8998->buck2_vol); j++) {
+ if (max8998->buck2_vol[j] == i) {
+ max8998->buck2_idx = j;
+ buck2_gpio_set(pdata->buck2_set3, j);
+ goto buck2_exit;
+ }
}
+
+ if (pdata->buck_voltage_lock)
+ return -EINVAL;
+
+ max8998_get_voltage_register(rdev,
+ ®, &shift, &mask);
+ ret = max8998_write_reg(i2c, reg, i);
+ max8998->buck2_vol[max8998->buck2_idx] = i;
+ buck2_gpio_set(pdata->buck2_set3, max8998->buck2_idx);
+buck2_exit:
dev_dbg(max8998->dev, "%s: SET3:%d\n", i2c->name,
gpio_get_value(pdata->buck2_set3));
} else {
@@ -707,6 +718,9 @@
platform_set_drvdata(pdev, max8998);
i2c = max8998->iodev->i2c;
+ max8998->buck1_idx = pdata->buck1_default_idx;
+ max8998->buck2_idx = pdata->buck2_default_idx;
+
/* NOTE: */
/* For unused GPIO NOT marked as -1 (thereof equal to 0) WARN_ON */
/* will be displayed */
@@ -739,23 +753,46 @@
i = 0;
while (buck12_voltage_map_desc.min +
buck12_voltage_map_desc.step*i
- != (pdata->buck1_max_voltage1 / 1000))
+ < (pdata->buck1_voltage1 / 1000))
i++;
- printk(KERN_ERR "i:%d, buck1_idx:%d\n", i, max8998->buck1_idx);
max8998->buck1_vol[0] = i;
ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE1, i);
+ if (ret)
+ return ret;
/* Set predefined value for BUCK1 register 2 */
i = 0;
while (buck12_voltage_map_desc.min +
buck12_voltage_map_desc.step*i
- != (pdata->buck1_max_voltage2 / 1000))
+ < (pdata->buck1_voltage2 / 1000))
i++;
max8998->buck1_vol[1] = i;
- printk(KERN_ERR "i:%d, buck1_idx:%d\n", i, max8998->buck1_idx);
- ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE2, i)
- + ret;
+ ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE2, i);
+ if (ret)
+ return ret;
+
+ /* Set predefined value for BUCK1 register 3 */
+ i = 0;
+ while (buck12_voltage_map_desc.min +
+ buck12_voltage_map_desc.step*i
+ < (pdata->buck1_voltage3 / 1000))
+ i++;
+
+ max8998->buck1_vol[2] = i;
+ ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE3, i);
+ if (ret)
+ return ret;
+
+ /* Set predefined value for BUCK1 register 4 */
+ i = 0;
+ while (buck12_voltage_map_desc.min +
+ buck12_voltage_map_desc.step*i
+ < (pdata->buck1_voltage4 / 1000))
+ i++;
+
+ max8998->buck1_vol[3] = i;
+ ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE4, i);
if (ret)
return ret;
@@ -772,18 +809,28 @@
gpio_direction_output(pdata->buck2_set3,
max8998->buck2_idx & 0x1);
- /* BUCK2 - set preset default voltage value to buck2_vol[0] */
+ /* BUCK2 register 1 */
i = 0;
while (buck12_voltage_map_desc.min +
buck12_voltage_map_desc.step*i
- != (pdata->buck2_max_voltage / 1000))
+ < (pdata->buck2_voltage1 / 1000))
i++;
- printk(KERN_ERR "i:%d, buck2_idx:%d\n", i, max8998->buck2_idx);
max8998->buck2_vol[0] = i;
ret = max8998_write_reg(i2c, MAX8998_REG_BUCK2_VOLTAGE1, i);
if (ret)
return ret;
+ /* BUCK2 register 2 */
+ i = 0;
+ while (buck12_voltage_map_desc.min +
+ buck12_voltage_map_desc.step*i
+ < (pdata->buck2_voltage2 / 1000))
+ i++;
+ printk(KERN_ERR "i2:%d, buck2_idx:%d\n", i, max8998->buck2_idx);
+ max8998->buck2_vol[1] = i;
+ ret = max8998_write_reg(i2c, MAX8998_REG_BUCK2_VOLTAGE2, i);
+ if (ret)
+ return ret;
}
for (i = 0; i < pdata->num_regulators; i++) {
@@ -835,6 +882,12 @@
return 0;
}
+static const struct platform_device_id max8998_pmic_id[] = {
+ { "max8998-pmic", TYPE_MAX8998 },
+ { "lp3974-pmic", TYPE_LP3974 },
+ { }
+};
+
static struct platform_driver max8998_pmic_driver = {
.driver = {
.name = "max8998-pmic",
@@ -842,6 +895,7 @@
},
.probe = max8998_pmic_probe,
.remove = __devexit_p(max8998_pmic_remove),
+ .id_table = max8998_pmic_id,
};
static int __init max8998_pmic_init(void)
diff --git a/drivers/rtc/rtc-max8998.c b/drivers/rtc/rtc-max8998.c
index f22dee3..3f7bc6b 100644
--- a/drivers/rtc/rtc-max8998.c
+++ b/drivers/rtc/rtc-max8998.c
@@ -20,6 +20,7 @@
#include <linux/platform_device.h>
#include <linux/mfd/max8998.h>
#include <linux/mfd/max8998-private.h>
+#include <linux/delay.h>
#define MAX8998_RTC_SEC 0x00
#define MAX8998_RTC_MIN 0x01
@@ -73,6 +74,7 @@
struct i2c_client *rtc;
struct rtc_device *rtc_dev;
int irq;
+ bool lp3974_bug_workaround;
};
static void max8998_data_to_tm(u8 *data, struct rtc_time *tm)
@@ -124,10 +126,16 @@
{
struct max8998_rtc_info *info = dev_get_drvdata(dev);
u8 data[8];
+ int ret;
max8998_tm_to_data(tm, data);
- return max8998_bulk_write(info->rtc, MAX8998_RTC_SEC, 8, data);
+ ret = max8998_bulk_write(info->rtc, MAX8998_RTC_SEC, 8, data);
+
+ if (info->lp3974_bug_workaround)
+ msleep(2000);
+
+ return ret;
}
static int max8998_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm)
@@ -163,12 +171,29 @@
static int max8998_rtc_stop_alarm(struct max8998_rtc_info *info)
{
- return max8998_write_reg(info->rtc, MAX8998_ALARM0_CONF, 0);
+ int ret = max8998_write_reg(info->rtc, MAX8998_ALARM0_CONF, 0);
+
+ if (info->lp3974_bug_workaround)
+ msleep(2000);
+
+ return ret;
}
static int max8998_rtc_start_alarm(struct max8998_rtc_info *info)
{
- return max8998_write_reg(info->rtc, MAX8998_ALARM0_CONF, 0x77);
+ int ret;
+ u8 alarm0_conf = 0x77;
+
+ /* LP3974 with delay bug chips has rtc alarm bugs with "MONTH" field */
+ if (info->lp3974_bug_workaround)
+ alarm0_conf = 0x57;
+
+ ret = max8998_write_reg(info->rtc, MAX8998_ALARM0_CONF, alarm0_conf);
+
+ if (info->lp3974_bug_workaround)
+ msleep(2000);
+
+ return ret;
}
static int max8998_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
@@ -187,10 +212,13 @@
if (ret < 0)
return ret;
- if (alrm->enabled)
- return max8998_rtc_start_alarm(info);
+ if (info->lp3974_bug_workaround)
+ msleep(2000);
- return 0;
+ if (alrm->enabled)
+ ret = max8998_rtc_start_alarm(info);
+
+ return ret;
}
static int max8998_rtc_alarm_irq_enable(struct device *dev,
@@ -224,6 +252,7 @@
static int __devinit max8998_rtc_probe(struct platform_device *pdev)
{
struct max8998_dev *max8998 = dev_get_drvdata(pdev->dev.parent);
+ struct max8998_platform_data *pdata = dev_get_platdata(max8998->dev);
struct max8998_rtc_info *info;
int ret;
@@ -249,10 +278,18 @@
ret = request_threaded_irq(info->irq, NULL, max8998_rtc_alarm_irq, 0,
"rtc-alarm0", info);
+
if (ret < 0)
dev_err(&pdev->dev, "Failed to request alarm IRQ: %d: %d\n",
info->irq, ret);
+ dev_info(&pdev->dev, "RTC CHIP NAME: %s\n", pdev->id_entry->name);
+ if (pdata->rtc_delay) {
+ info->lp3974_bug_workaround = true;
+ dev_warn(&pdev->dev, "LP3974 with RTC REGERR option."
+ " RTC updates will be extremely slow.\n");
+ }
+
return 0;
out_rtc:
@@ -273,6 +310,12 @@
return 0;
}
+static const struct platform_device_id max8998_rtc_id[] = {
+ { "max8998-rtc", TYPE_MAX8998 },
+ { "lp3974-rtc", TYPE_LP3974 },
+ { }
+};
+
static struct platform_driver max8998_rtc_driver = {
.driver = {
.name = "max8998-rtc",
@@ -280,6 +323,7 @@
},
.probe = max8998_rtc_probe,
.remove = __devexit_p(max8998_rtc_remove),
+ .id_table = max8998_rtc_id,
};
static int __init max8998_rtc_init(void)
diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c
index e8391b89..b7eaff9 100644
--- a/drivers/s390/cio/device.c
+++ b/drivers/s390/cio/device.c
@@ -1835,6 +1835,7 @@
* available again. Kick re-detection.
*/
cdev->private->flags.resuming = 1;
+ cdev->private->path_new_mask = LPM_ANYPATH;
css_schedule_eval(sch->schid);
spin_unlock_irq(sch->lock);
css_complete_work();
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index d3c5905..9c5c8be 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -7515,16 +7515,10 @@
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
volatile u32 int_reg;
- int rc;
ENTER;
ioa_cfg->pdev->state_saved = true;
- rc = pci_restore_state(ioa_cfg->pdev);
-
- if (rc != PCIBIOS_SUCCESSFUL) {
- ipr_cmd->s.ioasa.hdr.ioasc = cpu_to_be32(IPR_IOASC_PCI_ACCESS_ERROR);
- return IPR_RC_JOB_CONTINUE;
- }
+ pci_restore_state(ioa_cfg->pdev);
if (ipr_set_pcix_cmd_reg(ioa_cfg)) {
ipr_cmd->s.ioasa.hdr.ioasc = cpu_to_be32(IPR_IOASC_PCI_ACCESS_ERROR);
diff --git a/drivers/scsi/pmcraid.c b/drivers/scsi/pmcraid.c
index 300d59f..321cf3a 100644
--- a/drivers/scsi/pmcraid.c
+++ b/drivers/scsi/pmcraid.c
@@ -2228,12 +2228,7 @@
/* Once either bist or pci reset is done, restore PCI config
* space. If this fails, proceed with hard reset again
*/
- if (pci_restore_state(pinstance->pdev)) {
- pmcraid_info("config-space error resetting again\n");
- pinstance->ioa_state = IOA_STATE_IN_RESET_ALERT;
- pmcraid_reset_alert(cmd);
- break;
- }
+ pci_restore_state(pinstance->pdev);
/* fail all pending commands */
pmcraid_fail_outstanding_cmds(pinstance);
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index b65e65a..e567302 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -990,30 +990,51 @@
static void set_media_not_present(struct scsi_disk *sdkp)
{
- sdkp->media_present = 0;
- sdkp->capacity = 0;
- sdkp->device->changed = 1;
+ if (sdkp->media_present)
+ sdkp->device->changed = 1;
+
+ if (sdkp->device->removable) {
+ sdkp->media_present = 0;
+ sdkp->capacity = 0;
+ }
+}
+
+static int media_not_present(struct scsi_disk *sdkp,
+ struct scsi_sense_hdr *sshdr)
+{
+ if (!scsi_sense_valid(sshdr))
+ return 0;
+
+ /* not invoked for commands that could return deferred errors */
+ switch (sshdr->sense_key) {
+ case UNIT_ATTENTION:
+ case NOT_READY:
+ /* medium not present */
+ if (sshdr->asc == 0x3A) {
+ set_media_not_present(sdkp);
+ return 1;
+ }
+ }
+ return 0;
}
/**
- * sd_media_changed - check if our medium changed
- * @disk: kernel device descriptor
+ * sd_check_events - check media events
+ * @disk: kernel device descriptor
+ * @clearing: disk events currently being cleared
*
- * Returns 0 if not applicable or no change; 1 if change
+ * Returns mask of DISK_EVENT_*.
*
* Note: this function is invoked from the block subsystem.
**/
-static int sd_media_changed(struct gendisk *disk)
+static unsigned int sd_check_events(struct gendisk *disk, unsigned int clearing)
{
struct scsi_disk *sdkp = scsi_disk(disk);
struct scsi_device *sdp = sdkp->device;
struct scsi_sense_hdr *sshdr = NULL;
int retval;
- SCSI_LOG_HLQUEUE(3, sd_printk(KERN_INFO, sdkp, "sd_media_changed\n"));
-
- if (!sdp->removable)
- return 0;
+ SCSI_LOG_HLQUEUE(3, sd_printk(KERN_INFO, sdkp, "sd_check_events\n"));
/*
* If the device is offline, don't send any commands - just pretend as
@@ -1043,40 +1064,32 @@
sshdr);
}
- if (retval) {
+ /* failed to execute TUR, assume media not present */
+ if (host_byte(retval)) {
set_media_not_present(sdkp);
goto out;
}
+ if (media_not_present(sdkp, sshdr))
+ goto out;
+
/*
* For removable scsi disk we have to recognise the presence
- * of a disk in the drive. This is kept in the struct scsi_disk
- * struct and tested at open ! Daniel Roche (dan@lectra.fr)
+ * of a disk in the drive.
*/
+ if (!sdkp->media_present)
+ sdp->changed = 1;
sdkp->media_present = 1;
-
out:
/*
- * Report a media change under the following conditions:
+ * sdp->changed is set under the following conditions:
*
- * Medium is present now and wasn't present before.
- * Medium wasn't present before and is present now.
- * Medium was present at all times, but it changed while
- * we weren't looking (sdp->changed is set).
- *
- * If there was no medium before and there is no medium now then
- * don't report a change, even if a medium was inserted and removed
- * while we weren't looking.
+ * Medium present state has changed in either direction.
+ * Device has indicated UNIT_ATTENTION.
*/
- retval = (sdkp->media_present != sdkp->previous_state ||
- (sdkp->media_present && sdp->changed));
- if (retval)
- sdev_evt_send_simple(sdp, SDEV_EVT_MEDIA_CHANGE, GFP_KERNEL);
- sdkp->previous_state = sdkp->media_present;
-
- /* sdp->changed indicates medium was changed or is not present */
- sdp->changed = !sdkp->media_present;
kfree(sshdr);
+ retval = sdp->changed ? DISK_EVENT_MEDIA_CHANGE : 0;
+ sdp->changed = 0;
return retval;
}
@@ -1169,7 +1182,7 @@
#ifdef CONFIG_COMPAT
.compat_ioctl = sd_compat_ioctl,
#endif
- .media_changed = sd_media_changed,
+ .check_events = sd_check_events,
.revalidate_disk = sd_revalidate_disk,
.unlock_native_capacity = sd_unlock_native_capacity,
};
@@ -1312,23 +1325,6 @@
return good_bytes;
}
-static int media_not_present(struct scsi_disk *sdkp,
- struct scsi_sense_hdr *sshdr)
-{
-
- if (!scsi_sense_valid(sshdr))
- return 0;
- /* not invoked for commands that could return deferred errors */
- if (sshdr->sense_key != NOT_READY &&
- sshdr->sense_key != UNIT_ATTENTION)
- return 0;
- if (sshdr->asc != 0x3A) /* medium not present */
- return 0;
-
- set_media_not_present(sdkp);
- return 1;
-}
-
/*
* spinup disk - called only in sd_revalidate_disk()
*/
@@ -1503,7 +1499,7 @@
*/
if (sdp->removable &&
sense_valid && sshdr->sense_key == NOT_READY)
- sdp->changed = 1;
+ set_media_not_present(sdkp);
/*
* We used to set media_present to 0 here to indicate no media
@@ -2389,8 +2385,10 @@
gd->driverfs_dev = &sdp->sdev_gendev;
gd->flags = GENHD_FL_EXT_DEVT;
- if (sdp->removable)
+ if (sdp->removable) {
gd->flags |= GENHD_FL_REMOVABLE;
+ gd->events |= DISK_EVENT_MEDIA_CHANGE;
+ }
add_disk(gd);
sd_dif_config_host(sdkp);
@@ -2472,7 +2470,6 @@
sdkp->disk = gd;
sdkp->index = index;
atomic_set(&sdkp->openers, 0);
- sdkp->previous_state = 1;
if (!sdp->request_queue->rq_timeout) {
if (sdp->type != TYPE_MOD)
diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h
index 55488fa..c9d8f6c 100644
--- a/drivers/scsi/sd.h
+++ b/drivers/scsi/sd.h
@@ -55,7 +55,6 @@
u8 media_present;
u8 write_prot;
u8 protection_type;/* Data Integrity Field */
- unsigned previous_state : 1;
unsigned ATO : 1; /* state of disk ATO bit */
unsigned WCE : 1; /* state of disk WCE bit */
unsigned RCD : 1; /* state of disk RCD bit, unused */
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index be6baf8..aefadc6 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -249,10 +249,6 @@
cd->device->changed = 0;
}
- /* for backward compatibility */
- if (events & DISK_EVENT_MEDIA_CHANGE)
- sdev_evt_send_simple(cd->device, SDEV_EVT_MEDIA_CHANGE,
- GFP_KERNEL);
return events;
}
diff --git a/drivers/serial/atmel_serial.c b/drivers/serial/atmel_serial.c
index 3892666..2a1d52f 100644
--- a/drivers/serial/atmel_serial.c
+++ b/drivers/serial/atmel_serial.c
@@ -1732,6 +1732,11 @@
device_init_wakeup(&pdev->dev, 1);
platform_set_drvdata(pdev, port);
+ if (port->rs485.flags & SER_RS485_ENABLED) {
+ UART_PUT_MR(&port->uart, ATMEL_US_USMODE_NORMAL);
+ UART_PUT_CR(&port->uart, ATMEL_US_RTSEN);
+ }
+
return 0;
err_add_port:
diff --git a/drivers/serial/samsung.c b/drivers/serial/samsung.c
index 7ac2bf5..2335eda 100644
--- a/drivers/serial/samsung.c
+++ b/drivers/serial/samsung.c
@@ -883,10 +883,10 @@
static struct uart_driver s3c24xx_uart_drv = {
.owner = THIS_MODULE,
- .dev_name = "s3c2410_serial",
+ .driver_name = "s3c2410_serial",
.nr = CONFIG_SERIAL_SAMSUNG_UARTS,
.cons = S3C24XX_SERIAL_CONSOLE,
- .driver_name = S3C24XX_SERIAL_NAME,
+ .dev_name = S3C24XX_SERIAL_NAME,
.major = S3C24XX_SERIAL_MAJOR,
.minor = S3C24XX_SERIAL_MINOR,
};
diff --git a/drivers/sfi/sfi_core.c b/drivers/sfi/sfi_core.c
index ceba593..04113e5 100644
--- a/drivers/sfi/sfi_core.c
+++ b/drivers/sfi/sfi_core.c
@@ -101,7 +101,7 @@
return NULL;
if (sfi_use_ioremap)
- return ioremap(phys, size);
+ return ioremap_cache(phys, size);
else
return early_ioremap(phys, size);
}
diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 1906840..13bfa9d 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -156,10 +156,10 @@
def_bool y if ARCH_MX31
config SPI_IMX_VER_0_7
- def_bool y if ARCH_MX25 || ARCH_MX35 || ARCH_MX51
+ def_bool y if ARCH_MX25 || ARCH_MX35 || ARCH_MX51 || ARCH_MX53
config SPI_IMX_VER_2_3
- def_bool y if ARCH_MX51
+ def_bool y if ARCH_MX51 || ARCH_MX53
config SPI_IMX
tristate "Freescale i.MX SPI controllers"
@@ -310,8 +310,8 @@
config SPI_S3C64XX
tristate "Samsung S3C64XX series type SPI"
- depends on ARCH_S3C64XX && EXPERIMENTAL
- select S3C64XX_DMA
+ depends on (ARCH_S3C64XX || ARCH_S5P64X0)
+ select S3C64XX_DMA if ARCH_S3C64XX
help
SPI driver for Samsung S3C64XX and newer SoCs.
diff --git a/drivers/spi/amba-pl022.c b/drivers/spi/amba-pl022.c
index a2a5921..71a1219 100644
--- a/drivers/spi/amba-pl022.c
+++ b/drivers/spi/amba-pl022.c
@@ -1795,7 +1795,7 @@
{
struct pl022_config_chip const *chip_info;
struct chip_data *chip;
- struct ssp_clock_params clk_freq;
+ struct ssp_clock_params clk_freq = {0, };
int status = 0;
struct pl022 *pl022 = spi_master_get_devdata(spi->master);
unsigned int bits = spi->bits_per_word;
diff --git a/drivers/spi/dw_spi_mmio.c b/drivers/spi/dw_spi_mmio.c
index db35bd9..2fa012c1 100644
--- a/drivers/spi/dw_spi_mmio.c
+++ b/drivers/spi/dw_spi_mmio.c
@@ -9,6 +9,7 @@
*/
#include <linux/clk.h>
+#include <linux/err.h>
#include <linux/interrupt.h>
#include <linux/platform_device.h>
#include <linux/slab.h>
@@ -68,8 +69,8 @@
}
dwsmmio->clk = clk_get(&pdev->dev, NULL);
- if (!dwsmmio->clk) {
- ret = -ENODEV;
+ if (IS_ERR(dwsmmio->clk)) {
+ ret = PTR_ERR(dwsmmio->clk);
goto err_irq;
}
clk_enable(dwsmmio->clk);
diff --git a/drivers/spi/spi_imx.c b/drivers/spi/spi_imx.c
index 9469564..1cf9d5f 100644
--- a/drivers/spi/spi_imx.c
+++ b/drivers/spi/spi_imx.c
@@ -743,6 +743,12 @@
.name = "imx51-ecspi",
.driver_data = SPI_IMX_VER_2_3,
}, {
+ .name = "imx53-cspi",
+ .driver_data = SPI_IMX_VER_0_7,
+ }, {
+ .name = "imx53-ecspi",
+ .driver_data = SPI_IMX_VER_2_3,
+ }, {
/* sentinel */
}
};
diff --git a/drivers/spi/spi_tegra.c b/drivers/spi/spi_tegra.c
index bb7df02..891e590 100644
--- a/drivers/spi/spi_tegra.c
+++ b/drivers/spi/spi_tegra.c
@@ -513,7 +513,7 @@
}
tspi->clk = clk_get(&pdev->dev, NULL);
- if (IS_ERR_OR_NULL(tspi->clk)) {
+ if (IS_ERR(tspi->clk)) {
dev_err(&pdev->dev, "can not get clock\n");
ret = PTR_ERR(tspi->clk);
goto err2;
diff --git a/drivers/ssb/scan.c b/drivers/ssb/scan.c
index 5a0985d..29884c0 100644
--- a/drivers/ssb/scan.c
+++ b/drivers/ssb/scan.c
@@ -420,6 +420,16 @@
bus->pcicore.dev = dev;
#endif /* CONFIG_SSB_DRIVER_PCICORE */
break;
+ case SSB_DEV_ETHERNET:
+ if (bus->bustype == SSB_BUSTYPE_PCI) {
+ if (bus->host_pci->vendor == PCI_VENDOR_ID_BROADCOM &&
+ (bus->host_pci->device & 0xFF00) == 0x4300) {
+ /* This is a dangling ethernet core on a
+ * wireless device. Ignore it. */
+ continue;
+ }
+ }
+ break;
default:
break;
}
diff --git a/drivers/staging/autofs/dirhash.c b/drivers/staging/autofs/dirhash.c
index d3f42c8..a08bd73 100644
--- a/drivers/staging/autofs/dirhash.c
+++ b/drivers/staging/autofs/dirhash.c
@@ -88,14 +88,13 @@
}
path.mnt = mnt;
path_get(&path);
- if (!follow_down(&path)) {
+ if (!follow_down_one(&path)) {
path_put(&path);
DPRINTK(("autofs: not expirable\
(not a mounted directory): %s\n", ent->name));
continue;
}
- while (d_mountpoint(path.dentry) && follow_down(&path))
- ;
+ follow_down(&path, false); // TODO: need to check error
umount_ok = may_umount(path.mnt);
path_put(&path);
diff --git a/drivers/staging/sm7xx/smtcfb.c b/drivers/staging/sm7xx/smtcfb.c
index f4b163f..0bc113c 100644
--- a/drivers/staging/sm7xx/smtcfb.c
+++ b/drivers/staging/sm7xx/smtcfb.c
@@ -1071,7 +1071,7 @@
/* when resuming, restore pci data and fb cursor */
if (pdev->dev.power.power_state.event != PM_EVENT_FREEZE) {
retv = pci_set_power_state(pdev, PCI_D0);
- retv = pci_restore_state(pdev);
+ pci_restore_state(pdev);
if (pci_enable_device(pdev))
return -1;
pci_set_master(pdev);
diff --git a/drivers/target/Kconfig b/drivers/target/Kconfig
new file mode 100644
index 0000000..2fac3be
--- /dev/null
+++ b/drivers/target/Kconfig
@@ -0,0 +1,32 @@
+
+menuconfig TARGET_CORE
+ tristate "Generic Target Core Mod (TCM) and ConfigFS Infrastructure"
+ depends on SCSI && BLOCK
+ select CONFIGFS_FS
+ default n
+ help
+ Say Y or M here to enable the TCM Storage Engine and ConfigFS enabled
+ control path for target_core_mod. This includes built-in TCM RAMDISK
+ subsystem logic for virtual LUN 0 access
+
+if TARGET_CORE
+
+config TCM_IBLOCK
+ tristate "TCM/IBLOCK Subsystem Plugin for Linux/BLOCK"
+ help
+ Say Y here to enable the TCM/IBLOCK subsystem plugin for non-buffered
+ access to Linux/Block devices using BIO
+
+config TCM_FILEIO
+ tristate "TCM/FILEIO Subsystem Plugin for Linux/VFS"
+ help
+ Say Y here to enable the TCM/FILEIO subsystem plugin for buffered
+ access to Linux/VFS struct file or struct block_device
+
+config TCM_PSCSI
+ tristate "TCM/pSCSI Subsystem Plugin for Linux/SCSI"
+ help
+ Say Y here to enable the TCM/pSCSI subsystem plugin for non-buffered
+ passthrough access to Linux/SCSI device
+
+endif
diff --git a/drivers/target/Makefile b/drivers/target/Makefile
new file mode 100644
index 0000000..5cfd708
--- /dev/null
+++ b/drivers/target/Makefile
@@ -0,0 +1,24 @@
+EXTRA_CFLAGS += -I$(srctree)/drivers/target/ -I$(srctree)/drivers/scsi/
+
+target_core_mod-y := target_core_configfs.o \
+ target_core_device.o \
+ target_core_fabric_configfs.o \
+ target_core_fabric_lib.o \
+ target_core_hba.o \
+ target_core_pr.o \
+ target_core_alua.o \
+ target_core_scdb.o \
+ target_core_tmr.o \
+ target_core_tpg.o \
+ target_core_transport.o \
+ target_core_cdb.o \
+ target_core_ua.o \
+ target_core_rd.o \
+ target_core_mib.o
+
+obj-$(CONFIG_TARGET_CORE) += target_core_mod.o
+
+# Subsystem modules
+obj-$(CONFIG_TCM_IBLOCK) += target_core_iblock.o
+obj-$(CONFIG_TCM_FILEIO) += target_core_file.o
+obj-$(CONFIG_TCM_PSCSI) += target_core_pscsi.o
diff --git a/drivers/target/target_core_alua.c b/drivers/target/target_core_alua.c
new file mode 100644
index 0000000..2c5fcfe
--- /dev/null
+++ b/drivers/target/target_core_alua.c
@@ -0,0 +1,1991 @@
+/*******************************************************************************
+ * Filename: target_core_alua.c
+ *
+ * This file contains SPC-3 compliant asymmetric logical unit assigntment (ALUA)
+ *
+ * Copyright (c) 2009-2010 Rising Tide Systems
+ * Copyright (c) 2009-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/configfs.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_configfs.h>
+
+#include "target_core_alua.h"
+#include "target_core_hba.h"
+#include "target_core_ua.h"
+
+static int core_alua_check_transition(int state, int *primary);
+static int core_alua_set_tg_pt_secondary_state(
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem,
+ struct se_port *port, int explict, int offline);
+
+/*
+ * REPORT_TARGET_PORT_GROUPS
+ *
+ * See spc4r17 section 6.27
+ */
+int core_emulate_report_target_port_groups(struct se_cmd *cmd)
+{
+ struct se_subsystem_dev *su_dev = SE_DEV(cmd)->se_sub_dev;
+ struct se_port *port;
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ u32 rd_len = 0, off = 4; /* Skip over RESERVED area to first
+ Target port group descriptor */
+
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ list_for_each_entry(tg_pt_gp, &T10_ALUA(su_dev)->tg_pt_gps_list,
+ tg_pt_gp_list) {
+ /*
+ * PREF: Preferred target port bit, determine if this
+ * bit should be set for port group.
+ */
+ if (tg_pt_gp->tg_pt_gp_pref)
+ buf[off] = 0x80;
+ /*
+ * Set the ASYMMETRIC ACCESS State
+ */
+ buf[off++] |= (atomic_read(
+ &tg_pt_gp->tg_pt_gp_alua_access_state) & 0xff);
+ /*
+ * Set supported ASYMMETRIC ACCESS State bits
+ */
+ buf[off] = 0x80; /* T_SUP */
+ buf[off] |= 0x40; /* O_SUP */
+ buf[off] |= 0x8; /* U_SUP */
+ buf[off] |= 0x4; /* S_SUP */
+ buf[off] |= 0x2; /* AN_SUP */
+ buf[off++] |= 0x1; /* AO_SUP */
+ /*
+ * TARGET PORT GROUP
+ */
+ buf[off++] = ((tg_pt_gp->tg_pt_gp_id >> 8) & 0xff);
+ buf[off++] = (tg_pt_gp->tg_pt_gp_id & 0xff);
+
+ off++; /* Skip over Reserved */
+ /*
+ * STATUS CODE
+ */
+ buf[off++] = (tg_pt_gp->tg_pt_gp_alua_access_status & 0xff);
+ /*
+ * Vendor Specific field
+ */
+ buf[off++] = 0x00;
+ /*
+ * TARGET PORT COUNT
+ */
+ buf[off++] = (tg_pt_gp->tg_pt_gp_members & 0xff);
+ rd_len += 8;
+
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ list_for_each_entry(tg_pt_gp_mem, &tg_pt_gp->tg_pt_gp_mem_list,
+ tg_pt_gp_mem_list) {
+ port = tg_pt_gp_mem->tg_pt;
+ /*
+ * Start Target Port descriptor format
+ *
+ * See spc4r17 section 6.2.7 Table 247
+ */
+ off += 2; /* Skip over Obsolete */
+ /*
+ * Set RELATIVE TARGET PORT IDENTIFIER
+ */
+ buf[off++] = ((port->sep_rtpi >> 8) & 0xff);
+ buf[off++] = (port->sep_rtpi & 0xff);
+ rd_len += 4;
+ }
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+ }
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ /*
+ * Set the RETURN DATA LENGTH set in the header of the DataIN Payload
+ */
+ buf[0] = ((rd_len >> 24) & 0xff);
+ buf[1] = ((rd_len >> 16) & 0xff);
+ buf[2] = ((rd_len >> 8) & 0xff);
+ buf[3] = (rd_len & 0xff);
+
+ return 0;
+}
+
+/*
+ * SET_TARGET_PORT_GROUPS for explict ALUA operation.
+ *
+ * See spc4r17 section 6.35
+ */
+int core_emulate_set_target_port_groups(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_subsystem_dev *su_dev = SE_DEV(cmd)->se_sub_dev;
+ struct se_port *port, *l_port = SE_LUN(cmd)->lun_sep;
+ struct se_node_acl *nacl = SE_SESS(cmd)->se_node_acl;
+ struct t10_alua_tg_pt_gp *tg_pt_gp = NULL, *l_tg_pt_gp;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem, *l_tg_pt_gp_mem;
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ unsigned char *ptr = &buf[4]; /* Skip over RESERVED area in header */
+ u32 len = 4; /* Skip over RESERVED area in header */
+ int alua_access_state, primary = 0, rc;
+ u16 tg_pt_id, rtpi;
+
+ if (!(l_port))
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ /*
+ * Determine if explict ALUA via SET_TARGET_PORT_GROUPS is allowed
+ * for the local tg_pt_gp.
+ */
+ l_tg_pt_gp_mem = l_port->sep_alua_tg_pt_gp_mem;
+ if (!(l_tg_pt_gp_mem)) {
+ printk(KERN_ERR "Unable to access l_port->sep_alua_tg_pt_gp_mem\n");
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+ spin_lock(&l_tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ l_tg_pt_gp = l_tg_pt_gp_mem->tg_pt_gp;
+ if (!(l_tg_pt_gp)) {
+ spin_unlock(&l_tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ printk(KERN_ERR "Unable to access *l_tg_pt_gp_mem->tg_pt_gp\n");
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+ rc = (l_tg_pt_gp->tg_pt_gp_alua_access_type & TPGS_EXPLICT_ALUA);
+ spin_unlock(&l_tg_pt_gp_mem->tg_pt_gp_mem_lock);
+
+ if (!(rc)) {
+ printk(KERN_INFO "Unable to process SET_TARGET_PORT_GROUPS"
+ " while TPGS_EXPLICT_ALUA is disabled\n");
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+
+ while (len < cmd->data_length) {
+ alua_access_state = (ptr[0] & 0x0f);
+ /*
+ * Check the received ALUA access state, and determine if
+ * the state is a primary or secondary target port asymmetric
+ * access state.
+ */
+ rc = core_alua_check_transition(alua_access_state, &primary);
+ if (rc != 0) {
+ /*
+ * If the SET TARGET PORT GROUPS attempts to establish
+ * an invalid combination of target port asymmetric
+ * access states or attempts to establish an
+ * unsupported target port asymmetric access state,
+ * then the command shall be terminated with CHECK
+ * CONDITION status, with the sense key set to ILLEGAL
+ * REQUEST, and the additional sense code set to INVALID
+ * FIELD IN PARAMETER LIST.
+ */
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ }
+ rc = -1;
+ /*
+ * If the ASYMMETRIC ACCESS STATE field (see table 267)
+ * specifies a primary target port asymmetric access state,
+ * then the TARGET PORT GROUP OR TARGET PORT field specifies
+ * a primary target port group for which the primary target
+ * port asymmetric access state shall be changed. If the
+ * ASYMMETRIC ACCESS STATE field specifies a secondary target
+ * port asymmetric access state, then the TARGET PORT GROUP OR
+ * TARGET PORT field specifies the relative target port
+ * identifier (see 3.1.120) of the target port for which the
+ * secondary target port asymmetric access state shall be
+ * changed.
+ */
+ if (primary) {
+ tg_pt_id = ((ptr[2] << 8) & 0xff);
+ tg_pt_id |= (ptr[3] & 0xff);
+ /*
+ * Locate the matching target port group ID from
+ * the global tg_pt_gp list
+ */
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ list_for_each_entry(tg_pt_gp,
+ &T10_ALUA(su_dev)->tg_pt_gps_list,
+ tg_pt_gp_list) {
+ if (!(tg_pt_gp->tg_pt_gp_valid_id))
+ continue;
+
+ if (tg_pt_id != tg_pt_gp->tg_pt_gp_id)
+ continue;
+
+ atomic_inc(&tg_pt_gp->tg_pt_gp_ref_cnt);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+
+ rc = core_alua_do_port_transition(tg_pt_gp,
+ dev, l_port, nacl,
+ alua_access_state, 1);
+
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ atomic_dec(&tg_pt_gp->tg_pt_gp_ref_cnt);
+ smp_mb__after_atomic_dec();
+ break;
+ }
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ /*
+ * If not matching target port group ID can be located
+ * throw an exception with ASCQ: INVALID_PARAMETER_LIST
+ */
+ if (rc != 0)
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ } else {
+ /*
+ * Extact the RELATIVE TARGET PORT IDENTIFIER to identify
+ * the Target Port in question for the the incoming
+ * SET_TARGET_PORT_GROUPS op.
+ */
+ rtpi = ((ptr[2] << 8) & 0xff);
+ rtpi |= (ptr[3] & 0xff);
+ /*
+ * Locate the matching relative target port identifer
+ * for the struct se_device storage object.
+ */
+ spin_lock(&dev->se_port_lock);
+ list_for_each_entry(port, &dev->dev_sep_list,
+ sep_list) {
+ if (port->sep_rtpi != rtpi)
+ continue;
+
+ tg_pt_gp_mem = port->sep_alua_tg_pt_gp_mem;
+ spin_unlock(&dev->se_port_lock);
+
+ rc = core_alua_set_tg_pt_secondary_state(
+ tg_pt_gp_mem, port, 1, 1);
+
+ spin_lock(&dev->se_port_lock);
+ break;
+ }
+ spin_unlock(&dev->se_port_lock);
+ /*
+ * If not matching relative target port identifier can
+ * be located, throw an exception with ASCQ:
+ * INVALID_PARAMETER_LIST
+ */
+ if (rc != 0)
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ }
+
+ ptr += 4;
+ len += 4;
+ }
+
+ return 0;
+}
+
+static inline int core_alua_state_nonoptimized(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ int nonop_delay_msecs,
+ u8 *alua_ascq)
+{
+ /*
+ * Set SCF_ALUA_NON_OPTIMIZED here, this value will be checked
+ * later to determine if processing of this cmd needs to be
+ * temporarily delayed for the Active/NonOptimized primary access state.
+ */
+ cmd->se_cmd_flags |= SCF_ALUA_NON_OPTIMIZED;
+ cmd->alua_nonop_delay = nonop_delay_msecs;
+ return 0;
+}
+
+static inline int core_alua_state_standby(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ u8 *alua_ascq)
+{
+ /*
+ * Allowed CDBs for ALUA_ACCESS_STATE_STANDBY as defined by
+ * spc4r17 section 5.9.2.4.4
+ */
+ switch (cdb[0]) {
+ case INQUIRY:
+ case LOG_SELECT:
+ case LOG_SENSE:
+ case MODE_SELECT:
+ case MODE_SENSE:
+ case REPORT_LUNS:
+ case RECEIVE_DIAGNOSTIC:
+ case SEND_DIAGNOSTIC:
+ case MAINTENANCE_IN:
+ switch (cdb[1]) {
+ case MI_REPORT_TARGET_PGS:
+ return 0;
+ default:
+ *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ return 1;
+ }
+ case MAINTENANCE_OUT:
+ switch (cdb[1]) {
+ case MO_SET_TARGET_PGS:
+ return 0;
+ default:
+ *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ return 1;
+ }
+ case REQUEST_SENSE:
+ case PERSISTENT_RESERVE_IN:
+ case PERSISTENT_RESERVE_OUT:
+ case READ_BUFFER:
+ case WRITE_BUFFER:
+ return 0;
+ default:
+ *alua_ascq = ASCQ_04H_ALUA_TG_PT_STANDBY;
+ return 1;
+ }
+
+ return 0;
+}
+
+static inline int core_alua_state_unavailable(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ u8 *alua_ascq)
+{
+ /*
+ * Allowed CDBs for ALUA_ACCESS_STATE_UNAVAILABLE as defined by
+ * spc4r17 section 5.9.2.4.5
+ */
+ switch (cdb[0]) {
+ case INQUIRY:
+ case REPORT_LUNS:
+ case MAINTENANCE_IN:
+ switch (cdb[1]) {
+ case MI_REPORT_TARGET_PGS:
+ return 0;
+ default:
+ *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ return 1;
+ }
+ case MAINTENANCE_OUT:
+ switch (cdb[1]) {
+ case MO_SET_TARGET_PGS:
+ return 0;
+ default:
+ *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ return 1;
+ }
+ case REQUEST_SENSE:
+ case READ_BUFFER:
+ case WRITE_BUFFER:
+ return 0;
+ default:
+ *alua_ascq = ASCQ_04H_ALUA_TG_PT_UNAVAILABLE;
+ return 1;
+ }
+
+ return 0;
+}
+
+static inline int core_alua_state_transition(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ u8 *alua_ascq)
+{
+ /*
+ * Allowed CDBs for ALUA_ACCESS_STATE_TRANSITIO as defined by
+ * spc4r17 section 5.9.2.5
+ */
+ switch (cdb[0]) {
+ case INQUIRY:
+ case REPORT_LUNS:
+ case MAINTENANCE_IN:
+ switch (cdb[1]) {
+ case MI_REPORT_TARGET_PGS:
+ return 0;
+ default:
+ *alua_ascq = ASCQ_04H_ALUA_STATE_TRANSITION;
+ return 1;
+ }
+ case REQUEST_SENSE:
+ case READ_BUFFER:
+ case WRITE_BUFFER:
+ return 0;
+ default:
+ *alua_ascq = ASCQ_04H_ALUA_STATE_TRANSITION;
+ return 1;
+ }
+
+ return 0;
+}
+
+/*
+ * Used for alua_type SPC_ALUA_PASSTHROUGH and SPC2_ALUA_DISABLED
+ * in transport_cmd_sequencer(). This function is assigned to
+ * struct t10_alua *->state_check() in core_setup_alua()
+ */
+static int core_alua_state_check_nop(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ u8 *alua_ascq)
+{
+ return 0;
+}
+
+/*
+ * Used for alua_type SPC3_ALUA_EMULATED in transport_cmd_sequencer().
+ * This function is assigned to struct t10_alua *->state_check() in
+ * core_setup_alua()
+ *
+ * Also, this function can return three different return codes to
+ * signal transport_generic_cmd_sequencer()
+ *
+ * return 1: Is used to signal LUN not accecsable, and check condition/not ready
+ * return 0: Used to signal success
+ * reutrn -1: Used to signal failure, and invalid cdb field
+ */
+static int core_alua_state_check(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ u8 *alua_ascq)
+{
+ struct se_lun *lun = SE_LUN(cmd);
+ struct se_port *port = lun->lun_sep;
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+ int out_alua_state, nonop_delay_msecs;
+
+ if (!(port))
+ return 0;
+ /*
+ * First, check for a struct se_port specific secondary ALUA target port
+ * access state: OFFLINE
+ */
+ if (atomic_read(&port->sep_tg_pt_secondary_offline)) {
+ *alua_ascq = ASCQ_04H_ALUA_OFFLINE;
+ printk(KERN_INFO "ALUA: Got secondary offline status for local"
+ " target port\n");
+ *alua_ascq = ASCQ_04H_ALUA_OFFLINE;
+ return 1;
+ }
+ /*
+ * Second, obtain the struct t10_alua_tg_pt_gp_member pointer to the
+ * ALUA target port group, to obtain current ALUA access state.
+ * Otherwise look for the underlying struct se_device association with
+ * a ALUA logical unit group.
+ */
+ tg_pt_gp_mem = port->sep_alua_tg_pt_gp_mem;
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ tg_pt_gp = tg_pt_gp_mem->tg_pt_gp;
+ out_alua_state = atomic_read(&tg_pt_gp->tg_pt_gp_alua_access_state);
+ nonop_delay_msecs = tg_pt_gp->tg_pt_gp_nonop_delay_msecs;
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ /*
+ * Process ALUA_ACCESS_STATE_ACTIVE_OPTMIZED in a seperate conditional
+ * statement so the complier knows explictly to check this case first.
+ * For the Optimized ALUA access state case, we want to process the
+ * incoming fabric cmd ASAP..
+ */
+ if (out_alua_state == ALUA_ACCESS_STATE_ACTIVE_OPTMIZED)
+ return 0;
+
+ switch (out_alua_state) {
+ case ALUA_ACCESS_STATE_ACTIVE_NON_OPTIMIZED:
+ return core_alua_state_nonoptimized(cmd, cdb,
+ nonop_delay_msecs, alua_ascq);
+ case ALUA_ACCESS_STATE_STANDBY:
+ return core_alua_state_standby(cmd, cdb, alua_ascq);
+ case ALUA_ACCESS_STATE_UNAVAILABLE:
+ return core_alua_state_unavailable(cmd, cdb, alua_ascq);
+ case ALUA_ACCESS_STATE_TRANSITION:
+ return core_alua_state_transition(cmd, cdb, alua_ascq);
+ /*
+ * OFFLINE is a secondary ALUA target port group access state, that is
+ * handled above with struct se_port->sep_tg_pt_secondary_offline=1
+ */
+ case ALUA_ACCESS_STATE_OFFLINE:
+ default:
+ printk(KERN_ERR "Unknown ALUA access state: 0x%02x\n",
+ out_alua_state);
+ return -1;
+ }
+
+ return 0;
+}
+
+/*
+ * Check implict and explict ALUA state change request.
+ */
+static int core_alua_check_transition(int state, int *primary)
+{
+ switch (state) {
+ case ALUA_ACCESS_STATE_ACTIVE_OPTMIZED:
+ case ALUA_ACCESS_STATE_ACTIVE_NON_OPTIMIZED:
+ case ALUA_ACCESS_STATE_STANDBY:
+ case ALUA_ACCESS_STATE_UNAVAILABLE:
+ /*
+ * OPTIMIZED, NON-OPTIMIZED, STANDBY and UNAVAILABLE are
+ * defined as primary target port asymmetric access states.
+ */
+ *primary = 1;
+ break;
+ case ALUA_ACCESS_STATE_OFFLINE:
+ /*
+ * OFFLINE state is defined as a secondary target port
+ * asymmetric access state.
+ */
+ *primary = 0;
+ break;
+ default:
+ printk(KERN_ERR "Unknown ALUA access state: 0x%02x\n", state);
+ return -1;
+ }
+
+ return 0;
+}
+
+static char *core_alua_dump_state(int state)
+{
+ switch (state) {
+ case ALUA_ACCESS_STATE_ACTIVE_OPTMIZED:
+ return "Active/Optimized";
+ case ALUA_ACCESS_STATE_ACTIVE_NON_OPTIMIZED:
+ return "Active/NonOptimized";
+ case ALUA_ACCESS_STATE_STANDBY:
+ return "Standby";
+ case ALUA_ACCESS_STATE_UNAVAILABLE:
+ return "Unavailable";
+ case ALUA_ACCESS_STATE_OFFLINE:
+ return "Offline";
+ default:
+ return "Unknown";
+ }
+
+ return NULL;
+}
+
+char *core_alua_dump_status(int status)
+{
+ switch (status) {
+ case ALUA_STATUS_NONE:
+ return "None";
+ case ALUA_STATUS_ALTERED_BY_EXPLICT_STPG:
+ return "Altered by Explict STPG";
+ case ALUA_STATUS_ALTERED_BY_IMPLICT_ALUA:
+ return "Altered by Implict ALUA";
+ default:
+ return "Unknown";
+ }
+
+ return NULL;
+}
+
+/*
+ * Used by fabric modules to determine when we need to delay processing
+ * for the Active/NonOptimized paths..
+ */
+int core_alua_check_nonop_delay(
+ struct se_cmd *cmd)
+{
+ if (!(cmd->se_cmd_flags & SCF_ALUA_NON_OPTIMIZED))
+ return 0;
+ if (in_interrupt())
+ return 0;
+ /*
+ * The ALUA Active/NonOptimized access state delay can be disabled
+ * in via configfs with a value of zero
+ */
+ if (!(cmd->alua_nonop_delay))
+ return 0;
+ /*
+ * struct se_cmd->alua_nonop_delay gets set by a target port group
+ * defined interval in core_alua_state_nonoptimized()
+ */
+ msleep_interruptible(cmd->alua_nonop_delay);
+ return 0;
+}
+EXPORT_SYMBOL(core_alua_check_nonop_delay);
+
+/*
+ * Called with tg_pt_gp->tg_pt_gp_md_mutex or tg_pt_gp_mem->sep_tg_pt_md_mutex
+ *
+ */
+static int core_alua_write_tpg_metadata(
+ const char *path,
+ unsigned char *md_buf,
+ u32 md_buf_len)
+{
+ mm_segment_t old_fs;
+ struct file *file;
+ struct iovec iov[1];
+ int flags = O_RDWR | O_CREAT | O_TRUNC, ret;
+
+ memset(iov, 0, sizeof(struct iovec));
+
+ file = filp_open(path, flags, 0600);
+ if (IS_ERR(file) || !file || !file->f_dentry) {
+ printk(KERN_ERR "filp_open(%s) for ALUA metadata failed\n",
+ path);
+ return -ENODEV;
+ }
+
+ iov[0].iov_base = &md_buf[0];
+ iov[0].iov_len = md_buf_len;
+
+ old_fs = get_fs();
+ set_fs(get_ds());
+ ret = vfs_writev(file, &iov[0], 1, &file->f_pos);
+ set_fs(old_fs);
+
+ if (ret < 0) {
+ printk(KERN_ERR "Error writing ALUA metadata file: %s\n", path);
+ filp_close(file, NULL);
+ return -EIO;
+ }
+ filp_close(file, NULL);
+
+ return 0;
+}
+
+/*
+ * Called with tg_pt_gp->tg_pt_gp_md_mutex held
+ */
+static int core_alua_update_tpg_primary_metadata(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ int primary_state,
+ unsigned char *md_buf)
+{
+ struct se_subsystem_dev *su_dev = tg_pt_gp->tg_pt_gp_su_dev;
+ struct t10_wwn *wwn = &su_dev->t10_wwn;
+ char path[ALUA_METADATA_PATH_LEN];
+ int len;
+
+ memset(path, 0, ALUA_METADATA_PATH_LEN);
+
+ len = snprintf(md_buf, tg_pt_gp->tg_pt_gp_md_buf_len,
+ "tg_pt_gp_id=%hu\n"
+ "alua_access_state=0x%02x\n"
+ "alua_access_status=0x%02x\n",
+ tg_pt_gp->tg_pt_gp_id, primary_state,
+ tg_pt_gp->tg_pt_gp_alua_access_status);
+
+ snprintf(path, ALUA_METADATA_PATH_LEN,
+ "/var/target/alua/tpgs_%s/%s", &wwn->unit_serial[0],
+ config_item_name(&tg_pt_gp->tg_pt_gp_group.cg_item));
+
+ return core_alua_write_tpg_metadata(path, md_buf, len);
+}
+
+static int core_alua_do_transition_tg_pt(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ struct se_port *l_port,
+ struct se_node_acl *nacl,
+ unsigned char *md_buf,
+ int new_state,
+ int explict)
+{
+ struct se_dev_entry *se_deve;
+ struct se_lun_acl *lacl;
+ struct se_port *port;
+ struct t10_alua_tg_pt_gp_member *mem;
+ int old_state = 0;
+ /*
+ * Save the old primary ALUA access state, and set the current state
+ * to ALUA_ACCESS_STATE_TRANSITION.
+ */
+ old_state = atomic_read(&tg_pt_gp->tg_pt_gp_alua_access_state);
+ atomic_set(&tg_pt_gp->tg_pt_gp_alua_access_state,
+ ALUA_ACCESS_STATE_TRANSITION);
+ tg_pt_gp->tg_pt_gp_alua_access_status = (explict) ?
+ ALUA_STATUS_ALTERED_BY_EXPLICT_STPG :
+ ALUA_STATUS_ALTERED_BY_IMPLICT_ALUA;
+ /*
+ * Check for the optional ALUA primary state transition delay
+ */
+ if (tg_pt_gp->tg_pt_gp_trans_delay_msecs != 0)
+ msleep_interruptible(tg_pt_gp->tg_pt_gp_trans_delay_msecs);
+
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ list_for_each_entry(mem, &tg_pt_gp->tg_pt_gp_mem_list,
+ tg_pt_gp_mem_list) {
+ port = mem->tg_pt;
+ /*
+ * After an implicit target port asymmetric access state
+ * change, a device server shall establish a unit attention
+ * condition for the initiator port associated with every I_T
+ * nexus with the additional sense code set to ASYMMETRIC
+ * ACCESS STATE CHAGED.
+ *
+ * After an explicit target port asymmetric access state
+ * change, a device server shall establish a unit attention
+ * condition with the additional sense code set to ASYMMETRIC
+ * ACCESS STATE CHANGED for the initiator port associated with
+ * every I_T nexus other than the I_T nexus on which the SET
+ * TARGET PORT GROUPS command
+ */
+ atomic_inc(&mem->tg_pt_gp_mem_ref_cnt);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+
+ spin_lock_bh(&port->sep_alua_lock);
+ list_for_each_entry(se_deve, &port->sep_alua_list,
+ alua_port_list) {
+ lacl = se_deve->se_lun_acl;
+ /*
+ * se_deve->se_lun_acl pointer may be NULL for a
+ * entry created without explict Node+MappedLUN ACLs
+ */
+ if (!(lacl))
+ continue;
+
+ if (explict &&
+ (nacl != NULL) && (nacl == lacl->se_lun_nacl) &&
+ (l_port != NULL) && (l_port == port))
+ continue;
+
+ core_scsi3_ua_allocate(lacl->se_lun_nacl,
+ se_deve->mapped_lun, 0x2A,
+ ASCQ_2AH_ASYMMETRIC_ACCESS_STATE_CHANGED);
+ }
+ spin_unlock_bh(&port->sep_alua_lock);
+
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ atomic_dec(&mem->tg_pt_gp_mem_ref_cnt);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+ /*
+ * Update the ALUA metadata buf that has been allocated in
+ * core_alua_do_port_transition(), this metadata will be written
+ * to struct file.
+ *
+ * Note that there is the case where we do not want to update the
+ * metadata when the saved metadata is being parsed in userspace
+ * when setting the existing port access state and access status.
+ *
+ * Also note that the failure to write out the ALUA metadata to
+ * struct file does NOT affect the actual ALUA transition.
+ */
+ if (tg_pt_gp->tg_pt_gp_write_metadata) {
+ mutex_lock(&tg_pt_gp->tg_pt_gp_md_mutex);
+ core_alua_update_tpg_primary_metadata(tg_pt_gp,
+ new_state, md_buf);
+ mutex_unlock(&tg_pt_gp->tg_pt_gp_md_mutex);
+ }
+ /*
+ * Set the current primary ALUA access state to the requested new state
+ */
+ atomic_set(&tg_pt_gp->tg_pt_gp_alua_access_state, new_state);
+
+ printk(KERN_INFO "Successful %s ALUA transition TG PT Group: %s ID: %hu"
+ " from primary access state %s to %s\n", (explict) ? "explict" :
+ "implict", config_item_name(&tg_pt_gp->tg_pt_gp_group.cg_item),
+ tg_pt_gp->tg_pt_gp_id, core_alua_dump_state(old_state),
+ core_alua_dump_state(new_state));
+
+ return 0;
+}
+
+int core_alua_do_port_transition(
+ struct t10_alua_tg_pt_gp *l_tg_pt_gp,
+ struct se_device *l_dev,
+ struct se_port *l_port,
+ struct se_node_acl *l_nacl,
+ int new_state,
+ int explict)
+{
+ struct se_device *dev;
+ struct se_port *port;
+ struct se_subsystem_dev *su_dev;
+ struct se_node_acl *nacl;
+ struct t10_alua_lu_gp *lu_gp;
+ struct t10_alua_lu_gp_member *lu_gp_mem, *local_lu_gp_mem;
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ unsigned char *md_buf;
+ int primary;
+
+ if (core_alua_check_transition(new_state, &primary) != 0)
+ return -EINVAL;
+
+ md_buf = kzalloc(l_tg_pt_gp->tg_pt_gp_md_buf_len, GFP_KERNEL);
+ if (!(md_buf)) {
+ printk("Unable to allocate buf for ALUA metadata\n");
+ return -ENOMEM;
+ }
+
+ local_lu_gp_mem = l_dev->dev_alua_lu_gp_mem;
+ spin_lock(&local_lu_gp_mem->lu_gp_mem_lock);
+ lu_gp = local_lu_gp_mem->lu_gp;
+ atomic_inc(&lu_gp->lu_gp_ref_cnt);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&local_lu_gp_mem->lu_gp_mem_lock);
+ /*
+ * For storage objects that are members of the 'default_lu_gp',
+ * we only do transition on the passed *l_tp_pt_gp, and not
+ * on all of the matching target port groups IDs in default_lu_gp.
+ */
+ if (!(lu_gp->lu_gp_id)) {
+ /*
+ * core_alua_do_transition_tg_pt() will always return
+ * success.
+ */
+ core_alua_do_transition_tg_pt(l_tg_pt_gp, l_port, l_nacl,
+ md_buf, new_state, explict);
+ atomic_dec(&lu_gp->lu_gp_ref_cnt);
+ smp_mb__after_atomic_dec();
+ kfree(md_buf);
+ return 0;
+ }
+ /*
+ * For all other LU groups aside from 'default_lu_gp', walk all of
+ * the associated storage objects looking for a matching target port
+ * group ID from the local target port group.
+ */
+ spin_lock(&lu_gp->lu_gp_lock);
+ list_for_each_entry(lu_gp_mem, &lu_gp->lu_gp_mem_list,
+ lu_gp_mem_list) {
+
+ dev = lu_gp_mem->lu_gp_mem_dev;
+ su_dev = dev->se_sub_dev;
+ atomic_inc(&lu_gp_mem->lu_gp_mem_ref_cnt);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&lu_gp->lu_gp_lock);
+
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ list_for_each_entry(tg_pt_gp,
+ &T10_ALUA(su_dev)->tg_pt_gps_list,
+ tg_pt_gp_list) {
+
+ if (!(tg_pt_gp->tg_pt_gp_valid_id))
+ continue;
+ /*
+ * If the target behavior port asymmetric access state
+ * is changed for any target port group accessiable via
+ * a logical unit within a LU group, the target port
+ * behavior group asymmetric access states for the same
+ * target port group accessible via other logical units
+ * in that LU group will also change.
+ */
+ if (l_tg_pt_gp->tg_pt_gp_id != tg_pt_gp->tg_pt_gp_id)
+ continue;
+
+ if (l_tg_pt_gp == tg_pt_gp) {
+ port = l_port;
+ nacl = l_nacl;
+ } else {
+ port = NULL;
+ nacl = NULL;
+ }
+ atomic_inc(&tg_pt_gp->tg_pt_gp_ref_cnt);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ /*
+ * core_alua_do_transition_tg_pt() will always return
+ * success.
+ */
+ core_alua_do_transition_tg_pt(tg_pt_gp, port,
+ nacl, md_buf, new_state, explict);
+
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ atomic_dec(&tg_pt_gp->tg_pt_gp_ref_cnt);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+
+ spin_lock(&lu_gp->lu_gp_lock);
+ atomic_dec(&lu_gp_mem->lu_gp_mem_ref_cnt);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&lu_gp->lu_gp_lock);
+
+ printk(KERN_INFO "Successfully processed LU Group: %s all ALUA TG PT"
+ " Group IDs: %hu %s transition to primary state: %s\n",
+ config_item_name(&lu_gp->lu_gp_group.cg_item),
+ l_tg_pt_gp->tg_pt_gp_id, (explict) ? "explict" : "implict",
+ core_alua_dump_state(new_state));
+
+ atomic_dec(&lu_gp->lu_gp_ref_cnt);
+ smp_mb__after_atomic_dec();
+ kfree(md_buf);
+ return 0;
+}
+
+/*
+ * Called with tg_pt_gp_mem->sep_tg_pt_md_mutex held
+ */
+static int core_alua_update_tpg_secondary_metadata(
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem,
+ struct se_port *port,
+ unsigned char *md_buf,
+ u32 md_buf_len)
+{
+ struct se_portal_group *se_tpg = port->sep_tpg;
+ char path[ALUA_METADATA_PATH_LEN], wwn[ALUA_SECONDARY_METADATA_WWN_LEN];
+ int len;
+
+ memset(path, 0, ALUA_METADATA_PATH_LEN);
+ memset(wwn, 0, ALUA_SECONDARY_METADATA_WWN_LEN);
+
+ len = snprintf(wwn, ALUA_SECONDARY_METADATA_WWN_LEN, "%s",
+ TPG_TFO(se_tpg)->tpg_get_wwn(se_tpg));
+
+ if (TPG_TFO(se_tpg)->tpg_get_tag != NULL)
+ snprintf(wwn+len, ALUA_SECONDARY_METADATA_WWN_LEN-len, "+%hu",
+ TPG_TFO(se_tpg)->tpg_get_tag(se_tpg));
+
+ len = snprintf(md_buf, md_buf_len, "alua_tg_pt_offline=%d\n"
+ "alua_tg_pt_status=0x%02x\n",
+ atomic_read(&port->sep_tg_pt_secondary_offline),
+ port->sep_tg_pt_secondary_stat);
+
+ snprintf(path, ALUA_METADATA_PATH_LEN, "/var/target/alua/%s/%s/lun_%u",
+ TPG_TFO(se_tpg)->get_fabric_name(), wwn,
+ port->sep_lun->unpacked_lun);
+
+ return core_alua_write_tpg_metadata(path, md_buf, len);
+}
+
+static int core_alua_set_tg_pt_secondary_state(
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem,
+ struct se_port *port,
+ int explict,
+ int offline)
+{
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ unsigned char *md_buf;
+ u32 md_buf_len;
+ int trans_delay_msecs;
+
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ tg_pt_gp = tg_pt_gp_mem->tg_pt_gp;
+ if (!(tg_pt_gp)) {
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ printk(KERN_ERR "Unable to complete secondary state"
+ " transition\n");
+ return -1;
+ }
+ trans_delay_msecs = tg_pt_gp->tg_pt_gp_trans_delay_msecs;
+ /*
+ * Set the secondary ALUA target port access state to OFFLINE
+ * or release the previously secondary state for struct se_port
+ */
+ if (offline)
+ atomic_set(&port->sep_tg_pt_secondary_offline, 1);
+ else
+ atomic_set(&port->sep_tg_pt_secondary_offline, 0);
+
+ md_buf_len = tg_pt_gp->tg_pt_gp_md_buf_len;
+ port->sep_tg_pt_secondary_stat = (explict) ?
+ ALUA_STATUS_ALTERED_BY_EXPLICT_STPG :
+ ALUA_STATUS_ALTERED_BY_IMPLICT_ALUA;
+
+ printk(KERN_INFO "Successful %s ALUA transition TG PT Group: %s ID: %hu"
+ " to secondary access state: %s\n", (explict) ? "explict" :
+ "implict", config_item_name(&tg_pt_gp->tg_pt_gp_group.cg_item),
+ tg_pt_gp->tg_pt_gp_id, (offline) ? "OFFLINE" : "ONLINE");
+
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ /*
+ * Do the optional transition delay after we set the secondary
+ * ALUA access state.
+ */
+ if (trans_delay_msecs != 0)
+ msleep_interruptible(trans_delay_msecs);
+ /*
+ * See if we need to update the ALUA fabric port metadata for
+ * secondary state and status
+ */
+ if (port->sep_tg_pt_secondary_write_md) {
+ md_buf = kzalloc(md_buf_len, GFP_KERNEL);
+ if (!(md_buf)) {
+ printk(KERN_ERR "Unable to allocate md_buf for"
+ " secondary ALUA access metadata\n");
+ return -1;
+ }
+ mutex_lock(&port->sep_tg_pt_md_mutex);
+ core_alua_update_tpg_secondary_metadata(tg_pt_gp_mem, port,
+ md_buf, md_buf_len);
+ mutex_unlock(&port->sep_tg_pt_md_mutex);
+
+ kfree(md_buf);
+ }
+
+ return 0;
+}
+
+struct t10_alua_lu_gp *
+core_alua_allocate_lu_gp(const char *name, int def_group)
+{
+ struct t10_alua_lu_gp *lu_gp;
+
+ lu_gp = kmem_cache_zalloc(t10_alua_lu_gp_cache, GFP_KERNEL);
+ if (!(lu_gp)) {
+ printk(KERN_ERR "Unable to allocate struct t10_alua_lu_gp\n");
+ return ERR_PTR(-ENOMEM);;
+ }
+ INIT_LIST_HEAD(&lu_gp->lu_gp_list);
+ INIT_LIST_HEAD(&lu_gp->lu_gp_mem_list);
+ spin_lock_init(&lu_gp->lu_gp_lock);
+ atomic_set(&lu_gp->lu_gp_ref_cnt, 0);
+
+ if (def_group) {
+ lu_gp->lu_gp_id = se_global->alua_lu_gps_counter++;;
+ lu_gp->lu_gp_valid_id = 1;
+ se_global->alua_lu_gps_count++;
+ }
+
+ return lu_gp;
+}
+
+int core_alua_set_lu_gp_id(struct t10_alua_lu_gp *lu_gp, u16 lu_gp_id)
+{
+ struct t10_alua_lu_gp *lu_gp_tmp;
+ u16 lu_gp_id_tmp;
+ /*
+ * The lu_gp->lu_gp_id may only be set once..
+ */
+ if (lu_gp->lu_gp_valid_id) {
+ printk(KERN_WARNING "ALUA LU Group already has a valid ID,"
+ " ignoring request\n");
+ return -1;
+ }
+
+ spin_lock(&se_global->lu_gps_lock);
+ if (se_global->alua_lu_gps_count == 0x0000ffff) {
+ printk(KERN_ERR "Maximum ALUA se_global->alua_lu_gps_count:"
+ " 0x0000ffff reached\n");
+ spin_unlock(&se_global->lu_gps_lock);
+ kmem_cache_free(t10_alua_lu_gp_cache, lu_gp);
+ return -1;
+ }
+again:
+ lu_gp_id_tmp = (lu_gp_id != 0) ? lu_gp_id :
+ se_global->alua_lu_gps_counter++;
+
+ list_for_each_entry(lu_gp_tmp, &se_global->g_lu_gps_list, lu_gp_list) {
+ if (lu_gp_tmp->lu_gp_id == lu_gp_id_tmp) {
+ if (!(lu_gp_id))
+ goto again;
+
+ printk(KERN_WARNING "ALUA Logical Unit Group ID: %hu"
+ " already exists, ignoring request\n",
+ lu_gp_id);
+ spin_unlock(&se_global->lu_gps_lock);
+ return -1;
+ }
+ }
+
+ lu_gp->lu_gp_id = lu_gp_id_tmp;
+ lu_gp->lu_gp_valid_id = 1;
+ list_add_tail(&lu_gp->lu_gp_list, &se_global->g_lu_gps_list);
+ se_global->alua_lu_gps_count++;
+ spin_unlock(&se_global->lu_gps_lock);
+
+ return 0;
+}
+
+static struct t10_alua_lu_gp_member *
+core_alua_allocate_lu_gp_mem(struct se_device *dev)
+{
+ struct t10_alua_lu_gp_member *lu_gp_mem;
+
+ lu_gp_mem = kmem_cache_zalloc(t10_alua_lu_gp_mem_cache, GFP_KERNEL);
+ if (!(lu_gp_mem)) {
+ printk(KERN_ERR "Unable to allocate struct t10_alua_lu_gp_member\n");
+ return ERR_PTR(-ENOMEM);
+ }
+ INIT_LIST_HEAD(&lu_gp_mem->lu_gp_mem_list);
+ spin_lock_init(&lu_gp_mem->lu_gp_mem_lock);
+ atomic_set(&lu_gp_mem->lu_gp_mem_ref_cnt, 0);
+
+ lu_gp_mem->lu_gp_mem_dev = dev;
+ dev->dev_alua_lu_gp_mem = lu_gp_mem;
+
+ return lu_gp_mem;
+}
+
+void core_alua_free_lu_gp(struct t10_alua_lu_gp *lu_gp)
+{
+ struct t10_alua_lu_gp_member *lu_gp_mem, *lu_gp_mem_tmp;
+ /*
+ * Once we have reached this point, config_item_put() has
+ * already been called from target_core_alua_drop_lu_gp().
+ *
+ * Here, we remove the *lu_gp from the global list so that
+ * no associations can be made while we are releasing
+ * struct t10_alua_lu_gp.
+ */
+ spin_lock(&se_global->lu_gps_lock);
+ atomic_set(&lu_gp->lu_gp_shutdown, 1);
+ list_del(&lu_gp->lu_gp_list);
+ se_global->alua_lu_gps_count--;
+ spin_unlock(&se_global->lu_gps_lock);
+ /*
+ * Allow struct t10_alua_lu_gp * referenced by core_alua_get_lu_gp_by_name()
+ * in target_core_configfs.c:target_core_store_alua_lu_gp() to be
+ * released with core_alua_put_lu_gp_from_name()
+ */
+ while (atomic_read(&lu_gp->lu_gp_ref_cnt))
+ cpu_relax();
+ /*
+ * Release reference to struct t10_alua_lu_gp * from all associated
+ * struct se_device.
+ */
+ spin_lock(&lu_gp->lu_gp_lock);
+ list_for_each_entry_safe(lu_gp_mem, lu_gp_mem_tmp,
+ &lu_gp->lu_gp_mem_list, lu_gp_mem_list) {
+ if (lu_gp_mem->lu_gp_assoc) {
+ list_del(&lu_gp_mem->lu_gp_mem_list);
+ lu_gp->lu_gp_members--;
+ lu_gp_mem->lu_gp_assoc = 0;
+ }
+ spin_unlock(&lu_gp->lu_gp_lock);
+ /*
+ *
+ * lu_gp_mem is assoicated with a single
+ * struct se_device->dev_alua_lu_gp_mem, and is released when
+ * struct se_device is released via core_alua_free_lu_gp_mem().
+ *
+ * If the passed lu_gp does NOT match the default_lu_gp, assume
+ * we want to re-assocate a given lu_gp_mem with default_lu_gp.
+ */
+ spin_lock(&lu_gp_mem->lu_gp_mem_lock);
+ if (lu_gp != se_global->default_lu_gp)
+ __core_alua_attach_lu_gp_mem(lu_gp_mem,
+ se_global->default_lu_gp);
+ else
+ lu_gp_mem->lu_gp = NULL;
+ spin_unlock(&lu_gp_mem->lu_gp_mem_lock);
+
+ spin_lock(&lu_gp->lu_gp_lock);
+ }
+ spin_unlock(&lu_gp->lu_gp_lock);
+
+ kmem_cache_free(t10_alua_lu_gp_cache, lu_gp);
+}
+
+void core_alua_free_lu_gp_mem(struct se_device *dev)
+{
+ struct se_subsystem_dev *su_dev = dev->se_sub_dev;
+ struct t10_alua *alua = T10_ALUA(su_dev);
+ struct t10_alua_lu_gp *lu_gp;
+ struct t10_alua_lu_gp_member *lu_gp_mem;
+
+ if (alua->alua_type != SPC3_ALUA_EMULATED)
+ return;
+
+ lu_gp_mem = dev->dev_alua_lu_gp_mem;
+ if (!(lu_gp_mem))
+ return;
+
+ while (atomic_read(&lu_gp_mem->lu_gp_mem_ref_cnt))
+ cpu_relax();
+
+ spin_lock(&lu_gp_mem->lu_gp_mem_lock);
+ lu_gp = lu_gp_mem->lu_gp;
+ if ((lu_gp)) {
+ spin_lock(&lu_gp->lu_gp_lock);
+ if (lu_gp_mem->lu_gp_assoc) {
+ list_del(&lu_gp_mem->lu_gp_mem_list);
+ lu_gp->lu_gp_members--;
+ lu_gp_mem->lu_gp_assoc = 0;
+ }
+ spin_unlock(&lu_gp->lu_gp_lock);
+ lu_gp_mem->lu_gp = NULL;
+ }
+ spin_unlock(&lu_gp_mem->lu_gp_mem_lock);
+
+ kmem_cache_free(t10_alua_lu_gp_mem_cache, lu_gp_mem);
+}
+
+struct t10_alua_lu_gp *core_alua_get_lu_gp_by_name(const char *name)
+{
+ struct t10_alua_lu_gp *lu_gp;
+ struct config_item *ci;
+
+ spin_lock(&se_global->lu_gps_lock);
+ list_for_each_entry(lu_gp, &se_global->g_lu_gps_list, lu_gp_list) {
+ if (!(lu_gp->lu_gp_valid_id))
+ continue;
+ ci = &lu_gp->lu_gp_group.cg_item;
+ if (!(strcmp(config_item_name(ci), name))) {
+ atomic_inc(&lu_gp->lu_gp_ref_cnt);
+ spin_unlock(&se_global->lu_gps_lock);
+ return lu_gp;
+ }
+ }
+ spin_unlock(&se_global->lu_gps_lock);
+
+ return NULL;
+}
+
+void core_alua_put_lu_gp_from_name(struct t10_alua_lu_gp *lu_gp)
+{
+ spin_lock(&se_global->lu_gps_lock);
+ atomic_dec(&lu_gp->lu_gp_ref_cnt);
+ spin_unlock(&se_global->lu_gps_lock);
+}
+
+/*
+ * Called with struct t10_alua_lu_gp_member->lu_gp_mem_lock
+ */
+void __core_alua_attach_lu_gp_mem(
+ struct t10_alua_lu_gp_member *lu_gp_mem,
+ struct t10_alua_lu_gp *lu_gp)
+{
+ spin_lock(&lu_gp->lu_gp_lock);
+ lu_gp_mem->lu_gp = lu_gp;
+ lu_gp_mem->lu_gp_assoc = 1;
+ list_add_tail(&lu_gp_mem->lu_gp_mem_list, &lu_gp->lu_gp_mem_list);
+ lu_gp->lu_gp_members++;
+ spin_unlock(&lu_gp->lu_gp_lock);
+}
+
+/*
+ * Called with struct t10_alua_lu_gp_member->lu_gp_mem_lock
+ */
+void __core_alua_drop_lu_gp_mem(
+ struct t10_alua_lu_gp_member *lu_gp_mem,
+ struct t10_alua_lu_gp *lu_gp)
+{
+ spin_lock(&lu_gp->lu_gp_lock);
+ list_del(&lu_gp_mem->lu_gp_mem_list);
+ lu_gp_mem->lu_gp = NULL;
+ lu_gp_mem->lu_gp_assoc = 0;
+ lu_gp->lu_gp_members--;
+ spin_unlock(&lu_gp->lu_gp_lock);
+}
+
+struct t10_alua_tg_pt_gp *core_alua_allocate_tg_pt_gp(
+ struct se_subsystem_dev *su_dev,
+ const char *name,
+ int def_group)
+{
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+
+ tg_pt_gp = kmem_cache_zalloc(t10_alua_tg_pt_gp_cache, GFP_KERNEL);
+ if (!(tg_pt_gp)) {
+ printk(KERN_ERR "Unable to allocate struct t10_alua_tg_pt_gp\n");
+ return NULL;
+ }
+ INIT_LIST_HEAD(&tg_pt_gp->tg_pt_gp_list);
+ INIT_LIST_HEAD(&tg_pt_gp->tg_pt_gp_mem_list);
+ mutex_init(&tg_pt_gp->tg_pt_gp_md_mutex);
+ spin_lock_init(&tg_pt_gp->tg_pt_gp_lock);
+ atomic_set(&tg_pt_gp->tg_pt_gp_ref_cnt, 0);
+ tg_pt_gp->tg_pt_gp_su_dev = su_dev;
+ tg_pt_gp->tg_pt_gp_md_buf_len = ALUA_MD_BUF_LEN;
+ atomic_set(&tg_pt_gp->tg_pt_gp_alua_access_state,
+ ALUA_ACCESS_STATE_ACTIVE_OPTMIZED);
+ /*
+ * Enable both explict and implict ALUA support by default
+ */
+ tg_pt_gp->tg_pt_gp_alua_access_type =
+ TPGS_EXPLICT_ALUA | TPGS_IMPLICT_ALUA;
+ /*
+ * Set the default Active/NonOptimized Delay in milliseconds
+ */
+ tg_pt_gp->tg_pt_gp_nonop_delay_msecs = ALUA_DEFAULT_NONOP_DELAY_MSECS;
+ tg_pt_gp->tg_pt_gp_trans_delay_msecs = ALUA_DEFAULT_TRANS_DELAY_MSECS;
+
+ if (def_group) {
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ tg_pt_gp->tg_pt_gp_id =
+ T10_ALUA(su_dev)->alua_tg_pt_gps_counter++;
+ tg_pt_gp->tg_pt_gp_valid_id = 1;
+ T10_ALUA(su_dev)->alua_tg_pt_gps_count++;
+ list_add_tail(&tg_pt_gp->tg_pt_gp_list,
+ &T10_ALUA(su_dev)->tg_pt_gps_list);
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ }
+
+ return tg_pt_gp;
+}
+
+int core_alua_set_tg_pt_gp_id(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ u16 tg_pt_gp_id)
+{
+ struct se_subsystem_dev *su_dev = tg_pt_gp->tg_pt_gp_su_dev;
+ struct t10_alua_tg_pt_gp *tg_pt_gp_tmp;
+ u16 tg_pt_gp_id_tmp;
+ /*
+ * The tg_pt_gp->tg_pt_gp_id may only be set once..
+ */
+ if (tg_pt_gp->tg_pt_gp_valid_id) {
+ printk(KERN_WARNING "ALUA TG PT Group already has a valid ID,"
+ " ignoring request\n");
+ return -1;
+ }
+
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ if (T10_ALUA(su_dev)->alua_tg_pt_gps_count == 0x0000ffff) {
+ printk(KERN_ERR "Maximum ALUA alua_tg_pt_gps_count:"
+ " 0x0000ffff reached\n");
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ kmem_cache_free(t10_alua_tg_pt_gp_cache, tg_pt_gp);
+ return -1;
+ }
+again:
+ tg_pt_gp_id_tmp = (tg_pt_gp_id != 0) ? tg_pt_gp_id :
+ T10_ALUA(su_dev)->alua_tg_pt_gps_counter++;
+
+ list_for_each_entry(tg_pt_gp_tmp, &T10_ALUA(su_dev)->tg_pt_gps_list,
+ tg_pt_gp_list) {
+ if (tg_pt_gp_tmp->tg_pt_gp_id == tg_pt_gp_id_tmp) {
+ if (!(tg_pt_gp_id))
+ goto again;
+
+ printk(KERN_ERR "ALUA Target Port Group ID: %hu already"
+ " exists, ignoring request\n", tg_pt_gp_id);
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ return -1;
+ }
+ }
+
+ tg_pt_gp->tg_pt_gp_id = tg_pt_gp_id_tmp;
+ tg_pt_gp->tg_pt_gp_valid_id = 1;
+ list_add_tail(&tg_pt_gp->tg_pt_gp_list,
+ &T10_ALUA(su_dev)->tg_pt_gps_list);
+ T10_ALUA(su_dev)->alua_tg_pt_gps_count++;
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+
+ return 0;
+}
+
+struct t10_alua_tg_pt_gp_member *core_alua_allocate_tg_pt_gp_mem(
+ struct se_port *port)
+{
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+
+ tg_pt_gp_mem = kmem_cache_zalloc(t10_alua_tg_pt_gp_mem_cache,
+ GFP_KERNEL);
+ if (!(tg_pt_gp_mem)) {
+ printk(KERN_ERR "Unable to allocate struct t10_alua_tg_pt_gp_member\n");
+ return ERR_PTR(-ENOMEM);
+ }
+ INIT_LIST_HEAD(&tg_pt_gp_mem->tg_pt_gp_mem_list);
+ spin_lock_init(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ atomic_set(&tg_pt_gp_mem->tg_pt_gp_mem_ref_cnt, 0);
+
+ 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;
+}
+
+void core_alua_free_tg_pt_gp(
+ struct t10_alua_tg_pt_gp *tg_pt_gp)
+{
+ struct se_subsystem_dev *su_dev = tg_pt_gp->tg_pt_gp_su_dev;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem, *tg_pt_gp_mem_tmp;
+ /*
+ * Once we have reached this point, config_item_put() has already
+ * been called from target_core_alua_drop_tg_pt_gp().
+ *
+ * Here we remove *tg_pt_gp from the global list so that
+ * no assications *OR* explict ALUA via SET_TARGET_PORT_GROUPS
+ * can be made while we are releasing struct t10_alua_tg_pt_gp.
+ */
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ list_del(&tg_pt_gp->tg_pt_gp_list);
+ T10_ALUA(su_dev)->alua_tg_pt_gps_counter--;
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ /*
+ * Allow a struct t10_alua_tg_pt_gp_member * referenced by
+ * core_alua_get_tg_pt_gp_by_name() in
+ * target_core_configfs.c:target_core_store_alua_tg_pt_gp()
+ * to be released with core_alua_put_tg_pt_gp_from_name().
+ */
+ while (atomic_read(&tg_pt_gp->tg_pt_gp_ref_cnt))
+ cpu_relax();
+ /*
+ * Release reference to struct t10_alua_tg_pt_gp from all associated
+ * struct se_port.
+ */
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ list_for_each_entry_safe(tg_pt_gp_mem, tg_pt_gp_mem_tmp,
+ &tg_pt_gp->tg_pt_gp_mem_list, tg_pt_gp_mem_list) {
+ if (tg_pt_gp_mem->tg_pt_gp_assoc) {
+ list_del(&tg_pt_gp_mem->tg_pt_gp_mem_list);
+ tg_pt_gp->tg_pt_gp_members--;
+ tg_pt_gp_mem->tg_pt_gp_assoc = 0;
+ }
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+ /*
+ * tg_pt_gp_mem is assoicated with a single
+ * se_port->sep_alua_tg_pt_gp_mem, and is released via
+ * core_alua_free_tg_pt_gp_mem().
+ *
+ * If the passed tg_pt_gp does NOT match the default_tg_pt_gp,
+ * assume we want to re-assocate a given tg_pt_gp_mem with
+ * default_tg_pt_gp.
+ */
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ if (tg_pt_gp != T10_ALUA(su_dev)->default_tg_pt_gp) {
+ __core_alua_attach_tg_pt_gp_mem(tg_pt_gp_mem,
+ T10_ALUA(su_dev)->default_tg_pt_gp);
+ } else
+ tg_pt_gp_mem->tg_pt_gp = NULL;
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ }
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+
+ kmem_cache_free(t10_alua_tg_pt_gp_cache, tg_pt_gp);
+}
+
+void core_alua_free_tg_pt_gp_mem(struct se_port *port)
+{
+ struct se_subsystem_dev *su_dev = port->sep_lun->lun_se_dev->se_sub_dev;
+ struct t10_alua *alua = T10_ALUA(su_dev);
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+
+ if (alua->alua_type != SPC3_ALUA_EMULATED)
+ return;
+
+ tg_pt_gp_mem = port->sep_alua_tg_pt_gp_mem;
+ if (!(tg_pt_gp_mem))
+ return;
+
+ while (atomic_read(&tg_pt_gp_mem->tg_pt_gp_mem_ref_cnt))
+ cpu_relax();
+
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ tg_pt_gp = tg_pt_gp_mem->tg_pt_gp;
+ if ((tg_pt_gp)) {
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ if (tg_pt_gp_mem->tg_pt_gp_assoc) {
+ list_del(&tg_pt_gp_mem->tg_pt_gp_mem_list);
+ tg_pt_gp->tg_pt_gp_members--;
+ tg_pt_gp_mem->tg_pt_gp_assoc = 0;
+ }
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+ tg_pt_gp_mem->tg_pt_gp = NULL;
+ }
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+
+ kmem_cache_free(t10_alua_tg_pt_gp_mem_cache, tg_pt_gp_mem);
+}
+
+static struct t10_alua_tg_pt_gp *core_alua_get_tg_pt_gp_by_name(
+ struct se_subsystem_dev *su_dev,
+ const char *name)
+{
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct config_item *ci;
+
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ list_for_each_entry(tg_pt_gp, &T10_ALUA(su_dev)->tg_pt_gps_list,
+ tg_pt_gp_list) {
+ if (!(tg_pt_gp->tg_pt_gp_valid_id))
+ continue;
+ ci = &tg_pt_gp->tg_pt_gp_group.cg_item;
+ if (!(strcmp(config_item_name(ci), name))) {
+ atomic_inc(&tg_pt_gp->tg_pt_gp_ref_cnt);
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ return tg_pt_gp;
+ }
+ }
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+
+ return NULL;
+}
+
+static void core_alua_put_tg_pt_gp_from_name(
+ struct t10_alua_tg_pt_gp *tg_pt_gp)
+{
+ struct se_subsystem_dev *su_dev = tg_pt_gp->tg_pt_gp_su_dev;
+
+ spin_lock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+ atomic_dec(&tg_pt_gp->tg_pt_gp_ref_cnt);
+ spin_unlock(&T10_ALUA(su_dev)->tg_pt_gps_lock);
+}
+
+/*
+ * Called with struct t10_alua_tg_pt_gp_member->tg_pt_gp_mem_lock held
+ */
+void __core_alua_attach_tg_pt_gp_mem(
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem,
+ struct t10_alua_tg_pt_gp *tg_pt_gp)
+{
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ tg_pt_gp_mem->tg_pt_gp = tg_pt_gp;
+ tg_pt_gp_mem->tg_pt_gp_assoc = 1;
+ list_add_tail(&tg_pt_gp_mem->tg_pt_gp_mem_list,
+ &tg_pt_gp->tg_pt_gp_mem_list);
+ tg_pt_gp->tg_pt_gp_members++;
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+}
+
+/*
+ * Called with struct t10_alua_tg_pt_gp_member->tg_pt_gp_mem_lock held
+ */
+static void __core_alua_drop_tg_pt_gp_mem(
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem,
+ struct t10_alua_tg_pt_gp *tg_pt_gp)
+{
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ list_del(&tg_pt_gp_mem->tg_pt_gp_mem_list);
+ tg_pt_gp_mem->tg_pt_gp = NULL;
+ tg_pt_gp_mem->tg_pt_gp_assoc = 0;
+ tg_pt_gp->tg_pt_gp_members--;
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+}
+
+ssize_t core_alua_show_tg_pt_gp_info(struct se_port *port, char *page)
+{
+ struct se_subsystem_dev *su_dev = port->sep_lun->lun_se_dev->se_sub_dev;
+ struct config_item *tg_pt_ci;
+ struct t10_alua *alua = T10_ALUA(su_dev);
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+ ssize_t len = 0;
+
+ if (alua->alua_type != SPC3_ALUA_EMULATED)
+ return len;
+
+ tg_pt_gp_mem = port->sep_alua_tg_pt_gp_mem;
+ if (!(tg_pt_gp_mem))
+ return len;
+
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ tg_pt_gp = tg_pt_gp_mem->tg_pt_gp;
+ if ((tg_pt_gp)) {
+ tg_pt_ci = &tg_pt_gp->tg_pt_gp_group.cg_item;
+ len += sprintf(page, "TG Port Alias: %s\nTG Port Group ID:"
+ " %hu\nTG Port Primary Access State: %s\nTG Port "
+ "Primary Access Status: %s\nTG Port Secondary Access"
+ " State: %s\nTG Port Secondary Access Status: %s\n",
+ config_item_name(tg_pt_ci), tg_pt_gp->tg_pt_gp_id,
+ core_alua_dump_state(atomic_read(
+ &tg_pt_gp->tg_pt_gp_alua_access_state)),
+ core_alua_dump_status(
+ tg_pt_gp->tg_pt_gp_alua_access_status),
+ (atomic_read(&port->sep_tg_pt_secondary_offline)) ?
+ "Offline" : "None",
+ core_alua_dump_status(port->sep_tg_pt_secondary_stat));
+ }
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+
+ return len;
+}
+
+ssize_t core_alua_store_tg_pt_gp_info(
+ struct se_port *port,
+ const char *page,
+ size_t count)
+{
+ struct se_portal_group *tpg;
+ struct se_lun *lun;
+ struct se_subsystem_dev *su_dev = port->sep_lun->lun_se_dev->se_sub_dev;
+ struct t10_alua_tg_pt_gp *tg_pt_gp = NULL, *tg_pt_gp_new = NULL;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+ unsigned char buf[TG_PT_GROUP_NAME_BUF];
+ int move = 0;
+
+ tpg = port->sep_tpg;
+ lun = port->sep_lun;
+
+ if (T10_ALUA(su_dev)->alua_type != SPC3_ALUA_EMULATED) {
+ printk(KERN_WARNING "SPC3_ALUA_EMULATED not enabled for"
+ " %s/tpgt_%hu/%s\n", TPG_TFO(tpg)->tpg_get_wwn(tpg),
+ TPG_TFO(tpg)->tpg_get_tag(tpg),
+ config_item_name(&lun->lun_group.cg_item));
+ return -EINVAL;
+ }
+
+ if (count > TG_PT_GROUP_NAME_BUF) {
+ printk(KERN_ERR "ALUA Target Port Group alias too large!\n");
+ return -EINVAL;
+ }
+ memset(buf, 0, TG_PT_GROUP_NAME_BUF);
+ memcpy(buf, page, count);
+ /*
+ * Any ALUA target port group alias besides "NULL" means we will be
+ * making a new group association.
+ */
+ if (strcmp(strstrip(buf), "NULL")) {
+ /*
+ * core_alua_get_tg_pt_gp_by_name() will increment reference to
+ * struct t10_alua_tg_pt_gp. This reference is released with
+ * core_alua_put_tg_pt_gp_from_name() below.
+ */
+ tg_pt_gp_new = core_alua_get_tg_pt_gp_by_name(su_dev,
+ strstrip(buf));
+ if (!(tg_pt_gp_new))
+ return -ENODEV;
+ }
+ tg_pt_gp_mem = port->sep_alua_tg_pt_gp_mem;
+ if (!(tg_pt_gp_mem)) {
+ if (tg_pt_gp_new)
+ core_alua_put_tg_pt_gp_from_name(tg_pt_gp_new);
+ printk(KERN_ERR "NULL struct se_port->sep_alua_tg_pt_gp_mem pointer\n");
+ return -EINVAL;
+ }
+
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ tg_pt_gp = tg_pt_gp_mem->tg_pt_gp;
+ if ((tg_pt_gp)) {
+ /*
+ * Clearing an existing tg_pt_gp association, and replacing
+ * with the default_tg_pt_gp.
+ */
+ if (!(tg_pt_gp_new)) {
+ printk(KERN_INFO "Target_Core_ConfigFS: Moving"
+ " %s/tpgt_%hu/%s from ALUA Target Port Group:"
+ " alua/%s, ID: %hu back to"
+ " default_tg_pt_gp\n",
+ TPG_TFO(tpg)->tpg_get_wwn(tpg),
+ TPG_TFO(tpg)->tpg_get_tag(tpg),
+ config_item_name(&lun->lun_group.cg_item),
+ config_item_name(
+ &tg_pt_gp->tg_pt_gp_group.cg_item),
+ tg_pt_gp->tg_pt_gp_id);
+
+ __core_alua_drop_tg_pt_gp_mem(tg_pt_gp_mem, tg_pt_gp);
+ __core_alua_attach_tg_pt_gp_mem(tg_pt_gp_mem,
+ T10_ALUA(su_dev)->default_tg_pt_gp);
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+
+ return count;
+ }
+ /*
+ * Removing existing association of tg_pt_gp_mem with tg_pt_gp
+ */
+ __core_alua_drop_tg_pt_gp_mem(tg_pt_gp_mem, tg_pt_gp);
+ move = 1;
+ }
+ /*
+ * Associate tg_pt_gp_mem with tg_pt_gp_new.
+ */
+ __core_alua_attach_tg_pt_gp_mem(tg_pt_gp_mem, tg_pt_gp_new);
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ printk(KERN_INFO "Target_Core_ConfigFS: %s %s/tpgt_%hu/%s to ALUA"
+ " Target Port Group: alua/%s, ID: %hu\n", (move) ?
+ "Moving" : "Adding", TPG_TFO(tpg)->tpg_get_wwn(tpg),
+ TPG_TFO(tpg)->tpg_get_tag(tpg),
+ config_item_name(&lun->lun_group.cg_item),
+ config_item_name(&tg_pt_gp_new->tg_pt_gp_group.cg_item),
+ tg_pt_gp_new->tg_pt_gp_id);
+
+ core_alua_put_tg_pt_gp_from_name(tg_pt_gp_new);
+ return count;
+}
+
+ssize_t core_alua_show_access_type(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ if ((tg_pt_gp->tg_pt_gp_alua_access_type & TPGS_EXPLICT_ALUA) &&
+ (tg_pt_gp->tg_pt_gp_alua_access_type & TPGS_IMPLICT_ALUA))
+ return sprintf(page, "Implict and Explict\n");
+ else if (tg_pt_gp->tg_pt_gp_alua_access_type & TPGS_IMPLICT_ALUA)
+ return sprintf(page, "Implict\n");
+ else if (tg_pt_gp->tg_pt_gp_alua_access_type & TPGS_EXPLICT_ALUA)
+ return sprintf(page, "Explict\n");
+ else
+ return sprintf(page, "None\n");
+}
+
+ssize_t core_alua_store_access_type(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ unsigned long tmp;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract alua_access_type\n");
+ return -EINVAL;
+ }
+ if ((tmp != 0) && (tmp != 1) && (tmp != 2) && (tmp != 3)) {
+ printk(KERN_ERR "Illegal value for alua_access_type:"
+ " %lu\n", tmp);
+ return -EINVAL;
+ }
+ if (tmp == 3)
+ tg_pt_gp->tg_pt_gp_alua_access_type =
+ TPGS_IMPLICT_ALUA | TPGS_EXPLICT_ALUA;
+ else if (tmp == 2)
+ tg_pt_gp->tg_pt_gp_alua_access_type = TPGS_EXPLICT_ALUA;
+ else if (tmp == 1)
+ tg_pt_gp->tg_pt_gp_alua_access_type = TPGS_IMPLICT_ALUA;
+ else
+ tg_pt_gp->tg_pt_gp_alua_access_type = 0;
+
+ return count;
+}
+
+ssize_t core_alua_show_nonop_delay_msecs(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return sprintf(page, "%d\n", tg_pt_gp->tg_pt_gp_nonop_delay_msecs);
+}
+
+ssize_t core_alua_store_nonop_delay_msecs(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ unsigned long tmp;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract nonop_delay_msecs\n");
+ return -EINVAL;
+ }
+ if (tmp > ALUA_MAX_NONOP_DELAY_MSECS) {
+ printk(KERN_ERR "Passed nonop_delay_msecs: %lu, exceeds"
+ " ALUA_MAX_NONOP_DELAY_MSECS: %d\n", tmp,
+ ALUA_MAX_NONOP_DELAY_MSECS);
+ return -EINVAL;
+ }
+ tg_pt_gp->tg_pt_gp_nonop_delay_msecs = (int)tmp;
+
+ return count;
+}
+
+ssize_t core_alua_show_trans_delay_msecs(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return sprintf(page, "%d\n", tg_pt_gp->tg_pt_gp_trans_delay_msecs);
+}
+
+ssize_t core_alua_store_trans_delay_msecs(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ unsigned long tmp;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract trans_delay_msecs\n");
+ return -EINVAL;
+ }
+ if (tmp > ALUA_MAX_TRANS_DELAY_MSECS) {
+ printk(KERN_ERR "Passed trans_delay_msecs: %lu, exceeds"
+ " ALUA_MAX_TRANS_DELAY_MSECS: %d\n", tmp,
+ ALUA_MAX_TRANS_DELAY_MSECS);
+ return -EINVAL;
+ }
+ tg_pt_gp->tg_pt_gp_trans_delay_msecs = (int)tmp;
+
+ return count;
+}
+
+ssize_t core_alua_show_preferred_bit(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return sprintf(page, "%d\n", tg_pt_gp->tg_pt_gp_pref);
+}
+
+ssize_t core_alua_store_preferred_bit(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ unsigned long tmp;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract preferred ALUA value\n");
+ return -EINVAL;
+ }
+ if ((tmp != 0) && (tmp != 1)) {
+ printk(KERN_ERR "Illegal value for preferred ALUA: %lu\n", tmp);
+ return -EINVAL;
+ }
+ tg_pt_gp->tg_pt_gp_pref = (int)tmp;
+
+ return count;
+}
+
+ssize_t core_alua_show_offline_bit(struct se_lun *lun, char *page)
+{
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return sprintf(page, "%d\n",
+ atomic_read(&lun->lun_sep->sep_tg_pt_secondary_offline));
+}
+
+ssize_t core_alua_store_offline_bit(
+ struct se_lun *lun,
+ const char *page,
+ size_t count)
+{
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+ unsigned long tmp;
+ int ret;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract alua_tg_pt_offline value\n");
+ return -EINVAL;
+ }
+ if ((tmp != 0) && (tmp != 1)) {
+ printk(KERN_ERR "Illegal value for alua_tg_pt_offline: %lu\n",
+ tmp);
+ return -EINVAL;
+ }
+ tg_pt_gp_mem = lun->lun_sep->sep_alua_tg_pt_gp_mem;
+ if (!(tg_pt_gp_mem)) {
+ printk(KERN_ERR "Unable to locate *tg_pt_gp_mem\n");
+ return -EINVAL;
+ }
+
+ ret = core_alua_set_tg_pt_secondary_state(tg_pt_gp_mem,
+ lun->lun_sep, 0, (int)tmp);
+ if (ret < 0)
+ return -EINVAL;
+
+ return count;
+}
+
+ssize_t core_alua_show_secondary_status(
+ struct se_lun *lun,
+ char *page)
+{
+ return sprintf(page, "%d\n", lun->lun_sep->sep_tg_pt_secondary_stat);
+}
+
+ssize_t core_alua_store_secondary_status(
+ struct se_lun *lun,
+ const char *page,
+ size_t count)
+{
+ unsigned long tmp;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract alua_tg_pt_status\n");
+ return -EINVAL;
+ }
+ if ((tmp != ALUA_STATUS_NONE) &&
+ (tmp != ALUA_STATUS_ALTERED_BY_EXPLICT_STPG) &&
+ (tmp != ALUA_STATUS_ALTERED_BY_IMPLICT_ALUA)) {
+ printk(KERN_ERR "Illegal value for alua_tg_pt_status: %lu\n",
+ tmp);
+ return -EINVAL;
+ }
+ lun->lun_sep->sep_tg_pt_secondary_stat = (int)tmp;
+
+ return count;
+}
+
+ssize_t core_alua_show_secondary_write_metadata(
+ struct se_lun *lun,
+ char *page)
+{
+ return sprintf(page, "%d\n",
+ lun->lun_sep->sep_tg_pt_secondary_write_md);
+}
+
+ssize_t core_alua_store_secondary_write_metadata(
+ struct se_lun *lun,
+ const char *page,
+ size_t count)
+{
+ unsigned long tmp;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract alua_tg_pt_write_md\n");
+ return -EINVAL;
+ }
+ if ((tmp != 0) && (tmp != 1)) {
+ printk(KERN_ERR "Illegal value for alua_tg_pt_write_md:"
+ " %lu\n", tmp);
+ return -EINVAL;
+ }
+ lun->lun_sep->sep_tg_pt_secondary_write_md = (int)tmp;
+
+ return count;
+}
+
+int core_setup_alua(struct se_device *dev, int force_pt)
+{
+ struct se_subsystem_dev *su_dev = dev->se_sub_dev;
+ struct t10_alua *alua = T10_ALUA(su_dev);
+ struct t10_alua_lu_gp_member *lu_gp_mem;
+ /*
+ * If this device is from Target_Core_Mod/pSCSI, use the ALUA logic
+ * of the Underlying SCSI hardware. In Linux/SCSI terms, this can
+ * cause a problem because libata and some SATA RAID HBAs appear
+ * under Linux/SCSI, but emulate SCSI logic themselves.
+ */
+ if (((TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV) &&
+ !(DEV_ATTRIB(dev)->emulate_alua)) || force_pt) {
+ alua->alua_type = SPC_ALUA_PASSTHROUGH;
+ alua->alua_state_check = &core_alua_state_check_nop;
+ printk(KERN_INFO "%s: Using SPC_ALUA_PASSTHROUGH, no ALUA"
+ " emulation\n", TRANSPORT(dev)->name);
+ return 0;
+ }
+ /*
+ * If SPC-3 or above is reported by real or emulated struct se_device,
+ * use emulated ALUA.
+ */
+ if (TRANSPORT(dev)->get_device_rev(dev) >= SCSI_3) {
+ printk(KERN_INFO "%s: Enabling ALUA Emulation for SPC-3"
+ " device\n", TRANSPORT(dev)->name);
+ /*
+ * Assoicate this struct se_device with the default ALUA
+ * LUN Group.
+ */
+ lu_gp_mem = core_alua_allocate_lu_gp_mem(dev);
+ if (IS_ERR(lu_gp_mem) || !lu_gp_mem)
+ return -1;
+
+ alua->alua_type = SPC3_ALUA_EMULATED;
+ alua->alua_state_check = &core_alua_state_check;
+ spin_lock(&lu_gp_mem->lu_gp_mem_lock);
+ __core_alua_attach_lu_gp_mem(lu_gp_mem,
+ se_global->default_lu_gp);
+ spin_unlock(&lu_gp_mem->lu_gp_mem_lock);
+
+ printk(KERN_INFO "%s: Adding to default ALUA LU Group:"
+ " core/alua/lu_gps/default_lu_gp\n",
+ TRANSPORT(dev)->name);
+ } else {
+ alua->alua_type = SPC2_ALUA_DISABLED;
+ alua->alua_state_check = &core_alua_state_check_nop;
+ printk(KERN_INFO "%s: Disabling ALUA Emulation for SPC-2"
+ " device\n", TRANSPORT(dev)->name);
+ }
+
+ return 0;
+}
diff --git a/drivers/target/target_core_alua.h b/drivers/target/target_core_alua.h
new file mode 100644
index 0000000..c86f97a
--- /dev/null
+++ b/drivers/target/target_core_alua.h
@@ -0,0 +1,126 @@
+#ifndef TARGET_CORE_ALUA_H
+#define TARGET_CORE_ALUA_H
+
+/*
+ * INQUIRY response data, TPGS Field
+ *
+ * from spc4r17 section 6.4.2 Table 135
+ */
+#define TPGS_NO_ALUA 0x00
+#define TPGS_IMPLICT_ALUA 0x10
+#define TPGS_EXPLICT_ALUA 0x20
+
+/*
+ * ASYMMETRIC ACCESS STATE field
+ *
+ * from spc4r17 section 6.27 Table 245
+ */
+#define ALUA_ACCESS_STATE_ACTIVE_OPTMIZED 0x0
+#define ALUA_ACCESS_STATE_ACTIVE_NON_OPTIMIZED 0x1
+#define ALUA_ACCESS_STATE_STANDBY 0x2
+#define ALUA_ACCESS_STATE_UNAVAILABLE 0x3
+#define ALUA_ACCESS_STATE_OFFLINE 0xe
+#define ALUA_ACCESS_STATE_TRANSITION 0xf
+
+/*
+ * REPORT_TARGET_PORT_GROUP STATUS CODE
+ *
+ * from spc4r17 section 6.27 Table 246
+ */
+#define ALUA_STATUS_NONE 0x00
+#define ALUA_STATUS_ALTERED_BY_EXPLICT_STPG 0x01
+#define ALUA_STATUS_ALTERED_BY_IMPLICT_ALUA 0x02
+
+/*
+ * From spc4r17, Table D.1: ASC and ASCQ Assignement
+ */
+#define ASCQ_04H_ALUA_STATE_TRANSITION 0x0a
+#define ASCQ_04H_ALUA_TG_PT_STANDBY 0x0b
+#define ASCQ_04H_ALUA_TG_PT_UNAVAILABLE 0x0c
+#define ASCQ_04H_ALUA_OFFLINE 0x12
+
+/*
+ * Used as the default for Active/NonOptimized delay (in milliseconds)
+ * This can also be changed via configfs on a per target port group basis..
+ */
+#define ALUA_DEFAULT_NONOP_DELAY_MSECS 100
+#define ALUA_MAX_NONOP_DELAY_MSECS 10000 /* 10 seconds */
+/*
+ * Used for implict and explict ALUA transitional delay, that is disabled
+ * by default, and is intended to be used for debugging client side ALUA code.
+ */
+#define ALUA_DEFAULT_TRANS_DELAY_MSECS 0
+#define ALUA_MAX_TRANS_DELAY_MSECS 30000 /* 30 seconds */
+/*
+ * Used by core_alua_update_tpg_primary_metadata() and
+ * core_alua_update_tpg_secondary_metadata()
+ */
+#define ALUA_METADATA_PATH_LEN 512
+/*
+ * Used by core_alua_update_tpg_secondary_metadata()
+ */
+#define ALUA_SECONDARY_METADATA_WWN_LEN 256
+
+extern struct kmem_cache *t10_alua_lu_gp_cache;
+extern struct kmem_cache *t10_alua_lu_gp_mem_cache;
+extern struct kmem_cache *t10_alua_tg_pt_gp_cache;
+extern struct kmem_cache *t10_alua_tg_pt_gp_mem_cache;
+
+extern int core_emulate_report_target_port_groups(struct se_cmd *);
+extern int core_emulate_set_target_port_groups(struct se_cmd *);
+extern int core_alua_check_nonop_delay(struct se_cmd *);
+extern int core_alua_do_port_transition(struct t10_alua_tg_pt_gp *,
+ struct se_device *, struct se_port *,
+ struct se_node_acl *, int, int);
+extern char *core_alua_dump_status(int);
+extern struct t10_alua_lu_gp *core_alua_allocate_lu_gp(const char *, int);
+extern int core_alua_set_lu_gp_id(struct t10_alua_lu_gp *, u16);
+extern void core_alua_free_lu_gp(struct t10_alua_lu_gp *);
+extern void core_alua_free_lu_gp_mem(struct se_device *);
+extern struct t10_alua_lu_gp *core_alua_get_lu_gp_by_name(const char *);
+extern void core_alua_put_lu_gp_from_name(struct t10_alua_lu_gp *);
+extern void __core_alua_attach_lu_gp_mem(struct t10_alua_lu_gp_member *,
+ struct t10_alua_lu_gp *);
+extern void __core_alua_drop_lu_gp_mem(struct t10_alua_lu_gp_member *,
+ struct t10_alua_lu_gp *);
+extern void core_alua_drop_lu_gp_dev(struct se_device *);
+extern struct t10_alua_tg_pt_gp *core_alua_allocate_tg_pt_gp(
+ struct se_subsystem_dev *, const char *, int);
+extern int core_alua_set_tg_pt_gp_id(struct t10_alua_tg_pt_gp *, u16);
+extern struct t10_alua_tg_pt_gp_member *core_alua_allocate_tg_pt_gp_mem(
+ struct se_port *);
+extern void core_alua_free_tg_pt_gp(struct t10_alua_tg_pt_gp *);
+extern void core_alua_free_tg_pt_gp_mem(struct se_port *);
+extern void __core_alua_attach_tg_pt_gp_mem(struct t10_alua_tg_pt_gp_member *,
+ struct t10_alua_tg_pt_gp *);
+extern ssize_t core_alua_show_tg_pt_gp_info(struct se_port *, char *);
+extern ssize_t core_alua_store_tg_pt_gp_info(struct se_port *, const char *,
+ size_t);
+extern ssize_t core_alua_show_access_type(struct t10_alua_tg_pt_gp *, char *);
+extern ssize_t core_alua_store_access_type(struct t10_alua_tg_pt_gp *,
+ const char *, size_t);
+extern ssize_t core_alua_show_nonop_delay_msecs(struct t10_alua_tg_pt_gp *,
+ char *);
+extern ssize_t core_alua_store_nonop_delay_msecs(struct t10_alua_tg_pt_gp *,
+ const char *, size_t);
+extern ssize_t core_alua_show_trans_delay_msecs(struct t10_alua_tg_pt_gp *,
+ char *);
+extern ssize_t core_alua_store_trans_delay_msecs(struct t10_alua_tg_pt_gp *,
+ const char *, size_t);
+extern ssize_t core_alua_show_preferred_bit(struct t10_alua_tg_pt_gp *,
+ char *);
+extern ssize_t core_alua_store_preferred_bit(struct t10_alua_tg_pt_gp *,
+ const char *, size_t);
+extern ssize_t core_alua_show_offline_bit(struct se_lun *, char *);
+extern ssize_t core_alua_store_offline_bit(struct se_lun *, const char *,
+ size_t);
+extern ssize_t core_alua_show_secondary_status(struct se_lun *, char *);
+extern ssize_t core_alua_store_secondary_status(struct se_lun *,
+ const char *, size_t);
+extern ssize_t core_alua_show_secondary_write_metadata(struct se_lun *,
+ char *);
+extern ssize_t core_alua_store_secondary_write_metadata(struct se_lun *,
+ const char *, size_t);
+extern int core_setup_alua(struct se_device *, int);
+
+#endif /* TARGET_CORE_ALUA_H */
diff --git a/drivers/target/target_core_cdb.c b/drivers/target/target_core_cdb.c
new file mode 100644
index 0000000..366080b
--- /dev/null
+++ b/drivers/target/target_core_cdb.c
@@ -0,0 +1,1131 @@
+/*
+ * CDB emulation for non-READ/WRITE commands.
+ *
+ * Copyright (c) 2002, 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+#include <asm/unaligned.h>
+#include <scsi/scsi.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include "target_core_ua.h"
+
+static void
+target_fill_alua_data(struct se_port *port, unsigned char *buf)
+{
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+
+ /*
+ * Set SCCS for MAINTENANCE_IN + REPORT_TARGET_PORT_GROUPS.
+ */
+ buf[5] = 0x80;
+
+ /*
+ * Set TPGS field for explict and/or implict ALUA access type
+ * and opteration.
+ *
+ * See spc4r17 section 6.4.2 Table 135
+ */
+ if (!port)
+ return;
+ tg_pt_gp_mem = port->sep_alua_tg_pt_gp_mem;
+ if (!tg_pt_gp_mem)
+ return;
+
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ tg_pt_gp = tg_pt_gp_mem->tg_pt_gp;
+ if (tg_pt_gp)
+ buf[5] |= tg_pt_gp->tg_pt_gp_alua_access_type;
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+}
+
+static int
+target_emulate_inquiry_std(struct se_cmd *cmd)
+{
+ struct se_lun *lun = SE_LUN(cmd);
+ struct se_device *dev = SE_DEV(cmd);
+ unsigned char *buf = cmd->t_task->t_task_buf;
+
+ /*
+ * Make sure we at least have 6 bytes of INQUIRY response
+ * payload going back for EVPD=0
+ */
+ if (cmd->data_length < 6) {
+ printk(KERN_ERR "SCSI Inquiry payload length: %u"
+ " too small for EVPD=0\n", cmd->data_length);
+ return -1;
+ }
+
+ buf[0] = dev->transport->get_device_type(dev);
+ if (buf[0] == TYPE_TAPE)
+ buf[1] = 0x80;
+ buf[2] = dev->transport->get_device_rev(dev);
+
+ /*
+ * Enable SCCS and TPGS fields for Emulated ALUA
+ */
+ if (T10_ALUA(dev->se_sub_dev)->alua_type == SPC3_ALUA_EMULATED)
+ target_fill_alua_data(lun->lun_sep, buf);
+
+ if (cmd->data_length < 8) {
+ buf[4] = 1; /* Set additional length to 1 */
+ return 0;
+ }
+
+ buf[7] = 0x32; /* Sync=1 and CmdQue=1 */
+
+ /*
+ * Do not include vendor, product, reversion info in INQUIRY
+ * response payload for cdbs with a small allocation length.
+ */
+ if (cmd->data_length < 36) {
+ buf[4] = 3; /* Set additional length to 3 */
+ return 0;
+ }
+
+ snprintf((unsigned char *)&buf[8], 8, "LIO-ORG");
+ snprintf((unsigned char *)&buf[16], 16, "%s",
+ &DEV_T10_WWN(dev)->model[0]);
+ snprintf((unsigned char *)&buf[32], 4, "%s",
+ &DEV_T10_WWN(dev)->revision[0]);
+ buf[4] = 31; /* Set additional length to 31 */
+ return 0;
+}
+
+/* supported vital product data pages */
+static int
+target_emulate_evpd_00(struct se_cmd *cmd, unsigned char *buf)
+{
+ buf[1] = 0x00;
+ if (cmd->data_length < 8)
+ return 0;
+
+ buf[4] = 0x0;
+ /*
+ * Only report the INQUIRY EVPD=1 pages after a valid NAA
+ * Registered Extended LUN WWN has been set via ConfigFS
+ * during device creation/restart.
+ */
+ if (SE_DEV(cmd)->se_sub_dev->su_dev_flags &
+ SDF_EMULATED_VPD_UNIT_SERIAL) {
+ buf[3] = 3;
+ buf[5] = 0x80;
+ buf[6] = 0x83;
+ buf[7] = 0x86;
+ }
+
+ return 0;
+}
+
+/* unit serial number */
+static int
+target_emulate_evpd_80(struct se_cmd *cmd, unsigned char *buf)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ u16 len = 0;
+
+ buf[1] = 0x80;
+ if (dev->se_sub_dev->su_dev_flags &
+ SDF_EMULATED_VPD_UNIT_SERIAL) {
+ u32 unit_serial_len;
+
+ unit_serial_len =
+ strlen(&DEV_T10_WWN(dev)->unit_serial[0]);
+ unit_serial_len++; /* For NULL Terminator */
+
+ if (((len + 4) + unit_serial_len) > cmd->data_length) {
+ len += unit_serial_len;
+ buf[2] = ((len >> 8) & 0xff);
+ buf[3] = (len & 0xff);
+ return 0;
+ }
+ len += sprintf((unsigned char *)&buf[4], "%s",
+ &DEV_T10_WWN(dev)->unit_serial[0]);
+ len++; /* Extra Byte for NULL Terminator */
+ buf[3] = len;
+ }
+ return 0;
+}
+
+/*
+ * Device identification VPD, for a complete list of
+ * DESIGNATOR TYPEs see spc4r17 Table 459.
+ */
+static int
+target_emulate_evpd_83(struct se_cmd *cmd, unsigned char *buf)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_lun *lun = SE_LUN(cmd);
+ struct se_port *port = NULL;
+ struct se_portal_group *tpg = NULL;
+ struct t10_alua_lu_gp_member *lu_gp_mem;
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+ unsigned char binary, binary_new;
+ unsigned char *prod = &DEV_T10_WWN(dev)->model[0];
+ u32 prod_len;
+ u32 unit_serial_len, off = 0;
+ int i;
+ u16 len = 0, id_len;
+
+ buf[1] = 0x83;
+ off = 4;
+
+ /*
+ * NAA IEEE Registered Extended Assigned designator format, see
+ * spc4r17 section 7.7.3.6.5
+ *
+ * We depend upon a target_core_mod/ConfigFS provided
+ * /sys/kernel/config/target/core/$HBA/$DEV/wwn/vpd_unit_serial
+ * value in order to return the NAA id.
+ */
+ if (!(dev->se_sub_dev->su_dev_flags & SDF_EMULATED_VPD_UNIT_SERIAL))
+ goto check_t10_vend_desc;
+
+ if (off + 20 > cmd->data_length)
+ goto check_t10_vend_desc;
+
+ /* CODE SET == Binary */
+ buf[off++] = 0x1;
+
+ /* Set ASSOICATION == addressed logical unit: 0)b */
+ buf[off] = 0x00;
+
+ /* Identifier/Designator type == NAA identifier */
+ buf[off++] = 0x3;
+ off++;
+
+ /* Identifier/Designator length */
+ buf[off++] = 0x10;
+
+ /*
+ * Start NAA IEEE Registered Extended Identifier/Designator
+ */
+ buf[off++] = (0x6 << 4);
+
+ /*
+ * Use OpenFabrics IEEE Company ID: 00 14 05
+ */
+ buf[off++] = 0x01;
+ buf[off++] = 0x40;
+ buf[off] = (0x5 << 4);
+
+ /*
+ * Return ConfigFS Unit Serial Number information for
+ * VENDOR_SPECIFIC_IDENTIFIER and
+ * VENDOR_SPECIFIC_IDENTIFIER_EXTENTION
+ */
+ binary = transport_asciihex_to_binaryhex(
+ &DEV_T10_WWN(dev)->unit_serial[0]);
+ buf[off++] |= (binary & 0xf0) >> 4;
+ for (i = 0; i < 24; i += 2) {
+ binary_new = transport_asciihex_to_binaryhex(
+ &DEV_T10_WWN(dev)->unit_serial[i+2]);
+ buf[off] = (binary & 0x0f) << 4;
+ buf[off++] |= (binary_new & 0xf0) >> 4;
+ binary = binary_new;
+ }
+ len = 20;
+ off = (len + 4);
+
+check_t10_vend_desc:
+ /*
+ * T10 Vendor Identifier Page, see spc4r17 section 7.7.3.4
+ */
+ id_len = 8; /* For Vendor field */
+ prod_len = 4; /* For VPD Header */
+ prod_len += 8; /* For Vendor field */
+ prod_len += strlen(prod);
+ prod_len++; /* For : */
+
+ if (dev->se_sub_dev->su_dev_flags &
+ SDF_EMULATED_VPD_UNIT_SERIAL) {
+ unit_serial_len =
+ strlen(&DEV_T10_WWN(dev)->unit_serial[0]);
+ unit_serial_len++; /* For NULL Terminator */
+
+ if ((len + (id_len + 4) +
+ (prod_len + unit_serial_len)) >
+ cmd->data_length) {
+ len += (prod_len + unit_serial_len);
+ goto check_port;
+ }
+ id_len += sprintf((unsigned char *)&buf[off+12],
+ "%s:%s", prod,
+ &DEV_T10_WWN(dev)->unit_serial[0]);
+ }
+ buf[off] = 0x2; /* ASCII */
+ buf[off+1] = 0x1; /* T10 Vendor ID */
+ buf[off+2] = 0x0;
+ memcpy((unsigned char *)&buf[off+4], "LIO-ORG", 8);
+ /* Extra Byte for NULL Terminator */
+ id_len++;
+ /* Identifier Length */
+ buf[off+3] = id_len;
+ /* Header size for Designation descriptor */
+ len += (id_len + 4);
+ off += (id_len + 4);
+ /*
+ * struct se_port is only set for INQUIRY VPD=1 through $FABRIC_MOD
+ */
+check_port:
+ port = lun->lun_sep;
+ if (port) {
+ struct t10_alua_lu_gp *lu_gp;
+ u32 padding, scsi_name_len;
+ u16 lu_gp_id = 0;
+ u16 tg_pt_gp_id = 0;
+ u16 tpgt;
+
+ tpg = port->sep_tpg;
+ /*
+ * Relative target port identifer, see spc4r17
+ * section 7.7.3.7
+ *
+ * Get the PROTOCOL IDENTIFIER as defined by spc4r17
+ * section 7.5.1 Table 362
+ */
+ if (((len + 4) + 8) > cmd->data_length) {
+ len += 8;
+ goto check_tpgi;
+ }
+ buf[off] =
+ (TPG_TFO(tpg)->get_fabric_proto_ident(tpg) << 4);
+ buf[off++] |= 0x1; /* CODE SET == Binary */
+ buf[off] = 0x80; /* Set PIV=1 */
+ /* Set ASSOICATION == target port: 01b */
+ buf[off] |= 0x10;
+ /* DESIGNATOR TYPE == Relative target port identifer */
+ buf[off++] |= 0x4;
+ off++; /* Skip over Reserved */
+ buf[off++] = 4; /* DESIGNATOR LENGTH */
+ /* Skip over Obsolete field in RTPI payload
+ * in Table 472 */
+ off += 2;
+ buf[off++] = ((port->sep_rtpi >> 8) & 0xff);
+ buf[off++] = (port->sep_rtpi & 0xff);
+ len += 8; /* Header size + Designation descriptor */
+ /*
+ * Target port group identifier, see spc4r17
+ * section 7.7.3.8
+ *
+ * Get the PROTOCOL IDENTIFIER as defined by spc4r17
+ * section 7.5.1 Table 362
+ */
+check_tpgi:
+ if (T10_ALUA(dev->se_sub_dev)->alua_type !=
+ SPC3_ALUA_EMULATED)
+ goto check_scsi_name;
+
+ if (((len + 4) + 8) > cmd->data_length) {
+ len += 8;
+ goto check_lu_gp;
+ }
+ tg_pt_gp_mem = port->sep_alua_tg_pt_gp_mem;
+ if (!tg_pt_gp_mem)
+ goto check_lu_gp;
+
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ tg_pt_gp = tg_pt_gp_mem->tg_pt_gp;
+ if (!(tg_pt_gp)) {
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ goto check_lu_gp;
+ }
+ tg_pt_gp_id = tg_pt_gp->tg_pt_gp_id;
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+
+ buf[off] =
+ (TPG_TFO(tpg)->get_fabric_proto_ident(tpg) << 4);
+ buf[off++] |= 0x1; /* CODE SET == Binary */
+ buf[off] = 0x80; /* Set PIV=1 */
+ /* Set ASSOICATION == target port: 01b */
+ buf[off] |= 0x10;
+ /* DESIGNATOR TYPE == Target port group identifier */
+ buf[off++] |= 0x5;
+ off++; /* Skip over Reserved */
+ buf[off++] = 4; /* DESIGNATOR LENGTH */
+ off += 2; /* Skip over Reserved Field */
+ buf[off++] = ((tg_pt_gp_id >> 8) & 0xff);
+ buf[off++] = (tg_pt_gp_id & 0xff);
+ len += 8; /* Header size + Designation descriptor */
+ /*
+ * Logical Unit Group identifier, see spc4r17
+ * section 7.7.3.8
+ */
+check_lu_gp:
+ if (((len + 4) + 8) > cmd->data_length) {
+ len += 8;
+ goto check_scsi_name;
+ }
+ lu_gp_mem = dev->dev_alua_lu_gp_mem;
+ if (!(lu_gp_mem))
+ goto check_scsi_name;
+
+ spin_lock(&lu_gp_mem->lu_gp_mem_lock);
+ lu_gp = lu_gp_mem->lu_gp;
+ if (!(lu_gp)) {
+ spin_unlock(&lu_gp_mem->lu_gp_mem_lock);
+ goto check_scsi_name;
+ }
+ lu_gp_id = lu_gp->lu_gp_id;
+ spin_unlock(&lu_gp_mem->lu_gp_mem_lock);
+
+ buf[off++] |= 0x1; /* CODE SET == Binary */
+ /* DESIGNATOR TYPE == Logical Unit Group identifier */
+ buf[off++] |= 0x6;
+ off++; /* Skip over Reserved */
+ buf[off++] = 4; /* DESIGNATOR LENGTH */
+ off += 2; /* Skip over Reserved Field */
+ buf[off++] = ((lu_gp_id >> 8) & 0xff);
+ buf[off++] = (lu_gp_id & 0xff);
+ len += 8; /* Header size + Designation descriptor */
+ /*
+ * SCSI name string designator, see spc4r17
+ * section 7.7.3.11
+ *
+ * Get the PROTOCOL IDENTIFIER as defined by spc4r17
+ * section 7.5.1 Table 362
+ */
+check_scsi_name:
+ scsi_name_len = strlen(TPG_TFO(tpg)->tpg_get_wwn(tpg));
+ /* UTF-8 ",t,0x<16-bit TPGT>" + NULL Terminator */
+ scsi_name_len += 10;
+ /* Check for 4-byte padding */
+ padding = ((-scsi_name_len) & 3);
+ if (padding != 0)
+ scsi_name_len += padding;
+ /* Header size + Designation descriptor */
+ scsi_name_len += 4;
+
+ if (((len + 4) + scsi_name_len) > cmd->data_length) {
+ len += scsi_name_len;
+ goto set_len;
+ }
+ buf[off] =
+ (TPG_TFO(tpg)->get_fabric_proto_ident(tpg) << 4);
+ buf[off++] |= 0x3; /* CODE SET == UTF-8 */
+ buf[off] = 0x80; /* Set PIV=1 */
+ /* Set ASSOICATION == target port: 01b */
+ buf[off] |= 0x10;
+ /* DESIGNATOR TYPE == SCSI name string */
+ buf[off++] |= 0x8;
+ off += 2; /* Skip over Reserved and length */
+ /*
+ * SCSI name string identifer containing, $FABRIC_MOD
+ * dependent information. For LIO-Target and iSCSI
+ * Target Port, this means "<iSCSI name>,t,0x<TPGT> in
+ * UTF-8 encoding.
+ */
+ tpgt = TPG_TFO(tpg)->tpg_get_tag(tpg);
+ scsi_name_len = sprintf(&buf[off], "%s,t,0x%04x",
+ TPG_TFO(tpg)->tpg_get_wwn(tpg), tpgt);
+ scsi_name_len += 1 /* Include NULL terminator */;
+ /*
+ * The null-terminated, null-padded (see 4.4.2) SCSI
+ * NAME STRING field contains a UTF-8 format string.
+ * The number of bytes in the SCSI NAME STRING field
+ * (i.e., the value in the DESIGNATOR LENGTH field)
+ * shall be no larger than 256 and shall be a multiple
+ * of four.
+ */
+ if (padding)
+ scsi_name_len += padding;
+
+ buf[off-1] = scsi_name_len;
+ off += scsi_name_len;
+ /* Header size + Designation descriptor */
+ len += (scsi_name_len + 4);
+ }
+set_len:
+ buf[2] = ((len >> 8) & 0xff);
+ buf[3] = (len & 0xff); /* Page Length for VPD 0x83 */
+ return 0;
+}
+
+/* Extended INQUIRY Data VPD Page */
+static int
+target_emulate_evpd_86(struct se_cmd *cmd, unsigned char *buf)
+{
+ if (cmd->data_length < 60)
+ return 0;
+
+ buf[1] = 0x86;
+ buf[2] = 0x3c;
+ /* Set HEADSUP, ORDSUP, SIMPSUP */
+ buf[5] = 0x07;
+
+ /* If WriteCache emulation is enabled, set V_SUP */
+ if (DEV_ATTRIB(SE_DEV(cmd))->emulate_write_cache > 0)
+ buf[6] = 0x01;
+ return 0;
+}
+
+/* Block Limits VPD page */
+static int
+target_emulate_evpd_b0(struct se_cmd *cmd, unsigned char *buf)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ int have_tp = 0;
+
+ /*
+ * Following sbc3r22 section 6.5.3 Block Limits VPD page, when
+ * emulate_tpu=1 or emulate_tpws=1 we will be expect a
+ * different page length for Thin Provisioning.
+ */
+ if (DEV_ATTRIB(dev)->emulate_tpu || DEV_ATTRIB(dev)->emulate_tpws)
+ have_tp = 1;
+
+ if (cmd->data_length < (0x10 + 4)) {
+ printk(KERN_INFO "Received data_length: %u"
+ " too small for EVPD 0xb0\n",
+ cmd->data_length);
+ return -1;
+ }
+
+ if (have_tp && cmd->data_length < (0x3c + 4)) {
+ printk(KERN_INFO "Received data_length: %u"
+ " too small for TPE=1 EVPD 0xb0\n",
+ cmd->data_length);
+ have_tp = 0;
+ }
+
+ buf[0] = dev->transport->get_device_type(dev);
+ buf[1] = 0xb0;
+ buf[3] = have_tp ? 0x3c : 0x10;
+
+ /*
+ * Set OPTIMAL TRANSFER LENGTH GRANULARITY
+ */
+ put_unaligned_be16(1, &buf[6]);
+
+ /*
+ * Set MAXIMUM TRANSFER LENGTH
+ */
+ put_unaligned_be32(DEV_ATTRIB(dev)->max_sectors, &buf[8]);
+
+ /*
+ * Set OPTIMAL TRANSFER LENGTH
+ */
+ put_unaligned_be32(DEV_ATTRIB(dev)->optimal_sectors, &buf[12]);
+
+ /*
+ * Exit now if we don't support TP or the initiator sent a too
+ * short buffer.
+ */
+ if (!have_tp || cmd->data_length < (0x3c + 4))
+ return 0;
+
+ /*
+ * Set MAXIMUM UNMAP LBA COUNT
+ */
+ put_unaligned_be32(DEV_ATTRIB(dev)->max_unmap_lba_count, &buf[20]);
+
+ /*
+ * Set MAXIMUM UNMAP BLOCK DESCRIPTOR COUNT
+ */
+ put_unaligned_be32(DEV_ATTRIB(dev)->max_unmap_block_desc_count,
+ &buf[24]);
+
+ /*
+ * Set OPTIMAL UNMAP GRANULARITY
+ */
+ put_unaligned_be32(DEV_ATTRIB(dev)->unmap_granularity, &buf[28]);
+
+ /*
+ * UNMAP GRANULARITY ALIGNMENT
+ */
+ put_unaligned_be32(DEV_ATTRIB(dev)->unmap_granularity_alignment,
+ &buf[32]);
+ if (DEV_ATTRIB(dev)->unmap_granularity_alignment != 0)
+ buf[32] |= 0x80; /* Set the UGAVALID bit */
+
+ return 0;
+}
+
+/* Thin Provisioning VPD */
+static int
+target_emulate_evpd_b2(struct se_cmd *cmd, unsigned char *buf)
+{
+ struct se_device *dev = SE_DEV(cmd);
+
+ /*
+ * From sbc3r22 section 6.5.4 Thin Provisioning VPD page:
+ *
+ * The PAGE LENGTH field is defined in SPC-4. If the DP bit is set to
+ * zero, then the page length shall be set to 0004h. If the DP bit
+ * is set to one, then the page length shall be set to the value
+ * defined in table 162.
+ */
+ buf[0] = dev->transport->get_device_type(dev);
+ buf[1] = 0xb2;
+
+ /*
+ * Set Hardcoded length mentioned above for DP=0
+ */
+ put_unaligned_be16(0x0004, &buf[2]);
+
+ /*
+ * The THRESHOLD EXPONENT field indicates the threshold set size in
+ * LBAs as a power of 2 (i.e., the threshold set size is equal to
+ * 2(threshold exponent)).
+ *
+ * Note that this is currently set to 0x00 as mkp says it will be
+ * changing again. We can enable this once it has settled in T10
+ * and is actually used by Linux/SCSI ML code.
+ */
+ buf[4] = 0x00;
+
+ /*
+ * A TPU bit set to one indicates that the device server supports
+ * the UNMAP command (see 5.25). A TPU bit set to zero indicates
+ * that the device server does not support the UNMAP command.
+ */
+ if (DEV_ATTRIB(dev)->emulate_tpu != 0)
+ buf[5] = 0x80;
+
+ /*
+ * A TPWS bit set to one indicates that the device server supports
+ * the use of the WRITE SAME (16) command (see 5.42) to unmap LBAs.
+ * A TPWS bit set to zero indicates that the device server does not
+ * support the use of the WRITE SAME (16) command to unmap LBAs.
+ */
+ if (DEV_ATTRIB(dev)->emulate_tpws != 0)
+ buf[5] |= 0x40;
+
+ return 0;
+}
+
+static int
+target_emulate_inquiry(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ unsigned char *buf = cmd->t_task->t_task_buf;
+ unsigned char *cdb = cmd->t_task->t_task_cdb;
+
+ if (!(cdb[1] & 0x1))
+ return target_emulate_inquiry_std(cmd);
+
+ /*
+ * Make sure we at least have 4 bytes of INQUIRY response
+ * payload for 0x00 going back for EVPD=1. Note that 0x80
+ * and 0x83 will check for enough payload data length and
+ * jump to set_len: label when there is not enough inquiry EVPD
+ * payload length left for the next outgoing EVPD metadata
+ */
+ if (cmd->data_length < 4) {
+ printk(KERN_ERR "SCSI Inquiry payload length: %u"
+ " too small for EVPD=1\n", cmd->data_length);
+ return -1;
+ }
+ buf[0] = dev->transport->get_device_type(dev);
+
+ switch (cdb[2]) {
+ case 0x00:
+ return target_emulate_evpd_00(cmd, buf);
+ case 0x80:
+ return target_emulate_evpd_80(cmd, buf);
+ case 0x83:
+ return target_emulate_evpd_83(cmd, buf);
+ case 0x86:
+ return target_emulate_evpd_86(cmd, buf);
+ case 0xb0:
+ return target_emulate_evpd_b0(cmd, buf);
+ case 0xb2:
+ return target_emulate_evpd_b2(cmd, buf);
+ default:
+ printk(KERN_ERR "Unknown VPD Code: 0x%02x\n", cdb[2]);
+ return -1;
+ }
+
+ return 0;
+}
+
+static int
+target_emulate_readcapacity(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ unsigned char *buf = cmd->t_task->t_task_buf;
+ u32 blocks = dev->transport->get_blocks(dev);
+
+ buf[0] = (blocks >> 24) & 0xff;
+ buf[1] = (blocks >> 16) & 0xff;
+ buf[2] = (blocks >> 8) & 0xff;
+ buf[3] = blocks & 0xff;
+ buf[4] = (DEV_ATTRIB(dev)->block_size >> 24) & 0xff;
+ buf[5] = (DEV_ATTRIB(dev)->block_size >> 16) & 0xff;
+ buf[6] = (DEV_ATTRIB(dev)->block_size >> 8) & 0xff;
+ buf[7] = DEV_ATTRIB(dev)->block_size & 0xff;
+ /*
+ * Set max 32-bit blocks to signal SERVICE ACTION READ_CAPACITY_16
+ */
+ if (DEV_ATTRIB(dev)->emulate_tpu || DEV_ATTRIB(dev)->emulate_tpws)
+ put_unaligned_be32(0xFFFFFFFF, &buf[0]);
+
+ return 0;
+}
+
+static int
+target_emulate_readcapacity_16(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ unsigned char *buf = cmd->t_task->t_task_buf;
+ unsigned long long blocks = dev->transport->get_blocks(dev);
+
+ buf[0] = (blocks >> 56) & 0xff;
+ buf[1] = (blocks >> 48) & 0xff;
+ buf[2] = (blocks >> 40) & 0xff;
+ buf[3] = (blocks >> 32) & 0xff;
+ buf[4] = (blocks >> 24) & 0xff;
+ buf[5] = (blocks >> 16) & 0xff;
+ buf[6] = (blocks >> 8) & 0xff;
+ buf[7] = blocks & 0xff;
+ buf[8] = (DEV_ATTRIB(dev)->block_size >> 24) & 0xff;
+ buf[9] = (DEV_ATTRIB(dev)->block_size >> 16) & 0xff;
+ buf[10] = (DEV_ATTRIB(dev)->block_size >> 8) & 0xff;
+ buf[11] = DEV_ATTRIB(dev)->block_size & 0xff;
+ /*
+ * Set Thin Provisioning Enable bit following sbc3r22 in section
+ * READ CAPACITY (16) byte 14 if emulate_tpu or emulate_tpws is enabled.
+ */
+ if (DEV_ATTRIB(dev)->emulate_tpu || DEV_ATTRIB(dev)->emulate_tpws)
+ buf[14] = 0x80;
+
+ return 0;
+}
+
+static int
+target_modesense_rwrecovery(unsigned char *p)
+{
+ p[0] = 0x01;
+ p[1] = 0x0a;
+
+ return 12;
+}
+
+static int
+target_modesense_control(struct se_device *dev, unsigned char *p)
+{
+ p[0] = 0x0a;
+ p[1] = 0x0a;
+ p[2] = 2;
+ /*
+ * From spc4r17, section 7.4.6 Control mode Page
+ *
+ * Unit Attention interlocks control (UN_INTLCK_CTRL) to code 00b
+ *
+ * 00b: The logical unit shall clear any unit attention condition
+ * reported in the same I_T_L_Q nexus transaction as a CHECK CONDITION
+ * status and shall not establish a unit attention condition when a com-
+ * mand is completed with BUSY, TASK SET FULL, or RESERVATION CONFLICT
+ * status.
+ *
+ * 10b: The logical unit shall not clear any unit attention condition
+ * reported in the same I_T_L_Q nexus transaction as a CHECK CONDITION
+ * status and shall not establish a unit attention condition when
+ * a command is completed with BUSY, TASK SET FULL, or RESERVATION
+ * CONFLICT status.
+ *
+ * 11b a The logical unit shall not clear any unit attention condition
+ * reported in the same I_T_L_Q nexus transaction as a CHECK CONDITION
+ * status and shall establish a unit attention condition for the
+ * initiator port associated with the I_T nexus on which the BUSY,
+ * TASK SET FULL, or RESERVATION CONFLICT status is being returned.
+ * Depending on the status, the additional sense code shall be set to
+ * PREVIOUS BUSY STATUS, PREVIOUS TASK SET FULL STATUS, or PREVIOUS
+ * RESERVATION CONFLICT STATUS. Until it is cleared by a REQUEST SENSE
+ * command, a unit attention condition shall be established only once
+ * for a BUSY, TASK SET FULL, or RESERVATION CONFLICT status regardless
+ * to the number of commands completed with one of those status codes.
+ */
+ p[4] = (DEV_ATTRIB(dev)->emulate_ua_intlck_ctrl == 2) ? 0x30 :
+ (DEV_ATTRIB(dev)->emulate_ua_intlck_ctrl == 1) ? 0x20 : 0x00;
+ /*
+ * From spc4r17, section 7.4.6 Control mode Page
+ *
+ * Task Aborted Status (TAS) bit set to zero.
+ *
+ * A task aborted status (TAS) bit set to zero specifies that aborted
+ * tasks shall be terminated by the device server without any response
+ * to the application client. A TAS bit set to one specifies that tasks
+ * aborted by the actions of an I_T nexus other than the I_T nexus on
+ * which the command was received shall be completed with TASK ABORTED
+ * status (see SAM-4).
+ */
+ p[5] = (DEV_ATTRIB(dev)->emulate_tas) ? 0x40 : 0x00;
+ p[8] = 0xff;
+ p[9] = 0xff;
+ p[11] = 30;
+
+ return 12;
+}
+
+static int
+target_modesense_caching(struct se_device *dev, unsigned char *p)
+{
+ p[0] = 0x08;
+ p[1] = 0x12;
+ if (DEV_ATTRIB(dev)->emulate_write_cache > 0)
+ p[2] = 0x04; /* Write Cache Enable */
+ p[12] = 0x20; /* Disabled Read Ahead */
+
+ return 20;
+}
+
+static void
+target_modesense_write_protect(unsigned char *buf, int type)
+{
+ /*
+ * I believe that the WP bit (bit 7) in the mode header is the same for
+ * all device types..
+ */
+ switch (type) {
+ case TYPE_DISK:
+ case TYPE_TAPE:
+ default:
+ buf[0] |= 0x80; /* WP bit */
+ break;
+ }
+}
+
+static void
+target_modesense_dpofua(unsigned char *buf, int type)
+{
+ switch (type) {
+ case TYPE_DISK:
+ buf[0] |= 0x10; /* DPOFUA bit */
+ break;
+ default:
+ break;
+ }
+}
+
+static int
+target_emulate_modesense(struct se_cmd *cmd, int ten)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ char *cdb = cmd->t_task->t_task_cdb;
+ unsigned char *rbuf = cmd->t_task->t_task_buf;
+ int type = dev->transport->get_device_type(dev);
+ int offset = (ten) ? 8 : 4;
+ int length = 0;
+ unsigned char buf[SE_MODE_PAGE_BUF];
+
+ memset(buf, 0, SE_MODE_PAGE_BUF);
+
+ switch (cdb[2] & 0x3f) {
+ case 0x01:
+ length = target_modesense_rwrecovery(&buf[offset]);
+ break;
+ case 0x08:
+ length = target_modesense_caching(dev, &buf[offset]);
+ break;
+ case 0x0a:
+ length = target_modesense_control(dev, &buf[offset]);
+ break;
+ case 0x3f:
+ length = target_modesense_rwrecovery(&buf[offset]);
+ length += target_modesense_caching(dev, &buf[offset+length]);
+ length += target_modesense_control(dev, &buf[offset+length]);
+ break;
+ default:
+ printk(KERN_ERR "Got Unknown Mode Page: 0x%02x\n",
+ cdb[2] & 0x3f);
+ return PYX_TRANSPORT_UNKNOWN_MODE_PAGE;
+ }
+ offset += length;
+
+ if (ten) {
+ offset -= 2;
+ buf[0] = (offset >> 8) & 0xff;
+ buf[1] = offset & 0xff;
+
+ if ((SE_LUN(cmd)->lun_access & TRANSPORT_LUNFLAGS_READ_ONLY) ||
+ (cmd->se_deve &&
+ (cmd->se_deve->lun_flags & TRANSPORT_LUNFLAGS_READ_ONLY)))
+ target_modesense_write_protect(&buf[3], type);
+
+ if ((DEV_ATTRIB(dev)->emulate_write_cache > 0) &&
+ (DEV_ATTRIB(dev)->emulate_fua_write > 0))
+ target_modesense_dpofua(&buf[3], type);
+
+ if ((offset + 2) > cmd->data_length)
+ offset = cmd->data_length;
+
+ } else {
+ offset -= 1;
+ buf[0] = offset & 0xff;
+
+ if ((SE_LUN(cmd)->lun_access & TRANSPORT_LUNFLAGS_READ_ONLY) ||
+ (cmd->se_deve &&
+ (cmd->se_deve->lun_flags & TRANSPORT_LUNFLAGS_READ_ONLY)))
+ target_modesense_write_protect(&buf[2], type);
+
+ if ((DEV_ATTRIB(dev)->emulate_write_cache > 0) &&
+ (DEV_ATTRIB(dev)->emulate_fua_write > 0))
+ target_modesense_dpofua(&buf[2], type);
+
+ if ((offset + 1) > cmd->data_length)
+ offset = cmd->data_length;
+ }
+ memcpy(rbuf, buf, offset);
+
+ return 0;
+}
+
+static int
+target_emulate_request_sense(struct se_cmd *cmd)
+{
+ unsigned char *cdb = cmd->t_task->t_task_cdb;
+ unsigned char *buf = cmd->t_task->t_task_buf;
+ u8 ua_asc = 0, ua_ascq = 0;
+
+ if (cdb[1] & 0x01) {
+ printk(KERN_ERR "REQUEST_SENSE description emulation not"
+ " supported\n");
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+ if (!(core_scsi3_ua_clear_for_request_sense(cmd, &ua_asc, &ua_ascq))) {
+ /*
+ * CURRENT ERROR, UNIT ATTENTION
+ */
+ buf[0] = 0x70;
+ buf[SPC_SENSE_KEY_OFFSET] = UNIT_ATTENTION;
+ /*
+ * Make sure request data length is enough for additional
+ * sense data.
+ */
+ if (cmd->data_length <= 18) {
+ buf[7] = 0x00;
+ return 0;
+ }
+ /*
+ * The Additional Sense Code (ASC) from the UNIT ATTENTION
+ */
+ buf[SPC_ASC_KEY_OFFSET] = ua_asc;
+ buf[SPC_ASCQ_KEY_OFFSET] = ua_ascq;
+ buf[7] = 0x0A;
+ } else {
+ /*
+ * CURRENT ERROR, NO SENSE
+ */
+ buf[0] = 0x70;
+ buf[SPC_SENSE_KEY_OFFSET] = NO_SENSE;
+ /*
+ * Make sure request data length is enough for additional
+ * sense data.
+ */
+ if (cmd->data_length <= 18) {
+ buf[7] = 0x00;
+ return 0;
+ }
+ /*
+ * NO ADDITIONAL SENSE INFORMATION
+ */
+ buf[SPC_ASC_KEY_OFFSET] = 0x00;
+ buf[7] = 0x0A;
+ }
+
+ return 0;
+}
+
+/*
+ * Used for TCM/IBLOCK and TCM/FILEIO for block/blk-lib.c level discard support.
+ * Note this is not used for TCM/pSCSI passthrough
+ */
+static int
+target_emulate_unmap(struct se_task *task)
+{
+ struct se_cmd *cmd = TASK_CMD(task);
+ struct se_device *dev = SE_DEV(cmd);
+ unsigned char *buf = cmd->t_task->t_task_buf, *ptr = NULL;
+ unsigned char *cdb = &cmd->t_task->t_task_cdb[0];
+ sector_t lba;
+ unsigned int size = cmd->data_length, range;
+ int ret, offset;
+ unsigned short dl, bd_dl;
+
+ /* First UNMAP block descriptor starts at 8 byte offset */
+ offset = 8;
+ size -= 8;
+ dl = get_unaligned_be16(&cdb[0]);
+ bd_dl = get_unaligned_be16(&cdb[2]);
+ ptr = &buf[offset];
+ printk(KERN_INFO "UNMAP: Sub: %s Using dl: %hu bd_dl: %hu size: %hu"
+ " ptr: %p\n", dev->transport->name, dl, bd_dl, size, ptr);
+
+ while (size) {
+ lba = get_unaligned_be64(&ptr[0]);
+ range = get_unaligned_be32(&ptr[8]);
+ printk(KERN_INFO "UNMAP: Using lba: %llu and range: %u\n",
+ (unsigned long long)lba, range);
+
+ ret = dev->transport->do_discard(dev, lba, range);
+ if (ret < 0) {
+ printk(KERN_ERR "blkdev_issue_discard() failed: %d\n",
+ ret);
+ return -1;
+ }
+
+ ptr += 16;
+ size -= 16;
+ }
+
+ task->task_scsi_status = GOOD;
+ transport_complete_task(task, 1);
+ return 0;
+}
+
+/*
+ * Used for TCM/IBLOCK and TCM/FILEIO for block/blk-lib.c level discard support.
+ * Note this is not used for TCM/pSCSI passthrough
+ */
+static int
+target_emulate_write_same(struct se_task *task)
+{
+ struct se_cmd *cmd = TASK_CMD(task);
+ struct se_device *dev = SE_DEV(cmd);
+ sector_t lba = cmd->t_task->t_task_lba;
+ unsigned int range;
+ int ret;
+
+ range = (cmd->data_length / DEV_ATTRIB(dev)->block_size);
+
+ printk(KERN_INFO "WRITE_SAME UNMAP: LBA: %llu Range: %u\n",
+ (unsigned long long)lba, range);
+
+ ret = dev->transport->do_discard(dev, lba, range);
+ if (ret < 0) {
+ printk(KERN_INFO "blkdev_issue_discard() failed for WRITE_SAME\n");
+ return -1;
+ }
+
+ task->task_scsi_status = GOOD;
+ transport_complete_task(task, 1);
+ return 0;
+}
+
+int
+transport_emulate_control_cdb(struct se_task *task)
+{
+ struct se_cmd *cmd = TASK_CMD(task);
+ struct se_device *dev = SE_DEV(cmd);
+ unsigned short service_action;
+ int ret = 0;
+
+ switch (cmd->t_task->t_task_cdb[0]) {
+ case INQUIRY:
+ ret = target_emulate_inquiry(cmd);
+ break;
+ case READ_CAPACITY:
+ ret = target_emulate_readcapacity(cmd);
+ break;
+ case MODE_SENSE:
+ ret = target_emulate_modesense(cmd, 0);
+ break;
+ case MODE_SENSE_10:
+ ret = target_emulate_modesense(cmd, 1);
+ break;
+ case SERVICE_ACTION_IN:
+ switch (cmd->t_task->t_task_cdb[1] & 0x1f) {
+ case SAI_READ_CAPACITY_16:
+ ret = target_emulate_readcapacity_16(cmd);
+ break;
+ default:
+ printk(KERN_ERR "Unsupported SA: 0x%02x\n",
+ cmd->t_task->t_task_cdb[1] & 0x1f);
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+ break;
+ case REQUEST_SENSE:
+ ret = target_emulate_request_sense(cmd);
+ break;
+ case UNMAP:
+ if (!dev->transport->do_discard) {
+ printk(KERN_ERR "UNMAP emulation not supported for: %s\n",
+ dev->transport->name);
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+ ret = target_emulate_unmap(task);
+ break;
+ case WRITE_SAME_16:
+ if (!dev->transport->do_discard) {
+ printk(KERN_ERR "WRITE_SAME_16 emulation not supported"
+ " for: %s\n", dev->transport->name);
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+ ret = target_emulate_write_same(task);
+ break;
+ case VARIABLE_LENGTH_CMD:
+ service_action =
+ get_unaligned_be16(&cmd->t_task->t_task_cdb[8]);
+ switch (service_action) {
+ case WRITE_SAME_32:
+ if (!dev->transport->do_discard) {
+ printk(KERN_ERR "WRITE_SAME_32 SA emulation not"
+ " supported for: %s\n",
+ dev->transport->name);
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+ ret = target_emulate_write_same(task);
+ break;
+ default:
+ printk(KERN_ERR "Unsupported VARIABLE_LENGTH_CMD SA:"
+ " 0x%02x\n", service_action);
+ break;
+ }
+ break;
+ case SYNCHRONIZE_CACHE:
+ case 0x91: /* SYNCHRONIZE_CACHE_16: */
+ if (!dev->transport->do_sync_cache) {
+ printk(KERN_ERR
+ "SYNCHRONIZE_CACHE emulation not supported"
+ " for: %s\n", dev->transport->name);
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+ dev->transport->do_sync_cache(task);
+ break;
+ case ALLOW_MEDIUM_REMOVAL:
+ case ERASE:
+ case REZERO_UNIT:
+ case SEEK_10:
+ case SPACE:
+ case START_STOP:
+ case TEST_UNIT_READY:
+ case VERIFY:
+ case WRITE_FILEMARKS:
+ break;
+ default:
+ printk(KERN_ERR "Unsupported SCSI Opcode: 0x%02x for %s\n",
+ cmd->t_task->t_task_cdb[0], dev->transport->name);
+ return PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+
+ if (ret < 0)
+ return ret;
+ task->task_scsi_status = GOOD;
+ transport_complete_task(task, 1);
+
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+}
diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c
new file mode 100644
index 0000000..2764510
--- /dev/null
+++ b/drivers/target/target_core_configfs.c
@@ -0,0 +1,3225 @@
+/*******************************************************************************
+ * Filename: target_core_configfs.c
+ *
+ * This file contains ConfigFS logic for the Generic Target Engine project.
+ *
+ * Copyright (c) 2008-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * based on configfs Copyright (C) 2005 Oracle. All rights reserved.
+ *
+ * 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.
+ *
+ * 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.
+ ****************************************************************************/
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/version.h>
+#include <generated/utsrelease.h>
+#include <linux/utsname.h>
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/namei.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/delay.h>
+#include <linux/unistd.h>
+#include <linux/string.h>
+#include <linux/parser.h>
+#include <linux/syscalls.h>
+#include <linux/configfs.h>
+#include <linux/proc_fs.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_fabric_configfs.h>
+#include <target/target_core_configfs.h>
+#include <target/configfs_macros.h>
+
+#include "target_core_alua.h"
+#include "target_core_hba.h"
+#include "target_core_pr.h"
+#include "target_core_rd.h"
+
+static struct list_head g_tf_list;
+static struct mutex g_tf_lock;
+
+struct target_core_configfs_attribute {
+ struct configfs_attribute attr;
+ ssize_t (*show)(void *, char *);
+ ssize_t (*store)(void *, const char *, size_t);
+};
+
+static inline struct se_hba *
+item_to_hba(struct config_item *item)
+{
+ return container_of(to_config_group(item), struct se_hba, hba_group);
+}
+
+/*
+ * Attributes for /sys/kernel/config/target/
+ */
+static ssize_t target_core_attr_show(struct config_item *item,
+ struct configfs_attribute *attr,
+ char *page)
+{
+ return sprintf(page, "Target Engine Core ConfigFS Infrastructure %s"
+ " on %s/%s on "UTS_RELEASE"\n", TARGET_CORE_CONFIGFS_VERSION,
+ utsname()->sysname, utsname()->machine);
+}
+
+static struct configfs_item_operations target_core_fabric_item_ops = {
+ .show_attribute = target_core_attr_show,
+};
+
+static struct configfs_attribute target_core_item_attr_version = {
+ .ca_owner = THIS_MODULE,
+ .ca_name = "version",
+ .ca_mode = S_IRUGO,
+};
+
+static struct target_fabric_configfs *target_core_get_fabric(
+ const char *name)
+{
+ struct target_fabric_configfs *tf;
+
+ if (!(name))
+ return NULL;
+
+ mutex_lock(&g_tf_lock);
+ list_for_each_entry(tf, &g_tf_list, tf_list) {
+ if (!(strcmp(tf->tf_name, name))) {
+ atomic_inc(&tf->tf_access_cnt);
+ mutex_unlock(&g_tf_lock);
+ return tf;
+ }
+ }
+ mutex_unlock(&g_tf_lock);
+
+ return NULL;
+}
+
+/*
+ * Called from struct target_core_group_ops->make_group()
+ */
+static struct config_group *target_core_register_fabric(
+ struct config_group *group,
+ const char *name)
+{
+ struct target_fabric_configfs *tf;
+ int ret;
+
+ printk(KERN_INFO "Target_Core_ConfigFS: REGISTER -> group: %p name:"
+ " %s\n", group, name);
+ /*
+ * Ensure that TCM subsystem plugins are loaded at this point for
+ * using the RAMDISK_DR virtual LUN 0 and all other struct se_port
+ * LUN symlinks.
+ */
+ if (transport_subsystem_check_init() < 0)
+ return ERR_PTR(-EINVAL);
+
+ /*
+ * Below are some hardcoded request_module() calls to automatically
+ * local fabric modules when the following is called:
+ *
+ * mkdir -p /sys/kernel/config/target/$MODULE_NAME
+ *
+ * Note that this does not limit which TCM fabric module can be
+ * registered, but simply provids auto loading logic for modules with
+ * mkdir(2) system calls with known TCM fabric modules.
+ */
+ if (!(strncmp(name, "iscsi", 5))) {
+ /*
+ * Automatically load the LIO Target fabric module when the
+ * following is called:
+ *
+ * mkdir -p $CONFIGFS/target/iscsi
+ */
+ ret = request_module("iscsi_target_mod");
+ if (ret < 0) {
+ printk(KERN_ERR "request_module() failed for"
+ " iscsi_target_mod.ko: %d\n", ret);
+ return ERR_PTR(-EINVAL);
+ }
+ } else if (!(strncmp(name, "loopback", 8))) {
+ /*
+ * Automatically load the tcm_loop fabric module when the
+ * following is called:
+ *
+ * mkdir -p $CONFIGFS/target/loopback
+ */
+ ret = request_module("tcm_loop");
+ if (ret < 0) {
+ printk(KERN_ERR "request_module() failed for"
+ " tcm_loop.ko: %d\n", ret);
+ return ERR_PTR(-EINVAL);
+ }
+ }
+
+ tf = target_core_get_fabric(name);
+ if (!(tf)) {
+ printk(KERN_ERR "target_core_get_fabric() failed for %s\n",
+ name);
+ return ERR_PTR(-EINVAL);
+ }
+ printk(KERN_INFO "Target_Core_ConfigFS: REGISTER -> Located fabric:"
+ " %s\n", tf->tf_name);
+ /*
+ * On a successful target_core_get_fabric() look, the returned
+ * struct target_fabric_configfs *tf will contain a usage reference.
+ */
+ printk(KERN_INFO "Target_Core_ConfigFS: REGISTER tfc_wwn_cit -> %p\n",
+ &TF_CIT_TMPL(tf)->tfc_wwn_cit);
+
+ tf->tf_group.default_groups = tf->tf_default_groups;
+ tf->tf_group.default_groups[0] = &tf->tf_disc_group;
+ tf->tf_group.default_groups[1] = NULL;
+
+ config_group_init_type_name(&tf->tf_group, name,
+ &TF_CIT_TMPL(tf)->tfc_wwn_cit);
+ config_group_init_type_name(&tf->tf_disc_group, "discovery_auth",
+ &TF_CIT_TMPL(tf)->tfc_discovery_cit);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: REGISTER -> Allocated Fabric:"
+ " %s\n", tf->tf_group.cg_item.ci_name);
+ /*
+ * Setup tf_ops.tf_subsys pointer for usage with configfs_depend_item()
+ */
+ tf->tf_ops.tf_subsys = tf->tf_subsys;
+ tf->tf_fabric = &tf->tf_group.cg_item;
+ printk(KERN_INFO "Target_Core_ConfigFS: REGISTER -> Set tf->tf_fabric"
+ " for %s\n", name);
+
+ return &tf->tf_group;
+}
+
+/*
+ * Called from struct target_core_group_ops->drop_item()
+ */
+static void target_core_deregister_fabric(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct target_fabric_configfs *tf = container_of(
+ to_config_group(item), struct target_fabric_configfs, tf_group);
+ struct config_group *tf_group;
+ struct config_item *df_item;
+ int i;
+
+ printk(KERN_INFO "Target_Core_ConfigFS: DEREGISTER -> Looking up %s in"
+ " tf list\n", config_item_name(item));
+
+ printk(KERN_INFO "Target_Core_ConfigFS: DEREGISTER -> located fabric:"
+ " %s\n", tf->tf_name);
+ atomic_dec(&tf->tf_access_cnt);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: DEREGISTER -> Releasing"
+ " tf->tf_fabric for %s\n", tf->tf_name);
+ tf->tf_fabric = NULL;
+
+ printk(KERN_INFO "Target_Core_ConfigFS: DEREGISTER -> Releasing ci"
+ " %s\n", config_item_name(item));
+
+ tf_group = &tf->tf_group;
+ for (i = 0; tf_group->default_groups[i]; i++) {
+ df_item = &tf_group->default_groups[i]->cg_item;
+ tf_group->default_groups[i] = NULL;
+ config_item_put(df_item);
+ }
+ config_item_put(item);
+}
+
+static struct configfs_group_operations target_core_fabric_group_ops = {
+ .make_group = &target_core_register_fabric,
+ .drop_item = &target_core_deregister_fabric,
+};
+
+/*
+ * All item attributes appearing in /sys/kernel/target/ appear here.
+ */
+static struct configfs_attribute *target_core_fabric_item_attrs[] = {
+ &target_core_item_attr_version,
+ NULL,
+};
+
+/*
+ * Provides Fabrics Groups and Item Attributes for /sys/kernel/config/target/
+ */
+static struct config_item_type target_core_fabrics_item = {
+ .ct_item_ops = &target_core_fabric_item_ops,
+ .ct_group_ops = &target_core_fabric_group_ops,
+ .ct_attrs = target_core_fabric_item_attrs,
+ .ct_owner = THIS_MODULE,
+};
+
+static struct configfs_subsystem target_core_fabrics = {
+ .su_group = {
+ .cg_item = {
+ .ci_namebuf = "target",
+ .ci_type = &target_core_fabrics_item,
+ },
+ },
+};
+
+static struct configfs_subsystem *target_core_subsystem[] = {
+ &target_core_fabrics,
+ NULL,
+};
+
+/*##############################################################################
+// Start functions called by external Target Fabrics Modules
+//############################################################################*/
+
+/*
+ * First function called by fabric modules to:
+ *
+ * 1) Allocate a struct target_fabric_configfs and save the *fabric_cit pointer.
+ * 2) Add struct target_fabric_configfs to g_tf_list
+ * 3) Return struct target_fabric_configfs to fabric module to be passed
+ * into target_fabric_configfs_register().
+ */
+struct target_fabric_configfs *target_fabric_configfs_init(
+ struct module *fabric_mod,
+ const char *name)
+{
+ struct target_fabric_configfs *tf;
+
+ if (!(fabric_mod)) {
+ printk(KERN_ERR "Missing struct module *fabric_mod pointer\n");
+ return NULL;
+ }
+ if (!(name)) {
+ printk(KERN_ERR "Unable to locate passed fabric name\n");
+ return NULL;
+ }
+ if (strlen(name) > TARGET_FABRIC_NAME_SIZE) {
+ printk(KERN_ERR "Passed name: %s exceeds TARGET_FABRIC"
+ "_NAME_SIZE\n", name);
+ return NULL;
+ }
+
+ tf = kzalloc(sizeof(struct target_fabric_configfs), GFP_KERNEL);
+ if (!(tf))
+ return ERR_PTR(-ENOMEM);
+
+ INIT_LIST_HEAD(&tf->tf_list);
+ atomic_set(&tf->tf_access_cnt, 0);
+ /*
+ * Setup the default generic struct config_item_type's (cits) in
+ * struct target_fabric_configfs->tf_cit_tmpl
+ */
+ tf->tf_module = fabric_mod;
+ target_fabric_setup_cits(tf);
+
+ tf->tf_subsys = target_core_subsystem[0];
+ snprintf(tf->tf_name, TARGET_FABRIC_NAME_SIZE, "%s", name);
+
+ mutex_lock(&g_tf_lock);
+ list_add_tail(&tf->tf_list, &g_tf_list);
+ mutex_unlock(&g_tf_lock);
+
+ printk(KERN_INFO "<<<<<<<<<<<<<<<<<<<<<< BEGIN FABRIC API >>>>>>>>"
+ ">>>>>>>>>>>>>>\n");
+ printk(KERN_INFO "Initialized struct target_fabric_configfs: %p for"
+ " %s\n", tf, tf->tf_name);
+ return tf;
+}
+EXPORT_SYMBOL(target_fabric_configfs_init);
+
+/*
+ * Called by fabric plugins after FAILED target_fabric_configfs_register() call.
+ */
+void target_fabric_configfs_free(
+ struct target_fabric_configfs *tf)
+{
+ mutex_lock(&g_tf_lock);
+ list_del(&tf->tf_list);
+ mutex_unlock(&g_tf_lock);
+
+ kfree(tf);
+}
+EXPORT_SYMBOL(target_fabric_configfs_free);
+
+/*
+ * Perform a sanity check of the passed tf->tf_ops before completing
+ * TCM fabric module registration.
+ */
+static int target_fabric_tf_ops_check(
+ struct target_fabric_configfs *tf)
+{
+ struct target_core_fabric_ops *tfo = &tf->tf_ops;
+
+ if (!(tfo->get_fabric_name)) {
+ printk(KERN_ERR "Missing tfo->get_fabric_name()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->get_fabric_proto_ident)) {
+ printk(KERN_ERR "Missing tfo->get_fabric_proto_ident()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_get_wwn)) {
+ printk(KERN_ERR "Missing tfo->tpg_get_wwn()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_get_tag)) {
+ printk(KERN_ERR "Missing tfo->tpg_get_tag()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_get_default_depth)) {
+ printk(KERN_ERR "Missing tfo->tpg_get_default_depth()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_get_pr_transport_id)) {
+ printk(KERN_ERR "Missing tfo->tpg_get_pr_transport_id()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_get_pr_transport_id_len)) {
+ printk(KERN_ERR "Missing tfo->tpg_get_pr_transport_id_len()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_check_demo_mode)) {
+ printk(KERN_ERR "Missing tfo->tpg_check_demo_mode()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_check_demo_mode_cache)) {
+ printk(KERN_ERR "Missing tfo->tpg_check_demo_mode_cache()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_check_demo_mode_write_protect)) {
+ printk(KERN_ERR "Missing tfo->tpg_check_demo_mode_write_protect()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_check_prod_mode_write_protect)) {
+ printk(KERN_ERR "Missing tfo->tpg_check_prod_mode_write_protect()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_alloc_fabric_acl)) {
+ printk(KERN_ERR "Missing tfo->tpg_alloc_fabric_acl()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_release_fabric_acl)) {
+ printk(KERN_ERR "Missing tfo->tpg_release_fabric_acl()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->tpg_get_inst_index)) {
+ printk(KERN_ERR "Missing tfo->tpg_get_inst_index()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->release_cmd_to_pool)) {
+ printk(KERN_ERR "Missing tfo->release_cmd_to_pool()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->release_cmd_direct)) {
+ printk(KERN_ERR "Missing tfo->release_cmd_direct()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->shutdown_session)) {
+ printk(KERN_ERR "Missing tfo->shutdown_session()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->close_session)) {
+ printk(KERN_ERR "Missing tfo->close_session()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->stop_session)) {
+ printk(KERN_ERR "Missing tfo->stop_session()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->fall_back_to_erl0)) {
+ printk(KERN_ERR "Missing tfo->fall_back_to_erl0()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->sess_logged_in)) {
+ printk(KERN_ERR "Missing tfo->sess_logged_in()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->sess_get_index)) {
+ printk(KERN_ERR "Missing tfo->sess_get_index()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->write_pending)) {
+ printk(KERN_ERR "Missing tfo->write_pending()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->write_pending_status)) {
+ printk(KERN_ERR "Missing tfo->write_pending_status()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->set_default_node_attributes)) {
+ printk(KERN_ERR "Missing tfo->set_default_node_attributes()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->get_task_tag)) {
+ printk(KERN_ERR "Missing tfo->get_task_tag()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->get_cmd_state)) {
+ printk(KERN_ERR "Missing tfo->get_cmd_state()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->new_cmd_failure)) {
+ printk(KERN_ERR "Missing tfo->new_cmd_failure()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->queue_data_in)) {
+ printk(KERN_ERR "Missing tfo->queue_data_in()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->queue_status)) {
+ printk(KERN_ERR "Missing tfo->queue_status()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->queue_tm_rsp)) {
+ printk(KERN_ERR "Missing tfo->queue_tm_rsp()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->set_fabric_sense_len)) {
+ printk(KERN_ERR "Missing tfo->set_fabric_sense_len()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->get_fabric_sense_len)) {
+ printk(KERN_ERR "Missing tfo->get_fabric_sense_len()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->is_state_remove)) {
+ printk(KERN_ERR "Missing tfo->is_state_remove()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->pack_lun)) {
+ printk(KERN_ERR "Missing tfo->pack_lun()\n");
+ return -EINVAL;
+ }
+ /*
+ * We at least require tfo->fabric_make_wwn(), tfo->fabric_drop_wwn()
+ * tfo->fabric_make_tpg() and tfo->fabric_drop_tpg() in
+ * target_core_fabric_configfs.c WWN+TPG group context code.
+ */
+ if (!(tfo->fabric_make_wwn)) {
+ printk(KERN_ERR "Missing tfo->fabric_make_wwn()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->fabric_drop_wwn)) {
+ printk(KERN_ERR "Missing tfo->fabric_drop_wwn()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->fabric_make_tpg)) {
+ printk(KERN_ERR "Missing tfo->fabric_make_tpg()\n");
+ return -EINVAL;
+ }
+ if (!(tfo->fabric_drop_tpg)) {
+ printk(KERN_ERR "Missing tfo->fabric_drop_tpg()\n");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/*
+ * Called 2nd from fabric module with returned parameter of
+ * struct target_fabric_configfs * from target_fabric_configfs_init().
+ *
+ * Upon a successful registration, the new fabric's struct config_item is
+ * return. Also, a pointer to this struct is set in the passed
+ * struct target_fabric_configfs.
+ */
+int target_fabric_configfs_register(
+ struct target_fabric_configfs *tf)
+{
+ struct config_group *su_group;
+ int ret;
+
+ if (!(tf)) {
+ printk(KERN_ERR "Unable to locate target_fabric_configfs"
+ " pointer\n");
+ return -EINVAL;
+ }
+ if (!(tf->tf_subsys)) {
+ printk(KERN_ERR "Unable to target struct config_subsystem"
+ " pointer\n");
+ return -EINVAL;
+ }
+ su_group = &tf->tf_subsys->su_group;
+ if (!(su_group)) {
+ printk(KERN_ERR "Unable to locate target struct config_group"
+ " pointer\n");
+ return -EINVAL;
+ }
+ ret = target_fabric_tf_ops_check(tf);
+ if (ret < 0)
+ return ret;
+
+ printk(KERN_INFO "<<<<<<<<<<<<<<<<<<<<<< END FABRIC API >>>>>>>>>>>>"
+ ">>>>>>>>>>\n");
+ return 0;
+}
+EXPORT_SYMBOL(target_fabric_configfs_register);
+
+void target_fabric_configfs_deregister(
+ struct target_fabric_configfs *tf)
+{
+ struct config_group *su_group;
+ struct configfs_subsystem *su;
+
+ if (!(tf)) {
+ printk(KERN_ERR "Unable to locate passed target_fabric_"
+ "configfs\n");
+ return;
+ }
+ su = tf->tf_subsys;
+ if (!(su)) {
+ printk(KERN_ERR "Unable to locate passed tf->tf_subsys"
+ " pointer\n");
+ return;
+ }
+ su_group = &tf->tf_subsys->su_group;
+ if (!(su_group)) {
+ printk(KERN_ERR "Unable to locate target struct config_group"
+ " pointer\n");
+ return;
+ }
+
+ printk(KERN_INFO "<<<<<<<<<<<<<<<<<<<<<< BEGIN FABRIC API >>>>>>>>>>"
+ ">>>>>>>>>>>>\n");
+ mutex_lock(&g_tf_lock);
+ if (atomic_read(&tf->tf_access_cnt)) {
+ mutex_unlock(&g_tf_lock);
+ printk(KERN_ERR "Non zero tf->tf_access_cnt for fabric %s\n",
+ tf->tf_name);
+ BUG();
+ }
+ list_del(&tf->tf_list);
+ mutex_unlock(&g_tf_lock);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: DEREGISTER -> Releasing tf:"
+ " %s\n", tf->tf_name);
+ tf->tf_module = NULL;
+ tf->tf_subsys = NULL;
+ kfree(tf);
+
+ printk("<<<<<<<<<<<<<<<<<<<<<< END FABRIC API >>>>>>>>>>>>>>>>>"
+ ">>>>>\n");
+ return;
+}
+EXPORT_SYMBOL(target_fabric_configfs_deregister);
+
+/*##############################################################################
+// Stop functions called by external Target Fabrics Modules
+//############################################################################*/
+
+/* Start functions for struct config_item_type target_core_dev_attrib_cit */
+
+#define DEF_DEV_ATTRIB_SHOW(_name) \
+static ssize_t target_core_dev_show_attr_##_name( \
+ struct se_dev_attrib *da, \
+ char *page) \
+{ \
+ struct se_device *dev; \
+ struct se_subsystem_dev *se_dev = da->da_sub_dev; \
+ ssize_t rb; \
+ \
+ spin_lock(&se_dev->se_dev_lock); \
+ dev = se_dev->se_dev_ptr; \
+ if (!(dev)) { \
+ spin_unlock(&se_dev->se_dev_lock); \
+ return -ENODEV; \
+ } \
+ rb = snprintf(page, PAGE_SIZE, "%u\n", (u32)DEV_ATTRIB(dev)->_name); \
+ spin_unlock(&se_dev->se_dev_lock); \
+ \
+ return rb; \
+}
+
+#define DEF_DEV_ATTRIB_STORE(_name) \
+static ssize_t target_core_dev_store_attr_##_name( \
+ struct se_dev_attrib *da, \
+ const char *page, \
+ size_t count) \
+{ \
+ struct se_device *dev; \
+ struct se_subsystem_dev *se_dev = da->da_sub_dev; \
+ unsigned long val; \
+ int ret; \
+ \
+ spin_lock(&se_dev->se_dev_lock); \
+ dev = se_dev->se_dev_ptr; \
+ if (!(dev)) { \
+ spin_unlock(&se_dev->se_dev_lock); \
+ return -ENODEV; \
+ } \
+ ret = strict_strtoul(page, 0, &val); \
+ if (ret < 0) { \
+ spin_unlock(&se_dev->se_dev_lock); \
+ printk(KERN_ERR "strict_strtoul() failed with" \
+ " ret: %d\n", ret); \
+ return -EINVAL; \
+ } \
+ ret = se_dev_set_##_name(dev, (u32)val); \
+ spin_unlock(&se_dev->se_dev_lock); \
+ \
+ return (!ret) ? count : -EINVAL; \
+}
+
+#define DEF_DEV_ATTRIB(_name) \
+DEF_DEV_ATTRIB_SHOW(_name); \
+DEF_DEV_ATTRIB_STORE(_name);
+
+#define DEF_DEV_ATTRIB_RO(_name) \
+DEF_DEV_ATTRIB_SHOW(_name);
+
+CONFIGFS_EATTR_STRUCT(target_core_dev_attrib, se_dev_attrib);
+#define SE_DEV_ATTR(_name, _mode) \
+static struct target_core_dev_attrib_attribute \
+ target_core_dev_attrib_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ target_core_dev_show_attr_##_name, \
+ target_core_dev_store_attr_##_name);
+
+#define SE_DEV_ATTR_RO(_name); \
+static struct target_core_dev_attrib_attribute \
+ target_core_dev_attrib_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ target_core_dev_show_attr_##_name);
+
+DEF_DEV_ATTRIB(emulate_dpo);
+SE_DEV_ATTR(emulate_dpo, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(emulate_fua_write);
+SE_DEV_ATTR(emulate_fua_write, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(emulate_fua_read);
+SE_DEV_ATTR(emulate_fua_read, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(emulate_write_cache);
+SE_DEV_ATTR(emulate_write_cache, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(emulate_ua_intlck_ctrl);
+SE_DEV_ATTR(emulate_ua_intlck_ctrl, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(emulate_tas);
+SE_DEV_ATTR(emulate_tas, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(emulate_tpu);
+SE_DEV_ATTR(emulate_tpu, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(emulate_tpws);
+SE_DEV_ATTR(emulate_tpws, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(enforce_pr_isids);
+SE_DEV_ATTR(enforce_pr_isids, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB_RO(hw_block_size);
+SE_DEV_ATTR_RO(hw_block_size);
+
+DEF_DEV_ATTRIB(block_size);
+SE_DEV_ATTR(block_size, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB_RO(hw_max_sectors);
+SE_DEV_ATTR_RO(hw_max_sectors);
+
+DEF_DEV_ATTRIB(max_sectors);
+SE_DEV_ATTR(max_sectors, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(optimal_sectors);
+SE_DEV_ATTR(optimal_sectors, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB_RO(hw_queue_depth);
+SE_DEV_ATTR_RO(hw_queue_depth);
+
+DEF_DEV_ATTRIB(queue_depth);
+SE_DEV_ATTR(queue_depth, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(task_timeout);
+SE_DEV_ATTR(task_timeout, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(max_unmap_lba_count);
+SE_DEV_ATTR(max_unmap_lba_count, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(max_unmap_block_desc_count);
+SE_DEV_ATTR(max_unmap_block_desc_count, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(unmap_granularity);
+SE_DEV_ATTR(unmap_granularity, S_IRUGO | S_IWUSR);
+
+DEF_DEV_ATTRIB(unmap_granularity_alignment);
+SE_DEV_ATTR(unmap_granularity_alignment, S_IRUGO | S_IWUSR);
+
+CONFIGFS_EATTR_OPS(target_core_dev_attrib, se_dev_attrib, da_group);
+
+static struct configfs_attribute *target_core_dev_attrib_attrs[] = {
+ &target_core_dev_attrib_emulate_dpo.attr,
+ &target_core_dev_attrib_emulate_fua_write.attr,
+ &target_core_dev_attrib_emulate_fua_read.attr,
+ &target_core_dev_attrib_emulate_write_cache.attr,
+ &target_core_dev_attrib_emulate_ua_intlck_ctrl.attr,
+ &target_core_dev_attrib_emulate_tas.attr,
+ &target_core_dev_attrib_emulate_tpu.attr,
+ &target_core_dev_attrib_emulate_tpws.attr,
+ &target_core_dev_attrib_enforce_pr_isids.attr,
+ &target_core_dev_attrib_hw_block_size.attr,
+ &target_core_dev_attrib_block_size.attr,
+ &target_core_dev_attrib_hw_max_sectors.attr,
+ &target_core_dev_attrib_max_sectors.attr,
+ &target_core_dev_attrib_optimal_sectors.attr,
+ &target_core_dev_attrib_hw_queue_depth.attr,
+ &target_core_dev_attrib_queue_depth.attr,
+ &target_core_dev_attrib_task_timeout.attr,
+ &target_core_dev_attrib_max_unmap_lba_count.attr,
+ &target_core_dev_attrib_max_unmap_block_desc_count.attr,
+ &target_core_dev_attrib_unmap_granularity.attr,
+ &target_core_dev_attrib_unmap_granularity_alignment.attr,
+ NULL,
+};
+
+static struct configfs_item_operations target_core_dev_attrib_ops = {
+ .show_attribute = target_core_dev_attrib_attr_show,
+ .store_attribute = target_core_dev_attrib_attr_store,
+};
+
+static struct config_item_type target_core_dev_attrib_cit = {
+ .ct_item_ops = &target_core_dev_attrib_ops,
+ .ct_attrs = target_core_dev_attrib_attrs,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_dev_attrib_cit */
+
+/* Start functions for struct config_item_type target_core_dev_wwn_cit */
+
+CONFIGFS_EATTR_STRUCT(target_core_dev_wwn, t10_wwn);
+#define SE_DEV_WWN_ATTR(_name, _mode) \
+static struct target_core_dev_wwn_attribute target_core_dev_wwn_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ target_core_dev_wwn_show_attr_##_name, \
+ target_core_dev_wwn_store_attr_##_name);
+
+#define SE_DEV_WWN_ATTR_RO(_name); \
+do { \
+ static struct target_core_dev_wwn_attribute \
+ target_core_dev_wwn_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ target_core_dev_wwn_show_attr_##_name); \
+} while (0);
+
+/*
+ * VPD page 0x80 Unit serial
+ */
+static ssize_t target_core_dev_wwn_show_attr_vpd_unit_serial(
+ struct t10_wwn *t10_wwn,
+ char *page)
+{
+ struct se_subsystem_dev *se_dev = t10_wwn->t10_sub_dev;
+ struct se_device *dev;
+
+ dev = se_dev->se_dev_ptr;
+ if (!(dev))
+ return -ENODEV;
+
+ return sprintf(page, "T10 VPD Unit Serial Number: %s\n",
+ &t10_wwn->unit_serial[0]);
+}
+
+static ssize_t target_core_dev_wwn_store_attr_vpd_unit_serial(
+ struct t10_wwn *t10_wwn,
+ const char *page,
+ size_t count)
+{
+ struct se_subsystem_dev *su_dev = t10_wwn->t10_sub_dev;
+ struct se_device *dev;
+ unsigned char buf[INQUIRY_VPD_SERIAL_LEN];
+
+ /*
+ * If Linux/SCSI subsystem_api_t plugin got a VPD Unit Serial
+ * from the struct scsi_device level firmware, do not allow
+ * VPD Unit Serial to be emulated.
+ *
+ * Note this struct scsi_device could also be emulating VPD
+ * information from its drivers/scsi LLD. But for now we assume
+ * it is doing 'the right thing' wrt a world wide unique
+ * VPD Unit Serial Number that OS dependent multipath can depend on.
+ */
+ if (su_dev->su_dev_flags & SDF_FIRMWARE_VPD_UNIT_SERIAL) {
+ printk(KERN_ERR "Underlying SCSI device firmware provided VPD"
+ " Unit Serial, ignoring request\n");
+ return -EOPNOTSUPP;
+ }
+
+ if ((strlen(page) + 1) > INQUIRY_VPD_SERIAL_LEN) {
+ printk(KERN_ERR "Emulated VPD Unit Serial exceeds"
+ " INQUIRY_VPD_SERIAL_LEN: %d\n", INQUIRY_VPD_SERIAL_LEN);
+ return -EOVERFLOW;
+ }
+ /*
+ * Check to see if any active $FABRIC_MOD exports exist. If they
+ * do exist, fail here as changing this information on the fly
+ * (underneath the initiator side OS dependent multipath code)
+ * could cause negative effects.
+ */
+ dev = su_dev->se_dev_ptr;
+ if ((dev)) {
+ if (atomic_read(&dev->dev_export_obj.obj_access_count)) {
+ printk(KERN_ERR "Unable to set VPD Unit Serial while"
+ " active %d $FABRIC_MOD exports exist\n",
+ atomic_read(&dev->dev_export_obj.obj_access_count));
+ return -EINVAL;
+ }
+ }
+ /*
+ * This currently assumes ASCII encoding for emulated VPD Unit Serial.
+ *
+ * Also, strip any newline added from the userspace
+ * echo $UUID > $TARGET/$HBA/$STORAGE_OBJECT/wwn/vpd_unit_serial
+ */
+ memset(buf, 0, INQUIRY_VPD_SERIAL_LEN);
+ snprintf(buf, INQUIRY_VPD_SERIAL_LEN, "%s", page);
+ snprintf(su_dev->t10_wwn.unit_serial, INQUIRY_VPD_SERIAL_LEN,
+ "%s", strstrip(buf));
+ su_dev->su_dev_flags |= SDF_EMULATED_VPD_UNIT_SERIAL;
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Set emulated VPD Unit Serial:"
+ " %s\n", su_dev->t10_wwn.unit_serial);
+
+ return count;
+}
+
+SE_DEV_WWN_ATTR(vpd_unit_serial, S_IRUGO | S_IWUSR);
+
+/*
+ * VPD page 0x83 Protocol Identifier
+ */
+static ssize_t target_core_dev_wwn_show_attr_vpd_protocol_identifier(
+ struct t10_wwn *t10_wwn,
+ char *page)
+{
+ struct se_subsystem_dev *se_dev = t10_wwn->t10_sub_dev;
+ struct se_device *dev;
+ struct t10_vpd *vpd;
+ unsigned char buf[VPD_TMP_BUF_SIZE];
+ ssize_t len = 0;
+
+ dev = se_dev->se_dev_ptr;
+ if (!(dev))
+ return -ENODEV;
+
+ memset(buf, 0, VPD_TMP_BUF_SIZE);
+
+ spin_lock(&t10_wwn->t10_vpd_lock);
+ list_for_each_entry(vpd, &t10_wwn->t10_vpd_list, vpd_list) {
+ if (!(vpd->protocol_identifier_set))
+ continue;
+
+ transport_dump_vpd_proto_id(vpd, buf, VPD_TMP_BUF_SIZE);
+
+ if ((len + strlen(buf) > PAGE_SIZE))
+ break;
+
+ len += sprintf(page+len, "%s", buf);
+ }
+ spin_unlock(&t10_wwn->t10_vpd_lock);
+
+ return len;
+}
+
+static ssize_t target_core_dev_wwn_store_attr_vpd_protocol_identifier(
+ struct t10_wwn *t10_wwn,
+ const char *page,
+ size_t count)
+{
+ return -ENOSYS;
+}
+
+SE_DEV_WWN_ATTR(vpd_protocol_identifier, S_IRUGO | S_IWUSR);
+
+/*
+ * Generic wrapper for dumping VPD identifiers by association.
+ */
+#define DEF_DEV_WWN_ASSOC_SHOW(_name, _assoc) \
+static ssize_t target_core_dev_wwn_show_attr_##_name( \
+ struct t10_wwn *t10_wwn, \
+ char *page) \
+{ \
+ struct se_subsystem_dev *se_dev = t10_wwn->t10_sub_dev; \
+ struct se_device *dev; \
+ struct t10_vpd *vpd; \
+ unsigned char buf[VPD_TMP_BUF_SIZE]; \
+ ssize_t len = 0; \
+ \
+ dev = se_dev->se_dev_ptr; \
+ if (!(dev)) \
+ return -ENODEV; \
+ \
+ spin_lock(&t10_wwn->t10_vpd_lock); \
+ list_for_each_entry(vpd, &t10_wwn->t10_vpd_list, vpd_list) { \
+ if (vpd->association != _assoc) \
+ continue; \
+ \
+ memset(buf, 0, VPD_TMP_BUF_SIZE); \
+ transport_dump_vpd_assoc(vpd, buf, VPD_TMP_BUF_SIZE); \
+ if ((len + strlen(buf) > PAGE_SIZE)) \
+ break; \
+ len += sprintf(page+len, "%s", buf); \
+ \
+ memset(buf, 0, VPD_TMP_BUF_SIZE); \
+ transport_dump_vpd_ident_type(vpd, buf, VPD_TMP_BUF_SIZE); \
+ if ((len + strlen(buf) > PAGE_SIZE)) \
+ break; \
+ len += sprintf(page+len, "%s", buf); \
+ \
+ memset(buf, 0, VPD_TMP_BUF_SIZE); \
+ transport_dump_vpd_ident(vpd, buf, VPD_TMP_BUF_SIZE); \
+ if ((len + strlen(buf) > PAGE_SIZE)) \
+ break; \
+ len += sprintf(page+len, "%s", buf); \
+ } \
+ spin_unlock(&t10_wwn->t10_vpd_lock); \
+ \
+ return len; \
+}
+
+/*
+ * VPD page 0x83 Assoication: Logical Unit
+ */
+DEF_DEV_WWN_ASSOC_SHOW(vpd_assoc_logical_unit, 0x00);
+
+static ssize_t target_core_dev_wwn_store_attr_vpd_assoc_logical_unit(
+ struct t10_wwn *t10_wwn,
+ const char *page,
+ size_t count)
+{
+ return -ENOSYS;
+}
+
+SE_DEV_WWN_ATTR(vpd_assoc_logical_unit, S_IRUGO | S_IWUSR);
+
+/*
+ * VPD page 0x83 Association: Target Port
+ */
+DEF_DEV_WWN_ASSOC_SHOW(vpd_assoc_target_port, 0x10);
+
+static ssize_t target_core_dev_wwn_store_attr_vpd_assoc_target_port(
+ struct t10_wwn *t10_wwn,
+ const char *page,
+ size_t count)
+{
+ return -ENOSYS;
+}
+
+SE_DEV_WWN_ATTR(vpd_assoc_target_port, S_IRUGO | S_IWUSR);
+
+/*
+ * VPD page 0x83 Association: SCSI Target Device
+ */
+DEF_DEV_WWN_ASSOC_SHOW(vpd_assoc_scsi_target_device, 0x20);
+
+static ssize_t target_core_dev_wwn_store_attr_vpd_assoc_scsi_target_device(
+ struct t10_wwn *t10_wwn,
+ const char *page,
+ size_t count)
+{
+ return -ENOSYS;
+}
+
+SE_DEV_WWN_ATTR(vpd_assoc_scsi_target_device, S_IRUGO | S_IWUSR);
+
+CONFIGFS_EATTR_OPS(target_core_dev_wwn, t10_wwn, t10_wwn_group);
+
+static struct configfs_attribute *target_core_dev_wwn_attrs[] = {
+ &target_core_dev_wwn_vpd_unit_serial.attr,
+ &target_core_dev_wwn_vpd_protocol_identifier.attr,
+ &target_core_dev_wwn_vpd_assoc_logical_unit.attr,
+ &target_core_dev_wwn_vpd_assoc_target_port.attr,
+ &target_core_dev_wwn_vpd_assoc_scsi_target_device.attr,
+ NULL,
+};
+
+static struct configfs_item_operations target_core_dev_wwn_ops = {
+ .show_attribute = target_core_dev_wwn_attr_show,
+ .store_attribute = target_core_dev_wwn_attr_store,
+};
+
+static struct config_item_type target_core_dev_wwn_cit = {
+ .ct_item_ops = &target_core_dev_wwn_ops,
+ .ct_attrs = target_core_dev_wwn_attrs,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_dev_wwn_cit */
+
+/* Start functions for struct config_item_type target_core_dev_pr_cit */
+
+CONFIGFS_EATTR_STRUCT(target_core_dev_pr, se_subsystem_dev);
+#define SE_DEV_PR_ATTR(_name, _mode) \
+static struct target_core_dev_pr_attribute target_core_dev_pr_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ target_core_dev_pr_show_attr_##_name, \
+ target_core_dev_pr_store_attr_##_name);
+
+#define SE_DEV_PR_ATTR_RO(_name); \
+static struct target_core_dev_pr_attribute target_core_dev_pr_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ target_core_dev_pr_show_attr_##_name);
+
+/*
+ * res_holder
+ */
+static ssize_t target_core_dev_pr_show_spc3_res(
+ struct se_device *dev,
+ char *page,
+ ssize_t *len)
+{
+ struct se_node_acl *se_nacl;
+ struct t10_pr_registration *pr_reg;
+ char i_buf[PR_REG_ISID_ID_LEN];
+ int prf_isid;
+
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+
+ spin_lock(&dev->dev_reservation_lock);
+ pr_reg = dev->dev_pr_res_holder;
+ if (!(pr_reg)) {
+ *len += sprintf(page + *len, "No SPC-3 Reservation holder\n");
+ spin_unlock(&dev->dev_reservation_lock);
+ return *len;
+ }
+ se_nacl = pr_reg->pr_reg_nacl;
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+
+ *len += sprintf(page + *len, "SPC-3 Reservation: %s Initiator: %s%s\n",
+ TPG_TFO(se_nacl->se_tpg)->get_fabric_name(),
+ se_nacl->initiatorname, (prf_isid) ? &i_buf[0] : "");
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return *len;
+}
+
+static ssize_t target_core_dev_pr_show_spc2_res(
+ struct se_device *dev,
+ char *page,
+ ssize_t *len)
+{
+ struct se_node_acl *se_nacl;
+
+ spin_lock(&dev->dev_reservation_lock);
+ se_nacl = dev->dev_reserved_node_acl;
+ if (!(se_nacl)) {
+ *len += sprintf(page + *len, "No SPC-2 Reservation holder\n");
+ spin_unlock(&dev->dev_reservation_lock);
+ return *len;
+ }
+ *len += sprintf(page + *len, "SPC-2 Reservation: %s Initiator: %s\n",
+ TPG_TFO(se_nacl->se_tpg)->get_fabric_name(),
+ se_nacl->initiatorname);
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return *len;
+}
+
+static ssize_t target_core_dev_pr_show_attr_res_holder(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ ssize_t len = 0;
+
+ if (!(su_dev->se_dev_ptr))
+ return -ENODEV;
+
+ switch (T10_RES(su_dev)->res_type) {
+ case SPC3_PERSISTENT_RESERVATIONS:
+ target_core_dev_pr_show_spc3_res(su_dev->se_dev_ptr,
+ page, &len);
+ break;
+ case SPC2_RESERVATIONS:
+ target_core_dev_pr_show_spc2_res(su_dev->se_dev_ptr,
+ page, &len);
+ break;
+ case SPC_PASSTHROUGH:
+ len += sprintf(page+len, "Passthrough\n");
+ break;
+ default:
+ len += sprintf(page+len, "Unknown\n");
+ break;
+ }
+
+ return len;
+}
+
+SE_DEV_PR_ATTR_RO(res_holder);
+
+/*
+ * res_pr_all_tgt_pts
+ */
+static ssize_t target_core_dev_pr_show_attr_res_pr_all_tgt_pts(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ struct se_device *dev;
+ struct t10_pr_registration *pr_reg;
+ ssize_t len = 0;
+
+ dev = su_dev->se_dev_ptr;
+ if (!(dev))
+ return -ENODEV;
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return len;
+
+ spin_lock(&dev->dev_reservation_lock);
+ pr_reg = dev->dev_pr_res_holder;
+ if (!(pr_reg)) {
+ len = sprintf(page, "No SPC-3 Reservation holder\n");
+ spin_unlock(&dev->dev_reservation_lock);
+ return len;
+ }
+ /*
+ * See All Target Ports (ALL_TG_PT) bit in spcr17, section 6.14.3
+ * Basic PERSISTENT RESERVER OUT parameter list, page 290
+ */
+ if (pr_reg->pr_reg_all_tg_pt)
+ len = sprintf(page, "SPC-3 Reservation: All Target"
+ " Ports registration\n");
+ else
+ len = sprintf(page, "SPC-3 Reservation: Single"
+ " Target Port registration\n");
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return len;
+}
+
+SE_DEV_PR_ATTR_RO(res_pr_all_tgt_pts);
+
+/*
+ * res_pr_generation
+ */
+static ssize_t target_core_dev_pr_show_attr_res_pr_generation(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ if (!(su_dev->se_dev_ptr))
+ return -ENODEV;
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return 0;
+
+ return sprintf(page, "0x%08x\n", T10_RES(su_dev)->pr_generation);
+}
+
+SE_DEV_PR_ATTR_RO(res_pr_generation);
+
+/*
+ * res_pr_holder_tg_port
+ */
+static ssize_t target_core_dev_pr_show_attr_res_pr_holder_tg_port(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ struct se_device *dev;
+ struct se_node_acl *se_nacl;
+ struct se_lun *lun;
+ struct se_portal_group *se_tpg;
+ struct t10_pr_registration *pr_reg;
+ struct target_core_fabric_ops *tfo;
+ ssize_t len = 0;
+
+ dev = su_dev->se_dev_ptr;
+ if (!(dev))
+ return -ENODEV;
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return len;
+
+ spin_lock(&dev->dev_reservation_lock);
+ pr_reg = dev->dev_pr_res_holder;
+ if (!(pr_reg)) {
+ len = sprintf(page, "No SPC-3 Reservation holder\n");
+ spin_unlock(&dev->dev_reservation_lock);
+ return len;
+ }
+ se_nacl = pr_reg->pr_reg_nacl;
+ se_tpg = se_nacl->se_tpg;
+ lun = pr_reg->pr_reg_tg_pt_lun;
+ tfo = TPG_TFO(se_tpg);
+
+ len += sprintf(page+len, "SPC-3 Reservation: %s"
+ " Target Node Endpoint: %s\n", tfo->get_fabric_name(),
+ tfo->tpg_get_wwn(se_tpg));
+ len += sprintf(page+len, "SPC-3 Reservation: Relative Port"
+ " Identifer Tag: %hu %s Portal Group Tag: %hu"
+ " %s Logical Unit: %u\n", lun->lun_sep->sep_rtpi,
+ tfo->get_fabric_name(), tfo->tpg_get_tag(se_tpg),
+ tfo->get_fabric_name(), lun->unpacked_lun);
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return len;
+}
+
+SE_DEV_PR_ATTR_RO(res_pr_holder_tg_port);
+
+/*
+ * res_pr_registered_i_pts
+ */
+static ssize_t target_core_dev_pr_show_attr_res_pr_registered_i_pts(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ struct target_core_fabric_ops *tfo;
+ struct t10_pr_registration *pr_reg;
+ unsigned char buf[384];
+ char i_buf[PR_REG_ISID_ID_LEN];
+ ssize_t len = 0;
+ int reg_count = 0, prf_isid;
+
+ if (!(su_dev->se_dev_ptr))
+ return -ENODEV;
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return len;
+
+ len += sprintf(page+len, "SPC-3 PR Registrations:\n");
+
+ spin_lock(&T10_RES(su_dev)->registration_lock);
+ list_for_each_entry(pr_reg, &T10_RES(su_dev)->registration_list,
+ pr_reg_list) {
+
+ memset(buf, 0, 384);
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+ tfo = pr_reg->pr_reg_nacl->se_tpg->se_tpg_tfo;
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+ sprintf(buf, "%s Node: %s%s Key: 0x%016Lx PRgen: 0x%08x\n",
+ tfo->get_fabric_name(),
+ pr_reg->pr_reg_nacl->initiatorname, (prf_isid) ?
+ &i_buf[0] : "", pr_reg->pr_res_key,
+ pr_reg->pr_res_generation);
+
+ if ((len + strlen(buf) > PAGE_SIZE))
+ break;
+
+ len += sprintf(page+len, "%s", buf);
+ reg_count++;
+ }
+ spin_unlock(&T10_RES(su_dev)->registration_lock);
+
+ if (!(reg_count))
+ len += sprintf(page+len, "None\n");
+
+ return len;
+}
+
+SE_DEV_PR_ATTR_RO(res_pr_registered_i_pts);
+
+/*
+ * res_pr_type
+ */
+static ssize_t target_core_dev_pr_show_attr_res_pr_type(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ struct se_device *dev;
+ struct t10_pr_registration *pr_reg;
+ ssize_t len = 0;
+
+ dev = su_dev->se_dev_ptr;
+ if (!(dev))
+ return -ENODEV;
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return len;
+
+ spin_lock(&dev->dev_reservation_lock);
+ pr_reg = dev->dev_pr_res_holder;
+ if (!(pr_reg)) {
+ len = sprintf(page, "No SPC-3 Reservation holder\n");
+ spin_unlock(&dev->dev_reservation_lock);
+ return len;
+ }
+ len = sprintf(page, "SPC-3 Reservation Type: %s\n",
+ core_scsi3_pr_dump_type(pr_reg->pr_res_type));
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return len;
+}
+
+SE_DEV_PR_ATTR_RO(res_pr_type);
+
+/*
+ * res_type
+ */
+static ssize_t target_core_dev_pr_show_attr_res_type(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ ssize_t len = 0;
+
+ if (!(su_dev->se_dev_ptr))
+ return -ENODEV;
+
+ switch (T10_RES(su_dev)->res_type) {
+ case SPC3_PERSISTENT_RESERVATIONS:
+ len = sprintf(page, "SPC3_PERSISTENT_RESERVATIONS\n");
+ break;
+ case SPC2_RESERVATIONS:
+ len = sprintf(page, "SPC2_RESERVATIONS\n");
+ break;
+ case SPC_PASSTHROUGH:
+ len = sprintf(page, "SPC_PASSTHROUGH\n");
+ break;
+ default:
+ len = sprintf(page, "UNKNOWN\n");
+ break;
+ }
+
+ return len;
+}
+
+SE_DEV_PR_ATTR_RO(res_type);
+
+/*
+ * res_aptpl_active
+ */
+
+static ssize_t target_core_dev_pr_show_attr_res_aptpl_active(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ if (!(su_dev->se_dev_ptr))
+ return -ENODEV;
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return 0;
+
+ return sprintf(page, "APTPL Bit Status: %s\n",
+ (T10_RES(su_dev)->pr_aptpl_active) ? "Activated" : "Disabled");
+}
+
+SE_DEV_PR_ATTR_RO(res_aptpl_active);
+
+/*
+ * res_aptpl_metadata
+ */
+static ssize_t target_core_dev_pr_show_attr_res_aptpl_metadata(
+ struct se_subsystem_dev *su_dev,
+ char *page)
+{
+ if (!(su_dev->se_dev_ptr))
+ return -ENODEV;
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return 0;
+
+ return sprintf(page, "Ready to process PR APTPL metadata..\n");
+}
+
+enum {
+ Opt_initiator_fabric, Opt_initiator_node, Opt_initiator_sid,
+ Opt_sa_res_key, Opt_res_holder, Opt_res_type, Opt_res_scope,
+ Opt_res_all_tg_pt, Opt_mapped_lun, Opt_target_fabric,
+ Opt_target_node, Opt_tpgt, Opt_port_rtpi, Opt_target_lun, Opt_err
+};
+
+static match_table_t tokens = {
+ {Opt_initiator_fabric, "initiator_fabric=%s"},
+ {Opt_initiator_node, "initiator_node=%s"},
+ {Opt_initiator_sid, "initiator_sid=%s"},
+ {Opt_sa_res_key, "sa_res_key=%s"},
+ {Opt_res_holder, "res_holder=%d"},
+ {Opt_res_type, "res_type=%d"},
+ {Opt_res_scope, "res_scope=%d"},
+ {Opt_res_all_tg_pt, "res_all_tg_pt=%d"},
+ {Opt_mapped_lun, "mapped_lun=%d"},
+ {Opt_target_fabric, "target_fabric=%s"},
+ {Opt_target_node, "target_node=%s"},
+ {Opt_tpgt, "tpgt=%d"},
+ {Opt_port_rtpi, "port_rtpi=%d"},
+ {Opt_target_lun, "target_lun=%d"},
+ {Opt_err, NULL}
+};
+
+static ssize_t target_core_dev_pr_store_attr_res_aptpl_metadata(
+ struct se_subsystem_dev *su_dev,
+ const char *page,
+ size_t count)
+{
+ struct se_device *dev;
+ unsigned char *i_fabric, *t_fabric, *i_port = NULL, *t_port = NULL;
+ unsigned char *isid = NULL;
+ char *orig, *ptr, *arg_p, *opts;
+ substring_t args[MAX_OPT_ARGS];
+ unsigned long long tmp_ll;
+ u64 sa_res_key = 0;
+ u32 mapped_lun = 0, target_lun = 0;
+ int ret = -1, res_holder = 0, all_tg_pt = 0, arg, token;
+ u16 port_rpti = 0, tpgt = 0;
+ u8 type = 0, scope;
+
+ dev = su_dev->se_dev_ptr;
+ if (!(dev))
+ return -ENODEV;
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return 0;
+
+ if (atomic_read(&dev->dev_export_obj.obj_access_count)) {
+ printk(KERN_INFO "Unable to process APTPL metadata while"
+ " active fabric exports exist\n");
+ return -EINVAL;
+ }
+
+ opts = kstrdup(page, GFP_KERNEL);
+ if (!opts)
+ return -ENOMEM;
+
+ orig = opts;
+ while ((ptr = strsep(&opts, ",")) != NULL) {
+ if (!*ptr)
+ continue;
+
+ token = match_token(ptr, tokens, args);
+ switch (token) {
+ case Opt_initiator_fabric:
+ i_fabric = match_strdup(&args[0]);
+ break;
+ case Opt_initiator_node:
+ i_port = match_strdup(&args[0]);
+ if (strlen(i_port) > PR_APTPL_MAX_IPORT_LEN) {
+ printk(KERN_ERR "APTPL metadata initiator_node="
+ " exceeds PR_APTPL_MAX_IPORT_LEN: %d\n",
+ PR_APTPL_MAX_IPORT_LEN);
+ ret = -EINVAL;
+ break;
+ }
+ break;
+ case Opt_initiator_sid:
+ isid = match_strdup(&args[0]);
+ if (strlen(isid) > PR_REG_ISID_LEN) {
+ printk(KERN_ERR "APTPL metadata initiator_isid"
+ "= exceeds PR_REG_ISID_LEN: %d\n",
+ PR_REG_ISID_LEN);
+ ret = -EINVAL;
+ break;
+ }
+ break;
+ case Opt_sa_res_key:
+ arg_p = match_strdup(&args[0]);
+ ret = strict_strtoull(arg_p, 0, &tmp_ll);
+ if (ret < 0) {
+ printk(KERN_ERR "strict_strtoull() failed for"
+ " sa_res_key=\n");
+ goto out;
+ }
+ sa_res_key = (u64)tmp_ll;
+ break;
+ /*
+ * PR APTPL Metadata for Reservation
+ */
+ case Opt_res_holder:
+ match_int(args, &arg);
+ res_holder = arg;
+ break;
+ case Opt_res_type:
+ match_int(args, &arg);
+ type = (u8)arg;
+ break;
+ case Opt_res_scope:
+ match_int(args, &arg);
+ scope = (u8)arg;
+ break;
+ case Opt_res_all_tg_pt:
+ match_int(args, &arg);
+ all_tg_pt = (int)arg;
+ break;
+ case Opt_mapped_lun:
+ match_int(args, &arg);
+ mapped_lun = (u32)arg;
+ break;
+ /*
+ * PR APTPL Metadata for Target Port
+ */
+ case Opt_target_fabric:
+ t_fabric = match_strdup(&args[0]);
+ break;
+ case Opt_target_node:
+ t_port = match_strdup(&args[0]);
+ if (strlen(t_port) > PR_APTPL_MAX_TPORT_LEN) {
+ printk(KERN_ERR "APTPL metadata target_node="
+ " exceeds PR_APTPL_MAX_TPORT_LEN: %d\n",
+ PR_APTPL_MAX_TPORT_LEN);
+ ret = -EINVAL;
+ break;
+ }
+ break;
+ case Opt_tpgt:
+ match_int(args, &arg);
+ tpgt = (u16)arg;
+ break;
+ case Opt_port_rtpi:
+ match_int(args, &arg);
+ port_rpti = (u16)arg;
+ break;
+ case Opt_target_lun:
+ match_int(args, &arg);
+ target_lun = (u32)arg;
+ break;
+ default:
+ break;
+ }
+ }
+
+ if (!(i_port) || !(t_port) || !(sa_res_key)) {
+ printk(KERN_ERR "Illegal parameters for APTPL registration\n");
+ ret = -EINVAL;
+ goto out;
+ }
+
+ if (res_holder && !(type)) {
+ printk(KERN_ERR "Illegal PR type: 0x%02x for reservation"
+ " holder\n", type);
+ ret = -EINVAL;
+ goto out;
+ }
+
+ ret = core_scsi3_alloc_aptpl_registration(T10_RES(su_dev), sa_res_key,
+ i_port, isid, mapped_lun, t_port, tpgt, target_lun,
+ res_holder, all_tg_pt, type);
+out:
+ kfree(orig);
+ return (ret == 0) ? count : ret;
+}
+
+SE_DEV_PR_ATTR(res_aptpl_metadata, S_IRUGO | S_IWUSR);
+
+CONFIGFS_EATTR_OPS(target_core_dev_pr, se_subsystem_dev, se_dev_pr_group);
+
+static struct configfs_attribute *target_core_dev_pr_attrs[] = {
+ &target_core_dev_pr_res_holder.attr,
+ &target_core_dev_pr_res_pr_all_tgt_pts.attr,
+ &target_core_dev_pr_res_pr_generation.attr,
+ &target_core_dev_pr_res_pr_holder_tg_port.attr,
+ &target_core_dev_pr_res_pr_registered_i_pts.attr,
+ &target_core_dev_pr_res_pr_type.attr,
+ &target_core_dev_pr_res_type.attr,
+ &target_core_dev_pr_res_aptpl_active.attr,
+ &target_core_dev_pr_res_aptpl_metadata.attr,
+ NULL,
+};
+
+static struct configfs_item_operations target_core_dev_pr_ops = {
+ .show_attribute = target_core_dev_pr_attr_show,
+ .store_attribute = target_core_dev_pr_attr_store,
+};
+
+static struct config_item_type target_core_dev_pr_cit = {
+ .ct_item_ops = &target_core_dev_pr_ops,
+ .ct_attrs = target_core_dev_pr_attrs,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_dev_pr_cit */
+
+/* Start functions for struct config_item_type target_core_dev_cit */
+
+static ssize_t target_core_show_dev_info(void *p, char *page)
+{
+ struct se_subsystem_dev *se_dev = (struct se_subsystem_dev *)p;
+ struct se_hba *hba = se_dev->se_dev_hba;
+ struct se_subsystem_api *t = hba->transport;
+ int bl = 0;
+ ssize_t read_bytes = 0;
+
+ if (!(se_dev->se_dev_ptr))
+ return -ENODEV;
+
+ transport_dump_dev_state(se_dev->se_dev_ptr, page, &bl);
+ read_bytes += bl;
+ read_bytes += t->show_configfs_dev_params(hba, se_dev, page+read_bytes);
+ return read_bytes;
+}
+
+static struct target_core_configfs_attribute target_core_attr_dev_info = {
+ .attr = { .ca_owner = THIS_MODULE,
+ .ca_name = "info",
+ .ca_mode = S_IRUGO },
+ .show = target_core_show_dev_info,
+ .store = NULL,
+};
+
+static ssize_t target_core_store_dev_control(
+ void *p,
+ const char *page,
+ size_t count)
+{
+ struct se_subsystem_dev *se_dev = (struct se_subsystem_dev *)p;
+ struct se_hba *hba = se_dev->se_dev_hba;
+ struct se_subsystem_api *t = hba->transport;
+
+ if (!(se_dev->se_dev_su_ptr)) {
+ printk(KERN_ERR "Unable to locate struct se_subsystem_dev>se"
+ "_dev_su_ptr\n");
+ return -EINVAL;
+ }
+
+ return t->set_configfs_dev_params(hba, se_dev, page, count);
+}
+
+static struct target_core_configfs_attribute target_core_attr_dev_control = {
+ .attr = { .ca_owner = THIS_MODULE,
+ .ca_name = "control",
+ .ca_mode = S_IWUSR },
+ .show = NULL,
+ .store = target_core_store_dev_control,
+};
+
+static ssize_t target_core_show_dev_alias(void *p, char *page)
+{
+ struct se_subsystem_dev *se_dev = (struct se_subsystem_dev *)p;
+
+ if (!(se_dev->su_dev_flags & SDF_USING_ALIAS))
+ return 0;
+
+ return snprintf(page, PAGE_SIZE, "%s\n", se_dev->se_dev_alias);
+}
+
+static ssize_t target_core_store_dev_alias(
+ void *p,
+ const char *page,
+ size_t count)
+{
+ struct se_subsystem_dev *se_dev = (struct se_subsystem_dev *)p;
+ struct se_hba *hba = se_dev->se_dev_hba;
+ ssize_t read_bytes;
+
+ if (count > (SE_DEV_ALIAS_LEN-1)) {
+ printk(KERN_ERR "alias count: %d exceeds"
+ " SE_DEV_ALIAS_LEN-1: %u\n", (int)count,
+ SE_DEV_ALIAS_LEN-1);
+ return -EINVAL;
+ }
+
+ se_dev->su_dev_flags |= SDF_USING_ALIAS;
+ read_bytes = snprintf(&se_dev->se_dev_alias[0], SE_DEV_ALIAS_LEN,
+ "%s", page);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: %s/%s set alias: %s\n",
+ config_item_name(&hba->hba_group.cg_item),
+ config_item_name(&se_dev->se_dev_group.cg_item),
+ se_dev->se_dev_alias);
+
+ return read_bytes;
+}
+
+static struct target_core_configfs_attribute target_core_attr_dev_alias = {
+ .attr = { .ca_owner = THIS_MODULE,
+ .ca_name = "alias",
+ .ca_mode = S_IRUGO | S_IWUSR },
+ .show = target_core_show_dev_alias,
+ .store = target_core_store_dev_alias,
+};
+
+static ssize_t target_core_show_dev_udev_path(void *p, char *page)
+{
+ struct se_subsystem_dev *se_dev = (struct se_subsystem_dev *)p;
+
+ if (!(se_dev->su_dev_flags & SDF_USING_UDEV_PATH))
+ return 0;
+
+ return snprintf(page, PAGE_SIZE, "%s\n", se_dev->se_dev_udev_path);
+}
+
+static ssize_t target_core_store_dev_udev_path(
+ void *p,
+ const char *page,
+ size_t count)
+{
+ struct se_subsystem_dev *se_dev = (struct se_subsystem_dev *)p;
+ struct se_hba *hba = se_dev->se_dev_hba;
+ ssize_t read_bytes;
+
+ if (count > (SE_UDEV_PATH_LEN-1)) {
+ printk(KERN_ERR "udev_path count: %d exceeds"
+ " SE_UDEV_PATH_LEN-1: %u\n", (int)count,
+ SE_UDEV_PATH_LEN-1);
+ return -EINVAL;
+ }
+
+ se_dev->su_dev_flags |= SDF_USING_UDEV_PATH;
+ read_bytes = snprintf(&se_dev->se_dev_udev_path[0], SE_UDEV_PATH_LEN,
+ "%s", page);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: %s/%s set udev_path: %s\n",
+ config_item_name(&hba->hba_group.cg_item),
+ config_item_name(&se_dev->se_dev_group.cg_item),
+ se_dev->se_dev_udev_path);
+
+ return read_bytes;
+}
+
+static struct target_core_configfs_attribute target_core_attr_dev_udev_path = {
+ .attr = { .ca_owner = THIS_MODULE,
+ .ca_name = "udev_path",
+ .ca_mode = S_IRUGO | S_IWUSR },
+ .show = target_core_show_dev_udev_path,
+ .store = target_core_store_dev_udev_path,
+};
+
+static ssize_t target_core_store_dev_enable(
+ void *p,
+ const char *page,
+ size_t count)
+{
+ struct se_subsystem_dev *se_dev = (struct se_subsystem_dev *)p;
+ struct se_device *dev;
+ struct se_hba *hba = se_dev->se_dev_hba;
+ struct se_subsystem_api *t = hba->transport;
+ char *ptr;
+
+ ptr = strstr(page, "1");
+ if (!(ptr)) {
+ printk(KERN_ERR "For dev_enable ops, only valid value"
+ " is \"1\"\n");
+ return -EINVAL;
+ }
+ if ((se_dev->se_dev_ptr)) {
+ printk(KERN_ERR "se_dev->se_dev_ptr already set for storage"
+ " object\n");
+ return -EEXIST;
+ }
+
+ if (t->check_configfs_dev_params(hba, se_dev) < 0)
+ return -EINVAL;
+
+ dev = t->create_virtdevice(hba, se_dev, se_dev->se_dev_su_ptr);
+ if (!(dev) || IS_ERR(dev))
+ return -EINVAL;
+
+ se_dev->se_dev_ptr = dev;
+ printk(KERN_INFO "Target_Core_ConfigFS: Registered se_dev->se_dev_ptr:"
+ " %p\n", se_dev->se_dev_ptr);
+
+ return count;
+}
+
+static struct target_core_configfs_attribute target_core_attr_dev_enable = {
+ .attr = { .ca_owner = THIS_MODULE,
+ .ca_name = "enable",
+ .ca_mode = S_IWUSR },
+ .show = NULL,
+ .store = target_core_store_dev_enable,
+};
+
+static ssize_t target_core_show_alua_lu_gp(void *p, char *page)
+{
+ struct se_device *dev;
+ struct se_subsystem_dev *su_dev = (struct se_subsystem_dev *)p;
+ struct config_item *lu_ci;
+ struct t10_alua_lu_gp *lu_gp;
+ struct t10_alua_lu_gp_member *lu_gp_mem;
+ ssize_t len = 0;
+
+ dev = su_dev->se_dev_ptr;
+ if (!(dev))
+ return -ENODEV;
+
+ if (T10_ALUA(su_dev)->alua_type != SPC3_ALUA_EMULATED)
+ return len;
+
+ lu_gp_mem = dev->dev_alua_lu_gp_mem;
+ if (!(lu_gp_mem)) {
+ printk(KERN_ERR "NULL struct se_device->dev_alua_lu_gp_mem"
+ " pointer\n");
+ return -EINVAL;
+ }
+
+ spin_lock(&lu_gp_mem->lu_gp_mem_lock);
+ lu_gp = lu_gp_mem->lu_gp;
+ if ((lu_gp)) {
+ lu_ci = &lu_gp->lu_gp_group.cg_item;
+ len += sprintf(page, "LU Group Alias: %s\nLU Group ID: %hu\n",
+ config_item_name(lu_ci), lu_gp->lu_gp_id);
+ }
+ spin_unlock(&lu_gp_mem->lu_gp_mem_lock);
+
+ return len;
+}
+
+static ssize_t target_core_store_alua_lu_gp(
+ void *p,
+ const char *page,
+ size_t count)
+{
+ struct se_device *dev;
+ struct se_subsystem_dev *su_dev = (struct se_subsystem_dev *)p;
+ struct se_hba *hba = su_dev->se_dev_hba;
+ struct t10_alua_lu_gp *lu_gp = NULL, *lu_gp_new = NULL;
+ struct t10_alua_lu_gp_member *lu_gp_mem;
+ unsigned char buf[LU_GROUP_NAME_BUF];
+ int move = 0;
+
+ dev = su_dev->se_dev_ptr;
+ if (!(dev))
+ return -ENODEV;
+
+ if (T10_ALUA(su_dev)->alua_type != SPC3_ALUA_EMULATED) {
+ printk(KERN_WARNING "SPC3_ALUA_EMULATED not enabled for %s/%s\n",
+ config_item_name(&hba->hba_group.cg_item),
+ config_item_name(&su_dev->se_dev_group.cg_item));
+ return -EINVAL;
+ }
+ if (count > LU_GROUP_NAME_BUF) {
+ printk(KERN_ERR "ALUA LU Group Alias too large!\n");
+ return -EINVAL;
+ }
+ memset(buf, 0, LU_GROUP_NAME_BUF);
+ memcpy(buf, page, count);
+ /*
+ * Any ALUA logical unit alias besides "NULL" means we will be
+ * making a new group association.
+ */
+ if (strcmp(strstrip(buf), "NULL")) {
+ /*
+ * core_alua_get_lu_gp_by_name() will increment reference to
+ * struct t10_alua_lu_gp. This reference is released with
+ * core_alua_get_lu_gp_by_name below().
+ */
+ lu_gp_new = core_alua_get_lu_gp_by_name(strstrip(buf));
+ if (!(lu_gp_new))
+ return -ENODEV;
+ }
+ lu_gp_mem = dev->dev_alua_lu_gp_mem;
+ if (!(lu_gp_mem)) {
+ if (lu_gp_new)
+ core_alua_put_lu_gp_from_name(lu_gp_new);
+ printk(KERN_ERR "NULL struct se_device->dev_alua_lu_gp_mem"
+ " pointer\n");
+ return -EINVAL;
+ }
+
+ spin_lock(&lu_gp_mem->lu_gp_mem_lock);
+ lu_gp = lu_gp_mem->lu_gp;
+ if ((lu_gp)) {
+ /*
+ * Clearing an existing lu_gp association, and replacing
+ * with NULL
+ */
+ if (!(lu_gp_new)) {
+ printk(KERN_INFO "Target_Core_ConfigFS: Releasing %s/%s"
+ " from ALUA LU Group: core/alua/lu_gps/%s, ID:"
+ " %hu\n",
+ config_item_name(&hba->hba_group.cg_item),
+ config_item_name(&su_dev->se_dev_group.cg_item),
+ config_item_name(&lu_gp->lu_gp_group.cg_item),
+ lu_gp->lu_gp_id);
+
+ __core_alua_drop_lu_gp_mem(lu_gp_mem, lu_gp);
+ spin_unlock(&lu_gp_mem->lu_gp_mem_lock);
+
+ return count;
+ }
+ /*
+ * Removing existing association of lu_gp_mem with lu_gp
+ */
+ __core_alua_drop_lu_gp_mem(lu_gp_mem, lu_gp);
+ move = 1;
+ }
+ /*
+ * Associate lu_gp_mem with lu_gp_new.
+ */
+ __core_alua_attach_lu_gp_mem(lu_gp_mem, lu_gp_new);
+ spin_unlock(&lu_gp_mem->lu_gp_mem_lock);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: %s %s/%s to ALUA LU Group:"
+ " core/alua/lu_gps/%s, ID: %hu\n",
+ (move) ? "Moving" : "Adding",
+ config_item_name(&hba->hba_group.cg_item),
+ config_item_name(&su_dev->se_dev_group.cg_item),
+ config_item_name(&lu_gp_new->lu_gp_group.cg_item),
+ lu_gp_new->lu_gp_id);
+
+ core_alua_put_lu_gp_from_name(lu_gp_new);
+ return count;
+}
+
+static struct target_core_configfs_attribute target_core_attr_dev_alua_lu_gp = {
+ .attr = { .ca_owner = THIS_MODULE,
+ .ca_name = "alua_lu_gp",
+ .ca_mode = S_IRUGO | S_IWUSR },
+ .show = target_core_show_alua_lu_gp,
+ .store = target_core_store_alua_lu_gp,
+};
+
+static struct configfs_attribute *lio_core_dev_attrs[] = {
+ &target_core_attr_dev_info.attr,
+ &target_core_attr_dev_control.attr,
+ &target_core_attr_dev_alias.attr,
+ &target_core_attr_dev_udev_path.attr,
+ &target_core_attr_dev_enable.attr,
+ &target_core_attr_dev_alua_lu_gp.attr,
+ NULL,
+};
+
+static void target_core_dev_release(struct config_item *item)
+{
+ struct se_subsystem_dev *se_dev = container_of(to_config_group(item),
+ struct se_subsystem_dev, se_dev_group);
+ struct config_group *dev_cg;
+
+ if (!(se_dev))
+ return;
+
+ dev_cg = &se_dev->se_dev_group;
+ kfree(dev_cg->default_groups);
+}
+
+static ssize_t target_core_dev_show(struct config_item *item,
+ struct configfs_attribute *attr,
+ char *page)
+{
+ struct se_subsystem_dev *se_dev = container_of(
+ to_config_group(item), struct se_subsystem_dev,
+ se_dev_group);
+ struct target_core_configfs_attribute *tc_attr = container_of(
+ attr, struct target_core_configfs_attribute, attr);
+
+ if (!(tc_attr->show))
+ return -EINVAL;
+
+ return tc_attr->show((void *)se_dev, page);
+}
+
+static ssize_t target_core_dev_store(struct config_item *item,
+ struct configfs_attribute *attr,
+ const char *page, size_t count)
+{
+ struct se_subsystem_dev *se_dev = container_of(
+ to_config_group(item), struct se_subsystem_dev,
+ se_dev_group);
+ struct target_core_configfs_attribute *tc_attr = container_of(
+ attr, struct target_core_configfs_attribute, attr);
+
+ if (!(tc_attr->store))
+ return -EINVAL;
+
+ return tc_attr->store((void *)se_dev, page, count);
+}
+
+static struct configfs_item_operations target_core_dev_item_ops = {
+ .release = target_core_dev_release,
+ .show_attribute = target_core_dev_show,
+ .store_attribute = target_core_dev_store,
+};
+
+static struct config_item_type target_core_dev_cit = {
+ .ct_item_ops = &target_core_dev_item_ops,
+ .ct_attrs = lio_core_dev_attrs,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_dev_cit */
+
+/* Start functions for struct config_item_type target_core_alua_lu_gp_cit */
+
+CONFIGFS_EATTR_STRUCT(target_core_alua_lu_gp, t10_alua_lu_gp);
+#define SE_DEV_ALUA_LU_ATTR(_name, _mode) \
+static struct target_core_alua_lu_gp_attribute \
+ target_core_alua_lu_gp_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ target_core_alua_lu_gp_show_attr_##_name, \
+ target_core_alua_lu_gp_store_attr_##_name);
+
+#define SE_DEV_ALUA_LU_ATTR_RO(_name) \
+static struct target_core_alua_lu_gp_attribute \
+ target_core_alua_lu_gp_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ target_core_alua_lu_gp_show_attr_##_name);
+
+/*
+ * lu_gp_id
+ */
+static ssize_t target_core_alua_lu_gp_show_attr_lu_gp_id(
+ struct t10_alua_lu_gp *lu_gp,
+ char *page)
+{
+ if (!(lu_gp->lu_gp_valid_id))
+ return 0;
+
+ return sprintf(page, "%hu\n", lu_gp->lu_gp_id);
+}
+
+static ssize_t target_core_alua_lu_gp_store_attr_lu_gp_id(
+ struct t10_alua_lu_gp *lu_gp,
+ const char *page,
+ size_t count)
+{
+ struct config_group *alua_lu_gp_cg = &lu_gp->lu_gp_group;
+ unsigned long lu_gp_id;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &lu_gp_id);
+ if (ret < 0) {
+ printk(KERN_ERR "strict_strtoul() returned %d for"
+ " lu_gp_id\n", ret);
+ return -EINVAL;
+ }
+ if (lu_gp_id > 0x0000ffff) {
+ printk(KERN_ERR "ALUA lu_gp_id: %lu exceeds maximum:"
+ " 0x0000ffff\n", lu_gp_id);
+ return -EINVAL;
+ }
+
+ ret = core_alua_set_lu_gp_id(lu_gp, (u16)lu_gp_id);
+ if (ret < 0)
+ return -EINVAL;
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Set ALUA Logical Unit"
+ " Group: core/alua/lu_gps/%s to ID: %hu\n",
+ config_item_name(&alua_lu_gp_cg->cg_item),
+ lu_gp->lu_gp_id);
+
+ return count;
+}
+
+SE_DEV_ALUA_LU_ATTR(lu_gp_id, S_IRUGO | S_IWUSR);
+
+/*
+ * members
+ */
+static ssize_t target_core_alua_lu_gp_show_attr_members(
+ struct t10_alua_lu_gp *lu_gp,
+ char *page)
+{
+ struct se_device *dev;
+ struct se_hba *hba;
+ struct se_subsystem_dev *su_dev;
+ struct t10_alua_lu_gp_member *lu_gp_mem;
+ ssize_t len = 0, cur_len;
+ unsigned char buf[LU_GROUP_NAME_BUF];
+
+ memset(buf, 0, LU_GROUP_NAME_BUF);
+
+ spin_lock(&lu_gp->lu_gp_lock);
+ list_for_each_entry(lu_gp_mem, &lu_gp->lu_gp_mem_list, lu_gp_mem_list) {
+ dev = lu_gp_mem->lu_gp_mem_dev;
+ su_dev = dev->se_sub_dev;
+ hba = su_dev->se_dev_hba;
+
+ cur_len = snprintf(buf, LU_GROUP_NAME_BUF, "%s/%s\n",
+ config_item_name(&hba->hba_group.cg_item),
+ config_item_name(&su_dev->se_dev_group.cg_item));
+ cur_len++; /* Extra byte for NULL terminator */
+
+ if ((cur_len + len) > PAGE_SIZE) {
+ printk(KERN_WARNING "Ran out of lu_gp_show_attr"
+ "_members buffer\n");
+ break;
+ }
+ memcpy(page+len, buf, cur_len);
+ len += cur_len;
+ }
+ spin_unlock(&lu_gp->lu_gp_lock);
+
+ return len;
+}
+
+SE_DEV_ALUA_LU_ATTR_RO(members);
+
+CONFIGFS_EATTR_OPS(target_core_alua_lu_gp, t10_alua_lu_gp, lu_gp_group);
+
+static struct configfs_attribute *target_core_alua_lu_gp_attrs[] = {
+ &target_core_alua_lu_gp_lu_gp_id.attr,
+ &target_core_alua_lu_gp_members.attr,
+ NULL,
+};
+
+static struct configfs_item_operations target_core_alua_lu_gp_ops = {
+ .show_attribute = target_core_alua_lu_gp_attr_show,
+ .store_attribute = target_core_alua_lu_gp_attr_store,
+};
+
+static struct config_item_type target_core_alua_lu_gp_cit = {
+ .ct_item_ops = &target_core_alua_lu_gp_ops,
+ .ct_attrs = target_core_alua_lu_gp_attrs,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_alua_lu_gp_cit */
+
+/* Start functions for struct config_item_type target_core_alua_lu_gps_cit */
+
+static struct config_group *target_core_alua_create_lu_gp(
+ struct config_group *group,
+ const char *name)
+{
+ struct t10_alua_lu_gp *lu_gp;
+ struct config_group *alua_lu_gp_cg = NULL;
+ struct config_item *alua_lu_gp_ci = NULL;
+
+ lu_gp = core_alua_allocate_lu_gp(name, 0);
+ if (IS_ERR(lu_gp))
+ return NULL;
+
+ alua_lu_gp_cg = &lu_gp->lu_gp_group;
+ alua_lu_gp_ci = &alua_lu_gp_cg->cg_item;
+
+ config_group_init_type_name(alua_lu_gp_cg, name,
+ &target_core_alua_lu_gp_cit);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Allocated ALUA Logical Unit"
+ " Group: core/alua/lu_gps/%s\n",
+ config_item_name(alua_lu_gp_ci));
+
+ return alua_lu_gp_cg;
+
+}
+
+static void target_core_alua_drop_lu_gp(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct t10_alua_lu_gp *lu_gp = container_of(to_config_group(item),
+ struct t10_alua_lu_gp, lu_gp_group);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Releasing ALUA Logical Unit"
+ " Group: core/alua/lu_gps/%s, ID: %hu\n",
+ config_item_name(item), lu_gp->lu_gp_id);
+
+ config_item_put(item);
+ core_alua_free_lu_gp(lu_gp);
+}
+
+static struct configfs_group_operations target_core_alua_lu_gps_group_ops = {
+ .make_group = &target_core_alua_create_lu_gp,
+ .drop_item = &target_core_alua_drop_lu_gp,
+};
+
+static struct config_item_type target_core_alua_lu_gps_cit = {
+ .ct_item_ops = NULL,
+ .ct_group_ops = &target_core_alua_lu_gps_group_ops,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_alua_lu_gps_cit */
+
+/* Start functions for struct config_item_type target_core_alua_tg_pt_gp_cit */
+
+CONFIGFS_EATTR_STRUCT(target_core_alua_tg_pt_gp, t10_alua_tg_pt_gp);
+#define SE_DEV_ALUA_TG_PT_ATTR(_name, _mode) \
+static struct target_core_alua_tg_pt_gp_attribute \
+ target_core_alua_tg_pt_gp_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ target_core_alua_tg_pt_gp_show_attr_##_name, \
+ target_core_alua_tg_pt_gp_store_attr_##_name);
+
+#define SE_DEV_ALUA_TG_PT_ATTR_RO(_name) \
+static struct target_core_alua_tg_pt_gp_attribute \
+ target_core_alua_tg_pt_gp_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ target_core_alua_tg_pt_gp_show_attr_##_name);
+
+/*
+ * alua_access_state
+ */
+static ssize_t target_core_alua_tg_pt_gp_show_attr_alua_access_state(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return sprintf(page, "%d\n",
+ atomic_read(&tg_pt_gp->tg_pt_gp_alua_access_state));
+}
+
+static ssize_t target_core_alua_tg_pt_gp_store_attr_alua_access_state(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ struct se_subsystem_dev *su_dev = tg_pt_gp->tg_pt_gp_su_dev;
+ unsigned long tmp;
+ int new_state, ret;
+
+ if (!(tg_pt_gp->tg_pt_gp_valid_id)) {
+ printk(KERN_ERR "Unable to do implict ALUA on non valid"
+ " tg_pt_gp ID: %hu\n", tg_pt_gp->tg_pt_gp_valid_id);
+ return -EINVAL;
+ }
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk("Unable to extract new ALUA access state from"
+ " %s\n", page);
+ return -EINVAL;
+ }
+ new_state = (int)tmp;
+
+ if (!(tg_pt_gp->tg_pt_gp_alua_access_type & TPGS_IMPLICT_ALUA)) {
+ printk(KERN_ERR "Unable to process implict configfs ALUA"
+ " transition while TPGS_IMPLICT_ALUA is diabled\n");
+ return -EINVAL;
+ }
+
+ ret = core_alua_do_port_transition(tg_pt_gp, su_dev->se_dev_ptr,
+ NULL, NULL, new_state, 0);
+ return (!ret) ? count : -EINVAL;
+}
+
+SE_DEV_ALUA_TG_PT_ATTR(alua_access_state, S_IRUGO | S_IWUSR);
+
+/*
+ * alua_access_status
+ */
+static ssize_t target_core_alua_tg_pt_gp_show_attr_alua_access_status(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return sprintf(page, "%s\n",
+ core_alua_dump_status(tg_pt_gp->tg_pt_gp_alua_access_status));
+}
+
+static ssize_t target_core_alua_tg_pt_gp_store_attr_alua_access_status(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ unsigned long tmp;
+ int new_status, ret;
+
+ if (!(tg_pt_gp->tg_pt_gp_valid_id)) {
+ printk(KERN_ERR "Unable to do set ALUA access status on non"
+ " valid tg_pt_gp ID: %hu\n",
+ tg_pt_gp->tg_pt_gp_valid_id);
+ return -EINVAL;
+ }
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract new ALUA access status"
+ " from %s\n", page);
+ return -EINVAL;
+ }
+ new_status = (int)tmp;
+
+ if ((new_status != ALUA_STATUS_NONE) &&
+ (new_status != ALUA_STATUS_ALTERED_BY_EXPLICT_STPG) &&
+ (new_status != ALUA_STATUS_ALTERED_BY_IMPLICT_ALUA)) {
+ printk(KERN_ERR "Illegal ALUA access status: 0x%02x\n",
+ new_status);
+ return -EINVAL;
+ }
+
+ tg_pt_gp->tg_pt_gp_alua_access_status = new_status;
+ return count;
+}
+
+SE_DEV_ALUA_TG_PT_ATTR(alua_access_status, S_IRUGO | S_IWUSR);
+
+/*
+ * alua_access_type
+ */
+static ssize_t target_core_alua_tg_pt_gp_show_attr_alua_access_type(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return core_alua_show_access_type(tg_pt_gp, page);
+}
+
+static ssize_t target_core_alua_tg_pt_gp_store_attr_alua_access_type(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ return core_alua_store_access_type(tg_pt_gp, page, count);
+}
+
+SE_DEV_ALUA_TG_PT_ATTR(alua_access_type, S_IRUGO | S_IWUSR);
+
+/*
+ * alua_write_metadata
+ */
+static ssize_t target_core_alua_tg_pt_gp_show_attr_alua_write_metadata(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return sprintf(page, "%d\n", tg_pt_gp->tg_pt_gp_write_metadata);
+}
+
+static ssize_t target_core_alua_tg_pt_gp_store_attr_alua_write_metadata(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ unsigned long tmp;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract alua_write_metadata\n");
+ return -EINVAL;
+ }
+
+ if ((tmp != 0) && (tmp != 1)) {
+ printk(KERN_ERR "Illegal value for alua_write_metadata:"
+ " %lu\n", tmp);
+ return -EINVAL;
+ }
+ tg_pt_gp->tg_pt_gp_write_metadata = (int)tmp;
+
+ return count;
+}
+
+SE_DEV_ALUA_TG_PT_ATTR(alua_write_metadata, S_IRUGO | S_IWUSR);
+
+
+
+/*
+ * nonop_delay_msecs
+ */
+static ssize_t target_core_alua_tg_pt_gp_show_attr_nonop_delay_msecs(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return core_alua_show_nonop_delay_msecs(tg_pt_gp, page);
+
+}
+
+static ssize_t target_core_alua_tg_pt_gp_store_attr_nonop_delay_msecs(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ return core_alua_store_nonop_delay_msecs(tg_pt_gp, page, count);
+}
+
+SE_DEV_ALUA_TG_PT_ATTR(nonop_delay_msecs, S_IRUGO | S_IWUSR);
+
+/*
+ * trans_delay_msecs
+ */
+static ssize_t target_core_alua_tg_pt_gp_show_attr_trans_delay_msecs(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return core_alua_show_trans_delay_msecs(tg_pt_gp, page);
+}
+
+static ssize_t target_core_alua_tg_pt_gp_store_attr_trans_delay_msecs(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ return core_alua_store_trans_delay_msecs(tg_pt_gp, page, count);
+}
+
+SE_DEV_ALUA_TG_PT_ATTR(trans_delay_msecs, S_IRUGO | S_IWUSR);
+
+/*
+ * preferred
+ */
+
+static ssize_t target_core_alua_tg_pt_gp_show_attr_preferred(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ return core_alua_show_preferred_bit(tg_pt_gp, page);
+}
+
+static ssize_t target_core_alua_tg_pt_gp_store_attr_preferred(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ return core_alua_store_preferred_bit(tg_pt_gp, page, count);
+}
+
+SE_DEV_ALUA_TG_PT_ATTR(preferred, S_IRUGO | S_IWUSR);
+
+/*
+ * tg_pt_gp_id
+ */
+static ssize_t target_core_alua_tg_pt_gp_show_attr_tg_pt_gp_id(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ if (!(tg_pt_gp->tg_pt_gp_valid_id))
+ return 0;
+
+ return sprintf(page, "%hu\n", tg_pt_gp->tg_pt_gp_id);
+}
+
+static ssize_t target_core_alua_tg_pt_gp_store_attr_tg_pt_gp_id(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ const char *page,
+ size_t count)
+{
+ struct config_group *alua_tg_pt_gp_cg = &tg_pt_gp->tg_pt_gp_group;
+ unsigned long tg_pt_gp_id;
+ int ret;
+
+ ret = strict_strtoul(page, 0, &tg_pt_gp_id);
+ if (ret < 0) {
+ printk(KERN_ERR "strict_strtoul() returned %d for"
+ " tg_pt_gp_id\n", ret);
+ return -EINVAL;
+ }
+ if (tg_pt_gp_id > 0x0000ffff) {
+ printk(KERN_ERR "ALUA tg_pt_gp_id: %lu exceeds maximum:"
+ " 0x0000ffff\n", tg_pt_gp_id);
+ return -EINVAL;
+ }
+
+ ret = core_alua_set_tg_pt_gp_id(tg_pt_gp, (u16)tg_pt_gp_id);
+ if (ret < 0)
+ return -EINVAL;
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Set ALUA Target Port Group: "
+ "core/alua/tg_pt_gps/%s to ID: %hu\n",
+ config_item_name(&alua_tg_pt_gp_cg->cg_item),
+ tg_pt_gp->tg_pt_gp_id);
+
+ return count;
+}
+
+SE_DEV_ALUA_TG_PT_ATTR(tg_pt_gp_id, S_IRUGO | S_IWUSR);
+
+/*
+ * members
+ */
+static ssize_t target_core_alua_tg_pt_gp_show_attr_members(
+ struct t10_alua_tg_pt_gp *tg_pt_gp,
+ char *page)
+{
+ struct se_port *port;
+ struct se_portal_group *tpg;
+ struct se_lun *lun;
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem;
+ ssize_t len = 0, cur_len;
+ unsigned char buf[TG_PT_GROUP_NAME_BUF];
+
+ memset(buf, 0, TG_PT_GROUP_NAME_BUF);
+
+ spin_lock(&tg_pt_gp->tg_pt_gp_lock);
+ list_for_each_entry(tg_pt_gp_mem, &tg_pt_gp->tg_pt_gp_mem_list,
+ tg_pt_gp_mem_list) {
+ port = tg_pt_gp_mem->tg_pt;
+ tpg = port->sep_tpg;
+ lun = port->sep_lun;
+
+ cur_len = snprintf(buf, TG_PT_GROUP_NAME_BUF, "%s/%s/tpgt_%hu"
+ "/%s\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_wwn(tpg),
+ TPG_TFO(tpg)->tpg_get_tag(tpg),
+ config_item_name(&lun->lun_group.cg_item));
+ cur_len++; /* Extra byte for NULL terminator */
+
+ if ((cur_len + len) > PAGE_SIZE) {
+ printk(KERN_WARNING "Ran out of lu_gp_show_attr"
+ "_members buffer\n");
+ break;
+ }
+ memcpy(page+len, buf, cur_len);
+ len += cur_len;
+ }
+ spin_unlock(&tg_pt_gp->tg_pt_gp_lock);
+
+ return len;
+}
+
+SE_DEV_ALUA_TG_PT_ATTR_RO(members);
+
+CONFIGFS_EATTR_OPS(target_core_alua_tg_pt_gp, t10_alua_tg_pt_gp,
+ tg_pt_gp_group);
+
+static struct configfs_attribute *target_core_alua_tg_pt_gp_attrs[] = {
+ &target_core_alua_tg_pt_gp_alua_access_state.attr,
+ &target_core_alua_tg_pt_gp_alua_access_status.attr,
+ &target_core_alua_tg_pt_gp_alua_access_type.attr,
+ &target_core_alua_tg_pt_gp_alua_write_metadata.attr,
+ &target_core_alua_tg_pt_gp_nonop_delay_msecs.attr,
+ &target_core_alua_tg_pt_gp_trans_delay_msecs.attr,
+ &target_core_alua_tg_pt_gp_preferred.attr,
+ &target_core_alua_tg_pt_gp_tg_pt_gp_id.attr,
+ &target_core_alua_tg_pt_gp_members.attr,
+ NULL,
+};
+
+static struct configfs_item_operations target_core_alua_tg_pt_gp_ops = {
+ .show_attribute = target_core_alua_tg_pt_gp_attr_show,
+ .store_attribute = target_core_alua_tg_pt_gp_attr_store,
+};
+
+static struct config_item_type target_core_alua_tg_pt_gp_cit = {
+ .ct_item_ops = &target_core_alua_tg_pt_gp_ops,
+ .ct_attrs = target_core_alua_tg_pt_gp_attrs,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_alua_tg_pt_gp_cit */
+
+/* Start functions for struct config_item_type target_core_alua_tg_pt_gps_cit */
+
+static struct config_group *target_core_alua_create_tg_pt_gp(
+ struct config_group *group,
+ const char *name)
+{
+ struct t10_alua *alua = container_of(group, struct t10_alua,
+ alua_tg_pt_gps_group);
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct se_subsystem_dev *su_dev = alua->t10_sub_dev;
+ struct config_group *alua_tg_pt_gp_cg = NULL;
+ struct config_item *alua_tg_pt_gp_ci = NULL;
+
+ tg_pt_gp = core_alua_allocate_tg_pt_gp(su_dev, name, 0);
+ if (!(tg_pt_gp))
+ return NULL;
+
+ alua_tg_pt_gp_cg = &tg_pt_gp->tg_pt_gp_group;
+ alua_tg_pt_gp_ci = &alua_tg_pt_gp_cg->cg_item;
+
+ config_group_init_type_name(alua_tg_pt_gp_cg, name,
+ &target_core_alua_tg_pt_gp_cit);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Allocated ALUA Target Port"
+ " Group: alua/tg_pt_gps/%s\n",
+ config_item_name(alua_tg_pt_gp_ci));
+
+ return alua_tg_pt_gp_cg;
+}
+
+static void target_core_alua_drop_tg_pt_gp(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct t10_alua_tg_pt_gp *tg_pt_gp = container_of(to_config_group(item),
+ struct t10_alua_tg_pt_gp, tg_pt_gp_group);
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Releasing ALUA Target Port"
+ " Group: alua/tg_pt_gps/%s, ID: %hu\n",
+ config_item_name(item), tg_pt_gp->tg_pt_gp_id);
+
+ config_item_put(item);
+ core_alua_free_tg_pt_gp(tg_pt_gp);
+}
+
+static struct configfs_group_operations target_core_alua_tg_pt_gps_group_ops = {
+ .make_group = &target_core_alua_create_tg_pt_gp,
+ .drop_item = &target_core_alua_drop_tg_pt_gp,
+};
+
+static struct config_item_type target_core_alua_tg_pt_gps_cit = {
+ .ct_group_ops = &target_core_alua_tg_pt_gps_group_ops,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_alua_tg_pt_gps_cit */
+
+/* Start functions for struct config_item_type target_core_alua_cit */
+
+/*
+ * target_core_alua_cit is a ConfigFS group that lives under
+ * /sys/kernel/config/target/core/alua. There are default groups
+ * core/alua/lu_gps and core/alua/tg_pt_gps that are attached to
+ * target_core_alua_cit in target_core_init_configfs() below.
+ */
+static struct config_item_type target_core_alua_cit = {
+ .ct_item_ops = NULL,
+ .ct_attrs = NULL,
+ .ct_owner = THIS_MODULE,
+};
+
+/* End functions for struct config_item_type target_core_alua_cit */
+
+/* Start functions for struct config_item_type target_core_hba_cit */
+
+static struct config_group *target_core_make_subdev(
+ struct config_group *group,
+ const char *name)
+{
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct se_subsystem_dev *se_dev;
+ struct se_subsystem_api *t;
+ struct config_item *hba_ci = &group->cg_item;
+ struct se_hba *hba = item_to_hba(hba_ci);
+ struct config_group *dev_cg = NULL, *tg_pt_gp_cg = NULL;
+
+ if (mutex_lock_interruptible(&hba->hba_access_mutex))
+ return NULL;
+
+ /*
+ * Locate the struct se_subsystem_api from parent's struct se_hba.
+ */
+ t = hba->transport;
+
+ se_dev = kzalloc(sizeof(struct se_subsystem_dev), GFP_KERNEL);
+ if (!se_dev) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " struct se_subsystem_dev\n");
+ goto unlock;
+ }
+ INIT_LIST_HEAD(&se_dev->g_se_dev_list);
+ 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_reservation.registration_list);
+ INIT_LIST_HEAD(&se_dev->t10_reservation.aptpl_reg_list);
+ spin_lock_init(&se_dev->t10_reservation.registration_lock);
+ spin_lock_init(&se_dev->t10_reservation.aptpl_reg_lock);
+ INIT_LIST_HEAD(&se_dev->t10_alua.tg_pt_gps_list);
+ spin_lock_init(&se_dev->t10_alua.tg_pt_gps_lock);
+ spin_lock_init(&se_dev->se_dev_lock);
+ se_dev->t10_reservation.pr_aptpl_buf_len = PR_APTPL_BUF_LEN;
+ se_dev->t10_wwn.t10_sub_dev = se_dev;
+ se_dev->t10_alua.t10_sub_dev = se_dev;
+ se_dev->se_dev_attrib.da_sub_dev = se_dev;
+
+ se_dev->se_dev_hba = hba;
+ dev_cg = &se_dev->se_dev_group;
+
+ dev_cg->default_groups = kzalloc(sizeof(struct config_group) * 6,
+ GFP_KERNEL);
+ if (!(dev_cg->default_groups))
+ goto out;
+ /*
+ * Set se_dev_su_ptr from struct se_subsystem_api returned void ptr
+ * for ->allocate_virtdevice()
+ *
+ * se_dev->se_dev_ptr will be set after ->create_virtdev()
+ * has been called successfully in the next level up in the
+ * configfs tree for device object's struct config_group.
+ */
+ se_dev->se_dev_su_ptr = t->allocate_virtdevice(hba, name);
+ if (!(se_dev->se_dev_su_ptr)) {
+ printk(KERN_ERR "Unable to locate subsystem dependent pointer"
+ " from allocate_virtdevice()\n");
+ goto out;
+ }
+ spin_lock(&se_global->g_device_lock);
+ list_add_tail(&se_dev->g_se_dev_list, &se_global->g_se_dev_list);
+ spin_unlock(&se_global->g_device_lock);
+
+ config_group_init_type_name(&se_dev->se_dev_group, name,
+ &target_core_dev_cit);
+ config_group_init_type_name(&se_dev->se_dev_attrib.da_group, "attrib",
+ &target_core_dev_attrib_cit);
+ config_group_init_type_name(&se_dev->se_dev_pr_group, "pr",
+ &target_core_dev_pr_cit);
+ config_group_init_type_name(&se_dev->t10_wwn.t10_wwn_group, "wwn",
+ &target_core_dev_wwn_cit);
+ config_group_init_type_name(&se_dev->t10_alua.alua_tg_pt_gps_group,
+ "alua", &target_core_alua_tg_pt_gps_cit);
+ dev_cg->default_groups[0] = &se_dev->se_dev_attrib.da_group;
+ dev_cg->default_groups[1] = &se_dev->se_dev_pr_group;
+ dev_cg->default_groups[2] = &se_dev->t10_wwn.t10_wwn_group;
+ dev_cg->default_groups[3] = &se_dev->t10_alua.alua_tg_pt_gps_group;
+ dev_cg->default_groups[4] = NULL;
+ /*
+ * Add core/$HBA/$DEV/alua/tg_pt_gps/default_tg_pt_gp
+ */
+ tg_pt_gp = core_alua_allocate_tg_pt_gp(se_dev, "default_tg_pt_gp", 1);
+ if (!(tg_pt_gp))
+ goto out;
+
+ tg_pt_gp_cg = &T10_ALUA(se_dev)->alua_tg_pt_gps_group;
+ tg_pt_gp_cg->default_groups = kzalloc(sizeof(struct config_group) * 2,
+ GFP_KERNEL);
+ if (!(tg_pt_gp_cg->default_groups)) {
+ printk(KERN_ERR "Unable to allocate tg_pt_gp_cg->"
+ "default_groups\n");
+ goto out;
+ }
+
+ config_group_init_type_name(&tg_pt_gp->tg_pt_gp_group,
+ "default_tg_pt_gp", &target_core_alua_tg_pt_gp_cit);
+ tg_pt_gp_cg->default_groups[0] = &tg_pt_gp->tg_pt_gp_group;
+ tg_pt_gp_cg->default_groups[1] = NULL;
+ T10_ALUA(se_dev)->default_tg_pt_gp = tg_pt_gp;
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Allocated struct se_subsystem_dev:"
+ " %p se_dev_su_ptr: %p\n", se_dev, se_dev->se_dev_su_ptr);
+
+ mutex_unlock(&hba->hba_access_mutex);
+ return &se_dev->se_dev_group;
+out:
+ if (T10_ALUA(se_dev)->default_tg_pt_gp) {
+ core_alua_free_tg_pt_gp(T10_ALUA(se_dev)->default_tg_pt_gp);
+ T10_ALUA(se_dev)->default_tg_pt_gp = NULL;
+ }
+ if (tg_pt_gp_cg)
+ kfree(tg_pt_gp_cg->default_groups);
+ if (dev_cg)
+ kfree(dev_cg->default_groups);
+ if (se_dev->se_dev_su_ptr)
+ t->free_device(se_dev->se_dev_su_ptr);
+ kfree(se_dev);
+unlock:
+ mutex_unlock(&hba->hba_access_mutex);
+ return NULL;
+}
+
+static void target_core_drop_subdev(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct se_subsystem_dev *se_dev = container_of(to_config_group(item),
+ struct se_subsystem_dev, se_dev_group);
+ struct se_hba *hba;
+ struct se_subsystem_api *t;
+ struct config_item *df_item;
+ struct config_group *dev_cg, *tg_pt_gp_cg;
+ int i, ret;
+
+ hba = item_to_hba(&se_dev->se_dev_hba->hba_group.cg_item);
+
+ if (mutex_lock_interruptible(&hba->hba_access_mutex))
+ goto out;
+
+ t = hba->transport;
+
+ spin_lock(&se_global->g_device_lock);
+ list_del(&se_dev->g_se_dev_list);
+ spin_unlock(&se_global->g_device_lock);
+
+ tg_pt_gp_cg = &T10_ALUA(se_dev)->alua_tg_pt_gps_group;
+ for (i = 0; tg_pt_gp_cg->default_groups[i]; i++) {
+ df_item = &tg_pt_gp_cg->default_groups[i]->cg_item;
+ tg_pt_gp_cg->default_groups[i] = NULL;
+ config_item_put(df_item);
+ }
+ kfree(tg_pt_gp_cg->default_groups);
+ core_alua_free_tg_pt_gp(T10_ALUA(se_dev)->default_tg_pt_gp);
+ T10_ALUA(se_dev)->default_tg_pt_gp = NULL;
+
+ dev_cg = &se_dev->se_dev_group;
+ for (i = 0; dev_cg->default_groups[i]; i++) {
+ df_item = &dev_cg->default_groups[i]->cg_item;
+ dev_cg->default_groups[i] = NULL;
+ config_item_put(df_item);
+ }
+
+ config_item_put(item);
+ /*
+ * This pointer will set when the storage is enabled with:
+ * `echo 1 > $CONFIGFS/core/$HBA/$DEV/dev_enable`
+ */
+ if (se_dev->se_dev_ptr) {
+ printk(KERN_INFO "Target_Core_ConfigFS: Calling se_free_"
+ "virtual_device() for se_dev_ptr: %p\n",
+ se_dev->se_dev_ptr);
+
+ ret = se_free_virtual_device(se_dev->se_dev_ptr, hba);
+ if (ret < 0)
+ goto hba_out;
+ } else {
+ /*
+ * Release struct se_subsystem_dev->se_dev_su_ptr..
+ */
+ printk(KERN_INFO "Target_Core_ConfigFS: Calling t->free_"
+ "device() for se_dev_su_ptr: %p\n",
+ se_dev->se_dev_su_ptr);
+
+ t->free_device(se_dev->se_dev_su_ptr);
+ }
+
+ printk(KERN_INFO "Target_Core_ConfigFS: Deallocating se_subsystem"
+ "_dev_t: %p\n", se_dev);
+
+hba_out:
+ mutex_unlock(&hba->hba_access_mutex);
+out:
+ kfree(se_dev);
+}
+
+static struct configfs_group_operations target_core_hba_group_ops = {
+ .make_group = target_core_make_subdev,
+ .drop_item = target_core_drop_subdev,
+};
+
+CONFIGFS_EATTR_STRUCT(target_core_hba, se_hba);
+#define SE_HBA_ATTR(_name, _mode) \
+static struct target_core_hba_attribute \
+ target_core_hba_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ target_core_hba_show_attr_##_name, \
+ target_core_hba_store_attr_##_name);
+
+#define SE_HBA_ATTR_RO(_name) \
+static struct target_core_hba_attribute \
+ target_core_hba_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ target_core_hba_show_attr_##_name);
+
+static ssize_t target_core_hba_show_attr_hba_info(
+ struct se_hba *hba,
+ char *page)
+{
+ return sprintf(page, "HBA Index: %d plugin: %s version: %s\n",
+ hba->hba_id, hba->transport->name,
+ TARGET_CORE_CONFIGFS_VERSION);
+}
+
+SE_HBA_ATTR_RO(hba_info);
+
+static ssize_t target_core_hba_show_attr_hba_mode(struct se_hba *hba,
+ char *page)
+{
+ int hba_mode = 0;
+
+ if (hba->hba_flags & HBA_FLAGS_PSCSI_MODE)
+ hba_mode = 1;
+
+ return sprintf(page, "%d\n", hba_mode);
+}
+
+static ssize_t target_core_hba_store_attr_hba_mode(struct se_hba *hba,
+ const char *page, size_t count)
+{
+ struct se_subsystem_api *transport = hba->transport;
+ unsigned long mode_flag;
+ int ret;
+
+ if (transport->pmode_enable_hba == NULL)
+ return -EINVAL;
+
+ ret = strict_strtoul(page, 0, &mode_flag);
+ if (ret < 0) {
+ printk(KERN_ERR "Unable to extract hba mode flag: %d\n", ret);
+ return -EINVAL;
+ }
+
+ spin_lock(&hba->device_lock);
+ if (!(list_empty(&hba->hba_dev_list))) {
+ printk(KERN_ERR "Unable to set hba_mode with active devices\n");
+ spin_unlock(&hba->device_lock);
+ return -EINVAL;
+ }
+ spin_unlock(&hba->device_lock);
+
+ ret = transport->pmode_enable_hba(hba, mode_flag);
+ if (ret < 0)
+ return -EINVAL;
+ if (ret > 0)
+ hba->hba_flags |= HBA_FLAGS_PSCSI_MODE;
+ else if (ret == 0)
+ hba->hba_flags &= ~HBA_FLAGS_PSCSI_MODE;
+
+ return count;
+}
+
+SE_HBA_ATTR(hba_mode, S_IRUGO | S_IWUSR);
+
+CONFIGFS_EATTR_OPS(target_core_hba, se_hba, hba_group);
+
+static struct configfs_attribute *target_core_hba_attrs[] = {
+ &target_core_hba_hba_info.attr,
+ &target_core_hba_hba_mode.attr,
+ NULL,
+};
+
+static struct configfs_item_operations target_core_hba_item_ops = {
+ .show_attribute = target_core_hba_attr_show,
+ .store_attribute = target_core_hba_attr_store,
+};
+
+static struct config_item_type target_core_hba_cit = {
+ .ct_item_ops = &target_core_hba_item_ops,
+ .ct_group_ops = &target_core_hba_group_ops,
+ .ct_attrs = target_core_hba_attrs,
+ .ct_owner = THIS_MODULE,
+};
+
+static struct config_group *target_core_call_addhbatotarget(
+ struct config_group *group,
+ const char *name)
+{
+ char *se_plugin_str, *str, *str2;
+ struct se_hba *hba;
+ char buf[TARGET_CORE_NAME_MAX_LEN];
+ unsigned long plugin_dep_id = 0;
+ int ret;
+
+ memset(buf, 0, TARGET_CORE_NAME_MAX_LEN);
+ if (strlen(name) > TARGET_CORE_NAME_MAX_LEN) {
+ printk(KERN_ERR "Passed *name strlen(): %d exceeds"
+ " TARGET_CORE_NAME_MAX_LEN: %d\n", (int)strlen(name),
+ TARGET_CORE_NAME_MAX_LEN);
+ return ERR_PTR(-ENAMETOOLONG);
+ }
+ snprintf(buf, TARGET_CORE_NAME_MAX_LEN, "%s", name);
+
+ str = strstr(buf, "_");
+ if (!(str)) {
+ printk(KERN_ERR "Unable to locate \"_\" for $SUBSYSTEM_PLUGIN_$HOST_ID\n");
+ return ERR_PTR(-EINVAL);
+ }
+ se_plugin_str = buf;
+ /*
+ * Special case for subsystem plugins that have "_" in their names.
+ * Namely rd_direct and rd_mcp..
+ */
+ str2 = strstr(str+1, "_");
+ if ((str2)) {
+ *str2 = '\0'; /* Terminate for *se_plugin_str */
+ str2++; /* Skip to start of plugin dependent ID */
+ str = str2;
+ } else {
+ *str = '\0'; /* Terminate for *se_plugin_str */
+ str++; /* Skip to start of plugin dependent ID */
+ }
+
+ ret = strict_strtoul(str, 0, &plugin_dep_id);
+ if (ret < 0) {
+ printk(KERN_ERR "strict_strtoul() returned %d for"
+ " plugin_dep_id\n", ret);
+ return ERR_PTR(-EINVAL);
+ }
+ /*
+ * Load up TCM subsystem plugins if they have not already been loaded.
+ */
+ if (transport_subsystem_check_init() < 0)
+ return ERR_PTR(-EINVAL);
+
+ hba = core_alloc_hba(se_plugin_str, plugin_dep_id, 0);
+ if (IS_ERR(hba))
+ return ERR_CAST(hba);
+
+ config_group_init_type_name(&hba->hba_group, name,
+ &target_core_hba_cit);
+
+ return &hba->hba_group;
+}
+
+static void target_core_call_delhbafromtarget(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct se_hba *hba = item_to_hba(item);
+
+ config_item_put(item);
+ core_delete_hba(hba);
+}
+
+static struct configfs_group_operations target_core_group_ops = {
+ .make_group = target_core_call_addhbatotarget,
+ .drop_item = target_core_call_delhbafromtarget,
+};
+
+static struct config_item_type target_core_cit = {
+ .ct_item_ops = NULL,
+ .ct_group_ops = &target_core_group_ops,
+ .ct_attrs = NULL,
+ .ct_owner = THIS_MODULE,
+};
+
+/* Stop functions for struct config_item_type target_core_hba_cit */
+
+static int target_core_init_configfs(void)
+{
+ struct config_group *target_cg, *hba_cg = NULL, *alua_cg = NULL;
+ struct config_group *lu_gp_cg = NULL;
+ struct configfs_subsystem *subsys;
+ struct proc_dir_entry *scsi_target_proc = NULL;
+ struct t10_alua_lu_gp *lu_gp;
+ int ret;
+
+ printk(KERN_INFO "TARGET_CORE[0]: Loading Generic Kernel Storage"
+ " Engine: %s on %s/%s on "UTS_RELEASE"\n",
+ TARGET_CORE_VERSION, utsname()->sysname, utsname()->machine);
+
+ subsys = target_core_subsystem[0];
+ config_group_init(&subsys->su_group);
+ mutex_init(&subsys->su_mutex);
+
+ INIT_LIST_HEAD(&g_tf_list);
+ mutex_init(&g_tf_lock);
+ init_scsi_index_table();
+ ret = init_se_global();
+ if (ret < 0)
+ return -1;
+ /*
+ * Create $CONFIGFS/target/core default group for HBA <-> Storage Object
+ * and ALUA Logical Unit Group and Target Port Group infrastructure.
+ */
+ target_cg = &subsys->su_group;
+ target_cg->default_groups = kzalloc(sizeof(struct config_group) * 2,
+ GFP_KERNEL);
+ if (!(target_cg->default_groups)) {
+ printk(KERN_ERR "Unable to allocate target_cg->default_groups\n");
+ goto out_global;
+ }
+
+ config_group_init_type_name(&se_global->target_core_hbagroup,
+ "core", &target_core_cit);
+ target_cg->default_groups[0] = &se_global->target_core_hbagroup;
+ target_cg->default_groups[1] = NULL;
+ /*
+ * Create ALUA infrastructure under /sys/kernel/config/target/core/alua/
+ */
+ hba_cg = &se_global->target_core_hbagroup;
+ hba_cg->default_groups = kzalloc(sizeof(struct config_group) * 2,
+ GFP_KERNEL);
+ if (!(hba_cg->default_groups)) {
+ printk(KERN_ERR "Unable to allocate hba_cg->default_groups\n");
+ goto out_global;
+ }
+ config_group_init_type_name(&se_global->alua_group,
+ "alua", &target_core_alua_cit);
+ hba_cg->default_groups[0] = &se_global->alua_group;
+ hba_cg->default_groups[1] = NULL;
+ /*
+ * Add ALUA Logical Unit Group and Target Port Group ConfigFS
+ * groups under /sys/kernel/config/target/core/alua/
+ */
+ alua_cg = &se_global->alua_group;
+ alua_cg->default_groups = kzalloc(sizeof(struct config_group) * 2,
+ GFP_KERNEL);
+ if (!(alua_cg->default_groups)) {
+ printk(KERN_ERR "Unable to allocate alua_cg->default_groups\n");
+ goto out_global;
+ }
+
+ config_group_init_type_name(&se_global->alua_lu_gps_group,
+ "lu_gps", &target_core_alua_lu_gps_cit);
+ alua_cg->default_groups[0] = &se_global->alua_lu_gps_group;
+ alua_cg->default_groups[1] = NULL;
+ /*
+ * Add core/alua/lu_gps/default_lu_gp
+ */
+ lu_gp = core_alua_allocate_lu_gp("default_lu_gp", 1);
+ if (IS_ERR(lu_gp))
+ goto out_global;
+
+ lu_gp_cg = &se_global->alua_lu_gps_group;
+ lu_gp_cg->default_groups = kzalloc(sizeof(struct config_group) * 2,
+ GFP_KERNEL);
+ if (!(lu_gp_cg->default_groups)) {
+ printk(KERN_ERR "Unable to allocate lu_gp_cg->default_groups\n");
+ goto out_global;
+ }
+
+ config_group_init_type_name(&lu_gp->lu_gp_group, "default_lu_gp",
+ &target_core_alua_lu_gp_cit);
+ lu_gp_cg->default_groups[0] = &lu_gp->lu_gp_group;
+ lu_gp_cg->default_groups[1] = NULL;
+ se_global->default_lu_gp = lu_gp;
+ /*
+ * Register the target_core_mod subsystem with configfs.
+ */
+ ret = configfs_register_subsystem(subsys);
+ if (ret < 0) {
+ printk(KERN_ERR "Error %d while registering subsystem %s\n",
+ ret, subsys->su_group.cg_item.ci_namebuf);
+ goto out_global;
+ }
+ printk(KERN_INFO "TARGET_CORE[0]: Initialized ConfigFS Fabric"
+ " Infrastructure: "TARGET_CORE_CONFIGFS_VERSION" on %s/%s"
+ " on "UTS_RELEASE"\n", utsname()->sysname, utsname()->machine);
+ /*
+ * Register built-in RAMDISK subsystem logic for virtual LUN 0
+ */
+ ret = rd_module_init();
+ if (ret < 0)
+ goto out;
+
+ if (core_dev_setup_virtual_lun0() < 0)
+ goto out;
+
+ scsi_target_proc = proc_mkdir("scsi_target", 0);
+ if (!(scsi_target_proc)) {
+ printk(KERN_ERR "proc_mkdir(scsi_target, 0) failed\n");
+ goto out;
+ }
+ ret = init_scsi_target_mib();
+ if (ret < 0)
+ goto out;
+
+ return 0;
+
+out:
+ configfs_unregister_subsystem(subsys);
+ if (scsi_target_proc)
+ remove_proc_entry("scsi_target", 0);
+ core_dev_release_virtual_lun0();
+ rd_module_exit();
+out_global:
+ if (se_global->default_lu_gp) {
+ core_alua_free_lu_gp(se_global->default_lu_gp);
+ se_global->default_lu_gp = NULL;
+ }
+ if (lu_gp_cg)
+ kfree(lu_gp_cg->default_groups);
+ if (alua_cg)
+ kfree(alua_cg->default_groups);
+ if (hba_cg)
+ kfree(hba_cg->default_groups);
+ kfree(target_cg->default_groups);
+ release_se_global();
+ return -1;
+}
+
+static void target_core_exit_configfs(void)
+{
+ struct configfs_subsystem *subsys;
+ struct config_group *hba_cg, *alua_cg, *lu_gp_cg;
+ struct config_item *item;
+ int i;
+
+ se_global->in_shutdown = 1;
+ subsys = target_core_subsystem[0];
+
+ lu_gp_cg = &se_global->alua_lu_gps_group;
+ for (i = 0; lu_gp_cg->default_groups[i]; i++) {
+ item = &lu_gp_cg->default_groups[i]->cg_item;
+ lu_gp_cg->default_groups[i] = NULL;
+ config_item_put(item);
+ }
+ kfree(lu_gp_cg->default_groups);
+ core_alua_free_lu_gp(se_global->default_lu_gp);
+ se_global->default_lu_gp = NULL;
+
+ alua_cg = &se_global->alua_group;
+ for (i = 0; alua_cg->default_groups[i]; i++) {
+ item = &alua_cg->default_groups[i]->cg_item;
+ alua_cg->default_groups[i] = NULL;
+ config_item_put(item);
+ }
+ kfree(alua_cg->default_groups);
+
+ hba_cg = &se_global->target_core_hbagroup;
+ for (i = 0; hba_cg->default_groups[i]; i++) {
+ item = &hba_cg->default_groups[i]->cg_item;
+ hba_cg->default_groups[i] = NULL;
+ config_item_put(item);
+ }
+ kfree(hba_cg->default_groups);
+
+ for (i = 0; subsys->su_group.default_groups[i]; i++) {
+ item = &subsys->su_group.default_groups[i]->cg_item;
+ subsys->su_group.default_groups[i] = NULL;
+ config_item_put(item);
+ }
+ kfree(subsys->su_group.default_groups);
+
+ configfs_unregister_subsystem(subsys);
+ printk(KERN_INFO "TARGET_CORE[0]: Released ConfigFS Fabric"
+ " Infrastructure\n");
+
+ remove_scsi_target_mib();
+ remove_proc_entry("scsi_target", 0);
+ core_dev_release_virtual_lun0();
+ rd_module_exit();
+ release_se_global();
+
+ return;
+}
+
+MODULE_DESCRIPTION("Target_Core_Mod/ConfigFS");
+MODULE_AUTHOR("nab@Linux-iSCSI.org");
+MODULE_LICENSE("GPL");
+
+module_init(target_core_init_configfs);
+module_exit(target_core_exit_configfs);
diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
new file mode 100644
index 0000000..317ce58
--- /dev/null
+++ b/drivers/target/target_core_device.c
@@ -0,0 +1,1694 @@
+/*******************************************************************************
+ * Filename: target_core_device.c (based on iscsi_target_device.c)
+ *
+ * This file contains the iSCSI Virtual Device and Disk Transport
+ * agnostic related functions.
+ *
+ * Copyright (c) 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005-2006 SBE, Inc. All Rights Reserved.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/net.h>
+#include <linux/string.h>
+#include <linux/delay.h>
+#include <linux/timer.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <linux/kthread.h>
+#include <linux/in.h>
+#include <net/sock.h>
+#include <net/tcp.h>
+#include <scsi/scsi.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_tpg.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+
+#include "target_core_alua.h"
+#include "target_core_hba.h"
+#include "target_core_pr.h"
+#include "target_core_ua.h"
+
+static void se_dev_start(struct se_device *dev);
+static void se_dev_stop(struct se_device *dev);
+
+int transport_get_lun_for_cmd(
+ struct se_cmd *se_cmd,
+ unsigned char *cdb,
+ u32 unpacked_lun)
+{
+ struct se_dev_entry *deve;
+ struct se_lun *se_lun = NULL;
+ struct se_session *se_sess = SE_SESS(se_cmd);
+ unsigned long flags;
+ int read_only = 0;
+
+ spin_lock_irq(&SE_NODE_ACL(se_sess)->device_list_lock);
+ deve = se_cmd->se_deve =
+ &SE_NODE_ACL(se_sess)->device_list[unpacked_lun];
+ if (deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS) {
+ if (se_cmd) {
+ deve->total_cmds++;
+ deve->total_bytes += se_cmd->data_length;
+
+ if (se_cmd->data_direction == DMA_TO_DEVICE) {
+ if (deve->lun_flags &
+ TRANSPORT_LUNFLAGS_READ_ONLY) {
+ read_only = 1;
+ goto out;
+ }
+ deve->write_bytes += se_cmd->data_length;
+ } else if (se_cmd->data_direction ==
+ DMA_FROM_DEVICE) {
+ deve->read_bytes += se_cmd->data_length;
+ }
+ }
+ deve->deve_cmds++;
+
+ se_lun = 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_LUN(se_cmd)->lun_se_dev;
+ se_cmd->se_cmd_flags |= SCF_SE_LUN_CMD;
+ }
+out:
+ spin_unlock_irq(&SE_NODE_ACL(se_sess)->device_list_lock);
+
+ if (!se_lun) {
+ if (read_only) {
+ se_cmd->scsi_sense_reason = TCM_WRITE_PROTECTED;
+ se_cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ printk("TARGET_CORE[%s]: Detected WRITE_PROTECTED LUN"
+ " Access for 0x%08x\n",
+ CMD_TFO(se_cmd)->get_fabric_name(),
+ unpacked_lun);
+ return -1;
+ } else {
+ /*
+ * Use the se_portal_group->tpg_virt_lun0 to allow for
+ * REPORT_LUNS, et al to be returned when no active
+ * MappedLUN=0 exists for this Initiator Port.
+ */
+ if (unpacked_lun != 0) {
+ se_cmd->scsi_sense_reason = TCM_NON_EXISTENT_LUN;
+ se_cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ printk("TARGET_CORE[%s]: Detected NON_EXISTENT_LUN"
+ " Access for 0x%08x\n",
+ CMD_TFO(se_cmd)->get_fabric_name(),
+ unpacked_lun);
+ return -1;
+ }
+ /*
+ * Force WRITE PROTECT for virtual LUN 0
+ */
+ if ((se_cmd->data_direction != DMA_FROM_DEVICE) &&
+ (se_cmd->data_direction != DMA_NONE)) {
+ se_cmd->scsi_sense_reason = TCM_WRITE_PROTECTED;
+ se_cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ return -1;
+ }
+#if 0
+ printk("TARGET_CORE[%s]: Using virtual LUN0! :-)\n",
+ CMD_TFO(se_cmd)->get_fabric_name());
+#endif
+ se_lun = se_cmd->se_lun = &se_sess->se_tpg->tpg_virt_lun0;
+ se_cmd->orig_fe_lun = 0;
+ se_cmd->se_orig_obj_ptr = SE_LUN(se_cmd)->lun_se_dev;
+ se_cmd->se_cmd_flags |= SCF_SE_LUN_CMD;
+ }
+ }
+ /*
+ * Determine if the struct se_lun is online.
+ */
+/* #warning FIXME: Check for LUN_RESET + UNIT Attention */
+ if (se_dev_check_online(se_lun->lun_se_dev) != 0) {
+ se_cmd->scsi_sense_reason = TCM_NON_EXISTENT_LUN;
+ se_cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ return -1;
+ }
+
+ {
+ struct se_device *dev = se_lun->lun_se_dev;
+ spin_lock(&dev->stats_lock);
+ dev->num_cmds++;
+ if (se_cmd->data_direction == DMA_TO_DEVICE)
+ dev->write_bytes += se_cmd->data_length;
+ else if (se_cmd->data_direction == DMA_FROM_DEVICE)
+ dev->read_bytes += se_cmd->data_length;
+ spin_unlock(&dev->stats_lock);
+ }
+
+ /*
+ * Add the iscsi_cmd_t to the struct se_lun's cmd list. This list is used
+ * for tracking state of struct se_cmds during LUN shutdown events.
+ */
+ spin_lock_irqsave(&se_lun->lun_cmd_lock, flags);
+ list_add_tail(&se_cmd->se_lun_list, &se_lun->lun_cmd_list);
+ atomic_set(&T_TASK(se_cmd)->transport_lun_active, 1);
+#if 0
+ printk(KERN_INFO "Adding ITT: 0x%08x to LUN LIST[%d]\n",
+ CMD_TFO(se_cmd)->get_task_tag(se_cmd), se_lun->unpacked_lun);
+#endif
+ spin_unlock_irqrestore(&se_lun->lun_cmd_lock, flags);
+
+ return 0;
+}
+EXPORT_SYMBOL(transport_get_lun_for_cmd);
+
+int transport_get_lun_for_tmr(
+ struct se_cmd *se_cmd,
+ u32 unpacked_lun)
+{
+ struct se_device *dev = NULL;
+ struct se_dev_entry *deve;
+ struct se_lun *se_lun = NULL;
+ struct se_session *se_sess = SE_SESS(se_cmd);
+ struct se_tmr_req *se_tmr = se_cmd->se_tmr_req;
+
+ spin_lock_irq(&SE_NODE_ACL(se_sess)->device_list_lock);
+ deve = se_cmd->se_deve =
+ &SE_NODE_ACL(se_sess)->device_list[unpacked_lun];
+ if (deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS) {
+ se_lun = se_cmd->se_lun = se_tmr->tmr_lun = deve->se_lun;
+ dev = se_tmr->tmr_dev = se_lun->lun_se_dev;
+ se_cmd->pr_res_key = deve->pr_res_key;
+ se_cmd->orig_fe_lun = unpacked_lun;
+ se_cmd->se_orig_obj_ptr = SE_LUN(se_cmd)->lun_se_dev;
+/* se_cmd->se_cmd_flags |= SCF_SE_LUN_CMD; */
+ }
+ spin_unlock_irq(&SE_NODE_ACL(se_sess)->device_list_lock);
+
+ if (!se_lun) {
+ printk(KERN_INFO "TARGET_CORE[%s]: Detected NON_EXISTENT_LUN"
+ " Access for 0x%08x\n",
+ CMD_TFO(se_cmd)->get_fabric_name(),
+ unpacked_lun);
+ se_cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ return -1;
+ }
+ /*
+ * Determine if the struct se_lun is online.
+ */
+/* #warning FIXME: Check for LUN_RESET + UNIT Attention */
+ if (se_dev_check_online(se_lun->lun_se_dev) != 0) {
+ se_cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ return -1;
+ }
+
+ spin_lock(&dev->se_tmr_lock);
+ list_add_tail(&se_tmr->tmr_list, &dev->dev_tmr_list);
+ spin_unlock(&dev->se_tmr_lock);
+
+ return 0;
+}
+EXPORT_SYMBOL(transport_get_lun_for_tmr);
+
+/*
+ * This function is called from core_scsi3_emulate_pro_register_and_move()
+ * and core_scsi3_decode_spec_i_port(), and will increment &deve->pr_ref_count
+ * when a matching rtpi is found.
+ */
+struct se_dev_entry *core_get_se_deve_from_rtpi(
+ struct se_node_acl *nacl,
+ u16 rtpi)
+{
+ struct se_dev_entry *deve;
+ struct se_lun *lun;
+ struct se_port *port;
+ struct se_portal_group *tpg = nacl->se_tpg;
+ u32 i;
+
+ spin_lock_irq(&nacl->device_list_lock);
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ deve = &nacl->device_list[i];
+
+ if (!(deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS))
+ continue;
+
+ lun = deve->se_lun;
+ if (!(lun)) {
+ printk(KERN_ERR "%s device entries device pointer is"
+ " NULL, but Initiator has access.\n",
+ TPG_TFO(tpg)->get_fabric_name());
+ continue;
+ }
+ port = lun->lun_sep;
+ if (!(port)) {
+ printk(KERN_ERR "%s device entries device pointer is"
+ " NULL, but Initiator has access.\n",
+ TPG_TFO(tpg)->get_fabric_name());
+ continue;
+ }
+ if (port->sep_rtpi != rtpi)
+ continue;
+
+ atomic_inc(&deve->pr_ref_count);
+ smp_mb__after_atomic_inc();
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ return deve;
+ }
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ return NULL;
+}
+
+int core_free_device_list_for_node(
+ struct se_node_acl *nacl,
+ struct se_portal_group *tpg)
+{
+ struct se_dev_entry *deve;
+ struct se_lun *lun;
+ u32 i;
+
+ if (!nacl->device_list)
+ return 0;
+
+ spin_lock_irq(&nacl->device_list_lock);
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ deve = &nacl->device_list[i];
+
+ if (!(deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS))
+ continue;
+
+ if (!deve->se_lun) {
+ printk(KERN_ERR "%s device entries device pointer is"
+ " NULL, but Initiator has access.\n",
+ TPG_TFO(tpg)->get_fabric_name());
+ continue;
+ }
+ lun = deve->se_lun;
+
+ spin_unlock_irq(&nacl->device_list_lock);
+ core_update_device_list_for_node(lun, NULL, deve->mapped_lun,
+ TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg, 0);
+ spin_lock_irq(&nacl->device_list_lock);
+ }
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ kfree(nacl->device_list);
+ nacl->device_list = NULL;
+
+ return 0;
+}
+
+void core_dec_lacl_count(struct se_node_acl *se_nacl, struct se_cmd *se_cmd)
+{
+ struct se_dev_entry *deve;
+
+ spin_lock_irq(&se_nacl->device_list_lock);
+ deve = &se_nacl->device_list[se_cmd->orig_fe_lun];
+ deve->deve_cmds--;
+ spin_unlock_irq(&se_nacl->device_list_lock);
+
+ return;
+}
+
+void core_update_device_list_access(
+ u32 mapped_lun,
+ u32 lun_access,
+ struct se_node_acl *nacl)
+{
+ struct se_dev_entry *deve;
+
+ spin_lock_irq(&nacl->device_list_lock);
+ deve = &nacl->device_list[mapped_lun];
+ if (lun_access & TRANSPORT_LUNFLAGS_READ_WRITE) {
+ deve->lun_flags &= ~TRANSPORT_LUNFLAGS_READ_ONLY;
+ deve->lun_flags |= TRANSPORT_LUNFLAGS_READ_WRITE;
+ } else {
+ deve->lun_flags &= ~TRANSPORT_LUNFLAGS_READ_WRITE;
+ deve->lun_flags |= TRANSPORT_LUNFLAGS_READ_ONLY;
+ }
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ return;
+}
+
+/* core_update_device_list_for_node():
+ *
+ *
+ */
+int core_update_device_list_for_node(
+ struct se_lun *lun,
+ struct se_lun_acl *lun_acl,
+ u32 mapped_lun,
+ u32 lun_access,
+ struct se_node_acl *nacl,
+ struct se_portal_group *tpg,
+ int enable)
+{
+ struct se_port *port = lun->lun_sep;
+ struct se_dev_entry *deve = &nacl->device_list[mapped_lun];
+ int trans = 0;
+ /*
+ * If the MappedLUN entry is being disabled, the entry in
+ * port->sep_alua_list must be removed now before clearing the
+ * struct se_dev_entry pointers below as logic in
+ * core_alua_do_transition_tg_pt() depends on these being present.
+ */
+ if (!(enable)) {
+ /*
+ * deve->se_lun_acl will be NULL for demo-mode created LUNs
+ * that have not been explictly concerted to MappedLUNs ->
+ * struct se_lun_acl.
+ */
+ if (!(deve->se_lun_acl))
+ return 0;
+
+ spin_lock_bh(&port->sep_alua_lock);
+ list_del(&deve->alua_port_list);
+ spin_unlock_bh(&port->sep_alua_lock);
+ }
+
+ spin_lock_irq(&nacl->device_list_lock);
+ if (enable) {
+ /*
+ * Check if the call is handling demo mode -> explict LUN ACL
+ * transition. This transition must be for the same struct se_lun
+ * + mapped_lun that was setup in demo mode..
+ */
+ if (deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS) {
+ if (deve->se_lun_acl != NULL) {
+ printk(KERN_ERR "struct se_dev_entry->se_lun_acl"
+ " already set for demo mode -> explict"
+ " LUN ACL transition\n");
+ return -1;
+ }
+ if (deve->se_lun != lun) {
+ printk(KERN_ERR "struct se_dev_entry->se_lun does"
+ " match passed struct se_lun for demo mode"
+ " -> explict LUN ACL transition\n");
+ return -1;
+ }
+ deve->se_lun_acl = lun_acl;
+ trans = 1;
+ } else {
+ deve->se_lun = lun;
+ deve->se_lun_acl = lun_acl;
+ deve->mapped_lun = mapped_lun;
+ deve->lun_flags |= TRANSPORT_LUNFLAGS_INITIATOR_ACCESS;
+ }
+
+ if (lun_access & TRANSPORT_LUNFLAGS_READ_WRITE) {
+ deve->lun_flags &= ~TRANSPORT_LUNFLAGS_READ_ONLY;
+ deve->lun_flags |= TRANSPORT_LUNFLAGS_READ_WRITE;
+ } else {
+ deve->lun_flags &= ~TRANSPORT_LUNFLAGS_READ_WRITE;
+ deve->lun_flags |= TRANSPORT_LUNFLAGS_READ_ONLY;
+ }
+
+ if (trans) {
+ spin_unlock_irq(&nacl->device_list_lock);
+ return 0;
+ }
+ deve->creation_time = get_jiffies_64();
+ deve->attach_count++;
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ spin_lock_bh(&port->sep_alua_lock);
+ list_add_tail(&deve->alua_port_list, &port->sep_alua_list);
+ spin_unlock_bh(&port->sep_alua_lock);
+
+ return 0;
+ }
+ /*
+ * Wait for any in process SPEC_I_PT=1 or REGISTER_AND_MOVE
+ * PR operation to complete.
+ */
+ spin_unlock_irq(&nacl->device_list_lock);
+ while (atomic_read(&deve->pr_ref_count) != 0)
+ cpu_relax();
+ spin_lock_irq(&nacl->device_list_lock);
+ /*
+ * Disable struct se_dev_entry LUN ACL mapping
+ */
+ core_scsi3_ua_release_all(deve);
+ deve->se_lun = NULL;
+ deve->se_lun_acl = NULL;
+ deve->lun_flags = 0;
+ deve->creation_time = 0;
+ deve->attach_count--;
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ core_scsi3_free_pr_reg_from_nacl(lun->lun_se_dev, nacl);
+ return 0;
+}
+
+/* core_clear_lun_from_tpg():
+ *
+ *
+ */
+void core_clear_lun_from_tpg(struct se_lun *lun, struct se_portal_group *tpg)
+{
+ struct se_node_acl *nacl;
+ struct se_dev_entry *deve;
+ u32 i;
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ list_for_each_entry(nacl, &tpg->acl_node_list, acl_list) {
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+ spin_lock_irq(&nacl->device_list_lock);
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ deve = &nacl->device_list[i];
+ if (lun != deve->se_lun)
+ continue;
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ core_update_device_list_for_node(lun, NULL,
+ deve->mapped_lun, TRANSPORT_LUNFLAGS_NO_ACCESS,
+ nacl, tpg, 0);
+
+ spin_lock_irq(&nacl->device_list_lock);
+ }
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ }
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+ return;
+}
+
+static struct se_port *core_alloc_port(struct se_device *dev)
+{
+ struct se_port *port, *port_tmp;
+
+ port = kzalloc(sizeof(struct se_port), GFP_KERNEL);
+ if (!(port)) {
+ printk(KERN_ERR "Unable to allocate struct se_port\n");
+ return NULL;
+ }
+ INIT_LIST_HEAD(&port->sep_alua_list);
+ INIT_LIST_HEAD(&port->sep_list);
+ atomic_set(&port->sep_tg_pt_secondary_offline, 0);
+ spin_lock_init(&port->sep_alua_lock);
+ mutex_init(&port->sep_tg_pt_md_mutex);
+
+ spin_lock(&dev->se_port_lock);
+ if (dev->dev_port_count == 0x0000ffff) {
+ printk(KERN_WARNING "Reached dev->dev_port_count =="
+ " 0x0000ffff\n");
+ spin_unlock(&dev->se_port_lock);
+ return NULL;
+ }
+again:
+ /*
+ * Allocate the next RELATIVE TARGET PORT IDENTIFER for this struct se_device
+ * Here is the table from spc4r17 section 7.7.3.8.
+ *
+ * Table 473 -- RELATIVE TARGET PORT IDENTIFIER field
+ *
+ * Code Description
+ * 0h Reserved
+ * 1h Relative port 1, historically known as port A
+ * 2h Relative port 2, historically known as port B
+ * 3h to FFFFh Relative port 3 through 65 535
+ */
+ port->sep_rtpi = dev->dev_rpti_counter++;
+ if (!(port->sep_rtpi))
+ goto again;
+
+ list_for_each_entry(port_tmp, &dev->dev_sep_list, sep_list) {
+ /*
+ * Make sure RELATIVE TARGET PORT IDENTIFER is unique
+ * for 16-bit wrap..
+ */
+ if (port->sep_rtpi == port_tmp->sep_rtpi)
+ goto again;
+ }
+ spin_unlock(&dev->se_port_lock);
+
+ return port;
+}
+
+static void core_export_port(
+ struct se_device *dev,
+ struct se_portal_group *tpg,
+ struct se_port *port,
+ struct se_lun *lun)
+{
+ struct se_subsystem_dev *su_dev = SU_DEV(dev);
+ struct t10_alua_tg_pt_gp_member *tg_pt_gp_mem = NULL;
+
+ spin_lock(&dev->se_port_lock);
+ spin_lock(&lun->lun_sep_lock);
+ port->sep_tpg = tpg;
+ port->sep_lun = lun;
+ lun->lun_sep = port;
+ spin_unlock(&lun->lun_sep_lock);
+
+ list_add_tail(&port->sep_list, &dev->dev_sep_list);
+ spin_unlock(&dev->se_port_lock);
+
+ if (T10_ALUA(su_dev)->alua_type == SPC3_ALUA_EMULATED) {
+ tg_pt_gp_mem = core_alua_allocate_tg_pt_gp_mem(port);
+ if (IS_ERR(tg_pt_gp_mem) || !tg_pt_gp_mem) {
+ printk(KERN_ERR "Unable to allocate t10_alua_tg_pt"
+ "_gp_member_t\n");
+ return;
+ }
+ spin_lock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ __core_alua_attach_tg_pt_gp_mem(tg_pt_gp_mem,
+ T10_ALUA(su_dev)->default_tg_pt_gp);
+ spin_unlock(&tg_pt_gp_mem->tg_pt_gp_mem_lock);
+ printk(KERN_INFO "%s/%s: Adding to default ALUA Target Port"
+ " Group: alua/default_tg_pt_gp\n",
+ TRANSPORT(dev)->name, TPG_TFO(tpg)->get_fabric_name());
+ }
+
+ dev->dev_port_count++;
+ port->sep_index = port->sep_rtpi; /* RELATIVE TARGET PORT IDENTIFER */
+}
+
+/*
+ * Called with struct se_device->se_port_lock spinlock held.
+ */
+static void core_release_port(struct se_device *dev, struct se_port *port)
+{
+ /*
+ * Wait for any port reference for PR ALL_TG_PT=1 operation
+ * to complete in __core_scsi3_alloc_registration()
+ */
+ spin_unlock(&dev->se_port_lock);
+ if (atomic_read(&port->sep_tg_pt_ref_cnt))
+ cpu_relax();
+ spin_lock(&dev->se_port_lock);
+
+ core_alua_free_tg_pt_gp_mem(port);
+
+ list_del(&port->sep_list);
+ dev->dev_port_count--;
+ kfree(port);
+
+ return;
+}
+
+int core_dev_export(
+ struct se_device *dev,
+ struct se_portal_group *tpg,
+ struct se_lun *lun)
+{
+ struct se_port *port;
+
+ port = core_alloc_port(dev);
+ if (!(port))
+ return -1;
+
+ lun->lun_se_dev = dev;
+ se_dev_start(dev);
+
+ atomic_inc(&dev->dev_export_obj.obj_access_count);
+ core_export_port(dev, tpg, port, lun);
+ return 0;
+}
+
+void core_dev_unexport(
+ struct se_device *dev,
+ struct se_portal_group *tpg,
+ struct se_lun *lun)
+{
+ struct se_port *port = lun->lun_sep;
+
+ spin_lock(&lun->lun_sep_lock);
+ if (lun->lun_se_dev == NULL) {
+ spin_unlock(&lun->lun_sep_lock);
+ return;
+ }
+ spin_unlock(&lun->lun_sep_lock);
+
+ spin_lock(&dev->se_port_lock);
+ atomic_dec(&dev->dev_export_obj.obj_access_count);
+ core_release_port(dev, port);
+ spin_unlock(&dev->se_port_lock);
+
+ se_dev_stop(dev);
+ lun->lun_se_dev = NULL;
+}
+
+int transport_core_report_lun_response(struct se_cmd *se_cmd)
+{
+ struct se_dev_entry *deve;
+ struct se_lun *se_lun;
+ struct se_session *se_sess = SE_SESS(se_cmd);
+ struct se_task *se_task;
+ unsigned char *buf = (unsigned char *)T_TASK(se_cmd)->t_task_buf;
+ u32 cdb_offset = 0, lun_count = 0, offset = 8;
+ u64 i, lun;
+
+ list_for_each_entry(se_task, &T_TASK(se_cmd)->t_task_list, t_list)
+ break;
+
+ if (!(se_task)) {
+ printk(KERN_ERR "Unable to locate struct se_task for struct se_cmd\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+
+ /*
+ * If no struct se_session pointer is present, this struct se_cmd is
+ * coming via a target_core_mod PASSTHROUGH op, and not through
+ * a $FABRIC_MOD. In that case, report LUN=0 only.
+ */
+ if (!(se_sess)) {
+ lun = 0;
+ buf[offset++] = ((lun >> 56) & 0xff);
+ buf[offset++] = ((lun >> 48) & 0xff);
+ buf[offset++] = ((lun >> 40) & 0xff);
+ buf[offset++] = ((lun >> 32) & 0xff);
+ buf[offset++] = ((lun >> 24) & 0xff);
+ buf[offset++] = ((lun >> 16) & 0xff);
+ buf[offset++] = ((lun >> 8) & 0xff);
+ buf[offset++] = (lun & 0xff);
+ lun_count = 1;
+ goto done;
+ }
+
+ spin_lock_irq(&SE_NODE_ACL(se_sess)->device_list_lock);
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ deve = &SE_NODE_ACL(se_sess)->device_list[i];
+ if (!(deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS))
+ continue;
+ se_lun = deve->se_lun;
+ /*
+ * We determine the correct LUN LIST LENGTH even once we
+ * have reached the initial allocation length.
+ * See SPC2-R20 7.19.
+ */
+ lun_count++;
+ if ((cdb_offset + 8) >= se_cmd->data_length)
+ continue;
+
+ lun = cpu_to_be64(CMD_TFO(se_cmd)->pack_lun(deve->mapped_lun));
+ buf[offset++] = ((lun >> 56) & 0xff);
+ buf[offset++] = ((lun >> 48) & 0xff);
+ buf[offset++] = ((lun >> 40) & 0xff);
+ buf[offset++] = ((lun >> 32) & 0xff);
+ buf[offset++] = ((lun >> 24) & 0xff);
+ buf[offset++] = ((lun >> 16) & 0xff);
+ buf[offset++] = ((lun >> 8) & 0xff);
+ buf[offset++] = (lun & 0xff);
+ cdb_offset += 8;
+ }
+ spin_unlock_irq(&SE_NODE_ACL(se_sess)->device_list_lock);
+
+ /*
+ * See SPC3 r07, page 159.
+ */
+done:
+ lun_count *= 8;
+ buf[0] = ((lun_count >> 24) & 0xff);
+ buf[1] = ((lun_count >> 16) & 0xff);
+ buf[2] = ((lun_count >> 8) & 0xff);
+ buf[3] = (lun_count & 0xff);
+
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+}
+
+/* se_release_device_for_hba():
+ *
+ *
+ */
+void se_release_device_for_hba(struct se_device *dev)
+{
+ struct se_hba *hba = dev->se_hba;
+
+ if ((dev->dev_status & TRANSPORT_DEVICE_ACTIVATED) ||
+ (dev->dev_status & TRANSPORT_DEVICE_DEACTIVATED) ||
+ (dev->dev_status & TRANSPORT_DEVICE_SHUTDOWN) ||
+ (dev->dev_status & TRANSPORT_DEVICE_OFFLINE_ACTIVATED) ||
+ (dev->dev_status & TRANSPORT_DEVICE_OFFLINE_DEACTIVATED))
+ se_dev_stop(dev);
+
+ if (dev->dev_ptr) {
+ kthread_stop(dev->process_thread);
+ if (dev->transport->free_device)
+ dev->transport->free_device(dev->dev_ptr);
+ }
+
+ spin_lock(&hba->device_lock);
+ list_del(&dev->dev_list);
+ hba->dev_count--;
+ spin_unlock(&hba->device_lock);
+
+ core_scsi3_free_all_registrations(dev);
+ se_release_vpd_for_dev(dev);
+
+ kfree(dev->dev_status_queue_obj);
+ kfree(dev->dev_queue_obj);
+ kfree(dev);
+
+ return;
+}
+
+void se_release_vpd_for_dev(struct se_device *dev)
+{
+ struct t10_vpd *vpd, *vpd_tmp;
+
+ spin_lock(&DEV_T10_WWN(dev)->t10_vpd_lock);
+ list_for_each_entry_safe(vpd, vpd_tmp,
+ &DEV_T10_WWN(dev)->t10_vpd_list, vpd_list) {
+ list_del(&vpd->vpd_list);
+ kfree(vpd);
+ }
+ spin_unlock(&DEV_T10_WWN(dev)->t10_vpd_lock);
+
+ return;
+}
+
+/*
+ * Called with struct se_hba->device_lock held.
+ */
+void se_clear_dev_ports(struct se_device *dev)
+{
+ struct se_hba *hba = dev->se_hba;
+ struct se_lun *lun;
+ struct se_portal_group *tpg;
+ struct se_port *sep, *sep_tmp;
+
+ spin_lock(&dev->se_port_lock);
+ list_for_each_entry_safe(sep, sep_tmp, &dev->dev_sep_list, sep_list) {
+ spin_unlock(&dev->se_port_lock);
+ spin_unlock(&hba->device_lock);
+
+ lun = sep->sep_lun;
+ tpg = sep->sep_tpg;
+ spin_lock(&lun->lun_sep_lock);
+ if (lun->lun_se_dev == NULL) {
+ spin_unlock(&lun->lun_sep_lock);
+ continue;
+ }
+ spin_unlock(&lun->lun_sep_lock);
+
+ core_dev_del_lun(tpg, lun->unpacked_lun);
+
+ spin_lock(&hba->device_lock);
+ spin_lock(&dev->se_port_lock);
+ }
+ spin_unlock(&dev->se_port_lock);
+
+ return;
+}
+
+/* se_free_virtual_device():
+ *
+ * Used for IBLOCK, RAMDISK, and FILEIO Transport Drivers.
+ */
+int se_free_virtual_device(struct se_device *dev, struct se_hba *hba)
+{
+ spin_lock(&hba->device_lock);
+ se_clear_dev_ports(dev);
+ spin_unlock(&hba->device_lock);
+
+ core_alua_free_lu_gp_mem(dev);
+ se_release_device_for_hba(dev);
+
+ return 0;
+}
+
+static void se_dev_start(struct se_device *dev)
+{
+ struct se_hba *hba = dev->se_hba;
+
+ spin_lock(&hba->device_lock);
+ atomic_inc(&dev->dev_obj.obj_access_count);
+ if (atomic_read(&dev->dev_obj.obj_access_count) == 1) {
+ if (dev->dev_status & TRANSPORT_DEVICE_DEACTIVATED) {
+ dev->dev_status &= ~TRANSPORT_DEVICE_DEACTIVATED;
+ dev->dev_status |= TRANSPORT_DEVICE_ACTIVATED;
+ } else if (dev->dev_status &
+ TRANSPORT_DEVICE_OFFLINE_DEACTIVATED) {
+ dev->dev_status &=
+ ~TRANSPORT_DEVICE_OFFLINE_DEACTIVATED;
+ dev->dev_status |= TRANSPORT_DEVICE_OFFLINE_ACTIVATED;
+ }
+ }
+ spin_unlock(&hba->device_lock);
+}
+
+static void se_dev_stop(struct se_device *dev)
+{
+ struct se_hba *hba = dev->se_hba;
+
+ spin_lock(&hba->device_lock);
+ atomic_dec(&dev->dev_obj.obj_access_count);
+ if (atomic_read(&dev->dev_obj.obj_access_count) == 0) {
+ if (dev->dev_status & TRANSPORT_DEVICE_ACTIVATED) {
+ dev->dev_status &= ~TRANSPORT_DEVICE_ACTIVATED;
+ dev->dev_status |= TRANSPORT_DEVICE_DEACTIVATED;
+ } else if (dev->dev_status &
+ TRANSPORT_DEVICE_OFFLINE_ACTIVATED) {
+ dev->dev_status &= ~TRANSPORT_DEVICE_OFFLINE_ACTIVATED;
+ dev->dev_status |= TRANSPORT_DEVICE_OFFLINE_DEACTIVATED;
+ }
+ }
+ spin_unlock(&hba->device_lock);
+
+ while (atomic_read(&hba->dev_mib_access_count))
+ cpu_relax();
+}
+
+int se_dev_check_online(struct se_device *dev)
+{
+ int ret;
+
+ spin_lock_irq(&dev->dev_status_lock);
+ ret = ((dev->dev_status & TRANSPORT_DEVICE_ACTIVATED) ||
+ (dev->dev_status & TRANSPORT_DEVICE_DEACTIVATED)) ? 0 : 1;
+ spin_unlock_irq(&dev->dev_status_lock);
+
+ return ret;
+}
+
+int se_dev_check_shutdown(struct se_device *dev)
+{
+ int ret;
+
+ spin_lock_irq(&dev->dev_status_lock);
+ ret = (dev->dev_status & TRANSPORT_DEVICE_SHUTDOWN);
+ spin_unlock_irq(&dev->dev_status_lock);
+
+ return ret;
+}
+
+void se_dev_set_default_attribs(
+ struct se_device *dev,
+ struct se_dev_limits *dev_limits)
+{
+ struct queue_limits *limits = &dev_limits->limits;
+
+ DEV_ATTRIB(dev)->emulate_dpo = DA_EMULATE_DPO;
+ DEV_ATTRIB(dev)->emulate_fua_write = DA_EMULATE_FUA_WRITE;
+ DEV_ATTRIB(dev)->emulate_fua_read = DA_EMULATE_FUA_READ;
+ DEV_ATTRIB(dev)->emulate_write_cache = DA_EMULATE_WRITE_CACHE;
+ DEV_ATTRIB(dev)->emulate_ua_intlck_ctrl = DA_EMULATE_UA_INTLLCK_CTRL;
+ DEV_ATTRIB(dev)->emulate_tas = DA_EMULATE_TAS;
+ DEV_ATTRIB(dev)->emulate_tpu = DA_EMULATE_TPU;
+ DEV_ATTRIB(dev)->emulate_tpws = DA_EMULATE_TPWS;
+ DEV_ATTRIB(dev)->emulate_reservations = DA_EMULATE_RESERVATIONS;
+ DEV_ATTRIB(dev)->emulate_alua = DA_EMULATE_ALUA;
+ DEV_ATTRIB(dev)->enforce_pr_isids = DA_ENFORCE_PR_ISIDS;
+ /*
+ * The TPU=1 and TPWS=1 settings will be set in TCM/IBLOCK
+ * iblock_create_virtdevice() from struct queue_limits values
+ * if blk_queue_discard()==1
+ */
+ DEV_ATTRIB(dev)->max_unmap_lba_count = DA_MAX_UNMAP_LBA_COUNT;
+ DEV_ATTRIB(dev)->max_unmap_block_desc_count =
+ DA_MAX_UNMAP_BLOCK_DESC_COUNT;
+ DEV_ATTRIB(dev)->unmap_granularity = DA_UNMAP_GRANULARITY_DEFAULT;
+ DEV_ATTRIB(dev)->unmap_granularity_alignment =
+ DA_UNMAP_GRANULARITY_ALIGNMENT_DEFAULT;
+ /*
+ * block_size is based on subsystem plugin dependent requirements.
+ */
+ DEV_ATTRIB(dev)->hw_block_size = limits->logical_block_size;
+ DEV_ATTRIB(dev)->block_size = limits->logical_block_size;
+ /*
+ * max_sectors is based on subsystem plugin dependent requirements.
+ */
+ DEV_ATTRIB(dev)->hw_max_sectors = limits->max_hw_sectors;
+ DEV_ATTRIB(dev)->max_sectors = limits->max_sectors;
+ /*
+ * Set optimal_sectors from max_sectors, which can be lowered via
+ * configfs.
+ */
+ DEV_ATTRIB(dev)->optimal_sectors = limits->max_sectors;
+ /*
+ * queue_depth is based on subsystem plugin dependent requirements.
+ */
+ DEV_ATTRIB(dev)->hw_queue_depth = dev_limits->hw_queue_depth;
+ DEV_ATTRIB(dev)->queue_depth = dev_limits->queue_depth;
+}
+
+int se_dev_set_task_timeout(struct se_device *dev, u32 task_timeout)
+{
+ if (task_timeout > DA_TASK_TIMEOUT_MAX) {
+ printk(KERN_ERR "dev[%p]: Passed task_timeout: %u larger then"
+ " DA_TASK_TIMEOUT_MAX\n", dev, task_timeout);
+ return -1;
+ } else {
+ DEV_ATTRIB(dev)->task_timeout = task_timeout;
+ printk(KERN_INFO "dev[%p]: Set SE Device task_timeout: %u\n",
+ dev, task_timeout);
+ }
+
+ return 0;
+}
+
+int se_dev_set_max_unmap_lba_count(
+ struct se_device *dev,
+ u32 max_unmap_lba_count)
+{
+ DEV_ATTRIB(dev)->max_unmap_lba_count = max_unmap_lba_count;
+ printk(KERN_INFO "dev[%p]: Set max_unmap_lba_count: %u\n",
+ dev, DEV_ATTRIB(dev)->max_unmap_lba_count);
+ return 0;
+}
+
+int se_dev_set_max_unmap_block_desc_count(
+ struct se_device *dev,
+ u32 max_unmap_block_desc_count)
+{
+ DEV_ATTRIB(dev)->max_unmap_block_desc_count = max_unmap_block_desc_count;
+ printk(KERN_INFO "dev[%p]: Set max_unmap_block_desc_count: %u\n",
+ dev, DEV_ATTRIB(dev)->max_unmap_block_desc_count);
+ return 0;
+}
+
+int se_dev_set_unmap_granularity(
+ struct se_device *dev,
+ u32 unmap_granularity)
+{
+ DEV_ATTRIB(dev)->unmap_granularity = unmap_granularity;
+ printk(KERN_INFO "dev[%p]: Set unmap_granularity: %u\n",
+ dev, DEV_ATTRIB(dev)->unmap_granularity);
+ return 0;
+}
+
+int se_dev_set_unmap_granularity_alignment(
+ struct se_device *dev,
+ u32 unmap_granularity_alignment)
+{
+ DEV_ATTRIB(dev)->unmap_granularity_alignment = unmap_granularity_alignment;
+ printk(KERN_INFO "dev[%p]: Set unmap_granularity_alignment: %u\n",
+ dev, DEV_ATTRIB(dev)->unmap_granularity_alignment);
+ return 0;
+}
+
+int se_dev_set_emulate_dpo(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+ if (TRANSPORT(dev)->dpo_emulated == NULL) {
+ printk(KERN_ERR "TRANSPORT(dev)->dpo_emulated is NULL\n");
+ return -1;
+ }
+ if (TRANSPORT(dev)->dpo_emulated(dev) == 0) {
+ printk(KERN_ERR "TRANSPORT(dev)->dpo_emulated not supported\n");
+ return -1;
+ }
+ DEV_ATTRIB(dev)->emulate_dpo = flag;
+ printk(KERN_INFO "dev[%p]: SE Device Page Out (DPO) Emulation"
+ " bit: %d\n", dev, DEV_ATTRIB(dev)->emulate_dpo);
+ return 0;
+}
+
+int se_dev_set_emulate_fua_write(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+ if (TRANSPORT(dev)->fua_write_emulated == NULL) {
+ printk(KERN_ERR "TRANSPORT(dev)->fua_write_emulated is NULL\n");
+ return -1;
+ }
+ if (TRANSPORT(dev)->fua_write_emulated(dev) == 0) {
+ printk(KERN_ERR "TRANSPORT(dev)->fua_write_emulated not supported\n");
+ return -1;
+ }
+ DEV_ATTRIB(dev)->emulate_fua_write = flag;
+ printk(KERN_INFO "dev[%p]: SE Device Forced Unit Access WRITEs: %d\n",
+ dev, DEV_ATTRIB(dev)->emulate_fua_write);
+ return 0;
+}
+
+int se_dev_set_emulate_fua_read(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+ if (TRANSPORT(dev)->fua_read_emulated == NULL) {
+ printk(KERN_ERR "TRANSPORT(dev)->fua_read_emulated is NULL\n");
+ return -1;
+ }
+ if (TRANSPORT(dev)->fua_read_emulated(dev) == 0) {
+ printk(KERN_ERR "TRANSPORT(dev)->fua_read_emulated not supported\n");
+ return -1;
+ }
+ DEV_ATTRIB(dev)->emulate_fua_read = flag;
+ printk(KERN_INFO "dev[%p]: SE Device Forced Unit Access READs: %d\n",
+ dev, DEV_ATTRIB(dev)->emulate_fua_read);
+ return 0;
+}
+
+int se_dev_set_emulate_write_cache(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+ if (TRANSPORT(dev)->write_cache_emulated == NULL) {
+ printk(KERN_ERR "TRANSPORT(dev)->write_cache_emulated is NULL\n");
+ return -1;
+ }
+ if (TRANSPORT(dev)->write_cache_emulated(dev) == 0) {
+ printk(KERN_ERR "TRANSPORT(dev)->write_cache_emulated not supported\n");
+ return -1;
+ }
+ DEV_ATTRIB(dev)->emulate_write_cache = flag;
+ printk(KERN_INFO "dev[%p]: SE Device WRITE_CACHE_EMULATION flag: %d\n",
+ dev, DEV_ATTRIB(dev)->emulate_write_cache);
+ return 0;
+}
+
+int se_dev_set_emulate_ua_intlck_ctrl(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1) && (flag != 2)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+
+ if (atomic_read(&dev->dev_export_obj.obj_access_count)) {
+ printk(KERN_ERR "dev[%p]: Unable to change SE Device"
+ " UA_INTRLCK_CTRL while dev_export_obj: %d count"
+ " exists\n", dev,
+ atomic_read(&dev->dev_export_obj.obj_access_count));
+ return -1;
+ }
+ DEV_ATTRIB(dev)->emulate_ua_intlck_ctrl = flag;
+ printk(KERN_INFO "dev[%p]: SE Device UA_INTRLCK_CTRL flag: %d\n",
+ dev, DEV_ATTRIB(dev)->emulate_ua_intlck_ctrl);
+
+ return 0;
+}
+
+int se_dev_set_emulate_tas(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+
+ if (atomic_read(&dev->dev_export_obj.obj_access_count)) {
+ printk(KERN_ERR "dev[%p]: Unable to change SE Device TAS while"
+ " dev_export_obj: %d count exists\n", dev,
+ atomic_read(&dev->dev_export_obj.obj_access_count));
+ return -1;
+ }
+ DEV_ATTRIB(dev)->emulate_tas = flag;
+ printk(KERN_INFO "dev[%p]: SE Device TASK_ABORTED status bit: %s\n",
+ dev, (DEV_ATTRIB(dev)->emulate_tas) ? "Enabled" : "Disabled");
+
+ return 0;
+}
+
+int se_dev_set_emulate_tpu(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+ /*
+ * We expect this value to be non-zero when generic Block Layer
+ * Discard supported is detected iblock_create_virtdevice().
+ */
+ if (!(DEV_ATTRIB(dev)->max_unmap_block_desc_count)) {
+ printk(KERN_ERR "Generic Block Discard not supported\n");
+ return -ENOSYS;
+ }
+
+ DEV_ATTRIB(dev)->emulate_tpu = flag;
+ printk(KERN_INFO "dev[%p]: SE Device Thin Provisioning UNMAP bit: %d\n",
+ dev, flag);
+ return 0;
+}
+
+int se_dev_set_emulate_tpws(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+ /*
+ * We expect this value to be non-zero when generic Block Layer
+ * Discard supported is detected iblock_create_virtdevice().
+ */
+ if (!(DEV_ATTRIB(dev)->max_unmap_block_desc_count)) {
+ printk(KERN_ERR "Generic Block Discard not supported\n");
+ return -ENOSYS;
+ }
+
+ DEV_ATTRIB(dev)->emulate_tpws = flag;
+ printk(KERN_INFO "dev[%p]: SE Device Thin Provisioning WRITE_SAME: %d\n",
+ dev, flag);
+ return 0;
+}
+
+int se_dev_set_enforce_pr_isids(struct se_device *dev, int flag)
+{
+ if ((flag != 0) && (flag != 1)) {
+ printk(KERN_ERR "Illegal value %d\n", flag);
+ return -1;
+ }
+ DEV_ATTRIB(dev)->enforce_pr_isids = flag;
+ printk(KERN_INFO "dev[%p]: SE Device enforce_pr_isids bit: %s\n", dev,
+ (DEV_ATTRIB(dev)->enforce_pr_isids) ? "Enabled" : "Disabled");
+ return 0;
+}
+
+/*
+ * Note, this can only be called on unexported SE Device Object.
+ */
+int se_dev_set_queue_depth(struct se_device *dev, u32 queue_depth)
+{
+ u32 orig_queue_depth = dev->queue_depth;
+
+ if (atomic_read(&dev->dev_export_obj.obj_access_count)) {
+ printk(KERN_ERR "dev[%p]: Unable to change SE Device TCQ while"
+ " dev_export_obj: %d count exists\n", dev,
+ atomic_read(&dev->dev_export_obj.obj_access_count));
+ return -1;
+ }
+ if (!(queue_depth)) {
+ printk(KERN_ERR "dev[%p]: Illegal ZERO value for queue"
+ "_depth\n", dev);
+ return -1;
+ }
+
+ if (TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV) {
+ if (queue_depth > DEV_ATTRIB(dev)->hw_queue_depth) {
+ printk(KERN_ERR "dev[%p]: Passed queue_depth: %u"
+ " exceeds TCM/SE_Device TCQ: %u\n",
+ dev, queue_depth,
+ DEV_ATTRIB(dev)->hw_queue_depth);
+ return -1;
+ }
+ } else {
+ if (queue_depth > DEV_ATTRIB(dev)->queue_depth) {
+ if (queue_depth > DEV_ATTRIB(dev)->hw_queue_depth) {
+ printk(KERN_ERR "dev[%p]: Passed queue_depth:"
+ " %u exceeds TCM/SE_Device MAX"
+ " TCQ: %u\n", dev, queue_depth,
+ DEV_ATTRIB(dev)->hw_queue_depth);
+ return -1;
+ }
+ }
+ }
+
+ DEV_ATTRIB(dev)->queue_depth = dev->queue_depth = queue_depth;
+ if (queue_depth > orig_queue_depth)
+ atomic_add(queue_depth - orig_queue_depth, &dev->depth_left);
+ else if (queue_depth < orig_queue_depth)
+ atomic_sub(orig_queue_depth - queue_depth, &dev->depth_left);
+
+ printk(KERN_INFO "dev[%p]: SE Device TCQ Depth changed to: %u\n",
+ dev, queue_depth);
+ return 0;
+}
+
+int se_dev_set_max_sectors(struct se_device *dev, u32 max_sectors)
+{
+ int force = 0; /* Force setting for VDEVS */
+
+ if (atomic_read(&dev->dev_export_obj.obj_access_count)) {
+ printk(KERN_ERR "dev[%p]: Unable to change SE Device"
+ " max_sectors while dev_export_obj: %d count exists\n",
+ dev, atomic_read(&dev->dev_export_obj.obj_access_count));
+ return -1;
+ }
+ if (!(max_sectors)) {
+ printk(KERN_ERR "dev[%p]: Illegal ZERO value for"
+ " max_sectors\n", dev);
+ return -1;
+ }
+ if (max_sectors < DA_STATUS_MAX_SECTORS_MIN) {
+ printk(KERN_ERR "dev[%p]: Passed max_sectors: %u less than"
+ " DA_STATUS_MAX_SECTORS_MIN: %u\n", dev, max_sectors,
+ DA_STATUS_MAX_SECTORS_MIN);
+ return -1;
+ }
+ if (TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV) {
+ if (max_sectors > DEV_ATTRIB(dev)->hw_max_sectors) {
+ printk(KERN_ERR "dev[%p]: Passed max_sectors: %u"
+ " greater than TCM/SE_Device max_sectors:"
+ " %u\n", dev, max_sectors,
+ DEV_ATTRIB(dev)->hw_max_sectors);
+ return -1;
+ }
+ } else {
+ if (!(force) && (max_sectors >
+ DEV_ATTRIB(dev)->hw_max_sectors)) {
+ printk(KERN_ERR "dev[%p]: Passed max_sectors: %u"
+ " greater than TCM/SE_Device max_sectors"
+ ": %u, use force=1 to override.\n", dev,
+ max_sectors, DEV_ATTRIB(dev)->hw_max_sectors);
+ return -1;
+ }
+ if (max_sectors > DA_STATUS_MAX_SECTORS_MAX) {
+ printk(KERN_ERR "dev[%p]: Passed max_sectors: %u"
+ " greater than DA_STATUS_MAX_SECTORS_MAX:"
+ " %u\n", dev, max_sectors,
+ DA_STATUS_MAX_SECTORS_MAX);
+ return -1;
+ }
+ }
+
+ DEV_ATTRIB(dev)->max_sectors = max_sectors;
+ printk("dev[%p]: SE Device max_sectors changed to %u\n",
+ dev, max_sectors);
+ return 0;
+}
+
+int se_dev_set_optimal_sectors(struct se_device *dev, u32 optimal_sectors)
+{
+ if (atomic_read(&dev->dev_export_obj.obj_access_count)) {
+ printk(KERN_ERR "dev[%p]: Unable to change SE Device"
+ " optimal_sectors while dev_export_obj: %d count exists\n",
+ dev, atomic_read(&dev->dev_export_obj.obj_access_count));
+ return -EINVAL;
+ }
+ if (TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV) {
+ printk(KERN_ERR "dev[%p]: Passed optimal_sectors cannot be"
+ " changed for TCM/pSCSI\n", dev);
+ return -EINVAL;
+ }
+ if (optimal_sectors > DEV_ATTRIB(dev)->max_sectors) {
+ printk(KERN_ERR "dev[%p]: Passed optimal_sectors %u cannot be"
+ " greater than max_sectors: %u\n", dev,
+ optimal_sectors, DEV_ATTRIB(dev)->max_sectors);
+ return -EINVAL;
+ }
+
+ DEV_ATTRIB(dev)->optimal_sectors = optimal_sectors;
+ printk(KERN_INFO "dev[%p]: SE Device optimal_sectors changed to %u\n",
+ dev, optimal_sectors);
+ return 0;
+}
+
+int se_dev_set_block_size(struct se_device *dev, u32 block_size)
+{
+ if (atomic_read(&dev->dev_export_obj.obj_access_count)) {
+ printk(KERN_ERR "dev[%p]: Unable to change SE Device block_size"
+ " while dev_export_obj: %d count exists\n", dev,
+ atomic_read(&dev->dev_export_obj.obj_access_count));
+ return -1;
+ }
+
+ if ((block_size != 512) &&
+ (block_size != 1024) &&
+ (block_size != 2048) &&
+ (block_size != 4096)) {
+ printk(KERN_ERR "dev[%p]: Illegal value for block_device: %u"
+ " for SE device, must be 512, 1024, 2048 or 4096\n",
+ dev, block_size);
+ return -1;
+ }
+
+ if (TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV) {
+ printk(KERN_ERR "dev[%p]: Not allowed to change block_size for"
+ " Physical Device, use for Linux/SCSI to change"
+ " block_size for underlying hardware\n", dev);
+ return -1;
+ }
+
+ DEV_ATTRIB(dev)->block_size = block_size;
+ printk(KERN_INFO "dev[%p]: SE Device block_size changed to %u\n",
+ dev, block_size);
+ return 0;
+}
+
+struct se_lun *core_dev_add_lun(
+ struct se_portal_group *tpg,
+ struct se_hba *hba,
+ struct se_device *dev,
+ u32 lun)
+{
+ struct se_lun *lun_p;
+ u32 lun_access = 0;
+
+ if (atomic_read(&dev->dev_access_obj.obj_access_count) != 0) {
+ printk(KERN_ERR "Unable to export struct se_device while dev_access_obj: %d\n",
+ atomic_read(&dev->dev_access_obj.obj_access_count));
+ return NULL;
+ }
+
+ lun_p = core_tpg_pre_addlun(tpg, lun);
+ if ((IS_ERR(lun_p)) || !(lun_p))
+ return NULL;
+
+ if (dev->dev_flags & DF_READ_ONLY)
+ lun_access = TRANSPORT_LUNFLAGS_READ_ONLY;
+ else
+ lun_access = TRANSPORT_LUNFLAGS_READ_WRITE;
+
+ if (core_tpg_post_addlun(tpg, lun_p, lun_access, dev) < 0)
+ return NULL;
+
+ printk(KERN_INFO "%s_TPG[%u]_LUN[%u] - Activated %s Logical Unit from"
+ " CORE HBA: %u\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), lun_p->unpacked_lun,
+ TPG_TFO(tpg)->get_fabric_name(), hba->hba_id);
+ /*
+ * Update LUN maps for dynamically added initiators when
+ * generate_node_acl is enabled.
+ */
+ if (TPG_TFO(tpg)->tpg_check_demo_mode(tpg)) {
+ struct se_node_acl *acl;
+ spin_lock_bh(&tpg->acl_node_lock);
+ list_for_each_entry(acl, &tpg->acl_node_list, acl_list) {
+ if (acl->dynamic_node_acl) {
+ spin_unlock_bh(&tpg->acl_node_lock);
+ core_tpg_add_node_to_devs(acl, tpg);
+ spin_lock_bh(&tpg->acl_node_lock);
+ }
+ }
+ spin_unlock_bh(&tpg->acl_node_lock);
+ }
+
+ return lun_p;
+}
+
+/* core_dev_del_lun():
+ *
+ *
+ */
+int core_dev_del_lun(
+ struct se_portal_group *tpg,
+ u32 unpacked_lun)
+{
+ struct se_lun *lun;
+ int ret = 0;
+
+ lun = core_tpg_pre_dellun(tpg, unpacked_lun, &ret);
+ if (!(lun))
+ return ret;
+
+ core_tpg_post_dellun(tpg, lun);
+
+ printk(KERN_INFO "%s_TPG[%u]_LUN[%u] - Deactivated %s Logical Unit from"
+ " device object\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), unpacked_lun,
+ TPG_TFO(tpg)->get_fabric_name());
+
+ return 0;
+}
+
+struct se_lun *core_get_lun_from_tpg(struct se_portal_group *tpg, u32 unpacked_lun)
+{
+ struct se_lun *lun;
+
+ spin_lock(&tpg->tpg_lun_lock);
+ if (unpacked_lun > (TRANSPORT_MAX_LUNS_PER_TPG-1)) {
+ printk(KERN_ERR "%s LUN: %u exceeds TRANSPORT_MAX_LUNS"
+ "_PER_TPG-1: %u for Target Portal Group: %hu\n",
+ TPG_TFO(tpg)->get_fabric_name(), unpacked_lun,
+ TRANSPORT_MAX_LUNS_PER_TPG-1,
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ spin_unlock(&tpg->tpg_lun_lock);
+ return NULL;
+ }
+ lun = &tpg->tpg_lun_list[unpacked_lun];
+
+ if (lun->lun_status != TRANSPORT_LUN_STATUS_FREE) {
+ printk(KERN_ERR "%s Logical Unit Number: %u is not free on"
+ " Target Portal Group: %hu, ignoring request.\n",
+ TPG_TFO(tpg)->get_fabric_name(), unpacked_lun,
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ spin_unlock(&tpg->tpg_lun_lock);
+ return NULL;
+ }
+ spin_unlock(&tpg->tpg_lun_lock);
+
+ return lun;
+}
+
+/* core_dev_get_lun():
+ *
+ *
+ */
+static struct se_lun *core_dev_get_lun(struct se_portal_group *tpg, u32 unpacked_lun)
+{
+ struct se_lun *lun;
+
+ spin_lock(&tpg->tpg_lun_lock);
+ if (unpacked_lun > (TRANSPORT_MAX_LUNS_PER_TPG-1)) {
+ printk(KERN_ERR "%s LUN: %u exceeds TRANSPORT_MAX_LUNS_PER"
+ "_TPG-1: %u for Target Portal Group: %hu\n",
+ TPG_TFO(tpg)->get_fabric_name(), unpacked_lun,
+ TRANSPORT_MAX_LUNS_PER_TPG-1,
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ spin_unlock(&tpg->tpg_lun_lock);
+ return NULL;
+ }
+ lun = &tpg->tpg_lun_list[unpacked_lun];
+
+ if (lun->lun_status != TRANSPORT_LUN_STATUS_ACTIVE) {
+ printk(KERN_ERR "%s Logical Unit Number: %u is not active on"
+ " Target Portal Group: %hu, ignoring request.\n",
+ TPG_TFO(tpg)->get_fabric_name(), unpacked_lun,
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ spin_unlock(&tpg->tpg_lun_lock);
+ return NULL;
+ }
+ spin_unlock(&tpg->tpg_lun_lock);
+
+ return lun;
+}
+
+struct se_lun_acl *core_dev_init_initiator_node_lun_acl(
+ struct se_portal_group *tpg,
+ u32 mapped_lun,
+ char *initiatorname,
+ int *ret)
+{
+ struct se_lun_acl *lacl;
+ struct se_node_acl *nacl;
+
+ if (strlen(initiatorname) > TRANSPORT_IQN_LEN) {
+ printk(KERN_ERR "%s InitiatorName exceeds maximum size.\n",
+ TPG_TFO(tpg)->get_fabric_name());
+ *ret = -EOVERFLOW;
+ return NULL;
+ }
+ nacl = core_tpg_get_initiator_node_acl(tpg, initiatorname);
+ if (!(nacl)) {
+ *ret = -EINVAL;
+ return NULL;
+ }
+ lacl = kzalloc(sizeof(struct se_lun_acl), GFP_KERNEL);
+ if (!(lacl)) {
+ printk(KERN_ERR "Unable to allocate memory for struct se_lun_acl.\n");
+ *ret = -ENOMEM;
+ return NULL;
+ }
+
+ INIT_LIST_HEAD(&lacl->lacl_list);
+ lacl->mapped_lun = mapped_lun;
+ lacl->se_lun_nacl = nacl;
+ snprintf(lacl->initiatorname, TRANSPORT_IQN_LEN, "%s", initiatorname);
+
+ return lacl;
+}
+
+int core_dev_add_initiator_node_lun_acl(
+ struct se_portal_group *tpg,
+ struct se_lun_acl *lacl,
+ u32 unpacked_lun,
+ u32 lun_access)
+{
+ struct se_lun *lun;
+ struct se_node_acl *nacl;
+
+ lun = core_dev_get_lun(tpg, unpacked_lun);
+ if (!(lun)) {
+ printk(KERN_ERR "%s Logical Unit Number: %u is not active on"
+ " Target Portal Group: %hu, ignoring request.\n",
+ TPG_TFO(tpg)->get_fabric_name(), unpacked_lun,
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ return -EINVAL;
+ }
+
+ nacl = lacl->se_lun_nacl;
+ if (!(nacl))
+ return -EINVAL;
+
+ if ((lun->lun_access & TRANSPORT_LUNFLAGS_READ_ONLY) &&
+ (lun_access & TRANSPORT_LUNFLAGS_READ_WRITE))
+ lun_access = TRANSPORT_LUNFLAGS_READ_ONLY;
+
+ lacl->se_lun = lun;
+
+ if (core_update_device_list_for_node(lun, lacl, lacl->mapped_lun,
+ lun_access, nacl, tpg, 1) < 0)
+ return -EINVAL;
+
+ spin_lock(&lun->lun_acl_lock);
+ list_add_tail(&lacl->lacl_list, &lun->lun_acl_list);
+ atomic_inc(&lun->lun_acl_count);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&lun->lun_acl_lock);
+
+ printk(KERN_INFO "%s_TPG[%hu]_LUN[%u->%u] - Added %s ACL for "
+ " InitiatorNode: %s\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), unpacked_lun, lacl->mapped_lun,
+ (lun_access & TRANSPORT_LUNFLAGS_READ_WRITE) ? "RW" : "RO",
+ lacl->initiatorname);
+ /*
+ * Check to see if there are any existing persistent reservation APTPL
+ * pre-registrations that need to be enabled for this LUN ACL..
+ */
+ core_scsi3_check_aptpl_registration(lun->lun_se_dev, tpg, lun, lacl);
+ return 0;
+}
+
+/* core_dev_del_initiator_node_lun_acl():
+ *
+ *
+ */
+int core_dev_del_initiator_node_lun_acl(
+ struct se_portal_group *tpg,
+ struct se_lun *lun,
+ struct se_lun_acl *lacl)
+{
+ struct se_node_acl *nacl;
+
+ nacl = lacl->se_lun_nacl;
+ if (!(nacl))
+ return -EINVAL;
+
+ spin_lock(&lun->lun_acl_lock);
+ list_del(&lacl->lacl_list);
+ atomic_dec(&lun->lun_acl_count);
+ smp_mb__after_atomic_dec();
+ spin_unlock(&lun->lun_acl_lock);
+
+ core_update_device_list_for_node(lun, NULL, lacl->mapped_lun,
+ TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg, 0);
+
+ lacl->se_lun = NULL;
+
+ printk(KERN_INFO "%s_TPG[%hu]_LUN[%u] - Removed ACL for"
+ " InitiatorNode: %s Mapped LUN: %u\n",
+ TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), lun->unpacked_lun,
+ lacl->initiatorname, lacl->mapped_lun);
+
+ return 0;
+}
+
+void core_dev_free_initiator_node_lun_acl(
+ struct se_portal_group *tpg,
+ struct se_lun_acl *lacl)
+{
+ printk("%s_TPG[%hu] - Freeing ACL for %s InitiatorNode: %s"
+ " Mapped LUN: %u\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg),
+ TPG_TFO(tpg)->get_fabric_name(),
+ lacl->initiatorname, lacl->mapped_lun);
+
+ kfree(lacl);
+}
+
+int core_dev_setup_virtual_lun0(void)
+{
+ struct se_hba *hba;
+ struct se_device *dev;
+ struct se_subsystem_dev *se_dev = NULL;
+ struct se_subsystem_api *t;
+ char buf[16];
+ int ret;
+
+ hba = core_alloc_hba("rd_dr", 0, HBA_FLAGS_INTERNAL_USE);
+ if (IS_ERR(hba))
+ return PTR_ERR(hba);
+
+ se_global->g_lun0_hba = hba;
+ t = hba->transport;
+
+ se_dev = kzalloc(sizeof(struct se_subsystem_dev), GFP_KERNEL);
+ if (!(se_dev)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " struct se_subsystem_dev\n");
+ ret = -ENOMEM;
+ goto out;
+ }
+ INIT_LIST_HEAD(&se_dev->g_se_dev_list);
+ 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_reservation.registration_list);
+ INIT_LIST_HEAD(&se_dev->t10_reservation.aptpl_reg_list);
+ spin_lock_init(&se_dev->t10_reservation.registration_lock);
+ spin_lock_init(&se_dev->t10_reservation.aptpl_reg_lock);
+ INIT_LIST_HEAD(&se_dev->t10_alua.tg_pt_gps_list);
+ spin_lock_init(&se_dev->t10_alua.tg_pt_gps_lock);
+ spin_lock_init(&se_dev->se_dev_lock);
+ se_dev->t10_reservation.pr_aptpl_buf_len = PR_APTPL_BUF_LEN;
+ se_dev->t10_wwn.t10_sub_dev = se_dev;
+ se_dev->t10_alua.t10_sub_dev = se_dev;
+ se_dev->se_dev_attrib.da_sub_dev = se_dev;
+ se_dev->se_dev_hba = hba;
+
+ se_dev->se_dev_su_ptr = t->allocate_virtdevice(hba, "virt_lun0");
+ if (!(se_dev->se_dev_su_ptr)) {
+ printk(KERN_ERR "Unable to locate subsystem dependent pointer"
+ " from allocate_virtdevice()\n");
+ ret = -ENOMEM;
+ goto out;
+ }
+ se_global->g_lun0_su_dev = se_dev;
+
+ memset(buf, 0, 16);
+ sprintf(buf, "rd_pages=8");
+ t->set_configfs_dev_params(hba, se_dev, buf, sizeof(buf));
+
+ dev = t->create_virtdevice(hba, se_dev, se_dev->se_dev_su_ptr);
+ if (!(dev) || IS_ERR(dev)) {
+ ret = -ENOMEM;
+ goto out;
+ }
+ se_dev->se_dev_ptr = dev;
+ se_global->g_lun0_dev = dev;
+
+ return 0;
+out:
+ se_global->g_lun0_su_dev = NULL;
+ kfree(se_dev);
+ if (se_global->g_lun0_hba) {
+ core_delete_hba(se_global->g_lun0_hba);
+ se_global->g_lun0_hba = NULL;
+ }
+ return ret;
+}
+
+
+void core_dev_release_virtual_lun0(void)
+{
+ struct se_hba *hba = se_global->g_lun0_hba;
+ struct se_subsystem_dev *su_dev = se_global->g_lun0_su_dev;
+
+ if (!(hba))
+ return;
+
+ if (se_global->g_lun0_dev)
+ se_free_virtual_device(se_global->g_lun0_dev, hba);
+
+ kfree(su_dev);
+ core_delete_hba(hba);
+}
diff --git a/drivers/target/target_core_fabric_configfs.c b/drivers/target/target_core_fabric_configfs.c
new file mode 100644
index 0000000..32b148d
--- /dev/null
+++ b/drivers/target/target_core_fabric_configfs.c
@@ -0,0 +1,996 @@
+/*******************************************************************************
+* Filename: target_core_fabric_configfs.c
+ *
+ * This file contains generic fabric module configfs infrastructure for
+ * TCM v4.x code
+ *
+ * Copyright (c) 2010 Rising Tide Systems
+ * Copyright (c) 2010 Linux-iSCSI.org
+ *
+ * Copyright (c) 2010 Nicholas A. Bellinger <nab@linux-iscsi.org>
+*
+ * 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.
+ *
+ * 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.
+ ****************************************************************************/
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/version.h>
+#include <generated/utsrelease.h>
+#include <linux/utsname.h>
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/namei.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/delay.h>
+#include <linux/unistd.h>
+#include <linux/string.h>
+#include <linux/syscalls.h>
+#include <linux/configfs.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_tpg.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_fabric_configfs.h>
+#include <target/target_core_configfs.h>
+#include <target/configfs_macros.h>
+
+#include "target_core_alua.h"
+#include "target_core_hba.h"
+#include "target_core_pr.h"
+
+#define TF_CIT_SETUP(_name, _item_ops, _group_ops, _attrs) \
+static void target_fabric_setup_##_name##_cit(struct target_fabric_configfs *tf) \
+{ \
+ struct target_fabric_configfs_template *tfc = &tf->tf_cit_tmpl; \
+ struct config_item_type *cit = &tfc->tfc_##_name##_cit; \
+ \
+ cit->ct_item_ops = _item_ops; \
+ cit->ct_group_ops = _group_ops; \
+ cit->ct_attrs = _attrs; \
+ cit->ct_owner = tf->tf_module; \
+ printk("Setup generic %s\n", __stringify(_name)); \
+}
+
+/* Start of tfc_tpg_mappedlun_cit */
+
+static int target_fabric_mappedlun_link(
+ struct config_item *lun_acl_ci,
+ struct config_item *lun_ci)
+{
+ struct se_dev_entry *deve;
+ struct se_lun *lun = container_of(to_config_group(lun_ci),
+ struct se_lun, lun_group);
+ struct se_lun_acl *lacl = container_of(to_config_group(lun_acl_ci),
+ struct se_lun_acl, se_lun_group);
+ struct se_portal_group *se_tpg;
+ struct config_item *nacl_ci, *tpg_ci, *tpg_ci_s, *wwn_ci, *wwn_ci_s;
+ int ret = 0, lun_access;
+ /*
+ * Ensure that the source port exists
+ */
+ if (!(lun->lun_sep) || !(lun->lun_sep->sep_tpg)) {
+ printk(KERN_ERR "Source se_lun->lun_sep or lun->lun_sep->sep"
+ "_tpg does not exist\n");
+ return -EINVAL;
+ }
+ se_tpg = lun->lun_sep->sep_tpg;
+
+ nacl_ci = &lun_acl_ci->ci_parent->ci_group->cg_item;
+ tpg_ci = &nacl_ci->ci_group->cg_item;
+ wwn_ci = &tpg_ci->ci_group->cg_item;
+ tpg_ci_s = &lun_ci->ci_parent->ci_group->cg_item;
+ wwn_ci_s = &tpg_ci_s->ci_group->cg_item;
+ /*
+ * Make sure the SymLink is going to the same $FABRIC/$WWN/tpgt_$TPGT
+ */
+ if (strcmp(config_item_name(wwn_ci), config_item_name(wwn_ci_s))) {
+ printk(KERN_ERR "Illegal Initiator ACL SymLink outside of %s\n",
+ config_item_name(wwn_ci));
+ return -EINVAL;
+ }
+ if (strcmp(config_item_name(tpg_ci), config_item_name(tpg_ci_s))) {
+ printk(KERN_ERR "Illegal Initiator ACL Symlink outside of %s"
+ " TPGT: %s\n", config_item_name(wwn_ci),
+ config_item_name(tpg_ci));
+ return -EINVAL;
+ }
+ /*
+ * If this struct se_node_acl was dynamically generated with
+ * tpg_1/attrib/generate_node_acls=1, use the existing deve->lun_flags,
+ * which be will write protected (READ-ONLY) when
+ * tpg_1/attrib/demo_mode_write_protect=1
+ */
+ spin_lock_irq(&lacl->se_lun_nacl->device_list_lock);
+ deve = &lacl->se_lun_nacl->device_list[lacl->mapped_lun];
+ if (deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS)
+ lun_access = deve->lun_flags;
+ else
+ lun_access =
+ (TPG_TFO(se_tpg)->tpg_check_prod_mode_write_protect(
+ se_tpg)) ? TRANSPORT_LUNFLAGS_READ_ONLY :
+ TRANSPORT_LUNFLAGS_READ_WRITE;
+ spin_unlock_irq(&lacl->se_lun_nacl->device_list_lock);
+ /*
+ * Determine the actual mapped LUN value user wants..
+ *
+ * This value is what the SCSI Initiator actually sees the
+ * iscsi/$IQN/$TPGT/lun/lun_* as on their SCSI Initiator Ports.
+ */
+ ret = core_dev_add_initiator_node_lun_acl(se_tpg, lacl,
+ lun->unpacked_lun, lun_access);
+
+ return (ret < 0) ? -EINVAL : 0;
+}
+
+static int target_fabric_mappedlun_unlink(
+ struct config_item *lun_acl_ci,
+ struct config_item *lun_ci)
+{
+ struct se_lun *lun;
+ struct se_lun_acl *lacl = container_of(to_config_group(lun_acl_ci),
+ struct se_lun_acl, se_lun_group);
+ struct se_node_acl *nacl = lacl->se_lun_nacl;
+ struct se_dev_entry *deve = &nacl->device_list[lacl->mapped_lun];
+ struct se_portal_group *se_tpg;
+ /*
+ * Determine if the underlying MappedLUN has already been released..
+ */
+ if (!(deve->se_lun))
+ return 0;
+
+ lun = container_of(to_config_group(lun_ci), struct se_lun, lun_group);
+ se_tpg = lun->lun_sep->sep_tpg;
+
+ core_dev_del_initiator_node_lun_acl(se_tpg, lun, lacl);
+ return 0;
+}
+
+CONFIGFS_EATTR_STRUCT(target_fabric_mappedlun, se_lun_acl);
+#define TCM_MAPPEDLUN_ATTR(_name, _mode) \
+static struct target_fabric_mappedlun_attribute target_fabric_mappedlun_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ target_fabric_mappedlun_show_##_name, \
+ target_fabric_mappedlun_store_##_name);
+
+static ssize_t target_fabric_mappedlun_show_write_protect(
+ struct se_lun_acl *lacl,
+ char *page)
+{
+ struct se_node_acl *se_nacl = lacl->se_lun_nacl;
+ struct se_dev_entry *deve;
+ ssize_t len;
+
+ spin_lock_irq(&se_nacl->device_list_lock);
+ deve = &se_nacl->device_list[lacl->mapped_lun];
+ len = sprintf(page, "%d\n",
+ (deve->lun_flags & TRANSPORT_LUNFLAGS_READ_ONLY) ?
+ 1 : 0);
+ spin_unlock_irq(&se_nacl->device_list_lock);
+
+ return len;
+}
+
+static ssize_t target_fabric_mappedlun_store_write_protect(
+ struct se_lun_acl *lacl,
+ const char *page,
+ size_t count)
+{
+ struct se_node_acl *se_nacl = lacl->se_lun_nacl;
+ struct se_portal_group *se_tpg = se_nacl->se_tpg;
+ unsigned long op;
+
+ if (strict_strtoul(page, 0, &op))
+ return -EINVAL;
+
+ if ((op != 1) && (op != 0))
+ return -EINVAL;
+
+ core_update_device_list_access(lacl->mapped_lun, (op) ?
+ TRANSPORT_LUNFLAGS_READ_ONLY :
+ TRANSPORT_LUNFLAGS_READ_WRITE,
+ lacl->se_lun_nacl);
+
+ printk(KERN_INFO "%s_ConfigFS: Changed Initiator ACL: %s"
+ " Mapped LUN: %u Write Protect bit to %s\n",
+ TPG_TFO(se_tpg)->get_fabric_name(),
+ lacl->initiatorname, lacl->mapped_lun, (op) ? "ON" : "OFF");
+
+ return count;
+
+}
+
+TCM_MAPPEDLUN_ATTR(write_protect, S_IRUGO | S_IWUSR);
+
+CONFIGFS_EATTR_OPS(target_fabric_mappedlun, se_lun_acl, se_lun_group);
+
+static struct configfs_attribute *target_fabric_mappedlun_attrs[] = {
+ &target_fabric_mappedlun_write_protect.attr,
+ NULL,
+};
+
+static struct configfs_item_operations target_fabric_mappedlun_item_ops = {
+ .show_attribute = target_fabric_mappedlun_attr_show,
+ .store_attribute = target_fabric_mappedlun_attr_store,
+ .allow_link = target_fabric_mappedlun_link,
+ .drop_link = target_fabric_mappedlun_unlink,
+};
+
+TF_CIT_SETUP(tpg_mappedlun, &target_fabric_mappedlun_item_ops, NULL,
+ target_fabric_mappedlun_attrs);
+
+/* End of tfc_tpg_mappedlun_cit */
+
+/* Start of tfc_tpg_nacl_attrib_cit */
+
+CONFIGFS_EATTR_OPS(target_fabric_nacl_attrib, se_node_acl, acl_attrib_group);
+
+static struct configfs_item_operations target_fabric_nacl_attrib_item_ops = {
+ .show_attribute = target_fabric_nacl_attrib_attr_show,
+ .store_attribute = target_fabric_nacl_attrib_attr_store,
+};
+
+TF_CIT_SETUP(tpg_nacl_attrib, &target_fabric_nacl_attrib_item_ops, NULL, NULL);
+
+/* End of tfc_tpg_nacl_attrib_cit */
+
+/* Start of tfc_tpg_nacl_auth_cit */
+
+CONFIGFS_EATTR_OPS(target_fabric_nacl_auth, se_node_acl, acl_auth_group);
+
+static struct configfs_item_operations target_fabric_nacl_auth_item_ops = {
+ .show_attribute = target_fabric_nacl_auth_attr_show,
+ .store_attribute = target_fabric_nacl_auth_attr_store,
+};
+
+TF_CIT_SETUP(tpg_nacl_auth, &target_fabric_nacl_auth_item_ops, NULL, NULL);
+
+/* End of tfc_tpg_nacl_auth_cit */
+
+/* Start of tfc_tpg_nacl_param_cit */
+
+CONFIGFS_EATTR_OPS(target_fabric_nacl_param, se_node_acl, acl_param_group);
+
+static struct configfs_item_operations target_fabric_nacl_param_item_ops = {
+ .show_attribute = target_fabric_nacl_param_attr_show,
+ .store_attribute = target_fabric_nacl_param_attr_store,
+};
+
+TF_CIT_SETUP(tpg_nacl_param, &target_fabric_nacl_param_item_ops, NULL, NULL);
+
+/* End of tfc_tpg_nacl_param_cit */
+
+/* Start of tfc_tpg_nacl_base_cit */
+
+CONFIGFS_EATTR_OPS(target_fabric_nacl_base, se_node_acl, acl_group);
+
+static struct config_group *target_fabric_make_mappedlun(
+ struct config_group *group,
+ const char *name)
+{
+ struct se_node_acl *se_nacl = container_of(group,
+ struct se_node_acl, acl_group);
+ struct se_portal_group *se_tpg = se_nacl->se_tpg;
+ struct target_fabric_configfs *tf = se_tpg->se_tpg_wwn->wwn_tf;
+ struct se_lun_acl *lacl;
+ struct config_item *acl_ci;
+ char *buf;
+ unsigned long mapped_lun;
+ int ret = 0;
+
+ acl_ci = &group->cg_item;
+ if (!(acl_ci)) {
+ printk(KERN_ERR "Unable to locatel acl_ci\n");
+ return NULL;
+ }
+
+ buf = kzalloc(strlen(name) + 1, GFP_KERNEL);
+ if (!(buf)) {
+ printk(KERN_ERR "Unable to allocate memory for name buf\n");
+ return ERR_PTR(-ENOMEM);
+ }
+ snprintf(buf, strlen(name) + 1, "%s", name);
+ /*
+ * Make sure user is creating iscsi/$IQN/$TPGT/acls/$INITIATOR/lun_$ID.
+ */
+ if (strstr(buf, "lun_") != buf) {
+ printk(KERN_ERR "Unable to locate \"lun_\" from buf: %s"
+ " name: %s\n", buf, name);
+ ret = -EINVAL;
+ goto out;
+ }
+ /*
+ * Determine the Mapped LUN value. This is what the SCSI Initiator
+ * Port will actually see.
+ */
+ if (strict_strtoul(buf + 4, 0, &mapped_lun) || mapped_lun > UINT_MAX) {
+ ret = -EINVAL;
+ goto out;
+ }
+
+ lacl = core_dev_init_initiator_node_lun_acl(se_tpg, mapped_lun,
+ config_item_name(acl_ci), &ret);
+ if (!(lacl))
+ goto out;
+
+ config_group_init_type_name(&lacl->se_lun_group, name,
+ &TF_CIT_TMPL(tf)->tfc_tpg_mappedlun_cit);
+
+ kfree(buf);
+ return &lacl->se_lun_group;
+out:
+ kfree(buf);
+ return ERR_PTR(ret);
+}
+
+static void target_fabric_drop_mappedlun(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct se_lun_acl *lacl = container_of(to_config_group(item),
+ struct se_lun_acl, se_lun_group);
+ struct se_portal_group *se_tpg = lacl->se_lun_nacl->se_tpg;
+
+ config_item_put(item);
+ core_dev_free_initiator_node_lun_acl(se_tpg, lacl);
+}
+
+static struct configfs_item_operations target_fabric_nacl_base_item_ops = {
+ .show_attribute = target_fabric_nacl_base_attr_show,
+ .store_attribute = target_fabric_nacl_base_attr_store,
+};
+
+static struct configfs_group_operations target_fabric_nacl_base_group_ops = {
+ .make_group = target_fabric_make_mappedlun,
+ .drop_item = target_fabric_drop_mappedlun,
+};
+
+TF_CIT_SETUP(tpg_nacl_base, &target_fabric_nacl_base_item_ops,
+ &target_fabric_nacl_base_group_ops, NULL);
+
+/* End of tfc_tpg_nacl_base_cit */
+
+/* Start of tfc_tpg_nacl_cit */
+
+static struct config_group *target_fabric_make_nodeacl(
+ struct config_group *group,
+ const char *name)
+{
+ struct se_portal_group *se_tpg = container_of(group,
+ struct se_portal_group, tpg_acl_group);
+ struct target_fabric_configfs *tf = se_tpg->se_tpg_wwn->wwn_tf;
+ struct se_node_acl *se_nacl;
+ struct config_group *nacl_cg;
+
+ if (!(tf->tf_ops.fabric_make_nodeacl)) {
+ printk(KERN_ERR "tf->tf_ops.fabric_make_nodeacl is NULL\n");
+ return ERR_PTR(-ENOSYS);
+ }
+
+ se_nacl = tf->tf_ops.fabric_make_nodeacl(se_tpg, group, name);
+ if (IS_ERR(se_nacl))
+ return ERR_PTR(PTR_ERR(se_nacl));
+
+ nacl_cg = &se_nacl->acl_group;
+ nacl_cg->default_groups = se_nacl->acl_default_groups;
+ nacl_cg->default_groups[0] = &se_nacl->acl_attrib_group;
+ nacl_cg->default_groups[1] = &se_nacl->acl_auth_group;
+ nacl_cg->default_groups[2] = &se_nacl->acl_param_group;
+ nacl_cg->default_groups[3] = NULL;
+
+ config_group_init_type_name(&se_nacl->acl_group, name,
+ &TF_CIT_TMPL(tf)->tfc_tpg_nacl_base_cit);
+ config_group_init_type_name(&se_nacl->acl_attrib_group, "attrib",
+ &TF_CIT_TMPL(tf)->tfc_tpg_nacl_attrib_cit);
+ config_group_init_type_name(&se_nacl->acl_auth_group, "auth",
+ &TF_CIT_TMPL(tf)->tfc_tpg_nacl_auth_cit);
+ config_group_init_type_name(&se_nacl->acl_param_group, "param",
+ &TF_CIT_TMPL(tf)->tfc_tpg_nacl_param_cit);
+
+ return &se_nacl->acl_group;
+}
+
+static void target_fabric_drop_nodeacl(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct se_portal_group *se_tpg = container_of(group,
+ struct se_portal_group, tpg_acl_group);
+ struct target_fabric_configfs *tf = se_tpg->se_tpg_wwn->wwn_tf;
+ struct se_node_acl *se_nacl = container_of(to_config_group(item),
+ struct se_node_acl, acl_group);
+ struct config_item *df_item;
+ struct config_group *nacl_cg;
+ int i;
+
+ nacl_cg = &se_nacl->acl_group;
+ for (i = 0; nacl_cg->default_groups[i]; i++) {
+ df_item = &nacl_cg->default_groups[i]->cg_item;
+ nacl_cg->default_groups[i] = NULL;
+ config_item_put(df_item);
+ }
+
+ config_item_put(item);
+ tf->tf_ops.fabric_drop_nodeacl(se_nacl);
+}
+
+static struct configfs_group_operations target_fabric_nacl_group_ops = {
+ .make_group = target_fabric_make_nodeacl,
+ .drop_item = target_fabric_drop_nodeacl,
+};
+
+TF_CIT_SETUP(tpg_nacl, NULL, &target_fabric_nacl_group_ops, NULL);
+
+/* End of tfc_tpg_nacl_cit */
+
+/* Start of tfc_tpg_np_base_cit */
+
+CONFIGFS_EATTR_OPS(target_fabric_np_base, se_tpg_np, tpg_np_group);
+
+static struct configfs_item_operations target_fabric_np_base_item_ops = {
+ .show_attribute = target_fabric_np_base_attr_show,
+ .store_attribute = target_fabric_np_base_attr_store,
+};
+
+TF_CIT_SETUP(tpg_np_base, &target_fabric_np_base_item_ops, NULL, NULL);
+
+/* End of tfc_tpg_np_base_cit */
+
+/* Start of tfc_tpg_np_cit */
+
+static struct config_group *target_fabric_make_np(
+ struct config_group *group,
+ const char *name)
+{
+ struct se_portal_group *se_tpg = container_of(group,
+ struct se_portal_group, tpg_np_group);
+ struct target_fabric_configfs *tf = se_tpg->se_tpg_wwn->wwn_tf;
+ struct se_tpg_np *se_tpg_np;
+
+ if (!(tf->tf_ops.fabric_make_np)) {
+ printk(KERN_ERR "tf->tf_ops.fabric_make_np is NULL\n");
+ return ERR_PTR(-ENOSYS);
+ }
+
+ se_tpg_np = tf->tf_ops.fabric_make_np(se_tpg, group, name);
+ if (!(se_tpg_np) || IS_ERR(se_tpg_np))
+ return ERR_PTR(-EINVAL);
+
+ config_group_init_type_name(&se_tpg_np->tpg_np_group, name,
+ &TF_CIT_TMPL(tf)->tfc_tpg_np_base_cit);
+
+ return &se_tpg_np->tpg_np_group;
+}
+
+static void target_fabric_drop_np(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct se_portal_group *se_tpg = container_of(group,
+ struct se_portal_group, tpg_np_group);
+ struct target_fabric_configfs *tf = se_tpg->se_tpg_wwn->wwn_tf;
+ struct se_tpg_np *se_tpg_np = container_of(to_config_group(item),
+ struct se_tpg_np, tpg_np_group);
+
+ config_item_put(item);
+ tf->tf_ops.fabric_drop_np(se_tpg_np);
+}
+
+static struct configfs_group_operations target_fabric_np_group_ops = {
+ .make_group = &target_fabric_make_np,
+ .drop_item = &target_fabric_drop_np,
+};
+
+TF_CIT_SETUP(tpg_np, NULL, &target_fabric_np_group_ops, NULL);
+
+/* End of tfc_tpg_np_cit */
+
+/* Start of tfc_tpg_port_cit */
+
+CONFIGFS_EATTR_STRUCT(target_fabric_port, se_lun);
+#define TCM_PORT_ATTR(_name, _mode) \
+static struct target_fabric_port_attribute target_fabric_port_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ target_fabric_port_show_attr_##_name, \
+ target_fabric_port_store_attr_##_name);
+
+#define TCM_PORT_ATTOR_RO(_name) \
+ __CONFIGFS_EATTR_RO(_name, \
+ target_fabric_port_show_attr_##_name);
+
+/*
+ * alua_tg_pt_gp
+ */
+static ssize_t target_fabric_port_show_attr_alua_tg_pt_gp(
+ struct se_lun *lun,
+ char *page)
+{
+ if (!(lun))
+ return -ENODEV;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return core_alua_show_tg_pt_gp_info(lun->lun_sep, page);
+}
+
+static ssize_t target_fabric_port_store_attr_alua_tg_pt_gp(
+ struct se_lun *lun,
+ const char *page,
+ size_t count)
+{
+ if (!(lun))
+ return -ENODEV;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return core_alua_store_tg_pt_gp_info(lun->lun_sep, page, count);
+}
+
+TCM_PORT_ATTR(alua_tg_pt_gp, S_IRUGO | S_IWUSR);
+
+/*
+ * alua_tg_pt_offline
+ */
+static ssize_t target_fabric_port_show_attr_alua_tg_pt_offline(
+ struct se_lun *lun,
+ char *page)
+{
+ if (!(lun))
+ return -ENODEV;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return core_alua_show_offline_bit(lun, page);
+}
+
+static ssize_t target_fabric_port_store_attr_alua_tg_pt_offline(
+ struct se_lun *lun,
+ const char *page,
+ size_t count)
+{
+ if (!(lun))
+ return -ENODEV;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return core_alua_store_offline_bit(lun, page, count);
+}
+
+TCM_PORT_ATTR(alua_tg_pt_offline, S_IRUGO | S_IWUSR);
+
+/*
+ * alua_tg_pt_status
+ */
+static ssize_t target_fabric_port_show_attr_alua_tg_pt_status(
+ struct se_lun *lun,
+ char *page)
+{
+ if (!(lun))
+ return -ENODEV;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return core_alua_show_secondary_status(lun, page);
+}
+
+static ssize_t target_fabric_port_store_attr_alua_tg_pt_status(
+ struct se_lun *lun,
+ const char *page,
+ size_t count)
+{
+ if (!(lun))
+ return -ENODEV;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return core_alua_store_secondary_status(lun, page, count);
+}
+
+TCM_PORT_ATTR(alua_tg_pt_status, S_IRUGO | S_IWUSR);
+
+/*
+ * alua_tg_pt_write_md
+ */
+static ssize_t target_fabric_port_show_attr_alua_tg_pt_write_md(
+ struct se_lun *lun,
+ char *page)
+{
+ if (!(lun))
+ return -ENODEV;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return core_alua_show_secondary_write_metadata(lun, page);
+}
+
+static ssize_t target_fabric_port_store_attr_alua_tg_pt_write_md(
+ struct se_lun *lun,
+ const char *page,
+ size_t count)
+{
+ if (!(lun))
+ return -ENODEV;
+
+ if (!(lun->lun_sep))
+ return -ENODEV;
+
+ return core_alua_store_secondary_write_metadata(lun, page, count);
+}
+
+TCM_PORT_ATTR(alua_tg_pt_write_md, S_IRUGO | S_IWUSR);
+
+
+static struct configfs_attribute *target_fabric_port_attrs[] = {
+ &target_fabric_port_alua_tg_pt_gp.attr,
+ &target_fabric_port_alua_tg_pt_offline.attr,
+ &target_fabric_port_alua_tg_pt_status.attr,
+ &target_fabric_port_alua_tg_pt_write_md.attr,
+ NULL,
+};
+
+CONFIGFS_EATTR_OPS(target_fabric_port, se_lun, lun_group);
+
+static int target_fabric_port_link(
+ struct config_item *lun_ci,
+ struct config_item *se_dev_ci)
+{
+ struct config_item *tpg_ci;
+ struct se_device *dev;
+ struct se_lun *lun = container_of(to_config_group(lun_ci),
+ struct se_lun, lun_group);
+ struct se_lun *lun_p;
+ struct se_portal_group *se_tpg;
+ struct se_subsystem_dev *se_dev = container_of(
+ to_config_group(se_dev_ci), struct se_subsystem_dev,
+ se_dev_group);
+ struct target_fabric_configfs *tf;
+ int ret;
+
+ tpg_ci = &lun_ci->ci_parent->ci_group->cg_item;
+ se_tpg = container_of(to_config_group(tpg_ci),
+ struct se_portal_group, tpg_group);
+ tf = se_tpg->se_tpg_wwn->wwn_tf;
+
+ if (lun->lun_se_dev != NULL) {
+ printk(KERN_ERR "Port Symlink already exists\n");
+ return -EEXIST;
+ }
+
+ dev = se_dev->se_dev_ptr;
+ if (!(dev)) {
+ printk(KERN_ERR "Unable to locate struct se_device pointer from"
+ " %s\n", config_item_name(se_dev_ci));
+ ret = -ENODEV;
+ goto out;
+ }
+
+ lun_p = core_dev_add_lun(se_tpg, dev->se_hba, dev,
+ lun->unpacked_lun);
+ if ((IS_ERR(lun_p)) || !(lun_p)) {
+ printk(KERN_ERR "core_dev_add_lun() failed\n");
+ ret = -EINVAL;
+ goto out;
+ }
+
+ if (tf->tf_ops.fabric_post_link) {
+ /*
+ * Call the optional fabric_post_link() to allow a
+ * fabric module to setup any additional state once
+ * core_dev_add_lun() has been called..
+ */
+ tf->tf_ops.fabric_post_link(se_tpg, lun);
+ }
+
+ return 0;
+out:
+ return ret;
+}
+
+static int target_fabric_port_unlink(
+ struct config_item *lun_ci,
+ struct config_item *se_dev_ci)
+{
+ struct se_lun *lun = container_of(to_config_group(lun_ci),
+ struct se_lun, lun_group);
+ struct se_portal_group *se_tpg = lun->lun_sep->sep_tpg;
+ struct target_fabric_configfs *tf = se_tpg->se_tpg_wwn->wwn_tf;
+
+ if (tf->tf_ops.fabric_pre_unlink) {
+ /*
+ * Call the optional fabric_pre_unlink() to allow a
+ * fabric module to release any additional stat before
+ * core_dev_del_lun() is called.
+ */
+ tf->tf_ops.fabric_pre_unlink(se_tpg, lun);
+ }
+
+ core_dev_del_lun(se_tpg, lun->unpacked_lun);
+ return 0;
+}
+
+static struct configfs_item_operations target_fabric_port_item_ops = {
+ .show_attribute = target_fabric_port_attr_show,
+ .store_attribute = target_fabric_port_attr_store,
+ .allow_link = target_fabric_port_link,
+ .drop_link = target_fabric_port_unlink,
+};
+
+TF_CIT_SETUP(tpg_port, &target_fabric_port_item_ops, NULL, target_fabric_port_attrs);
+
+/* End of tfc_tpg_port_cit */
+
+/* Start of tfc_tpg_lun_cit */
+
+static struct config_group *target_fabric_make_lun(
+ struct config_group *group,
+ const char *name)
+{
+ struct se_lun *lun;
+ struct se_portal_group *se_tpg = container_of(group,
+ struct se_portal_group, tpg_lun_group);
+ struct target_fabric_configfs *tf = se_tpg->se_tpg_wwn->wwn_tf;
+ unsigned long unpacked_lun;
+
+ if (strstr(name, "lun_") != name) {
+ printk(KERN_ERR "Unable to locate \'_\" in"
+ " \"lun_$LUN_NUMBER\"\n");
+ return ERR_PTR(-EINVAL);
+ }
+ if (strict_strtoul(name + 4, 0, &unpacked_lun) || unpacked_lun > UINT_MAX)
+ return ERR_PTR(-EINVAL);
+
+ lun = core_get_lun_from_tpg(se_tpg, unpacked_lun);
+ if (!(lun))
+ return ERR_PTR(-EINVAL);
+
+ config_group_init_type_name(&lun->lun_group, name,
+ &TF_CIT_TMPL(tf)->tfc_tpg_port_cit);
+
+ return &lun->lun_group;
+}
+
+static void target_fabric_drop_lun(
+ struct config_group *group,
+ struct config_item *item)
+{
+ config_item_put(item);
+}
+
+static struct configfs_group_operations target_fabric_lun_group_ops = {
+ .make_group = &target_fabric_make_lun,
+ .drop_item = &target_fabric_drop_lun,
+};
+
+TF_CIT_SETUP(tpg_lun, NULL, &target_fabric_lun_group_ops, NULL);
+
+/* End of tfc_tpg_lun_cit */
+
+/* Start of tfc_tpg_attrib_cit */
+
+CONFIGFS_EATTR_OPS(target_fabric_tpg_attrib, se_portal_group, tpg_attrib_group);
+
+static struct configfs_item_operations target_fabric_tpg_attrib_item_ops = {
+ .show_attribute = target_fabric_tpg_attrib_attr_show,
+ .store_attribute = target_fabric_tpg_attrib_attr_store,
+};
+
+TF_CIT_SETUP(tpg_attrib, &target_fabric_tpg_attrib_item_ops, NULL, NULL);
+
+/* End of tfc_tpg_attrib_cit */
+
+/* Start of tfc_tpg_param_cit */
+
+CONFIGFS_EATTR_OPS(target_fabric_tpg_param, se_portal_group, tpg_param_group);
+
+static struct configfs_item_operations target_fabric_tpg_param_item_ops = {
+ .show_attribute = target_fabric_tpg_param_attr_show,
+ .store_attribute = target_fabric_tpg_param_attr_store,
+};
+
+TF_CIT_SETUP(tpg_param, &target_fabric_tpg_param_item_ops, NULL, NULL);
+
+/* End of tfc_tpg_param_cit */
+
+/* Start of tfc_tpg_base_cit */
+/*
+ * For use with TF_TPG_ATTR() and TF_TPG_ATTR_RO()
+ */
+CONFIGFS_EATTR_OPS(target_fabric_tpg, se_portal_group, tpg_group);
+
+static struct configfs_item_operations target_fabric_tpg_base_item_ops = {
+ .show_attribute = target_fabric_tpg_attr_show,
+ .store_attribute = target_fabric_tpg_attr_store,
+};
+
+TF_CIT_SETUP(tpg_base, &target_fabric_tpg_base_item_ops, NULL, NULL);
+
+/* End of tfc_tpg_base_cit */
+
+/* Start of tfc_tpg_cit */
+
+static struct config_group *target_fabric_make_tpg(
+ struct config_group *group,
+ const char *name)
+{
+ struct se_wwn *wwn = container_of(group, struct se_wwn, wwn_group);
+ struct target_fabric_configfs *tf = wwn->wwn_tf;
+ struct se_portal_group *se_tpg;
+
+ if (!(tf->tf_ops.fabric_make_tpg)) {
+ printk(KERN_ERR "tf->tf_ops.fabric_make_tpg is NULL\n");
+ return ERR_PTR(-ENOSYS);
+ }
+
+ se_tpg = tf->tf_ops.fabric_make_tpg(wwn, group, name);
+ if (!(se_tpg) || IS_ERR(se_tpg))
+ return ERR_PTR(-EINVAL);
+ /*
+ * Setup default groups from pre-allocated se_tpg->tpg_default_groups
+ */
+ se_tpg->tpg_group.default_groups = se_tpg->tpg_default_groups;
+ se_tpg->tpg_group.default_groups[0] = &se_tpg->tpg_lun_group;
+ se_tpg->tpg_group.default_groups[1] = &se_tpg->tpg_np_group;
+ se_tpg->tpg_group.default_groups[2] = &se_tpg->tpg_acl_group;
+ se_tpg->tpg_group.default_groups[3] = &se_tpg->tpg_attrib_group;
+ se_tpg->tpg_group.default_groups[4] = &se_tpg->tpg_param_group;
+ se_tpg->tpg_group.default_groups[5] = NULL;
+
+ config_group_init_type_name(&se_tpg->tpg_group, name,
+ &TF_CIT_TMPL(tf)->tfc_tpg_base_cit);
+ config_group_init_type_name(&se_tpg->tpg_lun_group, "lun",
+ &TF_CIT_TMPL(tf)->tfc_tpg_lun_cit);
+ config_group_init_type_name(&se_tpg->tpg_np_group, "np",
+ &TF_CIT_TMPL(tf)->tfc_tpg_np_cit);
+ config_group_init_type_name(&se_tpg->tpg_acl_group, "acls",
+ &TF_CIT_TMPL(tf)->tfc_tpg_nacl_cit);
+ config_group_init_type_name(&se_tpg->tpg_attrib_group, "attrib",
+ &TF_CIT_TMPL(tf)->tfc_tpg_attrib_cit);
+ config_group_init_type_name(&se_tpg->tpg_param_group, "param",
+ &TF_CIT_TMPL(tf)->tfc_tpg_param_cit);
+
+ return &se_tpg->tpg_group;
+}
+
+static void target_fabric_drop_tpg(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct se_wwn *wwn = container_of(group, struct se_wwn, wwn_group);
+ struct target_fabric_configfs *tf = wwn->wwn_tf;
+ struct se_portal_group *se_tpg = container_of(to_config_group(item),
+ struct se_portal_group, tpg_group);
+ struct config_group *tpg_cg = &se_tpg->tpg_group;
+ struct config_item *df_item;
+ int i;
+ /*
+ * Release default groups, but do not release tpg_cg->default_groups
+ * memory as it is statically allocated at se_tpg->tpg_default_groups.
+ */
+ for (i = 0; tpg_cg->default_groups[i]; i++) {
+ df_item = &tpg_cg->default_groups[i]->cg_item;
+ tpg_cg->default_groups[i] = NULL;
+ config_item_put(df_item);
+ }
+
+ config_item_put(item);
+ tf->tf_ops.fabric_drop_tpg(se_tpg);
+}
+
+static struct configfs_group_operations target_fabric_tpg_group_ops = {
+ .make_group = target_fabric_make_tpg,
+ .drop_item = target_fabric_drop_tpg,
+};
+
+TF_CIT_SETUP(tpg, NULL, &target_fabric_tpg_group_ops, NULL);
+
+/* End of tfc_tpg_cit */
+
+/* Start of tfc_wwn_cit */
+
+static struct config_group *target_fabric_make_wwn(
+ struct config_group *group,
+ const char *name)
+{
+ struct target_fabric_configfs *tf = container_of(group,
+ struct target_fabric_configfs, tf_group);
+ struct se_wwn *wwn;
+
+ if (!(tf->tf_ops.fabric_make_wwn)) {
+ printk(KERN_ERR "tf->tf_ops.fabric_make_wwn is NULL\n");
+ return ERR_PTR(-ENOSYS);
+ }
+
+ wwn = tf->tf_ops.fabric_make_wwn(tf, group, name);
+ if (!(wwn) || IS_ERR(wwn))
+ return ERR_PTR(-EINVAL);
+
+ wwn->wwn_tf = tf;
+ config_group_init_type_name(&wwn->wwn_group, name,
+ &TF_CIT_TMPL(tf)->tfc_tpg_cit);
+
+ return &wwn->wwn_group;
+}
+
+static void target_fabric_drop_wwn(
+ struct config_group *group,
+ struct config_item *item)
+{
+ struct target_fabric_configfs *tf = container_of(group,
+ struct target_fabric_configfs, tf_group);
+ struct se_wwn *wwn = container_of(to_config_group(item),
+ struct se_wwn, wwn_group);
+
+ config_item_put(item);
+ tf->tf_ops.fabric_drop_wwn(wwn);
+}
+
+static struct configfs_group_operations target_fabric_wwn_group_ops = {
+ .make_group = target_fabric_make_wwn,
+ .drop_item = target_fabric_drop_wwn,
+};
+/*
+ * For use with TF_WWN_ATTR() and TF_WWN_ATTR_RO()
+ */
+CONFIGFS_EATTR_OPS(target_fabric_wwn, target_fabric_configfs, tf_group);
+
+static struct configfs_item_operations target_fabric_wwn_item_ops = {
+ .show_attribute = target_fabric_wwn_attr_show,
+ .store_attribute = target_fabric_wwn_attr_store,
+};
+
+TF_CIT_SETUP(wwn, &target_fabric_wwn_item_ops, &target_fabric_wwn_group_ops, NULL);
+
+/* End of tfc_wwn_cit */
+
+/* Start of tfc_discovery_cit */
+
+CONFIGFS_EATTR_OPS(target_fabric_discovery, target_fabric_configfs,
+ tf_disc_group);
+
+static struct configfs_item_operations target_fabric_discovery_item_ops = {
+ .show_attribute = target_fabric_discovery_attr_show,
+ .store_attribute = target_fabric_discovery_attr_store,
+};
+
+TF_CIT_SETUP(discovery, &target_fabric_discovery_item_ops, NULL, NULL);
+
+/* End of tfc_discovery_cit */
+
+int target_fabric_setup_cits(struct target_fabric_configfs *tf)
+{
+ target_fabric_setup_discovery_cit(tf);
+ target_fabric_setup_wwn_cit(tf);
+ target_fabric_setup_tpg_cit(tf);
+ target_fabric_setup_tpg_base_cit(tf);
+ target_fabric_setup_tpg_port_cit(tf);
+ target_fabric_setup_tpg_lun_cit(tf);
+ target_fabric_setup_tpg_np_cit(tf);
+ target_fabric_setup_tpg_np_base_cit(tf);
+ target_fabric_setup_tpg_attrib_cit(tf);
+ target_fabric_setup_tpg_param_cit(tf);
+ target_fabric_setup_tpg_nacl_cit(tf);
+ target_fabric_setup_tpg_nacl_base_cit(tf);
+ target_fabric_setup_tpg_nacl_attrib_cit(tf);
+ target_fabric_setup_tpg_nacl_auth_cit(tf);
+ target_fabric_setup_tpg_nacl_param_cit(tf);
+ target_fabric_setup_tpg_mappedlun_cit(tf);
+
+ return 0;
+}
diff --git a/drivers/target/target_core_fabric_lib.c b/drivers/target/target_core_fabric_lib.c
new file mode 100644
index 0000000..2628564
--- /dev/null
+++ b/drivers/target/target_core_fabric_lib.c
@@ -0,0 +1,451 @@
+/*******************************************************************************
+ * Filename: target_core_fabric_lib.c
+ *
+ * This file contains generic high level protocol identifier and PR
+ * handlers for TCM fabric modules
+ *
+ * Copyright (c) 2010 Rising Tide Systems, Inc.
+ * Copyright (c) 2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@linux-iscsi.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_configfs.h>
+
+#include "target_core_hba.h"
+#include "target_core_pr.h"
+
+/*
+ * Handlers for Serial Attached SCSI (SAS)
+ */
+u8 sas_get_fabric_proto_ident(struct se_portal_group *se_tpg)
+{
+ /*
+ * Return a SAS Serial SCSI Protocol identifier for loopback operations
+ * This is defined in section 7.5.1 Table 362 in spc4r17
+ */
+ return 0x6;
+}
+EXPORT_SYMBOL(sas_get_fabric_proto_ident);
+
+u32 sas_get_pr_transport_id(
+ struct se_portal_group *se_tpg,
+ struct se_node_acl *se_nacl,
+ struct t10_pr_registration *pr_reg,
+ int *format_code,
+ unsigned char *buf)
+{
+ unsigned char binary, *ptr;
+ int i;
+ u32 off = 4;
+ /*
+ * Set PROTOCOL IDENTIFIER to 6h for SAS
+ */
+ buf[0] = 0x06;
+ /*
+ * From spc4r17, 7.5.4.7 TransportID for initiator ports using SCSI
+ * over SAS Serial SCSI Protocol
+ */
+ ptr = &se_nacl->initiatorname[4]; /* Skip over 'naa. prefix */
+
+ for (i = 0; i < 16; i += 2) {
+ binary = transport_asciihex_to_binaryhex(&ptr[i]);
+ buf[off++] = binary;
+ }
+ /*
+ * The SAS Transport ID is a hardcoded 24-byte length
+ */
+ return 24;
+}
+EXPORT_SYMBOL(sas_get_pr_transport_id);
+
+u32 sas_get_pr_transport_id_len(
+ struct se_portal_group *se_tpg,
+ struct se_node_acl *se_nacl,
+ struct t10_pr_registration *pr_reg,
+ int *format_code)
+{
+ *format_code = 0;
+ /*
+ * From spc4r17, 7.5.4.7 TransportID for initiator ports using SCSI
+ * over SAS Serial SCSI Protocol
+ *
+ * The SAS Transport ID is a hardcoded 24-byte length
+ */
+ return 24;
+}
+EXPORT_SYMBOL(sas_get_pr_transport_id_len);
+
+/*
+ * Used for handling SCSI fabric dependent TransportIDs in SPC-3 and above
+ * Persistent Reservation SPEC_I_PT=1 and PROUT REGISTER_AND_MOVE operations.
+ */
+char *sas_parse_pr_out_transport_id(
+ struct se_portal_group *se_tpg,
+ const char *buf,
+ u32 *out_tid_len,
+ char **port_nexus_ptr)
+{
+ /*
+ * Assume the FORMAT CODE 00b from spc4r17, 7.5.4.7 TransportID
+ * for initiator ports using SCSI over SAS Serial SCSI Protocol
+ *
+ * The TransportID for a SAS Initiator Port is of fixed size of
+ * 24 bytes, and SAS does not contain a I_T nexus identifier,
+ * so we return the **port_nexus_ptr set to NULL.
+ */
+ *port_nexus_ptr = NULL;
+ *out_tid_len = 24;
+
+ return (char *)&buf[4];
+}
+EXPORT_SYMBOL(sas_parse_pr_out_transport_id);
+
+/*
+ * Handlers for Fibre Channel Protocol (FCP)
+ */
+u8 fc_get_fabric_proto_ident(struct se_portal_group *se_tpg)
+{
+ return 0x0; /* 0 = fcp-2 per SPC4 section 7.5.1 */
+}
+EXPORT_SYMBOL(fc_get_fabric_proto_ident);
+
+u32 fc_get_pr_transport_id_len(
+ struct se_portal_group *se_tpg,
+ struct se_node_acl *se_nacl,
+ struct t10_pr_registration *pr_reg,
+ int *format_code)
+{
+ *format_code = 0;
+ /*
+ * The FC Transport ID is a hardcoded 24-byte length
+ */
+ return 24;
+}
+EXPORT_SYMBOL(fc_get_pr_transport_id_len);
+
+u32 fc_get_pr_transport_id(
+ struct se_portal_group *se_tpg,
+ struct se_node_acl *se_nacl,
+ struct t10_pr_registration *pr_reg,
+ int *format_code,
+ unsigned char *buf)
+{
+ unsigned char binary, *ptr;
+ int i;
+ u32 off = 8;
+ /*
+ * PROTOCOL IDENTIFIER is 0h for FCP-2
+ *
+ * From spc4r17, 7.5.4.2 TransportID for initiator ports using
+ * SCSI over Fibre Channel
+ *
+ * We convert the ASCII formatted N Port name into a binary
+ * encoded TransportID.
+ */
+ ptr = &se_nacl->initiatorname[0];
+
+ for (i = 0; i < 24; ) {
+ if (!(strncmp(&ptr[i], ":", 1))) {
+ i++;
+ continue;
+ }
+ binary = transport_asciihex_to_binaryhex(&ptr[i]);
+ buf[off++] = binary;
+ i += 2;
+ }
+ /*
+ * The FC Transport ID is a hardcoded 24-byte length
+ */
+ return 24;
+}
+EXPORT_SYMBOL(fc_get_pr_transport_id);
+
+char *fc_parse_pr_out_transport_id(
+ struct se_portal_group *se_tpg,
+ const char *buf,
+ u32 *out_tid_len,
+ char **port_nexus_ptr)
+{
+ /*
+ * The TransportID for a FC N Port is of fixed size of
+ * 24 bytes, and FC does not contain a I_T nexus identifier,
+ * so we return the **port_nexus_ptr set to NULL.
+ */
+ *port_nexus_ptr = NULL;
+ *out_tid_len = 24;
+
+ return (char *)&buf[8];
+}
+EXPORT_SYMBOL(fc_parse_pr_out_transport_id);
+
+/*
+ * Handlers for Internet Small Computer Systems Interface (iSCSI)
+ */
+
+u8 iscsi_get_fabric_proto_ident(struct se_portal_group *se_tpg)
+{
+ /*
+ * This value is defined for "Internet SCSI (iSCSI)"
+ * in spc4r17 section 7.5.1 Table 362
+ */
+ return 0x5;
+}
+EXPORT_SYMBOL(iscsi_get_fabric_proto_ident);
+
+u32 iscsi_get_pr_transport_id(
+ struct se_portal_group *se_tpg,
+ struct se_node_acl *se_nacl,
+ struct t10_pr_registration *pr_reg,
+ int *format_code,
+ unsigned char *buf)
+{
+ u32 off = 4, padding = 0;
+ u16 len = 0;
+
+ spin_lock_irq(&se_nacl->nacl_sess_lock);
+ /*
+ * Set PROTOCOL IDENTIFIER to 5h for iSCSI
+ */
+ buf[0] = 0x05;
+ /*
+ * From spc4r17 Section 7.5.4.6: TransportID for initiator
+ * ports using SCSI over iSCSI.
+ *
+ * The null-terminated, null-padded (see 4.4.2) ISCSI NAME field
+ * shall contain the iSCSI name of an iSCSI initiator node (see
+ * RFC 3720). The first ISCSI NAME field byte containing an ASCII
+ * null character terminates the ISCSI NAME field without regard for
+ * the specified length of the iSCSI TransportID or the contents of
+ * the ADDITIONAL LENGTH field.
+ */
+ len = sprintf(&buf[off], "%s", se_nacl->initiatorname);
+ /*
+ * Add Extra byte for NULL terminator
+ */
+ len++;
+ /*
+ * If there is ISID present with the registration and *format code == 1
+ * 1, use iSCSI Initiator port TransportID format.
+ *
+ * Otherwise use iSCSI Initiator device TransportID format that
+ * does not contain the ASCII encoded iSCSI Initiator iSID value
+ * provied by the iSCSi Initiator during the iSCSI login process.
+ */
+ if ((*format_code == 1) && (pr_reg->isid_present_at_reg)) {
+ /*
+ * Set FORMAT CODE 01b for iSCSI Initiator port TransportID
+ * format.
+ */
+ buf[0] |= 0x40;
+ /*
+ * From spc4r17 Section 7.5.4.6: TransportID for initiator
+ * ports using SCSI over iSCSI. Table 390
+ *
+ * The SEPARATOR field shall contain the five ASCII
+ * characters ",i,0x".
+ *
+ * The null-terminated, null-padded ISCSI INITIATOR SESSION ID
+ * field shall contain the iSCSI initiator session identifier
+ * (see RFC 3720) in the form of ASCII characters that are the
+ * hexadecimal digits converted from the binary iSCSI initiator
+ * session identifier value. The first ISCSI INITIATOR SESSION
+ * ID field byte containing an ASCII null character
+ */
+ buf[off+len] = 0x2c; off++; /* ASCII Character: "," */
+ buf[off+len] = 0x69; off++; /* ASCII Character: "i" */
+ buf[off+len] = 0x2c; off++; /* ASCII Character: "," */
+ buf[off+len] = 0x30; off++; /* ASCII Character: "0" */
+ buf[off+len] = 0x78; off++; /* ASCII Character: "x" */
+ len += 5;
+ buf[off+len] = pr_reg->pr_reg_isid[0]; off++;
+ buf[off+len] = pr_reg->pr_reg_isid[1]; off++;
+ buf[off+len] = pr_reg->pr_reg_isid[2]; off++;
+ buf[off+len] = pr_reg->pr_reg_isid[3]; off++;
+ buf[off+len] = pr_reg->pr_reg_isid[4]; off++;
+ buf[off+len] = pr_reg->pr_reg_isid[5]; off++;
+ buf[off+len] = '\0'; off++;
+ len += 7;
+ }
+ spin_unlock_irq(&se_nacl->nacl_sess_lock);
+ /*
+ * The ADDITIONAL LENGTH field specifies the number of bytes that follow
+ * in the TransportID. The additional length shall be at least 20 and
+ * shall be a multiple of four.
+ */
+ padding = ((-len) & 3);
+ if (padding != 0)
+ len += padding;
+
+ buf[2] = ((len >> 8) & 0xff);
+ buf[3] = (len & 0xff);
+ /*
+ * Increment value for total payload + header length for
+ * full status descriptor
+ */
+ len += 4;
+
+ return len;
+}
+EXPORT_SYMBOL(iscsi_get_pr_transport_id);
+
+u32 iscsi_get_pr_transport_id_len(
+ struct se_portal_group *se_tpg,
+ struct se_node_acl *se_nacl,
+ struct t10_pr_registration *pr_reg,
+ int *format_code)
+{
+ u32 len = 0, padding = 0;
+
+ spin_lock_irq(&se_nacl->nacl_sess_lock);
+ len = strlen(se_nacl->initiatorname);
+ /*
+ * Add extra byte for NULL terminator
+ */
+ len++;
+ /*
+ * If there is ISID present with the registration, use format code:
+ * 01b: iSCSI Initiator port TransportID format
+ *
+ * If there is not an active iSCSI session, use format code:
+ * 00b: iSCSI Initiator device TransportID format
+ */
+ if (pr_reg->isid_present_at_reg) {
+ len += 5; /* For ",i,0x" ASCII seperator */
+ len += 7; /* For iSCSI Initiator Session ID + Null terminator */
+ *format_code = 1;
+ } else
+ *format_code = 0;
+ spin_unlock_irq(&se_nacl->nacl_sess_lock);
+ /*
+ * The ADDITIONAL LENGTH field specifies the number of bytes that follow
+ * in the TransportID. The additional length shall be at least 20 and
+ * shall be a multiple of four.
+ */
+ padding = ((-len) & 3);
+ if (padding != 0)
+ len += padding;
+ /*
+ * Increment value for total payload + header length for
+ * full status descriptor
+ */
+ len += 4;
+
+ return len;
+}
+EXPORT_SYMBOL(iscsi_get_pr_transport_id_len);
+
+char *iscsi_parse_pr_out_transport_id(
+ struct se_portal_group *se_tpg,
+ const char *buf,
+ u32 *out_tid_len,
+ char **port_nexus_ptr)
+{
+ char *p;
+ u32 tid_len, padding;
+ int i;
+ u16 add_len;
+ u8 format_code = (buf[0] & 0xc0);
+ /*
+ * Check for FORMAT CODE 00b or 01b from spc4r17, section 7.5.4.6:
+ *
+ * TransportID for initiator ports using SCSI over iSCSI,
+ * from Table 388 -- iSCSI TransportID formats.
+ *
+ * 00b Initiator port is identified using the world wide unique
+ * SCSI device name of the iSCSI initiator
+ * device containing the initiator port (see table 389).
+ * 01b Initiator port is identified using the world wide unique
+ * initiator port identifier (see table 390).10b to 11b
+ * Reserved
+ */
+ if ((format_code != 0x00) && (format_code != 0x40)) {
+ printk(KERN_ERR "Illegal format code: 0x%02x for iSCSI"
+ " Initiator Transport ID\n", format_code);
+ return NULL;
+ }
+ /*
+ * If the caller wants the TransportID Length, we set that value for the
+ * entire iSCSI Tarnsport ID now.
+ */
+ if (out_tid_len != NULL) {
+ add_len = ((buf[2] >> 8) & 0xff);
+ add_len |= (buf[3] & 0xff);
+
+ tid_len = strlen((char *)&buf[4]);
+ tid_len += 4; /* Add four bytes for iSCSI Transport ID header */
+ tid_len += 1; /* Add one byte for NULL terminator */
+ padding = ((-tid_len) & 3);
+ if (padding != 0)
+ tid_len += padding;
+
+ if ((add_len + 4) != tid_len) {
+ printk(KERN_INFO "LIO-Target Extracted add_len: %hu "
+ "does not match calculated tid_len: %u,"
+ " using tid_len instead\n", add_len+4, tid_len);
+ *out_tid_len = tid_len;
+ } else
+ *out_tid_len = (add_len + 4);
+ }
+ /*
+ * Check for ',i,0x' seperator between iSCSI Name and iSCSI Initiator
+ * Session ID as defined in Table 390 - iSCSI initiator port TransportID
+ * format.
+ */
+ if (format_code == 0x40) {
+ p = strstr((char *)&buf[4], ",i,0x");
+ if (!(p)) {
+ printk(KERN_ERR "Unable to locate \",i,0x\" seperator"
+ " for Initiator port identifier: %s\n",
+ (char *)&buf[4]);
+ return NULL;
+ }
+ *p = '\0'; /* Terminate iSCSI Name */
+ p += 5; /* Skip over ",i,0x" seperator */
+
+ *port_nexus_ptr = p;
+ /*
+ * Go ahead and do the lower case conversion of the received
+ * 12 ASCII characters representing the ISID in the TransportID
+ * for comparision against the running iSCSI session's ISID from
+ * iscsi_target.c:lio_sess_get_initiator_sid()
+ */
+ for (i = 0; i < 12; i++) {
+ if (isdigit(*p)) {
+ p++;
+ continue;
+ }
+ *p = tolower(*p);
+ p++;
+ }
+ }
+
+ return (char *)&buf[4];
+}
+EXPORT_SYMBOL(iscsi_parse_pr_out_transport_id);
diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c
new file mode 100644
index 0000000..0aaca88
--- /dev/null
+++ b/drivers/target/target_core_file.c
@@ -0,0 +1,688 @@
+/*******************************************************************************
+ * Filename: target_core_file.c
+ *
+ * This file contains the Storage Engine <-> FILEIO transport specific functions
+ *
+ * Copyright (c) 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005-2006 SBE, Inc. All Rights Reserved.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/string.h>
+#include <linux/parser.h>
+#include <linux/timer.h>
+#include <linux/blkdev.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_host.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_transport.h>
+
+#include "target_core_file.h"
+
+#if 1
+#define DEBUG_FD_CACHE(x...) printk(x)
+#else
+#define DEBUG_FD_CACHE(x...)
+#endif
+
+#if 1
+#define DEBUG_FD_FUA(x...) printk(x)
+#else
+#define DEBUG_FD_FUA(x...)
+#endif
+
+static struct se_subsystem_api fileio_template;
+
+/* fd_attach_hba(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static int fd_attach_hba(struct se_hba *hba, u32 host_id)
+{
+ struct fd_host *fd_host;
+
+ fd_host = kzalloc(sizeof(struct fd_host), GFP_KERNEL);
+ if (!(fd_host)) {
+ printk(KERN_ERR "Unable to allocate memory for struct fd_host\n");
+ return -1;
+ }
+
+ fd_host->fd_host_id = host_id;
+
+ atomic_set(&hba->left_queue_depth, FD_HBA_QUEUE_DEPTH);
+ atomic_set(&hba->max_queue_depth, FD_HBA_QUEUE_DEPTH);
+ hba->hba_ptr = (void *) fd_host;
+
+ printk(KERN_INFO "CORE_HBA[%d] - TCM FILEIO HBA Driver %s on Generic"
+ " Target Core Stack %s\n", hba->hba_id, FD_VERSION,
+ TARGET_CORE_MOD_VERSION);
+ printk(KERN_INFO "CORE_HBA[%d] - Attached FILEIO HBA: %u to Generic"
+ " Target Core with TCQ Depth: %d MaxSectors: %u\n",
+ hba->hba_id, fd_host->fd_host_id,
+ atomic_read(&hba->max_queue_depth), FD_MAX_SECTORS);
+
+ return 0;
+}
+
+static void fd_detach_hba(struct se_hba *hba)
+{
+ struct fd_host *fd_host = hba->hba_ptr;
+
+ printk(KERN_INFO "CORE_HBA[%d] - Detached FILEIO HBA: %u from Generic"
+ " Target Core\n", hba->hba_id, fd_host->fd_host_id);
+
+ kfree(fd_host);
+ hba->hba_ptr = NULL;
+}
+
+static void *fd_allocate_virtdevice(struct se_hba *hba, const char *name)
+{
+ struct fd_dev *fd_dev;
+ struct fd_host *fd_host = (struct fd_host *) hba->hba_ptr;
+
+ fd_dev = kzalloc(sizeof(struct fd_dev), GFP_KERNEL);
+ if (!(fd_dev)) {
+ printk(KERN_ERR "Unable to allocate memory for struct fd_dev\n");
+ return NULL;
+ }
+
+ fd_dev->fd_host = fd_host;
+
+ printk(KERN_INFO "FILEIO: Allocated fd_dev for %p\n", name);
+
+ return fd_dev;
+}
+
+/* fd_create_virtdevice(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static struct se_device *fd_create_virtdevice(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ void *p)
+{
+ char *dev_p = NULL;
+ struct se_device *dev;
+ struct se_dev_limits dev_limits;
+ struct queue_limits *limits;
+ struct fd_dev *fd_dev = (struct fd_dev *) p;
+ struct fd_host *fd_host = (struct fd_host *) hba->hba_ptr;
+ mm_segment_t old_fs;
+ struct file *file;
+ struct inode *inode = NULL;
+ int dev_flags = 0, flags;
+
+ memset(&dev_limits, 0, sizeof(struct se_dev_limits));
+
+ old_fs = get_fs();
+ set_fs(get_ds());
+ dev_p = getname(fd_dev->fd_dev_name);
+ set_fs(old_fs);
+
+ if (IS_ERR(dev_p)) {
+ printk(KERN_ERR "getname(%s) failed: %lu\n",
+ fd_dev->fd_dev_name, IS_ERR(dev_p));
+ goto fail;
+ }
+#if 0
+ if (di->no_create_file)
+ flags = O_RDWR | O_LARGEFILE;
+ else
+ flags = O_RDWR | O_CREAT | O_LARGEFILE;
+#else
+ flags = O_RDWR | O_CREAT | O_LARGEFILE;
+#endif
+/* flags |= O_DIRECT; */
+ /*
+ * If fd_buffered_io=1 has not been set explictly (the default),
+ * use O_SYNC to force FILEIO writes to disk.
+ */
+ if (!(fd_dev->fbd_flags & FDBD_USE_BUFFERED_IO))
+ flags |= O_SYNC;
+
+ file = filp_open(dev_p, flags, 0600);
+
+ if (IS_ERR(file) || !file || !file->f_dentry) {
+ printk(KERN_ERR "filp_open(%s) failed\n", dev_p);
+ goto fail;
+ }
+ fd_dev->fd_file = file;
+ /*
+ * If using a block backend with this struct file, we extract
+ * fd_dev->fd_[block,dev]_size from struct block_device.
+ *
+ * Otherwise, we use the passed fd_size= from configfs
+ */
+ inode = file->f_mapping->host;
+ if (S_ISBLK(inode->i_mode)) {
+ struct request_queue *q;
+ /*
+ * Setup the local scope queue_limits from struct request_queue->limits
+ * to pass into transport_add_device_to_core_hba() as struct se_dev_limits.
+ */
+ q = bdev_get_queue(inode->i_bdev);
+ limits = &dev_limits.limits;
+ limits->logical_block_size = bdev_logical_block_size(inode->i_bdev);
+ limits->max_hw_sectors = queue_max_hw_sectors(q);
+ limits->max_sectors = queue_max_sectors(q);
+ /*
+ * Determine the number of bytes from i_size_read() minus
+ * one (1) logical sector from underlying struct block_device
+ */
+ fd_dev->fd_block_size = bdev_logical_block_size(inode->i_bdev);
+ fd_dev->fd_dev_size = (i_size_read(file->f_mapping->host) -
+ fd_dev->fd_block_size);
+
+ printk(KERN_INFO "FILEIO: Using size: %llu bytes from struct"
+ " block_device blocks: %llu logical_block_size: %d\n",
+ fd_dev->fd_dev_size,
+ div_u64(fd_dev->fd_dev_size, fd_dev->fd_block_size),
+ fd_dev->fd_block_size);
+ } else {
+ if (!(fd_dev->fbd_flags & FBDF_HAS_SIZE)) {
+ printk(KERN_ERR "FILEIO: Missing fd_dev_size="
+ " parameter, and no backing struct"
+ " block_device\n");
+ goto fail;
+ }
+
+ limits = &dev_limits.limits;
+ limits->logical_block_size = FD_BLOCKSIZE;
+ limits->max_hw_sectors = FD_MAX_SECTORS;
+ limits->max_sectors = FD_MAX_SECTORS;
+ fd_dev->fd_block_size = FD_BLOCKSIZE;
+ }
+
+ dev_limits.hw_queue_depth = FD_MAX_DEVICE_QUEUE_DEPTH;
+ dev_limits.queue_depth = FD_DEVICE_QUEUE_DEPTH;
+
+ dev = transport_add_device_to_core_hba(hba, &fileio_template,
+ se_dev, dev_flags, (void *)fd_dev,
+ &dev_limits, "FILEIO", FD_VERSION);
+ if (!(dev))
+ goto fail;
+
+ fd_dev->fd_dev_id = fd_host->fd_host_dev_id_count++;
+ fd_dev->fd_queue_depth = dev->queue_depth;
+
+ printk(KERN_INFO "CORE_FILE[%u] - Added TCM FILEIO Device ID: %u at %s,"
+ " %llu total bytes\n", fd_host->fd_host_id, fd_dev->fd_dev_id,
+ fd_dev->fd_dev_name, fd_dev->fd_dev_size);
+
+ putname(dev_p);
+ return dev;
+fail:
+ if (fd_dev->fd_file) {
+ filp_close(fd_dev->fd_file, NULL);
+ fd_dev->fd_file = NULL;
+ }
+ putname(dev_p);
+ return NULL;
+}
+
+/* fd_free_device(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static void fd_free_device(void *p)
+{
+ struct fd_dev *fd_dev = (struct fd_dev *) p;
+
+ if (fd_dev->fd_file) {
+ filp_close(fd_dev->fd_file, NULL);
+ fd_dev->fd_file = NULL;
+ }
+
+ kfree(fd_dev);
+}
+
+static inline struct fd_request *FILE_REQ(struct se_task *task)
+{
+ return container_of(task, struct fd_request, fd_task);
+}
+
+
+static struct se_task *
+fd_alloc_task(struct se_cmd *cmd)
+{
+ struct fd_request *fd_req;
+
+ fd_req = kzalloc(sizeof(struct fd_request), GFP_KERNEL);
+ if (!(fd_req)) {
+ printk(KERN_ERR "Unable to allocate struct fd_request\n");
+ return NULL;
+ }
+
+ fd_req->fd_dev = SE_DEV(cmd)->dev_ptr;
+
+ return &fd_req->fd_task;
+}
+
+static int fd_do_readv(struct se_task *task)
+{
+ struct fd_request *req = FILE_REQ(task);
+ struct file *fd = req->fd_dev->fd_file;
+ struct scatterlist *sg = task->task_sg;
+ struct iovec *iov;
+ mm_segment_t old_fs;
+ loff_t pos = (task->task_lba * DEV_ATTRIB(task->se_dev)->block_size);
+ int ret = 0, i;
+
+ iov = kzalloc(sizeof(struct iovec) * task->task_sg_num, GFP_KERNEL);
+ if (!(iov)) {
+ printk(KERN_ERR "Unable to allocate fd_do_readv iov[]\n");
+ return -1;
+ }
+
+ for (i = 0; i < task->task_sg_num; i++) {
+ iov[i].iov_len = sg[i].length;
+ iov[i].iov_base = sg_virt(&sg[i]);
+ }
+
+ old_fs = get_fs();
+ set_fs(get_ds());
+ ret = vfs_readv(fd, &iov[0], task->task_sg_num, &pos);
+ set_fs(old_fs);
+
+ kfree(iov);
+ /*
+ * Return zeros and GOOD status even if the READ did not return
+ * the expected virt_size for struct file w/o a backing struct
+ * block_device.
+ */
+ if (S_ISBLK(fd->f_dentry->d_inode->i_mode)) {
+ if (ret < 0 || ret != task->task_size) {
+ printk(KERN_ERR "vfs_readv() returned %d,"
+ " expecting %d for S_ISBLK\n", ret,
+ (int)task->task_size);
+ return -1;
+ }
+ } else {
+ if (ret < 0) {
+ printk(KERN_ERR "vfs_readv() returned %d for non"
+ " S_ISBLK\n", ret);
+ return -1;
+ }
+ }
+
+ return 1;
+}
+
+static int fd_do_writev(struct se_task *task)
+{
+ struct fd_request *req = FILE_REQ(task);
+ struct file *fd = req->fd_dev->fd_file;
+ struct scatterlist *sg = task->task_sg;
+ struct iovec *iov;
+ mm_segment_t old_fs;
+ loff_t pos = (task->task_lba * DEV_ATTRIB(task->se_dev)->block_size);
+ int ret, i = 0;
+
+ iov = kzalloc(sizeof(struct iovec) * task->task_sg_num, GFP_KERNEL);
+ if (!(iov)) {
+ printk(KERN_ERR "Unable to allocate fd_do_writev iov[]\n");
+ return -1;
+ }
+
+ for (i = 0; i < task->task_sg_num; i++) {
+ iov[i].iov_len = sg[i].length;
+ iov[i].iov_base = sg_virt(&sg[i]);
+ }
+
+ old_fs = get_fs();
+ set_fs(get_ds());
+ ret = vfs_writev(fd, &iov[0], task->task_sg_num, &pos);
+ set_fs(old_fs);
+
+ kfree(iov);
+
+ if (ret < 0 || ret != task->task_size) {
+ printk(KERN_ERR "vfs_writev() returned %d\n", ret);
+ return -1;
+ }
+
+ return 1;
+}
+
+static void fd_emulate_sync_cache(struct se_task *task)
+{
+ struct se_cmd *cmd = TASK_CMD(task);
+ struct se_device *dev = cmd->se_dev;
+ struct fd_dev *fd_dev = dev->dev_ptr;
+ int immed = (cmd->t_task->t_task_cdb[1] & 0x2);
+ loff_t start, end;
+ int ret;
+
+ /*
+ * If the Immediate bit is set, queue up the GOOD response
+ * for this SYNCHRONIZE_CACHE op
+ */
+ if (immed)
+ transport_complete_sync_cache(cmd, 1);
+
+ /*
+ * Determine if we will be flushing the entire device.
+ */
+ if (cmd->t_task->t_task_lba == 0 && cmd->data_length == 0) {
+ start = 0;
+ end = LLONG_MAX;
+ } else {
+ start = cmd->t_task->t_task_lba * DEV_ATTRIB(dev)->block_size;
+ if (cmd->data_length)
+ end = start + cmd->data_length;
+ else
+ end = LLONG_MAX;
+ }
+
+ ret = vfs_fsync_range(fd_dev->fd_file, start, end, 1);
+ if (ret != 0)
+ printk(KERN_ERR "FILEIO: vfs_fsync_range() failed: %d\n", ret);
+
+ if (!immed)
+ transport_complete_sync_cache(cmd, ret == 0);
+}
+
+/*
+ * Tell TCM Core that we are capable of WriteCache emulation for
+ * an underlying struct se_device.
+ */
+static int fd_emulated_write_cache(struct se_device *dev)
+{
+ return 1;
+}
+
+static int fd_emulated_dpo(struct se_device *dev)
+{
+ return 0;
+}
+/*
+ * Tell TCM Core that we will be emulating Forced Unit Access (FUA) for WRITEs
+ * for TYPE_DISK.
+ */
+static int fd_emulated_fua_write(struct se_device *dev)
+{
+ return 1;
+}
+
+static int fd_emulated_fua_read(struct se_device *dev)
+{
+ return 0;
+}
+
+/*
+ * WRITE Force Unit Access (FUA) emulation on a per struct se_task
+ * LBA range basis..
+ */
+static void fd_emulate_write_fua(struct se_cmd *cmd, struct se_task *task)
+{
+ struct se_device *dev = cmd->se_dev;
+ struct fd_dev *fd_dev = dev->dev_ptr;
+ loff_t start = task->task_lba * DEV_ATTRIB(dev)->block_size;
+ loff_t end = start + task->task_size;
+ int ret;
+
+ DEBUG_FD_CACHE("FILEIO: FUA WRITE LBA: %llu, bytes: %u\n",
+ task->task_lba, task->task_size);
+
+ ret = vfs_fsync_range(fd_dev->fd_file, start, end, 1);
+ if (ret != 0)
+ printk(KERN_ERR "FILEIO: vfs_fsync_range() failed: %d\n", ret);
+}
+
+static int fd_do_task(struct se_task *task)
+{
+ struct se_cmd *cmd = task->task_se_cmd;
+ struct se_device *dev = cmd->se_dev;
+ int ret = 0;
+
+ /*
+ * Call vectorized fileio functions to map struct scatterlist
+ * physical memory addresses to struct iovec virtual memory.
+ */
+ if (task->task_data_direction == DMA_FROM_DEVICE) {
+ ret = fd_do_readv(task);
+ } else {
+ ret = fd_do_writev(task);
+
+ if (ret > 0 &&
+ DEV_ATTRIB(dev)->emulate_write_cache > 0 &&
+ DEV_ATTRIB(dev)->emulate_fua_write > 0 &&
+ T_TASK(cmd)->t_tasks_fua) {
+ /*
+ * We might need to be a bit smarter here
+ * and return some sense data to let the initiator
+ * know the FUA WRITE cache sync failed..?
+ */
+ fd_emulate_write_fua(cmd, task);
+ }
+
+ }
+
+ if (ret < 0)
+ return ret;
+ if (ret) {
+ task->task_scsi_status = GOOD;
+ transport_complete_task(task, 1);
+ }
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+}
+
+/* fd_free_task(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static void fd_free_task(struct se_task *task)
+{
+ struct fd_request *req = FILE_REQ(task);
+
+ kfree(req);
+}
+
+enum {
+ Opt_fd_dev_name, Opt_fd_dev_size, Opt_fd_buffered_io, Opt_err
+};
+
+static match_table_t tokens = {
+ {Opt_fd_dev_name, "fd_dev_name=%s"},
+ {Opt_fd_dev_size, "fd_dev_size=%s"},
+ {Opt_fd_buffered_io, "fd_buffered_id=%d"},
+ {Opt_err, NULL}
+};
+
+static ssize_t fd_set_configfs_dev_params(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ const char *page, ssize_t count)
+{
+ struct fd_dev *fd_dev = se_dev->se_dev_su_ptr;
+ char *orig, *ptr, *arg_p, *opts;
+ substring_t args[MAX_OPT_ARGS];
+ int ret = 0, arg, token;
+
+ opts = kstrdup(page, GFP_KERNEL);
+ if (!opts)
+ return -ENOMEM;
+
+ orig = opts;
+
+ while ((ptr = strsep(&opts, ",")) != NULL) {
+ if (!*ptr)
+ continue;
+
+ token = match_token(ptr, tokens, args);
+ switch (token) {
+ case Opt_fd_dev_name:
+ snprintf(fd_dev->fd_dev_name, FD_MAX_DEV_NAME,
+ "%s", match_strdup(&args[0]));
+ printk(KERN_INFO "FILEIO: Referencing Path: %s\n",
+ fd_dev->fd_dev_name);
+ fd_dev->fbd_flags |= FBDF_HAS_PATH;
+ break;
+ case Opt_fd_dev_size:
+ arg_p = match_strdup(&args[0]);
+ ret = strict_strtoull(arg_p, 0, &fd_dev->fd_dev_size);
+ if (ret < 0) {
+ printk(KERN_ERR "strict_strtoull() failed for"
+ " fd_dev_size=\n");
+ goto out;
+ }
+ printk(KERN_INFO "FILEIO: Referencing Size: %llu"
+ " bytes\n", fd_dev->fd_dev_size);
+ fd_dev->fbd_flags |= FBDF_HAS_SIZE;
+ break;
+ case Opt_fd_buffered_io:
+ match_int(args, &arg);
+ if (arg != 1) {
+ printk(KERN_ERR "bogus fd_buffered_io=%d value\n", arg);
+ ret = -EINVAL;
+ goto out;
+ }
+
+ printk(KERN_INFO "FILEIO: Using buffered I/O"
+ " operations for struct fd_dev\n");
+
+ fd_dev->fbd_flags |= FDBD_USE_BUFFERED_IO;
+ break;
+ default:
+ break;
+ }
+ }
+
+out:
+ kfree(orig);
+ return (!ret) ? count : ret;
+}
+
+static ssize_t fd_check_configfs_dev_params(struct se_hba *hba, struct se_subsystem_dev *se_dev)
+{
+ struct fd_dev *fd_dev = (struct fd_dev *) se_dev->se_dev_su_ptr;
+
+ if (!(fd_dev->fbd_flags & FBDF_HAS_PATH)) {
+ printk(KERN_ERR "Missing fd_dev_name=\n");
+ return -1;
+ }
+
+ return 0;
+}
+
+static ssize_t fd_show_configfs_dev_params(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ char *b)
+{
+ struct fd_dev *fd_dev = se_dev->se_dev_su_ptr;
+ ssize_t bl = 0;
+
+ bl = sprintf(b + bl, "TCM FILEIO ID: %u", fd_dev->fd_dev_id);
+ bl += sprintf(b + bl, " File: %s Size: %llu Mode: %s\n",
+ fd_dev->fd_dev_name, fd_dev->fd_dev_size,
+ (fd_dev->fbd_flags & FDBD_USE_BUFFERED_IO) ?
+ "Buffered" : "Synchronous");
+ return bl;
+}
+
+/* fd_get_cdb(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static unsigned char *fd_get_cdb(struct se_task *task)
+{
+ struct fd_request *req = FILE_REQ(task);
+
+ return req->fd_scsi_cdb;
+}
+
+/* fd_get_device_rev(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static u32 fd_get_device_rev(struct se_device *dev)
+{
+ return SCSI_SPC_2; /* Returns SPC-3 in Initiator Data */
+}
+
+/* fd_get_device_type(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static u32 fd_get_device_type(struct se_device *dev)
+{
+ return TYPE_DISK;
+}
+
+static sector_t fd_get_blocks(struct se_device *dev)
+{
+ struct fd_dev *fd_dev = dev->dev_ptr;
+ unsigned long long blocks_long = div_u64(fd_dev->fd_dev_size,
+ DEV_ATTRIB(dev)->block_size);
+
+ return blocks_long;
+}
+
+static struct se_subsystem_api fileio_template = {
+ .name = "fileio",
+ .owner = THIS_MODULE,
+ .transport_type = TRANSPORT_PLUGIN_VHBA_PDEV,
+ .attach_hba = fd_attach_hba,
+ .detach_hba = fd_detach_hba,
+ .allocate_virtdevice = fd_allocate_virtdevice,
+ .create_virtdevice = fd_create_virtdevice,
+ .free_device = fd_free_device,
+ .dpo_emulated = fd_emulated_dpo,
+ .fua_write_emulated = fd_emulated_fua_write,
+ .fua_read_emulated = fd_emulated_fua_read,
+ .write_cache_emulated = fd_emulated_write_cache,
+ .alloc_task = fd_alloc_task,
+ .do_task = fd_do_task,
+ .do_sync_cache = fd_emulate_sync_cache,
+ .free_task = fd_free_task,
+ .check_configfs_dev_params = fd_check_configfs_dev_params,
+ .set_configfs_dev_params = fd_set_configfs_dev_params,
+ .show_configfs_dev_params = fd_show_configfs_dev_params,
+ .get_cdb = fd_get_cdb,
+ .get_device_rev = fd_get_device_rev,
+ .get_device_type = fd_get_device_type,
+ .get_blocks = fd_get_blocks,
+};
+
+static int __init fileio_module_init(void)
+{
+ return transport_subsystem_register(&fileio_template);
+}
+
+static void fileio_module_exit(void)
+{
+ transport_subsystem_release(&fileio_template);
+}
+
+MODULE_DESCRIPTION("TCM FILEIO subsystem plugin");
+MODULE_AUTHOR("nab@Linux-iSCSI.org");
+MODULE_LICENSE("GPL");
+
+module_init(fileio_module_init);
+module_exit(fileio_module_exit);
diff --git a/drivers/target/target_core_file.h b/drivers/target/target_core_file.h
new file mode 100644
index 0000000..ef4de2b
--- /dev/null
+++ b/drivers/target/target_core_file.h
@@ -0,0 +1,50 @@
+#ifndef TARGET_CORE_FILE_H
+#define TARGET_CORE_FILE_H
+
+#define FD_VERSION "4.0"
+
+#define FD_MAX_DEV_NAME 256
+/* Maximum queuedepth for the FILEIO HBA */
+#define FD_HBA_QUEUE_DEPTH 256
+#define FD_DEVICE_QUEUE_DEPTH 32
+#define FD_MAX_DEVICE_QUEUE_DEPTH 128
+#define FD_BLOCKSIZE 512
+#define FD_MAX_SECTORS 1024
+
+#define RRF_EMULATE_CDB 0x01
+#define RRF_GOT_LBA 0x02
+
+struct fd_request {
+ struct se_task fd_task;
+ /* SCSI CDB from iSCSI Command PDU */
+ unsigned char fd_scsi_cdb[TCM_MAX_COMMAND_SIZE];
+ /* FILEIO device */
+ struct fd_dev *fd_dev;
+} ____cacheline_aligned;
+
+#define FBDF_HAS_PATH 0x01
+#define FBDF_HAS_SIZE 0x02
+#define FDBD_USE_BUFFERED_IO 0x04
+
+struct fd_dev {
+ u32 fbd_flags;
+ unsigned char fd_dev_name[FD_MAX_DEV_NAME];
+ /* Unique Ramdisk Device ID in Ramdisk HBA */
+ u32 fd_dev_id;
+ /* Number of SG tables in sg_table_array */
+ u32 fd_table_count;
+ u32 fd_queue_depth;
+ u32 fd_block_size;
+ unsigned long long fd_dev_size;
+ struct file *fd_file;
+ /* FILEIO HBA device is connected to */
+ struct fd_host *fd_host;
+} ____cacheline_aligned;
+
+struct fd_host {
+ u32 fd_host_dev_id_count;
+ /* Unique FILEIO Host ID */
+ u32 fd_host_id;
+} ____cacheline_aligned;
+
+#endif /* TARGET_CORE_FILE_H */
diff --git a/drivers/target/target_core_hba.c b/drivers/target/target_core_hba.c
new file mode 100644
index 0000000..4bbe820
--- /dev/null
+++ b/drivers/target/target_core_hba.c
@@ -0,0 +1,185 @@
+/*******************************************************************************
+ * Filename: target_core_hba.c
+ *
+ * This file copntains the iSCSI HBA Transport related functions.
+ *
+ * Copyright (c) 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/net.h>
+#include <linux/string.h>
+#include <linux/timer.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <linux/in.h>
+#include <net/sock.h>
+#include <net/tcp.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_device.h>
+#include <target/target_core_tpg.h>
+#include <target/target_core_transport.h>
+
+#include "target_core_hba.h"
+
+static LIST_HEAD(subsystem_list);
+static DEFINE_MUTEX(subsystem_mutex);
+
+int transport_subsystem_register(struct se_subsystem_api *sub_api)
+{
+ struct se_subsystem_api *s;
+
+ INIT_LIST_HEAD(&sub_api->sub_api_list);
+
+ mutex_lock(&subsystem_mutex);
+ list_for_each_entry(s, &subsystem_list, sub_api_list) {
+ if (!(strcmp(s->name, sub_api->name))) {
+ printk(KERN_ERR "%p is already registered with"
+ " duplicate name %s, unable to process"
+ " request\n", s, s->name);
+ mutex_unlock(&subsystem_mutex);
+ return -EEXIST;
+ }
+ }
+ list_add_tail(&sub_api->sub_api_list, &subsystem_list);
+ mutex_unlock(&subsystem_mutex);
+
+ printk(KERN_INFO "TCM: Registered subsystem plugin: %s struct module:"
+ " %p\n", sub_api->name, sub_api->owner);
+ return 0;
+}
+EXPORT_SYMBOL(transport_subsystem_register);
+
+void transport_subsystem_release(struct se_subsystem_api *sub_api)
+{
+ mutex_lock(&subsystem_mutex);
+ list_del(&sub_api->sub_api_list);
+ mutex_unlock(&subsystem_mutex);
+}
+EXPORT_SYMBOL(transport_subsystem_release);
+
+static struct se_subsystem_api *core_get_backend(const char *sub_name)
+{
+ struct se_subsystem_api *s;
+
+ mutex_lock(&subsystem_mutex);
+ list_for_each_entry(s, &subsystem_list, sub_api_list) {
+ if (!strcmp(s->name, sub_name))
+ goto found;
+ }
+ mutex_unlock(&subsystem_mutex);
+ return NULL;
+found:
+ if (s->owner && !try_module_get(s->owner))
+ s = NULL;
+ mutex_unlock(&subsystem_mutex);
+ return s;
+}
+
+struct se_hba *
+core_alloc_hba(const char *plugin_name, u32 plugin_dep_id, u32 hba_flags)
+{
+ struct se_hba *hba;
+ int ret = 0;
+
+ hba = kzalloc(sizeof(*hba), GFP_KERNEL);
+ if (!hba) {
+ printk(KERN_ERR "Unable to allocate struct se_hba\n");
+ return ERR_PTR(-ENOMEM);
+ }
+
+ INIT_LIST_HEAD(&hba->hba_dev_list);
+ spin_lock_init(&hba->device_lock);
+ spin_lock_init(&hba->hba_queue_lock);
+ mutex_init(&hba->hba_access_mutex);
+
+ hba->hba_index = scsi_get_new_index(SCSI_INST_INDEX);
+ hba->hba_flags |= hba_flags;
+
+ atomic_set(&hba->max_queue_depth, 0);
+ atomic_set(&hba->left_queue_depth, 0);
+
+ hba->transport = core_get_backend(plugin_name);
+ if (!hba->transport) {
+ ret = -EINVAL;
+ goto out_free_hba;
+ }
+
+ ret = hba->transport->attach_hba(hba, plugin_dep_id);
+ if (ret < 0)
+ goto out_module_put;
+
+ spin_lock(&se_global->hba_lock);
+ hba->hba_id = se_global->g_hba_id_counter++;
+ list_add_tail(&hba->hba_list, &se_global->g_hba_list);
+ spin_unlock(&se_global->hba_lock);
+
+ printk(KERN_INFO "CORE_HBA[%d] - Attached HBA to Generic Target"
+ " Core\n", hba->hba_id);
+
+ return hba;
+
+out_module_put:
+ if (hba->transport->owner)
+ module_put(hba->transport->owner);
+ hba->transport = NULL;
+out_free_hba:
+ kfree(hba);
+ return ERR_PTR(ret);
+}
+
+int
+core_delete_hba(struct se_hba *hba)
+{
+ struct se_device *dev, *dev_tmp;
+
+ spin_lock(&hba->device_lock);
+ list_for_each_entry_safe(dev, dev_tmp, &hba->hba_dev_list, dev_list) {
+
+ se_clear_dev_ports(dev);
+ spin_unlock(&hba->device_lock);
+
+ se_release_device_for_hba(dev);
+
+ spin_lock(&hba->device_lock);
+ }
+ spin_unlock(&hba->device_lock);
+
+ hba->transport->detach_hba(hba);
+
+ spin_lock(&se_global->hba_lock);
+ list_del(&hba->hba_list);
+ spin_unlock(&se_global->hba_lock);
+
+ printk(KERN_INFO "CORE_HBA[%d] - Detached HBA from Generic Target"
+ " Core\n", hba->hba_id);
+
+ if (hba->transport->owner)
+ module_put(hba->transport->owner);
+
+ hba->transport = NULL;
+ kfree(hba);
+ return 0;
+}
diff --git a/drivers/target/target_core_hba.h b/drivers/target/target_core_hba.h
new file mode 100644
index 0000000..bb0fea5
--- /dev/null
+++ b/drivers/target/target_core_hba.h
@@ -0,0 +1,7 @@
+#ifndef TARGET_CORE_HBA_H
+#define TARGET_CORE_HBA_H
+
+extern struct se_hba *core_alloc_hba(const char *, u32, u32);
+extern int core_delete_hba(struct se_hba *);
+
+#endif /* TARGET_CORE_HBA_H */
diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c
new file mode 100644
index 0000000..c6e0d75
--- /dev/null
+++ b/drivers/target/target_core_iblock.c
@@ -0,0 +1,808 @@
+/*******************************************************************************
+ * Filename: target_core_iblock.c
+ *
+ * This file contains the Storage Engine <-> Linux BlockIO transport
+ * specific functions.
+ *
+ * Copyright (c) 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/string.h>
+#include <linux/parser.h>
+#include <linux/timer.h>
+#include <linux/fs.h>
+#include <linux/blkdev.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <linux/bio.h>
+#include <linux/genhd.h>
+#include <linux/file.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_host.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_transport.h>
+
+#include "target_core_iblock.h"
+
+#if 0
+#define DEBUG_IBLOCK(x...) printk(x)
+#else
+#define DEBUG_IBLOCK(x...)
+#endif
+
+static struct se_subsystem_api iblock_template;
+
+static void iblock_bio_done(struct bio *, int);
+
+/* iblock_attach_hba(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static int iblock_attach_hba(struct se_hba *hba, u32 host_id)
+{
+ struct iblock_hba *ib_host;
+
+ ib_host = kzalloc(sizeof(struct iblock_hba), GFP_KERNEL);
+ if (!(ib_host)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " struct iblock_hba\n");
+ return -ENOMEM;
+ }
+
+ ib_host->iblock_host_id = host_id;
+
+ atomic_set(&hba->left_queue_depth, IBLOCK_HBA_QUEUE_DEPTH);
+ atomic_set(&hba->max_queue_depth, IBLOCK_HBA_QUEUE_DEPTH);
+ hba->hba_ptr = (void *) ib_host;
+
+ printk(KERN_INFO "CORE_HBA[%d] - TCM iBlock HBA Driver %s on"
+ " Generic Target Core Stack %s\n", hba->hba_id,
+ IBLOCK_VERSION, TARGET_CORE_MOD_VERSION);
+
+ printk(KERN_INFO "CORE_HBA[%d] - Attached iBlock HBA: %u to Generic"
+ " Target Core TCQ Depth: %d\n", hba->hba_id,
+ ib_host->iblock_host_id, atomic_read(&hba->max_queue_depth));
+
+ return 0;
+}
+
+static void iblock_detach_hba(struct se_hba *hba)
+{
+ struct iblock_hba *ib_host = hba->hba_ptr;
+
+ printk(KERN_INFO "CORE_HBA[%d] - Detached iBlock HBA: %u from Generic"
+ " Target Core\n", hba->hba_id, ib_host->iblock_host_id);
+
+ kfree(ib_host);
+ hba->hba_ptr = NULL;
+}
+
+static void *iblock_allocate_virtdevice(struct se_hba *hba, const char *name)
+{
+ struct iblock_dev *ib_dev = NULL;
+ struct iblock_hba *ib_host = hba->hba_ptr;
+
+ ib_dev = kzalloc(sizeof(struct iblock_dev), GFP_KERNEL);
+ if (!(ib_dev)) {
+ printk(KERN_ERR "Unable to allocate struct iblock_dev\n");
+ return NULL;
+ }
+ ib_dev->ibd_host = ib_host;
+
+ printk(KERN_INFO "IBLOCK: Allocated ib_dev for %s\n", name);
+
+ return ib_dev;
+}
+
+static struct se_device *iblock_create_virtdevice(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ void *p)
+{
+ struct iblock_dev *ib_dev = p;
+ struct se_device *dev;
+ struct se_dev_limits dev_limits;
+ struct block_device *bd = NULL;
+ struct request_queue *q;
+ struct queue_limits *limits;
+ u32 dev_flags = 0;
+
+ if (!(ib_dev)) {
+ printk(KERN_ERR "Unable to locate struct iblock_dev parameter\n");
+ return 0;
+ }
+ memset(&dev_limits, 0, sizeof(struct se_dev_limits));
+ /*
+ * These settings need to be made tunable..
+ */
+ ib_dev->ibd_bio_set = bioset_create(32, 64);
+ if (!(ib_dev->ibd_bio_set)) {
+ printk(KERN_ERR "IBLOCK: Unable to create bioset()\n");
+ return 0;
+ }
+ printk(KERN_INFO "IBLOCK: Created bio_set()\n");
+ /*
+ * iblock_check_configfs_dev_params() ensures that ib_dev->ibd_udev_path
+ * must already have been set in order for echo 1 > $HBA/$DEV/enable to run.
+ */
+ printk(KERN_INFO "IBLOCK: Claiming struct block_device: %s\n",
+ ib_dev->ibd_udev_path);
+
+ bd = blkdev_get_by_path(ib_dev->ibd_udev_path,
+ FMODE_WRITE|FMODE_READ|FMODE_EXCL, ib_dev);
+ if (!(bd))
+ goto failed;
+ /*
+ * Setup the local scope queue_limits from struct request_queue->limits
+ * to pass into transport_add_device_to_core_hba() as struct se_dev_limits.
+ */
+ q = bdev_get_queue(bd);
+ limits = &dev_limits.limits;
+ limits->logical_block_size = bdev_logical_block_size(bd);
+ limits->max_hw_sectors = queue_max_hw_sectors(q);
+ limits->max_sectors = queue_max_sectors(q);
+ dev_limits.hw_queue_depth = IBLOCK_MAX_DEVICE_QUEUE_DEPTH;
+ dev_limits.queue_depth = IBLOCK_DEVICE_QUEUE_DEPTH;
+
+ ib_dev->ibd_major = MAJOR(bd->bd_dev);
+ ib_dev->ibd_minor = MINOR(bd->bd_dev);
+ ib_dev->ibd_bd = bd;
+
+ dev = transport_add_device_to_core_hba(hba,
+ &iblock_template, se_dev, dev_flags, (void *)ib_dev,
+ &dev_limits, "IBLOCK", IBLOCK_VERSION);
+ if (!(dev))
+ goto failed;
+
+ ib_dev->ibd_depth = dev->queue_depth;
+
+ /*
+ * Check if the underlying struct block_device request_queue supports
+ * the QUEUE_FLAG_DISCARD bit for UNMAP/WRITE_SAME in SCSI + TRIM
+ * in ATA and we need to set TPE=1
+ */
+ if (blk_queue_discard(bdev_get_queue(bd))) {
+ struct request_queue *q = bdev_get_queue(bd);
+
+ DEV_ATTRIB(dev)->max_unmap_lba_count =
+ q->limits.max_discard_sectors;
+ /*
+ * Currently hardcoded to 1 in Linux/SCSI code..
+ */
+ DEV_ATTRIB(dev)->max_unmap_block_desc_count = 1;
+ DEV_ATTRIB(dev)->unmap_granularity =
+ q->limits.discard_granularity;
+ DEV_ATTRIB(dev)->unmap_granularity_alignment =
+ q->limits.discard_alignment;
+
+ printk(KERN_INFO "IBLOCK: BLOCK Discard support available,"
+ " disabled by default\n");
+ }
+
+ return dev;
+
+failed:
+ if (ib_dev->ibd_bio_set) {
+ bioset_free(ib_dev->ibd_bio_set);
+ ib_dev->ibd_bio_set = NULL;
+ }
+ ib_dev->ibd_bd = NULL;
+ ib_dev->ibd_major = 0;
+ ib_dev->ibd_minor = 0;
+ return NULL;
+}
+
+static void iblock_free_device(void *p)
+{
+ struct iblock_dev *ib_dev = p;
+
+ blkdev_put(ib_dev->ibd_bd, FMODE_WRITE|FMODE_READ|FMODE_EXCL);
+ bioset_free(ib_dev->ibd_bio_set);
+ kfree(ib_dev);
+}
+
+static inline struct iblock_req *IBLOCK_REQ(struct se_task *task)
+{
+ return container_of(task, struct iblock_req, ib_task);
+}
+
+static struct se_task *
+iblock_alloc_task(struct se_cmd *cmd)
+{
+ struct iblock_req *ib_req;
+
+ ib_req = kzalloc(sizeof(struct iblock_req), GFP_KERNEL);
+ if (!(ib_req)) {
+ printk(KERN_ERR "Unable to allocate memory for struct iblock_req\n");
+ return NULL;
+ }
+
+ ib_req->ib_dev = SE_DEV(cmd)->dev_ptr;
+ atomic_set(&ib_req->ib_bio_cnt, 0);
+ return &ib_req->ib_task;
+}
+
+static unsigned long long iblock_emulate_read_cap_with_block_size(
+ struct se_device *dev,
+ struct block_device *bd,
+ struct request_queue *q)
+{
+ unsigned long long blocks_long = (div_u64(i_size_read(bd->bd_inode),
+ bdev_logical_block_size(bd)) - 1);
+ u32 block_size = bdev_logical_block_size(bd);
+
+ if (block_size == DEV_ATTRIB(dev)->block_size)
+ return blocks_long;
+
+ switch (block_size) {
+ case 4096:
+ switch (DEV_ATTRIB(dev)->block_size) {
+ case 2048:
+ blocks_long <<= 1;
+ break;
+ case 1024:
+ blocks_long <<= 2;
+ break;
+ case 512:
+ blocks_long <<= 3;
+ default:
+ break;
+ }
+ break;
+ case 2048:
+ switch (DEV_ATTRIB(dev)->block_size) {
+ case 4096:
+ blocks_long >>= 1;
+ break;
+ case 1024:
+ blocks_long <<= 1;
+ break;
+ case 512:
+ blocks_long <<= 2;
+ break;
+ default:
+ break;
+ }
+ break;
+ case 1024:
+ switch (DEV_ATTRIB(dev)->block_size) {
+ case 4096:
+ blocks_long >>= 2;
+ break;
+ case 2048:
+ blocks_long >>= 1;
+ break;
+ case 512:
+ blocks_long <<= 1;
+ break;
+ default:
+ break;
+ }
+ break;
+ case 512:
+ switch (DEV_ATTRIB(dev)->block_size) {
+ case 4096:
+ blocks_long >>= 3;
+ break;
+ case 2048:
+ blocks_long >>= 2;
+ break;
+ case 1024:
+ blocks_long >>= 1;
+ break;
+ default:
+ break;
+ }
+ break;
+ default:
+ break;
+ }
+
+ return blocks_long;
+}
+
+/*
+ * Emulate SYCHRONIZE_CACHE_*
+ */
+static void iblock_emulate_sync_cache(struct se_task *task)
+{
+ struct se_cmd *cmd = TASK_CMD(task);
+ struct iblock_dev *ib_dev = cmd->se_dev->dev_ptr;
+ int immed = (T_TASK(cmd)->t_task_cdb[1] & 0x2);
+ sector_t error_sector;
+ int ret;
+
+ /*
+ * If the Immediate bit is set, queue up the GOOD response
+ * for this SYNCHRONIZE_CACHE op
+ */
+ if (immed)
+ transport_complete_sync_cache(cmd, 1);
+
+ /*
+ * blkdev_issue_flush() does not support a specifying a range, so
+ * we have to flush the entire cache.
+ */
+ ret = blkdev_issue_flush(ib_dev->ibd_bd, GFP_KERNEL, &error_sector);
+ if (ret != 0) {
+ printk(KERN_ERR "IBLOCK: block_issue_flush() failed: %d "
+ " error_sector: %llu\n", ret,
+ (unsigned long long)error_sector);
+ }
+
+ if (!immed)
+ transport_complete_sync_cache(cmd, ret == 0);
+}
+
+/*
+ * Tell TCM Core that we are capable of WriteCache emulation for
+ * an underlying struct se_device.
+ */
+static int iblock_emulated_write_cache(struct se_device *dev)
+{
+ return 1;
+}
+
+static int iblock_emulated_dpo(struct se_device *dev)
+{
+ return 0;
+}
+
+/*
+ * Tell TCM Core that we will be emulating Forced Unit Access (FUA) for WRITEs
+ * for TYPE_DISK.
+ */
+static int iblock_emulated_fua_write(struct se_device *dev)
+{
+ return 1;
+}
+
+static int iblock_emulated_fua_read(struct se_device *dev)
+{
+ return 0;
+}
+
+static int iblock_do_task(struct se_task *task)
+{
+ struct se_device *dev = task->task_se_cmd->se_dev;
+ struct iblock_req *req = IBLOCK_REQ(task);
+ struct iblock_dev *ibd = (struct iblock_dev *)req->ib_dev;
+ struct request_queue *q = bdev_get_queue(ibd->ibd_bd);
+ struct bio *bio = req->ib_bio, *nbio = NULL;
+ int rw;
+
+ if (task->task_data_direction == DMA_TO_DEVICE) {
+ /*
+ * Force data to disk if we pretend to not have a volatile
+ * write cache, or the initiator set the Force Unit Access bit.
+ */
+ if (DEV_ATTRIB(dev)->emulate_write_cache == 0 ||
+ (DEV_ATTRIB(dev)->emulate_fua_write > 0 &&
+ T_TASK(task->task_se_cmd)->t_tasks_fua))
+ rw = WRITE_FUA;
+ else
+ rw = WRITE;
+ } else {
+ rw = READ;
+ }
+
+ while (bio) {
+ nbio = bio->bi_next;
+ bio->bi_next = NULL;
+ DEBUG_IBLOCK("Calling submit_bio() task: %p bio: %p"
+ " bio->bi_sector: %llu\n", task, bio, bio->bi_sector);
+
+ submit_bio(rw, bio);
+ bio = nbio;
+ }
+
+ if (q->unplug_fn)
+ q->unplug_fn(q);
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+}
+
+static int iblock_do_discard(struct se_device *dev, sector_t lba, u32 range)
+{
+ struct iblock_dev *ibd = dev->dev_ptr;
+ struct block_device *bd = ibd->ibd_bd;
+ int barrier = 0;
+
+ return blkdev_issue_discard(bd, lba, range, GFP_KERNEL, barrier);
+}
+
+static void iblock_free_task(struct se_task *task)
+{
+ struct iblock_req *req = IBLOCK_REQ(task);
+ struct bio *bio, *hbio = req->ib_bio;
+ /*
+ * We only release the bio(s) here if iblock_bio_done() has not called
+ * bio_put() -> iblock_bio_destructor().
+ */
+ while (hbio != NULL) {
+ bio = hbio;
+ hbio = hbio->bi_next;
+ bio->bi_next = NULL;
+ bio_put(bio);
+ }
+
+ kfree(req);
+}
+
+enum {
+ Opt_udev_path, Opt_force, Opt_err
+};
+
+static match_table_t tokens = {
+ {Opt_udev_path, "udev_path=%s"},
+ {Opt_force, "force=%d"},
+ {Opt_err, NULL}
+};
+
+static ssize_t iblock_set_configfs_dev_params(struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ const char *page, ssize_t count)
+{
+ struct iblock_dev *ib_dev = se_dev->se_dev_su_ptr;
+ char *orig, *ptr, *opts;
+ substring_t args[MAX_OPT_ARGS];
+ int ret = 0, arg, token;
+
+ opts = kstrdup(page, GFP_KERNEL);
+ if (!opts)
+ return -ENOMEM;
+
+ orig = opts;
+
+ while ((ptr = strsep(&opts, ",")) != NULL) {
+ if (!*ptr)
+ continue;
+
+ token = match_token(ptr, tokens, args);
+ switch (token) {
+ case Opt_udev_path:
+ if (ib_dev->ibd_bd) {
+ printk(KERN_ERR "Unable to set udev_path= while"
+ " ib_dev->ibd_bd exists\n");
+ ret = -EEXIST;
+ goto out;
+ }
+
+ ret = snprintf(ib_dev->ibd_udev_path, SE_UDEV_PATH_LEN,
+ "%s", match_strdup(&args[0]));
+ printk(KERN_INFO "IBLOCK: Referencing UDEV path: %s\n",
+ ib_dev->ibd_udev_path);
+ ib_dev->ibd_flags |= IBDF_HAS_UDEV_PATH;
+ break;
+ case Opt_force:
+ match_int(args, &arg);
+ ib_dev->ibd_force = arg;
+ printk(KERN_INFO "IBLOCK: Set force=%d\n",
+ ib_dev->ibd_force);
+ break;
+ default:
+ break;
+ }
+ }
+
+out:
+ kfree(orig);
+ return (!ret) ? count : ret;
+}
+
+static ssize_t iblock_check_configfs_dev_params(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev)
+{
+ struct iblock_dev *ibd = se_dev->se_dev_su_ptr;
+
+ if (!(ibd->ibd_flags & IBDF_HAS_UDEV_PATH)) {
+ printk(KERN_ERR "Missing udev_path= parameters for IBLOCK\n");
+ return -1;
+ }
+
+ return 0;
+}
+
+static ssize_t iblock_show_configfs_dev_params(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ char *b)
+{
+ struct iblock_dev *ibd = se_dev->se_dev_su_ptr;
+ struct block_device *bd = ibd->ibd_bd;
+ char buf[BDEVNAME_SIZE];
+ ssize_t bl = 0;
+
+ if (bd)
+ bl += sprintf(b + bl, "iBlock device: %s",
+ bdevname(bd, buf));
+ if (ibd->ibd_flags & IBDF_HAS_UDEV_PATH) {
+ bl += sprintf(b + bl, " UDEV PATH: %s\n",
+ ibd->ibd_udev_path);
+ } else
+ bl += sprintf(b + bl, "\n");
+
+ bl += sprintf(b + bl, " ");
+ if (bd) {
+ bl += sprintf(b + bl, "Major: %d Minor: %d %s\n",
+ ibd->ibd_major, ibd->ibd_minor, (!bd->bd_contains) ?
+ "" : (bd->bd_holder == (struct iblock_dev *)ibd) ?
+ "CLAIMED: IBLOCK" : "CLAIMED: OS");
+ } else {
+ bl += sprintf(b + bl, "Major: %d Minor: %d\n",
+ ibd->ibd_major, ibd->ibd_minor);
+ }
+
+ return bl;
+}
+
+static void iblock_bio_destructor(struct bio *bio)
+{
+ struct se_task *task = bio->bi_private;
+ struct iblock_dev *ib_dev = task->se_dev->dev_ptr;
+
+ bio_free(bio, ib_dev->ibd_bio_set);
+}
+
+static struct bio *iblock_get_bio(
+ struct se_task *task,
+ struct iblock_req *ib_req,
+ struct iblock_dev *ib_dev,
+ int *ret,
+ sector_t lba,
+ u32 sg_num)
+{
+ struct bio *bio;
+
+ bio = bio_alloc_bioset(GFP_NOIO, sg_num, ib_dev->ibd_bio_set);
+ if (!(bio)) {
+ printk(KERN_ERR "Unable to allocate memory for bio\n");
+ *ret = PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+ return NULL;
+ }
+
+ DEBUG_IBLOCK("Allocated bio: %p task_sg_num: %u using ibd_bio_set:"
+ " %p\n", bio, task->task_sg_num, ib_dev->ibd_bio_set);
+ DEBUG_IBLOCK("Allocated bio: %p task_size: %u\n", bio, task->task_size);
+
+ bio->bi_bdev = ib_dev->ibd_bd;
+ bio->bi_private = (void *) task;
+ bio->bi_destructor = iblock_bio_destructor;
+ bio->bi_end_io = &iblock_bio_done;
+ bio->bi_sector = lba;
+ atomic_inc(&ib_req->ib_bio_cnt);
+
+ DEBUG_IBLOCK("Set bio->bi_sector: %llu\n", bio->bi_sector);
+ DEBUG_IBLOCK("Set ib_req->ib_bio_cnt: %d\n",
+ atomic_read(&ib_req->ib_bio_cnt));
+ return bio;
+}
+
+static int iblock_map_task_SG(struct se_task *task)
+{
+ struct se_cmd *cmd = task->task_se_cmd;
+ struct se_device *dev = SE_DEV(cmd);
+ struct iblock_dev *ib_dev = task->se_dev->dev_ptr;
+ struct iblock_req *ib_req = IBLOCK_REQ(task);
+ struct bio *bio = NULL, *hbio = NULL, *tbio = NULL;
+ struct scatterlist *sg;
+ int ret = 0;
+ u32 i, sg_num = task->task_sg_num;
+ sector_t block_lba;
+ /*
+ * Do starting conversion up from non 512-byte blocksize with
+ * struct se_task SCSI blocksize into Linux/Block 512 units for BIO.
+ */
+ if (DEV_ATTRIB(dev)->block_size == 4096)
+ block_lba = (task->task_lba << 3);
+ else if (DEV_ATTRIB(dev)->block_size == 2048)
+ block_lba = (task->task_lba << 2);
+ else if (DEV_ATTRIB(dev)->block_size == 1024)
+ block_lba = (task->task_lba << 1);
+ else if (DEV_ATTRIB(dev)->block_size == 512)
+ block_lba = task->task_lba;
+ else {
+ printk(KERN_ERR "Unsupported SCSI -> BLOCK LBA conversion:"
+ " %u\n", DEV_ATTRIB(dev)->block_size);
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+
+ bio = iblock_get_bio(task, ib_req, ib_dev, &ret, block_lba, sg_num);
+ if (!(bio))
+ return ret;
+
+ ib_req->ib_bio = bio;
+ hbio = tbio = bio;
+ /*
+ * Use fs/bio.c:bio_add_pages() to setup the bio_vec maplist
+ * from TCM struct se_mem -> task->task_sg -> struct scatterlist memory.
+ */
+ for_each_sg(task->task_sg, sg, task->task_sg_num, i) {
+ DEBUG_IBLOCK("task: %p bio: %p Calling bio_add_page(): page:"
+ " %p len: %u offset: %u\n", task, bio, sg_page(sg),
+ sg->length, sg->offset);
+again:
+ ret = bio_add_page(bio, sg_page(sg), sg->length, sg->offset);
+ if (ret != sg->length) {
+
+ DEBUG_IBLOCK("*** Set bio->bi_sector: %llu\n",
+ bio->bi_sector);
+ DEBUG_IBLOCK("** task->task_size: %u\n",
+ task->task_size);
+ DEBUG_IBLOCK("*** bio->bi_max_vecs: %u\n",
+ bio->bi_max_vecs);
+ DEBUG_IBLOCK("*** bio->bi_vcnt: %u\n",
+ bio->bi_vcnt);
+
+ bio = iblock_get_bio(task, ib_req, ib_dev, &ret,
+ block_lba, sg_num);
+ if (!(bio))
+ goto fail;
+
+ tbio = tbio->bi_next = bio;
+ DEBUG_IBLOCK("-----------------> Added +1 bio: %p to"
+ " list, Going to again\n", bio);
+ goto again;
+ }
+ /* Always in 512 byte units for Linux/Block */
+ block_lba += sg->length >> IBLOCK_LBA_SHIFT;
+ sg_num--;
+ DEBUG_IBLOCK("task: %p bio-add_page() passed!, decremented"
+ " sg_num to %u\n", task, sg_num);
+ DEBUG_IBLOCK("task: %p bio_add_page() passed!, increased lba"
+ " to %llu\n", task, block_lba);
+ DEBUG_IBLOCK("task: %p bio_add_page() passed!, bio->bi_vcnt:"
+ " %u\n", task, bio->bi_vcnt);
+ }
+
+ return 0;
+fail:
+ while (hbio) {
+ bio = hbio;
+ hbio = hbio->bi_next;
+ bio->bi_next = NULL;
+ bio_put(bio);
+ }
+ return ret;
+}
+
+static unsigned char *iblock_get_cdb(struct se_task *task)
+{
+ return IBLOCK_REQ(task)->ib_scsi_cdb;
+}
+
+static u32 iblock_get_device_rev(struct se_device *dev)
+{
+ return SCSI_SPC_2; /* Returns SPC-3 in Initiator Data */
+}
+
+static u32 iblock_get_device_type(struct se_device *dev)
+{
+ return TYPE_DISK;
+}
+
+static sector_t iblock_get_blocks(struct se_device *dev)
+{
+ struct iblock_dev *ibd = dev->dev_ptr;
+ struct block_device *bd = ibd->ibd_bd;
+ struct request_queue *q = bdev_get_queue(bd);
+
+ return iblock_emulate_read_cap_with_block_size(dev, bd, q);
+}
+
+static void iblock_bio_done(struct bio *bio, int err)
+{
+ struct se_task *task = bio->bi_private;
+ struct iblock_req *ibr = IBLOCK_REQ(task);
+ /*
+ * Set -EIO if !BIO_UPTODATE and the passed is still err=0
+ */
+ if (!(test_bit(BIO_UPTODATE, &bio->bi_flags)) && !(err))
+ err = -EIO;
+
+ if (err != 0) {
+ printk(KERN_ERR "test_bit(BIO_UPTODATE) failed for bio: %p,"
+ " err: %d\n", bio, err);
+ /*
+ * Bump the ib_bio_err_cnt and release bio.
+ */
+ atomic_inc(&ibr->ib_bio_err_cnt);
+ smp_mb__after_atomic_inc();
+ bio_put(bio);
+ /*
+ * Wait to complete the task until the last bio as completed.
+ */
+ if (!(atomic_dec_and_test(&ibr->ib_bio_cnt)))
+ return;
+
+ ibr->ib_bio = NULL;
+ transport_complete_task(task, 0);
+ return;
+ }
+ DEBUG_IBLOCK("done[%p] bio: %p task_lba: %llu bio_lba: %llu err=%d\n",
+ task, bio, task->task_lba, bio->bi_sector, err);
+ /*
+ * bio_put() will call iblock_bio_destructor() to release the bio back
+ * to ibr->ib_bio_set.
+ */
+ bio_put(bio);
+ /*
+ * Wait to complete the task until the last bio as completed.
+ */
+ if (!(atomic_dec_and_test(&ibr->ib_bio_cnt)))
+ return;
+ /*
+ * Return GOOD status for task if zero ib_bio_err_cnt exists.
+ */
+ ibr->ib_bio = NULL;
+ transport_complete_task(task, (!atomic_read(&ibr->ib_bio_err_cnt)));
+}
+
+static struct se_subsystem_api iblock_template = {
+ .name = "iblock",
+ .owner = THIS_MODULE,
+ .transport_type = TRANSPORT_PLUGIN_VHBA_PDEV,
+ .map_task_SG = iblock_map_task_SG,
+ .attach_hba = iblock_attach_hba,
+ .detach_hba = iblock_detach_hba,
+ .allocate_virtdevice = iblock_allocate_virtdevice,
+ .create_virtdevice = iblock_create_virtdevice,
+ .free_device = iblock_free_device,
+ .dpo_emulated = iblock_emulated_dpo,
+ .fua_write_emulated = iblock_emulated_fua_write,
+ .fua_read_emulated = iblock_emulated_fua_read,
+ .write_cache_emulated = iblock_emulated_write_cache,
+ .alloc_task = iblock_alloc_task,
+ .do_task = iblock_do_task,
+ .do_discard = iblock_do_discard,
+ .do_sync_cache = iblock_emulate_sync_cache,
+ .free_task = iblock_free_task,
+ .check_configfs_dev_params = iblock_check_configfs_dev_params,
+ .set_configfs_dev_params = iblock_set_configfs_dev_params,
+ .show_configfs_dev_params = iblock_show_configfs_dev_params,
+ .get_cdb = iblock_get_cdb,
+ .get_device_rev = iblock_get_device_rev,
+ .get_device_type = iblock_get_device_type,
+ .get_blocks = iblock_get_blocks,
+};
+
+static int __init iblock_module_init(void)
+{
+ return transport_subsystem_register(&iblock_template);
+}
+
+static void iblock_module_exit(void)
+{
+ transport_subsystem_release(&iblock_template);
+}
+
+MODULE_DESCRIPTION("TCM IBLOCK subsystem plugin");
+MODULE_AUTHOR("nab@Linux-iSCSI.org");
+MODULE_LICENSE("GPL");
+
+module_init(iblock_module_init);
+module_exit(iblock_module_exit);
diff --git a/drivers/target/target_core_iblock.h b/drivers/target/target_core_iblock.h
new file mode 100644
index 0000000..64c1f4d
--- /dev/null
+++ b/drivers/target/target_core_iblock.h
@@ -0,0 +1,40 @@
+#ifndef TARGET_CORE_IBLOCK_H
+#define TARGET_CORE_IBLOCK_H
+
+#define IBLOCK_VERSION "4.0"
+
+#define IBLOCK_HBA_QUEUE_DEPTH 512
+#define IBLOCK_DEVICE_QUEUE_DEPTH 32
+#define IBLOCK_MAX_DEVICE_QUEUE_DEPTH 128
+#define IBLOCK_MAX_CDBS 16
+#define IBLOCK_LBA_SHIFT 9
+
+struct iblock_req {
+ struct se_task ib_task;
+ unsigned char ib_scsi_cdb[TCM_MAX_COMMAND_SIZE];
+ atomic_t ib_bio_cnt;
+ atomic_t ib_bio_err_cnt;
+ struct bio *ib_bio;
+ struct iblock_dev *ib_dev;
+} ____cacheline_aligned;
+
+#define IBDF_HAS_UDEV_PATH 0x01
+#define IBDF_HAS_FORCE 0x02
+
+struct iblock_dev {
+ unsigned char ibd_udev_path[SE_UDEV_PATH_LEN];
+ int ibd_force;
+ int ibd_major;
+ int ibd_minor;
+ u32 ibd_depth;
+ u32 ibd_flags;
+ struct bio_set *ibd_bio_set;
+ struct block_device *ibd_bd;
+ struct iblock_hba *ibd_host;
+} ____cacheline_aligned;
+
+struct iblock_hba {
+ int iblock_host_id;
+} ____cacheline_aligned;
+
+#endif /* TARGET_CORE_IBLOCK_H */
diff --git a/drivers/target/target_core_mib.c b/drivers/target/target_core_mib.c
new file mode 100644
index 0000000..d5a48aa
--- /dev/null
+++ b/drivers/target/target_core_mib.c
@@ -0,0 +1,1078 @@
+/*******************************************************************************
+ * Filename: target_core_mib.c
+ *
+ * Copyright (c) 2006-2007 SBE, Inc. All Rights Reserved.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@linux-iscsi.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/timer.h>
+#include <linux/string.h>
+#include <linux/version.h>
+#include <generated/utsrelease.h>
+#include <linux/utsname.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/blkdev.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_host.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_configfs.h>
+
+#include "target_core_hba.h"
+#include "target_core_mib.h"
+
+/* SCSI mib table index */
+static struct scsi_index_table scsi_index_table;
+
+#ifndef INITIAL_JIFFIES
+#define INITIAL_JIFFIES ((unsigned long)(unsigned int) (-300*HZ))
+#endif
+
+/* SCSI Instance Table */
+#define SCSI_INST_SW_INDEX 1
+#define SCSI_TRANSPORT_INDEX 1
+
+#define NONE "None"
+#define ISPRINT(a) ((a >= ' ') && (a <= '~'))
+
+static inline int list_is_first(const struct list_head *list,
+ const struct list_head *head)
+{
+ return list->prev == head;
+}
+
+static void *locate_hba_start(
+ struct seq_file *seq,
+ loff_t *pos)
+{
+ spin_lock(&se_global->g_device_lock);
+ return seq_list_start(&se_global->g_se_dev_list, *pos);
+}
+
+static void *locate_hba_next(
+ struct seq_file *seq,
+ void *v,
+ loff_t *pos)
+{
+ return seq_list_next(v, &se_global->g_se_dev_list, pos);
+}
+
+static void locate_hba_stop(struct seq_file *seq, void *v)
+{
+ spin_unlock(&se_global->g_device_lock);
+}
+
+/****************************************************************************
+ * SCSI MIB Tables
+ ****************************************************************************/
+
+/*
+ * SCSI Instance Table
+ */
+static void *scsi_inst_seq_start(
+ struct seq_file *seq,
+ loff_t *pos)
+{
+ spin_lock(&se_global->hba_lock);
+ return seq_list_start(&se_global->g_hba_list, *pos);
+}
+
+static void *scsi_inst_seq_next(
+ struct seq_file *seq,
+ void *v,
+ loff_t *pos)
+{
+ return seq_list_next(v, &se_global->g_hba_list, pos);
+}
+
+static void scsi_inst_seq_stop(struct seq_file *seq, void *v)
+{
+ spin_unlock(&se_global->hba_lock);
+}
+
+static int scsi_inst_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_hba *hba = list_entry(v, struct se_hba, hba_list);
+
+ if (list_is_first(&hba->hba_list, &se_global->g_hba_list))
+ seq_puts(seq, "inst sw_indx\n");
+
+ seq_printf(seq, "%u %u\n", hba->hba_index, SCSI_INST_SW_INDEX);
+ seq_printf(seq, "plugin: %s version: %s\n",
+ hba->transport->name, TARGET_CORE_VERSION);
+
+ return 0;
+}
+
+static const struct seq_operations scsi_inst_seq_ops = {
+ .start = scsi_inst_seq_start,
+ .next = scsi_inst_seq_next,
+ .stop = scsi_inst_seq_stop,
+ .show = scsi_inst_seq_show
+};
+
+static int scsi_inst_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_inst_seq_ops);
+}
+
+static const struct file_operations scsi_inst_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_inst_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/*
+ * SCSI Device Table
+ */
+static void *scsi_dev_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ return locate_hba_start(seq, pos);
+}
+
+static void *scsi_dev_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ return locate_hba_next(seq, v, pos);
+}
+
+static void scsi_dev_seq_stop(struct seq_file *seq, void *v)
+{
+ locate_hba_stop(seq, v);
+}
+
+static int scsi_dev_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_hba *hba;
+ struct se_subsystem_dev *se_dev = list_entry(v, struct se_subsystem_dev,
+ g_se_dev_list);
+ struct se_device *dev = se_dev->se_dev_ptr;
+ char str[28];
+ int k;
+
+ if (list_is_first(&se_dev->g_se_dev_list, &se_global->g_se_dev_list))
+ seq_puts(seq, "inst indx role ports\n");
+
+ if (!(dev))
+ return 0;
+
+ hba = dev->se_hba;
+ if (!(hba)) {
+ /* Log error ? */
+ return 0;
+ }
+
+ seq_printf(seq, "%u %u %s %u\n", hba->hba_index,
+ dev->dev_index, "Target", dev->dev_port_count);
+
+ memcpy(&str[0], (void *)DEV_T10_WWN(dev), 28);
+
+ /* vendor */
+ for (k = 0; k < 8; k++)
+ str[k] = ISPRINT(DEV_T10_WWN(dev)->vendor[k]) ?
+ DEV_T10_WWN(dev)->vendor[k] : 0x20;
+ str[k] = 0x20;
+
+ /* model */
+ for (k = 0; k < 16; k++)
+ str[k+9] = ISPRINT(DEV_T10_WWN(dev)->model[k]) ?
+ DEV_T10_WWN(dev)->model[k] : 0x20;
+ str[k + 9] = 0;
+
+ seq_printf(seq, "dev_alias: %s\n", str);
+
+ return 0;
+}
+
+static const struct seq_operations scsi_dev_seq_ops = {
+ .start = scsi_dev_seq_start,
+ .next = scsi_dev_seq_next,
+ .stop = scsi_dev_seq_stop,
+ .show = scsi_dev_seq_show
+};
+
+static int scsi_dev_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_dev_seq_ops);
+}
+
+static const struct file_operations scsi_dev_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_dev_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/*
+ * SCSI Port Table
+ */
+static void *scsi_port_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ return locate_hba_start(seq, pos);
+}
+
+static void *scsi_port_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ return locate_hba_next(seq, v, pos);
+}
+
+static void scsi_port_seq_stop(struct seq_file *seq, void *v)
+{
+ locate_hba_stop(seq, v);
+}
+
+static int scsi_port_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_hba *hba;
+ struct se_subsystem_dev *se_dev = list_entry(v, struct se_subsystem_dev,
+ g_se_dev_list);
+ struct se_device *dev = se_dev->se_dev_ptr;
+ struct se_port *sep, *sep_tmp;
+
+ if (list_is_first(&se_dev->g_se_dev_list, &se_global->g_se_dev_list))
+ seq_puts(seq, "inst device indx role busy_count\n");
+
+ if (!(dev))
+ return 0;
+
+ hba = dev->se_hba;
+ if (!(hba)) {
+ /* Log error ? */
+ return 0;
+ }
+
+ /* FIXME: scsiPortBusyStatuses count */
+ spin_lock(&dev->se_port_lock);
+ list_for_each_entry_safe(sep, sep_tmp, &dev->dev_sep_list, sep_list) {
+ seq_printf(seq, "%u %u %u %s%u %u\n", hba->hba_index,
+ dev->dev_index, sep->sep_index, "Device",
+ dev->dev_index, 0);
+ }
+ spin_unlock(&dev->se_port_lock);
+
+ return 0;
+}
+
+static const struct seq_operations scsi_port_seq_ops = {
+ .start = scsi_port_seq_start,
+ .next = scsi_port_seq_next,
+ .stop = scsi_port_seq_stop,
+ .show = scsi_port_seq_show
+};
+
+static int scsi_port_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_port_seq_ops);
+}
+
+static const struct file_operations scsi_port_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_port_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/*
+ * SCSI Transport Table
+ */
+static void *scsi_transport_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ return locate_hba_start(seq, pos);
+}
+
+static void *scsi_transport_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ return locate_hba_next(seq, v, pos);
+}
+
+static void scsi_transport_seq_stop(struct seq_file *seq, void *v)
+{
+ locate_hba_stop(seq, v);
+}
+
+static int scsi_transport_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_hba *hba;
+ struct se_subsystem_dev *se_dev = list_entry(v, struct se_subsystem_dev,
+ g_se_dev_list);
+ struct se_device *dev = se_dev->se_dev_ptr;
+ struct se_port *se, *se_tmp;
+ struct se_portal_group *tpg;
+ struct t10_wwn *wwn;
+ char buf[64];
+
+ if (list_is_first(&se_dev->g_se_dev_list, &se_global->g_se_dev_list))
+ seq_puts(seq, "inst device indx dev_name\n");
+
+ if (!(dev))
+ return 0;
+
+ hba = dev->se_hba;
+ if (!(hba)) {
+ /* Log error ? */
+ return 0;
+ }
+
+ wwn = DEV_T10_WWN(dev);
+
+ spin_lock(&dev->se_port_lock);
+ list_for_each_entry_safe(se, se_tmp, &dev->dev_sep_list, sep_list) {
+ tpg = se->sep_tpg;
+ sprintf(buf, "scsiTransport%s",
+ TPG_TFO(tpg)->get_fabric_name());
+
+ seq_printf(seq, "%u %s %u %s+%s\n",
+ hba->hba_index, /* scsiTransportIndex */
+ buf, /* scsiTransportType */
+ (TPG_TFO(tpg)->tpg_get_inst_index != NULL) ?
+ TPG_TFO(tpg)->tpg_get_inst_index(tpg) :
+ 0,
+ TPG_TFO(tpg)->tpg_get_wwn(tpg),
+ (strlen(wwn->unit_serial)) ?
+ /* scsiTransportDevName */
+ wwn->unit_serial : wwn->vendor);
+ }
+ spin_unlock(&dev->se_port_lock);
+
+ return 0;
+}
+
+static const struct seq_operations scsi_transport_seq_ops = {
+ .start = scsi_transport_seq_start,
+ .next = scsi_transport_seq_next,
+ .stop = scsi_transport_seq_stop,
+ .show = scsi_transport_seq_show
+};
+
+static int scsi_transport_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_transport_seq_ops);
+}
+
+static const struct file_operations scsi_transport_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_transport_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/*
+ * SCSI Target Device Table
+ */
+static void *scsi_tgt_dev_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ return locate_hba_start(seq, pos);
+}
+
+static void *scsi_tgt_dev_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ return locate_hba_next(seq, v, pos);
+}
+
+static void scsi_tgt_dev_seq_stop(struct seq_file *seq, void *v)
+{
+ locate_hba_stop(seq, v);
+}
+
+
+#define LU_COUNT 1 /* for now */
+static int scsi_tgt_dev_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_hba *hba;
+ struct se_subsystem_dev *se_dev = list_entry(v, struct se_subsystem_dev,
+ g_se_dev_list);
+ struct se_device *dev = se_dev->se_dev_ptr;
+ int non_accessible_lus = 0;
+ char status[16];
+
+ if (list_is_first(&se_dev->g_se_dev_list, &se_global->g_se_dev_list))
+ seq_puts(seq, "inst indx num_LUs status non_access_LUs"
+ " resets\n");
+
+ if (!(dev))
+ return 0;
+
+ hba = dev->se_hba;
+ if (!(hba)) {
+ /* Log error ? */
+ return 0;
+ }
+
+ switch (dev->dev_status) {
+ case TRANSPORT_DEVICE_ACTIVATED:
+ strcpy(status, "activated");
+ break;
+ case TRANSPORT_DEVICE_DEACTIVATED:
+ strcpy(status, "deactivated");
+ non_accessible_lus = 1;
+ break;
+ case TRANSPORT_DEVICE_SHUTDOWN:
+ strcpy(status, "shutdown");
+ non_accessible_lus = 1;
+ break;
+ case TRANSPORT_DEVICE_OFFLINE_ACTIVATED:
+ case TRANSPORT_DEVICE_OFFLINE_DEACTIVATED:
+ strcpy(status, "offline");
+ non_accessible_lus = 1;
+ break;
+ default:
+ sprintf(status, "unknown(%d)", dev->dev_status);
+ non_accessible_lus = 1;
+ }
+
+ seq_printf(seq, "%u %u %u %s %u %u\n",
+ hba->hba_index, dev->dev_index, LU_COUNT,
+ status, non_accessible_lus, dev->num_resets);
+
+ return 0;
+}
+
+static const struct seq_operations scsi_tgt_dev_seq_ops = {
+ .start = scsi_tgt_dev_seq_start,
+ .next = scsi_tgt_dev_seq_next,
+ .stop = scsi_tgt_dev_seq_stop,
+ .show = scsi_tgt_dev_seq_show
+};
+
+static int scsi_tgt_dev_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_tgt_dev_seq_ops);
+}
+
+static const struct file_operations scsi_tgt_dev_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_tgt_dev_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/*
+ * SCSI Target Port Table
+ */
+static void *scsi_tgt_port_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ return locate_hba_start(seq, pos);
+}
+
+static void *scsi_tgt_port_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ return locate_hba_next(seq, v, pos);
+}
+
+static void scsi_tgt_port_seq_stop(struct seq_file *seq, void *v)
+{
+ locate_hba_stop(seq, v);
+}
+
+static int scsi_tgt_port_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_hba *hba;
+ struct se_subsystem_dev *se_dev = list_entry(v, struct se_subsystem_dev,
+ g_se_dev_list);
+ struct se_device *dev = se_dev->se_dev_ptr;
+ struct se_port *sep, *sep_tmp;
+ struct se_portal_group *tpg;
+ u32 rx_mbytes, tx_mbytes;
+ unsigned long long num_cmds;
+ char buf[64];
+
+ if (list_is_first(&se_dev->g_se_dev_list, &se_global->g_se_dev_list))
+ seq_puts(seq, "inst device indx name port_index in_cmds"
+ " write_mbytes read_mbytes hs_in_cmds\n");
+
+ if (!(dev))
+ return 0;
+
+ hba = dev->se_hba;
+ if (!(hba)) {
+ /* Log error ? */
+ return 0;
+ }
+
+ spin_lock(&dev->se_port_lock);
+ list_for_each_entry_safe(sep, sep_tmp, &dev->dev_sep_list, sep_list) {
+ tpg = sep->sep_tpg;
+ sprintf(buf, "%sPort#",
+ TPG_TFO(tpg)->get_fabric_name());
+
+ seq_printf(seq, "%u %u %u %s%d %s%s%d ",
+ hba->hba_index,
+ dev->dev_index,
+ sep->sep_index,
+ buf, sep->sep_index,
+ TPG_TFO(tpg)->tpg_get_wwn(tpg), "+t+",
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+
+ spin_lock(&sep->sep_lun->lun_sep_lock);
+ num_cmds = sep->sep_stats.cmd_pdus;
+ rx_mbytes = (sep->sep_stats.rx_data_octets >> 20);
+ tx_mbytes = (sep->sep_stats.tx_data_octets >> 20);
+ spin_unlock(&sep->sep_lun->lun_sep_lock);
+
+ seq_printf(seq, "%llu %u %u %u\n", num_cmds,
+ rx_mbytes, tx_mbytes, 0);
+ }
+ spin_unlock(&dev->se_port_lock);
+
+ return 0;
+}
+
+static const struct seq_operations scsi_tgt_port_seq_ops = {
+ .start = scsi_tgt_port_seq_start,
+ .next = scsi_tgt_port_seq_next,
+ .stop = scsi_tgt_port_seq_stop,
+ .show = scsi_tgt_port_seq_show
+};
+
+static int scsi_tgt_port_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_tgt_port_seq_ops);
+}
+
+static const struct file_operations scsi_tgt_port_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_tgt_port_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/*
+ * SCSI Authorized Initiator Table:
+ * It contains the SCSI Initiators authorized to be attached to one of the
+ * local Target ports.
+ * Iterates through all active TPGs and extracts the info from the ACLs
+ */
+static void *scsi_auth_intr_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ spin_lock_bh(&se_global->se_tpg_lock);
+ return seq_list_start(&se_global->g_se_tpg_list, *pos);
+}
+
+static void *scsi_auth_intr_seq_next(struct seq_file *seq, void *v,
+ loff_t *pos)
+{
+ return seq_list_next(v, &se_global->g_se_tpg_list, pos);
+}
+
+static void scsi_auth_intr_seq_stop(struct seq_file *seq, void *v)
+{
+ spin_unlock_bh(&se_global->se_tpg_lock);
+}
+
+static int scsi_auth_intr_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_portal_group *se_tpg = list_entry(v, struct se_portal_group,
+ se_tpg_list);
+ struct se_dev_entry *deve;
+ struct se_lun *lun;
+ struct se_node_acl *se_nacl;
+ int j;
+
+ if (list_is_first(&se_tpg->se_tpg_list,
+ &se_global->g_se_tpg_list))
+ seq_puts(seq, "inst dev port indx dev_or_port intr_name "
+ "map_indx att_count num_cmds read_mbytes "
+ "write_mbytes hs_num_cmds creation_time row_status\n");
+
+ if (!(se_tpg))
+ return 0;
+
+ spin_lock(&se_tpg->acl_node_lock);
+ list_for_each_entry(se_nacl, &se_tpg->acl_node_list, acl_list) {
+
+ atomic_inc(&se_nacl->mib_ref_count);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&se_tpg->acl_node_lock);
+
+ spin_lock_irq(&se_nacl->device_list_lock);
+ for (j = 0; j < TRANSPORT_MAX_LUNS_PER_TPG; j++) {
+ deve = &se_nacl->device_list[j];
+ if (!(deve->lun_flags &
+ TRANSPORT_LUNFLAGS_INITIATOR_ACCESS) ||
+ (!deve->se_lun))
+ continue;
+ lun = deve->se_lun;
+ if (!lun->lun_se_dev)
+ continue;
+
+ seq_printf(seq, "%u %u %u %u %u %s %u %u %u %u %u %u"
+ " %u %s\n",
+ /* scsiInstIndex */
+ (TPG_TFO(se_tpg)->tpg_get_inst_index != NULL) ?
+ TPG_TFO(se_tpg)->tpg_get_inst_index(se_tpg) :
+ 0,
+ /* scsiDeviceIndex */
+ lun->lun_se_dev->dev_index,
+ /* scsiAuthIntrTgtPortIndex */
+ TPG_TFO(se_tpg)->tpg_get_tag(se_tpg),
+ /* scsiAuthIntrIndex */
+ se_nacl->acl_index,
+ /* scsiAuthIntrDevOrPort */
+ 1,
+ /* scsiAuthIntrName */
+ se_nacl->initiatorname[0] ?
+ se_nacl->initiatorname : NONE,
+ /* FIXME: scsiAuthIntrLunMapIndex */
+ 0,
+ /* scsiAuthIntrAttachedTimes */
+ deve->attach_count,
+ /* scsiAuthIntrOutCommands */
+ deve->total_cmds,
+ /* scsiAuthIntrReadMegaBytes */
+ (u32)(deve->read_bytes >> 20),
+ /* scsiAuthIntrWrittenMegaBytes */
+ (u32)(deve->write_bytes >> 20),
+ /* FIXME: scsiAuthIntrHSOutCommands */
+ 0,
+ /* scsiAuthIntrLastCreation */
+ (u32)(((u32)deve->creation_time -
+ INITIAL_JIFFIES) * 100 / HZ),
+ /* FIXME: scsiAuthIntrRowStatus */
+ "Ready");
+ }
+ spin_unlock_irq(&se_nacl->device_list_lock);
+
+ spin_lock(&se_tpg->acl_node_lock);
+ atomic_dec(&se_nacl->mib_ref_count);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&se_tpg->acl_node_lock);
+
+ return 0;
+}
+
+static const struct seq_operations scsi_auth_intr_seq_ops = {
+ .start = scsi_auth_intr_seq_start,
+ .next = scsi_auth_intr_seq_next,
+ .stop = scsi_auth_intr_seq_stop,
+ .show = scsi_auth_intr_seq_show
+};
+
+static int scsi_auth_intr_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_auth_intr_seq_ops);
+}
+
+static const struct file_operations scsi_auth_intr_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_auth_intr_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/*
+ * SCSI Attached Initiator Port Table:
+ * It lists the SCSI Initiators attached to one of the local Target ports.
+ * Iterates through all active TPGs and use active sessions from each TPG
+ * to list the info fo this table.
+ */
+static void *scsi_att_intr_port_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ spin_lock_bh(&se_global->se_tpg_lock);
+ return seq_list_start(&se_global->g_se_tpg_list, *pos);
+}
+
+static void *scsi_att_intr_port_seq_next(struct seq_file *seq, void *v,
+ loff_t *pos)
+{
+ return seq_list_next(v, &se_global->g_se_tpg_list, pos);
+}
+
+static void scsi_att_intr_port_seq_stop(struct seq_file *seq, void *v)
+{
+ spin_unlock_bh(&se_global->se_tpg_lock);
+}
+
+static int scsi_att_intr_port_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_portal_group *se_tpg = list_entry(v, struct se_portal_group,
+ se_tpg_list);
+ struct se_dev_entry *deve;
+ struct se_lun *lun;
+ struct se_node_acl *se_nacl;
+ struct se_session *se_sess;
+ unsigned char buf[64];
+ int j;
+
+ if (list_is_first(&se_tpg->se_tpg_list,
+ &se_global->g_se_tpg_list))
+ seq_puts(seq, "inst dev port indx port_auth_indx port_name"
+ " port_ident\n");
+
+ if (!(se_tpg))
+ return 0;
+
+ spin_lock(&se_tpg->session_lock);
+ list_for_each_entry(se_sess, &se_tpg->tpg_sess_list, sess_list) {
+ if ((TPG_TFO(se_tpg)->sess_logged_in(se_sess)) ||
+ (!se_sess->se_node_acl) ||
+ (!se_sess->se_node_acl->device_list))
+ continue;
+
+ atomic_inc(&se_sess->mib_ref_count);
+ smp_mb__after_atomic_inc();
+ se_nacl = se_sess->se_node_acl;
+ atomic_inc(&se_nacl->mib_ref_count);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&se_tpg->session_lock);
+
+ spin_lock_irq(&se_nacl->device_list_lock);
+ for (j = 0; j < TRANSPORT_MAX_LUNS_PER_TPG; j++) {
+ deve = &se_nacl->device_list[j];
+ if (!(deve->lun_flags &
+ TRANSPORT_LUNFLAGS_INITIATOR_ACCESS) ||
+ (!deve->se_lun))
+ continue;
+
+ lun = deve->se_lun;
+ if (!lun->lun_se_dev)
+ continue;
+
+ memset(buf, 0, 64);
+ if (TPG_TFO(se_tpg)->sess_get_initiator_sid != NULL)
+ TPG_TFO(se_tpg)->sess_get_initiator_sid(
+ se_sess, (unsigned char *)&buf[0], 64);
+
+ seq_printf(seq, "%u %u %u %u %u %s+i+%s\n",
+ /* scsiInstIndex */
+ (TPG_TFO(se_tpg)->tpg_get_inst_index != NULL) ?
+ TPG_TFO(se_tpg)->tpg_get_inst_index(se_tpg) :
+ 0,
+ /* scsiDeviceIndex */
+ lun->lun_se_dev->dev_index,
+ /* scsiPortIndex */
+ TPG_TFO(se_tpg)->tpg_get_tag(se_tpg),
+ /* scsiAttIntrPortIndex */
+ (TPG_TFO(se_tpg)->sess_get_index != NULL) ?
+ TPG_TFO(se_tpg)->sess_get_index(se_sess) :
+ 0,
+ /* scsiAttIntrPortAuthIntrIdx */
+ se_nacl->acl_index,
+ /* scsiAttIntrPortName */
+ se_nacl->initiatorname[0] ?
+ se_nacl->initiatorname : NONE,
+ /* scsiAttIntrPortIdentifier */
+ buf);
+ }
+ spin_unlock_irq(&se_nacl->device_list_lock);
+
+ spin_lock(&se_tpg->session_lock);
+ atomic_dec(&se_nacl->mib_ref_count);
+ smp_mb__after_atomic_dec();
+ atomic_dec(&se_sess->mib_ref_count);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&se_tpg->session_lock);
+
+ return 0;
+}
+
+static const struct seq_operations scsi_att_intr_port_seq_ops = {
+ .start = scsi_att_intr_port_seq_start,
+ .next = scsi_att_intr_port_seq_next,
+ .stop = scsi_att_intr_port_seq_stop,
+ .show = scsi_att_intr_port_seq_show
+};
+
+static int scsi_att_intr_port_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_att_intr_port_seq_ops);
+}
+
+static const struct file_operations scsi_att_intr_port_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_att_intr_port_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/*
+ * SCSI Logical Unit Table
+ */
+static void *scsi_lu_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ return locate_hba_start(seq, pos);
+}
+
+static void *scsi_lu_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ return locate_hba_next(seq, v, pos);
+}
+
+static void scsi_lu_seq_stop(struct seq_file *seq, void *v)
+{
+ locate_hba_stop(seq, v);
+}
+
+#define SCSI_LU_INDEX 1
+static int scsi_lu_seq_show(struct seq_file *seq, void *v)
+{
+ struct se_hba *hba;
+ struct se_subsystem_dev *se_dev = list_entry(v, struct se_subsystem_dev,
+ g_se_dev_list);
+ struct se_device *dev = se_dev->se_dev_ptr;
+ int j;
+ char str[28];
+
+ if (list_is_first(&se_dev->g_se_dev_list, &se_global->g_se_dev_list))
+ seq_puts(seq, "inst dev indx LUN lu_name vend prod rev"
+ " dev_type status state-bit num_cmds read_mbytes"
+ " write_mbytes resets full_stat hs_num_cmds creation_time\n");
+
+ if (!(dev))
+ return 0;
+
+ hba = dev->se_hba;
+ if (!(hba)) {
+ /* Log error ? */
+ return 0;
+ }
+
+ /* Fix LU state, if we can read it from the device */
+ seq_printf(seq, "%u %u %u %llu %s", hba->hba_index,
+ dev->dev_index, SCSI_LU_INDEX,
+ (unsigned long long)0, /* FIXME: scsiLuDefaultLun */
+ (strlen(DEV_T10_WWN(dev)->unit_serial)) ?
+ /* scsiLuWwnName */
+ (char *)&DEV_T10_WWN(dev)->unit_serial[0] :
+ "None");
+
+ memcpy(&str[0], (void *)DEV_T10_WWN(dev), 28);
+ /* scsiLuVendorId */
+ for (j = 0; j < 8; j++)
+ str[j] = ISPRINT(DEV_T10_WWN(dev)->vendor[j]) ?
+ DEV_T10_WWN(dev)->vendor[j] : 0x20;
+ str[8] = 0;
+ seq_printf(seq, " %s", str);
+
+ /* scsiLuProductId */
+ for (j = 0; j < 16; j++)
+ str[j] = ISPRINT(DEV_T10_WWN(dev)->model[j]) ?
+ DEV_T10_WWN(dev)->model[j] : 0x20;
+ str[16] = 0;
+ seq_printf(seq, " %s", str);
+
+ /* scsiLuRevisionId */
+ for (j = 0; j < 4; j++)
+ str[j] = ISPRINT(DEV_T10_WWN(dev)->revision[j]) ?
+ DEV_T10_WWN(dev)->revision[j] : 0x20;
+ str[4] = 0;
+ seq_printf(seq, " %s", str);
+
+ seq_printf(seq, " %u %s %s %llu %u %u %u %u %u %u\n",
+ /* scsiLuPeripheralType */
+ TRANSPORT(dev)->get_device_type(dev),
+ (dev->dev_status == TRANSPORT_DEVICE_ACTIVATED) ?
+ "available" : "notavailable", /* scsiLuStatus */
+ "exposed", /* scsiLuState */
+ (unsigned long long)dev->num_cmds,
+ /* scsiLuReadMegaBytes */
+ (u32)(dev->read_bytes >> 20),
+ /* scsiLuWrittenMegaBytes */
+ (u32)(dev->write_bytes >> 20),
+ dev->num_resets, /* scsiLuInResets */
+ 0, /* scsiLuOutTaskSetFullStatus */
+ 0, /* scsiLuHSInCommands */
+ (u32)(((u32)dev->creation_time - INITIAL_JIFFIES) *
+ 100 / HZ));
+
+ return 0;
+}
+
+static const struct seq_operations scsi_lu_seq_ops = {
+ .start = scsi_lu_seq_start,
+ .next = scsi_lu_seq_next,
+ .stop = scsi_lu_seq_stop,
+ .show = scsi_lu_seq_show
+};
+
+static int scsi_lu_seq_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &scsi_lu_seq_ops);
+}
+
+static const struct file_operations scsi_lu_seq_fops = {
+ .owner = THIS_MODULE,
+ .open = scsi_lu_seq_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+/****************************************************************************/
+
+/*
+ * Remove proc fs entries
+ */
+void remove_scsi_target_mib(void)
+{
+ remove_proc_entry("scsi_target/mib/scsi_inst", NULL);
+ remove_proc_entry("scsi_target/mib/scsi_dev", NULL);
+ remove_proc_entry("scsi_target/mib/scsi_port", NULL);
+ remove_proc_entry("scsi_target/mib/scsi_transport", NULL);
+ remove_proc_entry("scsi_target/mib/scsi_tgt_dev", NULL);
+ remove_proc_entry("scsi_target/mib/scsi_tgt_port", NULL);
+ remove_proc_entry("scsi_target/mib/scsi_auth_intr", NULL);
+ remove_proc_entry("scsi_target/mib/scsi_att_intr_port", NULL);
+ remove_proc_entry("scsi_target/mib/scsi_lu", NULL);
+ remove_proc_entry("scsi_target/mib", NULL);
+}
+
+/*
+ * Create proc fs entries for the mib tables
+ */
+int init_scsi_target_mib(void)
+{
+ struct proc_dir_entry *dir_entry;
+ struct proc_dir_entry *scsi_inst_entry;
+ struct proc_dir_entry *scsi_dev_entry;
+ struct proc_dir_entry *scsi_port_entry;
+ struct proc_dir_entry *scsi_transport_entry;
+ struct proc_dir_entry *scsi_tgt_dev_entry;
+ struct proc_dir_entry *scsi_tgt_port_entry;
+ struct proc_dir_entry *scsi_auth_intr_entry;
+ struct proc_dir_entry *scsi_att_intr_port_entry;
+ struct proc_dir_entry *scsi_lu_entry;
+
+ dir_entry = proc_mkdir("scsi_target/mib", NULL);
+ if (!(dir_entry)) {
+ printk(KERN_ERR "proc_mkdir() failed.\n");
+ return -1;
+ }
+
+ scsi_inst_entry =
+ create_proc_entry("scsi_target/mib/scsi_inst", 0, NULL);
+ if (scsi_inst_entry)
+ scsi_inst_entry->proc_fops = &scsi_inst_seq_fops;
+ else
+ goto error;
+
+ scsi_dev_entry =
+ create_proc_entry("scsi_target/mib/scsi_dev", 0, NULL);
+ if (scsi_dev_entry)
+ scsi_dev_entry->proc_fops = &scsi_dev_seq_fops;
+ else
+ goto error;
+
+ scsi_port_entry =
+ create_proc_entry("scsi_target/mib/scsi_port", 0, NULL);
+ if (scsi_port_entry)
+ scsi_port_entry->proc_fops = &scsi_port_seq_fops;
+ else
+ goto error;
+
+ scsi_transport_entry =
+ create_proc_entry("scsi_target/mib/scsi_transport", 0, NULL);
+ if (scsi_transport_entry)
+ scsi_transport_entry->proc_fops = &scsi_transport_seq_fops;
+ else
+ goto error;
+
+ scsi_tgt_dev_entry =
+ create_proc_entry("scsi_target/mib/scsi_tgt_dev", 0, NULL);
+ if (scsi_tgt_dev_entry)
+ scsi_tgt_dev_entry->proc_fops = &scsi_tgt_dev_seq_fops;
+ else
+ goto error;
+
+ scsi_tgt_port_entry =
+ create_proc_entry("scsi_target/mib/scsi_tgt_port", 0, NULL);
+ if (scsi_tgt_port_entry)
+ scsi_tgt_port_entry->proc_fops = &scsi_tgt_port_seq_fops;
+ else
+ goto error;
+
+ scsi_auth_intr_entry =
+ create_proc_entry("scsi_target/mib/scsi_auth_intr", 0, NULL);
+ if (scsi_auth_intr_entry)
+ scsi_auth_intr_entry->proc_fops = &scsi_auth_intr_seq_fops;
+ else
+ goto error;
+
+ scsi_att_intr_port_entry =
+ create_proc_entry("scsi_target/mib/scsi_att_intr_port", 0, NULL);
+ if (scsi_att_intr_port_entry)
+ scsi_att_intr_port_entry->proc_fops =
+ &scsi_att_intr_port_seq_fops;
+ else
+ goto error;
+
+ scsi_lu_entry = create_proc_entry("scsi_target/mib/scsi_lu", 0, NULL);
+ if (scsi_lu_entry)
+ scsi_lu_entry->proc_fops = &scsi_lu_seq_fops;
+ else
+ goto error;
+
+ return 0;
+
+error:
+ printk(KERN_ERR "create_proc_entry() failed.\n");
+ remove_scsi_target_mib();
+ return -1;
+}
+
+/*
+ * Initialize the index table for allocating unique row indexes to various mib
+ * tables
+ */
+void init_scsi_index_table(void)
+{
+ memset(&scsi_index_table, 0, sizeof(struct scsi_index_table));
+ spin_lock_init(&scsi_index_table.lock);
+}
+
+/*
+ * Allocate a new row index for the entry type specified
+ */
+u32 scsi_get_new_index(scsi_index_t type)
+{
+ u32 new_index;
+
+ if ((type < 0) || (type >= SCSI_INDEX_TYPE_MAX)) {
+ printk(KERN_ERR "Invalid index type %d\n", type);
+ return -1;
+ }
+
+ spin_lock(&scsi_index_table.lock);
+ new_index = ++scsi_index_table.scsi_mib_index[type];
+ if (new_index == 0)
+ new_index = ++scsi_index_table.scsi_mib_index[type];
+ spin_unlock(&scsi_index_table.lock);
+
+ return new_index;
+}
+EXPORT_SYMBOL(scsi_get_new_index);
diff --git a/drivers/target/target_core_mib.h b/drivers/target/target_core_mib.h
new file mode 100644
index 0000000..2772046
--- /dev/null
+++ b/drivers/target/target_core_mib.h
@@ -0,0 +1,28 @@
+#ifndef TARGET_CORE_MIB_H
+#define TARGET_CORE_MIB_H
+
+typedef enum {
+ SCSI_INST_INDEX,
+ SCSI_DEVICE_INDEX,
+ SCSI_AUTH_INTR_INDEX,
+ SCSI_INDEX_TYPE_MAX
+} scsi_index_t;
+
+struct scsi_index_table {
+ spinlock_t lock;
+ u32 scsi_mib_index[SCSI_INDEX_TYPE_MAX];
+} ____cacheline_aligned;
+
+/* SCSI Port stats */
+struct scsi_port_stats {
+ u64 cmd_pdus;
+ u64 tx_data_octets;
+ u64 rx_data_octets;
+} ____cacheline_aligned;
+
+extern int init_scsi_target_mib(void);
+extern void remove_scsi_target_mib(void);
+extern void init_scsi_index_table(void);
+extern u32 scsi_get_new_index(scsi_index_t);
+
+#endif /*** TARGET_CORE_MIB_H ***/
diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c
new file mode 100644
index 0000000..2521f75
--- /dev/null
+++ b/drivers/target/target_core_pr.c
@@ -0,0 +1,4252 @@
+/*******************************************************************************
+ * Filename: target_core_pr.c
+ *
+ * This file contains SPC-3 compliant persistent reservations and
+ * legacy SPC-2 reservations with compatible reservation handling (CRH=1)
+ *
+ * Copyright (c) 2009, 2010 Rising Tide Systems
+ * Copyright (c) 2009, 2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/list.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+#include <asm/unaligned.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_tmr.h>
+#include <target/target_core_tpg.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_configfs.h>
+
+#include "target_core_hba.h"
+#include "target_core_pr.h"
+#include "target_core_ua.h"
+
+/*
+ * Used for Specify Initiator Ports Capable Bit (SPEC_I_PT)
+ */
+struct pr_transport_id_holder {
+ int dest_local_nexus;
+ struct t10_pr_registration *dest_pr_reg;
+ struct se_portal_group *dest_tpg;
+ struct se_node_acl *dest_node_acl;
+ struct se_dev_entry *dest_se_deve;
+ struct list_head dest_list;
+};
+
+int core_pr_dump_initiator_port(
+ struct t10_pr_registration *pr_reg,
+ char *buf,
+ u32 size)
+{
+ if (!(pr_reg->isid_present_at_reg))
+ return 0;
+
+ snprintf(buf, size, ",i,0x%s", &pr_reg->pr_reg_isid[0]);
+ return 1;
+}
+
+static void __core_scsi3_complete_pro_release(struct se_device *, struct se_node_acl *,
+ struct t10_pr_registration *, int);
+
+static int core_scsi2_reservation_seq_non_holder(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ u32 pr_reg_type)
+{
+ switch (cdb[0]) {
+ case INQUIRY:
+ case RELEASE:
+ case RELEASE_10:
+ return 0;
+ default:
+ return 1;
+ }
+
+ return 1;
+}
+
+static int core_scsi2_reservation_check(struct se_cmd *cmd, u32 *pr_reg_type)
+{
+ struct se_device *dev = cmd->se_dev;
+ struct se_session *sess = cmd->se_sess;
+ int ret;
+
+ if (!(sess))
+ return 0;
+
+ spin_lock(&dev->dev_reservation_lock);
+ if (!dev->dev_reserved_node_acl || !sess) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return 0;
+ }
+ if (dev->dev_reserved_node_acl != sess->se_node_acl) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return -1;
+ }
+ if (!(dev->dev_flags & DF_SPC2_RESERVATIONS_WITH_ISID)) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return 0;
+ }
+ ret = (dev->dev_res_bin_isid == sess->sess_bin_isid) ? 0 : -1;
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return ret;
+}
+
+static int core_scsi2_reservation_release(struct se_cmd *cmd)
+{
+ struct se_device *dev = cmd->se_dev;
+ struct se_session *sess = cmd->se_sess;
+ struct se_portal_group *tpg = sess->se_tpg;
+
+ if (!(sess) || !(tpg))
+ return 0;
+
+ spin_lock(&dev->dev_reservation_lock);
+ if (!dev->dev_reserved_node_acl || !sess) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return 0;
+ }
+
+ if (dev->dev_reserved_node_acl != sess->se_node_acl) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return 0;
+ }
+ dev->dev_reserved_node_acl = NULL;
+ dev->dev_flags &= ~DF_SPC2_RESERVATIONS;
+ if (dev->dev_flags & DF_SPC2_RESERVATIONS_WITH_ISID) {
+ dev->dev_res_bin_isid = 0;
+ dev->dev_flags &= ~DF_SPC2_RESERVATIONS_WITH_ISID;
+ }
+ printk(KERN_INFO "SCSI-2 Released reservation for %s LUN: %u ->"
+ " MAPPED LUN: %u for %s\n", TPG_TFO(tpg)->get_fabric_name(),
+ SE_LUN(cmd)->unpacked_lun, cmd->se_deve->mapped_lun,
+ sess->se_node_acl->initiatorname);
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return 0;
+}
+
+static int core_scsi2_reservation_reserve(struct se_cmd *cmd)
+{
+ struct se_device *dev = cmd->se_dev;
+ struct se_session *sess = cmd->se_sess;
+ struct se_portal_group *tpg = sess->se_tpg;
+
+ if ((T_TASK(cmd)->t_task_cdb[1] & 0x01) &&
+ (T_TASK(cmd)->t_task_cdb[1] & 0x02)) {
+ printk(KERN_ERR "LongIO and Obselete Bits set, returning"
+ " ILLEGAL_REQUEST\n");
+ return PYX_TRANSPORT_ILLEGAL_REQUEST;
+ }
+ /*
+ * This is currently the case for target_core_mod passthrough struct se_cmd
+ * ops
+ */
+ if (!(sess) || !(tpg))
+ return 0;
+
+ spin_lock(&dev->dev_reservation_lock);
+ if (dev->dev_reserved_node_acl &&
+ (dev->dev_reserved_node_acl != sess->se_node_acl)) {
+ printk(KERN_ERR "SCSI-2 RESERVATION CONFLIFT for %s fabric\n",
+ TPG_TFO(tpg)->get_fabric_name());
+ printk(KERN_ERR "Original reserver LUN: %u %s\n",
+ SE_LUN(cmd)->unpacked_lun,
+ dev->dev_reserved_node_acl->initiatorname);
+ printk(KERN_ERR "Current attempt - LUN: %u -> MAPPED LUN: %u"
+ " from %s \n", SE_LUN(cmd)->unpacked_lun,
+ cmd->se_deve->mapped_lun,
+ sess->se_node_acl->initiatorname);
+ spin_unlock(&dev->dev_reservation_lock);
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+
+ dev->dev_reserved_node_acl = sess->se_node_acl;
+ dev->dev_flags |= DF_SPC2_RESERVATIONS;
+ if (sess->sess_bin_isid != 0) {
+ dev->dev_res_bin_isid = sess->sess_bin_isid;
+ dev->dev_flags |= DF_SPC2_RESERVATIONS_WITH_ISID;
+ }
+ printk(KERN_INFO "SCSI-2 Reserved %s LUN: %u -> MAPPED LUN: %u"
+ " for %s\n", TPG_TFO(tpg)->get_fabric_name(),
+ SE_LUN(cmd)->unpacked_lun, cmd->se_deve->mapped_lun,
+ sess->se_node_acl->initiatorname);
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return 0;
+}
+
+static struct t10_pr_registration *core_scsi3_locate_pr_reg(struct se_device *,
+ struct se_node_acl *, struct se_session *);
+static void core_scsi3_put_pr_reg(struct t10_pr_registration *);
+
+/*
+ * Setup in target_core_transport.c:transport_generic_cmd_sequencer()
+ * and called via struct se_cmd->transport_emulate_cdb() in TCM processing
+ * thread context.
+ */
+int core_scsi2_emulate_crh(struct se_cmd *cmd)
+{
+ struct se_session *se_sess = cmd->se_sess;
+ struct se_subsystem_dev *su_dev = cmd->se_dev->se_sub_dev;
+ struct t10_pr_registration *pr_reg;
+ struct t10_reservation_template *pr_tmpl = &su_dev->t10_reservation;
+ unsigned char *cdb = &T_TASK(cmd)->t_task_cdb[0];
+ int crh = (T10_RES(su_dev)->res_type == SPC3_PERSISTENT_RESERVATIONS);
+ int conflict = 0;
+
+ if (!(se_sess))
+ return 0;
+
+ if (!(crh))
+ goto after_crh;
+
+ pr_reg = core_scsi3_locate_pr_reg(cmd->se_dev, se_sess->se_node_acl,
+ se_sess);
+ if (pr_reg) {
+ /*
+ * From spc4r17 5.7.3 Exceptions to SPC-2 RESERVE and RELEASE
+ * behavior
+ *
+ * A RESERVE(6) or RESERVE(10) command shall complete with GOOD
+ * status, but no reservation shall be established and the
+ * persistent reservation shall not be changed, if the command
+ * is received from a) and b) below.
+ *
+ * A RELEASE(6) or RELEASE(10) command shall complete with GOOD
+ * status, but the persistent reservation shall not be released,
+ * if the command is received from a) and b)
+ *
+ * a) An I_T nexus that is a persistent reservation holder; or
+ * b) An I_T nexus that is registered if a registrants only or
+ * all registrants type persistent reservation is present.
+ *
+ * In all other cases, a RESERVE(6) command, RESERVE(10) command,
+ * RELEASE(6) command, or RELEASE(10) command shall be processed
+ * as defined in SPC-2.
+ */
+ if (pr_reg->pr_res_holder) {
+ core_scsi3_put_pr_reg(pr_reg);
+ return 0;
+ }
+ if ((pr_reg->pr_res_type == PR_TYPE_WRITE_EXCLUSIVE_REGONLY) ||
+ (pr_reg->pr_res_type == PR_TYPE_EXCLUSIVE_ACCESS_REGONLY) ||
+ (pr_reg->pr_res_type == PR_TYPE_WRITE_EXCLUSIVE_ALLREG) ||
+ (pr_reg->pr_res_type == PR_TYPE_EXCLUSIVE_ACCESS_ALLREG)) {
+ core_scsi3_put_pr_reg(pr_reg);
+ return 0;
+ }
+ core_scsi3_put_pr_reg(pr_reg);
+ conflict = 1;
+ } else {
+ /*
+ * Following spc2r20 5.5.1 Reservations overview:
+ *
+ * If a logical unit has executed a PERSISTENT RESERVE OUT
+ * command with the REGISTER or the REGISTER AND IGNORE
+ * EXISTING KEY service action and is still registered by any
+ * initiator, all RESERVE commands and all RELEASE commands
+ * regardless of initiator shall conflict and shall terminate
+ * with a RESERVATION CONFLICT status.
+ */
+ spin_lock(&pr_tmpl->registration_lock);
+ conflict = (list_empty(&pr_tmpl->registration_list)) ? 0 : 1;
+ spin_unlock(&pr_tmpl->registration_lock);
+ }
+
+ if (conflict) {
+ printk(KERN_ERR "Received legacy SPC-2 RESERVE/RELEASE"
+ " while active SPC-3 registrations exist,"
+ " returning RESERVATION_CONFLICT\n");
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+
+after_crh:
+ if ((cdb[0] == RESERVE) || (cdb[0] == RESERVE_10))
+ return core_scsi2_reservation_reserve(cmd);
+ else if ((cdb[0] == RELEASE) || (cdb[0] == RELEASE_10))
+ return core_scsi2_reservation_release(cmd);
+ else
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+}
+
+/*
+ * Begin SPC-3/SPC-4 Persistent Reservations emulation support
+ *
+ * This function is called by those initiator ports who are *NOT*
+ * the active PR reservation holder when a reservation is present.
+ */
+static int core_scsi3_pr_seq_non_holder(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ u32 pr_reg_type)
+{
+ struct se_dev_entry *se_deve;
+ struct se_session *se_sess = SE_SESS(cmd);
+ int other_cdb = 0, ignore_reg;
+ int registered_nexus = 0, ret = 1; /* Conflict by default */
+ int all_reg = 0, reg_only = 0; /* ALL_REG, REG_ONLY */
+ int we = 0; /* Write Exclusive */
+ int legacy = 0; /* Act like a legacy device and return
+ * RESERVATION CONFLICT on some CDBs */
+ /*
+ * A legacy SPC-2 reservation is being held.
+ */
+ if (cmd->se_dev->dev_flags & DF_SPC2_RESERVATIONS)
+ return core_scsi2_reservation_seq_non_holder(cmd,
+ cdb, pr_reg_type);
+
+ se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
+ /*
+ * Determine if the registration should be ignored due to
+ * non-matching ISIDs in core_scsi3_pr_reservation_check().
+ */
+ ignore_reg = (pr_reg_type & 0x80000000);
+ if (ignore_reg)
+ pr_reg_type &= ~0x80000000;
+
+ switch (pr_reg_type) {
+ case PR_TYPE_WRITE_EXCLUSIVE:
+ we = 1;
+ case PR_TYPE_EXCLUSIVE_ACCESS:
+ /*
+ * Some commands are only allowed for the persistent reservation
+ * holder.
+ */
+ if ((se_deve->def_pr_registered) && !(ignore_reg))
+ registered_nexus = 1;
+ break;
+ case PR_TYPE_WRITE_EXCLUSIVE_REGONLY:
+ we = 1;
+ case PR_TYPE_EXCLUSIVE_ACCESS_REGONLY:
+ /*
+ * Some commands are only allowed for registered I_T Nexuses.
+ */
+ reg_only = 1;
+ if ((se_deve->def_pr_registered) && !(ignore_reg))
+ registered_nexus = 1;
+ break;
+ case PR_TYPE_WRITE_EXCLUSIVE_ALLREG:
+ we = 1;
+ case PR_TYPE_EXCLUSIVE_ACCESS_ALLREG:
+ /*
+ * Each registered I_T Nexus is a reservation holder.
+ */
+ all_reg = 1;
+ if ((se_deve->def_pr_registered) && !(ignore_reg))
+ registered_nexus = 1;
+ break;
+ default:
+ return -1;
+ }
+ /*
+ * Referenced from spc4r17 table 45 for *NON* PR holder access
+ */
+ switch (cdb[0]) {
+ case SECURITY_PROTOCOL_IN:
+ if (registered_nexus)
+ return 0;
+ ret = (we) ? 0 : 1;
+ break;
+ case MODE_SENSE:
+ case MODE_SENSE_10:
+ case READ_ATTRIBUTE:
+ case READ_BUFFER:
+ case RECEIVE_DIAGNOSTIC:
+ if (legacy) {
+ ret = 1;
+ break;
+ }
+ if (registered_nexus) {
+ ret = 0;
+ break;
+ }
+ ret = (we) ? 0 : 1; /* Allowed Write Exclusive */
+ break;
+ case PERSISTENT_RESERVE_OUT:
+ /*
+ * This follows PERSISTENT_RESERVE_OUT service actions that
+ * are allowed in the presence of various reservations.
+ * See spc4r17, table 46
+ */
+ switch (cdb[1] & 0x1f) {
+ case PRO_CLEAR:
+ case PRO_PREEMPT:
+ case PRO_PREEMPT_AND_ABORT:
+ ret = (registered_nexus) ? 0 : 1;
+ break;
+ case PRO_REGISTER:
+ case PRO_REGISTER_AND_IGNORE_EXISTING_KEY:
+ ret = 0;
+ break;
+ case PRO_REGISTER_AND_MOVE:
+ case PRO_RESERVE:
+ ret = 1;
+ break;
+ case PRO_RELEASE:
+ ret = (registered_nexus) ? 0 : 1;
+ break;
+ default:
+ printk(KERN_ERR "Unknown PERSISTENT_RESERVE_OUT service"
+ " action: 0x%02x\n", cdb[1] & 0x1f);
+ return -1;
+ }
+ break;
+ case RELEASE:
+ case RELEASE_10:
+ /* Handled by CRH=1 in core_scsi2_emulate_crh() */
+ ret = 0;
+ break;
+ case RESERVE:
+ case RESERVE_10:
+ /* Handled by CRH=1 in core_scsi2_emulate_crh() */
+ ret = 0;
+ break;
+ case TEST_UNIT_READY:
+ ret = (legacy) ? 1 : 0; /* Conflict for legacy */
+ break;
+ case MAINTENANCE_IN:
+ switch (cdb[1] & 0x1f) {
+ case MI_MANAGEMENT_PROTOCOL_IN:
+ if (registered_nexus) {
+ ret = 0;
+ break;
+ }
+ ret = (we) ? 0 : 1; /* Allowed Write Exclusive */
+ break;
+ case MI_REPORT_SUPPORTED_OPERATION_CODES:
+ case MI_REPORT_SUPPORTED_TASK_MANAGEMENT_FUNCTIONS:
+ if (legacy) {
+ ret = 1;
+ break;
+ }
+ if (registered_nexus) {
+ ret = 0;
+ break;
+ }
+ ret = (we) ? 0 : 1; /* Allowed Write Exclusive */
+ break;
+ case MI_REPORT_ALIASES:
+ case MI_REPORT_IDENTIFYING_INFORMATION:
+ case MI_REPORT_PRIORITY:
+ case MI_REPORT_TARGET_PGS:
+ case MI_REPORT_TIMESTAMP:
+ ret = 0; /* Allowed */
+ break;
+ default:
+ printk(KERN_ERR "Unknown MI Service Action: 0x%02x\n",
+ (cdb[1] & 0x1f));
+ return -1;
+ }
+ break;
+ case ACCESS_CONTROL_IN:
+ case ACCESS_CONTROL_OUT:
+ case INQUIRY:
+ case LOG_SENSE:
+ case READ_MEDIA_SERIAL_NUMBER:
+ case REPORT_LUNS:
+ case REQUEST_SENSE:
+ ret = 0; /*/ Allowed CDBs */
+ break;
+ default:
+ other_cdb = 1;
+ break;
+ }
+ /*
+ * Case where the CDB is explictly allowed in the above switch
+ * statement.
+ */
+ if (!(ret) && !(other_cdb)) {
+#if 0
+ printk(KERN_INFO "Allowing explict CDB: 0x%02x for %s"
+ " reservation holder\n", cdb[0],
+ core_scsi3_pr_dump_type(pr_reg_type));
+#endif
+ return ret;
+ }
+ /*
+ * Check if write exclusive initiator ports *NOT* holding the
+ * WRITE_EXCLUSIVE_* reservation.
+ */
+ if ((we) && !(registered_nexus)) {
+ if (cmd->data_direction == DMA_TO_DEVICE) {
+ /*
+ * Conflict for write exclusive
+ */
+ printk(KERN_INFO "%s Conflict for unregistered nexus"
+ " %s CDB: 0x%02x to %s reservation\n",
+ transport_dump_cmd_direction(cmd),
+ se_sess->se_node_acl->initiatorname, cdb[0],
+ core_scsi3_pr_dump_type(pr_reg_type));
+ return 1;
+ } else {
+ /*
+ * Allow non WRITE CDBs for all Write Exclusive
+ * PR TYPEs to pass for registered and
+ * non-registered_nexuxes NOT holding the reservation.
+ *
+ * We only make noise for the unregisterd nexuses,
+ * as we expect registered non-reservation holding
+ * nexuses to issue CDBs.
+ */
+#if 0
+ if (!(registered_nexus)) {
+ printk(KERN_INFO "Allowing implict CDB: 0x%02x"
+ " for %s reservation on unregistered"
+ " nexus\n", cdb[0],
+ core_scsi3_pr_dump_type(pr_reg_type));
+ }
+#endif
+ return 0;
+ }
+ } else if ((reg_only) || (all_reg)) {
+ if (registered_nexus) {
+ /*
+ * For PR_*_REG_ONLY and PR_*_ALL_REG reservations,
+ * allow commands from registered nexuses.
+ */
+#if 0
+ printk(KERN_INFO "Allowing implict CDB: 0x%02x for %s"
+ " reservation\n", cdb[0],
+ core_scsi3_pr_dump_type(pr_reg_type));
+#endif
+ return 0;
+ }
+ }
+ printk(KERN_INFO "%s Conflict for %sregistered nexus %s CDB: 0x%2x"
+ " for %s reservation\n", transport_dump_cmd_direction(cmd),
+ (registered_nexus) ? "" : "un",
+ se_sess->se_node_acl->initiatorname, cdb[0],
+ core_scsi3_pr_dump_type(pr_reg_type));
+
+ return 1; /* Conflict by default */
+}
+
+static u32 core_scsi3_pr_generation(struct se_device *dev)
+{
+ struct se_subsystem_dev *su_dev = SU_DEV(dev);
+ u32 prg;
+ /*
+ * PRGeneration field shall contain the value of a 32-bit wrapping
+ * counter mainted by the device server.
+ *
+ * Note that this is done regardless of Active Persist across
+ * Target PowerLoss (APTPL)
+ *
+ * See spc4r17 section 6.3.12 READ_KEYS service action
+ */
+ spin_lock(&dev->dev_reservation_lock);
+ prg = T10_RES(su_dev)->pr_generation++;
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return prg;
+}
+
+static int core_scsi3_pr_reservation_check(
+ struct se_cmd *cmd,
+ u32 *pr_reg_type)
+{
+ struct se_device *dev = cmd->se_dev;
+ struct se_session *sess = cmd->se_sess;
+ int ret;
+
+ if (!(sess))
+ return 0;
+ /*
+ * A legacy SPC-2 reservation is being held.
+ */
+ if (dev->dev_flags & DF_SPC2_RESERVATIONS)
+ return core_scsi2_reservation_check(cmd, pr_reg_type);
+
+ spin_lock(&dev->dev_reservation_lock);
+ if (!(dev->dev_pr_res_holder)) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return 0;
+ }
+ *pr_reg_type = dev->dev_pr_res_holder->pr_res_type;
+ cmd->pr_res_key = dev->dev_pr_res_holder->pr_res_key;
+ if (dev->dev_pr_res_holder->pr_reg_nacl != sess->se_node_acl) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return -1;
+ }
+ if (!(dev->dev_pr_res_holder->isid_present_at_reg)) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return 0;
+ }
+ ret = (dev->dev_pr_res_holder->pr_reg_bin_isid ==
+ sess->sess_bin_isid) ? 0 : -1;
+ /*
+ * Use bit in *pr_reg_type to notify ISID mismatch in
+ * core_scsi3_pr_seq_non_holder().
+ */
+ if (ret != 0)
+ *pr_reg_type |= 0x80000000;
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return ret;
+}
+
+static struct t10_pr_registration *__core_scsi3_do_alloc_registration(
+ struct se_device *dev,
+ struct se_node_acl *nacl,
+ struct se_dev_entry *deve,
+ unsigned char *isid,
+ u64 sa_res_key,
+ int all_tg_pt,
+ int aptpl)
+{
+ struct se_subsystem_dev *su_dev = SU_DEV(dev);
+ struct t10_pr_registration *pr_reg;
+
+ pr_reg = kmem_cache_zalloc(t10_pr_reg_cache, GFP_ATOMIC);
+ if (!(pr_reg)) {
+ printk(KERN_ERR "Unable to allocate struct t10_pr_registration\n");
+ return NULL;
+ }
+
+ pr_reg->pr_aptpl_buf = kzalloc(T10_RES(su_dev)->pr_aptpl_buf_len,
+ GFP_ATOMIC);
+ if (!(pr_reg->pr_aptpl_buf)) {
+ printk(KERN_ERR "Unable to allocate pr_reg->pr_aptpl_buf\n");
+ kmem_cache_free(t10_pr_reg_cache, pr_reg);
+ return NULL;
+ }
+
+ INIT_LIST_HEAD(&pr_reg->pr_reg_list);
+ INIT_LIST_HEAD(&pr_reg->pr_reg_abort_list);
+ INIT_LIST_HEAD(&pr_reg->pr_reg_aptpl_list);
+ INIT_LIST_HEAD(&pr_reg->pr_reg_atp_list);
+ INIT_LIST_HEAD(&pr_reg->pr_reg_atp_mem_list);
+ atomic_set(&pr_reg->pr_res_holders, 0);
+ pr_reg->pr_reg_nacl = nacl;
+ pr_reg->pr_reg_deve = deve;
+ pr_reg->pr_res_mapped_lun = deve->mapped_lun;
+ pr_reg->pr_aptpl_target_lun = deve->se_lun->unpacked_lun;
+ pr_reg->pr_res_key = sa_res_key;
+ pr_reg->pr_reg_all_tg_pt = all_tg_pt;
+ pr_reg->pr_reg_aptpl = aptpl;
+ pr_reg->pr_reg_tg_pt_lun = deve->se_lun;
+ /*
+ * If an ISID value for this SCSI Initiator Port exists,
+ * save it to the registration now.
+ */
+ if (isid != NULL) {
+ pr_reg->pr_reg_bin_isid = get_unaligned_be64(isid);
+ snprintf(pr_reg->pr_reg_isid, PR_REG_ISID_LEN, "%s", isid);
+ pr_reg->isid_present_at_reg = 1;
+ }
+
+ return pr_reg;
+}
+
+static int core_scsi3_lunacl_depend_item(struct se_dev_entry *);
+static void core_scsi3_lunacl_undepend_item(struct se_dev_entry *);
+
+/*
+ * Function used for handling PR registrations for ALL_TG_PT=1 and ALL_TG_PT=0
+ * modes.
+ */
+static struct t10_pr_registration *__core_scsi3_alloc_registration(
+ struct se_device *dev,
+ struct se_node_acl *nacl,
+ struct se_dev_entry *deve,
+ unsigned char *isid,
+ u64 sa_res_key,
+ int all_tg_pt,
+ int aptpl)
+{
+ struct se_dev_entry *deve_tmp;
+ struct se_node_acl *nacl_tmp;
+ struct se_port *port, *port_tmp;
+ struct target_core_fabric_ops *tfo = nacl->se_tpg->se_tpg_tfo;
+ struct t10_pr_registration *pr_reg, *pr_reg_atp, *pr_reg_tmp, *pr_reg_tmp_safe;
+ int ret;
+ /*
+ * Create a registration for the I_T Nexus upon which the
+ * PROUT REGISTER was received.
+ */
+ pr_reg = __core_scsi3_do_alloc_registration(dev, nacl, deve, isid,
+ sa_res_key, all_tg_pt, aptpl);
+ if (!(pr_reg))
+ return NULL;
+ /*
+ * Return pointer to pr_reg for ALL_TG_PT=0
+ */
+ if (!(all_tg_pt))
+ return pr_reg;
+ /*
+ * Create list of matching SCSI Initiator Port registrations
+ * for ALL_TG_PT=1
+ */
+ spin_lock(&dev->se_port_lock);
+ list_for_each_entry_safe(port, port_tmp, &dev->dev_sep_list, sep_list) {
+ atomic_inc(&port->sep_tg_pt_ref_cnt);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&dev->se_port_lock);
+
+ spin_lock_bh(&port->sep_alua_lock);
+ list_for_each_entry(deve_tmp, &port->sep_alua_list,
+ alua_port_list) {
+ /*
+ * This pointer will be NULL for demo mode MappedLUNs
+ * that have not been make explict via a ConfigFS
+ * MappedLUN group for the SCSI Initiator Node ACL.
+ */
+ if (!(deve_tmp->se_lun_acl))
+ continue;
+
+ nacl_tmp = deve_tmp->se_lun_acl->se_lun_nacl;
+ /*
+ * Skip the matching struct se_node_acl that is allocated
+ * above..
+ */
+ if (nacl == nacl_tmp)
+ continue;
+ /*
+ * Only perform PR registrations for target ports on
+ * the same fabric module as the REGISTER w/ ALL_TG_PT=1
+ * arrived.
+ */
+ if (tfo != nacl_tmp->se_tpg->se_tpg_tfo)
+ continue;
+ /*
+ * Look for a matching Initiator Node ACL in ASCII format
+ */
+ if (strcmp(nacl->initiatorname, nacl_tmp->initiatorname))
+ continue;
+
+ atomic_inc(&deve_tmp->pr_ref_count);
+ smp_mb__after_atomic_inc();
+ spin_unlock_bh(&port->sep_alua_lock);
+ /*
+ * Grab a configfs group dependency that is released
+ * for the exception path at label out: below, or upon
+ * completion of adding ALL_TG_PT=1 registrations in
+ * __core_scsi3_add_registration()
+ */
+ ret = core_scsi3_lunacl_depend_item(deve_tmp);
+ if (ret < 0) {
+ printk(KERN_ERR "core_scsi3_lunacl_depend"
+ "_item() failed\n");
+ atomic_dec(&port->sep_tg_pt_ref_cnt);
+ smp_mb__after_atomic_dec();
+ atomic_dec(&deve_tmp->pr_ref_count);
+ smp_mb__after_atomic_dec();
+ goto out;
+ }
+ /*
+ * Located a matching SCSI Initiator Port on a different
+ * port, allocate the pr_reg_atp and attach it to the
+ * pr_reg->pr_reg_atp_list that will be processed once
+ * the original *pr_reg is processed in
+ * __core_scsi3_add_registration()
+ */
+ pr_reg_atp = __core_scsi3_do_alloc_registration(dev,
+ nacl_tmp, deve_tmp, NULL,
+ sa_res_key, all_tg_pt, aptpl);
+ if (!(pr_reg_atp)) {
+ atomic_dec(&port->sep_tg_pt_ref_cnt);
+ smp_mb__after_atomic_dec();
+ atomic_dec(&deve_tmp->pr_ref_count);
+ smp_mb__after_atomic_dec();
+ core_scsi3_lunacl_undepend_item(deve_tmp);
+ goto out;
+ }
+
+ list_add_tail(&pr_reg_atp->pr_reg_atp_mem_list,
+ &pr_reg->pr_reg_atp_list);
+ spin_lock_bh(&port->sep_alua_lock);
+ }
+ spin_unlock_bh(&port->sep_alua_lock);
+
+ spin_lock(&dev->se_port_lock);
+ atomic_dec(&port->sep_tg_pt_ref_cnt);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&dev->se_port_lock);
+
+ return pr_reg;
+out:
+ list_for_each_entry_safe(pr_reg_tmp, pr_reg_tmp_safe,
+ &pr_reg->pr_reg_atp_list, pr_reg_atp_mem_list) {
+ list_del(&pr_reg_tmp->pr_reg_atp_mem_list);
+ core_scsi3_lunacl_undepend_item(pr_reg_tmp->pr_reg_deve);
+ kmem_cache_free(t10_pr_reg_cache, pr_reg_tmp);
+ }
+ kmem_cache_free(t10_pr_reg_cache, pr_reg);
+ return NULL;
+}
+
+int core_scsi3_alloc_aptpl_registration(
+ struct t10_reservation_template *pr_tmpl,
+ u64 sa_res_key,
+ unsigned char *i_port,
+ unsigned char *isid,
+ u32 mapped_lun,
+ unsigned char *t_port,
+ u16 tpgt,
+ u32 target_lun,
+ int res_holder,
+ int all_tg_pt,
+ u8 type)
+{
+ struct t10_pr_registration *pr_reg;
+
+ if (!(i_port) || !(t_port) || !(sa_res_key)) {
+ printk(KERN_ERR "Illegal parameters for APTPL registration\n");
+ return -1;
+ }
+
+ pr_reg = kmem_cache_zalloc(t10_pr_reg_cache, GFP_KERNEL);
+ if (!(pr_reg)) {
+ printk(KERN_ERR "Unable to allocate struct t10_pr_registration\n");
+ return -1;
+ }
+ pr_reg->pr_aptpl_buf = kzalloc(pr_tmpl->pr_aptpl_buf_len, GFP_KERNEL);
+
+ INIT_LIST_HEAD(&pr_reg->pr_reg_list);
+ INIT_LIST_HEAD(&pr_reg->pr_reg_abort_list);
+ INIT_LIST_HEAD(&pr_reg->pr_reg_aptpl_list);
+ INIT_LIST_HEAD(&pr_reg->pr_reg_atp_list);
+ INIT_LIST_HEAD(&pr_reg->pr_reg_atp_mem_list);
+ atomic_set(&pr_reg->pr_res_holders, 0);
+ pr_reg->pr_reg_nacl = NULL;
+ pr_reg->pr_reg_deve = NULL;
+ pr_reg->pr_res_mapped_lun = mapped_lun;
+ pr_reg->pr_aptpl_target_lun = target_lun;
+ pr_reg->pr_res_key = sa_res_key;
+ pr_reg->pr_reg_all_tg_pt = all_tg_pt;
+ pr_reg->pr_reg_aptpl = 1;
+ pr_reg->pr_reg_tg_pt_lun = NULL;
+ pr_reg->pr_res_scope = 0; /* Always LUN_SCOPE */
+ pr_reg->pr_res_type = type;
+ /*
+ * If an ISID value had been saved in APTPL metadata for this
+ * SCSI Initiator Port, restore it now.
+ */
+ if (isid != NULL) {
+ pr_reg->pr_reg_bin_isid = get_unaligned_be64(isid);
+ snprintf(pr_reg->pr_reg_isid, PR_REG_ISID_LEN, "%s", isid);
+ pr_reg->isid_present_at_reg = 1;
+ }
+ /*
+ * Copy the i_port and t_port information from caller.
+ */
+ snprintf(pr_reg->pr_iport, PR_APTPL_MAX_IPORT_LEN, "%s", i_port);
+ snprintf(pr_reg->pr_tport, PR_APTPL_MAX_TPORT_LEN, "%s", t_port);
+ pr_reg->pr_reg_tpgt = tpgt;
+ /*
+ * Set pr_res_holder from caller, the pr_reg who is the reservation
+ * holder will get it's pointer set in core_scsi3_aptpl_reserve() once
+ * the Initiator Node LUN ACL from the fabric module is created for
+ * this registration.
+ */
+ pr_reg->pr_res_holder = res_holder;
+
+ list_add_tail(&pr_reg->pr_reg_aptpl_list, &pr_tmpl->aptpl_reg_list);
+ printk(KERN_INFO "SPC-3 PR APTPL Successfully added registration%s from"
+ " metadata\n", (res_holder) ? "+reservation" : "");
+ return 0;
+}
+
+static void core_scsi3_aptpl_reserve(
+ struct se_device *dev,
+ struct se_portal_group *tpg,
+ struct se_node_acl *node_acl,
+ struct t10_pr_registration *pr_reg)
+{
+ char i_buf[PR_REG_ISID_ID_LEN];
+ int prf_isid;
+
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+
+ spin_lock(&dev->dev_reservation_lock);
+ dev->dev_pr_res_holder = pr_reg;
+ spin_unlock(&dev->dev_reservation_lock);
+
+ printk(KERN_INFO "SPC-3 PR [%s] Service Action: APTPL RESERVE created"
+ " new reservation holder TYPE: %s ALL_TG_PT: %d\n",
+ TPG_TFO(tpg)->get_fabric_name(),
+ core_scsi3_pr_dump_type(pr_reg->pr_res_type),
+ (pr_reg->pr_reg_all_tg_pt) ? 1 : 0);
+ printk(KERN_INFO "SPC-3 PR [%s] RESERVE Node: %s%s\n",
+ TPG_TFO(tpg)->get_fabric_name(), node_acl->initiatorname,
+ (prf_isid) ? &i_buf[0] : "");
+}
+
+static void __core_scsi3_add_registration(struct se_device *, struct se_node_acl *,
+ struct t10_pr_registration *, int, int);
+
+static int __core_scsi3_check_aptpl_registration(
+ struct se_device *dev,
+ struct se_portal_group *tpg,
+ struct se_lun *lun,
+ u32 target_lun,
+ struct se_node_acl *nacl,
+ struct se_dev_entry *deve)
+{
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ unsigned char i_port[PR_APTPL_MAX_IPORT_LEN];
+ unsigned char t_port[PR_APTPL_MAX_TPORT_LEN];
+ u16 tpgt;
+
+ memset(i_port, 0, PR_APTPL_MAX_IPORT_LEN);
+ memset(t_port, 0, PR_APTPL_MAX_TPORT_LEN);
+ /*
+ * Copy Initiator Port information from struct se_node_acl
+ */
+ snprintf(i_port, PR_APTPL_MAX_IPORT_LEN, "%s", nacl->initiatorname);
+ snprintf(t_port, PR_APTPL_MAX_TPORT_LEN, "%s",
+ TPG_TFO(tpg)->tpg_get_wwn(tpg));
+ tpgt = TPG_TFO(tpg)->tpg_get_tag(tpg);
+ /*
+ * Look for the matching registrations+reservation from those
+ * created from APTPL metadata. Note that multiple registrations
+ * may exist for fabrics that use ISIDs in their SCSI Initiator Port
+ * TransportIDs.
+ */
+ spin_lock(&pr_tmpl->aptpl_reg_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp, &pr_tmpl->aptpl_reg_list,
+ pr_reg_aptpl_list) {
+ if (!(strcmp(pr_reg->pr_iport, i_port)) &&
+ (pr_reg->pr_res_mapped_lun == deve->mapped_lun) &&
+ !(strcmp(pr_reg->pr_tport, t_port)) &&
+ (pr_reg->pr_reg_tpgt == tpgt) &&
+ (pr_reg->pr_aptpl_target_lun == target_lun)) {
+
+ pr_reg->pr_reg_nacl = nacl;
+ pr_reg->pr_reg_deve = deve;
+ pr_reg->pr_reg_tg_pt_lun = lun;
+
+ list_del(&pr_reg->pr_reg_aptpl_list);
+ spin_unlock(&pr_tmpl->aptpl_reg_lock);
+ /*
+ * At this point all of the pointers in *pr_reg will
+ * be setup, so go ahead and add the registration.
+ */
+
+ __core_scsi3_add_registration(dev, nacl, pr_reg, 0, 0);
+ /*
+ * If this registration is the reservation holder,
+ * make that happen now..
+ */
+ if (pr_reg->pr_res_holder)
+ core_scsi3_aptpl_reserve(dev, tpg,
+ nacl, pr_reg);
+ /*
+ * Reenable pr_aptpl_active to accept new metadata
+ * updates once the SCSI device is active again..
+ */
+ spin_lock(&pr_tmpl->aptpl_reg_lock);
+ pr_tmpl->pr_aptpl_active = 1;
+ }
+ }
+ spin_unlock(&pr_tmpl->aptpl_reg_lock);
+
+ return 0;
+}
+
+int core_scsi3_check_aptpl_registration(
+ struct se_device *dev,
+ struct se_portal_group *tpg,
+ struct se_lun *lun,
+ struct se_lun_acl *lun_acl)
+{
+ struct se_subsystem_dev *su_dev = SU_DEV(dev);
+ struct se_node_acl *nacl = lun_acl->se_lun_nacl;
+ struct se_dev_entry *deve = &nacl->device_list[lun_acl->mapped_lun];
+
+ if (T10_RES(su_dev)->res_type != SPC3_PERSISTENT_RESERVATIONS)
+ return 0;
+
+ return __core_scsi3_check_aptpl_registration(dev, tpg, lun,
+ lun->unpacked_lun, nacl, deve);
+}
+
+static void __core_scsi3_dump_registration(
+ struct target_core_fabric_ops *tfo,
+ struct se_device *dev,
+ struct se_node_acl *nacl,
+ struct t10_pr_registration *pr_reg,
+ int register_type)
+{
+ struct se_portal_group *se_tpg = nacl->se_tpg;
+ char i_buf[PR_REG_ISID_ID_LEN];
+ int prf_isid;
+
+ memset(&i_buf[0], 0, PR_REG_ISID_ID_LEN);
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+
+ printk(KERN_INFO "SPC-3 PR [%s] Service Action: REGISTER%s Initiator"
+ " Node: %s%s\n", tfo->get_fabric_name(), (register_type == 2) ?
+ "_AND_MOVE" : (register_type == 1) ?
+ "_AND_IGNORE_EXISTING_KEY" : "", nacl->initiatorname,
+ (prf_isid) ? i_buf : "");
+ printk(KERN_INFO "SPC-3 PR [%s] registration on Target Port: %s,0x%04x\n",
+ tfo->get_fabric_name(), tfo->tpg_get_wwn(se_tpg),
+ tfo->tpg_get_tag(se_tpg));
+ printk(KERN_INFO "SPC-3 PR [%s] for %s TCM Subsystem %s Object Target"
+ " Port(s)\n", tfo->get_fabric_name(),
+ (pr_reg->pr_reg_all_tg_pt) ? "ALL" : "SINGLE",
+ TRANSPORT(dev)->name);
+ printk(KERN_INFO "SPC-3 PR [%s] SA Res Key: 0x%016Lx PRgeneration:"
+ " 0x%08x APTPL: %d\n", tfo->get_fabric_name(),
+ pr_reg->pr_res_key, pr_reg->pr_res_generation,
+ pr_reg->pr_reg_aptpl);
+}
+
+/*
+ * this function can be called with struct se_device->dev_reservation_lock
+ * when register_move = 1
+ */
+static void __core_scsi3_add_registration(
+ struct se_device *dev,
+ struct se_node_acl *nacl,
+ struct t10_pr_registration *pr_reg,
+ int register_type,
+ int register_move)
+{
+ struct se_subsystem_dev *su_dev = SU_DEV(dev);
+ struct target_core_fabric_ops *tfo = nacl->se_tpg->se_tpg_tfo;
+ struct t10_pr_registration *pr_reg_tmp, *pr_reg_tmp_safe;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+
+ /*
+ * Increment PRgeneration counter for struct se_device upon a successful
+ * REGISTER, see spc4r17 section 6.3.2 READ_KEYS service action
+ *
+ * Also, when register_move = 1 for PROUT REGISTER_AND_MOVE service
+ * action, the struct se_device->dev_reservation_lock will already be held,
+ * so we do not call core_scsi3_pr_generation() which grabs the lock
+ * for the REGISTER.
+ */
+ pr_reg->pr_res_generation = (register_move) ?
+ T10_RES(su_dev)->pr_generation++ :
+ core_scsi3_pr_generation(dev);
+
+ spin_lock(&pr_tmpl->registration_lock);
+ list_add_tail(&pr_reg->pr_reg_list, &pr_tmpl->registration_list);
+ pr_reg->pr_reg_deve->def_pr_registered = 1;
+
+ __core_scsi3_dump_registration(tfo, dev, nacl, pr_reg, register_type);
+ spin_unlock(&pr_tmpl->registration_lock);
+ /*
+ * Skip extra processing for ALL_TG_PT=0 or REGISTER_AND_MOVE.
+ */
+ if (!(pr_reg->pr_reg_all_tg_pt) || (register_move))
+ return;
+ /*
+ * Walk pr_reg->pr_reg_atp_list and add registrations for ALL_TG_PT=1
+ * allocated in __core_scsi3_alloc_registration()
+ */
+ list_for_each_entry_safe(pr_reg_tmp, pr_reg_tmp_safe,
+ &pr_reg->pr_reg_atp_list, pr_reg_atp_mem_list) {
+ list_del(&pr_reg_tmp->pr_reg_atp_mem_list);
+
+ pr_reg_tmp->pr_res_generation = core_scsi3_pr_generation(dev);
+
+ spin_lock(&pr_tmpl->registration_lock);
+ list_add_tail(&pr_reg_tmp->pr_reg_list,
+ &pr_tmpl->registration_list);
+ pr_reg_tmp->pr_reg_deve->def_pr_registered = 1;
+
+ __core_scsi3_dump_registration(tfo, dev,
+ pr_reg_tmp->pr_reg_nacl, pr_reg_tmp,
+ register_type);
+ spin_unlock(&pr_tmpl->registration_lock);
+ /*
+ * Drop configfs group dependency reference from
+ * __core_scsi3_alloc_registration()
+ */
+ core_scsi3_lunacl_undepend_item(pr_reg_tmp->pr_reg_deve);
+ }
+}
+
+static int core_scsi3_alloc_registration(
+ struct se_device *dev,
+ struct se_node_acl *nacl,
+ struct se_dev_entry *deve,
+ unsigned char *isid,
+ u64 sa_res_key,
+ int all_tg_pt,
+ int aptpl,
+ int register_type,
+ int register_move)
+{
+ struct t10_pr_registration *pr_reg;
+
+ pr_reg = __core_scsi3_alloc_registration(dev, nacl, deve, isid,
+ sa_res_key, all_tg_pt, aptpl);
+ if (!(pr_reg))
+ return -1;
+
+ __core_scsi3_add_registration(dev, nacl, pr_reg,
+ register_type, register_move);
+ return 0;
+}
+
+static struct t10_pr_registration *__core_scsi3_locate_pr_reg(
+ struct se_device *dev,
+ struct se_node_acl *nacl,
+ unsigned char *isid)
+{
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp;
+ struct se_portal_group *tpg;
+
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp,
+ &pr_tmpl->registration_list, pr_reg_list) {
+ /*
+ * First look for a matching struct se_node_acl
+ */
+ if (pr_reg->pr_reg_nacl != nacl)
+ continue;
+
+ tpg = pr_reg->pr_reg_nacl->se_tpg;
+ /*
+ * If this registration does NOT contain a fabric provided
+ * ISID, then we have found a match.
+ */
+ if (!(pr_reg->isid_present_at_reg)) {
+ /*
+ * Determine if this SCSI device server requires that
+ * SCSI Intiatior TransportID w/ ISIDs is enforced
+ * for fabric modules (iSCSI) requiring them.
+ */
+ if (TPG_TFO(tpg)->sess_get_initiator_sid != NULL) {
+ if (DEV_ATTRIB(dev)->enforce_pr_isids)
+ continue;
+ }
+ atomic_inc(&pr_reg->pr_res_holders);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&pr_tmpl->registration_lock);
+ return pr_reg;
+ }
+ /*
+ * If the *pr_reg contains a fabric defined ISID for multi-value
+ * SCSI Initiator Port TransportIDs, then we expect a valid
+ * matching ISID to be provided by the local SCSI Initiator Port.
+ */
+ if (!(isid))
+ continue;
+ if (strcmp(isid, pr_reg->pr_reg_isid))
+ continue;
+
+ atomic_inc(&pr_reg->pr_res_holders);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&pr_tmpl->registration_lock);
+ return pr_reg;
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+
+ return NULL;
+}
+
+static struct t10_pr_registration *core_scsi3_locate_pr_reg(
+ struct se_device *dev,
+ struct se_node_acl *nacl,
+ struct se_session *sess)
+{
+ struct se_portal_group *tpg = nacl->se_tpg;
+ unsigned char buf[PR_REG_ISID_LEN], *isid_ptr = NULL;
+
+ if (TPG_TFO(tpg)->sess_get_initiator_sid != NULL) {
+ memset(&buf[0], 0, PR_REG_ISID_LEN);
+ TPG_TFO(tpg)->sess_get_initiator_sid(sess, &buf[0],
+ PR_REG_ISID_LEN);
+ isid_ptr = &buf[0];
+ }
+
+ return __core_scsi3_locate_pr_reg(dev, nacl, isid_ptr);
+}
+
+static void core_scsi3_put_pr_reg(struct t10_pr_registration *pr_reg)
+{
+ atomic_dec(&pr_reg->pr_res_holders);
+ smp_mb__after_atomic_dec();
+}
+
+static int core_scsi3_check_implict_release(
+ struct se_device *dev,
+ struct t10_pr_registration *pr_reg)
+{
+ struct se_node_acl *nacl = pr_reg->pr_reg_nacl;
+ struct t10_pr_registration *pr_res_holder;
+ int ret = 0;
+
+ spin_lock(&dev->dev_reservation_lock);
+ pr_res_holder = dev->dev_pr_res_holder;
+ if (!(pr_res_holder)) {
+ spin_unlock(&dev->dev_reservation_lock);
+ return ret;
+ }
+ if (pr_res_holder == pr_reg) {
+ /*
+ * Perform an implict RELEASE if the registration that
+ * is being released is holding the reservation.
+ *
+ * From spc4r17, section 5.7.11.1:
+ *
+ * e) If the I_T nexus is the persistent reservation holder
+ * and the persistent reservation is not an all registrants
+ * type, then a PERSISTENT RESERVE OUT command with REGISTER
+ * service action or REGISTER AND IGNORE EXISTING KEY
+ * service action with the SERVICE ACTION RESERVATION KEY
+ * field set to zero (see 5.7.11.3).
+ */
+ __core_scsi3_complete_pro_release(dev, nacl, pr_reg, 0);
+ ret = 1;
+ /*
+ * For 'All Registrants' reservation types, all existing
+ * registrations are still processed as reservation holders
+ * in core_scsi3_pr_seq_non_holder() after the initial
+ * reservation holder is implictly released here.
+ */
+ } else if (pr_reg->pr_reg_all_tg_pt &&
+ (!strcmp(pr_res_holder->pr_reg_nacl->initiatorname,
+ pr_reg->pr_reg_nacl->initiatorname)) &&
+ (pr_res_holder->pr_res_key == pr_reg->pr_res_key)) {
+ printk(KERN_ERR "SPC-3 PR: Unable to perform ALL_TG_PT=1"
+ " UNREGISTER while existing reservation with matching"
+ " key 0x%016Lx is present from another SCSI Initiator"
+ " Port\n", pr_reg->pr_res_key);
+ ret = -1;
+ }
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return ret;
+}
+
+/*
+ * Called with struct t10_reservation_template->registration_lock held.
+ */
+static void __core_scsi3_free_registration(
+ struct se_device *dev,
+ struct t10_pr_registration *pr_reg,
+ struct list_head *preempt_and_abort_list,
+ int dec_holders)
+{
+ struct target_core_fabric_ops *tfo =
+ pr_reg->pr_reg_nacl->se_tpg->se_tpg_tfo;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ char i_buf[PR_REG_ISID_ID_LEN];
+ int prf_isid;
+
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+
+ pr_reg->pr_reg_deve->def_pr_registered = 0;
+ pr_reg->pr_reg_deve->pr_res_key = 0;
+ list_del(&pr_reg->pr_reg_list);
+ /*
+ * Caller accessing *pr_reg using core_scsi3_locate_pr_reg(),
+ * so call core_scsi3_put_pr_reg() to decrement our reference.
+ */
+ if (dec_holders)
+ core_scsi3_put_pr_reg(pr_reg);
+ /*
+ * Wait until all reference from any other I_T nexuses for this
+ * *pr_reg have been released. Because list_del() is called above,
+ * the last core_scsi3_put_pr_reg(pr_reg) will release this reference
+ * count back to zero, and we release *pr_reg.
+ */
+ while (atomic_read(&pr_reg->pr_res_holders) != 0) {
+ spin_unlock(&pr_tmpl->registration_lock);
+ printk("SPC-3 PR [%s] waiting for pr_res_holders\n",
+ tfo->get_fabric_name());
+ cpu_relax();
+ spin_lock(&pr_tmpl->registration_lock);
+ }
+
+ printk(KERN_INFO "SPC-3 PR [%s] Service Action: UNREGISTER Initiator"
+ " Node: %s%s\n", tfo->get_fabric_name(),
+ pr_reg->pr_reg_nacl->initiatorname,
+ (prf_isid) ? &i_buf[0] : "");
+ printk(KERN_INFO "SPC-3 PR [%s] for %s TCM Subsystem %s Object Target"
+ " Port(s)\n", tfo->get_fabric_name(),
+ (pr_reg->pr_reg_all_tg_pt) ? "ALL" : "SINGLE",
+ TRANSPORT(dev)->name);
+ printk(KERN_INFO "SPC-3 PR [%s] SA Res Key: 0x%016Lx PRgeneration:"
+ " 0x%08x\n", tfo->get_fabric_name(), pr_reg->pr_res_key,
+ pr_reg->pr_res_generation);
+
+ if (!(preempt_and_abort_list)) {
+ pr_reg->pr_reg_deve = NULL;
+ pr_reg->pr_reg_nacl = NULL;
+ kfree(pr_reg->pr_aptpl_buf);
+ kmem_cache_free(t10_pr_reg_cache, pr_reg);
+ return;
+ }
+ /*
+ * For PREEMPT_AND_ABORT, the list of *pr_reg in preempt_and_abort_list
+ * are released once the ABORT_TASK_SET has completed..
+ */
+ list_add_tail(&pr_reg->pr_reg_abort_list, preempt_and_abort_list);
+}
+
+void core_scsi3_free_pr_reg_from_nacl(
+ struct se_device *dev,
+ struct se_node_acl *nacl)
+{
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp, *pr_res_holder;
+ /*
+ * If the passed se_node_acl matches the reservation holder,
+ * release the reservation.
+ */
+ spin_lock(&dev->dev_reservation_lock);
+ pr_res_holder = dev->dev_pr_res_holder;
+ if ((pr_res_holder != NULL) &&
+ (pr_res_holder->pr_reg_nacl == nacl))
+ __core_scsi3_complete_pro_release(dev, nacl, pr_res_holder, 0);
+ spin_unlock(&dev->dev_reservation_lock);
+ /*
+ * Release any registration associated with the struct se_node_acl.
+ */
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp,
+ &pr_tmpl->registration_list, pr_reg_list) {
+
+ if (pr_reg->pr_reg_nacl != nacl)
+ continue;
+
+ __core_scsi3_free_registration(dev, pr_reg, NULL, 0);
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+}
+
+void core_scsi3_free_all_registrations(
+ struct se_device *dev)
+{
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp, *pr_res_holder;
+
+ spin_lock(&dev->dev_reservation_lock);
+ pr_res_holder = dev->dev_pr_res_holder;
+ if (pr_res_holder != NULL) {
+ struct se_node_acl *pr_res_nacl = pr_res_holder->pr_reg_nacl;
+ __core_scsi3_complete_pro_release(dev, pr_res_nacl,
+ pr_res_holder, 0);
+ }
+ spin_unlock(&dev->dev_reservation_lock);
+
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp,
+ &pr_tmpl->registration_list, pr_reg_list) {
+
+ __core_scsi3_free_registration(dev, pr_reg, NULL, 0);
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+
+ spin_lock(&pr_tmpl->aptpl_reg_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp, &pr_tmpl->aptpl_reg_list,
+ pr_reg_aptpl_list) {
+ list_del(&pr_reg->pr_reg_aptpl_list);
+ kfree(pr_reg->pr_aptpl_buf);
+ kmem_cache_free(t10_pr_reg_cache, pr_reg);
+ }
+ spin_unlock(&pr_tmpl->aptpl_reg_lock);
+}
+
+static int core_scsi3_tpg_depend_item(struct se_portal_group *tpg)
+{
+ return configfs_depend_item(TPG_TFO(tpg)->tf_subsys,
+ &tpg->tpg_group.cg_item);
+}
+
+static void core_scsi3_tpg_undepend_item(struct se_portal_group *tpg)
+{
+ configfs_undepend_item(TPG_TFO(tpg)->tf_subsys,
+ &tpg->tpg_group.cg_item);
+
+ atomic_dec(&tpg->tpg_pr_ref_count);
+ smp_mb__after_atomic_dec();
+}
+
+static int core_scsi3_nodeacl_depend_item(struct se_node_acl *nacl)
+{
+ struct se_portal_group *tpg = nacl->se_tpg;
+
+ if (nacl->dynamic_node_acl)
+ return 0;
+
+ return configfs_depend_item(TPG_TFO(tpg)->tf_subsys,
+ &nacl->acl_group.cg_item);
+}
+
+static void core_scsi3_nodeacl_undepend_item(struct se_node_acl *nacl)
+{
+ struct se_portal_group *tpg = nacl->se_tpg;
+
+ if (nacl->dynamic_node_acl) {
+ atomic_dec(&nacl->acl_pr_ref_count);
+ smp_mb__after_atomic_dec();
+ return;
+ }
+
+ configfs_undepend_item(TPG_TFO(tpg)->tf_subsys,
+ &nacl->acl_group.cg_item);
+
+ atomic_dec(&nacl->acl_pr_ref_count);
+ smp_mb__after_atomic_dec();
+}
+
+static int core_scsi3_lunacl_depend_item(struct se_dev_entry *se_deve)
+{
+ struct se_lun_acl *lun_acl = se_deve->se_lun_acl;
+ struct se_node_acl *nacl;
+ struct se_portal_group *tpg;
+ /*
+ * For nacl->dynamic_node_acl=1
+ */
+ if (!(lun_acl))
+ return 0;
+
+ nacl = lun_acl->se_lun_nacl;
+ tpg = nacl->se_tpg;
+
+ return configfs_depend_item(TPG_TFO(tpg)->tf_subsys,
+ &lun_acl->se_lun_group.cg_item);
+}
+
+static void core_scsi3_lunacl_undepend_item(struct se_dev_entry *se_deve)
+{
+ struct se_lun_acl *lun_acl = se_deve->se_lun_acl;
+ struct se_node_acl *nacl;
+ struct se_portal_group *tpg;
+ /*
+ * For nacl->dynamic_node_acl=1
+ */
+ if (!(lun_acl)) {
+ atomic_dec(&se_deve->pr_ref_count);
+ smp_mb__after_atomic_dec();
+ return;
+ }
+ nacl = lun_acl->se_lun_nacl;
+ tpg = nacl->se_tpg;
+
+ configfs_undepend_item(TPG_TFO(tpg)->tf_subsys,
+ &lun_acl->se_lun_group.cg_item);
+
+ atomic_dec(&se_deve->pr_ref_count);
+ smp_mb__after_atomic_dec();
+}
+
+static int core_scsi3_decode_spec_i_port(
+ struct se_cmd *cmd,
+ struct se_portal_group *tpg,
+ unsigned char *l_isid,
+ u64 sa_res_key,
+ int all_tg_pt,
+ int aptpl)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_port *tmp_port;
+ struct se_portal_group *dest_tpg = NULL, *tmp_tpg;
+ struct se_session *se_sess = SE_SESS(cmd);
+ struct se_node_acl *dest_node_acl = NULL;
+ struct se_dev_entry *dest_se_deve = NULL, *local_se_deve;
+ struct t10_pr_registration *dest_pr_reg, *local_pr_reg, *pr_reg_e;
+ struct t10_pr_registration *pr_reg_tmp, *pr_reg_tmp_safe;
+ struct list_head tid_dest_list;
+ struct pr_transport_id_holder *tidh_new, *tidh, *tidh_tmp;
+ struct target_core_fabric_ops *tmp_tf_ops;
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ unsigned char *ptr, *i_str = NULL, proto_ident, tmp_proto_ident;
+ char *iport_ptr = NULL, dest_iport[64], i_buf[PR_REG_ISID_ID_LEN];
+ u32 tpdl, tid_len = 0;
+ int ret, dest_local_nexus, prf_isid;
+ u32 dest_rtpi = 0;
+
+ memset(dest_iport, 0, 64);
+ INIT_LIST_HEAD(&tid_dest_list);
+
+ local_se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
+ /*
+ * Allocate a struct pr_transport_id_holder and setup the
+ * local_node_acl and local_se_deve pointers and add to
+ * struct list_head tid_dest_list for add registration
+ * processing in the loop of tid_dest_list below.
+ */
+ tidh_new = kzalloc(sizeof(struct pr_transport_id_holder), GFP_KERNEL);
+ if (!(tidh_new)) {
+ printk(KERN_ERR "Unable to allocate tidh_new\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ INIT_LIST_HEAD(&tidh_new->dest_list);
+ tidh_new->dest_tpg = tpg;
+ tidh_new->dest_node_acl = se_sess->se_node_acl;
+ tidh_new->dest_se_deve = local_se_deve;
+
+ local_pr_reg = __core_scsi3_alloc_registration(SE_DEV(cmd),
+ se_sess->se_node_acl, local_se_deve, l_isid,
+ sa_res_key, all_tg_pt, aptpl);
+ if (!(local_pr_reg)) {
+ kfree(tidh_new);
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ tidh_new->dest_pr_reg = local_pr_reg;
+ /*
+ * The local I_T nexus does not hold any configfs dependances,
+ * so we set tid_h->dest_local_nexus=1 to prevent the
+ * configfs_undepend_item() calls in the tid_dest_list loops below.
+ */
+ tidh_new->dest_local_nexus = 1;
+ list_add_tail(&tidh_new->dest_list, &tid_dest_list);
+ /*
+ * For a PERSISTENT RESERVE OUT specify initiator ports payload,
+ * first extract TransportID Parameter Data Length, and make sure
+ * the value matches up to the SCSI expected data transfer length.
+ */
+ tpdl = (buf[24] & 0xff) << 24;
+ tpdl |= (buf[25] & 0xff) << 16;
+ tpdl |= (buf[26] & 0xff) << 8;
+ tpdl |= buf[27] & 0xff;
+
+ if ((tpdl + 28) != cmd->data_length) {
+ printk(KERN_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;
+ goto out;
+ }
+ /*
+ * Start processing the received transport IDs using the
+ * receiving I_T Nexus portal's fabric dependent methods to
+ * obtain the SCSI Initiator Port/Device Identifiers.
+ */
+ ptr = &buf[28];
+
+ while (tpdl > 0) {
+ proto_ident = (ptr[0] & 0x0f);
+ dest_tpg = NULL;
+
+ spin_lock(&dev->se_port_lock);
+ list_for_each_entry(tmp_port, &dev->dev_sep_list, sep_list) {
+ tmp_tpg = tmp_port->sep_tpg;
+ if (!(tmp_tpg))
+ continue;
+ tmp_tf_ops = TPG_TFO(tmp_tpg);
+ if (!(tmp_tf_ops))
+ continue;
+ if (!(tmp_tf_ops->get_fabric_proto_ident) ||
+ !(tmp_tf_ops->tpg_parse_pr_out_transport_id))
+ continue;
+ /*
+ * Look for the matching proto_ident provided by
+ * the received TransportID
+ */
+ tmp_proto_ident = tmp_tf_ops->get_fabric_proto_ident(tmp_tpg);
+ if (tmp_proto_ident != proto_ident)
+ continue;
+ dest_rtpi = tmp_port->sep_rtpi;
+
+ i_str = tmp_tf_ops->tpg_parse_pr_out_transport_id(
+ tmp_tpg, (const char *)ptr, &tid_len,
+ &iport_ptr);
+ if (!(i_str))
+ continue;
+
+ atomic_inc(&tmp_tpg->tpg_pr_ref_count);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&dev->se_port_lock);
+
+ ret = core_scsi3_tpg_depend_item(tmp_tpg);
+ if (ret != 0) {
+ printk(KERN_ERR " core_scsi3_tpg_depend_item()"
+ " for tmp_tpg\n");
+ atomic_dec(&tmp_tpg->tpg_pr_ref_count);
+ smp_mb__after_atomic_dec();
+ ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+ goto out;
+ }
+ /*
+ * Locate the desination initiator ACL to be registered
+ * from the decoded fabric module specific TransportID
+ * at *i_str.
+ */
+ spin_lock_bh(&tmp_tpg->acl_node_lock);
+ dest_node_acl = __core_tpg_get_initiator_node_acl(
+ tmp_tpg, i_str);
+ if (dest_node_acl) {
+ atomic_inc(&dest_node_acl->acl_pr_ref_count);
+ smp_mb__after_atomic_inc();
+ }
+ spin_unlock_bh(&tmp_tpg->acl_node_lock);
+
+ if (!(dest_node_acl)) {
+ core_scsi3_tpg_undepend_item(tmp_tpg);
+ spin_lock(&dev->se_port_lock);
+ continue;
+ }
+
+ ret = core_scsi3_nodeacl_depend_item(dest_node_acl);
+ if (ret != 0) {
+ printk(KERN_ERR "configfs_depend_item() failed"
+ " for dest_node_acl->acl_group\n");
+ 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;
+ goto out;
+ }
+
+ dest_tpg = tmp_tpg;
+ printk(KERN_INFO "SPC-3 PR SPEC_I_PT: Located %s Node:"
+ " %s Port RTPI: %hu\n",
+ TPG_TFO(dest_tpg)->get_fabric_name(),
+ dest_node_acl->initiatorname, dest_rtpi);
+
+ spin_lock(&dev->se_port_lock);
+ break;
+ }
+ spin_unlock(&dev->se_port_lock);
+
+ if (!(dest_tpg)) {
+ printk(KERN_ERR "SPC-3 PR SPEC_I_PT: Unable to locate"
+ " dest_tpg\n");
+ ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ goto out;
+ }
+#if 0
+ printk("SPC-3 PR SPEC_I_PT: Got %s data_length: %u tpdl: %u"
+ " tid_len: %d for %s + %s\n",
+ TPG_TFO(dest_tpg)->get_fabric_name(), cmd->data_length,
+ tpdl, tid_len, i_str, iport_ptr);
+#endif
+ if (tid_len > tpdl) {
+ printk(KERN_ERR "SPC-3 PR SPEC_I_PT: Illegal tid_len:"
+ " %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;
+ goto out;
+ }
+ /*
+ * Locate the desintation struct se_dev_entry pointer for matching
+ * RELATIVE TARGET PORT IDENTIFIER on the receiving I_T Nexus
+ * Target Port.
+ */
+ dest_se_deve = core_get_se_deve_from_rtpi(dest_node_acl,
+ dest_rtpi);
+ if (!(dest_se_deve)) {
+ printk(KERN_ERR "Unable to locate %s dest_se_deve"
+ " from destination RTPI: %hu\n",
+ TPG_TFO(dest_tpg)->get_fabric_name(),
+ dest_rtpi);
+
+ core_scsi3_nodeacl_undepend_item(dest_node_acl);
+ core_scsi3_tpg_undepend_item(dest_tpg);
+ ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ goto out;
+ }
+
+ ret = core_scsi3_lunacl_depend_item(dest_se_deve);
+ if (ret < 0) {
+ printk(KERN_ERR "core_scsi3_lunacl_depend_item()"
+ " failed\n");
+ atomic_dec(&dest_se_deve->pr_ref_count);
+ 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;
+ goto out;
+ }
+#if 0
+ printk(KERN_INFO "SPC-3 PR SPEC_I_PT: Located %s Node: %s"
+ " dest_se_deve mapped_lun: %u\n",
+ TPG_TFO(dest_tpg)->get_fabric_name(),
+ dest_node_acl->initiatorname, dest_se_deve->mapped_lun);
+#endif
+ /*
+ * Skip any TransportIDs that already have a registration for
+ * this target port.
+ */
+ pr_reg_e = __core_scsi3_locate_pr_reg(dev, dest_node_acl,
+ iport_ptr);
+ if (pr_reg_e) {
+ core_scsi3_put_pr_reg(pr_reg_e);
+ core_scsi3_lunacl_undepend_item(dest_se_deve);
+ core_scsi3_nodeacl_undepend_item(dest_node_acl);
+ core_scsi3_tpg_undepend_item(dest_tpg);
+ ptr += tid_len;
+ tpdl -= tid_len;
+ tid_len = 0;
+ continue;
+ }
+ /*
+ * Allocate a struct pr_transport_id_holder and setup
+ * the dest_node_acl and dest_se_deve pointers for the
+ * loop below.
+ */
+ tidh_new = kzalloc(sizeof(struct pr_transport_id_holder),
+ GFP_KERNEL);
+ if (!(tidh_new)) {
+ printk(KERN_ERR "Unable to allocate tidh_new\n");
+ 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;
+ goto out;
+ }
+ INIT_LIST_HEAD(&tidh_new->dest_list);
+ tidh_new->dest_tpg = dest_tpg;
+ tidh_new->dest_node_acl = dest_node_acl;
+ tidh_new->dest_se_deve = dest_se_deve;
+
+ /*
+ * Allocate, but do NOT add the registration for the
+ * TransportID referenced SCSI Initiator port. This
+ * done because of the following from spc4r17 in section
+ * 6.14.3 wrt SPEC_I_PT:
+ *
+ * "If a registration fails for any initiator port (e.g., if th
+ * logical unit does not have enough resources available to
+ * hold the registration information), no registrations shall be
+ * made, and the command shall be terminated with
+ * CHECK CONDITION status."
+ *
+ * That means we call __core_scsi3_alloc_registration() here,
+ * and then call __core_scsi3_add_registration() in the
+ * 2nd loop which will never fail.
+ */
+ dest_pr_reg = __core_scsi3_alloc_registration(SE_DEV(cmd),
+ dest_node_acl, dest_se_deve, iport_ptr,
+ sa_res_key, all_tg_pt, aptpl);
+ if (!(dest_pr_reg)) {
+ core_scsi3_lunacl_undepend_item(dest_se_deve);
+ core_scsi3_nodeacl_undepend_item(dest_node_acl);
+ core_scsi3_tpg_undepend_item(dest_tpg);
+ kfree(tidh_new);
+ ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ goto out;
+ }
+ tidh_new->dest_pr_reg = dest_pr_reg;
+ list_add_tail(&tidh_new->dest_list, &tid_dest_list);
+
+ ptr += tid_len;
+ tpdl -= tid_len;
+ tid_len = 0;
+
+ }
+ /*
+ * Go ahead and create a registrations from tid_dest_list for the
+ * SPEC_I_PT provided TransportID for the *tidh referenced dest_node_acl
+ * and dest_se_deve.
+ *
+ * The SA Reservation Key from the PROUT is set for the
+ * registration, and ALL_TG_PT is also passed. ALL_TG_PT=1
+ * means that the TransportID Initiator port will be
+ * registered on all of the target ports in the SCSI target device
+ * ALL_TG_PT=0 means the registration will only be for the
+ * SCSI target port the PROUT REGISTER with SPEC_I_PT=1
+ * was received.
+ */
+ list_for_each_entry_safe(tidh, tidh_tmp, &tid_dest_list, dest_list) {
+ dest_tpg = tidh->dest_tpg;
+ dest_node_acl = tidh->dest_node_acl;
+ dest_se_deve = tidh->dest_se_deve;
+ dest_pr_reg = tidh->dest_pr_reg;
+ dest_local_nexus = tidh->dest_local_nexus;
+
+ list_del(&tidh->dest_list);
+ kfree(tidh);
+
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+ prf_isid = core_pr_dump_initiator_port(dest_pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+
+ __core_scsi3_add_registration(SE_DEV(cmd), dest_node_acl,
+ dest_pr_reg, 0, 0);
+
+ printk(KERN_INFO "SPC-3 PR [%s] SPEC_I_PT: Successfully"
+ " registered Transport ID for Node: %s%s Mapped LUN:"
+ " %u\n", TPG_TFO(dest_tpg)->get_fabric_name(),
+ dest_node_acl->initiatorname, (prf_isid) ?
+ &i_buf[0] : "", dest_se_deve->mapped_lun);
+
+ if (dest_local_nexus)
+ continue;
+
+ core_scsi3_lunacl_undepend_item(dest_se_deve);
+ core_scsi3_nodeacl_undepend_item(dest_node_acl);
+ core_scsi3_tpg_undepend_item(dest_tpg);
+ }
+
+ return 0;
+out:
+ /*
+ * For the failure case, release everything from tid_dest_list
+ * including *dest_pr_reg and the configfs dependances..
+ */
+ list_for_each_entry_safe(tidh, tidh_tmp, &tid_dest_list, dest_list) {
+ dest_tpg = tidh->dest_tpg;
+ dest_node_acl = tidh->dest_node_acl;
+ dest_se_deve = tidh->dest_se_deve;
+ dest_pr_reg = tidh->dest_pr_reg;
+ dest_local_nexus = tidh->dest_local_nexus;
+
+ list_del(&tidh->dest_list);
+ kfree(tidh);
+ /*
+ * Release any extra ALL_TG_PT=1 registrations for
+ * the SPEC_I_PT=1 case.
+ */
+ list_for_each_entry_safe(pr_reg_tmp, pr_reg_tmp_safe,
+ &dest_pr_reg->pr_reg_atp_list,
+ pr_reg_atp_mem_list) {
+ list_del(&pr_reg_tmp->pr_reg_atp_mem_list);
+ core_scsi3_lunacl_undepend_item(pr_reg_tmp->pr_reg_deve);
+ kmem_cache_free(t10_pr_reg_cache, pr_reg_tmp);
+ }
+
+ kfree(dest_pr_reg->pr_aptpl_buf);
+ kmem_cache_free(t10_pr_reg_cache, dest_pr_reg);
+
+ if (dest_local_nexus)
+ continue;
+
+ core_scsi3_lunacl_undepend_item(dest_se_deve);
+ core_scsi3_nodeacl_undepend_item(dest_node_acl);
+ core_scsi3_tpg_undepend_item(dest_tpg);
+ }
+ return ret;
+}
+
+/*
+ * Called with struct se_device->dev_reservation_lock held
+ */
+static int __core_scsi3_update_aptpl_buf(
+ struct se_device *dev,
+ unsigned char *buf,
+ u32 pr_aptpl_buf_len,
+ int clear_aptpl_metadata)
+{
+ struct se_lun *lun;
+ struct se_portal_group *tpg;
+ struct se_subsystem_dev *su_dev = SU_DEV(dev);
+ struct t10_pr_registration *pr_reg;
+ unsigned char tmp[512], isid_buf[32];
+ ssize_t len = 0;
+ int reg_count = 0;
+
+ memset(buf, 0, pr_aptpl_buf_len);
+ /*
+ * Called to clear metadata once APTPL has been deactivated.
+ */
+ if (clear_aptpl_metadata) {
+ snprintf(buf, pr_aptpl_buf_len,
+ "No Registrations or Reservations\n");
+ return 0;
+ }
+ /*
+ * Walk the registration list..
+ */
+ spin_lock(&T10_RES(su_dev)->registration_lock);
+ list_for_each_entry(pr_reg, &T10_RES(su_dev)->registration_list,
+ pr_reg_list) {
+
+ tmp[0] = '\0';
+ isid_buf[0] = '\0';
+ tpg = pr_reg->pr_reg_nacl->se_tpg;
+ lun = pr_reg->pr_reg_tg_pt_lun;
+ /*
+ * Write out any ISID value to APTPL metadata that was included
+ * in the original registration.
+ */
+ if (pr_reg->isid_present_at_reg)
+ snprintf(isid_buf, 32, "initiator_sid=%s\n",
+ pr_reg->pr_reg_isid);
+ /*
+ * Include special metadata if the pr_reg matches the
+ * reservation holder.
+ */
+ if (dev->dev_pr_res_holder == pr_reg) {
+ snprintf(tmp, 512, "PR_REG_START: %d"
+ "\ninitiator_fabric=%s\n"
+ "initiator_node=%s\n%s"
+ "sa_res_key=%llu\n"
+ "res_holder=1\nres_type=%02x\n"
+ "res_scope=%02x\nres_all_tg_pt=%d\n"
+ "mapped_lun=%u\n", reg_count,
+ TPG_TFO(tpg)->get_fabric_name(),
+ pr_reg->pr_reg_nacl->initiatorname, isid_buf,
+ pr_reg->pr_res_key, pr_reg->pr_res_type,
+ pr_reg->pr_res_scope, pr_reg->pr_reg_all_tg_pt,
+ pr_reg->pr_res_mapped_lun);
+ } else {
+ snprintf(tmp, 512, "PR_REG_START: %d\n"
+ "initiator_fabric=%s\ninitiator_node=%s\n%s"
+ "sa_res_key=%llu\nres_holder=0\n"
+ "res_all_tg_pt=%d\nmapped_lun=%u\n",
+ reg_count, TPG_TFO(tpg)->get_fabric_name(),
+ pr_reg->pr_reg_nacl->initiatorname, isid_buf,
+ pr_reg->pr_res_key, pr_reg->pr_reg_all_tg_pt,
+ pr_reg->pr_res_mapped_lun);
+ }
+
+ if ((len + strlen(tmp) > pr_aptpl_buf_len)) {
+ printk(KERN_ERR "Unable to update renaming"
+ " APTPL metadata\n");
+ spin_unlock(&T10_RES(su_dev)->registration_lock);
+ return -1;
+ }
+ len += sprintf(buf+len, "%s", tmp);
+
+ /*
+ * Include information about the associated SCSI target port.
+ */
+ snprintf(tmp, 512, "target_fabric=%s\ntarget_node=%s\n"
+ "tpgt=%hu\nport_rtpi=%hu\ntarget_lun=%u\nPR_REG_END:"
+ " %d\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_wwn(tpg),
+ TPG_TFO(tpg)->tpg_get_tag(tpg),
+ lun->lun_sep->sep_rtpi, lun->unpacked_lun, reg_count);
+
+ if ((len + strlen(tmp) > pr_aptpl_buf_len)) {
+ printk(KERN_ERR "Unable to update renaming"
+ " APTPL metadata\n");
+ spin_unlock(&T10_RES(su_dev)->registration_lock);
+ return -1;
+ }
+ len += sprintf(buf+len, "%s", tmp);
+ reg_count++;
+ }
+ spin_unlock(&T10_RES(su_dev)->registration_lock);
+
+ if (!(reg_count))
+ len += sprintf(buf+len, "No Registrations or Reservations");
+
+ return 0;
+}
+
+static int core_scsi3_update_aptpl_buf(
+ struct se_device *dev,
+ unsigned char *buf,
+ u32 pr_aptpl_buf_len,
+ int clear_aptpl_metadata)
+{
+ int ret;
+
+ spin_lock(&dev->dev_reservation_lock);
+ ret = __core_scsi3_update_aptpl_buf(dev, buf, pr_aptpl_buf_len,
+ clear_aptpl_metadata);
+ spin_unlock(&dev->dev_reservation_lock);
+
+ return ret;
+}
+
+/*
+ * Called with struct se_device->aptpl_file_mutex held
+ */
+static int __core_scsi3_write_aptpl_to_file(
+ struct se_device *dev,
+ unsigned char *buf,
+ u32 pr_aptpl_buf_len)
+{
+ struct t10_wwn *wwn = &SU_DEV(dev)->t10_wwn;
+ struct file *file;
+ struct iovec iov[1];
+ mm_segment_t old_fs;
+ int flags = O_RDWR | O_CREAT | O_TRUNC;
+ char path[512];
+ int ret;
+
+ memset(iov, 0, sizeof(struct iovec));
+ memset(path, 0, 512);
+
+ if (strlen(&wwn->unit_serial[0]) > 512) {
+ printk(KERN_ERR "WWN value for struct se_device does not fit"
+ " into path buffer\n");
+ return -1;
+ }
+
+ snprintf(path, 512, "/var/target/pr/aptpl_%s", &wwn->unit_serial[0]);
+ file = filp_open(path, flags, 0600);
+ if (IS_ERR(file) || !file || !file->f_dentry) {
+ printk(KERN_ERR "filp_open(%s) for APTPL metadata"
+ " failed\n", path);
+ return -1;
+ }
+
+ iov[0].iov_base = &buf[0];
+ if (!(pr_aptpl_buf_len))
+ iov[0].iov_len = (strlen(&buf[0]) + 1); /* Add extra for NULL */
+ else
+ iov[0].iov_len = pr_aptpl_buf_len;
+
+ old_fs = get_fs();
+ set_fs(get_ds());
+ ret = vfs_writev(file, &iov[0], 1, &file->f_pos);
+ set_fs(old_fs);
+
+ if (ret < 0) {
+ printk("Error writing APTPL metadata file: %s\n", path);
+ filp_close(file, NULL);
+ return -1;
+ }
+ filp_close(file, NULL);
+
+ return 0;
+}
+
+static int core_scsi3_update_and_write_aptpl(
+ struct se_device *dev,
+ unsigned char *in_buf,
+ u32 in_pr_aptpl_buf_len)
+{
+ unsigned char null_buf[64], *buf;
+ u32 pr_aptpl_buf_len;
+ int ret, clear_aptpl_metadata = 0;
+ /*
+ * Can be called with a NULL pointer from PROUT service action CLEAR
+ */
+ if (!(in_buf)) {
+ memset(null_buf, 0, 64);
+ buf = &null_buf[0];
+ /*
+ * This will clear the APTPL metadata to:
+ * "No Registrations or Reservations" status
+ */
+ pr_aptpl_buf_len = 64;
+ clear_aptpl_metadata = 1;
+ } else {
+ buf = in_buf;
+ pr_aptpl_buf_len = in_pr_aptpl_buf_len;
+ }
+
+ ret = core_scsi3_update_aptpl_buf(dev, buf, pr_aptpl_buf_len,
+ clear_aptpl_metadata);
+ if (ret != 0)
+ return -1;
+ /*
+ * __core_scsi3_write_aptpl_to_file() will call strlen()
+ * on the passed buf to determine pr_aptpl_buf_len.
+ */
+ ret = __core_scsi3_write_aptpl_to_file(dev, buf, 0);
+ if (ret != 0)
+ return -1;
+
+ return ret;
+}
+
+static int core_scsi3_emulate_pro_register(
+ struct se_cmd *cmd,
+ u64 res_key,
+ u64 sa_res_key,
+ int aptpl,
+ int all_tg_pt,
+ int spec_i_pt,
+ int ignore_key)
+{
+ struct se_session *se_sess = SE_SESS(cmd);
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_dev_entry *se_deve;
+ struct se_lun *se_lun = SE_LUN(cmd);
+ struct se_portal_group *se_tpg;
+ struct t10_pr_registration *pr_reg, *pr_reg_p, *pr_reg_tmp, *pr_reg_e;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ /* Used for APTPL metadata w/ UNREGISTER */
+ unsigned char *pr_aptpl_buf = NULL;
+ unsigned char isid_buf[PR_REG_ISID_LEN], *isid_ptr = NULL;
+ int pr_holder = 0, ret = 0, type;
+
+ if (!(se_sess) || !(se_lun)) {
+ printk(KERN_ERR "SPC-3 PR: se_sess || struct se_lun is NULL!\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ se_tpg = se_sess->se_tpg;
+ se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
+
+ if (TPG_TFO(se_tpg)->sess_get_initiator_sid != NULL) {
+ memset(&isid_buf[0], 0, PR_REG_ISID_LEN);
+ TPG_TFO(se_tpg)->sess_get_initiator_sid(se_sess, &isid_buf[0],
+ PR_REG_ISID_LEN);
+ isid_ptr = &isid_buf[0];
+ }
+ /*
+ * Follow logic from spc4r17 Section 5.7.7, Register Behaviors Table 47
+ */
+ pr_reg_e = core_scsi3_locate_pr_reg(dev, se_sess->se_node_acl, se_sess);
+ if (!(pr_reg_e)) {
+ if (res_key) {
+ printk(KERN_WARNING "SPC-3 PR: Reservation Key non-zero"
+ " for SA REGISTER, returning CONFLICT\n");
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+ /*
+ * Do nothing but return GOOD status.
+ */
+ if (!(sa_res_key))
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+
+ if (!(spec_i_pt)) {
+ /*
+ * Perform the Service Action REGISTER on the Initiator
+ * Port Endpoint that the PRO was received from on the
+ * Logical Unit of the SCSI device server.
+ */
+ ret = core_scsi3_alloc_registration(SE_DEV(cmd),
+ se_sess->se_node_acl, se_deve, isid_ptr,
+ sa_res_key, all_tg_pt, aptpl,
+ ignore_key, 0);
+ if (ret != 0) {
+ printk(KERN_ERR "Unable to allocate"
+ " struct t10_pr_registration\n");
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ }
+ } else {
+ /*
+ * Register both the Initiator port that received
+ * PROUT SA REGISTER + SPEC_I_PT=1 and extract SCSI
+ * TransportID from Parameter list and loop through
+ * fabric dependent parameter list while calling
+ * logic from of core_scsi3_alloc_registration() for
+ * each TransportID provided SCSI Initiator Port/Device
+ */
+ ret = core_scsi3_decode_spec_i_port(cmd, se_tpg,
+ isid_ptr, sa_res_key, all_tg_pt, aptpl);
+ if (ret != 0)
+ return ret;
+ }
+ /*
+ * Nothing left to do for the APTPL=0 case.
+ */
+ if (!(aptpl)) {
+ pr_tmpl->pr_aptpl_active = 0;
+ core_scsi3_update_and_write_aptpl(SE_DEV(cmd), NULL, 0);
+ printk("SPC-3 PR: Set APTPL Bit Deactivated for"
+ " REGISTER\n");
+ return 0;
+ }
+ /*
+ * Locate the newly allocated local I_T Nexus *pr_reg, and
+ * update the APTPL metadata information using its
+ * preallocated *pr_reg->pr_aptpl_buf.
+ */
+ pr_reg = core_scsi3_locate_pr_reg(SE_DEV(cmd),
+ se_sess->se_node_acl, se_sess);
+
+ ret = core_scsi3_update_and_write_aptpl(SE_DEV(cmd),
+ &pr_reg->pr_aptpl_buf[0],
+ pr_tmpl->pr_aptpl_buf_len);
+ if (!(ret)) {
+ pr_tmpl->pr_aptpl_active = 1;
+ printk("SPC-3 PR: Set APTPL Bit Activated for REGISTER\n");
+ }
+
+ core_scsi3_put_pr_reg(pr_reg);
+ return ret;
+ } else {
+ /*
+ * Locate the existing *pr_reg via struct se_node_acl pointers
+ */
+ pr_reg = pr_reg_e;
+ type = pr_reg->pr_res_type;
+
+ if (!(ignore_key)) {
+ if (res_key != pr_reg->pr_res_key) {
+ printk(KERN_ERR "SPC-3 PR REGISTER: Received"
+ " 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;
+ }
+ }
+ if (spec_i_pt) {
+ printk(KERN_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;
+ }
+ /*
+ * An existing ALL_TG_PT=1 registration being released
+ * must also set ALL_TG_PT=1 in the incoming PROUT.
+ */
+ if (pr_reg->pr_reg_all_tg_pt && !(all_tg_pt)) {
+ printk(KERN_ERR "SPC-3 PR UNREGISTER: ALL_TG_PT=1"
+ " 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;
+ }
+ /*
+ * Allocate APTPL metadata buffer used for UNREGISTER ops
+ */
+ if (aptpl) {
+ pr_aptpl_buf = kzalloc(pr_tmpl->pr_aptpl_buf_len,
+ GFP_KERNEL);
+ if (!(pr_aptpl_buf)) {
+ printk(KERN_ERR "Unable to allocate"
+ " pr_aptpl_buf\n");
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ }
+ /*
+ * sa_res_key=0 Unregister Reservation Key for registered I_T
+ * Nexus sa_res_key=1 Change Reservation Key for registered I_T
+ * Nexus.
+ */
+ if (!(sa_res_key)) {
+ pr_holder = core_scsi3_check_implict_release(
+ SE_DEV(cmd), pr_reg);
+ if (pr_holder < 0) {
+ kfree(pr_aptpl_buf);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+
+ spin_lock(&pr_tmpl->registration_lock);
+ /*
+ * Release all ALL_TG_PT=1 for the matching SCSI Initiator Port
+ * and matching pr_res_key.
+ */
+ if (pr_reg->pr_reg_all_tg_pt) {
+ list_for_each_entry_safe(pr_reg_p, pr_reg_tmp,
+ &pr_tmpl->registration_list,
+ pr_reg_list) {
+
+ if (!(pr_reg_p->pr_reg_all_tg_pt))
+ continue;
+
+ if (pr_reg_p->pr_res_key != res_key)
+ continue;
+
+ if (pr_reg == pr_reg_p)
+ continue;
+
+ if (strcmp(pr_reg->pr_reg_nacl->initiatorname,
+ pr_reg_p->pr_reg_nacl->initiatorname))
+ continue;
+
+ __core_scsi3_free_registration(dev,
+ pr_reg_p, NULL, 0);
+ }
+ }
+ /*
+ * Release the calling I_T Nexus registration now..
+ */
+ __core_scsi3_free_registration(SE_DEV(cmd), pr_reg,
+ NULL, 1);
+ /*
+ * From spc4r17, section 5.7.11.3 Unregistering
+ *
+ * If the persistent reservation is a registrants only
+ * type, the device server shall establish a unit
+ * attention condition for the initiator port associated
+ * with every registered I_T nexus except for the I_T
+ * nexus on which the PERSISTENT RESERVE OUT command was
+ * received, with the additional sense code set to
+ * RESERVATIONS RELEASED.
+ */
+ if (pr_holder &&
+ ((type == PR_TYPE_WRITE_EXCLUSIVE_REGONLY) ||
+ (type == PR_TYPE_EXCLUSIVE_ACCESS_REGONLY))) {
+ list_for_each_entry(pr_reg_p,
+ &pr_tmpl->registration_list,
+ pr_reg_list) {
+
+ core_scsi3_ua_allocate(
+ pr_reg_p->pr_reg_nacl,
+ pr_reg_p->pr_res_mapped_lun,
+ 0x2A,
+ ASCQ_2AH_RESERVATIONS_RELEASED);
+ }
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+
+ if (!(aptpl)) {
+ pr_tmpl->pr_aptpl_active = 0;
+ core_scsi3_update_and_write_aptpl(dev, NULL, 0);
+ printk("SPC-3 PR: Set APTPL Bit Deactivated"
+ " for UNREGISTER\n");
+ return 0;
+ }
+
+ ret = core_scsi3_update_and_write_aptpl(dev,
+ &pr_aptpl_buf[0],
+ pr_tmpl->pr_aptpl_buf_len);
+ if (!(ret)) {
+ pr_tmpl->pr_aptpl_active = 1;
+ printk("SPC-3 PR: Set APTPL Bit Activated"
+ " for UNREGISTER\n");
+ }
+
+ kfree(pr_aptpl_buf);
+ return ret;
+ } else {
+ /*
+ * Increment PRgeneration counter for struct se_device"
+ * upon a successful REGISTER, see spc4r17 section 6.3.2
+ * READ_KEYS service action.
+ */
+ pr_reg->pr_res_generation = core_scsi3_pr_generation(
+ SE_DEV(cmd));
+ pr_reg->pr_res_key = sa_res_key;
+ printk("SPC-3 PR [%s] REGISTER%s: Changed Reservation"
+ " Key for %s to: 0x%016Lx PRgeneration:"
+ " 0x%08x\n", CMD_TFO(cmd)->get_fabric_name(),
+ (ignore_key) ? "_AND_IGNORE_EXISTING_KEY" : "",
+ pr_reg->pr_reg_nacl->initiatorname,
+ pr_reg->pr_res_key, pr_reg->pr_res_generation);
+
+ if (!(aptpl)) {
+ pr_tmpl->pr_aptpl_active = 0;
+ core_scsi3_update_and_write_aptpl(dev, NULL, 0);
+ core_scsi3_put_pr_reg(pr_reg);
+ printk("SPC-3 PR: Set APTPL Bit Deactivated"
+ " for REGISTER\n");
+ return 0;
+ }
+
+ ret = core_scsi3_update_and_write_aptpl(dev,
+ &pr_aptpl_buf[0],
+ pr_tmpl->pr_aptpl_buf_len);
+ if (!(ret)) {
+ pr_tmpl->pr_aptpl_active = 1;
+ printk("SPC-3 PR: Set APTPL Bit Activated"
+ " for REGISTER\n");
+ }
+
+ kfree(pr_aptpl_buf);
+ core_scsi3_put_pr_reg(pr_reg);
+ }
+ }
+ return 0;
+}
+
+unsigned char *core_scsi3_pr_dump_type(int type)
+{
+ switch (type) {
+ case PR_TYPE_WRITE_EXCLUSIVE:
+ return "Write Exclusive Access";
+ case PR_TYPE_EXCLUSIVE_ACCESS:
+ return "Exclusive Access";
+ case PR_TYPE_WRITE_EXCLUSIVE_REGONLY:
+ return "Write Exclusive Access, Registrants Only";
+ case PR_TYPE_EXCLUSIVE_ACCESS_REGONLY:
+ return "Exclusive Access, Registrants Only";
+ case PR_TYPE_WRITE_EXCLUSIVE_ALLREG:
+ return "Write Exclusive Access, All Registrants";
+ case PR_TYPE_EXCLUSIVE_ACCESS_ALLREG:
+ return "Exclusive Access, All Registrants";
+ default:
+ break;
+ }
+
+ return "Unknown SPC-3 PR Type";
+}
+
+static int core_scsi3_pro_reserve(
+ struct se_cmd *cmd,
+ struct se_device *dev,
+ int type,
+ int scope,
+ u64 res_key)
+{
+ struct se_session *se_sess = SE_SESS(cmd);
+ struct se_dev_entry *se_deve;
+ struct se_lun *se_lun = SE_LUN(cmd);
+ struct se_portal_group *se_tpg;
+ struct t10_pr_registration *pr_reg, *pr_res_holder;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ char i_buf[PR_REG_ISID_ID_LEN];
+ int ret, prf_isid;
+
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+
+ if (!(se_sess) || !(se_lun)) {
+ printk(KERN_ERR "SPC-3 PR: se_sess || struct se_lun is NULL!\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ se_tpg = se_sess->se_tpg;
+ se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
+ /*
+ * Locate the existing *pr_reg via struct se_node_acl pointers
+ */
+ pr_reg = core_scsi3_locate_pr_reg(SE_DEV(cmd), se_sess->se_node_acl,
+ se_sess);
+ if (!(pr_reg)) {
+ printk(KERN_ERR "SPC-3 PR: Unable to locate"
+ " PR_REGISTERED *pr_reg for RESERVE\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ /*
+ * From spc4r17 Section 5.7.9: Reserving:
+ *
+ * An application client creates a persistent reservation by issuing
+ * a PERSISTENT RESERVE OUT command with RESERVE service action through
+ * a registered I_T nexus with the following parameters:
+ * a) RESERVATION KEY set to the value of the reservation key that is
+ * registered with the logical unit for the I_T nexus; and
+ */
+ if (res_key != pr_reg->pr_res_key) {
+ printk(KERN_ERR "SPC-3 PR RESERVE: Received 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;
+ }
+ /*
+ * From spc4r17 Section 5.7.9: Reserving:
+ *
+ * From above:
+ * b) TYPE field and SCOPE field set to the persistent reservation
+ * being created.
+ *
+ * Only one persistent reservation is allowed at a time per logical unit
+ * and that persistent reservation has a scope of LU_SCOPE.
+ */
+ if (scope != PR_SCOPE_LU_SCOPE) {
+ printk(KERN_ERR "SPC-3 PR: Illegal SCOPE: 0x%02x\n", scope);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ }
+ /*
+ * See if we have an existing PR reservation holder pointer at
+ * struct se_device->dev_pr_res_holder in the form struct t10_pr_registration
+ * *pr_res_holder.
+ */
+ spin_lock(&dev->dev_reservation_lock);
+ pr_res_holder = dev->dev_pr_res_holder;
+ if ((pr_res_holder)) {
+ /*
+ * From spc4r17 Section 5.7.9: Reserving:
+ *
+ * If the device server receives a PERSISTENT RESERVE OUT
+ * command from an I_T nexus other than a persistent reservation
+ * holder (see 5.7.10) that attempts to create a persistent
+ * reservation when a persistent reservation already exists for
+ * the logical unit, then the command shall be completed with
+ * RESERVATION CONFLICT status.
+ */
+ if (pr_res_holder != pr_reg) {
+ struct se_node_acl *pr_res_nacl = pr_res_holder->pr_reg_nacl;
+ printk(KERN_ERR "SPC-3 PR: Attempted RESERVE from"
+ " [%s]: %s while reservation already held by"
+ " [%s]: %s, returning RESERVATION_CONFLICT\n",
+ CMD_TFO(cmd)->get_fabric_name(),
+ se_sess->se_node_acl->initiatorname,
+ TPG_TFO(pr_res_nacl->se_tpg)->get_fabric_name(),
+ pr_res_holder->pr_reg_nacl->initiatorname);
+
+ spin_unlock(&dev->dev_reservation_lock);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+ /*
+ * From spc4r17 Section 5.7.9: Reserving:
+ *
+ * If a persistent reservation holder attempts to modify the
+ * type or scope of an existing persistent reservation, the
+ * command shall be completed with RESERVATION CONFLICT status.
+ */
+ if ((pr_res_holder->pr_res_type != type) ||
+ (pr_res_holder->pr_res_scope != scope)) {
+ struct se_node_acl *pr_res_nacl = pr_res_holder->pr_reg_nacl;
+ printk(KERN_ERR "SPC-3 PR: Attempted RESERVE from"
+ " [%s]: %s trying to change TYPE and/or SCOPE,"
+ " while reservation already held by [%s]: %s,"
+ " returning RESERVATION_CONFLICT\n",
+ CMD_TFO(cmd)->get_fabric_name(),
+ se_sess->se_node_acl->initiatorname,
+ TPG_TFO(pr_res_nacl->se_tpg)->get_fabric_name(),
+ pr_res_holder->pr_reg_nacl->initiatorname);
+
+ spin_unlock(&dev->dev_reservation_lock);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+ /*
+ * From spc4r17 Section 5.7.9: Reserving:
+ *
+ * If the device server receives a PERSISTENT RESERVE OUT
+ * command with RESERVE service action where the TYPE field and
+ * the SCOPE field contain the same values as the existing type
+ * and scope from a persistent reservation holder, it shall not
+ * make any change to the existing persistent reservation and
+ * shall completethe command with GOOD status.
+ */
+ spin_unlock(&dev->dev_reservation_lock);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+ }
+ /*
+ * Otherwise, our *pr_reg becomes the PR reservation holder for said
+ * TYPE/SCOPE. Also set the received scope and type in *pr_reg.
+ */
+ pr_reg->pr_res_scope = scope;
+ pr_reg->pr_res_type = type;
+ pr_reg->pr_res_holder = 1;
+ dev->dev_pr_res_holder = pr_reg;
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+
+ printk(KERN_INFO "SPC-3 PR [%s] Service Action: RESERVE created new"
+ " reservation holder TYPE: %s ALL_TG_PT: %d\n",
+ CMD_TFO(cmd)->get_fabric_name(), core_scsi3_pr_dump_type(type),
+ (pr_reg->pr_reg_all_tg_pt) ? 1 : 0);
+ printk(KERN_INFO "SPC-3 PR [%s] RESERVE Node: %s%s\n",
+ CMD_TFO(cmd)->get_fabric_name(),
+ se_sess->se_node_acl->initiatorname,
+ (prf_isid) ? &i_buf[0] : "");
+ spin_unlock(&dev->dev_reservation_lock);
+
+ if (pr_tmpl->pr_aptpl_active) {
+ ret = core_scsi3_update_and_write_aptpl(SE_DEV(cmd),
+ &pr_reg->pr_aptpl_buf[0],
+ pr_tmpl->pr_aptpl_buf_len);
+ if (!(ret))
+ printk(KERN_INFO "SPC-3 PR: Updated APTPL metadata"
+ " for RESERVE\n");
+ }
+
+ core_scsi3_put_pr_reg(pr_reg);
+ return 0;
+}
+
+static int core_scsi3_emulate_pro_reserve(
+ struct se_cmd *cmd,
+ int type,
+ int scope,
+ u64 res_key)
+{
+ struct se_device *dev = cmd->se_dev;
+ int ret = 0;
+
+ switch (type) {
+ case PR_TYPE_WRITE_EXCLUSIVE:
+ case PR_TYPE_EXCLUSIVE_ACCESS:
+ case PR_TYPE_WRITE_EXCLUSIVE_REGONLY:
+ case PR_TYPE_EXCLUSIVE_ACCESS_REGONLY:
+ case PR_TYPE_WRITE_EXCLUSIVE_ALLREG:
+ case PR_TYPE_EXCLUSIVE_ACCESS_ALLREG:
+ ret = core_scsi3_pro_reserve(cmd, dev, type, scope, res_key);
+ break;
+ default:
+ printk(KERN_ERR "SPC-3 PR: Unknown Service Action RESERVE Type:"
+ " 0x%02x\n", type);
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+
+ return ret;
+}
+
+/*
+ * Called with struct se_device->dev_reservation_lock held.
+ */
+static void __core_scsi3_complete_pro_release(
+ struct se_device *dev,
+ struct se_node_acl *se_nacl,
+ struct t10_pr_registration *pr_reg,
+ int explict)
+{
+ struct target_core_fabric_ops *tfo = se_nacl->se_tpg->se_tpg_tfo;
+ char i_buf[PR_REG_ISID_ID_LEN];
+ int prf_isid;
+
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+ /*
+ * Go ahead and release the current PR reservation holder.
+ */
+ dev->dev_pr_res_holder = NULL;
+
+ printk(KERN_INFO "SPC-3 PR [%s] Service Action: %s RELEASE cleared"
+ " reservation holder TYPE: %s ALL_TG_PT: %d\n",
+ tfo->get_fabric_name(), (explict) ? "explict" : "implict",
+ core_scsi3_pr_dump_type(pr_reg->pr_res_type),
+ (pr_reg->pr_reg_all_tg_pt) ? 1 : 0);
+ printk(KERN_INFO "SPC-3 PR [%s] RELEASE Node: %s%s\n",
+ tfo->get_fabric_name(), se_nacl->initiatorname,
+ (prf_isid) ? &i_buf[0] : "");
+ /*
+ * Clear TYPE and SCOPE for the next PROUT Service Action: RESERVE
+ */
+ pr_reg->pr_res_holder = pr_reg->pr_res_type = pr_reg->pr_res_scope = 0;
+}
+
+static int core_scsi3_emulate_pro_release(
+ struct se_cmd *cmd,
+ int type,
+ int scope,
+ u64 res_key)
+{
+ struct se_device *dev = cmd->se_dev;
+ struct se_session *se_sess = SE_SESS(cmd);
+ struct se_lun *se_lun = SE_LUN(cmd);
+ struct t10_pr_registration *pr_reg, *pr_reg_p, *pr_res_holder;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ int ret, all_reg = 0;
+
+ if (!(se_sess) || !(se_lun)) {
+ printk(KERN_ERR "SPC-3 PR: se_sess || struct se_lun is NULL!\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ /*
+ * Locate the existing *pr_reg via struct se_node_acl pointers
+ */
+ pr_reg = core_scsi3_locate_pr_reg(dev, se_sess->se_node_acl, se_sess);
+ if (!(pr_reg)) {
+ printk(KERN_ERR "SPC-3 PR: Unable to locate"
+ " PR_REGISTERED *pr_reg for RELEASE\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ /*
+ * From spc4r17 Section 5.7.11.2 Releasing:
+ *
+ * If there is no persistent reservation or in response to a persistent
+ * reservation release request from a registered I_T nexus that is not a
+ * persistent reservation holder (see 5.7.10), the device server shall
+ * do the following:
+ *
+ * a) Not release the persistent reservation, if any;
+ * b) Not remove any registrations; and
+ * c) Complete the command with GOOD status.
+ */
+ spin_lock(&dev->dev_reservation_lock);
+ pr_res_holder = dev->dev_pr_res_holder;
+ if (!(pr_res_holder)) {
+ /*
+ * No persistent reservation, return GOOD status.
+ */
+ spin_unlock(&dev->dev_reservation_lock);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+ }
+ if ((pr_res_holder->pr_res_type == PR_TYPE_WRITE_EXCLUSIVE_ALLREG) ||
+ (pr_res_holder->pr_res_type == PR_TYPE_EXCLUSIVE_ACCESS_ALLREG))
+ all_reg = 1;
+
+ if ((all_reg == 0) && (pr_res_holder != pr_reg)) {
+ /*
+ * Non 'All Registrants' PR Type cases..
+ * Release request from a registered I_T nexus that is not a
+ * persistent reservation holder. return GOOD status.
+ */
+ spin_unlock(&dev->dev_reservation_lock);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+ }
+ /*
+ * From spc4r17 Section 5.7.11.2 Releasing:
+ *
+ * Only the persistent reservation holder (see 5.7.10) is allowed to
+ * release a persistent reservation.
+ *
+ * An application client releases the persistent reservation by issuing
+ * a PERSISTENT RESERVE OUT command with RELEASE service action through
+ * an I_T nexus that is a persistent reservation holder with the
+ * following parameters:
+ *
+ * a) RESERVATION KEY field set to the value of the reservation key
+ * that is registered with the logical unit for the I_T nexus;
+ */
+ if (res_key != pr_reg->pr_res_key) {
+ printk(KERN_ERR "SPC-3 PR RELEASE: Received res_key: 0x%016Lx"
+ " does not match existing SA REGISTER res_key:"
+ " 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;
+ }
+ /*
+ * From spc4r17 Section 5.7.11.2 Releasing and above:
+ *
+ * b) TYPE field and SCOPE field set to match the persistent
+ * reservation being released.
+ */
+ if ((pr_res_holder->pr_res_type != type) ||
+ (pr_res_holder->pr_res_scope != scope)) {
+ struct se_node_acl *pr_res_nacl = pr_res_holder->pr_reg_nacl;
+ printk(KERN_ERR "SPC-3 PR RELEASE: Attempted to release"
+ " reservation from [%s]: %s with different TYPE "
+ "and/or SCOPE while reservation already held by"
+ " [%s]: %s, returning RESERVATION_CONFLICT\n",
+ CMD_TFO(cmd)->get_fabric_name(),
+ se_sess->se_node_acl->initiatorname,
+ TPG_TFO(pr_res_nacl->se_tpg)->get_fabric_name(),
+ pr_res_holder->pr_reg_nacl->initiatorname);
+
+ spin_unlock(&dev->dev_reservation_lock);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+ /*
+ * In response to a persistent reservation release request from the
+ * persistent reservation holder the device server shall perform a
+ * release by doing the following as an uninterrupted series of actions:
+ * a) Release the persistent reservation;
+ * b) Not remove any registration(s);
+ * c) If the released persistent reservation is a registrants only type
+ * or all registrants type persistent reservation,
+ * the device server shall establish a unit attention condition for
+ * the initiator port associated with every regis-
+ * tered I_T nexus other than I_T nexus on which the PERSISTENT
+ * RESERVE OUT command with RELEASE service action was received,
+ * with the additional sense code set to RESERVATIONS RELEASED; and
+ * d) If the persistent reservation is of any other type, the device
+ * server shall not establish a unit attention condition.
+ */
+ __core_scsi3_complete_pro_release(dev, se_sess->se_node_acl,
+ pr_reg, 1);
+
+ spin_unlock(&dev->dev_reservation_lock);
+
+ if ((type != PR_TYPE_WRITE_EXCLUSIVE_REGONLY) &&
+ (type != PR_TYPE_EXCLUSIVE_ACCESS_REGONLY) &&
+ (type != PR_TYPE_WRITE_EXCLUSIVE_ALLREG) &&
+ (type != PR_TYPE_EXCLUSIVE_ACCESS_ALLREG)) {
+ /*
+ * If no UNIT ATTENTION conditions will be established for
+ * PR_TYPE_WRITE_EXCLUSIVE or PR_TYPE_EXCLUSIVE_ACCESS
+ * go ahead and check for APTPL=1 update+write below
+ */
+ goto write_aptpl;
+ }
+
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry(pr_reg_p, &pr_tmpl->registration_list,
+ pr_reg_list) {
+ /*
+ * Do not establish a UNIT ATTENTION condition
+ * for the calling I_T Nexus
+ */
+ if (pr_reg_p == pr_reg)
+ continue;
+
+ core_scsi3_ua_allocate(pr_reg_p->pr_reg_nacl,
+ pr_reg_p->pr_res_mapped_lun,
+ 0x2A, ASCQ_2AH_RESERVATIONS_RELEASED);
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+
+write_aptpl:
+ if (pr_tmpl->pr_aptpl_active) {
+ ret = core_scsi3_update_and_write_aptpl(SE_DEV(cmd),
+ &pr_reg->pr_aptpl_buf[0],
+ pr_tmpl->pr_aptpl_buf_len);
+ if (!(ret))
+ printk("SPC-3 PR: Updated APTPL metadata for RELEASE\n");
+ }
+
+ core_scsi3_put_pr_reg(pr_reg);
+ return 0;
+}
+
+static int core_scsi3_emulate_pro_clear(
+ struct se_cmd *cmd,
+ u64 res_key)
+{
+ struct se_device *dev = cmd->se_dev;
+ struct se_node_acl *pr_reg_nacl;
+ struct se_session *se_sess = SE_SESS(cmd);
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp, *pr_reg_n, *pr_res_holder;
+ u32 pr_res_mapped_lun = 0;
+ int calling_it_nexus = 0;
+ /*
+ * Locate the existing *pr_reg via struct se_node_acl pointers
+ */
+ pr_reg_n = core_scsi3_locate_pr_reg(SE_DEV(cmd),
+ se_sess->se_node_acl, se_sess);
+ if (!(pr_reg_n)) {
+ printk(KERN_ERR "SPC-3 PR: Unable to locate"
+ " PR_REGISTERED *pr_reg for CLEAR\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ /*
+ * From spc4r17 section 5.7.11.6, Clearing:
+ *
+ * Any application client may release the persistent reservation and
+ * remove all registrations from a device server by issuing a
+ * PERSISTENT RESERVE OUT command with CLEAR service action through a
+ * registered I_T nexus with the following parameter:
+ *
+ * a) RESERVATION KEY field set to the value of the reservation key
+ * that is registered with the logical unit for the I_T nexus.
+ */
+ if (res_key != pr_reg_n->pr_res_key) {
+ printk(KERN_ERR "SPC-3 PR REGISTER: Received"
+ " res_key: 0x%016Lx does not match"
+ " 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;
+ }
+ /*
+ * a) Release the persistent reservation, if any;
+ */
+ spin_lock(&dev->dev_reservation_lock);
+ pr_res_holder = dev->dev_pr_res_holder;
+ if (pr_res_holder) {
+ struct se_node_acl *pr_res_nacl = pr_res_holder->pr_reg_nacl;
+ __core_scsi3_complete_pro_release(dev, pr_res_nacl,
+ pr_res_holder, 0);
+ }
+ spin_unlock(&dev->dev_reservation_lock);
+ /*
+ * b) Remove all registration(s) (see spc4r17 5.7.7);
+ */
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp,
+ &pr_tmpl->registration_list, pr_reg_list) {
+
+ calling_it_nexus = (pr_reg_n == pr_reg) ? 1 : 0;
+ pr_reg_nacl = pr_reg->pr_reg_nacl;
+ pr_res_mapped_lun = pr_reg->pr_res_mapped_lun;
+ __core_scsi3_free_registration(dev, pr_reg, NULL,
+ calling_it_nexus);
+ /*
+ * e) Establish a unit attention condition for the initiator
+ * port associated with every registered I_T nexus other
+ * than the I_T nexus on which the PERSISTENT RESERVE OUT
+ * command with CLEAR service action was received, with the
+ * additional sense code set to RESERVATIONS PREEMPTED.
+ */
+ if (!(calling_it_nexus))
+ core_scsi3_ua_allocate(pr_reg_nacl, pr_res_mapped_lun,
+ 0x2A, ASCQ_2AH_RESERVATIONS_PREEMPTED);
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+
+ printk(KERN_INFO "SPC-3 PR [%s] Service Action: CLEAR complete\n",
+ CMD_TFO(cmd)->get_fabric_name());
+
+ if (pr_tmpl->pr_aptpl_active) {
+ core_scsi3_update_and_write_aptpl(SE_DEV(cmd), NULL, 0);
+ printk(KERN_INFO "SPC-3 PR: Updated APTPL metadata"
+ " for CLEAR\n");
+ }
+
+ core_scsi3_pr_generation(dev);
+ return 0;
+}
+
+/*
+ * Called with struct se_device->dev_reservation_lock held.
+ */
+static void __core_scsi3_complete_pro_preempt(
+ struct se_device *dev,
+ struct t10_pr_registration *pr_reg,
+ struct list_head *preempt_and_abort_list,
+ int type,
+ int scope,
+ int abort)
+{
+ struct se_node_acl *nacl = pr_reg->pr_reg_nacl;
+ struct target_core_fabric_ops *tfo = nacl->se_tpg->se_tpg_tfo;
+ char i_buf[PR_REG_ISID_ID_LEN];
+ int prf_isid;
+
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+ /*
+ * Do an implict RELEASE of the existing reservation.
+ */
+ if (dev->dev_pr_res_holder)
+ __core_scsi3_complete_pro_release(dev, nacl,
+ dev->dev_pr_res_holder, 0);
+
+ dev->dev_pr_res_holder = pr_reg;
+ pr_reg->pr_res_holder = 1;
+ pr_reg->pr_res_type = type;
+ pr_reg->pr_res_scope = scope;
+
+ printk(KERN_INFO "SPC-3 PR [%s] Service Action: PREEMPT%s created new"
+ " reservation holder TYPE: %s ALL_TG_PT: %d\n",
+ tfo->get_fabric_name(), (abort) ? "_AND_ABORT" : "",
+ core_scsi3_pr_dump_type(type),
+ (pr_reg->pr_reg_all_tg_pt) ? 1 : 0);
+ printk(KERN_INFO "SPC-3 PR [%s] PREEMPT%s from Node: %s%s\n",
+ tfo->get_fabric_name(), (abort) ? "_AND_ABORT" : "",
+ nacl->initiatorname, (prf_isid) ? &i_buf[0] : "");
+ /*
+ * For PREEMPT_AND_ABORT, add the preempting reservation's
+ * struct t10_pr_registration to the list that will be compared
+ * against received CDBs..
+ */
+ if (preempt_and_abort_list)
+ list_add_tail(&pr_reg->pr_reg_abort_list,
+ preempt_and_abort_list);
+}
+
+static void core_scsi3_release_preempt_and_abort(
+ struct list_head *preempt_and_abort_list,
+ struct t10_pr_registration *pr_reg_holder)
+{
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp;
+
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp, preempt_and_abort_list,
+ pr_reg_abort_list) {
+
+ list_del(&pr_reg->pr_reg_abort_list);
+ if (pr_reg_holder == pr_reg)
+ continue;
+ if (pr_reg->pr_res_holder) {
+ printk(KERN_WARNING "pr_reg->pr_res_holder still set\n");
+ continue;
+ }
+
+ pr_reg->pr_reg_deve = NULL;
+ pr_reg->pr_reg_nacl = NULL;
+ kfree(pr_reg->pr_aptpl_buf);
+ kmem_cache_free(t10_pr_reg_cache, pr_reg);
+ }
+}
+
+int core_scsi3_check_cdb_abort_and_preempt(
+ struct list_head *preempt_and_abort_list,
+ struct se_cmd *cmd)
+{
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp;
+
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp, preempt_and_abort_list,
+ pr_reg_abort_list) {
+ if (pr_reg->pr_res_key == cmd->pr_res_key)
+ return 0;
+ }
+
+ return 1;
+}
+
+static int core_scsi3_pro_preempt(
+ struct se_cmd *cmd,
+ int type,
+ int scope,
+ u64 res_key,
+ u64 sa_res_key,
+ int abort)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_dev_entry *se_deve;
+ struct se_node_acl *pr_reg_nacl;
+ struct se_session *se_sess = SE_SESS(cmd);
+ struct list_head preempt_and_abort_list;
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp, *pr_reg_n, *pr_res_holder;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ u32 pr_res_mapped_lun = 0;
+ 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;
+
+ se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
+ pr_reg_n = core_scsi3_locate_pr_reg(SE_DEV(cmd), se_sess->se_node_acl,
+ se_sess);
+ if (!(pr_reg_n)) {
+ printk(KERN_ERR "SPC-3 PR: Unable to locate"
+ " PR_REGISTERED *pr_reg for PREEMPT%s\n",
+ (abort) ? "_AND_ABORT" : "");
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+ if (pr_reg_n->pr_res_key != res_key) {
+ core_scsi3_put_pr_reg(pr_reg_n);
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+ if (scope != PR_SCOPE_LU_SCOPE) {
+ printk(KERN_ERR "SPC-3 PR: Illegal SCOPE: 0x%02x\n", scope);
+ core_scsi3_put_pr_reg(pr_reg_n);
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ }
+ INIT_LIST_HEAD(&preempt_and_abort_list);
+
+ spin_lock(&dev->dev_reservation_lock);
+ pr_res_holder = dev->dev_pr_res_holder;
+ if (pr_res_holder &&
+ ((pr_res_holder->pr_res_type == PR_TYPE_WRITE_EXCLUSIVE_ALLREG) ||
+ (pr_res_holder->pr_res_type == PR_TYPE_EXCLUSIVE_ACCESS_ALLREG)))
+ all_reg = 1;
+
+ 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;
+ }
+ /*
+ * From spc4r17, section 5.7.11.4.4 Removing Registrations:
+ *
+ * If the SERVICE ACTION RESERVATION KEY field does not identify a
+ * persistent reservation holder or there is no persistent reservation
+ * holder (i.e., there is no persistent reservation), then the device
+ * server shall perform a preempt by doing the following in an
+ * uninterrupted series of actions. (See below..)
+ */
+ if (!(pr_res_holder) || (pr_res_holder->pr_res_key != sa_res_key)) {
+ /*
+ * No existing or SA Reservation Key matching reservations..
+ *
+ * PROUT SA PREEMPT with All Registrant type reservations are
+ * allowed to be processed without a matching SA Reservation Key
+ */
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp,
+ &pr_tmpl->registration_list, pr_reg_list) {
+ /*
+ * Removing of registrations in non all registrants
+ * type reservations without a matching SA reservation
+ * key.
+ *
+ * a) Remove the registrations for all I_T nexuses
+ * specified by the SERVICE ACTION RESERVATION KEY
+ * field;
+ * b) Ignore the contents of the SCOPE and TYPE fields;
+ * c) Process tasks as defined in 5.7.1; and
+ * d) Establish a unit attention condition for the
+ * initiator port associated with every I_T nexus
+ * that lost its registration other than the I_T
+ * nexus on which the PERSISTENT RESERVE OUT command
+ * was received, with the additional sense code set
+ * to REGISTRATIONS PREEMPTED.
+ */
+ if (!(all_reg)) {
+ if (pr_reg->pr_res_key != sa_res_key)
+ continue;
+
+ calling_it_nexus = (pr_reg_n == pr_reg) ? 1 : 0;
+ pr_reg_nacl = pr_reg->pr_reg_nacl;
+ pr_res_mapped_lun = pr_reg->pr_res_mapped_lun;
+ __core_scsi3_free_registration(dev, pr_reg,
+ (abort) ? &preempt_and_abort_list :
+ NULL, calling_it_nexus);
+ released_regs++;
+ } else {
+ /*
+ * Case for any existing all registrants type
+ * reservation, follow logic in spc4r17 section
+ * 5.7.11.4 Preempting, Table 52 and Figure 7.
+ *
+ * For a ZERO SA Reservation key, release
+ * all other registrations and do an implict
+ * release of active persistent reservation.
+ *
+ * For a non-ZERO SA Reservation key, only
+ * release the matching reservation key from
+ * registrations.
+ */
+ if ((sa_res_key) &&
+ (pr_reg->pr_res_key != sa_res_key))
+ continue;
+
+ calling_it_nexus = (pr_reg_n == pr_reg) ? 1 : 0;
+ if (calling_it_nexus)
+ continue;
+
+ pr_reg_nacl = pr_reg->pr_reg_nacl;
+ pr_res_mapped_lun = pr_reg->pr_res_mapped_lun;
+ __core_scsi3_free_registration(dev, pr_reg,
+ (abort) ? &preempt_and_abort_list :
+ NULL, 0);
+ released_regs++;
+ }
+ if (!(calling_it_nexus))
+ core_scsi3_ua_allocate(pr_reg_nacl,
+ pr_res_mapped_lun, 0x2A,
+ ASCQ_2AH_RESERVATIONS_PREEMPTED);
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+ /*
+ * If a PERSISTENT RESERVE OUT with a PREEMPT service action or
+ * a PREEMPT AND ABORT service action sets the SERVICE ACTION
+ * RESERVATION KEY field to a value that does not match any
+ * registered reservation key, then the device server shall
+ * complete the command with RESERVATION CONFLICT status.
+ */
+ if (!(released_regs)) {
+ spin_unlock(&dev->dev_reservation_lock);
+ core_scsi3_put_pr_reg(pr_reg_n);
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+ /*
+ * For an existing all registrants type reservation
+ * with a zero SA rservation key, preempt the existing
+ * reservation with the new PR type and scope.
+ */
+ if (pr_res_holder && all_reg && !(sa_res_key)) {
+ __core_scsi3_complete_pro_preempt(dev, pr_reg_n,
+ (abort) ? &preempt_and_abort_list : NULL,
+ type, scope, abort);
+
+ if (abort)
+ core_scsi3_release_preempt_and_abort(
+ &preempt_and_abort_list, pr_reg_n);
+ }
+ spin_unlock(&dev->dev_reservation_lock);
+
+ if (pr_tmpl->pr_aptpl_active) {
+ ret = core_scsi3_update_and_write_aptpl(SE_DEV(cmd),
+ &pr_reg_n->pr_aptpl_buf[0],
+ pr_tmpl->pr_aptpl_buf_len);
+ if (!(ret))
+ printk(KERN_INFO "SPC-3 PR: Updated APTPL"
+ " metadata for PREEMPT%s\n", (abort) ?
+ "_AND_ABORT" : "");
+ }
+
+ core_scsi3_put_pr_reg(pr_reg_n);
+ core_scsi3_pr_generation(SE_DEV(cmd));
+ return 0;
+ }
+ /*
+ * The PREEMPTing SA reservation key matches that of the
+ * existing persistent reservation, first, we check if
+ * we are preempting our own reservation.
+ * From spc4r17, section 5.7.11.4.3 Preempting
+ * persistent reservations and registration handling
+ *
+ * If an all registrants persistent reservation is not
+ * present, it is not an error for the persistent
+ * reservation holder to preempt itself (i.e., a
+ * PERSISTENT RESERVE OUT with a PREEMPT service action
+ * or a PREEMPT AND ABORT service action with the
+ * SERVICE ACTION RESERVATION KEY value equal to the
+ * persistent reservation holder's reservation key that
+ * is received from the persistent reservation holder).
+ * In that case, the device server shall establish the
+ * new persistent reservation and maintain the
+ * registration.
+ */
+ prh_type = pr_res_holder->pr_res_type;
+ prh_scope = pr_res_holder->pr_res_scope;
+ /*
+ * If the SERVICE ACTION RESERVATION KEY field identifies a
+ * persistent reservation holder (see 5.7.10), the device
+ * server shall perform a preempt by doing the following as
+ * an uninterrupted series of actions:
+ *
+ * a) Release the persistent reservation for the holder
+ * identified by the SERVICE ACTION RESERVATION KEY field;
+ */
+ if (pr_reg_n != pr_res_holder)
+ __core_scsi3_complete_pro_release(dev,
+ pr_res_holder->pr_reg_nacl,
+ dev->dev_pr_res_holder, 0);
+ /*
+ * b) Remove the registrations for all I_T nexuses identified
+ * by the SERVICE ACTION RESERVATION KEY field, except the
+ * I_T nexus that is being used for the PERSISTENT RESERVE
+ * OUT command. If an all registrants persistent reservation
+ * is present and the SERVICE ACTION RESERVATION KEY field
+ * is set to zero, then all registrations shall be removed
+ * except for that of the I_T nexus that is being used for
+ * the PERSISTENT RESERVE OUT command;
+ */
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp,
+ &pr_tmpl->registration_list, pr_reg_list) {
+
+ calling_it_nexus = (pr_reg_n == pr_reg) ? 1 : 0;
+ if (calling_it_nexus)
+ continue;
+
+ if (pr_reg->pr_res_key != sa_res_key)
+ continue;
+
+ pr_reg_nacl = pr_reg->pr_reg_nacl;
+ pr_res_mapped_lun = pr_reg->pr_res_mapped_lun;
+ __core_scsi3_free_registration(dev, pr_reg,
+ (abort) ? &preempt_and_abort_list : NULL,
+ calling_it_nexus);
+ /*
+ * e) Establish a unit attention condition for the initiator
+ * port associated with every I_T nexus that lost its
+ * persistent reservation and/or registration, with the
+ * additional sense code set to REGISTRATIONS PREEMPTED;
+ */
+ core_scsi3_ua_allocate(pr_reg_nacl, pr_res_mapped_lun, 0x2A,
+ ASCQ_2AH_RESERVATIONS_PREEMPTED);
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+ /*
+ * c) Establish a persistent reservation for the preempting
+ * I_T nexus using the contents of the SCOPE and TYPE fields;
+ */
+ __core_scsi3_complete_pro_preempt(dev, pr_reg_n,
+ (abort) ? &preempt_and_abort_list : NULL,
+ type, scope, abort);
+ /*
+ * d) Process tasks as defined in 5.7.1;
+ * e) See above..
+ * f) If the type or scope has changed, then for every I_T nexus
+ * whose reservation key was not removed, except for the I_T
+ * nexus on which the PERSISTENT RESERVE OUT command was
+ * received, the device server shall establish a unit
+ * attention condition for the initiator port associated with
+ * that I_T nexus, with the additional sense code set to
+ * RESERVATIONS RELEASED. If the type or scope have not
+ * changed, then no unit attention condition(s) shall be
+ * established for this reason.
+ */
+ if ((prh_type != type) || (prh_scope != scope)) {
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp,
+ &pr_tmpl->registration_list, pr_reg_list) {
+
+ calling_it_nexus = (pr_reg_n == pr_reg) ? 1 : 0;
+ if (calling_it_nexus)
+ continue;
+
+ core_scsi3_ua_allocate(pr_reg->pr_reg_nacl,
+ pr_reg->pr_res_mapped_lun, 0x2A,
+ ASCQ_2AH_RESERVATIONS_RELEASED);
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+ }
+ spin_unlock(&dev->dev_reservation_lock);
+ /*
+ * Call LUN_RESET logic upon list of struct t10_pr_registration,
+ * All received CDBs for the matching existing reservation and
+ * registrations undergo ABORT_TASK logic.
+ *
+ * From there, core_scsi3_release_preempt_and_abort() will
+ * release every registration in the list (which have already
+ * been removed from the primary pr_reg list), except the
+ * new persistent reservation holder, the calling Initiator Port.
+ */
+ if (abort) {
+ core_tmr_lun_reset(dev, NULL, &preempt_and_abort_list, cmd);
+ core_scsi3_release_preempt_and_abort(&preempt_and_abort_list,
+ pr_reg_n);
+ }
+
+ if (pr_tmpl->pr_aptpl_active) {
+ ret = core_scsi3_update_and_write_aptpl(SE_DEV(cmd),
+ &pr_reg_n->pr_aptpl_buf[0],
+ pr_tmpl->pr_aptpl_buf_len);
+ if (!(ret))
+ printk("SPC-3 PR: Updated APTPL metadata for PREEMPT"
+ "%s\n", (abort) ? "_AND_ABORT" : "");
+ }
+
+ core_scsi3_put_pr_reg(pr_reg_n);
+ core_scsi3_pr_generation(SE_DEV(cmd));
+ return 0;
+}
+
+static int core_scsi3_emulate_pro_preempt(
+ struct se_cmd *cmd,
+ int type,
+ int scope,
+ u64 res_key,
+ u64 sa_res_key,
+ int abort)
+{
+ int ret = 0;
+
+ switch (type) {
+ case PR_TYPE_WRITE_EXCLUSIVE:
+ case PR_TYPE_EXCLUSIVE_ACCESS:
+ case PR_TYPE_WRITE_EXCLUSIVE_REGONLY:
+ case PR_TYPE_EXCLUSIVE_ACCESS_REGONLY:
+ case PR_TYPE_WRITE_EXCLUSIVE_ALLREG:
+ case PR_TYPE_EXCLUSIVE_ACCESS_ALLREG:
+ ret = core_scsi3_pro_preempt(cmd, type, scope,
+ res_key, sa_res_key, abort);
+ break;
+ default:
+ printk(KERN_ERR "SPC-3 PR: Unknown Service Action PREEMPT%s"
+ " Type: 0x%02x\n", (abort) ? "_AND_ABORT" : "", type);
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+
+ return ret;
+}
+
+
+static int core_scsi3_emulate_pro_register_and_move(
+ struct se_cmd *cmd,
+ u64 res_key,
+ u64 sa_res_key,
+ int aptpl,
+ int unreg)
+{
+ struct se_session *se_sess = SE_SESS(cmd);
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_dev_entry *se_deve, *dest_se_deve = NULL;
+ struct se_lun *se_lun = SE_LUN(cmd);
+ struct se_node_acl *pr_res_nacl, *pr_reg_nacl, *dest_node_acl = NULL;
+ struct se_port *se_port;
+ struct se_portal_group *se_tpg, *dest_se_tpg = NULL;
+ struct target_core_fabric_ops *dest_tf_ops = NULL, *tf_ops;
+ struct t10_pr_registration *pr_reg, *pr_res_holder, *dest_pr_reg;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ unsigned char *initiator_str;
+ char *iport_ptr = NULL, dest_iport[64], i_buf[PR_REG_ISID_ID_LEN];
+ u32 tid_len, tmp_tid_len;
+ int new_reg = 0, type, scope, ret, matching_iname, prf_isid;
+ unsigned short rtpi;
+ unsigned char proto_ident;
+
+ if (!(se_sess) || !(se_lun)) {
+ printk(KERN_ERR "SPC-3 PR: se_sess || struct se_lun is NULL!\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ memset(dest_iport, 0, 64);
+ memset(i_buf, 0, PR_REG_ISID_ID_LEN);
+ se_tpg = se_sess->se_tpg;
+ tf_ops = TPG_TFO(se_tpg);
+ se_deve = &se_sess->se_node_acl->device_list[cmd->orig_fe_lun];
+ /*
+ * Follow logic from spc4r17 Section 5.7.8, Table 50 --
+ * Register behaviors for a REGISTER AND MOVE service action
+ *
+ * Locate the existing *pr_reg via struct se_node_acl pointers
+ */
+ pr_reg = core_scsi3_locate_pr_reg(SE_DEV(cmd), se_sess->se_node_acl,
+ se_sess);
+ if (!(pr_reg)) {
+ printk(KERN_ERR "SPC-3 PR: Unable to locate PR_REGISTERED"
+ " *pr_reg for REGISTER_AND_MOVE\n");
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ /*
+ * The provided reservation key much match the existing reservation key
+ * provided during this initiator's I_T nexus registration.
+ */
+ if (res_key != pr_reg->pr_res_key) {
+ printk(KERN_WARNING "SPC-3 PR REGISTER_AND_MOVE: Received"
+ " 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;
+ }
+ /*
+ * The service active reservation key needs to be non zero
+ */
+ if (!(sa_res_key)) {
+ printk(KERN_WARNING "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;
+ }
+ /*
+ * Determine the Relative Target Port Identifier where the reservation
+ * will be moved to for the TransportID containing SCSI initiator WWN
+ * information.
+ */
+ rtpi = (buf[18] & 0xff) << 8;
+ rtpi |= buf[19] & 0xff;
+ tid_len = (buf[20] & 0xff) << 24;
+ tid_len |= (buf[21] & 0xff) << 16;
+ tid_len |= (buf[22] & 0xff) << 8;
+ tid_len |= buf[23] & 0xff;
+
+ if ((tid_len + 24) != cmd->data_length) {
+ printk(KERN_ERR "SPC-3 PR: Illegal tid_len: %u + 24 byte header"
+ " 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;
+ }
+
+ spin_lock(&dev->se_port_lock);
+ list_for_each_entry(se_port, &dev->dev_sep_list, sep_list) {
+ if (se_port->sep_rtpi != rtpi)
+ continue;
+ dest_se_tpg = se_port->sep_tpg;
+ if (!(dest_se_tpg))
+ continue;
+ dest_tf_ops = TPG_TFO(dest_se_tpg);
+ if (!(dest_tf_ops))
+ continue;
+
+ atomic_inc(&dest_se_tpg->tpg_pr_ref_count);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&dev->se_port_lock);
+
+ ret = core_scsi3_tpg_depend_item(dest_se_tpg);
+ if (ret != 0) {
+ printk(KERN_ERR "core_scsi3_tpg_depend_item() failed"
+ " for dest_se_tpg\n");
+ 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;
+ }
+
+ spin_lock(&dev->se_port_lock);
+ break;
+ }
+ spin_unlock(&dev->se_port_lock);
+
+ if (!(dest_se_tpg) || (!dest_tf_ops)) {
+ printk(KERN_ERR "SPC-3 PR REGISTER_AND_MOVE: Unable to locate"
+ " fabric ops from Relative Target Port Identifier:"
+ " %hu\n", rtpi);
+ core_scsi3_put_pr_reg(pr_reg);
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ }
+ proto_ident = (buf[24] & 0x0f);
+#if 0
+ printk("SPC-3 PR REGISTER_AND_MOVE: Extracted Protocol Identifier:"
+ " 0x%02x\n", proto_ident);
+#endif
+ if (proto_ident != dest_tf_ops->get_fabric_proto_ident(dest_se_tpg)) {
+ printk(KERN_ERR "SPC-3 PR REGISTER_AND_MOVE: Received"
+ " proto_ident: 0x%02x does not match ident: 0x%02x"
+ " 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;
+ goto out;
+ }
+ if (dest_tf_ops->tpg_parse_pr_out_transport_id == NULL) {
+ printk(KERN_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;
+ goto out;
+ }
+ initiator_str = dest_tf_ops->tpg_parse_pr_out_transport_id(dest_se_tpg,
+ (const char *)&buf[24], &tmp_tid_len, &iport_ptr);
+ if (!(initiator_str)) {
+ printk(KERN_ERR "SPC-3 PR REGISTER_AND_MOVE: Unable to locate"
+ " initiator_str from Transport ID\n");
+ ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ goto out;
+ }
+
+ printk(KERN_INFO "SPC-3 PR [%s] Extracted initiator %s identifier: %s"
+ " %s\n", dest_tf_ops->get_fabric_name(), (iport_ptr != NULL) ?
+ "port" : "device", initiator_str, (iport_ptr != NULL) ?
+ iport_ptr : "");
+ /*
+ * If a PERSISTENT RESERVE OUT command with a REGISTER AND MOVE service
+ * action specifies a TransportID that is the same as the initiator port
+ * of the I_T nexus for the command received, then the command shall
+ * be terminated with CHECK CONDITION status, with the sense key set to
+ * ILLEGAL REQUEST, and the additional sense code set to INVALID FIELD
+ * IN PARAMETER LIST.
+ */
+ pr_reg_nacl = pr_reg->pr_reg_nacl;
+ matching_iname = (!strcmp(initiator_str,
+ pr_reg_nacl->initiatorname)) ? 1 : 0;
+ if (!(matching_iname))
+ goto after_iport_check;
+
+ if (!(iport_ptr) || !(pr_reg->isid_present_at_reg)) {
+ printk(KERN_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;
+ goto out;
+ }
+ if (!(strcmp(iport_ptr, pr_reg->pr_reg_isid))) {
+ printk(KERN_ERR "SPC-3 PR REGISTER_AND_MOVE: TransportID: %s %s"
+ " 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;
+ goto out;
+ }
+after_iport_check:
+ /*
+ * Locate the destination struct se_node_acl from the received Transport ID
+ */
+ spin_lock_bh(&dest_se_tpg->acl_node_lock);
+ dest_node_acl = __core_tpg_get_initiator_node_acl(dest_se_tpg,
+ initiator_str);
+ if (dest_node_acl) {
+ atomic_inc(&dest_node_acl->acl_pr_ref_count);
+ smp_mb__after_atomic_inc();
+ }
+ spin_unlock_bh(&dest_se_tpg->acl_node_lock);
+
+ if (!(dest_node_acl)) {
+ printk(KERN_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;
+ goto out;
+ }
+ ret = core_scsi3_nodeacl_depend_item(dest_node_acl);
+ if (ret != 0) {
+ printk(KERN_ERR "core_scsi3_nodeacl_depend_item() for"
+ " dest_node_acl\n");
+ atomic_dec(&dest_node_acl->acl_pr_ref_count);
+ smp_mb__after_atomic_dec();
+ dest_node_acl = NULL;
+ ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+ goto out;
+ }
+#if 0
+ printk(KERN_INFO "SPC-3 PR REGISTER_AND_MOVE: Found %s dest_node_acl:"
+ " %s from TransportID\n", dest_tf_ops->get_fabric_name(),
+ dest_node_acl->initiatorname);
+#endif
+ /*
+ * Locate the struct se_dev_entry pointer for the matching RELATIVE TARGET
+ * PORT IDENTIFIER.
+ */
+ dest_se_deve = core_get_se_deve_from_rtpi(dest_node_acl, rtpi);
+ if (!(dest_se_deve)) {
+ printk(KERN_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;
+ goto out;
+ }
+
+ ret = core_scsi3_lunacl_depend_item(dest_se_deve);
+ if (ret < 0) {
+ printk(KERN_ERR "core_scsi3_lunacl_depend_item() failed\n");
+ atomic_dec(&dest_se_deve->pr_ref_count);
+ smp_mb__after_atomic_dec();
+ dest_se_deve = NULL;
+ ret = PYX_TRANSPORT_LU_COMM_FAILURE;
+ goto out;
+ }
+#if 0
+ printk(KERN_INFO "SPC-3 PR REGISTER_AND_MOVE: Located %s node %s LUN"
+ " ACL for dest_se_deve->mapped_lun: %u\n",
+ dest_tf_ops->get_fabric_name(), dest_node_acl->initiatorname,
+ dest_se_deve->mapped_lun);
+#endif
+ /*
+ * A persistent reservation needs to already existing in order to
+ * successfully complete the REGISTER_AND_MOVE service action..
+ */
+ spin_lock(&dev->dev_reservation_lock);
+ pr_res_holder = dev->dev_pr_res_holder;
+ if (!(pr_res_holder)) {
+ printk(KERN_WARNING "SPC-3 PR REGISTER_AND_MOVE: No reservation"
+ " currently held\n");
+ spin_unlock(&dev->dev_reservation_lock);
+ ret = PYX_TRANSPORT_INVALID_CDB_FIELD;
+ goto out;
+ }
+ /*
+ * The received on I_T Nexus must be the reservation holder.
+ *
+ * From spc4r17 section 5.7.8 Table 50 --
+ * Register behaviors for a REGISTER AND MOVE service action
+ */
+ if (pr_res_holder != pr_reg) {
+ printk(KERN_WARNING "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;
+ goto out;
+ }
+ /*
+ * From spc4r17 section 5.7.8: registering and moving reservation
+ *
+ * If a PERSISTENT RESERVE OUT command with a REGISTER AND MOVE service
+ * action is received and the established persistent reservation is a
+ * Write Exclusive - All Registrants type or Exclusive Access -
+ * All Registrants type reservation, then the command shall be completed
+ * with RESERVATION CONFLICT status.
+ */
+ if ((pr_res_holder->pr_res_type == PR_TYPE_WRITE_EXCLUSIVE_ALLREG) ||
+ (pr_res_holder->pr_res_type == PR_TYPE_EXCLUSIVE_ACCESS_ALLREG)) {
+ printk(KERN_WARNING "SPC-3 PR REGISTER_AND_MOVE: Unable to move"
+ " 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;
+ goto out;
+ }
+ pr_res_nacl = pr_res_holder->pr_reg_nacl;
+ /*
+ * b) Ignore the contents of the (received) SCOPE and TYPE fields;
+ */
+ type = pr_res_holder->pr_res_type;
+ scope = pr_res_holder->pr_res_type;
+ /*
+ * c) Associate the reservation key specified in the SERVICE ACTION
+ * RESERVATION KEY field with the I_T nexus specified as the
+ * destination of the register and move, where:
+ * A) The I_T nexus is specified by the TransportID and the
+ * RELATIVE TARGET PORT IDENTIFIER field (see 6.14.4); and
+ * B) Regardless of the TransportID format used, the association for
+ * the initiator port is based on either the initiator port name
+ * (see 3.1.71) on SCSI transport protocols where port names are
+ * required or the initiator port identifier (see 3.1.70) on SCSI
+ * transport protocols where port names are not required;
+ * d) Register the reservation key specified in the SERVICE ACTION
+ * RESERVATION KEY field;
+ * e) Retain the reservation key specified in the SERVICE ACTION
+ * RESERVATION KEY field and associated information;
+ *
+ * Also, It is not an error for a REGISTER AND MOVE service action to
+ * register an I_T nexus that is already registered with the same
+ * reservation key or a different reservation key.
+ */
+ dest_pr_reg = __core_scsi3_locate_pr_reg(dev, dest_node_acl,
+ iport_ptr);
+ if (!(dest_pr_reg)) {
+ ret = core_scsi3_alloc_registration(SE_DEV(cmd),
+ dest_node_acl, dest_se_deve, iport_ptr,
+ sa_res_key, 0, aptpl, 2, 1);
+ if (ret != 0) {
+ spin_unlock(&dev->dev_reservation_lock);
+ ret = PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ goto out;
+ }
+ dest_pr_reg = __core_scsi3_locate_pr_reg(dev, dest_node_acl,
+ iport_ptr);
+ new_reg = 1;
+ }
+ /*
+ * f) Release the persistent reservation for the persistent reservation
+ * holder (i.e., the I_T nexus on which the
+ */
+ __core_scsi3_complete_pro_release(dev, pr_res_nacl,
+ dev->dev_pr_res_holder, 0);
+ /*
+ * g) Move the persistent reservation to the specified I_T nexus using
+ * the same scope and type as the persistent reservation released in
+ * item f); and
+ */
+ dev->dev_pr_res_holder = dest_pr_reg;
+ dest_pr_reg->pr_res_holder = 1;
+ dest_pr_reg->pr_res_type = type;
+ pr_reg->pr_res_scope = scope;
+ prf_isid = core_pr_dump_initiator_port(pr_reg, &i_buf[0],
+ PR_REG_ISID_ID_LEN);
+ /*
+ * Increment PRGeneration for existing registrations..
+ */
+ if (!(new_reg))
+ dest_pr_reg->pr_res_generation = pr_tmpl->pr_generation++;
+ spin_unlock(&dev->dev_reservation_lock);
+
+ printk(KERN_INFO "SPC-3 PR [%s] Service Action: REGISTER_AND_MOVE"
+ " created new reservation holder TYPE: %s on object RTPI:"
+ " %hu PRGeneration: 0x%08x\n", dest_tf_ops->get_fabric_name(),
+ core_scsi3_pr_dump_type(type), rtpi,
+ dest_pr_reg->pr_res_generation);
+ printk(KERN_INFO "SPC-3 PR Successfully moved reservation from"
+ " %s Fabric Node: %s%s -> %s Fabric Node: %s %s\n",
+ tf_ops->get_fabric_name(), pr_reg_nacl->initiatorname,
+ (prf_isid) ? &i_buf[0] : "", dest_tf_ops->get_fabric_name(),
+ dest_node_acl->initiatorname, (iport_ptr != NULL) ?
+ iport_ptr : "");
+ /*
+ * It is now safe to release configfs group dependencies for destination
+ * of Transport ID Initiator Device/Port Identifier
+ */
+ core_scsi3_lunacl_undepend_item(dest_se_deve);
+ core_scsi3_nodeacl_undepend_item(dest_node_acl);
+ core_scsi3_tpg_undepend_item(dest_se_tpg);
+ /*
+ * h) If the UNREG bit is set to one, unregister (see 5.7.11.3) the I_T
+ * nexus on which PERSISTENT RESERVE OUT command was received.
+ */
+ if (unreg) {
+ spin_lock(&pr_tmpl->registration_lock);
+ __core_scsi3_free_registration(dev, pr_reg, NULL, 1);
+ spin_unlock(&pr_tmpl->registration_lock);
+ } else
+ core_scsi3_put_pr_reg(pr_reg);
+
+ /*
+ * Clear the APTPL metadata if APTPL has been disabled, otherwise
+ * write out the updated metadata to struct file for this SCSI device.
+ */
+ if (!(aptpl)) {
+ pr_tmpl->pr_aptpl_active = 0;
+ core_scsi3_update_and_write_aptpl(SE_DEV(cmd), NULL, 0);
+ printk("SPC-3 PR: Set APTPL Bit Deactivated for"
+ " REGISTER_AND_MOVE\n");
+ } else {
+ pr_tmpl->pr_aptpl_active = 1;
+ ret = core_scsi3_update_and_write_aptpl(SE_DEV(cmd),
+ &dest_pr_reg->pr_aptpl_buf[0],
+ pr_tmpl->pr_aptpl_buf_len);
+ if (!(ret))
+ printk("SPC-3 PR: Set APTPL Bit Activated for"
+ " REGISTER_AND_MOVE\n");
+ }
+
+ core_scsi3_put_pr_reg(dest_pr_reg);
+ return 0;
+out:
+ if (dest_se_deve)
+ core_scsi3_lunacl_undepend_item(dest_se_deve);
+ if (dest_node_acl)
+ core_scsi3_nodeacl_undepend_item(dest_node_acl);
+ core_scsi3_tpg_undepend_item(dest_se_tpg);
+ core_scsi3_put_pr_reg(pr_reg);
+ return ret;
+}
+
+static unsigned long long core_scsi3_extract_reservation_key(unsigned char *cdb)
+{
+ unsigned int __v1, __v2;
+
+ __v1 = (cdb[0] << 24) | (cdb[1] << 16) | (cdb[2] << 8) | cdb[3];
+ __v2 = (cdb[4] << 24) | (cdb[5] << 16) | (cdb[6] << 8) | cdb[7];
+
+ return ((unsigned long long)__v2) | (unsigned long long)__v1 << 32;
+}
+
+/*
+ * See spc4r17 section 6.14 Table 170
+ */
+static int core_scsi3_emulate_pr_out(struct se_cmd *cmd, unsigned char *cdb)
+{
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ u64 res_key, sa_res_key;
+ int sa, scope, type, aptpl;
+ int spec_i_pt = 0, all_tg_pt = 0, unreg = 0;
+ /*
+ * 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 (!(SE_SESS(cmd)))
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+
+ if (cmd->data_length < 24) {
+ printk(KERN_WARNING "SPC-PR: Recieved PR OUT parameter list"
+ " length too small: %u\n", cmd->data_length);
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ }
+ /*
+ * From the PERSISTENT_RESERVE_OUT command descriptor block (CDB)
+ */
+ sa = (cdb[1] & 0x1f);
+ scope = (cdb[2] & 0xf0);
+ type = (cdb[2] & 0x0f);
+ /*
+ * From PERSISTENT_RESERVE_OUT parameter list (payload)
+ */
+ res_key = core_scsi3_extract_reservation_key(&buf[0]);
+ sa_res_key = core_scsi3_extract_reservation_key(&buf[8]);
+ /*
+ * REGISTER_AND_MOVE uses a different SA parameter list containing
+ * SCSI TransportIDs.
+ */
+ if (sa != PRO_REGISTER_AND_MOVE) {
+ spec_i_pt = (buf[20] & 0x08);
+ all_tg_pt = (buf[20] & 0x04);
+ aptpl = (buf[20] & 0x01);
+ } else {
+ aptpl = (buf[17] & 0x01);
+ unreg = (buf[17] & 0x02);
+ }
+ /*
+ * SPEC_I_PT=1 is only valid for Service action: REGISTER
+ */
+ if (spec_i_pt && ((cdb[1] & 0x1f) != PRO_REGISTER))
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ /*
+ * From spc4r17 section 6.14:
+ *
+ * If the SPEC_I_PT bit is set to zero, the service action is not
+ * REGISTER AND MOVE, and the parameter list length is not 24, then
+ * the command shall be terminated with CHECK CONDITION status, with
+ * the sense key set to ILLEGAL REQUEST, and the additional sense
+ * code set to PARAMETER LIST LENGTH ERROR.
+ */
+ if (!(spec_i_pt) && ((cdb[1] & 0x1f) != PRO_REGISTER_AND_MOVE) &&
+ (cmd->data_length != 24)) {
+ printk(KERN_WARNING "SPC-PR: Recieved PR OUT illegal parameter"
+ " list length: %u\n", cmd->data_length);
+ return PYX_TRANSPORT_INVALID_PARAMETER_LIST;
+ }
+ /*
+ * (core_scsi3_emulate_pro_* function parameters
+ * are defined by spc4r17 Table 174:
+ * PERSISTENT_RESERVE_OUT service actions and valid parameters.
+ */
+ switch (sa) {
+ case PRO_REGISTER:
+ return core_scsi3_emulate_pro_register(cmd,
+ res_key, sa_res_key, aptpl, all_tg_pt, spec_i_pt, 0);
+ case PRO_RESERVE:
+ return core_scsi3_emulate_pro_reserve(cmd,
+ type, scope, res_key);
+ case PRO_RELEASE:
+ return core_scsi3_emulate_pro_release(cmd,
+ type, scope, res_key);
+ case PRO_CLEAR:
+ return core_scsi3_emulate_pro_clear(cmd, res_key);
+ case PRO_PREEMPT:
+ return core_scsi3_emulate_pro_preempt(cmd, type, scope,
+ res_key, sa_res_key, 0);
+ case PRO_PREEMPT_AND_ABORT:
+ return core_scsi3_emulate_pro_preempt(cmd, type, scope,
+ res_key, sa_res_key, 1);
+ case PRO_REGISTER_AND_IGNORE_EXISTING_KEY:
+ return core_scsi3_emulate_pro_register(cmd,
+ 0, sa_res_key, aptpl, all_tg_pt, spec_i_pt, 1);
+ case PRO_REGISTER_AND_MOVE:
+ return core_scsi3_emulate_pro_register_and_move(cmd, res_key,
+ sa_res_key, aptpl, unreg);
+ default:
+ printk(KERN_ERR "Unknown PERSISTENT_RESERVE_OUT service"
+ " action: 0x%02x\n", cdb[1] & 0x1f);
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+}
+
+/*
+ * PERSISTENT_RESERVE_IN Service Action READ_KEYS
+ *
+ * See spc4r17 section 5.7.6.2 and section 6.13.2, Table 160
+ */
+static int core_scsi3_pri_read_keys(struct se_cmd *cmd)
+{
+ struct se_device *se_dev = SE_DEV(cmd);
+ struct se_subsystem_dev *su_dev = SU_DEV(se_dev);
+ struct t10_pr_registration *pr_reg;
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ u32 add_len = 0, off = 8;
+
+ if (cmd->data_length < 8) {
+ printk(KERN_ERR "PRIN SA READ_KEYS SCSI Data Length: %u"
+ " too small\n", cmd->data_length);
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+
+ buf[0] = ((T10_RES(su_dev)->pr_generation >> 24) & 0xff);
+ buf[1] = ((T10_RES(su_dev)->pr_generation >> 16) & 0xff);
+ buf[2] = ((T10_RES(su_dev)->pr_generation >> 8) & 0xff);
+ buf[3] = (T10_RES(su_dev)->pr_generation & 0xff);
+
+ spin_lock(&T10_RES(su_dev)->registration_lock);
+ list_for_each_entry(pr_reg, &T10_RES(su_dev)->registration_list,
+ pr_reg_list) {
+ /*
+ * Check for overflow of 8byte PRI READ_KEYS payload and
+ * next reservation key list descriptor.
+ */
+ if ((add_len + 8) > (cmd->data_length - 8))
+ break;
+
+ buf[off++] = ((pr_reg->pr_res_key >> 56) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 48) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 40) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 32) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 24) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 16) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 8) & 0xff);
+ buf[off++] = (pr_reg->pr_res_key & 0xff);
+
+ add_len += 8;
+ }
+ spin_unlock(&T10_RES(su_dev)->registration_lock);
+
+ buf[4] = ((add_len >> 24) & 0xff);
+ buf[5] = ((add_len >> 16) & 0xff);
+ buf[6] = ((add_len >> 8) & 0xff);
+ buf[7] = (add_len & 0xff);
+
+ return 0;
+}
+
+/*
+ * PERSISTENT_RESERVE_IN Service Action READ_RESERVATION
+ *
+ * See spc4r17 section 5.7.6.3 and section 6.13.3.2 Table 161 and 162
+ */
+static int core_scsi3_pri_read_reservation(struct se_cmd *cmd)
+{
+ struct se_device *se_dev = SE_DEV(cmd);
+ struct se_subsystem_dev *su_dev = SU_DEV(se_dev);
+ struct t10_pr_registration *pr_reg;
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ u64 pr_res_key;
+ u32 add_len = 16; /* Hardcoded to 16 when a reservation is held. */
+
+ if (cmd->data_length < 8) {
+ printk(KERN_ERR "PRIN SA READ_RESERVATIONS SCSI Data Length: %u"
+ " too small\n", cmd->data_length);
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+
+ buf[0] = ((T10_RES(su_dev)->pr_generation >> 24) & 0xff);
+ buf[1] = ((T10_RES(su_dev)->pr_generation >> 16) & 0xff);
+ buf[2] = ((T10_RES(su_dev)->pr_generation >> 8) & 0xff);
+ buf[3] = (T10_RES(su_dev)->pr_generation & 0xff);
+
+ spin_lock(&se_dev->dev_reservation_lock);
+ pr_reg = se_dev->dev_pr_res_holder;
+ if ((pr_reg)) {
+ /*
+ * Set the hardcoded Additional Length
+ */
+ buf[4] = ((add_len >> 24) & 0xff);
+ buf[5] = ((add_len >> 16) & 0xff);
+ buf[6] = ((add_len >> 8) & 0xff);
+ buf[7] = (add_len & 0xff);
+
+ if (cmd->data_length < 22) {
+ spin_unlock(&se_dev->dev_reservation_lock);
+ return 0;
+ }
+ /*
+ * Set the Reservation key.
+ *
+ * From spc4r17, section 5.7.10:
+ * A persistent reservation holder has its reservation key
+ * returned in the parameter data from a PERSISTENT
+ * RESERVE IN command with READ RESERVATION service action as
+ * follows:
+ * a) For a persistent reservation of the type Write Exclusive
+ * - All Registrants or Exclusive Access All Regitrants,
+ * the reservation key shall be set to zero; or
+ * b) For all other persistent reservation types, the
+ * reservation key shall be set to the registered
+ * reservation key for the I_T nexus that holds the
+ * persistent reservation.
+ */
+ if ((pr_reg->pr_res_type == PR_TYPE_WRITE_EXCLUSIVE_ALLREG) ||
+ (pr_reg->pr_res_type == PR_TYPE_EXCLUSIVE_ACCESS_ALLREG))
+ pr_res_key = 0;
+ else
+ pr_res_key = pr_reg->pr_res_key;
+
+ buf[8] = ((pr_res_key >> 56) & 0xff);
+ buf[9] = ((pr_res_key >> 48) & 0xff);
+ buf[10] = ((pr_res_key >> 40) & 0xff);
+ buf[11] = ((pr_res_key >> 32) & 0xff);
+ buf[12] = ((pr_res_key >> 24) & 0xff);
+ buf[13] = ((pr_res_key >> 16) & 0xff);
+ buf[14] = ((pr_res_key >> 8) & 0xff);
+ buf[15] = (pr_res_key & 0xff);
+ /*
+ * Set the SCOPE and TYPE
+ */
+ buf[21] = (pr_reg->pr_res_scope & 0xf0) |
+ (pr_reg->pr_res_type & 0x0f);
+ }
+ spin_unlock(&se_dev->dev_reservation_lock);
+
+ return 0;
+}
+
+/*
+ * PERSISTENT_RESERVE_IN Service Action REPORT_CAPABILITIES
+ *
+ * See spc4r17 section 6.13.4 Table 165
+ */
+static int core_scsi3_pri_report_capabilities(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(dev)->t10_reservation;
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ u16 add_len = 8; /* Hardcoded to 8. */
+
+ if (cmd->data_length < 6) {
+ printk(KERN_ERR "PRIN SA REPORT_CAPABILITIES SCSI Data Length:"
+ " %u too small\n", cmd->data_length);
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+
+ buf[0] = ((add_len << 8) & 0xff);
+ buf[1] = (add_len & 0xff);
+ buf[2] |= 0x10; /* CRH: Compatible Reservation Hanlding bit. */
+ buf[2] |= 0x08; /* SIP_C: Specify Initiator Ports Capable bit */
+ buf[2] |= 0x04; /* ATP_C: All Target Ports Capable bit */
+ buf[2] |= 0x01; /* PTPL_C: Persistence across Target Power Loss bit */
+ /*
+ * We are filling in the PERSISTENT RESERVATION TYPE MASK below, so
+ * set the TMV: Task Mask Valid bit.
+ */
+ buf[3] |= 0x80;
+ /*
+ * Change ALLOW COMMANDs to 0x20 or 0x40 later from Table 166
+ */
+ buf[3] |= 0x10; /* ALLOW COMMANDs field 001b */
+ /*
+ * PTPL_A: Persistence across Target Power Loss Active bit
+ */
+ if (pr_tmpl->pr_aptpl_active)
+ buf[3] |= 0x01;
+ /*
+ * Setup the PERSISTENT RESERVATION TYPE MASK from Table 167
+ */
+ buf[4] |= 0x80; /* PR_TYPE_EXCLUSIVE_ACCESS_ALLREG */
+ buf[4] |= 0x40; /* PR_TYPE_EXCLUSIVE_ACCESS_REGONLY */
+ buf[4] |= 0x20; /* PR_TYPE_WRITE_EXCLUSIVE_REGONLY */
+ buf[4] |= 0x08; /* PR_TYPE_EXCLUSIVE_ACCESS */
+ buf[4] |= 0x02; /* PR_TYPE_WRITE_EXCLUSIVE */
+ buf[5] |= 0x01; /* PR_TYPE_EXCLUSIVE_ACCESS_ALLREG */
+
+ return 0;
+}
+
+/*
+ * PERSISTENT_RESERVE_IN Service Action READ_FULL_STATUS
+ *
+ * See spc4r17 section 6.13.5 Table 168 and 169
+ */
+static int core_scsi3_pri_read_full_status(struct se_cmd *cmd)
+{
+ struct se_device *se_dev = SE_DEV(cmd);
+ struct se_node_acl *se_nacl;
+ struct se_subsystem_dev *su_dev = SU_DEV(se_dev);
+ struct se_portal_group *se_tpg;
+ struct t10_pr_registration *pr_reg, *pr_reg_tmp;
+ struct t10_reservation_template *pr_tmpl = &SU_DEV(se_dev)->t10_reservation;
+ unsigned char *buf = (unsigned char *)T_TASK(cmd)->t_task_buf;
+ u32 add_desc_len = 0, add_len = 0, desc_len, exp_desc_len;
+ u32 off = 8; /* off into first Full Status descriptor */
+ int format_code = 0;
+
+ if (cmd->data_length < 8) {
+ printk(KERN_ERR "PRIN SA READ_FULL_STATUS SCSI Data Length: %u"
+ " too small\n", cmd->data_length);
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+
+ buf[0] = ((T10_RES(su_dev)->pr_generation >> 24) & 0xff);
+ buf[1] = ((T10_RES(su_dev)->pr_generation >> 16) & 0xff);
+ buf[2] = ((T10_RES(su_dev)->pr_generation >> 8) & 0xff);
+ buf[3] = (T10_RES(su_dev)->pr_generation & 0xff);
+
+ spin_lock(&pr_tmpl->registration_lock);
+ list_for_each_entry_safe(pr_reg, pr_reg_tmp,
+ &pr_tmpl->registration_list, pr_reg_list) {
+
+ se_nacl = pr_reg->pr_reg_nacl;
+ se_tpg = pr_reg->pr_reg_nacl->se_tpg;
+ add_desc_len = 0;
+
+ atomic_inc(&pr_reg->pr_res_holders);
+ smp_mb__after_atomic_inc();
+ spin_unlock(&pr_tmpl->registration_lock);
+ /*
+ * Determine expected length of $FABRIC_MOD specific
+ * TransportID full status descriptor..
+ */
+ exp_desc_len = TPG_TFO(se_tpg)->tpg_get_pr_transport_id_len(
+ se_tpg, se_nacl, pr_reg, &format_code);
+
+ if ((exp_desc_len + add_len) > cmd->data_length) {
+ printk(KERN_WARNING "SPC-3 PRIN READ_FULL_STATUS ran"
+ " out of buffer: %d\n", cmd->data_length);
+ spin_lock(&pr_tmpl->registration_lock);
+ atomic_dec(&pr_reg->pr_res_holders);
+ smp_mb__after_atomic_dec();
+ break;
+ }
+ /*
+ * Set RESERVATION KEY
+ */
+ buf[off++] = ((pr_reg->pr_res_key >> 56) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 48) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 40) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 32) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 24) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 16) & 0xff);
+ buf[off++] = ((pr_reg->pr_res_key >> 8) & 0xff);
+ buf[off++] = (pr_reg->pr_res_key & 0xff);
+ off += 4; /* Skip Over Reserved area */
+
+ /*
+ * Set ALL_TG_PT bit if PROUT SA REGISTER had this set.
+ */
+ if (pr_reg->pr_reg_all_tg_pt)
+ buf[off] = 0x02;
+ /*
+ * The struct se_lun pointer will be present for the
+ * reservation holder for PR_HOLDER bit.
+ *
+ * Also, if this registration is the reservation
+ * holder, fill in SCOPE and TYPE in the next byte.
+ */
+ if (pr_reg->pr_res_holder) {
+ buf[off++] |= 0x01;
+ buf[off++] = (pr_reg->pr_res_scope & 0xf0) |
+ (pr_reg->pr_res_type & 0x0f);
+ } else
+ off += 2;
+
+ off += 4; /* Skip over reserved area */
+ /*
+ * From spc4r17 6.3.15:
+ *
+ * If the ALL_TG_PT bit set to zero, the RELATIVE TARGET PORT
+ * IDENTIFIER field contains the relative port identifier (see
+ * 3.1.120) of the target port that is part of the I_T nexus
+ * described by this full status descriptor. If the ALL_TG_PT
+ * bit is set to one, the contents of the RELATIVE TARGET PORT
+ * IDENTIFIER field are not defined by this standard.
+ */
+ if (!(pr_reg->pr_reg_all_tg_pt)) {
+ struct se_port *port = pr_reg->pr_reg_tg_pt_lun->lun_sep;
+
+ buf[off++] = ((port->sep_rtpi >> 8) & 0xff);
+ buf[off++] = (port->sep_rtpi & 0xff);
+ } else
+ off += 2; /* Skip over RELATIVE TARGET PORT IDENTIFER */
+
+ /*
+ * Now, have the $FABRIC_MOD fill in the protocol identifier
+ */
+ desc_len = TPG_TFO(se_tpg)->tpg_get_pr_transport_id(se_tpg,
+ se_nacl, pr_reg, &format_code, &buf[off+4]);
+
+ spin_lock(&pr_tmpl->registration_lock);
+ atomic_dec(&pr_reg->pr_res_holders);
+ smp_mb__after_atomic_dec();
+ /*
+ * Set the ADDITIONAL DESCRIPTOR LENGTH
+ */
+ buf[off++] = ((desc_len >> 24) & 0xff);
+ buf[off++] = ((desc_len >> 16) & 0xff);
+ buf[off++] = ((desc_len >> 8) & 0xff);
+ buf[off++] = (desc_len & 0xff);
+ /*
+ * Size of full desctipor header minus TransportID
+ * containing $FABRIC_MOD specific) initiator device/port
+ * WWN information.
+ *
+ * See spc4r17 Section 6.13.5 Table 169
+ */
+ add_desc_len = (24 + desc_len);
+
+ off += desc_len;
+ add_len += add_desc_len;
+ }
+ spin_unlock(&pr_tmpl->registration_lock);
+ /*
+ * Set ADDITIONAL_LENGTH
+ */
+ buf[4] = ((add_len >> 24) & 0xff);
+ buf[5] = ((add_len >> 16) & 0xff);
+ buf[6] = ((add_len >> 8) & 0xff);
+ buf[7] = (add_len & 0xff);
+
+ return 0;
+}
+
+static int core_scsi3_emulate_pr_in(struct se_cmd *cmd, unsigned char *cdb)
+{
+ switch (cdb[1] & 0x1f) {
+ case PRI_READ_KEYS:
+ return core_scsi3_pri_read_keys(cmd);
+ case PRI_READ_RESERVATION:
+ return core_scsi3_pri_read_reservation(cmd);
+ case PRI_REPORT_CAPABILITIES:
+ return core_scsi3_pri_report_capabilities(cmd);
+ case PRI_READ_FULL_STATUS:
+ return core_scsi3_pri_read_full_status(cmd);
+ default:
+ printk(KERN_ERR "Unknown PERSISTENT_RESERVE_IN service"
+ " action: 0x%02x\n", cdb[1] & 0x1f);
+ return PYX_TRANSPORT_INVALID_CDB_FIELD;
+ }
+
+}
+
+int core_scsi3_emulate_pr(struct se_cmd *cmd)
+{
+ unsigned char *cdb = &T_TASK(cmd)->t_task_cdb[0];
+ struct se_device *dev = cmd->se_dev;
+ /*
+ * Following spc2r20 5.5.1 Reservations overview:
+ *
+ * If a logical unit has been reserved by any RESERVE command and is
+ * still reserved by any initiator, all PERSISTENT RESERVE IN and all
+ * PERSISTENT RESERVE OUT commands shall conflict regardless of
+ * initiator or service action and shall terminate with a RESERVATION
+ * CONFLICT status.
+ */
+ if (dev->dev_flags & DF_SPC2_RESERVATIONS) {
+ printk(KERN_ERR "Received PERSISTENT_RESERVE CDB while legacy"
+ " SPC-2 reservation is held, returning"
+ " RESERVATION_CONFLICT\n");
+ return PYX_TRANSPORT_RESERVATION_CONFLICT;
+ }
+
+ return (cdb[0] == PERSISTENT_RESERVE_OUT) ?
+ core_scsi3_emulate_pr_out(cmd, cdb) :
+ core_scsi3_emulate_pr_in(cmd, cdb);
+}
+
+static int core_pt_reservation_check(struct se_cmd *cmd, u32 *pr_res_type)
+{
+ return 0;
+}
+
+static int core_pt_seq_non_holder(
+ struct se_cmd *cmd,
+ unsigned char *cdb,
+ u32 pr_reg_type)
+{
+ return 0;
+}
+
+int core_setup_reservations(struct se_device *dev, int force_pt)
+{
+ struct se_subsystem_dev *su_dev = dev->se_sub_dev;
+ struct t10_reservation_template *rest = &su_dev->t10_reservation;
+ /*
+ * If this device is from Target_Core_Mod/pSCSI, use the reservations
+ * of the Underlying SCSI hardware. In Linux/SCSI terms, this can
+ * cause a problem because libata and some SATA RAID HBAs appear
+ * under Linux/SCSI, but to emulate reservations themselves.
+ */
+ if (((TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV) &&
+ !(DEV_ATTRIB(dev)->emulate_reservations)) || force_pt) {
+ rest->res_type = SPC_PASSTHROUGH;
+ rest->pr_ops.t10_reservation_check = &core_pt_reservation_check;
+ rest->pr_ops.t10_seq_non_holder = &core_pt_seq_non_holder;
+ printk(KERN_INFO "%s: Using SPC_PASSTHROUGH, no reservation"
+ " emulation\n", TRANSPORT(dev)->name);
+ return 0;
+ }
+ /*
+ * If SPC-3 or above is reported by real or emulated struct se_device,
+ * use emulated Persistent Reservations.
+ */
+ if (TRANSPORT(dev)->get_device_rev(dev) >= SCSI_3) {
+ rest->res_type = SPC3_PERSISTENT_RESERVATIONS;
+ rest->pr_ops.t10_reservation_check = &core_scsi3_pr_reservation_check;
+ rest->pr_ops.t10_seq_non_holder = &core_scsi3_pr_seq_non_holder;
+ printk(KERN_INFO "%s: Using SPC3_PERSISTENT_RESERVATIONS"
+ " emulation\n", TRANSPORT(dev)->name);
+ } else {
+ rest->res_type = SPC2_RESERVATIONS;
+ rest->pr_ops.t10_reservation_check = &core_scsi2_reservation_check;
+ rest->pr_ops.t10_seq_non_holder =
+ &core_scsi2_reservation_seq_non_holder;
+ printk(KERN_INFO "%s: Using SPC2_RESERVATIONS emulation\n",
+ TRANSPORT(dev)->name);
+ }
+
+ return 0;
+}
diff --git a/drivers/target/target_core_pr.h b/drivers/target/target_core_pr.h
new file mode 100644
index 0000000..5603bcf
--- /dev/null
+++ b/drivers/target/target_core_pr.h
@@ -0,0 +1,67 @@
+#ifndef TARGET_CORE_PR_H
+#define TARGET_CORE_PR_H
+/*
+ * PERSISTENT_RESERVE_OUT service action codes
+ *
+ * spc4r17 section 6.14.2 Table 171
+ */
+#define PRO_REGISTER 0x00
+#define PRO_RESERVE 0x01
+#define PRO_RELEASE 0x02
+#define PRO_CLEAR 0x03
+#define PRO_PREEMPT 0x04
+#define PRO_PREEMPT_AND_ABORT 0x05
+#define PRO_REGISTER_AND_IGNORE_EXISTING_KEY 0x06
+#define PRO_REGISTER_AND_MOVE 0x07
+/*
+ * PERSISTENT_RESERVE_IN service action codes
+ *
+ * spc4r17 section 6.13.1 Table 159
+ */
+#define PRI_READ_KEYS 0x00
+#define PRI_READ_RESERVATION 0x01
+#define PRI_REPORT_CAPABILITIES 0x02
+#define PRI_READ_FULL_STATUS 0x03
+/*
+ * PERSISTENT_RESERVE_ SCOPE field
+ *
+ * spc4r17 section 6.13.3.3 Table 163
+ */
+#define PR_SCOPE_LU_SCOPE 0x00
+/*
+ * PERSISTENT_RESERVE_* TYPE field
+ *
+ * spc4r17 section 6.13.3.4 Table 164
+ */
+#define PR_TYPE_WRITE_EXCLUSIVE 0x01
+#define PR_TYPE_EXCLUSIVE_ACCESS 0x03
+#define PR_TYPE_WRITE_EXCLUSIVE_REGONLY 0x05
+#define PR_TYPE_EXCLUSIVE_ACCESS_REGONLY 0x06
+#define PR_TYPE_WRITE_EXCLUSIVE_ALLREG 0x07
+#define PR_TYPE_EXCLUSIVE_ACCESS_ALLREG 0x08
+
+#define PR_APTPL_MAX_IPORT_LEN 256
+#define PR_APTPL_MAX_TPORT_LEN 256
+
+extern struct kmem_cache *t10_pr_reg_cache;
+
+extern int core_pr_dump_initiator_port(struct t10_pr_registration *,
+ char *, u32);
+extern int core_scsi2_emulate_crh(struct se_cmd *);
+extern int core_scsi3_alloc_aptpl_registration(
+ struct t10_reservation_template *, u64,
+ unsigned char *, unsigned char *, u32,
+ unsigned char *, u16, u32, int, int, u8);
+extern int core_scsi3_check_aptpl_registration(struct se_device *,
+ struct se_portal_group *, struct se_lun *,
+ struct se_lun_acl *);
+extern void core_scsi3_free_pr_reg_from_nacl(struct se_device *,
+ struct se_node_acl *);
+extern void core_scsi3_free_all_registrations(struct se_device *);
+extern unsigned char *core_scsi3_pr_dump_type(int);
+extern int core_scsi3_check_cdb_abort_and_preempt(struct list_head *,
+ struct se_cmd *);
+extern int core_scsi3_emulate_pr(struct se_cmd *);
+extern int core_setup_reservations(struct se_device *, int);
+
+#endif /* TARGET_CORE_PR_H */
diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c
new file mode 100644
index 0000000..742d246
--- /dev/null
+++ b/drivers/target/target_core_pscsi.c
@@ -0,0 +1,1470 @@
+/*******************************************************************************
+ * Filename: target_core_pscsi.c
+ *
+ * This file contains the generic target mode <-> Linux SCSI subsystem plugin.
+ *
+ * Copyright (c) 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/string.h>
+#include <linux/parser.h>
+#include <linux/timer.h>
+#include <linux/blkdev.h>
+#include <linux/blk_types.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <linux/genhd.h>
+#include <linux/cdrom.h>
+#include <linux/file.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_host.h>
+#include <scsi/libsas.h> /* For TASK_ATTR_* */
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_transport.h>
+
+#include "target_core_pscsi.h"
+
+#define ISPRINT(a) ((a >= ' ') && (a <= '~'))
+
+static struct se_subsystem_api pscsi_template;
+
+static void pscsi_req_done(struct request *, int);
+
+/* pscsi_get_sh():
+ *
+ *
+ */
+static struct Scsi_Host *pscsi_get_sh(u32 host_no)
+{
+ struct Scsi_Host *sh = NULL;
+
+ sh = scsi_host_lookup(host_no);
+ if (IS_ERR(sh)) {
+ printk(KERN_ERR "Unable to locate SCSI HBA with Host ID:"
+ " %u\n", host_no);
+ return NULL;
+ }
+
+ return sh;
+}
+
+/* pscsi_attach_hba():
+ *
+ * pscsi_get_sh() used scsi_host_lookup() to locate struct Scsi_Host.
+ * from the passed SCSI Host ID.
+ */
+static int pscsi_attach_hba(struct se_hba *hba, u32 host_id)
+{
+ int hba_depth;
+ struct pscsi_hba_virt *phv;
+
+ phv = kzalloc(sizeof(struct pscsi_hba_virt), GFP_KERNEL);
+ if (!(phv)) {
+ printk(KERN_ERR "Unable to allocate struct pscsi_hba_virt\n");
+ return -1;
+ }
+ phv->phv_host_id = host_id;
+ phv->phv_mode = PHV_VIRUTAL_HOST_ID;
+ hba_depth = PSCSI_VIRTUAL_HBA_DEPTH;
+ atomic_set(&hba->left_queue_depth, hba_depth);
+ atomic_set(&hba->max_queue_depth, hba_depth);
+
+ hba->hba_ptr = (void *)phv;
+
+ printk(KERN_INFO "CORE_HBA[%d] - TCM SCSI HBA Driver %s on"
+ " Generic Target Core Stack %s\n", hba->hba_id,
+ PSCSI_VERSION, TARGET_CORE_MOD_VERSION);
+ printk(KERN_INFO "CORE_HBA[%d] - Attached SCSI HBA to Generic"
+ " Target Core with TCQ Depth: %d\n", hba->hba_id,
+ atomic_read(&hba->max_queue_depth));
+
+ return 0;
+}
+
+static void pscsi_detach_hba(struct se_hba *hba)
+{
+ struct pscsi_hba_virt *phv = hba->hba_ptr;
+ struct Scsi_Host *scsi_host = phv->phv_lld_host;
+
+ if (scsi_host) {
+ scsi_host_put(scsi_host);
+
+ printk(KERN_INFO "CORE_HBA[%d] - Detached SCSI HBA: %s from"
+ " Generic Target Core\n", hba->hba_id,
+ (scsi_host->hostt->name) ? (scsi_host->hostt->name) :
+ "Unknown");
+ } else
+ printk(KERN_INFO "CORE_HBA[%d] - Detached Virtual SCSI HBA"
+ " from Generic Target Core\n", hba->hba_id);
+
+ kfree(phv);
+ hba->hba_ptr = NULL;
+}
+
+static int pscsi_pmode_enable_hba(struct se_hba *hba, unsigned long mode_flag)
+{
+ struct pscsi_hba_virt *phv = (struct pscsi_hba_virt *)hba->hba_ptr;
+ struct Scsi_Host *sh = phv->phv_lld_host;
+ int hba_depth = PSCSI_VIRTUAL_HBA_DEPTH;
+ /*
+ * Release the struct Scsi_Host
+ */
+ if (!(mode_flag)) {
+ if (!(sh))
+ return 0;
+
+ phv->phv_lld_host = NULL;
+ phv->phv_mode = PHV_VIRUTAL_HOST_ID;
+ atomic_set(&hba->left_queue_depth, hba_depth);
+ atomic_set(&hba->max_queue_depth, hba_depth);
+
+ printk(KERN_INFO "CORE_HBA[%d] - Disabled pSCSI HBA Passthrough"
+ " %s\n", hba->hba_id, (sh->hostt->name) ?
+ (sh->hostt->name) : "Unknown");
+
+ scsi_host_put(sh);
+ return 0;
+ }
+ /*
+ * Otherwise, locate struct Scsi_Host from the original passed
+ * pSCSI Host ID and enable for phba mode
+ */
+ sh = pscsi_get_sh(phv->phv_host_id);
+ if (!(sh)) {
+ printk(KERN_ERR "pSCSI: Unable to locate SCSI Host for"
+ " phv_host_id: %d\n", phv->phv_host_id);
+ return -1;
+ }
+ /*
+ * Usually the SCSI LLD will use the hostt->can_queue value to define
+ * its HBA TCQ depth. Some other drivers (like 2.6 megaraid) don't set
+ * this at all and set sh->can_queue at runtime.
+ */
+ hba_depth = (sh->hostt->can_queue > sh->can_queue) ?
+ sh->hostt->can_queue : sh->can_queue;
+
+ atomic_set(&hba->left_queue_depth, hba_depth);
+ atomic_set(&hba->max_queue_depth, hba_depth);
+
+ phv->phv_lld_host = sh;
+ phv->phv_mode = PHV_LLD_SCSI_HOST_NO;
+
+ printk(KERN_INFO "CORE_HBA[%d] - Enabled pSCSI HBA Passthrough %s\n",
+ hba->hba_id, (sh->hostt->name) ? (sh->hostt->name) : "Unknown");
+
+ return 1;
+}
+
+static void pscsi_tape_read_blocksize(struct se_device *dev,
+ struct scsi_device *sdev)
+{
+ unsigned char cdb[MAX_COMMAND_SIZE], *buf;
+ int ret;
+
+ buf = kzalloc(12, GFP_KERNEL);
+ if (!buf)
+ return;
+
+ memset(cdb, 0, MAX_COMMAND_SIZE);
+ cdb[0] = MODE_SENSE;
+ cdb[4] = 0x0c; /* 12 bytes */
+
+ ret = scsi_execute_req(sdev, cdb, DMA_FROM_DEVICE, buf, 12, NULL,
+ HZ, 1, NULL);
+ if (ret)
+ goto out_free;
+
+ /*
+ * If MODE_SENSE still returns zero, set the default value to 1024.
+ */
+ sdev->sector_size = (buf[9] << 16) | (buf[10] << 8) | (buf[11]);
+ if (!sdev->sector_size)
+ sdev->sector_size = 1024;
+out_free:
+ kfree(buf);
+}
+
+static void
+pscsi_set_inquiry_info(struct scsi_device *sdev, struct t10_wwn *wwn)
+{
+ unsigned char *buf;
+
+ if (sdev->inquiry_len < INQUIRY_LEN)
+ return;
+
+ buf = sdev->inquiry;
+ if (!buf)
+ return;
+ /*
+ * Use sdev->inquiry from drivers/scsi/scsi_scan.c:scsi_alloc_sdev()
+ */
+ memcpy(&wwn->vendor[0], &buf[8], sizeof(wwn->vendor));
+ memcpy(&wwn->model[0], &buf[16], sizeof(wwn->model));
+ memcpy(&wwn->revision[0], &buf[32], sizeof(wwn->revision));
+}
+
+static int
+pscsi_get_inquiry_vpd_serial(struct scsi_device *sdev, struct t10_wwn *wwn)
+{
+ unsigned char cdb[MAX_COMMAND_SIZE], *buf;
+ int ret;
+
+ buf = kzalloc(INQUIRY_VPD_SERIAL_LEN, GFP_KERNEL);
+ if (!buf)
+ return -1;
+
+ memset(cdb, 0, MAX_COMMAND_SIZE);
+ cdb[0] = INQUIRY;
+ cdb[1] = 0x01; /* Query VPD */
+ cdb[2] = 0x80; /* Unit Serial Number */
+ cdb[3] = (INQUIRY_VPD_SERIAL_LEN >> 8) & 0xff;
+ cdb[4] = (INQUIRY_VPD_SERIAL_LEN & 0xff);
+
+ ret = scsi_execute_req(sdev, cdb, DMA_FROM_DEVICE, buf,
+ INQUIRY_VPD_SERIAL_LEN, NULL, HZ, 1, NULL);
+ if (ret)
+ goto out_free;
+
+ snprintf(&wwn->unit_serial[0], INQUIRY_VPD_SERIAL_LEN, "%s", &buf[4]);
+
+ wwn->t10_sub_dev->su_dev_flags |= SDF_FIRMWARE_VPD_UNIT_SERIAL;
+
+ kfree(buf);
+ return 0;
+
+out_free:
+ kfree(buf);
+ return -1;
+}
+
+static void
+pscsi_get_inquiry_vpd_device_ident(struct scsi_device *sdev,
+ struct t10_wwn *wwn)
+{
+ unsigned char cdb[MAX_COMMAND_SIZE], *buf, *page_83;
+ int ident_len, page_len, off = 4, ret;
+ struct t10_vpd *vpd;
+
+ buf = kzalloc(INQUIRY_VPD_SERIAL_LEN, GFP_KERNEL);
+ if (!buf)
+ return;
+
+ memset(cdb, 0, MAX_COMMAND_SIZE);
+ cdb[0] = INQUIRY;
+ cdb[1] = 0x01; /* Query VPD */
+ cdb[2] = 0x83; /* Device Identifier */
+ cdb[3] = (INQUIRY_VPD_DEVICE_IDENTIFIER_LEN >> 8) & 0xff;
+ cdb[4] = (INQUIRY_VPD_DEVICE_IDENTIFIER_LEN & 0xff);
+
+ ret = scsi_execute_req(sdev, cdb, DMA_FROM_DEVICE, buf,
+ INQUIRY_VPD_DEVICE_IDENTIFIER_LEN,
+ NULL, HZ, 1, NULL);
+ if (ret)
+ goto out;
+
+ page_len = (buf[2] << 8) | buf[3];
+ while (page_len > 0) {
+ /* Grab a pointer to the Identification descriptor */
+ page_83 = &buf[off];
+ ident_len = page_83[3];
+ if (!ident_len) {
+ printk(KERN_ERR "page_83[3]: identifier"
+ " length zero!\n");
+ break;
+ }
+ printk(KERN_INFO "T10 VPD Identifer Length: %d\n", ident_len);
+
+ vpd = kzalloc(sizeof(struct t10_vpd), GFP_KERNEL);
+ if (!vpd) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " struct t10_vpd\n");
+ goto out;
+ }
+ INIT_LIST_HEAD(&vpd->vpd_list);
+
+ transport_set_vpd_proto_id(vpd, page_83);
+ transport_set_vpd_assoc(vpd, page_83);
+
+ if (transport_set_vpd_ident_type(vpd, page_83) < 0) {
+ off += (ident_len + 4);
+ page_len -= (ident_len + 4);
+ kfree(vpd);
+ continue;
+ }
+ if (transport_set_vpd_ident(vpd, page_83) < 0) {
+ off += (ident_len + 4);
+ page_len -= (ident_len + 4);
+ kfree(vpd);
+ continue;
+ }
+
+ list_add_tail(&vpd->vpd_list, &wwn->t10_vpd_list);
+ off += (ident_len + 4);
+ page_len -= (ident_len + 4);
+ }
+
+out:
+ kfree(buf);
+}
+
+/* pscsi_add_device_to_list():
+ *
+ *
+ */
+static struct se_device *pscsi_add_device_to_list(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ struct pscsi_dev_virt *pdv,
+ struct scsi_device *sd,
+ int dev_flags)
+{
+ struct se_device *dev;
+ struct se_dev_limits dev_limits;
+ struct request_queue *q;
+ struct queue_limits *limits;
+
+ memset(&dev_limits, 0, sizeof(struct se_dev_limits));
+
+ if (!sd->queue_depth) {
+ sd->queue_depth = PSCSI_DEFAULT_QUEUEDEPTH;
+
+ printk(KERN_ERR "Set broken SCSI Device %d:%d:%d"
+ " queue_depth to %d\n", sd->channel, sd->id,
+ sd->lun, sd->queue_depth);
+ }
+ /*
+ * Setup the local scope queue_limits from struct request_queue->limits
+ * to pass into transport_add_device_to_core_hba() as struct se_dev_limits.
+ */
+ q = sd->request_queue;
+ limits = &dev_limits.limits;
+ limits->logical_block_size = sd->sector_size;
+ limits->max_hw_sectors = (sd->host->max_sectors > queue_max_hw_sectors(q)) ?
+ queue_max_hw_sectors(q) : sd->host->max_sectors;
+ limits->max_sectors = (sd->host->max_sectors > queue_max_sectors(q)) ?
+ queue_max_sectors(q) : sd->host->max_sectors;
+ dev_limits.hw_queue_depth = sd->queue_depth;
+ dev_limits.queue_depth = sd->queue_depth;
+ /*
+ * Setup our standard INQUIRY info into se_dev->t10_wwn
+ */
+ pscsi_set_inquiry_info(sd, &se_dev->t10_wwn);
+
+ /*
+ * Set the pointer pdv->pdv_sd to from passed struct scsi_device,
+ * which has already been referenced with Linux SCSI code with
+ * scsi_device_get() in this file's pscsi_create_virtdevice().
+ *
+ * The passthrough operations called by the transport_add_device_*
+ * function below will require this pointer to be set for passthroug
+ * ops.
+ *
+ * For the shutdown case in pscsi_free_device(), this struct
+ * scsi_device reference is released with Linux SCSI code
+ * scsi_device_put() and the pdv->pdv_sd cleared.
+ */
+ pdv->pdv_sd = sd;
+
+ dev = transport_add_device_to_core_hba(hba, &pscsi_template,
+ se_dev, dev_flags, (void *)pdv,
+ &dev_limits, NULL, NULL);
+ if (!(dev)) {
+ pdv->pdv_sd = NULL;
+ return NULL;
+ }
+
+ /*
+ * Locate VPD WWN Information used for various purposes within
+ * the Storage Engine.
+ */
+ if (!pscsi_get_inquiry_vpd_serial(sd, &se_dev->t10_wwn)) {
+ /*
+ * If VPD Unit Serial returned GOOD status, try
+ * VPD Device Identification page (0x83).
+ */
+ pscsi_get_inquiry_vpd_device_ident(sd, &se_dev->t10_wwn);
+ }
+
+ /*
+ * For TYPE_TAPE, attempt to determine blocksize with MODE_SENSE.
+ */
+ if (sd->type == TYPE_TAPE)
+ pscsi_tape_read_blocksize(dev, sd);
+ return dev;
+}
+
+static void *pscsi_allocate_virtdevice(struct se_hba *hba, const char *name)
+{
+ struct pscsi_dev_virt *pdv;
+
+ pdv = kzalloc(sizeof(struct pscsi_dev_virt), GFP_KERNEL);
+ if (!(pdv)) {
+ printk(KERN_ERR "Unable to allocate memory for struct pscsi_dev_virt\n");
+ return NULL;
+ }
+ pdv->pdv_se_hba = hba;
+
+ printk(KERN_INFO "PSCSI: Allocated pdv: %p for %s\n", pdv, name);
+ return (void *)pdv;
+}
+
+/*
+ * Called with struct Scsi_Host->host_lock called.
+ */
+static struct se_device *pscsi_create_type_disk(
+ struct scsi_device *sd,
+ struct pscsi_dev_virt *pdv,
+ struct se_subsystem_dev *se_dev,
+ struct se_hba *hba)
+{
+ struct se_device *dev;
+ struct pscsi_hba_virt *phv = (struct pscsi_hba_virt *)pdv->pdv_se_hba->hba_ptr;
+ struct Scsi_Host *sh = sd->host;
+ struct block_device *bd;
+ u32 dev_flags = 0;
+
+ if (scsi_device_get(sd)) {
+ printk(KERN_ERR "scsi_device_get() failed for %d:%d:%d:%d\n",
+ sh->host_no, sd->channel, sd->id, sd->lun);
+ spin_unlock_irq(sh->host_lock);
+ return NULL;
+ }
+ spin_unlock_irq(sh->host_lock);
+ /*
+ * Claim exclusive struct block_device access to struct scsi_device
+ * for TYPE_DISK using supplied udev_path
+ */
+ bd = blkdev_get_by_path(se_dev->se_dev_udev_path,
+ FMODE_WRITE|FMODE_READ|FMODE_EXCL, pdv);
+ if (!(bd)) {
+ printk("pSCSI: blkdev_get_by_path() failed\n");
+ scsi_device_put(sd);
+ return NULL;
+ }
+ pdv->pdv_bd = bd;
+
+ dev = pscsi_add_device_to_list(hba, se_dev, pdv, sd, dev_flags);
+ if (!(dev)) {
+ blkdev_put(pdv->pdv_bd, FMODE_WRITE|FMODE_READ|FMODE_EXCL);
+ scsi_device_put(sd);
+ return NULL;
+ }
+ printk(KERN_INFO "CORE_PSCSI[%d] - Added TYPE_DISK for %d:%d:%d:%d\n",
+ phv->phv_host_id, sh->host_no, sd->channel, sd->id, sd->lun);
+
+ return dev;
+}
+
+/*
+ * Called with struct Scsi_Host->host_lock called.
+ */
+static struct se_device *pscsi_create_type_rom(
+ struct scsi_device *sd,
+ struct pscsi_dev_virt *pdv,
+ struct se_subsystem_dev *se_dev,
+ struct se_hba *hba)
+{
+ struct se_device *dev;
+ struct pscsi_hba_virt *phv = (struct pscsi_hba_virt *)pdv->pdv_se_hba->hba_ptr;
+ struct Scsi_Host *sh = sd->host;
+ u32 dev_flags = 0;
+
+ if (scsi_device_get(sd)) {
+ printk(KERN_ERR "scsi_device_get() failed for %d:%d:%d:%d\n",
+ sh->host_no, sd->channel, sd->id, sd->lun);
+ spin_unlock_irq(sh->host_lock);
+ return NULL;
+ }
+ spin_unlock_irq(sh->host_lock);
+
+ dev = pscsi_add_device_to_list(hba, se_dev, pdv, sd, dev_flags);
+ if (!(dev)) {
+ scsi_device_put(sd);
+ return NULL;
+ }
+ printk(KERN_INFO "CORE_PSCSI[%d] - Added Type: %s for %d:%d:%d:%d\n",
+ phv->phv_host_id, scsi_device_type(sd->type), sh->host_no,
+ sd->channel, sd->id, sd->lun);
+
+ return dev;
+}
+
+/*
+ *Called with struct Scsi_Host->host_lock called.
+ */
+static struct se_device *pscsi_create_type_other(
+ struct scsi_device *sd,
+ struct pscsi_dev_virt *pdv,
+ struct se_subsystem_dev *se_dev,
+ struct se_hba *hba)
+{
+ struct se_device *dev;
+ struct pscsi_hba_virt *phv = (struct pscsi_hba_virt *)pdv->pdv_se_hba->hba_ptr;
+ struct Scsi_Host *sh = sd->host;
+ u32 dev_flags = 0;
+
+ spin_unlock_irq(sh->host_lock);
+ dev = pscsi_add_device_to_list(hba, se_dev, pdv, sd, dev_flags);
+ if (!(dev))
+ return NULL;
+
+ printk(KERN_INFO "CORE_PSCSI[%d] - Added Type: %s for %d:%d:%d:%d\n",
+ phv->phv_host_id, scsi_device_type(sd->type), sh->host_no,
+ sd->channel, sd->id, sd->lun);
+
+ return dev;
+}
+
+static struct se_device *pscsi_create_virtdevice(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ void *p)
+{
+ struct pscsi_dev_virt *pdv = (struct pscsi_dev_virt *)p;
+ struct se_device *dev;
+ struct scsi_device *sd;
+ struct pscsi_hba_virt *phv = (struct pscsi_hba_virt *)hba->hba_ptr;
+ struct Scsi_Host *sh = phv->phv_lld_host;
+ int legacy_mode_enable = 0;
+
+ if (!(pdv)) {
+ printk(KERN_ERR "Unable to locate struct pscsi_dev_virt"
+ " parameter\n");
+ return NULL;
+ }
+ /*
+ * If not running in PHV_LLD_SCSI_HOST_NO mode, locate the
+ * struct Scsi_Host we will need to bring the TCM/pSCSI object online
+ */
+ if (!(sh)) {
+ if (phv->phv_mode == PHV_LLD_SCSI_HOST_NO) {
+ printk(KERN_ERR "pSCSI: Unable to locate struct"
+ " Scsi_Host for PHV_LLD_SCSI_HOST_NO\n");
+ return NULL;
+ }
+ /*
+ * For the newer PHV_VIRUTAL_HOST_ID struct scsi_device
+ * reference, we enforce that udev_path has been set
+ */
+ if (!(se_dev->su_dev_flags & SDF_USING_UDEV_PATH)) {
+ printk(KERN_ERR "pSCSI: udev_path attribute has not"
+ " been set before ENABLE=1\n");
+ return NULL;
+ }
+ /*
+ * If no scsi_host_id= was passed for PHV_VIRUTAL_HOST_ID,
+ * use the original TCM hba ID to reference Linux/SCSI Host No
+ * and enable for PHV_LLD_SCSI_HOST_NO mode.
+ */
+ if (!(pdv->pdv_flags & PDF_HAS_VIRT_HOST_ID)) {
+ spin_lock(&hba->device_lock);
+ if (!(list_empty(&hba->hba_dev_list))) {
+ printk(KERN_ERR "pSCSI: Unable to set hba_mode"
+ " with active devices\n");
+ spin_unlock(&hba->device_lock);
+ return NULL;
+ }
+ spin_unlock(&hba->device_lock);
+
+ if (pscsi_pmode_enable_hba(hba, 1) != 1)
+ return NULL;
+
+ legacy_mode_enable = 1;
+ hba->hba_flags |= HBA_FLAGS_PSCSI_MODE;
+ sh = phv->phv_lld_host;
+ } else {
+ sh = pscsi_get_sh(pdv->pdv_host_id);
+ if (!(sh)) {
+ printk(KERN_ERR "pSCSI: Unable to locate"
+ " pdv_host_id: %d\n", pdv->pdv_host_id);
+ return NULL;
+ }
+ }
+ } else {
+ if (phv->phv_mode == PHV_VIRUTAL_HOST_ID) {
+ printk(KERN_ERR "pSCSI: PHV_VIRUTAL_HOST_ID set while"
+ " struct Scsi_Host exists\n");
+ return NULL;
+ }
+ }
+
+ spin_lock_irq(sh->host_lock);
+ list_for_each_entry(sd, &sh->__devices, siblings) {
+ if ((pdv->pdv_channel_id != sd->channel) ||
+ (pdv->pdv_target_id != sd->id) ||
+ (pdv->pdv_lun_id != sd->lun))
+ continue;
+ /*
+ * Functions will release the held struct scsi_host->host_lock
+ * before calling calling pscsi_add_device_to_list() to register
+ * struct scsi_device with target_core_mod.
+ */
+ switch (sd->type) {
+ case TYPE_DISK:
+ dev = pscsi_create_type_disk(sd, pdv, se_dev, hba);
+ break;
+ case TYPE_ROM:
+ dev = pscsi_create_type_rom(sd, pdv, se_dev, hba);
+ break;
+ default:
+ dev = pscsi_create_type_other(sd, pdv, se_dev, hba);
+ break;
+ }
+
+ if (!(dev)) {
+ if (phv->phv_mode == PHV_VIRUTAL_HOST_ID)
+ scsi_host_put(sh);
+ else if (legacy_mode_enable) {
+ pscsi_pmode_enable_hba(hba, 0);
+ hba->hba_flags &= ~HBA_FLAGS_PSCSI_MODE;
+ }
+ pdv->pdv_sd = NULL;
+ return NULL;
+ }
+ return dev;
+ }
+ spin_unlock_irq(sh->host_lock);
+
+ printk(KERN_ERR "pSCSI: Unable to locate %d:%d:%d:%d\n", sh->host_no,
+ pdv->pdv_channel_id, pdv->pdv_target_id, pdv->pdv_lun_id);
+
+ if (phv->phv_mode == PHV_VIRUTAL_HOST_ID)
+ scsi_host_put(sh);
+ else if (legacy_mode_enable) {
+ pscsi_pmode_enable_hba(hba, 0);
+ hba->hba_flags &= ~HBA_FLAGS_PSCSI_MODE;
+ }
+
+ return NULL;
+}
+
+/* pscsi_free_device(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static void pscsi_free_device(void *p)
+{
+ struct pscsi_dev_virt *pdv = p;
+ struct pscsi_hba_virt *phv = pdv->pdv_se_hba->hba_ptr;
+ struct scsi_device *sd = pdv->pdv_sd;
+
+ if (sd) {
+ /*
+ * Release exclusive pSCSI internal struct block_device claim for
+ * struct scsi_device with TYPE_DISK from pscsi_create_type_disk()
+ */
+ if ((sd->type == TYPE_DISK) && pdv->pdv_bd) {
+ blkdev_put(pdv->pdv_bd,
+ FMODE_WRITE|FMODE_READ|FMODE_EXCL);
+ pdv->pdv_bd = NULL;
+ }
+ /*
+ * For HBA mode PHV_LLD_SCSI_HOST_NO, release the reference
+ * to struct Scsi_Host now.
+ */
+ if ((phv->phv_mode == PHV_LLD_SCSI_HOST_NO) &&
+ (phv->phv_lld_host != NULL))
+ scsi_host_put(phv->phv_lld_host);
+
+ if ((sd->type == TYPE_DISK) || (sd->type == TYPE_ROM))
+ scsi_device_put(sd);
+
+ pdv->pdv_sd = NULL;
+ }
+
+ kfree(pdv);
+}
+
+static inline struct pscsi_plugin_task *PSCSI_TASK(struct se_task *task)
+{
+ return container_of(task, struct pscsi_plugin_task, pscsi_task);
+}
+
+
+/* pscsi_transport_complete():
+ *
+ *
+ */
+static int pscsi_transport_complete(struct se_task *task)
+{
+ struct pscsi_dev_virt *pdv = task->se_dev->dev_ptr;
+ struct scsi_device *sd = pdv->pdv_sd;
+ int result;
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+ unsigned char *cdb = &pt->pscsi_cdb[0];
+
+ result = pt->pscsi_result;
+ /*
+ * Hack to make sure that Write-Protect modepage is set if R/O mode is
+ * forced.
+ */
+ if (((cdb[0] == MODE_SENSE) || (cdb[0] == MODE_SENSE_10)) &&
+ (status_byte(result) << 1) == SAM_STAT_GOOD) {
+ if (!TASK_CMD(task)->se_deve)
+ goto after_mode_sense;
+
+ if (TASK_CMD(task)->se_deve->lun_flags &
+ TRANSPORT_LUNFLAGS_READ_ONLY) {
+ unsigned char *buf = (unsigned char *)
+ T_TASK(task->task_se_cmd)->t_task_buf;
+
+ if (cdb[0] == MODE_SENSE_10) {
+ if (!(buf[3] & 0x80))
+ buf[3] |= 0x80;
+ } else {
+ if (!(buf[2] & 0x80))
+ buf[2] |= 0x80;
+ }
+ }
+ }
+after_mode_sense:
+
+ if (sd->type != TYPE_TAPE)
+ goto after_mode_select;
+
+ /*
+ * Hack to correctly obtain the initiator requested blocksize for
+ * TYPE_TAPE. Since this value is dependent upon each tape media,
+ * struct scsi_device->sector_size will not contain the correct value
+ * by default, so we go ahead and set it so
+ * TRANSPORT(dev)->get_blockdev() returns the correct value to the
+ * storage engine.
+ */
+ if (((cdb[0] == MODE_SELECT) || (cdb[0] == MODE_SELECT_10)) &&
+ (status_byte(result) << 1) == SAM_STAT_GOOD) {
+ unsigned char *buf;
+ struct scatterlist *sg = task->task_sg;
+ u16 bdl;
+ u32 blocksize;
+
+ buf = sg_virt(&sg[0]);
+ if (!(buf)) {
+ printk(KERN_ERR "Unable to get buf for scatterlist\n");
+ goto after_mode_select;
+ }
+
+ if (cdb[0] == MODE_SELECT)
+ bdl = (buf[3]);
+ else
+ bdl = (buf[6] << 8) | (buf[7]);
+
+ if (!bdl)
+ goto after_mode_select;
+
+ if (cdb[0] == MODE_SELECT)
+ blocksize = (buf[9] << 16) | (buf[10] << 8) |
+ (buf[11]);
+ else
+ blocksize = (buf[13] << 16) | (buf[14] << 8) |
+ (buf[15]);
+
+ sd->sector_size = blocksize;
+ }
+after_mode_select:
+
+ if (status_byte(result) & CHECK_CONDITION)
+ return 1;
+
+ return 0;
+}
+
+static struct se_task *
+pscsi_alloc_task(struct se_cmd *cmd)
+{
+ struct pscsi_plugin_task *pt;
+ unsigned char *cdb = T_TASK(cmd)->t_task_cdb;
+
+ pt = kzalloc(sizeof(struct pscsi_plugin_task), GFP_KERNEL);
+ if (!pt) {
+ printk(KERN_ERR "Unable to allocate struct pscsi_plugin_task\n");
+ return NULL;
+ }
+
+ /*
+ * If TCM Core is signaling a > TCM_MAX_COMMAND_SIZE allocation,
+ * allocate the extended CDB buffer for per struct se_task context
+ * pt->pscsi_cdb now.
+ */
+ if (T_TASK(cmd)->t_task_cdb != T_TASK(cmd)->__t_task_cdb) {
+
+ pt->pscsi_cdb = kzalloc(scsi_command_size(cdb), GFP_KERNEL);
+ if (!(pt->pscsi_cdb)) {
+ printk(KERN_ERR "pSCSI: Unable to allocate extended"
+ " pt->pscsi_cdb\n");
+ return NULL;
+ }
+ } else
+ pt->pscsi_cdb = &pt->__pscsi_cdb[0];
+
+ return &pt->pscsi_task;
+}
+
+static inline void pscsi_blk_init_request(
+ struct se_task *task,
+ struct pscsi_plugin_task *pt,
+ struct request *req,
+ int bidi_read)
+{
+ /*
+ * Defined as "scsi command" in include/linux/blkdev.h.
+ */
+ req->cmd_type = REQ_TYPE_BLOCK_PC;
+ /*
+ * For the extra BIDI-COMMAND READ struct request we do not
+ * need to setup the remaining structure members
+ */
+ if (bidi_read)
+ return;
+ /*
+ * Setup the done function pointer for struct request,
+ * also set the end_io_data pointer.to struct se_task.
+ */
+ req->end_io = pscsi_req_done;
+ req->end_io_data = (void *)task;
+ /*
+ * Load the referenced struct se_task's SCSI CDB into
+ * include/linux/blkdev.h:struct request->cmd
+ */
+ req->cmd_len = scsi_command_size(pt->pscsi_cdb);
+ req->cmd = &pt->pscsi_cdb[0];
+ /*
+ * Setup pointer for outgoing sense data.
+ */
+ req->sense = (void *)&pt->pscsi_sense[0];
+ req->sense_len = 0;
+}
+
+/*
+ * Used for pSCSI data payloads for all *NON* SCF_SCSI_DATA_SG_IO_CDB
+*/
+static int pscsi_blk_get_request(struct se_task *task)
+{
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+ struct pscsi_dev_virt *pdv = task->se_dev->dev_ptr;
+
+ pt->pscsi_req = blk_get_request(pdv->pdv_sd->request_queue,
+ (task->task_data_direction == DMA_TO_DEVICE),
+ GFP_KERNEL);
+ if (!(pt->pscsi_req) || IS_ERR(pt->pscsi_req)) {
+ printk(KERN_ERR "PSCSI: blk_get_request() failed: %ld\n",
+ IS_ERR(pt->pscsi_req));
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ /*
+ * Setup the newly allocated struct request for REQ_TYPE_BLOCK_PC,
+ * and setup rq callback, CDB and sense.
+ */
+ pscsi_blk_init_request(task, pt, pt->pscsi_req, 0);
+ return 0;
+}
+
+/* pscsi_do_task(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static int pscsi_do_task(struct se_task *task)
+{
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+ struct pscsi_dev_virt *pdv = task->se_dev->dev_ptr;
+ /*
+ * Set the struct request->timeout value based on peripheral
+ * device type from SCSI.
+ */
+ if (pdv->pdv_sd->type == TYPE_DISK)
+ pt->pscsi_req->timeout = PS_TIMEOUT_DISK;
+ else
+ pt->pscsi_req->timeout = PS_TIMEOUT_OTHER;
+
+ pt->pscsi_req->retries = PS_RETRY;
+ /*
+ * Queue the struct request into the struct scsi_device->request_queue.
+ * Also check for HEAD_OF_QUEUE SAM TASK attr from received se_cmd
+ * descriptor
+ */
+ blk_execute_rq_nowait(pdv->pdv_sd->request_queue, NULL, pt->pscsi_req,
+ (task->task_se_cmd->sam_task_attr == TASK_ATTR_HOQ),
+ pscsi_req_done);
+
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+}
+
+static void pscsi_free_task(struct se_task *task)
+{
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+ struct se_cmd *cmd = task->task_se_cmd;
+
+ /*
+ * Release the extended CDB allocation from pscsi_alloc_task()
+ * if one exists.
+ */
+ if (T_TASK(cmd)->t_task_cdb != T_TASK(cmd)->__t_task_cdb)
+ kfree(pt->pscsi_cdb);
+ /*
+ * We do not release the bio(s) here associated with this task, as
+ * this is handled by bio_put() and pscsi_bi_endio().
+ */
+ kfree(pt);
+}
+
+enum {
+ Opt_scsi_host_id, Opt_scsi_channel_id, Opt_scsi_target_id,
+ Opt_scsi_lun_id, Opt_err
+};
+
+static match_table_t tokens = {
+ {Opt_scsi_host_id, "scsi_host_id=%d"},
+ {Opt_scsi_channel_id, "scsi_channel_id=%d"},
+ {Opt_scsi_target_id, "scsi_target_id=%d"},
+ {Opt_scsi_lun_id, "scsi_lun_id=%d"},
+ {Opt_err, NULL}
+};
+
+static ssize_t pscsi_set_configfs_dev_params(struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ const char *page,
+ ssize_t count)
+{
+ struct pscsi_dev_virt *pdv = se_dev->se_dev_su_ptr;
+ struct pscsi_hba_virt *phv = hba->hba_ptr;
+ char *orig, *ptr, *opts;
+ substring_t args[MAX_OPT_ARGS];
+ int ret = 0, arg, token;
+
+ opts = kstrdup(page, GFP_KERNEL);
+ if (!opts)
+ return -ENOMEM;
+
+ orig = opts;
+
+ while ((ptr = strsep(&opts, ",")) != NULL) {
+ if (!*ptr)
+ continue;
+
+ token = match_token(ptr, tokens, args);
+ switch (token) {
+ case Opt_scsi_host_id:
+ if (phv->phv_mode == PHV_LLD_SCSI_HOST_NO) {
+ printk(KERN_ERR "PSCSI[%d]: Unable to accept"
+ " scsi_host_id while phv_mode =="
+ " PHV_LLD_SCSI_HOST_NO\n",
+ phv->phv_host_id);
+ ret = -EINVAL;
+ goto out;
+ }
+ match_int(args, &arg);
+ pdv->pdv_host_id = arg;
+ printk(KERN_INFO "PSCSI[%d]: Referencing SCSI Host ID:"
+ " %d\n", phv->phv_host_id, pdv->pdv_host_id);
+ pdv->pdv_flags |= PDF_HAS_VIRT_HOST_ID;
+ break;
+ case Opt_scsi_channel_id:
+ match_int(args, &arg);
+ pdv->pdv_channel_id = arg;
+ printk(KERN_INFO "PSCSI[%d]: Referencing SCSI Channel"
+ " ID: %d\n", phv->phv_host_id,
+ pdv->pdv_channel_id);
+ pdv->pdv_flags |= PDF_HAS_CHANNEL_ID;
+ break;
+ case Opt_scsi_target_id:
+ match_int(args, &arg);
+ pdv->pdv_target_id = arg;
+ printk(KERN_INFO "PSCSI[%d]: Referencing SCSI Target"
+ " ID: %d\n", phv->phv_host_id,
+ pdv->pdv_target_id);
+ pdv->pdv_flags |= PDF_HAS_TARGET_ID;
+ break;
+ case Opt_scsi_lun_id:
+ match_int(args, &arg);
+ pdv->pdv_lun_id = arg;
+ printk(KERN_INFO "PSCSI[%d]: Referencing SCSI LUN ID:"
+ " %d\n", phv->phv_host_id, pdv->pdv_lun_id);
+ pdv->pdv_flags |= PDF_HAS_LUN_ID;
+ break;
+ default:
+ break;
+ }
+ }
+
+out:
+ kfree(orig);
+ return (!ret) ? count : ret;
+}
+
+static ssize_t pscsi_check_configfs_dev_params(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev)
+{
+ struct pscsi_dev_virt *pdv = se_dev->se_dev_su_ptr;
+
+ if (!(pdv->pdv_flags & PDF_HAS_CHANNEL_ID) ||
+ !(pdv->pdv_flags & PDF_HAS_TARGET_ID) ||
+ !(pdv->pdv_flags & PDF_HAS_LUN_ID)) {
+ printk(KERN_ERR "Missing scsi_channel_id=, scsi_target_id= and"
+ " scsi_lun_id= parameters\n");
+ return -1;
+ }
+
+ return 0;
+}
+
+static ssize_t pscsi_show_configfs_dev_params(struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ char *b)
+{
+ struct pscsi_hba_virt *phv = hba->hba_ptr;
+ struct pscsi_dev_virt *pdv = se_dev->se_dev_su_ptr;
+ struct scsi_device *sd = pdv->pdv_sd;
+ unsigned char host_id[16];
+ ssize_t bl;
+ int i;
+
+ if (phv->phv_mode == PHV_VIRUTAL_HOST_ID)
+ snprintf(host_id, 16, "%d", pdv->pdv_host_id);
+ else
+ snprintf(host_id, 16, "PHBA Mode");
+
+ bl = sprintf(b, "SCSI Device Bus Location:"
+ " Channel ID: %d Target ID: %d LUN: %d Host ID: %s\n",
+ pdv->pdv_channel_id, pdv->pdv_target_id, pdv->pdv_lun_id,
+ host_id);
+
+ if (sd) {
+ bl += sprintf(b + bl, " ");
+ bl += sprintf(b + bl, "Vendor: ");
+ for (i = 0; i < 8; i++) {
+ if (ISPRINT(sd->vendor[i])) /* printable character? */
+ bl += sprintf(b + bl, "%c", sd->vendor[i]);
+ else
+ bl += sprintf(b + bl, " ");
+ }
+ bl += sprintf(b + bl, " Model: ");
+ for (i = 0; i < 16; i++) {
+ if (ISPRINT(sd->model[i])) /* printable character ? */
+ bl += sprintf(b + bl, "%c", sd->model[i]);
+ else
+ bl += sprintf(b + bl, " ");
+ }
+ bl += sprintf(b + bl, " Rev: ");
+ for (i = 0; i < 4; i++) {
+ if (ISPRINT(sd->rev[i])) /* printable character ? */
+ bl += sprintf(b + bl, "%c", sd->rev[i]);
+ else
+ bl += sprintf(b + bl, " ");
+ }
+ bl += sprintf(b + bl, "\n");
+ }
+ return bl;
+}
+
+static void pscsi_bi_endio(struct bio *bio, int error)
+{
+ bio_put(bio);
+}
+
+static inline struct bio *pscsi_get_bio(struct pscsi_dev_virt *pdv, int sg_num)
+{
+ struct bio *bio;
+ /*
+ * Use bio_malloc() following the comment in for bio -> struct request
+ * in block/blk-core.c:blk_make_request()
+ */
+ bio = bio_kmalloc(GFP_KERNEL, sg_num);
+ if (!(bio)) {
+ printk(KERN_ERR "PSCSI: bio_kmalloc() failed\n");
+ return NULL;
+ }
+ bio->bi_end_io = pscsi_bi_endio;
+
+ return bio;
+}
+
+#if 0
+#define DEBUG_PSCSI(x...) printk(x)
+#else
+#define DEBUG_PSCSI(x...)
+#endif
+
+static int __pscsi_map_task_SG(
+ struct se_task *task,
+ struct scatterlist *task_sg,
+ u32 task_sg_num,
+ int bidi_read)
+{
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+ struct pscsi_dev_virt *pdv = task->se_dev->dev_ptr;
+ struct bio *bio = NULL, *hbio = NULL, *tbio = NULL;
+ struct page *page;
+ struct scatterlist *sg;
+ 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 rw = (task->task_data_direction == DMA_TO_DEVICE);
+
+ if (!task->task_size)
+ return 0;
+ /*
+ * For SCF_SCSI_DATA_SG_IO_CDB, Use fs/bio.c:bio_add_page() to setup
+ * the bio_vec maplist from TC< struct se_mem -> task->task_sg ->
+ * struct scatterlist memory. The struct se_task->task_sg[] currently needs
+ * to be attached to struct bios for submission to Linux/SCSI using
+ * struct request to struct scsi_device->request_queue.
+ *
+ * Note that this will be changing post v2.6.28 as Target_Core_Mod/pSCSI
+ * is ported to upstream SCSI passthrough functionality that accepts
+ * struct scatterlist->page_link or struct page as a paraemeter.
+ */
+ DEBUG_PSCSI("PSCSI: nr_pages: %d\n", nr_pages);
+
+ for_each_sg(task_sg, sg, task_sg_num, i) {
+ page = sg_page(sg);
+ off = sg->offset;
+ len = sg->length;
+
+ DEBUG_PSCSI("PSCSI: i: %d page: %p len: %d off: %d\n", i,
+ page, len, off);
+
+ while (len > 0 && data_len > 0) {
+ bytes = min_t(unsigned int, len, PAGE_SIZE - off);
+ bytes = min(bytes, data_len);
+
+ if (!(bio)) {
+ nr_vecs = min_t(int, BIO_MAX_PAGES, nr_pages);
+ nr_pages -= nr_vecs;
+ /*
+ * Calls bio_kmalloc() and sets bio->bi_end_io()
+ */
+ bio = pscsi_get_bio(pdv, nr_vecs);
+ if (!(bio))
+ goto fail;
+
+ if (rw)
+ bio->bi_rw |= REQ_WRITE;
+
+ DEBUG_PSCSI("PSCSI: Allocated bio: %p,"
+ " dir: %s nr_vecs: %d\n", bio,
+ (rw) ? "rw" : "r", nr_vecs);
+ /*
+ * Set *hbio pointer to handle the case:
+ * nr_pages > BIO_MAX_PAGES, where additional
+ * bios need to be added to complete a given
+ * struct se_task
+ */
+ if (!hbio)
+ hbio = tbio = bio;
+ else
+ tbio = tbio->bi_next = bio;
+ }
+
+ DEBUG_PSCSI("PSCSI: Calling bio_add_pc_page() i: %d"
+ " bio: %p page: %p len: %d off: %d\n", i, bio,
+ page, len, off);
+
+ rc = bio_add_pc_page(pdv->pdv_sd->request_queue,
+ bio, page, bytes, off);
+ if (rc != bytes)
+ goto fail;
+
+ DEBUG_PSCSI("PSCSI: bio->bi_vcnt: %d nr_vecs: %d\n",
+ bio->bi_vcnt, nr_vecs);
+
+ if (bio->bi_vcnt > nr_vecs) {
+ DEBUG_PSCSI("PSCSI: Reached bio->bi_vcnt max:"
+ " %d i: %d bio: %p, allocating another"
+ " bio\n", bio->bi_vcnt, i, bio);
+ /*
+ * Clear the pointer so that another bio will
+ * be allocated with pscsi_get_bio() above, the
+ * current bio has already been set *tbio and
+ * bio->bi_next.
+ */
+ bio = NULL;
+ }
+
+ page++;
+ len -= bytes;
+ data_len -= bytes;
+ off = 0;
+ }
+ }
+ /*
+ * Setup the primary pt->pscsi_req used for non BIDI and BIDI-COMMAND
+ * primary SCSI WRITE poayload mapped for struct se_task->task_sg[]
+ */
+ if (!(bidi_read)) {
+ /*
+ * Starting with v2.6.31, call blk_make_request() passing in *hbio to
+ * allocate the pSCSI task a struct request.
+ */
+ pt->pscsi_req = blk_make_request(pdv->pdv_sd->request_queue,
+ hbio, GFP_KERNEL);
+ if (!(pt->pscsi_req)) {
+ printk(KERN_ERR "pSCSI: blk_make_request() failed\n");
+ goto fail;
+ }
+ /*
+ * Setup the newly allocated struct request for REQ_TYPE_BLOCK_PC,
+ * and setup rq callback, CDB and sense.
+ */
+ pscsi_blk_init_request(task, pt, pt->pscsi_req, 0);
+
+ return task->task_sg_num;
+ }
+ /*
+ * Setup the secondary pt->pscsi_req->next_rq used for the extra BIDI-COMMAND
+ * SCSI READ paylaod mapped for struct se_task->task_sg_bidi[]
+ */
+ pt->pscsi_req->next_rq = blk_make_request(pdv->pdv_sd->request_queue,
+ hbio, GFP_KERNEL);
+ if (!(pt->pscsi_req->next_rq)) {
+ printk(KERN_ERR "pSCSI: blk_make_request() failed for BIDI\n");
+ goto fail;
+ }
+ pscsi_blk_init_request(task, pt, pt->pscsi_req->next_rq, 1);
+
+ return task->task_sg_num;
+fail:
+ while (hbio) {
+ bio = hbio;
+ hbio = hbio->bi_next;
+ bio->bi_next = NULL;
+ bio_endio(bio, 0);
+ }
+ return ret;
+}
+
+static int pscsi_map_task_SG(struct se_task *task)
+{
+ int ret;
+
+ /*
+ * Setup the main struct request for the task->task_sg[] payload
+ */
+
+ ret = __pscsi_map_task_SG(task, task->task_sg, task->task_sg_num, 0);
+ if (ret >= 0 && task->task_sg_bidi) {
+ /*
+ * If present, set up the extra BIDI-COMMAND SCSI READ
+ * struct request and payload.
+ */
+ ret = __pscsi_map_task_SG(task, task->task_sg_bidi,
+ task->task_sg_num, 1);
+ }
+
+ if (ret < 0)
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ return 0;
+}
+
+/* pscsi_map_task_non_SG():
+ *
+ *
+ */
+static int pscsi_map_task_non_SG(struct se_task *task)
+{
+ struct se_cmd *cmd = TASK_CMD(task);
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+ struct pscsi_dev_virt *pdv = task->se_dev->dev_ptr;
+ int ret = 0;
+
+ if (pscsi_blk_get_request(task) < 0)
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+
+ if (!task->task_size)
+ return 0;
+
+ ret = blk_rq_map_kern(pdv->pdv_sd->request_queue,
+ pt->pscsi_req, T_TASK(cmd)->t_task_buf,
+ task->task_size, GFP_KERNEL);
+ if (ret < 0) {
+ printk(KERN_ERR "PSCSI: blk_rq_map_kern() failed: %d\n", ret);
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ return 0;
+}
+
+static int pscsi_CDB_none(struct se_task *task)
+{
+ return pscsi_blk_get_request(task);
+}
+
+/* pscsi_get_cdb():
+ *
+ *
+ */
+static unsigned char *pscsi_get_cdb(struct se_task *task)
+{
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+
+ return pt->pscsi_cdb;
+}
+
+/* pscsi_get_sense_buffer():
+ *
+ *
+ */
+static unsigned char *pscsi_get_sense_buffer(struct se_task *task)
+{
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+
+ return (unsigned char *)&pt->pscsi_sense[0];
+}
+
+/* pscsi_get_device_rev():
+ *
+ *
+ */
+static u32 pscsi_get_device_rev(struct se_device *dev)
+{
+ struct pscsi_dev_virt *pdv = dev->dev_ptr;
+ struct scsi_device *sd = pdv->pdv_sd;
+
+ return (sd->scsi_level - 1) ? sd->scsi_level - 1 : 1;
+}
+
+/* pscsi_get_device_type():
+ *
+ *
+ */
+static u32 pscsi_get_device_type(struct se_device *dev)
+{
+ struct pscsi_dev_virt *pdv = dev->dev_ptr;
+ struct scsi_device *sd = pdv->pdv_sd;
+
+ return sd->type;
+}
+
+static sector_t pscsi_get_blocks(struct se_device *dev)
+{
+ struct pscsi_dev_virt *pdv = dev->dev_ptr;
+
+ if (pdv->pdv_bd && pdv->pdv_bd->bd_part)
+ return pdv->pdv_bd->bd_part->nr_sects;
+
+ dump_stack();
+ return 0;
+}
+
+/* pscsi_handle_SAM_STATUS_failures():
+ *
+ *
+ */
+static inline void pscsi_process_SAM_status(
+ struct se_task *task,
+ struct pscsi_plugin_task *pt)
+{
+ task->task_scsi_status = status_byte(pt->pscsi_result);
+ if ((task->task_scsi_status)) {
+ task->task_scsi_status <<= 1;
+ printk(KERN_INFO "PSCSI Status Byte exception at task: %p CDB:"
+ " 0x%02x Result: 0x%08x\n", task, pt->pscsi_cdb[0],
+ pt->pscsi_result);
+ }
+
+ switch (host_byte(pt->pscsi_result)) {
+ case DID_OK:
+ transport_complete_task(task, (!task->task_scsi_status));
+ break;
+ default:
+ printk(KERN_INFO "PSCSI Host Byte exception at task: %p CDB:"
+ " 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_CMD(task)->transport_error_status =
+ PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ transport_complete_task(task, 0);
+ break;
+ }
+
+ return;
+}
+
+static void pscsi_req_done(struct request *req, int uptodate)
+{
+ struct se_task *task = req->end_io_data;
+ struct pscsi_plugin_task *pt = PSCSI_TASK(task);
+
+ pt->pscsi_result = req->errors;
+ pt->pscsi_resid = req->resid_len;
+
+ pscsi_process_SAM_status(task, pt);
+ /*
+ * Release BIDI-READ if present
+ */
+ if (req->next_rq != NULL)
+ __blk_put_request(req->q, req->next_rq);
+
+ __blk_put_request(req->q, req);
+ pt->pscsi_req = NULL;
+}
+
+static struct se_subsystem_api pscsi_template = {
+ .name = "pscsi",
+ .owner = THIS_MODULE,
+ .transport_type = TRANSPORT_PLUGIN_PHBA_PDEV,
+ .cdb_none = pscsi_CDB_none,
+ .map_task_non_SG = pscsi_map_task_non_SG,
+ .map_task_SG = pscsi_map_task_SG,
+ .attach_hba = pscsi_attach_hba,
+ .detach_hba = pscsi_detach_hba,
+ .pmode_enable_hba = pscsi_pmode_enable_hba,
+ .allocate_virtdevice = pscsi_allocate_virtdevice,
+ .create_virtdevice = pscsi_create_virtdevice,
+ .free_device = pscsi_free_device,
+ .transport_complete = pscsi_transport_complete,
+ .alloc_task = pscsi_alloc_task,
+ .do_task = pscsi_do_task,
+ .free_task = pscsi_free_task,
+ .check_configfs_dev_params = pscsi_check_configfs_dev_params,
+ .set_configfs_dev_params = pscsi_set_configfs_dev_params,
+ .show_configfs_dev_params = pscsi_show_configfs_dev_params,
+ .get_cdb = pscsi_get_cdb,
+ .get_sense_buffer = pscsi_get_sense_buffer,
+ .get_device_rev = pscsi_get_device_rev,
+ .get_device_type = pscsi_get_device_type,
+ .get_blocks = pscsi_get_blocks,
+};
+
+static int __init pscsi_module_init(void)
+{
+ return transport_subsystem_register(&pscsi_template);
+}
+
+static void pscsi_module_exit(void)
+{
+ transport_subsystem_release(&pscsi_template);
+}
+
+MODULE_DESCRIPTION("TCM PSCSI subsystem plugin");
+MODULE_AUTHOR("nab@Linux-iSCSI.org");
+MODULE_LICENSE("GPL");
+
+module_init(pscsi_module_init);
+module_exit(pscsi_module_exit);
diff --git a/drivers/target/target_core_pscsi.h b/drivers/target/target_core_pscsi.h
new file mode 100644
index 0000000..a4cd5d3
--- /dev/null
+++ b/drivers/target/target_core_pscsi.h
@@ -0,0 +1,65 @@
+#ifndef TARGET_CORE_PSCSI_H
+#define TARGET_CORE_PSCSI_H
+
+#define PSCSI_VERSION "v4.0"
+#define PSCSI_VIRTUAL_HBA_DEPTH 2048
+
+/* used in pscsi_find_alloc_len() */
+#ifndef INQUIRY_DATA_SIZE
+#define INQUIRY_DATA_SIZE 0x24
+#endif
+
+/* used in pscsi_add_device_to_list() */
+#define PSCSI_DEFAULT_QUEUEDEPTH 1
+
+#define PS_RETRY 5
+#define PS_TIMEOUT_DISK (15*HZ)
+#define PS_TIMEOUT_OTHER (500*HZ)
+
+#include <linux/device.h>
+#include <scsi/scsi_driver.h>
+#include <scsi/scsi_device.h>
+#include <linux/kref.h>
+#include <linux/kobject.h>
+
+struct pscsi_plugin_task {
+ struct se_task pscsi_task;
+ unsigned char *pscsi_cdb;
+ unsigned char __pscsi_cdb[TCM_MAX_COMMAND_SIZE];
+ unsigned char pscsi_sense[SCSI_SENSE_BUFFERSIZE];
+ int pscsi_direction;
+ int pscsi_result;
+ u32 pscsi_resid;
+ struct request *pscsi_req;
+} ____cacheline_aligned;
+
+#define PDF_HAS_CHANNEL_ID 0x01
+#define PDF_HAS_TARGET_ID 0x02
+#define PDF_HAS_LUN_ID 0x04
+#define PDF_HAS_VPD_UNIT_SERIAL 0x08
+#define PDF_HAS_VPD_DEV_IDENT 0x10
+#define PDF_HAS_VIRT_HOST_ID 0x20
+
+struct pscsi_dev_virt {
+ int pdv_flags;
+ int pdv_host_id;
+ int pdv_channel_id;
+ int pdv_target_id;
+ int pdv_lun_id;
+ struct block_device *pdv_bd;
+ struct scsi_device *pdv_sd;
+ struct se_hba *pdv_se_hba;
+} ____cacheline_aligned;
+
+typedef enum phv_modes {
+ PHV_VIRUTAL_HOST_ID,
+ PHV_LLD_SCSI_HOST_NO
+} phv_modes_t;
+
+struct pscsi_hba_virt {
+ int phv_host_id;
+ phv_modes_t phv_mode;
+ struct Scsi_Host *phv_lld_host;
+} ____cacheline_aligned;
+
+#endif /*** TARGET_CORE_PSCSI_H ***/
diff --git a/drivers/target/target_core_rd.c b/drivers/target/target_core_rd.c
new file mode 100644
index 0000000..979aebf
--- /dev/null
+++ b/drivers/target/target_core_rd.c
@@ -0,0 +1,1091 @@
+/*******************************************************************************
+ * Filename: target_core_rd.c
+ *
+ * This file contains the Storage Engine <-> Ramdisk transport
+ * specific functions.
+ *
+ * Copyright (c) 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/string.h>
+#include <linux/parser.h>
+#include <linux/timer.h>
+#include <linux/blkdev.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_host.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+
+#include "target_core_rd.h"
+
+static struct se_subsystem_api rd_dr_template;
+static struct se_subsystem_api rd_mcp_template;
+
+/* #define DEBUG_RAMDISK_MCP */
+/* #define DEBUG_RAMDISK_DR */
+
+/* rd_attach_hba(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static int rd_attach_hba(struct se_hba *hba, u32 host_id)
+{
+ struct rd_host *rd_host;
+
+ rd_host = kzalloc(sizeof(struct rd_host), GFP_KERNEL);
+ if (!(rd_host)) {
+ printk(KERN_ERR "Unable to allocate memory for struct rd_host\n");
+ return -ENOMEM;
+ }
+
+ rd_host->rd_host_id = host_id;
+
+ atomic_set(&hba->left_queue_depth, RD_HBA_QUEUE_DEPTH);
+ atomic_set(&hba->max_queue_depth, RD_HBA_QUEUE_DEPTH);
+ hba->hba_ptr = (void *) rd_host;
+
+ printk(KERN_INFO "CORE_HBA[%d] - TCM Ramdisk HBA Driver %s on"
+ " Generic Target Core Stack %s\n", hba->hba_id,
+ RD_HBA_VERSION, TARGET_CORE_MOD_VERSION);
+ printk(KERN_INFO "CORE_HBA[%d] - Attached Ramdisk HBA: %u to Generic"
+ " Target Core TCQ Depth: %d MaxSectors: %u\n", hba->hba_id,
+ rd_host->rd_host_id, atomic_read(&hba->max_queue_depth),
+ RD_MAX_SECTORS);
+
+ return 0;
+}
+
+static void rd_detach_hba(struct se_hba *hba)
+{
+ struct rd_host *rd_host = hba->hba_ptr;
+
+ printk(KERN_INFO "CORE_HBA[%d] - Detached Ramdisk HBA: %u from"
+ " Generic Target Core\n", hba->hba_id, rd_host->rd_host_id);
+
+ kfree(rd_host);
+ hba->hba_ptr = NULL;
+}
+
+/* rd_release_device_space():
+ *
+ *
+ */
+static void rd_release_device_space(struct rd_dev *rd_dev)
+{
+ u32 i, j, page_count = 0, sg_per_table;
+ struct rd_dev_sg_table *sg_table;
+ struct page *pg;
+ struct scatterlist *sg;
+
+ if (!rd_dev->sg_table_array || !rd_dev->sg_table_count)
+ return;
+
+ sg_table = rd_dev->sg_table_array;
+
+ for (i = 0; i < rd_dev->sg_table_count; i++) {
+ sg = sg_table[i].sg_table;
+ sg_per_table = sg_table[i].rd_sg_count;
+
+ for (j = 0; j < sg_per_table; j++) {
+ pg = sg_page(&sg[j]);
+ if ((pg)) {
+ __free_page(pg);
+ page_count++;
+ }
+ }
+
+ kfree(sg);
+ }
+
+ printk(KERN_INFO "CORE_RD[%u] - Released device space for Ramdisk"
+ " Device ID: %u, pages %u in %u tables total bytes %lu\n",
+ rd_dev->rd_host->rd_host_id, rd_dev->rd_dev_id, page_count,
+ rd_dev->sg_table_count, (unsigned long)page_count * PAGE_SIZE);
+
+ kfree(sg_table);
+ rd_dev->sg_table_array = NULL;
+ rd_dev->sg_table_count = 0;
+}
+
+
+/* rd_build_device_space():
+ *
+ *
+ */
+static int rd_build_device_space(struct rd_dev *rd_dev)
+{
+ u32 i = 0, j, page_offset = 0, sg_per_table, sg_tables, total_sg_needed;
+ u32 max_sg_per_table = (RD_MAX_ALLOCATION_SIZE /
+ sizeof(struct scatterlist));
+ struct rd_dev_sg_table *sg_table;
+ struct page *pg;
+ struct scatterlist *sg;
+
+ if (rd_dev->rd_page_count <= 0) {
+ printk(KERN_ERR "Illegal page count: %u for Ramdisk device\n",
+ rd_dev->rd_page_count);
+ return -1;
+ }
+ total_sg_needed = rd_dev->rd_page_count;
+
+ sg_tables = (total_sg_needed / max_sg_per_table) + 1;
+
+ sg_table = kzalloc(sg_tables * sizeof(struct rd_dev_sg_table), GFP_KERNEL);
+ if (!(sg_table)) {
+ printk(KERN_ERR "Unable to allocate memory for Ramdisk"
+ " scatterlist tables\n");
+ return -1;
+ }
+
+ rd_dev->sg_table_array = sg_table;
+ rd_dev->sg_table_count = sg_tables;
+
+ while (total_sg_needed) {
+ sg_per_table = (total_sg_needed > max_sg_per_table) ?
+ max_sg_per_table : total_sg_needed;
+
+ sg = kzalloc(sg_per_table * sizeof(struct scatterlist),
+ GFP_KERNEL);
+ if (!(sg)) {
+ printk(KERN_ERR "Unable to allocate scatterlist array"
+ " for struct rd_dev\n");
+ return -1;
+ }
+
+ sg_init_table((struct scatterlist *)&sg[0], sg_per_table);
+
+ sg_table[i].sg_table = sg;
+ sg_table[i].rd_sg_count = sg_per_table;
+ sg_table[i].page_start_offset = page_offset;
+ sg_table[i++].page_end_offset = (page_offset + sg_per_table)
+ - 1;
+
+ for (j = 0; j < sg_per_table; j++) {
+ pg = alloc_pages(GFP_KERNEL, 0);
+ if (!(pg)) {
+ printk(KERN_ERR "Unable to allocate scatterlist"
+ " pages for struct rd_dev_sg_table\n");
+ return -1;
+ }
+ sg_assign_page(&sg[j], pg);
+ sg[j].length = PAGE_SIZE;
+ }
+
+ page_offset += sg_per_table;
+ total_sg_needed -= sg_per_table;
+ }
+
+ printk(KERN_INFO "CORE_RD[%u] - Built Ramdisk Device ID: %u space of"
+ " %u pages in %u tables\n", rd_dev->rd_host->rd_host_id,
+ rd_dev->rd_dev_id, rd_dev->rd_page_count,
+ rd_dev->sg_table_count);
+
+ return 0;
+}
+
+static void *rd_allocate_virtdevice(
+ struct se_hba *hba,
+ const char *name,
+ int rd_direct)
+{
+ struct rd_dev *rd_dev;
+ struct rd_host *rd_host = hba->hba_ptr;
+
+ rd_dev = kzalloc(sizeof(struct rd_dev), GFP_KERNEL);
+ if (!(rd_dev)) {
+ printk(KERN_ERR "Unable to allocate memory for struct rd_dev\n");
+ return NULL;
+ }
+
+ rd_dev->rd_host = rd_host;
+ rd_dev->rd_direct = rd_direct;
+
+ return rd_dev;
+}
+
+static void *rd_DIRECT_allocate_virtdevice(struct se_hba *hba, const char *name)
+{
+ return rd_allocate_virtdevice(hba, name, 1);
+}
+
+static void *rd_MEMCPY_allocate_virtdevice(struct se_hba *hba, const char *name)
+{
+ return rd_allocate_virtdevice(hba, name, 0);
+}
+
+/* rd_create_virtdevice():
+ *
+ *
+ */
+static struct se_device *rd_create_virtdevice(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ void *p,
+ int rd_direct)
+{
+ struct se_device *dev;
+ struct se_dev_limits dev_limits;
+ struct rd_dev *rd_dev = p;
+ struct rd_host *rd_host = hba->hba_ptr;
+ int dev_flags = 0;
+ char prod[16], rev[4];
+
+ memset(&dev_limits, 0, sizeof(struct se_dev_limits));
+
+ if (rd_build_device_space(rd_dev) < 0)
+ goto fail;
+
+ snprintf(prod, 16, "RAMDISK-%s", (rd_dev->rd_direct) ? "DR" : "MCP");
+ snprintf(rev, 4, "%s", (rd_dev->rd_direct) ? RD_DR_VERSION :
+ RD_MCP_VERSION);
+
+ dev_limits.limits.logical_block_size = RD_BLOCKSIZE;
+ dev_limits.limits.max_hw_sectors = RD_MAX_SECTORS;
+ dev_limits.limits.max_sectors = RD_MAX_SECTORS;
+ dev_limits.hw_queue_depth = RD_MAX_DEVICE_QUEUE_DEPTH;
+ dev_limits.queue_depth = RD_DEVICE_QUEUE_DEPTH;
+
+ dev = transport_add_device_to_core_hba(hba,
+ (rd_dev->rd_direct) ? &rd_dr_template :
+ &rd_mcp_template, se_dev, dev_flags, (void *)rd_dev,
+ &dev_limits, prod, rev);
+ if (!(dev))
+ goto fail;
+
+ rd_dev->rd_dev_id = rd_host->rd_host_dev_id_count++;
+ rd_dev->rd_queue_depth = dev->queue_depth;
+
+ printk(KERN_INFO "CORE_RD[%u] - Added TCM %s Ramdisk Device ID: %u of"
+ " %u pages in %u tables, %lu total bytes\n",
+ rd_host->rd_host_id, (!rd_dev->rd_direct) ? "MEMCPY" :
+ "DIRECT", rd_dev->rd_dev_id, rd_dev->rd_page_count,
+ rd_dev->sg_table_count,
+ (unsigned long)(rd_dev->rd_page_count * PAGE_SIZE));
+
+ return dev;
+
+fail:
+ rd_release_device_space(rd_dev);
+ return NULL;
+}
+
+static struct se_device *rd_DIRECT_create_virtdevice(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ void *p)
+{
+ return rd_create_virtdevice(hba, se_dev, p, 1);
+}
+
+static struct se_device *rd_MEMCPY_create_virtdevice(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ void *p)
+{
+ return rd_create_virtdevice(hba, se_dev, p, 0);
+}
+
+/* rd_free_device(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static void rd_free_device(void *p)
+{
+ struct rd_dev *rd_dev = p;
+
+ rd_release_device_space(rd_dev);
+ kfree(rd_dev);
+}
+
+static inline struct rd_request *RD_REQ(struct se_task *task)
+{
+ return container_of(task, struct rd_request, rd_task);
+}
+
+static struct se_task *
+rd_alloc_task(struct se_cmd *cmd)
+{
+ struct rd_request *rd_req;
+
+ rd_req = kzalloc(sizeof(struct rd_request), GFP_KERNEL);
+ if (!rd_req) {
+ printk(KERN_ERR "Unable to allocate struct rd_request\n");
+ return NULL;
+ }
+ rd_req->rd_dev = SE_DEV(cmd)->dev_ptr;
+
+ return &rd_req->rd_task;
+}
+
+/* rd_get_sg_table():
+ *
+ *
+ */
+static struct rd_dev_sg_table *rd_get_sg_table(struct rd_dev *rd_dev, u32 page)
+{
+ u32 i;
+ struct rd_dev_sg_table *sg_table;
+
+ for (i = 0; i < rd_dev->sg_table_count; i++) {
+ sg_table = &rd_dev->sg_table_array[i];
+ if ((sg_table->page_start_offset <= page) &&
+ (sg_table->page_end_offset >= page))
+ return sg_table;
+ }
+
+ printk(KERN_ERR "Unable to locate struct rd_dev_sg_table for page: %u\n",
+ page);
+
+ return NULL;
+}
+
+/* rd_MEMCPY_read():
+ *
+ *
+ */
+static int rd_MEMCPY_read(struct rd_request *req)
+{
+ struct se_task *task = &req->rd_task;
+ struct rd_dev *dev = req->rd_dev;
+ 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 -1;
+
+ 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];
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "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);
+#endif
+ src_offset = rd_offset;
+
+ while (req->rd_size) {
+ if ((sg_d[i].length - dst_offset) <
+ (sg_s[j].length - src_offset)) {
+ length = (sg_d[i].length - dst_offset);
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "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);
+ printk(KERN_INFO "Step 1 - length: %u dst_offset: %u"
+ " src_offset: %u\n", length, dst_offset,
+ src_offset);
+#endif
+ if (length > req->rd_size)
+ length = req->rd_size;
+
+ dst = sg_virt(&sg_d[i++]) + dst_offset;
+ if (!dst)
+ BUG();
+
+ src = sg_virt(&sg_s[j]) + src_offset;
+ if (!src)
+ BUG();
+
+ dst_offset = 0;
+ src_offset = length;
+ page_end = 0;
+ } else {
+ length = (sg_s[j].length - src_offset);
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "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);
+ printk(KERN_INFO "Step 2 - length: %u dst_offset: %u"
+ " src_offset: %u\n", length, dst_offset,
+ src_offset);
+#endif
+ if (length > req->rd_size)
+ length = req->rd_size;
+
+ dst = sg_virt(&sg_d[i]) + dst_offset;
+ if (!dst)
+ BUG();
+
+ if (sg_d[i].length == length) {
+ i++;
+ dst_offset = 0;
+ } else
+ dst_offset = length;
+
+ src = sg_virt(&sg_s[j++]) + src_offset;
+ if (!src)
+ BUG();
+
+ src_offset = 0;
+ page_end = 1;
+ }
+
+ memcpy(dst, src, length);
+
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "page: %u, remaining size: %u, length: %u,"
+ " i: %u, j: %u\n", req->rd_page,
+ (req->rd_size - length), length, i, j);
+#endif
+ req->rd_size -= length;
+ if (!(req->rd_size))
+ return 0;
+
+ if (!page_end)
+ continue;
+
+ if (++req->rd_page <= table->page_end_offset) {
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "page: %u in same page table\n",
+ req->rd_page);
+#endif
+ continue;
+ }
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "getting new page table for page: %u\n",
+ req->rd_page);
+#endif
+ table = rd_get_sg_table(dev, req->rd_page);
+ if (!(table))
+ return -1;
+
+ 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_dev;
+ 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 -1;
+
+ 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;
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "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);
+#endif
+ 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);
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "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);
+ printk(KERN_INFO "Step 1 - length: %u src_offset: %u"
+ " dst_offset: %u\n", length, src_offset,
+ dst_offset);
+#endif
+ if (length > req->rd_size)
+ length = req->rd_size;
+
+ src = sg_virt(&sg_s[i++]) + src_offset;
+ if (!src)
+ BUG();
+
+ dst = sg_virt(&sg_d[j]) + dst_offset;
+ if (!dst)
+ BUG();
+
+ src_offset = 0;
+ dst_offset = length;
+ page_end = 0;
+ } else {
+ length = (sg_d[j].length - dst_offset);
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "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);
+ printk(KERN_INFO "Step 2 - length: %u src_offset: %u"
+ " dst_offset: %u\n", length, src_offset,
+ dst_offset);
+#endif
+ if (length > req->rd_size)
+ length = req->rd_size;
+
+ src = sg_virt(&sg_s[i]) + src_offset;
+ if (!src)
+ BUG();
+
+ if (sg_s[i].length == length) {
+ i++;
+ src_offset = 0;
+ } else
+ src_offset = length;
+
+ dst = sg_virt(&sg_d[j++]) + dst_offset;
+ if (!dst)
+ BUG();
+
+ dst_offset = 0;
+ page_end = 1;
+ }
+
+ memcpy(dst, src, length);
+
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "page: %u, remaining size: %u, length: %u,"
+ " i: %u, j: %u\n", req->rd_page,
+ (req->rd_size - length), length, i, j);
+#endif
+ req->rd_size -= length;
+ if (!(req->rd_size))
+ return 0;
+
+ if (!page_end)
+ continue;
+
+ if (++req->rd_page <= table->page_end_offset) {
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "page: %u in same page table\n",
+ req->rd_page);
+#endif
+ continue;
+ }
+#ifdef DEBUG_RAMDISK_MCP
+ printk(KERN_INFO "getting new page table for page: %u\n",
+ req->rd_page);
+#endif
+ table = rd_get_sg_table(dev, req->rd_page);
+ if (!(table))
+ return -1;
+
+ sg_d = &table->sg_table[j = 0];
+ }
+
+ return 0;
+}
+
+/* rd_MEMCPY_do_task(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static int rd_MEMCPY_do_task(struct se_task *task)
+{
+ struct se_device *dev = task->se_dev;
+ struct rd_request *req = RD_REQ(task);
+ unsigned long long lba;
+ int ret;
+
+ req->rd_page = (task->task_lba * DEV_ATTRIB(dev)->block_size) / PAGE_SIZE;
+ lba = task->task_lba;
+ req->rd_offset = (do_div(lba,
+ (PAGE_SIZE / DEV_ATTRIB(dev)->block_size))) *
+ DEV_ATTRIB(dev)->block_size;
+ 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);
+
+ if (ret != 0)
+ return ret;
+
+ task->task_scsi_status = GOOD;
+ transport_complete_task(task, 1);
+
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+}
+
+/* rd_DIRECT_with_offset():
+ *
+ *
+ */
+static int rd_DIRECT_with_offset(
+ struct se_task *task,
+ struct list_head *se_mem_list,
+ u32 *se_mem_cnt,
+ u32 *task_offset)
+{
+ struct rd_request *req = RD_REQ(task);
+ struct rd_dev *dev = req->rd_dev;
+ struct rd_dev_sg_table *table;
+ struct se_mem *se_mem;
+ struct scatterlist *sg_s;
+ u32 j = 0, set_offset = 1;
+ u32 get_next_table = 0, offset_length, table_sg_end;
+
+ table = rd_get_sg_table(dev, req->rd_page);
+ if (!(table))
+ return -1;
+
+ table_sg_end = (table->page_end_offset - req->rd_page);
+ sg_s = &table->sg_table[req->rd_page - table->page_start_offset];
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "%s DIRECT LBA: %llu, Size: %u Page: %u, Offset: %u\n",
+ (task->task_data_direction == DMA_TO_DEVICE) ?
+ "Write" : "Read",
+ task->task_lba, req->rd_size, req->rd_page, req->rd_offset);
+#endif
+ while (req->rd_size) {
+ se_mem = kmem_cache_zalloc(se_mem_cache, GFP_KERNEL);
+ if (!(se_mem)) {
+ printk(KERN_ERR "Unable to allocate struct se_mem\n");
+ return -1;
+ }
+ INIT_LIST_HEAD(&se_mem->se_list);
+
+ if (set_offset) {
+ offset_length = sg_s[j].length - req->rd_offset;
+ if (offset_length > req->rd_size)
+ offset_length = req->rd_size;
+
+ se_mem->se_page = sg_page(&sg_s[j++]);
+ se_mem->se_off = req->rd_offset;
+ se_mem->se_len = offset_length;
+
+ set_offset = 0;
+ get_next_table = (j > table_sg_end);
+ goto check_eot;
+ }
+
+ offset_length = (req->rd_size < req->rd_offset) ?
+ req->rd_size : req->rd_offset;
+
+ se_mem->se_page = sg_page(&sg_s[j]);
+ se_mem->se_len = offset_length;
+
+ set_offset = 1;
+
+check_eot:
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "page: %u, size: %u, offset_length: %u, j: %u"
+ " se_mem: %p, se_page: %p se_off: %u se_len: %u\n",
+ req->rd_page, req->rd_size, offset_length, j, se_mem,
+ se_mem->se_page, se_mem->se_off, se_mem->se_len);
+#endif
+ list_add_tail(&se_mem->se_list, se_mem_list);
+ (*se_mem_cnt)++;
+
+ req->rd_size -= offset_length;
+ if (!(req->rd_size))
+ goto out;
+
+ if (!set_offset && !get_next_table)
+ continue;
+
+ if (++req->rd_page <= table->page_end_offset) {
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "page: %u in same page table\n",
+ req->rd_page);
+#endif
+ continue;
+ }
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "getting new page table for page: %u\n",
+ req->rd_page);
+#endif
+ table = rd_get_sg_table(dev, req->rd_page);
+ if (!(table))
+ return -1;
+
+ sg_s = &table->sg_table[j = 0];
+ }
+
+out:
+ T_TASK(task->task_se_cmd)->t_tasks_se_num += *se_mem_cnt;
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "RD_DR - Allocated %u struct se_mem segments for task\n",
+ *se_mem_cnt);
+#endif
+ return 0;
+}
+
+/* rd_DIRECT_without_offset():
+ *
+ *
+ */
+static int rd_DIRECT_without_offset(
+ struct se_task *task,
+ struct list_head *se_mem_list,
+ u32 *se_mem_cnt,
+ u32 *task_offset)
+{
+ struct rd_request *req = RD_REQ(task);
+ struct rd_dev *dev = req->rd_dev;
+ struct rd_dev_sg_table *table;
+ struct se_mem *se_mem;
+ struct scatterlist *sg_s;
+ u32 length, j = 0;
+
+ table = rd_get_sg_table(dev, req->rd_page);
+ if (!(table))
+ return -1;
+
+ sg_s = &table->sg_table[req->rd_page - table->page_start_offset];
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "%s DIRECT LBA: %llu, Size: %u, Page: %u\n",
+ (task->task_data_direction == DMA_TO_DEVICE) ?
+ "Write" : "Read",
+ task->task_lba, req->rd_size, req->rd_page);
+#endif
+ while (req->rd_size) {
+ se_mem = kmem_cache_zalloc(se_mem_cache, GFP_KERNEL);
+ if (!(se_mem)) {
+ printk(KERN_ERR "Unable to allocate struct se_mem\n");
+ return -1;
+ }
+ INIT_LIST_HEAD(&se_mem->se_list);
+
+ length = (req->rd_size < sg_s[j].length) ?
+ req->rd_size : sg_s[j].length;
+
+ se_mem->se_page = sg_page(&sg_s[j++]);
+ se_mem->se_len = length;
+
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "page: %u, size: %u, j: %u se_mem: %p,"
+ " se_page: %p se_off: %u se_len: %u\n", req->rd_page,
+ req->rd_size, j, se_mem, se_mem->se_page,
+ se_mem->se_off, se_mem->se_len);
+#endif
+ list_add_tail(&se_mem->se_list, se_mem_list);
+ (*se_mem_cnt)++;
+
+ req->rd_size -= length;
+ if (!(req->rd_size))
+ goto out;
+
+ if (++req->rd_page <= table->page_end_offset) {
+#ifdef DEBUG_RAMDISK_DR
+ printk("page: %u in same page table\n",
+ req->rd_page);
+#endif
+ continue;
+ }
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "getting new page table for page: %u\n",
+ req->rd_page);
+#endif
+ table = rd_get_sg_table(dev, req->rd_page);
+ if (!(table))
+ return -1;
+
+ sg_s = &table->sg_table[j = 0];
+ }
+
+out:
+ T_TASK(task->task_se_cmd)->t_tasks_se_num += *se_mem_cnt;
+#ifdef DEBUG_RAMDISK_DR
+ printk(KERN_INFO "RD_DR - Allocated %u struct se_mem segments for task\n",
+ *se_mem_cnt);
+#endif
+ return 0;
+}
+
+/* rd_DIRECT_do_se_mem_map():
+ *
+ *
+ */
+static int rd_DIRECT_do_se_mem_map(
+ struct se_task *task,
+ struct list_head *se_mem_list,
+ void *in_mem,
+ struct se_mem *in_se_mem,
+ struct se_mem **out_se_mem,
+ u32 *se_mem_cnt,
+ u32 *task_offset_in)
+{
+ struct se_cmd *cmd = task->task_se_cmd;
+ struct rd_request *req = RD_REQ(task);
+ u32 task_offset = *task_offset_in;
+ unsigned long long lba;
+ int ret;
+
+ req->rd_page = ((task->task_lba * DEV_ATTRIB(task->se_dev)->block_size) /
+ PAGE_SIZE);
+ lba = task->task_lba;
+ req->rd_offset = (do_div(lba,
+ (PAGE_SIZE / DEV_ATTRIB(task->se_dev)->block_size))) *
+ DEV_ATTRIB(task->se_dev)->block_size;
+ req->rd_size = task->task_size;
+
+ if (req->rd_offset)
+ ret = rd_DIRECT_with_offset(task, se_mem_list, se_mem_cnt,
+ task_offset_in);
+ else
+ ret = rd_DIRECT_without_offset(task, se_mem_list, se_mem_cnt,
+ task_offset_in);
+
+ if (ret < 0)
+ return ret;
+
+ if (CMD_TFO(cmd)->task_sg_chaining == 0)
+ return 0;
+ /*
+ * Currently prevent writers from multiple HW fabrics doing
+ * pci_map_sg() to RD_DR's internal scatterlist memory.
+ */
+ if (cmd->data_direction == DMA_TO_DEVICE) {
+ printk(KERN_ERR "DMA_TO_DEVICE not supported for"
+ " RAMDISK_DR with task_sg_chaining=1\n");
+ return -1;
+ }
+ /*
+ * Special case for if task_sg_chaining is enabled, then
+ * we setup struct se_task->task_sg[], as it will be used by
+ * transport_do_task_sg_chain() for creating chainged SGLs
+ * across multiple struct se_task->task_sg[].
+ */
+ if (!(transport_calc_sg_num(task,
+ list_entry(T_TASK(cmd)->t_mem_list->next,
+ struct se_mem, se_list),
+ task_offset)))
+ return -1;
+
+ return transport_map_mem_to_sg(task, se_mem_list, task->task_sg,
+ list_entry(T_TASK(cmd)->t_mem_list->next,
+ struct se_mem, se_list),
+ out_se_mem, se_mem_cnt, task_offset_in);
+}
+
+/* rd_DIRECT_do_task(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static int rd_DIRECT_do_task(struct se_task *task)
+{
+ /*
+ * At this point the locally allocated RD tables have been mapped
+ * to struct se_mem elements in rd_DIRECT_do_se_mem_map().
+ */
+ task->task_scsi_status = GOOD;
+ transport_complete_task(task, 1);
+
+ return PYX_TRANSPORT_SENT_TO_TRANSPORT;
+}
+
+/* rd_free_task(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static void rd_free_task(struct se_task *task)
+{
+ kfree(RD_REQ(task));
+}
+
+enum {
+ Opt_rd_pages, Opt_err
+};
+
+static match_table_t tokens = {
+ {Opt_rd_pages, "rd_pages=%d"},
+ {Opt_err, NULL}
+};
+
+static ssize_t rd_set_configfs_dev_params(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ const char *page,
+ ssize_t count)
+{
+ struct rd_dev *rd_dev = se_dev->se_dev_su_ptr;
+ char *orig, *ptr, *opts;
+ substring_t args[MAX_OPT_ARGS];
+ int ret = 0, arg, token;
+
+ opts = kstrdup(page, GFP_KERNEL);
+ if (!opts)
+ return -ENOMEM;
+
+ orig = opts;
+
+ while ((ptr = strsep(&opts, ",")) != NULL) {
+ if (!*ptr)
+ continue;
+
+ token = match_token(ptr, tokens, args);
+ switch (token) {
+ case Opt_rd_pages:
+ match_int(args, &arg);
+ rd_dev->rd_page_count = arg;
+ printk(KERN_INFO "RAMDISK: Referencing Page"
+ " Count: %u\n", rd_dev->rd_page_count);
+ rd_dev->rd_flags |= RDF_HAS_PAGE_COUNT;
+ break;
+ default:
+ break;
+ }
+ }
+
+ kfree(orig);
+ return (!ret) ? count : ret;
+}
+
+static ssize_t rd_check_configfs_dev_params(struct se_hba *hba, struct se_subsystem_dev *se_dev)
+{
+ struct rd_dev *rd_dev = se_dev->se_dev_su_ptr;
+
+ if (!(rd_dev->rd_flags & RDF_HAS_PAGE_COUNT)) {
+ printk(KERN_INFO "Missing rd_pages= parameter\n");
+ return -1;
+ }
+
+ return 0;
+}
+
+static ssize_t rd_show_configfs_dev_params(
+ struct se_hba *hba,
+ struct se_subsystem_dev *se_dev,
+ char *b)
+{
+ struct rd_dev *rd_dev = se_dev->se_dev_su_ptr;
+ ssize_t bl = sprintf(b, "TCM RamDisk ID: %u RamDisk Makeup: %s\n",
+ rd_dev->rd_dev_id, (rd_dev->rd_direct) ?
+ "rd_direct" : "rd_mcp");
+ bl += sprintf(b + bl, " PAGES/PAGE_SIZE: %u*%lu"
+ " SG_table_count: %u\n", rd_dev->rd_page_count,
+ PAGE_SIZE, rd_dev->sg_table_count);
+ return bl;
+}
+
+/* rd_get_cdb(): (Part of se_subsystem_api_t template)
+ *
+ *
+ */
+static unsigned char *rd_get_cdb(struct se_task *task)
+{
+ struct rd_request *req = RD_REQ(task);
+
+ return req->rd_scsi_cdb;
+}
+
+static u32 rd_get_device_rev(struct se_device *dev)
+{
+ return SCSI_SPC_2; /* Returns SPC-3 in Initiator Data */
+}
+
+static u32 rd_get_device_type(struct se_device *dev)
+{
+ return TYPE_DISK;
+}
+
+static sector_t rd_get_blocks(struct se_device *dev)
+{
+ struct rd_dev *rd_dev = dev->dev_ptr;
+ unsigned long long blocks_long = ((rd_dev->rd_page_count * PAGE_SIZE) /
+ DEV_ATTRIB(dev)->block_size) - 1;
+
+ return blocks_long;
+}
+
+static struct se_subsystem_api rd_dr_template = {
+ .name = "rd_dr",
+ .transport_type = TRANSPORT_PLUGIN_VHBA_VDEV,
+ .attach_hba = rd_attach_hba,
+ .detach_hba = rd_detach_hba,
+ .allocate_virtdevice = rd_DIRECT_allocate_virtdevice,
+ .create_virtdevice = rd_DIRECT_create_virtdevice,
+ .free_device = rd_free_device,
+ .alloc_task = rd_alloc_task,
+ .do_task = rd_DIRECT_do_task,
+ .free_task = rd_free_task,
+ .check_configfs_dev_params = rd_check_configfs_dev_params,
+ .set_configfs_dev_params = rd_set_configfs_dev_params,
+ .show_configfs_dev_params = rd_show_configfs_dev_params,
+ .get_cdb = rd_get_cdb,
+ .get_device_rev = rd_get_device_rev,
+ .get_device_type = rd_get_device_type,
+ .get_blocks = rd_get_blocks,
+ .do_se_mem_map = rd_DIRECT_do_se_mem_map,
+};
+
+static struct se_subsystem_api rd_mcp_template = {
+ .name = "rd_mcp",
+ .transport_type = TRANSPORT_PLUGIN_VHBA_VDEV,
+ .attach_hba = rd_attach_hba,
+ .detach_hba = rd_detach_hba,
+ .allocate_virtdevice = rd_MEMCPY_allocate_virtdevice,
+ .create_virtdevice = rd_MEMCPY_create_virtdevice,
+ .free_device = rd_free_device,
+ .alloc_task = rd_alloc_task,
+ .do_task = rd_MEMCPY_do_task,
+ .free_task = rd_free_task,
+ .check_configfs_dev_params = rd_check_configfs_dev_params,
+ .set_configfs_dev_params = rd_set_configfs_dev_params,
+ .show_configfs_dev_params = rd_show_configfs_dev_params,
+ .get_cdb = rd_get_cdb,
+ .get_device_rev = rd_get_device_rev,
+ .get_device_type = rd_get_device_type,
+ .get_blocks = rd_get_blocks,
+};
+
+int __init rd_module_init(void)
+{
+ int ret;
+
+ ret = transport_subsystem_register(&rd_dr_template);
+ if (ret < 0)
+ return ret;
+
+ ret = transport_subsystem_register(&rd_mcp_template);
+ if (ret < 0) {
+ transport_subsystem_release(&rd_dr_template);
+ return ret;
+ }
+
+ return 0;
+}
+
+void rd_module_exit(void)
+{
+ transport_subsystem_release(&rd_dr_template);
+ transport_subsystem_release(&rd_mcp_template);
+}
diff --git a/drivers/target/target_core_rd.h b/drivers/target/target_core_rd.h
new file mode 100644
index 0000000..13badfb
--- /dev/null
+++ b/drivers/target/target_core_rd.h
@@ -0,0 +1,73 @@
+#ifndef TARGET_CORE_RD_H
+#define TARGET_CORE_RD_H
+
+#define RD_HBA_VERSION "v4.0"
+#define RD_DR_VERSION "4.0"
+#define RD_MCP_VERSION "4.0"
+
+/* Largest piece of memory kmalloc can allocate */
+#define RD_MAX_ALLOCATION_SIZE 65536
+/* Maximum queuedepth for the Ramdisk HBA */
+#define RD_HBA_QUEUE_DEPTH 256
+#define RD_DEVICE_QUEUE_DEPTH 32
+#define RD_MAX_DEVICE_QUEUE_DEPTH 128
+#define RD_BLOCKSIZE 512
+#define RD_MAX_SECTORS 1024
+
+extern struct kmem_cache *se_mem_cache;
+
+/* Used in target_core_init_configfs() for virtual LUN 0 access */
+int __init rd_module_init(void);
+void rd_module_exit(void);
+
+#define RRF_EMULATE_CDB 0x01
+#define RRF_GOT_LBA 0x02
+
+struct rd_request {
+ struct se_task rd_task;
+
+ /* SCSI CDB from iSCSI Command PDU */
+ unsigned char rd_scsi_cdb[TCM_MAX_COMMAND_SIZE];
+ /* Offset from start of page */
+ u32 rd_offset;
+ /* Starting page in Ramdisk for request */
+ u32 rd_page;
+ /* Total number of pages needed for request */
+ u32 rd_page_count;
+ /* Scatterlist count */
+ u32 rd_size;
+ /* Ramdisk device */
+ struct rd_dev *rd_dev;
+} ____cacheline_aligned;
+
+struct rd_dev_sg_table {
+ u32 page_start_offset;
+ u32 page_end_offset;
+ u32 rd_sg_count;
+ struct scatterlist *sg_table;
+} ____cacheline_aligned;
+
+#define RDF_HAS_PAGE_COUNT 0x01
+
+struct rd_dev {
+ int rd_direct;
+ u32 rd_flags;
+ /* Unique Ramdisk Device ID in Ramdisk HBA */
+ u32 rd_dev_id;
+ /* Total page count for ramdisk device */
+ u32 rd_page_count;
+ /* Number of SG tables in sg_table_array */
+ u32 sg_table_count;
+ u32 rd_queue_depth;
+ /* Array of rd_dev_sg_table_t containing scatterlists */
+ struct rd_dev_sg_table *sg_table_array;
+ /* Ramdisk HBA device is connected to */
+ struct rd_host *rd_host;
+} ____cacheline_aligned;
+
+struct rd_host {
+ u32 rd_host_dev_id_count;
+ u32 rd_host_id; /* Unique Ramdisk Host ID */
+} ____cacheline_aligned;
+
+#endif /* TARGET_CORE_RD_H */
diff --git a/drivers/target/target_core_scdb.c b/drivers/target/target_core_scdb.c
new file mode 100644
index 0000000..dc6fed0
--- /dev/null
+++ b/drivers/target/target_core_scdb.c
@@ -0,0 +1,105 @@
+/*******************************************************************************
+ * Filename: target_core_scdb.c
+ *
+ * This file contains the generic target engine Split CDB related functions.
+ *
+ * Copyright (c) 2004-2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/net.h>
+#include <linux/string.h>
+#include <scsi/scsi.h>
+#include <asm/unaligned.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_transport.h>
+
+#include "target_core_scdb.h"
+
+/* split_cdb_XX_6():
+ *
+ * 21-bit LBA w/ 8-bit SECTORS
+ */
+void split_cdb_XX_6(
+ unsigned long long lba,
+ u32 *sectors,
+ unsigned char *cdb)
+{
+ cdb[1] = (lba >> 16) & 0x1f;
+ cdb[2] = (lba >> 8) & 0xff;
+ cdb[3] = lba & 0xff;
+ cdb[4] = *sectors & 0xff;
+}
+
+/* split_cdb_XX_10():
+ *
+ * 32-bit LBA w/ 16-bit SECTORS
+ */
+void split_cdb_XX_10(
+ unsigned long long lba,
+ u32 *sectors,
+ unsigned char *cdb)
+{
+ put_unaligned_be32(lba, &cdb[2]);
+ put_unaligned_be16(*sectors, &cdb[7]);
+}
+
+/* split_cdb_XX_12():
+ *
+ * 32-bit LBA w/ 32-bit SECTORS
+ */
+void split_cdb_XX_12(
+ unsigned long long lba,
+ u32 *sectors,
+ unsigned char *cdb)
+{
+ put_unaligned_be32(lba, &cdb[2]);
+ put_unaligned_be32(*sectors, &cdb[6]);
+}
+
+/* split_cdb_XX_16():
+ *
+ * 64-bit LBA w/ 32-bit SECTORS
+ */
+void split_cdb_XX_16(
+ unsigned long long lba,
+ u32 *sectors,
+ unsigned char *cdb)
+{
+ put_unaligned_be64(lba, &cdb[2]);
+ put_unaligned_be32(*sectors, &cdb[10]);
+}
+
+/*
+ * split_cdb_XX_32():
+ *
+ * 64-bit LBA w/ 32-bit SECTORS such as READ_32, WRITE_32 and emulated XDWRITEREAD_32
+ */
+void split_cdb_XX_32(
+ unsigned long long lba,
+ u32 *sectors,
+ unsigned char *cdb)
+{
+ put_unaligned_be64(lba, &cdb[12]);
+ put_unaligned_be32(*sectors, &cdb[28]);
+}
diff --git a/drivers/target/target_core_scdb.h b/drivers/target/target_core_scdb.h
new file mode 100644
index 0000000..98cd1c0
--- /dev/null
+++ b/drivers/target/target_core_scdb.h
@@ -0,0 +1,10 @@
+#ifndef TARGET_CORE_SCDB_H
+#define TARGET_CORE_SCDB_H
+
+extern void split_cdb_XX_6(unsigned long long, u32 *, unsigned char *);
+extern void split_cdb_XX_10(unsigned long long, u32 *, unsigned char *);
+extern void split_cdb_XX_12(unsigned long long, u32 *, unsigned char *);
+extern void split_cdb_XX_16(unsigned long long, u32 *, unsigned char *);
+extern void split_cdb_XX_32(unsigned long long, u32 *, unsigned char *);
+
+#endif /* TARGET_CORE_SCDB_H */
diff --git a/drivers/target/target_core_tmr.c b/drivers/target/target_core_tmr.c
new file mode 100644
index 0000000..158cecb
--- /dev/null
+++ b/drivers/target/target_core_tmr.c
@@ -0,0 +1,404 @@
+/*******************************************************************************
+ * Filename: target_core_tmr.c
+ *
+ * This file contains SPC-3 task management infrastructure
+ *
+ * Copyright (c) 2009,2010 Rising Tide Systems
+ * Copyright (c) 2009,2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/list.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_tmr.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_configfs.h>
+
+#include "target_core_alua.h"
+#include "target_core_pr.h"
+
+#define DEBUG_LUN_RESET
+#ifdef DEBUG_LUN_RESET
+#define DEBUG_LR(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_LR(x...)
+#endif
+
+struct se_tmr_req *core_tmr_alloc_req(
+ struct se_cmd *se_cmd,
+ void *fabric_tmr_ptr,
+ u8 function)
+{
+ struct se_tmr_req *tmr;
+
+ tmr = kmem_cache_zalloc(se_tmr_req_cache, GFP_KERNEL);
+ if (!(tmr)) {
+ printk(KERN_ERR "Unable to allocate struct se_tmr_req\n");
+ return ERR_PTR(-ENOMEM);
+ }
+ tmr->task_cmd = se_cmd;
+ tmr->fabric_tmr_ptr = fabric_tmr_ptr;
+ tmr->function = function;
+ INIT_LIST_HEAD(&tmr->tmr_list);
+
+ return tmr;
+}
+EXPORT_SYMBOL(core_tmr_alloc_req);
+
+void core_tmr_release_req(
+ struct se_tmr_req *tmr)
+{
+ struct se_device *dev = tmr->tmr_dev;
+
+ spin_lock(&dev->se_tmr_lock);
+ list_del(&tmr->tmr_list);
+ kmem_cache_free(se_tmr_req_cache, tmr);
+ spin_unlock(&dev->se_tmr_lock);
+}
+
+static void core_tmr_handle_tas_abort(
+ struct se_node_acl *tmr_nacl,
+ struct se_cmd *cmd,
+ int tas,
+ int fe_count)
+{
+ if (!(fe_count)) {
+ transport_cmd_finish_abort(cmd, 1);
+ return;
+ }
+ /*
+ * TASK ABORTED status (TAS) bit support
+ */
+ if (((tmr_nacl != NULL) &&
+ (tmr_nacl == cmd->se_sess->se_node_acl)) || tas)
+ transport_send_task_abort(cmd);
+
+ transport_cmd_finish_abort(cmd, 0);
+}
+
+int core_tmr_lun_reset(
+ struct se_device *dev,
+ struct se_tmr_req *tmr,
+ struct list_head *preempt_and_abort_list,
+ struct se_cmd *prout_cmd)
+{
+ struct se_cmd *cmd;
+ struct se_queue_req *qr, *qr_tmp;
+ struct se_node_acl *tmr_nacl = NULL;
+ struct se_portal_group *tmr_tpg = NULL;
+ struct se_queue_obj *qobj = dev->dev_queue_obj;
+ struct se_tmr_req *tmr_p, *tmr_pp;
+ struct se_task *task, *task_tmp;
+ unsigned long flags;
+ int fe_count, state, tas;
+ /*
+ * TASK_ABORTED status bit, this is configurable via ConfigFS
+ * struct se_device attributes. spc4r17 section 7.4.6 Control mode page
+ *
+ * A task aborted status (TAS) bit set to zero specifies that aborted
+ * tasks shall be terminated by the device server without any response
+ * to the application client. A TAS bit set to one specifies that tasks
+ * aborted by the actions of an I_T nexus other than the I_T nexus on
+ * which the command was received shall be completed with TASK ABORTED
+ * status (see SAM-4).
+ */
+ tas = DEV_ATTRIB(dev)->emulate_tas;
+ /*
+ * Determine if this se_tmr is coming from a $FABRIC_MOD
+ * or struct se_device passthrough..
+ */
+ if (tmr && tmr->task_cmd && tmr->task_cmd->se_sess) {
+ tmr_nacl = tmr->task_cmd->se_sess->se_node_acl;
+ tmr_tpg = tmr->task_cmd->se_sess->se_tpg;
+ if (tmr_nacl && tmr_tpg) {
+ DEBUG_LR("LUN_RESET: TMR caller fabric: %s"
+ " initiator port %s\n",
+ TPG_TFO(tmr_tpg)->get_fabric_name(),
+ tmr_nacl->initiatorname);
+ }
+ }
+ DEBUG_LR("LUN_RESET: %s starting for [%s], tas: %d\n",
+ (preempt_and_abort_list) ? "Preempt" : "TMR",
+ TRANSPORT(dev)->name, tas);
+ /*
+ * Release all pending and outgoing TMRs aside from the received
+ * LUN_RESET tmr..
+ */
+ spin_lock(&dev->se_tmr_lock);
+ list_for_each_entry_safe(tmr_p, tmr_pp, &dev->dev_tmr_list, tmr_list) {
+ /*
+ * Allow the received TMR to return with FUNCTION_COMPLETE.
+ */
+ if (tmr && (tmr_p == tmr))
+ continue;
+
+ cmd = tmr_p->task_cmd;
+ if (!(cmd)) {
+ printk(KERN_ERR "Unable to locate struct se_cmd for TMR\n");
+ continue;
+ }
+ /*
+ * If this function was called with a valid pr_res_key
+ * parameter (eg: for PROUT PREEMPT_AND_ABORT service action
+ * skip non regisration key matching TMRs.
+ */
+ if ((preempt_and_abort_list != NULL) &&
+ (core_scsi3_check_cdb_abort_and_preempt(
+ preempt_and_abort_list, cmd) != 0))
+ continue;
+ spin_unlock(&dev->se_tmr_lock);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (!(atomic_read(&T_TASK(cmd)->t_transport_active))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ spin_lock(&dev->se_tmr_lock);
+ continue;
+ }
+ if (cmd->t_state == TRANSPORT_ISTATE_PROCESSING) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ spin_lock(&dev->se_tmr_lock);
+ continue;
+ }
+ DEBUG_LR("LUN_RESET: %s releasing TMR %p Function: 0x%02x,"
+ " Response: 0x%02x, t_state: %d\n",
+ (preempt_and_abort_list) ? "Preempt" : "", tmr_p,
+ tmr_p->function, tmr_p->response, cmd->t_state);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_cmd_finish_abort_tmr(cmd);
+ spin_lock(&dev->se_tmr_lock);
+ }
+ spin_unlock(&dev->se_tmr_lock);
+ /*
+ * Complete outstanding struct se_task CDBs with TASK_ABORTED SAM status.
+ * This is following sam4r17, section 5.6 Aborting commands, Table 38
+ * for TMR LUN_RESET:
+ *
+ * a) "Yes" indicates that each command that is aborted on an I_T nexus
+ * other than the one that caused the SCSI device condition is
+ * completed with TASK ABORTED status, if the TAS bit is set to one in
+ * the Control mode page (see SPC-4). "No" indicates that no status is
+ * returned for aborted commands.
+ *
+ * d) If the logical unit reset is caused by a particular I_T nexus
+ * (e.g., by a LOGICAL UNIT RESET task management function), then "yes"
+ * (TASK_ABORTED status) applies.
+ *
+ * Otherwise (e.g., if triggered by a hard reset), "no"
+ * (no TASK_ABORTED SAM status) applies.
+ *
+ * Note that this seems to be independent of TAS (Task Aborted Status)
+ * in the Control Mode Page.
+ */
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ list_for_each_entry_safe(task, task_tmp, &dev->state_task_list,
+ t_state_list) {
+ if (!(TASK_CMD(task))) {
+ printk(KERN_ERR "TASK_CMD(task) is NULL!\n");
+ continue;
+ }
+ cmd = TASK_CMD(task);
+
+ if (!T_TASK(cmd)) {
+ printk(KERN_ERR "T_TASK(cmd) is NULL for task: %p cmd:"
+ " %p ITT: 0x%08x\n", task, cmd,
+ CMD_TFO(cmd)->get_task_tag(cmd));
+ continue;
+ }
+ /*
+ * For PREEMPT_AND_ABORT usage, only process commands
+ * with a matching reservation key.
+ */
+ if ((preempt_and_abort_list != NULL) &&
+ (core_scsi3_check_cdb_abort_and_preempt(
+ preempt_and_abort_list, cmd) != 0))
+ continue;
+ /*
+ * Not aborting PROUT PREEMPT_AND_ABORT CDB..
+ */
+ if (prout_cmd == cmd)
+ continue;
+
+ list_del(&task->t_state_list);
+ atomic_set(&task->task_state_active, 0);
+ spin_unlock_irqrestore(&dev->execute_task_lock, flags);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ DEBUG_LR("LUN_RESET: %s cmd: %p task: %p"
+ " ITT/CmdSN: 0x%08x/0x%08x, i_state: %d, t_state/"
+ "def_t_state: %d/%d cdb: 0x%02x\n",
+ (preempt_and_abort_list) ? "Preempt" : "", cmd, task,
+ CMD_TFO(cmd)->get_task_tag(cmd), 0,
+ CMD_TFO(cmd)->get_cmd_state(cmd), cmd->t_state,
+ cmd->deferred_t_state, T_TASK(cmd)->t_task_cdb[0]);
+ DEBUG_LR("LUN_RESET: ITT[0x%08x] - pr_res_key: 0x%016Lx"
+ " t_task_cdbs: %d t_task_cdbs_left: %d"
+ " t_task_cdbs_sent: %d -- t_transport_active: %d"
+ " t_transport_stop: %d t_transport_sent: %d\n",
+ CMD_TFO(cmd)->get_task_tag(cmd), cmd->pr_res_key,
+ T_TASK(cmd)->t_task_cdbs,
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_left),
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_sent),
+ atomic_read(&T_TASK(cmd)->t_transport_active),
+ atomic_read(&T_TASK(cmd)->t_transport_stop),
+ atomic_read(&T_TASK(cmd)->t_transport_sent));
+
+ if (atomic_read(&task->task_active)) {
+ atomic_set(&task->task_stop, 1);
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+
+ DEBUG_LR("LUN_RESET: Waiting for task: %p to shutdown"
+ " for dev: %p\n", task, dev);
+ wait_for_completion(&task->task_stop_comp);
+ DEBUG_LR("LUN_RESET Completed task: %p shutdown for"
+ " dev: %p\n", task, dev);
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ atomic_dec(&T_TASK(cmd)->t_task_cdbs_left);
+
+ atomic_set(&task->task_active, 0);
+ atomic_set(&task->task_stop, 0);
+ }
+ __transport_stop_task_timer(task, &flags);
+
+ if (!(atomic_dec_and_test(&T_TASK(cmd)->t_task_cdbs_ex_left))) {
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+ DEBUG_LR("LUN_RESET: Skipping task: %p, dev: %p for"
+ " t_task_cdbs_ex_left: %d\n", task, dev,
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_ex_left));
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ continue;
+ }
+ fe_count = atomic_read(&T_TASK(cmd)->t_fe_count);
+
+ if (atomic_read(&T_TASK(cmd)->t_transport_active)) {
+ DEBUG_LR("LUN_RESET: got t_transport_active = 1 for"
+ " task: %p, t_fe_count: %d dev: %p\n", task,
+ fe_count, dev);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock,
+ flags);
+ core_tmr_handle_tas_abort(tmr_nacl, cmd, tas, fe_count);
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ continue;
+ }
+ DEBUG_LR("LUN_RESET: Got t_transport_active = 0 for task: %p,"
+ " t_fe_count: %d dev: %p\n", task, fe_count, dev);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ core_tmr_handle_tas_abort(tmr_nacl, cmd, tas, fe_count);
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ }
+ spin_unlock_irqrestore(&dev->execute_task_lock, flags);
+ /*
+ * Release all commands remaining in the struct se_device cmd queue.
+ *
+ * This follows the same logic as above for the struct se_device
+ * struct se_task state list, where commands are returned with
+ * TASK_ABORTED status, if there is an outstanding $FABRIC_MOD
+ * reference, otherwise the struct se_cmd is released.
+ */
+ spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
+ list_for_each_entry_safe(qr, qr_tmp, &qobj->qobj_list, qr_list) {
+ cmd = (struct se_cmd *)qr->cmd;
+ if (!(cmd)) {
+ /*
+ * Skip these for non PREEMPT_AND_ABORT usage..
+ */
+ if (preempt_and_abort_list != NULL)
+ continue;
+
+ atomic_dec(&qobj->queue_cnt);
+ list_del(&qr->qr_list);
+ kfree(qr);
+ continue;
+ }
+ /*
+ * For PREEMPT_AND_ABORT usage, only process commands
+ * with a matching reservation key.
+ */
+ if ((preempt_and_abort_list != NULL) &&
+ (core_scsi3_check_cdb_abort_and_preempt(
+ preempt_and_abort_list, cmd) != 0))
+ continue;
+ /*
+ * Not aborting PROUT PREEMPT_AND_ABORT CDB..
+ */
+ if (prout_cmd == cmd)
+ continue;
+
+ atomic_dec(&T_TASK(cmd)->t_transport_queue_active);
+ atomic_dec(&qobj->queue_cnt);
+ list_del(&qr->qr_list);
+ spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
+
+ state = qr->state;
+ kfree(qr);
+
+ DEBUG_LR("LUN_RESET: %s from Device Queue: cmd: %p t_state:"
+ " %d t_fe_count: %d\n", (preempt_and_abort_list) ?
+ "Preempt" : "", cmd, state,
+ atomic_read(&T_TASK(cmd)->t_fe_count));
+ /*
+ * Signal that the command has failed via cmd->se_cmd_flags,
+ * and call TFO->new_cmd_failure() to wakeup any fabric
+ * dependent code used to wait for unsolicited data out
+ * allocation to complete. The fabric module is expected
+ * to dump any remaining unsolicited data out for the aborted
+ * command at this point.
+ */
+ transport_new_cmd_failure(cmd);
+
+ core_tmr_handle_tas_abort(tmr_nacl, cmd, tas,
+ atomic_read(&T_TASK(cmd)->t_fe_count));
+ spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
+ }
+ spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
+ /*
+ * Clear any legacy SPC-2 reservation when called during
+ * LOGICAL UNIT RESET
+ */
+ if (!(preempt_and_abort_list) &&
+ (dev->dev_flags & DF_SPC2_RESERVATIONS)) {
+ spin_lock(&dev->dev_reservation_lock);
+ dev->dev_reserved_node_acl = NULL;
+ dev->dev_flags &= ~DF_SPC2_RESERVATIONS;
+ spin_unlock(&dev->dev_reservation_lock);
+ printk(KERN_INFO "LUN_RESET: SCSI-2 Released reservation\n");
+ }
+
+ spin_lock(&dev->stats_lock);
+ dev->num_resets++;
+ spin_unlock(&dev->stats_lock);
+
+ DEBUG_LR("LUN_RESET: %s for [%s] Complete\n",
+ (preempt_and_abort_list) ? "Preempt" : "TMR",
+ TRANSPORT(dev)->name);
+ return 0;
+}
diff --git a/drivers/target/target_core_tpg.c b/drivers/target/target_core_tpg.c
new file mode 100644
index 0000000..abfa81a
--- /dev/null
+++ b/drivers/target/target_core_tpg.c
@@ -0,0 +1,826 @@
+/*******************************************************************************
+ * Filename: target_core_tpg.c
+ *
+ * This file contains generic Target Portal Group related functions.
+ *
+ * Copyright (c) 2002, 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/net.h>
+#include <linux/string.h>
+#include <linux/timer.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <linux/in.h>
+#include <net/sock.h>
+#include <net/tcp.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_tpg.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+
+#include "target_core_hba.h"
+
+/* core_clear_initiator_node_from_tpg():
+ *
+ *
+ */
+static void core_clear_initiator_node_from_tpg(
+ struct se_node_acl *nacl,
+ struct se_portal_group *tpg)
+{
+ int i;
+ struct se_dev_entry *deve;
+ struct se_lun *lun;
+ struct se_lun_acl *acl, *acl_tmp;
+
+ spin_lock_irq(&nacl->device_list_lock);
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ deve = &nacl->device_list[i];
+
+ if (!(deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS))
+ continue;
+
+ if (!deve->se_lun) {
+ printk(KERN_ERR "%s device entries device pointer is"
+ " NULL, but Initiator has access.\n",
+ TPG_TFO(tpg)->get_fabric_name());
+ continue;
+ }
+
+ lun = deve->se_lun;
+ spin_unlock_irq(&nacl->device_list_lock);
+ core_update_device_list_for_node(lun, NULL, deve->mapped_lun,
+ TRANSPORT_LUNFLAGS_NO_ACCESS, nacl, tpg, 0);
+
+ spin_lock(&lun->lun_acl_lock);
+ list_for_each_entry_safe(acl, acl_tmp,
+ &lun->lun_acl_list, lacl_list) {
+ if (!(strcmp(acl->initiatorname,
+ nacl->initiatorname)) &&
+ (acl->mapped_lun == deve->mapped_lun))
+ break;
+ }
+
+ if (!acl) {
+ printk(KERN_ERR "Unable to locate struct se_lun_acl for %s,"
+ " mapped_lun: %u\n", nacl->initiatorname,
+ deve->mapped_lun);
+ spin_unlock(&lun->lun_acl_lock);
+ spin_lock_irq(&nacl->device_list_lock);
+ continue;
+ }
+
+ list_del(&acl->lacl_list);
+ spin_unlock(&lun->lun_acl_lock);
+
+ spin_lock_irq(&nacl->device_list_lock);
+ kfree(acl);
+ }
+ spin_unlock_irq(&nacl->device_list_lock);
+}
+
+/* __core_tpg_get_initiator_node_acl():
+ *
+ * spin_lock_bh(&tpg->acl_node_lock); must be held when calling
+ */
+struct se_node_acl *__core_tpg_get_initiator_node_acl(
+ struct se_portal_group *tpg,
+ const char *initiatorname)
+{
+ struct se_node_acl *acl;
+
+ list_for_each_entry(acl, &tpg->acl_node_list, acl_list) {
+ if (!(strcmp(acl->initiatorname, initiatorname)))
+ return acl;
+ }
+
+ return NULL;
+}
+
+/* core_tpg_get_initiator_node_acl():
+ *
+ *
+ */
+struct se_node_acl *core_tpg_get_initiator_node_acl(
+ struct se_portal_group *tpg,
+ unsigned char *initiatorname)
+{
+ struct se_node_acl *acl;
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ list_for_each_entry(acl, &tpg->acl_node_list, acl_list) {
+ if (!(strcmp(acl->initiatorname, initiatorname)) &&
+ (!(acl->dynamic_node_acl))) {
+ spin_unlock_bh(&tpg->acl_node_lock);
+ return acl;
+ }
+ }
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+ return NULL;
+}
+
+/* core_tpg_add_node_to_devs():
+ *
+ *
+ */
+void core_tpg_add_node_to_devs(
+ struct se_node_acl *acl,
+ struct se_portal_group *tpg)
+{
+ int i = 0;
+ u32 lun_access = 0;
+ struct se_lun *lun;
+ struct se_device *dev;
+
+ spin_lock(&tpg->tpg_lun_lock);
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ lun = &tpg->tpg_lun_list[i];
+ if (lun->lun_status != TRANSPORT_LUN_STATUS_ACTIVE)
+ continue;
+
+ spin_unlock(&tpg->tpg_lun_lock);
+
+ dev = lun->lun_se_dev;
+ /*
+ * By default in LIO-Target $FABRIC_MOD,
+ * demo_mode_write_protect is ON, or READ_ONLY;
+ */
+ if (!(TPG_TFO(tpg)->tpg_check_demo_mode_write_protect(tpg))) {
+ if (dev->dev_flags & DF_READ_ONLY)
+ lun_access = TRANSPORT_LUNFLAGS_READ_ONLY;
+ else
+ lun_access = TRANSPORT_LUNFLAGS_READ_WRITE;
+ } else {
+ /*
+ * Allow only optical drives to issue R/W in default RO
+ * demo mode.
+ */
+ if (TRANSPORT(dev)->get_device_type(dev) == TYPE_DISK)
+ lun_access = TRANSPORT_LUNFLAGS_READ_ONLY;
+ else
+ lun_access = TRANSPORT_LUNFLAGS_READ_WRITE;
+ }
+
+ printk(KERN_INFO "TARGET_CORE[%s]->TPG[%u]_LUN[%u] - Adding %s"
+ " access for LUN in Demo Mode\n",
+ TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), lun->unpacked_lun,
+ (lun_access == TRANSPORT_LUNFLAGS_READ_WRITE) ?
+ "READ-WRITE" : "READ-ONLY");
+
+ core_update_device_list_for_node(lun, NULL, lun->unpacked_lun,
+ lun_access, acl, tpg, 1);
+ spin_lock(&tpg->tpg_lun_lock);
+ }
+ spin_unlock(&tpg->tpg_lun_lock);
+}
+
+/* core_set_queue_depth_for_node():
+ *
+ *
+ */
+static int core_set_queue_depth_for_node(
+ struct se_portal_group *tpg,
+ struct se_node_acl *acl)
+{
+ if (!acl->queue_depth) {
+ printk(KERN_ERR "Queue depth for %s Initiator Node: %s is 0,"
+ "defaulting to 1.\n", TPG_TFO(tpg)->get_fabric_name(),
+ acl->initiatorname);
+ acl->queue_depth = 1;
+ }
+
+ return 0;
+}
+
+/* core_create_device_list_for_node():
+ *
+ *
+ */
+static int core_create_device_list_for_node(struct se_node_acl *nacl)
+{
+ struct se_dev_entry *deve;
+ int i;
+
+ nacl->device_list = kzalloc(sizeof(struct se_dev_entry) *
+ TRANSPORT_MAX_LUNS_PER_TPG, GFP_KERNEL);
+ if (!(nacl->device_list)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " struct se_node_acl->device_list\n");
+ return -1;
+ }
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ deve = &nacl->device_list[i];
+
+ atomic_set(&deve->ua_count, 0);
+ atomic_set(&deve->pr_ref_count, 0);
+ spin_lock_init(&deve->ua_lock);
+ INIT_LIST_HEAD(&deve->alua_port_list);
+ INIT_LIST_HEAD(&deve->ua_list);
+ }
+
+ return 0;
+}
+
+/* core_tpg_check_initiator_node_acl()
+ *
+ *
+ */
+struct se_node_acl *core_tpg_check_initiator_node_acl(
+ struct se_portal_group *tpg,
+ unsigned char *initiatorname)
+{
+ struct se_node_acl *acl;
+
+ acl = core_tpg_get_initiator_node_acl(tpg, initiatorname);
+ if ((acl))
+ return acl;
+
+ if (!(TPG_TFO(tpg)->tpg_check_demo_mode(tpg)))
+ return NULL;
+
+ acl = TPG_TFO(tpg)->tpg_alloc_fabric_acl(tpg);
+ if (!(acl))
+ return NULL;
+
+ INIT_LIST_HEAD(&acl->acl_list);
+ INIT_LIST_HEAD(&acl->acl_sess_list);
+ spin_lock_init(&acl->device_list_lock);
+ spin_lock_init(&acl->nacl_sess_lock);
+ atomic_set(&acl->acl_pr_ref_count, 0);
+ atomic_set(&acl->mib_ref_count, 0);
+ acl->queue_depth = TPG_TFO(tpg)->tpg_get_default_depth(tpg);
+ snprintf(acl->initiatorname, TRANSPORT_IQN_LEN, "%s", initiatorname);
+ acl->se_tpg = tpg;
+ acl->acl_index = scsi_get_new_index(SCSI_AUTH_INTR_INDEX);
+ spin_lock_init(&acl->stats_lock);
+ acl->dynamic_node_acl = 1;
+
+ TPG_TFO(tpg)->set_default_node_attributes(acl);
+
+ if (core_create_device_list_for_node(acl) < 0) {
+ TPG_TFO(tpg)->tpg_release_fabric_acl(tpg, acl);
+ return NULL;
+ }
+
+ if (core_set_queue_depth_for_node(tpg, acl) < 0) {
+ core_free_device_list_for_node(acl, tpg);
+ TPG_TFO(tpg)->tpg_release_fabric_acl(tpg, acl);
+ return NULL;
+ }
+
+ core_tpg_add_node_to_devs(acl, tpg);
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ list_add_tail(&acl->acl_list, &tpg->acl_node_list);
+ tpg->num_node_acls++;
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+ printk("%s_TPG[%u] - Added DYNAMIC ACL with TCQ Depth: %d for %s"
+ " Initiator Node: %s\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), acl->queue_depth,
+ TPG_TFO(tpg)->get_fabric_name(), initiatorname);
+
+ return acl;
+}
+EXPORT_SYMBOL(core_tpg_check_initiator_node_acl);
+
+void core_tpg_wait_for_nacl_pr_ref(struct se_node_acl *nacl)
+{
+ while (atomic_read(&nacl->acl_pr_ref_count) != 0)
+ cpu_relax();
+}
+
+void core_tpg_wait_for_mib_ref(struct se_node_acl *nacl)
+{
+ while (atomic_read(&nacl->mib_ref_count) != 0)
+ cpu_relax();
+}
+
+void core_tpg_clear_object_luns(struct se_portal_group *tpg)
+{
+ int i, ret;
+ struct se_lun *lun;
+
+ spin_lock(&tpg->tpg_lun_lock);
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ lun = &tpg->tpg_lun_list[i];
+
+ if ((lun->lun_status != TRANSPORT_LUN_STATUS_ACTIVE) ||
+ (lun->lun_se_dev == NULL))
+ continue;
+
+ spin_unlock(&tpg->tpg_lun_lock);
+ ret = core_dev_del_lun(tpg, lun->unpacked_lun);
+ spin_lock(&tpg->tpg_lun_lock);
+ }
+ spin_unlock(&tpg->tpg_lun_lock);
+}
+EXPORT_SYMBOL(core_tpg_clear_object_luns);
+
+/* core_tpg_add_initiator_node_acl():
+ *
+ *
+ */
+struct se_node_acl *core_tpg_add_initiator_node_acl(
+ struct se_portal_group *tpg,
+ struct se_node_acl *se_nacl,
+ const char *initiatorname,
+ u32 queue_depth)
+{
+ struct se_node_acl *acl = NULL;
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ acl = __core_tpg_get_initiator_node_acl(tpg, initiatorname);
+ if ((acl)) {
+ if (acl->dynamic_node_acl) {
+ acl->dynamic_node_acl = 0;
+ printk(KERN_INFO "%s_TPG[%u] - Replacing dynamic ACL"
+ " for %s\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), initiatorname);
+ spin_unlock_bh(&tpg->acl_node_lock);
+ /*
+ * Release the locally allocated struct se_node_acl
+ * because * core_tpg_add_initiator_node_acl() returned
+ * a pointer to an existing demo mode node ACL.
+ */
+ if (se_nacl)
+ TPG_TFO(tpg)->tpg_release_fabric_acl(tpg,
+ se_nacl);
+ goto done;
+ }
+
+ printk(KERN_ERR "ACL entry for %s Initiator"
+ " Node %s already exists for TPG %u, ignoring"
+ " request.\n", TPG_TFO(tpg)->get_fabric_name(),
+ initiatorname, TPG_TFO(tpg)->tpg_get_tag(tpg));
+ spin_unlock_bh(&tpg->acl_node_lock);
+ return ERR_PTR(-EEXIST);
+ }
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+ if (!(se_nacl)) {
+ printk("struct se_node_acl pointer is NULL\n");
+ return ERR_PTR(-EINVAL);
+ }
+ /*
+ * For v4.x logic the se_node_acl_s is hanging off a fabric
+ * dependent structure allocated via
+ * struct target_core_fabric_ops->fabric_make_nodeacl()
+ */
+ acl = se_nacl;
+
+ INIT_LIST_HEAD(&acl->acl_list);
+ INIT_LIST_HEAD(&acl->acl_sess_list);
+ spin_lock_init(&acl->device_list_lock);
+ spin_lock_init(&acl->nacl_sess_lock);
+ atomic_set(&acl->acl_pr_ref_count, 0);
+ acl->queue_depth = queue_depth;
+ snprintf(acl->initiatorname, TRANSPORT_IQN_LEN, "%s", initiatorname);
+ acl->se_tpg = tpg;
+ acl->acl_index = scsi_get_new_index(SCSI_AUTH_INTR_INDEX);
+ spin_lock_init(&acl->stats_lock);
+
+ TPG_TFO(tpg)->set_default_node_attributes(acl);
+
+ if (core_create_device_list_for_node(acl) < 0) {
+ TPG_TFO(tpg)->tpg_release_fabric_acl(tpg, acl);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ if (core_set_queue_depth_for_node(tpg, acl) < 0) {
+ core_free_device_list_for_node(acl, tpg);
+ TPG_TFO(tpg)->tpg_release_fabric_acl(tpg, acl);
+ return ERR_PTR(-EINVAL);
+ }
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ list_add_tail(&acl->acl_list, &tpg->acl_node_list);
+ tpg->num_node_acls++;
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+done:
+ printk(KERN_INFO "%s_TPG[%hu] - Added ACL with TCQ Depth: %d for %s"
+ " Initiator Node: %s\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), acl->queue_depth,
+ TPG_TFO(tpg)->get_fabric_name(), initiatorname);
+
+ return acl;
+}
+EXPORT_SYMBOL(core_tpg_add_initiator_node_acl);
+
+/* core_tpg_del_initiator_node_acl():
+ *
+ *
+ */
+int core_tpg_del_initiator_node_acl(
+ struct se_portal_group *tpg,
+ struct se_node_acl *acl,
+ int force)
+{
+ struct se_session *sess, *sess_tmp;
+ int dynamic_acl = 0;
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ if (acl->dynamic_node_acl) {
+ acl->dynamic_node_acl = 0;
+ dynamic_acl = 1;
+ }
+ list_del(&acl->acl_list);
+ tpg->num_node_acls--;
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+ spin_lock_bh(&tpg->session_lock);
+ list_for_each_entry_safe(sess, sess_tmp,
+ &tpg->tpg_sess_list, sess_list) {
+ if (sess->se_node_acl != acl)
+ continue;
+ /*
+ * Determine if the session needs to be closed by our context.
+ */
+ if (!(TPG_TFO(tpg)->shutdown_session(sess)))
+ continue;
+
+ spin_unlock_bh(&tpg->session_lock);
+ /*
+ * If the $FABRIC_MOD session for the Initiator Node ACL exists,
+ * forcefully shutdown the $FABRIC_MOD session/nexus.
+ */
+ TPG_TFO(tpg)->close_session(sess);
+
+ spin_lock_bh(&tpg->session_lock);
+ }
+ spin_unlock_bh(&tpg->session_lock);
+
+ core_tpg_wait_for_nacl_pr_ref(acl);
+ core_tpg_wait_for_mib_ref(acl);
+ core_clear_initiator_node_from_tpg(acl, tpg);
+ core_free_device_list_for_node(acl, tpg);
+
+ printk(KERN_INFO "%s_TPG[%hu] - Deleted ACL with TCQ Depth: %d for %s"
+ " Initiator Node: %s\n", TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg), acl->queue_depth,
+ TPG_TFO(tpg)->get_fabric_name(), acl->initiatorname);
+
+ return 0;
+}
+EXPORT_SYMBOL(core_tpg_del_initiator_node_acl);
+
+/* core_tpg_set_initiator_node_queue_depth():
+ *
+ *
+ */
+int core_tpg_set_initiator_node_queue_depth(
+ struct se_portal_group *tpg,
+ unsigned char *initiatorname,
+ u32 queue_depth,
+ int force)
+{
+ struct se_session *sess, *init_sess = NULL;
+ struct se_node_acl *acl;
+ int dynamic_acl = 0;
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ acl = __core_tpg_get_initiator_node_acl(tpg, initiatorname);
+ if (!(acl)) {
+ printk(KERN_ERR "Access Control List entry for %s Initiator"
+ " Node %s does not exists for TPG %hu, ignoring"
+ " request.\n", TPG_TFO(tpg)->get_fabric_name(),
+ initiatorname, TPG_TFO(tpg)->tpg_get_tag(tpg));
+ spin_unlock_bh(&tpg->acl_node_lock);
+ return -ENODEV;
+ }
+ if (acl->dynamic_node_acl) {
+ acl->dynamic_node_acl = 0;
+ dynamic_acl = 1;
+ }
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+ spin_lock_bh(&tpg->session_lock);
+ list_for_each_entry(sess, &tpg->tpg_sess_list, sess_list) {
+ if (sess->se_node_acl != acl)
+ continue;
+
+ if (!force) {
+ printk(KERN_ERR "Unable to change queue depth for %s"
+ " Initiator Node: %s while session is"
+ " operational. To forcefully change the queue"
+ " depth and force session reinstatement"
+ " use the \"force=1\" parameter.\n",
+ TPG_TFO(tpg)->get_fabric_name(), initiatorname);
+ spin_unlock_bh(&tpg->session_lock);
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ if (dynamic_acl)
+ acl->dynamic_node_acl = 1;
+ spin_unlock_bh(&tpg->acl_node_lock);
+ return -EEXIST;
+ }
+ /*
+ * Determine if the session needs to be closed by our context.
+ */
+ if (!(TPG_TFO(tpg)->shutdown_session(sess)))
+ continue;
+
+ init_sess = sess;
+ break;
+ }
+
+ /*
+ * User has requested to change the queue depth for a Initiator Node.
+ * Change the value in the Node's struct se_node_acl, and call
+ * core_set_queue_depth_for_node() to add the requested queue depth.
+ *
+ * Finally call TPG_TFO(tpg)->close_session() to force session
+ * reinstatement to occur if there is an active session for the
+ * $FABRIC_MOD Initiator Node in question.
+ */
+ acl->queue_depth = queue_depth;
+
+ if (core_set_queue_depth_for_node(tpg, acl) < 0) {
+ spin_unlock_bh(&tpg->session_lock);
+ /*
+ * Force session reinstatement if
+ * core_set_queue_depth_for_node() failed, because we assume
+ * the $FABRIC_MOD has already the set session reinstatement
+ * bit from TPG_TFO(tpg)->shutdown_session() called above.
+ */
+ if (init_sess)
+ TPG_TFO(tpg)->close_session(init_sess);
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ if (dynamic_acl)
+ acl->dynamic_node_acl = 1;
+ spin_unlock_bh(&tpg->acl_node_lock);
+ return -EINVAL;
+ }
+ spin_unlock_bh(&tpg->session_lock);
+ /*
+ * If the $FABRIC_MOD session for the Initiator Node ACL exists,
+ * forcefully shutdown the $FABRIC_MOD session/nexus.
+ */
+ if (init_sess)
+ TPG_TFO(tpg)->close_session(init_sess);
+
+ printk(KERN_INFO "Successfuly changed queue depth to: %d for Initiator"
+ " Node: %s on %s Target Portal Group: %u\n", queue_depth,
+ initiatorname, TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+
+ spin_lock_bh(&tpg->acl_node_lock);
+ if (dynamic_acl)
+ acl->dynamic_node_acl = 1;
+ spin_unlock_bh(&tpg->acl_node_lock);
+
+ return 0;
+}
+EXPORT_SYMBOL(core_tpg_set_initiator_node_queue_depth);
+
+static int core_tpg_setup_virtual_lun0(struct se_portal_group *se_tpg)
+{
+ /* Set in core_dev_setup_virtual_lun0() */
+ struct se_device *dev = se_global->g_lun0_dev;
+ struct se_lun *lun = &se_tpg->tpg_virt_lun0;
+ u32 lun_access = TRANSPORT_LUNFLAGS_READ_ONLY;
+ int ret;
+
+ lun->unpacked_lun = 0;
+ lun->lun_status = TRANSPORT_LUN_STATUS_FREE;
+ atomic_set(&lun->lun_acl_count, 0);
+ init_completion(&lun->lun_shutdown_comp);
+ INIT_LIST_HEAD(&lun->lun_acl_list);
+ INIT_LIST_HEAD(&lun->lun_cmd_list);
+ spin_lock_init(&lun->lun_acl_lock);
+ spin_lock_init(&lun->lun_cmd_lock);
+ spin_lock_init(&lun->lun_sep_lock);
+
+ ret = core_tpg_post_addlun(se_tpg, lun, lun_access, dev);
+ if (ret < 0)
+ return -1;
+
+ return 0;
+}
+
+static void core_tpg_release_virtual_lun0(struct se_portal_group *se_tpg)
+{
+ struct se_lun *lun = &se_tpg->tpg_virt_lun0;
+
+ core_tpg_post_dellun(se_tpg, lun);
+}
+
+int core_tpg_register(
+ struct target_core_fabric_ops *tfo,
+ struct se_wwn *se_wwn,
+ struct se_portal_group *se_tpg,
+ void *tpg_fabric_ptr,
+ int se_tpg_type)
+{
+ struct se_lun *lun;
+ u32 i;
+
+ se_tpg->tpg_lun_list = kzalloc((sizeof(struct se_lun) *
+ TRANSPORT_MAX_LUNS_PER_TPG), GFP_KERNEL);
+ if (!(se_tpg->tpg_lun_list)) {
+ printk(KERN_ERR "Unable to allocate struct se_portal_group->"
+ "tpg_lun_list\n");
+ return -ENOMEM;
+ }
+
+ for (i = 0; i < TRANSPORT_MAX_LUNS_PER_TPG; i++) {
+ lun = &se_tpg->tpg_lun_list[i];
+ lun->unpacked_lun = i;
+ lun->lun_status = TRANSPORT_LUN_STATUS_FREE;
+ atomic_set(&lun->lun_acl_count, 0);
+ init_completion(&lun->lun_shutdown_comp);
+ INIT_LIST_HEAD(&lun->lun_acl_list);
+ INIT_LIST_HEAD(&lun->lun_cmd_list);
+ spin_lock_init(&lun->lun_acl_lock);
+ spin_lock_init(&lun->lun_cmd_lock);
+ spin_lock_init(&lun->lun_sep_lock);
+ }
+
+ se_tpg->se_tpg_type = se_tpg_type;
+ se_tpg->se_tpg_fabric_ptr = tpg_fabric_ptr;
+ se_tpg->se_tpg_tfo = tfo;
+ se_tpg->se_tpg_wwn = se_wwn;
+ atomic_set(&se_tpg->tpg_pr_ref_count, 0);
+ INIT_LIST_HEAD(&se_tpg->acl_node_list);
+ INIT_LIST_HEAD(&se_tpg->se_tpg_list);
+ INIT_LIST_HEAD(&se_tpg->tpg_sess_list);
+ spin_lock_init(&se_tpg->acl_node_lock);
+ spin_lock_init(&se_tpg->session_lock);
+ spin_lock_init(&se_tpg->tpg_lun_lock);
+
+ if (se_tpg->se_tpg_type == TRANSPORT_TPG_TYPE_NORMAL) {
+ if (core_tpg_setup_virtual_lun0(se_tpg) < 0) {
+ kfree(se_tpg);
+ return -ENOMEM;
+ }
+ }
+
+ spin_lock_bh(&se_global->se_tpg_lock);
+ list_add_tail(&se_tpg->se_tpg_list, &se_global->g_se_tpg_list);
+ spin_unlock_bh(&se_global->se_tpg_lock);
+
+ printk(KERN_INFO "TARGET_CORE[%s]: Allocated %s struct se_portal_group for"
+ " endpoint: %s, Portal Tag: %u\n", tfo->get_fabric_name(),
+ (se_tpg->se_tpg_type == TRANSPORT_TPG_TYPE_NORMAL) ?
+ "Normal" : "Discovery", (tfo->tpg_get_wwn(se_tpg) == NULL) ?
+ "None" : tfo->tpg_get_wwn(se_tpg), tfo->tpg_get_tag(se_tpg));
+
+ return 0;
+}
+EXPORT_SYMBOL(core_tpg_register);
+
+int core_tpg_deregister(struct se_portal_group *se_tpg)
+{
+ printk(KERN_INFO "TARGET_CORE[%s]: Deallocating %s struct se_portal_group"
+ " for endpoint: %s Portal Tag %u\n",
+ (se_tpg->se_tpg_type == TRANSPORT_TPG_TYPE_NORMAL) ?
+ "Normal" : "Discovery", TPG_TFO(se_tpg)->get_fabric_name(),
+ TPG_TFO(se_tpg)->tpg_get_wwn(se_tpg),
+ TPG_TFO(se_tpg)->tpg_get_tag(se_tpg));
+
+ spin_lock_bh(&se_global->se_tpg_lock);
+ list_del(&se_tpg->se_tpg_list);
+ spin_unlock_bh(&se_global->se_tpg_lock);
+
+ while (atomic_read(&se_tpg->tpg_pr_ref_count) != 0)
+ cpu_relax();
+
+ if (se_tpg->se_tpg_type == TRANSPORT_TPG_TYPE_NORMAL)
+ core_tpg_release_virtual_lun0(se_tpg);
+
+ se_tpg->se_tpg_fabric_ptr = NULL;
+ kfree(se_tpg->tpg_lun_list);
+ return 0;
+}
+EXPORT_SYMBOL(core_tpg_deregister);
+
+struct se_lun *core_tpg_pre_addlun(
+ struct se_portal_group *tpg,
+ u32 unpacked_lun)
+{
+ struct se_lun *lun;
+
+ if (unpacked_lun > (TRANSPORT_MAX_LUNS_PER_TPG-1)) {
+ printk(KERN_ERR "%s LUN: %u exceeds TRANSPORT_MAX_LUNS_PER_TPG"
+ "-1: %u for Target Portal Group: %u\n",
+ TPG_TFO(tpg)->get_fabric_name(),
+ unpacked_lun, TRANSPORT_MAX_LUNS_PER_TPG-1,
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ return ERR_PTR(-EOVERFLOW);
+ }
+
+ spin_lock(&tpg->tpg_lun_lock);
+ lun = &tpg->tpg_lun_list[unpacked_lun];
+ if (lun->lun_status == TRANSPORT_LUN_STATUS_ACTIVE) {
+ printk(KERN_ERR "TPG Logical Unit Number: %u is already active"
+ " on %s Target Portal Group: %u, ignoring request.\n",
+ unpacked_lun, TPG_TFO(tpg)->get_fabric_name(),
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ spin_unlock(&tpg->tpg_lun_lock);
+ return ERR_PTR(-EINVAL);
+ }
+ spin_unlock(&tpg->tpg_lun_lock);
+
+ return lun;
+}
+
+int core_tpg_post_addlun(
+ struct se_portal_group *tpg,
+ struct se_lun *lun,
+ u32 lun_access,
+ void *lun_ptr)
+{
+ if (core_dev_export(lun_ptr, tpg, lun) < 0)
+ return -1;
+
+ spin_lock(&tpg->tpg_lun_lock);
+ lun->lun_access = lun_access;
+ lun->lun_status = TRANSPORT_LUN_STATUS_ACTIVE;
+ spin_unlock(&tpg->tpg_lun_lock);
+
+ return 0;
+}
+
+static void core_tpg_shutdown_lun(
+ struct se_portal_group *tpg,
+ struct se_lun *lun)
+{
+ core_clear_lun_from_tpg(lun, tpg);
+ transport_clear_lun_from_sessions(lun);
+}
+
+struct se_lun *core_tpg_pre_dellun(
+ struct se_portal_group *tpg,
+ u32 unpacked_lun,
+ int *ret)
+{
+ struct se_lun *lun;
+
+ if (unpacked_lun > (TRANSPORT_MAX_LUNS_PER_TPG-1)) {
+ printk(KERN_ERR "%s LUN: %u exceeds TRANSPORT_MAX_LUNS_PER_TPG"
+ "-1: %u for Target Portal Group: %u\n",
+ TPG_TFO(tpg)->get_fabric_name(), unpacked_lun,
+ TRANSPORT_MAX_LUNS_PER_TPG-1,
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ return ERR_PTR(-EOVERFLOW);
+ }
+
+ spin_lock(&tpg->tpg_lun_lock);
+ lun = &tpg->tpg_lun_list[unpacked_lun];
+ if (lun->lun_status != TRANSPORT_LUN_STATUS_ACTIVE) {
+ printk(KERN_ERR "%s Logical Unit Number: %u is not active on"
+ " Target Portal Group: %u, ignoring request.\n",
+ TPG_TFO(tpg)->get_fabric_name(), unpacked_lun,
+ TPG_TFO(tpg)->tpg_get_tag(tpg));
+ spin_unlock(&tpg->tpg_lun_lock);
+ return ERR_PTR(-ENODEV);
+ }
+ spin_unlock(&tpg->tpg_lun_lock);
+
+ return lun;
+}
+
+int core_tpg_post_dellun(
+ struct se_portal_group *tpg,
+ struct se_lun *lun)
+{
+ core_tpg_shutdown_lun(tpg, lun);
+
+ core_dev_unexport(lun->lun_se_dev, tpg, lun);
+
+ spin_lock(&tpg->tpg_lun_lock);
+ lun->lun_status = TRANSPORT_LUN_STATUS_FREE;
+ spin_unlock(&tpg->tpg_lun_lock);
+
+ return 0;
+}
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
new file mode 100644
index 0000000..28b6292ff
--- /dev/null
+++ b/drivers/target/target_core_transport.c
@@ -0,0 +1,6134 @@
+/*******************************************************************************
+ * Filename: target_core_transport.c
+ *
+ * This file contains the Generic Target Engine Core.
+ *
+ * Copyright (c) 2002, 2003, 2004, 2005 PyX Technologies, Inc.
+ * Copyright (c) 2005, 2006, 2007 SBE, Inc.
+ * Copyright (c) 2007-2010 Rising Tide Systems
+ * Copyright (c) 2008-2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/net.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/timer.h>
+#include <linux/slab.h>
+#include <linux/blkdev.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <linux/kthread.h>
+#include <linux/in.h>
+#include <linux/cdrom.h>
+#include <asm/unaligned.h>
+#include <net/sock.h>
+#include <net/tcp.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/libsas.h> /* For TASK_ATTR_* */
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_tmr.h>
+#include <target/target_core_tpg.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_configfs.h>
+
+#include "target_core_alua.h"
+#include "target_core_hba.h"
+#include "target_core_pr.h"
+#include "target_core_scdb.h"
+#include "target_core_ua.h"
+
+/* #define DEBUG_CDB_HANDLER */
+#ifdef DEBUG_CDB_HANDLER
+#define DEBUG_CDB_H(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_CDB_H(x...)
+#endif
+
+/* #define DEBUG_CMD_MAP */
+#ifdef DEBUG_CMD_MAP
+#define DEBUG_CMD_M(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_CMD_M(x...)
+#endif
+
+/* #define DEBUG_MEM_ALLOC */
+#ifdef DEBUG_MEM_ALLOC
+#define DEBUG_MEM(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_MEM(x...)
+#endif
+
+/* #define DEBUG_MEM2_ALLOC */
+#ifdef DEBUG_MEM2_ALLOC
+#define DEBUG_MEM2(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_MEM2(x...)
+#endif
+
+/* #define DEBUG_SG_CALC */
+#ifdef DEBUG_SG_CALC
+#define DEBUG_SC(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_SC(x...)
+#endif
+
+/* #define DEBUG_SE_OBJ */
+#ifdef DEBUG_SE_OBJ
+#define DEBUG_SO(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_SO(x...)
+#endif
+
+/* #define DEBUG_CMD_VOL */
+#ifdef DEBUG_CMD_VOL
+#define DEBUG_VOL(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_VOL(x...)
+#endif
+
+/* #define DEBUG_CMD_STOP */
+#ifdef DEBUG_CMD_STOP
+#define DEBUG_CS(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_CS(x...)
+#endif
+
+/* #define DEBUG_PASSTHROUGH */
+#ifdef DEBUG_PASSTHROUGH
+#define DEBUG_PT(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_PT(x...)
+#endif
+
+/* #define DEBUG_TASK_STOP */
+#ifdef DEBUG_TASK_STOP
+#define DEBUG_TS(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_TS(x...)
+#endif
+
+/* #define DEBUG_TRANSPORT_STOP */
+#ifdef DEBUG_TRANSPORT_STOP
+#define DEBUG_TRANSPORT_S(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_TRANSPORT_S(x...)
+#endif
+
+/* #define DEBUG_TASK_FAILURE */
+#ifdef DEBUG_TASK_FAILURE
+#define DEBUG_TF(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_TF(x...)
+#endif
+
+/* #define DEBUG_DEV_OFFLINE */
+#ifdef DEBUG_DEV_OFFLINE
+#define DEBUG_DO(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_DO(x...)
+#endif
+
+/* #define DEBUG_TASK_STATE */
+#ifdef DEBUG_TASK_STATE
+#define DEBUG_TSTATE(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_TSTATE(x...)
+#endif
+
+/* #define DEBUG_STATUS_THR */
+#ifdef DEBUG_STATUS_THR
+#define DEBUG_ST(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_ST(x...)
+#endif
+
+/* #define DEBUG_TASK_TIMEOUT */
+#ifdef DEBUG_TASK_TIMEOUT
+#define DEBUG_TT(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_TT(x...)
+#endif
+
+/* #define DEBUG_GENERIC_REQUEST_FAILURE */
+#ifdef DEBUG_GENERIC_REQUEST_FAILURE
+#define DEBUG_GRF(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_GRF(x...)
+#endif
+
+/* #define DEBUG_SAM_TASK_ATTRS */
+#ifdef DEBUG_SAM_TASK_ATTRS
+#define DEBUG_STA(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_STA(x...)
+#endif
+
+struct se_global *se_global;
+
+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;
+struct kmem_cache *se_mem_cache;
+struct kmem_cache *t10_pr_reg_cache;
+struct kmem_cache *t10_alua_lu_gp_cache;
+struct kmem_cache *t10_alua_lu_gp_mem_cache;
+struct kmem_cache *t10_alua_tg_pt_gp_cache;
+struct kmem_cache *t10_alua_tg_pt_gp_mem_cache;
+
+/* Used for transport_dev_get_map_*() */
+typedef int (*map_func_t)(struct se_task *, u32);
+
+static int transport_generic_write_pending(struct se_cmd *);
+static int transport_processing_thread(void *);
+static int __transport_execute_tasks(struct se_device *dev);
+static void transport_complete_task_attr(struct se_cmd *cmd);
+static void transport_direct_request_timeout(struct se_cmd *cmd);
+static void transport_free_dev_tasks(struct se_cmd *cmd);
+static u32 transport_generic_get_cdb_count(struct se_cmd *cmd,
+ unsigned long long starting_lba, u32 sectors,
+ enum dma_data_direction data_direction,
+ struct list_head *mem_list, int set_counts);
+static int transport_generic_get_mem(struct se_cmd *cmd, u32 length,
+ u32 dma_size);
+static int transport_generic_remove(struct se_cmd *cmd,
+ int release_to_pool, int session_reinstatement);
+static int transport_get_sectors(struct se_cmd *cmd);
+static struct list_head *transport_init_se_mem_list(void);
+static int transport_map_sg_to_mem(struct se_cmd *cmd,
+ struct list_head *se_mem_list, void *in_mem,
+ u32 *se_mem_cnt);
+static void transport_memcpy_se_mem_read_contig(struct se_cmd *cmd,
+ unsigned char *dst, struct list_head *se_mem_list);
+static void transport_release_fe_cmd(struct se_cmd *cmd);
+static void transport_remove_cmd_from_queue(struct se_cmd *cmd,
+ struct se_queue_obj *qobj);
+static int transport_set_sense_codes(struct se_cmd *cmd, u8 asc, u8 ascq);
+static void transport_stop_all_task_timers(struct se_cmd *cmd);
+
+int transport_emulate_control_cdb(struct se_task *task);
+
+int init_se_global(void)
+{
+ struct se_global *global;
+
+ global = kzalloc(sizeof(struct se_global), GFP_KERNEL);
+ if (!(global)) {
+ printk(KERN_ERR "Unable to allocate memory for struct se_global\n");
+ return -1;
+ }
+
+ INIT_LIST_HEAD(&global->g_lu_gps_list);
+ INIT_LIST_HEAD(&global->g_se_tpg_list);
+ INIT_LIST_HEAD(&global->g_hba_list);
+ INIT_LIST_HEAD(&global->g_se_dev_list);
+ spin_lock_init(&global->g_device_lock);
+ spin_lock_init(&global->hba_lock);
+ spin_lock_init(&global->se_tpg_lock);
+ spin_lock_init(&global->lu_gps_lock);
+ spin_lock_init(&global->plugin_class_lock);
+
+ se_cmd_cache = kmem_cache_create("se_cmd_cache",
+ sizeof(struct se_cmd), __alignof__(struct se_cmd), 0, NULL);
+ if (!(se_cmd_cache)) {
+ printk(KERN_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)) {
+ printk(KERN_ERR "kmem_cache_create() for struct se_tmr_req"
+ " failed\n");
+ goto out;
+ }
+ se_sess_cache = kmem_cache_create("se_sess_cache",
+ sizeof(struct se_session), __alignof__(struct se_session),
+ 0, NULL);
+ if (!(se_sess_cache)) {
+ printk(KERN_ERR "kmem_cache_create() for struct se_session"
+ " failed\n");
+ goto out;
+ }
+ se_ua_cache = kmem_cache_create("se_ua_cache",
+ sizeof(struct se_ua), __alignof__(struct se_ua),
+ 0, NULL);
+ if (!(se_ua_cache)) {
+ printk(KERN_ERR "kmem_cache_create() for struct se_ua failed\n");
+ goto out;
+ }
+ se_mem_cache = kmem_cache_create("se_mem_cache",
+ sizeof(struct se_mem), __alignof__(struct se_mem), 0, NULL);
+ if (!(se_mem_cache)) {
+ printk(KERN_ERR "kmem_cache_create() for struct se_mem failed\n");
+ goto out;
+ }
+ t10_pr_reg_cache = kmem_cache_create("t10_pr_reg_cache",
+ sizeof(struct t10_pr_registration),
+ __alignof__(struct t10_pr_registration), 0, NULL);
+ if (!(t10_pr_reg_cache)) {
+ printk(KERN_ERR "kmem_cache_create() for struct t10_pr_registration"
+ " failed\n");
+ goto out;
+ }
+ t10_alua_lu_gp_cache = kmem_cache_create("t10_alua_lu_gp_cache",
+ sizeof(struct t10_alua_lu_gp), __alignof__(struct t10_alua_lu_gp),
+ 0, NULL);
+ if (!(t10_alua_lu_gp_cache)) {
+ printk(KERN_ERR "kmem_cache_create() for t10_alua_lu_gp_cache"
+ " failed\n");
+ goto out;
+ }
+ t10_alua_lu_gp_mem_cache = kmem_cache_create("t10_alua_lu_gp_mem_cache",
+ sizeof(struct t10_alua_lu_gp_member),
+ __alignof__(struct t10_alua_lu_gp_member), 0, NULL);
+ if (!(t10_alua_lu_gp_mem_cache)) {
+ printk(KERN_ERR "kmem_cache_create() for t10_alua_lu_gp_mem_"
+ "cache failed\n");
+ goto out;
+ }
+ t10_alua_tg_pt_gp_cache = kmem_cache_create("t10_alua_tg_pt_gp_cache",
+ sizeof(struct t10_alua_tg_pt_gp),
+ __alignof__(struct t10_alua_tg_pt_gp), 0, NULL);
+ if (!(t10_alua_tg_pt_gp_cache)) {
+ printk(KERN_ERR "kmem_cache_create() for t10_alua_tg_pt_gp_"
+ "cache failed\n");
+ goto out;
+ }
+ t10_alua_tg_pt_gp_mem_cache = kmem_cache_create(
+ "t10_alua_tg_pt_gp_mem_cache",
+ sizeof(struct t10_alua_tg_pt_gp_member),
+ __alignof__(struct t10_alua_tg_pt_gp_member),
+ 0, NULL);
+ if (!(t10_alua_tg_pt_gp_mem_cache)) {
+ printk(KERN_ERR "kmem_cache_create() for t10_alua_tg_pt_gp_"
+ "mem_t failed\n");
+ goto out;
+ }
+
+ se_global = global;
+
+ return 0;
+out:
+ if (se_cmd_cache)
+ kmem_cache_destroy(se_cmd_cache);
+ if (se_tmr_req_cache)
+ kmem_cache_destroy(se_tmr_req_cache);
+ if (se_sess_cache)
+ kmem_cache_destroy(se_sess_cache);
+ if (se_ua_cache)
+ kmem_cache_destroy(se_ua_cache);
+ if (se_mem_cache)
+ kmem_cache_destroy(se_mem_cache);
+ if (t10_pr_reg_cache)
+ kmem_cache_destroy(t10_pr_reg_cache);
+ if (t10_alua_lu_gp_cache)
+ kmem_cache_destroy(t10_alua_lu_gp_cache);
+ if (t10_alua_lu_gp_mem_cache)
+ kmem_cache_destroy(t10_alua_lu_gp_mem_cache);
+ if (t10_alua_tg_pt_gp_cache)
+ kmem_cache_destroy(t10_alua_tg_pt_gp_cache);
+ if (t10_alua_tg_pt_gp_mem_cache)
+ kmem_cache_destroy(t10_alua_tg_pt_gp_mem_cache);
+ kfree(global);
+ return -1;
+}
+
+void release_se_global(void)
+{
+ struct se_global *global;
+
+ global = se_global;
+ if (!(global))
+ return;
+
+ 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);
+ kmem_cache_destroy(se_mem_cache);
+ kmem_cache_destroy(t10_pr_reg_cache);
+ kmem_cache_destroy(t10_alua_lu_gp_cache);
+ kmem_cache_destroy(t10_alua_lu_gp_mem_cache);
+ kmem_cache_destroy(t10_alua_tg_pt_gp_cache);
+ kmem_cache_destroy(t10_alua_tg_pt_gp_mem_cache);
+ kfree(global);
+
+ se_global = NULL;
+}
+
+void transport_init_queue_obj(struct se_queue_obj *qobj)
+{
+ atomic_set(&qobj->queue_cnt, 0);
+ INIT_LIST_HEAD(&qobj->qobj_list);
+ init_waitqueue_head(&qobj->thread_wq);
+ spin_lock_init(&qobj->cmd_queue_lock);
+}
+EXPORT_SYMBOL(transport_init_queue_obj);
+
+static int transport_subsystem_reqmods(void)
+{
+ int ret;
+
+ ret = request_module("target_core_iblock");
+ if (ret != 0)
+ printk(KERN_ERR "Unable to load target_core_iblock\n");
+
+ ret = request_module("target_core_file");
+ if (ret != 0)
+ printk(KERN_ERR "Unable to load target_core_file\n");
+
+ ret = request_module("target_core_pscsi");
+ if (ret != 0)
+ printk(KERN_ERR "Unable to load target_core_pscsi\n");
+
+ ret = request_module("target_core_stgt");
+ if (ret != 0)
+ printk(KERN_ERR "Unable to load target_core_stgt\n");
+
+ return 0;
+}
+
+int transport_subsystem_check_init(void)
+{
+ if (se_global->g_sub_api_initialized)
+ return 0;
+ /*
+ * Request the loading of known TCM subsystem plugins..
+ */
+ if (transport_subsystem_reqmods() < 0)
+ return -1;
+
+ se_global->g_sub_api_initialized = 1;
+ return 0;
+}
+
+struct se_session *transport_init_session(void)
+{
+ struct se_session *se_sess;
+
+ se_sess = kmem_cache_zalloc(se_sess_cache, GFP_KERNEL);
+ if (!(se_sess)) {
+ printk(KERN_ERR "Unable to allocate struct se_session from"
+ " se_sess_cache\n");
+ return ERR_PTR(-ENOMEM);
+ }
+ INIT_LIST_HEAD(&se_sess->sess_list);
+ INIT_LIST_HEAD(&se_sess->sess_acl_list);
+ atomic_set(&se_sess->mib_ref_count, 0);
+
+ return se_sess;
+}
+EXPORT_SYMBOL(transport_init_session);
+
+/*
+ * Called with spin_lock_bh(&struct se_portal_group->session_lock called.
+ */
+void __transport_register_session(
+ struct se_portal_group *se_tpg,
+ struct se_node_acl *se_nacl,
+ struct se_session *se_sess,
+ void *fabric_sess_ptr)
+{
+ unsigned char buf[PR_REG_ISID_LEN];
+
+ se_sess->se_tpg = se_tpg;
+ se_sess->fabric_sess_ptr = fabric_sess_ptr;
+ /*
+ * Used by struct se_node_acl's under ConfigFS to locate active se_session-t
+ *
+ * Only set for struct se_session's that will actually be moving I/O.
+ * eg: *NOT* discovery sessions.
+ */
+ if (se_nacl) {
+ /*
+ * If the fabric module supports an ISID based TransportID,
+ * save this value in binary from the fabric I_T Nexus now.
+ */
+ if (TPG_TFO(se_tpg)->sess_get_initiator_sid != NULL) {
+ memset(&buf[0], 0, PR_REG_ISID_LEN);
+ TPG_TFO(se_tpg)->sess_get_initiator_sid(se_sess,
+ &buf[0], PR_REG_ISID_LEN);
+ se_sess->sess_bin_isid = get_unaligned_be64(&buf[0]);
+ }
+ spin_lock_irq(&se_nacl->nacl_sess_lock);
+ /*
+ * The se_nacl->nacl_sess pointer will be set to the
+ * last active I_T Nexus for each struct se_node_acl.
+ */
+ se_nacl->nacl_sess = se_sess;
+
+ list_add_tail(&se_sess->sess_acl_list,
+ &se_nacl->acl_sess_list);
+ spin_unlock_irq(&se_nacl->nacl_sess_lock);
+ }
+ list_add_tail(&se_sess->sess_list, &se_tpg->tpg_sess_list);
+
+ printk(KERN_INFO "TARGET_CORE[%s]: Registered fabric_sess_ptr: %p\n",
+ TPG_TFO(se_tpg)->get_fabric_name(), se_sess->fabric_sess_ptr);
+}
+EXPORT_SYMBOL(__transport_register_session);
+
+void transport_register_session(
+ struct se_portal_group *se_tpg,
+ struct se_node_acl *se_nacl,
+ struct se_session *se_sess,
+ void *fabric_sess_ptr)
+{
+ spin_lock_bh(&se_tpg->session_lock);
+ __transport_register_session(se_tpg, se_nacl, se_sess, fabric_sess_ptr);
+ spin_unlock_bh(&se_tpg->session_lock);
+}
+EXPORT_SYMBOL(transport_register_session);
+
+void transport_deregister_session_configfs(struct se_session *se_sess)
+{
+ struct se_node_acl *se_nacl;
+
+ /*
+ * Used by struct se_node_acl's under ConfigFS to locate active struct se_session
+ */
+ se_nacl = se_sess->se_node_acl;
+ if ((se_nacl)) {
+ spin_lock_irq(&se_nacl->nacl_sess_lock);
+ list_del(&se_sess->sess_acl_list);
+ /*
+ * If the session list is empty, then clear the pointer.
+ * Otherwise, set the struct se_session pointer from the tail
+ * element of the per struct se_node_acl active session list.
+ */
+ if (list_empty(&se_nacl->acl_sess_list))
+ se_nacl->nacl_sess = NULL;
+ else {
+ se_nacl->nacl_sess = container_of(
+ se_nacl->acl_sess_list.prev,
+ struct se_session, sess_acl_list);
+ }
+ spin_unlock_irq(&se_nacl->nacl_sess_lock);
+ }
+}
+EXPORT_SYMBOL(transport_deregister_session_configfs);
+
+void transport_free_session(struct se_session *se_sess)
+{
+ kmem_cache_free(se_sess_cache, se_sess);
+}
+EXPORT_SYMBOL(transport_free_session);
+
+void transport_deregister_session(struct se_session *se_sess)
+{
+ struct se_portal_group *se_tpg = se_sess->se_tpg;
+ struct se_node_acl *se_nacl;
+
+ if (!(se_tpg)) {
+ transport_free_session(se_sess);
+ return;
+ }
+ /*
+ * Wait for possible reference in drivers/target/target_core_mib.c:
+ * scsi_att_intr_port_seq_show()
+ */
+ while (atomic_read(&se_sess->mib_ref_count) != 0)
+ cpu_relax();
+
+ spin_lock_bh(&se_tpg->session_lock);
+ list_del(&se_sess->sess_list);
+ se_sess->se_tpg = NULL;
+ se_sess->fabric_sess_ptr = NULL;
+ spin_unlock_bh(&se_tpg->session_lock);
+
+ /*
+ * Determine if we need to do extra work for this initiator node's
+ * struct se_node_acl if it had been previously dynamically generated.
+ */
+ se_nacl = se_sess->se_node_acl;
+ if ((se_nacl)) {
+ spin_lock_bh(&se_tpg->acl_node_lock);
+ if (se_nacl->dynamic_node_acl) {
+ if (!(TPG_TFO(se_tpg)->tpg_check_demo_mode_cache(
+ se_tpg))) {
+ list_del(&se_nacl->acl_list);
+ se_tpg->num_node_acls--;
+ spin_unlock_bh(&se_tpg->acl_node_lock);
+
+ core_tpg_wait_for_nacl_pr_ref(se_nacl);
+ core_tpg_wait_for_mib_ref(se_nacl);
+ core_free_device_list_for_node(se_nacl, se_tpg);
+ TPG_TFO(se_tpg)->tpg_release_fabric_acl(se_tpg,
+ se_nacl);
+ spin_lock_bh(&se_tpg->acl_node_lock);
+ }
+ }
+ spin_unlock_bh(&se_tpg->acl_node_lock);
+ }
+
+ transport_free_session(se_sess);
+
+ printk(KERN_INFO "TARGET_CORE[%s]: Deregistered fabric_sess\n",
+ TPG_TFO(se_tpg)->get_fabric_name());
+}
+EXPORT_SYMBOL(transport_deregister_session);
+
+/*
+ * Called with T_TASK(cmd)->t_state_lock held.
+ */
+static void transport_all_task_dev_remove_state(struct se_cmd *cmd)
+{
+ struct se_device *dev;
+ struct se_task *task;
+ unsigned long flags;
+
+ if (!T_TASK(cmd))
+ return;
+
+ list_for_each_entry(task, &T_TASK(cmd)->t_task_list, t_list) {
+ dev = task->se_dev;
+ if (!(dev))
+ continue;
+
+ if (atomic_read(&task->task_active))
+ continue;
+
+ if (!(atomic_read(&task->task_state_active)))
+ continue;
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ list_del(&task->t_state_list);
+ DEBUG_TSTATE("Removed ITT: 0x%08x dev: %p task[%p]\n",
+ CMD_TFO(cmd)->tfo_get_task_tag(cmd), dev, task);
+ spin_unlock_irqrestore(&dev->execute_task_lock, flags);
+
+ atomic_set(&task->task_state_active, 0);
+ atomic_dec(&T_TASK(cmd)->t_task_cdbs_ex_left);
+ }
+}
+
+/* transport_cmd_check_stop():
+ *
+ * 'transport_off = 1' determines if t_transport_active should be cleared.
+ * 'transport_off = 2' determines if task_dev_state should be removed.
+ *
+ * A non-zero u8 t_state sets cmd->t_state.
+ * Returns 1 when command is stopped, else 0.
+ */
+static int transport_cmd_check_stop(
+ struct se_cmd *cmd,
+ int transport_off,
+ u8 t_state)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ /*
+ * Determine if IOCTL context caller in requesting the stopping of this
+ * command for LUN shutdown purposes.
+ */
+ if (atomic_read(&T_TASK(cmd)->transport_lun_stop)) {
+ DEBUG_CS("%s:%d atomic_read(&T_TASK(cmd)->transport_lun_stop)"
+ " == TRUE for ITT: 0x%08x\n", __func__, __LINE__,
+ CMD_TFO(cmd)->get_task_tag(cmd));
+
+ cmd->deferred_t_state = cmd->t_state;
+ cmd->t_state = TRANSPORT_DEFERRED_CMD;
+ atomic_set(&T_TASK(cmd)->t_transport_active, 0);
+ if (transport_off == 2)
+ transport_all_task_dev_remove_state(cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ complete(&T_TASK(cmd)->transport_lun_stop_comp);
+ return 1;
+ }
+ /*
+ * Determine if frontend context caller is requesting the stopping of
+ * this command for frontend excpections.
+ */
+ if (atomic_read(&T_TASK(cmd)->t_transport_stop)) {
+ DEBUG_CS("%s:%d atomic_read(&T_TASK(cmd)->t_transport_stop) =="
+ " TRUE for ITT: 0x%08x\n", __func__, __LINE__,
+ CMD_TFO(cmd)->get_task_tag(cmd));
+
+ cmd->deferred_t_state = cmd->t_state;
+ cmd->t_state = TRANSPORT_DEFERRED_CMD;
+ if (transport_off == 2)
+ transport_all_task_dev_remove_state(cmd);
+
+ /*
+ * Clear struct se_cmd->se_lun before the transport_off == 2 handoff
+ * to FE.
+ */
+ if (transport_off == 2)
+ cmd->se_lun = NULL;
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ complete(&T_TASK(cmd)->t_transport_stop_comp);
+ return 1;
+ }
+ if (transport_off) {
+ atomic_set(&T_TASK(cmd)->t_transport_active, 0);
+ if (transport_off == 2) {
+ transport_all_task_dev_remove_state(cmd);
+ /*
+ * Clear struct se_cmd->se_lun before the transport_off == 2
+ * handoff to fabric module.
+ */
+ cmd->se_lun = NULL;
+ /*
+ * Some fabric modules like tcm_loop can release
+ * their internally allocated I/O refrence now and
+ * struct se_cmd now.
+ */
+ if (CMD_TFO(cmd)->check_stop_free != NULL) {
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+
+ CMD_TFO(cmd)->check_stop_free(cmd);
+ return 1;
+ }
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ return 0;
+ } else if (t_state)
+ cmd->t_state = t_state;
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ return 0;
+}
+
+static int transport_cmd_check_stop_to_fabric(struct se_cmd *cmd)
+{
+ return transport_cmd_check_stop(cmd, 2, 0);
+}
+
+static void transport_lun_remove_cmd(struct se_cmd *cmd)
+{
+ struct se_lun *lun = SE_LUN(cmd);
+ unsigned long flags;
+
+ if (!lun)
+ return;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (!(atomic_read(&T_TASK(cmd)->transport_dev_active))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ goto check_lun;
+ }
+ atomic_set(&T_TASK(cmd)->transport_dev_active, 0);
+ transport_all_task_dev_remove_state(cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_free_dev_tasks(cmd);
+
+check_lun:
+ spin_lock_irqsave(&lun->lun_cmd_lock, flags);
+ if (atomic_read(&T_TASK(cmd)->transport_lun_active)) {
+ list_del(&cmd->se_lun_list);
+ atomic_set(&T_TASK(cmd)->transport_lun_active, 0);
+#if 0
+ printk(KERN_INFO "Removed ITT: 0x%08x from LUN LIST[%d]\n"
+ CMD_TFO(cmd)->get_task_tag(cmd), lun->unpacked_lun);
+#endif
+ }
+ spin_unlock_irqrestore(&lun->lun_cmd_lock, flags);
+}
+
+void transport_cmd_finish_abort(struct se_cmd *cmd, int remove)
+{
+ transport_remove_cmd_from_queue(cmd, SE_DEV(cmd)->dev_queue_obj);
+ transport_lun_remove_cmd(cmd);
+
+ if (transport_cmd_check_stop_to_fabric(cmd))
+ return;
+ if (remove)
+ transport_generic_remove(cmd, 0, 0);
+}
+
+void transport_cmd_finish_abort_tmr(struct se_cmd *cmd)
+{
+ transport_remove_cmd_from_queue(cmd, SE_DEV(cmd)->dev_queue_obj);
+
+ if (transport_cmd_check_stop_to_fabric(cmd))
+ return;
+
+ transport_generic_remove(cmd, 0, 0);
+}
+
+static int transport_add_cmd_to_queue(
+ struct se_cmd *cmd,
+ int t_state)
+{
+ struct se_device *dev = cmd->se_dev;
+ struct se_queue_obj *qobj = dev->dev_queue_obj;
+ struct se_queue_req *qr;
+ unsigned long flags;
+
+ qr = kzalloc(sizeof(struct se_queue_req), GFP_ATOMIC);
+ if (!(qr)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " struct se_queue_req\n");
+ return -1;
+ }
+ INIT_LIST_HEAD(&qr->qr_list);
+
+ qr->cmd = (void *)cmd;
+ qr->state = t_state;
+
+ if (t_state) {
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ cmd->t_state = t_state;
+ atomic_set(&T_TASK(cmd)->t_transport_active, 1);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ }
+
+ spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
+ list_add_tail(&qr->qr_list, &qobj->qobj_list);
+ atomic_inc(&T_TASK(cmd)->t_transport_queue_active);
+ spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
+
+ atomic_inc(&qobj->queue_cnt);
+ wake_up_interruptible(&qobj->thread_wq);
+ return 0;
+}
+
+/*
+ * Called with struct se_queue_obj->cmd_queue_lock held.
+ */
+static struct se_queue_req *
+__transport_get_qr_from_queue(struct se_queue_obj *qobj)
+{
+ struct se_cmd *cmd;
+ struct se_queue_req *qr = NULL;
+
+ if (list_empty(&qobj->qobj_list))
+ return NULL;
+
+ list_for_each_entry(qr, &qobj->qobj_list, qr_list)
+ break;
+
+ if (qr->cmd) {
+ cmd = (struct se_cmd *)qr->cmd;
+ atomic_dec(&T_TASK(cmd)->t_transport_queue_active);
+ }
+ list_del(&qr->qr_list);
+ atomic_dec(&qobj->queue_cnt);
+
+ return qr;
+}
+
+static struct se_queue_req *
+transport_get_qr_from_queue(struct se_queue_obj *qobj)
+{
+ struct se_cmd *cmd;
+ struct se_queue_req *qr;
+ unsigned long flags;
+
+ spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
+ if (list_empty(&qobj->qobj_list)) {
+ spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
+ return NULL;
+ }
+
+ list_for_each_entry(qr, &qobj->qobj_list, qr_list)
+ break;
+
+ if (qr->cmd) {
+ cmd = (struct se_cmd *)qr->cmd;
+ atomic_dec(&T_TASK(cmd)->t_transport_queue_active);
+ }
+ list_del(&qr->qr_list);
+ atomic_dec(&qobj->queue_cnt);
+ spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
+
+ return qr;
+}
+
+static void transport_remove_cmd_from_queue(struct se_cmd *cmd,
+ struct se_queue_obj *qobj)
+{
+ struct se_cmd *q_cmd;
+ struct se_queue_req *qr = NULL, *qr_p = NULL;
+ unsigned long flags;
+
+ spin_lock_irqsave(&qobj->cmd_queue_lock, flags);
+ if (!(atomic_read(&T_TASK(cmd)->t_transport_queue_active))) {
+ spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
+ return;
+ }
+
+ list_for_each_entry_safe(qr, qr_p, &qobj->qobj_list, qr_list) {
+ q_cmd = (struct se_cmd *)qr->cmd;
+ if (q_cmd != cmd)
+ continue;
+
+ atomic_dec(&T_TASK(q_cmd)->t_transport_queue_active);
+ atomic_dec(&qobj->queue_cnt);
+ list_del(&qr->qr_list);
+ kfree(qr);
+ }
+ spin_unlock_irqrestore(&qobj->cmd_queue_lock, flags);
+
+ if (atomic_read(&T_TASK(cmd)->t_transport_queue_active)) {
+ printk(KERN_ERR "ITT: 0x%08x t_transport_queue_active: %d\n",
+ CMD_TFO(cmd)->get_task_tag(cmd),
+ atomic_read(&T_TASK(cmd)->t_transport_queue_active));
+ }
+}
+
+/*
+ * Completion function used by TCM subsystem plugins (such as FILEIO)
+ * for queueing up response from struct se_subsystem_api->do_task()
+ */
+void transport_complete_sync_cache(struct se_cmd *cmd, int good)
+{
+ struct se_task *task = list_entry(T_TASK(cmd)->t_task_list.next,
+ struct se_task, t_list);
+
+ if (good) {
+ cmd->scsi_status = SAM_STAT_GOOD;
+ task->task_scsi_status = GOOD;
+ } else {
+ task->task_scsi_status = SAM_STAT_CHECK_CONDITION;
+ task->task_error_status = PYX_TRANSPORT_ILLEGAL_REQUEST;
+ TASK_CMD(task)->transport_error_status =
+ PYX_TRANSPORT_ILLEGAL_REQUEST;
+ }
+
+ transport_complete_task(task, good);
+}
+EXPORT_SYMBOL(transport_complete_sync_cache);
+
+/* transport_complete_task():
+ *
+ * Called from interrupt and non interrupt context depending
+ * on the transport plugin.
+ */
+void transport_complete_task(struct se_task *task, int success)
+{
+ struct se_cmd *cmd = TASK_CMD(task);
+ struct se_device *dev = task->se_dev;
+ int t_state;
+ unsigned long flags;
+#if 0
+ printk(KERN_INFO "task: %p CDB: 0x%02x obj_ptr: %p\n", task,
+ T_TASK(cmd)->t_task_cdb[0], dev);
+#endif
+ if (dev) {
+ spin_lock_irqsave(&SE_HBA(dev)->hba_queue_lock, flags);
+ atomic_inc(&dev->depth_left);
+ atomic_inc(&SE_HBA(dev)->left_queue_depth);
+ spin_unlock_irqrestore(&SE_HBA(dev)->hba_queue_lock, flags);
+ }
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ atomic_set(&task->task_active, 0);
+
+ /*
+ * See if any sense data exists, if so set the TASK_SENSE flag.
+ * Also check for any other post completion work that needs to be
+ * done by the plugins.
+ */
+ if (dev && dev->transport->transport_complete) {
+ if (dev->transport->transport_complete(task) != 0) {
+ cmd->se_cmd_flags |= SCF_TRANSPORT_TASK_SENSE;
+ task->task_sense = 1;
+ success = 1;
+ }
+ }
+
+ /*
+ * See if we are waiting for outstanding struct se_task
+ * to complete for an exception condition
+ */
+ if (atomic_read(&task->task_stop)) {
+ /*
+ * Decrement T_TASK(cmd)->t_se_count if this task had
+ * previously thrown its timeout exception handler.
+ */
+ if (atomic_read(&task->task_timeout)) {
+ atomic_dec(&T_TASK(cmd)->t_se_count);
+ atomic_set(&task->task_timeout, 0);
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ complete(&task->task_stop_comp);
+ return;
+ }
+ /*
+ * If the task's timeout handler has fired, use the t_task_cdbs_timeout
+ * left counter to determine when the struct se_cmd is ready to be queued to
+ * the processing thread.
+ */
+ if (atomic_read(&task->task_timeout)) {
+ if (!(atomic_dec_and_test(
+ &T_TASK(cmd)->t_task_cdbs_timeout_left))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock,
+ flags);
+ return;
+ }
+ t_state = TRANSPORT_COMPLETE_TIMEOUT;
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_add_cmd_to_queue(cmd, t_state);
+ return;
+ }
+ atomic_dec(&T_TASK(cmd)->t_task_cdbs_timeout_left);
+
+ /*
+ * Decrement the outstanding t_task_cdbs_left count. The last
+ * struct se_task from struct se_cmd will complete itself into the
+ * device queue depending upon int success.
+ */
+ if (!(atomic_dec_and_test(&T_TASK(cmd)->t_task_cdbs_left))) {
+ if (!success)
+ T_TASK(cmd)->t_tasks_failed = 1;
+
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ return;
+ }
+
+ if (!success || T_TASK(cmd)->t_tasks_failed) {
+ t_state = TRANSPORT_COMPLETE_FAILURE;
+ if (!task->task_error_status) {
+ task->task_error_status =
+ PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ cmd->transport_error_status =
+ PYX_TRANSPORT_UNKNOWN_SAM_OPCODE;
+ }
+ } else {
+ atomic_set(&T_TASK(cmd)->t_transport_complete, 1);
+ t_state = TRANSPORT_COMPLETE_OK;
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_add_cmd_to_queue(cmd, t_state);
+}
+EXPORT_SYMBOL(transport_complete_task);
+
+/*
+ * Called by transport_add_tasks_from_cmd() once a struct se_cmd's
+ * struct se_task list are ready to be added to the active execution list
+ * struct se_device
+
+ * Called with se_dev_t->execute_task_lock called.
+ */
+static inline int transport_add_task_check_sam_attr(
+ struct se_task *task,
+ struct se_task *task_prev,
+ struct se_device *dev)
+{
+ /*
+ * No SAM Task attribute emulation enabled, add to tail of
+ * execution queue
+ */
+ if (dev->dev_task_attr_type != SAM_TASK_ATTR_EMULATED) {
+ list_add_tail(&task->t_execute_list, &dev->execute_task_list);
+ return 0;
+ }
+ /*
+ * HEAD_OF_QUEUE attribute for received CDB, which means
+ * the first task that is associated with a struct se_cmd goes to
+ * head of the struct se_device->execute_task_list, and task_prev
+ * after that for each subsequent task
+ */
+ if (task->task_se_cmd->sam_task_attr == TASK_ATTR_HOQ) {
+ list_add(&task->t_execute_list,
+ (task_prev != NULL) ?
+ &task_prev->t_execute_list :
+ &dev->execute_task_list);
+
+ DEBUG_STA("Set HEAD_OF_QUEUE for task CDB: 0x%02x"
+ " in execution queue\n",
+ T_TASK(task->task_se_cmd)->t_task_cdb[0]);
+ return 1;
+ }
+ /*
+ * For ORDERED, SIMPLE or UNTAGGED attribute tasks once they have been
+ * transitioned from Dermant -> Active state, and are added to the end
+ * of the struct se_device->execute_task_list
+ */
+ list_add_tail(&task->t_execute_list, &dev->execute_task_list);
+ return 0;
+}
+
+/* __transport_add_task_to_execute_queue():
+ *
+ * Called with se_dev_t->execute_task_lock called.
+ */
+static void __transport_add_task_to_execute_queue(
+ struct se_task *task,
+ struct se_task *task_prev,
+ struct se_device *dev)
+{
+ int head_of_queue;
+
+ head_of_queue = transport_add_task_check_sam_attr(task, task_prev, dev);
+ atomic_inc(&dev->execute_tasks);
+
+ if (atomic_read(&task->task_state_active))
+ return;
+ /*
+ * Determine if this task needs to go to HEAD_OF_QUEUE for the
+ * state list as well. Running with SAM Task Attribute emulation
+ * will always return head_of_queue == 0 here
+ */
+ if (head_of_queue)
+ list_add(&task->t_state_list, (task_prev) ?
+ &task_prev->t_state_list :
+ &dev->state_task_list);
+ else
+ list_add_tail(&task->t_state_list, &dev->state_task_list);
+
+ atomic_set(&task->task_state_active, 1);
+
+ DEBUG_TSTATE("Added ITT: 0x%08x task[%p] to dev: %p\n",
+ CMD_TFO(task->task_se_cmd)->get_task_tag(task->task_se_cmd),
+ task, dev);
+}
+
+static void transport_add_tasks_to_state_queue(struct se_cmd *cmd)
+{
+ struct se_device *dev;
+ struct se_task *task;
+ unsigned long flags;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ list_for_each_entry(task, &T_TASK(cmd)->t_task_list, t_list) {
+ dev = task->se_dev;
+
+ if (atomic_read(&task->task_state_active))
+ continue;
+
+ spin_lock(&dev->execute_task_lock);
+ list_add_tail(&task->t_state_list, &dev->state_task_list);
+ atomic_set(&task->task_state_active, 1);
+
+ DEBUG_TSTATE("Added ITT: 0x%08x task[%p] to dev: %p\n",
+ CMD_TFO(task->task_se_cmd)->get_task_tag(
+ task->task_se_cmd), task, dev);
+
+ spin_unlock(&dev->execute_task_lock);
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+}
+
+static void transport_add_tasks_from_cmd(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_task *task, *task_prev = NULL;
+ unsigned long flags;
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ list_for_each_entry(task, &T_TASK(cmd)->t_task_list, t_list) {
+ if (atomic_read(&task->task_execute_queue))
+ continue;
+ /*
+ * __transport_add_task_to_execute_queue() handles the
+ * SAM Task Attribute emulation if enabled
+ */
+ __transport_add_task_to_execute_queue(task, task_prev, dev);
+ atomic_set(&task->task_execute_queue, 1);
+ task_prev = task;
+ }
+ spin_unlock_irqrestore(&dev->execute_task_lock, flags);
+
+ return;
+}
+
+/* transport_get_task_from_execute_queue():
+ *
+ * Called with dev->execute_task_lock held.
+ */
+static struct se_task *
+transport_get_task_from_execute_queue(struct se_device *dev)
+{
+ struct se_task *task;
+
+ if (list_empty(&dev->execute_task_list))
+ return NULL;
+
+ list_for_each_entry(task, &dev->execute_task_list, t_execute_list)
+ break;
+
+ list_del(&task->t_execute_list);
+ atomic_dec(&dev->execute_tasks);
+
+ return task;
+}
+
+/* transport_remove_task_from_execute_queue():
+ *
+ *
+ */
+static void transport_remove_task_from_execute_queue(
+ struct se_task *task,
+ struct se_device *dev)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ list_del(&task->t_execute_list);
+ atomic_dec(&dev->execute_tasks);
+ spin_unlock_irqrestore(&dev->execute_task_lock, flags);
+}
+
+unsigned char *transport_dump_cmd_direction(struct se_cmd *cmd)
+{
+ switch (cmd->data_direction) {
+ case DMA_NONE:
+ return "NONE";
+ case DMA_FROM_DEVICE:
+ return "READ";
+ case DMA_TO_DEVICE:
+ return "WRITE";
+ case DMA_BIDIRECTIONAL:
+ return "BIDI";
+ default:
+ break;
+ }
+
+ return "UNKNOWN";
+}
+
+void transport_dump_dev_state(
+ struct se_device *dev,
+ char *b,
+ int *bl)
+{
+ *bl += sprintf(b + *bl, "Status: ");
+ switch (dev->dev_status) {
+ case TRANSPORT_DEVICE_ACTIVATED:
+ *bl += sprintf(b + *bl, "ACTIVATED");
+ break;
+ case TRANSPORT_DEVICE_DEACTIVATED:
+ *bl += sprintf(b + *bl, "DEACTIVATED");
+ break;
+ case TRANSPORT_DEVICE_SHUTDOWN:
+ *bl += sprintf(b + *bl, "SHUTDOWN");
+ break;
+ case TRANSPORT_DEVICE_OFFLINE_ACTIVATED:
+ case TRANSPORT_DEVICE_OFFLINE_DEACTIVATED:
+ *bl += sprintf(b + *bl, "OFFLINE");
+ break;
+ default:
+ *bl += sprintf(b + *bl, "UNKNOWN=%d", dev->dev_status);
+ break;
+ }
+
+ *bl += sprintf(b + *bl, " Execute/Left/Max Queue Depth: %d/%d/%d",
+ atomic_read(&dev->execute_tasks), atomic_read(&dev->depth_left),
+ dev->queue_depth);
+ *bl += sprintf(b + *bl, " SectorSize: %u MaxSectors: %u\n",
+ DEV_ATTRIB(dev)->block_size, DEV_ATTRIB(dev)->max_sectors);
+ *bl += sprintf(b + *bl, " ");
+}
+
+/* transport_release_all_cmds():
+ *
+ *
+ */
+static void transport_release_all_cmds(struct se_device *dev)
+{
+ struct se_cmd *cmd = NULL;
+ struct se_queue_req *qr = NULL, *qr_p = NULL;
+ int bug_out = 0, t_state;
+ unsigned long flags;
+
+ spin_lock_irqsave(&dev->dev_queue_obj->cmd_queue_lock, flags);
+ list_for_each_entry_safe(qr, qr_p, &dev->dev_queue_obj->qobj_list,
+ qr_list) {
+
+ cmd = (struct se_cmd *)qr->cmd;
+ t_state = qr->state;
+ list_del(&qr->qr_list);
+ kfree(qr);
+ spin_unlock_irqrestore(&dev->dev_queue_obj->cmd_queue_lock,
+ flags);
+
+ printk(KERN_ERR "Releasing ITT: 0x%08x, i_state: %u,"
+ " t_state: %u directly\n",
+ CMD_TFO(cmd)->get_task_tag(cmd),
+ CMD_TFO(cmd)->get_cmd_state(cmd), t_state);
+
+ transport_release_fe_cmd(cmd);
+ bug_out = 1;
+
+ spin_lock_irqsave(&dev->dev_queue_obj->cmd_queue_lock, flags);
+ }
+ spin_unlock_irqrestore(&dev->dev_queue_obj->cmd_queue_lock, flags);
+#if 0
+ if (bug_out)
+ BUG();
+#endif
+}
+
+void transport_dump_vpd_proto_id(
+ struct t10_vpd *vpd,
+ unsigned char *p_buf,
+ int p_buf_len)
+{
+ unsigned char buf[VPD_TMP_BUF_SIZE];
+ int len;
+
+ memset(buf, 0, VPD_TMP_BUF_SIZE);
+ len = sprintf(buf, "T10 VPD Protocol Identifier: ");
+
+ switch (vpd->protocol_identifier) {
+ case 0x00:
+ sprintf(buf+len, "Fibre Channel\n");
+ break;
+ case 0x10:
+ sprintf(buf+len, "Parallel SCSI\n");
+ break;
+ case 0x20:
+ sprintf(buf+len, "SSA\n");
+ break;
+ case 0x30:
+ sprintf(buf+len, "IEEE 1394\n");
+ break;
+ case 0x40:
+ sprintf(buf+len, "SCSI Remote Direct Memory Access"
+ " Protocol\n");
+ break;
+ case 0x50:
+ sprintf(buf+len, "Internet SCSI (iSCSI)\n");
+ break;
+ case 0x60:
+ sprintf(buf+len, "SAS Serial SCSI Protocol\n");
+ break;
+ case 0x70:
+ sprintf(buf+len, "Automation/Drive Interface Transport"
+ " Protocol\n");
+ break;
+ case 0x80:
+ sprintf(buf+len, "AT Attachment Interface ATA/ATAPI\n");
+ break;
+ default:
+ sprintf(buf+len, "Unknown 0x%02x\n",
+ vpd->protocol_identifier);
+ break;
+ }
+
+ if (p_buf)
+ strncpy(p_buf, buf, p_buf_len);
+ else
+ printk(KERN_INFO "%s", buf);
+}
+
+void
+transport_set_vpd_proto_id(struct t10_vpd *vpd, unsigned char *page_83)
+{
+ /*
+ * Check if the Protocol Identifier Valid (PIV) bit is set..
+ *
+ * from spc3r23.pdf section 7.5.1
+ */
+ if (page_83[1] & 0x80) {
+ vpd->protocol_identifier = (page_83[0] & 0xf0);
+ vpd->protocol_identifier_set = 1;
+ transport_dump_vpd_proto_id(vpd, NULL, 0);
+ }
+}
+EXPORT_SYMBOL(transport_set_vpd_proto_id);
+
+int transport_dump_vpd_assoc(
+ struct t10_vpd *vpd,
+ unsigned char *p_buf,
+ int p_buf_len)
+{
+ unsigned char buf[VPD_TMP_BUF_SIZE];
+ int ret = 0, len;
+
+ memset(buf, 0, VPD_TMP_BUF_SIZE);
+ len = sprintf(buf, "T10 VPD Identifier Association: ");
+
+ switch (vpd->association) {
+ case 0x00:
+ sprintf(buf+len, "addressed logical unit\n");
+ break;
+ case 0x10:
+ sprintf(buf+len, "target port\n");
+ break;
+ case 0x20:
+ sprintf(buf+len, "SCSI target device\n");
+ break;
+ default:
+ sprintf(buf+len, "Unknown 0x%02x\n", vpd->association);
+ ret = -1;
+ break;
+ }
+
+ if (p_buf)
+ strncpy(p_buf, buf, p_buf_len);
+ else
+ printk("%s", buf);
+
+ return ret;
+}
+
+int transport_set_vpd_assoc(struct t10_vpd *vpd, unsigned char *page_83)
+{
+ /*
+ * The VPD identification association..
+ *
+ * from spc3r23.pdf Section 7.6.3.1 Table 297
+ */
+ vpd->association = (page_83[1] & 0x30);
+ return transport_dump_vpd_assoc(vpd, NULL, 0);
+}
+EXPORT_SYMBOL(transport_set_vpd_assoc);
+
+int transport_dump_vpd_ident_type(
+ struct t10_vpd *vpd,
+ unsigned char *p_buf,
+ int p_buf_len)
+{
+ unsigned char buf[VPD_TMP_BUF_SIZE];
+ int ret = 0, len;
+
+ memset(buf, 0, VPD_TMP_BUF_SIZE);
+ len = sprintf(buf, "T10 VPD Identifier Type: ");
+
+ switch (vpd->device_identifier_type) {
+ case 0x00:
+ sprintf(buf+len, "Vendor specific\n");
+ break;
+ case 0x01:
+ sprintf(buf+len, "T10 Vendor ID based\n");
+ break;
+ case 0x02:
+ sprintf(buf+len, "EUI-64 based\n");
+ break;
+ case 0x03:
+ sprintf(buf+len, "NAA\n");
+ break;
+ case 0x04:
+ sprintf(buf+len, "Relative target port identifier\n");
+ break;
+ case 0x08:
+ sprintf(buf+len, "SCSI name string\n");
+ break;
+ default:
+ sprintf(buf+len, "Unsupported: 0x%02x\n",
+ vpd->device_identifier_type);
+ ret = -1;
+ break;
+ }
+
+ if (p_buf)
+ strncpy(p_buf, buf, p_buf_len);
+ else
+ printk("%s", buf);
+
+ return ret;
+}
+
+int transport_set_vpd_ident_type(struct t10_vpd *vpd, unsigned char *page_83)
+{
+ /*
+ * The VPD identifier type..
+ *
+ * from spc3r23.pdf Section 7.6.3.1 Table 298
+ */
+ vpd->device_identifier_type = (page_83[1] & 0x0f);
+ return transport_dump_vpd_ident_type(vpd, NULL, 0);
+}
+EXPORT_SYMBOL(transport_set_vpd_ident_type);
+
+int transport_dump_vpd_ident(
+ struct t10_vpd *vpd,
+ unsigned char *p_buf,
+ int p_buf_len)
+{
+ unsigned char buf[VPD_TMP_BUF_SIZE];
+ int ret = 0;
+
+ memset(buf, 0, VPD_TMP_BUF_SIZE);
+
+ switch (vpd->device_identifier_code_set) {
+ case 0x01: /* Binary */
+ sprintf(buf, "T10 VPD Binary Device Identifier: %s\n",
+ &vpd->device_identifier[0]);
+ break;
+ case 0x02: /* ASCII */
+ sprintf(buf, "T10 VPD ASCII Device Identifier: %s\n",
+ &vpd->device_identifier[0]);
+ break;
+ case 0x03: /* UTF-8 */
+ sprintf(buf, "T10 VPD UTF-8 Device Identifier: %s\n",
+ &vpd->device_identifier[0]);
+ break;
+ default:
+ sprintf(buf, "T10 VPD Device Identifier encoding unsupported:"
+ " 0x%02x", vpd->device_identifier_code_set);
+ ret = -1;
+ break;
+ }
+
+ if (p_buf)
+ strncpy(p_buf, buf, p_buf_len);
+ else
+ printk("%s", buf);
+
+ return ret;
+}
+
+int
+transport_set_vpd_ident(struct t10_vpd *vpd, unsigned char *page_83)
+{
+ static const char hex_str[] = "0123456789abcdef";
+ int j = 0, i = 4; /* offset to start of the identifer */
+
+ /*
+ * The VPD Code Set (encoding)
+ *
+ * from spc3r23.pdf Section 7.6.3.1 Table 296
+ */
+ vpd->device_identifier_code_set = (page_83[0] & 0x0f);
+ switch (vpd->device_identifier_code_set) {
+ case 0x01: /* Binary */
+ vpd->device_identifier[j++] =
+ hex_str[vpd->device_identifier_type];
+ while (i < (4 + page_83[3])) {
+ vpd->device_identifier[j++] =
+ hex_str[(page_83[i] & 0xf0) >> 4];
+ vpd->device_identifier[j++] =
+ hex_str[page_83[i] & 0x0f];
+ i++;
+ }
+ break;
+ case 0x02: /* ASCII */
+ case 0x03: /* UTF-8 */
+ while (i < (4 + page_83[3]))
+ vpd->device_identifier[j++] = page_83[i++];
+ break;
+ default:
+ break;
+ }
+
+ return transport_dump_vpd_ident(vpd, NULL, 0);
+}
+EXPORT_SYMBOL(transport_set_vpd_ident);
+
+static void core_setup_task_attr_emulation(struct se_device *dev)
+{
+ /*
+ * If this device is from Target_Core_Mod/pSCSI, disable the
+ * SAM Task Attribute emulation.
+ *
+ * This is currently not available in upsream Linux/SCSI Target
+ * mode code, and is assumed to be disabled while using TCM/pSCSI.
+ */
+ if (TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV) {
+ dev->dev_task_attr_type = SAM_TASK_ATTR_PASSTHROUGH;
+ return;
+ }
+
+ dev->dev_task_attr_type = SAM_TASK_ATTR_EMULATED;
+ DEBUG_STA("%s: Using SAM_TASK_ATTR_EMULATED for SPC: 0x%02x"
+ " device\n", TRANSPORT(dev)->name,
+ TRANSPORT(dev)->get_device_rev(dev));
+}
+
+static void scsi_dump_inquiry(struct se_device *dev)
+{
+ struct t10_wwn *wwn = DEV_T10_WWN(dev);
+ int i, device_type;
+ /*
+ * Print Linux/SCSI style INQUIRY formatting to the kernel ring buffer
+ */
+ printk(" Vendor: ");
+ for (i = 0; i < 8; i++)
+ if (wwn->vendor[i] >= 0x20)
+ printk("%c", wwn->vendor[i]);
+ else
+ printk(" ");
+
+ printk(" Model: ");
+ for (i = 0; i < 16; i++)
+ if (wwn->model[i] >= 0x20)
+ printk("%c", wwn->model[i]);
+ else
+ printk(" ");
+
+ printk(" Revision: ");
+ for (i = 0; i < 4; i++)
+ if (wwn->revision[i] >= 0x20)
+ printk("%c", wwn->revision[i]);
+ else
+ printk(" ");
+
+ printk("\n");
+
+ device_type = TRANSPORT(dev)->get_device_type(dev);
+ printk(" Type: %s ", scsi_device_type(device_type));
+ printk(" ANSI SCSI revision: %02x\n",
+ TRANSPORT(dev)->get_device_rev(dev));
+}
+
+struct se_device *transport_add_device_to_core_hba(
+ struct se_hba *hba,
+ struct se_subsystem_api *transport,
+ struct se_subsystem_dev *se_dev,
+ u32 device_flags,
+ void *transport_dev,
+ struct se_dev_limits *dev_limits,
+ const char *inquiry_prod,
+ const char *inquiry_rev)
+{
+ int ret = 0, force_pt;
+ struct se_device *dev;
+
+ dev = kzalloc(sizeof(struct se_device), GFP_KERNEL);
+ if (!(dev)) {
+ printk(KERN_ERR "Unable to allocate memory for se_dev_t\n");
+ return NULL;
+ }
+ dev->dev_queue_obj = kzalloc(sizeof(struct se_queue_obj), GFP_KERNEL);
+ if (!(dev->dev_queue_obj)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " dev->dev_queue_obj\n");
+ kfree(dev);
+ return NULL;
+ }
+ transport_init_queue_obj(dev->dev_queue_obj);
+
+ dev->dev_status_queue_obj = kzalloc(sizeof(struct se_queue_obj),
+ GFP_KERNEL);
+ if (!(dev->dev_status_queue_obj)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " dev->dev_status_queue_obj\n");
+ kfree(dev->dev_queue_obj);
+ kfree(dev);
+ return NULL;
+ }
+ transport_init_queue_obj(dev->dev_status_queue_obj);
+
+ dev->dev_flags = device_flags;
+ dev->dev_status |= TRANSPORT_DEVICE_DEACTIVATED;
+ dev->dev_ptr = (void *) transport_dev;
+ 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);
+ 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);
+
+ dev->queue_depth = dev_limits->queue_depth;
+ atomic_set(&dev->depth_left, dev->queue_depth);
+ atomic_set(&dev->dev_ordered_id, 0);
+
+ se_dev_set_default_attribs(dev, dev_limits);
+
+ dev->dev_index = scsi_get_new_index(SCSI_DEVICE_INDEX);
+ dev->creation_time = get_jiffies_64();
+ spin_lock_init(&dev->stats_lock);
+
+ spin_lock(&hba->device_lock);
+ list_add_tail(&dev->dev_list, &hba->hba_dev_list);
+ hba->dev_count++;
+ spin_unlock(&hba->device_lock);
+ /*
+ * Setup the SAM Task Attribute emulation for struct se_device
+ */
+ core_setup_task_attr_emulation(dev);
+ /*
+ * Force PR and ALUA passthrough emulation with internal object use.
+ */
+ force_pt = (hba->hba_flags & HBA_FLAGS_INTERNAL_USE);
+ /*
+ * Setup the Reservations infrastructure for struct se_device
+ */
+ core_setup_reservations(dev, force_pt);
+ /*
+ * Setup the Asymmetric Logical Unit Assignment for struct se_device
+ */
+ if (core_setup_alua(dev, force_pt) < 0)
+ goto out;
+
+ /*
+ * Startup the struct se_device processing thread
+ */
+ dev->process_thread = kthread_run(transport_processing_thread, dev,
+ "LIO_%s", TRANSPORT(dev)->name);
+ if (IS_ERR(dev->process_thread)) {
+ printk(KERN_ERR "Unable to create kthread: LIO_%s\n",
+ TRANSPORT(dev)->name);
+ goto out;
+ }
+
+ /*
+ * Preload the initial INQUIRY const values if we are doing
+ * anything virtual (IBLOCK, FILEIO, RAMDISK), but not for TCM/pSCSI
+ * passthrough because this is being provided by the backend LLD.
+ * This is required so that transport_get_inquiry() copies these
+ * originals once back into DEV_T10_WWN(dev) for the virtual device
+ * setup.
+ */
+ if (TRANSPORT(dev)->transport_type != TRANSPORT_PLUGIN_PHBA_PDEV) {
+ if (!(inquiry_prod) || !(inquiry_prod)) {
+ printk(KERN_ERR "All non TCM/pSCSI plugins require"
+ " INQUIRY consts\n");
+ goto out;
+ }
+
+ strncpy(&DEV_T10_WWN(dev)->vendor[0], "LIO-ORG", 8);
+ strncpy(&DEV_T10_WWN(dev)->model[0], inquiry_prod, 16);
+ strncpy(&DEV_T10_WWN(dev)->revision[0], inquiry_rev, 4);
+ }
+ scsi_dump_inquiry(dev);
+
+out:
+ if (!ret)
+ return dev;
+ kthread_stop(dev->process_thread);
+
+ spin_lock(&hba->device_lock);
+ list_del(&dev->dev_list);
+ hba->dev_count--;
+ spin_unlock(&hba->device_lock);
+
+ se_release_vpd_for_dev(dev);
+
+ kfree(dev->dev_status_queue_obj);
+ kfree(dev->dev_queue_obj);
+ kfree(dev);
+
+ return NULL;
+}
+EXPORT_SYMBOL(transport_add_device_to_core_hba);
+
+/* transport_generic_prepare_cdb():
+ *
+ * Since the Initiator sees iSCSI devices as LUNs, the SCSI CDB will
+ * contain the iSCSI LUN in bits 7-5 of byte 1 as per SAM-2.
+ * The point of this is since we are mapping iSCSI LUNs to
+ * SCSI Target IDs having a non-zero LUN in the CDB will throw the
+ * devices and HBAs for a loop.
+ */
+static inline void transport_generic_prepare_cdb(
+ unsigned char *cdb)
+{
+ switch (cdb[0]) {
+ case READ_10: /* SBC - RDProtect */
+ case READ_12: /* SBC - RDProtect */
+ case READ_16: /* SBC - RDProtect */
+ case SEND_DIAGNOSTIC: /* SPC - SELF-TEST Code */
+ case VERIFY: /* SBC - VRProtect */
+ case VERIFY_16: /* SBC - VRProtect */
+ case WRITE_VERIFY: /* SBC - VRProtect */
+ case WRITE_VERIFY_12: /* SBC - VRProtect */
+ break;
+ default:
+ cdb[1] &= 0x1f; /* clear logical unit number */
+ break;
+ }
+}
+
+static struct se_task *
+transport_generic_get_task(struct se_cmd *cmd,
+ enum dma_data_direction data_direction)
+{
+ struct se_task *task;
+ struct se_device *dev = SE_DEV(cmd);
+ unsigned long flags;
+
+ task = dev->transport->alloc_task(cmd);
+ if (!task) {
+ printk(KERN_ERR "Unable to allocate struct se_task\n");
+ return NULL;
+ }
+
+ INIT_LIST_HEAD(&task->t_list);
+ INIT_LIST_HEAD(&task->t_execute_list);
+ INIT_LIST_HEAD(&task->t_state_list);
+ init_completion(&task->task_stop_comp);
+ task->task_no = T_TASK(cmd)->t_tasks_no++;
+ task->task_se_cmd = cmd;
+ task->se_dev = dev;
+ task->task_data_direction = data_direction;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ list_add_tail(&task->t_list, &T_TASK(cmd)->t_task_list);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ return task;
+}
+
+static int transport_generic_cmd_sequencer(struct se_cmd *, unsigned char *);
+
+void transport_device_setup_cmd(struct se_cmd *cmd)
+{
+ cmd->se_dev = SE_LUN(cmd)->lun_se_dev;
+}
+EXPORT_SYMBOL(transport_device_setup_cmd);
+
+/*
+ * Used by fabric modules containing a local struct se_cmd within their
+ * fabric dependent per I/O descriptor.
+ */
+void transport_init_se_cmd(
+ struct se_cmd *cmd,
+ struct target_core_fabric_ops *tfo,
+ struct se_session *se_sess,
+ u32 data_length,
+ int data_direction,
+ int task_attr,
+ unsigned char *sense_buffer)
+{
+ INIT_LIST_HEAD(&cmd->se_lun_list);
+ INIT_LIST_HEAD(&cmd->se_delayed_list);
+ INIT_LIST_HEAD(&cmd->se_ordered_list);
+ /*
+ * Setup t_task pointer to t_task_backstore
+ */
+ cmd->t_task = &cmd->t_task_backstore;
+
+ INIT_LIST_HEAD(&T_TASK(cmd)->t_task_list);
+ init_completion(&T_TASK(cmd)->transport_lun_fe_stop_comp);
+ init_completion(&T_TASK(cmd)->transport_lun_stop_comp);
+ init_completion(&T_TASK(cmd)->t_transport_stop_comp);
+ spin_lock_init(&T_TASK(cmd)->t_state_lock);
+ atomic_set(&T_TASK(cmd)->transport_dev_active, 1);
+
+ cmd->se_tfo = tfo;
+ cmd->se_sess = se_sess;
+ cmd->data_length = data_length;
+ cmd->data_direction = data_direction;
+ cmd->sam_task_attr = task_attr;
+ cmd->sense_buffer = sense_buffer;
+}
+EXPORT_SYMBOL(transport_init_se_cmd);
+
+static int transport_check_alloc_task_attr(struct se_cmd *cmd)
+{
+ /*
+ * Check if SAM Task Attribute emulation is enabled for this
+ * struct se_device storage object
+ */
+ if (SE_DEV(cmd)->dev_task_attr_type != SAM_TASK_ATTR_EMULATED)
+ return 0;
+
+ if (cmd->sam_task_attr == TASK_ATTR_ACA) {
+ DEBUG_STA("SAM Task Attribute ACA"
+ " emulation is not supported\n");
+ return -1;
+ }
+ /*
+ * Used to determine when ORDERED commands should go from
+ * Dormant to Active status.
+ */
+ cmd->se_ordered_id = atomic_inc_return(&SE_DEV(cmd)->dev_ordered_id);
+ smp_mb__after_atomic_inc();
+ DEBUG_STA("Allocated se_ordered_id: %u for Task Attr: 0x%02x on %s\n",
+ cmd->se_ordered_id, cmd->sam_task_attr,
+ TRANSPORT(cmd->se_dev)->name);
+ return 0;
+}
+
+void transport_free_se_cmd(
+ struct se_cmd *se_cmd)
+{
+ if (se_cmd->se_tmr_req)
+ core_tmr_release_req(se_cmd->se_tmr_req);
+ /*
+ * Check and free any extended CDB buffer that was allocated
+ */
+ if (T_TASK(se_cmd)->t_task_cdb != T_TASK(se_cmd)->__t_task_cdb)
+ kfree(T_TASK(se_cmd)->t_task_cdb);
+}
+EXPORT_SYMBOL(transport_free_se_cmd);
+
+static void transport_generic_wait_for_tasks(struct se_cmd *, int, int);
+
+/* transport_generic_allocate_tasks():
+ *
+ * Called from fabric RX Thread.
+ */
+int transport_generic_allocate_tasks(
+ struct se_cmd *cmd,
+ unsigned char *cdb)
+{
+ int ret;
+
+ transport_generic_prepare_cdb(cdb);
+
+ /*
+ * This is needed for early exceptions.
+ */
+ cmd->transport_wait_for_tasks = &transport_generic_wait_for_tasks;
+
+ transport_device_setup_cmd(cmd);
+ /*
+ * Ensure that the received CDB is less than the max (252 + 8) bytes
+ * for VARIABLE_LENGTH_CMD
+ */
+ if (scsi_command_size(cdb) > SCSI_MAX_VARLEN_CDB_SIZE) {
+ printk(KERN_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);
+ return -1;
+ }
+ /*
+ * If the received CDB is larger than TCM_MAX_COMMAND_SIZE,
+ * allocate the additional extended CDB buffer now.. Otherwise
+ * setup the pointer from __t_task_cdb to t_task_cdb.
+ */
+ if (scsi_command_size(cdb) > sizeof(T_TASK(cmd)->__t_task_cdb)) {
+ T_TASK(cmd)->t_task_cdb = kzalloc(scsi_command_size(cdb),
+ GFP_KERNEL);
+ if (!(T_TASK(cmd)->t_task_cdb)) {
+ printk(KERN_ERR "Unable to allocate T_TASK(cmd)->t_task_cdb"
+ " %u > sizeof(T_TASK(cmd)->__t_task_cdb): %lu ops\n",
+ scsi_command_size(cdb),
+ (unsigned long)sizeof(T_TASK(cmd)->__t_task_cdb));
+ return -1;
+ }
+ } else
+ T_TASK(cmd)->t_task_cdb = &T_TASK(cmd)->__t_task_cdb[0];
+ /*
+ * Copy the original CDB into T_TASK(cmd).
+ */
+ memcpy(T_TASK(cmd)->t_task_cdb, cdb, scsi_command_size(cdb));
+ /*
+ * Setup the received CDB based on SCSI defined opcodes and
+ * perform unit attention, persistent reservations and ALUA
+ * checks for virtual device backends. The T_TASK(cmd)->t_task_cdb
+ * pointer is expected to be setup before we reach this point.
+ */
+ ret = transport_generic_cmd_sequencer(cmd, cdb);
+ if (ret < 0)
+ return ret;
+ /*
+ * Check for SAM Task Attribute Emulation
+ */
+ if (transport_check_alloc_task_attr(cmd) < 0) {
+ cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+ return -2;
+ }
+ spin_lock(&cmd->se_lun->lun_sep_lock);
+ if (cmd->se_lun->lun_sep)
+ cmd->se_lun->lun_sep->sep_stats.cmd_pdus++;
+ spin_unlock(&cmd->se_lun->lun_sep_lock);
+ return 0;
+}
+EXPORT_SYMBOL(transport_generic_allocate_tasks);
+
+/*
+ * Used by fabric module frontends not defining a TFO->new_cmd_map()
+ * to queue up a newly setup se_cmd w/ TRANSPORT_NEW_CMD statis
+ */
+int transport_generic_handle_cdb(
+ struct se_cmd *cmd)
+{
+ if (!SE_LUN(cmd)) {
+ dump_stack();
+ printk(KERN_ERR "SE_LUN(cmd) is NULL\n");
+ return -1;
+ }
+
+ transport_add_cmd_to_queue(cmd, TRANSPORT_NEW_CMD);
+ return 0;
+}
+EXPORT_SYMBOL(transport_generic_handle_cdb);
+
+/*
+ * Used by fabric module frontends defining a TFO->new_cmd_map() caller
+ * to queue up a newly setup se_cmd w/ TRANSPORT_NEW_CMD_MAP in order to
+ * complete setup in TCM process context w/ TFO->new_cmd_map().
+ */
+int transport_generic_handle_cdb_map(
+ struct se_cmd *cmd)
+{
+ if (!SE_LUN(cmd)) {
+ dump_stack();
+ printk(KERN_ERR "SE_LUN(cmd) is NULL\n");
+ return -1;
+ }
+
+ transport_add_cmd_to_queue(cmd, TRANSPORT_NEW_CMD_MAP);
+ return 0;
+}
+EXPORT_SYMBOL(transport_generic_handle_cdb_map);
+
+/* transport_generic_handle_data():
+ *
+ *
+ */
+int transport_generic_handle_data(
+ struct se_cmd *cmd)
+{
+ /*
+ * For the software fabric case, then we assume the nexus is being
+ * failed/shutdown when signals are pending from the kthread context
+ * caller, so we return a failure. For the HW target mode case running
+ * in interrupt code, the signal_pending() check is skipped.
+ */
+ if (!in_interrupt() && signal_pending(current))
+ return -1;
+ /*
+ * If the received CDB has aleady been ABORTED by the generic
+ * target engine, we now call transport_check_aborted_status()
+ * to queue any delated TASK_ABORTED status for the received CDB to the
+ * fabric module as we are expecting no futher incoming DATA OUT
+ * sequences at this point.
+ */
+ if (transport_check_aborted_status(cmd, 1) != 0)
+ return 0;
+
+ transport_add_cmd_to_queue(cmd, TRANSPORT_PROCESS_WRITE);
+ return 0;
+}
+EXPORT_SYMBOL(transport_generic_handle_data);
+
+/* transport_generic_handle_tmr():
+ *
+ *
+ */
+int transport_generic_handle_tmr(
+ struct se_cmd *cmd)
+{
+ /*
+ * This is needed for early exceptions.
+ */
+ cmd->transport_wait_for_tasks = &transport_generic_wait_for_tasks;
+ transport_device_setup_cmd(cmd);
+
+ transport_add_cmd_to_queue(cmd, TRANSPORT_PROCESS_TMR);
+ return 0;
+}
+EXPORT_SYMBOL(transport_generic_handle_tmr);
+
+static int transport_stop_tasks_for_cmd(struct se_cmd *cmd)
+{
+ struct se_task *task, *task_tmp;
+ unsigned long flags;
+ int ret = 0;
+
+ DEBUG_TS("ITT[0x%08x] - Stopping tasks\n",
+ CMD_TFO(cmd)->get_task_tag(cmd));
+
+ /*
+ * No tasks remain in the execution queue
+ */
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ list_for_each_entry_safe(task, task_tmp,
+ &T_TASK(cmd)->t_task_list, t_list) {
+ DEBUG_TS("task_no[%d] - Processing task %p\n",
+ task->task_no, task);
+ /*
+ * If the struct se_task has not been sent and is not active,
+ * remove the struct se_task from the execution queue.
+ */
+ if (!atomic_read(&task->task_sent) &&
+ !atomic_read(&task->task_active)) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock,
+ flags);
+ transport_remove_task_from_execute_queue(task,
+ task->se_dev);
+
+ DEBUG_TS("task_no[%d] - Removed from execute queue\n",
+ task->task_no);
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ continue;
+ }
+
+ /*
+ * If the struct se_task is active, sleep until it is returned
+ * from the plugin.
+ */
+ if (atomic_read(&task->task_active)) {
+ atomic_set(&task->task_stop, 1);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock,
+ flags);
+
+ DEBUG_TS("task_no[%d] - Waiting to complete\n",
+ task->task_no);
+ wait_for_completion(&task->task_stop_comp);
+ DEBUG_TS("task_no[%d] - Stopped successfully\n",
+ task->task_no);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ atomic_dec(&T_TASK(cmd)->t_task_cdbs_left);
+
+ atomic_set(&task->task_active, 0);
+ atomic_set(&task->task_stop, 0);
+ } else {
+ DEBUG_TS("task_no[%d] - Did nothing\n", task->task_no);
+ ret++;
+ }
+
+ __transport_stop_task_timer(task, &flags);
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ return ret;
+}
+
+static void transport_failure_reset_queue_depth(struct se_device *dev)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&SE_HBA(dev)->hba_queue_lock, flags);;
+ atomic_inc(&dev->depth_left);
+ atomic_inc(&SE_HBA(dev)->left_queue_depth);
+ spin_unlock_irqrestore(&SE_HBA(dev)->hba_queue_lock, flags);
+}
+
+/*
+ * Handle SAM-esque emulation for generic transport request failures.
+ */
+static void transport_generic_request_failure(
+ struct se_cmd *cmd,
+ struct se_device *dev,
+ int complete,
+ int sc)
+{
+ DEBUG_GRF("-----[ Storage Engine Exception for cmd: %p ITT: 0x%08x"
+ " CDB: 0x%02x\n", cmd, CMD_TFO(cmd)->get_task_tag(cmd),
+ T_TASK(cmd)->t_task_cdb[0]);
+ DEBUG_GRF("-----[ i_state: %d t_state/def_t_state:"
+ " %d/%d transport_error_status: %d\n",
+ CMD_TFO(cmd)->get_cmd_state(cmd),
+ cmd->t_state, cmd->deferred_t_state,
+ cmd->transport_error_status);
+ DEBUG_GRF("-----[ t_task_cdbs: %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"
+ " t_transport_sent: %d\n", T_TASK(cmd)->t_task_cdbs,
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_left),
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_sent),
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_ex_left),
+ atomic_read(&T_TASK(cmd)->t_transport_active),
+ atomic_read(&T_TASK(cmd)->t_transport_stop),
+ atomic_read(&T_TASK(cmd)->t_transport_sent));
+
+ transport_stop_all_task_timers(cmd);
+
+ if (dev)
+ transport_failure_reset_queue_depth(dev);
+ /*
+ * For SAM Task Attribute emulation for failed struct se_cmd
+ */
+ if (cmd->se_dev->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
+ transport_complete_task_attr(cmd);
+
+ if (complete) {
+ transport_direct_request_timeout(cmd);
+ 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;
+ 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_TFO(cmd)->fall_back_to_erl0(cmd->se_sess);
+ CMD_TFO(cmd)->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:
+ /*
+ * No SENSE Data payload for this case, set SCSI Status
+ * and queue the response to $FABRIC_MOD.
+ *
+ * Uses linux/include/scsi/scsi.h SAM status codes defs
+ */
+ 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 (SE_SESS(cmd) &&
+ DEV_ATTRIB(cmd->se_dev)->emulate_ua_intlck_ctrl == 2)
+ core_scsi3_ua_allocate(SE_SESS(cmd)->se_node_acl,
+ cmd->orig_fe_lun, 0x2C,
+ ASCQ_2CH_PREVIOUS_RESERVATION_CONFLICT_STATUS);
+
+ CMD_TFO(cmd)->queue_status(cmd);
+ goto check_stop;
+ case PYX_TRANSPORT_USE_SENSE_REASON:
+ /*
+ * struct se_cmd->scsi_sense_reason already set
+ */
+ break;
+ default:
+ printk(KERN_ERR "Unknown transport error for CDB 0x%02x: %d\n",
+ T_TASK(cmd)->t_task_cdb[0],
+ cmd->transport_error_status);
+ cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+ break;
+ }
+
+ if (!sc)
+ transport_new_cmd_failure(cmd);
+ else
+ transport_send_check_condition_and_sense(cmd,
+ cmd->scsi_sense_reason, 0);
+check_stop:
+ transport_lun_remove_cmd(cmd);
+ if (!(transport_cmd_check_stop_to_fabric(cmd)))
+ ;
+}
+
+static void transport_direct_request_timeout(struct se_cmd *cmd)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (!(atomic_read(&T_TASK(cmd)->t_transport_timeout))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ return;
+ }
+ if (atomic_read(&T_TASK(cmd)->t_task_cdbs_timeout_left)) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ return;
+ }
+
+ atomic_sub(atomic_read(&T_TASK(cmd)->t_transport_timeout),
+ &T_TASK(cmd)->t_se_count);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+}
+
+static void transport_generic_request_timeout(struct se_cmd *cmd)
+{
+ unsigned long flags;
+
+ /*
+ * Reset T_TASK(cmd)->t_se_count to allow transport_generic_remove()
+ * to allow last call to free memory resources.
+ */
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (atomic_read(&T_TASK(cmd)->t_transport_timeout) > 1) {
+ int tmp = (atomic_read(&T_TASK(cmd)->t_transport_timeout) - 1);
+
+ atomic_sub(tmp, &T_TASK(cmd)->t_se_count);
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_generic_remove(cmd, 0, 0);
+}
+
+static int
+transport_generic_allocate_buf(struct se_cmd *cmd, u32 data_length)
+{
+ unsigned char *buf;
+
+ buf = kzalloc(data_length, GFP_KERNEL);
+ if (!(buf)) {
+ printk(KERN_ERR "Unable to allocate memory for buffer\n");
+ return -1;
+ }
+
+ T_TASK(cmd)->t_tasks_se_num = 0;
+ T_TASK(cmd)->t_task_buf = buf;
+
+ return 0;
+}
+
+static inline u32 transport_lba_21(unsigned char *cdb)
+{
+ return ((cdb[1] & 0x1f) << 16) | (cdb[2] << 8) | cdb[3];
+}
+
+static inline u32 transport_lba_32(unsigned char *cdb)
+{
+ return (cdb[2] << 24) | (cdb[3] << 16) | (cdb[4] << 8) | cdb[5];
+}
+
+static inline unsigned long long transport_lba_64(unsigned char *cdb)
+{
+ unsigned int __v1, __v2;
+
+ __v1 = (cdb[2] << 24) | (cdb[3] << 16) | (cdb[4] << 8) | cdb[5];
+ __v2 = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
+
+ return ((unsigned long long)__v2) | (unsigned long long)__v1 << 32;
+}
+
+/*
+ * For VARIABLE_LENGTH_CDB w/ 32 byte extended CDBs
+ */
+static inline unsigned long long transport_lba_64_ext(unsigned char *cdb)
+{
+ unsigned int __v1, __v2;
+
+ __v1 = (cdb[12] << 24) | (cdb[13] << 16) | (cdb[14] << 8) | cdb[15];
+ __v2 = (cdb[16] << 24) | (cdb[17] << 16) | (cdb[18] << 8) | cdb[19];
+
+ return ((unsigned long long)__v2) | (unsigned long long)__v1 << 32;
+}
+
+static void transport_set_supported_SAM_opcode(struct se_cmd *se_cmd)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&T_TASK(se_cmd)->t_state_lock, flags);
+ se_cmd->se_cmd_flags |= SCF_SUPPORTED_SAM_OPCODE;
+ spin_unlock_irqrestore(&T_TASK(se_cmd)->t_state_lock, flags);
+}
+
+/*
+ * Called from interrupt context.
+ */
+static void transport_task_timeout_handler(unsigned long data)
+{
+ struct se_task *task = (struct se_task *)data;
+ struct se_cmd *cmd = TASK_CMD(task);
+ unsigned long flags;
+
+ DEBUG_TT("transport task timeout fired! task: %p cmd: %p\n", task, cmd);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (task->task_flags & TF_STOP) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ return;
+ }
+ task->task_flags &= ~TF_RUNNING;
+
+ /*
+ * Determine if transport_complete_task() has already been called.
+ */
+ if (!(atomic_read(&task->task_active))) {
+ DEBUG_TT("transport task: %p cmd: %p timeout task_active"
+ " == 0\n", task, cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ return;
+ }
+
+ atomic_inc(&T_TASK(cmd)->t_se_count);
+ atomic_inc(&T_TASK(cmd)->t_transport_timeout);
+ T_TASK(cmd)->t_tasks_failed = 1;
+
+ atomic_set(&task->task_timeout, 1);
+ task->task_error_status = PYX_TRANSPORT_TASK_TIMEOUT;
+ task->task_scsi_status = 1;
+
+ if (atomic_read(&task->task_stop)) {
+ DEBUG_TT("transport task: %p cmd: %p timeout task_stop"
+ " == 1\n", task, cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ complete(&task->task_stop_comp);
+ return;
+ }
+
+ if (!(atomic_dec_and_test(&T_TASK(cmd)->t_task_cdbs_left))) {
+ DEBUG_TT("transport task: %p cmd: %p timeout non zero"
+ " t_task_cdbs_left\n", task, cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ return;
+ }
+ DEBUG_TT("transport task: %p cmd: %p timeout ZERO t_task_cdbs_left\n",
+ task, cmd);
+
+ cmd->t_state = TRANSPORT_COMPLETE_FAILURE;
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_add_cmd_to_queue(cmd, TRANSPORT_COMPLETE_FAILURE);
+}
+
+/*
+ * Called with T_TASK(cmd)->t_state_lock held.
+ */
+static void transport_start_task_timer(struct se_task *task)
+{
+ struct se_device *dev = task->se_dev;
+ int timeout;
+
+ if (task->task_flags & TF_RUNNING)
+ return;
+ /*
+ * If the task_timeout is disabled, exit now.
+ */
+ timeout = DEV_ATTRIB(dev)->task_timeout;
+ if (!(timeout))
+ return;
+
+ init_timer(&task->task_timer);
+ task->task_timer.expires = (get_jiffies_64() + timeout * HZ);
+ task->task_timer.data = (unsigned long) task;
+ task->task_timer.function = transport_task_timeout_handler;
+
+ task->task_flags |= TF_RUNNING;
+ add_timer(&task->task_timer);
+#if 0
+ printk(KERN_INFO "Starting task timer for cmd: %p task: %p seconds:"
+ " %d\n", task->task_se_cmd, task, timeout);
+#endif
+}
+
+/*
+ * Called with spin_lock_irq(&T_TASK(cmd)->t_state_lock) held.
+ */
+void __transport_stop_task_timer(struct se_task *task, unsigned long *flags)
+{
+ struct se_cmd *cmd = TASK_CMD(task);
+
+ if (!(task->task_flags & TF_RUNNING))
+ return;
+
+ task->task_flags |= TF_STOP;
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, *flags);
+
+ del_timer_sync(&task->task_timer);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, *flags);
+ task->task_flags &= ~TF_RUNNING;
+ task->task_flags &= ~TF_STOP;
+}
+
+static void transport_stop_all_task_timers(struct se_cmd *cmd)
+{
+ struct se_task *task = NULL, *task_tmp;
+ unsigned long flags;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ list_for_each_entry_safe(task, task_tmp,
+ &T_TASK(cmd)->t_task_list, t_list)
+ __transport_stop_task_timer(task, &flags);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+}
+
+static inline int transport_tcq_window_closed(struct se_device *dev)
+{
+ if (dev->dev_tcq_window_closed++ <
+ PYX_TRANSPORT_WINDOW_CLOSED_THRESHOLD) {
+ msleep(PYX_TRANSPORT_WINDOW_CLOSED_WAIT_SHORT);
+ } else
+ msleep(PYX_TRANSPORT_WINDOW_CLOSED_WAIT_LONG);
+
+ wake_up_interruptible(&dev->dev_queue_obj->thread_wq);
+ return 0;
+}
+
+/*
+ * Called from Fabric Module context from transport_execute_tasks()
+ *
+ * The return of this function determins if the tasks from struct se_cmd
+ * get added to the execution queue in transport_execute_tasks(),
+ * or are added to the delayed or ordered lists here.
+ */
+static inline int transport_execute_task_attr(struct se_cmd *cmd)
+{
+ if (SE_DEV(cmd)->dev_task_attr_type != SAM_TASK_ATTR_EMULATED)
+ return 1;
+ /*
+ * Check for the existance of HEAD_OF_QUEUE, and if true return 1
+ * to allow the passed struct se_cmd list of tasks to the front of the list.
+ */
+ if (cmd->sam_task_attr == TASK_ATTR_HOQ) {
+ atomic_inc(&SE_DEV(cmd)->dev_hoq_count);
+ smp_mb__after_atomic_inc();
+ DEBUG_STA("Added HEAD_OF_QUEUE for CDB:"
+ " 0x%02x, se_ordered_id: %u\n",
+ T_TASK(cmd)->t_task_cdb[0],
+ cmd->se_ordered_id);
+ return 1;
+ } else if (cmd->sam_task_attr == TASK_ATTR_ORDERED) {
+ spin_lock(&SE_DEV(cmd)->ordered_cmd_lock);
+ list_add_tail(&cmd->se_ordered_list,
+ &SE_DEV(cmd)->ordered_cmd_list);
+ spin_unlock(&SE_DEV(cmd)->ordered_cmd_lock);
+
+ atomic_inc(&SE_DEV(cmd)->dev_ordered_sync);
+ smp_mb__after_atomic_inc();
+
+ DEBUG_STA("Added ORDERED for CDB: 0x%02x to ordered"
+ " list, se_ordered_id: %u\n",
+ T_TASK(cmd)->t_task_cdb[0],
+ cmd->se_ordered_id);
+ /*
+ * Add ORDERED command to tail of execution queue if
+ * no other older commands exist that need to be
+ * completed first.
+ */
+ if (!(atomic_read(&SE_DEV(cmd)->simple_cmds)))
+ return 1;
+ } else {
+ /*
+ * For SIMPLE and UNTAGGED Task Attribute commands
+ */
+ atomic_inc(&SE_DEV(cmd)->simple_cmds);
+ smp_mb__after_atomic_inc();
+ }
+ /*
+ * Otherwise if one or more outstanding ORDERED task attribute exist,
+ * add the dormant task(s) built for the passed struct se_cmd to the
+ * execution queue and become in Active state for this struct se_device.
+ */
+ if (atomic_read(&SE_DEV(cmd)->dev_ordered_sync) != 0) {
+ /*
+ * Otherwise, add cmd w/ tasks to delayed cmd queue that
+ * will be drained upon competion of HEAD_OF_QUEUE task.
+ */
+ spin_lock(&SE_DEV(cmd)->delayed_cmd_lock);
+ cmd->se_cmd_flags |= SCF_DELAYED_CMD_FROM_SAM_ATTR;
+ list_add_tail(&cmd->se_delayed_list,
+ &SE_DEV(cmd)->delayed_cmd_list);
+ spin_unlock(&SE_DEV(cmd)->delayed_cmd_lock);
+
+ DEBUG_STA("Added CDB: 0x%02x Task Attr: 0x%02x to"
+ " delayed CMD list, se_ordered_id: %u\n",
+ T_TASK(cmd)->t_task_cdb[0], cmd->sam_task_attr,
+ cmd->se_ordered_id);
+ /*
+ * Return zero to let transport_execute_tasks() know
+ * not to add the delayed tasks to the execution list.
+ */
+ return 0;
+ }
+ /*
+ * Otherwise, no ORDERED task attributes exist..
+ */
+ return 1;
+}
+
+/*
+ * Called from fabric module context in transport_generic_new_cmd() and
+ * transport_generic_process_write()
+ */
+static int transport_execute_tasks(struct se_cmd *cmd)
+{
+ int add_tasks;
+
+ if (!(cmd->se_cmd_flags & SCF_SE_DISABLE_ONLINE_CHECK)) {
+ 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, NULL, 0, 1);
+ return 0;
+ }
+ }
+ /*
+ * Call transport_cmd_check_stop() to see if a fabric exception
+ * has occured that prevents execution.
+ */
+ if (!(transport_cmd_check_stop(cmd, 0, TRANSPORT_PROCESSING))) {
+ /*
+ * Check for SAM Task Attribute emulation and HEAD_OF_QUEUE
+ * attribute for the tasks of the received struct se_cmd CDB
+ */
+ add_tasks = transport_execute_task_attr(cmd);
+ if (add_tasks == 0)
+ goto execute_tasks;
+ /*
+ * This calls transport_add_tasks_from_cmd() to handle
+ * HEAD_OF_QUEUE ordering for SAM Task Attribute emulation
+ * (if enabled) in __transport_add_task_to_execute_queue() and
+ * transport_add_task_check_sam_attr().
+ */
+ transport_add_tasks_from_cmd(cmd);
+ }
+ /*
+ * Kick the execution queue for the cmd associated struct se_device
+ * storage object.
+ */
+execute_tasks:
+ __transport_execute_tasks(SE_DEV(cmd));
+ return 0;
+}
+
+/*
+ * Called to check struct se_device tcq depth window, and once open pull struct se_task
+ * from struct se_device->execute_task_list and
+ *
+ * Called from transport_processing_thread()
+ */
+static int __transport_execute_tasks(struct se_device *dev)
+{
+ int error;
+ struct se_cmd *cmd = NULL;
+ struct se_task *task;
+ unsigned long flags;
+
+ /*
+ * Check if there is enough room in the device and HBA queue to send
+ * struct se_transport_task's to the selected transport.
+ */
+check_depth:
+ spin_lock_irqsave(&SE_HBA(dev)->hba_queue_lock, flags);
+ if (!(atomic_read(&dev->depth_left)) ||
+ !(atomic_read(&SE_HBA(dev)->left_queue_depth))) {
+ spin_unlock_irqrestore(&SE_HBA(dev)->hba_queue_lock, flags);
+ return transport_tcq_window_closed(dev);
+ }
+ dev->dev_tcq_window_closed = 0;
+
+ spin_lock(&dev->execute_task_lock);
+ task = transport_get_task_from_execute_queue(dev);
+ spin_unlock(&dev->execute_task_lock);
+
+ if (!task) {
+ spin_unlock_irqrestore(&SE_HBA(dev)->hba_queue_lock, flags);
+ return 0;
+ }
+
+ atomic_dec(&dev->depth_left);
+ atomic_dec(&SE_HBA(dev)->left_queue_depth);
+ spin_unlock_irqrestore(&SE_HBA(dev)->hba_queue_lock, flags);
+
+ cmd = TASK_CMD(task);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ atomic_set(&task->task_active, 1);
+ atomic_set(&task->task_sent, 1);
+ atomic_inc(&T_TASK(cmd)->t_task_cdbs_sent);
+
+ if (atomic_read(&T_TASK(cmd)->t_task_cdbs_sent) ==
+ T_TASK(cmd)->t_task_cdbs)
+ atomic_set(&cmd->transport_sent, 1);
+
+ transport_start_task_timer(task);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ /*
+ * The struct se_cmd->transport_emulate_cdb() function pointer is used
+ * to grab REPORT_LUNS CDBs before they hit the
+ * struct se_subsystem_api->do_task() caller below.
+ */
+ if (cmd->transport_emulate_cdb) {
+ error = cmd->transport_emulate_cdb(cmd);
+ if (error != 0) {
+ cmd->transport_error_status = error;
+ atomic_set(&task->task_active, 0);
+ atomic_set(&cmd->transport_sent, 0);
+ transport_stop_tasks_for_cmd(cmd);
+ transport_generic_request_failure(cmd, dev, 0, 1);
+ goto check_depth;
+ }
+ /*
+ * Handle the successful completion for transport_emulate_cdb()
+ * for synchronous operation, following SCF_EMULATE_CDB_ASYNC
+ * Otherwise the caller is expected to complete the task with
+ * proper status.
+ */
+ if (!(cmd->se_cmd_flags & SCF_EMULATE_CDB_ASYNC)) {
+ cmd->scsi_status = SAM_STAT_GOOD;
+ task->task_scsi_status = GOOD;
+ transport_complete_task(task, 1);
+ }
+ } else {
+ /*
+ * Currently for all virtual TCM plugins including IBLOCK, FILEIO and
+ * RAMDISK we use the internal transport_emulate_control_cdb() logic
+ * with struct se_subsystem_api callers for the primary SPC-3 TYPE_DISK
+ * LUN emulation code.
+ *
+ * For TCM/pSCSI and all other SCF_SCSI_DATA_SG_IO_CDB I/O tasks we
+ * call ->do_task() directly and let the underlying TCM subsystem plugin
+ * code handle the CDB emulation.
+ */
+ if ((TRANSPORT(dev)->transport_type != TRANSPORT_PLUGIN_PHBA_PDEV) &&
+ (!(TASK_CMD(task)->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB)))
+ error = transport_emulate_control_cdb(task);
+ else
+ error = TRANSPORT(dev)->do_task(task);
+
+ if (error != 0) {
+ cmd->transport_error_status = error;
+ atomic_set(&task->task_active, 0);
+ atomic_set(&cmd->transport_sent, 0);
+ transport_stop_tasks_for_cmd(cmd);
+ transport_generic_request_failure(cmd, dev, 0, 1);
+ }
+ }
+
+ goto check_depth;
+
+ 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(&T_TASK(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(&T_TASK(se_cmd)->t_state_lock, flags);
+
+ CMD_TFO(se_cmd)->new_cmd_failure(se_cmd);
+}
+
+static void transport_nop_wait_for_tasks(struct se_cmd *, int, int);
+
+static inline u32 transport_get_sectors_6(
+ unsigned char *cdb,
+ struct se_cmd *cmd,
+ int *ret)
+{
+ struct se_device *dev = SE_LUN(cmd)->lun_se_dev;
+
+ /*
+ * Assume TYPE_DISK for non struct se_device objects.
+ * Use 8-bit sector value.
+ */
+ if (!dev)
+ goto type_disk;
+
+ /*
+ * Use 24-bit allocation length for TYPE_TAPE.
+ */
+ if (TRANSPORT(dev)->get_device_type(dev) == TYPE_TAPE)
+ return (u32)(cdb[2] << 16) + (cdb[3] << 8) + cdb[4];
+
+ /*
+ * Everything else assume TYPE_DISK Sector CDB location.
+ * Use 8-bit sector value.
+ */
+type_disk:
+ return (u32)cdb[4];
+}
+
+static inline u32 transport_get_sectors_10(
+ unsigned char *cdb,
+ struct se_cmd *cmd,
+ int *ret)
+{
+ struct se_device *dev = SE_LUN(cmd)->lun_se_dev;
+
+ /*
+ * Assume TYPE_DISK for non struct se_device objects.
+ * Use 16-bit sector value.
+ */
+ if (!dev)
+ goto type_disk;
+
+ /*
+ * XXX_10 is not defined in SSC, throw an exception
+ */
+ if (TRANSPORT(dev)->get_device_type(dev) == TYPE_TAPE) {
+ *ret = -1;
+ return 0;
+ }
+
+ /*
+ * Everything else assume TYPE_DISK Sector CDB location.
+ * Use 16-bit sector value.
+ */
+type_disk:
+ return (u32)(cdb[7] << 8) + cdb[8];
+}
+
+static inline u32 transport_get_sectors_12(
+ unsigned char *cdb,
+ struct se_cmd *cmd,
+ int *ret)
+{
+ struct se_device *dev = SE_LUN(cmd)->lun_se_dev;
+
+ /*
+ * Assume TYPE_DISK for non struct se_device objects.
+ * Use 32-bit sector value.
+ */
+ if (!dev)
+ goto type_disk;
+
+ /*
+ * XXX_12 is not defined in SSC, throw an exception
+ */
+ if (TRANSPORT(dev)->get_device_type(dev) == TYPE_TAPE) {
+ *ret = -1;
+ return 0;
+ }
+
+ /*
+ * Everything else assume TYPE_DISK Sector CDB location.
+ * Use 32-bit sector value.
+ */
+type_disk:
+ return (u32)(cdb[6] << 24) + (cdb[7] << 16) + (cdb[8] << 8) + cdb[9];
+}
+
+static inline u32 transport_get_sectors_16(
+ unsigned char *cdb,
+ struct se_cmd *cmd,
+ int *ret)
+{
+ struct se_device *dev = SE_LUN(cmd)->lun_se_dev;
+
+ /*
+ * Assume TYPE_DISK for non struct se_device objects.
+ * Use 32-bit sector value.
+ */
+ if (!dev)
+ goto type_disk;
+
+ /*
+ * Use 24-bit allocation length for TYPE_TAPE.
+ */
+ if (TRANSPORT(dev)->get_device_type(dev) == TYPE_TAPE)
+ return (u32)(cdb[12] << 16) + (cdb[13] << 8) + cdb[14];
+
+type_disk:
+ return (u32)(cdb[10] << 24) + (cdb[11] << 16) +
+ (cdb[12] << 8) + cdb[13];
+}
+
+/*
+ * Used for VARIABLE_LENGTH_CDB WRITE_32 and READ_32 variants
+ */
+static inline u32 transport_get_sectors_32(
+ unsigned char *cdb,
+ struct se_cmd *cmd,
+ int *ret)
+{
+ /*
+ * Assume TYPE_DISK for non struct se_device objects.
+ * Use 32-bit sector value.
+ */
+ return (u32)(cdb[28] << 24) + (cdb[29] << 16) +
+ (cdb[30] << 8) + cdb[31];
+
+}
+
+static inline u32 transport_get_size(
+ u32 sectors,
+ unsigned char *cdb,
+ struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+
+ if (TRANSPORT(dev)->get_device_type(dev) == TYPE_TAPE) {
+ if (cdb[1] & 1) { /* sectors */
+ return DEV_ATTRIB(dev)->block_size * sectors;
+ } else /* bytes */
+ return sectors;
+ }
+#if 0
+ printk(KERN_INFO "Returning block_size: %u, sectors: %u == %u for"
+ " %s object\n", DEV_ATTRIB(dev)->block_size, sectors,
+ DEV_ATTRIB(dev)->block_size * sectors,
+ TRANSPORT(dev)->name);
+#endif
+ return DEV_ATTRIB(dev)->block_size * sectors;
+}
+
+unsigned char transport_asciihex_to_binaryhex(unsigned char val[2])
+{
+ unsigned char result = 0;
+ /*
+ * MSB
+ */
+ if ((val[0] >= 'a') && (val[0] <= 'f'))
+ result = ((val[0] - 'a' + 10) & 0xf) << 4;
+ else
+ if ((val[0] >= 'A') && (val[0] <= 'F'))
+ result = ((val[0] - 'A' + 10) & 0xf) << 4;
+ else /* digit */
+ result = ((val[0] - '0') & 0xf) << 4;
+ /*
+ * LSB
+ */
+ if ((val[1] >= 'a') && (val[1] <= 'f'))
+ result |= ((val[1] - 'a' + 10) & 0xf);
+ else
+ if ((val[1] >= 'A') && (val[1] <= 'F'))
+ result |= ((val[1] - 'A' + 10) & 0xf);
+ else /* digit */
+ result |= ((val[1] - '0') & 0xf);
+
+ return result;
+}
+EXPORT_SYMBOL(transport_asciihex_to_binaryhex);
+
+static void transport_xor_callback(struct se_cmd *cmd)
+{
+ unsigned char *buf, *addr;
+ struct se_mem *se_mem;
+ unsigned int offset;
+ int i;
+ /*
+ * From sbc3r22.pdf section 5.48 XDWRITEREAD (10) command
+ *
+ * 1) read the specified logical block(s);
+ * 2) transfer logical blocks from the data-out buffer;
+ * 3) XOR the logical blocks transferred from the data-out buffer with
+ * the logical blocks read, storing the resulting XOR data in a buffer;
+ * 4) if the DISABLE WRITE bit is set to zero, then write the logical
+ * blocks transferred from the data-out buffer; and
+ * 5) transfer the resulting XOR data to the data-in buffer.
+ */
+ buf = kmalloc(cmd->data_length, GFP_KERNEL);
+ if (!(buf)) {
+ printk(KERN_ERR "Unable to allocate xor_callback buf\n");
+ return;
+ }
+ /*
+ * Copy the scatterlist WRITE buffer located at T_TASK(cmd)->t_mem_list
+ * into the locally allocated *buf
+ */
+ transport_memcpy_se_mem_read_contig(cmd, buf, T_TASK(cmd)->t_mem_list);
+ /*
+ * Now perform the XOR against the BIDI read memory located at
+ * T_TASK(cmd)->t_mem_bidi_list
+ */
+
+ offset = 0;
+ list_for_each_entry(se_mem, T_TASK(cmd)->t_mem_bidi_list, se_list) {
+ addr = (unsigned char *)kmap_atomic(se_mem->se_page, KM_USER0);
+ if (!(addr))
+ goto out;
+
+ for (i = 0; i < se_mem->se_len; i++)
+ *(addr + se_mem->se_off + i) ^= *(buf + offset + i);
+
+ offset += se_mem->se_len;
+ kunmap_atomic(addr, KM_USER0);
+ }
+out:
+ kfree(buf);
+}
+
+/*
+ * Used to obtain Sense Data from underlying Linux/SCSI struct scsi_cmnd
+ */
+static int transport_get_sense_data(struct se_cmd *cmd)
+{
+ unsigned char *buffer = cmd->sense_buffer, *sense_buffer = NULL;
+ struct se_device *dev;
+ struct se_task *task = NULL, *task_tmp;
+ unsigned long flags;
+ u32 offset = 0;
+
+ if (!SE_LUN(cmd)) {
+ printk(KERN_ERR "SE_LUN(cmd) is NULL\n");
+ return -1;
+ }
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (cmd->se_cmd_flags & SCF_SENT_CHECK_CONDITION) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ return 0;
+ }
+
+ list_for_each_entry_safe(task, task_tmp,
+ &T_TASK(cmd)->t_task_list, t_list) {
+
+ if (!task->task_sense)
+ continue;
+
+ dev = task->se_dev;
+ if (!(dev))
+ continue;
+
+ if (!TRANSPORT(dev)->get_sense_buffer) {
+ printk(KERN_ERR "TRANSPORT(dev)->get_sense_buffer"
+ " is NULL\n");
+ continue;
+ }
+
+ sense_buffer = TRANSPORT(dev)->get_sense_buffer(task);
+ if (!(sense_buffer)) {
+ printk(KERN_ERR "ITT[0x%08x]_TASK[%d]: Unable to locate"
+ " sense buffer for task with sense\n",
+ CMD_TFO(cmd)->get_task_tag(cmd), task->task_no);
+ continue;
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ offset = CMD_TFO(cmd)->set_fabric_sense_len(cmd,
+ TRANSPORT_SENSE_BUFFER);
+
+ memcpy((void *)&buffer[offset], (void *)sense_buffer,
+ TRANSPORT_SENSE_BUFFER);
+ cmd->scsi_status = task->task_scsi_status;
+ /* Automatically padded */
+ cmd->scsi_sense_length =
+ (TRANSPORT_SENSE_BUFFER + offset);
+
+ printk(KERN_INFO "HBA_[%u]_PLUG[%s]: Set SAM STATUS: 0x%02x"
+ " and sense\n",
+ dev->se_hba->hba_id, TRANSPORT(dev)->name,
+ cmd->scsi_status);
+ return 0;
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ return -1;
+}
+
+static int transport_allocate_resources(struct se_cmd *cmd)
+{
+ u32 length = cmd->data_length;
+
+ if ((cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB) ||
+ (cmd->se_cmd_flags & SCF_SCSI_CONTROL_SG_IO_CDB))
+ return transport_generic_get_mem(cmd, length, PAGE_SIZE);
+ else if (cmd->se_cmd_flags & SCF_SCSI_CONTROL_NONSG_IO_CDB)
+ return transport_generic_allocate_buf(cmd, length);
+ else
+ return 0;
+}
+
+static int
+transport_handle_reservation_conflict(struct se_cmd *cmd)
+{
+ cmd->transport_wait_for_tasks = &transport_nop_wait_for_tasks;
+ 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 (SE_SESS(cmd) &&
+ DEV_ATTRIB(cmd->se_dev)->emulate_ua_intlck_ctrl == 2)
+ core_scsi3_ua_allocate(SE_SESS(cmd)->se_node_acl,
+ cmd->orig_fe_lun, 0x2C,
+ ASCQ_2CH_PREVIOUS_RESERVATION_CONFLICT_STATUS);
+ return -2;
+}
+
+/* transport_generic_cmd_sequencer():
+ *
+ * Generic Command Sequencer that should work for most DAS transport
+ * drivers.
+ *
+ * Called from transport_generic_allocate_tasks() in the $FABRIC_MOD
+ * RX Thread.
+ *
+ * FIXME: Need to support other SCSI OPCODES where as well.
+ */
+static int transport_generic_cmd_sequencer(
+ struct se_cmd *cmd,
+ unsigned char *cdb)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_subsystem_dev *su_dev = dev->se_sub_dev;
+ int ret = 0, sector_ret = 0, passthrough;
+ u32 sectors = 0, size = 0, pr_reg_type = 0;
+ u16 service_action;
+ u8 alua_ascq = 0;
+ /*
+ * Check for an existing UNIT ATTENTION condition
+ */
+ if (core_scsi3_ua_check(cmd, cdb) < 0) {
+ cmd->transport_wait_for_tasks =
+ &transport_nop_wait_for_tasks;
+ cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ cmd->scsi_sense_reason = TCM_CHECK_CONDITION_UNIT_ATTENTION;
+ return -2;
+ }
+ /*
+ * Check status of Asymmetric Logical Unit Assignment port
+ */
+ ret = T10_ALUA(su_dev)->alua_state_check(cmd, cdb, &alua_ascq);
+ if (ret != 0) {
+ cmd->transport_wait_for_tasks = &transport_nop_wait_for_tasks;
+ /*
+ * Set SCSI additional sense code (ASC) to 'LUN Not Accessable';
+ * The ALUA additional sense code qualifier (ASCQ) is determined
+ * by the ALUA primary or secondary access state..
+ */
+ if (ret > 0) {
+#if 0
+ printk(KERN_INFO "[%s]: ALUA TG Port not available,"
+ " SenseKey: NOT_READY, ASC/ASCQ: 0x04/0x%02x\n",
+ CMD_TFO(cmd)->get_fabric_name(), alua_ascq);
+#endif
+ transport_set_sense_codes(cmd, 0x04, alua_ascq);
+ cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ cmd->scsi_sense_reason = TCM_CHECK_CONDITION_NOT_READY;
+ return -2;
+ }
+ goto out_invalid_cdb_field;
+ }
+ /*
+ * Check status for SPC-3 Persistent Reservations
+ */
+ if (T10_PR_OPS(su_dev)->t10_reservation_check(cmd, &pr_reg_type) != 0) {
+ if (T10_PR_OPS(su_dev)->t10_seq_non_holder(
+ cmd, cdb, pr_reg_type) != 0)
+ return transport_handle_reservation_conflict(cmd);
+ /*
+ * This means the CDB is allowed for the SCSI Initiator port
+ * when said port is *NOT* holding the legacy SPC-2 or
+ * SPC-3 Persistent Reservation.
+ */
+ }
+
+ switch (cdb[0]) {
+ case READ_6:
+ sectors = transport_get_sectors_6(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_6;
+ T_TASK(cmd)->t_task_lba = transport_lba_21(cdb);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ break;
+ case READ_10:
+ sectors = transport_get_sectors_10(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_10;
+ T_TASK(cmd)->t_task_lba = transport_lba_32(cdb);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ break;
+ case READ_12:
+ sectors = transport_get_sectors_12(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_12;
+ T_TASK(cmd)->t_task_lba = transport_lba_32(cdb);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ break;
+ case READ_16:
+ sectors = transport_get_sectors_16(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_16;
+ T_TASK(cmd)->t_task_lba = transport_lba_64(cdb);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ break;
+ case WRITE_6:
+ sectors = transport_get_sectors_6(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_6;
+ T_TASK(cmd)->t_task_lba = transport_lba_21(cdb);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ break;
+ case WRITE_10:
+ sectors = transport_get_sectors_10(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_10;
+ T_TASK(cmd)->t_task_lba = transport_lba_32(cdb);
+ T_TASK(cmd)->t_tasks_fua = (cdb[1] & 0x8);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ break;
+ case WRITE_12:
+ sectors = transport_get_sectors_12(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_12;
+ T_TASK(cmd)->t_task_lba = transport_lba_32(cdb);
+ T_TASK(cmd)->t_tasks_fua = (cdb[1] & 0x8);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ break;
+ case WRITE_16:
+ sectors = transport_get_sectors_16(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_16;
+ T_TASK(cmd)->t_task_lba = transport_lba_64(cdb);
+ T_TASK(cmd)->t_tasks_fua = (cdb[1] & 0x8);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ break;
+ case XDWRITEREAD_10:
+ if ((cmd->data_direction != DMA_TO_DEVICE) ||
+ !(T_TASK(cmd)->t_tasks_bidi))
+ goto out_invalid_cdb_field;
+ sectors = transport_get_sectors_10(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->transport_split_cdb = &split_cdb_XX_10;
+ T_TASK(cmd)->t_task_lba = transport_lba_32(cdb);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+ passthrough = (TRANSPORT(dev)->transport_type ==
+ TRANSPORT_PLUGIN_PHBA_PDEV);
+ /*
+ * Skip the remaining assignments for TCM/PSCSI passthrough
+ */
+ if (passthrough)
+ break;
+ /*
+ * Setup BIDI XOR callback to be run during transport_generic_complete_ok()
+ */
+ cmd->transport_complete_callback = &transport_xor_callback;
+ T_TASK(cmd)->t_tasks_fua = (cdb[1] & 0x8);
+ break;
+ case VARIABLE_LENGTH_CMD:
+ service_action = get_unaligned_be16(&cdb[8]);
+ /*
+ * Determine if this is TCM/PSCSI device and we should disable
+ * internal emulation for this CDB.
+ */
+ passthrough = (TRANSPORT(dev)->transport_type ==
+ TRANSPORT_PLUGIN_PHBA_PDEV);
+
+ switch (service_action) {
+ case XDWRITEREAD_32:
+ sectors = transport_get_sectors_32(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ /*
+ * Use WRITE_32 and READ_32 opcodes for the emulated
+ * XDWRITE_READ_32 logic.
+ */
+ cmd->transport_split_cdb = &split_cdb_XX_32;
+ T_TASK(cmd)->t_task_lba = transport_lba_64_ext(cdb);
+ cmd->se_cmd_flags |= SCF_SCSI_DATA_SG_IO_CDB;
+
+ /*
+ * Skip the remaining assignments for TCM/PSCSI passthrough
+ */
+ if (passthrough)
+ break;
+
+ /*
+ * Setup BIDI XOR callback to be run during
+ * transport_generic_complete_ok()
+ */
+ cmd->transport_complete_callback = &transport_xor_callback;
+ T_TASK(cmd)->t_tasks_fua = (cdb[10] & 0x8);
+ break;
+ case WRITE_SAME_32:
+ sectors = transport_get_sectors_32(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ T_TASK(cmd)->t_task_lba = get_unaligned_be64(&cdb[12]);
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
+
+ /*
+ * Skip the remaining assignments for TCM/PSCSI passthrough
+ */
+ if (passthrough)
+ break;
+
+ if ((cdb[10] & 0x04) || (cdb[10] & 0x02)) {
+ printk(KERN_ERR "WRITE_SAME PBDATA and LBDATA"
+ " bits not supported for Block Discard"
+ " Emulation\n");
+ goto out_invalid_cdb_field;
+ }
+ /*
+ * Currently for the emulated case we only accept
+ * tpws with the UNMAP=1 bit set.
+ */
+ if (!(cdb[10] & 0x08)) {
+ printk(KERN_ERR "WRITE_SAME w/o UNMAP bit not"
+ " supported for Block Discard Emulation\n");
+ goto out_invalid_cdb_field;
+ }
+ break;
+ default:
+ printk(KERN_ERR "VARIABLE_LENGTH_CMD service action"
+ " 0x%04x not supported\n", service_action);
+ goto out_unsupported_cdb;
+ }
+ break;
+ case 0xa3:
+ if (TRANSPORT(dev)->get_device_type(dev) != TYPE_ROM) {
+ /* MAINTENANCE_IN from SCC-2 */
+ /*
+ * Check for emulated MI_REPORT_TARGET_PGS.
+ */
+ if (cdb[1] == MI_REPORT_TARGET_PGS) {
+ cmd->transport_emulate_cdb =
+ (T10_ALUA(su_dev)->alua_type ==
+ SPC3_ALUA_EMULATED) ?
+ &core_emulate_report_target_port_groups :
+ NULL;
+ }
+ size = (cdb[6] << 24) | (cdb[7] << 16) |
+ (cdb[8] << 8) | cdb[9];
+ } else {
+ /* GPCMD_SEND_KEY from multi media commands */
+ size = (cdb[8] << 8) + cdb[9];
+ }
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case MODE_SELECT:
+ size = cdb[4];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
+ break;
+ case MODE_SELECT_10:
+ size = (cdb[7] << 8) + cdb[8];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
+ break;
+ case MODE_SENSE:
+ size = cdb[4];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case MODE_SENSE_10:
+ case GPCMD_READ_BUFFER_CAPACITY:
+ case GPCMD_SEND_OPC:
+ case LOG_SELECT:
+ case LOG_SENSE:
+ size = (cdb[7] << 8) + cdb[8];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case READ_BLOCK_LIMITS:
+ size = READ_BLOCK_LEN;
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case GPCMD_GET_CONFIGURATION:
+ case GPCMD_READ_FORMAT_CAPACITIES:
+ case GPCMD_READ_DISC_INFO:
+ case GPCMD_READ_TRACK_RZONE_INFO:
+ size = (cdb[7] << 8) + cdb[8];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
+ break;
+ case PERSISTENT_RESERVE_IN:
+ case PERSISTENT_RESERVE_OUT:
+ cmd->transport_emulate_cdb =
+ (T10_RES(su_dev)->res_type ==
+ SPC3_PERSISTENT_RESERVATIONS) ?
+ &core_scsi3_emulate_pr : NULL;
+ size = (cdb[7] << 8) + cdb[8];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case GPCMD_MECHANISM_STATUS:
+ case GPCMD_READ_DVD_STRUCTURE:
+ size = (cdb[8] << 8) + cdb[9];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
+ break;
+ case READ_POSITION:
+ size = READ_POSITION_LEN;
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case 0xa4:
+ if (TRANSPORT(dev)->get_device_type(dev) != TYPE_ROM) {
+ /* MAINTENANCE_OUT from SCC-2
+ *
+ * Check for emulated MO_SET_TARGET_PGS.
+ */
+ if (cdb[1] == MO_SET_TARGET_PGS) {
+ cmd->transport_emulate_cdb =
+ (T10_ALUA(su_dev)->alua_type ==
+ SPC3_ALUA_EMULATED) ?
+ &core_emulate_set_target_port_groups :
+ NULL;
+ }
+
+ size = (cdb[6] << 24) | (cdb[7] << 16) |
+ (cdb[8] << 8) | cdb[9];
+ } else {
+ /* GPCMD_REPORT_KEY from multi media commands */
+ size = (cdb[8] << 8) + cdb[9];
+ }
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case INQUIRY:
+ size = (cdb[3] << 8) + cdb[4];
+ /*
+ * Do implict HEAD_OF_QUEUE processing for INQUIRY.
+ * See spc4r17 section 5.3
+ */
+ if (SE_DEV(cmd)->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
+ cmd->sam_task_attr = TASK_ATTR_HOQ;
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case READ_BUFFER:
+ size = (cdb[6] << 16) + (cdb[7] << 8) + cdb[8];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case READ_CAPACITY:
+ size = READ_CAP_LEN;
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case READ_MEDIA_SERIAL_NUMBER:
+ case SECURITY_PROTOCOL_IN:
+ case SECURITY_PROTOCOL_OUT:
+ size = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case SERVICE_ACTION_IN:
+ case ACCESS_CONTROL_IN:
+ case ACCESS_CONTROL_OUT:
+ case EXTENDED_COPY:
+ case READ_ATTRIBUTE:
+ case RECEIVE_COPY_RESULTS:
+ case WRITE_ATTRIBUTE:
+ size = (cdb[10] << 24) | (cdb[11] << 16) |
+ (cdb[12] << 8) | cdb[13];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case RECEIVE_DIAGNOSTIC:
+ case SEND_DIAGNOSTIC:
+ size = (cdb[3] << 8) | cdb[4];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+/* #warning FIXME: Figure out correct GPCMD_READ_CD blocksize. */
+#if 0
+ case GPCMD_READ_CD:
+ sectors = (cdb[6] << 16) + (cdb[7] << 8) + cdb[8];
+ size = (2336 * sectors);
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+#endif
+ case READ_TOC:
+ size = cdb[8];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case REQUEST_SENSE:
+ size = cdb[4];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case READ_ELEMENT_STATUS:
+ size = 65536 * cdb[7] + 256 * cdb[8] + cdb[9];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case WRITE_BUFFER:
+ size = (cdb[6] << 16) + (cdb[7] << 8) + cdb[8];
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case RESERVE:
+ case RESERVE_10:
+ /*
+ * The SPC-2 RESERVE does not contain a size in the SCSI CDB.
+ * Assume the passthrough or $FABRIC_MOD will tell us about it.
+ */
+ if (cdb[0] == RESERVE_10)
+ size = (cdb[7] << 8) | cdb[8];
+ else
+ size = cmd->data_length;
+
+ /*
+ * Setup the legacy emulated handler for SPC-2 and
+ * >= SPC-3 compatible reservation handling (CRH=1)
+ * Otherwise, we assume the underlying SCSI logic is
+ * is running in SPC_PASSTHROUGH, and wants reservations
+ * emulation disabled.
+ */
+ cmd->transport_emulate_cdb =
+ (T10_RES(su_dev)->res_type !=
+ SPC_PASSTHROUGH) ?
+ &core_scsi2_emulate_crh : NULL;
+ cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
+ break;
+ case RELEASE:
+ case RELEASE_10:
+ /*
+ * The SPC-2 RELEASE does not contain a size in the SCSI CDB.
+ * Assume the passthrough or $FABRIC_MOD will tell us about it.
+ */
+ if (cdb[0] == RELEASE_10)
+ size = (cdb[7] << 8) | cdb[8];
+ else
+ size = cmd->data_length;
+
+ cmd->transport_emulate_cdb =
+ (T10_RES(su_dev)->res_type !=
+ SPC_PASSTHROUGH) ?
+ &core_scsi2_emulate_crh : NULL;
+ cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
+ break;
+ case SYNCHRONIZE_CACHE:
+ case 0x91: /* SYNCHRONIZE_CACHE_16: */
+ /*
+ * Extract LBA and range to be flushed for emulated SYNCHRONIZE_CACHE
+ */
+ if (cdb[0] == SYNCHRONIZE_CACHE) {
+ sectors = transport_get_sectors_10(cdb, cmd, §or_ret);
+ T_TASK(cmd)->t_task_lba = transport_lba_32(cdb);
+ } else {
+ sectors = transport_get_sectors_16(cdb, cmd, §or_ret);
+ T_TASK(cmd)->t_task_lba = transport_lba_64(cdb);
+ }
+ if (sector_ret)
+ goto out_unsupported_cdb;
+
+ size = transport_get_size(sectors, cdb, cmd);
+ cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
+
+ /*
+ * For TCM/pSCSI passthrough, skip cmd->transport_emulate_cdb()
+ */
+ if (TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV)
+ break;
+ /*
+ * Set SCF_EMULATE_CDB_ASYNC to ensure asynchronous operation
+ * for SYNCHRONIZE_CACHE* Immed=1 case in __transport_execute_tasks()
+ */
+ cmd->se_cmd_flags |= SCF_EMULATE_CDB_ASYNC;
+ /*
+ * Check to ensure that LBA + Range does not exceed past end of
+ * device.
+ */
+ if (transport_get_sectors(cmd) < 0)
+ goto out_invalid_cdb_field;
+ break;
+ case UNMAP:
+ size = get_unaligned_be16(&cdb[7]);
+ passthrough = (TRANSPORT(dev)->transport_type ==
+ TRANSPORT_PLUGIN_PHBA_PDEV);
+ /*
+ * Determine if the received UNMAP used to for direct passthrough
+ * into Linux/SCSI with struct request via TCM/pSCSI or we are
+ * signaling the use of internal transport_generic_unmap() emulation
+ * for UNMAP -> Linux/BLOCK disbard with TCM/IBLOCK and TCM/FILEIO
+ * subsystem plugin backstores.
+ */
+ if (!(passthrough))
+ cmd->se_cmd_flags |= SCF_EMULATE_SYNC_UNMAP;
+
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ case WRITE_SAME_16:
+ sectors = transport_get_sectors_16(cdb, cmd, §or_ret);
+ if (sector_ret)
+ goto out_unsupported_cdb;
+ size = transport_get_size(sectors, cdb, cmd);
+ T_TASK(cmd)->t_task_lba = get_unaligned_be16(&cdb[2]);
+ passthrough = (TRANSPORT(dev)->transport_type ==
+ TRANSPORT_PLUGIN_PHBA_PDEV);
+ /*
+ * Determine if the received WRITE_SAME_16 is used to for direct
+ * passthrough into Linux/SCSI with struct request via TCM/pSCSI
+ * or we are signaling the use of internal WRITE_SAME + UNMAP=1
+ * emulation for -> Linux/BLOCK disbard with TCM/IBLOCK and
+ * TCM/FILEIO subsystem plugin backstores.
+ */
+ if (!(passthrough)) {
+ if ((cdb[1] & 0x04) || (cdb[1] & 0x02)) {
+ printk(KERN_ERR "WRITE_SAME PBDATA and LBDATA"
+ " bits not supported for Block Discard"
+ " Emulation\n");
+ goto out_invalid_cdb_field;
+ }
+ /*
+ * Currently for the emulated case we only accept
+ * tpws with the UNMAP=1 bit set.
+ */
+ if (!(cdb[1] & 0x08)) {
+ printk(KERN_ERR "WRITE_SAME w/o UNMAP bit not "
+ " supported for Block Discard Emulation\n");
+ goto out_invalid_cdb_field;
+ }
+ }
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_SG_IO_CDB;
+ break;
+ case ALLOW_MEDIUM_REMOVAL:
+ case GPCMD_CLOSE_TRACK:
+ case ERASE:
+ case INITIALIZE_ELEMENT_STATUS:
+ case GPCMD_LOAD_UNLOAD:
+ case REZERO_UNIT:
+ case SEEK_10:
+ case GPCMD_SET_SPEED:
+ case SPACE:
+ case START_STOP:
+ case TEST_UNIT_READY:
+ case VERIFY:
+ case WRITE_FILEMARKS:
+ case MOVE_MEDIUM:
+ cmd->se_cmd_flags |= SCF_SCSI_NON_DATA_CDB;
+ break;
+ case REPORT_LUNS:
+ cmd->transport_emulate_cdb =
+ &transport_core_report_lun_response;
+ size = (cdb[6] << 24) | (cdb[7] << 16) | (cdb[8] << 8) | cdb[9];
+ /*
+ * Do implict HEAD_OF_QUEUE processing for REPORT_LUNS
+ * See spc4r17 section 5.3
+ */
+ if (SE_DEV(cmd)->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
+ cmd->sam_task_attr = TASK_ATTR_HOQ;
+ cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB;
+ break;
+ default:
+ printk(KERN_WARNING "TARGET_CORE[%s]: Unsupported SCSI Opcode"
+ " 0x%02x, sending CHECK_CONDITION.\n",
+ CMD_TFO(cmd)->get_fabric_name(), cdb[0]);
+ cmd->transport_wait_for_tasks = &transport_nop_wait_for_tasks;
+ goto out_unsupported_cdb;
+ }
+
+ if (size != cmd->data_length) {
+ printk(KERN_WARNING "TARGET_CORE[%s]: Expected Transfer Length:"
+ " %u does not match SCSI CDB Length: %u for SAM Opcode:"
+ " 0x%02x\n", CMD_TFO(cmd)->get_fabric_name(),
+ cmd->data_length, size, cdb[0]);
+
+ cmd->cmd_spdtl = size;
+
+ if (cmd->data_direction == DMA_TO_DEVICE) {
+ printk(KERN_ERR "Rejecting underflow/overflow"
+ " WRITE data\n");
+ goto out_invalid_cdb_field;
+ }
+ /*
+ * Reject READ_* or WRITE_* with overflow/underflow for
+ * type SCF_SCSI_DATA_SG_IO_CDB.
+ */
+ if (!(ret) && (DEV_ATTRIB(dev)->block_size != 512)) {
+ printk(KERN_ERR "Failing OVERFLOW/UNDERFLOW for LBA op"
+ " CDB on non 512-byte sector setup subsystem"
+ " plugin: %s\n", TRANSPORT(dev)->name);
+ /* Returns CHECK_CONDITION + INVALID_CDB_FIELD */
+ goto out_invalid_cdb_field;
+ }
+
+ if (size > cmd->data_length) {
+ cmd->se_cmd_flags |= SCF_OVERFLOW_BIT;
+ cmd->residual_count = (size - cmd->data_length);
+ } else {
+ cmd->se_cmd_flags |= SCF_UNDERFLOW_BIT;
+ cmd->residual_count = (cmd->data_length - size);
+ }
+ cmd->data_length = size;
+ }
+
+ transport_set_supported_SAM_opcode(cmd);
+ return ret;
+
+out_unsupported_cdb:
+ cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ cmd->scsi_sense_reason = TCM_UNSUPPORTED_SCSI_OPCODE;
+ return -2;
+out_invalid_cdb_field:
+ cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ cmd->scsi_sense_reason = TCM_INVALID_CDB_FIELD;
+ return -2;
+}
+
+static inline void transport_release_tasks(struct se_cmd *);
+
+/*
+ * This function will copy a contiguous *src buffer into a destination
+ * struct scatterlist array.
+ */
+static void transport_memcpy_write_contig(
+ struct se_cmd *cmd,
+ struct scatterlist *sg_d,
+ unsigned char *src)
+{
+ u32 i = 0, length = 0, total_length = cmd->data_length;
+ void *dst;
+
+ while (total_length) {
+ length = sg_d[i].length;
+
+ if (length > total_length)
+ length = total_length;
+
+ dst = sg_virt(&sg_d[i]);
+
+ memcpy(dst, src, length);
+
+ if (!(total_length -= length))
+ return;
+
+ src += length;
+ i++;
+ }
+}
+
+/*
+ * This function will copy a struct scatterlist array *sg_s into a destination
+ * contiguous *dst buffer.
+ */
+static void transport_memcpy_read_contig(
+ struct se_cmd *cmd,
+ unsigned char *dst,
+ struct scatterlist *sg_s)
+{
+ u32 i = 0, length = 0, total_length = cmd->data_length;
+ void *src;
+
+ while (total_length) {
+ length = sg_s[i].length;
+
+ if (length > total_length)
+ length = total_length;
+
+ src = sg_virt(&sg_s[i]);
+
+ memcpy(dst, src, length);
+
+ if (!(total_length -= length))
+ return;
+
+ dst += length;
+ i++;
+ }
+}
+
+static void transport_memcpy_se_mem_read_contig(
+ struct se_cmd *cmd,
+ unsigned char *dst,
+ struct list_head *se_mem_list)
+{
+ struct se_mem *se_mem;
+ void *src;
+ u32 length = 0, total_length = cmd->data_length;
+
+ list_for_each_entry(se_mem, se_mem_list, se_list) {
+ length = se_mem->se_len;
+
+ if (length > total_length)
+ length = total_length;
+
+ src = page_address(se_mem->se_page) + se_mem->se_off;
+
+ memcpy(dst, src, length);
+
+ if (!(total_length -= length))
+ return;
+
+ dst += length;
+ }
+}
+
+/*
+ * Called from transport_generic_complete_ok() and
+ * transport_generic_request_failure() to determine which dormant/delayed
+ * and ordered cmds need to have their tasks added to the execution queue.
+ */
+static void transport_complete_task_attr(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_cmd *cmd_p, *cmd_tmp;
+ int new_active_tasks = 0;
+
+ if (cmd->sam_task_attr == TASK_ATTR_SIMPLE) {
+ atomic_dec(&dev->simple_cmds);
+ smp_mb__after_atomic_dec();
+ dev->dev_cur_ordered_id++;
+ DEBUG_STA("Incremented dev->dev_cur_ordered_id: %u for"
+ " SIMPLE: %u\n", dev->dev_cur_ordered_id,
+ cmd->se_ordered_id);
+ } else if (cmd->sam_task_attr == TASK_ATTR_HOQ) {
+ atomic_dec(&dev->dev_hoq_count);
+ smp_mb__after_atomic_dec();
+ dev->dev_cur_ordered_id++;
+ DEBUG_STA("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 == TASK_ATTR_ORDERED) {
+ spin_lock(&dev->ordered_cmd_lock);
+ list_del(&cmd->se_ordered_list);
+ atomic_dec(&dev->dev_ordered_sync);
+ smp_mb__after_atomic_dec();
+ spin_unlock(&dev->ordered_cmd_lock);
+
+ dev->dev_cur_ordered_id++;
+ DEBUG_STA("Incremented dev_cur_ordered_id: %u for ORDERED:"
+ " %u\n", dev->dev_cur_ordered_id, cmd->se_ordered_id);
+ }
+ /*
+ * Process all commands up to the last received
+ * ORDERED task attribute which requires another blocking
+ * boundary
+ */
+ spin_lock(&dev->delayed_cmd_lock);
+ list_for_each_entry_safe(cmd_p, cmd_tmp,
+ &dev->delayed_cmd_list, se_delayed_list) {
+
+ list_del(&cmd_p->se_delayed_list);
+ spin_unlock(&dev->delayed_cmd_lock);
+
+ DEBUG_STA("Calling add_tasks() for"
+ " cmd_p: 0x%02x Task Attr: 0x%02x"
+ " Dormant -> Active, se_ordered_id: %u\n",
+ T_TASK(cmd_p)->t_task_cdb[0],
+ cmd_p->sam_task_attr, cmd_p->se_ordered_id);
+
+ transport_add_tasks_from_cmd(cmd_p);
+ new_active_tasks++;
+
+ spin_lock(&dev->delayed_cmd_lock);
+ if (cmd_p->sam_task_attr == TASK_ATTR_ORDERED)
+ break;
+ }
+ spin_unlock(&dev->delayed_cmd_lock);
+ /*
+ * If new tasks have become active, wake up the transport thread
+ * to do the processing of the Active tasks.
+ */
+ if (new_active_tasks != 0)
+ wake_up_interruptible(&dev->dev_queue_obj->thread_wq);
+}
+
+static void transport_generic_complete_ok(struct se_cmd *cmd)
+{
+ int reason = 0;
+ /*
+ * Check if we need to move delayed/dormant tasks from cmds on the
+ * delayed execution list after a HEAD_OF_QUEUE or ORDERED Task
+ * Attribute.
+ */
+ if (SE_DEV(cmd)->dev_task_attr_type == SAM_TASK_ATTR_EMULATED)
+ transport_complete_task_attr(cmd);
+ /*
+ * Check if we need to retrieve a sense buffer from
+ * the struct se_cmd in question.
+ */
+ if (cmd->se_cmd_flags & SCF_TRANSPORT_TASK_SENSE) {
+ if (transport_get_sense_data(cmd) < 0)
+ reason = TCM_NON_EXISTENT_LUN;
+
+ /*
+ * Only set when an struct se_task->task_scsi_status returned
+ * a non GOOD status.
+ */
+ if (cmd->scsi_status) {
+ transport_send_check_condition_and_sense(
+ cmd, reason, 1);
+ transport_lun_remove_cmd(cmd);
+ transport_cmd_check_stop_to_fabric(cmd);
+ return;
+ }
+ }
+ /*
+ * Check for a callback, used by amoungst other things
+ * XDWRITE_READ_10 emulation.
+ */
+ if (cmd->transport_complete_callback)
+ cmd->transport_complete_callback(cmd);
+
+ switch (cmd->data_direction) {
+ case DMA_FROM_DEVICE:
+ spin_lock(&cmd->se_lun->lun_sep_lock);
+ if (SE_LUN(cmd)->lun_sep) {
+ SE_LUN(cmd)->lun_sep->sep_stats.tx_data_octets +=
+ cmd->data_length;
+ }
+ spin_unlock(&cmd->se_lun->lun_sep_lock);
+ /*
+ * If enabled by TCM fabirc module pre-registered SGL
+ * memory, perform the memcpy() from the TCM internal
+ * contigious buffer back to the original SGL.
+ */
+ if (cmd->se_cmd_flags & SCF_PASSTHROUGH_CONTIG_TO_SG)
+ transport_memcpy_write_contig(cmd,
+ T_TASK(cmd)->t_task_pt_sgl,
+ T_TASK(cmd)->t_task_buf);
+
+ CMD_TFO(cmd)->queue_data_in(cmd);
+ break;
+ case DMA_TO_DEVICE:
+ spin_lock(&cmd->se_lun->lun_sep_lock);
+ if (SE_LUN(cmd)->lun_sep) {
+ SE_LUN(cmd)->lun_sep->sep_stats.rx_data_octets +=
+ cmd->data_length;
+ }
+ spin_unlock(&cmd->se_lun->lun_sep_lock);
+ /*
+ * Check if we need to send READ payload for BIDI-COMMAND
+ */
+ if (T_TASK(cmd)->t_mem_bidi_list != NULL) {
+ spin_lock(&cmd->se_lun->lun_sep_lock);
+ if (SE_LUN(cmd)->lun_sep) {
+ SE_LUN(cmd)->lun_sep->sep_stats.tx_data_octets +=
+ cmd->data_length;
+ }
+ spin_unlock(&cmd->se_lun->lun_sep_lock);
+ CMD_TFO(cmd)->queue_data_in(cmd);
+ break;
+ }
+ /* Fall through for DMA_TO_DEVICE */
+ case DMA_NONE:
+ CMD_TFO(cmd)->queue_status(cmd);
+ break;
+ default:
+ break;
+ }
+
+ transport_lun_remove_cmd(cmd);
+ transport_cmd_check_stop_to_fabric(cmd);
+}
+
+static void transport_free_dev_tasks(struct se_cmd *cmd)
+{
+ struct se_task *task, *task_tmp;
+ unsigned long flags;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ list_for_each_entry_safe(task, task_tmp,
+ &T_TASK(cmd)->t_task_list, t_list) {
+ if (atomic_read(&task->task_active))
+ continue;
+
+ kfree(task->task_sg_bidi);
+ kfree(task->task_sg);
+
+ list_del(&task->t_list);
+
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ if (task->se_dev)
+ TRANSPORT(task->se_dev)->free_task(task);
+ else
+ printk(KERN_ERR "task[%u] - task->se_dev is NULL\n",
+ task->task_no);
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+}
+
+static inline void transport_free_pages(struct se_cmd *cmd)
+{
+ struct se_mem *se_mem, *se_mem_tmp;
+ int free_page = 1;
+
+ if (cmd->se_cmd_flags & SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC)
+ free_page = 0;
+ if (cmd->se_dev->transport->do_se_mem_map)
+ free_page = 0;
+
+ if (T_TASK(cmd)->t_task_buf) {
+ kfree(T_TASK(cmd)->t_task_buf);
+ T_TASK(cmd)->t_task_buf = NULL;
+ return;
+ }
+
+ /*
+ * Caller will handle releasing of struct se_mem.
+ */
+ if (cmd->se_cmd_flags & SCF_CMD_PASSTHROUGH_NOALLOC)
+ return;
+
+ if (!(T_TASK(cmd)->t_tasks_se_num))
+ return;
+
+ list_for_each_entry_safe(se_mem, se_mem_tmp,
+ T_TASK(cmd)->t_mem_list, se_list) {
+ /*
+ * We only release call __free_page(struct se_mem->se_page) when
+ * SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC is NOT in use,
+ */
+ if (free_page)
+ __free_page(se_mem->se_page);
+
+ list_del(&se_mem->se_list);
+ kmem_cache_free(se_mem_cache, se_mem);
+ }
+
+ if (T_TASK(cmd)->t_mem_bidi_list && T_TASK(cmd)->t_tasks_se_bidi_num) {
+ list_for_each_entry_safe(se_mem, se_mem_tmp,
+ T_TASK(cmd)->t_mem_bidi_list, se_list) {
+ /*
+ * We only release call __free_page(struct se_mem->se_page) when
+ * SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC is NOT in use,
+ */
+ if (free_page)
+ __free_page(se_mem->se_page);
+
+ list_del(&se_mem->se_list);
+ kmem_cache_free(se_mem_cache, se_mem);
+ }
+ }
+
+ kfree(T_TASK(cmd)->t_mem_bidi_list);
+ T_TASK(cmd)->t_mem_bidi_list = NULL;
+ kfree(T_TASK(cmd)->t_mem_list);
+ T_TASK(cmd)->t_mem_list = NULL;
+ T_TASK(cmd)->t_tasks_se_num = 0;
+}
+
+static inline void transport_release_tasks(struct se_cmd *cmd)
+{
+ transport_free_dev_tasks(cmd);
+}
+
+static inline int transport_dec_and_check(struct se_cmd *cmd)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (atomic_read(&T_TASK(cmd)->t_fe_count)) {
+ if (!(atomic_dec_and_test(&T_TASK(cmd)->t_fe_count))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock,
+ flags);
+ return 1;
+ }
+ }
+
+ if (atomic_read(&T_TASK(cmd)->t_se_count)) {
+ if (!(atomic_dec_and_test(&T_TASK(cmd)->t_se_count))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock,
+ flags);
+ return 1;
+ }
+ }
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ return 0;
+}
+
+static void transport_release_fe_cmd(struct se_cmd *cmd)
+{
+ unsigned long flags;
+
+ if (transport_dec_and_check(cmd))
+ return;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (!(atomic_read(&T_TASK(cmd)->transport_dev_active))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ goto free_pages;
+ }
+ atomic_set(&T_TASK(cmd)->transport_dev_active, 0);
+ transport_all_task_dev_remove_state(cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_release_tasks(cmd);
+free_pages:
+ transport_free_pages(cmd);
+ transport_free_se_cmd(cmd);
+ CMD_TFO(cmd)->release_cmd_direct(cmd);
+}
+
+static int transport_generic_remove(
+ struct se_cmd *cmd,
+ int release_to_pool,
+ int session_reinstatement)
+{
+ unsigned long flags;
+
+ if (!(T_TASK(cmd)))
+ goto release_cmd;
+
+ if (transport_dec_and_check(cmd)) {
+ if (session_reinstatement) {
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ transport_all_task_dev_remove_state(cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock,
+ flags);
+ }
+ return 1;
+ }
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (!(atomic_read(&T_TASK(cmd)->transport_dev_active))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ goto free_pages;
+ }
+ atomic_set(&T_TASK(cmd)->transport_dev_active, 0);
+ transport_all_task_dev_remove_state(cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_release_tasks(cmd);
+free_pages:
+ transport_free_pages(cmd);
+
+release_cmd:
+ if (release_to_pool) {
+ transport_release_cmd_to_pool(cmd);
+ } else {
+ transport_free_se_cmd(cmd);
+ CMD_TFO(cmd)->release_cmd_direct(cmd);
+ }
+
+ return 0;
+}
+
+/*
+ * transport_generic_map_mem_to_cmd - Perform SGL -> struct se_mem map
+ * @cmd: Associated se_cmd descriptor
+ * @mem: SGL style memory for TCM WRITE / READ
+ * @sg_mem_num: Number of SGL elements
+ * @mem_bidi_in: SGL style memory for TCM BIDI READ
+ * @sg_mem_bidi_num: Number of BIDI READ SGL elements
+ *
+ * Return: nonzero return cmd was rejected for -ENOMEM or inproper usage
+ * of parameters.
+ */
+int transport_generic_map_mem_to_cmd(
+ struct se_cmd *cmd,
+ struct scatterlist *mem,
+ u32 sg_mem_num,
+ struct scatterlist *mem_bidi_in,
+ u32 sg_mem_bidi_num)
+{
+ u32 se_mem_cnt_out = 0;
+ int ret;
+
+ if (!(mem) || !(sg_mem_num))
+ return 0;
+ /*
+ * Passed *mem will contain a list_head containing preformatted
+ * struct se_mem elements...
+ */
+ if (!(cmd->se_cmd_flags & SCF_PASSTHROUGH_SG_TO_MEM)) {
+ if ((mem_bidi_in) || (sg_mem_bidi_num)) {
+ printk(KERN_ERR "SCF_CMD_PASSTHROUGH_NOALLOC not supported"
+ " with BIDI-COMMAND\n");
+ return -ENOSYS;
+ }
+
+ T_TASK(cmd)->t_mem_list = (struct list_head *)mem;
+ T_TASK(cmd)->t_tasks_se_num = sg_mem_num;
+ cmd->se_cmd_flags |= SCF_CMD_PASSTHROUGH_NOALLOC;
+ return 0;
+ }
+ /*
+ * Otherwise, assume the caller is passing a struct scatterlist
+ * array from include/linux/scatterlist.h
+ */
+ if ((cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB) ||
+ (cmd->se_cmd_flags & SCF_SCSI_CONTROL_SG_IO_CDB)) {
+ /*
+ * For CDB using TCM struct se_mem linked list scatterlist memory
+ * processed into a TCM struct se_subsystem_dev, we do the mapping
+ * from the passed physical memory to struct se_mem->se_page here.
+ */
+ T_TASK(cmd)->t_mem_list = transport_init_se_mem_list();
+ if (!(T_TASK(cmd)->t_mem_list))
+ return -ENOMEM;
+
+ ret = transport_map_sg_to_mem(cmd,
+ T_TASK(cmd)->t_mem_list, mem, &se_mem_cnt_out);
+ if (ret < 0)
+ return -ENOMEM;
+
+ T_TASK(cmd)->t_tasks_se_num = se_mem_cnt_out;
+ /*
+ * Setup BIDI READ list of struct se_mem elements
+ */
+ if ((mem_bidi_in) && (sg_mem_bidi_num)) {
+ T_TASK(cmd)->t_mem_bidi_list = transport_init_se_mem_list();
+ if (!(T_TASK(cmd)->t_mem_bidi_list)) {
+ kfree(T_TASK(cmd)->t_mem_list);
+ return -ENOMEM;
+ }
+ se_mem_cnt_out = 0;
+
+ ret = transport_map_sg_to_mem(cmd,
+ T_TASK(cmd)->t_mem_bidi_list, mem_bidi_in,
+ &se_mem_cnt_out);
+ if (ret < 0) {
+ kfree(T_TASK(cmd)->t_mem_list);
+ return -ENOMEM;
+ }
+
+ T_TASK(cmd)->t_tasks_se_bidi_num = se_mem_cnt_out;
+ }
+ cmd->se_cmd_flags |= SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC;
+
+ } else if (cmd->se_cmd_flags & SCF_SCSI_CONTROL_NONSG_IO_CDB) {
+ if (mem_bidi_in || sg_mem_bidi_num) {
+ printk(KERN_ERR "BIDI-Commands not supported using "
+ "SCF_SCSI_CONTROL_NONSG_IO_CDB\n");
+ return -ENOSYS;
+ }
+ /*
+ * For incoming CDBs using a contiguous buffer internall with TCM,
+ * save the passed struct scatterlist memory. After TCM storage object
+ * processing has completed for this struct se_cmd, TCM core will call
+ * transport_memcpy_[write,read]_contig() as necessary from
+ * transport_generic_complete_ok() and transport_write_pending() in order
+ * to copy the TCM buffer to/from the original passed *mem in SGL ->
+ * struct scatterlist format.
+ */
+ cmd->se_cmd_flags |= SCF_PASSTHROUGH_CONTIG_TO_SG;
+ T_TASK(cmd)->t_task_pt_sgl = mem;
+ }
+
+ return 0;
+}
+EXPORT_SYMBOL(transport_generic_map_mem_to_cmd);
+
+
+static inline long long transport_dev_end_lba(struct se_device *dev)
+{
+ return dev->transport->get_blocks(dev) + 1;
+}
+
+static int transport_get_sectors(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+
+ T_TASK(cmd)->t_tasks_sectors =
+ (cmd->data_length / DEV_ATTRIB(dev)->block_size);
+ if (!(T_TASK(cmd)->t_tasks_sectors))
+ T_TASK(cmd)->t_tasks_sectors = 1;
+
+ if (TRANSPORT(dev)->get_device_type(dev) != TYPE_DISK)
+ return 0;
+
+ if ((T_TASK(cmd)->t_task_lba + T_TASK(cmd)->t_tasks_sectors) >
+ transport_dev_end_lba(dev)) {
+ printk(KERN_ERR "LBA: %llu Sectors: %u exceeds"
+ " transport_dev_end_lba(): %llu\n",
+ T_TASK(cmd)->t_task_lba, T_TASK(cmd)->t_tasks_sectors,
+ transport_dev_end_lba(dev));
+ cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ cmd->scsi_sense_reason = TCM_SECTOR_COUNT_TOO_MANY;
+ return PYX_TRANSPORT_REQ_TOO_MANY_SECTORS;
+ }
+
+ return 0;
+}
+
+static int transport_new_cmd_obj(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ u32 task_cdbs = 0, rc;
+
+ if (!(cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB)) {
+ task_cdbs++;
+ T_TASK(cmd)->t_task_cdbs++;
+ } else {
+ int set_counts = 1;
+
+ /*
+ * Setup any BIDI READ tasks and memory from
+ * T_TASK(cmd)->t_mem_bidi_list so the READ struct se_tasks
+ * are queued first for the non pSCSI passthrough case.
+ */
+ if ((T_TASK(cmd)->t_mem_bidi_list != NULL) &&
+ (TRANSPORT(dev)->transport_type != TRANSPORT_PLUGIN_PHBA_PDEV)) {
+ rc = transport_generic_get_cdb_count(cmd,
+ T_TASK(cmd)->t_task_lba,
+ T_TASK(cmd)->t_tasks_sectors,
+ DMA_FROM_DEVICE, T_TASK(cmd)->t_mem_bidi_list,
+ set_counts);
+ if (!(rc)) {
+ cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ cmd->scsi_sense_reason =
+ TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ set_counts = 0;
+ }
+ /*
+ * Setup the tasks and memory from T_TASK(cmd)->t_mem_list
+ * Note for BIDI transfers this will contain the WRITE payload
+ */
+ task_cdbs = transport_generic_get_cdb_count(cmd,
+ T_TASK(cmd)->t_task_lba,
+ T_TASK(cmd)->t_tasks_sectors,
+ cmd->data_direction, T_TASK(cmd)->t_mem_list,
+ set_counts);
+ if (!(task_cdbs)) {
+ cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION;
+ cmd->scsi_sense_reason =
+ TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
+ return PYX_TRANSPORT_LU_COMM_FAILURE;
+ }
+ T_TASK(cmd)->t_task_cdbs += task_cdbs;
+
+#if 0
+ printk(KERN_INFO "data_length: %u, LBA: %llu t_tasks_sectors:"
+ " %u, t_task_cdbs: %u\n", obj_ptr, cmd->data_length,
+ T_TASK(cmd)->t_task_lba, T_TASK(cmd)->t_tasks_sectors,
+ T_TASK(cmd)->t_task_cdbs);
+#endif
+ }
+
+ atomic_set(&T_TASK(cmd)->t_task_cdbs_left, task_cdbs);
+ atomic_set(&T_TASK(cmd)->t_task_cdbs_ex_left, task_cdbs);
+ atomic_set(&T_TASK(cmd)->t_task_cdbs_timeout_left, task_cdbs);
+ return 0;
+}
+
+static struct list_head *transport_init_se_mem_list(void)
+{
+ struct list_head *se_mem_list;
+
+ se_mem_list = kzalloc(sizeof(struct list_head), GFP_KERNEL);
+ if (!(se_mem_list)) {
+ printk(KERN_ERR "Unable to allocate memory for se_mem_list\n");
+ return NULL;
+ }
+ INIT_LIST_HEAD(se_mem_list);
+
+ return se_mem_list;
+}
+
+static int
+transport_generic_get_mem(struct se_cmd *cmd, u32 length, u32 dma_size)
+{
+ unsigned char *buf;
+ struct se_mem *se_mem;
+
+ T_TASK(cmd)->t_mem_list = transport_init_se_mem_list();
+ if (!(T_TASK(cmd)->t_mem_list))
+ return -ENOMEM;
+
+ /*
+ * If the device uses memory mapping this is enough.
+ */
+ if (cmd->se_dev->transport->do_se_mem_map)
+ return 0;
+
+ /*
+ * Setup BIDI-COMMAND READ list of struct se_mem elements
+ */
+ if (T_TASK(cmd)->t_tasks_bidi) {
+ T_TASK(cmd)->t_mem_bidi_list = transport_init_se_mem_list();
+ if (!(T_TASK(cmd)->t_mem_bidi_list)) {
+ kfree(T_TASK(cmd)->t_mem_list);
+ return -ENOMEM;
+ }
+ }
+
+ while (length) {
+ se_mem = kmem_cache_zalloc(se_mem_cache, GFP_KERNEL);
+ if (!(se_mem)) {
+ printk(KERN_ERR "Unable to allocate struct se_mem\n");
+ goto out;
+ }
+ INIT_LIST_HEAD(&se_mem->se_list);
+ se_mem->se_len = (length > dma_size) ? dma_size : length;
+
+/* #warning FIXME Allocate contigous pages for struct se_mem elements */
+ se_mem->se_page = (struct page *) alloc_pages(GFP_KERNEL, 0);
+ if (!(se_mem->se_page)) {
+ printk(KERN_ERR "alloc_pages() failed\n");
+ goto out;
+ }
+
+ buf = kmap_atomic(se_mem->se_page, KM_IRQ0);
+ if (!(buf)) {
+ printk(KERN_ERR "kmap_atomic() failed\n");
+ goto out;
+ }
+ memset(buf, 0, se_mem->se_len);
+ kunmap_atomic(buf, KM_IRQ0);
+
+ list_add_tail(&se_mem->se_list, T_TASK(cmd)->t_mem_list);
+ T_TASK(cmd)->t_tasks_se_num++;
+
+ DEBUG_MEM("Allocated struct se_mem page(%p) Length(%u)"
+ " Offset(%u)\n", se_mem->se_page, se_mem->se_len,
+ se_mem->se_off);
+
+ length -= se_mem->se_len;
+ }
+
+ DEBUG_MEM("Allocated total struct se_mem elements(%u)\n",
+ T_TASK(cmd)->t_tasks_se_num);
+
+ return 0;
+out:
+ return -1;
+}
+
+extern u32 transport_calc_sg_num(
+ struct se_task *task,
+ struct se_mem *in_se_mem,
+ u32 task_offset)
+{
+ struct se_cmd *se_cmd = task->task_se_cmd;
+ struct se_device *se_dev = SE_DEV(se_cmd);
+ struct se_mem *se_mem = in_se_mem;
+ struct target_core_fabric_ops *tfo = CMD_TFO(se_cmd);
+ u32 sg_length, task_size = task->task_size, task_sg_num_padded;
+
+ while (task_size != 0) {
+ DEBUG_SC("se_mem->se_page(%p) se_mem->se_len(%u)"
+ " se_mem->se_off(%u) task_offset(%u)\n",
+ se_mem->se_page, se_mem->se_len,
+ se_mem->se_off, task_offset);
+
+ if (task_offset == 0) {
+ if (task_size >= se_mem->se_len) {
+ sg_length = se_mem->se_len;
+
+ if (!(list_is_last(&se_mem->se_list,
+ T_TASK(se_cmd)->t_mem_list)))
+ se_mem = list_entry(se_mem->se_list.next,
+ struct se_mem, se_list);
+ } else {
+ sg_length = task_size;
+ task_size -= sg_length;
+ goto next;
+ }
+
+ DEBUG_SC("sg_length(%u) task_size(%u)\n",
+ sg_length, task_size);
+ } else {
+ if ((se_mem->se_len - task_offset) > task_size) {
+ sg_length = task_size;
+ task_size -= sg_length;
+ goto next;
+ } else {
+ sg_length = (se_mem->se_len - task_offset);
+
+ if (!(list_is_last(&se_mem->se_list,
+ T_TASK(se_cmd)->t_mem_list)))
+ se_mem = list_entry(se_mem->se_list.next,
+ struct se_mem, se_list);
+ }
+
+ DEBUG_SC("sg_length(%u) task_size(%u)\n",
+ sg_length, task_size);
+
+ task_offset = 0;
+ }
+ task_size -= sg_length;
+next:
+ DEBUG_SC("task[%u] - Reducing task_size to(%u)\n",
+ task->task_no, task_size);
+
+ task->task_sg_num++;
+ }
+ /*
+ * Check if the fabric module driver is requesting that all
+ * struct se_task->task_sg[] be chained together.. If so,
+ * then allocate an extra padding SG entry for linking and
+ * marking the end of the chained SGL.
+ */
+ if (tfo->task_sg_chaining) {
+ task_sg_num_padded = (task->task_sg_num + 1);
+ task->task_padded_sg = 1;
+ } else
+ task_sg_num_padded = task->task_sg_num;
+
+ task->task_sg = kzalloc(task_sg_num_padded *
+ sizeof(struct scatterlist), GFP_KERNEL);
+ if (!(task->task_sg)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " task->task_sg\n");
+ return 0;
+ }
+ sg_init_table(&task->task_sg[0], task_sg_num_padded);
+ /*
+ * Setup task->task_sg_bidi for SCSI READ payload for
+ * TCM/pSCSI passthrough if present for BIDI-COMMAND
+ */
+ if ((T_TASK(se_cmd)->t_mem_bidi_list != NULL) &&
+ (TRANSPORT(se_dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV)) {
+ task->task_sg_bidi = kzalloc(task_sg_num_padded *
+ sizeof(struct scatterlist), GFP_KERNEL);
+ if (!(task->task_sg_bidi)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " task->task_sg_bidi\n");
+ return 0;
+ }
+ sg_init_table(&task->task_sg_bidi[0], task_sg_num_padded);
+ }
+ /*
+ * For the chaining case, setup the proper end of SGL for the
+ * initial submission struct task into struct se_subsystem_api.
+ * This will be cleared later by transport_do_task_sg_chain()
+ */
+ if (task->task_padded_sg) {
+ sg_mark_end(&task->task_sg[task->task_sg_num - 1]);
+ /*
+ * Added the 'if' check before marking end of bi-directional
+ * scatterlist (which gets created only in case of request
+ * (RD + WR).
+ */
+ if (task->task_sg_bidi)
+ sg_mark_end(&task->task_sg_bidi[task->task_sg_num - 1]);
+ }
+
+ DEBUG_SC("Successfully allocated task->task_sg_num(%u),"
+ " task_sg_num_padded(%u)\n", task->task_sg_num,
+ task_sg_num_padded);
+
+ return task->task_sg_num;
+}
+
+static inline int transport_set_tasks_sectors_disk(
+ struct se_task *task,
+ struct se_device *dev,
+ unsigned long long lba,
+ u32 sectors,
+ int *max_sectors_set)
+{
+ if ((lba + sectors) > transport_dev_end_lba(dev)) {
+ task->task_sectors = ((transport_dev_end_lba(dev) - lba) + 1);
+
+ if (task->task_sectors > DEV_ATTRIB(dev)->max_sectors) {
+ task->task_sectors = DEV_ATTRIB(dev)->max_sectors;
+ *max_sectors_set = 1;
+ }
+ } else {
+ if (sectors > DEV_ATTRIB(dev)->max_sectors) {
+ task->task_sectors = DEV_ATTRIB(dev)->max_sectors;
+ *max_sectors_set = 1;
+ } else
+ task->task_sectors = sectors;
+ }
+
+ return 0;
+}
+
+static inline int transport_set_tasks_sectors_non_disk(
+ struct se_task *task,
+ struct se_device *dev,
+ unsigned long long lba,
+ u32 sectors,
+ int *max_sectors_set)
+{
+ if (sectors > DEV_ATTRIB(dev)->max_sectors) {
+ task->task_sectors = DEV_ATTRIB(dev)->max_sectors;
+ *max_sectors_set = 1;
+ } else
+ task->task_sectors = sectors;
+
+ return 0;
+}
+
+static inline int transport_set_tasks_sectors(
+ struct se_task *task,
+ struct se_device *dev,
+ unsigned long long lba,
+ u32 sectors,
+ int *max_sectors_set)
+{
+ return (TRANSPORT(dev)->get_device_type(dev) == TYPE_DISK) ?
+ transport_set_tasks_sectors_disk(task, dev, lba, sectors,
+ max_sectors_set) :
+ transport_set_tasks_sectors_non_disk(task, dev, lba, sectors,
+ max_sectors_set);
+}
+
+static int transport_map_sg_to_mem(
+ struct se_cmd *cmd,
+ struct list_head *se_mem_list,
+ void *in_mem,
+ u32 *se_mem_cnt)
+{
+ struct se_mem *se_mem;
+ struct scatterlist *sg;
+ u32 sg_count = 1, cmd_size = cmd->data_length;
+
+ if (!in_mem) {
+ printk(KERN_ERR "No source scatterlist\n");
+ return -1;
+ }
+ sg = (struct scatterlist *)in_mem;
+
+ while (cmd_size) {
+ se_mem = kmem_cache_zalloc(se_mem_cache, GFP_KERNEL);
+ if (!(se_mem)) {
+ printk(KERN_ERR "Unable to allocate struct se_mem\n");
+ return -1;
+ }
+ INIT_LIST_HEAD(&se_mem->se_list);
+ DEBUG_MEM("sg_to_mem: Starting loop with cmd_size: %u"
+ " sg_page: %p offset: %d length: %d\n", cmd_size,
+ sg_page(sg), sg->offset, sg->length);
+
+ se_mem->se_page = sg_page(sg);
+ se_mem->se_off = sg->offset;
+
+ if (cmd_size > sg->length) {
+ se_mem->se_len = sg->length;
+ sg = sg_next(sg);
+ sg_count++;
+ } else
+ se_mem->se_len = cmd_size;
+
+ cmd_size -= se_mem->se_len;
+
+ DEBUG_MEM("sg_to_mem: *se_mem_cnt: %u cmd_size: %u\n",
+ *se_mem_cnt, cmd_size);
+ DEBUG_MEM("sg_to_mem: Final se_page: %p se_off: %d se_len: %d\n",
+ se_mem->se_page, se_mem->se_off, se_mem->se_len);
+
+ list_add_tail(&se_mem->se_list, se_mem_list);
+ (*se_mem_cnt)++;
+ }
+
+ DEBUG_MEM("task[0] - Mapped(%u) struct scatterlist segments to(%u)"
+ " struct se_mem\n", sg_count, *se_mem_cnt);
+
+ if (sg_count != *se_mem_cnt)
+ BUG();
+
+ return 0;
+}
+
+/* transport_map_mem_to_sg():
+ *
+ *
+ */
+int transport_map_mem_to_sg(
+ struct se_task *task,
+ struct list_head *se_mem_list,
+ void *in_mem,
+ struct se_mem *in_se_mem,
+ struct se_mem **out_se_mem,
+ u32 *se_mem_cnt,
+ u32 *task_offset)
+{
+ struct se_cmd *se_cmd = task->task_se_cmd;
+ struct se_mem *se_mem = in_se_mem;
+ struct scatterlist *sg = (struct scatterlist *)in_mem;
+ u32 task_size = task->task_size, sg_no = 0;
+
+ if (!sg) {
+ printk(KERN_ERR "Unable to locate valid struct"
+ " scatterlist pointer\n");
+ return -1;
+ }
+
+ while (task_size != 0) {
+ /*
+ * Setup the contigious array of scatterlists for
+ * this struct se_task.
+ */
+ sg_assign_page(sg, se_mem->se_page);
+
+ if (*task_offset == 0) {
+ sg->offset = se_mem->se_off;
+
+ if (task_size >= se_mem->se_len) {
+ sg->length = se_mem->se_len;
+
+ if (!(list_is_last(&se_mem->se_list,
+ T_TASK(se_cmd)->t_mem_list))) {
+ se_mem = list_entry(se_mem->se_list.next,
+ struct se_mem, se_list);
+ (*se_mem_cnt)++;
+ }
+ } else {
+ sg->length = task_size;
+ /*
+ * Determine if we need to calculate an offset
+ * into the struct se_mem on the next go around..
+ */
+ task_size -= sg->length;
+ if (!(task_size))
+ *task_offset = sg->length;
+
+ goto next;
+ }
+
+ } else {
+ sg->offset = (*task_offset + se_mem->se_off);
+
+ if ((se_mem->se_len - *task_offset) > task_size) {
+ sg->length = task_size;
+ /*
+ * Determine if we need to calculate an offset
+ * into the struct se_mem on the next go around..
+ */
+ task_size -= sg->length;
+ if (!(task_size))
+ *task_offset += sg->length;
+
+ goto next;
+ } else {
+ sg->length = (se_mem->se_len - *task_offset);
+
+ if (!(list_is_last(&se_mem->se_list,
+ T_TASK(se_cmd)->t_mem_list))) {
+ se_mem = list_entry(se_mem->se_list.next,
+ struct se_mem, se_list);
+ (*se_mem_cnt)++;
+ }
+ }
+
+ *task_offset = 0;
+ }
+ task_size -= sg->length;
+next:
+ DEBUG_MEM("task[%u] mem_to_sg - sg[%u](%p)(%u)(%u) - Reducing"
+ " task_size to(%u), task_offset: %u\n", task->task_no, sg_no,
+ sg_page(sg), sg->length, sg->offset, task_size, *task_offset);
+
+ sg_no++;
+ if (!(task_size))
+ break;
+
+ sg = sg_next(sg);
+
+ if (task_size > se_cmd->data_length)
+ BUG();
+ }
+ *out_se_mem = se_mem;
+
+ DEBUG_MEM("task[%u] - Mapped(%u) struct se_mem segments to total(%u)"
+ " SGs\n", task->task_no, *se_mem_cnt, sg_no);
+
+ return 0;
+}
+
+/*
+ * This function can be used by HW target mode drivers to create a linked
+ * scatterlist from all contiguously allocated struct se_task->task_sg[].
+ * This is intended to be called during the completion path by TCM Core
+ * when struct target_core_fabric_ops->check_task_sg_chaining is enabled.
+ */
+void transport_do_task_sg_chain(struct se_cmd *cmd)
+{
+ struct scatterlist *sg_head = NULL, *sg_link = NULL, *sg_first = NULL;
+ struct scatterlist *sg_head_cur = NULL, *sg_link_cur = NULL;
+ struct scatterlist *sg, *sg_end = NULL, *sg_end_cur = NULL;
+ struct se_task *task;
+ struct target_core_fabric_ops *tfo = CMD_TFO(cmd);
+ u32 task_sg_num = 0, sg_count = 0;
+ int i;
+
+ if (tfo->task_sg_chaining == 0) {
+ printk(KERN_ERR "task_sg_chaining is diabled for fabric module:"
+ " %s\n", tfo->get_fabric_name());
+ dump_stack();
+ return;
+ }
+ /*
+ * Walk the struct se_task list and setup scatterlist chains
+ * for each contiguosly allocated struct se_task->task_sg[].
+ */
+ list_for_each_entry(task, &T_TASK(cmd)->t_task_list, t_list) {
+ if (!(task->task_sg) || !(task->task_padded_sg))
+ continue;
+
+ if (sg_head && sg_link) {
+ sg_head_cur = &task->task_sg[0];
+ sg_link_cur = &task->task_sg[task->task_sg_num];
+ /*
+ * Either add chain or mark end of scatterlist
+ */
+ if (!(list_is_last(&task->t_list,
+ &T_TASK(cmd)->t_task_list))) {
+ /*
+ * Clear existing SGL termination bit set in
+ * transport_calc_sg_num(), see sg_mark_end()
+ */
+ sg_end_cur = &task->task_sg[task->task_sg_num - 1];
+ sg_end_cur->page_link &= ~0x02;
+
+ sg_chain(sg_head, task_sg_num, sg_head_cur);
+ sg_count += (task->task_sg_num + 1);
+ } else
+ sg_count += task->task_sg_num;
+
+ sg_head = sg_head_cur;
+ sg_link = sg_link_cur;
+ task_sg_num = task->task_sg_num;
+ continue;
+ }
+ sg_head = sg_first = &task->task_sg[0];
+ sg_link = &task->task_sg[task->task_sg_num];
+ task_sg_num = task->task_sg_num;
+ /*
+ * Check for single task..
+ */
+ if (!(list_is_last(&task->t_list, &T_TASK(cmd)->t_task_list))) {
+ /*
+ * Clear existing SGL termination bit set in
+ * transport_calc_sg_num(), see sg_mark_end()
+ */
+ sg_end = &task->task_sg[task->task_sg_num - 1];
+ sg_end->page_link &= ~0x02;
+ sg_count += (task->task_sg_num + 1);
+ } else
+ sg_count += task->task_sg_num;
+ }
+ /*
+ * Setup the starting pointer and total t_tasks_sg_linked_no including
+ * padding SGs for linking and to mark the end.
+ */
+ T_TASK(cmd)->t_tasks_sg_chained = sg_first;
+ T_TASK(cmd)->t_tasks_sg_chained_no = sg_count;
+
+ DEBUG_CMD_M("Setup T_TASK(cmd)->t_tasks_sg_chained: %p and"
+ " t_tasks_sg_chained_no: %u\n", T_TASK(cmd)->t_tasks_sg_chained,
+ T_TASK(cmd)->t_tasks_sg_chained_no);
+
+ for_each_sg(T_TASK(cmd)->t_tasks_sg_chained, sg,
+ T_TASK(cmd)->t_tasks_sg_chained_no, i) {
+
+ DEBUG_CMD_M("SG: %p page: %p length: %d offset: %d\n",
+ sg, sg_page(sg), sg->length, sg->offset);
+ if (sg_is_chain(sg))
+ DEBUG_CMD_M("SG: %p sg_is_chain=1\n", sg);
+ if (sg_is_last(sg))
+ DEBUG_CMD_M("SG: %p sg_is_last=1\n", sg);
+ }
+
+}
+EXPORT_SYMBOL(transport_do_task_sg_chain);
+
+static int transport_do_se_mem_map(
+ struct se_device *dev,
+ struct se_task *task,
+ struct list_head *se_mem_list,
+ void *in_mem,
+ struct se_mem *in_se_mem,
+ struct se_mem **out_se_mem,
+ u32 *se_mem_cnt,
+ u32 *task_offset_in)
+{
+ u32 task_offset = *task_offset_in;
+ int ret = 0;
+ /*
+ * se_subsystem_api_t->do_se_mem_map is used when internal allocation
+ * has been done by the transport plugin.
+ */
+ if (TRANSPORT(dev)->do_se_mem_map) {
+ ret = TRANSPORT(dev)->do_se_mem_map(task, se_mem_list,
+ in_mem, in_se_mem, out_se_mem, se_mem_cnt,
+ task_offset_in);
+ if (ret == 0)
+ T_TASK(task->task_se_cmd)->t_tasks_se_num += *se_mem_cnt;
+
+ return ret;
+ }
+ /*
+ * This is the normal path for all normal non BIDI and BIDI-COMMAND
+ * WRITE payloads.. If we need to do BIDI READ passthrough for
+ * TCM/pSCSI the first call to transport_do_se_mem_map ->
+ * transport_calc_sg_num() -> transport_map_mem_to_sg() will do the
+ * allocation for task->task_sg_bidi, and the subsequent call to
+ * transport_do_se_mem_map() from transport_generic_get_cdb_count()
+ */
+ if (!(task->task_sg_bidi)) {
+ /*
+ * Assume default that transport plugin speaks preallocated
+ * scatterlists.
+ */
+ if (!(transport_calc_sg_num(task, in_se_mem, task_offset)))
+ return -1;
+ /*
+ * struct se_task->task_sg now contains the struct scatterlist array.
+ */
+ return transport_map_mem_to_sg(task, se_mem_list, task->task_sg,
+ in_se_mem, out_se_mem, se_mem_cnt,
+ task_offset_in);
+ }
+ /*
+ * Handle the se_mem_list -> struct task->task_sg_bidi
+ * memory map for the extra BIDI READ payload
+ */
+ return transport_map_mem_to_sg(task, se_mem_list, task->task_sg_bidi,
+ in_se_mem, out_se_mem, se_mem_cnt,
+ task_offset_in);
+}
+
+static u32 transport_generic_get_cdb_count(
+ struct se_cmd *cmd,
+ unsigned long long lba,
+ u32 sectors,
+ enum dma_data_direction data_direction,
+ struct list_head *mem_list,
+ int set_counts)
+{
+ unsigned char *cdb = NULL;
+ struct se_task *task;
+ struct se_mem *se_mem = NULL, *se_mem_lout = NULL;
+ struct se_mem *se_mem_bidi = NULL, *se_mem_bidi_lout = NULL;
+ struct se_device *dev = SE_DEV(cmd);
+ int max_sectors_set = 0, ret;
+ u32 task_offset_in = 0, se_mem_cnt = 0, se_mem_bidi_cnt = 0, task_cdbs = 0;
+
+ if (!mem_list) {
+ printk(KERN_ERR "mem_list is NULL in transport_generic_get"
+ "_cdb_count()\n");
+ return 0;
+ }
+ /*
+ * While using RAMDISK_DR backstores is the only case where
+ * mem_list will ever be empty at this point.
+ */
+ if (!(list_empty(mem_list)))
+ se_mem = list_entry(mem_list->next, struct se_mem, se_list);
+ /*
+ * Check for extra se_mem_bidi mapping for BIDI-COMMANDs to
+ * struct se_task->task_sg_bidi for TCM/pSCSI passthrough operation
+ */
+ if ((T_TASK(cmd)->t_mem_bidi_list != NULL) &&
+ !(list_empty(T_TASK(cmd)->t_mem_bidi_list)) &&
+ (TRANSPORT(dev)->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV))
+ se_mem_bidi = list_entry(T_TASK(cmd)->t_mem_bidi_list->next,
+ struct se_mem, se_list);
+
+ while (sectors) {
+ DEBUG_VOL("ITT[0x%08x] LBA(%llu) SectorsLeft(%u) EOBJ(%llu)\n",
+ CMD_TFO(cmd)->get_task_tag(cmd), lba, sectors,
+ transport_dev_end_lba(dev));
+
+ task = transport_generic_get_task(cmd, data_direction);
+ if (!(task))
+ goto out;
+
+ transport_set_tasks_sectors(task, dev, lba, sectors,
+ &max_sectors_set);
+
+ task->task_lba = lba;
+ lba += task->task_sectors;
+ sectors -= task->task_sectors;
+ task->task_size = (task->task_sectors *
+ DEV_ATTRIB(dev)->block_size);
+
+ cdb = TRANSPORT(dev)->get_cdb(task);
+ if ((cdb)) {
+ memcpy(cdb, T_TASK(cmd)->t_task_cdb,
+ scsi_command_size(T_TASK(cmd)->t_task_cdb));
+ cmd->transport_split_cdb(task->task_lba,
+ &task->task_sectors, cdb);
+ }
+
+ /*
+ * Perform the SE OBJ plugin and/or Transport plugin specific
+ * mapping for T_TASK(cmd)->t_mem_list. And setup the
+ * task->task_sg and if necessary task->task_sg_bidi
+ */
+ ret = transport_do_se_mem_map(dev, task, mem_list,
+ NULL, se_mem, &se_mem_lout, &se_mem_cnt,
+ &task_offset_in);
+ if (ret < 0)
+ goto out;
+
+ se_mem = se_mem_lout;
+ /*
+ * Setup the T_TASK(cmd)->t_mem_bidi_list -> task->task_sg_bidi
+ * mapping for SCSI READ for BIDI-COMMAND passthrough with TCM/pSCSI
+ *
+ * Note that the first call to transport_do_se_mem_map() above will
+ * allocate struct se_task->task_sg_bidi in transport_do_se_mem_map()
+ * -> transport_calc_sg_num(), and the second here will do the
+ * mapping for SCSI READ for BIDI-COMMAND passthrough with TCM/pSCSI.
+ */
+ if (task->task_sg_bidi != NULL) {
+ ret = transport_do_se_mem_map(dev, task,
+ T_TASK(cmd)->t_mem_bidi_list, NULL,
+ se_mem_bidi, &se_mem_bidi_lout, &se_mem_bidi_cnt,
+ &task_offset_in);
+ if (ret < 0)
+ goto out;
+
+ se_mem_bidi = se_mem_bidi_lout;
+ }
+ task_cdbs++;
+
+ DEBUG_VOL("Incremented task_cdbs(%u) task->task_sg_num(%u)\n",
+ task_cdbs, task->task_sg_num);
+
+ if (max_sectors_set) {
+ max_sectors_set = 0;
+ continue;
+ }
+
+ if (!sectors)
+ break;
+ }
+
+ if (set_counts) {
+ atomic_inc(&T_TASK(cmd)->t_fe_count);
+ atomic_inc(&T_TASK(cmd)->t_se_count);
+ }
+
+ DEBUG_VOL("ITT[0x%08x] total %s cdbs(%u)\n",
+ CMD_TFO(cmd)->get_task_tag(cmd), (data_direction == DMA_TO_DEVICE)
+ ? "DMA_TO_DEVICE" : "DMA_FROM_DEVICE", task_cdbs);
+
+ return task_cdbs;
+out:
+ return 0;
+}
+
+static int
+transport_map_control_cmd_to_task(struct se_cmd *cmd)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ unsigned char *cdb;
+ struct se_task *task;
+ int ret;
+
+ task = transport_generic_get_task(cmd, cmd->data_direction);
+ if (!task)
+ return PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+
+ cdb = TRANSPORT(dev)->get_cdb(task);
+ if (cdb)
+ memcpy(cdb, cmd->t_task->t_task_cdb,
+ scsi_command_size(cmd->t_task->t_task_cdb));
+
+ task->task_size = cmd->data_length;
+ task->task_sg_num =
+ (cmd->se_cmd_flags & SCF_SCSI_CONTROL_SG_IO_CDB) ? 1 : 0;
+
+ atomic_inc(&cmd->t_task->t_fe_count);
+ atomic_inc(&cmd->t_task->t_se_count);
+
+ if (cmd->se_cmd_flags & SCF_SCSI_CONTROL_SG_IO_CDB) {
+ struct se_mem *se_mem = NULL, *se_mem_lout = NULL;
+ u32 se_mem_cnt = 0, task_offset = 0;
+
+ BUG_ON(list_empty(cmd->t_task->t_mem_list));
+
+ ret = transport_do_se_mem_map(dev, task,
+ cmd->t_task->t_mem_list, NULL, se_mem,
+ &se_mem_lout, &se_mem_cnt, &task_offset);
+ if (ret < 0)
+ return PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+
+ if (dev->transport->map_task_SG)
+ return dev->transport->map_task_SG(task);
+ return 0;
+ } else if (cmd->se_cmd_flags & SCF_SCSI_CONTROL_NONSG_IO_CDB) {
+ if (dev->transport->map_task_non_SG)
+ return dev->transport->map_task_non_SG(task);
+ return 0;
+ } else if (cmd->se_cmd_flags & SCF_SCSI_NON_DATA_CDB) {
+ if (dev->transport->cdb_none)
+ return dev->transport->cdb_none(task);
+ return 0;
+ } else {
+ BUG();
+ return PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+ }
+}
+
+/* transport_generic_new_cmd(): Called from transport_processing_thread()
+ *
+ * Allocate storage transport resources from a set of values predefined
+ * by transport_generic_cmd_sequencer() from the iSCSI Target RX process.
+ * Any non zero return here is treated as an "out of resource' op here.
+ */
+ /*
+ * Generate struct se_task(s) and/or their payloads for this CDB.
+ */
+static int transport_generic_new_cmd(struct se_cmd *cmd)
+{
+ struct se_portal_group *se_tpg;
+ struct se_task *task;
+ struct se_device *dev = SE_DEV(cmd);
+ int ret = 0;
+
+ /*
+ * Determine is the TCM fabric module has already allocated physical
+ * memory, and is directly calling transport_generic_map_mem_to_cmd()
+ * to setup beforehand the linked list of physical memory at
+ * T_TASK(cmd)->t_mem_list of struct se_mem->se_page
+ */
+ if (!(cmd->se_cmd_flags & SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC)) {
+ ret = transport_allocate_resources(cmd);
+ if (ret < 0)
+ return ret;
+ }
+
+ ret = transport_get_sectors(cmd);
+ if (ret < 0)
+ return ret;
+
+ ret = transport_new_cmd_obj(cmd);
+ if (ret < 0)
+ return ret;
+
+ /*
+ * Determine if the calling TCM fabric module is talking to
+ * Linux/NET via kernel sockets and needs to allocate a
+ * struct iovec array to complete the struct se_cmd
+ */
+ se_tpg = SE_LUN(cmd)->lun_sep->sep_tpg;
+ if (TPG_TFO(se_tpg)->alloc_cmd_iovecs != NULL) {
+ ret = TPG_TFO(se_tpg)->alloc_cmd_iovecs(cmd);
+ if (ret < 0)
+ return PYX_TRANSPORT_OUT_OF_MEMORY_RESOURCES;
+ }
+
+ if (cmd->se_cmd_flags & SCF_SCSI_DATA_SG_IO_CDB) {
+ list_for_each_entry(task, &T_TASK(cmd)->t_task_list, t_list) {
+ if (atomic_read(&task->task_sent))
+ continue;
+ if (!dev->transport->map_task_SG)
+ continue;
+
+ ret = dev->transport->map_task_SG(task);
+ if (ret < 0)
+ return ret;
+ }
+ } else {
+ ret = transport_map_control_cmd_to_task(cmd);
+ if (ret < 0)
+ return ret;
+ }
+
+ /*
+ * For WRITEs, let the iSCSI Target RX Thread know its buffer is ready..
+ * This WRITE struct se_cmd (and all of its associated struct se_task's)
+ * will be added to the struct se_device execution queue after its WRITE
+ * data has arrived. (ie: It gets handled by the transport processing
+ * thread a second time)
+ */
+ if (cmd->data_direction == DMA_TO_DEVICE) {
+ transport_add_tasks_to_state_queue(cmd);
+ return transport_generic_write_pending(cmd);
+ }
+ /*
+ * Everything else but a WRITE, add the struct se_cmd's struct se_task's
+ * to the execution queue.
+ */
+ transport_execute_tasks(cmd);
+ return 0;
+}
+
+/* transport_generic_process_write():
+ *
+ *
+ */
+void transport_generic_process_write(struct se_cmd *cmd)
+{
+#if 0
+ /*
+ * Copy SCSI Presented DTL sector(s) from received buffers allocated to
+ * original EDTL
+ */
+ if (cmd->se_cmd_flags & SCF_UNDERFLOW_BIT) {
+ if (!T_TASK(cmd)->t_tasks_se_num) {
+ unsigned char *dst, *buf =
+ (unsigned char *)T_TASK(cmd)->t_task_buf;
+
+ dst = kzalloc(cmd->cmd_spdtl), GFP_KERNEL);
+ if (!(dst)) {
+ printk(KERN_ERR "Unable to allocate memory for"
+ " WRITE underflow\n");
+ transport_generic_request_failure(cmd, NULL,
+ PYX_TRANSPORT_REQ_TOO_MANY_SECTORS, 1);
+ return;
+ }
+ memcpy(dst, buf, cmd->cmd_spdtl);
+
+ kfree(T_TASK(cmd)->t_task_buf);
+ T_TASK(cmd)->t_task_buf = dst;
+ } else {
+ struct scatterlist *sg =
+ (struct scatterlist *sg)T_TASK(cmd)->t_task_buf;
+ struct scatterlist *orig_sg;
+
+ orig_sg = kzalloc(sizeof(struct scatterlist) *
+ T_TASK(cmd)->t_tasks_se_num,
+ GFP_KERNEL))) {
+ if (!(orig_sg)) {
+ printk(KERN_ERR "Unable to allocate memory"
+ " for WRITE underflow\n");
+ transport_generic_request_failure(cmd, NULL,
+ PYX_TRANSPORT_REQ_TOO_MANY_SECTORS, 1);
+ return;
+ }
+
+ memcpy(orig_sg, T_TASK(cmd)->t_task_buf,
+ sizeof(struct scatterlist) *
+ T_TASK(cmd)->t_tasks_se_num);
+
+ cmd->data_length = cmd->cmd_spdtl;
+ /*
+ * FIXME, clear out original struct se_task and state
+ * information.
+ */
+ if (transport_generic_new_cmd(cmd) < 0) {
+ transport_generic_request_failure(cmd, NULL,
+ PYX_TRANSPORT_REQ_TOO_MANY_SECTORS, 1);
+ kfree(orig_sg);
+ return;
+ }
+
+ transport_memcpy_write_sg(cmd, orig_sg);
+ }
+ }
+#endif
+ transport_execute_tasks(cmd);
+}
+EXPORT_SYMBOL(transport_generic_process_write);
+
+/* transport_generic_write_pending():
+ *
+ *
+ */
+static int transport_generic_write_pending(struct se_cmd *cmd)
+{
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ cmd->t_state = TRANSPORT_WRITE_PENDING;
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ /*
+ * For the TCM control CDBs using a contiguous buffer, do the memcpy
+ * from the passed Linux/SCSI struct scatterlist located at
+ * T_TASK(se_cmd)->t_task_pt_buf to the contiguous buffer at
+ * T_TASK(se_cmd)->t_task_buf.
+ */
+ if (cmd->se_cmd_flags & SCF_PASSTHROUGH_CONTIG_TO_SG)
+ transport_memcpy_read_contig(cmd,
+ T_TASK(cmd)->t_task_buf,
+ T_TASK(cmd)->t_task_pt_sgl);
+ /*
+ * Clear the se_cmd for WRITE_PENDING status in order to set
+ * T_TASK(cmd)->t_transport_active=0 so that transport_generic_handle_data
+ * can be called from HW target mode interrupt code. This is safe
+ * to be called with transport_off=1 before the CMD_TFO(cmd)->write_pending
+ * because the se_cmd->se_lun pointer is not being cleared.
+ */
+ transport_cmd_check_stop(cmd, 1, 0);
+
+ /*
+ * Call the fabric write_pending function here to let the
+ * frontend know that WRITE buffers are ready.
+ */
+ ret = CMD_TFO(cmd)->write_pending(cmd);
+ if (ret < 0)
+ return ret;
+
+ return PYX_TRANSPORT_WRITE_PENDING;
+}
+
+/* transport_release_cmd_to_pool():
+ *
+ *
+ */
+void transport_release_cmd_to_pool(struct se_cmd *cmd)
+{
+ BUG_ON(!T_TASK(cmd));
+ BUG_ON(!CMD_TFO(cmd));
+
+ transport_free_se_cmd(cmd);
+ CMD_TFO(cmd)->release_cmd_to_pool(cmd);
+}
+EXPORT_SYMBOL(transport_release_cmd_to_pool);
+
+/* transport_generic_free_cmd():
+ *
+ * Called from processing frontend to release storage engine resources
+ */
+void transport_generic_free_cmd(
+ struct se_cmd *cmd,
+ int wait_for_tasks,
+ int release_to_pool,
+ int session_reinstatement)
+{
+ if (!(cmd->se_cmd_flags & SCF_SE_LUN_CMD) || !T_TASK(cmd))
+ transport_release_cmd_to_pool(cmd);
+ else {
+ core_dec_lacl_count(cmd->se_sess->se_node_acl, cmd);
+
+ if (SE_LUN(cmd)) {
+#if 0
+ printk(KERN_INFO "cmd: %p ITT: 0x%08x contains"
+ " SE_LUN(cmd)\n", cmd,
+ CMD_TFO(cmd)->get_task_tag(cmd));
+#endif
+ transport_lun_remove_cmd(cmd);
+ }
+
+ if (wait_for_tasks && cmd->transport_wait_for_tasks)
+ cmd->transport_wait_for_tasks(cmd, 0, 0);
+
+ transport_generic_remove(cmd, release_to_pool,
+ session_reinstatement);
+ }
+}
+EXPORT_SYMBOL(transport_generic_free_cmd);
+
+static void transport_nop_wait_for_tasks(
+ struct se_cmd *cmd,
+ int remove_cmd,
+ int session_reinstatement)
+{
+ return;
+}
+
+/* transport_lun_wait_for_tasks():
+ *
+ * Called from ConfigFS context to stop the passed struct se_cmd to allow
+ * an struct se_lun to be successfully shutdown.
+ */
+static int transport_lun_wait_for_tasks(struct se_cmd *cmd, struct se_lun *lun)
+{
+ unsigned long flags;
+ int ret;
+ /*
+ * If the frontend has already requested this struct se_cmd to
+ * be stopped, we can safely ignore this struct se_cmd.
+ */
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (atomic_read(&T_TASK(cmd)->t_transport_stop)) {
+ atomic_set(&T_TASK(cmd)->transport_lun_stop, 0);
+ DEBUG_TRANSPORT_S("ConfigFS ITT[0x%08x] - t_transport_stop =="
+ " TRUE, skipping\n", CMD_TFO(cmd)->get_task_tag(cmd));
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ transport_cmd_check_stop(cmd, 1, 0);
+ return -1;
+ }
+ atomic_set(&T_TASK(cmd)->transport_lun_fe_stop, 1);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ wake_up_interruptible(&SE_DEV(cmd)->dev_queue_obj->thread_wq);
+
+ ret = transport_stop_tasks_for_cmd(cmd);
+
+ DEBUG_TRANSPORT_S("ConfigFS: cmd: %p t_task_cdbs: %d stop tasks ret:"
+ " %d\n", cmd, T_TASK(cmd)->t_task_cdbs, ret);
+ if (!ret) {
+ DEBUG_TRANSPORT_S("ConfigFS: ITT[0x%08x] - stopping cmd....\n",
+ CMD_TFO(cmd)->get_task_tag(cmd));
+ wait_for_completion(&T_TASK(cmd)->transport_lun_stop_comp);
+ DEBUG_TRANSPORT_S("ConfigFS: ITT[0x%08x] - stopped cmd....\n",
+ CMD_TFO(cmd)->get_task_tag(cmd));
+ }
+ transport_remove_cmd_from_queue(cmd, SE_DEV(cmd)->dev_queue_obj);
+
+ return 0;
+}
+
+/* #define DEBUG_CLEAR_LUN */
+#ifdef DEBUG_CLEAR_LUN
+#define DEBUG_CLEAR_L(x...) printk(KERN_INFO x)
+#else
+#define DEBUG_CLEAR_L(x...)
+#endif
+
+static void __transport_clear_lun_from_sessions(struct se_lun *lun)
+{
+ struct se_cmd *cmd = NULL;
+ unsigned long lun_flags, cmd_flags;
+ /*
+ * Do exception processing and return CHECK_CONDITION status to the
+ * Initiator Port.
+ */
+ spin_lock_irqsave(&lun->lun_cmd_lock, lun_flags);
+ while (!list_empty_careful(&lun->lun_cmd_list)) {
+ cmd = list_entry(lun->lun_cmd_list.next,
+ struct se_cmd, se_lun_list);
+ list_del(&cmd->se_lun_list);
+
+ if (!(T_TASK(cmd))) {
+ printk(KERN_ERR "ITT: 0x%08x, T_TASK(cmd) = NULL"
+ "[i,t]_state: %u/%u\n",
+ CMD_TFO(cmd)->get_task_tag(cmd),
+ CMD_TFO(cmd)->get_cmd_state(cmd), cmd->t_state);
+ BUG();
+ }
+ atomic_set(&T_TASK(cmd)->transport_lun_active, 0);
+ /*
+ * This will notify iscsi_target_transport.c:
+ * transport_cmd_check_stop() that a LUN shutdown is in
+ * progress for the iscsi_cmd_t.
+ */
+ spin_lock(&T_TASK(cmd)->t_state_lock);
+ DEBUG_CLEAR_L("SE_LUN[%d] - Setting T_TASK(cmd)->transport"
+ "_lun_stop for ITT: 0x%08x\n",
+ SE_LUN(cmd)->unpacked_lun,
+ CMD_TFO(cmd)->get_task_tag(cmd));
+ atomic_set(&T_TASK(cmd)->transport_lun_stop, 1);
+ spin_unlock(&T_TASK(cmd)->t_state_lock);
+
+ spin_unlock_irqrestore(&lun->lun_cmd_lock, lun_flags);
+
+ if (!(SE_LUN(cmd))) {
+ printk(KERN_ERR "ITT: 0x%08x, [i,t]_state: %u/%u\n",
+ CMD_TFO(cmd)->get_task_tag(cmd),
+ CMD_TFO(cmd)->get_cmd_state(cmd), cmd->t_state);
+ BUG();
+ }
+ /*
+ * If the Storage engine still owns the iscsi_cmd_t, determine
+ * and/or stop its context.
+ */
+ DEBUG_CLEAR_L("SE_LUN[%d] - ITT: 0x%08x before transport"
+ "_lun_wait_for_tasks()\n", SE_LUN(cmd)->unpacked_lun,
+ CMD_TFO(cmd)->get_task_tag(cmd));
+
+ if (transport_lun_wait_for_tasks(cmd, SE_LUN(cmd)) < 0) {
+ spin_lock_irqsave(&lun->lun_cmd_lock, lun_flags);
+ continue;
+ }
+
+ DEBUG_CLEAR_L("SE_LUN[%d] - ITT: 0x%08x after transport_lun"
+ "_wait_for_tasks(): SUCCESS\n",
+ SE_LUN(cmd)->unpacked_lun,
+ CMD_TFO(cmd)->get_task_tag(cmd));
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, cmd_flags);
+ if (!(atomic_read(&T_TASK(cmd)->transport_dev_active))) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, cmd_flags);
+ goto check_cond;
+ }
+ atomic_set(&T_TASK(cmd)->transport_dev_active, 0);
+ transport_all_task_dev_remove_state(cmd);
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, cmd_flags);
+
+ transport_free_dev_tasks(cmd);
+ /*
+ * The Storage engine stopped this struct se_cmd before it was
+ * send to the fabric frontend for delivery back to the
+ * Initiator Node. Return this SCSI CDB back with an
+ * CHECK_CONDITION status.
+ */
+check_cond:
+ transport_send_check_condition_and_sense(cmd,
+ TCM_NON_EXISTENT_LUN, 0);
+ /*
+ * If the fabric frontend is waiting for this iscsi_cmd_t to
+ * be released, notify the waiting thread now that LU has
+ * finished accessing it.
+ */
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, cmd_flags);
+ if (atomic_read(&T_TASK(cmd)->transport_lun_fe_stop)) {
+ DEBUG_CLEAR_L("SE_LUN[%d] - Detected FE stop for"
+ " struct se_cmd: %p ITT: 0x%08x\n",
+ lun->unpacked_lun,
+ cmd, CMD_TFO(cmd)->get_task_tag(cmd));
+
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock,
+ cmd_flags);
+ transport_cmd_check_stop(cmd, 1, 0);
+ complete(&T_TASK(cmd)->transport_lun_fe_stop_comp);
+ spin_lock_irqsave(&lun->lun_cmd_lock, lun_flags);
+ continue;
+ }
+ DEBUG_CLEAR_L("SE_LUN[%d] - ITT: 0x%08x finished processing\n",
+ lun->unpacked_lun, CMD_TFO(cmd)->get_task_tag(cmd));
+
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, cmd_flags);
+ spin_lock_irqsave(&lun->lun_cmd_lock, lun_flags);
+ }
+ spin_unlock_irqrestore(&lun->lun_cmd_lock, lun_flags);
+}
+
+static int transport_clear_lun_thread(void *p)
+{
+ struct se_lun *lun = (struct se_lun *)p;
+
+ __transport_clear_lun_from_sessions(lun);
+ complete(&lun->lun_shutdown_comp);
+
+ return 0;
+}
+
+int transport_clear_lun_from_sessions(struct se_lun *lun)
+{
+ struct task_struct *kt;
+
+ kt = kthread_run(transport_clear_lun_thread, (void *)lun,
+ "tcm_cl_%u", lun->unpacked_lun);
+ if (IS_ERR(kt)) {
+ printk(KERN_ERR "Unable to start clear_lun thread\n");
+ return -1;
+ }
+ wait_for_completion(&lun->lun_shutdown_comp);
+
+ return 0;
+}
+
+/* transport_generic_wait_for_tasks():
+ *
+ * Called from frontend or passthrough context to wait for storage engine
+ * to pause and/or release frontend generated struct se_cmd.
+ */
+static void transport_generic_wait_for_tasks(
+ struct se_cmd *cmd,
+ int remove_cmd,
+ int session_reinstatement)
+{
+ unsigned long flags;
+
+ if (!(cmd->se_cmd_flags & SCF_SE_LUN_CMD) && !(cmd->se_tmr_req))
+ return;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ /*
+ * If we are already stopped due to an external event (ie: LUN shutdown)
+ * sleep until the connection can have the passed struct se_cmd back.
+ * The T_TASK(cmd)->transport_lun_stopped_sem will be upped by
+ * transport_clear_lun_from_sessions() once the ConfigFS context caller
+ * has completed its operation on the struct se_cmd.
+ */
+ if (atomic_read(&T_TASK(cmd)->transport_lun_stop)) {
+
+ DEBUG_TRANSPORT_S("wait_for_tasks: Stopping"
+ " wait_for_completion(&T_TASK(cmd)transport_lun_fe"
+ "_stop_comp); for ITT: 0x%08x\n",
+ CMD_TFO(cmd)->get_task_tag(cmd));
+ /*
+ * There is a special case for WRITES where a FE exception +
+ * LUN shutdown means ConfigFS context is still sleeping on
+ * transport_lun_stop_comp in transport_lun_wait_for_tasks().
+ * We go ahead and up transport_lun_stop_comp just to be sure
+ * here.
+ */
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ complete(&T_TASK(cmd)->transport_lun_stop_comp);
+ wait_for_completion(&T_TASK(cmd)->transport_lun_fe_stop_comp);
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+
+ transport_all_task_dev_remove_state(cmd);
+ /*
+ * At this point, the frontend who was the originator of this
+ * struct se_cmd, now owns the structure and can be released through
+ * normal means below.
+ */
+ DEBUG_TRANSPORT_S("wait_for_tasks: Stopped"
+ " wait_for_completion(&T_TASK(cmd)transport_lun_fe_"
+ "stop_comp); for ITT: 0x%08x\n",
+ CMD_TFO(cmd)->get_task_tag(cmd));
+
+ atomic_set(&T_TASK(cmd)->transport_lun_stop, 0);
+ }
+ if (!atomic_read(&T_TASK(cmd)->t_transport_active))
+ goto remove;
+
+ atomic_set(&T_TASK(cmd)->t_transport_stop, 1);
+
+ DEBUG_TRANSPORT_S("wait_for_tasks: Stopping %p ITT: 0x%08x"
+ " i_state: %d, t_state/def_t_state: %d/%d, t_transport_stop"
+ " = TRUE\n", cmd, CMD_TFO(cmd)->get_task_tag(cmd),
+ CMD_TFO(cmd)->get_cmd_state(cmd), cmd->t_state,
+ cmd->deferred_t_state);
+
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ wake_up_interruptible(&SE_DEV(cmd)->dev_queue_obj->thread_wq);
+
+ wait_for_completion(&T_TASK(cmd)->t_transport_stop_comp);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ atomic_set(&T_TASK(cmd)->t_transport_active, 0);
+ atomic_set(&T_TASK(cmd)->t_transport_stop, 0);
+
+ DEBUG_TRANSPORT_S("wait_for_tasks: Stopped wait_for_compltion("
+ "&T_TASK(cmd)->t_transport_stop_comp) for ITT: 0x%08x\n",
+ CMD_TFO(cmd)->get_task_tag(cmd));
+remove:
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ if (!remove_cmd)
+ return;
+
+ transport_generic_free_cmd(cmd, 0, 0, session_reinstatement);
+}
+
+static int transport_get_sense_codes(
+ struct se_cmd *cmd,
+ u8 *asc,
+ u8 *ascq)
+{
+ *asc = cmd->scsi_asc;
+ *ascq = cmd->scsi_ascq;
+
+ return 0;
+}
+
+static int transport_set_sense_codes(
+ struct se_cmd *cmd,
+ u8 asc,
+ u8 ascq)
+{
+ cmd->scsi_asc = asc;
+ cmd->scsi_ascq = ascq;
+
+ return 0;
+}
+
+int transport_send_check_condition_and_sense(
+ struct se_cmd *cmd,
+ u8 reason,
+ int from_transport)
+{
+ unsigned char *buffer = cmd->sense_buffer;
+ unsigned long flags;
+ int offset;
+ u8 asc = 0, ascq = 0;
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ if (cmd->se_cmd_flags & SCF_SENT_CHECK_CONDITION) {
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+ return 0;
+ }
+ cmd->se_cmd_flags |= SCF_SENT_CHECK_CONDITION;
+ spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
+
+ if (!reason && from_transport)
+ goto after_reason;
+
+ if (!from_transport)
+ cmd->se_cmd_flags |= SCF_EMULATED_TASK_SENSE;
+ /*
+ * Data Segment and SenseLength of the fabric response PDU.
+ *
+ * TRANSPORT_SENSE_BUFFER is now set to SCSI_SENSE_BUFFERSIZE
+ * from include/scsi/scsi_cmnd.h
+ */
+ offset = CMD_TFO(cmd)->set_fabric_sense_len(cmd,
+ TRANSPORT_SENSE_BUFFER);
+ /*
+ * Actual SENSE DATA, see SPC-3 7.23.2 SPC_SENSE_KEY_OFFSET uses
+ * SENSE KEY values from include/scsi/scsi.h
+ */
+ switch (reason) {
+ case TCM_NON_EXISTENT_LUN:
+ case TCM_UNSUPPORTED_SCSI_OPCODE:
+ case TCM_SECTOR_COUNT_TOO_MANY:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ILLEGAL REQUEST */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ILLEGAL_REQUEST;
+ /* INVALID COMMAND OPERATION CODE */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x20;
+ break;
+ case TCM_UNKNOWN_MODE_PAGE:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ILLEGAL REQUEST */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ILLEGAL_REQUEST;
+ /* INVALID FIELD IN CDB */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x24;
+ break;
+ case TCM_CHECK_CONDITION_ABORT_CMD:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ABORTED COMMAND */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ABORTED_COMMAND;
+ /* BUS DEVICE RESET FUNCTION OCCURRED */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x29;
+ buffer[offset+SPC_ASCQ_KEY_OFFSET] = 0x03;
+ break;
+ case TCM_INCORRECT_AMOUNT_OF_DATA:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ABORTED COMMAND */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ABORTED_COMMAND;
+ /* WRITE ERROR */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x0c;
+ /* NOT ENOUGH UNSOLICITED DATA */
+ buffer[offset+SPC_ASCQ_KEY_OFFSET] = 0x0d;
+ break;
+ case TCM_INVALID_CDB_FIELD:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ABORTED COMMAND */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ABORTED_COMMAND;
+ /* INVALID FIELD IN CDB */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x24;
+ break;
+ case TCM_INVALID_PARAMETER_LIST:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ABORTED COMMAND */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ABORTED_COMMAND;
+ /* INVALID FIELD IN PARAMETER LIST */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x26;
+ break;
+ case TCM_UNEXPECTED_UNSOLICITED_DATA:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ABORTED COMMAND */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ABORTED_COMMAND;
+ /* WRITE ERROR */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x0c;
+ /* UNEXPECTED_UNSOLICITED_DATA */
+ buffer[offset+SPC_ASCQ_KEY_OFFSET] = 0x0c;
+ break;
+ case TCM_SERVICE_CRC_ERROR:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ABORTED COMMAND */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ABORTED_COMMAND;
+ /* PROTOCOL SERVICE CRC ERROR */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x47;
+ /* N/A */
+ buffer[offset+SPC_ASCQ_KEY_OFFSET] = 0x05;
+ break;
+ case TCM_SNACK_REJECTED:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ABORTED COMMAND */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ABORTED_COMMAND;
+ /* READ ERROR */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x11;
+ /* FAILED RETRANSMISSION REQUEST */
+ buffer[offset+SPC_ASCQ_KEY_OFFSET] = 0x13;
+ break;
+ case TCM_WRITE_PROTECTED:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* DATA PROTECT */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = DATA_PROTECT;
+ /* WRITE PROTECTED */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x27;
+ break;
+ case TCM_CHECK_CONDITION_UNIT_ATTENTION:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* UNIT ATTENTION */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = UNIT_ATTENTION;
+ core_scsi3_ua_for_check_condition(cmd, &asc, &ascq);
+ buffer[offset+SPC_ASC_KEY_OFFSET] = asc;
+ buffer[offset+SPC_ASCQ_KEY_OFFSET] = ascq;
+ break;
+ case TCM_CHECK_CONDITION_NOT_READY:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* Not Ready */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = NOT_READY;
+ transport_get_sense_codes(cmd, &asc, &ascq);
+ buffer[offset+SPC_ASC_KEY_OFFSET] = asc;
+ buffer[offset+SPC_ASCQ_KEY_OFFSET] = ascq;
+ break;
+ case TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE:
+ default:
+ /* CURRENT ERROR */
+ buffer[offset] = 0x70;
+ /* ILLEGAL REQUEST */
+ buffer[offset+SPC_SENSE_KEY_OFFSET] = ILLEGAL_REQUEST;
+ /* LOGICAL UNIT COMMUNICATION FAILURE */
+ buffer[offset+SPC_ASC_KEY_OFFSET] = 0x80;
+ break;
+ }
+ /*
+ * This code uses linux/include/scsi/scsi.h SAM status codes!
+ */
+ cmd->scsi_status = SAM_STAT_CHECK_CONDITION;
+ /*
+ * Automatically padded, this value is encoded in the fabric's
+ * data_length response PDU containing the SCSI defined sense data.
+ */
+ cmd->scsi_sense_length = TRANSPORT_SENSE_BUFFER + offset;
+
+after_reason:
+ CMD_TFO(cmd)->queue_status(cmd);
+ return 0;
+}
+EXPORT_SYMBOL(transport_send_check_condition_and_sense);
+
+int transport_check_aborted_status(struct se_cmd *cmd, int send_status)
+{
+ int ret = 0;
+
+ if (atomic_read(&T_TASK(cmd)->t_transport_aborted) != 0) {
+ if (!(send_status) ||
+ (cmd->se_cmd_flags & SCF_SENT_DELAYED_TAS))
+ return 1;
+#if 0
+ printk(KERN_INFO "Sending delayed SAM_STAT_TASK_ABORTED"
+ " status for CDB: 0x%02x ITT: 0x%08x\n",
+ T_TASK(cmd)->t_task_cdb[0],
+ CMD_TFO(cmd)->get_task_tag(cmd));
+#endif
+ cmd->se_cmd_flags |= SCF_SENT_DELAYED_TAS;
+ CMD_TFO(cmd)->queue_status(cmd);
+ ret = 1;
+ }
+ return ret;
+}
+EXPORT_SYMBOL(transport_check_aborted_status);
+
+void transport_send_task_abort(struct se_cmd *cmd)
+{
+ /*
+ * If there are still expected incoming fabric WRITEs, we wait
+ * until until they have completed before sending a TASK_ABORTED
+ * response. This response with TASK_ABORTED status will be
+ * queued back to fabric module by transport_check_aborted_status().
+ */
+ if (cmd->data_direction == DMA_TO_DEVICE) {
+ if (CMD_TFO(cmd)->write_pending_status(cmd) != 0) {
+ atomic_inc(&T_TASK(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;
+#if 0
+ printk(KERN_INFO "Setting SAM_STAT_TASK_ABORTED status for CDB: 0x%02x,"
+ " ITT: 0x%08x\n", T_TASK(cmd)->t_task_cdb[0],
+ CMD_TFO(cmd)->get_task_tag(cmd));
+#endif
+ CMD_TFO(cmd)->queue_status(cmd);
+}
+
+/* transport_generic_do_tmr():
+ *
+ *
+ */
+int transport_generic_do_tmr(struct se_cmd *cmd)
+{
+ struct se_cmd *ref_cmd;
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_tmr_req *tmr = cmd->se_tmr_req;
+ int ret;
+
+ switch (tmr->function) {
+ case ABORT_TASK:
+ ref_cmd = tmr->ref_cmd;
+ tmr->response = TMR_FUNCTION_REJECTED;
+ break;
+ case ABORT_TASK_SET:
+ case CLEAR_ACA:
+ case CLEAR_TASK_SET:
+ tmr->response = TMR_TASK_MGMT_FUNCTION_NOT_SUPPORTED;
+ break;
+ case LUN_RESET:
+ ret = core_tmr_lun_reset(dev, tmr, NULL, NULL);
+ tmr->response = (!ret) ? TMR_FUNCTION_COMPLETE :
+ TMR_FUNCTION_REJECTED;
+ break;
+#if 0
+ case TARGET_WARM_RESET:
+ transport_generic_host_reset(dev->se_hba);
+ tmr->response = TMR_FUNCTION_REJECTED;
+ break;
+ case TARGET_COLD_RESET:
+ transport_generic_host_reset(dev->se_hba);
+ transport_generic_cold_reset(dev->se_hba);
+ tmr->response = TMR_FUNCTION_REJECTED;
+ break;
+#endif
+ default:
+ printk(KERN_ERR "Uknown TMR function: 0x%02x.\n",
+ tmr->function);
+ tmr->response = TMR_FUNCTION_REJECTED;
+ break;
+ }
+
+ cmd->t_state = TRANSPORT_ISTATE_PROCESSING;
+ CMD_TFO(cmd)->queue_tm_rsp(cmd);
+
+ transport_cmd_check_stop(cmd, 2, 0);
+ return 0;
+}
+
+/*
+ * Called with spin_lock_irq(&dev->execute_task_lock); held
+ *
+ */
+static struct se_task *
+transport_get_task_from_state_list(struct se_device *dev)
+{
+ struct se_task *task;
+
+ if (list_empty(&dev->state_task_list))
+ return NULL;
+
+ list_for_each_entry(task, &dev->state_task_list, t_state_list)
+ break;
+
+ list_del(&task->t_state_list);
+ atomic_set(&task->task_state_active, 0);
+
+ return task;
+}
+
+static void transport_processing_shutdown(struct se_device *dev)
+{
+ struct se_cmd *cmd;
+ struct se_queue_req *qr;
+ struct se_task *task;
+ u8 state;
+ unsigned long flags;
+ /*
+ * Empty the struct se_device's struct se_task state list.
+ */
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ while ((task = transport_get_task_from_state_list(dev))) {
+ if (!(TASK_CMD(task))) {
+ printk(KERN_ERR "TASK_CMD(task) is NULL!\n");
+ continue;
+ }
+ cmd = TASK_CMD(task);
+
+ if (!T_TASK(cmd)) {
+ printk(KERN_ERR "T_TASK(cmd) is NULL for task: %p cmd:"
+ " %p ITT: 0x%08x\n", task, cmd,
+ CMD_TFO(cmd)->get_task_tag(cmd));
+ continue;
+ }
+ spin_unlock_irqrestore(&dev->execute_task_lock, flags);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+
+ DEBUG_DO("PT: cmd: %p task: %p ITT/CmdSN: 0x%08x/0x%08x,"
+ " i_state/def_i_state: %d/%d, t_state/def_t_state:"
+ " %d/%d cdb: 0x%02x\n", cmd, task,
+ CMD_TFO(cmd)->get_task_tag(cmd), cmd->cmd_sn,
+ CMD_TFO(cmd)->get_cmd_state(cmd), cmd->deferred_i_state,
+ cmd->t_state, cmd->deferred_t_state,
+ T_TASK(cmd)->t_task_cdb[0]);
+ DEBUG_DO("PT: ITT[0x%08x] - t_task_cdbs: %d t_task_cdbs_left:"
+ " %d t_task_cdbs_sent: %d -- t_transport_active: %d"
+ " t_transport_stop: %d t_transport_sent: %d\n",
+ CMD_TFO(cmd)->get_task_tag(cmd),
+ T_TASK(cmd)->t_task_cdbs,
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_left),
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_sent),
+ atomic_read(&T_TASK(cmd)->t_transport_active),
+ atomic_read(&T_TASK(cmd)->t_transport_stop),
+ atomic_read(&T_TASK(cmd)->t_transport_sent));
+
+ if (atomic_read(&task->task_active)) {
+ atomic_set(&task->task_stop, 1);
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+
+ DEBUG_DO("Waiting for task: %p to shutdown for dev:"
+ " %p\n", task, dev);
+ wait_for_completion(&task->task_stop_comp);
+ DEBUG_DO("Completed task: %p shutdown for dev: %p\n",
+ task, dev);
+
+ spin_lock_irqsave(&T_TASK(cmd)->t_state_lock, flags);
+ atomic_dec(&T_TASK(cmd)->t_task_cdbs_left);
+
+ atomic_set(&task->task_active, 0);
+ atomic_set(&task->task_stop, 0);
+ }
+ __transport_stop_task_timer(task, &flags);
+
+ if (!(atomic_dec_and_test(&T_TASK(cmd)->t_task_cdbs_ex_left))) {
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+
+ DEBUG_DO("Skipping task: %p, dev: %p for"
+ " t_task_cdbs_ex_left: %d\n", task, dev,
+ atomic_read(&T_TASK(cmd)->t_task_cdbs_ex_left));
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ continue;
+ }
+
+ if (atomic_read(&T_TASK(cmd)->t_transport_active)) {
+ DEBUG_DO("got t_transport_active = 1 for task: %p, dev:"
+ " %p\n", task, dev);
+
+ if (atomic_read(&T_TASK(cmd)->t_fe_count)) {
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+ transport_send_check_condition_and_sense(
+ cmd, TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE,
+ 0);
+ transport_remove_cmd_from_queue(cmd,
+ SE_DEV(cmd)->dev_queue_obj);
+
+ transport_lun_remove_cmd(cmd);
+ transport_cmd_check_stop(cmd, 1, 0);
+ } else {
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+
+ transport_remove_cmd_from_queue(cmd,
+ SE_DEV(cmd)->dev_queue_obj);
+
+ transport_lun_remove_cmd(cmd);
+
+ if (transport_cmd_check_stop(cmd, 1, 0))
+ transport_generic_remove(cmd, 0, 0);
+ }
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ continue;
+ }
+ DEBUG_DO("Got t_transport_active = 0 for task: %p, dev: %p\n",
+ task, dev);
+
+ if (atomic_read(&T_TASK(cmd)->t_fe_count)) {
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+ transport_send_check_condition_and_sense(cmd,
+ TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE, 0);
+ transport_remove_cmd_from_queue(cmd,
+ SE_DEV(cmd)->dev_queue_obj);
+
+ transport_lun_remove_cmd(cmd);
+ transport_cmd_check_stop(cmd, 1, 0);
+ } else {
+ spin_unlock_irqrestore(
+ &T_TASK(cmd)->t_state_lock, flags);
+
+ transport_remove_cmd_from_queue(cmd,
+ SE_DEV(cmd)->dev_queue_obj);
+ transport_lun_remove_cmd(cmd);
+
+ if (transport_cmd_check_stop(cmd, 1, 0))
+ transport_generic_remove(cmd, 0, 0);
+ }
+
+ spin_lock_irqsave(&dev->execute_task_lock, flags);
+ }
+ spin_unlock_irqrestore(&dev->execute_task_lock, flags);
+ /*
+ * Empty the struct se_device's struct se_cmd list.
+ */
+ spin_lock_irqsave(&dev->dev_queue_obj->cmd_queue_lock, flags);
+ while ((qr = __transport_get_qr_from_queue(dev->dev_queue_obj))) {
+ spin_unlock_irqrestore(
+ &dev->dev_queue_obj->cmd_queue_lock, flags);
+ cmd = (struct se_cmd *)qr->cmd;
+ state = qr->state;
+ kfree(qr);
+
+ DEBUG_DO("From Device Queue: cmd: %p t_state: %d\n",
+ cmd, state);
+
+ if (atomic_read(&T_TASK(cmd)->t_fe_count)) {
+ transport_send_check_condition_and_sense(cmd,
+ TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE, 0);
+
+ transport_lun_remove_cmd(cmd);
+ transport_cmd_check_stop(cmd, 1, 0);
+ } else {
+ transport_lun_remove_cmd(cmd);
+ if (transport_cmd_check_stop(cmd, 1, 0))
+ transport_generic_remove(cmd, 0, 0);
+ }
+ spin_lock_irqsave(&dev->dev_queue_obj->cmd_queue_lock, flags);
+ }
+ spin_unlock_irqrestore(&dev->dev_queue_obj->cmd_queue_lock, flags);
+}
+
+/* transport_processing_thread():
+ *
+ *
+ */
+static int transport_processing_thread(void *param)
+{
+ int ret, t_state;
+ struct se_cmd *cmd;
+ struct se_device *dev = (struct se_device *) param;
+ struct se_queue_req *qr;
+
+ 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) ||
+ kthread_should_stop());
+ if (ret < 0)
+ goto out;
+
+ spin_lock_irq(&dev->dev_status_lock);
+ if (dev->dev_status & TRANSPORT_DEVICE_SHUTDOWN) {
+ spin_unlock_irq(&dev->dev_status_lock);
+ transport_processing_shutdown(dev);
+ continue;
+ }
+ spin_unlock_irq(&dev->dev_status_lock);
+
+get_cmd:
+ __transport_execute_tasks(dev);
+
+ qr = transport_get_qr_from_queue(dev->dev_queue_obj);
+ if (!(qr))
+ continue;
+
+ cmd = (struct se_cmd *)qr->cmd;
+ t_state = qr->state;
+ kfree(qr);
+
+ switch (t_state) {
+ case TRANSPORT_NEW_CMD_MAP:
+ if (!(CMD_TFO(cmd)->new_cmd_map)) {
+ printk(KERN_ERR "CMD_TFO(cmd)->new_cmd_map is"
+ " NULL for TRANSPORT_NEW_CMD_MAP\n");
+ BUG();
+ }
+ ret = CMD_TFO(cmd)->new_cmd_map(cmd);
+ if (ret < 0) {
+ cmd->transport_error_status = ret;
+ transport_generic_request_failure(cmd, NULL,
+ 0, (cmd->data_direction !=
+ DMA_TO_DEVICE));
+ break;
+ }
+ /* Fall through */
+ case TRANSPORT_NEW_CMD:
+ ret = transport_generic_new_cmd(cmd);
+ if (ret < 0) {
+ cmd->transport_error_status = ret;
+ transport_generic_request_failure(cmd, NULL,
+ 0, (cmd->data_direction !=
+ DMA_TO_DEVICE));
+ }
+ break;
+ case TRANSPORT_PROCESS_WRITE:
+ transport_generic_process_write(cmd);
+ break;
+ case TRANSPORT_COMPLETE_OK:
+ transport_stop_all_task_timers(cmd);
+ transport_generic_complete_ok(cmd);
+ break;
+ case TRANSPORT_REMOVE:
+ transport_generic_remove(cmd, 1, 0);
+ break;
+ case TRANSPORT_PROCESS_TMR:
+ transport_generic_do_tmr(cmd);
+ break;
+ case TRANSPORT_COMPLETE_FAILURE:
+ transport_generic_request_failure(cmd, NULL, 1, 1);
+ break;
+ case TRANSPORT_COMPLETE_TIMEOUT:
+ transport_stop_all_task_timers(cmd);
+ transport_generic_request_timeout(cmd);
+ break;
+ default:
+ printk(KERN_ERR "Unknown t_state: %d deferred_t_state:"
+ " %d for ITT: 0x%08x i_state: %d on SE LUN:"
+ " %u\n", t_state, cmd->deferred_t_state,
+ CMD_TFO(cmd)->get_task_tag(cmd),
+ CMD_TFO(cmd)->get_cmd_state(cmd),
+ SE_LUN(cmd)->unpacked_lun);
+ BUG();
+ }
+
+ goto get_cmd;
+ }
+
+out:
+ transport_release_all_cmds(dev);
+ dev->process_thread = NULL;
+ return 0;
+}
diff --git a/drivers/target/target_core_ua.c b/drivers/target/target_core_ua.c
new file mode 100644
index 0000000..a2ef346
--- /dev/null
+++ b/drivers/target/target_core_ua.c
@@ -0,0 +1,332 @@
+/*******************************************************************************
+ * Filename: target_core_ua.c
+ *
+ * This file contains logic for SPC-3 Unit Attention emulation
+ *
+ * Copyright (c) 2009,2010 Rising Tide Systems
+ * Copyright (c) 2009,2010 Linux-iSCSI.org
+ *
+ * Nicholas A. Bellinger <nab@kernel.org>
+ *
+ * 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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ ******************************************************************************/
+
+#include <linux/version.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+
+#include <target/target_core_base.h>
+#include <target/target_core_device.h>
+#include <target/target_core_transport.h>
+#include <target/target_core_fabric_ops.h>
+#include <target/target_core_configfs.h>
+
+#include "target_core_alua.h"
+#include "target_core_hba.h"
+#include "target_core_pr.h"
+#include "target_core_ua.h"
+
+int core_scsi3_ua_check(
+ struct se_cmd *cmd,
+ unsigned char *cdb)
+{
+ struct se_dev_entry *deve;
+ struct se_session *sess = cmd->se_sess;
+ struct se_node_acl *nacl;
+
+ if (!(sess))
+ return 0;
+
+ nacl = sess->se_node_acl;
+ if (!(nacl))
+ return 0;
+
+ deve = &nacl->device_list[cmd->orig_fe_lun];
+ if (!(atomic_read(&deve->ua_count)))
+ return 0;
+ /*
+ * From sam4r14, section 5.14 Unit attention condition:
+ *
+ * a) if an INQUIRY command enters the enabled command state, the
+ * device server shall process the INQUIRY command and shall neither
+ * report nor clear any unit attention condition;
+ * b) if a REPORT LUNS command enters the enabled command state, the
+ * device server shall process the REPORT LUNS command and shall not
+ * report any unit attention condition;
+ * e) if a REQUEST SENSE command enters the enabled command state while
+ * a unit attention condition exists for the SCSI initiator port
+ * associated with the I_T nexus on which the REQUEST SENSE command
+ * was received, then the device server shall process the command
+ * and either:
+ */
+ switch (cdb[0]) {
+ case INQUIRY:
+ case REPORT_LUNS:
+ case REQUEST_SENSE:
+ return 0;
+ default:
+ return -1;
+ }
+
+ return -1;
+}
+
+int core_scsi3_ua_allocate(
+ struct se_node_acl *nacl,
+ u32 unpacked_lun,
+ u8 asc,
+ u8 ascq)
+{
+ struct se_dev_entry *deve;
+ struct se_ua *ua, *ua_p, *ua_tmp;
+ /*
+ * PASSTHROUGH OPS
+ */
+ if (!(nacl))
+ return -1;
+
+ ua = kmem_cache_zalloc(se_ua_cache, GFP_ATOMIC);
+ if (!(ua)) {
+ printk(KERN_ERR "Unable to allocate struct se_ua\n");
+ return -1;
+ }
+ INIT_LIST_HEAD(&ua->ua_dev_list);
+ INIT_LIST_HEAD(&ua->ua_nacl_list);
+
+ ua->ua_nacl = nacl;
+ ua->ua_asc = asc;
+ ua->ua_ascq = ascq;
+
+ spin_lock_irq(&nacl->device_list_lock);
+ deve = &nacl->device_list[unpacked_lun];
+
+ spin_lock(&deve->ua_lock);
+ list_for_each_entry_safe(ua_p, ua_tmp, &deve->ua_list, ua_nacl_list) {
+ /*
+ * Do not report the same UNIT ATTENTION twice..
+ */
+ if ((ua_p->ua_asc == asc) && (ua_p->ua_ascq == ascq)) {
+ spin_unlock(&deve->ua_lock);
+ spin_unlock_irq(&nacl->device_list_lock);
+ kmem_cache_free(se_ua_cache, ua);
+ return 0;
+ }
+ /*
+ * Attach the highest priority Unit Attention to
+ * the head of the list following sam4r14,
+ * Section 5.14 Unit Attention Condition:
+ *
+ * POWER ON, RESET, OR BUS DEVICE RESET OCCURRED highest
+ * POWER ON OCCURRED or
+ * DEVICE INTERNAL RESET
+ * SCSI BUS RESET OCCURRED or
+ * MICROCODE HAS BEEN CHANGED or
+ * protocol specific
+ * BUS DEVICE RESET FUNCTION OCCURRED
+ * I_T NEXUS LOSS OCCURRED
+ * COMMANDS CLEARED BY POWER LOSS NOTIFICATION
+ * all others Lowest
+ *
+ * Each of the ASCQ codes listed above are defined in
+ * the 29h ASC family, see spc4r17 Table D.1
+ */
+ if (ua_p->ua_asc == 0x29) {
+ if ((asc == 0x29) && (ascq > ua_p->ua_ascq))
+ list_add(&ua->ua_nacl_list,
+ &deve->ua_list);
+ else
+ list_add_tail(&ua->ua_nacl_list,
+ &deve->ua_list);
+ } else if (ua_p->ua_asc == 0x2a) {
+ /*
+ * Incoming Family 29h ASCQ codes will override
+ * Family 2AHh ASCQ codes for Unit Attention condition.
+ */
+ if ((asc == 0x29) || (ascq > ua_p->ua_asc))
+ list_add(&ua->ua_nacl_list,
+ &deve->ua_list);
+ else
+ list_add_tail(&ua->ua_nacl_list,
+ &deve->ua_list);
+ } else
+ list_add_tail(&ua->ua_nacl_list,
+ &deve->ua_list);
+ spin_unlock(&deve->ua_lock);
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ atomic_inc(&deve->ua_count);
+ smp_mb__after_atomic_inc();
+ return 0;
+ }
+ list_add_tail(&ua->ua_nacl_list, &deve->ua_list);
+ spin_unlock(&deve->ua_lock);
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ printk(KERN_INFO "[%s]: Allocated UNIT ATTENTION, mapped LUN: %u, ASC:"
+ " 0x%02x, ASCQ: 0x%02x\n",
+ TPG_TFO(nacl->se_tpg)->get_fabric_name(), unpacked_lun,
+ asc, ascq);
+
+ atomic_inc(&deve->ua_count);
+ smp_mb__after_atomic_inc();
+ return 0;
+}
+
+void core_scsi3_ua_release_all(
+ struct se_dev_entry *deve)
+{
+ struct se_ua *ua, *ua_p;
+
+ spin_lock(&deve->ua_lock);
+ list_for_each_entry_safe(ua, ua_p, &deve->ua_list, ua_nacl_list) {
+ list_del(&ua->ua_nacl_list);
+ kmem_cache_free(se_ua_cache, ua);
+
+ atomic_dec(&deve->ua_count);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&deve->ua_lock);
+}
+
+void core_scsi3_ua_for_check_condition(
+ struct se_cmd *cmd,
+ u8 *asc,
+ u8 *ascq)
+{
+ struct se_device *dev = SE_DEV(cmd);
+ struct se_dev_entry *deve;
+ struct se_session *sess = cmd->se_sess;
+ struct se_node_acl *nacl;
+ struct se_ua *ua = NULL, *ua_p;
+ int head = 1;
+
+ if (!(sess))
+ return;
+
+ nacl = sess->se_node_acl;
+ if (!(nacl))
+ return;
+
+ spin_lock_irq(&nacl->device_list_lock);
+ deve = &nacl->device_list[cmd->orig_fe_lun];
+ if (!(atomic_read(&deve->ua_count))) {
+ spin_unlock_irq(&nacl->device_list_lock);
+ return;
+ }
+ /*
+ * The highest priority Unit Attentions are placed at the head of the
+ * struct se_dev_entry->ua_list, and will be returned in CHECK_CONDITION +
+ * sense data for the received CDB.
+ */
+ spin_lock(&deve->ua_lock);
+ list_for_each_entry_safe(ua, ua_p, &deve->ua_list, ua_nacl_list) {
+ /*
+ * For ua_intlck_ctrl code not equal to 00b, only report the
+ * highest priority UNIT_ATTENTION and ASC/ASCQ without
+ * clearing it.
+ */
+ if (DEV_ATTRIB(dev)->emulate_ua_intlck_ctrl != 0) {
+ *asc = ua->ua_asc;
+ *ascq = ua->ua_ascq;
+ break;
+ }
+ /*
+ * Otherwise for the default 00b, release the UNIT ATTENTION
+ * condition. Return the ASC/ASCQ of the higest priority UA
+ * (head of the list) in the outgoing CHECK_CONDITION + sense.
+ */
+ if (head) {
+ *asc = ua->ua_asc;
+ *ascq = ua->ua_ascq;
+ head = 0;
+ }
+ list_del(&ua->ua_nacl_list);
+ kmem_cache_free(se_ua_cache, ua);
+
+ atomic_dec(&deve->ua_count);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&deve->ua_lock);
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ printk(KERN_INFO "[%s]: %s UNIT ATTENTION condition with"
+ " INTLCK_CTRL: %d, mapped LUN: %u, got CDB: 0x%02x"
+ " reported ASC: 0x%02x, ASCQ: 0x%02x\n",
+ TPG_TFO(nacl->se_tpg)->get_fabric_name(),
+ (DEV_ATTRIB(dev)->emulate_ua_intlck_ctrl != 0) ? "Reporting" :
+ "Releasing", DEV_ATTRIB(dev)->emulate_ua_intlck_ctrl,
+ cmd->orig_fe_lun, T_TASK(cmd)->t_task_cdb[0], *asc, *ascq);
+}
+
+int core_scsi3_ua_clear_for_request_sense(
+ struct se_cmd *cmd,
+ u8 *asc,
+ u8 *ascq)
+{
+ struct se_dev_entry *deve;
+ struct se_session *sess = cmd->se_sess;
+ struct se_node_acl *nacl;
+ struct se_ua *ua = NULL, *ua_p;
+ int head = 1;
+
+ if (!(sess))
+ return -1;
+
+ nacl = sess->se_node_acl;
+ if (!(nacl))
+ return -1;
+
+ spin_lock_irq(&nacl->device_list_lock);
+ deve = &nacl->device_list[cmd->orig_fe_lun];
+ if (!(atomic_read(&deve->ua_count))) {
+ spin_unlock_irq(&nacl->device_list_lock);
+ return -1;
+ }
+ /*
+ * The highest priority Unit Attentions are placed at the head of the
+ * struct se_dev_entry->ua_list. The First (and hence highest priority)
+ * ASC/ASCQ will be returned in REQUEST_SENSE payload data for the
+ * matching struct se_lun.
+ *
+ * Once the returning ASC/ASCQ values are set, we go ahead and
+ * release all of the Unit Attention conditions for the assoicated
+ * struct se_lun.
+ */
+ spin_lock(&deve->ua_lock);
+ list_for_each_entry_safe(ua, ua_p, &deve->ua_list, ua_nacl_list) {
+ if (head) {
+ *asc = ua->ua_asc;
+ *ascq = ua->ua_ascq;
+ head = 0;
+ }
+ list_del(&ua->ua_nacl_list);
+ kmem_cache_free(se_ua_cache, ua);
+
+ atomic_dec(&deve->ua_count);
+ smp_mb__after_atomic_dec();
+ }
+ spin_unlock(&deve->ua_lock);
+ spin_unlock_irq(&nacl->device_list_lock);
+
+ printk(KERN_INFO "[%s]: Released UNIT ATTENTION condition, mapped"
+ " LUN: %u, got REQUEST_SENSE reported ASC: 0x%02x,"
+ " ASCQ: 0x%02x\n", TPG_TFO(nacl->se_tpg)->get_fabric_name(),
+ cmd->orig_fe_lun, *asc, *ascq);
+
+ return (head) ? -1 : 0;
+}
diff --git a/drivers/target/target_core_ua.h b/drivers/target/target_core_ua.h
new file mode 100644
index 0000000..6e6b034
--- /dev/null
+++ b/drivers/target/target_core_ua.h
@@ -0,0 +1,36 @@
+#ifndef TARGET_CORE_UA_H
+
+/*
+ * From spc4r17, Table D.1: ASC and ASCQ Assignement
+ */
+#define ASCQ_29H_POWER_ON_RESET_OR_BUS_DEVICE_RESET_OCCURED 0x00
+#define ASCQ_29H_POWER_ON_OCCURRED 0x01
+#define ASCQ_29H_SCSI_BUS_RESET_OCCURED 0x02
+#define ASCQ_29H_BUS_DEVICE_RESET_FUNCTION_OCCURRED 0x03
+#define ASCQ_29H_DEVICE_INTERNAL_RESET 0x04
+#define ASCQ_29H_TRANSCEIVER_MODE_CHANGED_TO_SINGLE_ENDED 0x05
+#define ASCQ_29H_TRANSCEIVER_MODE_CHANGED_TO_LVD 0x06
+#define ASCQ_29H_NEXUS_LOSS_OCCURRED 0x07
+
+#define ASCQ_2AH_PARAMETERS_CHANGED 0x00
+#define ASCQ_2AH_MODE_PARAMETERS_CHANGED 0x01
+#define ASCQ_2AH_LOG_PARAMETERS_CHANGED 0x02
+#define ASCQ_2AH_RESERVATIONS_PREEMPTED 0x03
+#define ASCQ_2AH_RESERVATIONS_RELEASED 0x04
+#define ASCQ_2AH_REGISTRATIONS_PREEMPTED 0x05
+#define ASCQ_2AH_ASYMMETRIC_ACCESS_STATE_CHANGED 0x06
+#define ASCQ_2AH_IMPLICT_ASYMMETRIC_ACCESS_STATE_TRANSITION_FAILED 0x07
+#define ASCQ_2AH_PRIORITY_CHANGED 0x08
+
+#define ASCQ_2CH_PREVIOUS_RESERVATION_CONFLICT_STATUS 0x09
+
+extern struct kmem_cache *se_ua_cache;
+
+extern int core_scsi3_ua_check(struct se_cmd *, unsigned char *);
+extern int core_scsi3_ua_allocate(struct se_node_acl *, u32, u8, u8);
+extern void core_scsi3_ua_release_all(struct se_dev_entry *);
+extern void core_scsi3_ua_for_check_condition(struct se_cmd *, u8 *, u8 *);
+extern int core_scsi3_ua_clear_for_request_sense(struct se_cmd *,
+ u8 *, u8 *);
+
+#endif /* TARGET_CORE_UA_H */
diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
index bf7c687..f7a5dba 100644
--- a/drivers/thermal/Kconfig
+++ b/drivers/thermal/Kconfig
@@ -4,6 +4,7 @@
menuconfig THERMAL
tristate "Generic Thermal sysfs driver"
+ depends on NET
help
Generic Thermal Sysfs driver offers a generic mechanism for
thermal management. Usually it's made up of one or more thermal
diff --git a/drivers/thermal/thermal_sys.c b/drivers/thermal/thermal_sys.c
index 13c72c6..7d0e63c 100644
--- a/drivers/thermal/thermal_sys.c
+++ b/drivers/thermal/thermal_sys.c
@@ -32,6 +32,8 @@
#include <linux/thermal.h>
#include <linux/spinlock.h>
#include <linux/reboot.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
MODULE_AUTHOR("Zhang Rui");
MODULE_DESCRIPTION("Generic thermal management sysfs support");
@@ -58,6 +60,22 @@
static LIST_HEAD(thermal_cdev_list);
static DEFINE_MUTEX(thermal_list_lock);
+static unsigned int thermal_event_seqnum;
+
+static struct genl_family thermal_event_genl_family = {
+ .id = GENL_ID_GENERATE,
+ .name = THERMAL_GENL_FAMILY_NAME,
+ .version = THERMAL_GENL_VERSION,
+ .maxattr = THERMAL_GENL_ATTR_MAX,
+};
+
+static struct genl_multicast_group thermal_event_mcgrp = {
+ .name = THERMAL_GENL_MCAST_GROUP_NAME,
+};
+
+static int genetlink_init(void);
+static void genetlink_exit(void);
+
static int get_idr(struct idr *idr, struct mutex *lock, int *id)
{
int err;
@@ -823,11 +841,8 @@
* @devdata: device private data.
* @ops: standard thermal cooling devices callbacks.
*/
-struct thermal_cooling_device *thermal_cooling_device_register(char *type,
- void *devdata,
- struct
- thermal_cooling_device_ops
- *ops)
+struct thermal_cooling_device *thermal_cooling_device_register(
+ char *type, void *devdata, const struct thermal_cooling_device_ops *ops)
{
struct thermal_cooling_device *cdev;
struct thermal_zone_device *pos;
@@ -1048,13 +1063,9 @@
* section 11.1.5.1 of the ACPI specification 3.0.
*/
struct thermal_zone_device *thermal_zone_device_register(char *type,
- int trips,
- void *devdata, struct
- thermal_zone_device_ops
- *ops, int tc1, int
- tc2,
- int passive_delay,
- int polling_delay)
+ int trips, void *devdata,
+ const struct thermal_zone_device_ops *ops,
+ int tc1, int tc2, int passive_delay, int polling_delay)
{
struct thermal_zone_device *tz;
struct thermal_cooling_device *pos;
@@ -1214,6 +1225,82 @@
EXPORT_SYMBOL(thermal_zone_device_unregister);
+int generate_netlink_event(u32 orig, enum events event)
+{
+ struct sk_buff *skb;
+ struct nlattr *attr;
+ struct thermal_genl_event *thermal_event;
+ void *msg_header;
+ int size;
+ int result;
+
+ /* allocate memory */
+ size = nla_total_size(sizeof(struct thermal_genl_event)) + \
+ nla_total_size(0);
+
+ skb = genlmsg_new(size, GFP_ATOMIC);
+ if (!skb)
+ return -ENOMEM;
+
+ /* add the genetlink message header */
+ msg_header = genlmsg_put(skb, 0, thermal_event_seqnum++,
+ &thermal_event_genl_family, 0,
+ THERMAL_GENL_CMD_EVENT);
+ if (!msg_header) {
+ nlmsg_free(skb);
+ return -ENOMEM;
+ }
+
+ /* fill the data */
+ attr = nla_reserve(skb, THERMAL_GENL_ATTR_EVENT, \
+ sizeof(struct thermal_genl_event));
+
+ if (!attr) {
+ nlmsg_free(skb);
+ return -EINVAL;
+ }
+
+ thermal_event = nla_data(attr);
+ if (!thermal_event) {
+ nlmsg_free(skb);
+ return -EINVAL;
+ }
+
+ memset(thermal_event, 0, sizeof(struct thermal_genl_event));
+
+ thermal_event->orig = orig;
+ thermal_event->event = event;
+
+ /* send multicast genetlink message */
+ result = genlmsg_end(skb, msg_header);
+ if (result < 0) {
+ nlmsg_free(skb);
+ return result;
+ }
+
+ result = genlmsg_multicast(skb, 0, thermal_event_mcgrp.id, GFP_ATOMIC);
+ if (result)
+ printk(KERN_INFO "failed to send netlink event:%d", result);
+
+ return result;
+}
+EXPORT_SYMBOL(generate_netlink_event);
+
+static int genetlink_init(void)
+{
+ int result;
+
+ result = genl_register_family(&thermal_event_genl_family);
+ if (result)
+ return result;
+
+ result = genl_register_mc_group(&thermal_event_genl_family,
+ &thermal_event_mcgrp);
+ if (result)
+ genl_unregister_family(&thermal_event_genl_family);
+ return result;
+}
+
static int __init thermal_init(void)
{
int result = 0;
@@ -1225,9 +1312,15 @@
mutex_destroy(&thermal_idr_lock);
mutex_destroy(&thermal_list_lock);
}
+ result = genetlink_init();
return result;
}
+static void genetlink_exit(void)
+{
+ genl_unregister_family(&thermal_event_genl_family);
+}
+
static void __exit thermal_exit(void)
{
class_unregister(&thermal_class);
@@ -1235,7 +1328,8 @@
idr_destroy(&thermal_cdev_idr);
mutex_destroy(&thermal_idr_lock);
mutex_destroy(&thermal_list_lock);
+ genetlink_exit();
}
-subsys_initcall(thermal_init);
+fs_initcall(thermal_init);
module_exit(thermal_exit);
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 38244f5..ade0568 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -97,22 +97,26 @@
remove_wait_queue(poll->wqh, &poll->wait);
}
+static bool vhost_work_seq_done(struct vhost_dev *dev, struct vhost_work *work,
+ unsigned seq)
+{
+ int left;
+ spin_lock_irq(&dev->work_lock);
+ left = seq - work->done_seq;
+ spin_unlock_irq(&dev->work_lock);
+ return left <= 0;
+}
+
static void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work)
{
unsigned seq;
- int left;
int flushing;
spin_lock_irq(&dev->work_lock);
seq = work->queue_seq;
work->flushing++;
spin_unlock_irq(&dev->work_lock);
- wait_event(work->done, ({
- spin_lock_irq(&dev->work_lock);
- left = seq - work->done_seq <= 0;
- spin_unlock_irq(&dev->work_lock);
- left;
- }));
+ wait_event(work->done, vhost_work_seq_done(dev, work, seq));
spin_lock_irq(&dev->work_lock);
flushing = --work->flushing;
spin_unlock_irq(&dev->work_lock);
diff --git a/drivers/video/ep93xx-fb.c b/drivers/video/ep93xx-fb.c
index 0c99de0..b358d04 100644
--- a/drivers/video/ep93xx-fb.c
+++ b/drivers/video/ep93xx-fb.c
@@ -483,7 +483,7 @@
info->screen_base, info->fix.smem_start);
}
-static int __init ep93xxfb_probe(struct platform_device *pdev)
+static int __devinit ep93xxfb_probe(struct platform_device *pdev)
{
struct ep93xxfb_mach_info *mach_info = pdev->dev.platform_data;
struct fb_info *info;
@@ -598,7 +598,7 @@
return err;
}
-static int ep93xxfb_remove(struct platform_device *pdev)
+static int __devexit ep93xxfb_remove(struct platform_device *pdev)
{
struct fb_info *info = platform_get_drvdata(pdev);
struct ep93xx_fbi *fbi = info->par;
@@ -622,7 +622,7 @@
static struct platform_driver ep93xxfb_driver = {
.probe = ep93xxfb_probe,
- .remove = ep93xxfb_remove,
+ .remove = __devexit_p(ep93xxfb_remove),
.driver = {
.name = "ep93xx-fb",
.owner = THIS_MODULE,
diff --git a/drivers/xen/Kconfig b/drivers/xen/Kconfig
index 5a48ce9..07bec09 100644
--- a/drivers/xen/Kconfig
+++ b/drivers/xen/Kconfig
@@ -71,11 +71,18 @@
but will have no xen contents.
config XEN_XENBUS_FRONTEND
- tristate
+ tristate
+
+config XEN_GNTDEV
+ tristate "userspace grant access device driver"
+ depends on XEN
+ select MMU_NOTIFIER
+ help
+ Allows userspace processes to use grants.
config XEN_PLATFORM_PCI
tristate "xen platform pci device driver"
- depends on XEN_PVHVM
+ depends on XEN_PVHVM && PCI
default m
help
Driver for the Xen PCI Platform device: it is responsible for
diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index 533a199..5088cc2 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -9,11 +9,14 @@
obj-$(CONFIG_XEN_XENCOMM) += xencomm.o
obj-$(CONFIG_XEN_BALLOON) += balloon.o
obj-$(CONFIG_XEN_DEV_EVTCHN) += xen-evtchn.o
+obj-$(CONFIG_XEN_GNTDEV) += xen-gntdev.o
obj-$(CONFIG_XENFS) += xenfs/
obj-$(CONFIG_XEN_SYS_HYPERVISOR) += sys-hypervisor.o
-obj-$(CONFIG_XEN_PLATFORM_PCI) += platform-pci.o
+obj-$(CONFIG_XEN_PLATFORM_PCI) += xen-platform-pci.o
obj-$(CONFIG_SWIOTLB_XEN) += swiotlb-xen.o
obj-$(CONFIG_XEN_DOM0) += pci.o
xen-evtchn-y := evtchn.o
+xen-gntdev-y := gntdev.o
+xen-platform-pci-y := platform-pci.o
diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
new file mode 100644
index 0000000..1e31cdc
--- /dev/null
+++ b/drivers/xen/gntdev.c
@@ -0,0 +1,665 @@
+/******************************************************************************
+ * gntdev.c
+ *
+ * Device for accessing (in user-space) pages that have been granted by other
+ * domains.
+ *
+ * Copyright (c) 2006-2007, D G Murray.
+ * (c) 2009 Gerd Hoffmann <kraxel@redhat.com>
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#undef DEBUG
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/miscdevice.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/mman.h>
+#include <linux/mmu_notifier.h>
+#include <linux/types.h>
+#include <linux/uaccess.h>
+#include <linux/sched.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+
+#include <xen/xen.h>
+#include <xen/grant_table.h>
+#include <xen/gntdev.h>
+#include <asm/xen/hypervisor.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Derek G. Murray <Derek.Murray@cl.cam.ac.uk>, "
+ "Gerd Hoffmann <kraxel@redhat.com>");
+MODULE_DESCRIPTION("User-space granted page access driver");
+
+static int limit = 1024;
+module_param(limit, int, 0644);
+MODULE_PARM_DESC(limit, "Maximum number of grants that may be mapped at "
+ "once by a gntdev instance");
+
+struct gntdev_priv {
+ struct list_head maps;
+ uint32_t used;
+ uint32_t limit;
+ /* lock protects maps from concurrent changes */
+ spinlock_t lock;
+ struct mm_struct *mm;
+ struct mmu_notifier mn;
+};
+
+struct grant_map {
+ struct list_head next;
+ struct gntdev_priv *priv;
+ struct vm_area_struct *vma;
+ int index;
+ int count;
+ int flags;
+ int is_mapped;
+ struct ioctl_gntdev_grant_ref *grants;
+ struct gnttab_map_grant_ref *map_ops;
+ struct gnttab_unmap_grant_ref *unmap_ops;
+ struct page **pages;
+};
+
+/* ------------------------------------------------------------------ */
+
+static void gntdev_print_maps(struct gntdev_priv *priv,
+ char *text, int text_index)
+{
+#ifdef DEBUG
+ struct grant_map *map;
+
+ pr_debug("maps list (priv %p, usage %d/%d)\n",
+ priv, priv->used, priv->limit);
+
+ list_for_each_entry(map, &priv->maps, next)
+ pr_debug(" index %2d, count %2d %s\n",
+ map->index, map->count,
+ map->index == text_index && text ? text : "");
+#endif
+}
+
+static struct grant_map *gntdev_alloc_map(struct gntdev_priv *priv, int count)
+{
+ struct grant_map *add;
+ int i;
+
+ add = kzalloc(sizeof(struct grant_map), GFP_KERNEL);
+ 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->pages = kzalloc(sizeof(add->pages[0]) * count, GFP_KERNEL);
+ if (NULL == add->grants ||
+ NULL == add->map_ops ||
+ NULL == add->unmap_ops ||
+ NULL == add->pages)
+ goto err;
+
+ for (i = 0; i < count; i++) {
+ add->pages[i] = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
+ if (add->pages[i] == NULL)
+ goto err;
+ }
+
+ add->index = 0;
+ add->count = count;
+ add->priv = priv;
+
+ if (add->count + priv->used > priv->limit)
+ goto err;
+
+ return add;
+
+err:
+ if (add->pages)
+ for (i = 0; i < count; i++) {
+ if (add->pages[i])
+ __free_page(add->pages[i]);
+ }
+ kfree(add->pages);
+ kfree(add->grants);
+ kfree(add->map_ops);
+ kfree(add->unmap_ops);
+ kfree(add);
+ return NULL;
+}
+
+static void gntdev_add_map(struct gntdev_priv *priv, struct grant_map *add)
+{
+ struct grant_map *map;
+
+ list_for_each_entry(map, &priv->maps, next) {
+ if (add->index + add->count < map->index) {
+ list_add_tail(&add->next, &map->next);
+ goto done;
+ }
+ add->index = map->index + map->count;
+ }
+ list_add_tail(&add->next, &priv->maps);
+
+done:
+ priv->used += add->count;
+ gntdev_print_maps(priv, "[new]", add->index);
+}
+
+static struct grant_map *gntdev_find_map_index(struct gntdev_priv *priv,
+ int index, int count)
+{
+ struct grant_map *map;
+
+ list_for_each_entry(map, &priv->maps, next) {
+ if (map->index != index)
+ continue;
+ if (map->count != count)
+ continue;
+ return map;
+ }
+ return NULL;
+}
+
+static struct grant_map *gntdev_find_map_vaddr(struct gntdev_priv *priv,
+ unsigned long vaddr)
+{
+ struct grant_map *map;
+
+ list_for_each_entry(map, &priv->maps, next) {
+ if (!map->vma)
+ continue;
+ if (vaddr < map->vma->vm_start)
+ continue;
+ if (vaddr >= map->vma->vm_end)
+ continue;
+ return map;
+ }
+ return NULL;
+}
+
+static int gntdev_del_map(struct grant_map *map)
+{
+ int i;
+
+ if (map->vma)
+ return -EBUSY;
+ for (i = 0; i < map->count; i++)
+ if (map->unmap_ops[i].handle)
+ return -EBUSY;
+
+ map->priv->used -= map->count;
+ list_del(&map->next);
+ return 0;
+}
+
+static void gntdev_free_map(struct grant_map *map)
+{
+ int i;
+
+ if (!map)
+ return;
+
+ if (map->pages)
+ for (i = 0; i < map->count; i++) {
+ if (map->pages[i])
+ __free_page(map->pages[i]);
+ }
+ kfree(map->pages);
+ kfree(map->grants);
+ kfree(map->map_ops);
+ kfree(map->unmap_ops);
+ kfree(map);
+}
+
+/* ------------------------------------------------------------------ */
+
+static int find_grant_ptes(pte_t *pte, pgtable_t token,
+ unsigned long addr, void *data)
+{
+ struct grant_map *map = data;
+ unsigned int pgnr = (addr - map->vma->vm_start) >> PAGE_SHIFT;
+ u64 pte_maddr;
+
+ BUG_ON(pgnr >= map->count);
+ pte_maddr = arbitrary_virt_to_machine(pte).maddr;
+
+ gnttab_set_map_op(&map->map_ops[pgnr], pte_maddr,
+ GNTMAP_contains_pte | map->flags,
+ map->grants[pgnr].ref,
+ map->grants[pgnr].domid);
+ gnttab_set_unmap_op(&map->unmap_ops[pgnr], pte_maddr,
+ GNTMAP_contains_pte | map->flags,
+ 0 /* handle */);
+ return 0;
+}
+
+static int map_grant_pages(struct grant_map *map)
+{
+ int i, err = 0;
+
+ pr_debug("map %d+%d\n", map->index, map->count);
+ err = gnttab_map_refs(map->map_ops, map->pages, map->count);
+ if (err)
+ return err;
+
+ for (i = 0; i < map->count; i++) {
+ if (map->map_ops[i].status)
+ err = -EINVAL;
+ map->unmap_ops[i].handle = map->map_ops[i].handle;
+ }
+ return err;
+}
+
+static int unmap_grant_pages(struct grant_map *map, int offset, int pages)
+{
+ int i, err = 0;
+
+ pr_debug("map %d+%d [%d+%d]\n", map->index, map->count, offset, pages);
+ err = gnttab_unmap_refs(map->unmap_ops + offset, map->pages, pages);
+ if (err)
+ return err;
+
+ for (i = 0; i < pages; i++) {
+ if (map->unmap_ops[offset+i].status)
+ err = -EINVAL;
+ map->unmap_ops[offset+i].handle = 0;
+ }
+ return err;
+}
+
+/* ------------------------------------------------------------------ */
+
+static void gntdev_vma_close(struct vm_area_struct *vma)
+{
+ struct grant_map *map = vma->vm_private_data;
+
+ pr_debug("close %p\n", vma);
+ map->is_mapped = 0;
+ map->vma = NULL;
+ vma->vm_private_data = NULL;
+}
+
+static int gntdev_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+ pr_debug("vaddr %p, pgoff %ld (shouldn't happen)\n",
+ vmf->virtual_address, vmf->pgoff);
+ vmf->flags = VM_FAULT_ERROR;
+ return 0;
+}
+
+static struct vm_operations_struct gntdev_vmops = {
+ .close = gntdev_vma_close,
+ .fault = gntdev_vma_fault,
+};
+
+/* ------------------------------------------------------------------ */
+
+static void mn_invl_range_start(struct mmu_notifier *mn,
+ struct mm_struct *mm,
+ unsigned long start, unsigned long end)
+{
+ struct gntdev_priv *priv = container_of(mn, struct gntdev_priv, mn);
+ struct grant_map *map;
+ unsigned long mstart, mend;
+ int err;
+
+ spin_lock(&priv->lock);
+ list_for_each_entry(map, &priv->maps, next) {
+ if (!map->vma)
+ continue;
+ if (!map->is_mapped)
+ continue;
+ if (map->vma->vm_start >= end)
+ continue;
+ if (map->vma->vm_end <= start)
+ continue;
+ mstart = max(start, map->vma->vm_start);
+ mend = min(end, map->vma->vm_end);
+ pr_debug("map %d+%d (%lx %lx), range %lx %lx, mrange %lx %lx\n",
+ map->index, map->count,
+ map->vma->vm_start, map->vma->vm_end,
+ start, end, mstart, mend);
+ err = unmap_grant_pages(map,
+ (mstart - map->vma->vm_start) >> PAGE_SHIFT,
+ (mend - mstart) >> PAGE_SHIFT);
+ WARN_ON(err);
+ }
+ spin_unlock(&priv->lock);
+}
+
+static void mn_invl_page(struct mmu_notifier *mn,
+ struct mm_struct *mm,
+ unsigned long address)
+{
+ mn_invl_range_start(mn, mm, address, address + PAGE_SIZE);
+}
+
+static void mn_release(struct mmu_notifier *mn,
+ struct mm_struct *mm)
+{
+ struct gntdev_priv *priv = container_of(mn, struct gntdev_priv, mn);
+ struct grant_map *map;
+ int err;
+
+ spin_lock(&priv->lock);
+ list_for_each_entry(map, &priv->maps, next) {
+ if (!map->vma)
+ continue;
+ pr_debug("map %d+%d (%lx %lx)\n",
+ map->index, map->count,
+ map->vma->vm_start, map->vma->vm_end);
+ err = unmap_grant_pages(map, /* offset */ 0, map->count);
+ WARN_ON(err);
+ }
+ spin_unlock(&priv->lock);
+}
+
+struct mmu_notifier_ops gntdev_mmu_ops = {
+ .release = mn_release,
+ .invalidate_page = mn_invl_page,
+ .invalidate_range_start = mn_invl_range_start,
+};
+
+/* ------------------------------------------------------------------ */
+
+static int gntdev_open(struct inode *inode, struct file *flip)
+{
+ struct gntdev_priv *priv;
+ int ret = 0;
+
+ priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+ if (!priv)
+ return -ENOMEM;
+
+ INIT_LIST_HEAD(&priv->maps);
+ spin_lock_init(&priv->lock);
+ priv->limit = limit;
+
+ priv->mm = get_task_mm(current);
+ if (!priv->mm) {
+ kfree(priv);
+ return -ENOMEM;
+ }
+ priv->mn.ops = &gntdev_mmu_ops;
+ ret = mmu_notifier_register(&priv->mn, priv->mm);
+ mmput(priv->mm);
+
+ if (ret) {
+ kfree(priv);
+ return ret;
+ }
+
+ flip->private_data = priv;
+ pr_debug("priv %p\n", priv);
+
+ return 0;
+}
+
+static int gntdev_release(struct inode *inode, struct file *flip)
+{
+ struct gntdev_priv *priv = flip->private_data;
+ struct grant_map *map;
+ int err;
+
+ pr_debug("priv %p\n", priv);
+
+ spin_lock(&priv->lock);
+ while (!list_empty(&priv->maps)) {
+ map = list_entry(priv->maps.next, struct grant_map, next);
+ err = gntdev_del_map(map);
+ if (WARN_ON(err))
+ gntdev_free_map(map);
+
+ }
+ spin_unlock(&priv->lock);
+
+ mmu_notifier_unregister(&priv->mn, priv->mm);
+ kfree(priv);
+ return 0;
+}
+
+static long gntdev_ioctl_map_grant_ref(struct gntdev_priv *priv,
+ struct ioctl_gntdev_map_grant_ref __user *u)
+{
+ struct ioctl_gntdev_map_grant_ref op;
+ struct grant_map *map;
+ int err;
+
+ if (copy_from_user(&op, u, sizeof(op)) != 0)
+ return -EFAULT;
+ pr_debug("priv %p, add %d\n", priv, op.count);
+ if (unlikely(op.count <= 0))
+ return -EINVAL;
+ if (unlikely(op.count > priv->limit))
+ return -EINVAL;
+
+ err = -ENOMEM;
+ map = gntdev_alloc_map(priv, op.count);
+ if (!map)
+ return err;
+ if (copy_from_user(map->grants, &u->refs,
+ sizeof(map->grants[0]) * op.count) != 0) {
+ gntdev_free_map(map);
+ return err;
+ }
+
+ spin_lock(&priv->lock);
+ gntdev_add_map(priv, map);
+ op.index = map->index << PAGE_SHIFT;
+ spin_unlock(&priv->lock);
+
+ if (copy_to_user(u, &op, sizeof(op)) != 0) {
+ spin_lock(&priv->lock);
+ gntdev_del_map(map);
+ spin_unlock(&priv->lock);
+ gntdev_free_map(map);
+ return err;
+ }
+ return 0;
+}
+
+static long gntdev_ioctl_unmap_grant_ref(struct gntdev_priv *priv,
+ struct ioctl_gntdev_unmap_grant_ref __user *u)
+{
+ struct ioctl_gntdev_unmap_grant_ref op;
+ struct grant_map *map;
+ int err = -ENOENT;
+
+ if (copy_from_user(&op, u, sizeof(op)) != 0)
+ return -EFAULT;
+ pr_debug("priv %p, del %d+%d\n", priv, (int)op.index, (int)op.count);
+
+ spin_lock(&priv->lock);
+ map = gntdev_find_map_index(priv, op.index >> PAGE_SHIFT, op.count);
+ if (map)
+ err = gntdev_del_map(map);
+ spin_unlock(&priv->lock);
+ if (!err)
+ gntdev_free_map(map);
+ return err;
+}
+
+static long gntdev_ioctl_get_offset_for_vaddr(struct gntdev_priv *priv,
+ struct ioctl_gntdev_get_offset_for_vaddr __user *u)
+{
+ struct ioctl_gntdev_get_offset_for_vaddr op;
+ struct grant_map *map;
+
+ if (copy_from_user(&op, u, sizeof(op)) != 0)
+ return -EFAULT;
+ pr_debug("priv %p, offset for vaddr %lx\n", priv, (unsigned long)op.vaddr);
+
+ spin_lock(&priv->lock);
+ map = gntdev_find_map_vaddr(priv, op.vaddr);
+ if (map == NULL ||
+ map->vma->vm_start != op.vaddr) {
+ spin_unlock(&priv->lock);
+ return -EINVAL;
+ }
+ op.offset = map->index << PAGE_SHIFT;
+ op.count = map->count;
+ spin_unlock(&priv->lock);
+
+ if (copy_to_user(u, &op, sizeof(op)) != 0)
+ return -EFAULT;
+ return 0;
+}
+
+static long gntdev_ioctl_set_max_grants(struct gntdev_priv *priv,
+ struct ioctl_gntdev_set_max_grants __user *u)
+{
+ struct ioctl_gntdev_set_max_grants op;
+
+ if (copy_from_user(&op, u, sizeof(op)) != 0)
+ return -EFAULT;
+ pr_debug("priv %p, limit %d\n", priv, op.count);
+ if (op.count > limit)
+ return -E2BIG;
+
+ spin_lock(&priv->lock);
+ priv->limit = op.count;
+ spin_unlock(&priv->lock);
+ return 0;
+}
+
+static long gntdev_ioctl(struct file *flip,
+ unsigned int cmd, unsigned long arg)
+{
+ struct gntdev_priv *priv = flip->private_data;
+ void __user *ptr = (void __user *)arg;
+
+ switch (cmd) {
+ case IOCTL_GNTDEV_MAP_GRANT_REF:
+ return gntdev_ioctl_map_grant_ref(priv, ptr);
+
+ case IOCTL_GNTDEV_UNMAP_GRANT_REF:
+ return gntdev_ioctl_unmap_grant_ref(priv, ptr);
+
+ case IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR:
+ return gntdev_ioctl_get_offset_for_vaddr(priv, ptr);
+
+ case IOCTL_GNTDEV_SET_MAX_GRANTS:
+ return gntdev_ioctl_set_max_grants(priv, ptr);
+
+ default:
+ pr_debug("priv %p, unknown cmd %x\n", priv, cmd);
+ return -ENOIOCTLCMD;
+ }
+
+ return 0;
+}
+
+static int gntdev_mmap(struct file *flip, struct vm_area_struct *vma)
+{
+ struct gntdev_priv *priv = flip->private_data;
+ int index = vma->vm_pgoff;
+ int count = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
+ struct grant_map *map;
+ int err = -EINVAL;
+
+ if ((vma->vm_flags & VM_WRITE) && !(vma->vm_flags & VM_SHARED))
+ return -EINVAL;
+
+ pr_debug("map %d+%d at %lx (pgoff %lx)\n",
+ index, count, vma->vm_start, vma->vm_pgoff);
+
+ spin_lock(&priv->lock);
+ map = gntdev_find_map_index(priv, index, count);
+ if (!map)
+ goto unlock_out;
+ if (map->vma)
+ goto unlock_out;
+ if (priv->mm != vma->vm_mm) {
+ printk(KERN_WARNING "Huh? Other mm?\n");
+ goto unlock_out;
+ }
+
+ vma->vm_ops = &gntdev_vmops;
+
+ vma->vm_flags |= VM_RESERVED|VM_DONTCOPY|VM_DONTEXPAND|VM_PFNMAP;
+
+ vma->vm_private_data = map;
+ map->vma = vma;
+
+ map->flags = GNTMAP_host_map | GNTMAP_application_map;
+ if (!(vma->vm_flags & VM_WRITE))
+ map->flags |= GNTMAP_readonly;
+
+ spin_unlock(&priv->lock);
+
+ err = apply_to_page_range(vma->vm_mm, vma->vm_start,
+ vma->vm_end - vma->vm_start,
+ find_grant_ptes, map);
+ if (err) {
+ printk(KERN_WARNING "find_grant_ptes() failure.\n");
+ return err;
+ }
+
+ err = map_grant_pages(map);
+ if (err) {
+ printk(KERN_WARNING "map_grant_pages() failure.\n");
+ return err;
+ }
+
+ map->is_mapped = 1;
+
+ return 0;
+
+unlock_out:
+ spin_unlock(&priv->lock);
+ return err;
+}
+
+static const struct file_operations gntdev_fops = {
+ .owner = THIS_MODULE,
+ .open = gntdev_open,
+ .release = gntdev_release,
+ .mmap = gntdev_mmap,
+ .unlocked_ioctl = gntdev_ioctl
+};
+
+static struct miscdevice gntdev_miscdev = {
+ .minor = MISC_DYNAMIC_MINOR,
+ .name = "xen/gntdev",
+ .fops = &gntdev_fops,
+};
+
+/* ------------------------------------------------------------------ */
+
+static int __init gntdev_init(void)
+{
+ int err;
+
+ if (!xen_domain())
+ return -ENODEV;
+
+ err = misc_register(&gntdev_miscdev);
+ if (err != 0) {
+ printk(KERN_ERR "Could not register gntdev device\n");
+ return err;
+ }
+ return 0;
+}
+
+static void __exit gntdev_exit(void)
+{
+ misc_deregister(&gntdev_miscdev);
+}
+
+module_init(gntdev_init);
+module_exit(gntdev_exit);
+
+/* ------------------------------------------------------------------ */
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 6c45318..9ef54eb 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -447,6 +447,52 @@
}
EXPORT_SYMBOL_GPL(gnttab_max_grant_frames);
+int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
+ struct page **pages, unsigned int count)
+{
+ int i, ret;
+ pte_t *pte;
+ unsigned long mfn;
+
+ ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, map_ops, count);
+ if (ret)
+ return ret;
+
+ for (i = 0; i < count; i++) {
+ /* m2p override only supported for GNTMAP_contains_pte mappings */
+ if (!(map_ops[i].flags & GNTMAP_contains_pte))
+ continue;
+ pte = (pte_t *) (mfn_to_virt(PFN_DOWN(map_ops[i].host_addr)) +
+ (map_ops[i].host_addr & ~PAGE_MASK));
+ mfn = pte_mfn(*pte);
+ ret = m2p_add_override(mfn, pages[i]);
+ if (ret)
+ return ret;
+ }
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(gnttab_map_refs);
+
+int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
+ struct page **pages, unsigned int count)
+{
+ int i, ret;
+
+ ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, unmap_ops, count);
+ if (ret)
+ return ret;
+
+ for (i = 0; i < count; i++) {
+ ret = m2p_remove_override(pages[i]);
+ if (ret)
+ return ret;
+ }
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(gnttab_unmap_refs);
+
static int gnttab_map(unsigned int start_idx, unsigned int end_idx)
{
struct gnttab_setup_table setup;
diff --git a/drivers/xen/platform-pci.c b/drivers/xen/platform-pci.c
index c01b5dd..afbe041 100644
--- a/drivers/xen/platform-pci.c
+++ b/drivers/xen/platform-pci.c
@@ -105,7 +105,7 @@
const struct pci_device_id *ent)
{
int i, ret;
- long ioaddr, iolen;
+ long ioaddr;
long mmio_addr, mmio_len;
unsigned int max_nr_gframes;
@@ -114,7 +114,6 @@
return i;
ioaddr = pci_resource_start(pdev, 0);
- iolen = pci_resource_len(pdev, 0);
mmio_addr = pci_resource_start(pdev, 1);
mmio_len = pci_resource_len(pdev, 1);
@@ -125,19 +124,13 @@
goto pci_out;
}
- if (request_mem_region(mmio_addr, mmio_len, DRV_NAME) == NULL) {
- dev_err(&pdev->dev, "MEM I/O resource 0x%lx @ 0x%lx busy\n",
- mmio_addr, mmio_len);
- ret = -EBUSY;
+ ret = pci_request_region(pdev, 1, DRV_NAME);
+ if (ret < 0)
goto pci_out;
- }
- if (request_region(ioaddr, iolen, DRV_NAME) == NULL) {
- dev_err(&pdev->dev, "I/O resource 0x%lx @ 0x%lx busy\n",
- iolen, ioaddr);
- ret = -EBUSY;
+ ret = pci_request_region(pdev, 0, DRV_NAME);
+ if (ret < 0)
goto mem_out;
- }
platform_mmio = mmio_addr;
platform_mmiolen = mmio_len;
@@ -169,9 +162,9 @@
return 0;
out:
- release_region(ioaddr, iolen);
+ pci_release_region(pdev, 0);
mem_out:
- release_mem_region(mmio_addr, mmio_len);
+ pci_release_region(pdev, 1);
pci_out:
pci_disable_device(pdev);
return ret;
diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c
index a3bcec7..1c8c6cc 100644
--- a/fs/afs/cmservice.c
+++ b/fs/afs/cmservice.c
@@ -289,7 +289,7 @@
call->server = server;
INIT_WORK(&call->work, SRXAFSCB_CallBack);
- schedule_work(&call->work);
+ queue_work(afs_wq, &call->work);
return 0;
}
@@ -336,7 +336,7 @@
call->server = server;
INIT_WORK(&call->work, SRXAFSCB_InitCallBackState);
- schedule_work(&call->work);
+ queue_work(afs_wq, &call->work);
return 0;
}
@@ -367,7 +367,7 @@
call->server = server;
INIT_WORK(&call->work, SRXAFSCB_InitCallBackState);
- schedule_work(&call->work);
+ queue_work(afs_wq, &call->work);
return 0;
}
@@ -400,7 +400,7 @@
call->state = AFS_CALL_REPLYING;
INIT_WORK(&call->work, SRXAFSCB_Probe);
- schedule_work(&call->work);
+ queue_work(afs_wq, &call->work);
return 0;
}
@@ -496,7 +496,7 @@
call->state = AFS_CALL_REPLYING;
INIT_WORK(&call->work, SRXAFSCB_ProbeUuid);
- schedule_work(&call->work);
+ queue_work(afs_wq, &call->work);
return 0;
}
@@ -580,6 +580,6 @@
call->state = AFS_CALL_REPLYING;
INIT_WORK(&call->work, SRXAFSCB_TellMeAboutYourself);
- schedule_work(&call->work);
+ queue_work(afs_wq, &call->work);
return 0;
}
diff --git a/fs/afs/dir.c b/fs/afs/dir.c
index e6a4ab9..20c106f 100644
--- a/fs/afs/dir.c
+++ b/fs/afs/dir.c
@@ -66,6 +66,7 @@
.d_revalidate = afs_d_revalidate,
.d_delete = afs_d_delete,
.d_release = afs_d_release,
+ .d_automount = afs_d_automount,
};
#define AFS_DIR_HASHTBL_SIZE 128
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index 0747339..db66c52 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -184,7 +184,8 @@
inode->i_generation = 0;
set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags);
- inode->i_flags |= S_NOATIME;
+ set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags);
+ inode->i_flags |= S_AUTOMOUNT | S_NOATIME;
unlock_new_inode(inode);
_leave(" = %p", inode);
return inode;
diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index ab6db5a..5a9b684 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -577,6 +577,7 @@
/*
* main.c
*/
+extern struct workqueue_struct *afs_wq;
extern struct afs_uuid afs_uuid;
/*
@@ -591,6 +592,7 @@
extern const struct inode_operations afs_autocell_inode_operations;
extern const struct file_operations afs_mntpt_file_operations;
+extern struct vfsmount *afs_d_automount(struct path *);
extern int afs_mntpt_check_symlink(struct afs_vnode *, struct key *);
extern void afs_mntpt_kill_timer(void);
diff --git a/fs/afs/main.c b/fs/afs/main.c
index cfd1cbe..42dd2e4 100644
--- a/fs/afs/main.c
+++ b/fs/afs/main.c
@@ -30,6 +30,7 @@
MODULE_PARM_DESC(rootcell, "root AFS cell name and VL server IP addr list");
struct afs_uuid afs_uuid;
+struct workqueue_struct *afs_wq;
/*
* get a client UUID
@@ -87,10 +88,16 @@
if (ret < 0)
return ret;
+ /* create workqueue */
+ ret = -ENOMEM;
+ afs_wq = alloc_workqueue("afs", 0, 0);
+ if (!afs_wq)
+ return ret;
+
/* register the /proc stuff */
ret = afs_proc_init();
if (ret < 0)
- return ret;
+ goto error_proc;
#ifdef CONFIG_AFS_FSCACHE
/* we want to be able to cache */
@@ -140,6 +147,8 @@
error_cache:
#endif
afs_proc_cleanup();
+error_proc:
+ destroy_workqueue(afs_wq);
rcu_barrier();
printk(KERN_ERR "kAFS: failed to register: %d\n", ret);
return ret;
@@ -163,7 +172,7 @@
afs_purge_servers();
afs_callback_update_kill();
afs_vlocation_purge();
- flush_scheduled_work();
+ destroy_workqueue(afs_wq);
afs_cell_purge();
#ifdef CONFIG_AFS_FSCACHE
fscache_unregister_netfs(&afs_cache_netfs);
diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c
index 6153417..aa59184 100644
--- a/fs/afs/mntpt.c
+++ b/fs/afs/mntpt.c
@@ -24,7 +24,6 @@
struct dentry *dentry,
struct nameidata *nd);
static int afs_mntpt_open(struct inode *inode, struct file *file);
-static void *afs_mntpt_follow_link(struct dentry *dentry, struct nameidata *nd);
static void afs_mntpt_expiry_timed_out(struct work_struct *work);
const struct file_operations afs_mntpt_file_operations = {
@@ -34,13 +33,11 @@
const struct inode_operations afs_mntpt_inode_operations = {
.lookup = afs_mntpt_lookup,
- .follow_link = afs_mntpt_follow_link,
.readlink = page_readlink,
.getattr = afs_getattr,
};
const struct inode_operations afs_autocell_inode_operations = {
- .follow_link = afs_mntpt_follow_link,
.getattr = afs_getattr,
};
@@ -88,6 +85,7 @@
_debug("symlink is a mountpoint");
spin_lock(&vnode->lock);
set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags);
+ vnode->vfs_inode.i_flags |= S_AUTOMOUNT;
spin_unlock(&vnode->lock);
}
@@ -238,52 +236,24 @@
}
/*
- * follow a link from a mountpoint directory, thus causing it to be mounted
+ * handle an automount point
*/
-static void *afs_mntpt_follow_link(struct dentry *dentry, struct nameidata *nd)
+struct vfsmount *afs_d_automount(struct path *path)
{
struct vfsmount *newmnt;
- int err;
- _enter("%p{%s},{%s:%p{%s},}",
- dentry,
- dentry->d_name.name,
- nd->path.mnt->mnt_devname,
- dentry,
- nd->path.dentry->d_name.name);
+ _enter("{%s,%s}", path->mnt->mnt_devname, path->dentry->d_name.name);
- dput(nd->path.dentry);
- nd->path.dentry = dget(dentry);
+ newmnt = afs_mntpt_do_automount(path->dentry);
+ if (IS_ERR(newmnt))
+ return newmnt;
- newmnt = afs_mntpt_do_automount(nd->path.dentry);
- if (IS_ERR(newmnt)) {
- path_put(&nd->path);
- return (void *)newmnt;
- }
-
- mntget(newmnt);
- err = do_add_mount(newmnt, &nd->path, MNT_SHRINKABLE, &afs_vfsmounts);
- switch (err) {
- case 0:
- path_put(&nd->path);
- nd->path.mnt = newmnt;
- nd->path.dentry = dget(newmnt->mnt_root);
- schedule_delayed_work(&afs_mntpt_expiry_timer,
- afs_mntpt_expiry_timeout * HZ);
- break;
- case -EBUSY:
- /* someone else made a mount here whilst we were busy */
- while (d_mountpoint(nd->path.dentry) &&
- follow_down(&nd->path))
- ;
- err = 0;
- default:
- mntput(newmnt);
- break;
- }
-
- _leave(" = %d", err);
- return ERR_PTR(err);
+ mntget(newmnt); /* prevent immediate expiration */
+ mnt_set_expiry(newmnt, &afs_vfsmounts);
+ queue_delayed_work(afs_wq, &afs_mntpt_expiry_timer,
+ afs_mntpt_expiry_timeout * HZ);
+ _leave(" = %p {%s}", newmnt, newmnt->mnt_devname);
+ return newmnt;
}
/*
@@ -295,8 +265,8 @@
if (!list_empty(&afs_vfsmounts)) {
mark_mounts_for_expiry(&afs_vfsmounts);
- schedule_delayed_work(&afs_mntpt_expiry_timer,
- afs_mntpt_expiry_timeout * HZ);
+ queue_delayed_work(afs_wq, &afs_mntpt_expiry_timer,
+ afs_mntpt_expiry_timeout * HZ);
}
_leave("");
@@ -310,6 +280,5 @@
_enter("");
ASSERT(list_empty(&afs_vfsmounts));
- cancel_delayed_work(&afs_mntpt_expiry_timer);
- flush_scheduled_work();
+ cancel_delayed_work_sync(&afs_mntpt_expiry_timer);
}
diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c
index 654d8fd..e45a323 100644
--- a/fs/afs/rxrpc.c
+++ b/fs/afs/rxrpc.c
@@ -410,7 +410,7 @@
if (!call) {
/* its an incoming call for our callback service */
skb_queue_tail(&afs_incoming_calls, skb);
- schedule_work(&afs_collect_incoming_call_work);
+ queue_work(afs_wq, &afs_collect_incoming_call_work);
} else {
/* route the messages directly to the appropriate call */
skb_queue_tail(&call->rx_queue, skb);
diff --git a/fs/afs/server.c b/fs/afs/server.c
index 9fdc7fe..d59b751 100644
--- a/fs/afs/server.c
+++ b/fs/afs/server.c
@@ -238,8 +238,8 @@
if (atomic_read(&server->usage) == 0) {
list_move_tail(&server->grave, &afs_server_graveyard);
server->time_of_death = get_seconds();
- schedule_delayed_work(&afs_server_reaper,
- afs_server_timeout * HZ);
+ queue_delayed_work(afs_wq, &afs_server_reaper,
+ afs_server_timeout * HZ);
}
spin_unlock(&afs_server_graveyard_lock);
_leave(" [dead]");
@@ -285,10 +285,11 @@
expiry = server->time_of_death + afs_server_timeout;
if (expiry > now) {
delay = (expiry - now) * HZ;
- if (!schedule_delayed_work(&afs_server_reaper, delay)) {
+ if (!queue_delayed_work(afs_wq, &afs_server_reaper,
+ delay)) {
cancel_delayed_work(&afs_server_reaper);
- schedule_delayed_work(&afs_server_reaper,
- delay);
+ queue_delayed_work(afs_wq, &afs_server_reaper,
+ delay);
}
break;
}
@@ -323,5 +324,5 @@
{
afs_server_timeout = 0;
cancel_delayed_work(&afs_server_reaper);
- schedule_delayed_work(&afs_server_reaper, 0);
+ queue_delayed_work(afs_wq, &afs_server_reaper, 0);
}
diff --git a/fs/afs/vlocation.c b/fs/afs/vlocation.c
index 9ac260d..431984d 100644
--- a/fs/afs/vlocation.c
+++ b/fs/afs/vlocation.c
@@ -507,8 +507,8 @@
_debug("buried");
list_move_tail(&vl->grave, &afs_vlocation_graveyard);
vl->time_of_death = get_seconds();
- schedule_delayed_work(&afs_vlocation_reap,
- afs_vlocation_timeout * HZ);
+ queue_delayed_work(afs_wq, &afs_vlocation_reap,
+ afs_vlocation_timeout * HZ);
/* suspend updates on this record */
if (!list_empty(&vl->update)) {
@@ -561,11 +561,11 @@
if (expiry > now) {
delay = (expiry - now) * HZ;
_debug("delay %lu", delay);
- if (!schedule_delayed_work(&afs_vlocation_reap,
- delay)) {
+ if (!queue_delayed_work(afs_wq, &afs_vlocation_reap,
+ delay)) {
cancel_delayed_work(&afs_vlocation_reap);
- schedule_delayed_work(&afs_vlocation_reap,
- delay);
+ queue_delayed_work(afs_wq, &afs_vlocation_reap,
+ delay);
}
break;
}
@@ -620,7 +620,7 @@
destroy_workqueue(afs_vlocation_update_worker);
cancel_delayed_work(&afs_vlocation_reap);
- schedule_delayed_work(&afs_vlocation_reap, 0);
+ queue_delayed_work(afs_wq, &afs_vlocation_reap, 0);
}
/*
diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c
index cbe57f3..c5567cb 100644
--- a/fs/anon_inodes.c
+++ b/fs/anon_inodes.c
@@ -233,7 +233,7 @@
return 0;
err_mntput:
- mntput_long(anon_inode_mnt);
+ mntput(anon_inode_mnt);
err_unregister_filesystem:
unregister_filesystem(&anon_inode_fs_type);
err_exit:
diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 0fffe1c..1f016bf 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -99,7 +99,6 @@
};
#define AUTOFS_INF_EXPIRING (1<<0) /* dentry is in the process of expiring */
-#define AUTOFS_INF_MOUNTPOINT (1<<1) /* mountpoint status for direct expire */
#define AUTOFS_INF_PENDING (1<<2) /* dentry pending mount */
struct autofs_wait_queue {
@@ -176,13 +175,6 @@
return 0;
}
-static inline void autofs4_copy_atime(struct file *src, struct file *dst)
-{
- dst->f_path.dentry->d_inode->i_atime =
- src->f_path.dentry->d_inode->i_atime;
- return;
-}
-
struct inode *autofs4_get_inode(struct super_block *, struct autofs_info *);
void autofs4_free_ino(struct autofs_info *);
@@ -212,11 +204,83 @@
extern const struct inode_operations autofs4_symlink_inode_operations;
extern const struct inode_operations autofs4_dir_inode_operations;
-extern const struct inode_operations autofs4_root_inode_operations;
-extern const struct inode_operations autofs4_indirect_root_inode_operations;
-extern const struct inode_operations autofs4_direct_root_inode_operations;
extern const struct file_operations autofs4_dir_operations;
extern const struct file_operations autofs4_root_operations;
+extern const struct dentry_operations autofs4_dentry_operations;
+
+/* VFS automount flags management functions */
+
+static inline void __managed_dentry_set_automount(struct dentry *dentry)
+{
+ dentry->d_flags |= DCACHE_NEED_AUTOMOUNT;
+}
+
+static inline void managed_dentry_set_automount(struct dentry *dentry)
+{
+ spin_lock(&dentry->d_lock);
+ __managed_dentry_set_automount(dentry);
+ spin_unlock(&dentry->d_lock);
+}
+
+static inline void __managed_dentry_clear_automount(struct dentry *dentry)
+{
+ dentry->d_flags &= ~DCACHE_NEED_AUTOMOUNT;
+}
+
+static inline void managed_dentry_clear_automount(struct dentry *dentry)
+{
+ spin_lock(&dentry->d_lock);
+ __managed_dentry_clear_automount(dentry);
+ spin_unlock(&dentry->d_lock);
+}
+
+static inline void __managed_dentry_set_transit(struct dentry *dentry)
+{
+ dentry->d_flags |= DCACHE_MANAGE_TRANSIT;
+}
+
+static inline void managed_dentry_set_transit(struct dentry *dentry)
+{
+ spin_lock(&dentry->d_lock);
+ __managed_dentry_set_transit(dentry);
+ spin_unlock(&dentry->d_lock);
+}
+
+static inline void __managed_dentry_clear_transit(struct dentry *dentry)
+{
+ dentry->d_flags &= ~DCACHE_MANAGE_TRANSIT;
+}
+
+static inline void managed_dentry_clear_transit(struct dentry *dentry)
+{
+ spin_lock(&dentry->d_lock);
+ __managed_dentry_clear_transit(dentry);
+ spin_unlock(&dentry->d_lock);
+}
+
+static inline void __managed_dentry_set_managed(struct dentry *dentry)
+{
+ dentry->d_flags |= (DCACHE_NEED_AUTOMOUNT|DCACHE_MANAGE_TRANSIT);
+}
+
+static inline void managed_dentry_set_managed(struct dentry *dentry)
+{
+ spin_lock(&dentry->d_lock);
+ __managed_dentry_set_managed(dentry);
+ spin_unlock(&dentry->d_lock);
+}
+
+static inline void __managed_dentry_clear_managed(struct dentry *dentry)
+{
+ dentry->d_flags &= ~(DCACHE_NEED_AUTOMOUNT|DCACHE_MANAGE_TRANSIT);
+}
+
+static inline void managed_dentry_clear_managed(struct dentry *dentry)
+{
+ spin_lock(&dentry->d_lock);
+ __managed_dentry_clear_managed(dentry);
+ spin_unlock(&dentry->d_lock);
+}
/* Initializing function */
@@ -229,19 +293,6 @@
int autofs4_wait_release(struct autofs_sb_info *,autofs_wqt_t,int);
void autofs4_catatonic_mode(struct autofs_sb_info *);
-static inline int autofs4_follow_mount(struct path *path)
-{
- int res = 0;
-
- while (d_mountpoint(path->dentry)) {
- int followed = follow_down(path);
- if (!followed)
- break;
- res = 1;
- }
- return res;
-}
-
static inline u32 autofs4_get_dev(struct autofs_sb_info *sbi)
{
return new_encode_dev(sbi->sb->s_dev);
diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c
index eff9a41..1442da4 100644
--- a/fs/autofs4/dev-ioctl.c
+++ b/fs/autofs4/dev-ioctl.c
@@ -551,7 +551,7 @@
err = have_submounts(path.dentry);
- if (follow_down(&path))
+ if (follow_down_one(&path))
magic = path.mnt->mnt_sb->s_magic;
}
diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index cc1d013..3ed79d7 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -26,10 +26,6 @@
if (ino == NULL)
return 0;
- /* No point expiring a pending mount */
- if (ino->flags & AUTOFS_INF_PENDING)
- return 0;
-
if (!do_now) {
/* Too young to die */
if (!timeout || time_after(ino->last_used + timeout, now))
@@ -56,7 +52,7 @@
path_get(&path);
- if (!follow_down(&path))
+ if (!follow_down_one(&path))
goto done;
if (is_autofs4_dentry(path.dentry)) {
@@ -283,6 +279,7 @@
unsigned long timeout;
struct dentry *root = dget(sb->s_root);
int do_now = how & AUTOFS_EXP_IMMEDIATE;
+ struct autofs_info *ino;
if (!root)
return NULL;
@@ -291,19 +288,21 @@
timeout = sbi->exp_timeout;
spin_lock(&sbi->fs_lock);
+ ino = autofs4_dentry_ino(root);
+ /* No point expiring a pending mount */
+ if (ino->flags & AUTOFS_INF_PENDING) {
+ spin_unlock(&sbi->fs_lock);
+ return NULL;
+ }
+ managed_dentry_set_transit(root);
if (!autofs4_direct_busy(mnt, root, timeout, do_now)) {
struct autofs_info *ino = autofs4_dentry_ino(root);
- if (d_mountpoint(root)) {
- ino->flags |= AUTOFS_INF_MOUNTPOINT;
- spin_lock(&root->d_lock);
- root->d_flags &= ~DCACHE_MOUNTED;
- spin_unlock(&root->d_lock);
- }
ino->flags |= AUTOFS_INF_EXPIRING;
init_completion(&ino->expire_complete);
spin_unlock(&sbi->fs_lock);
return root;
}
+ managed_dentry_clear_transit(root);
spin_unlock(&sbi->fs_lock);
dput(root);
@@ -340,6 +339,10 @@
while ((dentry = get_next_positive_dentry(dentry, root))) {
spin_lock(&sbi->fs_lock);
ino = autofs4_dentry_ino(dentry);
+ /* No point expiring a pending mount */
+ if (ino->flags & AUTOFS_INF_PENDING)
+ goto cont;
+ managed_dentry_set_transit(dentry);
/*
* Case 1: (i) indirect mount or top level pseudo direct mount
@@ -399,6 +402,8 @@
}
}
next:
+ managed_dentry_clear_transit(dentry);
+cont:
spin_unlock(&sbi->fs_lock);
}
return NULL;
@@ -479,6 +484,8 @@
spin_lock(&sbi->fs_lock);
ino = autofs4_dentry_ino(dentry);
ino->flags &= ~AUTOFS_INF_EXPIRING;
+ if (!d_unhashed(dentry))
+ managed_dentry_clear_transit(dentry);
complete_all(&ino->expire_complete);
spin_unlock(&sbi->fs_lock);
@@ -504,18 +511,18 @@
ret = autofs4_wait(sbi, dentry, NFY_EXPIRE);
spin_lock(&sbi->fs_lock);
- if (ino->flags & AUTOFS_INF_MOUNTPOINT) {
- spin_lock(&sb->s_root->d_lock);
- /*
- * If we haven't been expired away, then reset
- * mounted status.
- */
- if (mnt->mnt_parent != mnt)
- sb->s_root->d_flags |= DCACHE_MOUNTED;
- spin_unlock(&sb->s_root->d_lock);
- ino->flags &= ~AUTOFS_INF_MOUNTPOINT;
- }
ino->flags &= ~AUTOFS_INF_EXPIRING;
+ spin_lock(&dentry->d_lock);
+ if (ret)
+ __managed_dentry_clear_transit(dentry);
+ else {
+ if ((IS_ROOT(dentry) ||
+ (autofs_type_indirect(sbi->type) &&
+ IS_ROOT(dentry->d_parent))) &&
+ !(dentry->d_flags & DCACHE_NEED_AUTOMOUNT))
+ __managed_dentry_set_automount(dentry);
+ }
+ spin_unlock(&dentry->d_lock);
complete_all(&ino->expire_complete);
spin_unlock(&sbi->fs_lock);
dput(dentry);
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index a7bdb9d..9e1a9da 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -45,7 +45,6 @@
if (!reinit) {
ino->flags = 0;
- ino->inode = NULL;
ino->dentry = NULL;
ino->size = 0;
INIT_LIST_HEAD(&ino->active);
@@ -76,19 +75,8 @@
void autofs4_free_ino(struct autofs_info *ino)
{
- struct autofs_info *p_ino;
-
if (ino->dentry) {
ino->dentry->d_fsdata = NULL;
- if (ino->dentry->d_inode) {
- struct dentry *parent = ino->dentry->d_parent;
- if (atomic_dec_and_test(&ino->count)) {
- p_ino = autofs4_dentry_ino(parent);
- if (p_ino && parent != ino->dentry)
- atomic_dec(&p_ino->count);
- }
- dput(ino->dentry);
- }
ino->dentry = NULL;
}
if (ino->free)
@@ -251,10 +239,6 @@
return ino;
}
-static const struct dentry_operations autofs4_sb_dentry_operations = {
- .d_release = autofs4_dentry_release,
-};
-
int autofs4_fill_super(struct super_block *s, void *data, int silent)
{
struct inode * root_inode;
@@ -292,6 +276,7 @@
s->s_blocksize_bits = 10;
s->s_magic = AUTOFS_SUPER_MAGIC;
s->s_op = &autofs4_sops;
+ s->s_d_op = &autofs4_dentry_operations;
s->s_time_gran = 1;
/*
@@ -309,7 +294,6 @@
goto fail_iput;
pipe = NULL;
- d_set_d_op(root, &autofs4_sb_dentry_operations);
root->d_fsdata = ino;
/* Can this call block? */
@@ -320,10 +304,11 @@
goto fail_dput;
}
+ if (autofs_type_trigger(sbi->type))
+ __managed_dentry_set_managed(root);
+
root_inode->i_fop = &autofs4_root_operations;
- root_inode->i_op = autofs_type_trigger(sbi->type) ?
- &autofs4_direct_root_inode_operations :
- &autofs4_indirect_root_inode_operations;
+ root_inode->i_op = &autofs4_dir_inode_operations;
/* Couldn't this be tested earlier? */
if (sbi->max_proto < AUTOFS_MIN_PROTO_VERSION ||
@@ -391,7 +376,6 @@
if (inode == NULL)
return NULL;
- inf->inode = inode;
inode->i_mode = inf->mode;
if (sb->s_root) {
inode->i_uid = sb->s_root->d_inode->i_uid;
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 651e4ef..1dba035 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -35,10 +35,8 @@
#endif
static int autofs4_dir_open(struct inode *inode, struct file *file);
static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *);
-static void *autofs4_follow_link(struct dentry *, struct nameidata *);
-
-#define TRIGGER_FLAGS (LOOKUP_CONTINUE | LOOKUP_DIRECTORY)
-#define TRIGGER_INTENTS (LOOKUP_OPEN | LOOKUP_CREATE)
+static struct vfsmount *autofs4_d_automount(struct path *);
+static int autofs4_d_manage(struct dentry *, bool, bool);
const struct file_operations autofs4_root_operations = {
.open = dcache_dir_open,
@@ -60,22 +58,6 @@
.llseek = dcache_dir_lseek,
};
-const struct inode_operations autofs4_indirect_root_inode_operations = {
- .lookup = autofs4_lookup,
- .unlink = autofs4_dir_unlink,
- .symlink = autofs4_dir_symlink,
- .mkdir = autofs4_dir_mkdir,
- .rmdir = autofs4_dir_rmdir,
-};
-
-const struct inode_operations autofs4_direct_root_inode_operations = {
- .lookup = autofs4_lookup,
- .unlink = autofs4_dir_unlink,
- .mkdir = autofs4_dir_mkdir,
- .rmdir = autofs4_dir_rmdir,
- .follow_link = autofs4_follow_link,
-};
-
const struct inode_operations autofs4_dir_inode_operations = {
.lookup = autofs4_lookup,
.unlink = autofs4_dir_unlink,
@@ -84,6 +66,12 @@
.rmdir = autofs4_dir_rmdir,
};
+const struct dentry_operations autofs4_dentry_operations = {
+ .d_automount = autofs4_d_automount,
+ .d_manage = autofs4_d_manage,
+ .d_release = autofs4_dentry_release,
+};
+
static void autofs4_add_active(struct dentry *dentry)
{
struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
@@ -116,14 +104,6 @@
return;
}
-static unsigned int autofs4_need_mount(unsigned int flags)
-{
- unsigned int res = 0;
- if (flags & (TRIGGER_FLAGS | TRIGGER_INTENTS))
- res = 1;
- return res;
-}
-
static int autofs4_dir_open(struct inode *inode, struct file *file)
{
struct dentry *dentry = file->f_path.dentry;
@@ -158,239 +138,6 @@
return dcache_dir_open(inode, file);
}
-static int try_to_fill_dentry(struct dentry *dentry, int flags)
-{
- struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
- struct autofs_info *ino = autofs4_dentry_ino(dentry);
- int status;
-
- DPRINTK("dentry=%p %.*s ino=%p",
- dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode);
-
- /*
- * Wait for a pending mount, triggering one if there
- * isn't one already
- */
- if (dentry->d_inode == NULL) {
- DPRINTK("waiting for mount name=%.*s",
- dentry->d_name.len, dentry->d_name.name);
-
- status = autofs4_wait(sbi, dentry, NFY_MOUNT);
-
- DPRINTK("mount done status=%d", status);
-
- /* Turn this into a real negative dentry? */
- if (status == -ENOENT) {
- spin_lock(&sbi->fs_lock);
- ino->flags &= ~AUTOFS_INF_PENDING;
- spin_unlock(&sbi->fs_lock);
- return status;
- } else if (status) {
- /* Return a negative dentry, but leave it "pending" */
- return status;
- }
- /* Trigger mount for path component or follow link */
- } else if (ino->flags & AUTOFS_INF_PENDING ||
- autofs4_need_mount(flags)) {
- DPRINTK("waiting for mount name=%.*s",
- dentry->d_name.len, dentry->d_name.name);
-
- spin_lock(&sbi->fs_lock);
- ino->flags |= AUTOFS_INF_PENDING;
- spin_unlock(&sbi->fs_lock);
- status = autofs4_wait(sbi, dentry, NFY_MOUNT);
-
- DPRINTK("mount done status=%d", status);
-
- if (status) {
- spin_lock(&sbi->fs_lock);
- ino->flags &= ~AUTOFS_INF_PENDING;
- spin_unlock(&sbi->fs_lock);
- return status;
- }
- }
-
- /* Initialize expiry counter after successful mount */
- ino->last_used = jiffies;
-
- spin_lock(&sbi->fs_lock);
- ino->flags &= ~AUTOFS_INF_PENDING;
- spin_unlock(&sbi->fs_lock);
-
- return 0;
-}
-
-/* For autofs direct mounts the follow link triggers the mount */
-static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
-{
- struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
- struct autofs_info *ino = autofs4_dentry_ino(dentry);
- int oz_mode = autofs4_oz_mode(sbi);
- unsigned int lookup_type;
- int status;
-
- DPRINTK("dentry=%p %.*s oz_mode=%d nd->flags=%d",
- dentry, dentry->d_name.len, dentry->d_name.name, oz_mode,
- nd->flags);
- /*
- * For an expire of a covered direct or offset mount we need
- * to break out of follow_down() at the autofs mount trigger
- * (d_mounted--), so we can see the expiring flag, and manage
- * the blocking and following here until the expire is completed.
- */
- if (oz_mode) {
- spin_lock(&sbi->fs_lock);
- if (ino->flags & AUTOFS_INF_EXPIRING) {
- spin_unlock(&sbi->fs_lock);
- /* Follow down to our covering mount. */
- if (!follow_down(&nd->path))
- goto done;
- goto follow;
- }
- spin_unlock(&sbi->fs_lock);
- goto done;
- }
-
- /* If an expire request is pending everyone must wait. */
- autofs4_expire_wait(dentry);
-
- /* We trigger a mount for almost all flags */
- lookup_type = autofs4_need_mount(nd->flags);
- spin_lock(&sbi->fs_lock);
- spin_lock(&autofs4_lock);
- spin_lock(&dentry->d_lock);
- if (!(lookup_type || ino->flags & AUTOFS_INF_PENDING)) {
- spin_unlock(&dentry->d_lock);
- spin_unlock(&autofs4_lock);
- spin_unlock(&sbi->fs_lock);
- goto follow;
- }
-
- /*
- * If the dentry contains directories then it is an autofs
- * multi-mount with no root mount offset. So don't try to
- * mount it again.
- */
- if (ino->flags & AUTOFS_INF_PENDING ||
- (!d_mountpoint(dentry) && list_empty(&dentry->d_subdirs))) {
- spin_unlock(&dentry->d_lock);
- spin_unlock(&autofs4_lock);
- spin_unlock(&sbi->fs_lock);
-
- status = try_to_fill_dentry(dentry, nd->flags);
- if (status)
- goto out_error;
-
- goto follow;
- }
- spin_unlock(&dentry->d_lock);
- spin_unlock(&autofs4_lock);
- spin_unlock(&sbi->fs_lock);
-follow:
- /*
- * If there is no root mount it must be an autofs
- * multi-mount with no root offset so we don't need
- * to follow it.
- */
- if (d_mountpoint(dentry)) {
- if (!autofs4_follow_mount(&nd->path)) {
- status = -ENOENT;
- goto out_error;
- }
- }
-
-done:
- return NULL;
-
-out_error:
- path_put(&nd->path);
- return ERR_PTR(status);
-}
-
-/*
- * Revalidate is called on every cache lookup. Some of those
- * cache lookups may actually happen while the dentry is not
- * yet completely filled in, and revalidate has to delay such
- * lookups..
- */
-static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
-{
- struct inode *dir;
- struct autofs_sb_info *sbi;
- int oz_mode;
- int flags = nd ? nd->flags : 0;
- int status = 1;
-
- if (flags & LOOKUP_RCU)
- return -ECHILD;
-
- dir = dentry->d_parent->d_inode;
- sbi = autofs4_sbi(dir->i_sb);
- oz_mode = autofs4_oz_mode(sbi);
-
- /* Pending dentry */
- spin_lock(&sbi->fs_lock);
- if (autofs4_ispending(dentry)) {
- /* The daemon never causes a mount to trigger */
- spin_unlock(&sbi->fs_lock);
-
- if (oz_mode)
- return 1;
-
- /*
- * If the directory has gone away due to an expire
- * we have been called as ->d_revalidate() and so
- * we need to return false and proceed to ->lookup().
- */
- if (autofs4_expire_wait(dentry) == -EAGAIN)
- return 0;
-
- /*
- * A zero status is success otherwise we have a
- * negative error code.
- */
- status = try_to_fill_dentry(dentry, flags);
- if (status == 0)
- return 1;
-
- return status;
- }
- spin_unlock(&sbi->fs_lock);
-
- /* Negative dentry.. invalidate if "old" */
- if (dentry->d_inode == NULL)
- return 0;
-
- /* Check for a non-mountpoint directory with no contents */
- spin_lock(&autofs4_lock);
- spin_lock(&dentry->d_lock);
- if (S_ISDIR(dentry->d_inode->i_mode) &&
- !d_mountpoint(dentry) && list_empty(&dentry->d_subdirs)) {
- DPRINTK("dentry=%p %.*s, emptydir",
- dentry, dentry->d_name.len, dentry->d_name.name);
- spin_unlock(&dentry->d_lock);
- spin_unlock(&autofs4_lock);
-
- /* The daemon never causes a mount to trigger */
- if (oz_mode)
- return 1;
-
- /*
- * A zero status is success otherwise we have a
- * negative error code.
- */
- status = try_to_fill_dentry(dentry, flags);
- if (status == 0)
- return 1;
-
- return status;
- }
- spin_unlock(&dentry->d_lock);
- spin_unlock(&autofs4_lock);
-
- return 1;
-}
-
void autofs4_dentry_release(struct dentry *de)
{
struct autofs_info *inf;
@@ -398,11 +145,8 @@
DPRINTK("releasing %p", de);
inf = autofs4_dentry_ino(de);
- de->d_fsdata = NULL;
-
if (inf) {
struct autofs_sb_info *sbi = autofs4_sbi(de->d_sb);
-
if (sbi) {
spin_lock(&sbi->lookup_lock);
if (!list_empty(&inf->active))
@@ -411,26 +155,10 @@
list_del(&inf->expiring);
spin_unlock(&sbi->lookup_lock);
}
-
- inf->dentry = NULL;
- inf->inode = NULL;
-
autofs4_free_ino(inf);
}
}
-/* For dentries of directories in the root dir */
-static const struct dentry_operations autofs4_root_dentry_operations = {
- .d_revalidate = autofs4_revalidate,
- .d_release = autofs4_dentry_release,
-};
-
-/* For other dentries */
-static const struct dentry_operations autofs4_dentry_operations = {
- .d_revalidate = autofs4_revalidate,
- .d_release = autofs4_dentry_release,
-};
-
static struct dentry *autofs4_lookup_active(struct dentry *dentry)
{
struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
@@ -541,50 +269,244 @@
return NULL;
}
+static int autofs4_mount_wait(struct dentry *dentry)
+{
+ struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+ struct autofs_info *ino = autofs4_dentry_ino(dentry);
+ int status;
+
+ if (ino->flags & AUTOFS_INF_PENDING) {
+ DPRINTK("waiting for mount name=%.*s",
+ dentry->d_name.len, dentry->d_name.name);
+ status = autofs4_wait(sbi, dentry, NFY_MOUNT);
+ DPRINTK("mount wait done status=%d", status);
+ ino->last_used = jiffies;
+ return status;
+ }
+ return 0;
+}
+
+static int do_expire_wait(struct dentry *dentry)
+{
+ struct dentry *expiring;
+
+ expiring = autofs4_lookup_expiring(dentry);
+ if (!expiring)
+ return autofs4_expire_wait(dentry);
+ else {
+ /*
+ * If we are racing with expire the request might not
+ * be quite complete, but the directory has been removed
+ * so it must have been successful, just wait for it.
+ */
+ autofs4_expire_wait(expiring);
+ autofs4_del_expiring(expiring);
+ dput(expiring);
+ }
+ return 0;
+}
+
+static struct dentry *autofs4_mountpoint_changed(struct path *path)
+{
+ struct dentry *dentry = path->dentry;
+ struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+
+ /*
+ * If this is an indirect mount the dentry could have gone away
+ * as a result of an expire and a new one created.
+ */
+ if (autofs_type_indirect(sbi->type) && d_unhashed(dentry)) {
+ struct dentry *parent = dentry->d_parent;
+ struct dentry *new = d_lookup(parent, &dentry->d_name);
+ if (!new)
+ return NULL;
+ dput(path->dentry);
+ path->dentry = new;
+ }
+ return path->dentry;
+}
+
+static struct vfsmount *autofs4_d_automount(struct path *path)
+{
+ struct dentry *dentry = path->dentry;
+ struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+ struct autofs_info *ino = autofs4_dentry_ino(dentry);
+ int status;
+
+ DPRINTK("dentry=%p %.*s",
+ dentry, dentry->d_name.len, dentry->d_name.name);
+
+ /*
+ * Someone may have manually umounted this or it was a submount
+ * that has gone away.
+ */
+ spin_lock(&dentry->d_lock);
+ if (!d_mountpoint(dentry) && list_empty(&dentry->d_subdirs)) {
+ if (!(dentry->d_flags & DCACHE_MANAGE_TRANSIT) &&
+ (dentry->d_flags & DCACHE_NEED_AUTOMOUNT))
+ __managed_dentry_set_transit(path->dentry);
+ }
+ spin_unlock(&dentry->d_lock);
+
+ /* The daemon never triggers a mount. */
+ if (autofs4_oz_mode(sbi))
+ return NULL;
+
+ /*
+ * If an expire request is pending everyone must wait.
+ * If the expire fails we're still mounted so continue
+ * the follow and return. A return of -EAGAIN (which only
+ * happens with indirect mounts) means the expire completed
+ * and the directory was removed, so just go ahead and try
+ * the mount.
+ */
+ status = do_expire_wait(dentry);
+ if (status && status != -EAGAIN)
+ return NULL;
+
+ /* Callback to the daemon to perform the mount or wait */
+ spin_lock(&sbi->fs_lock);
+ if (ino->flags & AUTOFS_INF_PENDING) {
+ spin_unlock(&sbi->fs_lock);
+ status = autofs4_mount_wait(dentry);
+ if (status)
+ return ERR_PTR(status);
+ spin_lock(&sbi->fs_lock);
+ goto done;
+ }
+
+ /*
+ * If the dentry is a symlink it's equivalent to a directory
+ * having d_mountpoint() true, so there's no need to call back
+ * to the daemon.
+ */
+ if (dentry->d_inode && S_ISLNK(dentry->d_inode->i_mode))
+ goto done;
+ if (!d_mountpoint(dentry)) {
+ /*
+ * It's possible that user space hasn't removed directories
+ * after umounting a rootless multi-mount, although it
+ * should. For v5 have_submounts() is sufficient to handle
+ * this because the leaves of the directory tree under the
+ * mount never trigger mounts themselves (they have an autofs
+ * trigger mount mounted on them). But v4 pseudo direct mounts
+ * do need the leaves to to trigger mounts. In this case we
+ * have no choice but to use the list_empty() check and
+ * require user space behave.
+ */
+ if (sbi->version > 4) {
+ if (have_submounts(dentry))
+ goto done;
+ } else {
+ spin_lock(&dentry->d_lock);
+ if (!list_empty(&dentry->d_subdirs)) {
+ spin_unlock(&dentry->d_lock);
+ goto done;
+ }
+ spin_unlock(&dentry->d_lock);
+ }
+ ino->flags |= AUTOFS_INF_PENDING;
+ spin_unlock(&sbi->fs_lock);
+ status = autofs4_mount_wait(dentry);
+ if (status)
+ return ERR_PTR(status);
+ spin_lock(&sbi->fs_lock);
+ ino->flags &= ~AUTOFS_INF_PENDING;
+ }
+done:
+ if (!(ino->flags & AUTOFS_INF_EXPIRING)) {
+ /*
+ * Any needed mounting has been completed and the path updated
+ * so turn this into a normal dentry so we don't continually
+ * call ->d_automount() and ->d_manage().
+ */
+ spin_lock(&dentry->d_lock);
+ __managed_dentry_clear_transit(dentry);
+ /*
+ * Only clear DMANAGED_AUTOMOUNT for rootless multi-mounts and
+ * symlinks as in all other cases the dentry will be covered by
+ * an actual mount so ->d_automount() won't be called during
+ * the follow.
+ */
+ if ((!d_mountpoint(dentry) &&
+ !list_empty(&dentry->d_subdirs)) ||
+ (dentry->d_inode && S_ISLNK(dentry->d_inode->i_mode)))
+ __managed_dentry_clear_automount(dentry);
+ spin_unlock(&dentry->d_lock);
+ }
+ spin_unlock(&sbi->fs_lock);
+
+ /* Mount succeeded, check if we ended up with a new dentry */
+ dentry = autofs4_mountpoint_changed(path);
+ if (!dentry)
+ return ERR_PTR(-ENOENT);
+
+ return NULL;
+}
+
+int autofs4_d_manage(struct dentry *dentry, bool mounting_here, bool rcu_walk)
+{
+ struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+
+ DPRINTK("dentry=%p %.*s",
+ dentry, dentry->d_name.len, dentry->d_name.name);
+
+ /* The daemon never waits. */
+ if (autofs4_oz_mode(sbi) || mounting_here) {
+ if (!d_mountpoint(dentry))
+ return -EISDIR;
+ return 0;
+ }
+
+ /* We need to sleep, so we need pathwalk to be in ref-mode */
+ if (rcu_walk)
+ return -ECHILD;
+
+ /* Wait for pending expires */
+ do_expire_wait(dentry);
+
+ /*
+ * This dentry may be under construction so wait on mount
+ * completion.
+ */
+ return autofs4_mount_wait(dentry);
+}
+
/* Lookups in the root directory */
static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
{
struct autofs_sb_info *sbi;
struct autofs_info *ino;
- struct dentry *expiring, *active;
- int oz_mode;
+ struct dentry *active;
- DPRINTK("name = %.*s",
- dentry->d_name.len, dentry->d_name.name);
+ DPRINTK("name = %.*s", dentry->d_name.len, dentry->d_name.name);
/* File name too long to exist */
if (dentry->d_name.len > NAME_MAX)
return ERR_PTR(-ENAMETOOLONG);
sbi = autofs4_sbi(dir->i_sb);
- oz_mode = autofs4_oz_mode(sbi);
DPRINTK("pid = %u, pgrp = %u, catatonic = %d, oz_mode = %d",
- current->pid, task_pgrp_nr(current), sbi->catatonic, oz_mode);
+ current->pid, task_pgrp_nr(current), sbi->catatonic, oz_mode);
active = autofs4_lookup_active(dentry);
if (active) {
- dentry = active;
- ino = autofs4_dentry_ino(dentry);
+ return active;
} else {
/*
- * Mark the dentry incomplete but don't hash it. We do this
- * to serialize our inode creation operations (symlink and
- * mkdir) which prevents deadlock during the callback to
- * the daemon. Subsequent user space lookups for the same
- * dentry are placed on the wait queue while the daemon
- * itself is allowed passage unresticted so the create
- * operation itself can then hash the dentry. Finally,
- * we check for the hashed dentry and return the newly
- * hashed dentry.
+ * A dentry that is not within the root can never trigger a
+ * mount operation, unless the directory already exists, so we
+ * can return fail immediately. The daemon however does need
+ * to create directories within the file system.
*/
- d_set_d_op(dentry, &autofs4_root_dentry_operations);
+ if (!autofs4_oz_mode(sbi) && !IS_ROOT(dentry->d_parent))
+ return ERR_PTR(-ENOENT);
- /*
- * And we need to ensure that the same dentry is used for
- * all following lookup calls until it is hashed so that
- * the dentry flags are persistent throughout the request.
- */
+ /* Mark entries in the root as mount triggers */
+ if (autofs_type_indirect(sbi->type) && IS_ROOT(dentry->d_parent))
+ __managed_dentry_set_managed(dentry);
+
ino = autofs4_init_ino(NULL, sbi, 0555);
if (!ino)
return ERR_PTR(-ENOMEM);
@@ -596,82 +518,6 @@
d_instantiate(dentry, NULL);
}
-
- if (!oz_mode) {
- mutex_unlock(&dir->i_mutex);
- expiring = autofs4_lookup_expiring(dentry);
- if (expiring) {
- /*
- * If we are racing with expire the request might not
- * be quite complete but the directory has been removed
- * so it must have been successful, so just wait for it.
- */
- autofs4_expire_wait(expiring);
- autofs4_del_expiring(expiring);
- dput(expiring);
- }
-
- spin_lock(&sbi->fs_lock);
- ino->flags |= AUTOFS_INF_PENDING;
- spin_unlock(&sbi->fs_lock);
- if (dentry->d_op && dentry->d_op->d_revalidate)
- (dentry->d_op->d_revalidate)(dentry, nd);
- mutex_lock(&dir->i_mutex);
- }
-
- /*
- * If we are still pending, check if we had to handle
- * a signal. If so we can force a restart..
- */
- if (ino->flags & AUTOFS_INF_PENDING) {
- /* See if we were interrupted */
- if (signal_pending(current)) {
- sigset_t *sigset = ¤t->pending.signal;
- if (sigismember (sigset, SIGKILL) ||
- sigismember (sigset, SIGQUIT) ||
- sigismember (sigset, SIGINT)) {
- if (active)
- dput(active);
- return ERR_PTR(-ERESTARTNOINTR);
- }
- }
- if (!oz_mode) {
- spin_lock(&sbi->fs_lock);
- ino->flags &= ~AUTOFS_INF_PENDING;
- spin_unlock(&sbi->fs_lock);
- }
- }
-
- /*
- * If this dentry is unhashed, then we shouldn't honour this
- * lookup. Returning ENOENT here doesn't do the right thing
- * for all system calls, but it should be OK for the operations
- * we permit from an autofs.
- */
- if (!oz_mode && d_unhashed(dentry)) {
- /*
- * A user space application can (and has done in the past)
- * remove and re-create this directory during the callback.
- * This can leave us with an unhashed dentry, but a
- * successful mount! So we need to perform another
- * cached lookup in case the dentry now exists.
- */
- struct dentry *parent = dentry->d_parent;
- struct dentry *new = d_lookup(parent, &dentry->d_name);
- if (new != NULL)
- dentry = new;
- else
- dentry = ERR_PTR(-ENOENT);
-
- if (active)
- dput(active);
-
- return dentry;
- }
-
- if (active)
- return active;
-
return NULL;
}
@@ -716,18 +562,12 @@
}
d_add(dentry, inode);
- if (dir == dir->i_sb->s_root->d_inode)
- d_set_d_op(dentry, &autofs4_root_dentry_operations);
- else
- d_set_d_op(dentry, &autofs4_dentry_operations);
-
dentry->d_fsdata = ino;
ino->dentry = dget(dentry);
atomic_inc(&ino->count);
p_ino = autofs4_dentry_ino(dentry->d_parent);
if (p_ino && dentry->d_parent != dentry)
atomic_inc(&p_ino->count);
- ino->inode = inode;
ino->u.symlink = cp;
dir->i_mtime = CURRENT_TIME;
@@ -782,6 +622,58 @@
return 0;
}
+/*
+ * Version 4 of autofs provides a pseudo direct mount implementation
+ * that relies on directories at the leaves of a directory tree under
+ * an indirect mount to trigger mounts. To allow for this we need to
+ * set the DMANAGED_AUTOMOUNT and DMANAGED_TRANSIT flags on the leaves
+ * of the directory tree. There is no need to clear the automount flag
+ * following a mount or restore it after an expire because these mounts
+ * are always covered. However, it is neccessary to ensure that these
+ * flags are clear on non-empty directories to avoid unnecessary calls
+ * during path walks.
+ */
+static void autofs_set_leaf_automount_flags(struct dentry *dentry)
+{
+ struct dentry *parent;
+
+ /* root and dentrys in the root are already handled */
+ if (IS_ROOT(dentry->d_parent))
+ return;
+
+ managed_dentry_set_managed(dentry);
+
+ parent = dentry->d_parent;
+ /* only consider parents below dentrys in the root */
+ if (IS_ROOT(parent->d_parent))
+ return;
+ managed_dentry_clear_managed(parent);
+ return;
+}
+
+static void autofs_clear_leaf_automount_flags(struct dentry *dentry)
+{
+ struct list_head *d_child;
+ struct dentry *parent;
+
+ /* flags for dentrys in the root are handled elsewhere */
+ if (IS_ROOT(dentry->d_parent))
+ return;
+
+ managed_dentry_clear_managed(dentry);
+
+ parent = dentry->d_parent;
+ /* only consider parents below dentrys in the root */
+ if (IS_ROOT(parent->d_parent))
+ return;
+ d_child = &dentry->d_u.d_child;
+ /* Set parent managed if it's becoming empty */
+ if (d_child->next == &parent->d_subdirs &&
+ d_child->prev == &parent->d_subdirs)
+ managed_dentry_set_managed(parent);
+ return;
+}
+
static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry)
{
struct autofs_sb_info *sbi = autofs4_sbi(dir->i_sb);
@@ -809,6 +701,9 @@
spin_unlock(&dentry->d_lock);
spin_unlock(&autofs4_lock);
+ if (sbi->version < 5)
+ autofs_clear_leaf_automount_flags(dentry);
+
if (atomic_dec_and_test(&ino->count)) {
p_ino = autofs4_dentry_ino(dentry->d_parent);
if (p_ino && dentry->d_parent != dentry)
@@ -851,10 +746,8 @@
}
d_add(dentry, inode);
- if (dir == dir->i_sb->s_root->d_inode)
- d_set_d_op(dentry, &autofs4_root_dentry_operations);
- else
- d_set_d_op(dentry, &autofs4_dentry_operations);
+ if (sbi->version < 5)
+ autofs_set_leaf_automount_flags(dentry);
dentry->d_fsdata = ino;
ino->dentry = dget(dentry);
@@ -862,7 +755,6 @@
p_ino = autofs4_dentry_ino(dentry->d_parent);
if (p_ino && dentry->d_parent != dentry)
atomic_inc(&p_ino->count);
- ino->inode = inode;
inc_nlink(dir);
dir->i_mtime = CURRENT_TIME;
@@ -944,8 +836,7 @@
int is_autofs4_dentry(struct dentry *dentry)
{
return dentry && dentry->d_inode &&
- (dentry->d_op == &autofs4_root_dentry_operations ||
- dentry->d_op == &autofs4_dentry_operations) &&
+ dentry->d_op == &autofs4_dentry_operations &&
dentry->d_fsdata != NULL;
}
diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index c5f8459..5601005 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -309,6 +309,9 @@
* completed while we waited on the mutex ...
*/
if (notify == NFY_MOUNT) {
+ struct dentry *new = NULL;
+ int valid = 1;
+
/*
* If the dentry was successfully mounted while we slept
* on the wait queue mutex we can return success. If it
@@ -316,8 +319,20 @@
* a multi-mount with no mount at it's base) we can
* continue on and create a new request.
*/
+ if (!IS_ROOT(dentry)) {
+ if (dentry->d_inode && d_unhashed(dentry)) {
+ struct dentry *parent = dentry->d_parent;
+ new = d_lookup(parent, &dentry->d_name);
+ if (new)
+ dentry = new;
+ }
+ }
if (have_submounts(dentry))
- return 0;
+ valid = 0;
+
+ if (new)
+ dput(new);
+ return valid;
}
return 1;
diff --git a/fs/block_dev.c b/fs/block_dev.c
index fe3f59c1..333a7bb 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -432,6 +432,9 @@
mutex_init(&bdev->bd_mutex);
INIT_LIST_HEAD(&bdev->bd_inodes);
INIT_LIST_HEAD(&bdev->bd_list);
+#ifdef CONFIG_SYSFS
+ INIT_LIST_HEAD(&bdev->bd_holder_disks);
+#endif
inode_init_once(&ei->vfs_inode);
/* Initialize mutex for freeze. */
mutex_init(&bdev->bd_fsfreeze_mutex);
@@ -779,6 +782,23 @@
}
#ifdef CONFIG_SYSFS
+struct bd_holder_disk {
+ struct list_head list;
+ struct gendisk *disk;
+ int refcnt;
+};
+
+static struct bd_holder_disk *bd_find_holder_disk(struct block_device *bdev,
+ struct gendisk *disk)
+{
+ struct bd_holder_disk *holder;
+
+ list_for_each_entry(holder, &bdev->bd_holder_disks, list)
+ if (holder->disk == disk)
+ return holder;
+ return NULL;
+}
+
static int add_symlink(struct kobject *from, struct kobject *to)
{
return sysfs_create_link(from, to, kobject_name(to));
@@ -794,6 +814,8 @@
* @bdev: the claimed slave bdev
* @disk: the holding disk
*
+ * DON'T USE THIS UNLESS YOU'RE ALREADY USING IT.
+ *
* This functions creates the following sysfs symlinks.
*
* - from "slaves" directory of the holder @disk to the claimed @bdev
@@ -817,47 +839,83 @@
*/
int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk)
{
+ struct bd_holder_disk *holder;
int ret = 0;
mutex_lock(&bdev->bd_mutex);
- WARN_ON_ONCE(!bdev->bd_holder || bdev->bd_holder_disk);
+ WARN_ON_ONCE(!bdev->bd_holder);
/* FIXME: remove the following once add_disk() handles errors */
if (WARN_ON(!disk->slave_dir || !bdev->bd_part->holder_dir))
goto out_unlock;
- ret = add_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj);
- if (ret)
- goto out_unlock;
-
- ret = add_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj);
- if (ret) {
- del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj);
+ holder = bd_find_holder_disk(bdev, disk);
+ if (holder) {
+ holder->refcnt++;
goto out_unlock;
}
- bdev->bd_holder_disk = disk;
+ holder = kzalloc(sizeof(*holder), GFP_KERNEL);
+ if (!holder) {
+ ret = -ENOMEM;
+ goto out_unlock;
+ }
+
+ INIT_LIST_HEAD(&holder->list);
+ holder->disk = disk;
+ holder->refcnt = 1;
+
+ ret = add_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj);
+ if (ret)
+ goto out_free;
+
+ ret = add_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj);
+ if (ret)
+ goto out_del;
+
+ list_add(&holder->list, &bdev->bd_holder_disks);
+ goto out_unlock;
+
+out_del:
+ del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj);
+out_free:
+ kfree(holder);
out_unlock:
mutex_unlock(&bdev->bd_mutex);
return ret;
}
EXPORT_SYMBOL_GPL(bd_link_disk_holder);
-static void bd_unlink_disk_holder(struct block_device *bdev)
+/**
+ * bd_unlink_disk_holder - destroy symlinks created by bd_link_disk_holder()
+ * @bdev: the calimed slave bdev
+ * @disk: the holding disk
+ *
+ * DON'T USE THIS UNLESS YOU'RE ALREADY USING IT.
+ *
+ * CONTEXT:
+ * Might sleep.
+ */
+void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk)
{
- struct gendisk *disk = bdev->bd_holder_disk;
+ struct bd_holder_disk *holder;
- bdev->bd_holder_disk = NULL;
- if (!disk)
- return;
+ mutex_lock(&bdev->bd_mutex);
- del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj);
- del_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj);
+ holder = bd_find_holder_disk(bdev, disk);
+
+ if (!WARN_ON_ONCE(holder == NULL) && !--holder->refcnt) {
+ del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj);
+ del_symlink(bdev->bd_part->holder_dir,
+ &disk_to_dev(disk)->kobj);
+ list_del_init(&holder->list);
+ kfree(holder);
+ }
+
+ mutex_unlock(&bdev->bd_mutex);
}
-#else
-static inline void bd_unlink_disk_holder(struct block_device *bdev)
-{ }
+EXPORT_SYMBOL_GPL(bd_unlink_disk_holder);
#endif
/**
@@ -1380,7 +1438,6 @@
* unblock evpoll if it was a write holder.
*/
if (bdev_free) {
- bd_unlink_disk_holder(bdev);
if (bdev->bd_write_holder) {
disk_unblock_events(bdev->bd_disk);
bdev->bd_write_holder = false;
diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c
index c68a056..7ed3653 100644
--- a/fs/cifs/cifs_dfs_ref.c
+++ b/fs/cifs/cifs_dfs_ref.c
@@ -255,35 +255,6 @@
}
-static int add_mount_helper(struct vfsmount *newmnt, struct nameidata *nd,
- struct list_head *mntlist)
-{
- /* stolen from afs code */
- int err;
-
- mntget(newmnt);
- err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags | MNT_SHRINKABLE, mntlist);
- switch (err) {
- case 0:
- path_put(&nd->path);
- nd->path.mnt = newmnt;
- nd->path.dentry = dget(newmnt->mnt_root);
- schedule_delayed_work(&cifs_dfs_automount_task,
- cifs_dfs_mountpoint_expiry_timeout);
- break;
- case -EBUSY:
- /* someone else made a mount here whilst we were busy */
- while (d_mountpoint(nd->path.dentry) &&
- follow_down(&nd->path))
- ;
- err = 0;
- default:
- mntput(newmnt);
- break;
- }
- return err;
-}
-
static void dump_referral(const struct dfs_info3_param *ref)
{
cFYI(1, "DFS: ref path: %s", ref->path_name);
@@ -293,45 +264,43 @@
ref->path_consumed);
}
-
-static void*
-cifs_dfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
+/*
+ * Create a vfsmount that we can automount
+ */
+static struct vfsmount *cifs_dfs_do_automount(struct dentry *mntpt)
{
struct dfs_info3_param *referrals = NULL;
unsigned int num_referrals = 0;
struct cifs_sb_info *cifs_sb;
struct cifsSesInfo *ses;
- char *full_path = NULL;
+ char *full_path;
int xid, i;
- int rc = 0;
- struct vfsmount *mnt = ERR_PTR(-ENOENT);
+ int rc;
+ struct vfsmount *mnt;
struct tcon_link *tlink;
cFYI(1, "in %s", __func__);
- BUG_ON(IS_ROOT(dentry));
+ BUG_ON(IS_ROOT(mntpt));
xid = GetXid();
- dput(nd->path.dentry);
- nd->path.dentry = dget(dentry);
-
/*
* The MSDFS spec states that paths in DFS referral requests and
* responses must be prefixed by a single '\' character instead of
* the double backslashes usually used in the UNC. This function
* gives us the latter, so we must adjust the result.
*/
- full_path = build_path_from_dentry(dentry);
- if (full_path == NULL) {
- rc = -ENOMEM;
- goto out_err;
- }
+ mnt = ERR_PTR(-ENOMEM);
+ full_path = build_path_from_dentry(mntpt);
+ if (full_path == NULL)
+ goto free_xid;
- cifs_sb = CIFS_SB(dentry->d_inode->i_sb);
+ cifs_sb = CIFS_SB(mntpt->d_inode->i_sb);
tlink = cifs_sb_tlink(cifs_sb);
+ mnt = ERR_PTR(-EINVAL);
if (IS_ERR(tlink)) {
- rc = PTR_ERR(tlink);
- goto out_err;
+ mnt = ERR_CAST(tlink);
+ goto free_full_path;
}
ses = tlink_tcon(tlink)->ses;
@@ -341,46 +310,63 @@
cifs_put_tlink(tlink);
+ mnt = ERR_PTR(-ENOENT);
for (i = 0; i < num_referrals; i++) {
int len;
- dump_referral(referrals+i);
+ dump_referral(referrals + i);
/* connect to a node */
len = strlen(referrals[i].node_name);
if (len < 2) {
cERROR(1, "%s: Net Address path too short: %s",
__func__, referrals[i].node_name);
- rc = -EINVAL;
- goto out_err;
+ mnt = ERR_PTR(-EINVAL);
+ break;
}
mnt = cifs_dfs_do_refmount(cifs_sb,
full_path, referrals + i);
cFYI(1, "%s: cifs_dfs_do_refmount:%s , mnt:%p", __func__,
referrals[i].node_name, mnt);
-
- /* complete mount procedure if we accured submount */
if (!IS_ERR(mnt))
- break;
+ goto success;
}
- /* we need it cause for() above could exit without valid submount */
- rc = PTR_ERR(mnt);
- if (IS_ERR(mnt))
- goto out_err;
+ /* no valid submounts were found; return error from get_dfs_path() by
+ * preference */
+ if (rc != 0)
+ mnt = ERR_PTR(rc);
- rc = add_mount_helper(mnt, nd, &cifs_dfs_automount_list);
-
-out:
- FreeXid(xid);
+success:
free_dfs_info_array(referrals, num_referrals);
+free_full_path:
kfree(full_path);
+free_xid:
+ FreeXid(xid);
cFYI(1, "leaving %s" , __func__);
- return ERR_PTR(rc);
-out_err:
- path_put(&nd->path);
- goto out;
+ return mnt;
+}
+
+/*
+ * Attempt to automount the referral
+ */
+struct vfsmount *cifs_dfs_d_automount(struct path *path)
+{
+ struct vfsmount *newmnt;
+
+ cFYI(1, "in %s", __func__);
+
+ newmnt = cifs_dfs_do_automount(path->dentry);
+ if (IS_ERR(newmnt)) {
+ cFYI(1, "leaving %s [automount failed]" , __func__);
+ return newmnt;
+ }
+
+ mntget(newmnt); /* prevent immediate expiration */
+ mnt_set_expiry(newmnt, &cifs_dfs_automount_list);
+ schedule_delayed_work(&cifs_dfs_automount_task,
+ cifs_dfs_mountpoint_expiry_timeout);
+ cFYI(1, "leaving %s [ok]" , __func__);
+ return newmnt;
}
const struct inode_operations cifs_dfs_referral_inode_operations = {
- .follow_link = cifs_dfs_follow_mountpoint,
};
-
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 897b2b2..851030f 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -93,6 +93,12 @@
extern const struct dentry_operations cifs_dentry_ops;
extern const struct dentry_operations cifs_ci_dentry_ops;
+#ifdef CONFIG_CIFS_DFS_UPCALL
+extern struct vfsmount *cifs_dfs_d_automount(struct path *path);
+#else
+#define cifs_dfs_d_automount NULL
+#endif
+
/* Functions related to symlinks */
extern void *cifs_follow_link(struct dentry *direntry, struct nameidata *nd);
extern void cifs_put_link(struct dentry *direntry,
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 1e95dd6..dd5f229 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -675,6 +675,7 @@
const struct dentry_operations cifs_dentry_ops = {
.d_revalidate = cifs_d_revalidate,
+ .d_automount = cifs_dfs_d_automount,
/* d_delete: cifs_d_delete, */ /* not needed except for debugging */
};
@@ -711,4 +712,5 @@
.d_revalidate = cifs_d_revalidate,
.d_hash = cifs_ci_hash,
.d_compare = cifs_ci_compare,
+ .d_automount = cifs_dfs_d_automount,
};
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index b06b606..6c9ee80 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -32,7 +32,7 @@
#include "fscache.h"
-static void cifs_set_ops(struct inode *inode, const bool is_dfs_referral)
+static void cifs_set_ops(struct inode *inode)
{
struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
@@ -60,7 +60,7 @@
break;
case S_IFDIR:
#ifdef CONFIG_CIFS_DFS_UPCALL
- if (is_dfs_referral) {
+ if (IS_AUTOMOUNT(inode)) {
inode->i_op = &cifs_dfs_referral_inode_operations;
} else {
#else /* NO DFS support, treat as a directory */
@@ -167,7 +167,9 @@
}
spin_unlock(&inode->i_lock);
- cifs_set_ops(inode, fattr->cf_flags & CIFS_FATTR_DFS_REFERRAL);
+ if (fattr->cf_flags & CIFS_FATTR_DFS_REFERRAL)
+ inode->i_flags |= S_AUTOMOUNT;
+ cifs_set_ops(inode);
}
void
diff --git a/fs/dcache.c b/fs/dcache.c
index 0c6d5c5..9f493ee 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -1357,8 +1357,8 @@
void d_set_d_op(struct dentry *dentry, const struct dentry_operations *op)
{
- BUG_ON(dentry->d_op);
- BUG_ON(dentry->d_flags & (DCACHE_OP_HASH |
+ WARN_ON_ONCE(dentry->d_op);
+ WARN_ON_ONCE(dentry->d_flags & (DCACHE_OP_HASH |
DCACHE_OP_COMPARE |
DCACHE_OP_REVALIDATE |
DCACHE_OP_DELETE ));
@@ -1380,8 +1380,11 @@
static void __d_instantiate(struct dentry *dentry, struct inode *inode)
{
spin_lock(&dentry->d_lock);
- if (inode)
+ if (inode) {
+ if (unlikely(IS_AUTOMOUNT(inode)))
+ dentry->d_flags |= DCACHE_NEED_AUTOMOUNT;
list_add(&dentry->d_alias, &inode->i_dentry);
+ }
dentry->d_inode = inode;
dentry_rcuwalk_barrier(dentry);
spin_unlock(&dentry->d_lock);
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index 9ed4769..d3b28ab 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -141,13 +141,12 @@
return rc;
}
-static inode *ecryptfs_get_inode(struct inode *lower_inode,
+static struct inode *ecryptfs_get_inode(struct inode *lower_inode,
struct super_block *sb)
{
struct inode *inode;
int rc = 0;
- lower_inode = lower_dentry->d_inode;
if (lower_inode->i_sb != ecryptfs_superblock_to_lower(sb)) {
rc = -EXDEV;
goto out;
@@ -202,7 +201,7 @@
{
struct inode *lower_inode = lower_dentry->d_inode;
struct inode *inode = ecryptfs_get_inode(lower_inode, sb);
- if (IS_ERR(inode)
+ if (IS_ERR(inode))
return PTR_ERR(inode);
if (flags & ECRYPTFS_INTERPOSE_FLAG_D_ADD)
d_add(dentry, inode);
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 3d06ccc..59c6e49 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -84,13 +84,9 @@
return list_entry(head, struct inode, i_wb_list);
}
-static void bdi_queue_work(struct backing_dev_info *bdi,
- struct wb_writeback_work *work)
+/* Wakeup flusher thread or forker thread to fork it. Requires bdi->wb_lock. */
+static void bdi_wakeup_flusher(struct backing_dev_info *bdi)
{
- trace_writeback_queue(bdi, work);
-
- spin_lock_bh(&bdi->wb_lock);
- list_add_tail(&work->list, &bdi->work_list);
if (bdi->wb.task) {
wake_up_process(bdi->wb.task);
} else {
@@ -98,15 +94,26 @@
* The bdi thread isn't there, wake up the forker thread which
* will create and run it.
*/
- trace_writeback_nothread(bdi, work);
wake_up_process(default_backing_dev_info.wb.task);
}
+}
+
+static void bdi_queue_work(struct backing_dev_info *bdi,
+ struct wb_writeback_work *work)
+{
+ trace_writeback_queue(bdi, work);
+
+ spin_lock_bh(&bdi->wb_lock);
+ list_add_tail(&work->list, &bdi->work_list);
+ if (!bdi->wb.task)
+ trace_writeback_nothread(bdi, work);
+ bdi_wakeup_flusher(bdi);
spin_unlock_bh(&bdi->wb_lock);
}
static void
__bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
- bool range_cyclic, bool for_background)
+ bool range_cyclic)
{
struct wb_writeback_work *work;
@@ -126,7 +133,6 @@
work->sync_mode = WB_SYNC_NONE;
work->nr_pages = nr_pages;
work->range_cyclic = range_cyclic;
- work->for_background = for_background;
bdi_queue_work(bdi, work);
}
@@ -144,7 +150,7 @@
*/
void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages)
{
- __bdi_start_writeback(bdi, nr_pages, true, false);
+ __bdi_start_writeback(bdi, nr_pages, true);
}
/**
@@ -152,13 +158,21 @@
* @bdi: the backing device to write from
*
* Description:
- * This does WB_SYNC_NONE background writeback. The IO is only
- * started when this function returns, we make no guarentees on
- * completion. Caller need not hold sb s_umount semaphore.
+ * This makes sure WB_SYNC_NONE background writeback happens. When
+ * this function returns, it is only guaranteed that for given BDI
+ * some IO is happening if we are over background dirty threshold.
+ * Caller need not hold sb s_umount semaphore.
*/
void bdi_start_background_writeback(struct backing_dev_info *bdi)
{
- __bdi_start_writeback(bdi, LONG_MAX, true, true);
+ /*
+ * We just wake up the flusher thread. It will perform background
+ * writeback as soon as there is no other work to do.
+ */
+ trace_writeback_wake_background(bdi);
+ spin_lock_bh(&bdi->wb_lock);
+ bdi_wakeup_flusher(bdi);
+ spin_unlock_bh(&bdi->wb_lock);
}
/*
@@ -616,6 +630,7 @@
};
unsigned long oldest_jif;
long wrote = 0;
+ long write_chunk;
struct inode *inode;
if (wbc.for_kupdate) {
@@ -628,6 +643,24 @@
wbc.range_end = LLONG_MAX;
}
+ /*
+ * WB_SYNC_ALL mode does livelock avoidance by syncing dirty
+ * inodes/pages in one big loop. Setting wbc.nr_to_write=LONG_MAX
+ * here avoids calling into writeback_inodes_wb() more than once.
+ *
+ * The intended call sequence for WB_SYNC_ALL writeback is:
+ *
+ * wb_writeback()
+ * __writeback_inodes_sb() <== called only once
+ * write_cache_pages() <== called once for each inode
+ * (quickly) tag currently dirty pages
+ * (maybe slowly) sync all tagged pages
+ */
+ if (wbc.sync_mode == WB_SYNC_NONE)
+ write_chunk = MAX_WRITEBACK_PAGES;
+ else
+ write_chunk = LONG_MAX;
+
wbc.wb_start = jiffies; /* livelock avoidance */
for (;;) {
/*
@@ -637,6 +670,16 @@
break;
/*
+ * Background writeout and kupdate-style writeback may
+ * run forever. Stop them if there is other work to do
+ * so that e.g. sync can proceed. They'll be restarted
+ * after the other works are all done.
+ */
+ if ((work->for_background || work->for_kupdate) &&
+ !list_empty(&wb->bdi->work_list))
+ break;
+
+ /*
* For background writeout, stop when we are below the
* background dirty threshold
*/
@@ -644,7 +687,7 @@
break;
wbc.more_io = 0;
- wbc.nr_to_write = MAX_WRITEBACK_PAGES;
+ wbc.nr_to_write = write_chunk;
wbc.pages_skipped = 0;
trace_wbc_writeback_start(&wbc, wb->bdi);
@@ -654,8 +697,8 @@
writeback_inodes_wb(wb, &wbc);
trace_wbc_writeback_written(&wbc, wb->bdi);
- work->nr_pages -= MAX_WRITEBACK_PAGES - wbc.nr_to_write;
- wrote += MAX_WRITEBACK_PAGES - wbc.nr_to_write;
+ work->nr_pages -= write_chunk - wbc.nr_to_write;
+ wrote += write_chunk - wbc.nr_to_write;
/*
* If we consumed everything, see if we have more
@@ -670,7 +713,7 @@
/*
* Did we write something? Try for more
*/
- if (wbc.nr_to_write < MAX_WRITEBACK_PAGES)
+ if (wbc.nr_to_write < write_chunk)
continue;
/*
* Nothing written. Wait for some inode to
@@ -718,6 +761,23 @@
get_nr_dirty_inodes();
}
+static long wb_check_background_flush(struct bdi_writeback *wb)
+{
+ if (over_bground_thresh()) {
+
+ struct wb_writeback_work work = {
+ .nr_pages = LONG_MAX,
+ .sync_mode = WB_SYNC_NONE,
+ .for_background = 1,
+ .range_cyclic = 1,
+ };
+
+ return wb_writeback(wb, &work);
+ }
+
+ return 0;
+}
+
static long wb_check_old_data_flush(struct bdi_writeback *wb)
{
unsigned long expired;
@@ -787,6 +847,7 @@
* Check for periodic writeback, kupdated() style
*/
wrote += wb_check_old_data_flush(wb);
+ wrote += wb_check_background_flush(wb);
clear_bit(BDI_writeback_running, &wb->bdi->state);
return wrote;
@@ -873,7 +934,7 @@
list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
if (!bdi_has_dirty_io(bdi))
continue;
- __bdi_start_writeback(bdi, nr_pages, false, false);
+ __bdi_start_writeback(bdi, nr_pages, false);
}
rcu_read_unlock();
}
@@ -1164,7 +1225,7 @@
* @sb: the superblock
*
* This function writes and waits on any dirty inode belonging to this
- * super_block. The number of pages synced is returned.
+ * super_block.
*/
void sync_inodes_sb(struct super_block *sb)
{
@@ -1242,11 +1303,11 @@
EXPORT_SYMBOL(sync_inode);
/**
- * sync_inode - write an inode to disk
+ * sync_inode_metadata - write an inode to disk
* @inode: the inode to sync
* @wait: wait for I/O to complete.
*
- * Write an inode to disk and adjust it's dirty state after completion.
+ * Write an inode to disk and adjust its dirty state after completion.
*
* Note: only writes the actual inode, no associated data or other metadata.
*/
diff --git a/fs/fs_struct.c b/fs/fs_struct.c
index 68ca487..78b519c 100644
--- a/fs/fs_struct.c
+++ b/fs/fs_struct.c
@@ -4,6 +4,19 @@
#include <linux/path.h>
#include <linux/slab.h>
#include <linux/fs_struct.h>
+#include "internal.h"
+
+static inline void path_get_longterm(struct path *path)
+{
+ path_get(path);
+ mnt_make_longterm(path->mnt);
+}
+
+static inline void path_put_longterm(struct path *path)
+{
+ mnt_make_shortterm(path->mnt);
+ path_put(path);
+}
/*
* Replace the fs->{rootmnt,root} with {mnt,dentry}. Put the old values.
@@ -17,11 +30,11 @@
write_seqcount_begin(&fs->seq);
old_root = fs->root;
fs->root = *path;
- path_get_long(path);
+ path_get_longterm(path);
write_seqcount_end(&fs->seq);
spin_unlock(&fs->lock);
if (old_root.dentry)
- path_put_long(&old_root);
+ path_put_longterm(&old_root);
}
/*
@@ -36,12 +49,12 @@
write_seqcount_begin(&fs->seq);
old_pwd = fs->pwd;
fs->pwd = *path;
- path_get_long(path);
+ path_get_longterm(path);
write_seqcount_end(&fs->seq);
spin_unlock(&fs->lock);
if (old_pwd.dentry)
- path_put_long(&old_pwd);
+ path_put_longterm(&old_pwd);
}
void chroot_fs_refs(struct path *old_root, struct path *new_root)
@@ -59,13 +72,13 @@
write_seqcount_begin(&fs->seq);
if (fs->root.dentry == old_root->dentry
&& fs->root.mnt == old_root->mnt) {
- path_get_long(new_root);
+ path_get_longterm(new_root);
fs->root = *new_root;
count++;
}
if (fs->pwd.dentry == old_root->dentry
&& fs->pwd.mnt == old_root->mnt) {
- path_get_long(new_root);
+ path_get_longterm(new_root);
fs->pwd = *new_root;
count++;
}
@@ -76,13 +89,13 @@
} while_each_thread(g, p);
read_unlock(&tasklist_lock);
while (count--)
- path_put_long(old_root);
+ path_put_longterm(old_root);
}
void free_fs_struct(struct fs_struct *fs)
{
- path_put_long(&fs->root);
- path_put_long(&fs->pwd);
+ path_put_longterm(&fs->root);
+ path_put_longterm(&fs->pwd);
kmem_cache_free(fs_cachep, fs);
}
@@ -118,9 +131,9 @@
spin_lock(&old->lock);
fs->root = old->root;
- path_get_long(&fs->root);
+ path_get_longterm(&fs->root);
fs->pwd = old->pwd;
- path_get_long(&fs->pwd);
+ path_get_longterm(&fs->pwd);
spin_unlock(&old->lock);
}
return fs;
diff --git a/fs/fscache/operation.c b/fs/fscache/operation.c
index b9f34ea..48a18f1 100644
--- a/fs/fscache/operation.c
+++ b/fs/fscache/operation.c
@@ -101,7 +101,7 @@
object->n_ops++;
object->n_exclusive++; /* reads and writes must wait */
- if (object->n_ops > 0) {
+ if (object->n_ops > 1) {
atomic_inc(&op->usage);
list_add_tail(&op->pend_link, &object->pending_ops);
fscache_stat(&fscache_n_op_pend);
diff --git a/fs/internal.h b/fs/internal.h
index 9687c2e..12ccb86 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -70,6 +70,11 @@
extern void release_mounts(struct list_head *);
extern void umount_tree(struct vfsmount *, int, struct list_head *);
extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int);
+extern int do_add_mount(struct vfsmount *, struct path *, int);
+extern void mnt_clear_expiry(struct vfsmount *);
+
+extern void mnt_make_longterm(struct vfsmount *);
+extern void mnt_make_shortterm(struct vfsmount *);
extern void __init mnt_init(void);
diff --git a/fs/locks.c b/fs/locks.c
index 08415b2..0f39982 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -444,15 +444,9 @@
fl->fl_file->f_owner.signum = 0;
}
-static int lease_mylease_callback(struct file_lock *fl, struct file_lock *try)
-{
- return fl->fl_file == try->fl_file;
-}
-
static const struct lock_manager_operations lease_manager_ops = {
.fl_break = lease_break_callback,
.fl_release_private = lease_release_private_callback,
- .fl_mylease = lease_mylease_callback,
.fl_change = lease_modify,
};
@@ -1405,7 +1399,7 @@
for (before = &inode->i_flock;
((fl = *before) != NULL) && IS_LEASE(fl);
before = &fl->fl_next) {
- if (lease->fl_lmops->fl_mylease(fl, lease))
+ if (fl->fl_file == filp)
my_before = before;
else if (fl->fl_type == (F_INPROGRESS | F_UNLCK))
/*
diff --git a/fs/mpage.c b/fs/mpage.c
index fd56ca2..d78455a 100644
--- a/fs/mpage.c
+++ b/fs/mpage.c
@@ -40,7 +40,7 @@
* status of that page is hard. See end_buffer_async_read() for the details.
* There is no point in duplicating all that complexity.
*/
-static void mpage_end_io_read(struct bio *bio, int err)
+static void mpage_end_io(struct bio *bio, int err)
{
const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
@@ -50,44 +50,29 @@
if (--bvec >= bio->bi_io_vec)
prefetchw(&bvec->bv_page->flags);
-
- if (uptodate) {
- SetPageUptodate(page);
- } else {
- ClearPageUptodate(page);
- SetPageError(page);
+ if (bio_data_dir(bio) == READ) {
+ if (uptodate) {
+ SetPageUptodate(page);
+ } else {
+ ClearPageUptodate(page);
+ SetPageError(page);
+ }
+ unlock_page(page);
+ } else { /* bio_data_dir(bio) == WRITE */
+ if (!uptodate) {
+ SetPageError(page);
+ if (page->mapping)
+ set_bit(AS_EIO, &page->mapping->flags);
+ }
+ end_page_writeback(page);
}
- unlock_page(page);
- } while (bvec >= bio->bi_io_vec);
- bio_put(bio);
-}
-
-static void mpage_end_io_write(struct bio *bio, int err)
-{
- const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
- struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
-
- do {
- struct page *page = bvec->bv_page;
-
- if (--bvec >= bio->bi_io_vec)
- prefetchw(&bvec->bv_page->flags);
-
- if (!uptodate){
- SetPageError(page);
- if (page->mapping)
- set_bit(AS_EIO, &page->mapping->flags);
- }
- end_page_writeback(page);
} while (bvec >= bio->bi_io_vec);
bio_put(bio);
}
static struct bio *mpage_bio_submit(int rw, struct bio *bio)
{
- bio->bi_end_io = mpage_end_io_read;
- if (rw == WRITE)
- bio->bi_end_io = mpage_end_io_write;
+ bio->bi_end_io = mpage_end_io;
submit_bio(rw, bio);
return NULL;
}
diff --git a/fs/namei.c b/fs/namei.c
index 0b14f69..8f7b41a 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -368,18 +368,6 @@
EXPORT_SYMBOL(path_get);
/**
- * path_get_long - get a long reference to a path
- * @path: path to get the reference to
- *
- * Given a path increment the reference count to the dentry and the vfsmount.
- */
-void path_get_long(struct path *path)
-{
- mntget_long(path->mnt);
- dget(path->dentry);
-}
-
-/**
* path_put - put a reference to a path
* @path: path to put the reference to
*
@@ -393,18 +381,6 @@
EXPORT_SYMBOL(path_put);
/**
- * path_put_long - put a long reference to a path
- * @path: path to put the reference to
- *
- * Given a path decrement the reference count to the dentry and the vfsmount.
- */
-void path_put_long(struct path *path)
-{
- dput(path->dentry);
- mntput_long(path->mnt);
-}
-
-/**
* nameidata_drop_rcu - drop this nameidata out of rcu-walk
* @nd: nameidata pathwalk data to drop
* Returns: 0 on success, -ECHILD on failure
@@ -479,6 +455,14 @@
struct fs_struct *fs = current->fs;
struct dentry *parent = nd->path.dentry;
+ /*
+ * It can be possible to revalidate the dentry that we started
+ * the path walk with. force_reval_path may also revalidate the
+ * dentry already committed to the nameidata.
+ */
+ if (unlikely(parent == dentry))
+ return nameidata_drop_rcu(nd);
+
BUG_ON(!(nd->flags & LOOKUP_RCU));
if (nd->root.mnt) {
spin_lock(&fs->lock);
@@ -583,6 +567,13 @@
fput(nd->intent.open.file);
}
+/*
+ * Call d_revalidate and handle filesystems that request rcu-walk
+ * to be dropped. This may be called and return in rcu-walk mode,
+ * regardless of success or error. If -ECHILD is returned, the caller
+ * must return -ECHILD back up the path walk stack so path walk may
+ * be restarted in ref-walk mode.
+ */
static int d_revalidate(struct dentry *dentry, struct nameidata *nd)
{
int status;
@@ -673,6 +664,9 @@
return 0;
if (!status) {
+ /* Don't d_invalidate in rcu-walk mode */
+ if (nameidata_drop_rcu(nd))
+ return -ECHILD;
d_invalidate(dentry);
status = -ESTALE;
}
@@ -761,7 +755,8 @@
mntput(path->mnt);
}
-static inline void path_to_nameidata(struct path *path, struct nameidata *nd)
+static inline void path_to_nameidata(const struct path *path,
+ struct nameidata *nd)
{
if (!(nd->flags & LOOKUP_RCU)) {
dput(nd->path.dentry);
@@ -773,20 +768,16 @@
}
static __always_inline int
-__do_follow_link(struct path *path, struct nameidata *nd, void **p)
+__do_follow_link(const struct path *link, struct nameidata *nd, void **p)
{
int error;
- struct dentry *dentry = path->dentry;
+ struct dentry *dentry = link->dentry;
- touch_atime(path->mnt, dentry);
+ touch_atime(link->mnt, dentry);
nd_set_link(nd, NULL);
- if (path->mnt != nd->path.mnt) {
- path_to_nameidata(path, nd);
- nd->inode = nd->path.dentry->d_inode;
- dget(dentry);
- }
- mntget(path->mnt);
+ if (link->mnt == nd->path.mnt)
+ mntget(link->mnt);
nd->last_type = LAST_BIND;
*p = dentry->d_inode->i_op->follow_link(dentry, nd);
@@ -877,54 +868,169 @@
}
/*
- * serialization is taken care of in namespace.c
+ * Perform an automount
+ * - return -EISDIR to tell follow_managed() to stop and return the path we
+ * were called with.
*/
-static void __follow_mount_rcu(struct nameidata *nd, struct path *path,
- struct inode **inode)
+static int follow_automount(struct path *path, unsigned flags,
+ bool *need_mntput)
{
- while (d_mountpoint(path->dentry)) {
- struct vfsmount *mounted;
- mounted = __lookup_mnt(path->mnt, path->dentry, 1);
- if (!mounted)
- return;
- path->mnt = mounted;
- path->dentry = mounted->mnt_root;
- nd->seq = read_seqcount_begin(&path->dentry->d_seq);
- *inode = path->dentry->d_inode;
- }
-}
+ struct vfsmount *mnt;
+ int err;
-static int __follow_mount(struct path *path)
-{
- int res = 0;
- while (d_mountpoint(path->dentry)) {
- struct vfsmount *mounted = lookup_mnt(path);
- if (!mounted)
- break;
+ if (!path->dentry->d_op || !path->dentry->d_op->d_automount)
+ return -EREMOTE;
+
+ /* We don't want to mount if someone supplied AT_NO_AUTOMOUNT
+ * and this is the terminal part of the path.
+ */
+ if ((flags & LOOKUP_NO_AUTOMOUNT) && !(flags & LOOKUP_CONTINUE))
+ return -EISDIR; /* we actually want to stop here */
+
+ /* We want to mount if someone is trying to open/create a file of any
+ * type under the mountpoint, wants to traverse through the mountpoint
+ * or wants to open the mounted directory.
+ *
+ * We don't want to mount if someone's just doing a stat and they've
+ * set AT_SYMLINK_NOFOLLOW - unless they're stat'ing a directory and
+ * appended a '/' to the name.
+ */
+ if (!(flags & LOOKUP_FOLLOW) &&
+ !(flags & (LOOKUP_CONTINUE | LOOKUP_DIRECTORY |
+ LOOKUP_OPEN | LOOKUP_CREATE)))
+ return -EISDIR;
+
+ current->total_link_count++;
+ if (current->total_link_count >= 40)
+ return -ELOOP;
+
+ mnt = path->dentry->d_op->d_automount(path);
+ if (IS_ERR(mnt)) {
+ /*
+ * The filesystem is allowed to return -EISDIR here to indicate
+ * it doesn't want to automount. For instance, autofs would do
+ * this so that its userspace daemon can mount on this dentry.
+ *
+ * However, we can only permit this if it's a terminal point in
+ * the path being looked up; if it wasn't then the remainder of
+ * the path is inaccessible and we should say so.
+ */
+ if (PTR_ERR(mnt) == -EISDIR && (flags & LOOKUP_CONTINUE))
+ return -EREMOTE;
+ return PTR_ERR(mnt);
+ }
+
+ if (!mnt) /* mount collision */
+ return 0;
+
+ /* The new mount record should have at least 2 refs to prevent it being
+ * expired before we get a chance to add it
+ */
+ BUG_ON(mnt_get_count(mnt) < 2);
+
+ if (mnt->mnt_sb == path->mnt->mnt_sb &&
+ mnt->mnt_root == path->dentry) {
+ mnt_clear_expiry(mnt);
+ mntput(mnt);
+ mntput(mnt);
+ return -ELOOP;
+ }
+
+ /* We need to add the mountpoint to the parent. The filesystem may
+ * have placed it on an expiry list, and so we need to make sure it
+ * won't be expired under us if do_add_mount() fails (do_add_mount()
+ * will eat a reference unconditionally).
+ */
+ mntget(mnt);
+ err = do_add_mount(mnt, path, path->mnt->mnt_flags | MNT_SHRINKABLE);
+ switch (err) {
+ case -EBUSY:
+ /* Someone else made a mount here whilst we were busy */
+ err = 0;
+ default:
+ mnt_clear_expiry(mnt);
+ mntput(mnt);
+ mntput(mnt);
+ return err;
+ case 0:
+ mntput(mnt);
dput(path->dentry);
- if (res)
+ if (*need_mntput)
mntput(path->mnt);
- path->mnt = mounted;
- path->dentry = dget(mounted->mnt_root);
- res = 1;
+ path->mnt = mnt;
+ path->dentry = dget(mnt->mnt_root);
+ *need_mntput = true;
+ return 0;
}
- return res;
}
-static void follow_mount(struct path *path)
+/*
+ * Handle a dentry that is managed in some way.
+ * - Flagged for transit management (autofs)
+ * - Flagged as mountpoint
+ * - Flagged as automount point
+ *
+ * This may only be called in refwalk mode.
+ *
+ * Serialization is taken care of in namespace.c
+ */
+static int follow_managed(struct path *path, unsigned flags)
{
- while (d_mountpoint(path->dentry)) {
- struct vfsmount *mounted = lookup_mnt(path);
- if (!mounted)
- break;
- dput(path->dentry);
- mntput(path->mnt);
- path->mnt = mounted;
- path->dentry = dget(mounted->mnt_root);
+ unsigned managed;
+ bool need_mntput = false;
+ int ret;
+
+ /* Given that we're not holding a lock here, we retain the value in a
+ * local variable for each dentry as we look at it so that we don't see
+ * the components of that value change under us */
+ while (managed = ACCESS_ONCE(path->dentry->d_flags),
+ managed &= DCACHE_MANAGED_DENTRY,
+ unlikely(managed != 0)) {
+ /* Allow the filesystem to manage the transit without i_mutex
+ * being held. */
+ if (managed & DCACHE_MANAGE_TRANSIT) {
+ BUG_ON(!path->dentry->d_op);
+ BUG_ON(!path->dentry->d_op->d_manage);
+ ret = path->dentry->d_op->d_manage(path->dentry,
+ false, false);
+ if (ret < 0)
+ return ret == -EISDIR ? 0 : ret;
+ }
+
+ /* Transit to a mounted filesystem. */
+ if (managed & DCACHE_MOUNTED) {
+ struct vfsmount *mounted = lookup_mnt(path);
+ if (mounted) {
+ dput(path->dentry);
+ if (need_mntput)
+ mntput(path->mnt);
+ path->mnt = mounted;
+ path->dentry = dget(mounted->mnt_root);
+ need_mntput = true;
+ continue;
+ }
+
+ /* Something is mounted on this dentry in another
+ * namespace and/or whatever was mounted there in this
+ * namespace got unmounted before we managed to get the
+ * vfsmount_lock */
+ }
+
+ /* Handle an automount point */
+ if (managed & DCACHE_NEED_AUTOMOUNT) {
+ ret = follow_automount(path, flags, &need_mntput);
+ if (ret < 0)
+ return ret == -EISDIR ? 0 : ret;
+ continue;
+ }
+
+ /* We didn't change the current path point */
+ break;
}
+ return 0;
}
-int follow_down(struct path *path)
+int follow_down_one(struct path *path)
{
struct vfsmount *mounted;
@@ -939,13 +1045,41 @@
return 0;
}
+/*
+ * Skip to top of mountpoint pile in rcuwalk mode. We abort the rcu-walk if we
+ * meet a managed dentry and we're not walking to "..". True is returned to
+ * continue, false to abort.
+ */
+static bool __follow_mount_rcu(struct nameidata *nd, struct path *path,
+ struct inode **inode, bool reverse_transit)
+{
+ while (d_mountpoint(path->dentry)) {
+ struct vfsmount *mounted;
+ if (unlikely(path->dentry->d_flags & DCACHE_MANAGE_TRANSIT) &&
+ !reverse_transit &&
+ path->dentry->d_op->d_manage(path->dentry, false, true) < 0)
+ return false;
+ mounted = __lookup_mnt(path->mnt, path->dentry, 1);
+ if (!mounted)
+ break;
+ path->mnt = mounted;
+ path->dentry = mounted->mnt_root;
+ nd->seq = read_seqcount_begin(&path->dentry->d_seq);
+ *inode = path->dentry->d_inode;
+ }
+
+ if (unlikely(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT))
+ return reverse_transit;
+ return true;
+}
+
static int follow_dotdot_rcu(struct nameidata *nd)
{
struct inode *inode = nd->inode;
set_root_rcu(nd);
- while(1) {
+ while (1) {
if (nd->path.dentry == nd->root.dentry &&
nd->path.mnt == nd->root.mnt) {
break;
@@ -968,12 +1102,80 @@
nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq);
inode = nd->path.dentry->d_inode;
}
- __follow_mount_rcu(nd, &nd->path, &inode);
+ __follow_mount_rcu(nd, &nd->path, &inode, true);
nd->inode = inode;
return 0;
}
+/*
+ * Follow down to the covering mount currently visible to userspace. At each
+ * point, the filesystem owning that dentry may be queried as to whether the
+ * caller is permitted to proceed or not.
+ *
+ * Care must be taken as namespace_sem may be held (indicated by mounting_here
+ * being true).
+ */
+int follow_down(struct path *path, bool mounting_here)
+{
+ unsigned managed;
+ int ret;
+
+ while (managed = ACCESS_ONCE(path->dentry->d_flags),
+ unlikely(managed & DCACHE_MANAGED_DENTRY)) {
+ /* Allow the filesystem to manage the transit without i_mutex
+ * being held.
+ *
+ * We indicate to the filesystem if someone is trying to mount
+ * something here. This gives autofs the chance to deny anyone
+ * other than its daemon the right to mount on its
+ * superstructure.
+ *
+ * The filesystem may sleep at this point.
+ */
+ if (managed & DCACHE_MANAGE_TRANSIT) {
+ BUG_ON(!path->dentry->d_op);
+ BUG_ON(!path->dentry->d_op->d_manage);
+ ret = path->dentry->d_op->d_manage(
+ path->dentry, mounting_here, false);
+ if (ret < 0)
+ return ret == -EISDIR ? 0 : ret;
+ }
+
+ /* Transit to a mounted filesystem. */
+ if (managed & DCACHE_MOUNTED) {
+ struct vfsmount *mounted = lookup_mnt(path);
+ if (!mounted)
+ break;
+ dput(path->dentry);
+ mntput(path->mnt);
+ path->mnt = mounted;
+ path->dentry = dget(mounted->mnt_root);
+ continue;
+ }
+
+ /* Don't handle automount points here */
+ break;
+ }
+ return 0;
+}
+
+/*
+ * Skip to top of mountpoint pile in refwalk mode for follow_dotdot()
+ */
+static void follow_mount(struct path *path)
+{
+ while (d_mountpoint(path->dentry)) {
+ struct vfsmount *mounted = lookup_mnt(path);
+ if (!mounted)
+ break;
+ dput(path->dentry);
+ mntput(path->mnt);
+ path->mnt = mounted;
+ path->dentry = dget(mounted->mnt_root);
+ }
+}
+
static void follow_dotdot(struct nameidata *nd)
{
set_root(nd);
@@ -1038,12 +1240,14 @@
struct vfsmount *mnt = nd->path.mnt;
struct dentry *dentry, *parent = nd->path.dentry;
struct inode *dir;
+ int err;
+
/*
* See if the low-level filesystem might want
* to use its own hash..
*/
if (unlikely(parent->d_flags & DCACHE_OP_HASH)) {
- int err = parent->d_op->d_hash(parent, nd->inode, name);
+ err = parent->d_op->d_hash(parent, nd->inode, name);
if (err < 0)
return err;
}
@@ -1070,22 +1274,28 @@
nd->seq = seq;
if (dentry->d_flags & DCACHE_OP_REVALIDATE)
goto need_revalidate;
+done2:
path->mnt = mnt;
path->dentry = dentry;
- __follow_mount_rcu(nd, path, inode);
- } else {
- dentry = __d_lookup(parent, name);
- if (!dentry)
- goto need_lookup;
-found:
- if (dentry->d_flags & DCACHE_OP_REVALIDATE)
- goto need_revalidate;
-done:
- path->mnt = mnt;
- path->dentry = dentry;
- __follow_mount(path);
- *inode = path->dentry->d_inode;
+ if (likely(__follow_mount_rcu(nd, path, inode, false)))
+ return 0;
+ if (nameidata_drop_rcu(nd))
+ return -ECHILD;
+ /* fallthru */
}
+ dentry = __d_lookup(parent, name);
+ if (!dentry)
+ goto need_lookup;
+found:
+ if (dentry->d_flags & DCACHE_OP_REVALIDATE)
+ goto need_revalidate;
+done:
+ path->mnt = mnt;
+ path->dentry = dentry;
+ err = follow_managed(path, nd->flags);
+ if (unlikely(err < 0))
+ return err;
+ *inode = path->dentry->d_inode;
return 0;
need_lookup:
@@ -1124,6 +1334,8 @@
goto need_lookup;
if (IS_ERR(dentry))
goto fail;
+ if (nd->flags & LOOKUP_RCU)
+ goto done2;
goto done;
fail:
@@ -1131,17 +1343,6 @@
}
/*
- * This is a temporary kludge to deal with "automount" symlinks; proper
- * solution is to trigger them on follow_mount(), so that do_lookup()
- * would DTRT. To be killed before 2.6.34-final.
- */
-static inline int follow_on_final(struct inode *inode, unsigned lookup_flags)
-{
- return inode && unlikely(inode->i_op->follow_link) &&
- ((lookup_flags & LOOKUP_FOLLOW) || S_ISDIR(inode->i_mode));
-}
-
-/*
* Name resolution.
* This is the basic name resolution function, turning a pathname into
* the final dentry. We expect 'base' to be positive and a directory.
@@ -1279,7 +1480,8 @@
err = do_lookup(nd, &this, &next, &inode);
if (err)
break;
- if (follow_on_final(inode, lookup_flags)) {
+ if (inode && unlikely(inode->i_op->follow_link) &&
+ (lookup_flags & LOOKUP_FOLLOW)) {
if (nameidata_dentry_drop_rcu_maybe(nd, next.dentry))
return -ECHILD;
BUG_ON(inode != next.dentry->d_inode);
@@ -2105,11 +2307,13 @@
dir = nd->path.dentry;
case LAST_DOT:
if (need_reval_dot(dir)) {
- error = d_revalidate(nd->path.dentry, nd);
- if (!error)
- error = -ESTALE;
- if (error < 0)
+ int status = d_revalidate(nd->path.dentry, nd);
+ if (!status)
+ status = -ESTALE;
+ if (status < 0) {
+ error = status;
goto exit;
+ }
}
/* fallthrough */
case LAST_ROOT:
@@ -2179,11 +2383,9 @@
if (open_flag & O_EXCL)
goto exit_dput;
- if (__follow_mount(path)) {
- error = -ELOOP;
- if (open_flag & O_NOFOLLOW)
- goto exit_dput;
- }
+ error = follow_managed(path, nd->flags);
+ if (error < 0)
+ goto exit_dput;
error = -ENOENT;
if (!path->dentry->d_inode)
@@ -2328,11 +2530,11 @@
nd.flags = flags;
filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname);
while (unlikely(!filp)) { /* trailing symlink */
- struct path holder;
+ struct path link = path;
+ struct inode *linki = link.dentry->d_inode;
void *cookie;
error = -ELOOP;
- /* S_ISDIR part is a temporary automount kludge */
- if (!(nd.flags & LOOKUP_FOLLOW) && !S_ISDIR(nd.inode->i_mode))
+ if (!(nd.flags & LOOKUP_FOLLOW))
goto exit_dput;
if (count++ == 32)
goto exit_dput;
@@ -2348,23 +2550,22 @@
* just set LAST_BIND.
*/
nd.flags |= LOOKUP_PARENT;
- error = security_inode_follow_link(path.dentry, &nd);
+ error = security_inode_follow_link(link.dentry, &nd);
if (error)
goto exit_dput;
- error = __do_follow_link(&path, &nd, &cookie);
+ error = __do_follow_link(&link, &nd, &cookie);
if (unlikely(error)) {
- if (!IS_ERR(cookie) && nd.inode->i_op->put_link)
- nd.inode->i_op->put_link(path.dentry, &nd, cookie);
+ if (!IS_ERR(cookie) && linki->i_op->put_link)
+ linki->i_op->put_link(link.dentry, &nd, cookie);
/* nd.path had been dropped */
- nd.path = path;
+ nd.path = link;
goto out_path;
}
- holder = path;
nd.flags &= ~LOOKUP_PARENT;
filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname);
- if (nd.inode->i_op->put_link)
- nd.inode->i_op->put_link(holder.dentry, &nd, cookie);
- path_put(&holder);
+ if (linki->i_op->put_link)
+ linki->i_op->put_link(link.dentry, &nd, cookie);
+ path_put(&link);
}
out:
if (nd.root.mnt)
@@ -3392,6 +3593,7 @@
};
EXPORT_SYMBOL(user_path_at);
+EXPORT_SYMBOL(follow_down_one);
EXPORT_SYMBOL(follow_down);
EXPORT_SYMBOL(follow_up);
EXPORT_SYMBOL(get_write_access); /* binfmt_aout */
diff --git a/fs/namespace.c b/fs/namespace.c
index 3ddfd90..48809e2 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -183,7 +183,7 @@
unsigned int mnt_get_count(struct vfsmount *mnt)
{
#ifdef CONFIG_SMP
- unsigned int count = atomic_read(&mnt->mnt_longrefs);
+ unsigned int count = 0;
int cpu;
for_each_possible_cpu(cpu) {
@@ -217,7 +217,7 @@
if (!mnt->mnt_pcp)
goto out_free_devname;
- atomic_set(&mnt->mnt_longrefs, 1);
+ this_cpu_add(mnt->mnt_pcp->mnt_count, 1);
#else
mnt->mnt_count = 1;
mnt->mnt_writers = 0;
@@ -624,8 +624,11 @@
BUG_ON(parent == mnt);
list_add_tail(&head, &mnt->mnt_list);
- list_for_each_entry(m, &head, mnt_list)
+ list_for_each_entry(m, &head, mnt_list) {
m->mnt_ns = n;
+ atomic_inc(&m->mnt_longterm);
+ }
+
list_splice(&head, n->list.prev);
list_add_tail(&mnt->mnt_hash, mount_hashtable +
@@ -734,51 +737,30 @@
deactivate_super(sb);
}
-#ifdef CONFIG_SMP
-static inline void __mntput(struct vfsmount *mnt, int longrefs)
+static void mntput_no_expire(struct vfsmount *mnt)
{
- if (!longrefs) {
put_again:
- br_read_lock(vfsmount_lock);
- if (likely(atomic_read(&mnt->mnt_longrefs))) {
- mnt_dec_count(mnt);
- br_read_unlock(vfsmount_lock);
- return;
- }
+#ifdef CONFIG_SMP
+ br_read_lock(vfsmount_lock);
+ if (likely(atomic_read(&mnt->mnt_longterm))) {
+ mnt_dec_count(mnt);
br_read_unlock(vfsmount_lock);
- } else {
- BUG_ON(!atomic_read(&mnt->mnt_longrefs));
- if (atomic_add_unless(&mnt->mnt_longrefs, -1, 1))
- return;
+ return;
}
+ br_read_unlock(vfsmount_lock);
br_write_lock(vfsmount_lock);
- if (!longrefs)
- mnt_dec_count(mnt);
- else
- atomic_dec(&mnt->mnt_longrefs);
+ mnt_dec_count(mnt);
if (mnt_get_count(mnt)) {
br_write_unlock(vfsmount_lock);
return;
}
- if (unlikely(mnt->mnt_pinned)) {
- mnt_add_count(mnt, mnt->mnt_pinned + 1);
- mnt->mnt_pinned = 0;
- br_write_unlock(vfsmount_lock);
- acct_auto_close_mnt(mnt);
- goto put_again;
- }
- br_write_unlock(vfsmount_lock);
- mntfree(mnt);
-}
#else
-static inline void __mntput(struct vfsmount *mnt, int longrefs)
-{
-put_again:
mnt_dec_count(mnt);
if (likely(mnt_get_count(mnt)))
return;
br_write_lock(vfsmount_lock);
+#endif
if (unlikely(mnt->mnt_pinned)) {
mnt_add_count(mnt, mnt->mnt_pinned + 1);
mnt->mnt_pinned = 0;
@@ -789,12 +771,6 @@
br_write_unlock(vfsmount_lock);
mntfree(mnt);
}
-#endif
-
-static void mntput_no_expire(struct vfsmount *mnt)
-{
- __mntput(mnt, 0);
-}
void mntput(struct vfsmount *mnt)
{
@@ -802,7 +778,7 @@
/* avoid cacheline pingpong, hope gcc doesn't get "smart" */
if (unlikely(mnt->mnt_expiry_mark))
mnt->mnt_expiry_mark = 0;
- __mntput(mnt, 0);
+ mntput_no_expire(mnt);
}
}
EXPORT_SYMBOL(mntput);
@@ -815,33 +791,6 @@
}
EXPORT_SYMBOL(mntget);
-void mntput_long(struct vfsmount *mnt)
-{
-#ifdef CONFIG_SMP
- if (mnt) {
- /* avoid cacheline pingpong, hope gcc doesn't get "smart" */
- if (unlikely(mnt->mnt_expiry_mark))
- mnt->mnt_expiry_mark = 0;
- __mntput(mnt, 1);
- }
-#else
- mntput(mnt);
-#endif
-}
-EXPORT_SYMBOL(mntput_long);
-
-struct vfsmount *mntget_long(struct vfsmount *mnt)
-{
-#ifdef CONFIG_SMP
- if (mnt)
- atomic_inc(&mnt->mnt_longrefs);
- return mnt;
-#else
- return mntget(mnt);
-#endif
-}
-EXPORT_SYMBOL(mntget_long);
-
void mnt_pin(struct vfsmount *mnt)
{
br_write_lock(vfsmount_lock);
@@ -1216,7 +1165,7 @@
dput(dentry);
mntput(m);
}
- mntput_long(mnt);
+ mntput(mnt);
}
}
@@ -1226,19 +1175,21 @@
*/
void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill)
{
+ LIST_HEAD(tmp_list);
struct vfsmount *p;
for (p = mnt; p; p = next_mnt(p, mnt))
- list_move(&p->mnt_hash, kill);
+ list_move(&p->mnt_hash, &tmp_list);
if (propagate)
- propagate_umount(kill);
+ propagate_umount(&tmp_list);
- list_for_each_entry(p, kill, mnt_hash) {
+ list_for_each_entry(p, &tmp_list, mnt_hash) {
list_del_init(&p->mnt_expire);
list_del_init(&p->mnt_list);
__touch_mnt_namespace(p->mnt_ns);
p->mnt_ns = NULL;
+ atomic_dec(&p->mnt_longterm);
list_del_init(&p->mnt_child);
if (p->mnt_parent != p) {
p->mnt_parent->mnt_ghosts++;
@@ -1246,6 +1197,7 @@
}
change_mnt_propagation(p, MS_PRIVATE);
}
+ list_splice(&tmp_list, kill);
}
static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts);
@@ -1844,9 +1796,10 @@
return err;
down_write(&namespace_sem);
- while (d_mountpoint(path->dentry) &&
- follow_down(path))
- ;
+ err = follow_down(path, true);
+ if (err < 0)
+ goto out;
+
err = -EINVAL;
if (!check_mnt(path->mnt) || !check_mnt(old_path.mnt))
goto out;
@@ -1924,15 +1877,14 @@
if (IS_ERR(mnt))
return PTR_ERR(mnt);
- return do_add_mount(mnt, path, mnt_flags, NULL);
+ return do_add_mount(mnt, path, mnt_flags);
}
/*
* add a mount into a namespace's mount tree
- * - provide the option of adding the new mount to an expiration list
+ * - this unconditionally eats one of the caller's references to newmnt.
*/
-int do_add_mount(struct vfsmount *newmnt, struct path *path,
- int mnt_flags, struct list_head *fslist)
+int do_add_mount(struct vfsmount *newmnt, struct path *path, int mnt_flags)
{
int err;
@@ -1940,9 +1892,10 @@
down_write(&namespace_sem);
/* Something was mounted here while we slept */
- while (d_mountpoint(path->dentry) &&
- follow_down(path))
- ;
+ err = follow_down(path, true);
+ if (err < 0)
+ goto unlock;
+
err = -EINVAL;
if (!(mnt_flags & MNT_SHRINKABLE) && !check_mnt(path->mnt))
goto unlock;
@@ -1961,19 +1914,45 @@
if ((err = graft_tree(newmnt, path)))
goto unlock;
- if (fslist) /* add to the specified expiration list */
- list_add_tail(&newmnt->mnt_expire, fslist);
-
up_write(&namespace_sem);
return 0;
unlock:
up_write(&namespace_sem);
- mntput_long(newmnt);
+ mntput(newmnt);
return err;
}
-EXPORT_SYMBOL_GPL(do_add_mount);
+/**
+ * mnt_set_expiry - Put a mount on an expiration list
+ * @mnt: The mount to list.
+ * @expiry_list: The list to add the mount to.
+ */
+void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list)
+{
+ down_write(&namespace_sem);
+ br_write_lock(vfsmount_lock);
+
+ list_add_tail(&mnt->mnt_expire, expiry_list);
+
+ br_write_unlock(vfsmount_lock);
+ up_write(&namespace_sem);
+}
+EXPORT_SYMBOL(mnt_set_expiry);
+
+/*
+ * Remove a vfsmount from any expiration list it may be on
+ */
+void mnt_clear_expiry(struct vfsmount *mnt)
+{
+ if (!list_empty(&mnt->mnt_expire)) {
+ down_write(&namespace_sem);
+ br_write_lock(vfsmount_lock);
+ list_del_init(&mnt->mnt_expire);
+ br_write_unlock(vfsmount_lock);
+ up_write(&namespace_sem);
+ }
+}
/*
* process a list of expirable mountpoints with the intent of discarding any
@@ -2262,6 +2241,20 @@
return new_ns;
}
+void mnt_make_longterm(struct vfsmount *mnt)
+{
+ atomic_inc(&mnt->mnt_longterm);
+}
+
+void mnt_make_shortterm(struct vfsmount *mnt)
+{
+ if (atomic_add_unless(&mnt->mnt_longterm, -1, 1))
+ return;
+ br_write_lock(vfsmount_lock);
+ atomic_dec(&mnt->mnt_longterm);
+ br_write_unlock(vfsmount_lock);
+}
+
/*
* Allocate a new namespace structure and populate it with contents
* copied from the namespace of the passed in task structure.
@@ -2299,14 +2292,19 @@
q = new_ns->root;
while (p) {
q->mnt_ns = new_ns;
+ atomic_inc(&q->mnt_longterm);
if (fs) {
if (p == fs->root.mnt) {
+ fs->root.mnt = mntget(q);
+ atomic_inc(&q->mnt_longterm);
+ mnt_make_shortterm(p);
rootmnt = p;
- fs->root.mnt = mntget_long(q);
}
if (p == fs->pwd.mnt) {
+ fs->pwd.mnt = mntget(q);
+ atomic_inc(&q->mnt_longterm);
+ mnt_make_shortterm(p);
pwdmnt = p;
- fs->pwd.mnt = mntget_long(q);
}
}
p = next_mnt(p, mnt_ns->root);
@@ -2315,9 +2313,9 @@
up_write(&namespace_sem);
if (rootmnt)
- mntput_long(rootmnt);
+ mntput(rootmnt);
if (pwdmnt)
- mntput_long(pwdmnt);
+ mntput(pwdmnt);
return new_ns;
}
@@ -2350,6 +2348,7 @@
new_ns = alloc_mnt_ns();
if (!IS_ERR(new_ns)) {
mnt->mnt_ns = new_ns;
+ atomic_inc(&mnt->mnt_longterm);
new_ns->root = mnt;
list_add(&new_ns->list, &new_ns->root->mnt_list);
}
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 95b081b..2c3eb33 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -970,7 +970,7 @@
{
struct nfs_server *server = NFS_SERVER(inode);
- if (test_bit(NFS_INO_MOUNTPOINT, &NFS_I(inode)->flags))
+ if (IS_AUTOMOUNT(inode))
return 0;
if (nd != NULL) {
/* VFS wants an on-the-wire revalidation */
@@ -1173,6 +1173,7 @@
.d_revalidate = nfs_lookup_revalidate,
.d_delete = nfs_dentry_delete,
.d_iput = nfs_dentry_iput,
+ .d_automount = nfs_d_automount,
};
static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd)
@@ -1246,6 +1247,7 @@
.d_revalidate = nfs_open_revalidate,
.d_delete = nfs_dentry_delete,
.d_iput = nfs_dentry_iput,
+ .d_automount = nfs_d_automount,
};
/*
@@ -1406,11 +1408,15 @@
static int nfs_open_revalidate(struct dentry *dentry, struct nameidata *nd)
{
struct dentry *parent = NULL;
- struct inode *inode = dentry->d_inode;
+ struct inode *inode;
struct inode *dir;
struct nfs_open_context *ctx;
int openflags, ret = 0;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ inode = dentry->d_inode;
if (!is_atomic_open(nd) || d_mountpoint(dentry))
goto no_open;
@@ -1579,6 +1585,7 @@
{
struct iattr attr;
int error;
+ int open_flags = 0;
dfprintk(VFS, "NFS: create(%s/%ld), %s\n",
dir->i_sb->s_id, dir->i_ino, dentry->d_name.name);
@@ -1586,7 +1593,10 @@
attr.ia_mode = mode;
attr.ia_valid = ATTR_MODE;
- error = NFS_PROTO(dir)->create(dir, dentry, &attr, 0, NULL);
+ if ((nd->flags & LOOKUP_CREATE) != 0)
+ open_flags = nd->intent.open.flags;
+
+ error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags, NULL);
if (error != 0)
goto out_err;
return 0;
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index ce00b70..d851242 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -300,7 +300,7 @@
else
inode->i_op = &nfs_mountpoint_inode_operations;
inode->i_fop = NULL;
- set_bit(NFS_INO_MOUNTPOINT, &nfsi->flags);
+ inode->i_flags |= S_AUTOMOUNT;
}
} else if (S_ISLNK(inode->i_mode))
inode->i_op = &nfs_symlink_inode_operations;
@@ -1208,7 +1208,7 @@
/* Update the fsid? */
if (S_ISDIR(inode->i_mode) && (fattr->valid & NFS_ATTR_FATTR_FSID) &&
!nfs_fsid_equal(&server->fsid, &fattr->fsid) &&
- !test_bit(NFS_INO_MOUNTPOINT, &nfsi->flags))
+ !IS_AUTOMOUNT(inode))
server->fsid = fattr->fsid;
/*
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index bfa3a34..4644f04 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -252,6 +252,7 @@
const struct dentry *droot,
const struct dentry *dentry,
char *buffer, ssize_t buflen);
+extern struct vfsmount *nfs_d_automount(struct path *path);
/* getroot.c */
extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *);
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index 74aaf39..f32b860 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -97,9 +97,8 @@
}
/*
- * nfs_follow_mountpoint - handle crossing a mountpoint on the server
- * @dentry - dentry of mountpoint
- * @nd - nameidata info
+ * nfs_d_automount - Handle crossing a mountpoint on the server
+ * @path - The mountpoint
*
* When we encounter a mountpoint on the server, we want to set up
* a mountpoint on the client too, to prevent inode numbers from
@@ -109,87 +108,65 @@
* situation, and that different filesystems may want to use
* different security flavours.
*/
-static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
+struct vfsmount *nfs_d_automount(struct path *path)
{
struct vfsmount *mnt;
- struct nfs_server *server = NFS_SERVER(dentry->d_inode);
+ struct nfs_server *server = NFS_SERVER(path->dentry->d_inode);
struct dentry *parent;
struct nfs_fh *fh = NULL;
struct nfs_fattr *fattr = NULL;
int err;
- dprintk("--> nfs_follow_mountpoint()\n");
+ dprintk("--> nfs_d_automount()\n");
- err = -ESTALE;
- if (IS_ROOT(dentry))
- goto out_err;
+ mnt = ERR_PTR(-ESTALE);
+ if (IS_ROOT(path->dentry))
+ goto out_nofree;
- err = -ENOMEM;
+ mnt = ERR_PTR(-ENOMEM);
fh = nfs_alloc_fhandle();
fattr = nfs_alloc_fattr();
if (fh == NULL || fattr == NULL)
- goto out_err;
+ goto out;
dprintk("%s: enter\n", __func__);
- dput(nd->path.dentry);
- nd->path.dentry = dget(dentry);
- /* Look it up again */
- parent = dget_parent(nd->path.dentry);
+ /* Look it up again to get its attributes */
+ parent = dget_parent(path->dentry);
err = server->nfs_client->rpc_ops->lookup(parent->d_inode,
- &nd->path.dentry->d_name,
+ &path->dentry->d_name,
fh, fattr);
dput(parent);
- if (err != 0)
- goto out_err;
+ if (err != 0) {
+ mnt = ERR_PTR(err);
+ goto out;
+ }
if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL)
- mnt = nfs_do_refmount(nd->path.mnt, nd->path.dentry);
+ mnt = nfs_do_refmount(path->mnt, path->dentry);
else
- mnt = nfs_do_submount(nd->path.mnt, nd->path.dentry, fh,
- fattr);
- err = PTR_ERR(mnt);
+ mnt = nfs_do_submount(path->mnt, path->dentry, fh, fattr);
if (IS_ERR(mnt))
- goto out_err;
+ goto out;
- mntget(mnt);
- err = do_add_mount(mnt, &nd->path, nd->path.mnt->mnt_flags|MNT_SHRINKABLE,
- &nfs_automount_list);
- if (err < 0) {
- mntput(mnt);
- if (err == -EBUSY)
- goto out_follow;
- goto out_err;
- }
- path_put(&nd->path);
- nd->path.mnt = mnt;
- nd->path.dentry = dget(mnt->mnt_root);
+ dprintk("%s: done, success\n", __func__);
+ mntget(mnt); /* prevent immediate expiration */
+ mnt_set_expiry(mnt, &nfs_automount_list);
schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout);
+
out:
nfs_free_fattr(fattr);
nfs_free_fhandle(fh);
- dprintk("%s: done, returned %d\n", __func__, err);
-
- dprintk("<-- nfs_follow_mountpoint() = %d\n", err);
- return ERR_PTR(err);
-out_err:
- path_put(&nd->path);
- goto out;
-out_follow:
- while (d_mountpoint(nd->path.dentry) &&
- follow_down(&nd->path))
- ;
- err = 0;
- goto out;
+out_nofree:
+ dprintk("<-- nfs_follow_mountpoint() = %p\n", mnt);
+ return mnt;
}
const struct inode_operations nfs_mountpoint_inode_operations = {
- .follow_link = nfs_follow_mountpoint,
.getattr = nfs_getattr,
};
const struct inode_operations nfs_referral_inode_operations = {
- .follow_link = nfs_follow_mountpoint,
};
static void nfs_expire_automounts(struct work_struct *work)
diff --git a/include/linux/nfs4_acl.h b/fs/nfsd/acl.h
similarity index 97%
rename from include/linux/nfs4_acl.h
rename to fs/nfsd/acl.h
index c9c05a7..34e5c40 100644
--- a/include/linux/nfs4_acl.h
+++ b/fs/nfsd/acl.h
@@ -1,6 +1,4 @@
/*
- * include/linux/nfs4_acl.c
- *
* Common NFSv4 ACL handling definitions.
*
* Copyright (c) 2002 The Regents of the University of Michigan.
diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index c0fcb7a..8b31e5f 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -1,4 +1,3 @@
-#define MSNFS /* HACK HACK */
/*
* NFS exporting and validation.
*
@@ -1444,9 +1443,6 @@
{ NFSEXP_NOSUBTREECHECK, {"no_subtree_check", ""}},
{ NFSEXP_NOAUTHNLM, {"insecure_locks", ""}},
{ NFSEXP_V4ROOT, {"v4root", ""}},
-#ifdef MSNFS
- { NFSEXP_MSNFS, {"msnfs", ""}},
-#endif
{ 0, {"", ""}}
};
diff --git a/include/linux/nfsd_idmap.h b/fs/nfsd/idmap.h
similarity index 91%
rename from include/linux/nfsd_idmap.h
rename to fs/nfsd/idmap.h
index d4a2ac1..2f3be13 100644
--- a/include/linux/nfsd_idmap.h
+++ b/fs/nfsd/idmap.h
@@ -1,6 +1,4 @@
/*
- * include/linux/nfsd_idmap.h
- *
* Mapping of UID to name and vice versa.
*
* Copyright (c) 2002, 2003 The Regents of the University of
@@ -56,8 +54,8 @@
}
#endif
-int nfsd_map_name_to_uid(struct svc_rqst *, const char *, size_t, __u32 *);
-int nfsd_map_name_to_gid(struct svc_rqst *, const char *, size_t, __u32 *);
+__be32 nfsd_map_name_to_uid(struct svc_rqst *, const char *, size_t, __u32 *);
+__be32 nfsd_map_name_to_gid(struct svc_rqst *, const char *, size_t, __u32 *);
int nfsd_map_uid_to_name(struct svc_rqst *, __u32, char *);
int nfsd_map_gid_to_name(struct svc_rqst *, __u32, char *);
diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index 5b7e302..2247fc9 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -151,10 +151,10 @@
__be32 nfserr;
u32 max_blocksize = svc_max_payload(rqstp);
- dprintk("nfsd: READ(3) %s %lu bytes at %lu\n",
+ dprintk("nfsd: READ(3) %s %lu bytes at %Lu\n",
SVCFH_fmt(&argp->fh),
(unsigned long) argp->count,
- (unsigned long) argp->offset);
+ (unsigned long long) argp->offset);
/* Obtain buffer pointer for payload.
* 1 (status) + 22 (post_op_attr) + 1 (count) + 1 (eof)
@@ -191,10 +191,10 @@
__be32 nfserr;
unsigned long cnt = argp->len;
- dprintk("nfsd: WRITE(3) %s %d bytes at %ld%s\n",
+ dprintk("nfsd: WRITE(3) %s %d bytes at %Lu%s\n",
SVCFH_fmt(&argp->fh),
argp->len,
- (unsigned long) argp->offset,
+ (unsigned long long) argp->offset,
argp->stable? " stable" : "");
fh_copy(&resp->fh, &argp->fh);
diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c
index e480526..ad88f1c 100644
--- a/fs/nfsd/nfs4acl.c
+++ b/fs/nfsd/nfs4acl.c
@@ -36,7 +36,7 @@
#include <linux/slab.h>
#include <linux/nfs_fs.h>
-#include <linux/nfs4_acl.h>
+#include "acl.h"
/* mode bit translations: */
diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 21a63da..3be975e 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -628,10 +628,8 @@
return max(nfsd4_lease/10, (time_t)1) * HZ;
}
-/* Reference counting, callback cleanup, etc., all look racy as heck.
- * And why is cl_cb_set an atomic? */
-int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *conn)
+static int setup_callback_client(struct nfs4_client *clp, struct nfs4_cb_conn *conn, struct nfsd4_session *ses)
{
struct rpc_timeout timeparms = {
.to_initval = max_cb_time(),
@@ -641,6 +639,7 @@
.net = &init_net,
.address = (struct sockaddr *) &conn->cb_addr,
.addrsize = conn->cb_addrlen,
+ .saddress = (struct sockaddr *) &conn->cb_saddr,
.timeout = &timeparms,
.program = &cb_program,
.version = 0,
@@ -657,6 +656,10 @@
args.protocol = XPRT_TRANSPORT_TCP;
clp->cl_cb_ident = conn->cb_ident;
} else {
+ if (!conn->cb_xprt)
+ return -EINVAL;
+ clp->cl_cb_conn.cb_xprt = conn->cb_xprt;
+ clp->cl_cb_session = ses;
args.bc_xprt = conn->cb_xprt;
args.prognumber = clp->cl_cb_session->se_cb_prog;
args.protocol = XPRT_TRANSPORT_BC_TCP;
@@ -679,14 +682,20 @@
(int)clp->cl_name.len, clp->cl_name.data, reason);
}
+static void nfsd4_mark_cb_down(struct nfs4_client *clp, int reason)
+{
+ clp->cl_cb_state = NFSD4_CB_DOWN;
+ warn_no_callback_path(clp, reason);
+}
+
static void nfsd4_cb_probe_done(struct rpc_task *task, void *calldata)
{
struct nfs4_client *clp = container_of(calldata, struct nfs4_client, cl_cb_null);
if (task->tk_status)
- warn_no_callback_path(clp, task->tk_status);
+ nfsd4_mark_cb_down(clp, task->tk_status);
else
- atomic_set(&clp->cl_cb_set, 1);
+ clp->cl_cb_state = NFSD4_CB_UP;
}
static const struct rpc_call_ops nfsd4_cb_probe_ops = {
@@ -709,6 +718,11 @@
static struct workqueue_struct *callback_wq;
+static void run_nfsd4_cb(struct nfsd4_callback *cb)
+{
+ queue_work(callback_wq, &cb->cb_work);
+}
+
static void do_probe_callback(struct nfs4_client *clp)
{
struct nfsd4_callback *cb = &clp->cl_cb_null;
@@ -723,7 +737,7 @@
cb->cb_ops = &nfsd4_cb_probe_ops;
- queue_work(callback_wq, &cb->cb_work);
+ run_nfsd4_cb(cb);
}
/*
@@ -732,14 +746,21 @@
*/
void nfsd4_probe_callback(struct nfs4_client *clp)
{
+ /* XXX: atomicity? Also, should we be using cl_cb_flags? */
+ clp->cl_cb_state = NFSD4_CB_UNKNOWN;
set_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_cb_flags);
do_probe_callback(clp);
}
+void nfsd4_probe_callback_sync(struct nfs4_client *clp)
+{
+ nfsd4_probe_callback(clp);
+ flush_workqueue(callback_wq);
+}
+
void nfsd4_change_callback(struct nfs4_client *clp, struct nfs4_cb_conn *conn)
{
- BUG_ON(atomic_read(&clp->cl_cb_set));
-
+ clp->cl_cb_state = NFSD4_CB_UNKNOWN;
spin_lock(&clp->cl_lock);
memcpy(&clp->cl_cb_conn, conn, sizeof(struct nfs4_cb_conn));
spin_unlock(&clp->cl_lock);
@@ -750,24 +771,14 @@
* If the slot is available, then mark it busy. Otherwise, set the
* thread for sleeping on the callback RPC wait queue.
*/
-static int nfsd41_cb_setup_sequence(struct nfs4_client *clp,
- struct rpc_task *task)
+static bool nfsd41_cb_get_slot(struct nfs4_client *clp, struct rpc_task *task)
{
- u32 *ptr = (u32 *)clp->cl_cb_session->se_sessionid.data;
- int status = 0;
-
- dprintk("%s: %u:%u:%u:%u\n", __func__,
- ptr[0], ptr[1], ptr[2], ptr[3]);
-
if (test_and_set_bit(0, &clp->cl_cb_slot_busy) != 0) {
rpc_sleep_on(&clp->cl_cb_waitq, task, NULL);
dprintk("%s slot is busy\n", __func__);
- status = -EAGAIN;
- goto out;
+ return false;
}
-out:
- dprintk("%s status=%d\n", __func__, status);
- return status;
+ return true;
}
/*
@@ -780,20 +791,19 @@
struct nfs4_delegation *dp = container_of(cb, struct nfs4_delegation, dl_recall);
struct nfs4_client *clp = dp->dl_client;
u32 minorversion = clp->cl_minorversion;
- int status = 0;
cb->cb_minorversion = minorversion;
if (minorversion) {
- status = nfsd41_cb_setup_sequence(clp, task);
- if (status) {
- if (status != -EAGAIN) {
- /* terminate rpc task */
- task->tk_status = status;
- task->tk_action = NULL;
- }
+ if (!nfsd41_cb_get_slot(clp, task))
return;
- }
}
+ spin_lock(&clp->cl_lock);
+ if (list_empty(&cb->cb_per_client)) {
+ /* This is the first call, not a restart */
+ cb->cb_done = false;
+ list_add(&cb->cb_per_client, &clp->cl_callbacks);
+ }
+ spin_unlock(&clp->cl_lock);
rpc_call_start(task);
}
@@ -829,15 +839,18 @@
nfsd4_cb_done(task, calldata);
- if (current_rpc_client == NULL) {
- /* We're shutting down; give up. */
- /* XXX: err, or is it ok just to fall through
- * and rpc_restart_call? */
+ if (current_rpc_client != task->tk_client) {
+ /* We're shutting down or changing cl_cb_client; leave
+ * it to nfsd4_process_cb_update to restart the call if
+ * necessary. */
return;
}
+ if (cb->cb_done)
+ return;
switch (task->tk_status) {
case 0:
+ cb->cb_done = true;
return;
case -EBADHANDLE:
case -NFS4ERR_BAD_STATEID:
@@ -846,32 +859,30 @@
break;
default:
/* Network partition? */
- atomic_set(&clp->cl_cb_set, 0);
- warn_no_callback_path(clp, task->tk_status);
- if (current_rpc_client != task->tk_client) {
- /* queue a callback on the new connection: */
- atomic_inc(&dp->dl_count);
- nfsd4_cb_recall(dp);
- return;
- }
+ nfsd4_mark_cb_down(clp, task->tk_status);
}
if (dp->dl_retries--) {
rpc_delay(task, 2*HZ);
task->tk_status = 0;
rpc_restart_call_prepare(task);
return;
- } else {
- atomic_set(&clp->cl_cb_set, 0);
- warn_no_callback_path(clp, task->tk_status);
}
+ nfsd4_mark_cb_down(clp, task->tk_status);
+ cb->cb_done = true;
}
static void nfsd4_cb_recall_release(void *calldata)
{
struct nfsd4_callback *cb = calldata;
+ struct nfs4_client *clp = cb->cb_clp;
struct nfs4_delegation *dp = container_of(cb, struct nfs4_delegation, dl_recall);
- nfs4_put_delegation(dp);
+ if (cb->cb_done) {
+ spin_lock(&clp->cl_lock);
+ list_del(&cb->cb_per_client);
+ spin_unlock(&clp->cl_lock);
+ nfs4_put_delegation(dp);
+ }
}
static const struct rpc_call_ops nfsd4_cb_recall_ops = {
@@ -906,16 +917,33 @@
flush_workqueue(callback_wq);
}
-void nfsd4_release_cb(struct nfsd4_callback *cb)
+static void nfsd4_release_cb(struct nfsd4_callback *cb)
{
if (cb->cb_ops->rpc_release)
cb->cb_ops->rpc_release(cb);
}
-void nfsd4_process_cb_update(struct nfsd4_callback *cb)
+/* requires cl_lock: */
+static struct nfsd4_conn * __nfsd4_find_backchannel(struct nfs4_client *clp)
+{
+ struct nfsd4_session *s;
+ struct nfsd4_conn *c;
+
+ list_for_each_entry(s, &clp->cl_sessions, se_perclnt) {
+ list_for_each_entry(c, &s->se_conns, cn_persession) {
+ if (c->cn_flags & NFS4_CDFC4_BACK)
+ return c;
+ }
+ }
+ return NULL;
+}
+
+static void nfsd4_process_cb_update(struct nfsd4_callback *cb)
{
struct nfs4_cb_conn conn;
struct nfs4_client *clp = cb->cb_clp;
+ struct nfsd4_session *ses = NULL;
+ struct nfsd4_conn *c;
int err;
/*
@@ -926,6 +954,10 @@
rpc_shutdown_client(clp->cl_cb_client);
clp->cl_cb_client = NULL;
}
+ if (clp->cl_cb_conn.cb_xprt) {
+ svc_xprt_put(clp->cl_cb_conn.cb_xprt);
+ clp->cl_cb_conn.cb_xprt = NULL;
+ }
if (test_bit(NFSD4_CLIENT_KILL, &clp->cl_cb_flags))
return;
spin_lock(&clp->cl_lock);
@@ -936,11 +968,22 @@
BUG_ON(!clp->cl_cb_flags);
clear_bit(NFSD4_CLIENT_CB_UPDATE, &clp->cl_cb_flags);
memcpy(&conn, &cb->cb_clp->cl_cb_conn, sizeof(struct nfs4_cb_conn));
+ c = __nfsd4_find_backchannel(clp);
+ if (c) {
+ svc_xprt_get(c->cn_xprt);
+ conn.cb_xprt = c->cn_xprt;
+ ses = c->cn_session;
+ }
spin_unlock(&clp->cl_lock);
- err = setup_callback_client(clp, &conn);
- if (err)
+ err = setup_callback_client(clp, &conn, ses);
+ if (err) {
warn_no_callback_path(clp, err);
+ return;
+ }
+ /* Yay, the callback channel's back! Restart any callbacks: */
+ list_for_each_entry(cb, &clp->cl_callbacks, cb_per_client)
+ run_nfsd4_cb(cb);
}
void nfsd4_do_callback_rpc(struct work_struct *w)
@@ -965,10 +1008,11 @@
void nfsd4_cb_recall(struct nfs4_delegation *dp)
{
struct nfsd4_callback *cb = &dp->dl_recall;
+ struct nfs4_client *clp = dp->dl_client;
dp->dl_retries = 1;
cb->cb_op = dp;
- cb->cb_clp = dp->dl_client;
+ cb->cb_clp = clp;
cb->cb_msg.rpc_proc = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_RECALL];
cb->cb_msg.rpc_argp = cb;
cb->cb_msg.rpc_resp = cb;
@@ -977,5 +1021,8 @@
cb->cb_ops = &nfsd4_cb_recall_ops;
dp->dl_retries = 1;
- queue_work(callback_wq, &dp->dl_recall.cb_work);
+ INIT_LIST_HEAD(&cb->cb_per_client);
+ cb->cb_done = true;
+
+ run_nfsd4_cb(&dp->dl_recall);
}
diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c
index f0695e8..6d2c397 100644
--- a/fs/nfsd/nfs4idmap.c
+++ b/fs/nfsd/nfs4idmap.c
@@ -33,10 +33,11 @@
*/
#include <linux/module.h>
-#include <linux/nfsd_idmap.h>
#include <linux/seq_file.h>
#include <linux/sched.h>
#include <linux/slab.h>
+#include "idmap.h"
+#include "nfsd.h"
/*
* Cache entry
@@ -514,7 +515,7 @@
return clp->name;
}
-static int
+static __be32
idmap_name_to_id(struct svc_rqst *rqstp, int type, const char *name, u32 namelen,
uid_t *id)
{
@@ -524,15 +525,15 @@
int ret;
if (namelen + 1 > sizeof(key.name))
- return -EINVAL;
+ return nfserr_badowner;
memcpy(key.name, name, namelen);
key.name[namelen] = '\0';
strlcpy(key.authname, rqst_authname(rqstp), sizeof(key.authname));
ret = idmap_lookup(rqstp, nametoid_lookup, &key, &nametoid_cache, &item);
if (ret == -ENOENT)
- ret = -ESRCH; /* nfserr_badname */
+ return nfserr_badowner;
if (ret)
- return ret;
+ return nfserrno(ret);
*id = item->id;
cache_put(&item->h, &nametoid_cache);
return 0;
@@ -560,14 +561,14 @@
return ret;
}
-int
+__be32
nfsd_map_name_to_uid(struct svc_rqst *rqstp, const char *name, size_t namelen,
__u32 *id)
{
return idmap_name_to_id(rqstp, IDMAP_TYPE_USER, name, namelen, id);
}
-int
+__be32
nfsd_map_name_to_gid(struct svc_rqst *rqstp, const char *name, size_t namelen,
__u32 *id)
{
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 0cdfd02..db52546 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -604,9 +604,7 @@
return status;
}
-static __be32
-nfsd4_lookupp(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
- void *arg)
+static __be32 nfsd4_do_lookupp(struct svc_rqst *rqstp, struct svc_fh *fh)
{
struct svc_fh tmp_fh;
__be32 ret;
@@ -615,13 +613,19 @@
ret = exp_pseudoroot(rqstp, &tmp_fh);
if (ret)
return ret;
- if (tmp_fh.fh_dentry == cstate->current_fh.fh_dentry) {
+ if (tmp_fh.fh_dentry == fh->fh_dentry) {
fh_put(&tmp_fh);
return nfserr_noent;
}
fh_put(&tmp_fh);
- return nfsd_lookup(rqstp, &cstate->current_fh,
- "..", 2, &cstate->current_fh);
+ return nfsd_lookup(rqstp, fh, "..", 2, fh);
+}
+
+static __be32
+nfsd4_lookupp(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
+ void *arg)
+{
+ return nfsd4_do_lookupp(rqstp, &cstate->current_fh);
}
static __be32
@@ -769,10 +773,36 @@
} else
secinfo->si_exp = exp;
dput(dentry);
+ if (cstate->minorversion)
+ /* See rfc 5661 section 2.6.3.1.1.8 */
+ fh_put(&cstate->current_fh);
return err;
}
static __be32
+nfsd4_secinfo_no_name(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
+ struct nfsd4_secinfo_no_name *sin)
+{
+ __be32 err;
+
+ switch (sin->sin_style) {
+ case NFS4_SECINFO_STYLE4_CURRENT_FH:
+ break;
+ case NFS4_SECINFO_STYLE4_PARENT:
+ err = nfsd4_do_lookupp(rqstp, &cstate->current_fh);
+ if (err)
+ return err;
+ break;
+ default:
+ return nfserr_inval;
+ }
+ exp_get(cstate->current_fh.fh_export);
+ sin->sin_exp = cstate->current_fh.fh_export;
+ fh_put(&cstate->current_fh);
+ return nfs_ok;
+}
+
+static __be32
nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
struct nfsd4_setattr *setattr)
{
@@ -974,8 +1004,8 @@
* Also note, enforced elsewhere:
* - SEQUENCE other than as first op results in
* NFS4ERR_SEQUENCE_POS. (Enforced in nfsd4_sequence().)
- * - BIND_CONN_TO_SESSION must be the only op in its compound
- * (Will be enforced in nfsd4_bind_conn_to_session().)
+ * - BIND_CONN_TO_SESSION must be the only op in its compound.
+ * (Enforced in nfsd4_bind_conn_to_session().)
* - DESTROY_SESSION must be the final operation in a compound, if
* sessionid's in SEQUENCE and DESTROY_SESSION are the same.
* (Enforced in nfsd4_destroy_session().)
@@ -1126,10 +1156,6 @@
nfsd4_increment_op_stats(op->opnum);
}
- if (!rqstp->rq_usedeferral && status == nfserr_dropit) {
- dprintk("%s Dropit - send NFS4ERR_DELAY\n", __func__);
- status = nfserr_jukebox;
- }
resp->cstate.status = status;
fh_put(&resp->cstate.current_fh);
@@ -1300,6 +1326,11 @@
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
.op_name = "OP_EXCHANGE_ID",
},
+ [OP_BIND_CONN_TO_SESSION] = {
+ .op_func = (nfsd4op_func)nfsd4_bind_conn_to_session,
+ .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
+ .op_name = "OP_BIND_CONN_TO_SESSION",
+ },
[OP_CREATE_SESSION] = {
.op_func = (nfsd4op_func)nfsd4_create_session,
.op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP,
@@ -1320,6 +1351,10 @@
.op_flags = ALLOWED_WITHOUT_FH,
.op_name = "OP_RECLAIM_COMPLETE",
},
+ [OP_SECINFO_NO_NAME] = {
+ .op_func = (nfsd4op_func)nfsd4_secinfo_no_name,
+ .op_name = "OP_SECINFO_NO_NAME",
+ },
};
static const char *nfsd4_op_name(unsigned opnum)
diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index 7e26caa..ffb59ef 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -302,7 +302,6 @@
{
int status;
- /* note: we currently use this path only for minorversion 0 */
if (nfs4_has_reclaimed_state(child->d_name.name, false))
return 0;
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index fbd18c3..d98d021 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -230,7 +230,8 @@
dp->dl_client = clp;
get_nfs4_file(fp);
dp->dl_file = fp;
- nfs4_file_get_access(fp, O_RDONLY);
+ dp->dl_vfs_file = find_readable_file(fp);
+ get_file(dp->dl_vfs_file);
dp->dl_flock = NULL;
dp->dl_type = type;
dp->dl_stateid.si_boot = boot_time;
@@ -252,6 +253,7 @@
if (atomic_dec_and_test(&dp->dl_count)) {
dprintk("NFSD: freeing dp %p\n",dp);
put_nfs4_file(dp->dl_file);
+ fput(dp->dl_vfs_file);
kmem_cache_free(deleg_slab, dp);
num_delegations--;
}
@@ -265,12 +267,10 @@
static void
nfs4_close_delegation(struct nfs4_delegation *dp)
{
- struct file *filp = find_readable_file(dp->dl_file);
-
dprintk("NFSD: close_delegation dp %p\n",dp);
+ /* XXX: do we even need this check?: */
if (dp->dl_flock)
- vfs_setlease(filp, F_UNLCK, &dp->dl_flock);
- nfs4_file_put_access(dp->dl_file, O_RDONLY);
+ vfs_setlease(dp->dl_vfs_file, F_UNLCK, &dp->dl_flock);
}
/* Called under the state lock. */
@@ -642,6 +642,7 @@
free_conn(c);
}
spin_unlock(&clp->cl_lock);
+ nfsd4_probe_callback(clp);
}
static struct nfsd4_conn *alloc_conn(struct svc_rqst *rqstp, u32 flags)
@@ -679,15 +680,12 @@
return register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user);
}
-static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses)
+static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses, u32 dir)
{
struct nfsd4_conn *conn;
- u32 flags = NFS4_CDFC4_FORE;
int ret;
- if (ses->se_flags & SESSION4_BACK_CHAN)
- flags |= NFS4_CDFC4_BACK;
- conn = alloc_conn(rqstp, flags);
+ conn = alloc_conn(rqstp, dir);
if (!conn)
return nfserr_jukebox;
nfsd4_hash_conn(conn, ses);
@@ -698,6 +696,17 @@
return nfs_ok;
}
+static __be32 nfsd4_new_conn_from_crses(struct svc_rqst *rqstp, struct nfsd4_session *ses)
+{
+ u32 dir = NFS4_CDFC4_FORE;
+
+ if (ses->se_flags & SESSION4_BACK_CHAN)
+ dir |= NFS4_CDFC4_BACK;
+
+ return nfsd4_new_conn(rqstp, ses, dir);
+}
+
+/* must be called under client_lock */
static void nfsd4_del_conns(struct nfsd4_session *s)
{
struct nfs4_client *clp = s->se_client;
@@ -749,6 +758,8 @@
*/
slotsize = nfsd4_sanitize_slot_size(fchan->maxresp_cached);
numslots = nfsd4_get_drc_mem(slotsize, fchan->maxreqs);
+ if (numslots < 1)
+ return NULL;
new = alloc_session(slotsize, numslots);
if (!new) {
@@ -769,25 +780,30 @@
idx = hash_sessionid(&new->se_sessionid);
spin_lock(&client_lock);
list_add(&new->se_hash, &sessionid_hashtbl[idx]);
+ spin_lock(&clp->cl_lock);
list_add(&new->se_perclnt, &clp->cl_sessions);
+ spin_unlock(&clp->cl_lock);
spin_unlock(&client_lock);
- status = nfsd4_new_conn(rqstp, new);
+ status = nfsd4_new_conn_from_crses(rqstp, new);
/* whoops: benny points out, status is ignored! (err, or bogus) */
if (status) {
free_session(&new->se_ref);
return NULL;
}
- if (!clp->cl_cb_session && (cses->flags & SESSION4_BACK_CHAN)) {
+ if (cses->flags & SESSION4_BACK_CHAN) {
struct sockaddr *sa = svc_addr(rqstp);
-
- clp->cl_cb_session = new;
- clp->cl_cb_conn.cb_xprt = rqstp->rq_xprt;
- svc_xprt_get(rqstp->rq_xprt);
+ /*
+ * This is a little silly; with sessions there's no real
+ * use for the callback address. Use the peer address
+ * as a reasonable default for now, but consider fixing
+ * the rpc client not to require an address in the
+ * future:
+ */
rpc_copy_addr((struct sockaddr *)&clp->cl_cb_conn.cb_addr, sa);
clp->cl_cb_conn.cb_addrlen = svc_addr_len(sa);
- nfsd4_probe_callback(clp);
}
+ nfsd4_probe_callback(clp);
return new;
}
@@ -817,7 +833,9 @@
unhash_session(struct nfsd4_session *ses)
{
list_del(&ses->se_hash);
+ spin_lock(&ses->se_client->cl_lock);
list_del(&ses->se_perclnt);
+ spin_unlock(&ses->se_client->cl_lock);
}
/* must be called under the client_lock */
@@ -923,8 +941,10 @@
mark_client_expired(clp);
list_del(&clp->cl_lru);
+ spin_lock(&clp->cl_lock);
list_for_each_entry(ses, &clp->cl_sessions, se_perclnt)
list_del_init(&ses->se_hash);
+ spin_unlock(&clp->cl_lock);
}
static void
@@ -1051,12 +1071,13 @@
memcpy(clp->cl_recdir, recdir, HEXDIR_LEN);
atomic_set(&clp->cl_refcount, 0);
- atomic_set(&clp->cl_cb_set, 0);
+ clp->cl_cb_state = NFSD4_CB_UNKNOWN;
INIT_LIST_HEAD(&clp->cl_idhash);
INIT_LIST_HEAD(&clp->cl_strhash);
INIT_LIST_HEAD(&clp->cl_openowners);
INIT_LIST_HEAD(&clp->cl_delegations);
INIT_LIST_HEAD(&clp->cl_lru);
+ INIT_LIST_HEAD(&clp->cl_callbacks);
spin_lock_init(&clp->cl_lock);
INIT_WORK(&clp->cl_cb_null.cb_work, nfsd4_do_callback_rpc);
clp->cl_time = get_seconds();
@@ -1132,54 +1153,55 @@
return NULL;
}
-/*
- * Return 1 iff clp's clientid establishment method matches the use_exchange_id
- * parameter. Matching is based on the fact the at least one of the
- * EXCHGID4_FLAG_USE_{NON_PNFS,PNFS_MDS,PNFS_DS} flags must be set for v4.1
- *
- * FIXME: we need to unify the clientid namespaces for nfsv4.x
- * and correctly deal with client upgrade/downgrade in EXCHANGE_ID
- * and SET_CLIENTID{,_CONFIRM}
- */
-static inline int
-match_clientid_establishment(struct nfs4_client *clp, bool use_exchange_id)
+static bool clp_used_exchangeid(struct nfs4_client *clp)
{
- bool has_exchange_flags = (clp->cl_exchange_flags != 0);
- return use_exchange_id == has_exchange_flags;
-}
+ return clp->cl_exchange_flags != 0;
+}
static struct nfs4_client *
-find_confirmed_client_by_str(const char *dname, unsigned int hashval,
- bool use_exchange_id)
+find_confirmed_client_by_str(const char *dname, unsigned int hashval)
{
struct nfs4_client *clp;
list_for_each_entry(clp, &conf_str_hashtbl[hashval], cl_strhash) {
- if (same_name(clp->cl_recdir, dname) &&
- match_clientid_establishment(clp, use_exchange_id))
+ if (same_name(clp->cl_recdir, dname))
return clp;
}
return NULL;
}
static struct nfs4_client *
-find_unconfirmed_client_by_str(const char *dname, unsigned int hashval,
- bool use_exchange_id)
+find_unconfirmed_client_by_str(const char *dname, unsigned int hashval)
{
struct nfs4_client *clp;
list_for_each_entry(clp, &unconf_str_hashtbl[hashval], cl_strhash) {
- if (same_name(clp->cl_recdir, dname) &&
- match_clientid_establishment(clp, use_exchange_id))
+ if (same_name(clp->cl_recdir, dname))
return clp;
}
return NULL;
}
+static void rpc_svcaddr2sockaddr(struct sockaddr *sa, unsigned short family, union svc_addr_u *svcaddr)
+{
+ switch (family) {
+ case AF_INET:
+ ((struct sockaddr_in *)sa)->sin_family = AF_INET;
+ ((struct sockaddr_in *)sa)->sin_addr = svcaddr->addr;
+ return;
+ case AF_INET6:
+ ((struct sockaddr_in6 *)sa)->sin6_family = AF_INET6;
+ ((struct sockaddr_in6 *)sa)->sin6_addr = svcaddr->addr6;
+ return;
+ }
+}
+
static void
-gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se, u32 scopeid)
+gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se, struct svc_rqst *rqstp)
{
struct nfs4_cb_conn *conn = &clp->cl_cb_conn;
+ struct sockaddr *sa = svc_addr(rqstp);
+ u32 scopeid = rpc_get_scope_id(sa);
unsigned short expected_family;
/* Currently, we only support tcp and tcp6 for the callback channel */
@@ -1205,6 +1227,7 @@
conn->cb_prog = se->se_callback_prog;
conn->cb_ident = se->se_callback_ident;
+ rpc_svcaddr2sockaddr((struct sockaddr *)&conn->cb_saddr, expected_family, &rqstp->rq_daddr);
return;
out_err:
conn->cb_addr.ss_family = AF_UNSPEC;
@@ -1344,7 +1367,7 @@
case SP4_NONE:
break;
case SP4_SSV:
- return nfserr_encr_alg_unsupp;
+ return nfserr_serverfault;
default:
BUG(); /* checked by xdr code */
case SP4_MACH_CRED:
@@ -1361,8 +1384,12 @@
nfs4_lock_state();
status = nfs_ok;
- conf = find_confirmed_client_by_str(dname, strhashval, true);
+ conf = find_confirmed_client_by_str(dname, strhashval);
if (conf) {
+ if (!clp_used_exchangeid(conf)) {
+ status = nfserr_clid_inuse; /* XXX: ? */
+ goto out;
+ }
if (!same_verf(&verf, &conf->cl_verifier)) {
/* 18.35.4 case 8 */
if (exid->flags & EXCHGID4_FLAG_UPD_CONFIRMED_REC_A) {
@@ -1403,7 +1430,7 @@
goto out;
}
- unconf = find_unconfirmed_client_by_str(dname, strhashval, true);
+ unconf = find_unconfirmed_client_by_str(dname, strhashval);
if (unconf) {
/*
* Possible retry or client restart. Per 18.35.4 case 4,
@@ -1560,6 +1587,8 @@
status = nfs_ok;
memcpy(cr_ses->sessionid.data, new->se_sessionid.data,
NFS4_MAX_SESSIONID_LEN);
+ memcpy(&cr_ses->fore_channel, &new->se_fchannel,
+ sizeof(struct nfsd4_channel_attrs));
cs_slot->sl_seqid++;
cr_ses->seqid = cs_slot->sl_seqid;
@@ -1581,6 +1610,45 @@
return argp->opcnt == resp->opcnt;
}
+static __be32 nfsd4_map_bcts_dir(u32 *dir)
+{
+ switch (*dir) {
+ case NFS4_CDFC4_FORE:
+ case NFS4_CDFC4_BACK:
+ return nfs_ok;
+ case NFS4_CDFC4_FORE_OR_BOTH:
+ case NFS4_CDFC4_BACK_OR_BOTH:
+ *dir = NFS4_CDFC4_BOTH;
+ return nfs_ok;
+ };
+ return nfserr_inval;
+}
+
+__be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp,
+ struct nfsd4_compound_state *cstate,
+ struct nfsd4_bind_conn_to_session *bcts)
+{
+ __be32 status;
+
+ if (!nfsd4_last_compound_op(rqstp))
+ return nfserr_not_only_op;
+ spin_lock(&client_lock);
+ cstate->session = find_in_sessionid_hashtbl(&bcts->sessionid);
+ /* Sorta weird: we only need the refcnt'ing because new_conn acquires
+ * client_lock iself: */
+ if (cstate->session) {
+ nfsd4_get_session(cstate->session);
+ atomic_inc(&cstate->session->se_client->cl_refcount);
+ }
+ spin_unlock(&client_lock);
+ if (!cstate->session)
+ return nfserr_badsession;
+
+ status = nfsd4_map_bcts_dir(&bcts->dir);
+ nfsd4_new_conn(rqstp, cstate->session, bcts->dir);
+ return nfs_ok;
+}
+
static bool nfsd4_compound_in_session(struct nfsd4_session *session, struct nfs4_sessionid *sid)
{
if (!session)
@@ -1619,8 +1687,7 @@
spin_unlock(&client_lock);
nfs4_lock_state();
- /* wait for callbacks */
- nfsd4_shutdown_callback(ses->se_client);
+ nfsd4_probe_callback_sync(ses->se_client);
nfs4_unlock_state();
nfsd4_del_conns(ses);
@@ -1733,8 +1800,12 @@
out:
/* Hold a session reference until done processing the compound. */
if (cstate->session) {
+ struct nfs4_client *clp = session->se_client;
+
nfsd4_get_session(cstate->session);
- atomic_inc(&session->se_client->cl_refcount);
+ atomic_inc(&clp->cl_refcount);
+ if (clp->cl_cb_state == NFSD4_CB_DOWN)
+ seq->status_flags |= SEQ4_STATUS_CB_PATH_DOWN;
}
kfree(conn);
spin_unlock(&client_lock);
@@ -1775,7 +1846,6 @@
nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
struct nfsd4_setclientid *setclid)
{
- struct sockaddr *sa = svc_addr(rqstp);
struct xdr_netobj clname = {
.len = setclid->se_namelen,
.data = setclid->se_name,
@@ -1801,10 +1871,12 @@
strhashval = clientstr_hashval(dname);
nfs4_lock_state();
- conf = find_confirmed_client_by_str(dname, strhashval, false);
+ conf = find_confirmed_client_by_str(dname, strhashval);
if (conf) {
/* RFC 3530 14.2.33 CASE 0: */
status = nfserr_clid_inuse;
+ if (clp_used_exchangeid(conf))
+ goto out;
if (!same_creds(&conf->cl_cred, &rqstp->rq_cred)) {
char addr_str[INET6_ADDRSTRLEN];
rpc_ntop((struct sockaddr *) &conf->cl_addr, addr_str,
@@ -1819,7 +1891,7 @@
* has a description of SETCLIENTID request processing consisting
* of 5 bullet points, labeled as CASE0 - CASE4 below.
*/
- unconf = find_unconfirmed_client_by_str(dname, strhashval, false);
+ unconf = find_unconfirmed_client_by_str(dname, strhashval);
status = nfserr_resource;
if (!conf) {
/*
@@ -1876,7 +1948,7 @@
* for consistent minorversion use throughout:
*/
new->cl_minorversion = 0;
- gen_callback(new, setclid, rpc_get_scope_id(sa));
+ gen_callback(new, setclid, rqstp);
add_to_unconfirmed(new, strhashval);
setclid->se_clientid.cl_boot = new->cl_clientid.cl_boot;
setclid->se_clientid.cl_id = new->cl_clientid.cl_id;
@@ -1935,7 +2007,6 @@
if (!same_creds(&conf->cl_cred, &unconf->cl_cred))
status = nfserr_clid_inuse;
else {
- atomic_set(&conf->cl_cb_set, 0);
nfsd4_change_callback(conf, &unconf->cl_cb_conn);
nfsd4_probe_callback(conf);
expire_client(unconf);
@@ -1964,7 +2035,7 @@
unsigned int hash =
clientstr_hashval(unconf->cl_recdir);
conf = find_confirmed_client_by_str(unconf->cl_recdir,
- hash, false);
+ hash);
if (conf) {
nfsd4_remove_clid_dir(conf);
expire_client(conf);
@@ -2300,41 +2371,6 @@
nfsd4_cb_recall(dp);
}
-/*
- * The file_lock is being reapd.
- *
- * Called by locks_free_lock() with lock_flocks() held.
- */
-static
-void nfsd_release_deleg_cb(struct file_lock *fl)
-{
- struct nfs4_delegation *dp = (struct nfs4_delegation *)fl->fl_owner;
-
- dprintk("NFSD nfsd_release_deleg_cb: fl %p dp %p dl_count %d\n", fl,dp, atomic_read(&dp->dl_count));
-
- if (!(fl->fl_flags & FL_LEASE) || !dp)
- return;
- dp->dl_flock = NULL;
-}
-
-/*
- * Called from setlease() with lock_flocks() held
- */
-static
-int nfsd_same_client_deleg_cb(struct file_lock *onlist, struct file_lock *try)
-{
- struct nfs4_delegation *onlistd =
- (struct nfs4_delegation *)onlist->fl_owner;
- struct nfs4_delegation *tryd =
- (struct nfs4_delegation *)try->fl_owner;
-
- if (onlist->fl_lmops != try->fl_lmops)
- return 0;
-
- return onlistd->dl_client == tryd->dl_client;
-}
-
-
static
int nfsd_change_deleg_cb(struct file_lock **onlist, int arg)
{
@@ -2346,8 +2382,6 @@
static const struct lock_manager_operations nfsd_lease_mng_ops = {
.fl_break = nfsd_break_deleg_cb,
- .fl_release_private = nfsd_release_deleg_cb,
- .fl_mylease = nfsd_same_client_deleg_cb,
.fl_change = nfsd_change_deleg_cb,
};
@@ -2514,8 +2548,6 @@
if (!fp->fi_fds[oflag]) {
status = nfsd_open(rqstp, cur_fh, S_IFREG, access,
&fp->fi_fds[oflag]);
- if (status == nfserr_dropit)
- status = nfserr_jukebox;
if (status)
return status;
}
@@ -2596,6 +2628,19 @@
open->op_stateowner->so_client->cl_firststate = 1;
}
+/* Should we give out recallable state?: */
+static bool nfsd4_cb_channel_good(struct nfs4_client *clp)
+{
+ if (clp->cl_cb_state == NFSD4_CB_UP)
+ return true;
+ /*
+ * In the sessions case, since we don't have to establish a
+ * separate connection for callbacks, we assume it's OK
+ * until we hear otherwise:
+ */
+ return clp->cl_minorversion && clp->cl_cb_state == NFSD4_CB_UNKNOWN;
+}
+
/*
* Attempt to hand out a delegation.
*/
@@ -2604,10 +2649,11 @@
{
struct nfs4_delegation *dp;
struct nfs4_stateowner *sop = stp->st_stateowner;
- int cb_up = atomic_read(&sop->so_client->cl_cb_set);
+ int cb_up;
struct file_lock *fl;
int status, flag = 0;
+ cb_up = nfsd4_cb_channel_good(sop->so_client);
flag = NFS4_OPEN_DELEGATE_NONE;
open->op_recall = 0;
switch (open->op_claim_type) {
@@ -2655,7 +2701,7 @@
dp->dl_flock = fl;
/* vfs_setlease checks to see if delegation should be handed out.
- * the lock_manager callbacks fl_mylease and fl_change are used
+ * the lock_manager callback fl_change is used
*/
if ((status = vfs_setlease(fl->fl_file, fl->fl_type, &fl))) {
dprintk("NFSD: setlease failed [%d], no delegation\n", status);
@@ -2794,7 +2840,7 @@
renew_client(clp);
status = nfserr_cb_path_down;
if (!list_empty(&clp->cl_delegations)
- && !atomic_read(&clp->cl_cb_set))
+ && clp->cl_cb_state != NFSD4_CB_UP)
goto out;
status = nfs_ok;
out:
@@ -3081,9 +3127,10 @@
if (status)
goto out;
renew_client(dp->dl_client);
- if (filpp)
+ if (filpp) {
*filpp = find_readable_file(dp->dl_file);
- BUG_ON(!*filpp);
+ BUG_ON(!*filpp);
+ }
} else { /* open or lock stateid */
stp = find_stateid(stateid, flags);
if (!stp)
@@ -4107,7 +4154,7 @@
unsigned int strhashval = clientstr_hashval(name);
struct nfs4_client *clp;
- clp = find_confirmed_client_by_str(name, strhashval, use_exchange_id);
+ clp = find_confirmed_client_by_str(name, strhashval);
return clp ? 1 : 0;
}
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index f35a94a..956629b 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -44,13 +44,14 @@
#include <linux/namei.h>
#include <linux/statfs.h>
#include <linux/utsname.h>
-#include <linux/nfsd_idmap.h>
-#include <linux/nfs4_acl.h>
#include <linux/sunrpc/svcauth_gss.h>
+#include "idmap.h"
+#include "acl.h"
#include "xdr4.h"
#include "vfs.h"
+
#define NFSDDBG_FACILITY NFSDDBG_XDR
/*
@@ -288,17 +289,17 @@
len += XDR_QUADLEN(dummy32) << 2;
READMEM(buf, dummy32);
ace->whotype = nfs4_acl_get_whotype(buf, dummy32);
- host_err = 0;
+ status = nfs_ok;
if (ace->whotype != NFS4_ACL_WHO_NAMED)
ace->who = 0;
else if (ace->flag & NFS4_ACE_IDENTIFIER_GROUP)
- host_err = nfsd_map_name_to_gid(argp->rqstp,
+ status = nfsd_map_name_to_gid(argp->rqstp,
buf, dummy32, &ace->who);
else
- host_err = nfsd_map_name_to_uid(argp->rqstp,
+ status = nfsd_map_name_to_uid(argp->rqstp,
buf, dummy32, &ace->who);
- if (host_err)
- goto out_nfserr;
+ if (status)
+ return status;
}
} else
*acl = NULL;
@@ -420,6 +421,21 @@
DECODE_TAIL;
}
+static __be32 nfsd4_decode_bind_conn_to_session(struct nfsd4_compoundargs *argp, struct nfsd4_bind_conn_to_session *bcts)
+{
+ DECODE_HEAD;
+ u32 dummy;
+
+ READ_BUF(NFS4_MAX_SESSIONID_LEN + 8);
+ COPYMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN);
+ READ32(bcts->dir);
+ /* XXX: Perhaps Tom Tucker could help us figure out how we
+ * should be using ctsa_use_conn_in_rdma_mode: */
+ READ32(dummy);
+
+ DECODE_TAIL;
+}
+
static __be32
nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close)
{
@@ -847,6 +863,17 @@
}
static __be32
+nfsd4_decode_secinfo_no_name(struct nfsd4_compoundargs *argp,
+ struct nfsd4_secinfo_no_name *sin)
+{
+ DECODE_HEAD;
+
+ READ_BUF(4);
+ READ32(sin->sin_style);
+ DECODE_TAIL;
+}
+
+static __be32
nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *setattr)
{
__be32 status;
@@ -1005,7 +1032,7 @@
nfsd4_decode_exchange_id(struct nfsd4_compoundargs *argp,
struct nfsd4_exchange_id *exid)
{
- int dummy;
+ int dummy, tmp;
DECODE_HEAD;
READ_BUF(NFS4_VERIFIER_SIZE);
@@ -1053,15 +1080,23 @@
/* ssp_hash_algs<> */
READ_BUF(4);
- READ32(dummy);
- READ_BUF(dummy);
- p += XDR_QUADLEN(dummy);
+ READ32(tmp);
+ while (tmp--) {
+ READ_BUF(4);
+ READ32(dummy);
+ READ_BUF(dummy);
+ p += XDR_QUADLEN(dummy);
+ }
/* ssp_encr_algs<> */
READ_BUF(4);
- READ32(dummy);
- READ_BUF(dummy);
- p += XDR_QUADLEN(dummy);
+ READ32(tmp);
+ while (tmp--) {
+ READ_BUF(4);
+ READ32(dummy);
+ READ_BUF(dummy);
+ p += XDR_QUADLEN(dummy);
+ }
/* ssp_window and ssp_num_gss_handles */
READ_BUF(8);
@@ -1339,7 +1374,7 @@
/* new operations for NFSv4.1 */
[OP_BACKCHANNEL_CTL] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_BIND_CONN_TO_SESSION]= (nfsd4_dec)nfsd4_decode_notsupp,
+ [OP_BIND_CONN_TO_SESSION]= (nfsd4_dec)nfsd4_decode_bind_conn_to_session,
[OP_EXCHANGE_ID] = (nfsd4_dec)nfsd4_decode_exchange_id,
[OP_CREATE_SESSION] = (nfsd4_dec)nfsd4_decode_create_session,
[OP_DESTROY_SESSION] = (nfsd4_dec)nfsd4_decode_destroy_session,
@@ -1350,7 +1385,7 @@
[OP_LAYOUTCOMMIT] = (nfsd4_dec)nfsd4_decode_notsupp,
[OP_LAYOUTGET] = (nfsd4_dec)nfsd4_decode_notsupp,
[OP_LAYOUTRETURN] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_SECINFO_NO_NAME] = (nfsd4_dec)nfsd4_decode_notsupp,
+ [OP_SECINFO_NO_NAME] = (nfsd4_dec)nfsd4_decode_secinfo_no_name,
[OP_SEQUENCE] = (nfsd4_dec)nfsd4_decode_sequence,
[OP_SET_SSV] = (nfsd4_dec)nfsd4_decode_notsupp,
[OP_TEST_STATEID] = (nfsd4_dec)nfsd4_decode_notsupp,
@@ -2309,8 +2344,6 @@
case nfserr_resource:
nfserr = nfserr_toosmall;
goto fail;
- case nfserr_dropit:
- goto fail;
case nfserr_noent:
goto skip_entry;
default:
@@ -2365,6 +2398,21 @@
return nfserr;
}
+static __be32 nfsd4_encode_bind_conn_to_session(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_bind_conn_to_session *bcts)
+{
+ __be32 *p;
+
+ if (!nfserr) {
+ RESERVE_SPACE(NFS4_MAX_SESSIONID_LEN + 8);
+ WRITEMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN);
+ WRITE32(bcts->dir);
+ /* XXX: ? */
+ WRITE32(0);
+ ADJUST_ARGS();
+ }
+ return nfserr;
+}
+
static __be32
nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_close *close)
{
@@ -2826,11 +2874,10 @@
}
static __be32
-nfsd4_encode_secinfo(struct nfsd4_compoundres *resp, __be32 nfserr,
- struct nfsd4_secinfo *secinfo)
+nfsd4_do_encode_secinfo(struct nfsd4_compoundres *resp,
+ __be32 nfserr,struct svc_export *exp)
{
int i = 0;
- struct svc_export *exp = secinfo->si_exp;
u32 nflavs;
struct exp_flavor_info *flavs;
struct exp_flavor_info def_flavs[2];
@@ -2892,6 +2939,20 @@
return nfserr;
}
+static __be32
+nfsd4_encode_secinfo(struct nfsd4_compoundres *resp, __be32 nfserr,
+ struct nfsd4_secinfo *secinfo)
+{
+ return nfsd4_do_encode_secinfo(resp, nfserr, secinfo->si_exp);
+}
+
+static __be32
+nfsd4_encode_secinfo_no_name(struct nfsd4_compoundres *resp, __be32 nfserr,
+ struct nfsd4_secinfo_no_name *secinfo)
+{
+ return nfsd4_do_encode_secinfo(resp, nfserr, secinfo->sin_exp);
+}
+
/*
* The SETATTR encode routine is special -- it always encodes a bitmap,
* regardless of the error status.
@@ -3076,13 +3137,9 @@
WRITE32(seq->seqid);
WRITE32(seq->slotid);
WRITE32(seq->maxslots);
- /*
- * FIXME: for now:
- * target_maxslots = maxslots
- * status_flags = 0
- */
+ /* For now: target_maxslots = maxslots */
WRITE32(seq->maxslots);
- WRITE32(0);
+ WRITE32(seq->status_flags);
ADJUST_ARGS();
resp->cstate.datap = p; /* DRC cache data pointer */
@@ -3143,7 +3200,7 @@
/* NFSv4.1 operations */
[OP_BACKCHANNEL_CTL] = (nfsd4_enc)nfsd4_encode_noop,
- [OP_BIND_CONN_TO_SESSION] = (nfsd4_enc)nfsd4_encode_noop,
+ [OP_BIND_CONN_TO_SESSION] = (nfsd4_enc)nfsd4_encode_bind_conn_to_session,
[OP_EXCHANGE_ID] = (nfsd4_enc)nfsd4_encode_exchange_id,
[OP_CREATE_SESSION] = (nfsd4_enc)nfsd4_encode_create_session,
[OP_DESTROY_SESSION] = (nfsd4_enc)nfsd4_encode_destroy_session,
@@ -3154,7 +3211,7 @@
[OP_LAYOUTCOMMIT] = (nfsd4_enc)nfsd4_encode_noop,
[OP_LAYOUTGET] = (nfsd4_enc)nfsd4_encode_noop,
[OP_LAYOUTRETURN] = (nfsd4_enc)nfsd4_encode_noop,
- [OP_SECINFO_NO_NAME] = (nfsd4_enc)nfsd4_encode_noop,
+ [OP_SECINFO_NO_NAME] = (nfsd4_enc)nfsd4_encode_secinfo_no_name,
[OP_SEQUENCE] = (nfsd4_enc)nfsd4_encode_sequence,
[OP_SET_SSV] = (nfsd4_enc)nfsd4_encode_noop,
[OP_TEST_STATEID] = (nfsd4_enc)nfsd4_encode_noop,
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 4514ebb..33b3e2b 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -8,12 +8,12 @@
#include <linux/namei.h>
#include <linux/ctype.h>
-#include <linux/nfsd_idmap.h>
#include <linux/sunrpc/svcsock.h>
#include <linux/nfsd/syscall.h>
#include <linux/lockd/lockd.h>
#include <linux/sunrpc/clnt.h>
+#include "idmap.h"
#include "nfsd.h"
#include "cache.h"
@@ -127,6 +127,7 @@
static ssize_t nfsctl_transaction_read(struct file *file, char __user *buf, size_t size, loff_t *pos)
{
+#ifdef CONFIG_NFSD_DEPRECATED
static int warned;
if (file->f_dentry->d_name.name[0] == '.' && !warned) {
printk(KERN_INFO
@@ -135,6 +136,7 @@
current->comm, file->f_dentry->d_name.name);
warned = 1;
}
+#endif
if (! file->private_data) {
/* An attempt to read a transaction file without writing
* causes a 0-byte write so that the file can return
diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h
index 6b641cf..7ecfa24 100644
--- a/fs/nfsd/nfsd.h
+++ b/fs/nfsd/nfsd.h
@@ -158,6 +158,7 @@
#define nfserr_attrnotsupp cpu_to_be32(NFSERR_ATTRNOTSUPP)
#define nfserr_bad_xdr cpu_to_be32(NFSERR_BAD_XDR)
#define nfserr_openmode cpu_to_be32(NFSERR_OPENMODE)
+#define nfserr_badowner cpu_to_be32(NFSERR_BADOWNER)
#define nfserr_locks_held cpu_to_be32(NFSERR_LOCKS_HELD)
#define nfserr_op_illegal cpu_to_be32(NFSERR_OP_ILLEGAL)
#define nfserr_grace cpu_to_be32(NFSERR_GRACE)
diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c
index 08e1726..e15dc45 100644
--- a/fs/nfsd/nfsproc.c
+++ b/fs/nfsd/nfsproc.c
@@ -735,9 +735,9 @@
{ nfserr_stale, -ESTALE },
{ nfserr_jukebox, -ETIMEDOUT },
{ nfserr_jukebox, -ERESTARTSYS },
- { nfserr_dropit, -EAGAIN },
- { nfserr_dropit, -ENOMEM },
- { nfserr_badname, -ESRCH },
+ { nfserr_jukebox, -EAGAIN },
+ { nfserr_jukebox, -EWOULDBLOCK },
+ { nfserr_jukebox, -ENOMEM },
{ nfserr_io, -ETXTBSY },
{ nfserr_notsupp, -EOPNOTSUPP },
{ nfserr_toosmall, -ETOOSMALL },
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 2bae1d8..18743c4 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -608,7 +608,7 @@
/* Now call the procedure handler, and encode NFS status. */
nfserr = proc->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp);
nfserr = map_new_errors(rqstp->rq_vers, nfserr);
- if (nfserr == nfserr_dropit) {
+ if (nfserr == nfserr_dropit || rqstp->rq_dropme) {
dprintk("nfsd: Dropping request; may be revisited later\n");
nfsd_cache_update(rqstp, RC_NOCACHE, NULL);
return 0;
diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h
index 39adc27..3074656 100644
--- a/fs/nfsd/state.h
+++ b/fs/nfsd/state.h
@@ -68,10 +68,12 @@
struct nfsd4_callback {
void *cb_op;
struct nfs4_client *cb_clp;
+ struct list_head cb_per_client;
u32 cb_minorversion;
struct rpc_message cb_msg;
const struct rpc_call_ops *cb_ops;
struct work_struct cb_work;
+ bool cb_done;
};
struct nfs4_delegation {
@@ -81,6 +83,7 @@
atomic_t dl_count; /* ref count */
struct nfs4_client *dl_client;
struct nfs4_file *dl_file;
+ struct file *dl_vfs_file;
struct file_lock *dl_flock;
u32 dl_type;
time_t dl_time;
@@ -95,6 +98,7 @@
struct nfs4_cb_conn {
/* SETCLIENTID info */
struct sockaddr_storage cb_addr;
+ struct sockaddr_storage cb_saddr;
size_t cb_addrlen;
u32 cb_prog; /* used only in 4.0 case;
per-session otherwise */
@@ -146,6 +150,11 @@
u32 gid;
};
+struct nfsd4_bind_conn_to_session {
+ struct nfs4_sessionid sessionid;
+ u32 dir;
+};
+
/* The single slot clientid cache structure */
struct nfsd4_clid_slot {
u32 sl_seqid;
@@ -235,9 +244,13 @@
unsigned long cl_cb_flags;
struct rpc_clnt *cl_cb_client;
u32 cl_cb_ident;
- atomic_t cl_cb_set;
+#define NFSD4_CB_UP 0
+#define NFSD4_CB_UNKNOWN 1
+#define NFSD4_CB_DOWN 2
+ int cl_cb_state;
struct nfsd4_callback cl_cb_null;
struct nfsd4_session *cl_cb_session;
+ struct list_head cl_callbacks; /* list of in-progress callbacks */
/* for all client information that callback code might need: */
spinlock_t cl_lock;
@@ -454,6 +467,7 @@
extern void nfs4_free_stateowner(struct kref *kref);
extern int set_callback_cred(void);
extern void nfsd4_probe_callback(struct nfs4_client *clp);
+extern void nfsd4_probe_callback_sync(struct nfs4_client *clp);
extern void nfsd4_change_callback(struct nfs4_client *clp, struct nfs4_cb_conn *);
extern void nfsd4_do_callback_rpc(struct work_struct *);
extern void nfsd4_cb_recall(struct nfs4_delegation *dp);
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 230b79f..641117f 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1,4 +1,3 @@
-#define MSNFS /* HACK HACK */
/*
* File operations used by nfsd. Some of these have been ripped from
* other parts of the kernel because they weren't exported, others
@@ -35,8 +34,8 @@
#endif /* CONFIG_NFSD_V3 */
#ifdef CONFIG_NFSD_V4
-#include <linux/nfs4_acl.h>
-#include <linux/nfsd_idmap.h>
+#include "acl.h"
+#include "idmap.h"
#endif /* CONFIG_NFSD_V4 */
#include "nfsd.h"
@@ -88,8 +87,9 @@
.dentry = dget(dentry)};
int err = 0;
- while (d_mountpoint(path.dentry) && follow_down(&path))
- ;
+ err = follow_down(&path, false);
+ if (err < 0)
+ goto out;
exp2 = rqst_exp_get_by_name(rqstp, &path);
if (IS_ERR(exp2)) {
@@ -273,6 +273,13 @@
return err;
}
+static int nfsd_break_lease(struct inode *inode)
+{
+ if (!S_ISREG(inode->i_mode))
+ return 0;
+ return break_lease(inode, O_WRONLY | O_NONBLOCK);
+}
+
/*
* Commit metadata changes to stable storage.
*/
@@ -375,16 +382,6 @@
goto out;
}
- /*
- * If we are changing the size of the file, then
- * we need to break all leases.
- */
- host_err = break_lease(inode, O_WRONLY | O_NONBLOCK);
- if (host_err == -EWOULDBLOCK)
- host_err = -ETIMEDOUT;
- if (host_err) /* ENOMEM or EWOULDBLOCK */
- goto out_nfserr;
-
host_err = get_write_access(inode);
if (host_err)
goto out_nfserr;
@@ -425,7 +422,11 @@
err = nfserr_notsync;
if (!check_guard || guardtime == inode->i_ctime.tv_sec) {
+ host_err = nfsd_break_lease(inode);
+ if (host_err)
+ goto out_nfserr;
fh_lock(fhp);
+
host_err = notify_change(dentry, iap);
err = nfserrno(host_err);
fh_unlock(fhp);
@@ -752,8 +753,6 @@
*/
if (!(access & NFSD_MAY_NOT_BREAK_LEASE))
host_err = break_lease(inode, O_NONBLOCK | ((access & NFSD_MAY_WRITE) ? O_WRONLY : 0));
- if (host_err == -EWOULDBLOCK)
- host_err = -ETIMEDOUT;
if (host_err) /* NOMEM or WOULDBLOCK */
goto out_nfserr;
@@ -874,15 +873,6 @@
return __splice_from_pipe(pipe, sd, nfsd_splice_actor);
}
-static inline int svc_msnfs(struct svc_fh *ffhp)
-{
-#ifdef MSNFS
- return (ffhp->fh_export->ex_flags & NFSEXP_MSNFS);
-#else
- return 0;
-#endif
-}
-
static __be32
nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
loff_t offset, struct kvec *vec, int vlen, unsigned long *count)
@@ -895,9 +885,6 @@
err = nfserr_perm;
inode = file->f_path.dentry->d_inode;
- if (svc_msnfs(fhp) && !lock_may_read(inode, offset, *count))
- goto out;
-
if (file->f_op->splice_read && rqstp->rq_splice_ok) {
struct splice_desc sd = {
.len = 0,
@@ -922,7 +909,6 @@
fsnotify_access(file);
} else
err = nfserrno(host_err);
-out:
return err;
}
@@ -987,14 +973,6 @@
int stable = *stablep;
int use_wgather;
-#ifdef MSNFS
- err = nfserr_perm;
-
- if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) &&
- (!lock_may_write(file->f_path.dentry->d_inode, offset, *cnt)))
- goto out;
-#endif
-
dentry = file->f_path.dentry;
inode = dentry->d_inode;
exp = fhp->fh_export;
@@ -1045,7 +1023,6 @@
err = 0;
else
err = nfserrno(host_err);
-out:
return err;
}
@@ -1665,6 +1642,12 @@
err = nfserrno(host_err);
goto out_dput;
}
+ err = nfserr_noent;
+ if (!dold->d_inode)
+ goto out_drop_write;
+ host_err = nfsd_break_lease(dold->d_inode);
+ if (host_err)
+ goto out_drop_write;
host_err = vfs_link(dold, dirp, dnew);
if (!host_err) {
err = nfserrno(commit_metadata(ffhp));
@@ -1676,6 +1659,7 @@
else
err = nfserrno(host_err);
}
+out_drop_write:
mnt_drop_write(tfhp->fh_export->ex_path.mnt);
out_dput:
dput(dnew);
@@ -1750,12 +1734,6 @@
if (ndentry == trap)
goto out_dput_new;
- if (svc_msnfs(ffhp) &&
- ((odentry->d_count > 1) || (ndentry->d_count > 1))) {
- host_err = -EPERM;
- goto out_dput_new;
- }
-
host_err = -EXDEV;
if (ffhp->fh_export->ex_path.mnt != tfhp->fh_export->ex_path.mnt)
goto out_dput_new;
@@ -1763,15 +1741,17 @@
if (host_err)
goto out_dput_new;
+ host_err = nfsd_break_lease(odentry->d_inode);
+ if (host_err)
+ goto out_drop_write;
host_err = vfs_rename(fdir, odentry, tdir, ndentry);
if (!host_err) {
host_err = commit_metadata(tfhp);
if (!host_err)
host_err = commit_metadata(ffhp);
}
-
+out_drop_write:
mnt_drop_write(ffhp->fh_export->ex_path.mnt);
-
out_dput_new:
dput(ndentry);
out_dput_old:
@@ -1834,18 +1814,14 @@
if (host_err)
goto out_nfserr;
- if (type != S_IFDIR) { /* It's UNLINK */
-#ifdef MSNFS
- if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) &&
- (rdentry->d_count > 1)) {
- host_err = -EPERM;
- } else
-#endif
+ host_err = nfsd_break_lease(rdentry->d_inode);
+ if (host_err)
+ goto out_put;
+ if (type != S_IFDIR)
host_err = vfs_unlink(dirp, rdentry);
- } else { /* It's RMDIR */
+ else
host_err = vfs_rmdir(dirp, rdentry);
- }
-
+out_put:
dput(rdentry);
if (!host_err)
diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h
index 60fce3d..366401e 100644
--- a/fs/nfsd/xdr4.h
+++ b/fs/nfsd/xdr4.h
@@ -311,6 +311,11 @@
struct svc_export *si_exp; /* response */
};
+struct nfsd4_secinfo_no_name {
+ u32 sin_style; /* request */
+ struct svc_export *sin_exp; /* response */
+};
+
struct nfsd4_setattr {
stateid_t sa_stateid; /* request */
u32 sa_bmval[3]; /* request */
@@ -373,8 +378,8 @@
u32 cachethis; /* request */
#if 0
u32 target_maxslots; /* response */
- u32 status_flags; /* response */
#endif /* not yet */
+ u32 status_flags; /* response */
};
struct nfsd4_destroy_session {
@@ -422,6 +427,7 @@
/* NFSv4.1 */
struct nfsd4_exchange_id exchange_id;
+ struct nfsd4_bind_conn_to_session bind_conn_to_session;
struct nfsd4_create_session create_session;
struct nfsd4_destroy_session destroy_session;
struct nfsd4_sequence sequence;
@@ -518,6 +524,7 @@
struct nfsd4_sequence *seq);
extern __be32 nfsd4_exchange_id(struct svc_rqst *rqstp,
struct nfsd4_compound_state *, struct nfsd4_exchange_id *);
+extern __be32 nfsd4_bind_conn_to_session(struct svc_rqst *, struct nfsd4_compound_state *, struct nfsd4_bind_conn_to_session *);
extern __be32 nfsd4_create_session(struct svc_rqst *,
struct nfsd4_compound_state *,
struct nfsd4_create_session *);
diff --git a/fs/pipe.c b/fs/pipe.c
index e2e95fb..89e9e19 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -1292,7 +1292,7 @@
static void __exit exit_pipe_fs(void)
{
unregister_filesystem(&pipe_fs_type);
- mntput_long(pipe_mnt);
+ mntput(pipe_mnt);
}
fs_initcall(init_pipe_fs);
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 93f1cdd..9d096e8 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1151,7 +1151,7 @@
goto err_task_lock;
}
- if (oom_score_adj < task->signal->oom_score_adj &&
+ if (oom_score_adj < task->signal->oom_score_adj_min &&
!capable(CAP_SYS_RESOURCE)) {
err = -EACCES;
goto err_sighand;
@@ -1164,6 +1164,8 @@
atomic_dec(&task->mm->oom_disable_count);
}
task->signal->oom_score_adj = oom_score_adj;
+ if (has_capability_noaudit(current, CAP_SYS_RESOURCE))
+ task->signal->oom_score_adj_min = oom_score_adj;
/*
* Scale /proc/pid/oom_adj appropriately ensuring that OOM_DISABLE is
* always attainable.
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index a65239c..ed257d1 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -101,6 +101,9 @@
#ifdef CONFIG_MEMORY_FAILURE
"HardwareCorrupted: %5lu kB\n"
#endif
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ "AnonHugePages: %8lu kB\n"
+#endif
,
K(i.totalram),
K(i.freeram),
@@ -128,7 +131,12 @@
K(i.freeswap),
K(global_page_state(NR_FILE_DIRTY)),
K(global_page_state(NR_WRITEBACK)),
- K(global_page_state(NR_ANON_PAGES)),
+ K(global_page_state(NR_ANON_PAGES)
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ + global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) *
+ HPAGE_PMD_NR
+#endif
+ ),
K(global_page_state(NR_FILE_MAPPED)),
K(global_page_state(NR_SHMEM)),
K(global_page_state(NR_SLAB_RECLAIMABLE) +
@@ -151,6 +159,10 @@
#ifdef CONFIG_MEMORY_FAILURE
,atomic_long_read(&mce_bad_pages) << (PAGE_SHIFT - 10)
#endif
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ ,K(global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) *
+ HPAGE_PMD_NR)
+#endif
);
hugetlb_report_meminfo(m);
diff --git a/fs/proc/page.c b/fs/proc/page.c
index b06c674..6d8e6a9 100644
--- a/fs/proc/page.c
+++ b/fs/proc/page.c
@@ -116,15 +116,17 @@
if (PageHuge(page))
u |= 1 << KPF_HUGE;
+ /*
+ * Caveats on high order pages: page->_count will only be set
+ * -1 on the head page; SLUB/SLQB do the same for PG_slab;
+ * SLOB won't set PG_slab at all on compound pages.
+ */
+ if (PageBuddy(page))
+ u |= 1 << KPF_BUDDY;
+
u |= kpf_copy_bit(k, KPF_LOCKED, PG_locked);
- /*
- * Caveats on high order pages:
- * PG_buddy will only be set on the head page; SLUB/SLQB do the same
- * for PG_slab; SLOB won't set PG_slab at all on compound pages.
- */
u |= kpf_copy_bit(k, KPF_SLAB, PG_slab);
- u |= kpf_copy_bit(k, KPF_BUDDY, PG_buddy);
u |= kpf_copy_bit(k, KPF_ERROR, PG_error);
u |= kpf_copy_bit(k, KPF_DIRTY, PG_dirty);
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index c3755bd..60b9148 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -418,7 +418,8 @@
"Anonymous: %8lu kB\n"
"Swap: %8lu kB\n"
"KernelPageSize: %8lu kB\n"
- "MMUPageSize: %8lu kB\n",
+ "MMUPageSize: %8lu kB\n"
+ "Locked: %8lu kB\n",
(vma->vm_end - vma->vm_start) >> 10,
mss.resident >> 10,
(unsigned long)(mss.pss >> (10 + PSS_SHIFT)),
@@ -430,7 +431,9 @@
mss.anonymous >> 10,
mss.swap >> 10,
vma_kernel_pagesize(vma) >> 10,
- vma_mmu_pagesize(vma) >> 10);
+ vma_mmu_pagesize(vma) >> 10,
+ (vma->vm_flags & VM_LOCKED) ?
+ (unsigned long)(mss.pss >> (10 + PSS_SHIFT)) : 0);
if (m->count < m->size) /* vma is copied successfully */
m->version = (vma != get_gate_vma(task)) ? vma->vm_start : 0;
diff --git a/fs/stat.c b/fs/stat.c
index 12e90e2..d5c61cf 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -75,11 +75,13 @@
int error = -EINVAL;
int lookup_flags = 0;
- if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
+ if ((flag & ~(AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT)) != 0)
goto out;
if (!(flag & AT_SYMLINK_NOFOLLOW))
lookup_flags |= LOOKUP_FOLLOW;
+ if (flag & AT_NO_AUTOMOUNT)
+ lookup_flags |= LOOKUP_NO_AUTOMOUNT;
error = user_path_at(dfd, filename, lookup_flags, &path);
if (error)
diff --git a/fs/super.c b/fs/super.c
index 4f6a357..74e149e 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -1141,7 +1141,7 @@
return mnt;
err:
- mntput_long(mnt);
+ mntput(mnt);
return ERR_PTR(err);
}
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 0dce969..faca449 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -98,6 +98,7 @@
kmem.o \
xfs_aops.o \
xfs_buf.o \
+ xfs_discard.o \
xfs_export.o \
xfs_file.o \
xfs_fs_subr.o \
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 92f1f2a..ac1c7e8 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -896,7 +896,6 @@
trace_xfs_buf_rele(bp, _RET_IP_);
if (!pag) {
- ASSERT(!bp->b_relse);
ASSERT(list_empty(&bp->b_lru));
ASSERT(RB_EMPTY_NODE(&bp->b_rbnode));
if (atomic_dec_and_test(&bp->b_hold))
@@ -908,11 +907,7 @@
ASSERT(atomic_read(&bp->b_hold) > 0);
if (atomic_dec_and_lock(&bp->b_hold, &pag->pag_buf_lock)) {
- if (bp->b_relse) {
- atomic_inc(&bp->b_hold);
- spin_unlock(&pag->pag_buf_lock);
- bp->b_relse(bp);
- } else if (!(bp->b_flags & XBF_STALE) &&
+ if (!(bp->b_flags & XBF_STALE) &&
atomic_read(&bp->b_lru_ref)) {
xfs_buf_lru_add(bp);
spin_unlock(&pag->pag_buf_lock);
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index a76c242..cbe6595 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -152,8 +152,6 @@
struct xfs_buf;
typedef void (*xfs_buf_iodone_t)(struct xfs_buf *);
-typedef void (*xfs_buf_relse_t)(struct xfs_buf *);
-typedef int (*xfs_buf_bdstrat_t)(struct xfs_buf *);
#define XB_PAGES 2
@@ -183,7 +181,6 @@
void *b_addr; /* virtual address of buffer */
struct work_struct b_iodone_work;
xfs_buf_iodone_t b_iodone; /* I/O completion function */
- xfs_buf_relse_t b_relse; /* releasing function */
struct completion b_iowait; /* queue for I/O waiters */
void *b_fspriv;
void *b_fspriv2;
@@ -323,7 +320,6 @@
#define XFS_BUF_FSPRIVATE2(bp, type) ((type)(bp)->b_fspriv2)
#define XFS_BUF_SET_FSPRIVATE2(bp, val) ((bp)->b_fspriv2 = (void*)(val))
#define XFS_BUF_SET_START(bp) do { } while (0)
-#define XFS_BUF_SET_BRELSE_FUNC(bp, func) ((bp)->b_relse = (func))
#define XFS_BUF_PTR(bp) (xfs_caddr_t)((bp)->b_addr)
#define XFS_BUF_SET_PTR(bp, val, cnt) xfs_buf_associate_memory(bp, val, cnt)
@@ -360,8 +356,7 @@
static inline void xfs_buf_relse(xfs_buf_t *bp)
{
- if (!bp->b_relse)
- xfs_buf_unlock(bp);
+ xfs_buf_unlock(bp);
xfs_buf_rele(bp);
}
diff --git a/fs/xfs/linux-2.6/xfs_discard.c b/fs/xfs/linux-2.6/xfs_discard.c
new file mode 100644
index 0000000..05201ae
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_discard.c
@@ -0,0 +1,191 @@
+/*
+ * Copyright (C) 2010 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it would 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 the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#include "xfs.h"
+#include "xfs_sb.h"
+#include "xfs_inum.h"
+#include "xfs_log.h"
+#include "xfs_ag.h"
+#include "xfs_mount.h"
+#include "xfs_quota.h"
+#include "xfs_trans.h"
+#include "xfs_alloc_btree.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_ialloc_btree.h"
+#include "xfs_btree.h"
+#include "xfs_inode.h"
+#include "xfs_alloc.h"
+#include "xfs_error.h"
+#include "xfs_discard.h"
+#include "xfs_trace.h"
+
+STATIC int
+xfs_trim_extents(
+ struct xfs_mount *mp,
+ xfs_agnumber_t agno,
+ xfs_fsblock_t start,
+ xfs_fsblock_t len,
+ xfs_fsblock_t minlen,
+ __uint64_t *blocks_trimmed)
+{
+ struct block_device *bdev = mp->m_ddev_targp->bt_bdev;
+ struct xfs_btree_cur *cur;
+ struct xfs_buf *agbp;
+ struct xfs_perag *pag;
+ int error;
+ int i;
+
+ pag = xfs_perag_get(mp, agno);
+
+ error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp);
+ if (error || !agbp)
+ goto out_put_perag;
+
+ cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
+
+ /*
+ * Force out the log. This means any transactions that might have freed
+ * space before we took the AGF buffer lock are now on disk, and the
+ * volatile disk cache is flushed.
+ */
+ xfs_log_force(mp, XFS_LOG_SYNC);
+
+ /*
+ * Look up the longest btree in the AGF and start with it.
+ */
+ error = xfs_alloc_lookup_le(cur, 0,
+ XFS_BUF_TO_AGF(agbp)->agf_longest, &i);
+ if (error)
+ goto out_del_cursor;
+
+ /*
+ * Loop until we are done with all extents that are large
+ * enough to be worth discarding.
+ */
+ while (i) {
+ xfs_agblock_t fbno;
+ xfs_extlen_t flen;
+
+ error = xfs_alloc_get_rec(cur, &fbno, &flen, &i);
+ if (error)
+ goto out_del_cursor;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor);
+ ASSERT(flen <= XFS_BUF_TO_AGF(agbp)->agf_longest);
+
+ /*
+ * Too small? Give up.
+ */
+ if (flen < minlen) {
+ trace_xfs_discard_toosmall(mp, agno, fbno, flen);
+ goto out_del_cursor;
+ }
+
+ /*
+ * If the extent is entirely outside of the range we are
+ * supposed to discard skip it. Do not bother to trim
+ * down partially overlapping ranges for now.
+ */
+ if (XFS_AGB_TO_FSB(mp, agno, fbno) + flen < start ||
+ XFS_AGB_TO_FSB(mp, agno, fbno) >= start + len) {
+ trace_xfs_discard_exclude(mp, agno, fbno, flen);
+ goto next_extent;
+ }
+
+ /*
+ * If any blocks in the range are still busy, skip the
+ * discard and try again the next time.
+ */
+ if (xfs_alloc_busy_search(mp, agno, fbno, flen)) {
+ trace_xfs_discard_busy(mp, agno, fbno, flen);
+ goto next_extent;
+ }
+
+ trace_xfs_discard_extent(mp, agno, fbno, flen);
+ error = -blkdev_issue_discard(bdev,
+ XFS_AGB_TO_DADDR(mp, agno, fbno),
+ XFS_FSB_TO_BB(mp, flen),
+ GFP_NOFS, 0);
+ if (error)
+ goto out_del_cursor;
+ *blocks_trimmed += flen;
+
+next_extent:
+ error = xfs_btree_decrement(cur, 0, &i);
+ if (error)
+ goto out_del_cursor;
+ }
+
+out_del_cursor:
+ xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
+ xfs_buf_relse(agbp);
+out_put_perag:
+ xfs_perag_put(pag);
+ return error;
+}
+
+int
+xfs_ioc_trim(
+ struct xfs_mount *mp,
+ struct fstrim_range __user *urange)
+{
+ struct request_queue *q = mp->m_ddev_targp->bt_bdev->bd_disk->queue;
+ unsigned int granularity = q->limits.discard_granularity;
+ struct fstrim_range range;
+ xfs_fsblock_t start, len, minlen;
+ xfs_agnumber_t start_agno, end_agno, agno;
+ __uint64_t blocks_trimmed = 0;
+ int error, last_error = 0;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -XFS_ERROR(EPERM);
+ if (copy_from_user(&range, urange, sizeof(range)))
+ return -XFS_ERROR(EFAULT);
+
+ /*
+ * Truncating down the len isn't actually quite correct, but using
+ * XFS_B_TO_FSB would mean we trivially get overflows for values
+ * of ULLONG_MAX or slightly lower. And ULLONG_MAX is the default
+ * used by the fstrim application. In the end it really doesn't
+ * matter as trimming blocks is an advisory interface.
+ */
+ start = XFS_B_TO_FSBT(mp, range.start);
+ len = XFS_B_TO_FSBT(mp, range.len);
+ minlen = XFS_B_TO_FSB(mp, max_t(u64, granularity, range.minlen));
+
+ start_agno = XFS_FSB_TO_AGNO(mp, start);
+ if (start_agno >= mp->m_sb.sb_agcount)
+ return -XFS_ERROR(EINVAL);
+
+ end_agno = XFS_FSB_TO_AGNO(mp, start + len);
+ if (end_agno >= mp->m_sb.sb_agcount)
+ end_agno = mp->m_sb.sb_agcount - 1;
+
+ for (agno = start_agno; agno <= end_agno; agno++) {
+ error = -xfs_trim_extents(mp, agno, start, len, minlen,
+ &blocks_trimmed);
+ if (error)
+ last_error = error;
+ }
+
+ if (last_error)
+ return last_error;
+
+ range.len = XFS_FSB_TO_B(mp, blocks_trimmed);
+ if (copy_to_user(urange, &range, sizeof(range)))
+ return -XFS_ERROR(EFAULT);
+ return 0;
+}
diff --git a/fs/xfs/linux-2.6/xfs_discard.h b/fs/xfs/linux-2.6/xfs_discard.h
new file mode 100644
index 0000000..e82b6dd
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_discard.h
@@ -0,0 +1,8 @@
+#ifndef XFS_DISCARD_H
+#define XFS_DISCARD_H 1
+
+struct fstrim_range;
+
+extern int xfs_ioc_trim(struct xfs_mount *, struct fstrim_range __user *);
+
+#endif /* XFS_DISCARD_H */
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index ba8ad42..ef51eb43e 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -41,6 +41,40 @@
static const struct vm_operations_struct xfs_file_vm_ops;
/*
+ * Locking primitives for read and write IO paths to ensure we consistently use
+ * and order the inode->i_mutex, ip->i_lock and ip->i_iolock.
+ */
+static inline void
+xfs_rw_ilock(
+ struct xfs_inode *ip,
+ int type)
+{
+ if (type & XFS_IOLOCK_EXCL)
+ mutex_lock(&VFS_I(ip)->i_mutex);
+ xfs_ilock(ip, type);
+}
+
+static inline void
+xfs_rw_iunlock(
+ struct xfs_inode *ip,
+ int type)
+{
+ xfs_iunlock(ip, type);
+ if (type & XFS_IOLOCK_EXCL)
+ mutex_unlock(&VFS_I(ip)->i_mutex);
+}
+
+static inline void
+xfs_rw_ilock_demote(
+ struct xfs_inode *ip,
+ int type)
+{
+ xfs_ilock_demote(ip, type);
+ if (type & XFS_IOLOCK_EXCL)
+ mutex_unlock(&VFS_I(ip)->i_mutex);
+}
+
+/*
* xfs_iozero
*
* xfs_iozero clears the specified range of buffer supplied,
@@ -262,22 +296,21 @@
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
- if (unlikely(ioflags & IO_ISDIRECT))
- mutex_lock(&inode->i_mutex);
- xfs_ilock(ip, XFS_IOLOCK_SHARED);
-
if (unlikely(ioflags & IO_ISDIRECT)) {
+ xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
+
if (inode->i_mapping->nrpages) {
ret = -xfs_flushinval_pages(ip,
(iocb->ki_pos & PAGE_CACHE_MASK),
-1, FI_REMAPF_LOCKED);
+ if (ret) {
+ xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
+ return ret;
+ }
}
- mutex_unlock(&inode->i_mutex);
- if (ret) {
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return ret;
- }
- }
+ xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
+ } else
+ xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags);
@@ -285,7 +318,7 @@
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
@@ -309,7 +342,7 @@
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
- xfs_ilock(ip, XFS_IOLOCK_SHARED);
+ xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
@@ -317,10 +350,61 @@
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
+STATIC void
+xfs_aio_write_isize_update(
+ struct inode *inode,
+ loff_t *ppos,
+ ssize_t bytes_written)
+{
+ struct xfs_inode *ip = XFS_I(inode);
+ xfs_fsize_t isize = i_size_read(inode);
+
+ if (bytes_written > 0)
+ XFS_STATS_ADD(xs_write_bytes, bytes_written);
+
+ if (unlikely(bytes_written < 0 && bytes_written != -EFAULT &&
+ *ppos > isize))
+ *ppos = isize;
+
+ if (*ppos > ip->i_size) {
+ xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
+ if (*ppos > ip->i_size)
+ ip->i_size = *ppos;
+ xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+}
+
+/*
+ * If this was a direct or synchronous I/O that failed (such as ENOSPC) then
+ * part of the I/O may have been written to disk before the error occured. In
+ * this case the on-disk file size may have been adjusted beyond the in-memory
+ * file size and now needs to be truncated back.
+ */
+STATIC void
+xfs_aio_write_newsize_update(
+ struct xfs_inode *ip)
+{
+ if (ip->i_new_size) {
+ xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
+ ip->i_new_size = 0;
+ if (ip->i_d.di_size > ip->i_size)
+ ip->i_d.di_size = ip->i_size;
+ xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+}
+
+/*
+ * xfs_file_splice_write() does not use xfs_rw_ilock() because
+ * generic_file_splice_write() takes the i_mutex itself. This, in theory,
+ * couuld cause lock inversions between the aio_write path and the splice path
+ * if someone is doing concurrent splice(2) based writes and write(2) based
+ * writes to the same inode. The only real way to fix this is to re-implement
+ * the generic code here with correct locking orders.
+ */
STATIC ssize_t
xfs_file_splice_write(
struct pipe_inode_info *pipe,
@@ -331,7 +415,7 @@
{
struct inode *inode = outfilp->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
- xfs_fsize_t isize, new_size;
+ xfs_fsize_t new_size;
int ioflags = 0;
ssize_t ret;
@@ -355,27 +439,9 @@
trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
- if (ret > 0)
- XFS_STATS_ADD(xs_write_bytes, ret);
- isize = i_size_read(inode);
- if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize))
- *ppos = isize;
-
- if (*ppos > ip->i_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (*ppos > ip->i_size)
- ip->i_size = *ppos;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
-
- if (ip->i_new_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ip->i_new_size = 0;
- if (ip->i_d.di_size > ip->i_size)
- ip->i_d.di_size = ip->i_size;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
+ xfs_aio_write_isize_update(inode, ppos, ret);
+ xfs_aio_write_newsize_update(ip);
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
@@ -562,6 +628,194 @@
return error;
}
+/*
+ * Common pre-write limit and setup checks.
+ *
+ * Returns with iolock held according to @iolock.
+ */
+STATIC ssize_t
+xfs_file_aio_write_checks(
+ struct file *file,
+ loff_t *pos,
+ size_t *count,
+ int *iolock)
+{
+ struct inode *inode = file->f_mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ xfs_fsize_t new_size;
+ int error = 0;
+
+ error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode));
+ if (error) {
+ xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
+ *iolock = 0;
+ return error;
+ }
+
+ new_size = *pos + *count;
+ if (new_size > ip->i_size)
+ ip->i_new_size = new_size;
+
+ if (likely(!(file->f_mode & FMODE_NOCMTIME)))
+ file_update_time(file);
+
+ /*
+ * If the offset is beyond the size of the file, we need to zero any
+ * blocks that fall between the existing EOF and the start of this
+ * write.
+ */
+ if (*pos > ip->i_size)
+ error = -xfs_zero_eof(ip, *pos, ip->i_size);
+
+ xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
+ if (error)
+ return error;
+
+ /*
+ * If we're writing the file then make sure to clear the setuid and
+ * setgid bits if the process is not being run by root. This keeps
+ * people from modifying setuid and setgid binaries.
+ */
+ return file_remove_suid(file);
+
+}
+
+/*
+ * xfs_file_dio_aio_write - handle direct IO writes
+ *
+ * Lock the inode appropriately to prepare for and issue a direct IO write.
+ * By separating it from the buffered write path we remove all the tricky to
+ * follow locking changes and looping.
+ *
+ * If there are cached pages or we're extending the file, we need IOLOCK_EXCL
+ * until we're sure the bytes at the new EOF have been zeroed and/or the cached
+ * pages are flushed out.
+ *
+ * In most cases the direct IO writes will be done holding IOLOCK_SHARED
+ * allowing them to be done in parallel with reads and other direct IO writes.
+ * However, if the IO is not aligned to filesystem blocks, the direct IO layer
+ * needs to do sub-block zeroing and that requires serialisation against other
+ * direct IOs to the same block. In this case we need to serialise the
+ * submission of the unaligned IOs so that we don't get racing block zeroing in
+ * the dio layer. To avoid the problem with aio, we also need to wait for
+ * outstanding IOs to complete so that unwritten extent conversion is completed
+ * before we try to map the overlapping block. This is currently implemented by
+ * hitting it with a big hammer (i.e. xfs_ioend_wait()).
+ *
+ * Returns with locks held indicated by @iolock and errors indicated by
+ * negative return values.
+ */
+STATIC ssize_t
+xfs_file_dio_aio_write(
+ struct kiocb *iocb,
+ const struct iovec *iovp,
+ unsigned long nr_segs,
+ loff_t pos,
+ size_t ocount,
+ int *iolock)
+{
+ struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
+ struct inode *inode = mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ ssize_t ret = 0;
+ size_t count = ocount;
+ int unaligned_io = 0;
+ struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ?
+ mp->m_rtdev_targp : mp->m_ddev_targp;
+
+ *iolock = 0;
+ if ((pos & target->bt_smask) || (count & target->bt_smask))
+ return -XFS_ERROR(EINVAL);
+
+ if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
+ unaligned_io = 1;
+
+ if (unaligned_io || mapping->nrpages || pos > ip->i_size)
+ *iolock = XFS_IOLOCK_EXCL;
+ else
+ *iolock = XFS_IOLOCK_SHARED;
+ xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock);
+
+ ret = xfs_file_aio_write_checks(file, &pos, &count, iolock);
+ if (ret)
+ return ret;
+
+ if (mapping->nrpages) {
+ WARN_ON(*iolock != XFS_IOLOCK_EXCL);
+ ret = -xfs_flushinval_pages(ip, (pos & PAGE_CACHE_MASK), -1,
+ FI_REMAPF_LOCKED);
+ if (ret)
+ return ret;
+ }
+
+ /*
+ * If we are doing unaligned IO, wait for all other IO to drain,
+ * otherwise demote the lock if we had to flush cached pages
+ */
+ if (unaligned_io)
+ xfs_ioend_wait(ip);
+ else if (*iolock == XFS_IOLOCK_EXCL) {
+ xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
+ *iolock = XFS_IOLOCK_SHARED;
+ }
+
+ trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
+ ret = generic_file_direct_write(iocb, iovp,
+ &nr_segs, pos, &iocb->ki_pos, count, ocount);
+
+ /* No fallback to buffered IO on errors for XFS. */
+ ASSERT(ret < 0 || ret == count);
+ return ret;
+}
+
+STATIC ssize_t
+xfs_file_buffered_aio_write(
+ struct kiocb *iocb,
+ const struct iovec *iovp,
+ unsigned long nr_segs,
+ loff_t pos,
+ size_t ocount,
+ int *iolock)
+{
+ struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
+ struct inode *inode = mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ ssize_t ret;
+ int enospc = 0;
+ size_t count = ocount;
+
+ *iolock = XFS_IOLOCK_EXCL;
+ xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock);
+
+ ret = xfs_file_aio_write_checks(file, &pos, &count, iolock);
+ if (ret)
+ return ret;
+
+ /* We can write back this queue in page reclaim */
+ current->backing_dev_info = mapping->backing_dev_info;
+
+write_retry:
+ trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0);
+ ret = generic_file_buffered_write(iocb, iovp, nr_segs,
+ pos, &iocb->ki_pos, count, ret);
+ /*
+ * if we just got an ENOSPC, flush the inode now we aren't holding any
+ * page locks and retry *once*
+ */
+ if (ret == -ENOSPC && !enospc) {
+ ret = -xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
+ if (ret)
+ return ret;
+ enospc = 1;
+ goto write_retry;
+ }
+ current->backing_dev_info = NULL;
+ return ret;
+}
+
STATIC ssize_t
xfs_file_aio_write(
struct kiocb *iocb,
@@ -573,234 +827,59 @@
struct address_space *mapping = file->f_mapping;
struct inode *inode = mapping->host;
struct xfs_inode *ip = XFS_I(inode);
- struct xfs_mount *mp = ip->i_mount;
- ssize_t ret = 0, error = 0;
- int ioflags = 0;
- xfs_fsize_t isize, new_size;
+ ssize_t ret;
int iolock;
- size_t ocount = 0, count;
- int need_i_mutex;
+ size_t ocount = 0;
XFS_STATS_INC(xs_write_calls);
BUG_ON(iocb->ki_pos != pos);
- if (unlikely(file->f_flags & O_DIRECT))
- ioflags |= IO_ISDIRECT;
- if (file->f_mode & FMODE_NOCMTIME)
- ioflags |= IO_INVIS;
+ ret = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
+ if (ret)
+ return ret;
- error = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
- if (error)
- return error;
-
- count = ocount;
- if (count == 0)
+ if (ocount == 0)
return 0;
- xfs_wait_for_freeze(mp, SB_FREEZE_WRITE);
+ xfs_wait_for_freeze(ip->i_mount, SB_FREEZE_WRITE);
- if (XFS_FORCED_SHUTDOWN(mp))
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
-relock:
- if (ioflags & IO_ISDIRECT) {
- iolock = XFS_IOLOCK_SHARED;
- need_i_mutex = 0;
- } else {
- iolock = XFS_IOLOCK_EXCL;
- need_i_mutex = 1;
- mutex_lock(&inode->i_mutex);
- }
+ if (unlikely(file->f_flags & O_DIRECT))
+ ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos,
+ ocount, &iolock);
+ else
+ ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos,
+ ocount, &iolock);
- xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
+ xfs_aio_write_isize_update(inode, &iocb->ki_pos, ret);
-start:
- error = -generic_write_checks(file, &pos, &count,
- S_ISBLK(inode->i_mode));
- if (error) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
- goto out_unlock_mutex;
- }
-
- if (ioflags & IO_ISDIRECT) {
- xfs_buftarg_t *target =
- XFS_IS_REALTIME_INODE(ip) ?
- mp->m_rtdev_targp : mp->m_ddev_targp;
-
- if ((pos & target->bt_smask) || (count & target->bt_smask)) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
- return XFS_ERROR(-EINVAL);
- }
-
- if (!need_i_mutex && (mapping->nrpages || pos > ip->i_size)) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
- iolock = XFS_IOLOCK_EXCL;
- need_i_mutex = 1;
- mutex_lock(&inode->i_mutex);
- xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
- goto start;
- }
- }
-
- new_size = pos + count;
- if (new_size > ip->i_size)
- ip->i_new_size = new_size;
-
- if (likely(!(ioflags & IO_INVIS)))
- file_update_time(file);
-
- /*
- * If the offset is beyond the size of the file, we have a couple
- * of things to do. First, if there is already space allocated
- * we need to either create holes or zero the disk or ...
- *
- * If there is a page where the previous size lands, we need
- * to zero it out up to the new size.
- */
-
- if (pos > ip->i_size) {
- error = xfs_zero_eof(ip, pos, ip->i_size);
- if (error) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- goto out_unlock_internal;
- }
- }
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
- /*
- * If we're writing the file then make sure to clear the
- * setuid and setgid bits if the process is not being run
- * by root. This keeps people from modifying setuid and
- * setgid binaries.
- */
- error = -file_remove_suid(file);
- if (unlikely(error))
- goto out_unlock_internal;
-
- /* We can write back this queue in page reclaim */
- current->backing_dev_info = mapping->backing_dev_info;
-
- if ((ioflags & IO_ISDIRECT)) {
- if (mapping->nrpages) {
- WARN_ON(need_i_mutex == 0);
- error = xfs_flushinval_pages(ip,
- (pos & PAGE_CACHE_MASK),
- -1, FI_REMAPF_LOCKED);
- if (error)
- goto out_unlock_internal;
- }
-
- if (need_i_mutex) {
- /* demote the lock now the cached pages are gone */
- xfs_ilock_demote(ip, XFS_IOLOCK_EXCL);
- mutex_unlock(&inode->i_mutex);
-
- iolock = XFS_IOLOCK_SHARED;
- need_i_mutex = 0;
- }
-
- trace_xfs_file_direct_write(ip, count, iocb->ki_pos, ioflags);
- ret = generic_file_direct_write(iocb, iovp,
- &nr_segs, pos, &iocb->ki_pos, count, ocount);
-
- /*
- * direct-io write to a hole: fall through to buffered I/O
- * for completing the rest of the request.
- */
- if (ret >= 0 && ret != count) {
- XFS_STATS_ADD(xs_write_bytes, ret);
-
- pos += ret;
- count -= ret;
-
- ioflags &= ~IO_ISDIRECT;
- xfs_iunlock(ip, iolock);
- goto relock;
- }
- } else {
- int enospc = 0;
- ssize_t ret2 = 0;
-
-write_retry:
- trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, ioflags);
- ret2 = generic_file_buffered_write(iocb, iovp, nr_segs,
- pos, &iocb->ki_pos, count, ret);
- /*
- * if we just got an ENOSPC, flush the inode now we
- * aren't holding any page locks and retry *once*
- */
- if (ret2 == -ENOSPC && !enospc) {
- error = xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
- if (error)
- goto out_unlock_internal;
- enospc = 1;
- goto write_retry;
- }
- ret = ret2;
- }
-
- current->backing_dev_info = NULL;
-
- isize = i_size_read(inode);
- if (unlikely(ret < 0 && ret != -EFAULT && iocb->ki_pos > isize))
- iocb->ki_pos = isize;
-
- if (iocb->ki_pos > ip->i_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (iocb->ki_pos > ip->i_size)
- ip->i_size = iocb->ki_pos;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
-
- error = -ret;
if (ret <= 0)
- goto out_unlock_internal;
-
- XFS_STATS_ADD(xs_write_bytes, ret);
+ goto out_unlock;
/* Handle various SYNC-type writes */
if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
loff_t end = pos + ret - 1;
- int error2;
+ int error, error2;
- xfs_iunlock(ip, iolock);
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
-
- error2 = filemap_write_and_wait_range(mapping, pos, end);
- if (!error)
- error = error2;
- if (need_i_mutex)
- mutex_lock(&inode->i_mutex);
- xfs_ilock(ip, iolock);
+ xfs_rw_iunlock(ip, iolock);
+ error = filemap_write_and_wait_range(mapping, pos, end);
+ xfs_rw_ilock(ip, iolock);
error2 = -xfs_file_fsync(file,
(file->f_flags & __O_SYNC) ? 0 : 1);
- if (!error)
- error = error2;
+ if (error)
+ ret = error;
+ else if (error2)
+ ret = error2;
}
- out_unlock_internal:
- if (ip->i_new_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ip->i_new_size = 0;
- /*
- * If this was a direct or synchronous I/O that failed (such
- * as ENOSPC) then part of the I/O may have been written to
- * disk before the error occured. In this case the on-disk
- * file size may have been adjusted beyond the in-memory file
- * size and now needs to be truncated back.
- */
- if (ip->i_d.di_size > ip->i_size)
- ip->i_d.di_size = ip->i_size;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
- xfs_iunlock(ip, iolock);
- out_unlock_mutex:
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
- return -error;
+out_unlock:
+ xfs_aio_write_newsize_update(ip);
+ xfs_rw_iunlock(ip, iolock);
+ return ret;
}
STATIC int
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index ad442d9..b06ede1 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -39,6 +39,7 @@
#include "xfs_dfrag.h"
#include "xfs_fsops.h"
#include "xfs_vnodeops.h"
+#include "xfs_discard.h"
#include "xfs_quota.h"
#include "xfs_inode_item.h"
#include "xfs_export.h"
@@ -1294,6 +1295,8 @@
trace_xfs_file_ioctl(ip);
switch (cmd) {
+ case FITRIM:
+ return xfs_ioc_trim(mp, arg);
case XFS_IOC_ALLOCSP:
case XFS_IOC_FREESP:
case XFS_IOC_RESVSP:
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index bd07f73..9731898 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -1414,7 +1414,7 @@
xfs_save_resvblks(mp);
xfs_quiesce_attr(mp);
- return -xfs_fs_log_dummy(mp, SYNC_WAIT);
+ return -xfs_fs_log_dummy(mp);
}
STATIC int
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index a02480d..e22f005 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -362,7 +362,7 @@
/* mark the log as covered if needed */
if (xfs_log_need_covered(mp))
- error2 = xfs_fs_log_dummy(mp, SYNC_WAIT);
+ error2 = xfs_fs_log_dummy(mp);
/* flush data-only devices */
if (mp->m_rtdev_targp)
@@ -503,13 +503,14 @@
int error;
if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
- xfs_log_force(mp, 0);
- xfs_reclaim_inodes(mp, 0);
/* dgc: errors ignored here */
- error = xfs_qm_sync(mp, SYNC_TRYLOCK);
if (mp->m_super->s_frozen == SB_UNFROZEN &&
xfs_log_need_covered(mp))
- error = xfs_fs_log_dummy(mp, 0);
+ error = xfs_fs_log_dummy(mp);
+ else
+ xfs_log_force(mp, 0);
+ xfs_reclaim_inodes(mp, 0);
+ error = xfs_qm_sync(mp, SYNC_TRYLOCK);
}
mp->m_sync_seq++;
wake_up(&mp->m_wait_single_sync_task);
diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c
index 7bb5092..ee3cee0 100644
--- a/fs/xfs/linux-2.6/xfs_sysctl.c
+++ b/fs/xfs/linux-2.6/xfs_sysctl.c
@@ -18,6 +18,7 @@
#include "xfs.h"
#include <linux/sysctl.h>
#include <linux/proc_fs.h>
+#include "xfs_error.h"
static struct ctl_table_header *xfs_table_header;
@@ -51,6 +52,26 @@
return ret;
}
+
+STATIC int
+xfs_panic_mask_proc_handler(
+ ctl_table *ctl,
+ int write,
+ void __user *buffer,
+ size_t *lenp,
+ loff_t *ppos)
+{
+ int ret, *valp = ctl->data;
+
+ ret = proc_dointvec_minmax(ctl, write, buffer, lenp, ppos);
+ if (!ret && write) {
+ xfs_panic_mask = *valp;
+#ifdef DEBUG
+ xfs_panic_mask |= (XFS_PTAG_SHUTDOWN_CORRUPT | XFS_PTAG_LOGRES);
+#endif
+ }
+ return ret;
+}
#endif /* CONFIG_PROC_FS */
static ctl_table xfs_table[] = {
@@ -77,7 +98,7 @@
.data = &xfs_params.panic_mask.val,
.maxlen = sizeof(int),
.mode = 0644,
- .proc_handler = proc_dointvec_minmax,
+ .proc_handler = xfs_panic_mask_proc_handler,
.extra1 = &xfs_params.panic_mask.min,
.extra2 = &xfs_params.panic_mask.max
},
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
index 647af2a..2d0bcb4 100644
--- a/fs/xfs/linux-2.6/xfs_trace.h
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -1759,6 +1759,39 @@
DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_cancel);
DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_skip);
+DECLARE_EVENT_CLASS(xfs_discard_class,
+ TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
+ xfs_agblock_t agbno, xfs_extlen_t len),
+ TP_ARGS(mp, agno, agbno, len),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(xfs_agnumber_t, agno)
+ __field(xfs_agblock_t, agbno)
+ __field(xfs_extlen_t, len)
+ ),
+ TP_fast_assign(
+ __entry->dev = mp->m_super->s_dev;
+ __entry->agno = agno;
+ __entry->agbno = agbno;
+ __entry->len = len;
+ ),
+ TP_printk("dev %d:%d agno %u agbno %u len %u\n",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->agno,
+ __entry->agbno,
+ __entry->len)
+)
+
+#define DEFINE_DISCARD_EVENT(name) \
+DEFINE_EVENT(xfs_discard_class, name, \
+ TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \
+ xfs_agblock_t agbno, xfs_extlen_t len), \
+ TP_ARGS(mp, agno, agbno, len))
+DEFINE_DISCARD_EVENT(xfs_discard_extent);
+DEFINE_DISCARD_EVENT(xfs_discard_toosmall);
+DEFINE_DISCARD_EVENT(xfs_discard_exclude);
+DEFINE_DISCARD_EVENT(xfs_discard_busy);
+
#endif /* _TRACE_XFS_H */
#undef TRACE_INCLUDE_PATH
diff --git a/fs/xfs/support/debug.c b/fs/xfs/support/debug.c
index 975aa10..e6cf955 100644
--- a/fs/xfs/support/debug.c
+++ b/fs/xfs/support/debug.c
@@ -25,86 +25,78 @@
#include "xfs_mount.h"
#include "xfs_error.h"
-static char message[1024]; /* keep it off the stack */
-static DEFINE_SPINLOCK(xfs_err_lock);
-
-/* Translate from CE_FOO to KERN_FOO, err_level(CE_FOO) == KERN_FOO */
-#define XFS_MAX_ERR_LEVEL 7
-#define XFS_ERR_MASK ((1 << 3) - 1)
-static const char * const err_level[XFS_MAX_ERR_LEVEL+1] =
- {KERN_EMERG, KERN_ALERT, KERN_CRIT,
- KERN_ERR, KERN_WARNING, KERN_NOTICE,
- KERN_INFO, KERN_DEBUG};
-
void
-cmn_err(register int level, char *fmt, ...)
+cmn_err(
+ const char *lvl,
+ const char *fmt,
+ ...)
{
- char *fp = fmt;
- int len;
- ulong flags;
- va_list ap;
+ struct va_format vaf;
+ va_list args;
- level &= XFS_ERR_MASK;
- if (level > XFS_MAX_ERR_LEVEL)
- level = XFS_MAX_ERR_LEVEL;
- spin_lock_irqsave(&xfs_err_lock,flags);
- va_start(ap, fmt);
- if (*fmt == '!') fp++;
- len = vsnprintf(message, sizeof(message), fp, ap);
- if (len >= sizeof(message))
- len = sizeof(message) - 1;
- if (message[len-1] == '\n')
- message[len-1] = 0;
- printk("%s%s\n", err_level[level], message);
- va_end(ap);
- spin_unlock_irqrestore(&xfs_err_lock,flags);
- BUG_ON(level == CE_PANIC);
+ va_start(args, fmt);
+ vaf.fmt = fmt;
+ vaf.va = &args;
+
+ printk("%s%pV", lvl, &vaf);
+ va_end(args);
+
+ BUG_ON(strncmp(lvl, KERN_EMERG, strlen(KERN_EMERG)) == 0);
}
void
-xfs_fs_vcmn_err(
- int level,
+xfs_fs_cmn_err(
+ const char *lvl,
struct xfs_mount *mp,
- char *fmt,
- va_list ap)
+ const char *fmt,
+ ...)
{
- unsigned long flags;
- int len = 0;
+ struct va_format vaf;
+ va_list args;
- level &= XFS_ERR_MASK;
- if (level > XFS_MAX_ERR_LEVEL)
- level = XFS_MAX_ERR_LEVEL;
+ va_start(args, fmt);
+ vaf.fmt = fmt;
+ vaf.va = &args;
- spin_lock_irqsave(&xfs_err_lock,flags);
+ printk("%sFilesystem %s: %pV", lvl, mp->m_fsname, &vaf);
+ va_end(args);
- if (mp) {
- len = sprintf(message, "Filesystem \"%s\": ", mp->m_fsname);
+ BUG_ON(strncmp(lvl, KERN_EMERG, strlen(KERN_EMERG)) == 0);
+}
- /*
- * Skip the printk if we can't print anything useful
- * due to an over-long device name.
- */
- if (len >= sizeof(message))
- goto out;
+/* All callers to xfs_cmn_err use CE_ALERT, so don't bother testing lvl */
+void
+xfs_cmn_err(
+ int panic_tag,
+ const char *lvl,
+ struct xfs_mount *mp,
+ const char *fmt,
+ ...)
+{
+ struct va_format vaf;
+ va_list args;
+ int panic = 0;
+
+ if (xfs_panic_mask && (xfs_panic_mask & panic_tag)) {
+ printk(KERN_ALERT "XFS: Transforming an alert into a BUG.");
+ panic = 1;
}
- len = vsnprintf(message + len, sizeof(message) - len, fmt, ap);
- if (len >= sizeof(message))
- len = sizeof(message) - 1;
- if (message[len-1] == '\n')
- message[len-1] = 0;
+ va_start(args, fmt);
+ vaf.fmt = fmt;
+ vaf.va = &args;
- printk("%s%s\n", err_level[level], message);
- out:
- spin_unlock_irqrestore(&xfs_err_lock,flags);
+ printk(KERN_ALERT "Filesystem %s: %pV", mp->m_fsname, &vaf);
+ va_end(args);
- BUG_ON(level == CE_PANIC);
+ BUG_ON(panic);
}
void
assfail(char *expr, char *file, int line)
{
- printk("Assertion failed: %s, file: %s, line: %d\n", expr, file, line);
+ printk(KERN_CRIT "Assertion failed: %s, file: %s, line: %d\n", expr,
+ file, line);
BUG();
}
diff --git a/fs/xfs/support/debug.h b/fs/xfs/support/debug.h
index d2d2046..05699f6 100644
--- a/fs/xfs/support/debug.h
+++ b/fs/xfs/support/debug.h
@@ -20,15 +20,22 @@
#include <stdarg.h>
-#define CE_DEBUG 7 /* debug */
-#define CE_CONT 6 /* continuation */
-#define CE_NOTE 5 /* notice */
-#define CE_WARN 4 /* warning */
-#define CE_ALERT 1 /* alert */
-#define CE_PANIC 0 /* panic */
+struct xfs_mount;
-extern void cmn_err(int, char *, ...)
- __attribute__ ((format (printf, 2, 3)));
+#define CE_DEBUG KERN_DEBUG
+#define CE_CONT KERN_INFO
+#define CE_NOTE KERN_NOTICE
+#define CE_WARN KERN_WARNING
+#define CE_ALERT KERN_ALERT
+#define CE_PANIC KERN_EMERG
+
+void cmn_err(const char *lvl, const char *fmt, ...)
+ __attribute__ ((format (printf, 2, 3)));
+void xfs_fs_cmn_err( const char *lvl, struct xfs_mount *mp,
+ const char *fmt, ...) __attribute__ ((format (printf, 3, 4)));
+void xfs_cmn_err( int panic_tag, const char *lvl, struct xfs_mount *mp,
+ const char *fmt, ...) __attribute__ ((format (printf, 4, 5)));
+
extern void assfail(char *expr, char *f, int l);
#define ASSERT_ALWAYS(expr) \
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index fa8723f..f322798 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -41,10 +41,6 @@
#define XFSA_FIXUP_BNO_OK 1
#define XFSA_FIXUP_CNT_OK 2
-static int
-xfs_alloc_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
- xfs_agblock_t bno, xfs_extlen_t len);
-
/*
* Prototypes for per-ag allocation routines
*/
@@ -94,7 +90,7 @@
* Lookup the first record less than or equal to [bno, len]
* in the btree given by cur.
*/
-STATIC int /* error */
+int /* error */
xfs_alloc_lookup_le(
struct xfs_btree_cur *cur, /* btree cursor */
xfs_agblock_t bno, /* starting block of extent */
@@ -127,7 +123,7 @@
/*
* Get the data from the pointed-to record.
*/
-STATIC int /* error */
+int /* error */
xfs_alloc_get_rec(
struct xfs_btree_cur *cur, /* btree cursor */
xfs_agblock_t *bno, /* output: starting block of extent */
@@ -2615,7 +2611,7 @@
* will require a synchronous transaction, but it can still be
* used to distinguish between a partial or exact match.
*/
-static int
+int
xfs_alloc_busy_search(
struct xfs_mount *mp,
xfs_agnumber_t agno,
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h
index 895009a..0ab56b3 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/xfs_alloc.h
@@ -19,6 +19,7 @@
#define __XFS_ALLOC_H__
struct xfs_buf;
+struct xfs_btree_cur;
struct xfs_mount;
struct xfs_perag;
struct xfs_trans;
@@ -118,16 +119,16 @@
struct xfs_perag *pag);
#ifdef __KERNEL__
-
void
-xfs_alloc_busy_insert(xfs_trans_t *tp,
- xfs_agnumber_t agno,
- xfs_agblock_t bno,
- xfs_extlen_t len);
+xfs_alloc_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno,
+ xfs_agblock_t bno, xfs_extlen_t len);
void
xfs_alloc_busy_clear(struct xfs_mount *mp, struct xfs_busy_extent *busyp);
+int
+xfs_alloc_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
+ xfs_agblock_t bno, xfs_extlen_t len);
#endif /* __KERNEL__ */
/*
@@ -205,4 +206,18 @@
xfs_fsblock_t bno, /* starting block number of extent */
xfs_extlen_t len); /* length of extent */
+int /* error */
+xfs_alloc_lookup_le(
+ struct xfs_btree_cur *cur, /* btree cursor */
+ xfs_agblock_t bno, /* starting block of extent */
+ xfs_extlen_t len, /* length of extent */
+ int *stat); /* success/failure */
+
+int /* error */
+xfs_alloc_get_rec(
+ struct xfs_btree_cur *cur, /* btree cursor */
+ xfs_agblock_t *bno, /* output: starting block of extent */
+ xfs_extlen_t *len, /* output: length of extent */
+ int *stat); /* output: success/failure */
+
#endif /* __XFS_ALLOC_H__ */
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index ed2b65f..98c6f73 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -141,7 +141,6 @@
#define xfs_buf_item_log_check(x)
#endif
-STATIC void xfs_buf_error_relse(xfs_buf_t *bp);
STATIC void xfs_buf_do_callbacks(struct xfs_buf *bp);
/*
@@ -959,86 +958,70 @@
*/
void
xfs_buf_iodone_callbacks(
- xfs_buf_t *bp)
+ struct xfs_buf *bp)
{
- xfs_log_item_t *lip;
- static ulong lasttime;
- static xfs_buftarg_t *lasttarg;
- xfs_mount_t *mp;
+ struct xfs_log_item *lip = bp->b_fspriv;
+ struct xfs_mount *mp = lip->li_mountp;
+ static ulong lasttime;
+ static xfs_buftarg_t *lasttarg;
- ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
- lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
+ if (likely(!XFS_BUF_GETERROR(bp)))
+ goto do_callbacks;
- if (XFS_BUF_GETERROR(bp) != 0) {
- /*
- * If we've already decided to shutdown the filesystem
- * because of IO errors, there's no point in giving this
- * a retry.
- */
- mp = lip->li_mountp;
- if (XFS_FORCED_SHUTDOWN(mp)) {
- ASSERT(XFS_BUF_TARGET(bp) == mp->m_ddev_targp);
- XFS_BUF_SUPER_STALE(bp);
- trace_xfs_buf_item_iodone(bp, _RET_IP_);
- xfs_buf_do_callbacks(bp);
- XFS_BUF_SET_FSPRIVATE(bp, NULL);
- XFS_BUF_CLR_IODONE_FUNC(bp);
- xfs_buf_ioend(bp, 0);
- return;
- }
+ /*
+ * If we've already decided to shutdown the filesystem because of
+ * I/O errors, there's no point in giving this a retry.
+ */
+ if (XFS_FORCED_SHUTDOWN(mp)) {
+ XFS_BUF_SUPER_STALE(bp);
+ trace_xfs_buf_item_iodone(bp, _RET_IP_);
+ goto do_callbacks;
+ }
- if ((XFS_BUF_TARGET(bp) != lasttarg) ||
- (time_after(jiffies, (lasttime + 5*HZ)))) {
- lasttime = jiffies;
- cmn_err(CE_ALERT, "Device %s, XFS metadata write error"
- " block 0x%llx in %s",
- XFS_BUFTARG_NAME(XFS_BUF_TARGET(bp)),
- (__uint64_t)XFS_BUF_ADDR(bp), mp->m_fsname);
- }
- lasttarg = XFS_BUF_TARGET(bp);
+ if (XFS_BUF_TARGET(bp) != lasttarg ||
+ time_after(jiffies, (lasttime + 5*HZ))) {
+ lasttime = jiffies;
+ cmn_err(CE_ALERT, "Device %s, XFS metadata write error"
+ " block 0x%llx in %s",
+ XFS_BUFTARG_NAME(XFS_BUF_TARGET(bp)),
+ (__uint64_t)XFS_BUF_ADDR(bp), mp->m_fsname);
+ }
+ lasttarg = XFS_BUF_TARGET(bp);
- if (XFS_BUF_ISASYNC(bp)) {
- /*
- * If the write was asynchronous then noone will be
- * looking for the error. Clear the error state
- * and write the buffer out again delayed write.
- *
- * XXXsup This is OK, so long as we catch these
- * before we start the umount; we don't want these
- * DELWRI metadata bufs to be hanging around.
- */
- XFS_BUF_ERROR(bp,0); /* errno of 0 unsets the flag */
+ /*
+ * If the write was asynchronous then noone will be looking for the
+ * error. Clear the error state and write the buffer out again.
+ *
+ * During sync or umount we'll write all pending buffers again
+ * synchronous, which will catch these errors if they keep hanging
+ * around.
+ */
+ if (XFS_BUF_ISASYNC(bp)) {
+ XFS_BUF_ERROR(bp, 0); /* errno of 0 unsets the flag */
- if (!(XFS_BUF_ISSTALE(bp))) {
- XFS_BUF_DELAYWRITE(bp);
- XFS_BUF_DONE(bp);
- XFS_BUF_SET_START(bp);
- }
- ASSERT(XFS_BUF_IODONE_FUNC(bp));
- trace_xfs_buf_item_iodone_async(bp, _RET_IP_);
- xfs_buf_relse(bp);
- } else {
- /*
- * If the write of the buffer was not asynchronous,
- * then we want to make sure to return the error
- * to the caller of bwrite(). Because of this we
- * cannot clear the B_ERROR state at this point.
- * Instead we install a callback function that
- * will be called when the buffer is released, and
- * that routine will clear the error state and
- * set the buffer to be written out again after
- * some delay.
- */
- /* We actually overwrite the existing b-relse
- function at times, but we're gonna be shutting down
- anyway. */
- XFS_BUF_SET_BRELSE_FUNC(bp,xfs_buf_error_relse);
+ if (!XFS_BUF_ISSTALE(bp)) {
+ XFS_BUF_DELAYWRITE(bp);
XFS_BUF_DONE(bp);
- XFS_BUF_FINISH_IOWAIT(bp);
+ XFS_BUF_SET_START(bp);
}
+ ASSERT(XFS_BUF_IODONE_FUNC(bp));
+ trace_xfs_buf_item_iodone_async(bp, _RET_IP_);
+ xfs_buf_relse(bp);
return;
}
+ /*
+ * If the write of the buffer was synchronous, we want to make
+ * sure to return the error to the caller of xfs_bwrite().
+ */
+ XFS_BUF_STALE(bp);
+ XFS_BUF_DONE(bp);
+ XFS_BUF_UNDELAYWRITE(bp);
+
+ trace_xfs_buf_error_relse(bp, _RET_IP_);
+ xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
+
+do_callbacks:
xfs_buf_do_callbacks(bp);
XFS_BUF_SET_FSPRIVATE(bp, NULL);
XFS_BUF_CLR_IODONE_FUNC(bp);
@@ -1046,42 +1029,6 @@
}
/*
- * This is a callback routine attached to a buffer which gets an error
- * when being written out synchronously.
- */
-STATIC void
-xfs_buf_error_relse(
- xfs_buf_t *bp)
-{
- xfs_log_item_t *lip;
- xfs_mount_t *mp;
-
- lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
- mp = (xfs_mount_t *)lip->li_mountp;
- ASSERT(XFS_BUF_TARGET(bp) == mp->m_ddev_targp);
-
- XFS_BUF_STALE(bp);
- XFS_BUF_DONE(bp);
- XFS_BUF_UNDELAYWRITE(bp);
- XFS_BUF_ERROR(bp,0);
-
- trace_xfs_buf_error_relse(bp, _RET_IP_);
-
- if (! XFS_FORCED_SHUTDOWN(mp))
- xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
- /*
- * We have to unpin the pinned buffers so do the
- * callbacks.
- */
- xfs_buf_do_callbacks(bp);
- XFS_BUF_SET_FSPRIVATE(bp, NULL);
- XFS_BUF_CLR_IODONE_FUNC(bp);
- XFS_BUF_SET_BRELSE_FUNC(bp,NULL);
- xfs_buf_relse(bp);
-}
-
-
-/*
* This is the iodone() function for buffers which have been
* logged. It is called when they are eventually flushed out.
* It should remove the buf item from the AIL, and free the buf item.
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index c78cc6a..4c7db74 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -152,37 +152,6 @@
}
#endif /* DEBUG */
-
-void
-xfs_fs_cmn_err(int level, xfs_mount_t *mp, char *fmt, ...)
-{
- va_list ap;
-
- va_start(ap, fmt);
- xfs_fs_vcmn_err(level, mp, fmt, ap);
- va_end(ap);
-}
-
-void
-xfs_cmn_err(int panic_tag, int level, xfs_mount_t *mp, char *fmt, ...)
-{
- va_list ap;
-
-#ifdef DEBUG
- xfs_panic_mask |= (XFS_PTAG_SHUTDOWN_CORRUPT | XFS_PTAG_LOGRES);
-#endif
-
- if (xfs_panic_mask && (xfs_panic_mask & panic_tag)
- && (level & CE_ALERT)) {
- level &= ~CE_ALERT;
- level |= CE_PANIC;
- cmn_err(CE_ALERT, "XFS: Transforming an alert into a BUG.");
- }
- va_start(ap, fmt);
- xfs_fs_vcmn_err(level, mp, fmt, ap);
- va_end(ap);
-}
-
void
xfs_error_report(
const char *tag,
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index f338847..10dce54 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -136,8 +136,8 @@
xfs_error_test((tag), (mp)->m_fixedfsid, "expr", __LINE__, __FILE__, \
(rf))))
-extern int xfs_errortag_add(int error_tag, xfs_mount_t *mp);
-extern int xfs_errortag_clearall(xfs_mount_t *mp, int loud);
+extern int xfs_errortag_add(int error_tag, struct xfs_mount *mp);
+extern int xfs_errortag_clearall(struct xfs_mount *mp, int loud);
#else
#define XFS_TEST_ERROR(expr, mp, tag, rf) (expr)
#define xfs_errortag_add(tag, mp) (ENOSYS)
@@ -162,21 +162,15 @@
struct xfs_mount;
-extern void xfs_fs_vcmn_err(int level, struct xfs_mount *mp,
- char *fmt, va_list ap)
- __attribute__ ((format (printf, 3, 0)));
-extern void xfs_cmn_err(int panic_tag, int level, struct xfs_mount *mp,
- char *fmt, ...)
- __attribute__ ((format (printf, 4, 5)));
-extern void xfs_fs_cmn_err(int level, struct xfs_mount *mp, char *fmt, ...)
- __attribute__ ((format (printf, 3, 4)));
-
extern void xfs_hex_dump(void *p, int length);
#define xfs_fs_repair_cmn_err(level, mp, fmt, args...) \
xfs_fs_cmn_err(level, mp, fmt " Unmount and run xfs_repair.", ## args)
#define xfs_fs_mount_cmn_err(f, fmt, args...) \
- ((f & XFS_MFSI_QUIET)? (void)0 : cmn_err(CE_WARN, "XFS: " fmt, ## args))
+ do { \
+ if (!(f & XFS_MFSI_QUIET)) \
+ cmn_err(CE_WARN, "XFS: " fmt, ## args); \
+ } while (0)
#endif /* __XFS_ERROR_H__ */
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index f56d30e..cec89dd 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -612,12 +612,13 @@
*
* We cannot use an inode here for this - that will push dirty state back up
* into the VFS and then periodic inode flushing will prevent log covering from
- * making progress. Hence we log a field in the superblock instead.
+ * making progress. Hence we log a field in the superblock instead and use a
+ * synchronous transaction to ensure the superblock is immediately unpinned
+ * and can be written back.
*/
int
xfs_fs_log_dummy(
- xfs_mount_t *mp,
- int flags)
+ xfs_mount_t *mp)
{
xfs_trans_t *tp;
int error;
@@ -632,8 +633,7 @@
/* log the UUID because it is an unchanging field */
xfs_mod_sb(tp, XFS_SB_UUID);
- if (flags & SYNC_WAIT)
- xfs_trans_set_sync(tp);
+ xfs_trans_set_sync(tp);
return xfs_trans_commit(tp, 0);
}
diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h
index a786c52..1b6a98b 100644
--- a/fs/xfs/xfs_fsops.h
+++ b/fs/xfs/xfs_fsops.h
@@ -25,6 +25,6 @@
extern int xfs_reserve_blocks(xfs_mount_t *mp, __uint64_t *inval,
xfs_fsop_resblks_t *outval);
extern int xfs_fs_goingdown(xfs_mount_t *mp, __uint32_t inflags);
-extern int xfs_fs_log_dummy(xfs_mount_t *mp, int flags);
+extern int xfs_fs_log_dummy(struct xfs_mount *mp);
#endif /* __XFS_FSOPS_H__ */
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 0bf24b1..ae6fef1f 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -377,7 +377,7 @@
cmn_err(CE_NOTE, "XFS mounting filesystem %s", mp->m_fsname);
else {
cmn_err(CE_NOTE,
- "!Mounting filesystem \"%s\" in no-recovery mode. Filesystem will be inconsistent.",
+ "Mounting filesystem \"%s\" in no-recovery mode. Filesystem will be inconsistent.",
mp->m_fsname);
ASSERT(mp->m_flags & XFS_MOUNT_RDONLY);
}
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 204d8e5..aa0ebb7 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -3800,7 +3800,7 @@
log->l_flags &= ~XLOG_RECOVERY_NEEDED;
} else {
cmn_err(CE_DEBUG,
- "!Ending clean XFS mount for filesystem: %s\n",
+ "Ending clean XFS mount for filesystem: %s\n",
log->l_mp->m_fsname);
}
return 0;
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index f80a067..33dbc4e 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -1137,7 +1137,7 @@
if (blkdelta)
xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, -blkdelta, rsvd);
out:
- ASSERT(error = 0);
+ ASSERT(error == 0);
return;
}
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index 359ef11..78ca429 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -148,9 +148,7 @@
u32 suprise_removal_ok:1;
u32 power_manageable:1;
u32 performance_manageable:1;
- u32 wake_capable:1; /* Wakeup(_PRW) supported? */
- u32 force_power_state:1;
- u32 reserved:22;
+ u32 reserved:24;
};
/* File System */
@@ -242,20 +240,14 @@
struct acpi_device_wakeup_flags {
u8 valid:1; /* Can successfully enable wakeup? */
u8 run_wake:1; /* Run-Wake GPE devices */
- u8 always_enabled:1; /* Run-wake devices that are always enabled */
u8 notifier_present:1; /* Wake-up notify handler has been installed */
};
-struct acpi_device_wakeup_state {
- u8 enabled:1;
-};
-
struct acpi_device_wakeup {
acpi_handle gpe_device;
u64 gpe_number;
u64 sleep_state;
struct acpi_handle_list resources;
- struct acpi_device_wakeup_state state;
struct acpi_device_wakeup_flags flags;
int prepare_count;
int run_wake_count;
@@ -328,8 +320,8 @@
acpi_status acpi_bus_get_status_handle(acpi_handle handle,
unsigned long long *sta);
int acpi_bus_get_status(struct acpi_device *device);
-int acpi_bus_get_power(acpi_handle handle, int *state);
int acpi_bus_set_power(acpi_handle handle, int state);
+int acpi_bus_update_power(acpi_handle handle, int *state_p);
bool acpi_bus_power_manageable(acpi_handle handle);
bool acpi_bus_can_wakeup(acpi_handle handle);
#ifdef CONFIG_ACPI_PROC_EVENT
diff --git a/include/acpi/acpixf.h b/include/acpi/acpixf.h
index 53b7cfd..241b8a0 100644
--- a/include/acpi/acpixf.h
+++ b/include/acpi/acpixf.h
@@ -47,7 +47,7 @@
/* Current ACPICA subsystem version in YYYYMMDD format */
-#define ACPI_CA_VERSION 0x20101013
+#define ACPI_CA_VERSION 0x20101209
#include "actypes.h"
#include "actbl.h"
@@ -229,6 +229,10 @@
acpi_install_initialization_handler(acpi_init_handler handler, u32 function);
acpi_status
+acpi_install_global_event_handler(ACPI_GBL_EVENT_HANDLER handler,
+ void *context);
+
+acpi_status
acpi_install_fixed_event_handler(u32 acpi_event,
acpi_event_handler handler, void *context);
@@ -258,11 +262,11 @@
acpi_status
acpi_install_gpe_handler(acpi_handle gpe_device,
u32 gpe_number,
- u32 type, acpi_event_handler address, void *context);
+ u32 type, acpi_gpe_handler address, void *context);
acpi_status
acpi_remove_gpe_handler(acpi_handle gpe_device,
- u32 gpe_number, acpi_event_handler address);
+ u32 gpe_number, acpi_gpe_handler address);
#ifdef ACPI_FUTURE_USAGE
acpi_status acpi_install_exception_handler(acpi_exception_handler handler);
@@ -292,11 +296,13 @@
acpi_status acpi_disable_gpe(acpi_handle gpe_device, u32 gpe_number);
-acpi_status acpi_gpe_can_wake(acpi_handle gpe_device, u32 gpe_number);
-
acpi_status acpi_clear_gpe(acpi_handle gpe_device, u32 gpe_number);
-acpi_status acpi_gpe_wakeup(acpi_handle gpe_device, u32 gpe_number, u8 action);
+acpi_status
+acpi_setup_gpe_for_wake(acpi_handle parent_device,
+ acpi_handle gpe_device, u32 gpe_number);
+
+acpi_status acpi_set_gpe_wake_mask(acpi_handle gpe_device, u32 gpe_number, u8 action);
acpi_status
acpi_get_gpe_status(acpi_handle gpe_device,
@@ -315,7 +321,7 @@
acpi_status acpi_remove_gpe_block(acpi_handle gpe_device);
-acpi_status acpi_update_gpes(void);
+acpi_status acpi_update_all_gpes(void);
/*
* Resource interfaces
diff --git a/include/acpi/actypes.h b/include/acpi/actypes.h
index 2b134b6..939a431 100644
--- a/include/acpi/actypes.h
+++ b/include/acpi/actypes.h
@@ -656,33 +656,34 @@
#define ACPI_GPE_MAX 0xFF
#define ACPI_NUM_GPE 256
-/* Actions for acpi_gpe_wakeup, acpi_hw_low_set_gpe */
+/* Actions for acpi_set_gpe_wake_mask, acpi_hw_low_set_gpe */
#define ACPI_GPE_ENABLE 0
#define ACPI_GPE_DISABLE 1
-#define ACPI_GPE_COND_ENABLE 2
+#define ACPI_GPE_CONDITIONAL_ENABLE 2
/*
* GPE info flags - Per GPE
- * +-------+---+-+-+
- * | 7:4 |3:2|1|0|
- * +-------+---+-+-+
- * | | | |
- * | | | +--- Interrupt type: edge or level triggered
- * | | +----- GPE can wake the system
- * | +-------- Type of dispatch:to method, handler, or none
- * +-------------- <Reserved>
+ * +-------+-+-+---+
+ * | 7:4 |3|2|1:0|
+ * +-------+-+-+---+
+ * | | | |
+ * | | | +-- Type of dispatch:to method, handler, notify, or none
+ * | | +----- Interrupt type: edge or level triggered
+ * | +------- Is a Wake GPE
+ * +------------ <Reserved>
*/
-#define ACPI_GPE_XRUPT_TYPE_MASK (u8) 0x01
-#define ACPI_GPE_LEVEL_TRIGGERED (u8) 0x01
+#define ACPI_GPE_DISPATCH_NONE (u8) 0x00
+#define ACPI_GPE_DISPATCH_METHOD (u8) 0x01
+#define ACPI_GPE_DISPATCH_HANDLER (u8) 0x02
+#define ACPI_GPE_DISPATCH_NOTIFY (u8) 0x03
+#define ACPI_GPE_DISPATCH_MASK (u8) 0x03
+
+#define ACPI_GPE_LEVEL_TRIGGERED (u8) 0x04
#define ACPI_GPE_EDGE_TRIGGERED (u8) 0x00
+#define ACPI_GPE_XRUPT_TYPE_MASK (u8) 0x04
-#define ACPI_GPE_CAN_WAKE (u8) 0x02
-
-#define ACPI_GPE_DISPATCH_MASK (u8) 0x0C
-#define ACPI_GPE_DISPATCH_HANDLER (u8) 0x04
-#define ACPI_GPE_DISPATCH_METHOD (u8) 0x08
-#define ACPI_GPE_DISPATCH_NOT_USED (u8) 0x00
+#define ACPI_GPE_CAN_WAKE (u8) 0x08
/*
* Flags for GPE and Lock interfaces
@@ -894,9 +895,20 @@
/*
* Various handlers and callback procedures
*/
+typedef
+void (*ACPI_GBL_EVENT_HANDLER) (u32 event_type,
+ acpi_handle device,
+ u32 event_number, void *context);
+
+#define ACPI_EVENT_TYPE_GPE 0
+#define ACPI_EVENT_TYPE_FIXED 1
+
typedef u32(*acpi_event_handler) (void *context);
typedef
+u32 (*acpi_gpe_handler) (acpi_handle gpe_device, u32 gpe_number, void *context);
+
+typedef
void (*acpi_notify_handler) (acpi_handle device, u32 value, void *context);
typedef
@@ -951,6 +963,10 @@
#define ACPI_INTERRUPT_NOT_HANDLED 0x00
#define ACPI_INTERRUPT_HANDLED 0x01
+/* GPE handler return values */
+
+#define ACPI_REENABLE_GPE 0x80
+
/* Length of 32-bit EISAID values when converted back to a string */
#define ACPI_EISAID_STRING_SIZE 8 /* Includes null terminator */
diff --git a/include/acpi/apei.h b/include/acpi/apei.h
index b336502..c4dbb13 100644
--- a/include/acpi/apei.h
+++ b/include/acpi/apei.h
@@ -19,6 +19,12 @@
extern int hest_disable;
extern int erst_disable;
+#ifdef CONFIG_ACPI_APEI
+void __init acpi_hest_init(void);
+#else
+static inline void acpi_hest_init(void) { return; }
+#endif
+
typedef int (*apei_hest_func_t)(struct acpi_hest_header *hest_hdr, void *data);
int apei_hest_parse(apei_hest_func_t func, void *data);
diff --git a/include/acpi/processor.h b/include/acpi/processor.h
index 1b62102..55192ac 100644
--- a/include/acpi/processor.h
+++ b/include/acpi/processor.h
@@ -324,6 +324,12 @@
int acpi_processor_get_throttling_info(struct acpi_processor *pr);
extern int acpi_processor_set_throttling(struct acpi_processor *pr,
int state, bool force);
+/*
+ * Reevaluate whether the T-state is invalid after one cpu is
+ * onlined/offlined. In such case the flags.throttling will be updated.
+ */
+extern void acpi_processor_reevaluate_tstate(struct acpi_processor *pr,
+ unsigned long action);
extern const struct file_operations acpi_processor_throttling_fops;
extern void acpi_processor_throttling_init(void);
/* in processor_idle.c */
diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h
index 6098cae..ff5c660 100644
--- a/include/asm-generic/gpio.h
+++ b/include/asm-generic/gpio.h
@@ -147,11 +147,11 @@
/* Always use the library code for GPIO management calls,
* or when sleeping may be involved.
*/
-extern int __must_check gpio_request(unsigned gpio, const char *label);
+extern int gpio_request(unsigned gpio, const char *label);
extern void gpio_free(unsigned gpio);
-extern int __must_check gpio_direction_input(unsigned gpio);
-extern int __must_check gpio_direction_output(unsigned gpio, int value);
+extern int gpio_direction_input(unsigned gpio);
+extern int gpio_direction_output(unsigned gpio, int value);
extern int gpio_set_debounce(unsigned gpio, unsigned debounce);
@@ -192,8 +192,8 @@
const char *label;
};
-extern int __must_check gpio_request_one(unsigned gpio, unsigned long flags, const char *label);
-extern int __must_check gpio_request_array(struct gpio *array, size_t num);
+extern int gpio_request_one(unsigned gpio, unsigned long flags, const char *label);
+extern int gpio_request_array(struct gpio *array, size_t num);
extern void gpio_free_array(struct gpio *array, size_t num);
#ifdef CONFIG_GPIO_SYSFS
diff --git a/include/asm-generic/mman-common.h b/include/asm-generic/mman-common.h
index 3da9e27..787abbb 100644
--- a/include/asm-generic/mman-common.h
+++ b/include/asm-generic/mman-common.h
@@ -45,6 +45,9 @@
#define MADV_MERGEABLE 12 /* KSM may merge identical pages */
#define MADV_UNMERGEABLE 13 /* KSM may not merge identical pages */
+#define MADV_HUGEPAGE 14 /* Worth backing with hugepages */
+#define MADV_NOHUGEPAGE 15 /* Not worth backing with hugepages */
+
/* compatibility flags */
#define MAP_FILE 0
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 6f3c6ae..f1eddf7 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -5,67 +5,108 @@
#ifdef CONFIG_MMU
#ifndef __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
-/*
- * Largely same as above, but only sets the access flags (dirty,
- * accessed, and writable). Furthermore, we know it always gets set
- * to a "more permissive" setting, which allows most architectures
- * to optimize this. We return whether the PTE actually changed, which
- * in turn instructs the caller to do things like update__mmu_cache.
- * This used to be done in the caller, but sparc needs minor faults to
- * force that call on sun4c so we changed this macro slightly
- */
-#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
-({ \
- int __changed = !pte_same(*(__ptep), __entry); \
- if (__changed) { \
- set_pte_at((__vma)->vm_mm, (__address), __ptep, __entry); \
- flush_tlb_page(__vma, __address); \
- } \
- __changed; \
-})
+extern int ptep_set_access_flags(struct vm_area_struct *vma,
+ unsigned long address, pte_t *ptep,
+ pte_t entry, int dirty);
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS
+extern int pmdp_set_access_flags(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp,
+ pmd_t entry, int dirty);
#endif
#ifndef __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
-#define ptep_test_and_clear_young(__vma, __address, __ptep) \
-({ \
- pte_t __pte = *(__ptep); \
- int r = 1; \
- if (!pte_young(__pte)) \
- r = 0; \
- else \
- set_pte_at((__vma)->vm_mm, (__address), \
- (__ptep), pte_mkold(__pte)); \
- r; \
-})
+static inline int ptep_test_and_clear_young(struct vm_area_struct *vma,
+ unsigned long address,
+ pte_t *ptep)
+{
+ pte_t pte = *ptep;
+ int r = 1;
+ if (!pte_young(pte))
+ r = 0;
+ else
+ set_pte_at(vma->vm_mm, address, ptep, pte_mkold(pte));
+ return r;
+}
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_TEST_AND_CLEAR_YOUNG
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline int pmdp_test_and_clear_young(struct vm_area_struct *vma,
+ unsigned long address,
+ pmd_t *pmdp)
+{
+ pmd_t pmd = *pmdp;
+ int r = 1;
+ if (!pmd_young(pmd))
+ r = 0;
+ else
+ set_pmd_at(vma->vm_mm, address, pmdp, pmd_mkold(pmd));
+ return r;
+}
+#else /* CONFIG_TRANSPARENT_HUGEPAGE */
+static inline int pmdp_test_and_clear_young(struct vm_area_struct *vma,
+ unsigned long address,
+ pmd_t *pmdp)
+{
+ BUG();
+ return 0;
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
#endif
#ifndef __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
-#define ptep_clear_flush_young(__vma, __address, __ptep) \
-({ \
- int __young; \
- __young = ptep_test_and_clear_young(__vma, __address, __ptep); \
- if (__young) \
- flush_tlb_page(__vma, __address); \
- __young; \
-})
+int ptep_clear_flush_young(struct vm_area_struct *vma,
+ unsigned long address, pte_t *ptep);
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH
+int pmdp_clear_flush_young(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp);
#endif
#ifndef __HAVE_ARCH_PTEP_GET_AND_CLEAR
-#define ptep_get_and_clear(__mm, __address, __ptep) \
-({ \
- pte_t __pte = *(__ptep); \
- pte_clear((__mm), (__address), (__ptep)); \
- __pte; \
+static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
+ unsigned long address,
+ pte_t *ptep)
+{
+ pte_t pte = *ptep;
+ pte_clear(mm, address, ptep);
+ return pte;
+}
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_GET_AND_CLEAR
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm,
+ unsigned long address,
+ pmd_t *pmdp)
+{
+ pmd_t pmd = *pmdp;
+ pmd_clear(mm, address, pmdp);
+ return pmd;
})
+#else /* CONFIG_TRANSPARENT_HUGEPAGE */
+static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm,
+ unsigned long address,
+ pmd_t *pmdp)
+{
+ BUG();
+ return __pmd(0);
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
#endif
#ifndef __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
-#define ptep_get_and_clear_full(__mm, __address, __ptep, __full) \
-({ \
- pte_t __pte; \
- __pte = ptep_get_and_clear((__mm), (__address), (__ptep)); \
- __pte; \
-})
+static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm,
+ unsigned long address, pte_t *ptep,
+ int full)
+{
+ pte_t pte;
+ pte = ptep_get_and_clear(mm, address, ptep);
+ return pte;
+}
#endif
/*
@@ -74,20 +115,25 @@
* not present, or in the process of an address space destruction.
*/
#ifndef __HAVE_ARCH_PTE_CLEAR_NOT_PRESENT_FULL
-#define pte_clear_not_present_full(__mm, __address, __ptep, __full) \
-do { \
- pte_clear((__mm), (__address), (__ptep)); \
-} while (0)
+static inline void pte_clear_not_present_full(struct mm_struct *mm,
+ unsigned long address,
+ pte_t *ptep,
+ int full)
+{
+ pte_clear(mm, address, ptep);
+}
#endif
#ifndef __HAVE_ARCH_PTEP_CLEAR_FLUSH
-#define ptep_clear_flush(__vma, __address, __ptep) \
-({ \
- pte_t __pte; \
- __pte = ptep_get_and_clear((__vma)->vm_mm, __address, __ptep); \
- flush_tlb_page(__vma, __address); \
- __pte; \
-})
+extern pte_t ptep_clear_flush(struct vm_area_struct *vma,
+ unsigned long address,
+ pte_t *ptep);
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_CLEAR_FLUSH
+extern pmd_t pmdp_clear_flush(struct vm_area_struct *vma,
+ unsigned long address,
+ pmd_t *pmdp);
#endif
#ifndef __HAVE_ARCH_PTEP_SET_WRPROTECT
@@ -99,8 +145,49 @@
}
#endif
+#ifndef __HAVE_ARCH_PMDP_SET_WRPROTECT
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline void pmdp_set_wrprotect(struct mm_struct *mm,
+ unsigned long address, pmd_t *pmdp)
+{
+ pmd_t old_pmd = *pmdp;
+ set_pmd_at(mm, address, pmdp, pmd_wrprotect(old_pmd));
+}
+#else /* CONFIG_TRANSPARENT_HUGEPAGE */
+static inline void pmdp_set_wrprotect(struct mm_struct *mm,
+ unsigned long address, pmd_t *pmdp)
+{
+ BUG();
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_SPLITTING_FLUSH
+extern pmd_t pmdp_clear_flush(struct vm_area_struct *vma,
+ unsigned long address,
+ pmd_t *pmdp);
+#endif
+
#ifndef __HAVE_ARCH_PTE_SAME
-#define pte_same(A,B) (pte_val(A) == pte_val(B))
+static inline int pte_same(pte_t pte_a, pte_t pte_b)
+{
+ return pte_val(pte_a) == pte_val(pte_b);
+}
+#endif
+
+#ifndef __HAVE_ARCH_PMD_SAME
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline int pmd_same(pmd_t pmd_a, pmd_t pmd_b)
+{
+ return pmd_val(pmd_a) == pmd_val(pmd_b);
+}
+#else /* CONFIG_TRANSPARENT_HUGEPAGE */
+static inline int pmd_same(pmd_t pmd_a, pmd_t pmd_b)
+{
+ BUG();
+ return 0;
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
#endif
#ifndef __HAVE_ARCH_PAGE_TEST_DIRTY
@@ -348,6 +435,24 @@
unsigned long size);
#endif
+#ifndef CONFIG_TRANSPARENT_HUGEPAGE
+static inline int pmd_trans_huge(pmd_t pmd)
+{
+ return 0;
+}
+static inline int pmd_trans_splitting(pmd_t pmd)
+{
+ return 0;
+}
+#ifndef __HAVE_ARCH_PMD_WRITE
+static inline int pmd_write(pmd_t pmd)
+{
+ BUG();
+ return 0;
+}
+#endif /* __HAVE_ARCH_PMD_WRITE */
+#endif
+
#endif /* !__ASSEMBLY__ */
#endif /* _ASM_GENERIC_PGTABLE_H */
diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index aac27bd..f22e7fe 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -121,6 +121,9 @@
void drm_fb_helper_restore(void);
void drm_fb_helper_fill_var(struct fb_info *info, struct drm_fb_helper *fb_helper,
uint32_t fb_width, uint32_t fb_height);
+void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
+ uint32_t depth);
+
int drm_fb_helper_setcmap(struct fb_cmap *cmap, struct fb_info *info);
bool drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper);
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 67c91b4..eb176bb 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -352,4 +352,14 @@
return -1;
}
#endif /* !CONFIG_ACPI */
+
+#ifdef CONFIG_ACPI_SLEEP
+int suspend_nvs_register(unsigned long start, unsigned long size);
+#else
+static inline int suspend_nvs_register(unsigned long a, unsigned long b)
+{
+ return 0;
+}
+#endif
+
#endif /*_LINUX_ACPI_H*/
diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h
index 8b49ac4..e02982f 100644
--- a/include/linux/auto_fs4.h
+++ b/include/linux/auto_fs4.h
@@ -24,7 +24,7 @@
#define AUTOFS_MIN_PROTO_VERSION 3
#define AUTOFS_MAX_PROTO_VERSION 5
-#define AUTOFS_PROTO_SUBVERSION 1
+#define AUTOFS_PROTO_SUBVERSION 2
/* Mask for expire behaviour */
#define AUTOFS_EXP_IMMEDIATE 1
diff --git a/include/linux/compaction.h b/include/linux/compaction.h
index 5ac5155..dfa2ed4 100644
--- a/include/linux/compaction.h
+++ b/include/linux/compaction.h
@@ -11,6 +11,9 @@
/* The full zone was compacted */
#define COMPACT_COMPLETE 3
+#define COMPACT_MODE_DIRECT_RECLAIM 0
+#define COMPACT_MODE_KSWAPD 1
+
#ifdef CONFIG_COMPACTION
extern int sysctl_compact_memory;
extern int sysctl_compaction_handler(struct ctl_table *table, int write,
@@ -21,7 +24,12 @@
extern int fragmentation_index(struct zone *zone, unsigned int order);
extern unsigned long try_to_compact_pages(struct zonelist *zonelist,
- int order, gfp_t gfp_mask, nodemask_t *mask);
+ int order, gfp_t gfp_mask, nodemask_t *mask,
+ bool sync);
+extern unsigned long compaction_suitable(struct zone *zone, int order);
+extern unsigned long compact_zone_order(struct zone *zone, int order,
+ gfp_t gfp_mask, bool sync,
+ int compact_mode);
/* Do not skip compaction more than 64 times */
#define COMPACT_MAX_DEFER_SHIFT 6
@@ -54,7 +62,20 @@
#else
static inline unsigned long try_to_compact_pages(struct zonelist *zonelist,
- int order, gfp_t gfp_mask, nodemask_t *nodemask)
+ int order, gfp_t gfp_mask, nodemask_t *nodemask,
+ bool sync)
+{
+ return COMPACT_CONTINUE;
+}
+
+static inline unsigned long compaction_suitable(struct zone *zone, int order)
+{
+ return COMPACT_SKIPPED;
+}
+
+static inline unsigned long compact_zone_order(struct zone *zone, int order,
+ gfp_t gfp_mask, bool sync,
+ int compact_mode)
{
return COMPACT_CONTINUE;
}
diff --git a/include/linux/cper.h b/include/linux/cper.h
index bf972f8..3104aaf 100644
--- a/include/linux/cper.h
+++ b/include/linux/cper.h
@@ -39,10 +39,12 @@
* Severity difinition for error_severity in struct cper_record_header
* and section_severity in struct cper_section_descriptor
*/
-#define CPER_SEV_RECOVERABLE 0x0
-#define CPER_SEV_FATAL 0x1
-#define CPER_SEV_CORRECTED 0x2
-#define CPER_SEV_INFORMATIONAL 0x3
+enum {
+ CPER_SEV_RECOVERABLE,
+ CPER_SEV_FATAL,
+ CPER_SEV_CORRECTED,
+ CPER_SEV_INFORMATIONAL,
+};
/*
* Validation bits difinition for validation_bits in struct
@@ -201,6 +203,47 @@
UUID_LE(0x036F84E1, 0x7F37, 0x428c, 0xA7, 0x9E, 0x57, 0x5F, \
0xDF, 0xAA, 0x84, 0xEC)
+#define CPER_PROC_VALID_TYPE 0x0001
+#define CPER_PROC_VALID_ISA 0x0002
+#define CPER_PROC_VALID_ERROR_TYPE 0x0004
+#define CPER_PROC_VALID_OPERATION 0x0008
+#define CPER_PROC_VALID_FLAGS 0x0010
+#define CPER_PROC_VALID_LEVEL 0x0020
+#define CPER_PROC_VALID_VERSION 0x0040
+#define CPER_PROC_VALID_BRAND_INFO 0x0080
+#define CPER_PROC_VALID_ID 0x0100
+#define CPER_PROC_VALID_TARGET_ADDRESS 0x0200
+#define CPER_PROC_VALID_REQUESTOR_ID 0x0400
+#define CPER_PROC_VALID_RESPONDER_ID 0x0800
+#define CPER_PROC_VALID_IP 0x1000
+
+#define CPER_MEM_VALID_ERROR_STATUS 0x0001
+#define CPER_MEM_VALID_PHYSICAL_ADDRESS 0x0002
+#define CPER_MEM_VALID_PHYSICAL_ADDRESS_MASK 0x0004
+#define CPER_MEM_VALID_NODE 0x0008
+#define CPER_MEM_VALID_CARD 0x0010
+#define CPER_MEM_VALID_MODULE 0x0020
+#define CPER_MEM_VALID_BANK 0x0040
+#define CPER_MEM_VALID_DEVICE 0x0080
+#define CPER_MEM_VALID_ROW 0x0100
+#define CPER_MEM_VALID_COLUMN 0x0200
+#define CPER_MEM_VALID_BIT_POSITION 0x0400
+#define CPER_MEM_VALID_REQUESTOR_ID 0x0800
+#define CPER_MEM_VALID_RESPONDER_ID 0x1000
+#define CPER_MEM_VALID_TARGET_ID 0x2000
+#define CPER_MEM_VALID_ERROR_TYPE 0x4000
+
+#define CPER_PCIE_VALID_PORT_TYPE 0x0001
+#define CPER_PCIE_VALID_VERSION 0x0002
+#define CPER_PCIE_VALID_COMMAND_STATUS 0x0004
+#define CPER_PCIE_VALID_DEVICE_ID 0x0008
+#define CPER_PCIE_VALID_SERIAL_NUMBER 0x0010
+#define CPER_PCIE_VALID_BRIDGE_CONTROL_STATUS 0x0020
+#define CPER_PCIE_VALID_CAPABILITY 0x0040
+#define CPER_PCIE_VALID_AER_INFO 0x0080
+
+#define CPER_PCIE_SLOT_SHIFT 3
+
/*
* All tables and structs must be byte-packed to match CPER
* specification, since the tables are provided by the system BIOS
@@ -306,6 +349,41 @@
__u8 error_type;
};
+struct cper_sec_pcie {
+ __u64 validation_bits;
+ __u32 port_type;
+ struct {
+ __u8 minor;
+ __u8 major;
+ __u8 reserved[2];
+ } version;
+ __u16 command;
+ __u16 status;
+ __u32 reserved;
+ struct {
+ __u16 vendor_id;
+ __u16 device_id;
+ __u8 class_code[3];
+ __u8 function;
+ __u8 device;
+ __u16 segment;
+ __u8 bus;
+ __u8 secondary_bus;
+ __u16 slot;
+ __u8 reserved;
+ } device_id;
+ struct {
+ __u32 lower;
+ __u32 upper;
+ } serial_number;
+ struct {
+ __u16 secondary_status;
+ __u16 control;
+ } bridge;
+ __u8 capability[60];
+ __u8 aer_info[96];
+};
+
/* Reset to default packing */
#pragma pack()
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index 1be416b..36719ea 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -47,13 +47,7 @@
/* Idle State Flags */
#define CPUIDLE_FLAG_TIME_VALID (0x01) /* is residency time measurable? */
-#define CPUIDLE_FLAG_CHECK_BM (0x02) /* BM activity will exit state */
-#define CPUIDLE_FLAG_POLL (0x10) /* no latency, no savings */
-#define CPUIDLE_FLAG_SHALLOW (0x20) /* low latency, minimal savings */
-#define CPUIDLE_FLAG_BALANCED (0x40) /* medium latency, moderate savings */
-#define CPUIDLE_FLAG_DEEP (0x80) /* high latency, large savings */
#define CPUIDLE_FLAG_IGNORE (0x100) /* ignore during this idle period */
-#define CPUIDLE_FLAG_TLB_FLUSHED (0x200) /* tlb will be flushed */
#define CPUIDLE_DRIVER_FLAGS_MASK (0xFFFF0000)
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index 59fcd24..f958c19 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -167,6 +167,8 @@
void (*d_release)(struct dentry *);
void (*d_iput)(struct dentry *, struct inode *);
char *(*d_dname)(struct dentry *, char *, int);
+ struct vfsmount *(*d_automount)(struct path *);
+ int (*d_manage)(struct dentry *, bool, bool);
} ____cacheline_aligned;
/*
@@ -205,13 +207,18 @@
#define DCACHE_CANT_MOUNT 0x0100
#define DCACHE_GENOCIDE 0x0200
-#define DCACHE_MOUNTED 0x0400 /* is a mountpoint */
#define DCACHE_OP_HASH 0x1000
#define DCACHE_OP_COMPARE 0x2000
#define DCACHE_OP_REVALIDATE 0x4000
#define DCACHE_OP_DELETE 0x8000
+#define DCACHE_MOUNTED 0x10000 /* is a mountpoint */
+#define DCACHE_NEED_AUTOMOUNT 0x20000 /* handle automount on this dir */
+#define DCACHE_MANAGE_TRANSIT 0x40000 /* manage transit from this dirent */
+#define DCACHE_MANAGED_DENTRY \
+ (DCACHE_MOUNTED|DCACHE_NEED_AUTOMOUNT|DCACHE_MANAGE_TRANSIT)
+
extern seqlock_t rename_lock;
static inline int dname_external(struct dentry *dentry)
@@ -399,7 +406,12 @@
extern void dput(struct dentry *);
-static inline int d_mountpoint(struct dentry *dentry)
+static inline bool d_managed(struct dentry *dentry)
+{
+ return dentry->d_flags & DCACHE_MANAGED_DENTRY;
+}
+
+static inline bool d_mountpoint(struct dentry *dentry)
{
return dentry->d_flags & DCACHE_MOUNTED;
}
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 2970022..272496d 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -193,6 +193,13 @@
char *error;
};
+/* Each target can link one of these into the table */
+struct dm_target_callbacks {
+ struct list_head list;
+ int (*congested_fn) (struct dm_target_callbacks *, int);
+ void (*unplug_fn)(struct dm_target_callbacks *);
+};
+
int dm_register_target(struct target_type *t);
void dm_unregister_target(struct target_type *t);
@@ -269,6 +276,11 @@
sector_t start, sector_t len, char *params);
/*
+ * Target_ctr should call this if it needs to add any callbacks.
+ */
+void dm_table_add_target_callbacks(struct dm_table *t, struct dm_target_callbacks *cb);
+
+/*
* Finally call this to make the table ready for use.
*/
int dm_table_complete(struct dm_table *t);
diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h
index 49eab36..78bbf47 100644
--- a/include/linux/dm-ioctl.h
+++ b/include/linux/dm-ioctl.h
@@ -44,7 +44,7 @@
* Remove a device, destroy any tables.
*
* DM_DEV_RENAME:
- * Rename a device.
+ * Rename a device or set its uuid if none was previously supplied.
*
* DM_SUSPEND:
* This performs both suspend and resume, depending which flag is
@@ -267,9 +267,9 @@
#define DM_DEV_SET_GEOMETRY _IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl)
#define DM_VERSION_MAJOR 4
-#define DM_VERSION_MINOR 18
-#define DM_VERSION_PATCHLEVEL 0
-#define DM_VERSION_EXTRA "-ioctl (2010-06-29)"
+#define DM_VERSION_MINOR 19
+#define DM_VERSION_PATCHLEVEL 1
+#define DM_VERSION_EXTRA "-ioctl (2011-01-07)"
/* Status bits */
#define DM_READONLY_FLAG (1 << 0) /* In/Out */
@@ -322,4 +322,10 @@
*/
#define DM_UEVENT_GENERATED_FLAG (1 << 13) /* Out */
+/*
+ * If set, rename changes the uuid not the name. Only permitted
+ * if no uuid was previously supplied: an existing uuid cannot be changed.
+ */
+#define DM_UUID_FLAG (1 << 14) /* In */
+
#endif /* _LINUX_DM_IOCTL_H */
diff --git a/include/linux/dm-log-userspace.h b/include/linux/dm-log-userspace.h
index 0c3c3a2..eeace7d 100644
--- a/include/linux/dm-log-userspace.h
+++ b/include/linux/dm-log-userspace.h
@@ -370,6 +370,16 @@
#define DM_ULOG_REQUEST_TYPE(request_type) \
(DM_ULOG_REQUEST_MASK & (request_type))
+/*
+ * DM_ULOG_REQUEST_VERSION is incremented when there is a
+ * change to the way information is passed between kernel
+ * and userspace. This could be a structure change of
+ * dm_ulog_request or a change in the way requests are
+ * issued/handled. Changes are outlined here:
+ * version 1: Initial implementation
+ */
+#define DM_ULOG_REQUEST_VERSION 1
+
struct dm_ulog_request {
/*
* The local unique identifier (luid) and the universally unique
@@ -383,8 +393,9 @@
*/
uint64_t luid;
char uuid[DM_UUID_LEN];
- char padding[7]; /* Padding because DM_UUID_LEN = 129 */
+ char padding[3]; /* Padding because DM_UUID_LEN = 129 */
+ uint32_t version; /* See DM_ULOG_REQUEST_VERSION */
int32_t error; /* Used to report back processing errors */
uint32_t seq; /* Sequence number for request */
diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h
index bec8b82..ab68f78 100644
--- a/include/linux/etherdevice.h
+++ b/include/linux/etherdevice.h
@@ -99,6 +99,17 @@
}
/**
+ * is_unicast_ether_addr - Determine if the Ethernet address is unicast
+ * @addr: Pointer to a six-byte array containing the Ethernet address
+ *
+ * Return true if the address is a unicast address.
+ */
+static inline int is_unicast_ether_addr(const u8 *addr)
+{
+ return !is_multicast_ether_addr(addr);
+}
+
+/**
* is_valid_ether_addr - Determine if the given Ethernet address is valid
* @addr: Pointer to a six-byte array containing the Ethernet address
*
diff --git a/include/linux/fcntl.h b/include/linux/fcntl.h
index afc00af..a562fa5 100644
--- a/include/linux/fcntl.h
+++ b/include/linux/fcntl.h
@@ -45,6 +45,7 @@
#define AT_REMOVEDIR 0x200 /* Remove directory instead of
unlinking file. */
#define AT_SYMLINK_FOLLOW 0x400 /* Follow symbolic links. */
+#define AT_NO_AUTOMOUNT 0x800 /* Suppress terminal automount traversal */
#ifdef __KERNEL__
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 3984f23..177b4dde 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -242,6 +242,7 @@
#define S_SWAPFILE 256 /* Do not truncate: swapon got its bmaps */
#define S_PRIVATE 512 /* Inode is fs-internal */
#define S_IMA 1024 /* Inode has an associated IMA struct */
+#define S_AUTOMOUNT 2048 /* Automount/referral quasi-directory */
/*
* Note that nosuid etc flags are inode-specific: setting some file-system
@@ -277,6 +278,7 @@
#define IS_SWAPFILE(inode) ((inode)->i_flags & S_SWAPFILE)
#define IS_PRIVATE(inode) ((inode)->i_flags & S_PRIVATE)
#define IS_IMA(inode) ((inode)->i_flags & S_IMA)
+#define IS_AUTOMOUNT(inode) ((inode)->i_flags & S_AUTOMOUNT)
/* the read-only stuff doesn't really belong here, but any other place is
probably as bad and I don't want to create yet another include file. */
@@ -666,7 +668,7 @@
int bd_holders;
bool bd_write_holder;
#ifdef CONFIG_SYSFS
- struct gendisk * bd_holder_disk; /* for sysfs slave linkng */
+ struct list_head bd_holder_disks;
#endif
struct block_device * bd_contains;
unsigned bd_block_size;
@@ -1066,7 +1068,6 @@
int (*fl_grant)(struct file_lock *, struct file_lock *, int);
void (*fl_release_private)(struct file_lock *);
void (*fl_break)(struct file_lock *);
- int (*fl_mylease)(struct file_lock *, struct file_lock *);
int (*fl_change)(struct file_lock **, int);
};
@@ -2058,12 +2059,18 @@
extern int blkdev_put(struct block_device *bdev, fmode_t mode);
#ifdef CONFIG_SYSFS
extern int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk);
+extern void bd_unlink_disk_holder(struct block_device *bdev,
+ struct gendisk *disk);
#else
static inline int bd_link_disk_holder(struct block_device *bdev,
struct gendisk *disk)
{
return 0;
}
+static inline void bd_unlink_disk_holder(struct block_device *bdev,
+ struct gendisk *disk)
+{
+}
#endif
#endif
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index f54adfc..a3b148a 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -34,6 +34,7 @@
#else
#define ___GFP_NOTRACK 0
#endif
+#define ___GFP_NO_KSWAPD 0x400000u
/*
* GFP bitmasks..
@@ -81,13 +82,15 @@
#define __GFP_RECLAIMABLE ((__force gfp_t)___GFP_RECLAIMABLE) /* Page is reclaimable */
#define __GFP_NOTRACK ((__force gfp_t)___GFP_NOTRACK) /* Don't track with kmemcheck */
+#define __GFP_NO_KSWAPD ((__force gfp_t)___GFP_NO_KSWAPD)
+
/*
* This may seem redundant, but it's a way of annotating false positives vs.
* allocations that simply cannot be supported (e.g. page tables).
*/
#define __GFP_NOTRACK_FALSE_POSITIVE (__GFP_NOTRACK)
-#define __GFP_BITS_SHIFT 22 /* Room for 22 __GFP_FOO bits */
+#define __GFP_BITS_SHIFT 23 /* Room for 23 __GFP_FOO bits */
#define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
/* This equals 0, but use constants in case they ever change */
@@ -106,6 +109,9 @@
__GFP_HARDWALL | __GFP_HIGHMEM | \
__GFP_MOVABLE)
#define GFP_IOFS (__GFP_IO | __GFP_FS)
+#define GFP_TRANSHUGE (GFP_HIGHUSER_MOVABLE | __GFP_COMP | \
+ __GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN | \
+ __GFP_NO_KSWAPD)
#ifdef CONFIG_NUMA
#define GFP_THISNODE (__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
@@ -325,14 +331,17 @@
{
return alloc_pages_current(gfp_mask, order);
}
-extern struct page *alloc_page_vma(gfp_t gfp_mask,
+extern struct page *alloc_pages_vma(gfp_t gfp_mask, int order,
struct vm_area_struct *vma, unsigned long addr);
#else
#define alloc_pages(gfp_mask, order) \
alloc_pages_node(numa_node_id(), gfp_mask, order)
-#define alloc_page_vma(gfp_mask, vma, addr) alloc_pages(gfp_mask, 0)
+#define alloc_pages_vma(gfp_mask, order, vma, addr) \
+ alloc_pages(gfp_mask, order)
#endif
#define alloc_page(gfp_mask) alloc_pages(gfp_mask, 0)
+#define alloc_page_vma(gfp_mask, vma, addr) \
+ alloc_pages_vma(gfp_mask, 0, vma, addr)
extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order);
extern unsigned long get_zeroed_page(gfp_t gfp_mask);
diff --git a/include/linux/gpio.h b/include/linux/gpio.h
index f79d67f..32720ba 100644
--- a/include/linux/gpio.h
+++ b/include/linux/gpio.h
@@ -30,18 +30,18 @@
return 0;
}
-static inline int __must_check gpio_request(unsigned gpio, const char *label)
+static inline int gpio_request(unsigned gpio, const char *label)
{
return -ENOSYS;
}
-static inline int __must_check gpio_request_one(unsigned gpio,
+static inline int gpio_request_one(unsigned gpio,
unsigned long flags, const char *label)
{
return -ENOSYS;
}
-static inline int __must_check gpio_request_array(struct gpio *array, size_t num)
+static inline int gpio_request_array(struct gpio *array, size_t num)
{
return -ENOSYS;
}
@@ -62,12 +62,12 @@
WARN_ON(1);
}
-static inline int __must_check gpio_direction_input(unsigned gpio)
+static inline int gpio_direction_input(unsigned gpio)
{
return -ENOSYS;
}
-static inline int __must_check gpio_direction_output(unsigned gpio, int value)
+static inline int gpio_direction_output(unsigned gpio, int value)
{
return -ENOSYS;
}
diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
new file mode 100644
index 0000000..8e6c8c4
--- /dev/null
+++ b/include/linux/huge_mm.h
@@ -0,0 +1,179 @@
+#ifndef _LINUX_HUGE_MM_H
+#define _LINUX_HUGE_MM_H
+
+extern int do_huge_pmd_anonymous_page(struct mm_struct *mm,
+ struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmd,
+ unsigned int flags);
+extern int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
+ pmd_t *dst_pmd, pmd_t *src_pmd, unsigned long addr,
+ struct vm_area_struct *vma);
+extern int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmd,
+ pmd_t orig_pmd);
+extern pgtable_t get_pmd_huge_pte(struct mm_struct *mm);
+extern struct page *follow_trans_huge_pmd(struct mm_struct *mm,
+ unsigned long addr,
+ pmd_t *pmd,
+ unsigned int flags);
+extern int zap_huge_pmd(struct mmu_gather *tlb,
+ struct vm_area_struct *vma,
+ pmd_t *pmd);
+extern int mincore_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
+ unsigned long addr, unsigned long end,
+ unsigned char *vec);
+extern int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
+ unsigned long addr, pgprot_t newprot);
+
+enum transparent_hugepage_flag {
+ TRANSPARENT_HUGEPAGE_FLAG,
+ TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG,
+ TRANSPARENT_HUGEPAGE_DEFRAG_FLAG,
+ TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG,
+ TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG,
+#ifdef CONFIG_DEBUG_VM
+ TRANSPARENT_HUGEPAGE_DEBUG_COW_FLAG,
+#endif
+};
+
+enum page_check_address_pmd_flag {
+ PAGE_CHECK_ADDRESS_PMD_FLAG,
+ PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG,
+ PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG,
+};
+extern pmd_t *page_check_address_pmd(struct page *page,
+ struct mm_struct *mm,
+ unsigned long address,
+ enum page_check_address_pmd_flag flag);
+
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+#define HPAGE_PMD_SHIFT HPAGE_SHIFT
+#define HPAGE_PMD_MASK HPAGE_MASK
+#define HPAGE_PMD_SIZE HPAGE_SIZE
+
+#define transparent_hugepage_enabled(__vma) \
+ ((transparent_hugepage_flags & \
+ (1<<TRANSPARENT_HUGEPAGE_FLAG) || \
+ (transparent_hugepage_flags & \
+ (1<<TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG) && \
+ ((__vma)->vm_flags & VM_HUGEPAGE))) && \
+ !((__vma)->vm_flags & VM_NOHUGEPAGE))
+#define transparent_hugepage_defrag(__vma) \
+ ((transparent_hugepage_flags & \
+ (1<<TRANSPARENT_HUGEPAGE_DEFRAG_FLAG)) || \
+ (transparent_hugepage_flags & \
+ (1<<TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG) && \
+ (__vma)->vm_flags & VM_HUGEPAGE))
+#ifdef CONFIG_DEBUG_VM
+#define transparent_hugepage_debug_cow() \
+ (transparent_hugepage_flags & \
+ (1<<TRANSPARENT_HUGEPAGE_DEBUG_COW_FLAG))
+#else /* CONFIG_DEBUG_VM */
+#define transparent_hugepage_debug_cow() 0
+#endif /* CONFIG_DEBUG_VM */
+
+extern unsigned long transparent_hugepage_flags;
+extern int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
+ pmd_t *dst_pmd, pmd_t *src_pmd,
+ struct vm_area_struct *vma,
+ unsigned long addr, unsigned long end);
+extern int handle_pte_fault(struct mm_struct *mm,
+ struct vm_area_struct *vma, unsigned long address,
+ pte_t *pte, pmd_t *pmd, unsigned int flags);
+extern int split_huge_page(struct page *page);
+extern void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd);
+#define split_huge_page_pmd(__mm, __pmd) \
+ do { \
+ pmd_t *____pmd = (__pmd); \
+ if (unlikely(pmd_trans_huge(*____pmd))) \
+ __split_huge_page_pmd(__mm, ____pmd); \
+ } while (0)
+#define wait_split_huge_page(__anon_vma, __pmd) \
+ do { \
+ pmd_t *____pmd = (__pmd); \
+ spin_unlock_wait(&(__anon_vma)->root->lock); \
+ /* \
+ * spin_unlock_wait() is just a loop in C and so the \
+ * CPU can reorder anything around it. \
+ */ \
+ smp_mb(); \
+ BUG_ON(pmd_trans_splitting(*____pmd) || \
+ pmd_trans_huge(*____pmd)); \
+ } while (0)
+#define HPAGE_PMD_ORDER (HPAGE_PMD_SHIFT-PAGE_SHIFT)
+#define HPAGE_PMD_NR (1<<HPAGE_PMD_ORDER)
+#if HPAGE_PMD_ORDER > MAX_ORDER
+#error "hugepages can't be allocated by the buddy allocator"
+#endif
+extern int hugepage_madvise(struct vm_area_struct *vma,
+ unsigned long *vm_flags, int advice);
+extern void __vma_adjust_trans_huge(struct vm_area_struct *vma,
+ unsigned long start,
+ unsigned long end,
+ long adjust_next);
+static inline void vma_adjust_trans_huge(struct vm_area_struct *vma,
+ unsigned long start,
+ unsigned long end,
+ long adjust_next)
+{
+ if (!vma->anon_vma || vma->vm_ops || vma->vm_file)
+ return;
+ __vma_adjust_trans_huge(vma, start, end, adjust_next);
+}
+static inline int hpage_nr_pages(struct page *page)
+{
+ if (unlikely(PageTransHuge(page)))
+ return HPAGE_PMD_NR;
+ return 1;
+}
+static inline struct page *compound_trans_head(struct page *page)
+{
+ if (PageTail(page)) {
+ struct page *head;
+ head = page->first_page;
+ smp_rmb();
+ /*
+ * head may be a dangling pointer.
+ * __split_huge_page_refcount clears PageTail before
+ * overwriting first_page, so if PageTail is still
+ * there it means the head pointer isn't dangling.
+ */
+ if (PageTail(page))
+ return head;
+ }
+ return page;
+}
+#else /* CONFIG_TRANSPARENT_HUGEPAGE */
+#define HPAGE_PMD_SHIFT ({ BUG(); 0; })
+#define HPAGE_PMD_MASK ({ BUG(); 0; })
+#define HPAGE_PMD_SIZE ({ BUG(); 0; })
+
+#define hpage_nr_pages(x) 1
+
+#define transparent_hugepage_enabled(__vma) 0
+
+#define transparent_hugepage_flags 0UL
+static inline int split_huge_page(struct page *page)
+{
+ return 0;
+}
+#define split_huge_page_pmd(__mm, __pmd) \
+ do { } while (0)
+#define wait_split_huge_page(__anon_vma, __pmd) \
+ do { } while (0)
+#define compound_trans_head(page) compound_head(page)
+static inline int hugepage_madvise(struct vm_area_struct *vma,
+ unsigned long *vm_flags, int advice)
+{
+ BUG();
+ return 0;
+}
+static inline void vma_adjust_trans_huge(struct vm_area_struct *vma,
+ unsigned long start,
+ unsigned long end,
+ long adjust_next)
+{
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+
+#endif /* _LINUX_HUGE_MM_H */
diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h
index 65aae34..045f2f2 100644
--- a/include/linux/ipmi.h
+++ b/include/linux/ipmi.h
@@ -454,6 +454,44 @@
/* Validate that the given IPMI address is valid. */
int ipmi_validate_addr(struct ipmi_addr *addr, int len);
+/*
+ * How did the IPMI driver find out about the device?
+ */
+enum ipmi_addr_src {
+ SI_INVALID = 0, SI_HOTMOD, SI_HARDCODED, SI_SPMI, SI_ACPI, SI_SMBIOS,
+ SI_PCI, SI_DEVICETREE, SI_DEFAULT
+};
+
+union ipmi_smi_info_union {
+ /*
+ * the acpi_info element is defined for the SI_ACPI
+ * address type
+ */
+ struct {
+ void *acpi_handle;
+ } acpi_info;
+};
+
+struct ipmi_smi_info {
+ enum ipmi_addr_src addr_src;
+
+ /*
+ * Base device for the interface. Don't forget to put this when
+ * you are done.
+ */
+ struct device *dev;
+
+ /*
+ * The addr_info provides more detailed info for some IPMI
+ * devices, depending on the addr_src. Currently only SI_ACPI
+ * info is provided.
+ */
+ union ipmi_smi_info_union addr_info;
+};
+
+/* This is to get the private info of ipmi_smi_t */
+extern int ipmi_get_smi_info(int if_num, struct ipmi_smi_info *data);
+
#endif /* __KERNEL__ */
diff --git a/include/linux/ipmi_smi.h b/include/linux/ipmi_smi.h
index 4b48318..906590a 100644
--- a/include/linux/ipmi_smi.h
+++ b/include/linux/ipmi_smi.h
@@ -39,6 +39,7 @@
#include <linux/module.h>
#include <linux/device.h>
#include <linux/platform_device.h>
+#include <linux/ipmi.h>
/* This files describes the interface for IPMI system management interface
drivers to bind into the IPMI message handler. */
@@ -86,6 +87,13 @@
int (*start_processing)(void *send_info,
ipmi_smi_t new_intf);
+ /*
+ * Get the detailed private info of the low level interface and store
+ * it into the structure of ipmi_smi_data. For example: the
+ * ACPI device handle will be returned for the pnp_acpi IPMI device.
+ */
+ int (*get_smi_info)(void *send_info, struct ipmi_smi_info *data);
+
/* Called to enqueue an SMI message to be sent. This
operation is not allowed to fail. If an error occurs, it
should report back the error in a received message. It may
diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h
index 979c68c..6a64c6f 100644
--- a/include/linux/irqdesc.h
+++ b/include/linux/irqdesc.h
@@ -57,7 +57,7 @@
#endif
struct timer_rand_state *timer_rand_state;
- unsigned int *kstat_irqs;
+ unsigned int __percpu *kstat_irqs;
irq_flow_handler_t handle_irq;
struct irqaction *action; /* IRQ action list */
unsigned int status; /* IRQ status */
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 57dac70..5a9d905 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -600,6 +600,13 @@
#define NUMA_BUILD 0
#endif
+/* This helps us avoid #ifdef CONFIG_COMPACTION */
+#ifdef CONFIG_COMPACTION
+#define COMPACTION_BUILD 1
+#else
+#define COMPACTION_BUILD 0
+#endif
+
/* Rebuild everything on CONFIG_FTRACE_MCOUNT_RECORD */
#ifdef CONFIG_FTRACE_MCOUNT_RECORD
# define REBUILD_DUE_TO_FTRACE_MCOUNT_RECORD
diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h
index 44e83ba..0cce2db 100644
--- a/include/linux/kernel_stat.h
+++ b/include/linux/kernel_stat.h
@@ -46,16 +46,14 @@
extern unsigned long long nr_context_switches(void);
#ifndef CONFIG_GENERIC_HARDIRQS
-#define kstat_irqs_this_cpu(irq) \
- (this_cpu_read(kstat.irqs[irq])
struct irq_desc;
static inline void kstat_incr_irqs_this_cpu(unsigned int irq,
struct irq_desc *desc)
{
- kstat_this_cpu.irqs[irq]++;
- kstat_this_cpu.irqs_sum++;
+ __this_cpu_inc(kstat.irqs[irq]);
+ __this_cpu_inc(kstat.irqs_sum);
}
static inline unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
@@ -65,17 +63,18 @@
#else
#include <linux/irq.h>
extern unsigned int kstat_irqs_cpu(unsigned int irq, int cpu);
-#define kstat_irqs_this_cpu(DESC) \
- ((DESC)->kstat_irqs[smp_processor_id()])
-#define kstat_incr_irqs_this_cpu(irqno, DESC) do {\
- ((DESC)->kstat_irqs[smp_processor_id()]++);\
- kstat_this_cpu.irqs_sum++; } while (0)
+
+#define kstat_incr_irqs_this_cpu(irqno, DESC) \
+do { \
+ __this_cpu_inc(*(DESC)->kstat_irqs); \
+ __this_cpu_inc(kstat.irqs_sum); \
+} while (0)
#endif
static inline void kstat_incr_softirqs_this_cpu(unsigned int irq)
{
- kstat_this_cpu.softirqs[irq]++;
+ __this_cpu_inc(kstat.softirqs[irq]);
}
static inline unsigned int kstat_softirqs_cpu(unsigned int irq, int cpu)
diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h
new file mode 100644
index 0000000..6b394f0
--- /dev/null
+++ b/include/linux/khugepaged.h
@@ -0,0 +1,67 @@
+#ifndef _LINUX_KHUGEPAGED_H
+#define _LINUX_KHUGEPAGED_H
+
+#include <linux/sched.h> /* MMF_VM_HUGEPAGE */
+
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+extern int __khugepaged_enter(struct mm_struct *mm);
+extern void __khugepaged_exit(struct mm_struct *mm);
+extern int khugepaged_enter_vma_merge(struct vm_area_struct *vma);
+
+#define khugepaged_enabled() \
+ (transparent_hugepage_flags & \
+ ((1<<TRANSPARENT_HUGEPAGE_FLAG) | \
+ (1<<TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG)))
+#define khugepaged_always() \
+ (transparent_hugepage_flags & \
+ (1<<TRANSPARENT_HUGEPAGE_FLAG))
+#define khugepaged_req_madv() \
+ (transparent_hugepage_flags & \
+ (1<<TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG))
+#define khugepaged_defrag() \
+ (transparent_hugepage_flags & \
+ (1<<TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG))
+
+static inline int khugepaged_fork(struct mm_struct *mm, struct mm_struct *oldmm)
+{
+ if (test_bit(MMF_VM_HUGEPAGE, &oldmm->flags))
+ return __khugepaged_enter(mm);
+ return 0;
+}
+
+static inline void khugepaged_exit(struct mm_struct *mm)
+{
+ if (test_bit(MMF_VM_HUGEPAGE, &mm->flags))
+ __khugepaged_exit(mm);
+}
+
+static inline int khugepaged_enter(struct vm_area_struct *vma)
+{
+ if (!test_bit(MMF_VM_HUGEPAGE, &vma->vm_mm->flags))
+ if ((khugepaged_always() ||
+ (khugepaged_req_madv() &&
+ vma->vm_flags & VM_HUGEPAGE)) &&
+ !(vma->vm_flags & VM_NOHUGEPAGE))
+ if (__khugepaged_enter(vma->vm_mm))
+ return -ENOMEM;
+ return 0;
+}
+#else /* CONFIG_TRANSPARENT_HUGEPAGE */
+static inline int khugepaged_fork(struct mm_struct *mm, struct mm_struct *oldmm)
+{
+ return 0;
+}
+static inline void khugepaged_exit(struct mm_struct *mm)
+{
+}
+static inline int khugepaged_enter(struct vm_area_struct *vma)
+{
+ return 0;
+}
+static inline int khugepaged_enter_vma_merge(struct vm_area_struct *vma)
+{
+ return 0;
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+
+#endif /* _LINUX_KHUGEPAGED_H */
diff --git a/include/linux/list_bl.h b/include/linux/list_bl.h
index 9ee97e7..5bad17d1 100644
--- a/include/linux/list_bl.h
+++ b/include/linux/list_bl.h
@@ -62,7 +62,8 @@
struct hlist_bl_node *n)
{
LIST_BL_BUG_ON((unsigned long)n & LIST_BL_LOCKMASK);
- LIST_BL_BUG_ON(!((unsigned long)h->first & LIST_BL_LOCKMASK));
+ LIST_BL_BUG_ON(((unsigned long)h->first & LIST_BL_LOCKMASK) !=
+ LIST_BL_LOCKMASK);
h->first = (struct hlist_bl_node *)((unsigned long)n | LIST_BL_LOCKMASK);
}
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 159a0762..6a576f9 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -25,6 +25,11 @@
struct page;
struct mm_struct;
+/* Stats that can be updated by kernel. */
+enum mem_cgroup_page_stat_item {
+ MEMCG_NR_FILE_MAPPED, /* # of pages charged as file rss */
+};
+
extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
struct list_head *dst,
unsigned long *scanned, int order,
@@ -93,7 +98,7 @@
mem_cgroup_prepare_migration(struct page *page,
struct page *newpage, struct mem_cgroup **ptr);
extern void mem_cgroup_end_migration(struct mem_cgroup *mem,
- struct page *oldpage, struct page *newpage);
+ struct page *oldpage, struct page *newpage, bool migration_ok);
/*
* For memory reclaim.
@@ -121,7 +126,22 @@
return false;
}
-void mem_cgroup_update_file_mapped(struct page *page, int val);
+void mem_cgroup_update_page_stat(struct page *page,
+ enum mem_cgroup_page_stat_item idx,
+ int val);
+
+static inline void mem_cgroup_inc_page_stat(struct page *page,
+ enum mem_cgroup_page_stat_item idx)
+{
+ mem_cgroup_update_page_stat(page, idx, 1);
+}
+
+static inline void mem_cgroup_dec_page_stat(struct page *page,
+ enum mem_cgroup_page_stat_item idx)
+{
+ mem_cgroup_update_page_stat(page, idx, -1);
+}
+
unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order,
gfp_t gfp_mask);
u64 mem_cgroup_get_limit(struct mem_cgroup *mem);
@@ -231,8 +251,7 @@
}
static inline void mem_cgroup_end_migration(struct mem_cgroup *mem,
- struct page *oldpage,
- struct page *newpage)
+ struct page *oldpage, struct page *newpage, bool migration_ok)
{
}
@@ -293,8 +312,13 @@
{
}
-static inline void mem_cgroup_update_file_mapped(struct page *page,
- int val)
+static inline void mem_cgroup_inc_page_stat(struct page *page,
+ enum mem_cgroup_page_stat_item idx)
+{
+}
+
+static inline void mem_cgroup_dec_page_stat(struct page *page,
+ enum mem_cgroup_page_stat_item idx)
{
}
diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h
index 31c237a..8122018d 100644
--- a/include/linux/memory_hotplug.h
+++ b/include/linux/memory_hotplug.h
@@ -13,12 +13,16 @@
#ifdef CONFIG_MEMORY_HOTPLUG
/*
- * Types for free bootmem.
- * The normal smallest mapcount is -1. Here is smaller value than it.
+ * Types for free bootmem stored in page->lru.next. These have to be in
+ * some random range in unsigned long space for debugging purposes.
*/
-#define SECTION_INFO (-1 - 1)
-#define MIX_SECTION_INFO (-1 - 2)
-#define NODE_INFO (-1 - 3)
+enum {
+ MEMORY_HOTPLUG_MIN_BOOTMEM_TYPE = 12,
+ SECTION_INFO = MEMORY_HOTPLUG_MIN_BOOTMEM_TYPE,
+ MIX_SECTION_INFO,
+ NODE_INFO,
+ MEMORY_HOTPLUG_MAX_BOOTMEM_TYPE = NODE_INFO,
+};
/*
* pgdat resizing functions
@@ -161,6 +165,12 @@
extern void put_page_bootmem(struct page *page);
#endif
+/*
+ * Lock for memory hotplug guarantees 1) all callbacks for memory hotplug
+ * notifier will be called under this. 2) offline/online/add/remove memory
+ * will not run simultaneously.
+ */
+
void lock_memory_hotplug(void);
void unlock_memory_hotplug(void);
diff --git a/include/linux/mfd/ab8500.h b/include/linux/mfd/ab8500.h
index 85cf2c28..37f56b7 100644
--- a/include/linux/mfd/ab8500.h
+++ b/include/linux/mfd/ab8500.h
@@ -74,30 +74,37 @@
#define AB8500_INT_ACC_DETECT_21DB_F 37
#define AB8500_INT_ACC_DETECT_21DB_R 38
#define AB8500_INT_GP_SW_ADC_CONV_END 39
-#define AB8500_INT_BTEMP_LOW 72
-#define AB8500_INT_BTEMP_LOW_MEDIUM 73
-#define AB8500_INT_BTEMP_MEDIUM_HIGH 74
-#define AB8500_INT_BTEMP_HIGH 75
-#define AB8500_INT_USB_CHARGER_NOT_OK 81
-#define AB8500_INT_ID_WAKEUP_R 82
-#define AB8500_INT_ID_DET_R1R 84
-#define AB8500_INT_ID_DET_R2R 85
-#define AB8500_INT_ID_DET_R3R 86
-#define AB8500_INT_ID_DET_R4R 87
-#define AB8500_INT_ID_WAKEUP_F 88
-#define AB8500_INT_ID_DET_R1F 90
-#define AB8500_INT_ID_DET_R2F 91
-#define AB8500_INT_ID_DET_R3F 92
-#define AB8500_INT_ID_DET_R4F 93
-#define AB8500_INT_USB_CHG_DET_DONE 94
-#define AB8500_INT_USB_CH_TH_PROT_F 96
-#define AB8500_INT_USB_CH_TH_PROP_R 97
-#define AB8500_INT_MAIN_CH_TH_PROP_F 98
-#define AB8500_INT_MAIN_CH_TH_PROT_R 99
-#define AB8500_INT_USB_CHARGER_NOT_OKF 103
+#define AB8500_INT_ADP_SOURCE_ERROR 72
+#define AB8500_INT_ADP_SINK_ERROR 73
+#define AB8500_INT_ADP_PROBE_PLUG 74
+#define AB8500_INT_ADP_PROBE_UNPLUG 75
+#define AB8500_INT_ADP_SENSE_OFF 76
+#define AB8500_INT_USB_PHY_POWER_ERR 78
+#define AB8500_INT_USB_LINK_STATUS 79
+#define AB8500_INT_BTEMP_LOW 80
+#define AB8500_INT_BTEMP_LOW_MEDIUM 81
+#define AB8500_INT_BTEMP_MEDIUM_HIGH 82
+#define AB8500_INT_BTEMP_HIGH 83
+#define AB8500_INT_USB_CHARGER_NOT_OK 89
+#define AB8500_INT_ID_WAKEUP_R 90
+#define AB8500_INT_ID_DET_R1R 92
+#define AB8500_INT_ID_DET_R2R 93
+#define AB8500_INT_ID_DET_R3R 94
+#define AB8500_INT_ID_DET_R4R 95
+#define AB8500_INT_ID_WAKEUP_F 96
+#define AB8500_INT_ID_DET_R1F 98
+#define AB8500_INT_ID_DET_R2F 99
+#define AB8500_INT_ID_DET_R3F 100
+#define AB8500_INT_ID_DET_R4F 101
+#define AB8500_INT_USB_CHG_DET_DONE 102
+#define AB8500_INT_USB_CH_TH_PROT_F 104
+#define AB8500_INT_USB_CH_TH_PROT_R 105
+#define AB8500_INT_MAIN_CH_TH_PROT_F 106
+#define AB8500_INT_MAIN_CH_TH_PROT_R 107
+#define AB8500_INT_USB_CHARGER_NOT_OKF 111
-#define AB8500_NR_IRQS 104
-#define AB8500_NUM_IRQ_REGS 13
+#define AB8500_NR_IRQS 112
+#define AB8500_NUM_IRQ_REGS 14
/**
* struct ab8500 - ab8500 internal structure
diff --git a/include/linux/mfd/core.h b/include/linux/mfd/core.h
index 5582ab3..835996e 100644
--- a/include/linux/mfd/core.h
+++ b/include/linux/mfd/core.h
@@ -47,6 +47,12 @@
/* don't check for resource conflicts */
bool ignore_resource_conflicts;
+
+ /*
+ * Disable runtime PM callbacks for this subdevice - see
+ * pm_runtime_no_callbacks().
+ */
+ bool pm_runtime_no_callbacks;
};
extern int mfd_add_devices(struct device *parent, int id,
diff --git a/include/linux/mfd/max8998-private.h b/include/linux/mfd/max8998-private.h
index 7363dea..effa5d3 100644
--- a/include/linux/mfd/max8998-private.h
+++ b/include/linux/mfd/max8998-private.h
@@ -159,10 +159,12 @@
u8 irq_masks_cur[MAX8998_NUM_IRQ_REGS];
u8 irq_masks_cache[MAX8998_NUM_IRQ_REGS];
int type;
+ bool wakeup;
};
int max8998_irq_init(struct max8998_dev *max8998);
void max8998_irq_exit(struct max8998_dev *max8998);
+int max8998_irq_resume(struct max8998_dev *max8998);
extern int max8998_read_reg(struct i2c_client *i2c, u8 reg, u8 *dest);
extern int max8998_bulk_read(struct i2c_client *i2c, u8 reg, int count,
diff --git a/include/linux/mfd/max8998.h b/include/linux/mfd/max8998.h
index f8c9f88..61daa16 100644
--- a/include/linux/mfd/max8998.h
+++ b/include/linux/mfd/max8998.h
@@ -70,24 +70,43 @@
* @num_regulators: number of regultors used
* @irq_base: base IRQ number for max8998, required for IRQs
* @ono: power onoff IRQ number for max8998
- * @buck1_max_voltage1: BUCK1 maximum alowed voltage register 1
- * @buck1_max_voltage2: BUCK1 maximum alowed voltage register 2
- * @buck2_max_voltage: BUCK2 maximum alowed voltage
+ * @buck_voltage_lock: Do NOT change the values of the following six
+ * registers set by buck?_voltage?. The voltage of BUCK1/2 cannot
+ * be other than the preset values.
+ * @buck1_voltage1: BUCK1 DVS mode 1 voltage register
+ * @buck1_voltage2: BUCK1 DVS mode 2 voltage register
+ * @buck1_voltage3: BUCK1 DVS mode 3 voltage register
+ * @buck1_voltage4: BUCK1 DVS mode 4 voltage register
+ * @buck2_voltage1: BUCK2 DVS mode 1 voltage register
+ * @buck2_voltage2: BUCK2 DVS mode 2 voltage register
* @buck1_set1: BUCK1 gpio pin 1 to set output voltage
* @buck1_set2: BUCK1 gpio pin 2 to set output voltage
+ * @buck1_default_idx: Default for BUCK1 gpio pin 1, 2
* @buck2_set3: BUCK2 gpio pin to set output voltage
+ * @buck2_default_idx: Default for BUCK2 gpio pin.
+ * @wakeup: Allow to wake up from suspend
+ * @rtc_delay: LP3974 RTC chip bug that requires delay after a register
+ * write before reading it.
*/
struct max8998_platform_data {
struct max8998_regulator_data *regulators;
int num_regulators;
int irq_base;
int ono;
- int buck1_max_voltage1;
- int buck1_max_voltage2;
- int buck2_max_voltage;
+ bool buck_voltage_lock;
+ int buck1_voltage1;
+ int buck1_voltage2;
+ int buck1_voltage3;
+ int buck1_voltage4;
+ int buck2_voltage1;
+ int buck2_voltage2;
int buck1_set1;
int buck1_set2;
+ int buck1_default_idx;
int buck2_set3;
+ int buck2_default_idx;
+ bool wakeup;
+ bool rtc_delay;
};
#endif /* __LINUX_MFD_MAX8998_H */
diff --git a/include/linux/mfd/wm831x/core.h b/include/linux/mfd/wm831x/core.h
index a1239c4..903280d 100644
--- a/include/linux/mfd/wm831x/core.h
+++ b/include/linux/mfd/wm831x/core.h
@@ -245,6 +245,7 @@
WM8320 = 0x8320,
WM8321 = 0x8321,
WM8325 = 0x8325,
+ WM8326 = 0x8326,
};
struct wm831x {
diff --git a/include/linux/migrate.h b/include/linux/migrate.h
index 085527f..e39aeec 100644
--- a/include/linux/migrate.h
+++ b/include/linux/migrate.h
@@ -13,9 +13,11 @@
extern int migrate_page(struct address_space *,
struct page *, struct page *);
extern int migrate_pages(struct list_head *l, new_page_t x,
- unsigned long private, int offlining);
+ unsigned long private, bool offlining,
+ bool sync);
extern int migrate_huge_pages(struct list_head *l, new_page_t x,
- unsigned long private, int offlining);
+ unsigned long private, bool offlining,
+ bool sync);
extern int fail_migrate_page(struct address_space *,
struct page *, struct page *);
@@ -33,9 +35,11 @@
static inline void putback_lru_pages(struct list_head *l) {}
static inline int migrate_pages(struct list_head *l, new_page_t x,
- unsigned long private, int offlining) { return -ENOSYS; }
+ unsigned long private, bool offlining,
+ bool sync) { return -ENOSYS; }
static inline int migrate_huge_pages(struct list_head *l, new_page_t x,
- unsigned long private, int offlining) { return -ENOSYS; }
+ unsigned long private, bool offlining,
+ bool sync) { return -ENOSYS; }
static inline int migrate_prep(void) { return -ENOSYS; }
static inline int migrate_prep_local(void) { return -ENOSYS; }
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 721f451..956a355 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -14,6 +14,7 @@
#include <linux/mm_types.h>
#include <linux/range.h>
#include <linux/pfn.h>
+#include <linux/bit_spinlock.h>
struct mempolicy;
struct anon_vma;
@@ -82,6 +83,7 @@
#define VM_GROWSUP 0x00000200
#else
#define VM_GROWSUP 0x00000000
+#define VM_NOHUGEPAGE 0x00000200 /* MADV_NOHUGEPAGE marked this vma */
#endif
#define VM_PFNMAP 0x00000400 /* Page-ranges managed without "struct page", just pure PFN */
#define VM_DENYWRITE 0x00000800 /* ETXTBSY on write attempts.. */
@@ -101,7 +103,11 @@
#define VM_NORESERVE 0x00200000 /* should the VM suppress accounting */
#define VM_HUGETLB 0x00400000 /* Huge TLB Page VM */
#define VM_NONLINEAR 0x00800000 /* Is non-linear (remap_file_pages) */
+#ifndef CONFIG_TRANSPARENT_HUGEPAGE
#define VM_MAPPED_COPY 0x01000000 /* T if mapped copy of data (nommu mmap) */
+#else
+#define VM_HUGEPAGE 0x01000000 /* MADV_HUGEPAGE marked this vma */
+#endif
#define VM_INSERTPAGE 0x02000000 /* The vma has had "vm_insert_page()" done on it */
#define VM_ALWAYSDUMP 0x04000000 /* Always include in core dumps */
@@ -242,6 +248,7 @@
* files which need it (119 of them)
*/
#include <linux/page-flags.h>
+#include <linux/huge_mm.h>
/*
* Methods to modify the page usage count.
@@ -305,6 +312,39 @@
}
#endif
+static inline void compound_lock(struct page *page)
+{
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ bit_spin_lock(PG_compound_lock, &page->flags);
+#endif
+}
+
+static inline void compound_unlock(struct page *page)
+{
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ bit_spin_unlock(PG_compound_lock, &page->flags);
+#endif
+}
+
+static inline unsigned long compound_lock_irqsave(struct page *page)
+{
+ unsigned long uninitialized_var(flags);
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ local_irq_save(flags);
+ compound_lock(page);
+#endif
+ return flags;
+}
+
+static inline void compound_unlock_irqrestore(struct page *page,
+ unsigned long flags)
+{
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ compound_unlock(page);
+ local_irq_restore(flags);
+#endif
+}
+
static inline struct page *compound_head(struct page *page)
{
if (unlikely(PageTail(page)))
@@ -319,9 +359,29 @@
static inline void get_page(struct page *page)
{
- page = compound_head(page);
- VM_BUG_ON(atomic_read(&page->_count) == 0);
+ /*
+ * Getting a normal page or the head of a compound page
+ * requires to already have an elevated page->_count. Only if
+ * we're getting a tail page, the elevated page->_count is
+ * required only in the head page, so for tail pages the
+ * bugcheck only verifies that the page->_count isn't
+ * negative.
+ */
+ VM_BUG_ON(atomic_read(&page->_count) < !PageTail(page));
atomic_inc(&page->_count);
+ /*
+ * Getting a tail page will elevate both the head and tail
+ * page->_count(s).
+ */
+ if (unlikely(PageTail(page))) {
+ /*
+ * This is safe only because
+ * __split_huge_page_refcount can't run under
+ * get_page().
+ */
+ VM_BUG_ON(atomic_read(&page->first_page->_count) <= 0);
+ atomic_inc(&page->first_page->_count);
+ }
}
static inline struct page *virt_to_head_page(const void *x)
@@ -339,6 +399,27 @@
atomic_set(&page->_count, 1);
}
+/*
+ * PageBuddy() indicate that the page is free and in the buddy system
+ * (see mm/page_alloc.c).
+ */
+static inline int PageBuddy(struct page *page)
+{
+ return atomic_read(&page->_mapcount) == -2;
+}
+
+static inline void __SetPageBuddy(struct page *page)
+{
+ VM_BUG_ON(atomic_read(&page->_mapcount) != -1);
+ atomic_set(&page->_mapcount, -2);
+}
+
+static inline void __ClearPageBuddy(struct page *page)
+{
+ VM_BUG_ON(!PageBuddy(page));
+ atomic_set(&page->_mapcount, -1);
+}
+
void put_page(struct page *page);
void put_pages_list(struct list_head *pages);
@@ -370,12 +451,39 @@
return (unsigned long)page[1].lru.prev;
}
+static inline int compound_trans_order(struct page *page)
+{
+ int order;
+ unsigned long flags;
+
+ if (!PageHead(page))
+ return 0;
+
+ flags = compound_lock_irqsave(page);
+ order = compound_order(page);
+ compound_unlock_irqrestore(page, flags);
+ return order;
+}
+
static inline void set_compound_order(struct page *page, unsigned long order)
{
page[1].lru.prev = (void *)order;
}
/*
+ * Do pte_mkwrite, but only if the vma says VM_WRITE. We do this when
+ * servicing faults for write access. In the normal case, do always want
+ * pte_mkwrite. But get_user_pages can cause write faults for mappings
+ * that do not have writing enabled, when used by access_process_vm.
+ */
+static inline pte_t maybe_mkwrite(pte_t pte, struct vm_area_struct *vma)
+{
+ if (likely(vma->vm_flags & VM_WRITE))
+ pte = pte_mkwrite(pte);
+ return pte;
+}
+
+/*
* Multiple processes may "see" the same page. E.g. for untouched
* mappings of /dev/null, all processes see the same page full of
* zeroes, and text pages of executables and shared libraries have
@@ -657,7 +765,7 @@
VM_BUG_ON(PageSlab(page));
if (unlikely(PageSwapCache(page)))
mapping = &swapper_space;
- else if (unlikely((unsigned long)mapping & PAGE_MAPPING_ANON))
+ else if ((unsigned long)mapping & PAGE_MAPPING_ANON)
mapping = NULL;
return mapping;
}
@@ -1064,7 +1172,8 @@
int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address);
#endif
-int __pte_alloc(struct mm_struct *mm, pmd_t *pmd, unsigned long address);
+int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma,
+ pmd_t *pmd, unsigned long address);
int __pte_alloc_kernel(pmd_t *pmd, unsigned long address);
/*
@@ -1133,16 +1242,18 @@
pte_unmap(pte); \
} while (0)
-#define pte_alloc_map(mm, pmd, address) \
- ((unlikely(!pmd_present(*(pmd))) && __pte_alloc(mm, pmd, address))? \
- NULL: pte_offset_map(pmd, address))
+#define pte_alloc_map(mm, vma, pmd, address) \
+ ((unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, vma, \
+ pmd, address))? \
+ NULL: pte_offset_map(pmd, address))
#define pte_alloc_map_lock(mm, pmd, address, ptlp) \
- ((unlikely(!pmd_present(*(pmd))) && __pte_alloc(mm, pmd, address))? \
+ ((unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, NULL, \
+ pmd, address))? \
NULL: pte_offset_map_lock(mm, pmd, address, ptlp))
#define pte_alloc_kernel(pmd, address) \
- ((unlikely(!pmd_present(*(pmd))) && __pte_alloc_kernel(pmd, address))? \
+ ((unlikely(pmd_none(*(pmd))) && __pte_alloc_kernel(pmd, address))? \
NULL: pte_offset_kernel(pmd, address))
extern void free_area_init(unsigned long * zones_size);
@@ -1415,6 +1526,8 @@
#define FOLL_GET 0x04 /* do get_page on page */
#define FOLL_DUMP 0x08 /* give error on hole if it would be zero */
#define FOLL_FORCE 0x10 /* get_user_pages read/write w/o permission */
+#define FOLL_MLOCK 0x40 /* mark page as mlocked */
+#define FOLL_SPLIT 0x80 /* don't return transhuge pages, split them */
typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
void *data);
@@ -1518,5 +1631,14 @@
extern void dump_page(struct page *page);
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS)
+extern void clear_huge_page(struct page *page,
+ unsigned long addr,
+ unsigned int pages_per_huge_page);
+extern void copy_user_huge_page(struct page *dst, struct page *src,
+ unsigned long addr, struct vm_area_struct *vma,
+ unsigned int pages_per_huge_page);
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */
+
#endif /* __KERNEL__ */
#endif /* _LINUX_MM_H */
diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h
index 8835b87..8f7d247 100644
--- a/include/linux/mm_inline.h
+++ b/include/linux/mm_inline.h
@@ -1,6 +1,8 @@
#ifndef LINUX_MM_INLINE_H
#define LINUX_MM_INLINE_H
+#include <linux/huge_mm.h>
+
/**
* page_is_file_cache - should the page be on a file LRU or anon LRU?
* @page: the page to test
@@ -20,18 +22,25 @@
}
static inline void
+__add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l,
+ struct list_head *head)
+{
+ list_add(&page->lru, head);
+ __mod_zone_page_state(zone, NR_LRU_BASE + l, hpage_nr_pages(page));
+ mem_cgroup_add_lru_list(page, l);
+}
+
+static inline void
add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l)
{
- list_add(&page->lru, &zone->lru[l].list);
- __inc_zone_state(zone, NR_LRU_BASE + l);
- mem_cgroup_add_lru_list(page, l);
+ __add_page_to_lru_list(zone, page, l, &zone->lru[l].list);
}
static inline void
del_page_from_lru_list(struct zone *zone, struct page *page, enum lru_list l)
{
list_del(&page->lru);
- __dec_zone_state(zone, NR_LRU_BASE + l);
+ __mod_zone_page_state(zone, NR_LRU_BASE + l, -hpage_nr_pages(page));
mem_cgroup_del_lru_list(page, l);
}
@@ -66,7 +75,7 @@
l += LRU_ACTIVE;
}
}
- __dec_zone_state(zone, NR_LRU_BASE + l);
+ __mod_zone_page_state(zone, NR_LRU_BASE + l, -hpage_nr_pages(page));
mem_cgroup_del_lru_list(page, l);
}
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index bb7288a..26bc4e2 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -310,6 +310,9 @@
#ifdef CONFIG_MMU_NOTIFIER
struct mmu_notifier_mm *mmu_notifier_mm;
#endif
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ pgtable_t pmd_huge_pte; /* protected by page_table_lock */
+#endif
/* How many tasks sharing this mm are OOM_DISABLE */
atomic_t oom_disable_count;
};
diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
index 43dcfbd..cc2e7df 100644
--- a/include/linux/mmu_notifier.h
+++ b/include/linux/mmu_notifier.h
@@ -62,6 +62,16 @@
unsigned long address);
/*
+ * test_young is called to check the young/accessed bitflag in
+ * the secondary pte. This is used to know if the page is
+ * frequently used without actually clearing the flag or tearing
+ * down the secondary mapping on the page.
+ */
+ int (*test_young)(struct mmu_notifier *mn,
+ struct mm_struct *mm,
+ unsigned long address);
+
+ /*
* change_pte is called in cases that pte mapping to page is changed:
* for example, when ksm remaps pte to point to a new shared page.
*/
@@ -163,6 +173,8 @@
extern void __mmu_notifier_release(struct mm_struct *mm);
extern int __mmu_notifier_clear_flush_young(struct mm_struct *mm,
unsigned long address);
+extern int __mmu_notifier_test_young(struct mm_struct *mm,
+ unsigned long address);
extern void __mmu_notifier_change_pte(struct mm_struct *mm,
unsigned long address, pte_t pte);
extern void __mmu_notifier_invalidate_page(struct mm_struct *mm,
@@ -186,6 +198,14 @@
return 0;
}
+static inline int mmu_notifier_test_young(struct mm_struct *mm,
+ unsigned long address)
+{
+ if (mm_has_notifiers(mm))
+ return __mmu_notifier_test_young(mm, address);
+ return 0;
+}
+
static inline void mmu_notifier_change_pte(struct mm_struct *mm,
unsigned long address, pte_t pte)
{
@@ -243,6 +263,32 @@
__pte; \
})
+#define pmdp_clear_flush_notify(__vma, __address, __pmdp) \
+({ \
+ pmd_t __pmd; \
+ struct vm_area_struct *___vma = __vma; \
+ unsigned long ___address = __address; \
+ VM_BUG_ON(__address & ~HPAGE_PMD_MASK); \
+ mmu_notifier_invalidate_range_start(___vma->vm_mm, ___address, \
+ (__address)+HPAGE_PMD_SIZE);\
+ __pmd = pmdp_clear_flush(___vma, ___address, __pmdp); \
+ mmu_notifier_invalidate_range_end(___vma->vm_mm, ___address, \
+ (__address)+HPAGE_PMD_SIZE); \
+ __pmd; \
+})
+
+#define pmdp_splitting_flush_notify(__vma, __address, __pmdp) \
+({ \
+ struct vm_area_struct *___vma = __vma; \
+ unsigned long ___address = __address; \
+ VM_BUG_ON(__address & ~HPAGE_PMD_MASK); \
+ mmu_notifier_invalidate_range_start(___vma->vm_mm, ___address, \
+ (__address)+HPAGE_PMD_SIZE);\
+ pmdp_splitting_flush(___vma, ___address, __pmdp); \
+ mmu_notifier_invalidate_range_end(___vma->vm_mm, ___address, \
+ (__address)+HPAGE_PMD_SIZE); \
+})
+
#define ptep_clear_flush_young_notify(__vma, __address, __ptep) \
({ \
int __young; \
@@ -254,6 +300,17 @@
__young; \
})
+#define pmdp_clear_flush_young_notify(__vma, __address, __pmdp) \
+({ \
+ int __young; \
+ struct vm_area_struct *___vma = __vma; \
+ unsigned long ___address = __address; \
+ __young = pmdp_clear_flush_young(___vma, ___address, __pmdp); \
+ __young |= mmu_notifier_clear_flush_young(___vma->vm_mm, \
+ ___address); \
+ __young; \
+})
+
#define set_pte_at_notify(__mm, __address, __ptep, __pte) \
({ \
struct mm_struct *___mm = __mm; \
@@ -276,6 +333,12 @@
return 0;
}
+static inline int mmu_notifier_test_young(struct mm_struct *mm,
+ unsigned long address)
+{
+ return 0;
+}
+
static inline void mmu_notifier_change_pte(struct mm_struct *mm,
unsigned long address, pte_t pte)
{
@@ -305,7 +368,10 @@
}
#define ptep_clear_flush_young_notify ptep_clear_flush_young
+#define pmdp_clear_flush_young_notify pmdp_clear_flush_young
#define ptep_clear_flush_notify ptep_clear_flush
+#define pmdp_clear_flush_notify pmdp_clear_flush
+#define pmdp_splitting_flush_notify pmdp_splitting_flush
#define set_pte_at_notify set_pte_at
#endif /* CONFIG_MMU_NOTIFIER */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 39c24eb..02ecb01 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -114,6 +114,7 @@
NUMA_LOCAL, /* allocation from local node */
NUMA_OTHER, /* allocation from other node */
#endif
+ NR_ANON_TRANSPARENT_HUGEPAGES,
NR_VM_ZONE_STAT_ITEMS };
/*
@@ -458,12 +459,6 @@
return test_bit(ZONE_OOM_LOCKED, &zone->flags);
}
-#ifdef CONFIG_SMP
-unsigned long zone_nr_free_pages(struct zone *zone);
-#else
-#define zone_nr_free_pages(zone) zone_page_state(zone, NR_FREE_PAGES)
-#endif /* CONFIG_SMP */
-
/*
* The "priority" of VM scanning is how much of the queues we will scan in one
* go. A value of 12 for DEF_PRIORITY implies that we will scan 1/4096th of the
@@ -645,6 +640,7 @@
wait_queue_head_t kswapd_wait;
struct task_struct *kswapd;
int kswapd_max_order;
+ enum zone_type classzone_idx;
} pg_data_t;
#define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages)
@@ -660,8 +656,10 @@
extern struct mutex zonelists_mutex;
void build_all_zonelists(void *data);
-void wakeup_kswapd(struct zone *zone, int order);
-int zone_watermark_ok(struct zone *z, int order, unsigned long mark,
+void wakeup_kswapd(struct zone *zone, int order, enum zone_type classzone_idx);
+bool zone_watermark_ok(struct zone *z, int order, unsigned long mark,
+ int classzone_idx, int alloc_flags);
+bool zone_watermark_ok_safe(struct zone *z, int order, unsigned long mark,
int classzone_idx, int alloc_flags);
enum memmap_context {
MEMMAP_EARLY,
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 1869ea2..604f122 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -60,7 +60,7 @@
struct super_block *mnt_sb; /* pointer to superblock */
#ifdef CONFIG_SMP
struct mnt_pcp __percpu *mnt_pcp;
- atomic_t mnt_longrefs;
+ atomic_t mnt_longterm; /* how many of the refs are longterm */
#else
int mnt_count;
int mnt_writers;
@@ -96,8 +96,6 @@
extern void mnt_drop_write(struct vfsmount *mnt);
extern void mntput(struct vfsmount *mnt);
extern struct vfsmount *mntget(struct vfsmount *mnt);
-extern void mntput_long(struct vfsmount *mnt);
-extern struct vfsmount *mntget_long(struct vfsmount *mnt);
extern void mnt_pin(struct vfsmount *mnt);
extern void mnt_unpin(struct vfsmount *mnt);
extern int __mnt_is_readonly(struct vfsmount *mnt);
@@ -110,12 +108,7 @@
int flags, const char *name,
void *data);
-struct nameidata;
-
-struct path;
-extern int do_add_mount(struct vfsmount *newmnt, struct path *path,
- int mnt_flags, struct list_head *fslist);
-
+extern void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list);
extern void mark_mounts_for_expiry(struct list_head *mounts);
extern dev_t name_to_dev_t(char *name);
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 18d06ad..f276d4f 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -45,6 +45,7 @@
* - ending slashes ok even for nonexistent files
* - internal "there are more path components" flag
* - dentry cache is untrusted; force a real lookup
+ * - suppress terminal automount
*/
#define LOOKUP_FOLLOW 0x0001
#define LOOKUP_DIRECTORY 0x0002
@@ -53,6 +54,7 @@
#define LOOKUP_PARENT 0x0010
#define LOOKUP_REVAL 0x0020
#define LOOKUP_RCU 0x0040
+#define LOOKUP_NO_AUTOMOUNT 0x0080
/*
* Intent data
*/
@@ -79,7 +81,8 @@
extern struct dentry *lookup_one_len(const char *, struct dentry *, int);
-extern int follow_down(struct path *);
+extern int follow_down_one(struct path *);
+extern int follow_down(struct path *, bool);
extern int follow_up(struct path *);
extern struct dentry *lock_rename(struct dentry *, struct dentry *);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index be4957c..d971346 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -520,9 +520,6 @@
* please use this field instead of dev->trans_start
*/
unsigned long trans_start;
- u64 tx_bytes;
- u64 tx_packets;
- u64 tx_dropped;
} ____cacheline_aligned_in_smp;
static inline int netdev_queue_numa_node_read(const struct netdev_queue *q)
@@ -2265,8 +2262,6 @@
extern void dev_mcast_init(void);
extern struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
struct rtnl_link_stats64 *storage);
-extern void dev_txq_stats_fold(const struct net_device *dev,
- struct rtnl_link_stats64 *stats);
extern int netdev_max_backlog;
extern int netdev_tstamp_prequeue;
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 9b46300..134716e 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -65,6 +65,9 @@
#define NFS4_CDFC4_FORE 0x1
#define NFS4_CDFC4_BACK 0x2
+#define NFS4_CDFC4_BOTH 0x3
+#define NFS4_CDFC4_FORE_OR_BOTH 0x3
+#define NFS4_CDFC4_BACK_OR_BOTH 0x7
#define NFS4_SET_TO_SERVER_TIME 0
#define NFS4_SET_TO_CLIENT_TIME 1
@@ -140,6 +143,9 @@
#define SEQ4_STATUS_CB_PATH_DOWN_SESSION 0x00000200
#define SEQ4_STATUS_BACKCHANNEL_FAULT 0x00000400
+#define NFS4_SECINFO_STYLE4_CURRENT_FH 0
+#define NFS4_SECINFO_STYLE4_PARENT 1
+
#define NFS4_MAX_UINT64 (~(u64)0)
/* An NFS4 sessions server must support at least NFS4_MAX_OPS operations.
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 0779bb8..6023efa 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -215,7 +215,6 @@
#define NFS_INO_ADVISE_RDPLUS (0) /* advise readdirplus */
#define NFS_INO_STALE (1) /* possible stale inode */
#define NFS_INO_ACL_LRU_SET (2) /* Inode is on the LRU list */
-#define NFS_INO_MOUNTPOINT (3) /* inode is remote mountpoint */
#define NFS_INO_FLUSHING (4) /* inode is flushing out data */
#define NFS_INO_FSCACHE (5) /* inode can be cached by FS-Cache */
#define NFS_INO_FSCACHE_LOCK (6) /* FS-Cache cookie management lock */
diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h
index 8ae78a6..bd31615 100644
--- a/include/linux/nfsd/export.h
+++ b/include/linux/nfsd/export.h
@@ -35,7 +35,7 @@
#define NFSEXP_NOHIDE 0x0200
#define NFSEXP_NOSUBTREECHECK 0x0400
#define NFSEXP_NOAUTHNLM 0x0800 /* Don't authenticate NLM requests - just trust */
-#define NFSEXP_MSNFS 0x1000 /* do silly things that MS clients expect */
+#define NFSEXP_MSNFS 0x1000 /* do silly things that MS clients expect; no longer supported */
#define NFSEXP_FSID 0x2000
#define NFSEXP_CROSSMOUNT 0x4000
#define NFSEXP_NOACL 0x8000 /* reserved for possible ACL related use */
diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h
index 2b89b71..821ffb9 100644
--- a/include/linux/nl80211.h
+++ b/include/linux/nl80211.h
@@ -148,6 +148,10 @@
* @NL80211_CMD_SET_MPATH: Set mesh path attributes for mesh path to
* destination %NL80211_ATTR_MAC on the interface identified by
* %NL80211_ATTR_IFINDEX.
+ * @NL80211_CMD_NEW_MPATH: Create a new mesh path for the destination given by
+ * %NL80211_ATTR_MAC via %NL80211_ATTR_MPATH_NEXT_HOP.
+ * @NL80211_CMD_DEL_MPATH: Delete a mesh path to the destination given by
+ * %NL80211_ATTR_MAC.
* @NL80211_CMD_NEW_PATH: Add a mesh path with given attributes to the
* the interface identified by %NL80211_ATTR_IFINDEX.
* @NL80211_CMD_DEL_PATH: Remove a mesh path identified by %NL80211_ATTR_MAC
@@ -612,7 +616,7 @@
* consisting of a nested array.
*
* @NL80211_ATTR_MESH_ID: mesh id (1-32 bytes).
- * @NL80211_ATTR_PLINK_ACTION: action to perform on the mesh peer link.
+ * @NL80211_ATTR_STA_PLINK_ACTION: action to perform on the mesh peer link.
* @NL80211_ATTR_MPATH_NEXT_HOP: MAC address of the next hop for a mesh path.
* @NL80211_ATTR_MPATH_INFO: information about a mesh_path, part of mesh path
* info given for %NL80211_CMD_GET_MPATH, nested attribute described at
@@ -879,7 +883,9 @@
* See &enum nl80211_key_default_types.
*
* @NL80211_ATTR_MESH_SETUP: Optional mesh setup parameters. These cannot be
- * changed once the mesh is active.
+ * changed once the mesh is active.
+ * @NL80211_ATTR_MESH_CONFIG: Mesh configuration parameters, a nested attribute
+ * containing attributes from &enum nl80211_meshconf_params.
*
* @NL80211_ATTR_MAX: highest attribute number currently defined
* @__NL80211_ATTR_AFTER_LAST: internal use
@@ -1225,8 +1231,6 @@
* @NL80211_STA_INFO_INACTIVE_TIME: time since last activity (u32, msecs)
* @NL80211_STA_INFO_RX_BYTES: total received bytes (u32, from this station)
* @NL80211_STA_INFO_TX_BYTES: total transmitted bytes (u32, to this station)
- * @__NL80211_STA_INFO_AFTER_LAST: internal
- * @NL80211_STA_INFO_MAX: highest possible station info attribute
* @NL80211_STA_INFO_SIGNAL: signal strength of last received PPDU (u8, dBm)
* @NL80211_STA_INFO_TX_BITRATE: current unicast tx rate, nested attribute
* containing info as possible, see &enum nl80211_sta_info_txrate.
@@ -1236,6 +1240,11 @@
* @NL80211_STA_INFO_TX_RETRIES: total retries (u32, to this station)
* @NL80211_STA_INFO_TX_FAILED: total failed packets (u32, to this station)
* @NL80211_STA_INFO_SIGNAL_AVG: signal strength average (u8, dBm)
+ * @NL80211_STA_INFO_LLID: the station's mesh LLID
+ * @NL80211_STA_INFO_PLID: the station's mesh PLID
+ * @NL80211_STA_INFO_PLINK_STATE: peer link state for the station
+ * @__NL80211_STA_INFO_AFTER_LAST: internal
+ * @NL80211_STA_INFO_MAX: highest possible station info attribute
*/
enum nl80211_sta_info {
__NL80211_STA_INFO_INVALID,
@@ -1626,7 +1635,7 @@
* @NL80211_MESHCONF_HWMP_NET_DIAM_TRVS_TIME: The interval of time (in TUs)
* that it takes for an HWMP information element to propagate across the mesh
*
- * @NL80211_MESHCONF_ROOTMODE: whether root mode is enabled or not
+ * @NL80211_MESHCONF_HWMP_ROOTMODE: whether root mode is enabled or not
*
* @NL80211_MESHCONF_ELEMENT_TTL: specifies the value of TTL field set at a
* source mesh point for path selection elements.
@@ -1678,6 +1687,7 @@
* element that vendors will use to identify the path selection methods and
* metrics in use.
*
+ * @NL80211_MESH_SETUP_ATTR_MAX: highest possible mesh setup attribute number
* @__NL80211_MESH_SETUP_ATTR_AFTER_LAST: Internal use
*/
enum nl80211_mesh_setup_params {
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 5f38c46..0db8037 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -48,9 +48,6 @@
* struct page (these bits with information) are always mapped into kernel
* address space...
*
- * PG_buddy is set to indicate that the page is free and in the buddy system
- * (see mm/page_alloc.c).
- *
* PG_hwpoison indicates that a page got corrupted in hardware and contains
* data with incorrect ECC bits that triggered a machine check. Accessing is
* not safe since it may cause another machine check. Don't touch!
@@ -96,7 +93,6 @@
PG_swapcache, /* Swap page: swp_entry_t in private */
PG_mappedtodisk, /* Has blocks allocated on-disk */
PG_reclaim, /* To be reclaimed asap */
- PG_buddy, /* Page is free, on buddy lists */
PG_swapbacked, /* Page is backed by RAM/swap */
PG_unevictable, /* Page is "unevictable" */
#ifdef CONFIG_MMU
@@ -108,6 +104,9 @@
#ifdef CONFIG_MEMORY_FAILURE
PG_hwpoison, /* hardware poisoned page. Don't touch */
#endif
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ PG_compound_lock,
+#endif
__NR_PAGEFLAGS,
/* Filesystems */
@@ -198,7 +197,7 @@
struct page; /* forward declaration */
TESTPAGEFLAG(Locked, locked) TESTSETFLAG(Locked, locked)
-PAGEFLAG(Error, error)
+PAGEFLAG(Error, error) TESTCLEARFLAG(Error, error)
PAGEFLAG(Referenced, referenced) TESTCLEARFLAG(Referenced, referenced)
PAGEFLAG(Dirty, dirty) TESTSCFLAG(Dirty, dirty) __CLEARPAGEFLAG(Dirty, dirty)
PAGEFLAG(LRU, lru) __CLEARPAGEFLAG(LRU, lru)
@@ -230,7 +229,6 @@
* risky: they bypass page accounting.
*/
TESTPAGEFLAG(Writeback, writeback) TESTSCFLAG(Writeback, writeback)
-__PAGEFLAG(Buddy, buddy)
PAGEFLAG(MappedToDisk, mappedtodisk)
/* PG_readahead is only used for file reads; PG_reclaim is only for writes */
@@ -344,7 +342,7 @@
* tests can be used in performance sensitive paths. PageCompound is
* generally not used in hot code paths.
*/
-__PAGEFLAG(Head, head)
+__PAGEFLAG(Head, head) CLEARPAGEFLAG(Head, head)
__PAGEFLAG(Tail, tail)
static inline int PageCompound(struct page *page)
@@ -352,6 +350,13 @@
return page->flags & ((1L << PG_head) | (1L << PG_tail));
}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline void ClearPageCompound(struct page *page)
+{
+ BUG_ON(!PageHead(page));
+ ClearPageHead(page);
+}
+#endif
#else
/*
* Reduce page flag use as much as possible by overlapping
@@ -389,14 +394,61 @@
page->flags &= ~PG_head_tail_mask;
}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+static inline void ClearPageCompound(struct page *page)
+{
+ BUG_ON((page->flags & PG_head_tail_mask) != (1 << PG_compound));
+ clear_bit(PG_compound, &page->flags);
+}
+#endif
+
#endif /* !PAGEFLAGS_EXTENDED */
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+/*
+ * PageHuge() only returns true for hugetlbfs pages, but not for
+ * normal or transparent huge pages.
+ *
+ * PageTransHuge() returns true for both transparent huge and
+ * hugetlbfs pages, but not normal pages. PageTransHuge() can only be
+ * called only in the core VM paths where hugetlbfs pages can't exist.
+ */
+static inline int PageTransHuge(struct page *page)
+{
+ VM_BUG_ON(PageTail(page));
+ return PageHead(page);
+}
+
+static inline int PageTransCompound(struct page *page)
+{
+ return PageCompound(page);
+}
+
+#else
+
+static inline int PageTransHuge(struct page *page)
+{
+ return 0;
+}
+
+static inline int PageTransCompound(struct page *page)
+{
+ return 0;
+}
+#endif
+
#ifdef CONFIG_MMU
#define __PG_MLOCKED (1 << PG_mlocked)
#else
#define __PG_MLOCKED 0
#endif
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+#define __PG_COMPOUND_LOCK (1 << PG_compound_lock)
+#else
+#define __PG_COMPOUND_LOCK 0
+#endif
+
/*
* Flags checked when a page is freed. Pages being freed should not have
* these flags set. It they are, there is a problem.
@@ -404,9 +456,10 @@
#define PAGE_FLAGS_CHECK_AT_FREE \
(1 << PG_lru | 1 << PG_locked | \
1 << PG_private | 1 << PG_private_2 | \
- 1 << PG_buddy | 1 << PG_writeback | 1 << PG_reserved | \
+ 1 << PG_writeback | 1 << PG_reserved | \
1 << PG_slab | 1 << PG_swapcache | 1 << PG_active | \
- 1 << PG_unevictable | __PG_MLOCKED | __PG_HWPOISON)
+ 1 << PG_unevictable | __PG_MLOCKED | __PG_HWPOISON | \
+ __PG_COMPOUND_LOCK)
/*
* Flags checked when a page is prepped for return by the page allocator.
diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h
index b02195d..6d6cb7a 100644
--- a/include/linux/page_cgroup.h
+++ b/include/linux/page_cgroup.h
@@ -35,12 +35,15 @@
enum {
/* flags for mem_cgroup */
- PCG_LOCK, /* page cgroup is locked */
+ PCG_LOCK, /* Lock for pc->mem_cgroup and following bits. */
PCG_CACHE, /* charged as cache */
PCG_USED, /* this object is in use. */
- PCG_ACCT_LRU, /* page has been accounted for */
- PCG_FILE_MAPPED, /* page is accounted as "mapped" */
PCG_MIGRATION, /* under page migration */
+ /* flags for mem_cgroup and file and I/O status */
+ PCG_MOVE_LOCK, /* For race between move_account v.s. following bits */
+ PCG_FILE_MAPPED, /* page is accounted as "mapped" */
+ /* No lock in page_cgroup */
+ PCG_ACCT_LRU, /* page has been accounted for (under lru_lock) */
};
#define TESTPCGFLAG(uname, lname) \
@@ -94,6 +97,10 @@
static inline void lock_page_cgroup(struct page_cgroup *pc)
{
+ /*
+ * Don't take this lock in IRQ context.
+ * This lock is for pc->mem_cgroup, USED, CACHE, MIGRATION
+ */
bit_spin_lock(PCG_LOCK, &pc->flags);
}
@@ -107,6 +114,24 @@
return bit_spin_is_locked(PCG_LOCK, &pc->flags);
}
+static inline void move_lock_page_cgroup(struct page_cgroup *pc,
+ unsigned long *flags)
+{
+ /*
+ * We know updates to pc->flags of page cache's stats are from both of
+ * usual context or IRQ context. Disable IRQ to avoid deadlock.
+ */
+ local_irq_save(*flags);
+ bit_spin_lock(PCG_MOVE_LOCK, &pc->flags);
+}
+
+static inline void move_unlock_page_cgroup(struct page_cgroup *pc,
+ unsigned long *flags)
+{
+ bit_spin_unlock(PCG_MOVE_LOCK, &pc->flags);
+ local_irq_restore(*flags);
+}
+
#else /* CONFIG_CGROUP_MEM_RES_CTLR */
struct page_cgroup;
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 2d1ffe3..9c66e99 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -48,7 +48,7 @@
static inline int mapping_unevictable(struct address_space *mapping)
{
- if (likely(mapping))
+ if (mapping)
return test_bit(AS_UNEVICTABLE, &mapping->flags);
return !!mapping;
}
diff --git a/include/linux/path.h b/include/linux/path.h
index a581e8c..edc98de 100644
--- a/include/linux/path.h
+++ b/include/linux/path.h
@@ -10,9 +10,7 @@
};
extern void path_get(struct path *);
-extern void path_get_long(struct path *);
extern void path_put(struct path *);
-extern void path_put_long(struct path *);
static inline int path_equal(const struct path *path1, const struct path *path2)
{
diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h
index c8b6473..4462350 100644
--- a/include/linux/pci-acpi.h
+++ b/include/linux/pci-acpi.h
@@ -35,9 +35,12 @@
return acpi_get_pci_rootbridge_handle(pci_domain_nr(pbus),
pbus->number);
}
+#endif
+
+#ifdef CONFIG_ACPI_APEI
+extern bool aer_acpi_firmware_first(void);
#else
-static inline acpi_handle acpi_find_root_bridge_handle(struct pci_dev *pdev)
-{ return NULL; }
+static inline bool aer_acpi_firmware_first(void) { return false; }
#endif
#endif /* _PCI_ACPI_H_ */
diff --git a/include/linux/pci-aspm.h b/include/linux/pci-aspm.h
index 91ba0b3..ce681051 100644
--- a/include/linux/pci-aspm.h
+++ b/include/linux/pci-aspm.h
@@ -27,6 +27,7 @@
extern void pcie_aspm_exit_link_state(struct pci_dev *pdev);
extern void pcie_aspm_pm_state_change(struct pci_dev *pdev);
extern void pci_disable_link_state(struct pci_dev *pdev, int state);
+extern void pcie_clear_aspm(void);
extern void pcie_no_aspm(void);
#else
static inline void pcie_aspm_init_link_state(struct pci_dev *pdev)
@@ -41,7 +42,9 @@
static inline void pci_disable_link_state(struct pci_dev *pdev, int state)
{
}
-
+static inline void pcie_clear_aspm(void)
+{
+}
static inline void pcie_no_aspm(void)
{
}
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 7454408..559d028 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -806,7 +806,7 @@
/* Power management related routines */
int pci_save_state(struct pci_dev *dev);
-int pci_restore_state(struct pci_dev *dev);
+void pci_restore_state(struct pci_dev *dev);
int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state);
int pci_set_power_state(struct pci_dev *dev, pci_power_t state);
pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state);
@@ -820,7 +820,6 @@
int pci_back_from_sleep(struct pci_dev *dev);
bool pci_dev_run_wake(struct pci_dev *dev);
bool pci_check_pme_status(struct pci_dev *dev);
-void pci_wakeup_event(struct pci_dev *dev);
void pci_pme_wakeup_bus(struct pci_bus *bus);
static inline int pci_enable_wake(struct pci_dev *dev, pci_power_t state,
@@ -994,6 +993,14 @@
extern int pci_msi_enabled(void);
#endif
+#ifdef CONFIG_PCIEPORTBUS
+extern bool pcie_ports_disabled;
+extern bool pcie_ports_auto;
+#else
+#define pcie_ports_disabled true
+#define pcie_ports_auto false
+#endif
+
#ifndef CONFIG_PCIEASPM
static inline int pcie_aspm_enabled(void)
{
@@ -1003,6 +1010,14 @@
extern int pcie_aspm_enabled(void);
#endif
+#ifdef CONFIG_PCIEAER
+void pci_no_aer(void);
+bool pci_aer_available(void);
+#else
+static inline void pci_no_aer(void) { }
+static inline bool pci_aer_available(void) { return false; }
+#endif
+
#ifndef CONFIG_PCIE_ECRC
static inline void pcie_set_ecrc_checking(struct pci_dev *dev)
{
@@ -1168,10 +1183,8 @@
return 0;
}
-static inline int pci_restore_state(struct pci_dev *dev)
-{
- return 0;
-}
+static inline void pci_restore_state(struct pci_dev *dev)
+{ }
static inline int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
{
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index ae0dc45..3adb06e 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2478,7 +2478,8 @@
#define PCI_DEVICE_ID_INTEL_COUGARPOINT_LPC_MIN 0x1c41
#define PCI_DEVICE_ID_INTEL_COUGARPOINT_LPC_MAX 0x1c5f
#define PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS 0x1d22
-#define PCI_DEVICE_ID_INTEL_PATSBURG_LPC 0x1d40
+#define PCI_DEVICE_ID_INTEL_PATSBURG_LPC_0 0x1d40
+#define PCI_DEVICE_ID_INTEL_PATSBURG_LPC_1 0x1d41
#define PCI_DEVICE_ID_INTEL_82801AA_0 0x2410
#define PCI_DEVICE_ID_INTEL_82801AA_1 0x2411
#define PCI_DEVICE_ID_INTEL_82801AA_3 0x2413
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index af83076..5b7e6b1 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -309,6 +309,14 @@
#define PCI_MSIX_PBA 8
#define PCI_MSIX_FLAGS_BIRMASK (7 << 0)
+/* MSI-X entry's format */
+#define PCI_MSIX_ENTRY_SIZE 16
+#define PCI_MSIX_ENTRY_LOWER_ADDR 0
+#define PCI_MSIX_ENTRY_UPPER_ADDR 4
+#define PCI_MSIX_ENTRY_DATA 8
+#define PCI_MSIX_ENTRY_VECTOR_CTRL 12
+#define PCI_MSIX_ENTRY_CTRL_MASKBIT 1
+
/* CompactPCI Hotswap Register */
#define PCI_CHSWP_CSR 2 /* Control and Status Register */
@@ -496,6 +504,8 @@
#define PCI_EXP_RTCTL_CRSSVE 0x10 /* CRS Software Visibility Enable */
#define PCI_EXP_RTCAP 30 /* Root Capabilities */
#define PCI_EXP_RTSTA 32 /* Root Status */
+#define PCI_EXP_RTSTA_PME 0x10000 /* PME status */
+#define PCI_EXP_RTSTA_PENDING 0x20000 /* PME pending */
#define PCI_EXP_DEVCAP2 36 /* Device Capabilities 2 */
#define PCI_EXP_DEVCAP2_ARI 0x20 /* Alternative Routing-ID */
#define PCI_EXP_DEVCTL2 40 /* Device Control 2 */
diff --git a/include/linux/power/gpio-charger.h b/include/linux/power/gpio-charger.h
new file mode 100644
index 0000000..de1dfe0
--- /dev/null
+++ b/include/linux/power/gpio-charger.h
@@ -0,0 +1,41 @@
+/*
+ * Copyright (C) 2010, Lars-Peter Clausen <lars@metafoo.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.
+ *
+ * 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.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+#ifndef __LINUX_POWER_GPIO_CHARGER_H__
+#define __LINUX_POWER_GPIO_CHARGER_H__
+
+#include <linux/power_supply.h>
+#include <linux/types.h>
+
+/**
+ * struct gpio_charger_platform_data - platform_data for gpio_charger devices
+ * @name: Name for the chargers power_supply device
+ * @type: Type of the charger
+ * @gpio: GPIO which is used to indicate the chargers status
+ * @gpio_active_low: Should be set to 1 if the GPIO is active low otherwise 0
+ * @supplied_to: Array of battery names to which this chargers supplies power
+ * @num_supplicants: Number of entries in the supplied_to array
+ */
+struct gpio_charger_platform_data {
+ const char *name;
+ enum power_supply_type type;
+
+ int gpio;
+ int gpio_active_low;
+
+ char **supplied_to;
+ size_t num_supplicants;
+};
+
+#endif
diff --git a/include/linux/power/max17042_battery.h b/include/linux/power/max17042_battery.h
new file mode 100644
index 0000000..7995deb
--- /dev/null
+++ b/include/linux/power/max17042_battery.h
@@ -0,0 +1,30 @@
+/*
+ * Fuel gauge driver for Maxim 17042 / 8966 / 8997
+ * Note that Maxim 8966 and 8997 are mfd and this is its subdevice.
+ *
+ * Copyright (C) 2011 Samsung Electronics
+ * MyungJoo Ham <myungjoo.ham@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.
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#ifndef __MAX17042_BATTERY_H_
+#define __MAX17042_BATTERY_H_
+
+struct max17042_platform_data {
+ bool enable_current_sense;
+};
+
+#endif /* __MAX17042_BATTERY_H_ */
diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h
index ab2baa5..23241c2 100644
--- a/include/linux/radix-tree.h
+++ b/include/linux/radix-tree.h
@@ -146,6 +146,22 @@
}
/**
+ * radix_tree_deref_slot_protected - dereference a slot without RCU lock but with tree lock held
+ * @pslot: pointer to slot, returned by radix_tree_lookup_slot
+ * Returns: item that was stored in that slot with any direct pointer flag
+ * removed.
+ *
+ * Similar to radix_tree_deref_slot but only used during migration when a pages
+ * mapping is being moved. The caller does not hold the RCU read lock but it
+ * must hold the tree lock to prevent parallel updates.
+ */
+static inline void *radix_tree_deref_slot_protected(void **pslot,
+ spinlock_t *treelock)
+{
+ return rcu_dereference_protected(*pslot, lockdep_is_held(treelock));
+}
+
+/**
* radix_tree_deref_retry - check radix_tree_deref_slot
* @arg: pointer returned by radix_tree_deref_slot
* Returns: 0 if retry is not required, otherwise retry is required
diff --git a/include/linux/rculist_bl.h b/include/linux/rculist_bl.h
index b872b49..cf1244f 100644
--- a/include/linux/rculist_bl.h
+++ b/include/linux/rculist_bl.h
@@ -11,7 +11,8 @@
struct hlist_bl_node *n)
{
LIST_BL_BUG_ON((unsigned long)n & LIST_BL_LOCKMASK);
- LIST_BL_BUG_ON(!((unsigned long)h->first & LIST_BL_LOCKMASK));
+ LIST_BL_BUG_ON(((unsigned long)h->first & LIST_BL_LOCKMASK) !=
+ LIST_BL_LOCKMASK);
rcu_assign_pointer(h->first,
(struct hlist_bl_node *)((unsigned long)n | LIST_BL_LOCKMASK));
}
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index bb83c0d..e9fd04c 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -198,6 +198,8 @@
};
#define TTU_ACTION(x) ((x) & TTU_ACTION_MASK)
+bool is_vma_temporary_stack(struct vm_area_struct *vma);
+
int try_to_unmap(struct page *, enum ttu_flags flags);
int try_to_unmap_one(struct page *, struct vm_area_struct *,
unsigned long address, enum ttu_flags flags);
diff --git a/include/linux/s3c_adc_battery.h b/include/linux/s3c_adc_battery.h
index dbce22f..fbe58b7 100644
--- a/include/linux/s3c_adc_battery.h
+++ b/include/linux/s3c_adc_battery.h
@@ -14,6 +14,7 @@
void (*disable_charger)(void);
int gpio_charge_finished;
+ int gpio_inverted;
const struct s3c_adc_bat_thresh *lut_noac;
unsigned int lut_noac_cnt;
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 96e2321..d747f94 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -21,7 +21,8 @@
#define CLONE_DETACHED 0x00400000 /* Unused, ignored */
#define CLONE_UNTRACED 0x00800000 /* set if the tracing process can't force CLONE_PTRACE on this clone */
#define CLONE_CHILD_SETTID 0x01000000 /* set the TID in the child */
-#define CLONE_STOPPED 0x02000000 /* Start in stopped state */
+/* 0x02000000 was previously the unused CLONE_STOPPED (Start in stopped state)
+ and is now available for re-use. */
#define CLONE_NEWUTS 0x04000000 /* New utsname group? */
#define CLONE_NEWIPC 0x08000000 /* New ipcs */
#define CLONE_NEWUSER 0x10000000 /* New user namespace */
@@ -433,6 +434,7 @@
#endif
/* leave room for more dump flags */
#define MMF_VM_MERGEABLE 16 /* KSM may merge identical pages */
+#define MMF_VM_HUGEPAGE 17 /* set when VM_HUGEPAGE is set on vma */
#define MMF_INIT_MASK (MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK)
@@ -633,6 +635,8 @@
int oom_adj; /* OOM kill score adjustment (bit shift) */
int oom_score_adj; /* OOM kill score adjustment */
+ int oom_score_adj_min; /* OOM kill score adjustment minimum value.
+ * Only settable by CAP_SYS_RESOURCE. */
struct mutex cred_guard_mutex; /* guard against foreign influences on
* credential calculations
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 20ec0a64..bf221d6 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -255,6 +255,11 @@
typedef unsigned char *sk_buff_data_t;
#endif
+#if defined(CONFIG_NF_DEFRAG_IPV4) || defined(CONFIG_NF_DEFRAG_IPV4_MODULE) || \
+ defined(CONFIG_NF_DEFRAG_IPV6) || defined(CONFIG_NF_DEFRAG_IPV6_MODULE)
+#define NET_SKBUFF_NF_DEFRAG_NEEDED 1
+#endif
+
/**
* struct sk_buff - socket buffer
* @next: Next buffer in list
@@ -362,6 +367,8 @@
void (*destructor)(struct sk_buff *skb);
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
struct nf_conntrack *nfct;
+#endif
+#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
struct sk_buff *nfct_reasm;
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
@@ -2057,6 +2064,8 @@
if (nfct)
atomic_inc(&nfct->use);
}
+#endif
+#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
static inline void nf_conntrack_get_reasm(struct sk_buff *skb)
{
if (skb)
@@ -2085,6 +2094,8 @@
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
nf_conntrack_put(skb->nfct);
skb->nfct = NULL;
+#endif
+#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
nf_conntrack_put_reasm(skb->nfct_reasm);
skb->nfct_reasm = NULL;
#endif
@@ -2101,6 +2112,8 @@
dst->nfct = src->nfct;
nf_conntrack_get(src->nfct);
dst->nfctinfo = src->nfctinfo;
+#endif
+#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
dst->nfct_reasm = src->nfct_reasm;
nf_conntrack_get_reasm(src->nfct_reasm);
#endif
@@ -2114,6 +2127,8 @@
{
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
nf_conntrack_put(dst->nfct);
+#endif
+#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
nf_conntrack_put_reasm(dst->nfct_reasm);
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
diff --git a/include/linux/sunrpc/cache.h b/include/linux/sunrpc/cache.h
index 78aa104..7898ea1 100644
--- a/include/linux/sunrpc/cache.h
+++ b/include/linux/sunrpc/cache.h
@@ -256,10 +256,13 @@
return rv - boot.tv_sec;
}
+#ifdef CONFIG_NFSD_DEPRECATED
static inline void sunrpc_invalidate(struct cache_head *h,
struct cache_detail *detail)
{
h->expiry_time = seconds_since_boot() - 1;
detail->nextcheck = seconds_since_boot();
}
+#endif /* CONFIG_NFSD_DEPRECATED */
+
#endif /* _LINUX_SUNRPC_CACHE_H_ */
diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index c81d4d8..ea29330 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -269,6 +269,7 @@
struct cache_req rq_chandle; /* handle passed to caches for
* request delaying
*/
+ bool rq_dropme;
/* Catering to nfsd */
struct auth_domain * rq_client; /* RPC peer info */
struct auth_domain * rq_gssclient; /* "gss/"-style peer info */
diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h
index 357da5e..059877b 100644
--- a/include/linux/sunrpc/svc_xprt.h
+++ b/include/linux/sunrpc/svc_xprt.h
@@ -63,7 +63,6 @@
#define XPT_LISTENER 11 /* listening endpoint */
#define XPT_CACHE_AUTH 12 /* cache auth info */
- struct svc_pool *xpt_pool; /* current pool iff queued */
struct svc_serv *xpt_server; /* service for transport */
atomic_t xpt_reserved; /* space on outq that is rsvd */
struct mutex xpt_mutex; /* to serialize sending data */
@@ -81,6 +80,7 @@
void *xpt_bc_sid; /* back channel session ID */
struct net *xpt_net;
+ struct rpc_xprt *xpt_bc_xprt; /* NFSv4.1 backchannel */
};
static inline void unregister_xpt_user(struct svc_xprt *xpt, struct svc_xpt_user *u)
diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h
index 1b353a7..04dba23 100644
--- a/include/linux/sunrpc/svcsock.h
+++ b/include/linux/sunrpc/svcsock.h
@@ -28,7 +28,6 @@
/* private TCP part */
u32 sk_reclen; /* length of record */
u32 sk_tcplen; /* current read length */
- struct rpc_xprt *sk_bc_xprt; /* NFSv4.1 backchannel xprt */
};
/*
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 89d10d2..bef0f53 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -321,6 +321,7 @@
#define XPRT_CLOSING (6)
#define XPRT_CONNECTION_ABORT (7)
#define XPRT_CONNECTION_CLOSE (8)
+#define XPRT_INITIALIZED (9)
static inline void xprt_set_connected(struct rpc_xprt *xprt)
{
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index c1f4998..5a89e36 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -258,23 +258,6 @@
static inline bool system_entering_hibernation(void) { return false; }
#endif /* CONFIG_HIBERNATION */
-#ifdef CONFIG_SUSPEND_NVS
-extern int suspend_nvs_register(unsigned long start, unsigned long size);
-extern int suspend_nvs_alloc(void);
-extern void suspend_nvs_free(void);
-extern void suspend_nvs_save(void);
-extern void suspend_nvs_restore(void);
-#else /* CONFIG_SUSPEND_NVS */
-static inline int suspend_nvs_register(unsigned long a, unsigned long b)
-{
- return 0;
-}
-static inline int suspend_nvs_alloc(void) { return 0; }
-static inline void suspend_nvs_free(void) {}
-static inline void suspend_nvs_save(void) {}
-static inline void suspend_nvs_restore(void) {}
-#endif /* CONFIG_SUSPEND_NVS */
-
#ifdef CONFIG_PM_SLEEP
void save_processor_state(void);
void restore_processor_state(void);
diff --git a/include/linux/swap.h b/include/linux/swap.h
index eba53e7..4d55932 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -208,6 +208,8 @@
/* linux/mm/swap.c */
extern void __lru_cache_add(struct page *, enum lru_list lru);
extern void lru_cache_add_lru(struct page *, enum lru_list lru);
+extern void lru_add_page_tail(struct zone* zone,
+ struct page *page, struct page *page_tail);
extern void activate_page(struct page *);
extern void mark_page_accessed(struct page *);
extern void lru_add_drain(void);
diff --git a/include/linux/thermal.h b/include/linux/thermal.h
index 1de8b9eb..8651556 100644
--- a/include/linux/thermal.h
+++ b/include/linux/thermal.h
@@ -77,7 +77,7 @@
char type[THERMAL_NAME_LENGTH];
struct device device;
void *devdata;
- struct thermal_cooling_device_ops *ops;
+ const struct thermal_cooling_device_ops *ops;
struct list_head node;
};
@@ -114,7 +114,7 @@
int last_temperature;
bool passive;
unsigned int forced_passive;
- struct thermal_zone_device_ops *ops;
+ const struct thermal_zone_device_ops *ops;
struct list_head cooling_devices;
struct idr idr;
struct mutex lock; /* protect cooling devices list */
@@ -127,13 +127,41 @@
struct thermal_hwmon_attr temp_crit; /* hwmon sys attr */
#endif
};
+/* Adding event notification support elements */
+#define THERMAL_GENL_FAMILY_NAME "thermal_event"
+#define THERMAL_GENL_VERSION 0x01
+#define THERMAL_GENL_MCAST_GROUP_NAME "thermal_mc_group"
+
+enum events {
+ THERMAL_AUX0,
+ THERMAL_AUX1,
+ THERMAL_CRITICAL,
+ THERMAL_DEV_FAULT,
+};
+
+struct thermal_genl_event {
+ u32 orig;
+ enum events event;
+};
+/* attributes of thermal_genl_family */
+enum {
+ THERMAL_GENL_ATTR_UNSPEC,
+ THERMAL_GENL_ATTR_EVENT,
+ __THERMAL_GENL_ATTR_MAX,
+};
+#define THERMAL_GENL_ATTR_MAX (__THERMAL_GENL_ATTR_MAX - 1)
+
+/* commands supported by the thermal_genl_family */
+enum {
+ THERMAL_GENL_CMD_UNSPEC,
+ THERMAL_GENL_CMD_EVENT,
+ __THERMAL_GENL_CMD_MAX,
+};
+#define THERMAL_GENL_CMD_MAX (__THERMAL_GENL_CMD_MAX - 1)
struct thermal_zone_device *thermal_zone_device_register(char *, int, void *,
- struct
- thermal_zone_device_ops
- *, int tc1, int tc2,
- int passive_freq,
- int polling_freq);
+ const struct thermal_zone_device_ops *, int tc1, int tc2,
+ int passive_freq, int polling_freq);
void thermal_zone_device_unregister(struct thermal_zone_device *);
int thermal_zone_bind_cooling_device(struct thermal_zone_device *, int,
@@ -142,9 +170,8 @@
struct thermal_cooling_device *);
void thermal_zone_device_update(struct thermal_zone_device *);
struct thermal_cooling_device *thermal_cooling_device_register(char *, void *,
- struct
- thermal_cooling_device_ops
- *);
+ const struct thermal_cooling_device_ops *);
void thermal_cooling_device_unregister(struct thermal_cooling_device *);
+extern int generate_netlink_event(u32 orig, enum events event);
#endif /* __THERMAL_H__ */
diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index 44b54f6..4ed6fcd 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -59,8 +59,9 @@
extern void *vmalloc_32(unsigned long size);
extern void *vmalloc_32_user(unsigned long size);
extern void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot);
-extern void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask,
- pgprot_t prot);
+extern void *__vmalloc_node_range(unsigned long size, unsigned long align,
+ unsigned long start, unsigned long end, gfp_t gfp_mask,
+ pgprot_t prot, int node, void *caller);
extern void vfree(const void *addr);
extern void *vmap(struct page **pages, unsigned int count,
@@ -90,9 +91,6 @@
unsigned long flags,
unsigned long start, unsigned long end,
void *caller);
-extern struct vm_struct *get_vm_area_node(unsigned long size,
- unsigned long flags, int node,
- gfp_t gfp_mask);
extern struct vm_struct *remove_vm_area(const void *addr);
extern int map_vm_area(struct vm_struct *area, pgprot_t prot,
@@ -120,7 +118,7 @@
#ifdef CONFIG_SMP
struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
const size_t *sizes, int nr_vms,
- size_t align, gfp_t gfp_mask);
+ size_t align);
void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms);
#endif
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index eaaea37..833e676 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -254,6 +254,11 @@
extern void __dec_zone_state(struct zone *, enum zone_stat_item);
void refresh_cpu_vm_stats(int);
+
+int calculate_pressure_threshold(struct zone *zone);
+int calculate_normal_threshold(struct zone *zone);
+void set_pgdat_percpu_threshold(pg_data_t *pgdat,
+ int (*calculate_pressure)(struct zone *));
#else /* CONFIG_SMP */
/*
@@ -298,6 +303,8 @@
#define dec_zone_page_state __dec_zone_page_state
#define mod_zone_page_state __mod_zone_page_state
+#define set_pgdat_percpu_threshold(pgdat, callback) { }
+
static inline void refresh_cpu_vm_stats(int cpu) { }
#endif
diff --git a/include/net/ah.h b/include/net/ah.h
index be7798d..ca95b98 100644
--- a/include/net/ah.h
+++ b/include/net/ah.h
@@ -4,7 +4,7 @@
#include <linux/skbuff.h>
/* This is the maximum truncated ICV length that we know of. */
-#define MAX_AH_AUTH_LEN 16
+#define MAX_AH_AUTH_LEN 64
struct crypto_ahash;
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index bcc9f44..1322695 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -1103,6 +1103,8 @@
* @change_mpath: change a given mesh path
* @get_mpath: get a mesh path for the given parameters
* @dump_mpath: dump mesh path callback -- resume dump at index @idx
+ * @join_mesh: join the mesh network with the specified parameters
+ * @leave_mesh: leave the current mesh network
*
* @get_mesh_config: Get the current mesh configuration
*
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 5b3fd5a..62c0ce2 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -337,6 +337,10 @@
* @IEEE80211_TX_CTL_LDPC: tells the driver to use LDPC for this frame
* @IEEE80211_TX_CTL_STBC: Enables Space-Time Block Coding (STBC) for this
* frame and selects the maximum number of streams that it can use.
+ * @IEEE80211_TX_CTL_TX_OFFCHAN: Marks this packet to be transmitted on
+ * the off-channel channel when a remain-on-channel offload is done
+ * in hardware -- normal packets still flow and are expected to be
+ * handled properly by the device.
*
* Note: If you have to add new flags to the enumeration, then don't
* forget to update %IEEE80211_TX_TEMPORARY_FLAGS when necessary.
@@ -1753,6 +1757,16 @@
* (also see nl80211.h @NL80211_ATTR_WIPHY_ANTENNA_TX).
*
* @get_antenna: Get current antenna configuration from device (tx_ant, rx_ant).
+ *
+ * @remain_on_channel: Starts an off-channel period on the given channel, must
+ * call back to ieee80211_ready_on_channel() when on that channel. Note
+ * that normal channel traffic is not stopped as this is intended for hw
+ * offload. Frames to transmit on the off-channel channel are transmitted
+ * normally except for the %IEEE80211_TX_CTL_TX_OFFCHAN flag. When the
+ * duration (which will always be non-zero) expires, the driver must call
+ * ieee80211_remain_on_channel_expired(). This callback may sleep.
+ * @cancel_remain_on_channel: Requests that an ongoing off-channel period is
+ * aborted before it expires. This callback may sleep.
*/
struct ieee80211_ops {
int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb);
diff --git a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
index 1ee717e..a4c9936 100644
--- a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
+++ b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h
@@ -7,16 +7,6 @@
extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6;
extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmpv6;
-extern int nf_ct_frag6_init(void);
-extern void nf_ct_frag6_cleanup(void);
-extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb, u32 user);
-extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb,
- struct net_device *in,
- struct net_device *out,
- int (*okfn)(struct sk_buff *));
-
-struct inet_frags_ctl;
-
#include <linux/sysctl.h>
extern struct ctl_table nf_ct_ipv6_sysctl_table[];
diff --git a/include/net/netfilter/ipv6/nf_defrag_ipv6.h b/include/net/netfilter/ipv6/nf_defrag_ipv6.h
index 94dd54d..fd79c9a 100644
--- a/include/net/netfilter/ipv6/nf_defrag_ipv6.h
+++ b/include/net/netfilter/ipv6/nf_defrag_ipv6.h
@@ -3,4 +3,14 @@
extern void nf_defrag_ipv6_enable(void);
+extern int nf_ct_frag6_init(void);
+extern void nf_ct_frag6_cleanup(void);
+extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb, u32 user);
+extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb,
+ struct net_device *in,
+ struct net_device *out,
+ int (*okfn)(struct sk_buff *));
+
+struct inet_frags_ctl;
+
#endif /* _NF_DEFRAG_IPV6_H */
diff --git a/include/net/red.h b/include/net/red.h
index 995108e..3319f16 100644
--- a/include/net/red.h
+++ b/include/net/red.h
@@ -97,7 +97,6 @@
u32 forced_mark; /* Forced marks, qavg > max_thresh */
u32 pdrop; /* Drops due to queue limits */
u32 other; /* Drops due to drop() calls */
- u32 backlog;
};
struct red_parms {
diff --git a/include/target/configfs_macros.h b/include/target/configfs_macros.h
new file mode 100644
index 0000000..7fe7460
--- /dev/null
+++ b/include/target/configfs_macros.h
@@ -0,0 +1,147 @@
+/* -*- mode: c; c-basic-offset: 8; -*-
+ * vim: noexpandtab sw=8 ts=8 sts=0:
+ *
+ * configfs_macros.h - extends macros for configfs
+ *
+ * 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.
+ *
+ * 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, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ *
+ * Based on sysfs:
+ * sysfs is Copyright (C) 2001, 2002, 2003 Patrick Mochel
+ *
+ * Based on kobject.h:
+ * Copyright (c) 2002-2003 Patrick Mochel
+ * Copyright (c) 2002-2003 Open Source Development Labs
+ *
+ * configfs Copyright (C) 2005 Oracle. All rights reserved.
+ *
+ * Added CONFIGFS_EATTR() macros from original configfs.h macros
+ * Copright (C) 2008-2009 Nicholas A. Bellinger <nab@linux-iscsi.org>
+ *
+ * Please read Documentation/filesystems/configfs.txt before using the
+ * configfs interface, ESPECIALLY the parts about reference counts and
+ * item destructors.
+ */
+
+#ifndef _CONFIGFS_MACROS_H_
+#define _CONFIGFS_MACROS_H_
+
+#include <linux/configfs.h>
+
+/*
+ * Users often need to create attribute structures for their configurable
+ * attributes, containing a configfs_attribute member and function pointers
+ * for the show() and store() operations on that attribute. If they don't
+ * need anything else on the extended attribute structure, they can use
+ * this macro to define it. The argument _name isends up as
+ * 'struct _name_attribute, as well as names of to CONFIGFS_ATTR_OPS() below.
+ * The argument _item is the name of the structure containing the
+ * struct config_item or struct config_group structure members
+ */
+#define CONFIGFS_EATTR_STRUCT(_name, _item) \
+struct _name##_attribute { \
+ struct configfs_attribute attr; \
+ ssize_t (*show)(struct _item *, char *); \
+ ssize_t (*store)(struct _item *, const char *, size_t); \
+}
+
+/*
+ * With the extended attribute structure, users can use this macro
+ * (similar to sysfs' __ATTR) to make defining attributes easier.
+ * An example:
+ * #define MYITEM_EATTR(_name, _mode, _show, _store) \
+ * struct myitem_attribute childless_attr_##_name = \
+ * __CONFIGFS_EATTR(_name, _mode, _show, _store)
+ */
+#define __CONFIGFS_EATTR(_name, _mode, _show, _store) \
+{ \
+ .attr = { \
+ .ca_name = __stringify(_name), \
+ .ca_mode = _mode, \
+ .ca_owner = THIS_MODULE, \
+ }, \
+ .show = _show, \
+ .store = _store, \
+}
+/* Here is a readonly version, only requiring a show() operation */
+#define __CONFIGFS_EATTR_RO(_name, _show) \
+{ \
+ .attr = { \
+ .ca_name = __stringify(_name), \
+ .ca_mode = 0444, \
+ .ca_owner = THIS_MODULE, \
+ }, \
+ .show = _show, \
+}
+
+/*
+ * With these extended attributes, the simple show_attribute() and
+ * store_attribute() operations need to call the show() and store() of the
+ * attributes. This is a common pattern, so we provide a macro to define
+ * them. The argument _name is the name of the attribute defined by
+ * CONFIGFS_ATTR_STRUCT(). The argument _item is the name of the structure
+ * containing the struct config_item or struct config_group structure member.
+ * The argument _item_member is the actual name of the struct config_* struct
+ * in your _item structure. Meaning my_structure->some_config_group.
+ * ^^_item^^^^^ ^^_item_member^^^
+ * This macro expects the attributes to be named "struct <name>_attribute".
+ */
+#define CONFIGFS_EATTR_OPS_TO_FUNC(_name, _item, _item_member) \
+static struct _item *to_##_name(struct config_item *ci) \
+{ \
+ return (ci) ? container_of(to_config_group(ci), struct _item, \
+ _item_member) : NULL; \
+}
+
+#define CONFIGFS_EATTR_OPS_SHOW(_name, _item) \
+static ssize_t _name##_attr_show(struct config_item *item, \
+ struct configfs_attribute *attr, \
+ char *page) \
+{ \
+ struct _item *_item = to_##_name(item); \
+ struct _name##_attribute * _name##_attr = \
+ container_of(attr, struct _name##_attribute, attr); \
+ ssize_t ret = 0; \
+ \
+ if (_name##_attr->show) \
+ ret = _name##_attr->show(_item, page); \
+ return ret; \
+}
+
+#define CONFIGFS_EATTR_OPS_STORE(_name, _item) \
+static ssize_t _name##_attr_store(struct config_item *item, \
+ struct configfs_attribute *attr, \
+ const char *page, size_t count) \
+{ \
+ struct _item *_item = to_##_name(item); \
+ struct _name##_attribute * _name##_attr = \
+ container_of(attr, struct _name##_attribute, attr); \
+ ssize_t ret = -EINVAL; \
+ \
+ if (_name##_attr->store) \
+ ret = _name##_attr->store(_item, page, count); \
+ return ret; \
+}
+
+#define CONFIGFS_EATTR_OPS(_name, _item, _item_member) \
+ CONFIGFS_EATTR_OPS_TO_FUNC(_name, _item, _item_member); \
+ CONFIGFS_EATTR_OPS_SHOW(_name, _item); \
+ CONFIGFS_EATTR_OPS_STORE(_name, _item);
+
+#define CONFIGFS_EATTR_OPS_RO(_name, _item, _item_member) \
+ CONFIGFS_EATTR_OPS_TO_FUNC(_name, _item, _item_member); \
+ CONFIGFS_EATTR_OPS_SHOW(_name, _item);
+
+#endif /* _CONFIGFS_MACROS_H_ */
diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h
new file mode 100644
index 0000000..07fdfb6
--- /dev/null
+++ b/include/target/target_core_base.h
@@ -0,0 +1,937 @@
+#ifndef TARGET_CORE_BASE_H
+#define TARGET_CORE_BASE_H
+
+#include <linux/in.h>
+#include <linux/configfs.h>
+#include <linux/dma-mapping.h>
+#include <linux/blkdev.h>
+#include <scsi/scsi_cmnd.h>
+#include <net/sock.h>
+#include <net/tcp.h>
+#include "target_core_mib.h"
+
+#define TARGET_CORE_MOD_VERSION "v4.0.0-rc6"
+#define SHUTDOWN_SIGS (sigmask(SIGKILL)|sigmask(SIGINT)|sigmask(SIGABRT))
+
+/* Used by transport_generic_allocate_iovecs() */
+#define TRANSPORT_IOV_DATA_BUFFER 5
+/* Maximum Number of LUNs per Target Portal Group */
+#define TRANSPORT_MAX_LUNS_PER_TPG 256
+/*
+ * By default we use 32-byte CDBs in TCM Core and subsystem plugin code.
+ *
+ * Note that both include/scsi/scsi_cmnd.h:MAX_COMMAND_SIZE and
+ * include/linux/blkdev.h:BLOCK_MAX_CDB as of v2.6.36-rc4 still use
+ * 16-byte CDBs by default and require an extra allocation for
+ * 32-byte CDBs to becasue of legacy issues.
+ *
+ * Within TCM Core there are no such legacy limitiations, so we go ahead
+ * use 32-byte CDBs by default and use include/scsi/scsi.h:scsi_command_size()
+ * within all TCM Core and subsystem plugin code.
+ */
+#define TCM_MAX_COMMAND_SIZE 32
+/*
+ * From include/scsi/scsi_cmnd.h:SCSI_SENSE_BUFFERSIZE, currently
+ * defined 96, but the real limit is 252 (or 260 including the header)
+ */
+#define TRANSPORT_SENSE_BUFFER SCSI_SENSE_BUFFERSIZE
+/* Used by transport_send_check_condition_and_sense() */
+#define SPC_SENSE_KEY_OFFSET 2
+#define SPC_ASC_KEY_OFFSET 12
+#define SPC_ASCQ_KEY_OFFSET 13
+#define TRANSPORT_IQN_LEN 224
+/* Used by target_core_store_alua_lu_gp() and target_core_alua_lu_gp_show_attr_members() */
+#define LU_GROUP_NAME_BUF 256
+/* Used by core_alua_store_tg_pt_gp_info() and target_core_alua_tg_pt_gp_show_attr_members() */
+#define TG_PT_GROUP_NAME_BUF 256
+/* Used to parse VPD into struct t10_vpd */
+#define VPD_TMP_BUF_SIZE 128
+/* Used by transport_generic_cmd_sequencer() */
+#define READ_BLOCK_LEN 6
+#define READ_CAP_LEN 8
+#define READ_POSITION_LEN 20
+#define INQUIRY_LEN 36
+/* Used by transport_get_inquiry_vpd_serial() */
+#define INQUIRY_VPD_SERIAL_LEN 254
+/* Used by transport_get_inquiry_vpd_device_ident() */
+#define INQUIRY_VPD_DEVICE_IDENTIFIER_LEN 254
+
+/* struct se_hba->hba_flags */
+enum hba_flags_table {
+ HBA_FLAGS_INTERNAL_USE = 0x01,
+ HBA_FLAGS_PSCSI_MODE = 0x02,
+};
+
+/* struct se_lun->lun_status */
+enum transport_lun_status_table {
+ TRANSPORT_LUN_STATUS_FREE = 0,
+ TRANSPORT_LUN_STATUS_ACTIVE = 1,
+};
+
+/* struct se_portal_group->se_tpg_type */
+enum transport_tpg_type_table {
+ TRANSPORT_TPG_TYPE_NORMAL = 0,
+ TRANSPORT_TPG_TYPE_DISCOVERY = 1,
+};
+
+/* Used for generate timer flags */
+enum timer_flags_table {
+ TF_RUNNING = 0x01,
+ TF_STOP = 0x02,
+};
+
+/* Special transport agnostic struct se_cmd->t_states */
+enum transport_state_table {
+ TRANSPORT_NO_STATE = 0,
+ TRANSPORT_NEW_CMD = 1,
+ TRANSPORT_DEFERRED_CMD = 2,
+ TRANSPORT_WRITE_PENDING = 3,
+ TRANSPORT_PROCESS_WRITE = 4,
+ TRANSPORT_PROCESSING = 5,
+ TRANSPORT_COMPLETE_OK = 6,
+ TRANSPORT_COMPLETE_FAILURE = 7,
+ TRANSPORT_COMPLETE_TIMEOUT = 8,
+ TRANSPORT_PROCESS_TMR = 9,
+ TRANSPORT_TMR_COMPLETE = 10,
+ TRANSPORT_ISTATE_PROCESSING = 11,
+ TRANSPORT_ISTATE_PROCESSED = 12,
+ TRANSPORT_KILL = 13,
+ TRANSPORT_REMOVE = 14,
+ TRANSPORT_FREE = 15,
+ TRANSPORT_NEW_CMD_MAP = 16,
+};
+
+/* Used for struct se_cmd->se_cmd_flags */
+enum se_cmd_flags_table {
+ SCF_SUPPORTED_SAM_OPCODE = 0x00000001,
+ SCF_TRANSPORT_TASK_SENSE = 0x00000002,
+ SCF_EMULATED_TASK_SENSE = 0x00000004,
+ SCF_SCSI_DATA_SG_IO_CDB = 0x00000008,
+ SCF_SCSI_CONTROL_SG_IO_CDB = 0x00000010,
+ SCF_SCSI_CONTROL_NONSG_IO_CDB = 0x00000020,
+ SCF_SCSI_NON_DATA_CDB = 0x00000040,
+ SCF_SCSI_CDB_EXCEPTION = 0x00000080,
+ SCF_SCSI_RESERVATION_CONFLICT = 0x00000100,
+ SCF_CMD_PASSTHROUGH_NOALLOC = 0x00000200,
+ SCF_SE_CMD_FAILED = 0x00000400,
+ SCF_SE_LUN_CMD = 0x00000800,
+ SCF_SE_ALLOW_EOO = 0x00001000,
+ SCF_SE_DISABLE_ONLINE_CHECK = 0x00002000,
+ SCF_SENT_CHECK_CONDITION = 0x00004000,
+ SCF_OVERFLOW_BIT = 0x00008000,
+ SCF_UNDERFLOW_BIT = 0x00010000,
+ SCF_SENT_DELAYED_TAS = 0x00020000,
+ SCF_ALUA_NON_OPTIMIZED = 0x00040000,
+ SCF_DELAYED_CMD_FROM_SAM_ATTR = 0x00080000,
+ SCF_PASSTHROUGH_SG_TO_MEM = 0x00100000,
+ SCF_PASSTHROUGH_CONTIG_TO_SG = 0x00200000,
+ SCF_PASSTHROUGH_SG_TO_MEM_NOALLOC = 0x00400000,
+ SCF_EMULATE_SYNC_CACHE = 0x00800000,
+ SCF_EMULATE_CDB_ASYNC = 0x01000000,
+ SCF_EMULATE_SYNC_UNMAP = 0x02000000
+};
+
+/* struct se_dev_entry->lun_flags and struct se_lun->lun_access */
+enum transport_lunflags_table {
+ TRANSPORT_LUNFLAGS_NO_ACCESS = 0x00,
+ TRANSPORT_LUNFLAGS_INITIATOR_ACCESS = 0x01,
+ TRANSPORT_LUNFLAGS_READ_ONLY = 0x02,
+ TRANSPORT_LUNFLAGS_READ_WRITE = 0x04,
+};
+
+/* struct se_device->dev_status */
+enum transport_device_status_table {
+ TRANSPORT_DEVICE_ACTIVATED = 0x01,
+ TRANSPORT_DEVICE_DEACTIVATED = 0x02,
+ TRANSPORT_DEVICE_QUEUE_FULL = 0x04,
+ TRANSPORT_DEVICE_SHUTDOWN = 0x08,
+ TRANSPORT_DEVICE_OFFLINE_ACTIVATED = 0x10,
+ TRANSPORT_DEVICE_OFFLINE_DEACTIVATED = 0x20,
+};
+
+/*
+ * Used by transport_send_check_condition_and_sense() and se_cmd->scsi_sense_reason
+ * to signal which ASC/ASCQ sense payload should be built.
+ */
+enum tcm_sense_reason_table {
+ TCM_NON_EXISTENT_LUN = 0x01,
+ TCM_UNSUPPORTED_SCSI_OPCODE = 0x02,
+ TCM_INCORRECT_AMOUNT_OF_DATA = 0x03,
+ TCM_UNEXPECTED_UNSOLICITED_DATA = 0x04,
+ TCM_SERVICE_CRC_ERROR = 0x05,
+ TCM_SNACK_REJECTED = 0x06,
+ TCM_SECTOR_COUNT_TOO_MANY = 0x07,
+ TCM_INVALID_CDB_FIELD = 0x08,
+ TCM_INVALID_PARAMETER_LIST = 0x09,
+ TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE = 0x0a,
+ TCM_UNKNOWN_MODE_PAGE = 0x0b,
+ TCM_WRITE_PROTECTED = 0x0c,
+ TCM_CHECK_CONDITION_ABORT_CMD = 0x0d,
+ TCM_CHECK_CONDITION_UNIT_ATTENTION = 0x0e,
+ TCM_CHECK_CONDITION_NOT_READY = 0x0f,
+};
+
+struct se_obj {
+ atomic_t obj_access_count;
+} ____cacheline_aligned;
+
+/*
+ * Used by TCM Core internally to signal if ALUA emulation is enabled or
+ * disabled, or running in with TCM/pSCSI passthrough mode
+ */
+typedef enum {
+ SPC_ALUA_PASSTHROUGH,
+ SPC2_ALUA_DISABLED,
+ SPC3_ALUA_EMULATED
+} t10_alua_index_t;
+
+/*
+ * Used by TCM Core internally to signal if SAM Task Attribute emulation
+ * is enabled or disabled, or running in with TCM/pSCSI passthrough mode
+ */
+typedef enum {
+ SAM_TASK_ATTR_PASSTHROUGH,
+ SAM_TASK_ATTR_UNTAGGED,
+ SAM_TASK_ATTR_EMULATED
+} t10_task_attr_index_t;
+
+struct se_cmd;
+
+struct t10_alua {
+ t10_alua_index_t alua_type;
+ /* ALUA Target Port Group ID */
+ u16 alua_tg_pt_gps_counter;
+ u32 alua_tg_pt_gps_count;
+ spinlock_t tg_pt_gps_lock;
+ struct se_subsystem_dev *t10_sub_dev;
+ /* Used for default ALUA Target Port Group */
+ struct t10_alua_tg_pt_gp *default_tg_pt_gp;
+ /* Used for default ALUA Target Port Group ConfigFS group */
+ struct config_group alua_tg_pt_gps_group;
+ int (*alua_state_check)(struct se_cmd *, unsigned char *, u8 *);
+ struct list_head tg_pt_gps_list;
+} ____cacheline_aligned;
+
+struct t10_alua_lu_gp {
+ 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;
+ struct list_head lu_gp_list;
+ struct list_head lu_gp_mem_list;
+} ____cacheline_aligned;
+
+struct t10_alua_lu_gp_member {
+ int lu_gp_assoc:1;
+ atomic_t lu_gp_mem_ref_cnt;
+ spinlock_t lu_gp_mem_lock;
+ struct t10_alua_lu_gp *lu_gp;
+ struct se_device *lu_gp_mem_dev;
+ struct list_head lu_gp_mem_list;
+} ____cacheline_aligned;
+
+struct t10_alua_tg_pt_gp {
+ u16 tg_pt_gp_id;
+ int tg_pt_gp_valid_id;
+ int tg_pt_gp_alua_access_status;
+ int tg_pt_gp_alua_access_type;
+ int tg_pt_gp_nonop_delay_msecs;
+ int tg_pt_gp_trans_delay_msecs;
+ int tg_pt_gp_pref;
+ int tg_pt_gp_write_metadata;
+ /* Used by struct t10_alua_tg_pt_gp->tg_pt_gp_md_buf_len */
+#define ALUA_MD_BUF_LEN 1024
+ u32 tg_pt_gp_md_buf_len;
+ u32 tg_pt_gp_members;
+ atomic_t tg_pt_gp_alua_access_state;
+ atomic_t tg_pt_gp_ref_cnt;
+ spinlock_t tg_pt_gp_lock;
+ struct mutex tg_pt_gp_md_mutex;
+ struct se_subsystem_dev *tg_pt_gp_su_dev;
+ struct config_group tg_pt_gp_group;
+ struct list_head tg_pt_gp_list;
+ struct list_head tg_pt_gp_mem_list;
+} ____cacheline_aligned;
+
+struct t10_alua_tg_pt_gp_member {
+ int tg_pt_gp_assoc:1;
+ atomic_t tg_pt_gp_mem_ref_cnt;
+ spinlock_t tg_pt_gp_mem_lock;
+ struct t10_alua_tg_pt_gp *tg_pt_gp;
+ struct se_port *tg_pt;
+ struct list_head tg_pt_gp_mem_list;
+} ____cacheline_aligned;
+
+struct t10_vpd {
+ unsigned char device_identifier[INQUIRY_VPD_DEVICE_IDENTIFIER_LEN];
+ int protocol_identifier_set;
+ u32 protocol_identifier;
+ u32 device_identifier_code_set;
+ u32 association;
+ u32 device_identifier_type;
+ struct list_head vpd_list;
+} ____cacheline_aligned;
+
+struct t10_wwn {
+ unsigned char vendor[8];
+ unsigned char model[16];
+ unsigned char revision[4];
+ unsigned char unit_serial[INQUIRY_VPD_SERIAL_LEN];
+ spinlock_t t10_vpd_lock;
+ struct se_subsystem_dev *t10_sub_dev;
+ struct config_group t10_wwn_group;
+ struct list_head t10_vpd_list;
+} ____cacheline_aligned;
+
+
+/*
+ * Used by TCM Core internally to signal if >= SPC-3 peristent reservations
+ * emulation is enabled or disabled, or running in with TCM/pSCSI passthrough
+ * mode
+ */
+typedef enum {
+ SPC_PASSTHROUGH,
+ SPC2_RESERVATIONS,
+ SPC3_PERSISTENT_RESERVATIONS
+} t10_reservations_index_t;
+
+struct t10_pr_registration {
+ /* Used for fabrics that contain WWN+ISID */
+#define PR_REG_ISID_LEN 16
+ /* PR_REG_ISID_LEN + ',i,0x' */
+#define PR_REG_ISID_ID_LEN (PR_REG_ISID_LEN + 5)
+ char pr_reg_isid[PR_REG_ISID_LEN];
+ /* Used during APTPL metadata reading */
+#define PR_APTPL_MAX_IPORT_LEN 256
+ unsigned char pr_iport[PR_APTPL_MAX_IPORT_LEN];
+ /* Used during APTPL metadata reading */
+#define PR_APTPL_MAX_TPORT_LEN 256
+ unsigned char pr_tport[PR_APTPL_MAX_TPORT_LEN];
+ /* For writing out live meta data */
+ unsigned char *pr_aptpl_buf;
+ u16 pr_aptpl_rpti;
+ u16 pr_reg_tpgt;
+ /* Reservation effects all target ports */
+ int pr_reg_all_tg_pt;
+ /* Activate Persistence across Target Power Loss */
+ int pr_reg_aptpl;
+ int pr_res_holder;
+ int pr_res_type;
+ int pr_res_scope;
+ /* Used for fabric initiator WWPNs using a ISID */
+ int isid_present_at_reg:1;
+ u32 pr_res_mapped_lun;
+ u32 pr_aptpl_target_lun;
+ u32 pr_res_generation;
+ u64 pr_reg_bin_isid;
+ u64 pr_res_key;
+ atomic_t pr_res_holders;
+ struct se_node_acl *pr_reg_nacl;
+ struct se_dev_entry *pr_reg_deve;
+ struct se_lun *pr_reg_tg_pt_lun;
+ struct list_head pr_reg_list;
+ struct list_head pr_reg_abort_list;
+ struct list_head pr_reg_aptpl_list;
+ struct list_head pr_reg_atp_list;
+ struct list_head pr_reg_atp_mem_list;
+} ____cacheline_aligned;
+
+/*
+ * This set of function pointer ops is set based upon SPC3_PERSISTENT_RESERVATIONS,
+ * SPC2_RESERVATIONS or SPC_PASSTHROUGH in drivers/target/target_core_pr.c:
+ * core_setup_reservations()
+ */
+struct t10_reservation_ops {
+ int (*t10_reservation_check)(struct se_cmd *, u32 *);
+ int (*t10_seq_non_holder)(struct se_cmd *, unsigned char *, u32);
+ int (*t10_pr_register)(struct se_cmd *);
+ int (*t10_pr_clear)(struct se_cmd *);
+};
+
+struct t10_reservation_template {
+ /* Reservation effects all target ports */
+ int pr_all_tg_pt;
+ /* Activate Persistence across Target Power Loss enabled
+ * for SCSI device */
+ int pr_aptpl_active;
+ /* Used by struct t10_reservation_template->pr_aptpl_buf_len */
+#define PR_APTPL_BUF_LEN 8192
+ u32 pr_aptpl_buf_len;
+ u32 pr_generation;
+ t10_reservations_index_t res_type;
+ spinlock_t registration_lock;
+ spinlock_t aptpl_reg_lock;
+ /*
+ * This will always be set by one individual I_T Nexus.
+ * However with all_tg_pt=1, other I_T Nexus from the
+ * same initiator can access PR reg/res info on a different
+ * target port.
+ *
+ * There is also the 'All Registrants' case, where there is
+ * a single *pr_res_holder of the reservation, but all
+ * registrations are considered reservation holders.
+ */
+ struct se_node_acl *pr_res_holder;
+ struct list_head registration_list;
+ struct list_head aptpl_reg_list;
+ struct t10_reservation_ops pr_ops;
+} ____cacheline_aligned;
+
+struct se_queue_req {
+ int state;
+ void *cmd;
+ struct list_head qr_list;
+} ____cacheline_aligned;
+
+struct se_queue_obj {
+ atomic_t queue_cnt;
+ spinlock_t cmd_queue_lock;
+ struct list_head qobj_list;
+ wait_queue_head_t thread_wq;
+} ____cacheline_aligned;
+
+/*
+ * Used one per struct se_cmd to hold all extra struct se_task
+ * metadata. This structure is setup and allocated in
+ * drivers/target/target_core_transport.c:__transport_alloc_se_cmd()
+ */
+struct se_transport_task {
+ unsigned char *t_task_cdb;
+ unsigned char __t_task_cdb[TCM_MAX_COMMAND_SIZE];
+ unsigned long long t_task_lba;
+ int t_tasks_failed;
+ int t_tasks_fua;
+ int t_tasks_bidi:1;
+ u32 t_task_cdbs;
+ u32 t_tasks_check;
+ u32 t_tasks_no;
+ u32 t_tasks_sectors;
+ u32 t_tasks_se_num;
+ u32 t_tasks_se_bidi_num;
+ u32 t_tasks_sg_chained_no;
+ atomic_t t_fe_count;
+ atomic_t t_se_count;
+ atomic_t t_task_cdbs_left;
+ atomic_t t_task_cdbs_ex_left;
+ atomic_t t_task_cdbs_timeout_left;
+ atomic_t t_task_cdbs_sent;
+ atomic_t t_transport_aborted;
+ atomic_t t_transport_active;
+ atomic_t t_transport_complete;
+ atomic_t t_transport_queue_active;
+ atomic_t t_transport_sent;
+ atomic_t t_transport_stop;
+ atomic_t t_transport_timeout;
+ atomic_t transport_dev_active;
+ atomic_t transport_lun_active;
+ atomic_t transport_lun_fe_stop;
+ atomic_t transport_lun_stop;
+ spinlock_t t_state_lock;
+ struct completion t_transport_stop_comp;
+ struct completion transport_lun_fe_stop_comp;
+ struct completion transport_lun_stop_comp;
+ struct scatterlist *t_tasks_sg_chained;
+ struct scatterlist t_tasks_sg_bounce;
+ void *t_task_buf;
+ /*
+ * 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;
+ struct list_head *t_mem_list;
+ /* Used for BIDI READ */
+ struct list_head *t_mem_bidi_list;
+ struct list_head t_task_list;
+} ____cacheline_aligned;
+
+struct se_task {
+ unsigned char task_sense;
+ struct scatterlist *task_sg;
+ struct scatterlist *task_sg_bidi;
+ u8 task_scsi_status;
+ u8 task_flags;
+ int task_error_status;
+ int task_state_flags;
+ int task_padded_sg:1;
+ unsigned long long task_lba;
+ u32 task_no;
+ u32 task_sectors;
+ u32 task_size;
+ u32 task_sg_num;
+ u32 task_sg_offset;
+ enum dma_data_direction task_data_direction;
+ struct se_cmd *task_se_cmd;
+ struct se_device *se_dev;
+ struct completion task_stop_comp;
+ atomic_t task_active;
+ atomic_t task_execute_queue;
+ atomic_t task_timeout;
+ atomic_t task_sent;
+ atomic_t task_stop;
+ atomic_t task_state_active;
+ struct timer_list task_timer;
+ struct se_device *se_obj_ptr;
+ struct list_head t_list;
+ struct list_head t_execute_list;
+ struct list_head t_state_list;
+} ____cacheline_aligned;
+
+#define TASK_CMD(task) ((struct se_cmd *)task->task_se_cmd)
+#define TASK_DEV(task) ((struct se_device *)task->se_dev)
+
+struct se_cmd {
+ /* SAM response code being sent to initiator */
+ u8 scsi_status;
+ u8 scsi_asc;
+ u8 scsi_ascq;
+ u8 scsi_sense_reason;
+ u16 scsi_sense_length;
+ /* Delay for ALUA Active/NonOptimized state access in milliseconds */
+ int alua_nonop_delay;
+ /* See include/linux/dma-mapping.h */
+ enum dma_data_direction data_direction;
+ /* For SAM Task Attribute */
+ int sam_task_attr;
+ /* Transport protocol dependent state, see transport_state_table */
+ enum transport_state_table t_state;
+ /* Transport protocol dependent state for out of order CmdSNs */
+ int deferred_t_state;
+ /* Transport specific error status */
+ int transport_error_status;
+ /* See se_cmd_flags_table */
+ u32 se_cmd_flags;
+ u32 se_ordered_id;
+ /* Total size in bytes associated with command */
+ u32 data_length;
+ /* SCSI Presented Data Transfer Length */
+ u32 cmd_spdtl;
+ u32 residual_count;
+ u32 orig_fe_lun;
+ /* Persistent Reservation key */
+ u64 pr_res_key;
+ atomic_t transport_sent;
+ /* Used for sense data */
+ void *sense_buffer;
+ struct list_head se_delayed_list;
+ struct list_head se_ordered_list;
+ struct list_head se_lun_list;
+ 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;
+ struct se_tmr_req *se_tmr_req;
+ /* t_task is setup to t_task_backstore in transport_init_se_cmd() */
+ struct se_transport_task *t_task;
+ struct se_transport_task t_task_backstore;
+ struct target_core_fabric_ops *se_tfo;
+ int (*transport_emulate_cdb)(struct se_cmd *);
+ void (*transport_split_cdb)(unsigned long long, u32 *, unsigned char *);
+ void (*transport_wait_for_tasks)(struct se_cmd *, int, int);
+ void (*transport_complete_callback)(struct se_cmd *);
+} ____cacheline_aligned;
+
+#define T_TASK(cmd) ((struct se_transport_task *)(cmd->t_task))
+#define CMD_TFO(cmd) ((struct target_core_fabric_ops *)cmd->se_tfo)
+
+struct se_tmr_req {
+ /* Task Management function to be preformed */
+ u8 function;
+ /* Task Management response to send */
+ u8 response;
+ int call_transport;
+ /* Reference to ITT that Task Mgmt should be preformed */
+ u32 ref_task_tag;
+ /* 64-bit encoded SAM LUN from $FABRIC_MOD TMR header */
+ u64 ref_task_lun;
+ void *fabric_tmr_ptr;
+ struct se_cmd *task_cmd;
+ struct se_cmd *ref_cmd;
+ struct se_device *tmr_dev;
+ struct se_lun *tmr_lun;
+ struct list_head tmr_list;
+} ____cacheline_aligned;
+
+struct se_ua {
+ u8 ua_asc;
+ u8 ua_ascq;
+ struct se_node_acl *ua_nacl;
+ struct list_head ua_dev_list;
+ struct list_head ua_nacl_list;
+} ____cacheline_aligned;
+
+struct se_node_acl {
+ char initiatorname[TRANSPORT_IQN_LEN];
+ /* Used to signal demo mode created ACL, disabled by default */
+ int dynamic_node_acl:1;
+ u32 queue_depth;
+ u32 acl_index;
+ u64 num_cmds;
+ u64 read_bytes;
+ u64 write_bytes;
+ spinlock_t stats_lock;
+ /* Used for PR SPEC_I_PT=1 and REGISTER_AND_MOVE */
+ atomic_t acl_pr_ref_count;
+ /* Used for MIB access */
+ atomic_t mib_ref_count;
+ struct se_dev_entry *device_list;
+ struct se_session *nacl_sess;
+ struct se_portal_group *se_tpg;
+ spinlock_t device_list_lock;
+ spinlock_t nacl_sess_lock;
+ struct config_group acl_group;
+ struct config_group acl_attrib_group;
+ struct config_group acl_auth_group;
+ struct config_group acl_param_group;
+ struct config_group *acl_default_groups[4];
+ struct list_head acl_list;
+ struct list_head acl_sess_list;
+} ____cacheline_aligned;
+
+struct se_session {
+ /* Used for MIB access */
+ atomic_t mib_ref_count;
+ u64 sess_bin_isid;
+ struct se_node_acl *se_node_acl;
+ struct se_portal_group *se_tpg;
+ void *fabric_sess_ptr;
+ struct list_head sess_list;
+ struct list_head sess_acl_list;
+} ____cacheline_aligned;
+
+#define SE_SESS(cmd) ((struct se_session *)(cmd)->se_sess)
+#define SE_NODE_ACL(sess) ((struct se_node_acl *)(sess)->se_node_acl)
+
+struct se_device;
+struct se_transform_info;
+struct scatterlist;
+
+struct se_lun_acl {
+ char initiatorname[TRANSPORT_IQN_LEN];
+ u32 mapped_lun;
+ struct se_node_acl *se_lun_nacl;
+ struct se_lun *se_lun;
+ struct list_head lacl_list;
+ struct config_group se_lun_group;
+} ____cacheline_aligned;
+
+struct se_dev_entry {
+ int def_pr_registered:1;
+ /* See transport_lunflags_table */
+ u32 lun_flags;
+ u32 deve_cmds;
+ u32 mapped_lun;
+ u32 average_bytes;
+ u32 last_byte_count;
+ u32 total_cmds;
+ u32 total_bytes;
+ u64 pr_res_key;
+ u64 creation_time;
+ u32 attach_count;
+ u64 read_bytes;
+ u64 write_bytes;
+ atomic_t ua_count;
+ /* Used for PR SPEC_I_PT=1 and REGISTER_AND_MOVE */
+ atomic_t pr_ref_count;
+ struct se_lun_acl *se_lun_acl;
+ spinlock_t ua_lock;
+ struct se_lun *se_lun;
+ struct list_head alua_port_list;
+ struct list_head ua_list;
+} ____cacheline_aligned;
+
+struct se_dev_limits {
+ /* Max supported HW queue depth */
+ u32 hw_queue_depth;
+ /* Max supported virtual queue depth */
+ u32 queue_depth;
+ /* From include/linux/blkdev.h for the other HW/SW limits. */
+ struct queue_limits limits;
+} ____cacheline_aligned;
+
+struct se_dev_attrib {
+ int emulate_dpo;
+ int emulate_fua_write;
+ int emulate_fua_read;
+ int emulate_write_cache;
+ int emulate_ua_intlck_ctrl;
+ int emulate_tas;
+ int emulate_tpu;
+ int emulate_tpws;
+ int emulate_reservations;
+ int emulate_alua;
+ int enforce_pr_isids;
+ u32 hw_block_size;
+ u32 block_size;
+ u32 hw_max_sectors;
+ u32 max_sectors;
+ u32 optimal_sectors;
+ u32 hw_queue_depth;
+ u32 queue_depth;
+ u32 task_timeout;
+ u32 max_unmap_lba_count;
+ u32 max_unmap_block_desc_count;
+ u32 unmap_granularity;
+ u32 unmap_granularity_alignment;
+ struct se_subsystem_dev *da_sub_dev;
+ struct config_group da_group;
+} ____cacheline_aligned;
+
+struct se_subsystem_dev {
+/* Used for struct se_subsystem_dev-->se_dev_alias, must be less than PAGE_SIZE */
+#define SE_DEV_ALIAS_LEN 512
+ unsigned char se_dev_alias[SE_DEV_ALIAS_LEN];
+/* Used for struct se_subsystem_dev->se_dev_udev_path[], must be less than PAGE_SIZE */
+#define SE_UDEV_PATH_LEN 512
+ unsigned char se_dev_udev_path[SE_UDEV_PATH_LEN];
+ u32 su_dev_flags;
+ struct se_hba *se_dev_hba;
+ struct se_device *se_dev_ptr;
+ struct se_dev_attrib se_dev_attrib;
+ /* T10 Asymmetric Logical Unit Assignment for Target Ports */
+ struct t10_alua t10_alua;
+ /* T10 Inquiry and VPD WWN Information */
+ struct t10_wwn t10_wwn;
+ /* T10 SPC-2 + SPC-3 Reservations */
+ struct t10_reservation_template t10_reservation;
+ spinlock_t se_dev_lock;
+ void *se_dev_su_ptr;
+ struct list_head g_se_dev_list;
+ struct config_group se_dev_group;
+ /* For T10 Reservations */
+ struct config_group se_dev_pr_group;
+} ____cacheline_aligned;
+
+#define T10_ALUA(su_dev) (&(su_dev)->t10_alua)
+#define T10_RES(su_dev) (&(su_dev)->t10_reservation)
+#define T10_PR_OPS(su_dev) (&(su_dev)->t10_reservation.pr_ops)
+
+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 */
+ u32 dev_cur_ordered_id;
+ u32 dev_flags;
+ u32 dev_port_count;
+ /* See transport_device_status_table */
+ u32 dev_status;
+ u32 dev_tcq_window_closed;
+ /* Physical device queue depth */
+ u32 queue_depth;
+ /* Used for SPC-2 reservations enforce of ISIDs */
+ u64 dev_res_bin_isid;
+ t10_task_attr_index_t dev_task_attr_type;
+ /* Pointer to transport specific device structure */
+ void *dev_ptr;
+ u32 dev_index;
+ u64 creation_time;
+ u32 num_resets;
+ u64 num_cmds;
+ u64 read_bytes;
+ 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;
+ struct se_obj dev_obj;
+ struct se_obj dev_access_obj;
+ struct se_obj dev_export_obj;
+ struct se_queue_obj *dev_queue_obj;
+ struct se_queue_obj *dev_status_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;
+ /* Used for legacy SPC-2 reservationsa */
+ struct se_node_acl *dev_reserved_node_acl;
+ /* Used for ALUA Logical Unit Group membership */
+ struct t10_alua_lu_gp_member *dev_alua_lu_gp_mem;
+ /* Used for SPC-3 Persistent Reservations */
+ 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 list_head delayed_cmd_list;
+ struct list_head ordered_cmd_list;
+ struct list_head execute_task_list;
+ struct list_head state_task_list;
+ /* Pointer to associated SE HBA */
+ struct se_hba *se_hba;
+ struct se_subsystem_dev *se_sub_dev;
+ /* Pointer to template of function pointers for transport */
+ 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;
+
+#define SE_DEV(cmd) ((struct se_device *)(cmd)->se_lun->lun_se_dev)
+#define SU_DEV(dev) ((struct se_subsystem_dev *)(dev)->se_sub_dev)
+#define DEV_ATTRIB(dev) (&(dev)->se_sub_dev->se_dev_attrib)
+#define DEV_T10_WWN(dev) (&(dev)->se_sub_dev->t10_wwn)
+
+struct se_hba {
+ u16 hba_tpgt;
+ u32 hba_id;
+ /* See hba_flags_table */
+ u32 hba_flags;
+ /* Virtual iSCSI devices attached. */
+ u32 dev_count;
+ u32 hba_index;
+ atomic_t dev_mib_access_count;
+ atomic_t load_balance_queue;
+ atomic_t left_queue_depth;
+ /* Maximum queue depth the HBA can handle. */
+ atomic_t max_queue_depth;
+ /* Pointer to transport specific host structure. */
+ void *hba_ptr;
+ /* Linked list for struct se_device */
+ struct list_head hba_dev_list;
+ struct list_head hba_list;
+ spinlock_t device_lock;
+ spinlock_t hba_queue_lock;
+ struct config_group hba_group;
+ struct mutex hba_access_mutex;
+ struct se_subsystem_api *transport;
+} ____cacheline_aligned;
+
+#define SE_HBA(d) ((struct se_hba *)(d)->se_hba)
+
+struct se_lun {
+ /* See transport_lun_status_table */
+ enum transport_lun_status_table lun_status;
+ u32 lun_access;
+ u32 lun_flags;
+ u32 unpacked_lun;
+ atomic_t lun_acl_count;
+ spinlock_t lun_acl_lock;
+ spinlock_t lun_cmd_lock;
+ spinlock_t lun_sep_lock;
+ struct completion lun_shutdown_comp;
+ struct list_head lun_cmd_list;
+ struct list_head lun_acl_list;
+ struct se_device *lun_se_dev;
+ struct config_group lun_group;
+ struct se_port *lun_sep;
+} ____cacheline_aligned;
+
+#define SE_LUN(c) ((struct se_lun *)(c)->se_lun)
+
+struct se_port {
+ /* RELATIVE TARGET PORT IDENTIFER */
+ u16 sep_rtpi;
+ int sep_tg_pt_secondary_stat;
+ int sep_tg_pt_secondary_write_md;
+ 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;
+ spinlock_t sep_alua_lock;
+ struct mutex sep_tg_pt_md_mutex;
+ struct t10_alua_tg_pt_gp_member *sep_alua_tg_pt_gp_mem;
+ struct se_lun *sep_lun;
+ struct se_portal_group *sep_tpg;
+ struct list_head sep_alua_list;
+ struct list_head sep_list;
+} ____cacheline_aligned;
+
+struct se_tpg_np {
+ struct config_group tpg_np_group;
+} ____cacheline_aligned;
+
+struct se_portal_group {
+ /* Type of target portal group, see transport_tpg_type_table */
+ enum transport_tpg_type_table se_tpg_type;
+ /* Number of ACLed Initiator Nodes for this TPG */
+ u32 num_node_acls;
+ /* Used for PR SPEC_I_PT=1 and REGISTER_AND_MOVE */
+ atomic_t tpg_pr_ref_count;
+ /* Spinlock for adding/removing ACLed Nodes */
+ spinlock_t acl_node_lock;
+ /* Spinlock for adding/removing sessions */
+ spinlock_t session_lock;
+ spinlock_t tpg_lun_lock;
+ /* Pointer to $FABRIC_MOD portal group */
+ void *se_tpg_fabric_ptr;
+ struct list_head se_tpg_list;
+ /* linked list for initiator ACL list */
+ struct list_head acl_node_list;
+ struct se_lun *tpg_lun_list;
+ struct se_lun tpg_virt_lun0;
+ /* List of TCM sessions assoicated wth this TPG */
+ struct list_head tpg_sess_list;
+ /* Pointer to $FABRIC_MOD dependent code */
+ struct target_core_fabric_ops *se_tpg_tfo;
+ struct se_wwn *se_tpg_wwn;
+ struct config_group tpg_group;
+ struct config_group *tpg_default_groups[6];
+ struct config_group tpg_lun_group;
+ struct config_group tpg_np_group;
+ struct config_group tpg_acl_group;
+ struct config_group tpg_attrib_group;
+ struct config_group tpg_param_group;
+} ____cacheline_aligned;
+
+#define TPG_TFO(se_tpg) ((struct target_core_fabric_ops *)(se_tpg)->se_tpg_tfo)
+
+struct se_wwn {
+ struct target_fabric_configfs *wwn_tf;
+ struct config_group wwn_group;
+} ____cacheline_aligned;
+
+struct se_global {
+ u16 alua_lu_gps_counter;
+ int g_sub_api_initialized;
+ u32 in_shutdown;
+ u32 alua_lu_gps_count;
+ u32 g_hba_id_counter;
+ struct config_group target_core_hbagroup;
+ struct config_group alua_group;
+ struct config_group alua_lu_gps_group;
+ struct list_head g_lu_gps_list;
+ struct list_head g_se_tpg_list;
+ struct list_head g_hba_list;
+ struct list_head g_se_dev_list;
+ struct se_hba *g_lun0_hba;
+ struct se_subsystem_dev *g_lun0_su_dev;
+ struct se_device *g_lun0_dev;
+ struct t10_alua_lu_gp *default_lu_gp;
+ spinlock_t g_device_lock;
+ spinlock_t hba_lock;
+ spinlock_t se_tpg_lock;
+ spinlock_t lu_gps_lock;
+ spinlock_t plugin_class_lock;
+} ____cacheline_aligned;
+
+#endif /* TARGET_CORE_BASE_H */
diff --git a/include/target/target_core_configfs.h b/include/target/target_core_configfs.h
new file mode 100644
index 0000000..40e6e74
--- /dev/null
+++ b/include/target/target_core_configfs.h
@@ -0,0 +1,52 @@
+#define TARGET_CORE_CONFIGFS_VERSION TARGET_CORE_MOD_VERSION
+
+#define TARGET_CORE_CONFIG_ROOT "/sys/kernel/config"
+
+#define TARGET_CORE_NAME_MAX_LEN 64
+#define TARGET_FABRIC_NAME_SIZE 32
+
+extern struct target_fabric_configfs *target_fabric_configfs_init(
+ struct module *, const char *);
+extern void target_fabric_configfs_free(struct target_fabric_configfs *);
+extern int target_fabric_configfs_register(struct target_fabric_configfs *);
+extern void target_fabric_configfs_deregister(struct target_fabric_configfs *);
+
+struct target_fabric_configfs_template {
+ struct config_item_type tfc_discovery_cit;
+ struct config_item_type tfc_wwn_cit;
+ struct config_item_type tfc_tpg_cit;
+ struct config_item_type tfc_tpg_base_cit;
+ struct config_item_type tfc_tpg_lun_cit;
+ struct config_item_type tfc_tpg_port_cit;
+ struct config_item_type tfc_tpg_np_cit;
+ struct config_item_type tfc_tpg_np_base_cit;
+ struct config_item_type tfc_tpg_attrib_cit;
+ struct config_item_type tfc_tpg_param_cit;
+ struct config_item_type tfc_tpg_nacl_cit;
+ struct config_item_type tfc_tpg_nacl_base_cit;
+ struct config_item_type tfc_tpg_nacl_attrib_cit;
+ struct config_item_type tfc_tpg_nacl_auth_cit;
+ struct config_item_type tfc_tpg_nacl_param_cit;
+ struct config_item_type tfc_tpg_mappedlun_cit;
+};
+
+struct target_fabric_configfs {
+ char tf_name[TARGET_FABRIC_NAME_SIZE];
+ atomic_t tf_access_cnt;
+ struct list_head tf_list;
+ struct config_group tf_group;
+ struct config_group tf_disc_group;
+ struct config_group *tf_default_groups[2];
+ /* Pointer to fabric's config_item */
+ struct config_item *tf_fabric;
+ /* Passed from fabric modules */
+ struct config_item_type *tf_fabric_cit;
+ /* Pointer to target core subsystem */
+ struct configfs_subsystem *tf_subsys;
+ /* Pointer to fabric's struct module */
+ struct module *tf_module;
+ struct target_core_fabric_ops tf_ops;
+ struct target_fabric_configfs_template tf_cit_tmpl;
+};
+
+#define TF_CIT_TMPL(tf) (&(tf)->tf_cit_tmpl)
diff --git a/include/target/target_core_device.h b/include/target/target_core_device.h
new file mode 100644
index 0000000..52b18a5
--- /dev/null
+++ b/include/target/target_core_device.h
@@ -0,0 +1,61 @@
+#ifndef TARGET_CORE_DEVICE_H
+#define TARGET_CORE_DEVICE_H
+
+extern int transport_get_lun_for_cmd(struct se_cmd *, unsigned char *, u32);
+extern int transport_get_lun_for_tmr(struct se_cmd *, u32);
+extern struct se_dev_entry *core_get_se_deve_from_rtpi(
+ struct se_node_acl *, u16);
+extern int core_free_device_list_for_node(struct se_node_acl *,
+ struct se_portal_group *);
+extern void core_dec_lacl_count(struct se_node_acl *, struct se_cmd *);
+extern void core_update_device_list_access(u32, u32, struct se_node_acl *);
+extern int core_update_device_list_for_node(struct se_lun *, struct se_lun_acl *, u32,
+ u32, struct se_node_acl *,
+ struct se_portal_group *, int);
+extern void core_clear_lun_from_tpg(struct se_lun *, struct se_portal_group *);
+extern int core_dev_export(struct se_device *, struct se_portal_group *,
+ struct se_lun *);
+extern void core_dev_unexport(struct se_device *, struct se_portal_group *,
+ struct se_lun *);
+extern int transport_core_report_lun_response(struct se_cmd *);
+extern void se_release_device_for_hba(struct se_device *);
+extern void se_release_vpd_for_dev(struct se_device *);
+extern void se_clear_dev_ports(struct se_device *);
+extern int se_free_virtual_device(struct se_device *, struct se_hba *);
+extern int se_dev_check_online(struct se_device *);
+extern int se_dev_check_shutdown(struct se_device *);
+extern void se_dev_set_default_attribs(struct se_device *, struct se_dev_limits *);
+extern int se_dev_set_task_timeout(struct se_device *, u32);
+extern int se_dev_set_max_unmap_lba_count(struct se_device *, u32);
+extern int se_dev_set_max_unmap_block_desc_count(struct se_device *, u32);
+extern int se_dev_set_unmap_granularity(struct se_device *, u32);
+extern int se_dev_set_unmap_granularity_alignment(struct se_device *, u32);
+extern int se_dev_set_emulate_dpo(struct se_device *, int);
+extern int se_dev_set_emulate_fua_write(struct se_device *, int);
+extern int se_dev_set_emulate_fua_read(struct se_device *, int);
+extern int se_dev_set_emulate_write_cache(struct se_device *, int);
+extern int se_dev_set_emulate_ua_intlck_ctrl(struct se_device *, int);
+extern int se_dev_set_emulate_tas(struct se_device *, int);
+extern int se_dev_set_emulate_tpu(struct se_device *, int);
+extern int se_dev_set_emulate_tpws(struct se_device *, int);
+extern int se_dev_set_enforce_pr_isids(struct se_device *, int);
+extern int se_dev_set_queue_depth(struct se_device *, u32);
+extern int se_dev_set_max_sectors(struct se_device *, u32);
+extern int se_dev_set_optimal_sectors(struct se_device *, u32);
+extern int se_dev_set_block_size(struct se_device *, u32);
+extern struct se_lun *core_dev_add_lun(struct se_portal_group *, struct se_hba *,
+ struct se_device *, u32);
+extern int core_dev_del_lun(struct se_portal_group *, u32);
+extern struct se_lun *core_get_lun_from_tpg(struct se_portal_group *, u32);
+extern struct se_lun_acl *core_dev_init_initiator_node_lun_acl(struct se_portal_group *,
+ u32, char *, int *);
+extern int core_dev_add_initiator_node_lun_acl(struct se_portal_group *,
+ struct se_lun_acl *, u32, u32);
+extern int core_dev_del_initiator_node_lun_acl(struct se_portal_group *,
+ struct se_lun *, struct se_lun_acl *);
+extern void core_dev_free_initiator_node_lun_acl(struct se_portal_group *,
+ struct se_lun_acl *lacl);
+extern int core_dev_setup_virtual_lun0(void);
+extern void core_dev_release_virtual_lun0(void);
+
+#endif /* TARGET_CORE_DEVICE_H */
diff --git a/include/target/target_core_fabric_configfs.h b/include/target/target_core_fabric_configfs.h
new file mode 100644
index 0000000..a26fb75
--- /dev/null
+++ b/include/target/target_core_fabric_configfs.h
@@ -0,0 +1,106 @@
+/*
+ * Used for tfc_wwn_cit attributes
+ */
+
+#include <target/configfs_macros.h>
+
+CONFIGFS_EATTR_STRUCT(target_fabric_nacl_attrib, se_node_acl);
+#define TF_NACL_ATTRIB_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_nacl_attrib_attribute _fabric##_nacl_attrib_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_nacl_attrib_show_##_name, \
+ _fabric##_nacl_attrib_store_##_name);
+
+CONFIGFS_EATTR_STRUCT(target_fabric_nacl_auth, se_node_acl);
+#define TF_NACL_AUTH_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_nacl_auth_attribute _fabric##_nacl_auth_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_nacl_auth_show_##_name, \
+ _fabric##_nacl_auth_store_##_name);
+
+#define TF_NACL_AUTH_ATTR_RO(_fabric, _name) \
+static struct target_fabric_nacl_auth_attribute _fabric##_nacl_auth_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ _fabric##_nacl_auth_show_##_name);
+
+CONFIGFS_EATTR_STRUCT(target_fabric_nacl_param, se_node_acl);
+#define TF_NACL_PARAM_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_nacl_param_attribute _fabric##_nacl_param_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_nacl_param_show_##_name, \
+ _fabric##_nacl_param_store_##_name);
+
+#define TF_NACL_PARAM_ATTR_RO(_fabric, _name) \
+static struct target_fabric_nacl_param_attribute _fabric##_nacl_param_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ _fabric##_nacl_param_show_##_name);
+
+
+CONFIGFS_EATTR_STRUCT(target_fabric_nacl_base, se_node_acl);
+#define TF_NACL_BASE_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_nacl_base_attribute _fabric##_nacl_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_nacl_show_##_name, \
+ _fabric##_nacl_store_##_name);
+
+#define TF_NACL_BASE_ATTR_RO(_fabric, _name) \
+static struct target_fabric_nacl_base_attribute _fabric##_nacl_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ _fabric##_nacl_show_##_name);
+
+CONFIGFS_EATTR_STRUCT(target_fabric_np_base, se_tpg_np);
+#define TF_NP_BASE_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_np_base_attribute _fabric##_np_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_np_show_##_name, \
+ _fabric##_np_store_##_name);
+
+CONFIGFS_EATTR_STRUCT(target_fabric_tpg_attrib, se_portal_group);
+#define TF_TPG_ATTRIB_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_tpg_attrib_attribute _fabric##_tpg_attrib_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_tpg_attrib_show_##_name, \
+ _fabric##_tpg_attrib_store_##_name);
+
+
+CONFIGFS_EATTR_STRUCT(target_fabric_tpg_param, se_portal_group);
+#define TF_TPG_PARAM_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_tpg_param_attribute _fabric##_tpg_param_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_tpg_param_show_##_name, \
+ _fabric##_tpg_param_store_##_name);
+
+
+CONFIGFS_EATTR_STRUCT(target_fabric_tpg, se_portal_group);
+#define TF_TPG_BASE_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_tpg_attribute _fabric##_tpg_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_tpg_show_##_name, \
+ _fabric##_tpg_store_##_name);
+
+
+CONFIGFS_EATTR_STRUCT(target_fabric_wwn, target_fabric_configfs);
+#define TF_WWN_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_wwn_attribute _fabric##_wwn_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_wwn_show_attr_##_name, \
+ _fabric##_wwn_store_attr_##_name);
+
+#define TF_WWN_ATTR_RO(_fabric, _name) \
+static struct target_fabric_wwn_attribute _fabric##_wwn_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ _fabric##_wwn_show_attr_##_name);
+
+CONFIGFS_EATTR_STRUCT(target_fabric_discovery, target_fabric_configfs);
+#define TF_DISC_ATTR(_fabric, _name, _mode) \
+static struct target_fabric_discovery_attribute _fabric##_disc_##_name = \
+ __CONFIGFS_EATTR(_name, _mode, \
+ _fabric##_disc_show_##_name, \
+ _fabric##_disc_store_##_name);
+
+#define TF_DISC_ATTR_RO(_fabric, _name) \
+static struct target_fabric_discovery_attribute _fabric##_disc_##_name = \
+ __CONFIGFS_EATTR_RO(_name, \
+ _fabric##_disc_show_##_name);
+
+extern int target_fabric_setup_cits(struct target_fabric_configfs *);
diff --git a/include/target/target_core_fabric_lib.h b/include/target/target_core_fabric_lib.h
new file mode 100644
index 0000000..c2f8d0e
--- /dev/null
+++ b/include/target/target_core_fabric_lib.h
@@ -0,0 +1,28 @@
+#ifndef TARGET_CORE_FABRIC_LIB_H
+#define TARGET_CORE_FABRIC_LIB_H
+
+extern u8 sas_get_fabric_proto_ident(struct se_portal_group *);
+extern u32 sas_get_pr_transport_id(struct se_portal_group *, struct se_node_acl *,
+ struct t10_pr_registration *, int *, unsigned char *);
+extern u32 sas_get_pr_transport_id_len(struct se_portal_group *, struct se_node_acl *,
+ struct t10_pr_registration *, int *);
+extern char *sas_parse_pr_out_transport_id(struct se_portal_group *,
+ const char *, u32 *, char **);
+
+extern u8 fc_get_fabric_proto_ident(struct se_portal_group *);
+extern u32 fc_get_pr_transport_id(struct se_portal_group *, struct se_node_acl *,
+ struct t10_pr_registration *, int *, unsigned char *);
+extern u32 fc_get_pr_transport_id_len(struct se_portal_group *, struct se_node_acl *,
+ struct t10_pr_registration *, int *);
+extern char *fc_parse_pr_out_transport_id(struct se_portal_group *,
+ const char *, u32 *, char **);
+
+extern u8 iscsi_get_fabric_proto_ident(struct se_portal_group *);
+extern u32 iscsi_get_pr_transport_id(struct se_portal_group *, struct se_node_acl *,
+ struct t10_pr_registration *, int *, unsigned char *);
+extern u32 iscsi_get_pr_transport_id_len(struct se_portal_group *, struct se_node_acl *,
+ struct t10_pr_registration *, int *);
+extern char *iscsi_parse_pr_out_transport_id(struct se_portal_group *,
+ const char *, u32 *, char **);
+
+#endif /* TARGET_CORE_FABRIC_LIB_H */
diff --git a/include/target/target_core_fabric_ops.h b/include/target/target_core_fabric_ops.h
new file mode 100644
index 0000000..f3ac12b
--- /dev/null
+++ b/include/target/target_core_fabric_ops.h
@@ -0,0 +1,100 @@
+/* Defined in target_core_configfs.h */
+struct target_fabric_configfs;
+
+struct target_core_fabric_ops {
+ struct configfs_subsystem *tf_subsys;
+ /*
+ * Optional to signal struct se_task->task_sg[] padding entries
+ * for scatterlist chaining using transport_do_task_sg_link(),
+ * disabled by default
+ */
+ int task_sg_chaining:1;
+ char *(*get_fabric_name)(void);
+ u8 (*get_fabric_proto_ident)(struct se_portal_group *);
+ char *(*tpg_get_wwn)(struct se_portal_group *);
+ u16 (*tpg_get_tag)(struct se_portal_group *);
+ u32 (*tpg_get_default_depth)(struct se_portal_group *);
+ u32 (*tpg_get_pr_transport_id)(struct se_portal_group *,
+ struct se_node_acl *,
+ struct t10_pr_registration *, int *,
+ unsigned char *);
+ u32 (*tpg_get_pr_transport_id_len)(struct se_portal_group *,
+ struct se_node_acl *,
+ struct t10_pr_registration *, int *);
+ char *(*tpg_parse_pr_out_transport_id)(struct se_portal_group *,
+ const char *, u32 *, char **);
+ int (*tpg_check_demo_mode)(struct se_portal_group *);
+ int (*tpg_check_demo_mode_cache)(struct se_portal_group *);
+ int (*tpg_check_demo_mode_write_protect)(struct se_portal_group *);
+ int (*tpg_check_prod_mode_write_protect)(struct se_portal_group *);
+ struct se_node_acl *(*tpg_alloc_fabric_acl)(
+ struct se_portal_group *);
+ void (*tpg_release_fabric_acl)(struct se_portal_group *,
+ struct se_node_acl *);
+ u32 (*tpg_get_inst_index)(struct se_portal_group *);
+ /*
+ * Optional function pointer for TCM to perform command map
+ * from TCM processing thread context, for those struct se_cmd
+ * initally allocated in interrupt context.
+ */
+ int (*new_cmd_map)(struct se_cmd *);
+ /*
+ * Optional function pointer for TCM fabric modules that use
+ * Linux/NET sockets to allocate struct iovec array to struct se_cmd
+ */
+ int (*alloc_cmd_iovecs)(struct se_cmd *);
+ /*
+ * Optional to release struct se_cmd and fabric dependent allocated
+ * I/O descriptor in transport_cmd_check_stop()
+ */
+ void (*check_stop_free)(struct se_cmd *);
+ void (*release_cmd_to_pool)(struct se_cmd *);
+ void (*release_cmd_direct)(struct se_cmd *);
+ /*
+ * Called with spin_lock_bh(struct se_portal_group->session_lock held.
+ */
+ int (*shutdown_session)(struct se_session *);
+ void (*close_session)(struct se_session *);
+ void (*stop_session)(struct se_session *, int, int);
+ void (*fall_back_to_erl0)(struct se_session *);
+ int (*sess_logged_in)(struct se_session *);
+ u32 (*sess_get_index)(struct se_session *);
+ /*
+ * Used only for SCSI fabrics that contain multi-value TransportIDs
+ * (like iSCSI). All other SCSI fabrics should set this to NULL.
+ */
+ u32 (*sess_get_initiator_sid)(struct se_session *,
+ unsigned char *, u32);
+ int (*write_pending)(struct se_cmd *);
+ int (*write_pending_status)(struct se_cmd *);
+ void (*set_default_node_attributes)(struct se_node_acl *);
+ u32 (*get_task_tag)(struct se_cmd *);
+ int (*get_cmd_state)(struct se_cmd *);
+ void (*new_cmd_failure)(struct se_cmd *);
+ int (*queue_data_in)(struct se_cmd *);
+ int (*queue_status)(struct se_cmd *);
+ int (*queue_tm_rsp)(struct se_cmd *);
+ u16 (*set_fabric_sense_len)(struct se_cmd *, u32);
+ u16 (*get_fabric_sense_len)(void);
+ int (*is_state_remove)(struct se_cmd *);
+ u64 (*pack_lun)(unsigned int);
+ /*
+ * fabric module calls for target_core_fabric_configfs.c
+ */
+ struct se_wwn *(*fabric_make_wwn)(struct target_fabric_configfs *,
+ struct config_group *, const char *);
+ void (*fabric_drop_wwn)(struct se_wwn *);
+ struct se_portal_group *(*fabric_make_tpg)(struct se_wwn *,
+ struct config_group *, const char *);
+ void (*fabric_drop_tpg)(struct se_portal_group *);
+ int (*fabric_post_link)(struct se_portal_group *,
+ struct se_lun *);
+ void (*fabric_pre_unlink)(struct se_portal_group *,
+ struct se_lun *);
+ struct se_tpg_np *(*fabric_make_np)(struct se_portal_group *,
+ struct config_group *, const char *);
+ void (*fabric_drop_np)(struct se_tpg_np *);
+ struct se_node_acl *(*fabric_make_nodeacl)(struct se_portal_group *,
+ struct config_group *, const char *);
+ void (*fabric_drop_nodeacl)(struct se_node_acl *);
+};
diff --git a/include/target/target_core_tmr.h b/include/target/target_core_tmr.h
new file mode 100644
index 0000000..6c8248b
--- /dev/null
+++ b/include/target/target_core_tmr.h
@@ -0,0 +1,43 @@
+#ifndef TARGET_CORE_TMR_H
+#define TARGET_CORE_TMR_H
+
+/* task management function values */
+#ifdef ABORT_TASK
+#undef ABORT_TASK
+#endif /* ABORT_TASK */
+#define ABORT_TASK 1
+#ifdef ABORT_TASK_SET
+#undef ABORT_TASK_SET
+#endif /* ABORT_TASK_SET */
+#define ABORT_TASK_SET 2
+#ifdef CLEAR_ACA
+#undef CLEAR_ACA
+#endif /* CLEAR_ACA */
+#define CLEAR_ACA 3
+#ifdef CLEAR_TASK_SET
+#undef CLEAR_TASK_SET
+#endif /* CLEAR_TASK_SET */
+#define CLEAR_TASK_SET 4
+#define LUN_RESET 5
+#define TARGET_WARM_RESET 6
+#define TARGET_COLD_RESET 7
+#define TASK_REASSIGN 8
+
+/* task management response values */
+#define TMR_FUNCTION_COMPLETE 0
+#define TMR_TASK_DOES_NOT_EXIST 1
+#define TMR_LUN_DOES_NOT_EXIST 2
+#define TMR_TASK_STILL_ALLEGIANT 3
+#define TMR_TASK_FAILOVER_NOT_SUPPORTED 4
+#define TMR_TASK_MGMT_FUNCTION_NOT_SUPPORTED 5
+#define TMR_FUNCTION_AUTHORIZATION_FAILED 6
+#define TMR_FUNCTION_REJECTED 255
+
+extern struct kmem_cache *se_tmr_req_cache;
+
+extern struct se_tmr_req *core_tmr_alloc_req(struct se_cmd *, void *, u8);
+extern void core_tmr_release_req(struct se_tmr_req *);
+extern int core_tmr_lun_reset(struct se_device *, struct se_tmr_req *,
+ struct list_head *, struct se_cmd *);
+
+#endif /* TARGET_CORE_TMR_H */
diff --git a/include/target/target_core_tpg.h b/include/target/target_core_tpg.h
new file mode 100644
index 0000000..77e1872
--- /dev/null
+++ b/include/target/target_core_tpg.h
@@ -0,0 +1,35 @@
+#ifndef TARGET_CORE_TPG_H
+#define TARGET_CORE_TPG_H
+
+extern struct se_node_acl *__core_tpg_get_initiator_node_acl(struct se_portal_group *tpg,
+ const char *);
+extern struct se_node_acl *core_tpg_get_initiator_node_acl(struct se_portal_group *tpg,
+ unsigned char *);
+extern void core_tpg_add_node_to_devs(struct se_node_acl *,
+ struct se_portal_group *);
+extern struct se_node_acl *core_tpg_check_initiator_node_acl(
+ struct se_portal_group *,
+ unsigned char *);
+extern void core_tpg_wait_for_nacl_pr_ref(struct se_node_acl *);
+extern void core_tpg_wait_for_mib_ref(struct se_node_acl *);
+extern void core_tpg_clear_object_luns(struct se_portal_group *);
+extern struct se_node_acl *core_tpg_add_initiator_node_acl(
+ struct se_portal_group *,
+ struct se_node_acl *,
+ const char *, u32);
+extern int core_tpg_del_initiator_node_acl(struct se_portal_group *,
+ struct se_node_acl *, int);
+extern int core_tpg_set_initiator_node_queue_depth(struct se_portal_group *,
+ unsigned char *, u32, int);
+extern int core_tpg_register(struct target_core_fabric_ops *,
+ struct se_wwn *,
+ struct se_portal_group *, void *,
+ int);
+extern int core_tpg_deregister(struct se_portal_group *);
+extern struct se_lun *core_tpg_pre_addlun(struct se_portal_group *, u32);
+extern int core_tpg_post_addlun(struct se_portal_group *, struct se_lun *, u32,
+ void *);
+extern struct se_lun *core_tpg_pre_dellun(struct se_portal_group *, u32, int *);
+extern int core_tpg_post_dellun(struct se_portal_group *, struct se_lun *);
+
+#endif /* TARGET_CORE_TPG_H */
diff --git a/include/target/target_core_transport.h b/include/target/target_core_transport.h
new file mode 100644
index 0000000..66f44e5
--- /dev/null
+++ b/include/target/target_core_transport.h
@@ -0,0 +1,351 @@
+#ifndef TARGET_CORE_TRANSPORT_H
+#define TARGET_CORE_TRANSPORT_H
+
+#define TARGET_CORE_VERSION TARGET_CORE_MOD_VERSION
+
+/* Attempts before moving from SHORT to LONG */
+#define PYX_TRANSPORT_WINDOW_CLOSED_THRESHOLD 3
+#define PYX_TRANSPORT_WINDOW_CLOSED_WAIT_SHORT 3 /* In milliseconds */
+#define PYX_TRANSPORT_WINDOW_CLOSED_WAIT_LONG 10 /* In milliseconds */
+
+#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_TASK_TIMEOUT -10
+#define PYX_TRANSPORT_RESERVATION_CONFLICT -11
+#define PYX_TRANSPORT_ILLEGAL_REQUEST -12
+#define PYX_TRANSPORT_USE_SENSE_REASON -13
+
+#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
+
+/* For SE OBJ Plugins, in seconds */
+#define TRANSPORT_TIMEOUT_TUR 10
+#define TRANSPORT_TIMEOUT_TYPE_DISK 60
+#define TRANSPORT_TIMEOUT_TYPE_ROM 120
+#define TRANSPORT_TIMEOUT_TYPE_TAPE 600
+#define TRANSPORT_TIMEOUT_TYPE_OTHER 300
+
+/* For se_task->task_state_flags */
+#define TSF_EXCEPTION_CLEARED 0x01
+
+/*
+ * struct se_subsystem_dev->su_dev_flags
+*/
+#define SDF_FIRMWARE_VPD_UNIT_SERIAL 0x00000001
+#define SDF_EMULATED_VPD_UNIT_SERIAL 0x00000002
+#define SDF_USING_UDEV_PATH 0x00000004
+#define SDF_USING_ALIAS 0x00000008
+
+/*
+ * struct se_device->dev_flags
+ */
+#define DF_READ_ONLY 0x00000001
+#define DF_SPC2_RESERVATIONS 0x00000002
+#define DF_SPC2_RESERVATIONS_WITH_ISID 0x00000004
+
+/* struct se_dev_attrib sanity values */
+/* 10 Minutes */
+#define DA_TASK_TIMEOUT_MAX 600
+/* Default max_unmap_lba_count */
+#define DA_MAX_UNMAP_LBA_COUNT 0
+/* Default max_unmap_block_desc_count */
+#define DA_MAX_UNMAP_BLOCK_DESC_COUNT 0
+/* Default unmap_granularity */
+#define DA_UNMAP_GRANULARITY_DEFAULT 0
+/* Default unmap_granularity_alignment */
+#define DA_UNMAP_GRANULARITY_ALIGNMENT_DEFAULT 0
+/* Emulation for Direct Page Out */
+#define DA_EMULATE_DPO 0
+/* Emulation for Forced Unit Access WRITEs */
+#define DA_EMULATE_FUA_WRITE 1
+/* Emulation for Forced Unit Access READs */
+#define DA_EMULATE_FUA_READ 0
+/* Emulation for WriteCache and SYNCHRONIZE_CACHE */
+#define DA_EMULATE_WRITE_CACHE 0
+/* Emulation for UNIT ATTENTION Interlock Control */
+#define DA_EMULATE_UA_INTLLCK_CTRL 0
+/* Emulation for TASK_ABORTED status (TAS) by default */
+#define DA_EMULATE_TAS 1
+/* Emulation for Thin Provisioning UNMAP using block/blk-lib.c:blkdev_issue_discard() */
+#define DA_EMULATE_TPU 0
+/*
+ * Emulation for Thin Provisioning WRITE_SAME w/ UNMAP=1 bit using
+ * block/blk-lib.c:blkdev_issue_discard()
+ */
+#define DA_EMULATE_TPWS 0
+/* No Emulation for PSCSI by default */
+#define DA_EMULATE_RESERVATIONS 0
+/* No Emulation for PSCSI by default */
+#define DA_EMULATE_ALUA 0
+/* Enforce SCSI Initiator Port TransportID with 'ISID' for PR */
+#define DA_ENFORCE_PR_ISIDS 1
+#define DA_STATUS_MAX_SECTORS_MIN 16
+#define DA_STATUS_MAX_SECTORS_MAX 8192
+
+#define SE_MODE_PAGE_BUF 512
+
+#define MOD_MAX_SECTORS(ms, bs) (ms % (PAGE_SIZE / bs))
+
+struct se_mem;
+struct se_subsystem_api;
+
+extern int init_se_global(void);
+extern void release_se_global(void);
+extern void transport_init_queue_obj(struct se_queue_obj *);
+extern int transport_subsystem_check_init(void);
+extern int transport_subsystem_register(struct se_subsystem_api *);
+extern void transport_subsystem_release(struct se_subsystem_api *);
+extern void transport_load_plugins(void);
+extern struct se_session *transport_init_session(void);
+extern void __transport_register_session(struct se_portal_group *,
+ struct se_node_acl *,
+ struct se_session *, void *);
+extern void transport_register_session(struct se_portal_group *,
+ struct se_node_acl *,
+ struct se_session *, void *);
+extern void transport_free_session(struct se_session *);
+extern void transport_deregister_session_configfs(struct se_session *);
+extern void transport_deregister_session(struct se_session *);
+extern void transport_cmd_finish_abort(struct se_cmd *, int);
+extern void transport_cmd_finish_abort_tmr(struct se_cmd *);
+extern void transport_complete_sync_cache(struct se_cmd *, int);
+extern void transport_complete_task(struct se_task *, int);
+extern void transport_add_task_to_execute_queue(struct se_task *,
+ struct se_task *,
+ struct se_device *);
+unsigned char *transport_dump_cmd_direction(struct se_cmd *);
+extern void transport_dump_dev_state(struct se_device *, char *, int *);
+extern void transport_dump_dev_info(struct se_device *, struct se_lun *,
+ unsigned long long, char *, int *);
+extern void transport_dump_vpd_proto_id(struct t10_vpd *,
+ unsigned char *, int);
+extern void transport_set_vpd_proto_id(struct t10_vpd *, unsigned char *);
+extern int transport_dump_vpd_assoc(struct t10_vpd *,
+ unsigned char *, int);
+extern int transport_set_vpd_assoc(struct t10_vpd *, unsigned char *);
+extern int transport_dump_vpd_ident_type(struct t10_vpd *,
+ unsigned char *, int);
+extern int transport_set_vpd_ident_type(struct t10_vpd *, unsigned char *);
+extern int transport_dump_vpd_ident(struct t10_vpd *,
+ unsigned char *, int);
+extern int transport_set_vpd_ident(struct t10_vpd *, unsigned char *);
+extern struct se_device *transport_add_device_to_core_hba(struct se_hba *,
+ struct se_subsystem_api *,
+ struct se_subsystem_dev *, u32,
+ void *, struct se_dev_limits *,
+ const char *, const char *);
+extern void transport_device_setup_cmd(struct se_cmd *);
+extern void transport_init_se_cmd(struct se_cmd *,
+ struct target_core_fabric_ops *,
+ struct se_session *, u32, int, int,
+ unsigned char *);
+extern void transport_free_se_cmd(struct se_cmd *);
+extern int transport_generic_allocate_tasks(struct se_cmd *, unsigned char *);
+extern int transport_generic_handle_cdb(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 void __transport_stop_task_timer(struct se_task *, unsigned long *);
+extern unsigned char transport_asciihex_to_binaryhex(unsigned char val[2]);
+extern int transport_generic_map_mem_to_cmd(struct se_cmd *cmd, struct scatterlist *, u32,
+ struct scatterlist *, u32);
+extern int transport_clear_lun_from_sessions(struct se_lun *);
+extern int transport_check_aborted_status(struct se_cmd *, int);
+extern int transport_send_check_condition_and_sense(struct se_cmd *, u8, int);
+extern void transport_send_task_abort(struct se_cmd *);
+extern void transport_release_cmd_to_pool(struct se_cmd *);
+extern void transport_generic_free_cmd(struct se_cmd *, int, int, int);
+extern void transport_generic_wait_for_cmds(struct se_cmd *, int);
+extern u32 transport_calc_sg_num(struct se_task *, struct se_mem *, u32);
+extern int transport_map_mem_to_sg(struct se_task *, struct list_head *,
+ void *, struct se_mem *,
+ struct se_mem **, u32 *, u32 *);
+extern void transport_do_task_sg_chain(struct se_cmd *);
+extern void transport_generic_process_write(struct se_cmd *);
+extern int transport_generic_do_tmr(struct se_cmd *);
+/* From target_core_alua.c */
+extern int core_alua_check_nonop_delay(struct se_cmd *);
+
+/*
+ * Each se_transport_task_t can have N number of possible struct se_task's
+ * for the storage transport(s) to possibly execute.
+ * Used primarily for splitting up CDBs that exceed the physical storage
+ * HBA's maximum sector count per task.
+ */
+struct se_mem {
+ struct page *se_page;
+ u32 se_len;
+ u32 se_off;
+ struct list_head se_list;
+} ____cacheline_aligned;
+
+/*
+ * Each type of disk transport supported MUST have a template defined
+ * within its .h file.
+ */
+struct se_subsystem_api {
+ /*
+ * The Name. :-)
+ */
+ char name[16];
+ /*
+ * Transport Type.
+ */
+ u8 transport_type;
+ /*
+ * struct module for struct se_hba references
+ */
+ struct module *owner;
+ /*
+ * Used for global se_subsystem_api list_head
+ */
+ struct list_head sub_api_list;
+ /*
+ * For SCF_SCSI_NON_DATA_CDB
+ */
+ int (*cdb_none)(struct se_task *);
+ /*
+ * For SCF_SCSI_CONTROL_NONSG_IO_CDB
+ */
+ int (*map_task_non_SG)(struct se_task *);
+ /*
+ * For SCF_SCSI_DATA_SG_IO_CDB and SCF_SCSI_CONTROL_SG_IO_CDB
+ */
+ int (*map_task_SG)(struct se_task *);
+ /*
+ * attach_hba():
+ */
+ int (*attach_hba)(struct se_hba *, u32);
+ /*
+ * detach_hba():
+ */
+ void (*detach_hba)(struct se_hba *);
+ /*
+ * pmode_hba(): Used for TCM/pSCSI subsystem plugin HBA ->
+ * Linux/SCSI struct Scsi_Host passthrough
+ */
+ int (*pmode_enable_hba)(struct se_hba *, unsigned long);
+ /*
+ * allocate_virtdevice():
+ */
+ void *(*allocate_virtdevice)(struct se_hba *, const char *);
+ /*
+ * create_virtdevice(): Only for Virtual HBAs
+ */
+ struct se_device *(*create_virtdevice)(struct se_hba *,
+ struct se_subsystem_dev *, void *);
+ /*
+ * free_device():
+ */
+ void (*free_device)(void *);
+
+ /*
+ * dpo_emulated():
+ */
+ int (*dpo_emulated)(struct se_device *);
+ /*
+ * fua_write_emulated():
+ */
+ int (*fua_write_emulated)(struct se_device *);
+ /*
+ * fua_read_emulated():
+ */
+ int (*fua_read_emulated)(struct se_device *);
+ /*
+ * write_cache_emulated():
+ */
+ int (*write_cache_emulated)(struct se_device *);
+ /*
+ * transport_complete():
+ *
+ * Use transport_generic_complete() for majority of DAS transport
+ * drivers. Provided out of convenience.
+ */
+ int (*transport_complete)(struct se_task *task);
+ struct se_task *(*alloc_task)(struct se_cmd *);
+ /*
+ * do_task():
+ */
+ int (*do_task)(struct se_task *);
+ /*
+ * Used by virtual subsystem plugins IBLOCK and FILEIO to emulate
+ * UNMAP and WRITE_SAME_* w/ UNMAP=1 <-> Linux/Block Discard
+ */
+ int (*do_discard)(struct se_device *, sector_t, u32);
+ /*
+ * Used by virtual subsystem plugins IBLOCK and FILEIO to emulate
+ * SYNCHRONIZE_CACHE_* <-> Linux/Block blkdev_issue_flush()
+ */
+ void (*do_sync_cache)(struct se_task *);
+ /*
+ * free_task():
+ */
+ void (*free_task)(struct se_task *);
+ /*
+ * check_configfs_dev_params():
+ */
+ ssize_t (*check_configfs_dev_params)(struct se_hba *, struct se_subsystem_dev *);
+ /*
+ * set_configfs_dev_params():
+ */
+ ssize_t (*set_configfs_dev_params)(struct se_hba *, struct se_subsystem_dev *,
+ const char *, ssize_t);
+ /*
+ * show_configfs_dev_params():
+ */
+ ssize_t (*show_configfs_dev_params)(struct se_hba *, struct se_subsystem_dev *,
+ char *);
+ /*
+ * get_cdb():
+ */
+ unsigned char *(*get_cdb)(struct se_task *);
+ /*
+ * get_device_rev():
+ */
+ u32 (*get_device_rev)(struct se_device *);
+ /*
+ * get_device_type():
+ */
+ u32 (*get_device_type)(struct se_device *);
+ /*
+ * Get the sector_t from a subsystem backstore..
+ */
+ sector_t (*get_blocks)(struct se_device *);
+ /*
+ * do_se_mem_map():
+ */
+ int (*do_se_mem_map)(struct se_task *, struct list_head *, void *,
+ struct se_mem *, struct se_mem **, u32 *, u32 *);
+ /*
+ * get_sense_buffer():
+ */
+ unsigned char *(*get_sense_buffer)(struct se_task *);
+} ____cacheline_aligned;
+
+#define TRANSPORT(dev) ((dev)->transport)
+#define HBA_TRANSPORT(hba) ((hba)->transport)
+
+extern struct se_global *se_global;
+
+#endif /* TARGET_CORE_TRANSPORT_H */
diff --git a/include/trace/events/compaction.h b/include/trace/events/compaction.h
new file mode 100644
index 0000000..388bcdd
--- /dev/null
+++ b/include/trace/events/compaction.h
@@ -0,0 +1,74 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM compaction
+
+#if !defined(_TRACE_COMPACTION_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_COMPACTION_H
+
+#include <linux/types.h>
+#include <linux/tracepoint.h>
+#include "gfpflags.h"
+
+DECLARE_EVENT_CLASS(mm_compaction_isolate_template,
+
+ TP_PROTO(unsigned long nr_scanned,
+ unsigned long nr_taken),
+
+ TP_ARGS(nr_scanned, nr_taken),
+
+ TP_STRUCT__entry(
+ __field(unsigned long, nr_scanned)
+ __field(unsigned long, nr_taken)
+ ),
+
+ TP_fast_assign(
+ __entry->nr_scanned = nr_scanned;
+ __entry->nr_taken = nr_taken;
+ ),
+
+ TP_printk("nr_scanned=%lu nr_taken=%lu",
+ __entry->nr_scanned,
+ __entry->nr_taken)
+);
+
+DEFINE_EVENT(mm_compaction_isolate_template, mm_compaction_isolate_migratepages,
+
+ TP_PROTO(unsigned long nr_scanned,
+ unsigned long nr_taken),
+
+ TP_ARGS(nr_scanned, nr_taken)
+);
+
+DEFINE_EVENT(mm_compaction_isolate_template, mm_compaction_isolate_freepages,
+ TP_PROTO(unsigned long nr_scanned,
+ unsigned long nr_taken),
+
+ TP_ARGS(nr_scanned, nr_taken)
+);
+
+TRACE_EVENT(mm_compaction_migratepages,
+
+ TP_PROTO(unsigned long nr_migrated,
+ unsigned long nr_failed),
+
+ TP_ARGS(nr_migrated, nr_failed),
+
+ TP_STRUCT__entry(
+ __field(unsigned long, nr_migrated)
+ __field(unsigned long, nr_failed)
+ ),
+
+ TP_fast_assign(
+ __entry->nr_migrated = nr_migrated;
+ __entry->nr_failed = nr_failed;
+ ),
+
+ TP_printk("nr_migrated=%lu nr_failed=%lu",
+ __entry->nr_migrated,
+ __entry->nr_failed)
+);
+
+
+#endif /* _TRACE_COMPACTION_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/events/module.h b/include/trace/events/module.h
index c7bb2f0..c6bae36 100644
--- a/include/trace/events/module.h
+++ b/include/trace/events/module.h
@@ -1,5 +1,15 @@
+/*
+ * Because linux/module.h has tracepoints in the header, and ftrace.h
+ * eventually includes this file, define_trace.h includes linux/module.h
+ * But we do not want the module.h to override the TRACE_SYSTEM macro
+ * variable that define_trace.h is processing, so we only set it
+ * when module events are being processed, which would happen when
+ * CREATE_TRACE_POINTS is defined.
+ */
+#ifdef CREATE_TRACE_POINTS
#undef TRACE_SYSTEM
#define TRACE_SYSTEM module
+#endif
#if !defined(_TRACE_MODULE_H) || defined(TRACE_HEADER_MULTI_READ)
#define _TRACE_MODULE_H
diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h
index c255fcc..ea422aa 100644
--- a/include/trace/events/vmscan.h
+++ b/include/trace/events/vmscan.h
@@ -25,13 +25,13 @@
#define trace_reclaim_flags(page, sync) ( \
(page_is_file_cache(page) ? RECLAIM_WB_FILE : RECLAIM_WB_ANON) | \
- (sync == LUMPY_MODE_SYNC ? RECLAIM_WB_SYNC : RECLAIM_WB_ASYNC) \
+ (sync & RECLAIM_MODE_SYNC ? RECLAIM_WB_SYNC : RECLAIM_WB_ASYNC) \
)
#define trace_shrink_flags(file, sync) ( \
- (sync == LUMPY_MODE_SYNC ? RECLAIM_WB_MIXED : \
+ (sync & RECLAIM_MODE_SYNC ? RECLAIM_WB_MIXED : \
(file ? RECLAIM_WB_FILE : RECLAIM_WB_ANON)) | \
- (sync == LUMPY_MODE_SYNC ? RECLAIM_WB_SYNC : RECLAIM_WB_ASYNC) \
+ (sync & RECLAIM_MODE_SYNC ? RECLAIM_WB_SYNC : RECLAIM_WB_ASYNC) \
)
TRACE_EVENT(mm_vmscan_kswapd_sleep,
diff --git a/include/trace/events/writeback.h b/include/trace/events/writeback.h
index 89a2b2d..4e249b9 100644
--- a/include/trace/events/writeback.h
+++ b/include/trace/events/writeback.h
@@ -81,6 +81,7 @@
TP_ARGS(bdi))
DEFINE_WRITEBACK_EVENT(writeback_nowork);
+DEFINE_WRITEBACK_EVENT(writeback_wake_background);
DEFINE_WRITEBACK_EVENT(writeback_wake_thread);
DEFINE_WRITEBACK_EVENT(writeback_wake_forker_thread);
DEFINE_WRITEBACK_EVENT(writeback_bdi_register);
diff --git a/include/xen/gntdev.h b/include/xen/gntdev.h
new file mode 100644
index 0000000..eb23f41
--- /dev/null
+++ b/include/xen/gntdev.h
@@ -0,0 +1,119 @@
+/******************************************************************************
+ * gntdev.h
+ *
+ * Interface to /dev/xen/gntdev.
+ *
+ * Copyright (c) 2007, D G Murray
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (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
+ * 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.
+ */
+
+#ifndef __LINUX_PUBLIC_GNTDEV_H__
+#define __LINUX_PUBLIC_GNTDEV_H__
+
+struct ioctl_gntdev_grant_ref {
+ /* The domain ID of the grant to be mapped. */
+ uint32_t domid;
+ /* The grant reference of the grant to be mapped. */
+ uint32_t ref;
+};
+
+/*
+ * Inserts the grant references into the mapping table of an instance
+ * of gntdev. N.B. This does not perform the mapping, which is deferred
+ * until mmap() is called with @index as the offset.
+ */
+#define IOCTL_GNTDEV_MAP_GRANT_REF \
+_IOC(_IOC_NONE, 'G', 0, sizeof(struct ioctl_gntdev_map_grant_ref))
+struct ioctl_gntdev_map_grant_ref {
+ /* IN parameters */
+ /* The number of grants to be mapped. */
+ uint32_t count;
+ uint32_t pad;
+ /* OUT parameters */
+ /* The offset to be used on a subsequent call to mmap(). */
+ uint64_t index;
+ /* Variable IN parameter. */
+ /* Array of grant references, of size @count. */
+ struct ioctl_gntdev_grant_ref refs[1];
+};
+
+/*
+ * Removes the grant references from the mapping table of an instance of
+ * of gntdev. N.B. munmap() must be called on the relevant virtual address(es)
+ * before this ioctl is called, or an error will result.
+ */
+#define IOCTL_GNTDEV_UNMAP_GRANT_REF \
+_IOC(_IOC_NONE, 'G', 1, sizeof(struct ioctl_gntdev_unmap_grant_ref))
+struct ioctl_gntdev_unmap_grant_ref {
+ /* IN parameters */
+ /* The offset was returned by the corresponding map operation. */
+ uint64_t index;
+ /* The number of pages to be unmapped. */
+ uint32_t count;
+ uint32_t pad;
+};
+
+/*
+ * Returns the offset in the driver's address space that corresponds
+ * to @vaddr. This can be used to perform a munmap(), followed by an
+ * UNMAP_GRANT_REF ioctl, where no state about the offset is retained by
+ * the caller. The number of pages that were allocated at the same time as
+ * @vaddr is returned in @count.
+ *
+ * N.B. Where more than one page has been mapped into a contiguous range, the
+ * supplied @vaddr must correspond to the start of the range; otherwise
+ * an error will result. It is only possible to munmap() the entire
+ * contiguously-allocated range at once, and not any subrange thereof.
+ */
+#define IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR \
+_IOC(_IOC_NONE, 'G', 2, sizeof(struct ioctl_gntdev_get_offset_for_vaddr))
+struct ioctl_gntdev_get_offset_for_vaddr {
+ /* IN parameters */
+ /* The virtual address of the first mapped page in a range. */
+ uint64_t vaddr;
+ /* OUT parameters */
+ /* The offset that was used in the initial mmap() operation. */
+ uint64_t offset;
+ /* The number of pages mapped in the VM area that begins at @vaddr. */
+ uint32_t count;
+ uint32_t pad;
+};
+
+/*
+ * Sets the maximum number of grants that may mapped at once by this gntdev
+ * instance.
+ *
+ * N.B. This must be called before any other ioctl is performed on the device.
+ */
+#define IOCTL_GNTDEV_SET_MAX_GRANTS \
+_IOC(_IOC_NONE, 'G', 3, sizeof(struct ioctl_gntdev_set_max_grants))
+struct ioctl_gntdev_set_max_grants {
+ /* IN parameter */
+ /* The maximum number of grants that may be mapped at once. */
+ uint32_t count;
+};
+
+#endif /* __LINUX_PUBLIC_GNTDEV_H__ */
diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h
index 9a73170..b1fab6b 100644
--- a/include/xen/grant_table.h
+++ b/include/xen/grant_table.h
@@ -37,10 +37,16 @@
#ifndef __ASM_GNTTAB_H__
#define __ASM_GNTTAB_H__
-#include <asm/xen/hypervisor.h>
+#include <asm/page.h>
+
+#include <xen/interface/xen.h>
#include <xen/interface/grant_table.h>
+
+#include <asm/xen/hypervisor.h>
#include <asm/xen/grant_table.h>
+#include <xen/features.h>
+
/* NR_GRANT_FRAMES must be less than or equal to that configured in Xen */
#define NR_GRANT_FRAMES 4
@@ -107,6 +113,37 @@
void gnttab_grant_foreign_transfer_ref(grant_ref_t, domid_t domid,
unsigned long pfn);
+static inline void
+gnttab_set_map_op(struct gnttab_map_grant_ref *map, phys_addr_t addr,
+ uint32_t flags, grant_ref_t ref, domid_t domid)
+{
+ if (flags & GNTMAP_contains_pte)
+ map->host_addr = addr;
+ else if (xen_feature(XENFEAT_auto_translated_physmap))
+ map->host_addr = __pa(addr);
+ else
+ map->host_addr = addr;
+
+ map->flags = flags;
+ map->ref = ref;
+ map->dom = domid;
+}
+
+static inline void
+gnttab_set_unmap_op(struct gnttab_unmap_grant_ref *unmap, phys_addr_t addr,
+ uint32_t flags, grant_handle_t handle)
+{
+ if (flags & GNTMAP_contains_pte)
+ unmap->host_addr = addr;
+ else if (xen_feature(XENFEAT_auto_translated_physmap))
+ unmap->host_addr = __pa(addr);
+ else
+ unmap->host_addr = addr;
+
+ unmap->handle = handle;
+ unmap->dev_bus_addr = 0;
+}
+
int arch_gnttab_map_shared(unsigned long *frames, unsigned long nr_gframes,
unsigned long max_nr_gframes,
struct grant_entry **__shared);
@@ -118,4 +155,9 @@
#define gnttab_map_vaddr(map) ((void *)(map.host_virt_addr))
+int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
+ struct page **pages, unsigned int count);
+int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
+ struct page **pages, unsigned int count);
+
#endif /* __ASM_GNTTAB_H__ */
diff --git a/init/Kconfig b/init/Kconfig
index 4f6cdbf..4e33790 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -515,21 +515,6 @@
Accept the default if unsure.
-config SRCU_SYNCHRONIZE_DELAY
- int "Microseconds to delay before waiting for readers"
- range 0 20
- default 10
- help
- This option controls how long SRCU delays before entering its
- loop waiting on SRCU readers. The purpose of this loop is
- to avoid the unconditional context-switch penalty that would
- otherwise be incurred if there was an active SRCU reader,
- in a manner similar to adaptive locking schemes. This should
- be set to be a bit longer than the common-case SRCU read-side
- critical-section overhead.
-
- Accept the default if unsure.
-
endmenu # "RCU Subsystem"
config IKCONFIG
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 5c5f4cc..b24d702 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -764,6 +764,7 @@
*/
static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, int mode);
+static struct dentry *cgroup_lookup(struct inode *, struct dentry *, struct nameidata *);
static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry);
static int cgroup_populate_dir(struct cgroup *cgrp);
static const struct inode_operations cgroup_dir_inode_operations;
@@ -860,6 +861,11 @@
iput(inode);
}
+static int cgroup_delete(const struct dentry *d)
+{
+ return 1;
+}
+
static void remove_dir(struct dentry *d)
{
struct dentry *parent = dget(d->d_parent);
@@ -910,7 +916,7 @@
parent = dentry->d_parent;
spin_lock(&parent->d_lock);
- spin_lock(&dentry->d_lock);
+ spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED);
list_del_init(&dentry->d_u.d_child);
spin_unlock(&dentry->d_lock);
spin_unlock(&parent->d_lock);
@@ -1451,6 +1457,7 @@
{
static const struct dentry_operations cgroup_dops = {
.d_iput = cgroup_diput,
+ .d_delete = cgroup_delete,
};
struct inode *inode =
@@ -2195,12 +2202,20 @@
};
static const struct inode_operations cgroup_dir_inode_operations = {
- .lookup = simple_lookup,
+ .lookup = cgroup_lookup,
.mkdir = cgroup_mkdir,
.rmdir = cgroup_rmdir,
.rename = cgroup_rename,
};
+static struct dentry *cgroup_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
+{
+ if (dentry->d_name.len > NAME_MAX)
+ return ERR_PTR(-ENAMETOOLONG);
+ d_add(dentry, NULL);
+ return NULL;
+}
+
/*
* Check if a file is a control file
*/
diff --git a/kernel/fork.c b/kernel/fork.c
index d9b44f2..25e4291 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -66,6 +66,7 @@
#include <linux/posix-timers.h>
#include <linux/user-return-notifier.h>
#include <linux/oom.h>
+#include <linux/khugepaged.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
@@ -330,6 +331,9 @@
retval = ksm_fork(mm, oldmm);
if (retval)
goto out;
+ retval = khugepaged_fork(mm, oldmm);
+ if (retval)
+ goto out;
prev = NULL;
for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) {
@@ -529,6 +533,9 @@
mm_free_pgd(mm);
destroy_context(mm);
mmu_notifier_mm_destroy(mm);
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ VM_BUG_ON(mm->pmd_huge_pte);
+#endif
free_mm(mm);
}
EXPORT_SYMBOL_GPL(__mmdrop);
@@ -543,6 +550,7 @@
if (atomic_dec_and_test(&mm->mm_users)) {
exit_aio(mm);
ksm_exit(mm);
+ khugepaged_exit(mm); /* must run before exit_mmap */
exit_mmap(mm);
set_mm_exe_file(mm, NULL);
if (!list_empty(&mm->mmlist)) {
@@ -669,6 +677,10 @@
mm->token_priority = 0;
mm->last_interval = 0;
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ mm->pmd_huge_pte = NULL;
+#endif
+
if (!mm_init(mm, tsk))
goto fail_nomem;
@@ -910,6 +922,7 @@
sig->oom_adj = current->signal->oom_adj;
sig->oom_score_adj = current->signal->oom_score_adj;
+ sig->oom_score_adj_min = current->signal->oom_score_adj_min;
mutex_init(&sig->cred_guard_mutex);
@@ -1410,23 +1423,6 @@
}
/*
- * We hope to recycle these flags after 2.6.26
- */
- if (unlikely(clone_flags & CLONE_STOPPED)) {
- static int __read_mostly count = 100;
-
- if (count > 0 && printk_ratelimit()) {
- char comm[TASK_COMM_LEN];
-
- count--;
- printk(KERN_INFO "fork(): process `%s' used deprecated "
- "clone flags 0x%lx\n",
- get_task_comm(comm, current),
- clone_flags & CLONE_STOPPED);
- }
- }
-
- /*
* When called from kernel_thread, don't do user tracing stuff.
*/
if (likely(user_mode(regs)))
@@ -1464,16 +1460,7 @@
*/
p->flags &= ~PF_STARTING;
- if (unlikely(clone_flags & CLONE_STOPPED)) {
- /*
- * We'll start up with an immediate SIGSTOP.
- */
- sigaddset(&p->pending.signal, SIGSTOP);
- set_tsk_thread_flag(p, TIF_SIGPENDING);
- __set_task_state(p, TASK_STOPPED);
- } else {
- wake_up_new_task(p, clone_flags);
- }
+ wake_up_new_task(p, clone_flags);
tracehook_report_clone_complete(trace, regs,
clone_flags, nr, p);
diff --git a/kernel/futex.c b/kernel/futex.c
index 3019b92..b766d28 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -233,7 +233,7 @@
{
unsigned long address = (unsigned long)uaddr;
struct mm_struct *mm = current->mm;
- struct page *page;
+ struct page *page, *page_head;
int err;
/*
@@ -265,11 +265,46 @@
if (err < 0)
return err;
- page = compound_head(page);
- lock_page(page);
- if (!page->mapping) {
- unlock_page(page);
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ page_head = page;
+ if (unlikely(PageTail(page))) {
put_page(page);
+ /* serialize against __split_huge_page_splitting() */
+ local_irq_disable();
+ if (likely(__get_user_pages_fast(address, 1, 1, &page) == 1)) {
+ page_head = compound_head(page);
+ /*
+ * page_head is valid pointer but we must pin
+ * it before taking the PG_lock and/or
+ * PG_compound_lock. The moment we re-enable
+ * irqs __split_huge_page_splitting() can
+ * return and the head page can be freed from
+ * under us. We can't take the PG_lock and/or
+ * PG_compound_lock on a page that could be
+ * freed from under us.
+ */
+ if (page != page_head) {
+ get_page(page_head);
+ put_page(page);
+ }
+ local_irq_enable();
+ } else {
+ local_irq_enable();
+ goto again;
+ }
+ }
+#else
+ page_head = compound_head(page);
+ if (page != page_head) {
+ get_page(page_head);
+ put_page(page);
+ }
+#endif
+
+ lock_page(page_head);
+ if (!page_head->mapping) {
+ unlock_page(page_head);
+ put_page(page_head);
goto again;
}
@@ -280,20 +315,20 @@
* it's a read-only handle, it's expected that futexes attach to
* the object not the particular process.
*/
- if (PageAnon(page)) {
+ if (PageAnon(page_head)) {
key->both.offset |= FUT_OFF_MMSHARED; /* ref taken on mm */
key->private.mm = mm;
key->private.address = address;
} else {
key->both.offset |= FUT_OFF_INODE; /* inode-based key */
- key->shared.inode = page->mapping->host;
- key->shared.pgoff = page->index;
+ key->shared.inode = page_head->mapping->host;
+ key->shared.pgoff = page_head->index;
}
get_futex_key_refs(key);
- unlock_page(page);
- put_page(page);
+ unlock_page(page_head);
+ put_page(page_head);
return 0;
}
@@ -791,10 +826,9 @@
new_owner = rt_mutex_next_owner(&pi_state->pi_mutex);
/*
- * This happens when we have stolen the lock and the original
- * pending owner did not enqueue itself back on the rt_mutex.
- * Thats not a tragedy. We know that way, that a lock waiter
- * is on the fly. We make the futex_q waiter the pending owner.
+ * It is possible that the next waiter (the one that brought
+ * this owner to the kernel) timed out and is no longer
+ * waiting on the lock.
*/
if (!new_owner)
new_owner = this->task;
diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
index 9988d03..282f202 100644
--- a/kernel/irq/irqdesc.c
+++ b/kernel/irq/irqdesc.c
@@ -72,6 +72,8 @@
static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node)
{
+ int cpu;
+
desc->irq_data.irq = irq;
desc->irq_data.chip = &no_irq_chip;
desc->irq_data.chip_data = NULL;
@@ -83,7 +85,8 @@
desc->irq_count = 0;
desc->irqs_unhandled = 0;
desc->name = NULL;
- memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs)));
+ for_each_possible_cpu(cpu)
+ *per_cpu_ptr(desc->kstat_irqs, cpu) = 0;
desc_smp_init(desc, node);
}
@@ -133,8 +136,7 @@
if (!desc)
return NULL;
/* allocate based on nr_cpu_ids */
- desc->kstat_irqs = kzalloc_node(nr_cpu_ids * sizeof(*desc->kstat_irqs),
- gfp, node);
+ desc->kstat_irqs = alloc_percpu(unsigned int);
if (!desc->kstat_irqs)
goto err_desc;
@@ -149,7 +151,7 @@
return desc;
err_kstat:
- kfree(desc->kstat_irqs);
+ free_percpu(desc->kstat_irqs);
err_desc:
kfree(desc);
return NULL;
@@ -166,7 +168,7 @@
mutex_unlock(&sparse_irq_lock);
free_masks(desc);
- kfree(desc->kstat_irqs);
+ free_percpu(desc->kstat_irqs);
kfree(desc);
}
@@ -234,7 +236,6 @@
}
};
-static unsigned int kstat_irqs_all[NR_IRQS][NR_CPUS];
int __init early_irq_init(void)
{
int count, i, node = first_online_node;
@@ -250,7 +251,8 @@
for (i = 0; i < count; i++) {
desc[i].irq_data.irq = i;
desc[i].irq_data.chip = &no_irq_chip;
- desc[i].kstat_irqs = kstat_irqs_all[i];
+ /* TODO : do this allocation on-demand ... */
+ desc[i].kstat_irqs = alloc_percpu(unsigned int);
alloc_masks(desc + i, GFP_KERNEL, node);
desc_smp_init(desc + i, node);
lockdep_set_class(&desc[i].lock, &irq_desc_lock_class);
@@ -275,6 +277,22 @@
static inline int alloc_descs(unsigned int start, unsigned int cnt, int node)
{
+#if defined(CONFIG_KSTAT_IRQS_ONDEMAND)
+ struct irq_desc *desc;
+ unsigned int i;
+
+ for (i = 0; i < cnt; i++) {
+ desc = irq_to_desc(start + i);
+ if (desc && !desc->kstat_irqs) {
+ unsigned int __percpu *stats = alloc_percpu(unsigned int);
+
+ if (!stats)
+ return -1;
+ if (cmpxchg(&desc->kstat_irqs, NULL, stats) != NULL)
+ free_percpu(stats);
+ }
+ }
+#endif
return start;
}
#endif /* !CONFIG_SPARSE_IRQ */
@@ -391,7 +409,9 @@
unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
{
struct irq_desc *desc = irq_to_desc(irq);
- return desc ? desc->kstat_irqs[cpu] : 0;
+
+ return desc && desc->kstat_irqs ?
+ *per_cpu_ptr(desc->kstat_irqs, cpu) : 0;
}
#ifdef CONFIG_GENERIC_HARDIRQS
@@ -401,10 +421,10 @@
int cpu;
int sum = 0;
- if (!desc)
+ if (!desc || !desc->kstat_irqs)
return 0;
for_each_possible_cpu(cpu)
- sum += desc->kstat_irqs[cpu];
+ sum += *per_cpu_ptr(desc->kstat_irqs, cpu);
return sum;
}
#endif /* CONFIG_GENERIC_HARDIRQS */
diff --git a/kernel/panic.c b/kernel/panic.c
index 4c13b1a..991bb87 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -34,6 +34,7 @@
static DEFINE_SPINLOCK(pause_on_oops_lock);
int panic_timeout;
+EXPORT_SYMBOL_GPL(panic_timeout);
ATOMIC_NOTIFIER_HEAD(panic_notifier_list);
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index a5aff3e..2657299 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -100,13 +100,9 @@
depends on PM_ADVANCED_DEBUG
default n
-config SUSPEND_NVS
- bool
-
config SUSPEND
bool "Suspend to RAM and standby"
depends on PM && ARCH_SUSPEND_POSSIBLE
- select SUSPEND_NVS if HAS_IOMEM
default y
---help---
Allow the system to enter sleep states in which main memory is
@@ -140,7 +136,6 @@
depends on PM && SWAP && ARCH_HIBERNATION_POSSIBLE
select LZO_COMPRESS
select LZO_DECOMPRESS
- select SUSPEND_NVS if HAS_IOMEM
---help---
Enable the suspend to disk (STD) functionality, which is usually
called "hibernation" in user interfaces. STD checkpoints the
diff --git a/kernel/power/Makefile b/kernel/power/Makefile
index b755972..c350e18 100644
--- a/kernel/power/Makefile
+++ b/kernel/power/Makefile
@@ -7,6 +7,5 @@
obj-$(CONFIG_PM_TEST_SUSPEND) += suspend_test.o
obj-$(CONFIG_HIBERNATION) += hibernate.o snapshot.o swap.o user.o \
block_io.o
-obj-$(CONFIG_SUSPEND_NVS) += nvs.o
obj-$(CONFIG_MAGIC_SYSRQ) += poweroff.o
diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c
index 0344937..0c343b9 100644
--- a/kernel/rcutiny.c
+++ b/kernel/rcutiny.c
@@ -189,7 +189,8 @@
unsigned long flags;
for (;;) {
- wait_event(rcu_kthread_wq, have_rcu_kthread_work != 0);
+ wait_event_interruptible(rcu_kthread_wq,
+ have_rcu_kthread_work != 0);
morework = rcu_boost();
local_irq_save(flags);
work = have_rcu_kthread_work;
diff --git a/kernel/srcu.c b/kernel/srcu.c
index 98d8c1e..73ce23f 100644
--- a/kernel/srcu.c
+++ b/kernel/srcu.c
@@ -156,6 +156,16 @@
EXPORT_SYMBOL_GPL(__srcu_read_unlock);
/*
+ * We use an adaptive strategy for synchronize_srcu() and especially for
+ * synchronize_srcu_expedited(). We spin for a fixed time period
+ * (defined below) to allow SRCU readers to exit their read-side critical
+ * sections. If there are still some readers after 10 microseconds,
+ * we repeatedly block for 1-millisecond time periods. This approach
+ * has done well in testing, so there is no need for a config parameter.
+ */
+#define SYNCHRONIZE_SRCU_READER_DELAY 10
+
+/*
* Helper function for synchronize_srcu() and synchronize_srcu_expedited().
*/
static void __synchronize_srcu(struct srcu_struct *sp, void (*sync_func)(void))
@@ -207,11 +217,12 @@
* will have finished executing. We initially give readers
* an arbitrarily chosen 10 microseconds to get out of their
* SRCU read-side critical sections, then loop waiting 1/HZ
- * seconds per iteration.
+ * seconds per iteration. The 10-microsecond value has done
+ * very well in testing.
*/
if (srcu_readers_active_idx(sp, idx))
- udelay(CONFIG_SRCU_SYNCHRONIZE_DELAY);
+ udelay(SYNCHRONIZE_SRCU_READER_DELAY);
while (srcu_readers_active_idx(sp, idx))
schedule_timeout_interruptible(1);
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index c50a034..6519cf62 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -113,7 +113,7 @@
* @shift: pointer to shift variable
* @from: frequency to convert from
* @to: frequency to convert to
- * @minsec: guaranteed runtime conversion range in seconds
+ * @maxsec: guaranteed runtime conversion range in seconds
*
* The function evaluates the shift/mult pair for the scaled math
* operations of clocksources and clockevents.
@@ -122,7 +122,7 @@
* NSEC_PER_SEC == 1GHz and @from is the counter frequency. For clock
* event @to is the counter frequency and @from is NSEC_PER_SEC.
*
- * The @minsec conversion range argument controls the time frame in
+ * The @maxsec conversion range argument controls the time frame in
* seconds which must be covered by the runtime conversion with the
* calculated mult and shift factors. This guarantees that no 64bit
* overflow happens when the input value of the conversion is
@@ -131,7 +131,7 @@
* factors.
*/
void
-clocks_calc_mult_shift(u32 *mult, u32 *shift, u32 from, u32 to, u32 minsec)
+clocks_calc_mult_shift(u32 *mult, u32 *shift, u32 from, u32 to, u32 maxsec)
{
u64 tmp;
u32 sft, sftacc= 32;
@@ -140,7 +140,7 @@
* Calculate the shift factor which is limiting the conversion
* range:
*/
- tmp = ((u64)minsec * from) >> 32;
+ tmp = ((u64)maxsec * from) >> 32;
while (tmp) {
tmp >>=1;
sftacc--;
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 5536aaf..d27c756 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -49,7 +49,7 @@
u32 mult;
};
-struct timekeeper timekeeper;
+static struct timekeeper timekeeper;
/**
* timekeeper_setup_internals - Set up internals to use clocksource clock.
@@ -164,7 +164,7 @@
/*
* The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock.
*/
-struct timespec raw_time;
+static struct timespec raw_time;
/* flag for if timekeeping is suspended */
int __read_mostly timekeeping_suspended;
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index bac752f..b706529 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -23,9 +23,6 @@
static int syscall_enter_define_fields(struct ftrace_event_call *call);
static int syscall_exit_define_fields(struct ftrace_event_call *call);
-/* All syscall exit events have the same fields */
-static LIST_HEAD(syscall_exit_fields);
-
static struct list_head *
syscall_get_enter_fields(struct ftrace_event_call *call)
{
@@ -34,34 +31,28 @@
return &entry->enter_fields;
}
-static struct list_head *
-syscall_get_exit_fields(struct ftrace_event_call *call)
-{
- return &syscall_exit_fields;
-}
-
struct trace_event_functions enter_syscall_print_funcs = {
- .trace = print_syscall_enter,
+ .trace = print_syscall_enter,
};
struct trace_event_functions exit_syscall_print_funcs = {
- .trace = print_syscall_exit,
+ .trace = print_syscall_exit,
};
struct ftrace_event_class event_class_syscall_enter = {
- .system = "syscalls",
- .reg = syscall_enter_register,
- .define_fields = syscall_enter_define_fields,
- .get_fields = syscall_get_enter_fields,
- .raw_init = init_syscall_trace,
+ .system = "syscalls",
+ .reg = syscall_enter_register,
+ .define_fields = syscall_enter_define_fields,
+ .get_fields = syscall_get_enter_fields,
+ .raw_init = init_syscall_trace,
};
struct ftrace_event_class event_class_syscall_exit = {
- .system = "syscalls",
- .reg = syscall_exit_register,
- .define_fields = syscall_exit_define_fields,
- .get_fields = syscall_get_exit_fields,
- .raw_init = init_syscall_trace,
+ .system = "syscalls",
+ .reg = syscall_exit_register,
+ .define_fields = syscall_exit_define_fields,
+ .fields = LIST_HEAD_INIT(event_class_syscall_exit.fields),
+ .raw_init = init_syscall_trace,
};
extern unsigned long __start_syscalls_metadata[];
diff --git a/lib/ioremap.c b/lib/ioremap.c
index 5730ecd..da4e2ad 100644
--- a/lib/ioremap.c
+++ b/lib/ioremap.c
@@ -9,6 +9,7 @@
#include <linux/mm.h>
#include <linux/sched.h>
#include <linux/io.h>
+#include <linux/module.h>
#include <asm/cacheflush.h>
#include <asm/pgtable.h>
@@ -90,3 +91,4 @@
return err;
}
+EXPORT_SYMBOL_GPL(ioremap_page_range);
diff --git a/mm/Kconfig b/mm/Kconfig
index c2c8a4a..3ad483b 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -302,6 +302,44 @@
See Documentation/nommu-mmap.txt for more information.
+config TRANSPARENT_HUGEPAGE
+ bool "Transparent Hugepage Support"
+ depends on X86 && MMU
+ select COMPACTION
+ help
+ Transparent Hugepages allows the kernel to use huge pages and
+ huge tlb transparently to the applications whenever possible.
+ This feature can improve computing performance to certain
+ applications by speeding up page faults during memory
+ allocation, by reducing the number of tlb misses and by speeding
+ up the pagetable walking.
+
+ If memory constrained on embedded, you may want to say N.
+
+choice
+ prompt "Transparent Hugepage Support sysfs defaults"
+ depends on TRANSPARENT_HUGEPAGE
+ default TRANSPARENT_HUGEPAGE_ALWAYS
+ help
+ Selects the sysfs defaults for Transparent Hugepage Support.
+
+ config TRANSPARENT_HUGEPAGE_ALWAYS
+ bool "always"
+ help
+ Enabling Transparent Hugepage always, can increase the
+ memory footprint of applications without a guaranteed
+ benefit but it will work automatically for all applications.
+
+ config TRANSPARENT_HUGEPAGE_MADVISE
+ bool "madvise"
+ help
+ Enabling Transparent Hugepage madvise, will only provide a
+ performance improvement benefit to the applications using
+ madvise(MADV_HUGEPAGE) but it won't risk to increase the
+ memory footprint of applications without a guaranteed
+ benefit.
+endchoice
+
#
# UP and nommu archs use km based percpu allocator
#
diff --git a/mm/Makefile b/mm/Makefile
index f73f75a..2b1b575 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -5,7 +5,7 @@
mmu-y := nommu.o
mmu-$(CONFIG_MMU) := fremap.o highmem.o madvise.o memory.o mincore.o \
mlock.o mmap.o mprotect.o mremap.o msync.o rmap.o \
- vmalloc.o pagewalk.o
+ vmalloc.o pagewalk.o pgtable-generic.o
obj-y := bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \
maccess.o page_alloc.o page-writeback.o \
@@ -37,6 +37,7 @@
obj-$(CONFIG_FS_XIP) += filemap_xip.o
obj-$(CONFIG_MIGRATION) += migrate.o
obj-$(CONFIG_QUICKLIST) += quicklist.o
+obj-$(CONFIG_TRANSPARENT_HUGEPAGE) += huge_memory.o
obj-$(CONFIG_CGROUP_MEM_RES_CTLR) += memcontrol.o page_cgroup.o
obj-$(CONFIG_MEMORY_FAILURE) += memory-failure.o
obj-$(CONFIG_HWPOISON_INJECT) += hwpoison-inject.o
diff --git a/mm/compaction.c b/mm/compaction.c
index 1a8894e..6d592a0 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -16,6 +16,9 @@
#include <linux/sysfs.h>
#include "internal.h"
+#define CREATE_TRACE_POINTS
+#include <trace/events/compaction.h>
+
/*
* compact_control is used to track pages being migrated and the free pages
* they are being migrated to during memory compaction. The free_pfn starts
@@ -30,6 +33,7 @@
unsigned long nr_migratepages; /* Number of pages to migrate */
unsigned long free_pfn; /* isolate_freepages search base */
unsigned long migrate_pfn; /* isolate_migratepages search base */
+ bool sync; /* Synchronous migration */
/* Account for isolated anon and file pages */
unsigned long nr_anon;
@@ -38,6 +42,8 @@
unsigned int order; /* order a direct compactor needs */
int migratetype; /* MOVABLE, RECLAIMABLE etc */
struct zone *zone;
+
+ int compact_mode;
};
static unsigned long release_freepages(struct list_head *freelist)
@@ -60,7 +66,7 @@
struct list_head *freelist)
{
unsigned long zone_end_pfn, end_pfn;
- int total_isolated = 0;
+ int nr_scanned = 0, total_isolated = 0;
struct page *cursor;
/* Get the last PFN we should scan for free pages at */
@@ -81,6 +87,7 @@
if (!pfn_valid_within(blockpfn))
continue;
+ nr_scanned++;
if (!PageBuddy(page))
continue;
@@ -100,6 +107,7 @@
}
}
+ trace_mm_compaction_isolate_freepages(nr_scanned, total_isolated);
return total_isolated;
}
@@ -234,6 +242,8 @@
struct compact_control *cc)
{
unsigned long low_pfn, end_pfn;
+ unsigned long last_pageblock_nr = 0, pageblock_nr;
+ unsigned long nr_scanned = 0, nr_isolated = 0;
struct list_head *migratelist = &cc->migratepages;
/* Do not scan outside zone boundaries */
@@ -266,20 +276,51 @@
struct page *page;
if (!pfn_valid_within(low_pfn))
continue;
+ nr_scanned++;
/* Get the page and skip if free */
page = pfn_to_page(low_pfn);
if (PageBuddy(page))
continue;
+ /*
+ * For async migration, also only scan in MOVABLE blocks. Async
+ * migration is optimistic to see if the minimum amount of work
+ * satisfies the allocation
+ */
+ pageblock_nr = low_pfn >> pageblock_order;
+ if (!cc->sync && last_pageblock_nr != pageblock_nr &&
+ get_pageblock_migratetype(page) != MIGRATE_MOVABLE) {
+ low_pfn += pageblock_nr_pages;
+ low_pfn = ALIGN(low_pfn, pageblock_nr_pages) - 1;
+ last_pageblock_nr = pageblock_nr;
+ continue;
+ }
+
+ if (!PageLRU(page))
+ continue;
+
+ /*
+ * PageLRU is set, and lru_lock excludes isolation,
+ * splitting and collapsing (collapsing has already
+ * happened if PageLRU is set).
+ */
+ if (PageTransHuge(page)) {
+ low_pfn += (1 << compound_order(page)) - 1;
+ continue;
+ }
+
/* Try isolate the page */
if (__isolate_lru_page(page, ISOLATE_BOTH, 0) != 0)
continue;
+ VM_BUG_ON(PageTransCompound(page));
+
/* Successfully isolated */
del_page_from_lru_list(zone, page, page_lru(page));
list_add(&page->lru, migratelist);
cc->nr_migratepages++;
+ nr_isolated++;
/* Avoid isolating too much */
if (cc->nr_migratepages == COMPACT_CLUSTER_MAX)
@@ -291,6 +332,8 @@
spin_unlock_irq(&zone->lru_lock);
cc->migrate_pfn = low_pfn;
+ trace_mm_compaction_isolate_migratepages(nr_scanned, nr_isolated);
+
return cc->nr_migratepages;
}
@@ -341,10 +384,10 @@
}
static int compact_finished(struct zone *zone,
- struct compact_control *cc)
+ struct compact_control *cc)
{
unsigned int order;
- unsigned long watermark = low_wmark_pages(zone) + (1 << cc->order);
+ unsigned long watermark;
if (fatal_signal_pending(current))
return COMPACT_PARTIAL;
@@ -354,12 +397,27 @@
return COMPACT_COMPLETE;
/* Compaction run is not finished if the watermark is not met */
+ if (cc->compact_mode != COMPACT_MODE_KSWAPD)
+ watermark = low_wmark_pages(zone);
+ else
+ watermark = high_wmark_pages(zone);
+ watermark += (1 << cc->order);
+
if (!zone_watermark_ok(zone, cc->order, watermark, 0, 0))
return COMPACT_CONTINUE;
if (cc->order == -1)
return COMPACT_CONTINUE;
+ /*
+ * Generating only one page of the right order is not enough
+ * for kswapd, we must continue until we're above the high
+ * watermark as a pool for high order GFP_ATOMIC allocations
+ * too.
+ */
+ if (cc->compact_mode == COMPACT_MODE_KSWAPD)
+ return COMPACT_CONTINUE;
+
/* Direct compactor: Is a suitable page free? */
for (order = cc->order; order < MAX_ORDER; order++) {
/* Job done if page is free of the right migratetype */
@@ -374,10 +432,62 @@
return COMPACT_CONTINUE;
}
+/*
+ * compaction_suitable: Is this suitable to run compaction on this zone now?
+ * Returns
+ * COMPACT_SKIPPED - If there are too few free pages for compaction
+ * COMPACT_PARTIAL - If the allocation would succeed without compaction
+ * COMPACT_CONTINUE - If compaction should run now
+ */
+unsigned long compaction_suitable(struct zone *zone, int order)
+{
+ int fragindex;
+ unsigned long watermark;
+
+ /*
+ * Watermarks for order-0 must be met for compaction. Note the 2UL.
+ * This is because during migration, copies of pages need to be
+ * allocated and for a short time, the footprint is higher
+ */
+ watermark = low_wmark_pages(zone) + (2UL << order);
+ if (!zone_watermark_ok(zone, 0, watermark, 0, 0))
+ return COMPACT_SKIPPED;
+
+ /*
+ * fragmentation index determines if allocation failures are due to
+ * low memory or external fragmentation
+ *
+ * index of -1 implies allocations might succeed dependingon watermarks
+ * index towards 0 implies failure is due to lack of memory
+ * index towards 1000 implies failure is due to fragmentation
+ *
+ * Only compact if a failure would be due to fragmentation.
+ */
+ fragindex = fragmentation_index(zone, order);
+ if (fragindex >= 0 && fragindex <= sysctl_extfrag_threshold)
+ return COMPACT_SKIPPED;
+
+ if (fragindex == -1 && zone_watermark_ok(zone, order, watermark, 0, 0))
+ return COMPACT_PARTIAL;
+
+ return COMPACT_CONTINUE;
+}
+
static int compact_zone(struct zone *zone, struct compact_control *cc)
{
int ret;
+ ret = compaction_suitable(zone, cc->order);
+ switch (ret) {
+ case COMPACT_PARTIAL:
+ case COMPACT_SKIPPED:
+ /* Compaction is likely to fail */
+ return ret;
+ case COMPACT_CONTINUE:
+ /* Fall through to compaction */
+ ;
+ }
+
/* Setup to move all movable pages to the end of the zone */
cc->migrate_pfn = zone->zone_start_pfn;
cc->free_pfn = cc->migrate_pfn + zone->spanned_pages;
@@ -393,7 +503,8 @@
nr_migrate = cc->nr_migratepages;
migrate_pages(&cc->migratepages, compaction_alloc,
- (unsigned long)cc, 0);
+ (unsigned long)cc, false,
+ cc->sync);
update_nr_listpages(cc);
nr_remaining = cc->nr_migratepages;
@@ -401,6 +512,8 @@
count_vm_events(COMPACTPAGES, nr_migrate - nr_remaining);
if (nr_remaining)
count_vm_events(COMPACTPAGEFAILED, nr_remaining);
+ trace_mm_compaction_migratepages(nr_migrate - nr_remaining,
+ nr_remaining);
/* Release LRU pages not migrated */
if (!list_empty(&cc->migratepages)) {
@@ -417,8 +530,10 @@
return ret;
}
-static unsigned long compact_zone_order(struct zone *zone,
- int order, gfp_t gfp_mask)
+unsigned long compact_zone_order(struct zone *zone,
+ int order, gfp_t gfp_mask,
+ bool sync,
+ int compact_mode)
{
struct compact_control cc = {
.nr_freepages = 0,
@@ -426,6 +541,8 @@
.order = order,
.migratetype = allocflags_to_migratetype(gfp_mask),
.zone = zone,
+ .sync = sync,
+ .compact_mode = compact_mode,
};
INIT_LIST_HEAD(&cc.freepages);
INIT_LIST_HEAD(&cc.migratepages);
@@ -441,16 +558,17 @@
* @order: The order of the current allocation
* @gfp_mask: The GFP mask of the current allocation
* @nodemask: The allowed nodes to allocate from
+ * @sync: Whether migration is synchronous or not
*
* This is the main entry point for direct page compaction.
*/
unsigned long try_to_compact_pages(struct zonelist *zonelist,
- int order, gfp_t gfp_mask, nodemask_t *nodemask)
+ int order, gfp_t gfp_mask, nodemask_t *nodemask,
+ bool sync)
{
enum zone_type high_zoneidx = gfp_zone(gfp_mask);
int may_enter_fs = gfp_mask & __GFP_FS;
int may_perform_io = gfp_mask & __GFP_IO;
- unsigned long watermark;
struct zoneref *z;
struct zone *zone;
int rc = COMPACT_SKIPPED;
@@ -460,7 +578,7 @@
* made because an assumption is made that the page allocator can satisfy
* the "cheaper" orders without taking special steps
*/
- if (order <= PAGE_ALLOC_COSTLY_ORDER || !may_enter_fs || !may_perform_io)
+ if (!order || !may_enter_fs || !may_perform_io)
return rc;
count_vm_event(COMPACTSTALL);
@@ -468,43 +586,14 @@
/* Compact each zone in the list */
for_each_zone_zonelist_nodemask(zone, z, zonelist, high_zoneidx,
nodemask) {
- int fragindex;
int status;
- /*
- * Watermarks for order-0 must be met for compaction. Note
- * the 2UL. This is because during migration, copies of
- * pages need to be allocated and for a short time, the
- * footprint is higher
- */
- watermark = low_wmark_pages(zone) + (2UL << order);
- if (!zone_watermark_ok(zone, 0, watermark, 0, 0))
- continue;
-
- /*
- * fragmentation index determines if allocation failures are
- * due to low memory or external fragmentation
- *
- * index of -1 implies allocations might succeed depending
- * on watermarks
- * index towards 0 implies failure is due to lack of memory
- * index towards 1000 implies failure is due to fragmentation
- *
- * Only compact if a failure would be due to fragmentation.
- */
- fragindex = fragmentation_index(zone, order);
- if (fragindex >= 0 && fragindex <= sysctl_extfrag_threshold)
- continue;
-
- if (fragindex == -1 && zone_watermark_ok(zone, order, watermark, 0, 0)) {
- rc = COMPACT_PARTIAL;
- break;
- }
-
- status = compact_zone_order(zone, order, gfp_mask);
+ status = compact_zone_order(zone, order, gfp_mask, sync,
+ COMPACT_MODE_DIRECT_RECLAIM);
rc = max(status, rc);
- if (zone_watermark_ok(zone, order, watermark, 0, 0))
+ /* If a normal allocation would succeed, stop compacting */
+ if (zone_watermark_ok(zone, order, low_wmark_pages(zone), 0, 0))
break;
}
@@ -531,6 +620,7 @@
.nr_freepages = 0,
.nr_migratepages = 0,
.order = -1,
+ .compact_mode = COMPACT_MODE_DIRECT_RECLAIM,
};
zone = &pgdat->node_zones[zoneid];
diff --git a/mm/dmapool.c b/mm/dmapool.c
index 4df2de7..03bf3bb 100644
--- a/mm/dmapool.c
+++ b/mm/dmapool.c
@@ -324,7 +324,7 @@
if (mem_flags & __GFP_WAIT) {
DECLARE_WAITQUEUE(wait, current);
- __set_current_state(TASK_INTERRUPTIBLE);
+ __set_current_state(TASK_UNINTERRUPTIBLE);
__add_wait_queue(&pool->waitq, &wait);
spin_unlock_irqrestore(&pool->lock, flags);
@@ -355,20 +355,15 @@
static struct dma_page *pool_find_page(struct dma_pool *pool, dma_addr_t dma)
{
- unsigned long flags;
struct dma_page *page;
- spin_lock_irqsave(&pool->lock, flags);
list_for_each_entry(page, &pool->page_list, page_list) {
if (dma < page->dma)
continue;
if (dma < (page->dma + pool->allocation))
- goto done;
+ return page;
}
- page = NULL;
- done:
- spin_unlock_irqrestore(&pool->lock, flags);
- return page;
+ return NULL;
}
/**
@@ -386,8 +381,10 @@
unsigned long flags;
unsigned int offset;
+ spin_lock_irqsave(&pool->lock, flags);
page = pool_find_page(pool, dma);
if (!page) {
+ spin_unlock_irqrestore(&pool->lock, flags);
if (pool->dev)
dev_err(pool->dev,
"dma_pool_free %s, %p/%lx (bad dma)\n",
@@ -401,6 +398,7 @@
offset = vaddr - page->vaddr;
#ifdef DMAPOOL_DEBUG
if ((dma - page->dma) != offset) {
+ spin_unlock_irqrestore(&pool->lock, flags);
if (pool->dev)
dev_err(pool->dev,
"dma_pool_free %s, %p (bad vaddr)/%Lx\n",
@@ -418,6 +416,7 @@
chain = *(int *)(page->vaddr + chain);
continue;
}
+ spin_unlock_irqrestore(&pool->lock, flags);
if (pool->dev)
dev_err(pool->dev, "dma_pool_free %s, dma %Lx "
"already free\n", pool->name,
@@ -432,7 +431,6 @@
memset(vaddr, POOL_POISON_FREED, pool->size);
#endif
- spin_lock_irqsave(&pool->lock, flags);
page->in_use--;
*(int *)vaddr = page->offset;
page->offset = offset;
diff --git a/mm/filemap.c b/mm/filemap.c
index ca38939..83a45d3 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -298,7 +298,7 @@
continue;
wait_on_page_writeback(page);
- if (PageError(page))
+ if (TestClearPageError(page))
ret = -EIO;
}
pagevec_release(&pvec);
@@ -837,9 +837,6 @@
if (radix_tree_deref_retry(page))
goto restart;
- if (page->mapping == NULL || page->index != index)
- break;
-
if (!page_cache_get_speculative(page))
goto repeat;
@@ -849,6 +846,16 @@
goto repeat;
}
+ /*
+ * must check mapping and index after taking the ref.
+ * otherwise we can get both false positives and false
+ * negatives, which is just confusing to the caller.
+ */
+ if (page->mapping == NULL || page->index != index) {
+ page_cache_release(page);
+ break;
+ }
+
pages[ret] = page;
ret++;
index++;
@@ -2220,7 +2227,7 @@
gfp_notmask = __GFP_FS;
repeat:
page = find_lock_page(mapping, index);
- if (likely(page))
+ if (page)
return page;
page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~gfp_notmask);
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
new file mode 100644
index 0000000..004c9c2
--- /dev/null
+++ b/mm/huge_memory.c
@@ -0,0 +1,2346 @@
+/*
+ * Copyright (C) 2009 Red Hat, Inc.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ */
+
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/highmem.h>
+#include <linux/hugetlb.h>
+#include <linux/mmu_notifier.h>
+#include <linux/rmap.h>
+#include <linux/swap.h>
+#include <linux/mm_inline.h>
+#include <linux/kthread.h>
+#include <linux/khugepaged.h>
+#include <linux/freezer.h>
+#include <linux/mman.h>
+#include <asm/tlb.h>
+#include <asm/pgalloc.h>
+#include "internal.h"
+
+/*
+ * By default transparent hugepage support is enabled for all mappings
+ * and khugepaged scans all mappings. Defrag is only invoked by
+ * khugepaged hugepage allocations and by page faults inside
+ * MADV_HUGEPAGE regions to avoid the risk of slowing down short lived
+ * allocations.
+ */
+unsigned long transparent_hugepage_flags __read_mostly =
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE_ALWAYS
+ (1<<TRANSPARENT_HUGEPAGE_FLAG)|
+#endif
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE_MADVISE
+ (1<<TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG)|
+#endif
+ (1<<TRANSPARENT_HUGEPAGE_DEFRAG_FLAG)|
+ (1<<TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG);
+
+/* default scan 8*512 pte (or vmas) every 30 second */
+static unsigned int khugepaged_pages_to_scan __read_mostly = HPAGE_PMD_NR*8;
+static unsigned int khugepaged_pages_collapsed;
+static unsigned int khugepaged_full_scans;
+static unsigned int khugepaged_scan_sleep_millisecs __read_mostly = 10000;
+/* during fragmentation poll the hugepage allocator once every minute */
+static unsigned int khugepaged_alloc_sleep_millisecs __read_mostly = 60000;
+static struct task_struct *khugepaged_thread __read_mostly;
+static DEFINE_MUTEX(khugepaged_mutex);
+static DEFINE_SPINLOCK(khugepaged_mm_lock);
+static DECLARE_WAIT_QUEUE_HEAD(khugepaged_wait);
+/*
+ * default collapse hugepages if there is at least one pte mapped like
+ * it would have happened if the vma was large enough during page
+ * fault.
+ */
+static unsigned int khugepaged_max_ptes_none __read_mostly = HPAGE_PMD_NR-1;
+
+static int khugepaged(void *none);
+static int mm_slots_hash_init(void);
+static int khugepaged_slab_init(void);
+static void khugepaged_slab_free(void);
+
+#define MM_SLOTS_HASH_HEADS 1024
+static struct hlist_head *mm_slots_hash __read_mostly;
+static struct kmem_cache *mm_slot_cache __read_mostly;
+
+/**
+ * struct mm_slot - hash lookup from mm to mm_slot
+ * @hash: hash collision list
+ * @mm_node: khugepaged scan list headed in khugepaged_scan.mm_head
+ * @mm: the mm that this information is valid for
+ */
+struct mm_slot {
+ struct hlist_node hash;
+ struct list_head mm_node;
+ struct mm_struct *mm;
+};
+
+/**
+ * struct khugepaged_scan - cursor for scanning
+ * @mm_head: the head of the mm list to scan
+ * @mm_slot: the current mm_slot we are scanning
+ * @address: the next address inside that to be scanned
+ *
+ * There is only the one khugepaged_scan instance of this cursor structure.
+ */
+struct khugepaged_scan {
+ struct list_head mm_head;
+ struct mm_slot *mm_slot;
+ unsigned long address;
+} khugepaged_scan = {
+ .mm_head = LIST_HEAD_INIT(khugepaged_scan.mm_head),
+};
+
+
+static int set_recommended_min_free_kbytes(void)
+{
+ struct zone *zone;
+ int nr_zones = 0;
+ unsigned long recommended_min;
+ extern int min_free_kbytes;
+
+ if (!test_bit(TRANSPARENT_HUGEPAGE_FLAG,
+ &transparent_hugepage_flags) &&
+ !test_bit(TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG,
+ &transparent_hugepage_flags))
+ return 0;
+
+ for_each_populated_zone(zone)
+ nr_zones++;
+
+ /* Make sure at least 2 hugepages are free for MIGRATE_RESERVE */
+ recommended_min = pageblock_nr_pages * nr_zones * 2;
+
+ /*
+ * Make sure that on average at least two pageblocks are almost free
+ * of another type, one for a migratetype to fall back to and a
+ * second to avoid subsequent fallbacks of other types There are 3
+ * MIGRATE_TYPES we care about.
+ */
+ recommended_min += pageblock_nr_pages * nr_zones *
+ MIGRATE_PCPTYPES * MIGRATE_PCPTYPES;
+
+ /* don't ever allow to reserve more than 5% of the lowmem */
+ recommended_min = min(recommended_min,
+ (unsigned long) nr_free_buffer_pages() / 20);
+ recommended_min <<= (PAGE_SHIFT-10);
+
+ if (recommended_min > min_free_kbytes)
+ min_free_kbytes = recommended_min;
+ setup_per_zone_wmarks();
+ return 0;
+}
+late_initcall(set_recommended_min_free_kbytes);
+
+static int start_khugepaged(void)
+{
+ int err = 0;
+ if (khugepaged_enabled()) {
+ int wakeup;
+ if (unlikely(!mm_slot_cache || !mm_slots_hash)) {
+ err = -ENOMEM;
+ goto out;
+ }
+ mutex_lock(&khugepaged_mutex);
+ if (!khugepaged_thread)
+ khugepaged_thread = kthread_run(khugepaged, NULL,
+ "khugepaged");
+ if (unlikely(IS_ERR(khugepaged_thread))) {
+ printk(KERN_ERR
+ "khugepaged: kthread_run(khugepaged) failed\n");
+ err = PTR_ERR(khugepaged_thread);
+ khugepaged_thread = NULL;
+ }
+ wakeup = !list_empty(&khugepaged_scan.mm_head);
+ mutex_unlock(&khugepaged_mutex);
+ if (wakeup)
+ wake_up_interruptible(&khugepaged_wait);
+
+ set_recommended_min_free_kbytes();
+ } else
+ /* wakeup to exit */
+ wake_up_interruptible(&khugepaged_wait);
+out:
+ return err;
+}
+
+#ifdef CONFIG_SYSFS
+
+static ssize_t double_flag_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf,
+ enum transparent_hugepage_flag enabled,
+ enum transparent_hugepage_flag req_madv)
+{
+ if (test_bit(enabled, &transparent_hugepage_flags)) {
+ VM_BUG_ON(test_bit(req_madv, &transparent_hugepage_flags));
+ return sprintf(buf, "[always] madvise never\n");
+ } else if (test_bit(req_madv, &transparent_hugepage_flags))
+ return sprintf(buf, "always [madvise] never\n");
+ else
+ return sprintf(buf, "always madvise [never]\n");
+}
+static ssize_t double_flag_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count,
+ enum transparent_hugepage_flag enabled,
+ enum transparent_hugepage_flag req_madv)
+{
+ if (!memcmp("always", buf,
+ min(sizeof("always")-1, count))) {
+ set_bit(enabled, &transparent_hugepage_flags);
+ clear_bit(req_madv, &transparent_hugepage_flags);
+ } else if (!memcmp("madvise", buf,
+ min(sizeof("madvise")-1, count))) {
+ clear_bit(enabled, &transparent_hugepage_flags);
+ set_bit(req_madv, &transparent_hugepage_flags);
+ } else if (!memcmp("never", buf,
+ min(sizeof("never")-1, count))) {
+ clear_bit(enabled, &transparent_hugepage_flags);
+ clear_bit(req_madv, &transparent_hugepage_flags);
+ } else
+ return -EINVAL;
+
+ return count;
+}
+
+static ssize_t enabled_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return double_flag_show(kobj, attr, buf,
+ TRANSPARENT_HUGEPAGE_FLAG,
+ TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG);
+}
+static ssize_t enabled_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ ssize_t ret;
+
+ ret = double_flag_store(kobj, attr, buf, count,
+ TRANSPARENT_HUGEPAGE_FLAG,
+ TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG);
+
+ if (ret > 0) {
+ int err = start_khugepaged();
+ if (err)
+ ret = err;
+ }
+
+ if (ret > 0 &&
+ (test_bit(TRANSPARENT_HUGEPAGE_FLAG,
+ &transparent_hugepage_flags) ||
+ test_bit(TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG,
+ &transparent_hugepage_flags)))
+ set_recommended_min_free_kbytes();
+
+ return ret;
+}
+static struct kobj_attribute enabled_attr =
+ __ATTR(enabled, 0644, enabled_show, enabled_store);
+
+static ssize_t single_flag_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf,
+ enum transparent_hugepage_flag flag)
+{
+ if (test_bit(flag, &transparent_hugepage_flags))
+ return sprintf(buf, "[yes] no\n");
+ else
+ return sprintf(buf, "yes [no]\n");
+}
+static ssize_t single_flag_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count,
+ enum transparent_hugepage_flag flag)
+{
+ if (!memcmp("yes", buf,
+ min(sizeof("yes")-1, count))) {
+ set_bit(flag, &transparent_hugepage_flags);
+ } else if (!memcmp("no", buf,
+ min(sizeof("no")-1, count))) {
+ clear_bit(flag, &transparent_hugepage_flags);
+ } else
+ return -EINVAL;
+
+ return count;
+}
+
+/*
+ * Currently defrag only disables __GFP_NOWAIT for allocation. A blind
+ * __GFP_REPEAT is too aggressive, it's never worth swapping tons of
+ * memory just to allocate one more hugepage.
+ */
+static ssize_t defrag_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return double_flag_show(kobj, attr, buf,
+ TRANSPARENT_HUGEPAGE_DEFRAG_FLAG,
+ TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG);
+}
+static ssize_t defrag_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ return double_flag_store(kobj, attr, buf, count,
+ TRANSPARENT_HUGEPAGE_DEFRAG_FLAG,
+ TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG);
+}
+static struct kobj_attribute defrag_attr =
+ __ATTR(defrag, 0644, defrag_show, defrag_store);
+
+#ifdef CONFIG_DEBUG_VM
+static ssize_t debug_cow_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return single_flag_show(kobj, attr, buf,
+ TRANSPARENT_HUGEPAGE_DEBUG_COW_FLAG);
+}
+static ssize_t debug_cow_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ return single_flag_store(kobj, attr, buf, count,
+ TRANSPARENT_HUGEPAGE_DEBUG_COW_FLAG);
+}
+static struct kobj_attribute debug_cow_attr =
+ __ATTR(debug_cow, 0644, debug_cow_show, debug_cow_store);
+#endif /* CONFIG_DEBUG_VM */
+
+static struct attribute *hugepage_attr[] = {
+ &enabled_attr.attr,
+ &defrag_attr.attr,
+#ifdef CONFIG_DEBUG_VM
+ &debug_cow_attr.attr,
+#endif
+ NULL,
+};
+
+static struct attribute_group hugepage_attr_group = {
+ .attrs = hugepage_attr,
+};
+
+static ssize_t scan_sleep_millisecs_show(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%u\n", khugepaged_scan_sleep_millisecs);
+}
+
+static ssize_t scan_sleep_millisecs_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ unsigned long msecs;
+ int err;
+
+ err = strict_strtoul(buf, 10, &msecs);
+ if (err || msecs > UINT_MAX)
+ return -EINVAL;
+
+ khugepaged_scan_sleep_millisecs = msecs;
+ wake_up_interruptible(&khugepaged_wait);
+
+ return count;
+}
+static struct kobj_attribute scan_sleep_millisecs_attr =
+ __ATTR(scan_sleep_millisecs, 0644, scan_sleep_millisecs_show,
+ scan_sleep_millisecs_store);
+
+static ssize_t alloc_sleep_millisecs_show(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%u\n", khugepaged_alloc_sleep_millisecs);
+}
+
+static ssize_t alloc_sleep_millisecs_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ unsigned long msecs;
+ int err;
+
+ err = strict_strtoul(buf, 10, &msecs);
+ if (err || msecs > UINT_MAX)
+ return -EINVAL;
+
+ khugepaged_alloc_sleep_millisecs = msecs;
+ wake_up_interruptible(&khugepaged_wait);
+
+ return count;
+}
+static struct kobj_attribute alloc_sleep_millisecs_attr =
+ __ATTR(alloc_sleep_millisecs, 0644, alloc_sleep_millisecs_show,
+ alloc_sleep_millisecs_store);
+
+static ssize_t pages_to_scan_show(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%u\n", khugepaged_pages_to_scan);
+}
+static ssize_t pages_to_scan_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ int err;
+ unsigned long pages;
+
+ err = strict_strtoul(buf, 10, &pages);
+ if (err || !pages || pages > UINT_MAX)
+ return -EINVAL;
+
+ khugepaged_pages_to_scan = pages;
+
+ return count;
+}
+static struct kobj_attribute pages_to_scan_attr =
+ __ATTR(pages_to_scan, 0644, pages_to_scan_show,
+ pages_to_scan_store);
+
+static ssize_t pages_collapsed_show(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%u\n", khugepaged_pages_collapsed);
+}
+static struct kobj_attribute pages_collapsed_attr =
+ __ATTR_RO(pages_collapsed);
+
+static ssize_t full_scans_show(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%u\n", khugepaged_full_scans);
+}
+static struct kobj_attribute full_scans_attr =
+ __ATTR_RO(full_scans);
+
+static ssize_t khugepaged_defrag_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return single_flag_show(kobj, attr, buf,
+ TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG);
+}
+static ssize_t khugepaged_defrag_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ return single_flag_store(kobj, attr, buf, count,
+ TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG);
+}
+static struct kobj_attribute khugepaged_defrag_attr =
+ __ATTR(defrag, 0644, khugepaged_defrag_show,
+ khugepaged_defrag_store);
+
+/*
+ * max_ptes_none controls if khugepaged should collapse hugepages over
+ * any unmapped ptes in turn potentially increasing the memory
+ * footprint of the vmas. When max_ptes_none is 0 khugepaged will not
+ * reduce the available free memory in the system as it
+ * runs. Increasing max_ptes_none will instead potentially reduce the
+ * free memory in the system during the khugepaged scan.
+ */
+static ssize_t khugepaged_max_ptes_none_show(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%u\n", khugepaged_max_ptes_none);
+}
+static ssize_t khugepaged_max_ptes_none_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ int err;
+ unsigned long max_ptes_none;
+
+ err = strict_strtoul(buf, 10, &max_ptes_none);
+ if (err || max_ptes_none > HPAGE_PMD_NR-1)
+ return -EINVAL;
+
+ khugepaged_max_ptes_none = max_ptes_none;
+
+ return count;
+}
+static struct kobj_attribute khugepaged_max_ptes_none_attr =
+ __ATTR(max_ptes_none, 0644, khugepaged_max_ptes_none_show,
+ khugepaged_max_ptes_none_store);
+
+static struct attribute *khugepaged_attr[] = {
+ &khugepaged_defrag_attr.attr,
+ &khugepaged_max_ptes_none_attr.attr,
+ &pages_to_scan_attr.attr,
+ &pages_collapsed_attr.attr,
+ &full_scans_attr.attr,
+ &scan_sleep_millisecs_attr.attr,
+ &alloc_sleep_millisecs_attr.attr,
+ NULL,
+};
+
+static struct attribute_group khugepaged_attr_group = {
+ .attrs = khugepaged_attr,
+ .name = "khugepaged",
+};
+#endif /* CONFIG_SYSFS */
+
+static int __init hugepage_init(void)
+{
+ int err;
+#ifdef CONFIG_SYSFS
+ static struct kobject *hugepage_kobj;
+#endif
+
+ err = -EINVAL;
+ if (!has_transparent_hugepage()) {
+ transparent_hugepage_flags = 0;
+ goto out;
+ }
+
+#ifdef CONFIG_SYSFS
+ err = -ENOMEM;
+ hugepage_kobj = kobject_create_and_add("transparent_hugepage", mm_kobj);
+ if (unlikely(!hugepage_kobj)) {
+ printk(KERN_ERR "hugepage: failed kobject create\n");
+ goto out;
+ }
+
+ err = sysfs_create_group(hugepage_kobj, &hugepage_attr_group);
+ if (err) {
+ printk(KERN_ERR "hugepage: failed register hugeage group\n");
+ goto out;
+ }
+
+ err = sysfs_create_group(hugepage_kobj, &khugepaged_attr_group);
+ if (err) {
+ printk(KERN_ERR "hugepage: failed register hugeage group\n");
+ goto out;
+ }
+#endif
+
+ err = khugepaged_slab_init();
+ if (err)
+ goto out;
+
+ err = mm_slots_hash_init();
+ if (err) {
+ khugepaged_slab_free();
+ goto out;
+ }
+
+ /*
+ * By default disable transparent hugepages on smaller systems,
+ * where the extra memory used could hurt more than TLB overhead
+ * is likely to save. The admin can still enable it through /sys.
+ */
+ if (totalram_pages < (512 << (20 - PAGE_SHIFT)))
+ transparent_hugepage_flags = 0;
+
+ start_khugepaged();
+
+ set_recommended_min_free_kbytes();
+
+out:
+ return err;
+}
+module_init(hugepage_init)
+
+static int __init setup_transparent_hugepage(char *str)
+{
+ int ret = 0;
+ if (!str)
+ goto out;
+ if (!strcmp(str, "always")) {
+ set_bit(TRANSPARENT_HUGEPAGE_FLAG,
+ &transparent_hugepage_flags);
+ clear_bit(TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG,
+ &transparent_hugepage_flags);
+ ret = 1;
+ } else if (!strcmp(str, "madvise")) {
+ clear_bit(TRANSPARENT_HUGEPAGE_FLAG,
+ &transparent_hugepage_flags);
+ set_bit(TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG,
+ &transparent_hugepage_flags);
+ ret = 1;
+ } else if (!strcmp(str, "never")) {
+ clear_bit(TRANSPARENT_HUGEPAGE_FLAG,
+ &transparent_hugepage_flags);
+ clear_bit(TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG,
+ &transparent_hugepage_flags);
+ ret = 1;
+ }
+out:
+ if (!ret)
+ printk(KERN_WARNING
+ "transparent_hugepage= cannot parse, ignored\n");
+ return ret;
+}
+__setup("transparent_hugepage=", setup_transparent_hugepage);
+
+static void prepare_pmd_huge_pte(pgtable_t pgtable,
+ struct mm_struct *mm)
+{
+ assert_spin_locked(&mm->page_table_lock);
+
+ /* FIFO */
+ if (!mm->pmd_huge_pte)
+ INIT_LIST_HEAD(&pgtable->lru);
+ else
+ list_add(&pgtable->lru, &mm->pmd_huge_pte->lru);
+ mm->pmd_huge_pte = pgtable;
+}
+
+static inline pmd_t maybe_pmd_mkwrite(pmd_t pmd, struct vm_area_struct *vma)
+{
+ if (likely(vma->vm_flags & VM_WRITE))
+ pmd = pmd_mkwrite(pmd);
+ return pmd;
+}
+
+static int __do_huge_pmd_anonymous_page(struct mm_struct *mm,
+ struct vm_area_struct *vma,
+ unsigned long haddr, pmd_t *pmd,
+ struct page *page)
+{
+ int ret = 0;
+ pgtable_t pgtable;
+
+ VM_BUG_ON(!PageCompound(page));
+ pgtable = pte_alloc_one(mm, haddr);
+ if (unlikely(!pgtable)) {
+ mem_cgroup_uncharge_page(page);
+ put_page(page);
+ return VM_FAULT_OOM;
+ }
+
+ clear_huge_page(page, haddr, HPAGE_PMD_NR);
+ __SetPageUptodate(page);
+
+ spin_lock(&mm->page_table_lock);
+ if (unlikely(!pmd_none(*pmd))) {
+ spin_unlock(&mm->page_table_lock);
+ mem_cgroup_uncharge_page(page);
+ put_page(page);
+ pte_free(mm, pgtable);
+ } else {
+ pmd_t entry;
+ entry = mk_pmd(page, vma->vm_page_prot);
+ entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma);
+ entry = pmd_mkhuge(entry);
+ /*
+ * The spinlocking to take the lru_lock inside
+ * page_add_new_anon_rmap() acts as a full memory
+ * barrier to be sure clear_huge_page writes become
+ * visible after the set_pmd_at() write.
+ */
+ page_add_new_anon_rmap(page, vma, haddr);
+ set_pmd_at(mm, haddr, pmd, entry);
+ prepare_pmd_huge_pte(pgtable, mm);
+ add_mm_counter(mm, MM_ANONPAGES, HPAGE_PMD_NR);
+ spin_unlock(&mm->page_table_lock);
+ }
+
+ return ret;
+}
+
+static inline gfp_t alloc_hugepage_gfpmask(int defrag)
+{
+ return GFP_TRANSHUGE & ~(defrag ? 0 : __GFP_WAIT);
+}
+
+static inline struct page *alloc_hugepage_vma(int defrag,
+ struct vm_area_struct *vma,
+ unsigned long haddr)
+{
+ return alloc_pages_vma(alloc_hugepage_gfpmask(defrag),
+ HPAGE_PMD_ORDER, vma, haddr);
+}
+
+#ifndef CONFIG_NUMA
+static inline struct page *alloc_hugepage(int defrag)
+{
+ return alloc_pages(alloc_hugepage_gfpmask(defrag),
+ HPAGE_PMD_ORDER);
+}
+#endif
+
+int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmd,
+ unsigned int flags)
+{
+ struct page *page;
+ unsigned long haddr = address & HPAGE_PMD_MASK;
+ pte_t *pte;
+
+ if (haddr >= vma->vm_start && haddr + HPAGE_PMD_SIZE <= vma->vm_end) {
+ if (unlikely(anon_vma_prepare(vma)))
+ return VM_FAULT_OOM;
+ if (unlikely(khugepaged_enter(vma)))
+ return VM_FAULT_OOM;
+ page = alloc_hugepage_vma(transparent_hugepage_defrag(vma),
+ vma, haddr);
+ if (unlikely(!page))
+ goto out;
+ if (unlikely(mem_cgroup_newpage_charge(page, mm, GFP_KERNEL))) {
+ put_page(page);
+ goto out;
+ }
+
+ return __do_huge_pmd_anonymous_page(mm, vma, haddr, pmd, page);
+ }
+out:
+ /*
+ * Use __pte_alloc instead of pte_alloc_map, because we can't
+ * run pte_offset_map on the pmd, if an huge pmd could
+ * materialize from under us from a different thread.
+ */
+ if (unlikely(__pte_alloc(mm, vma, pmd, address)))
+ return VM_FAULT_OOM;
+ /* if an huge pmd materialized from under us just retry later */
+ if (unlikely(pmd_trans_huge(*pmd)))
+ return 0;
+ /*
+ * A regular pmd is established and it can't morph into a huge pmd
+ * from under us anymore at this point because we hold the mmap_sem
+ * read mode and khugepaged takes it in write mode. So now it's
+ * safe to run pte_offset_map().
+ */
+ pte = pte_offset_map(pmd, address);
+ return handle_pte_fault(mm, vma, address, pte, pmd, flags);
+}
+
+int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
+ pmd_t *dst_pmd, pmd_t *src_pmd, unsigned long addr,
+ struct vm_area_struct *vma)
+{
+ struct page *src_page;
+ pmd_t pmd;
+ pgtable_t pgtable;
+ int ret;
+
+ ret = -ENOMEM;
+ pgtable = pte_alloc_one(dst_mm, addr);
+ if (unlikely(!pgtable))
+ goto out;
+
+ spin_lock(&dst_mm->page_table_lock);
+ spin_lock_nested(&src_mm->page_table_lock, SINGLE_DEPTH_NESTING);
+
+ ret = -EAGAIN;
+ pmd = *src_pmd;
+ if (unlikely(!pmd_trans_huge(pmd))) {
+ pte_free(dst_mm, pgtable);
+ goto out_unlock;
+ }
+ if (unlikely(pmd_trans_splitting(pmd))) {
+ /* split huge page running from under us */
+ spin_unlock(&src_mm->page_table_lock);
+ spin_unlock(&dst_mm->page_table_lock);
+ pte_free(dst_mm, pgtable);
+
+ wait_split_huge_page(vma->anon_vma, src_pmd); /* src_vma */
+ goto out;
+ }
+ src_page = pmd_page(pmd);
+ VM_BUG_ON(!PageHead(src_page));
+ get_page(src_page);
+ page_dup_rmap(src_page);
+ add_mm_counter(dst_mm, MM_ANONPAGES, HPAGE_PMD_NR);
+
+ pmdp_set_wrprotect(src_mm, addr, src_pmd);
+ pmd = pmd_mkold(pmd_wrprotect(pmd));
+ set_pmd_at(dst_mm, addr, dst_pmd, pmd);
+ prepare_pmd_huge_pte(pgtable, dst_mm);
+
+ ret = 0;
+out_unlock:
+ spin_unlock(&src_mm->page_table_lock);
+ spin_unlock(&dst_mm->page_table_lock);
+out:
+ return ret;
+}
+
+/* no "address" argument so destroys page coloring of some arch */
+pgtable_t get_pmd_huge_pte(struct mm_struct *mm)
+{
+ pgtable_t pgtable;
+
+ assert_spin_locked(&mm->page_table_lock);
+
+ /* FIFO */
+ pgtable = mm->pmd_huge_pte;
+ if (list_empty(&pgtable->lru))
+ mm->pmd_huge_pte = NULL;
+ else {
+ mm->pmd_huge_pte = list_entry(pgtable->lru.next,
+ struct page, lru);
+ list_del(&pgtable->lru);
+ }
+ return pgtable;
+}
+
+static int do_huge_pmd_wp_page_fallback(struct mm_struct *mm,
+ struct vm_area_struct *vma,
+ unsigned long address,
+ pmd_t *pmd, pmd_t orig_pmd,
+ struct page *page,
+ unsigned long haddr)
+{
+ pgtable_t pgtable;
+ pmd_t _pmd;
+ int ret = 0, i;
+ struct page **pages;
+
+ pages = kmalloc(sizeof(struct page *) * HPAGE_PMD_NR,
+ GFP_KERNEL);
+ if (unlikely(!pages)) {
+ ret |= VM_FAULT_OOM;
+ goto out;
+ }
+
+ for (i = 0; i < HPAGE_PMD_NR; i++) {
+ pages[i] = alloc_page_vma(GFP_HIGHUSER_MOVABLE,
+ vma, address);
+ if (unlikely(!pages[i] ||
+ mem_cgroup_newpage_charge(pages[i], mm,
+ GFP_KERNEL))) {
+ if (pages[i])
+ put_page(pages[i]);
+ mem_cgroup_uncharge_start();
+ while (--i >= 0) {
+ mem_cgroup_uncharge_page(pages[i]);
+ put_page(pages[i]);
+ }
+ mem_cgroup_uncharge_end();
+ kfree(pages);
+ ret |= VM_FAULT_OOM;
+ goto out;
+ }
+ }
+
+ for (i = 0; i < HPAGE_PMD_NR; i++) {
+ copy_user_highpage(pages[i], page + i,
+ haddr + PAGE_SHIFT*i, vma);
+ __SetPageUptodate(pages[i]);
+ cond_resched();
+ }
+
+ spin_lock(&mm->page_table_lock);
+ if (unlikely(!pmd_same(*pmd, orig_pmd)))
+ goto out_free_pages;
+ VM_BUG_ON(!PageHead(page));
+
+ pmdp_clear_flush_notify(vma, haddr, pmd);
+ /* leave pmd empty until pte is filled */
+
+ pgtable = get_pmd_huge_pte(mm);
+ pmd_populate(mm, &_pmd, pgtable);
+
+ for (i = 0; i < HPAGE_PMD_NR; i++, haddr += PAGE_SIZE) {
+ pte_t *pte, entry;
+ entry = mk_pte(pages[i], vma->vm_page_prot);
+ entry = maybe_mkwrite(pte_mkdirty(entry), vma);
+ page_add_new_anon_rmap(pages[i], vma, haddr);
+ pte = pte_offset_map(&_pmd, haddr);
+ VM_BUG_ON(!pte_none(*pte));
+ set_pte_at(mm, haddr, pte, entry);
+ pte_unmap(pte);
+ }
+ kfree(pages);
+
+ mm->nr_ptes++;
+ smp_wmb(); /* make pte visible before pmd */
+ pmd_populate(mm, pmd, pgtable);
+ page_remove_rmap(page);
+ spin_unlock(&mm->page_table_lock);
+
+ ret |= VM_FAULT_WRITE;
+ put_page(page);
+
+out:
+ return ret;
+
+out_free_pages:
+ spin_unlock(&mm->page_table_lock);
+ mem_cgroup_uncharge_start();
+ for (i = 0; i < HPAGE_PMD_NR; i++) {
+ mem_cgroup_uncharge_page(pages[i]);
+ put_page(pages[i]);
+ }
+ mem_cgroup_uncharge_end();
+ kfree(pages);
+ goto out;
+}
+
+int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmd, pmd_t orig_pmd)
+{
+ int ret = 0;
+ struct page *page, *new_page;
+ unsigned long haddr;
+
+ VM_BUG_ON(!vma->anon_vma);
+ spin_lock(&mm->page_table_lock);
+ if (unlikely(!pmd_same(*pmd, orig_pmd)))
+ goto out_unlock;
+
+ page = pmd_page(orig_pmd);
+ VM_BUG_ON(!PageCompound(page) || !PageHead(page));
+ haddr = address & HPAGE_PMD_MASK;
+ if (page_mapcount(page) == 1) {
+ pmd_t entry;
+ entry = pmd_mkyoung(orig_pmd);
+ entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma);
+ if (pmdp_set_access_flags(vma, haddr, pmd, entry, 1))
+ update_mmu_cache(vma, address, entry);
+ ret |= VM_FAULT_WRITE;
+ goto out_unlock;
+ }
+ get_page(page);
+ spin_unlock(&mm->page_table_lock);
+
+ if (transparent_hugepage_enabled(vma) &&
+ !transparent_hugepage_debug_cow())
+ new_page = alloc_hugepage_vma(transparent_hugepage_defrag(vma),
+ vma, haddr);
+ else
+ new_page = NULL;
+
+ if (unlikely(!new_page)) {
+ ret = do_huge_pmd_wp_page_fallback(mm, vma, address,
+ pmd, orig_pmd, page, haddr);
+ put_page(page);
+ goto out;
+ }
+
+ if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) {
+ put_page(new_page);
+ put_page(page);
+ ret |= VM_FAULT_OOM;
+ goto out;
+ }
+
+ copy_user_huge_page(new_page, page, haddr, vma, HPAGE_PMD_NR);
+ __SetPageUptodate(new_page);
+
+ spin_lock(&mm->page_table_lock);
+ put_page(page);
+ if (unlikely(!pmd_same(*pmd, orig_pmd))) {
+ mem_cgroup_uncharge_page(new_page);
+ put_page(new_page);
+ } else {
+ pmd_t entry;
+ VM_BUG_ON(!PageHead(page));
+ entry = mk_pmd(new_page, vma->vm_page_prot);
+ entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma);
+ entry = pmd_mkhuge(entry);
+ pmdp_clear_flush_notify(vma, haddr, pmd);
+ page_add_new_anon_rmap(new_page, vma, haddr);
+ set_pmd_at(mm, haddr, pmd, entry);
+ update_mmu_cache(vma, address, entry);
+ page_remove_rmap(page);
+ put_page(page);
+ ret |= VM_FAULT_WRITE;
+ }
+out_unlock:
+ spin_unlock(&mm->page_table_lock);
+out:
+ return ret;
+}
+
+struct page *follow_trans_huge_pmd(struct mm_struct *mm,
+ unsigned long addr,
+ pmd_t *pmd,
+ unsigned int flags)
+{
+ struct page *page = NULL;
+
+ assert_spin_locked(&mm->page_table_lock);
+
+ if (flags & FOLL_WRITE && !pmd_write(*pmd))
+ goto out;
+
+ page = pmd_page(*pmd);
+ VM_BUG_ON(!PageHead(page));
+ if (flags & FOLL_TOUCH) {
+ pmd_t _pmd;
+ /*
+ * We should set the dirty bit only for FOLL_WRITE but
+ * for now the dirty bit in the pmd is meaningless.
+ * And if the dirty bit will become meaningful and
+ * we'll only set it with FOLL_WRITE, an atomic
+ * set_bit will be required on the pmd to set the
+ * young bit, instead of the current set_pmd_at.
+ */
+ _pmd = pmd_mkyoung(pmd_mkdirty(*pmd));
+ set_pmd_at(mm, addr & HPAGE_PMD_MASK, pmd, _pmd);
+ }
+ page += (addr & ~HPAGE_PMD_MASK) >> PAGE_SHIFT;
+ VM_BUG_ON(!PageCompound(page));
+ if (flags & FOLL_GET)
+ get_page(page);
+
+out:
+ return page;
+}
+
+int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
+ pmd_t *pmd)
+{
+ int ret = 0;
+
+ spin_lock(&tlb->mm->page_table_lock);
+ if (likely(pmd_trans_huge(*pmd))) {
+ if (unlikely(pmd_trans_splitting(*pmd))) {
+ spin_unlock(&tlb->mm->page_table_lock);
+ wait_split_huge_page(vma->anon_vma,
+ pmd);
+ } else {
+ struct page *page;
+ pgtable_t pgtable;
+ pgtable = get_pmd_huge_pte(tlb->mm);
+ page = pmd_page(*pmd);
+ pmd_clear(pmd);
+ page_remove_rmap(page);
+ VM_BUG_ON(page_mapcount(page) < 0);
+ add_mm_counter(tlb->mm, MM_ANONPAGES, -HPAGE_PMD_NR);
+ VM_BUG_ON(!PageHead(page));
+ spin_unlock(&tlb->mm->page_table_lock);
+ tlb_remove_page(tlb, page);
+ pte_free(tlb->mm, pgtable);
+ ret = 1;
+ }
+ } else
+ spin_unlock(&tlb->mm->page_table_lock);
+
+ return ret;
+}
+
+int mincore_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
+ unsigned long addr, unsigned long end,
+ unsigned char *vec)
+{
+ int ret = 0;
+
+ spin_lock(&vma->vm_mm->page_table_lock);
+ if (likely(pmd_trans_huge(*pmd))) {
+ ret = !pmd_trans_splitting(*pmd);
+ spin_unlock(&vma->vm_mm->page_table_lock);
+ if (unlikely(!ret))
+ wait_split_huge_page(vma->anon_vma, pmd);
+ else {
+ /*
+ * All logical pages in the range are present
+ * if backed by a huge page.
+ */
+ memset(vec, 1, (end - addr) >> PAGE_SHIFT);
+ }
+ } else
+ spin_unlock(&vma->vm_mm->page_table_lock);
+
+ return ret;
+}
+
+int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
+ unsigned long addr, pgprot_t newprot)
+{
+ struct mm_struct *mm = vma->vm_mm;
+ int ret = 0;
+
+ spin_lock(&mm->page_table_lock);
+ if (likely(pmd_trans_huge(*pmd))) {
+ if (unlikely(pmd_trans_splitting(*pmd))) {
+ spin_unlock(&mm->page_table_lock);
+ wait_split_huge_page(vma->anon_vma, pmd);
+ } else {
+ pmd_t entry;
+
+ entry = pmdp_get_and_clear(mm, addr, pmd);
+ entry = pmd_modify(entry, newprot);
+ set_pmd_at(mm, addr, pmd, entry);
+ spin_unlock(&vma->vm_mm->page_table_lock);
+ flush_tlb_range(vma, addr, addr + HPAGE_PMD_SIZE);
+ ret = 1;
+ }
+ } else
+ spin_unlock(&vma->vm_mm->page_table_lock);
+
+ return ret;
+}
+
+pmd_t *page_check_address_pmd(struct page *page,
+ struct mm_struct *mm,
+ unsigned long address,
+ enum page_check_address_pmd_flag flag)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd, *ret = NULL;
+
+ if (address & ~HPAGE_PMD_MASK)
+ goto out;
+
+ pgd = pgd_offset(mm, address);
+ if (!pgd_present(*pgd))
+ goto out;
+
+ pud = pud_offset(pgd, address);
+ if (!pud_present(*pud))
+ goto out;
+
+ pmd = pmd_offset(pud, address);
+ if (pmd_none(*pmd))
+ goto out;
+ if (pmd_page(*pmd) != page)
+ goto out;
+ /*
+ * split_vma() may create temporary aliased mappings. There is
+ * no risk as long as all huge pmd are found and have their
+ * splitting bit set before __split_huge_page_refcount
+ * runs. Finding the same huge pmd more than once during the
+ * same rmap walk is not a problem.
+ */
+ if (flag == PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG &&
+ pmd_trans_splitting(*pmd))
+ goto out;
+ if (pmd_trans_huge(*pmd)) {
+ VM_BUG_ON(flag == PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG &&
+ !pmd_trans_splitting(*pmd));
+ ret = pmd;
+ }
+out:
+ return ret;
+}
+
+static int __split_huge_page_splitting(struct page *page,
+ struct vm_area_struct *vma,
+ unsigned long address)
+{
+ struct mm_struct *mm = vma->vm_mm;
+ pmd_t *pmd;
+ int ret = 0;
+
+ spin_lock(&mm->page_table_lock);
+ pmd = page_check_address_pmd(page, mm, address,
+ PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG);
+ if (pmd) {
+ /*
+ * We can't temporarily set the pmd to null in order
+ * to split it, the pmd must remain marked huge at all
+ * times or the VM won't take the pmd_trans_huge paths
+ * and it won't wait on the anon_vma->root->lock to
+ * serialize against split_huge_page*.
+ */
+ pmdp_splitting_flush_notify(vma, address, pmd);
+ ret = 1;
+ }
+ spin_unlock(&mm->page_table_lock);
+
+ return ret;
+}
+
+static void __split_huge_page_refcount(struct page *page)
+{
+ int i;
+ unsigned long head_index = page->index;
+ struct zone *zone = page_zone(page);
+ int zonestat;
+
+ /* prevent PageLRU to go away from under us, and freeze lru stats */
+ spin_lock_irq(&zone->lru_lock);
+ compound_lock(page);
+
+ for (i = 1; i < HPAGE_PMD_NR; i++) {
+ struct page *page_tail = page + i;
+
+ /* tail_page->_count cannot change */
+ atomic_sub(atomic_read(&page_tail->_count), &page->_count);
+ BUG_ON(page_count(page) <= 0);
+ atomic_add(page_mapcount(page) + 1, &page_tail->_count);
+ BUG_ON(atomic_read(&page_tail->_count) <= 0);
+
+ /* after clearing PageTail the gup refcount can be released */
+ smp_mb();
+
+ page_tail->flags &= ~PAGE_FLAGS_CHECK_AT_PREP;
+ page_tail->flags |= (page->flags &
+ ((1L << PG_referenced) |
+ (1L << PG_swapbacked) |
+ (1L << PG_mlocked) |
+ (1L << PG_uptodate)));
+ page_tail->flags |= (1L << PG_dirty);
+
+ /*
+ * 1) clear PageTail before overwriting first_page
+ * 2) clear PageTail before clearing PageHead for VM_BUG_ON
+ */
+ smp_wmb();
+
+ /*
+ * __split_huge_page_splitting() already set the
+ * splitting bit in all pmd that could map this
+ * hugepage, that will ensure no CPU can alter the
+ * mapcount on the head page. The mapcount is only
+ * accounted in the head page and it has to be
+ * transferred to all tail pages in the below code. So
+ * for this code to be safe, the split the mapcount
+ * can't change. But that doesn't mean userland can't
+ * keep changing and reading the page contents while
+ * we transfer the mapcount, so the pmd splitting
+ * status is achieved setting a reserved bit in the
+ * pmd, not by clearing the present bit.
+ */
+ BUG_ON(page_mapcount(page_tail));
+ page_tail->_mapcount = page->_mapcount;
+
+ BUG_ON(page_tail->mapping);
+ page_tail->mapping = page->mapping;
+
+ page_tail->index = ++head_index;
+
+ BUG_ON(!PageAnon(page_tail));
+ BUG_ON(!PageUptodate(page_tail));
+ BUG_ON(!PageDirty(page_tail));
+ BUG_ON(!PageSwapBacked(page_tail));
+
+ lru_add_page_tail(zone, page, page_tail);
+ }
+
+ __dec_zone_page_state(page, NR_ANON_TRANSPARENT_HUGEPAGES);
+ __mod_zone_page_state(zone, NR_ANON_PAGES, HPAGE_PMD_NR);
+
+ /*
+ * A hugepage counts for HPAGE_PMD_NR pages on the LRU statistics,
+ * so adjust those appropriately if this page is on the LRU.
+ */
+ if (PageLRU(page)) {
+ zonestat = NR_LRU_BASE + page_lru(page);
+ __mod_zone_page_state(zone, zonestat, -(HPAGE_PMD_NR-1));
+ }
+
+ ClearPageCompound(page);
+ compound_unlock(page);
+ spin_unlock_irq(&zone->lru_lock);
+
+ for (i = 1; i < HPAGE_PMD_NR; i++) {
+ struct page *page_tail = page + i;
+ BUG_ON(page_count(page_tail) <= 0);
+ /*
+ * Tail pages may be freed if there wasn't any mapping
+ * like if add_to_swap() is running on a lru page that
+ * had its mapping zapped. And freeing these pages
+ * requires taking the lru_lock so we do the put_page
+ * of the tail pages after the split is complete.
+ */
+ put_page(page_tail);
+ }
+
+ /*
+ * Only the head page (now become a regular page) is required
+ * to be pinned by the caller.
+ */
+ BUG_ON(page_count(page) <= 0);
+}
+
+static int __split_huge_page_map(struct page *page,
+ struct vm_area_struct *vma,
+ unsigned long address)
+{
+ struct mm_struct *mm = vma->vm_mm;
+ pmd_t *pmd, _pmd;
+ int ret = 0, i;
+ pgtable_t pgtable;
+ unsigned long haddr;
+
+ spin_lock(&mm->page_table_lock);
+ pmd = page_check_address_pmd(page, mm, address,
+ PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG);
+ if (pmd) {
+ pgtable = get_pmd_huge_pte(mm);
+ pmd_populate(mm, &_pmd, pgtable);
+
+ for (i = 0, haddr = address; i < HPAGE_PMD_NR;
+ i++, haddr += PAGE_SIZE) {
+ pte_t *pte, entry;
+ BUG_ON(PageCompound(page+i));
+ entry = mk_pte(page + i, vma->vm_page_prot);
+ entry = maybe_mkwrite(pte_mkdirty(entry), vma);
+ if (!pmd_write(*pmd))
+ entry = pte_wrprotect(entry);
+ else
+ BUG_ON(page_mapcount(page) != 1);
+ if (!pmd_young(*pmd))
+ entry = pte_mkold(entry);
+ pte = pte_offset_map(&_pmd, haddr);
+ BUG_ON(!pte_none(*pte));
+ set_pte_at(mm, haddr, pte, entry);
+ pte_unmap(pte);
+ }
+
+ mm->nr_ptes++;
+ smp_wmb(); /* make pte visible before pmd */
+ /*
+ * Up to this point the pmd is present and huge and
+ * userland has the whole access to the hugepage
+ * during the split (which happens in place). If we
+ * overwrite the pmd with the not-huge version
+ * pointing to the pte here (which of course we could
+ * if all CPUs were bug free), userland could trigger
+ * a small page size TLB miss on the small sized TLB
+ * while the hugepage TLB entry is still established
+ * in the huge TLB. Some CPU doesn't like that. See
+ * http://support.amd.com/us/Processor_TechDocs/41322.pdf,
+ * Erratum 383 on page 93. Intel should be safe but is
+ * also warns that it's only safe if the permission
+ * and cache attributes of the two entries loaded in
+ * the two TLB is identical (which should be the case
+ * here). But it is generally safer to never allow
+ * small and huge TLB entries for the same virtual
+ * address to be loaded simultaneously. So instead of
+ * doing "pmd_populate(); flush_tlb_range();" we first
+ * mark the current pmd notpresent (atomically because
+ * here the pmd_trans_huge and pmd_trans_splitting
+ * must remain set at all times on the pmd until the
+ * split is complete for this pmd), then we flush the
+ * SMP TLB and finally we write the non-huge version
+ * of the pmd entry with pmd_populate.
+ */
+ set_pmd_at(mm, address, pmd, pmd_mknotpresent(*pmd));
+ flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+ pmd_populate(mm, pmd, pgtable);
+ ret = 1;
+ }
+ spin_unlock(&mm->page_table_lock);
+
+ return ret;
+}
+
+/* must be called with anon_vma->root->lock hold */
+static void __split_huge_page(struct page *page,
+ struct anon_vma *anon_vma)
+{
+ int mapcount, mapcount2;
+ struct anon_vma_chain *avc;
+
+ BUG_ON(!PageHead(page));
+ BUG_ON(PageTail(page));
+
+ mapcount = 0;
+ list_for_each_entry(avc, &anon_vma->head, same_anon_vma) {
+ struct vm_area_struct *vma = avc->vma;
+ unsigned long addr = vma_address(page, vma);
+ BUG_ON(is_vma_temporary_stack(vma));
+ if (addr == -EFAULT)
+ continue;
+ mapcount += __split_huge_page_splitting(page, vma, addr);
+ }
+ /*
+ * It is critical that new vmas are added to the tail of the
+ * anon_vma list. This guarantes that if copy_huge_pmd() runs
+ * and establishes a child pmd before
+ * __split_huge_page_splitting() freezes the parent pmd (so if
+ * we fail to prevent copy_huge_pmd() from running until the
+ * whole __split_huge_page() is complete), we will still see
+ * the newly established pmd of the child later during the
+ * walk, to be able to set it as pmd_trans_splitting too.
+ */
+ if (mapcount != page_mapcount(page))
+ printk(KERN_ERR "mapcount %d page_mapcount %d\n",
+ mapcount, page_mapcount(page));
+ BUG_ON(mapcount != page_mapcount(page));
+
+ __split_huge_page_refcount(page);
+
+ mapcount2 = 0;
+ list_for_each_entry(avc, &anon_vma->head, same_anon_vma) {
+ struct vm_area_struct *vma = avc->vma;
+ unsigned long addr = vma_address(page, vma);
+ BUG_ON(is_vma_temporary_stack(vma));
+ if (addr == -EFAULT)
+ continue;
+ mapcount2 += __split_huge_page_map(page, vma, addr);
+ }
+ if (mapcount != mapcount2)
+ printk(KERN_ERR "mapcount %d mapcount2 %d page_mapcount %d\n",
+ mapcount, mapcount2, page_mapcount(page));
+ BUG_ON(mapcount != mapcount2);
+}
+
+int split_huge_page(struct page *page)
+{
+ struct anon_vma *anon_vma;
+ int ret = 1;
+
+ BUG_ON(!PageAnon(page));
+ anon_vma = page_lock_anon_vma(page);
+ if (!anon_vma)
+ goto out;
+ ret = 0;
+ if (!PageCompound(page))
+ goto out_unlock;
+
+ BUG_ON(!PageSwapBacked(page));
+ __split_huge_page(page, anon_vma);
+
+ BUG_ON(PageCompound(page));
+out_unlock:
+ page_unlock_anon_vma(anon_vma);
+out:
+ return ret;
+}
+
+int hugepage_madvise(struct vm_area_struct *vma,
+ unsigned long *vm_flags, int advice)
+{
+ switch (advice) {
+ case MADV_HUGEPAGE:
+ /*
+ * Be somewhat over-protective like KSM for now!
+ */
+ if (*vm_flags & (VM_HUGEPAGE |
+ VM_SHARED | VM_MAYSHARE |
+ VM_PFNMAP | VM_IO | VM_DONTEXPAND |
+ VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE |
+ VM_MIXEDMAP | VM_SAO))
+ return -EINVAL;
+ *vm_flags &= ~VM_NOHUGEPAGE;
+ *vm_flags |= VM_HUGEPAGE;
+ /*
+ * If the vma become good for khugepaged to scan,
+ * register it here without waiting a page fault that
+ * may not happen any time soon.
+ */
+ if (unlikely(khugepaged_enter_vma_merge(vma)))
+ return -ENOMEM;
+ break;
+ case MADV_NOHUGEPAGE:
+ /*
+ * Be somewhat over-protective like KSM for now!
+ */
+ if (*vm_flags & (VM_NOHUGEPAGE |
+ VM_SHARED | VM_MAYSHARE |
+ VM_PFNMAP | VM_IO | VM_DONTEXPAND |
+ VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE |
+ VM_MIXEDMAP | VM_SAO))
+ return -EINVAL;
+ *vm_flags &= ~VM_HUGEPAGE;
+ *vm_flags |= VM_NOHUGEPAGE;
+ /*
+ * Setting VM_NOHUGEPAGE will prevent khugepaged from scanning
+ * this vma even if we leave the mm registered in khugepaged if
+ * it got registered before VM_NOHUGEPAGE was set.
+ */
+ break;
+ }
+
+ return 0;
+}
+
+static int __init khugepaged_slab_init(void)
+{
+ mm_slot_cache = kmem_cache_create("khugepaged_mm_slot",
+ sizeof(struct mm_slot),
+ __alignof__(struct mm_slot), 0, NULL);
+ if (!mm_slot_cache)
+ return -ENOMEM;
+
+ return 0;
+}
+
+static void __init khugepaged_slab_free(void)
+{
+ kmem_cache_destroy(mm_slot_cache);
+ mm_slot_cache = NULL;
+}
+
+static inline struct mm_slot *alloc_mm_slot(void)
+{
+ if (!mm_slot_cache) /* initialization failed */
+ return NULL;
+ return kmem_cache_zalloc(mm_slot_cache, GFP_KERNEL);
+}
+
+static inline void free_mm_slot(struct mm_slot *mm_slot)
+{
+ kmem_cache_free(mm_slot_cache, mm_slot);
+}
+
+static int __init mm_slots_hash_init(void)
+{
+ mm_slots_hash = kzalloc(MM_SLOTS_HASH_HEADS * sizeof(struct hlist_head),
+ GFP_KERNEL);
+ if (!mm_slots_hash)
+ return -ENOMEM;
+ return 0;
+}
+
+#if 0
+static void __init mm_slots_hash_free(void)
+{
+ kfree(mm_slots_hash);
+ mm_slots_hash = NULL;
+}
+#endif
+
+static struct mm_slot *get_mm_slot(struct mm_struct *mm)
+{
+ struct mm_slot *mm_slot;
+ struct hlist_head *bucket;
+ struct hlist_node *node;
+
+ bucket = &mm_slots_hash[((unsigned long)mm / sizeof(struct mm_struct))
+ % MM_SLOTS_HASH_HEADS];
+ hlist_for_each_entry(mm_slot, node, bucket, hash) {
+ if (mm == mm_slot->mm)
+ return mm_slot;
+ }
+ return NULL;
+}
+
+static void insert_to_mm_slots_hash(struct mm_struct *mm,
+ struct mm_slot *mm_slot)
+{
+ struct hlist_head *bucket;
+
+ bucket = &mm_slots_hash[((unsigned long)mm / sizeof(struct mm_struct))
+ % MM_SLOTS_HASH_HEADS];
+ mm_slot->mm = mm;
+ hlist_add_head(&mm_slot->hash, bucket);
+}
+
+static inline int khugepaged_test_exit(struct mm_struct *mm)
+{
+ return atomic_read(&mm->mm_users) == 0;
+}
+
+int __khugepaged_enter(struct mm_struct *mm)
+{
+ struct mm_slot *mm_slot;
+ int wakeup;
+
+ mm_slot = alloc_mm_slot();
+ if (!mm_slot)
+ return -ENOMEM;
+
+ /* __khugepaged_exit() must not run from under us */
+ VM_BUG_ON(khugepaged_test_exit(mm));
+ if (unlikely(test_and_set_bit(MMF_VM_HUGEPAGE, &mm->flags))) {
+ free_mm_slot(mm_slot);
+ return 0;
+ }
+
+ spin_lock(&khugepaged_mm_lock);
+ insert_to_mm_slots_hash(mm, mm_slot);
+ /*
+ * Insert just behind the scanning cursor, to let the area settle
+ * down a little.
+ */
+ wakeup = list_empty(&khugepaged_scan.mm_head);
+ list_add_tail(&mm_slot->mm_node, &khugepaged_scan.mm_head);
+ spin_unlock(&khugepaged_mm_lock);
+
+ atomic_inc(&mm->mm_count);
+ if (wakeup)
+ wake_up_interruptible(&khugepaged_wait);
+
+ return 0;
+}
+
+int khugepaged_enter_vma_merge(struct vm_area_struct *vma)
+{
+ unsigned long hstart, hend;
+ if (!vma->anon_vma)
+ /*
+ * Not yet faulted in so we will register later in the
+ * page fault if needed.
+ */
+ return 0;
+ if (vma->vm_file || vma->vm_ops)
+ /* khugepaged not yet working on file or special mappings */
+ return 0;
+ VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma));
+ hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK;
+ hend = vma->vm_end & HPAGE_PMD_MASK;
+ if (hstart < hend)
+ return khugepaged_enter(vma);
+ return 0;
+}
+
+void __khugepaged_exit(struct mm_struct *mm)
+{
+ struct mm_slot *mm_slot;
+ int free = 0;
+
+ spin_lock(&khugepaged_mm_lock);
+ mm_slot = get_mm_slot(mm);
+ if (mm_slot && khugepaged_scan.mm_slot != mm_slot) {
+ hlist_del(&mm_slot->hash);
+ list_del(&mm_slot->mm_node);
+ free = 1;
+ }
+
+ if (free) {
+ spin_unlock(&khugepaged_mm_lock);
+ clear_bit(MMF_VM_HUGEPAGE, &mm->flags);
+ free_mm_slot(mm_slot);
+ mmdrop(mm);
+ } else if (mm_slot) {
+ spin_unlock(&khugepaged_mm_lock);
+ /*
+ * This is required to serialize against
+ * khugepaged_test_exit() (which is guaranteed to run
+ * under mmap sem read mode). Stop here (after we
+ * return all pagetables will be destroyed) until
+ * khugepaged has finished working on the pagetables
+ * under the mmap_sem.
+ */
+ down_write(&mm->mmap_sem);
+ up_write(&mm->mmap_sem);
+ } else
+ spin_unlock(&khugepaged_mm_lock);
+}
+
+static void release_pte_page(struct page *page)
+{
+ /* 0 stands for page_is_file_cache(page) == false */
+ dec_zone_page_state(page, NR_ISOLATED_ANON + 0);
+ unlock_page(page);
+ putback_lru_page(page);
+}
+
+static void release_pte_pages(pte_t *pte, pte_t *_pte)
+{
+ while (--_pte >= pte) {
+ pte_t pteval = *_pte;
+ if (!pte_none(pteval))
+ release_pte_page(pte_page(pteval));
+ }
+}
+
+static void release_all_pte_pages(pte_t *pte)
+{
+ release_pte_pages(pte, pte + HPAGE_PMD_NR);
+}
+
+static int __collapse_huge_page_isolate(struct vm_area_struct *vma,
+ unsigned long address,
+ pte_t *pte)
+{
+ struct page *page;
+ pte_t *_pte;
+ int referenced = 0, isolated = 0, none = 0;
+ for (_pte = pte; _pte < pte+HPAGE_PMD_NR;
+ _pte++, address += PAGE_SIZE) {
+ pte_t pteval = *_pte;
+ if (pte_none(pteval)) {
+ if (++none <= khugepaged_max_ptes_none)
+ continue;
+ else {
+ release_pte_pages(pte, _pte);
+ goto out;
+ }
+ }
+ if (!pte_present(pteval) || !pte_write(pteval)) {
+ release_pte_pages(pte, _pte);
+ goto out;
+ }
+ page = vm_normal_page(vma, address, pteval);
+ if (unlikely(!page)) {
+ release_pte_pages(pte, _pte);
+ goto out;
+ }
+ VM_BUG_ON(PageCompound(page));
+ BUG_ON(!PageAnon(page));
+ VM_BUG_ON(!PageSwapBacked(page));
+
+ /* cannot use mapcount: can't collapse if there's a gup pin */
+ if (page_count(page) != 1) {
+ release_pte_pages(pte, _pte);
+ goto out;
+ }
+ /*
+ * We can do it before isolate_lru_page because the
+ * page can't be freed from under us. NOTE: PG_lock
+ * is needed to serialize against split_huge_page
+ * when invoked from the VM.
+ */
+ if (!trylock_page(page)) {
+ release_pte_pages(pte, _pte);
+ goto out;
+ }
+ /*
+ * Isolate the page to avoid collapsing an hugepage
+ * currently in use by the VM.
+ */
+ if (isolate_lru_page(page)) {
+ unlock_page(page);
+ release_pte_pages(pte, _pte);
+ goto out;
+ }
+ /* 0 stands for page_is_file_cache(page) == false */
+ inc_zone_page_state(page, NR_ISOLATED_ANON + 0);
+ VM_BUG_ON(!PageLocked(page));
+ VM_BUG_ON(PageLRU(page));
+
+ /* If there is no mapped pte young don't collapse the page */
+ if (pte_young(pteval) || PageReferenced(page) ||
+ mmu_notifier_test_young(vma->vm_mm, address))
+ referenced = 1;
+ }
+ if (unlikely(!referenced))
+ release_all_pte_pages(pte);
+ else
+ isolated = 1;
+out:
+ return isolated;
+}
+
+static void __collapse_huge_page_copy(pte_t *pte, struct page *page,
+ struct vm_area_struct *vma,
+ unsigned long address,
+ spinlock_t *ptl)
+{
+ pte_t *_pte;
+ for (_pte = pte; _pte < pte+HPAGE_PMD_NR; _pte++) {
+ pte_t pteval = *_pte;
+ struct page *src_page;
+
+ if (pte_none(pteval)) {
+ clear_user_highpage(page, address);
+ add_mm_counter(vma->vm_mm, MM_ANONPAGES, 1);
+ } else {
+ src_page = pte_page(pteval);
+ copy_user_highpage(page, src_page, address, vma);
+ VM_BUG_ON(page_mapcount(src_page) != 1);
+ VM_BUG_ON(page_count(src_page) != 2);
+ release_pte_page(src_page);
+ /*
+ * ptl mostly unnecessary, but preempt has to
+ * be disabled to update the per-cpu stats
+ * inside page_remove_rmap().
+ */
+ spin_lock(ptl);
+ /*
+ * paravirt calls inside pte_clear here are
+ * superfluous.
+ */
+ pte_clear(vma->vm_mm, address, _pte);
+ page_remove_rmap(src_page);
+ spin_unlock(ptl);
+ free_page_and_swap_cache(src_page);
+ }
+
+ address += PAGE_SIZE;
+ page++;
+ }
+}
+
+static void collapse_huge_page(struct mm_struct *mm,
+ unsigned long address,
+ struct page **hpage,
+ struct vm_area_struct *vma)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd, _pmd;
+ pte_t *pte;
+ pgtable_t pgtable;
+ struct page *new_page;
+ spinlock_t *ptl;
+ int isolated;
+ unsigned long hstart, hend;
+
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+#ifndef CONFIG_NUMA
+ VM_BUG_ON(!*hpage);
+ new_page = *hpage;
+#else
+ VM_BUG_ON(*hpage);
+ /*
+ * Allocate the page while the vma is still valid and under
+ * the mmap_sem read mode so there is no memory allocation
+ * later when we take the mmap_sem in write mode. This is more
+ * friendly behavior (OTOH it may actually hide bugs) to
+ * filesystems in userland with daemons allocating memory in
+ * the userland I/O paths. Allocating memory with the
+ * mmap_sem in read mode is good idea also to allow greater
+ * scalability.
+ */
+ new_page = alloc_hugepage_vma(khugepaged_defrag(), vma, address);
+ if (unlikely(!new_page)) {
+ up_read(&mm->mmap_sem);
+ *hpage = ERR_PTR(-ENOMEM);
+ return;
+ }
+#endif
+ if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) {
+ up_read(&mm->mmap_sem);
+ put_page(new_page);
+ return;
+ }
+
+ /* after allocating the hugepage upgrade to mmap_sem write mode */
+ up_read(&mm->mmap_sem);
+
+ /*
+ * Prevent all access to pagetables with the exception of
+ * gup_fast later hanlded by the ptep_clear_flush and the VM
+ * handled by the anon_vma lock + PG_lock.
+ */
+ down_write(&mm->mmap_sem);
+ if (unlikely(khugepaged_test_exit(mm)))
+ goto out;
+
+ vma = find_vma(mm, address);
+ hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK;
+ hend = vma->vm_end & HPAGE_PMD_MASK;
+ if (address < hstart || address + HPAGE_PMD_SIZE > hend)
+ goto out;
+
+ if ((!(vma->vm_flags & VM_HUGEPAGE) && !khugepaged_always()) ||
+ (vma->vm_flags & VM_NOHUGEPAGE))
+ goto out;
+
+ /* VM_PFNMAP vmas may have vm_ops null but vm_file set */
+ if (!vma->anon_vma || vma->vm_ops || vma->vm_file)
+ goto out;
+ VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma));
+
+ pgd = pgd_offset(mm, address);
+ if (!pgd_present(*pgd))
+ goto out;
+
+ pud = pud_offset(pgd, address);
+ if (!pud_present(*pud))
+ goto out;
+
+ pmd = pmd_offset(pud, address);
+ /* pmd can't go away or become huge under us */
+ if (!pmd_present(*pmd) || pmd_trans_huge(*pmd))
+ goto out;
+
+ anon_vma_lock(vma->anon_vma);
+
+ pte = pte_offset_map(pmd, address);
+ ptl = pte_lockptr(mm, pmd);
+
+ spin_lock(&mm->page_table_lock); /* probably unnecessary */
+ /*
+ * After this gup_fast can't run anymore. This also removes
+ * any huge TLB entry from the CPU so we won't allow
+ * huge and small TLB entries for the same virtual address
+ * to avoid the risk of CPU bugs in that area.
+ */
+ _pmd = pmdp_clear_flush_notify(vma, address, pmd);
+ spin_unlock(&mm->page_table_lock);
+
+ spin_lock(ptl);
+ isolated = __collapse_huge_page_isolate(vma, address, pte);
+ spin_unlock(ptl);
+ pte_unmap(pte);
+
+ if (unlikely(!isolated)) {
+ spin_lock(&mm->page_table_lock);
+ BUG_ON(!pmd_none(*pmd));
+ set_pmd_at(mm, address, pmd, _pmd);
+ spin_unlock(&mm->page_table_lock);
+ anon_vma_unlock(vma->anon_vma);
+ mem_cgroup_uncharge_page(new_page);
+ goto out;
+ }
+
+ /*
+ * All pages are isolated and locked so anon_vma rmap
+ * can't run anymore.
+ */
+ anon_vma_unlock(vma->anon_vma);
+
+ __collapse_huge_page_copy(pte, new_page, vma, address, ptl);
+ __SetPageUptodate(new_page);
+ pgtable = pmd_pgtable(_pmd);
+ VM_BUG_ON(page_count(pgtable) != 1);
+ VM_BUG_ON(page_mapcount(pgtable) != 0);
+
+ _pmd = mk_pmd(new_page, vma->vm_page_prot);
+ _pmd = maybe_pmd_mkwrite(pmd_mkdirty(_pmd), vma);
+ _pmd = pmd_mkhuge(_pmd);
+
+ /*
+ * spin_lock() below is not the equivalent of smp_wmb(), so
+ * this is needed to avoid the copy_huge_page writes to become
+ * visible after the set_pmd_at() write.
+ */
+ smp_wmb();
+
+ spin_lock(&mm->page_table_lock);
+ BUG_ON(!pmd_none(*pmd));
+ page_add_new_anon_rmap(new_page, vma, address);
+ set_pmd_at(mm, address, pmd, _pmd);
+ update_mmu_cache(vma, address, entry);
+ prepare_pmd_huge_pte(pgtable, mm);
+ mm->nr_ptes--;
+ spin_unlock(&mm->page_table_lock);
+
+#ifndef CONFIG_NUMA
+ *hpage = NULL;
+#endif
+ khugepaged_pages_collapsed++;
+out_up_write:
+ up_write(&mm->mmap_sem);
+ return;
+
+out:
+#ifdef CONFIG_NUMA
+ put_page(new_page);
+#endif
+ goto out_up_write;
+}
+
+static int khugepaged_scan_pmd(struct mm_struct *mm,
+ struct vm_area_struct *vma,
+ unsigned long address,
+ struct page **hpage)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte, *_pte;
+ int ret = 0, referenced = 0, none = 0;
+ struct page *page;
+ unsigned long _address;
+ spinlock_t *ptl;
+
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+
+ pgd = pgd_offset(mm, address);
+ if (!pgd_present(*pgd))
+ goto out;
+
+ pud = pud_offset(pgd, address);
+ if (!pud_present(*pud))
+ goto out;
+
+ pmd = pmd_offset(pud, address);
+ if (!pmd_present(*pmd) || pmd_trans_huge(*pmd))
+ goto out;
+
+ pte = pte_offset_map_lock(mm, pmd, address, &ptl);
+ for (_address = address, _pte = pte; _pte < pte+HPAGE_PMD_NR;
+ _pte++, _address += PAGE_SIZE) {
+ pte_t pteval = *_pte;
+ if (pte_none(pteval)) {
+ if (++none <= khugepaged_max_ptes_none)
+ continue;
+ else
+ goto out_unmap;
+ }
+ if (!pte_present(pteval) || !pte_write(pteval))
+ goto out_unmap;
+ page = vm_normal_page(vma, _address, pteval);
+ if (unlikely(!page))
+ goto out_unmap;
+ VM_BUG_ON(PageCompound(page));
+ if (!PageLRU(page) || PageLocked(page) || !PageAnon(page))
+ goto out_unmap;
+ /* cannot use mapcount: can't collapse if there's a gup pin */
+ if (page_count(page) != 1)
+ goto out_unmap;
+ if (pte_young(pteval) || PageReferenced(page) ||
+ mmu_notifier_test_young(vma->vm_mm, address))
+ referenced = 1;
+ }
+ if (referenced)
+ ret = 1;
+out_unmap:
+ pte_unmap_unlock(pte, ptl);
+ if (ret)
+ /* collapse_huge_page will return with the mmap_sem released */
+ collapse_huge_page(mm, address, hpage, vma);
+out:
+ return ret;
+}
+
+static void collect_mm_slot(struct mm_slot *mm_slot)
+{
+ struct mm_struct *mm = mm_slot->mm;
+
+ VM_BUG_ON(!spin_is_locked(&khugepaged_mm_lock));
+
+ if (khugepaged_test_exit(mm)) {
+ /* free mm_slot */
+ hlist_del(&mm_slot->hash);
+ list_del(&mm_slot->mm_node);
+
+ /*
+ * Not strictly needed because the mm exited already.
+ *
+ * clear_bit(MMF_VM_HUGEPAGE, &mm->flags);
+ */
+
+ /* khugepaged_mm_lock actually not necessary for the below */
+ free_mm_slot(mm_slot);
+ mmdrop(mm);
+ }
+}
+
+static unsigned int khugepaged_scan_mm_slot(unsigned int pages,
+ struct page **hpage)
+{
+ struct mm_slot *mm_slot;
+ struct mm_struct *mm;
+ struct vm_area_struct *vma;
+ int progress = 0;
+
+ VM_BUG_ON(!pages);
+ VM_BUG_ON(!spin_is_locked(&khugepaged_mm_lock));
+
+ if (khugepaged_scan.mm_slot)
+ mm_slot = khugepaged_scan.mm_slot;
+ else {
+ mm_slot = list_entry(khugepaged_scan.mm_head.next,
+ struct mm_slot, mm_node);
+ khugepaged_scan.address = 0;
+ khugepaged_scan.mm_slot = mm_slot;
+ }
+ spin_unlock(&khugepaged_mm_lock);
+
+ mm = mm_slot->mm;
+ down_read(&mm->mmap_sem);
+ if (unlikely(khugepaged_test_exit(mm)))
+ vma = NULL;
+ else
+ vma = find_vma(mm, khugepaged_scan.address);
+
+ progress++;
+ for (; vma; vma = vma->vm_next) {
+ unsigned long hstart, hend;
+
+ cond_resched();
+ if (unlikely(khugepaged_test_exit(mm))) {
+ progress++;
+ break;
+ }
+
+ if ((!(vma->vm_flags & VM_HUGEPAGE) &&
+ !khugepaged_always()) ||
+ (vma->vm_flags & VM_NOHUGEPAGE)) {
+ progress++;
+ continue;
+ }
+
+ /* VM_PFNMAP vmas may have vm_ops null but vm_file set */
+ if (!vma->anon_vma || vma->vm_ops || vma->vm_file) {
+ khugepaged_scan.address = vma->vm_end;
+ progress++;
+ continue;
+ }
+ VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma));
+
+ hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK;
+ hend = vma->vm_end & HPAGE_PMD_MASK;
+ if (hstart >= hend) {
+ progress++;
+ continue;
+ }
+ if (khugepaged_scan.address < hstart)
+ khugepaged_scan.address = hstart;
+ if (khugepaged_scan.address > hend) {
+ khugepaged_scan.address = hend + HPAGE_PMD_SIZE;
+ progress++;
+ continue;
+ }
+ BUG_ON(khugepaged_scan.address & ~HPAGE_PMD_MASK);
+
+ while (khugepaged_scan.address < hend) {
+ int ret;
+ cond_resched();
+ if (unlikely(khugepaged_test_exit(mm)))
+ goto breakouterloop;
+
+ VM_BUG_ON(khugepaged_scan.address < hstart ||
+ khugepaged_scan.address + HPAGE_PMD_SIZE >
+ hend);
+ ret = khugepaged_scan_pmd(mm, vma,
+ khugepaged_scan.address,
+ hpage);
+ /* move to next address */
+ khugepaged_scan.address += HPAGE_PMD_SIZE;
+ progress += HPAGE_PMD_NR;
+ if (ret)
+ /* we released mmap_sem so break loop */
+ goto breakouterloop_mmap_sem;
+ if (progress >= pages)
+ goto breakouterloop;
+ }
+ }
+breakouterloop:
+ up_read(&mm->mmap_sem); /* exit_mmap will destroy ptes after this */
+breakouterloop_mmap_sem:
+
+ spin_lock(&khugepaged_mm_lock);
+ BUG_ON(khugepaged_scan.mm_slot != mm_slot);
+ /*
+ * Release the current mm_slot if this mm is about to die, or
+ * if we scanned all vmas of this mm.
+ */
+ if (khugepaged_test_exit(mm) || !vma) {
+ /*
+ * Make sure that if mm_users is reaching zero while
+ * khugepaged runs here, khugepaged_exit will find
+ * mm_slot not pointing to the exiting mm.
+ */
+ if (mm_slot->mm_node.next != &khugepaged_scan.mm_head) {
+ khugepaged_scan.mm_slot = list_entry(
+ mm_slot->mm_node.next,
+ struct mm_slot, mm_node);
+ khugepaged_scan.address = 0;
+ } else {
+ khugepaged_scan.mm_slot = NULL;
+ khugepaged_full_scans++;
+ }
+
+ collect_mm_slot(mm_slot);
+ }
+
+ return progress;
+}
+
+static int khugepaged_has_work(void)
+{
+ return !list_empty(&khugepaged_scan.mm_head) &&
+ khugepaged_enabled();
+}
+
+static int khugepaged_wait_event(void)
+{
+ return !list_empty(&khugepaged_scan.mm_head) ||
+ !khugepaged_enabled();
+}
+
+static void khugepaged_do_scan(struct page **hpage)
+{
+ unsigned int progress = 0, pass_through_head = 0;
+ unsigned int pages = khugepaged_pages_to_scan;
+
+ barrier(); /* write khugepaged_pages_to_scan to local stack */
+
+ while (progress < pages) {
+ cond_resched();
+
+#ifndef CONFIG_NUMA
+ if (!*hpage) {
+ *hpage = alloc_hugepage(khugepaged_defrag());
+ if (unlikely(!*hpage))
+ break;
+ }
+#else
+ if (IS_ERR(*hpage))
+ break;
+#endif
+
+ if (unlikely(kthread_should_stop() || freezing(current)))
+ break;
+
+ spin_lock(&khugepaged_mm_lock);
+ if (!khugepaged_scan.mm_slot)
+ pass_through_head++;
+ if (khugepaged_has_work() &&
+ pass_through_head < 2)
+ progress += khugepaged_scan_mm_slot(pages - progress,
+ hpage);
+ else
+ progress = pages;
+ spin_unlock(&khugepaged_mm_lock);
+ }
+}
+
+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);
+}
+
+#ifndef CONFIG_NUMA
+static struct page *khugepaged_alloc_hugepage(void)
+{
+ struct page *hpage;
+
+ do {
+ hpage = alloc_hugepage(khugepaged_defrag());
+ if (!hpage)
+ khugepaged_alloc_sleep();
+ } while (unlikely(!hpage) &&
+ likely(khugepaged_enabled()));
+ return hpage;
+}
+#endif
+
+static void khugepaged_loop(void)
+{
+ struct page *hpage;
+
+#ifdef CONFIG_NUMA
+ hpage = NULL;
+#endif
+ while (likely(khugepaged_enabled())) {
+#ifndef CONFIG_NUMA
+ hpage = khugepaged_alloc_hugepage();
+ if (unlikely(!hpage))
+ break;
+#else
+ if (IS_ERR(hpage)) {
+ khugepaged_alloc_sleep();
+ hpage = NULL;
+ }
+#endif
+
+ khugepaged_do_scan(&hpage);
+#ifndef CONFIG_NUMA
+ if (hpage)
+ put_page(hpage);
+#endif
+ try_to_freeze();
+ 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);
+ } else if (khugepaged_enabled())
+ wait_event_freezable(khugepaged_wait,
+ khugepaged_wait_event());
+ }
+}
+
+static int khugepaged(void *none)
+{
+ struct mm_slot *mm_slot;
+
+ set_freezable();
+ set_user_nice(current, 19);
+
+ /* serialize with start_khugepaged() */
+ mutex_lock(&khugepaged_mutex);
+
+ for (;;) {
+ mutex_unlock(&khugepaged_mutex);
+ BUG_ON(khugepaged_thread != current);
+ khugepaged_loop();
+ BUG_ON(khugepaged_thread != current);
+
+ mutex_lock(&khugepaged_mutex);
+ if (!khugepaged_enabled())
+ break;
+ if (unlikely(kthread_should_stop()))
+ break;
+ }
+
+ spin_lock(&khugepaged_mm_lock);
+ mm_slot = khugepaged_scan.mm_slot;
+ khugepaged_scan.mm_slot = NULL;
+ if (mm_slot)
+ collect_mm_slot(mm_slot);
+ spin_unlock(&khugepaged_mm_lock);
+
+ khugepaged_thread = NULL;
+ mutex_unlock(&khugepaged_mutex);
+
+ return 0;
+}
+
+void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd)
+{
+ struct page *page;
+
+ spin_lock(&mm->page_table_lock);
+ if (unlikely(!pmd_trans_huge(*pmd))) {
+ spin_unlock(&mm->page_table_lock);
+ return;
+ }
+ page = pmd_page(*pmd);
+ VM_BUG_ON(!page_count(page));
+ get_page(page);
+ spin_unlock(&mm->page_table_lock);
+
+ split_huge_page(page);
+
+ put_page(page);
+ BUG_ON(pmd_trans_huge(*pmd));
+}
+
+static void split_huge_page_address(struct mm_struct *mm,
+ unsigned long address)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+
+ VM_BUG_ON(!(address & ~HPAGE_PMD_MASK));
+
+ pgd = pgd_offset(mm, address);
+ if (!pgd_present(*pgd))
+ return;
+
+ pud = pud_offset(pgd, address);
+ if (!pud_present(*pud))
+ return;
+
+ pmd = pmd_offset(pud, address);
+ if (!pmd_present(*pmd))
+ return;
+ /*
+ * Caller holds the mmap_sem write mode, so a huge pmd cannot
+ * materialize from under us.
+ */
+ split_huge_page_pmd(mm, pmd);
+}
+
+void __vma_adjust_trans_huge(struct vm_area_struct *vma,
+ unsigned long start,
+ unsigned long end,
+ long adjust_next)
+{
+ /*
+ * If the new start address isn't hpage aligned and it could
+ * previously contain an hugepage: check if we need to split
+ * an huge pmd.
+ */
+ if (start & ~HPAGE_PMD_MASK &&
+ (start & HPAGE_PMD_MASK) >= vma->vm_start &&
+ (start & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE <= vma->vm_end)
+ split_huge_page_address(vma->vm_mm, start);
+
+ /*
+ * If the new end address isn't hpage aligned and it could
+ * previously contain an hugepage: check if we need to split
+ * an huge pmd.
+ */
+ if (end & ~HPAGE_PMD_MASK &&
+ (end & HPAGE_PMD_MASK) >= vma->vm_start &&
+ (end & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE <= vma->vm_end)
+ split_huge_page_address(vma->vm_mm, end);
+
+ /*
+ * If we're also updating the vma->vm_next->vm_start, if the new
+ * vm_next->vm_start isn't page aligned and it could previously
+ * contain an hugepage: check if we need to split an huge pmd.
+ */
+ if (adjust_next > 0) {
+ struct vm_area_struct *next = vma->vm_next;
+ unsigned long nstart = next->vm_start;
+ nstart += adjust_next << PAGE_SHIFT;
+ if (nstart & ~HPAGE_PMD_MASK &&
+ (nstart & HPAGE_PMD_MASK) >= next->vm_start &&
+ (nstart & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE <= next->vm_end)
+ split_huge_page_address(next->vm_mm, nstart);
+ }
+}
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 8585524..bb0b7c1 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -394,71 +394,6 @@
return 0;
}
-static void clear_gigantic_page(struct page *page,
- unsigned long addr, unsigned long sz)
-{
- int i;
- struct page *p = page;
-
- might_sleep();
- for (i = 0; i < sz/PAGE_SIZE; i++, p = mem_map_next(p, page, i)) {
- cond_resched();
- clear_user_highpage(p, addr + i * PAGE_SIZE);
- }
-}
-static void clear_huge_page(struct page *page,
- unsigned long addr, unsigned long sz)
-{
- int i;
-
- if (unlikely(sz/PAGE_SIZE > MAX_ORDER_NR_PAGES)) {
- clear_gigantic_page(page, addr, sz);
- return;
- }
-
- might_sleep();
- for (i = 0; i < sz/PAGE_SIZE; i++) {
- cond_resched();
- clear_user_highpage(page + i, addr + i * PAGE_SIZE);
- }
-}
-
-static void copy_user_gigantic_page(struct page *dst, struct page *src,
- unsigned long addr, struct vm_area_struct *vma)
-{
- int i;
- struct hstate *h = hstate_vma(vma);
- struct page *dst_base = dst;
- struct page *src_base = src;
-
- for (i = 0; i < pages_per_huge_page(h); ) {
- cond_resched();
- copy_user_highpage(dst, src, addr + i*PAGE_SIZE, vma);
-
- i++;
- dst = mem_map_next(dst, dst_base, i);
- src = mem_map_next(src, src_base, i);
- }
-}
-
-static void copy_user_huge_page(struct page *dst, struct page *src,
- unsigned long addr, struct vm_area_struct *vma)
-{
- int i;
- struct hstate *h = hstate_vma(vma);
-
- if (unlikely(pages_per_huge_page(h) > MAX_ORDER_NR_PAGES)) {
- copy_user_gigantic_page(dst, src, addr, vma);
- return;
- }
-
- might_sleep();
- for (i = 0; i < pages_per_huge_page(h); i++) {
- cond_resched();
- copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma);
- }
-}
-
static void copy_gigantic_page(struct page *dst, struct page *src)
{
int i;
@@ -1428,6 +1363,7 @@
return sprintf(buf, "%lu\n", nr_huge_pages);
}
+
static ssize_t nr_hugepages_store_common(bool obey_mempolicy,
struct kobject *kobj, struct kobj_attribute *attr,
const char *buf, size_t len)
@@ -1440,9 +1376,14 @@
err = strict_strtoul(buf, 10, &count);
if (err)
- return 0;
+ goto out;
h = kobj_to_hstate(kobj, &nid);
+ if (h->order >= MAX_ORDER) {
+ err = -EINVAL;
+ goto out;
+ }
+
if (nid == NUMA_NO_NODE) {
/*
* global hstate attribute
@@ -1468,6 +1409,9 @@
NODEMASK_FREE(nodes_allowed);
return len;
+out:
+ NODEMASK_FREE(nodes_allowed);
+ return err;
}
static ssize_t nr_hugepages_show(struct kobject *kobj,
@@ -1510,6 +1454,7 @@
struct hstate *h = kobj_to_hstate(kobj, NULL);
return sprintf(buf, "%lu\n", h->nr_overcommit_huge_pages);
}
+
static ssize_t nr_overcommit_hugepages_store(struct kobject *kobj,
struct kobj_attribute *attr, const char *buf, size_t count)
{
@@ -1517,9 +1462,12 @@
unsigned long input;
struct hstate *h = kobj_to_hstate(kobj, NULL);
+ if (h->order >= MAX_ORDER)
+ return -EINVAL;
+
err = strict_strtoul(buf, 10, &input);
if (err)
- return 0;
+ return err;
spin_lock(&hugetlb_lock);
h->nr_overcommit_huge_pages = input;
@@ -1922,13 +1870,19 @@
{
struct hstate *h = &default_hstate;
unsigned long tmp;
+ int ret;
if (!write)
tmp = h->max_huge_pages;
+ if (write && h->order >= MAX_ORDER)
+ return -EINVAL;
+
table->data = &tmp;
table->maxlen = sizeof(unsigned long);
- proc_doulongvec_minmax(table, write, buffer, length, ppos);
+ ret = proc_doulongvec_minmax(table, write, buffer, length, ppos);
+ if (ret)
+ goto out;
if (write) {
NODEMASK_ALLOC(nodemask_t, nodes_allowed,
@@ -1943,8 +1897,8 @@
if (nodes_allowed != &node_states[N_HIGH_MEMORY])
NODEMASK_FREE(nodes_allowed);
}
-
- return 0;
+out:
+ return ret;
}
int hugetlb_sysctl_handler(struct ctl_table *table, int write,
@@ -1982,21 +1936,27 @@
{
struct hstate *h = &default_hstate;
unsigned long tmp;
+ int ret;
if (!write)
tmp = h->nr_overcommit_huge_pages;
+ if (write && h->order >= MAX_ORDER)
+ return -EINVAL;
+
table->data = &tmp;
table->maxlen = sizeof(unsigned long);
- proc_doulongvec_minmax(table, write, buffer, length, ppos);
+ ret = proc_doulongvec_minmax(table, write, buffer, length, ppos);
+ if (ret)
+ goto out;
if (write) {
spin_lock(&hugetlb_lock);
h->nr_overcommit_huge_pages = tmp;
spin_unlock(&hugetlb_lock);
}
-
- return 0;
+out:
+ return ret;
}
#endif /* CONFIG_SYSCTL */
@@ -2454,7 +2414,8 @@
return VM_FAULT_OOM;
}
- copy_user_huge_page(new_page, old_page, address, vma);
+ copy_user_huge_page(new_page, old_page, address, vma,
+ pages_per_huge_page(h));
__SetPageUptodate(new_page);
/*
@@ -2558,7 +2519,7 @@
ret = -PTR_ERR(page);
goto out;
}
- clear_huge_page(page, address, huge_page_size(h));
+ clear_huge_page(page, address, pages_per_huge_page(h));
__SetPageUptodate(page);
if (vma->vm_flags & VM_MAYSHARE) {
diff --git a/mm/internal.h b/mm/internal.h
index dedb0af..4c98630 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -39,6 +39,15 @@
extern unsigned long highest_memmap_pfn;
+#ifdef CONFIG_SMP
+extern int putback_active_lru_page(struct zone *zone, struct page *page);
+#else
+static inline int putback_active_lru_page(struct zone *zone, struct page *page)
+{
+ return 0;
+}
+#endif
+
/*
* in mm/vmscan.c:
*/
@@ -134,6 +143,10 @@
}
}
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+extern unsigned long vma_address(struct page *page,
+ struct vm_area_struct *vma);
+#endif
#else /* !CONFIG_MMU */
static inline int is_mlocked_vma(struct vm_area_struct *v, struct page *p)
{
@@ -243,7 +256,8 @@
int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, int len, unsigned int foll_flags,
- struct page **pages, struct vm_area_struct **vmas);
+ struct page **pages, struct vm_area_struct **vmas,
+ int *nonblocking);
#define ZONE_RECLAIM_NOSCAN -2
#define ZONE_RECLAIM_FULL -1
diff --git a/mm/ksm.c b/mm/ksm.c
index 43bc893..c2b2a94 100644
--- a/mm/ksm.c
+++ b/mm/ksm.c
@@ -34,6 +34,7 @@
#include <linux/swap.h>
#include <linux/ksm.h>
#include <linux/hash.h>
+#include <linux/freezer.h>
#include <asm/tlbflush.h>
#include "internal.h"
@@ -411,6 +412,20 @@
up_read(&mm->mmap_sem);
}
+static struct page *page_trans_compound_anon(struct page *page)
+{
+ if (PageTransCompound(page)) {
+ struct page *head = compound_trans_head(page);
+ /*
+ * head may actually be splitted and freed from under
+ * us but it's ok here.
+ */
+ if (PageAnon(head))
+ return head;
+ }
+ return NULL;
+}
+
static struct page *get_mergeable_page(struct rmap_item *rmap_item)
{
struct mm_struct *mm = rmap_item->mm;
@@ -430,7 +445,7 @@
page = follow_page(vma, addr, FOLL_GET);
if (IS_ERR_OR_NULL(page))
goto out;
- if (PageAnon(page)) {
+ if (PageAnon(page) || page_trans_compound_anon(page)) {
flush_anon_page(vma, page, addr);
flush_dcache_page(page);
} else {
@@ -708,6 +723,7 @@
if (addr == -EFAULT)
goto out;
+ BUG_ON(PageTransCompound(page));
ptep = page_check_address(page, mm, addr, &ptl, 0);
if (!ptep)
goto out;
@@ -783,6 +799,7 @@
goto out;
pmd = pmd_offset(pud, addr);
+ BUG_ON(pmd_trans_huge(*pmd));
if (!pmd_present(*pmd))
goto out;
@@ -800,6 +817,8 @@
set_pte_at_notify(mm, addr, ptep, mk_pte(kpage, vma->vm_page_prot));
page_remove_rmap(page);
+ if (!page_mapped(page))
+ try_to_free_swap(page);
put_page(page);
pte_unmap_unlock(ptep, ptl);
@@ -808,6 +827,33 @@
return err;
}
+static int page_trans_compound_anon_split(struct page *page)
+{
+ int ret = 0;
+ struct page *transhuge_head = page_trans_compound_anon(page);
+ if (transhuge_head) {
+ /* Get the reference on the head to split it. */
+ if (get_page_unless_zero(transhuge_head)) {
+ /*
+ * Recheck we got the reference while the head
+ * was still anonymous.
+ */
+ if (PageAnon(transhuge_head))
+ ret = split_huge_page(transhuge_head);
+ else
+ /*
+ * Retry later if split_huge_page run
+ * from under us.
+ */
+ ret = 1;
+ put_page(transhuge_head);
+ } else
+ /* Retry later if split_huge_page run from under us. */
+ ret = 1;
+ }
+ return ret;
+}
+
/*
* try_to_merge_one_page - take two pages and merge them into one
* @vma: the vma that holds the pte pointing to page
@@ -828,6 +874,9 @@
if (!(vma->vm_flags & VM_MERGEABLE))
goto out;
+ if (PageTransCompound(page) && page_trans_compound_anon_split(page))
+ goto out;
+ BUG_ON(PageTransCompound(page));
if (!PageAnon(page))
goto out;
@@ -1247,6 +1296,18 @@
slot = ksm_scan.mm_slot;
if (slot == &ksm_mm_head) {
+ /*
+ * A number of pages can hang around indefinitely on per-cpu
+ * pagevecs, raised page count preventing write_protect_page
+ * from merging them. Though it doesn't really matter much,
+ * it is puzzling to see some stuck in pages_volatile until
+ * other activity jostles them out, and they also prevented
+ * LTP's KSM test from succeeding deterministically; so drain
+ * them here (here rather than on entry to ksm_do_scan(),
+ * so we don't IPI too often when pages_to_scan is set low).
+ */
+ lru_add_drain_all();
+
root_unstable_tree = RB_ROOT;
spin_lock(&ksm_mmlist_lock);
@@ -1277,7 +1338,13 @@
if (ksm_test_exit(mm))
break;
*page = follow_page(vma, ksm_scan.address, FOLL_GET);
- if (!IS_ERR_OR_NULL(*page) && PageAnon(*page)) {
+ if (IS_ERR_OR_NULL(*page)) {
+ ksm_scan.address += PAGE_SIZE;
+ cond_resched();
+ continue;
+ }
+ if (PageAnon(*page) ||
+ page_trans_compound_anon(*page)) {
flush_anon_page(vma, *page, ksm_scan.address);
flush_dcache_page(*page);
rmap_item = get_next_rmap_item(slot,
@@ -1291,8 +1358,7 @@
up_read(&mm->mmap_sem);
return rmap_item;
}
- if (!IS_ERR_OR_NULL(*page))
- put_page(*page);
+ put_page(*page);
ksm_scan.address += PAGE_SIZE;
cond_resched();
}
@@ -1352,7 +1418,7 @@
struct rmap_item *rmap_item;
struct page *uninitialized_var(page);
- while (scan_npages--) {
+ while (scan_npages-- && likely(!freezing(current))) {
cond_resched();
rmap_item = scan_get_next_rmap_item(&page);
if (!rmap_item)
@@ -1370,6 +1436,7 @@
static int ksm_scan_thread(void *nothing)
{
+ set_freezable();
set_user_nice(current, 5);
while (!kthread_should_stop()) {
@@ -1378,11 +1445,13 @@
ksm_do_scan(ksm_thread_pages_to_scan);
mutex_unlock(&ksm_thread_mutex);
+ try_to_freeze();
+
if (ksmd_should_run()) {
schedule_timeout_interruptible(
msecs_to_jiffies(ksm_thread_sleep_millisecs));
} else {
- wait_event_interruptible(ksm_thread_wait,
+ wait_event_freezable(ksm_thread_wait,
ksmd_should_run() || kthread_should_stop());
}
}
diff --git a/mm/madvise.c b/mm/madvise.c
index 319528b..2221491 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -71,6 +71,12 @@
if (error)
goto out;
break;
+ case MADV_HUGEPAGE:
+ case MADV_NOHUGEPAGE:
+ error = hugepage_madvise(vma, &new_flags, behavior);
+ if (error)
+ goto out;
+ break;
}
if (new_flags == vma->vm_flags) {
@@ -283,6 +289,10 @@
case MADV_MERGEABLE:
case MADV_UNMERGEABLE:
#endif
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ case MADV_HUGEPAGE:
+ case MADV_NOHUGEPAGE:
+#endif
return 1;
default:
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 00bb8a6..8ab8410 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -292,7 +292,6 @@
unsigned long moved_charge;
unsigned long moved_swap;
struct task_struct *moving_task; /* a task moving charges */
- struct mm_struct *mm;
wait_queue_head_t waitq; /* a waitq for other context */
} mc = {
.lock = __SPIN_LOCK_UNLOCKED(mc.lock),
@@ -821,7 +820,6 @@
return;
VM_BUG_ON(list_empty(&pc->lru));
list_del_init(&pc->lru);
- return;
}
void mem_cgroup_del_lru(struct page *page)
@@ -1087,7 +1085,7 @@
case 0:
list_move(&page->lru, dst);
mem_cgroup_del_lru(page);
- nr_taken++;
+ nr_taken += hpage_nr_pages(page);
break;
case -EBUSY:
/* we don't affect global LRU but rotate in our LRU */
@@ -1312,8 +1310,9 @@
u64 limit;
u64 memsw;
- limit = res_counter_read_u64(&memcg->res, RES_LIMIT) +
- total_swap_pages;
+ limit = res_counter_read_u64(&memcg->res, RES_LIMIT);
+ limit += total_swap_pages << PAGE_SHIFT;
+
memsw = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
/*
* If memsw is finite and limits the amount of swap space available
@@ -1600,11 +1599,13 @@
* possibility of race condition. If there is, we take a lock.
*/
-static void mem_cgroup_update_file_stat(struct page *page, int idx, int val)
+void mem_cgroup_update_page_stat(struct page *page,
+ enum mem_cgroup_page_stat_item idx, int val)
{
struct mem_cgroup *mem;
struct page_cgroup *pc = lookup_page_cgroup(page);
bool need_unlock = false;
+ unsigned long uninitialized_var(flags);
if (unlikely(!pc))
return;
@@ -1616,37 +1617,34 @@
/* pc->mem_cgroup is unstable ? */
if (unlikely(mem_cgroup_stealed(mem))) {
/* take a lock against to access pc->mem_cgroup */
- lock_page_cgroup(pc);
+ move_lock_page_cgroup(pc, &flags);
need_unlock = true;
mem = pc->mem_cgroup;
if (!mem || !PageCgroupUsed(pc))
goto out;
}
- this_cpu_add(mem->stat->count[idx], val);
-
switch (idx) {
- case MEM_CGROUP_STAT_FILE_MAPPED:
+ case MEMCG_NR_FILE_MAPPED:
if (val > 0)
SetPageCgroupFileMapped(pc);
else if (!page_mapped(page))
ClearPageCgroupFileMapped(pc);
+ idx = MEM_CGROUP_STAT_FILE_MAPPED;
break;
default:
BUG();
}
+ this_cpu_add(mem->stat->count[idx], val);
+
out:
if (unlikely(need_unlock))
- unlock_page_cgroup(pc);
+ move_unlock_page_cgroup(pc, &flags);
rcu_read_unlock();
return;
}
-
-void mem_cgroup_update_file_mapped(struct page *page, int val)
-{
- mem_cgroup_update_file_stat(page, MEM_CGROUP_STAT_FILE_MAPPED, val);
-}
+EXPORT_SYMBOL(mem_cgroup_update_page_stat);
/*
* size of first charge trial. "32" comes from vmscan.c's magic value.
@@ -1887,12 +1885,14 @@
* oom-killer can be invoked.
*/
static int __mem_cgroup_try_charge(struct mm_struct *mm,
- gfp_t gfp_mask, struct mem_cgroup **memcg, bool oom)
+ gfp_t gfp_mask,
+ struct mem_cgroup **memcg, bool oom,
+ int page_size)
{
int nr_oom_retries = MEM_CGROUP_RECLAIM_RETRIES;
struct mem_cgroup *mem = NULL;
int ret;
- int csize = CHARGE_SIZE;
+ int csize = max(CHARGE_SIZE, (unsigned long) page_size);
/*
* Unlike gloval-vm's OOM-kill, we're not in memory shortage
@@ -1917,7 +1917,7 @@
VM_BUG_ON(css_is_removed(&mem->css));
if (mem_cgroup_is_root(mem))
goto done;
- if (consume_stock(mem))
+ if (page_size == PAGE_SIZE && consume_stock(mem))
goto done;
css_get(&mem->css);
} else {
@@ -1940,7 +1940,7 @@
rcu_read_unlock();
goto done;
}
- if (consume_stock(mem)) {
+ if (page_size == PAGE_SIZE && consume_stock(mem)) {
/*
* It seems dagerous to access memcg without css_get().
* But considering how consume_stok works, it's not
@@ -1981,7 +1981,7 @@
case CHARGE_OK:
break;
case CHARGE_RETRY: /* not in OOM situation but retry */
- csize = PAGE_SIZE;
+ csize = page_size;
css_put(&mem->css);
mem = NULL;
goto again;
@@ -2002,8 +2002,8 @@
}
} while (ret != CHARGE_OK);
- if (csize > PAGE_SIZE)
- refill_stock(mem, csize - PAGE_SIZE);
+ if (csize > page_size)
+ refill_stock(mem, csize - page_size);
css_put(&mem->css);
done:
*memcg = mem;
@@ -2031,9 +2031,10 @@
}
}
-static void mem_cgroup_cancel_charge(struct mem_cgroup *mem)
+static void mem_cgroup_cancel_charge(struct mem_cgroup *mem,
+ int page_size)
{
- __mem_cgroup_cancel_charge(mem, 1);
+ __mem_cgroup_cancel_charge(mem, page_size >> PAGE_SHIFT);
}
/*
@@ -2087,22 +2088,10 @@
* commit a charge got by __mem_cgroup_try_charge() and makes page_cgroup to be
* USED state. If already USED, uncharge and return.
*/
-
-static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
- struct page_cgroup *pc,
- enum charge_type ctype)
+static void ____mem_cgroup_commit_charge(struct mem_cgroup *mem,
+ struct page_cgroup *pc,
+ enum charge_type ctype)
{
- /* try_charge() can return NULL to *memcg, taking care of it. */
- if (!mem)
- return;
-
- lock_page_cgroup(pc);
- if (unlikely(PageCgroupUsed(pc))) {
- unlock_page_cgroup(pc);
- mem_cgroup_cancel_charge(mem);
- return;
- }
-
pc->mem_cgroup = mem;
/*
* We access a page_cgroup asynchronously without lock_page_cgroup().
@@ -2127,6 +2116,33 @@
}
mem_cgroup_charge_statistics(mem, pc, true);
+}
+
+static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
+ struct page_cgroup *pc,
+ enum charge_type ctype,
+ int page_size)
+{
+ int i;
+ int count = page_size >> PAGE_SHIFT;
+
+ /* try_charge() can return NULL to *memcg, taking care of it. */
+ if (!mem)
+ return;
+
+ lock_page_cgroup(pc);
+ if (unlikely(PageCgroupUsed(pc))) {
+ unlock_page_cgroup(pc);
+ mem_cgroup_cancel_charge(mem, page_size);
+ return;
+ }
+
+ /*
+ * we don't need page_cgroup_lock about tail pages, becase they are not
+ * accessed by any other context at this point.
+ */
+ for (i = 0; i < count; i++)
+ ____mem_cgroup_commit_charge(mem, pc + i, ctype);
unlock_page_cgroup(pc);
/*
@@ -2173,7 +2189,7 @@
mem_cgroup_charge_statistics(from, pc, false);
if (uncharge)
/* This is not "cancel", but cancel_charge does all we need. */
- mem_cgroup_cancel_charge(from);
+ mem_cgroup_cancel_charge(from, PAGE_SIZE);
/* caller should have done css_get */
pc->mem_cgroup = to;
@@ -2195,9 +2211,13 @@
struct mem_cgroup *from, struct mem_cgroup *to, bool uncharge)
{
int ret = -EINVAL;
+ unsigned long flags;
+
lock_page_cgroup(pc);
if (PageCgroupUsed(pc) && pc->mem_cgroup == from) {
+ move_lock_page_cgroup(pc, &flags);
__mem_cgroup_move_account(pc, from, to, uncharge);
+ move_unlock_page_cgroup(pc, &flags);
ret = 0;
}
unlock_page_cgroup(pc);
@@ -2234,13 +2254,14 @@
goto put;
parent = mem_cgroup_from_cont(pcg);
- ret = __mem_cgroup_try_charge(NULL, gfp_mask, &parent, false);
+ ret = __mem_cgroup_try_charge(NULL, gfp_mask, &parent, false,
+ PAGE_SIZE);
if (ret || !parent)
goto put_back;
ret = mem_cgroup_move_account(pc, child, parent, true);
if (ret)
- mem_cgroup_cancel_charge(parent);
+ mem_cgroup_cancel_charge(parent, PAGE_SIZE);
put_back:
putback_lru_page(page);
put:
@@ -2261,6 +2282,12 @@
struct mem_cgroup *mem = NULL;
struct page_cgroup *pc;
int ret;
+ int page_size = PAGE_SIZE;
+
+ if (PageTransHuge(page)) {
+ page_size <<= compound_order(page);
+ VM_BUG_ON(!PageTransHuge(page));
+ }
pc = lookup_page_cgroup(page);
/* can happen at boot */
@@ -2268,11 +2295,11 @@
return 0;
prefetchw(pc);
- ret = __mem_cgroup_try_charge(mm, gfp_mask, &mem, true);
+ ret = __mem_cgroup_try_charge(mm, gfp_mask, &mem, true, page_size);
if (ret || !mem)
return ret;
- __mem_cgroup_commit_charge(mem, pc, ctype);
+ __mem_cgroup_commit_charge(mem, pc, ctype, page_size);
return 0;
}
@@ -2281,8 +2308,6 @@
{
if (mem_cgroup_disabled())
return 0;
- if (PageCompound(page))
- return 0;
/*
* If already mapped, we don't have to account.
* If page cache, page->mapping has address_space.
@@ -2388,13 +2413,13 @@
if (!mem)
goto charge_cur_mm;
*ptr = mem;
- ret = __mem_cgroup_try_charge(NULL, mask, ptr, true);
+ ret = __mem_cgroup_try_charge(NULL, mask, ptr, true, PAGE_SIZE);
css_put(&mem->css);
return ret;
charge_cur_mm:
if (unlikely(!mm))
mm = &init_mm;
- return __mem_cgroup_try_charge(mm, mask, ptr, true);
+ return __mem_cgroup_try_charge(mm, mask, ptr, true, PAGE_SIZE);
}
static void
@@ -2410,7 +2435,7 @@
cgroup_exclude_rmdir(&ptr->css);
pc = lookup_page_cgroup(page);
mem_cgroup_lru_del_before_commit_swapcache(page);
- __mem_cgroup_commit_charge(ptr, pc, ctype);
+ __mem_cgroup_commit_charge(ptr, pc, ctype, PAGE_SIZE);
mem_cgroup_lru_add_after_commit_swapcache(page);
/*
* Now swap is on-memory. This means this page may be
@@ -2459,11 +2484,12 @@
return;
if (!mem)
return;
- mem_cgroup_cancel_charge(mem);
+ mem_cgroup_cancel_charge(mem, PAGE_SIZE);
}
static void
-__do_uncharge(struct mem_cgroup *mem, const enum charge_type ctype)
+__do_uncharge(struct mem_cgroup *mem, const enum charge_type ctype,
+ int page_size)
{
struct memcg_batch_info *batch = NULL;
bool uncharge_memsw = true;
@@ -2490,6 +2516,9 @@
if (!batch->do_batch || test_thread_flag(TIF_MEMDIE))
goto direct_uncharge;
+ if (page_size != PAGE_SIZE)
+ goto direct_uncharge;
+
/*
* In typical case, batch->memcg == mem. This means we can
* merge a series of uncharges to an uncharge of res_counter.
@@ -2503,9 +2532,9 @@
batch->memsw_bytes += PAGE_SIZE;
return;
direct_uncharge:
- res_counter_uncharge(&mem->res, PAGE_SIZE);
+ res_counter_uncharge(&mem->res, page_size);
if (uncharge_memsw)
- res_counter_uncharge(&mem->memsw, PAGE_SIZE);
+ res_counter_uncharge(&mem->memsw, page_size);
if (unlikely(batch->memcg != mem))
memcg_oom_recover(mem);
return;
@@ -2517,8 +2546,11 @@
static struct mem_cgroup *
__mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
{
+ int i;
+ int count;
struct page_cgroup *pc;
struct mem_cgroup *mem = NULL;
+ int page_size = PAGE_SIZE;
if (mem_cgroup_disabled())
return NULL;
@@ -2526,6 +2558,12 @@
if (PageSwapCache(page))
return NULL;
+ if (PageTransHuge(page)) {
+ page_size <<= compound_order(page);
+ VM_BUG_ON(!PageTransHuge(page));
+ }
+
+ count = page_size >> PAGE_SHIFT;
/*
* Check if our page_cgroup is valid
*/
@@ -2558,7 +2596,8 @@
break;
}
- mem_cgroup_charge_statistics(mem, pc, false);
+ for (i = 0; i < count; i++)
+ mem_cgroup_charge_statistics(mem, pc + i, false);
ClearPageCgroupUsed(pc);
/*
@@ -2579,7 +2618,7 @@
mem_cgroup_get(mem);
}
if (!mem_cgroup_is_root(mem))
- __do_uncharge(mem, ctype);
+ __do_uncharge(mem, ctype, page_size);
return mem;
@@ -2774,6 +2813,7 @@
enum charge_type ctype;
int ret = 0;
+ VM_BUG_ON(PageTransHuge(page));
if (mem_cgroup_disabled())
return 0;
@@ -2823,7 +2863,7 @@
return 0;
*ptr = mem;
- ret = __mem_cgroup_try_charge(NULL, GFP_KERNEL, ptr, false);
+ ret = __mem_cgroup_try_charge(NULL, GFP_KERNEL, ptr, false, PAGE_SIZE);
css_put(&mem->css);/* drop extra refcnt */
if (ret || *ptr == NULL) {
if (PageAnon(page)) {
@@ -2850,13 +2890,13 @@
ctype = MEM_CGROUP_CHARGE_TYPE_CACHE;
else
ctype = MEM_CGROUP_CHARGE_TYPE_SHMEM;
- __mem_cgroup_commit_charge(mem, pc, ctype);
+ __mem_cgroup_commit_charge(mem, pc, ctype, PAGE_SIZE);
return ret;
}
/* remove redundant charge if migration failed*/
void mem_cgroup_end_migration(struct mem_cgroup *mem,
- struct page *oldpage, struct page *newpage)
+ struct page *oldpage, struct page *newpage, bool migration_ok)
{
struct page *used, *unused;
struct page_cgroup *pc;
@@ -2865,8 +2905,7 @@
return;
/* blocks rmdir() */
cgroup_exclude_rmdir(&mem->css);
- /* at migration success, oldpage->mapping is NULL. */
- if (oldpage->mapping) {
+ if (!migration_ok) {
used = oldpage;
unused = newpage;
} else {
@@ -4176,13 +4215,11 @@
*/
if (!node_state(node, N_NORMAL_MEMORY))
tmp = -1;
- pn = kmalloc_node(sizeof(*pn), GFP_KERNEL, tmp);
+ pn = kzalloc_node(sizeof(*pn), GFP_KERNEL, tmp);
if (!pn)
return 1;
mem->info.nodeinfo[node] = pn;
- memset(pn, 0, sizeof(*pn));
-
for (zone = 0; zone < MAX_NR_ZONES; zone++) {
mz = &pn->zoneinfo[zone];
for_each_lru(l)
@@ -4206,14 +4243,13 @@
/* Can be very big if MAX_NUMNODES is very big */
if (size < PAGE_SIZE)
- mem = kmalloc(size, GFP_KERNEL);
+ mem = kzalloc(size, GFP_KERNEL);
else
- mem = vmalloc(size);
+ mem = vzalloc(size);
if (!mem)
return NULL;
- memset(mem, 0, size);
mem->stat = alloc_percpu(struct mem_cgroup_stat_cpu);
if (!mem->stat)
goto out_free;
@@ -4461,7 +4497,8 @@
batch_count = PRECHARGE_COUNT_AT_ONCE;
cond_resched();
}
- ret = __mem_cgroup_try_charge(NULL, GFP_KERNEL, &mem, false);
+ ret = __mem_cgroup_try_charge(NULL, GFP_KERNEL, &mem, false,
+ PAGE_SIZE);
if (ret || !mem)
/* mem_cgroup_clear_mc() will do uncharge later */
return -ENOMEM;
@@ -4623,6 +4660,7 @@
pte_t *pte;
spinlock_t *ptl;
+ VM_BUG_ON(pmd_trans_huge(*pmd));
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
for (; addr != end; pte++, addr += PAGE_SIZE)
if (is_target_pte_for_mc(vma, addr, *pte, NULL))
@@ -4638,7 +4676,7 @@
unsigned long precharge;
struct vm_area_struct *vma;
- /* We've already held the mmap_sem */
+ down_read(&mm->mmap_sem);
for (vma = mm->mmap; vma; vma = vma->vm_next) {
struct mm_walk mem_cgroup_count_precharge_walk = {
.pmd_entry = mem_cgroup_count_precharge_pte_range,
@@ -4650,6 +4688,7 @@
walk_page_range(vma->vm_start, vma->vm_end,
&mem_cgroup_count_precharge_walk);
}
+ up_read(&mm->mmap_sem);
precharge = mc.precharge;
mc.precharge = 0;
@@ -4659,10 +4698,15 @@
static int mem_cgroup_precharge_mc(struct mm_struct *mm)
{
- return mem_cgroup_do_precharge(mem_cgroup_count_precharge(mm));
+ unsigned long precharge = mem_cgroup_count_precharge(mm);
+
+ VM_BUG_ON(mc.moving_task);
+ mc.moving_task = current;
+ return mem_cgroup_do_precharge(precharge);
}
-static void mem_cgroup_clear_mc(void)
+/* cancels all extra charges on mc.from and mc.to, and wakes up all waiters. */
+static void __mem_cgroup_clear_mc(void)
{
struct mem_cgroup *from = mc.from;
struct mem_cgroup *to = mc.to;
@@ -4697,23 +4741,28 @@
PAGE_SIZE * mc.moved_swap);
}
/* we've already done mem_cgroup_get(mc.to) */
-
mc.moved_swap = 0;
}
- if (mc.mm) {
- up_read(&mc.mm->mmap_sem);
- mmput(mc.mm);
- }
+ memcg_oom_recover(from);
+ memcg_oom_recover(to);
+ wake_up_all(&mc.waitq);
+}
+
+static void mem_cgroup_clear_mc(void)
+{
+ struct mem_cgroup *from = mc.from;
+
+ /*
+ * we must clear moving_task before waking up waiters at the end of
+ * task migration.
+ */
+ mc.moving_task = NULL;
+ __mem_cgroup_clear_mc();
spin_lock(&mc.lock);
mc.from = NULL;
mc.to = NULL;
spin_unlock(&mc.lock);
- mc.moving_task = NULL;
- mc.mm = NULL;
mem_cgroup_end_move(from);
- memcg_oom_recover(from);
- memcg_oom_recover(to);
- wake_up_all(&mc.waitq);
}
static int mem_cgroup_can_attach(struct cgroup_subsys *ss,
@@ -4735,38 +4784,23 @@
return 0;
/* We move charges only when we move a owner of the mm */
if (mm->owner == p) {
- /*
- * We do all the move charge works under one mmap_sem to
- * avoid deadlock with down_write(&mmap_sem)
- * -> try_charge() -> if (mc.moving_task) -> sleep.
- */
- down_read(&mm->mmap_sem);
-
VM_BUG_ON(mc.from);
VM_BUG_ON(mc.to);
VM_BUG_ON(mc.precharge);
VM_BUG_ON(mc.moved_charge);
VM_BUG_ON(mc.moved_swap);
- VM_BUG_ON(mc.moving_task);
- VM_BUG_ON(mc.mm);
-
mem_cgroup_start_move(from);
spin_lock(&mc.lock);
mc.from = from;
mc.to = mem;
- mc.precharge = 0;
- mc.moved_charge = 0;
- mc.moved_swap = 0;
spin_unlock(&mc.lock);
- mc.moving_task = current;
- mc.mm = mm;
+ /* We set mc.moving_task later */
ret = mem_cgroup_precharge_mc(mm);
if (ret)
mem_cgroup_clear_mc();
- /* We call up_read() and mmput() in clear_mc(). */
- } else
- mmput(mm);
+ }
+ mmput(mm);
}
return ret;
}
@@ -4789,6 +4823,7 @@
spinlock_t *ptl;
retry:
+ VM_BUG_ON(pmd_trans_huge(*pmd));
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
for (; addr != end; addr += PAGE_SIZE) {
pte_t ptent = *(pte++);
@@ -4854,7 +4889,19 @@
struct vm_area_struct *vma;
lru_add_drain_all();
- /* We've already held the mmap_sem */
+retry:
+ if (unlikely(!down_read_trylock(&mm->mmap_sem))) {
+ /*
+ * Someone who are holding the mmap_sem might be waiting in
+ * waitq. So we cancel all extra charges, wake up all waiters,
+ * and retry. Because we cancel precharges, we might not be able
+ * to move enough charges, but moving charge is a best-effort
+ * feature anyway, so it wouldn't be a big problem.
+ */
+ __mem_cgroup_clear_mc();
+ cond_resched();
+ goto retry;
+ }
for (vma = mm->mmap; vma; vma = vma->vm_next) {
int ret;
struct mm_walk mem_cgroup_move_charge_walk = {
@@ -4873,6 +4920,7 @@
*/
break;
}
+ up_read(&mm->mmap_sem);
}
static void mem_cgroup_move_task(struct cgroup_subsys *ss,
@@ -4881,11 +4929,17 @@
struct task_struct *p,
bool threadgroup)
{
- if (!mc.mm)
+ struct mm_struct *mm;
+
+ if (!mc.to)
/* no need to move charge */
return;
- mem_cgroup_move_charge(mc.mm);
+ mm = get_task_mm(p);
+ if (mm) {
+ mem_cgroup_move_charge(mm);
+ mmput(mm);
+ }
mem_cgroup_clear_mc();
}
#else /* !CONFIG_MMU */
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 46ab2c0..548fbd7 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -203,7 +203,7 @@
#ifdef __ARCH_SI_TRAPNO
si.si_trapno = trapno;
#endif
- si.si_addr_lsb = compound_order(compound_head(page)) + PAGE_SHIFT;
+ si.si_addr_lsb = compound_trans_order(compound_head(page)) + PAGE_SHIFT;
/*
* Don't use force here, it's convenient if the signal
* can be temporarily blocked.
@@ -386,6 +386,8 @@
struct task_struct *tsk;
struct anon_vma *av;
+ if (!PageHuge(page) && unlikely(split_huge_page(page)))
+ return;
read_lock(&tasklist_lock);
av = page_lock_anon_vma(page);
if (av == NULL) /* Not actually mapped anymore */
@@ -928,7 +930,7 @@
static void set_page_hwpoison_huge_page(struct page *hpage)
{
int i;
- int nr_pages = 1 << compound_order(hpage);
+ int nr_pages = 1 << compound_trans_order(hpage);
for (i = 0; i < nr_pages; i++)
SetPageHWPoison(hpage + i);
}
@@ -936,7 +938,7 @@
static void clear_page_hwpoison_huge_page(struct page *hpage)
{
int i;
- int nr_pages = 1 << compound_order(hpage);
+ int nr_pages = 1 << compound_trans_order(hpage);
for (i = 0; i < nr_pages; i++)
ClearPageHWPoison(hpage + i);
}
@@ -966,7 +968,7 @@
return 0;
}
- nr_pages = 1 << compound_order(hpage);
+ nr_pages = 1 << compound_trans_order(hpage);
atomic_long_add(nr_pages, &mce_bad_pages);
/*
@@ -1164,7 +1166,7 @@
return 0;
}
- nr_pages = 1 << compound_order(page);
+ nr_pages = 1 << compound_trans_order(page);
if (!get_page_unless_zero(page)) {
/*
@@ -1290,9 +1292,10 @@
/* Keep page count to indicate a given hugepage is isolated. */
list_add(&hpage->lru, &pagelist);
- ret = migrate_huge_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0);
+ ret = migrate_huge_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0,
+ true);
if (ret) {
- putback_lru_pages(&pagelist);
+ putback_lru_pages(&pagelist);
pr_debug("soft offline: %#lx: migration failed %d, type %lx\n",
pfn, ret, page->flags);
if (ret > 0)
@@ -1301,7 +1304,7 @@
}
done:
if (!PageHWPoison(hpage))
- atomic_long_add(1 << compound_order(hpage), &mce_bad_pages);
+ atomic_long_add(1 << compound_trans_order(hpage), &mce_bad_pages);
set_page_hwpoison_huge_page(hpage);
dequeue_hwpoisoned_huge_page(hpage);
/* keep elevated page count for bad page */
@@ -1413,7 +1416,8 @@
LIST_HEAD(pagelist);
list_add(&page->lru, &pagelist);
- ret = migrate_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0);
+ ret = migrate_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL,
+ 0, true);
if (ret) {
pr_info("soft offline: %#lx: migration failed %d, type %lx\n",
pfn, ret, page->flags);
diff --git a/mm/memory.c b/mm/memory.c
index 02e48aa..31250fa 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -394,9 +394,11 @@
}
}
-int __pte_alloc(struct mm_struct *mm, pmd_t *pmd, unsigned long address)
+int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma,
+ pmd_t *pmd, unsigned long address)
{
pgtable_t new = pte_alloc_one(mm, address);
+ int wait_split_huge_page;
if (!new)
return -ENOMEM;
@@ -416,14 +418,18 @@
smp_wmb(); /* Could be smp_wmb__xxx(before|after)_spin_lock */
spin_lock(&mm->page_table_lock);
- if (!pmd_present(*pmd)) { /* Has another populated it ? */
+ wait_split_huge_page = 0;
+ if (likely(pmd_none(*pmd))) { /* Has another populated it ? */
mm->nr_ptes++;
pmd_populate(mm, pmd, new);
new = NULL;
- }
+ } else if (unlikely(pmd_trans_splitting(*pmd)))
+ wait_split_huge_page = 1;
spin_unlock(&mm->page_table_lock);
if (new)
pte_free(mm, new);
+ if (wait_split_huge_page)
+ wait_split_huge_page(vma->anon_vma, pmd);
return 0;
}
@@ -436,10 +442,11 @@
smp_wmb(); /* See comment in __pte_alloc */
spin_lock(&init_mm.page_table_lock);
- if (!pmd_present(*pmd)) { /* Has another populated it ? */
+ if (likely(pmd_none(*pmd))) { /* Has another populated it ? */
pmd_populate_kernel(&init_mm, pmd, new);
new = NULL;
- }
+ } else
+ VM_BUG_ON(pmd_trans_splitting(*pmd));
spin_unlock(&init_mm.page_table_lock);
if (new)
pte_free_kernel(&init_mm, new);
@@ -719,9 +726,9 @@
return 0;
}
-static int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
- pmd_t *dst_pmd, pmd_t *src_pmd, struct vm_area_struct *vma,
- unsigned long addr, unsigned long end)
+int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
+ pmd_t *dst_pmd, pmd_t *src_pmd, struct vm_area_struct *vma,
+ unsigned long addr, unsigned long end)
{
pte_t *orig_src_pte, *orig_dst_pte;
pte_t *src_pte, *dst_pte;
@@ -795,6 +802,17 @@
src_pmd = pmd_offset(src_pud, addr);
do {
next = pmd_addr_end(addr, end);
+ if (pmd_trans_huge(*src_pmd)) {
+ int err;
+ VM_BUG_ON(next-addr != HPAGE_PMD_SIZE);
+ err = copy_huge_pmd(dst_mm, src_mm,
+ dst_pmd, src_pmd, addr, vma);
+ if (err == -ENOMEM)
+ return -ENOMEM;
+ if (!err)
+ continue;
+ /* fall through */
+ }
if (pmd_none_or_clear_bad(src_pmd))
continue;
if (copy_pte_range(dst_mm, src_mm, dst_pmd, src_pmd,
@@ -997,6 +1015,16 @@
pmd = pmd_offset(pud, addr);
do {
next = pmd_addr_end(addr, end);
+ if (pmd_trans_huge(*pmd)) {
+ if (next-addr != HPAGE_PMD_SIZE) {
+ VM_BUG_ON(!rwsem_is_locked(&tlb->mm->mmap_sem));
+ split_huge_page_pmd(vma->vm_mm, pmd);
+ } else if (zap_huge_pmd(tlb, vma, pmd)) {
+ (*zap_work)--;
+ continue;
+ }
+ /* fall through */
+ }
if (pmd_none_or_clear_bad(pmd)) {
(*zap_work)--;
continue;
@@ -1262,7 +1290,7 @@
pud = pud_offset(pgd, address);
if (pud_none(*pud))
goto no_page_table;
- if (pud_huge(*pud)) {
+ if (pud_huge(*pud) && vma->vm_flags & VM_HUGETLB) {
BUG_ON(flags & FOLL_GET);
page = follow_huge_pud(mm, address, pud, flags & FOLL_WRITE);
goto out;
@@ -1273,11 +1301,32 @@
pmd = pmd_offset(pud, address);
if (pmd_none(*pmd))
goto no_page_table;
- if (pmd_huge(*pmd)) {
+ if (pmd_huge(*pmd) && vma->vm_flags & VM_HUGETLB) {
BUG_ON(flags & FOLL_GET);
page = follow_huge_pmd(mm, address, pmd, flags & FOLL_WRITE);
goto out;
}
+ if (pmd_trans_huge(*pmd)) {
+ if (flags & FOLL_SPLIT) {
+ split_huge_page_pmd(mm, pmd);
+ goto split_fallthrough;
+ }
+ spin_lock(&mm->page_table_lock);
+ if (likely(pmd_trans_huge(*pmd))) {
+ if (unlikely(pmd_trans_splitting(*pmd))) {
+ spin_unlock(&mm->page_table_lock);
+ wait_split_huge_page(vma->anon_vma, pmd);
+ } else {
+ page = follow_trans_huge_pmd(mm, address,
+ pmd, flags);
+ spin_unlock(&mm->page_table_lock);
+ goto out;
+ }
+ } else
+ spin_unlock(&mm->page_table_lock);
+ /* fall through */
+ }
+split_fallthrough:
if (unlikely(pmd_bad(*pmd)))
goto no_page_table;
@@ -1310,6 +1359,28 @@
*/
mark_page_accessed(page);
}
+ if (flags & FOLL_MLOCK) {
+ /*
+ * The preliminary mapping check is mainly to avoid the
+ * pointless overhead of lock_page on the ZERO_PAGE
+ * which might bounce very badly if there is contention.
+ *
+ * If the page is already locked, we don't need to
+ * handle it now - vmscan will handle it later if and
+ * when it attempts to reclaim the page.
+ */
+ if (page->mapping && trylock_page(page)) {
+ lru_add_drain(); /* push cached pages to LRU */
+ /*
+ * Because we lock page here and migration is
+ * blocked by the pte's page reference, we need
+ * only check for file-cache page truncation.
+ */
+ if (page->mapping)
+ mlock_vma_page(page);
+ unlock_page(page);
+ }
+ }
unlock:
pte_unmap_unlock(ptep, ptl);
out:
@@ -1341,7 +1412,8 @@
int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, int nr_pages, unsigned int gup_flags,
- struct page **pages, struct vm_area_struct **vmas)
+ struct page **pages, struct vm_area_struct **vmas,
+ int *nonblocking)
{
int i;
unsigned long vm_flags;
@@ -1386,6 +1458,7 @@
pmd = pmd_offset(pud, pg);
if (pmd_none(*pmd))
return i ? : -EFAULT;
+ VM_BUG_ON(pmd_trans_huge(*pmd));
pte = pte_offset_map(pmd, pg);
if (pte_none(*pte)) {
pte_unmap(pte);
@@ -1441,10 +1514,15 @@
cond_resched();
while (!(page = follow_page(vma, start, foll_flags))) {
int ret;
+ unsigned int fault_flags = 0;
+
+ if (foll_flags & FOLL_WRITE)
+ fault_flags |= FAULT_FLAG_WRITE;
+ if (nonblocking)
+ fault_flags |= FAULT_FLAG_ALLOW_RETRY;
ret = handle_mm_fault(mm, vma, start,
- (foll_flags & FOLL_WRITE) ?
- FAULT_FLAG_WRITE : 0);
+ fault_flags);
if (ret & VM_FAULT_ERROR) {
if (ret & VM_FAULT_OOM)
@@ -1460,6 +1538,11 @@
else
tsk->min_flt++;
+ if (ret & VM_FAULT_RETRY) {
+ *nonblocking = 0;
+ return i;
+ }
+
/*
* The VM_FAULT_WRITE bit tells us that
* do_wp_page has broken COW when necessary,
@@ -1559,7 +1642,8 @@
if (force)
flags |= FOLL_FORCE;
- return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas);
+ return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas,
+ NULL);
}
EXPORT_SYMBOL(get_user_pages);
@@ -1584,7 +1668,8 @@
struct page *page;
if (__get_user_pages(current, current->mm, addr, 1,
- FOLL_FORCE | FOLL_DUMP | FOLL_GET, &page, &vma) < 1)
+ FOLL_FORCE | FOLL_DUMP | FOLL_GET, &page, &vma,
+ NULL) < 1)
return NULL;
flush_cache_page(vma, addr, page_to_pfn(page));
return page;
@@ -1598,8 +1683,10 @@
pud_t * pud = pud_alloc(mm, pgd, addr);
if (pud) {
pmd_t * pmd = pmd_alloc(mm, pud, addr);
- if (pmd)
+ if (pmd) {
+ VM_BUG_ON(pmd_trans_huge(*pmd));
return pte_alloc_map_lock(mm, pmd, addr, ptl);
+ }
}
return NULL;
}
@@ -1818,6 +1905,7 @@
pmd = pmd_alloc(mm, pud, addr);
if (!pmd)
return -ENOMEM;
+ VM_BUG_ON(pmd_trans_huge(*pmd));
do {
next = pmd_addr_end(addr, end);
if (remap_pte_range(mm, pmd, addr, next,
@@ -2048,19 +2136,6 @@
return same;
}
-/*
- * Do pte_mkwrite, but only if the vma says VM_WRITE. We do this when
- * servicing faults for write access. In the normal case, do always want
- * pte_mkwrite. But get_user_pages can cause write faults for mappings
- * that do not have writing enabled, when used by access_process_vm.
- */
-static inline pte_t maybe_mkwrite(pte_t pte, struct vm_area_struct *vma)
-{
- if (likely(vma->vm_flags & VM_WRITE))
- pte = pte_mkwrite(pte);
- return pte;
-}
-
static inline void cow_user_page(struct page *dst, struct page *src, unsigned long va, struct vm_area_struct *vma)
{
/*
@@ -2112,7 +2187,7 @@
{
struct page *old_page, *new_page;
pte_t entry;
- int reuse = 0, ret = 0;
+ int ret = 0;
int page_mkwrite = 0;
struct page *dirty_page = NULL;
@@ -2149,14 +2224,16 @@
}
page_cache_release(old_page);
}
- reuse = reuse_swap_page(old_page);
- if (reuse)
+ if (reuse_swap_page(old_page)) {
/*
* The page is all ours. Move it to our anon_vma so
* the rmap code will not search our parent or siblings.
* Protected against the rmap code by the page lock.
*/
page_move_anon_rmap(old_page, vma, address);
+ unlock_page(old_page);
+ goto reuse;
+ }
unlock_page(old_page);
} else if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
(VM_WRITE|VM_SHARED))) {
@@ -2220,18 +2297,52 @@
}
dirty_page = old_page;
get_page(dirty_page);
- reuse = 1;
- }
- if (reuse) {
reuse:
flush_cache_page(vma, address, pte_pfn(orig_pte));
entry = pte_mkyoung(orig_pte);
entry = maybe_mkwrite(pte_mkdirty(entry), vma);
if (ptep_set_access_flags(vma, address, page_table, entry,1))
update_mmu_cache(vma, address, page_table);
+ pte_unmap_unlock(page_table, ptl);
ret |= VM_FAULT_WRITE;
- goto unlock;
+
+ if (!dirty_page)
+ return ret;
+
+ /*
+ * Yes, Virginia, this is actually required to prevent a race
+ * with clear_page_dirty_for_io() from clearing the page dirty
+ * bit after it clear all dirty ptes, but before a racing
+ * do_wp_page installs a dirty pte.
+ *
+ * do_no_page is protected similarly.
+ */
+ if (!page_mkwrite) {
+ wait_on_page_locked(dirty_page);
+ set_page_dirty_balance(dirty_page, page_mkwrite);
+ }
+ put_page(dirty_page);
+ if (page_mkwrite) {
+ struct address_space *mapping = dirty_page->mapping;
+
+ set_page_dirty(dirty_page);
+ unlock_page(dirty_page);
+ page_cache_release(dirty_page);
+ if (mapping) {
+ /*
+ * Some device drivers do not set page.mapping
+ * but still dirty their pages
+ */
+ balance_dirty_pages_ratelimited(mapping);
+ }
+ }
+
+ /* file_update_time outside page_lock */
+ if (vma->vm_file)
+ file_update_time(vma->vm_file);
+
+ return ret;
}
/*
@@ -2337,39 +2448,6 @@
page_cache_release(old_page);
unlock:
pte_unmap_unlock(page_table, ptl);
- if (dirty_page) {
- /*
- * Yes, Virginia, this is actually required to prevent a race
- * with clear_page_dirty_for_io() from clearing the page dirty
- * bit after it clear all dirty ptes, but before a racing
- * do_wp_page installs a dirty pte.
- *
- * do_no_page is protected similarly.
- */
- if (!page_mkwrite) {
- wait_on_page_locked(dirty_page);
- set_page_dirty_balance(dirty_page, page_mkwrite);
- }
- put_page(dirty_page);
- if (page_mkwrite) {
- struct address_space *mapping = dirty_page->mapping;
-
- set_page_dirty(dirty_page);
- unlock_page(dirty_page);
- page_cache_release(dirty_page);
- if (mapping) {
- /*
- * Some device drivers do not set page.mapping
- * but still dirty their pages
- */
- balance_dirty_pages_ratelimited(mapping);
- }
- }
-
- /* file_update_time outside page_lock */
- if (vma->vm_file)
- file_update_time(vma->vm_file);
- }
return ret;
oom_free_new:
page_cache_release(new_page);
@@ -3147,9 +3225,9 @@
* but allow concurrent faults), and pte mapped but not yet locked.
* We return with mmap_sem still held, but pte unmapped and unlocked.
*/
-static inline int handle_pte_fault(struct mm_struct *mm,
- struct vm_area_struct *vma, unsigned long address,
- pte_t *pte, pmd_t *pmd, unsigned int flags)
+int handle_pte_fault(struct mm_struct *mm,
+ struct vm_area_struct *vma, unsigned long address,
+ pte_t *pte, pmd_t *pmd, unsigned int flags)
{
pte_t entry;
spinlock_t *ptl;
@@ -3228,9 +3306,40 @@
pmd = pmd_alloc(mm, pud, address);
if (!pmd)
return VM_FAULT_OOM;
- pte = pte_alloc_map(mm, pmd, address);
- if (!pte)
+ if (pmd_none(*pmd) && transparent_hugepage_enabled(vma)) {
+ if (!vma->vm_ops)
+ return do_huge_pmd_anonymous_page(mm, vma, address,
+ pmd, flags);
+ } else {
+ pmd_t orig_pmd = *pmd;
+ barrier();
+ if (pmd_trans_huge(orig_pmd)) {
+ if (flags & FAULT_FLAG_WRITE &&
+ !pmd_write(orig_pmd) &&
+ !pmd_trans_splitting(orig_pmd))
+ return do_huge_pmd_wp_page(mm, vma, address,
+ pmd, orig_pmd);
+ return 0;
+ }
+ }
+
+ /*
+ * Use __pte_alloc instead of pte_alloc_map, because we can't
+ * run pte_offset_map on the pmd, if an huge pmd could
+ * materialize from under us from a different thread.
+ */
+ if (unlikely(__pte_alloc(mm, vma, pmd, address)))
return VM_FAULT_OOM;
+ /* if an huge pmd materialized from under us just retry later */
+ if (unlikely(pmd_trans_huge(*pmd)))
+ return 0;
+ /*
+ * A regular pmd is established and it can't morph into a huge pmd
+ * from under us anymore at this point because we hold the mmap_sem
+ * read mode and khugepaged takes it in write mode. So now it's
+ * safe to run pte_offset_map().
+ */
+ pte = pte_offset_map(pmd, address);
return handle_pte_fault(mm, vma, address, pte, pmd, flags);
}
@@ -3296,7 +3405,12 @@
vma = find_vma(current->mm, addr);
if (!vma)
return -ENOMEM;
- write = (vma->vm_flags & VM_WRITE) != 0;
+ /*
+ * We want to touch writable mappings with a write fault in order
+ * to break COW, except for shared mappings because these don't COW
+ * and we would not want to dirty them for nothing.
+ */
+ write = (vma->vm_flags & (VM_WRITE | VM_SHARED)) == VM_WRITE;
BUG_ON(addr >= end);
BUG_ON(end > vma->vm_end);
len = DIV_ROUND_UP(end, PAGE_SIZE) - addr/PAGE_SIZE;
@@ -3368,6 +3482,7 @@
goto out;
pmd = pmd_offset(pud, address);
+ VM_BUG_ON(pmd_trans_huge(*pmd));
if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd)))
goto out;
@@ -3608,3 +3723,74 @@
}
EXPORT_SYMBOL(might_fault);
#endif
+
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS)
+static void clear_gigantic_page(struct page *page,
+ unsigned long addr,
+ unsigned int pages_per_huge_page)
+{
+ int i;
+ struct page *p = page;
+
+ might_sleep();
+ for (i = 0; i < pages_per_huge_page;
+ i++, p = mem_map_next(p, page, i)) {
+ cond_resched();
+ clear_user_highpage(p, addr + i * PAGE_SIZE);
+ }
+}
+void clear_huge_page(struct page *page,
+ unsigned long addr, unsigned int pages_per_huge_page)
+{
+ int i;
+
+ if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) {
+ clear_gigantic_page(page, addr, pages_per_huge_page);
+ return;
+ }
+
+ might_sleep();
+ for (i = 0; i < pages_per_huge_page; i++) {
+ cond_resched();
+ clear_user_highpage(page + i, addr + i * PAGE_SIZE);
+ }
+}
+
+static void copy_user_gigantic_page(struct page *dst, struct page *src,
+ unsigned long addr,
+ struct vm_area_struct *vma,
+ unsigned int pages_per_huge_page)
+{
+ int i;
+ struct page *dst_base = dst;
+ struct page *src_base = src;
+
+ for (i = 0; i < pages_per_huge_page; ) {
+ cond_resched();
+ copy_user_highpage(dst, src, addr + i*PAGE_SIZE, vma);
+
+ i++;
+ dst = mem_map_next(dst, dst_base, i);
+ src = mem_map_next(src, src_base, i);
+ }
+}
+
+void copy_user_huge_page(struct page *dst, struct page *src,
+ unsigned long addr, struct vm_area_struct *vma,
+ unsigned int pages_per_huge_page)
+{
+ int i;
+
+ if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) {
+ copy_user_gigantic_page(dst, src, addr, vma,
+ pages_per_huge_page);
+ return;
+ }
+
+ might_sleep();
+ for (i = 0; i < pages_per_huge_page; i++) {
+ cond_resched();
+ copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma);
+ }
+}
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 2c6523a..321fc74 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -82,9 +82,10 @@
#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
#ifndef CONFIG_SPARSEMEM_VMEMMAP
-static void get_page_bootmem(unsigned long info, struct page *page, int type)
+static void get_page_bootmem(unsigned long info, struct page *page,
+ unsigned long type)
{
- atomic_set(&page->_mapcount, type);
+ page->lru.next = (struct list_head *) type;
SetPagePrivate(page);
set_page_private(page, info);
atomic_inc(&page->_count);
@@ -94,15 +95,16 @@
* so use __ref to tell modpost not to generate a warning */
void __ref put_page_bootmem(struct page *page)
{
- int type;
+ unsigned long type;
- type = atomic_read(&page->_mapcount);
- BUG_ON(type >= -1);
+ type = (unsigned long) page->lru.next;
+ BUG_ON(type < MEMORY_HOTPLUG_MIN_BOOTMEM_TYPE ||
+ type > MEMORY_HOTPLUG_MAX_BOOTMEM_TYPE);
if (atomic_dec_return(&page->_count) == 1) {
ClearPagePrivate(page);
set_page_private(page, 0);
- reset_page_mapcount(page);
+ INIT_LIST_HEAD(&page->lru);
__free_pages_bootmem(page, 0);
}
@@ -407,6 +409,7 @@
int ret;
struct memory_notify arg;
+ lock_memory_hotplug();
arg.start_pfn = pfn;
arg.nr_pages = nr_pages;
arg.status_change_nid = -1;
@@ -419,6 +422,7 @@
ret = notifier_to_errno(ret);
if (ret) {
memory_notify(MEM_CANCEL_ONLINE, &arg);
+ unlock_memory_hotplug();
return ret;
}
/*
@@ -443,6 +447,7 @@
printk(KERN_DEBUG "online_pages %lx at %lx failed\n",
nr_pages, pfn);
memory_notify(MEM_CANCEL_ONLINE, &arg);
+ unlock_memory_hotplug();
return ret;
}
@@ -467,6 +472,7 @@
if (onlined_pages)
memory_notify(MEM_ONLINE, &arg);
+ unlock_memory_hotplug();
return 0;
}
@@ -733,7 +739,8 @@
goto out;
}
/* this function returns # of failed pages */
- ret = migrate_pages(&source, hotremove_migrate_alloc, 0, 1);
+ ret = migrate_pages(&source, hotremove_migrate_alloc, 0,
+ true, true);
if (ret)
putback_lru_pages(&source);
}
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 11ff260..368fc9d 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -514,6 +514,7 @@
pmd = pmd_offset(pud, addr);
do {
next = pmd_addr_end(addr, end);
+ split_huge_page_pmd(vma->vm_mm, pmd);
if (pmd_none_or_clear_bad(pmd))
continue;
if (check_pte_range(vma, pmd, addr, next, nodes,
@@ -935,7 +936,8 @@
return PTR_ERR(vma);
if (!list_empty(&pagelist)) {
- err = migrate_pages(&pagelist, new_node_page, dest, 0);
+ err = migrate_pages(&pagelist, new_node_page, dest,
+ false, true);
if (err)
putback_lru_pages(&pagelist);
}
@@ -1155,7 +1157,8 @@
if (!list_empty(&pagelist)) {
nr_failed = migrate_pages(&pagelist, new_vma_page,
- (unsigned long)vma, 0);
+ (unsigned long)vma,
+ false, true);
if (nr_failed)
putback_lru_pages(&pagelist);
}
@@ -1308,16 +1311,13 @@
/* Find the mm_struct */
rcu_read_lock();
- read_lock(&tasklist_lock);
task = pid ? find_task_by_vpid(pid) : current;
if (!task) {
- read_unlock(&tasklist_lock);
rcu_read_unlock();
err = -ESRCH;
goto out;
}
mm = get_task_mm(task);
- read_unlock(&tasklist_lock);
rcu_read_unlock();
err = -EINVAL;
@@ -1796,7 +1796,7 @@
}
/**
- * alloc_page_vma - Allocate a page for a VMA.
+ * alloc_pages_vma - Allocate a page for a VMA.
*
* @gfp:
* %GFP_USER user allocation.
@@ -1805,6 +1805,7 @@
* %GFP_FS allocation should not call back into a file system.
* %GFP_ATOMIC don't sleep.
*
+ * @order:Order of the GFP allocation.
* @vma: Pointer to VMA or NULL if not available.
* @addr: Virtual Address of the allocation. Must be inside the VMA.
*
@@ -1818,7 +1819,8 @@
* Should be called with the mm_sem of the vma hold.
*/
struct page *
-alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr)
+alloc_pages_vma(gfp_t gfp, int order, struct vm_area_struct *vma,
+ unsigned long addr)
{
struct mempolicy *pol = get_vma_policy(current, vma, addr);
struct zonelist *zl;
@@ -1830,7 +1832,7 @@
nid = interleave_nid(pol, vma, addr, PAGE_SHIFT);
mpol_cond_put(pol);
- page = alloc_page_interleave(gfp, 0, nid);
+ page = alloc_page_interleave(gfp, order, nid);
put_mems_allowed();
return page;
}
@@ -1839,7 +1841,7 @@
/*
* slow path: ref counted shared policy
*/
- struct page *page = __alloc_pages_nodemask(gfp, 0,
+ struct page *page = __alloc_pages_nodemask(gfp, order,
zl, policy_nodemask(gfp, pol));
__mpol_put(pol);
put_mems_allowed();
@@ -1848,7 +1850,8 @@
/*
* fast path: default or task policy
*/
- page = __alloc_pages_nodemask(gfp, 0, zl, policy_nodemask(gfp, pol));
+ page = __alloc_pages_nodemask(gfp, order, zl,
+ policy_nodemask(gfp, pol));
put_mems_allowed();
return page;
}
diff --git a/mm/migrate.c b/mm/migrate.c
index 6ae8a66..46fe8cc 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -113,6 +113,8 @@
goto out;
pmd = pmd_offset(pud, addr);
+ if (pmd_trans_huge(*pmd))
+ goto out;
if (!pmd_present(*pmd))
goto out;
@@ -246,7 +248,7 @@
expected_count = 2 + page_has_private(page);
if (page_count(page) != expected_count ||
- (struct page *)radix_tree_deref_slot(pslot) != page) {
+ radix_tree_deref_slot_protected(pslot, &mapping->tree_lock) != page) {
spin_unlock_irq(&mapping->tree_lock);
return -EAGAIN;
}
@@ -318,7 +320,7 @@
expected_count = 2 + page_has_private(page);
if (page_count(page) != expected_count ||
- (struct page *)radix_tree_deref_slot(pslot) != page) {
+ radix_tree_deref_slot_protected(pslot, &mapping->tree_lock) != page) {
spin_unlock_irq(&mapping->tree_lock);
return -EAGAIN;
}
@@ -614,13 +616,12 @@
* to the newly allocated page in newpage.
*/
static int unmap_and_move(new_page_t get_new_page, unsigned long private,
- struct page *page, int force, int offlining)
+ struct page *page, int force, bool offlining, bool sync)
{
int rc = 0;
int *result = NULL;
struct page *newpage = get_new_page(page, private, &result);
int remap_swapcache = 1;
- int rcu_locked = 0;
int charge = 0;
struct mem_cgroup *mem = NULL;
struct anon_vma *anon_vma = NULL;
@@ -632,6 +633,9 @@
/* page was freed from under us. So we are done. */
goto move_newpage;
}
+ if (unlikely(PageTransHuge(page)))
+ if (unlikely(split_huge_page(page)))
+ goto move_newpage;
/* prepare cgroup just returns 0 or -ENOMEM */
rc = -EAGAIN;
@@ -639,6 +643,23 @@
if (!trylock_page(page)) {
if (!force)
goto move_newpage;
+
+ /*
+ * It's not safe for direct compaction to call lock_page.
+ * For example, during page readahead pages are added locked
+ * to the LRU. Later, when the IO completes the pages are
+ * marked uptodate and unlocked. However, the queueing
+ * could be merging multiple pages for one bio (e.g.
+ * mpage_readpages). If an allocation happens for the
+ * second or third page, the process can end up locking
+ * the same page twice and deadlocking. Rather than
+ * trying to be clever about what pages can be locked,
+ * avoid the use of lock_page for direct compaction
+ * altogether.
+ */
+ if (current->flags & PF_MEMALLOC)
+ goto move_newpage;
+
lock_page(page);
}
@@ -665,27 +686,33 @@
BUG_ON(charge);
if (PageWriteback(page)) {
- if (!force)
+ if (!force || !sync)
goto uncharge;
wait_on_page_writeback(page);
}
/*
* By try_to_unmap(), page->mapcount goes down to 0 here. In this case,
* we cannot notice that anon_vma is freed while we migrates a page.
- * This rcu_read_lock() delays freeing anon_vma pointer until the end
+ * This get_anon_vma() delays freeing anon_vma pointer until the end
* of migration. File cache pages are no problem because of page_lock()
* File Caches may use write_page() or lock_page() in migration, then,
* just care Anon page here.
*/
if (PageAnon(page)) {
- rcu_read_lock();
- rcu_locked = 1;
-
- /* Determine how to safely use anon_vma */
- if (!page_mapped(page)) {
- if (!PageSwapCache(page))
- goto rcu_unlock;
-
+ /*
+ * Only page_lock_anon_vma() understands the subtleties of
+ * getting a hold on an anon_vma from outside one of its mms.
+ */
+ anon_vma = page_lock_anon_vma(page);
+ if (anon_vma) {
+ /*
+ * Take a reference count on the anon_vma if the
+ * page is mapped so that it is guaranteed to
+ * exist when the page is remapped later
+ */
+ get_anon_vma(anon_vma);
+ page_unlock_anon_vma(anon_vma);
+ } else if (PageSwapCache(page)) {
/*
* We cannot be sure that the anon_vma of an unmapped
* swapcache page is safe to use because we don't
@@ -700,13 +727,7 @@
*/
remap_swapcache = 0;
} else {
- /*
- * Take a reference count on the anon_vma if the
- * page is mapped so that it is guaranteed to
- * exist when the page is remapped later
- */
- anon_vma = page_anon_vma(page);
- get_anon_vma(anon_vma);
+ goto uncharge;
}
}
@@ -723,16 +744,10 @@
* free the metadata, so the page can be freed.
*/
if (!page->mapping) {
- if (!PageAnon(page) && page_has_private(page)) {
- /*
- * Go direct to try_to_free_buffers() here because
- * a) that's what try_to_release_page() would do anyway
- * b) we may be under rcu_read_lock() here, so we can't
- * use GFP_KERNEL which is what try_to_release_page()
- * needs to be effective.
- */
+ VM_BUG_ON(PageAnon(page));
+ if (page_has_private(page)) {
try_to_free_buffers(page);
- goto rcu_unlock;
+ goto uncharge;
}
goto skip_unmap;
}
@@ -746,17 +761,14 @@
if (rc && remap_swapcache)
remove_migration_ptes(page, page);
-rcu_unlock:
/* Drop an anon_vma reference if we took one */
if (anon_vma)
drop_anon_vma(anon_vma);
- if (rcu_locked)
- rcu_read_unlock();
uncharge:
if (!charge)
- mem_cgroup_end_migration(mem, page, newpage);
+ mem_cgroup_end_migration(mem, page, newpage, rc == 0);
unlock:
unlock_page(page);
@@ -810,12 +822,11 @@
*/
static int unmap_and_move_huge_page(new_page_t get_new_page,
unsigned long private, struct page *hpage,
- int force, int offlining)
+ int force, bool offlining, bool sync)
{
int rc = 0;
int *result = NULL;
struct page *new_hpage = get_new_page(hpage, private, &result);
- int rcu_locked = 0;
struct anon_vma *anon_vma = NULL;
if (!new_hpage)
@@ -824,18 +835,16 @@
rc = -EAGAIN;
if (!trylock_page(hpage)) {
- if (!force)
+ if (!force || !sync)
goto out;
lock_page(hpage);
}
if (PageAnon(hpage)) {
- rcu_read_lock();
- rcu_locked = 1;
-
- if (page_mapped(hpage)) {
- anon_vma = page_anon_vma(hpage);
- atomic_inc(&anon_vma->external_refcount);
+ anon_vma = page_lock_anon_vma(hpage);
+ if (anon_vma) {
+ get_anon_vma(anon_vma);
+ page_unlock_anon_vma(anon_vma);
}
}
@@ -847,16 +856,8 @@
if (rc)
remove_migration_ptes(hpage, hpage);
- if (anon_vma && atomic_dec_and_lock(&anon_vma->external_refcount,
- &anon_vma->lock)) {
- int empty = list_empty(&anon_vma->head);
- spin_unlock(&anon_vma->lock);
- if (empty)
- anon_vma_free(anon_vma);
- }
-
- if (rcu_locked)
- rcu_read_unlock();
+ if (anon_vma)
+ drop_anon_vma(anon_vma);
out:
unlock_page(hpage);
@@ -892,7 +893,8 @@
* Return: Number of pages not migrated or error code.
*/
int migrate_pages(struct list_head *from,
- new_page_t get_new_page, unsigned long private, int offlining)
+ new_page_t get_new_page, unsigned long private, bool offlining,
+ bool sync)
{
int retry = 1;
int nr_failed = 0;
@@ -912,7 +914,8 @@
cond_resched();
rc = unmap_and_move(get_new_page, private,
- page, pass > 2, offlining);
+ page, pass > 2, offlining,
+ sync);
switch(rc) {
case -ENOMEM:
@@ -941,7 +944,8 @@
}
int migrate_huge_pages(struct list_head *from,
- new_page_t get_new_page, unsigned long private, int offlining)
+ new_page_t get_new_page, unsigned long private, bool offlining,
+ bool sync)
{
int retry = 1;
int nr_failed = 0;
@@ -957,7 +961,8 @@
cond_resched();
rc = unmap_and_move_huge_page(get_new_page,
- private, page, pass > 2, offlining);
+ private, page, pass > 2, offlining,
+ sync);
switch(rc) {
case -ENOMEM:
@@ -1042,7 +1047,7 @@
if (!vma || pp->addr < vma->vm_start || !vma_migratable(vma))
goto set_status;
- page = follow_page(vma, pp->addr, FOLL_GET);
+ page = follow_page(vma, pp->addr, FOLL_GET|FOLL_SPLIT);
err = PTR_ERR(page);
if (IS_ERR(page))
@@ -1090,7 +1095,7 @@
err = 0;
if (!list_empty(&pagelist)) {
err = migrate_pages(&pagelist, new_page_node,
- (unsigned long)pm, 0);
+ (unsigned long)pm, 0, true);
if (err)
putback_lru_pages(&pagelist);
}
diff --git a/mm/mincore.c b/mm/mincore.c
index 9ac42dc..a4e6b9d 100644
--- a/mm/mincore.c
+++ b/mm/mincore.c
@@ -154,6 +154,13 @@
pmd = pmd_offset(pud, addr);
do {
next = pmd_addr_end(addr, end);
+ if (pmd_trans_huge(*pmd)) {
+ if (mincore_huge_pmd(vma, pmd, addr, next, vec)) {
+ vec += (next - addr) >> PAGE_SHIFT;
+ continue;
+ }
+ /* fall through */
+ }
if (pmd_none_or_clear_bad(pmd))
mincore_unmapped_range(vma, addr, next, vec);
else
diff --git a/mm/mlock.c b/mm/mlock.c
index b70919c..13e81ee 100644
--- a/mm/mlock.c
+++ b/mm/mlock.c
@@ -155,13 +155,12 @@
* vma->vm_mm->mmap_sem must be held for at least read.
*/
static long __mlock_vma_pages_range(struct vm_area_struct *vma,
- unsigned long start, unsigned long end)
+ unsigned long start, unsigned long end,
+ int *nonblocking)
{
struct mm_struct *mm = vma->vm_mm;
unsigned long addr = start;
- struct page *pages[16]; /* 16 gives a reasonable batch */
int nr_pages = (end - start) / PAGE_SIZE;
- int ret = 0;
int gup_flags;
VM_BUG_ON(start & ~PAGE_MASK);
@@ -170,73 +169,26 @@
VM_BUG_ON(end > vma->vm_end);
VM_BUG_ON(!rwsem_is_locked(&mm->mmap_sem));
- gup_flags = FOLL_TOUCH | FOLL_GET;
- if (vma->vm_flags & VM_WRITE)
+ gup_flags = FOLL_TOUCH;
+ /*
+ * We want to touch writable mappings with a write fault in order
+ * to break COW, except for shared mappings because these don't COW
+ * and we would not want to dirty them for nothing.
+ */
+ if ((vma->vm_flags & (VM_WRITE | VM_SHARED)) == VM_WRITE)
gup_flags |= FOLL_WRITE;
+ if (vma->vm_flags & VM_LOCKED)
+ gup_flags |= FOLL_MLOCK;
+
/* We don't try to access the guard page of a stack vma */
if (stack_guard_page(vma, start)) {
addr += PAGE_SIZE;
nr_pages--;
}
- while (nr_pages > 0) {
- int i;
-
- cond_resched();
-
- /*
- * get_user_pages makes pages present if we are
- * setting mlock. and this extra reference count will
- * disable migration of this page. However, page may
- * still be truncated out from under us.
- */
- ret = __get_user_pages(current, mm, addr,
- min_t(int, nr_pages, ARRAY_SIZE(pages)),
- gup_flags, pages, NULL);
- /*
- * This can happen for, e.g., VM_NONLINEAR regions before
- * a page has been allocated and mapped at a given offset,
- * or for addresses that map beyond end of a file.
- * We'll mlock the pages if/when they get faulted in.
- */
- if (ret < 0)
- break;
-
- lru_add_drain(); /* push cached pages to LRU */
-
- for (i = 0; i < ret; i++) {
- struct page *page = pages[i];
-
- if (page->mapping) {
- /*
- * That preliminary check is mainly to avoid
- * the pointless overhead of lock_page on the
- * ZERO_PAGE: which might bounce very badly if
- * there is contention. However, we're still
- * dirtying its cacheline with get/put_page:
- * we'll add another __get_user_pages flag to
- * avoid it if that case turns out to matter.
- */
- lock_page(page);
- /*
- * Because we lock page here and migration is
- * blocked by the elevated reference, we need
- * only check for file-cache page truncation.
- */
- if (page->mapping)
- mlock_vma_page(page);
- unlock_page(page);
- }
- put_page(page); /* ref from get_user_pages() */
- }
-
- addr += ret * PAGE_SIZE;
- nr_pages -= ret;
- ret = 0;
- }
-
- return ret; /* 0 or negative error code */
+ return __get_user_pages(current, mm, addr, nr_pages, gup_flags,
+ NULL, NULL, nonblocking);
}
/*
@@ -280,7 +232,7 @@
is_vm_hugetlb_page(vma) ||
vma == get_gate_vma(current))) {
- __mlock_vma_pages_range(vma, start, end);
+ __mlock_vma_pages_range(vma, start, end, NULL);
/* Hide errors from mmap() and other callers */
return 0;
@@ -372,18 +324,10 @@
int ret = 0;
int lock = newflags & VM_LOCKED;
- if (newflags == vma->vm_flags ||
- (vma->vm_flags & (VM_IO | VM_PFNMAP)))
+ if (newflags == vma->vm_flags || (vma->vm_flags & VM_SPECIAL) ||
+ is_vm_hugetlb_page(vma) || vma == get_gate_vma(current))
goto out; /* don't set VM_LOCKED, don't count */
- if ((vma->vm_flags & (VM_DONTEXPAND | VM_RESERVED)) ||
- is_vm_hugetlb_page(vma) ||
- vma == get_gate_vma(current)) {
- if (lock)
- make_pages_present(start, end);
- goto out; /* don't set VM_LOCKED, don't count */
- }
-
pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT);
*prev = vma_merge(mm, *prev, start, end, newflags, vma->anon_vma,
vma->vm_file, pgoff, vma_policy(vma));
@@ -419,14 +363,10 @@
* set VM_LOCKED, __mlock_vma_pages_range will bring it back.
*/
- if (lock) {
+ if (lock)
vma->vm_flags = newflags;
- ret = __mlock_vma_pages_range(vma, start, end);
- if (ret < 0)
- ret = __mlock_posix_error_return(ret);
- } else {
+ else
munlock_vma_pages_range(vma, start, end);
- }
out:
*prev = vma;
@@ -439,7 +379,8 @@
struct vm_area_struct * vma, * prev;
int error;
- len = PAGE_ALIGN(len);
+ VM_BUG_ON(start & ~PAGE_MASK);
+ VM_BUG_ON(len != PAGE_ALIGN(len));
end = start + len;
if (end < start)
return -EINVAL;
@@ -482,6 +423,62 @@
return error;
}
+static int do_mlock_pages(unsigned long start, size_t len, int ignore_errors)
+{
+ struct mm_struct *mm = current->mm;
+ unsigned long end, nstart, nend;
+ struct vm_area_struct *vma = NULL;
+ int locked = 0;
+ int ret = 0;
+
+ VM_BUG_ON(start & ~PAGE_MASK);
+ VM_BUG_ON(len != PAGE_ALIGN(len));
+ end = start + len;
+
+ for (nstart = start; nstart < end; nstart = nend) {
+ /*
+ * We want to fault in pages for [nstart; end) address range.
+ * Find first corresponding VMA.
+ */
+ if (!locked) {
+ locked = 1;
+ down_read(&mm->mmap_sem);
+ vma = find_vma(mm, nstart);
+ } else if (nstart >= vma->vm_end)
+ vma = vma->vm_next;
+ if (!vma || vma->vm_start >= end)
+ break;
+ /*
+ * Set [nstart; nend) to intersection of desired address
+ * range with the first VMA. Also, skip undesirable VMA types.
+ */
+ nend = min(end, vma->vm_end);
+ if (vma->vm_flags & (VM_IO | VM_PFNMAP))
+ continue;
+ if (nstart < vma->vm_start)
+ nstart = vma->vm_start;
+ /*
+ * Now fault in a range of pages. __mlock_vma_pages_range()
+ * double checks the vma flags, so that it won't mlock pages
+ * if the vma was already munlocked.
+ */
+ ret = __mlock_vma_pages_range(vma, nstart, nend, &locked);
+ if (ret < 0) {
+ if (ignore_errors) {
+ ret = 0;
+ continue; /* continue at next VMA */
+ }
+ ret = __mlock_posix_error_return(ret);
+ break;
+ }
+ nend = nstart + ret * PAGE_SIZE;
+ ret = 0;
+ }
+ if (locked)
+ up_read(&mm->mmap_sem);
+ return ret; /* 0 or negative error code */
+}
+
SYSCALL_DEFINE2(mlock, unsigned long, start, size_t, len)
{
unsigned long locked;
@@ -507,6 +504,8 @@
if ((locked <= lock_limit) || capable(CAP_IPC_LOCK))
error = do_mlock(start, len, 1);
up_write(¤t->mm->mmap_sem);
+ if (!error)
+ error = do_mlock_pages(start, len, 0);
return error;
}
@@ -571,6 +570,10 @@
capable(CAP_IPC_LOCK))
ret = do_mlockall(flags);
up_write(¤t->mm->mmap_sem);
+ if (!ret && (flags & MCL_CURRENT)) {
+ /* Ignore errors */
+ do_mlock_pages(0, TASK_SIZE, 1);
+ }
out:
return ret;
}
diff --git a/mm/mmap.c b/mm/mmap.c
index 50a4aa0..2ec8eb5 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -29,6 +29,7 @@
#include <linux/mmu_notifier.h>
#include <linux/perf_event.h>
#include <linux/audit.h>
+#include <linux/khugepaged.h>
#include <asm/uaccess.h>
#include <asm/cacheflush.h>
@@ -253,7 +254,15 @@
down_write(&mm->mmap_sem);
#ifdef CONFIG_COMPAT_BRK
- min_brk = mm->end_code;
+ /*
+ * CONFIG_COMPAT_BRK can still be overridden by setting
+ * randomize_va_space to 2, which will still cause mm->start_brk
+ * to be arbitrarily shifted
+ */
+ if (mm->start_brk > PAGE_ALIGN(mm->end_data))
+ min_brk = mm->start_brk;
+ else
+ min_brk = mm->end_data;
#else
min_brk = mm->start_brk;
#endif
@@ -588,6 +597,8 @@
}
}
+ vma_adjust_trans_huge(vma, start, end, adjust_next);
+
/*
* When changing only vma->vm_end, we don't really need anon_vma
* lock. This is a fairly rare case by itself, but the anon_vma
@@ -815,6 +826,7 @@
end, prev->vm_pgoff, NULL);
if (err)
return NULL;
+ khugepaged_enter_vma_merge(prev);
return prev;
}
@@ -833,6 +845,7 @@
next->vm_pgoff - pglen, NULL);
if (err)
return NULL;
+ khugepaged_enter_vma_merge(area);
return area;
}
@@ -1761,6 +1774,7 @@
}
}
vma_unlock_anon_vma(vma);
+ khugepaged_enter_vma_merge(vma);
return error;
}
#endif /* CONFIG_STACK_GROWSUP || CONFIG_IA64 */
@@ -1808,6 +1822,7 @@
}
}
vma_unlock_anon_vma(vma);
+ khugepaged_enter_vma_merge(vma);
return error;
}
diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
index 438951d..8d032de 100644
--- a/mm/mmu_notifier.c
+++ b/mm/mmu_notifier.c
@@ -100,6 +100,26 @@
return young;
}
+int __mmu_notifier_test_young(struct mm_struct *mm,
+ unsigned long address)
+{
+ struct mmu_notifier *mn;
+ struct hlist_node *n;
+ int young = 0;
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(mn, n, &mm->mmu_notifier_mm->list, hlist) {
+ if (mn->ops->test_young) {
+ young = mn->ops->test_young(mn, mm, address);
+ if (young)
+ break;
+ }
+ }
+ rcu_read_unlock();
+
+ return young;
+}
+
void __mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address,
pte_t pte)
{
diff --git a/mm/mmzone.c b/mm/mmzone.c
index e35bfb8..f5b7d17 100644
--- a/mm/mmzone.c
+++ b/mm/mmzone.c
@@ -87,24 +87,3 @@
return 1;
}
#endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */
-
-#ifdef CONFIG_SMP
-/* Called when a more accurate view of NR_FREE_PAGES is needed */
-unsigned long zone_nr_free_pages(struct zone *zone)
-{
- unsigned long nr_free_pages = zone_page_state(zone, NR_FREE_PAGES);
-
- /*
- * While kswapd is awake, it is considered the zone is under some
- * memory pressure. Under pressure, there is a risk that
- * per-cpu-counter-drift will allow the min watermark to be breached
- * potentially causing a live-lock. While kswapd is awake and
- * free pages are low, get a better estimate for free pages
- */
- if (nr_free_pages < zone->percpu_drift_mark &&
- !waitqueue_active(&zone->zone_pgdat->kswapd_wait))
- return zone_page_state_snapshot(zone, NR_FREE_PAGES);
-
- return nr_free_pages;
-}
-#endif /* CONFIG_SMP */
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 4c51338..5a688a2 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -78,7 +78,7 @@
pte_unmap_unlock(pte - 1, ptl);
}
-static inline void change_pmd_range(struct mm_struct *mm, pud_t *pud,
+static inline void change_pmd_range(struct vm_area_struct *vma, pud_t *pud,
unsigned long addr, unsigned long end, pgprot_t newprot,
int dirty_accountable)
{
@@ -88,13 +88,21 @@
pmd = pmd_offset(pud, addr);
do {
next = pmd_addr_end(addr, end);
+ if (pmd_trans_huge(*pmd)) {
+ if (next - addr != HPAGE_PMD_SIZE)
+ split_huge_page_pmd(vma->vm_mm, pmd);
+ else if (change_huge_pmd(vma, pmd, addr, newprot))
+ continue;
+ /* fall through */
+ }
if (pmd_none_or_clear_bad(pmd))
continue;
- change_pte_range(mm, pmd, addr, next, newprot, dirty_accountable);
+ change_pte_range(vma->vm_mm, pmd, addr, next, newprot,
+ dirty_accountable);
} while (pmd++, addr = next, addr != end);
}
-static inline void change_pud_range(struct mm_struct *mm, pgd_t *pgd,
+static inline void change_pud_range(struct vm_area_struct *vma, pgd_t *pgd,
unsigned long addr, unsigned long end, pgprot_t newprot,
int dirty_accountable)
{
@@ -106,7 +114,8 @@
next = pud_addr_end(addr, end);
if (pud_none_or_clear_bad(pud))
continue;
- change_pmd_range(mm, pud, addr, next, newprot, dirty_accountable);
+ change_pmd_range(vma, pud, addr, next, newprot,
+ dirty_accountable);
} while (pud++, addr = next, addr != end);
}
@@ -126,7 +135,8 @@
next = pgd_addr_end(addr, end);
if (pgd_none_or_clear_bad(pgd))
continue;
- change_pud_range(mm, pgd, addr, next, newprot, dirty_accountable);
+ change_pud_range(vma, pgd, addr, next, newprot,
+ dirty_accountable);
} while (pgd++, addr = next, addr != end);
flush_tlb_range(vma, start, end);
}
diff --git a/mm/mremap.c b/mm/mremap.c
index 563fbdd..9925b63 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -41,13 +41,15 @@
return NULL;
pmd = pmd_offset(pud, addr);
+ split_huge_page_pmd(mm, pmd);
if (pmd_none_or_clear_bad(pmd))
return NULL;
return pmd;
}
-static pmd_t *alloc_new_pmd(struct mm_struct *mm, unsigned long addr)
+static pmd_t *alloc_new_pmd(struct mm_struct *mm, struct vm_area_struct *vma,
+ unsigned long addr)
{
pgd_t *pgd;
pud_t *pud;
@@ -62,7 +64,8 @@
if (!pmd)
return NULL;
- if (!pmd_present(*pmd) && __pte_alloc(mm, pmd, addr))
+ VM_BUG_ON(pmd_trans_huge(*pmd));
+ if (pmd_none(*pmd) && __pte_alloc(mm, vma, pmd, addr))
return NULL;
return pmd;
@@ -147,7 +150,7 @@
old_pmd = get_old_pmd(vma->vm_mm, old_addr);
if (!old_pmd)
continue;
- new_pmd = alloc_new_pmd(vma->vm_mm, new_addr);
+ new_pmd = alloc_new_pmd(vma->vm_mm, vma, new_addr);
if (!new_pmd)
break;
next = (new_addr + PMD_SIZE) & PMD_MASK;
diff --git a/mm/nommu.c b/mm/nommu.c
index ef4045d..f59e142 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -127,7 +127,8 @@
int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, int nr_pages, unsigned int foll_flags,
- struct page **pages, struct vm_area_struct **vmas)
+ struct page **pages, struct vm_area_struct **vmas,
+ int *retry)
{
struct vm_area_struct *vma;
unsigned long vm_flags;
@@ -185,7 +186,8 @@
if (force)
flags |= FOLL_FORCE;
- return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas);
+ return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas,
+ NULL);
}
EXPORT_SYMBOL(get_user_pages);
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index b5d8a1f..2cb01f6 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -410,9 +410,12 @@
{
unsigned long background;
unsigned long dirty;
- unsigned long available_memory = determine_dirtyable_memory();
+ unsigned long uninitialized_var(available_memory);
struct task_struct *tsk;
+ if (!vm_dirty_bytes || !dirty_background_bytes)
+ available_memory = determine_dirtyable_memory();
+
if (vm_dirty_bytes)
dirty = DIV_ROUND_UP(vm_dirty_bytes, PAGE_SIZE);
else
@@ -1103,7 +1106,7 @@
int __set_page_dirty_no_writeback(struct page *page)
{
if (!PageDirty(page))
- SetPageDirty(page);
+ return !TestSetPageDirty(page);
return 0;
}
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 826ba69..90c1439 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -357,6 +357,7 @@
}
}
+/* update __split_huge_page_refcount if you change this function */
static int destroy_compound_page(struct page *page, unsigned long order)
{
int i;
@@ -426,18 +427,10 @@
*
* Assumption: *_mem_map is contiguous at least up to MAX_ORDER
*/
-static inline struct page *
-__page_find_buddy(struct page *page, unsigned long page_idx, unsigned int order)
-{
- unsigned long buddy_idx = page_idx ^ (1 << order);
-
- return page + (buddy_idx - page_idx);
-}
-
static inline unsigned long
-__find_combined_index(unsigned long page_idx, unsigned int order)
+__find_buddy_index(unsigned long page_idx, unsigned int order)
{
- return (page_idx & ~(1 << order));
+ return page_idx ^ (1 << order);
}
/*
@@ -448,8 +441,8 @@
* (c) a page and its buddy have the same order &&
* (d) a page and its buddy are in the same zone.
*
- * For recording whether a page is in the buddy system, we use PG_buddy.
- * Setting, clearing, and testing PG_buddy is serialized by zone->lock.
+ * For recording whether a page is in the buddy system, we set ->_mapcount -2.
+ * Setting, clearing, and testing _mapcount -2 is serialized by zone->lock.
*
* For recording page's order, we use page_private(page).
*/
@@ -482,7 +475,7 @@
* as necessary, plus some accounting needed to play nicely with other
* parts of the VM system.
* At each level, we keep a list of pages, which are heads of continuous
- * free pages of length of (1 << order) and marked with PG_buddy. Page's
+ * free pages of length of (1 << order) and marked with _mapcount -2. Page's
* order is recorded in page_private(page) field.
* So when we are allocating or freeing one, we can derive the state of the
* other. That is, if we allocate a small block, and both were
@@ -499,6 +492,7 @@
{
unsigned long page_idx;
unsigned long combined_idx;
+ unsigned long uninitialized_var(buddy_idx);
struct page *buddy;
if (unlikely(PageCompound(page)))
@@ -513,7 +507,8 @@
VM_BUG_ON(bad_range(zone, page));
while (order < MAX_ORDER-1) {
- buddy = __page_find_buddy(page, page_idx, order);
+ buddy_idx = __find_buddy_index(page_idx, order);
+ buddy = page + (buddy_idx - page_idx);
if (!page_is_buddy(page, buddy, order))
break;
@@ -521,7 +516,7 @@
list_del(&buddy->lru);
zone->free_area[order].nr_free--;
rmv_page_order(buddy);
- combined_idx = __find_combined_index(page_idx, order);
+ combined_idx = buddy_idx & page_idx;
page = page + (combined_idx - page_idx);
page_idx = combined_idx;
order++;
@@ -538,9 +533,10 @@
*/
if ((order < MAX_ORDER-2) && pfn_valid_within(page_to_pfn(buddy))) {
struct page *higher_page, *higher_buddy;
- combined_idx = __find_combined_index(page_idx, order);
- higher_page = page + combined_idx - page_idx;
- higher_buddy = __page_find_buddy(higher_page, combined_idx, order + 1);
+ combined_idx = buddy_idx & page_idx;
+ higher_page = page + (combined_idx - page_idx);
+ buddy_idx = __find_buddy_index(combined_idx, order + 1);
+ higher_buddy = page + (buddy_idx - combined_idx);
if (page_is_buddy(higher_page, higher_buddy, order + 1)) {
list_add_tail(&page->lru,
&zone->free_area[order].free_list[migratetype]);
@@ -651,13 +647,10 @@
trace_mm_page_free_direct(page, order);
kmemcheck_free_shadow(page, order);
- for (i = 0; i < (1 << order); i++) {
- struct page *pg = page + i;
-
- if (PageAnon(pg))
- pg->mapping = NULL;
- bad += free_pages_check(pg);
- }
+ if (PageAnon(page))
+ page->mapping = NULL;
+ for (i = 0; i < (1 << order); i++)
+ bad += free_pages_check(page + i);
if (bad)
return false;
@@ -1460,24 +1453,24 @@
#endif /* CONFIG_FAIL_PAGE_ALLOC */
/*
- * Return 1 if free pages are above 'mark'. This takes into account the order
+ * Return true if free pages are above 'mark'. This takes into account the order
* of the allocation.
*/
-int zone_watermark_ok(struct zone *z, int order, unsigned long mark,
- int classzone_idx, int alloc_flags)
+static bool __zone_watermark_ok(struct zone *z, int order, unsigned long mark,
+ int classzone_idx, int alloc_flags, long free_pages)
{
/* free_pages my go negative - that's OK */
long min = mark;
- long free_pages = zone_nr_free_pages(z) - (1 << order) + 1;
int o;
+ free_pages -= (1 << order) + 1;
if (alloc_flags & ALLOC_HIGH)
min -= min / 2;
if (alloc_flags & ALLOC_HARDER)
min -= min / 4;
if (free_pages <= min + z->lowmem_reserve[classzone_idx])
- return 0;
+ return false;
for (o = 0; o < order; o++) {
/* At the next order, this order's pages become unavailable */
free_pages -= z->free_area[o].nr_free << o;
@@ -1486,9 +1479,28 @@
min >>= 1;
if (free_pages <= min)
- return 0;
+ return false;
}
- return 1;
+ return true;
+}
+
+bool zone_watermark_ok(struct zone *z, int order, unsigned long mark,
+ int classzone_idx, int alloc_flags)
+{
+ return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags,
+ zone_page_state(z, NR_FREE_PAGES));
+}
+
+bool zone_watermark_ok_safe(struct zone *z, int order, unsigned long mark,
+ int classzone_idx, int alloc_flags)
+{
+ long free_pages = zone_page_state(z, NR_FREE_PAGES);
+
+ if (z->percpu_drift_mark && free_pages < z->percpu_drift_mark)
+ free_pages = zone_page_state_snapshot(z, NR_FREE_PAGES);
+
+ return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags,
+ free_pages);
}
#ifdef CONFIG_NUMA
@@ -1793,15 +1805,18 @@
__alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
struct zonelist *zonelist, enum zone_type high_zoneidx,
nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone,
- int migratetype, unsigned long *did_some_progress)
+ int migratetype, unsigned long *did_some_progress,
+ bool sync_migration)
{
struct page *page;
if (!order || compaction_deferred(preferred_zone))
return NULL;
+ current->flags |= PF_MEMALLOC;
*did_some_progress = try_to_compact_pages(zonelist, order, gfp_mask,
- nodemask);
+ nodemask, sync_migration);
+ current->flags &= ~PF_MEMALLOC;
if (*did_some_progress != COMPACT_SKIPPED) {
/* Page migration frees to the PCP lists but we want merging */
@@ -1837,7 +1852,8 @@
__alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
struct zonelist *zonelist, enum zone_type high_zoneidx,
nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone,
- int migratetype, unsigned long *did_some_progress)
+ int migratetype, unsigned long *did_some_progress,
+ bool sync_migration)
{
return NULL;
}
@@ -1852,23 +1868,22 @@
{
struct page *page = NULL;
struct reclaim_state reclaim_state;
- struct task_struct *p = current;
bool drained = false;
cond_resched();
/* We now go into synchronous reclaim */
cpuset_memory_pressure_bump();
- p->flags |= PF_MEMALLOC;
+ current->flags |= PF_MEMALLOC;
lockdep_set_current_reclaim_state(gfp_mask);
reclaim_state.reclaimed_slab = 0;
- p->reclaim_state = &reclaim_state;
+ current->reclaim_state = &reclaim_state;
*did_some_progress = try_to_free_pages(zonelist, order, gfp_mask, nodemask);
- p->reclaim_state = NULL;
+ current->reclaim_state = NULL;
lockdep_clear_current_reclaim_state();
- p->flags &= ~PF_MEMALLOC;
+ current->flags &= ~PF_MEMALLOC;
cond_resched();
@@ -1920,19 +1935,19 @@
static inline
void wake_all_kswapd(unsigned int order, struct zonelist *zonelist,
- enum zone_type high_zoneidx)
+ enum zone_type high_zoneidx,
+ enum zone_type classzone_idx)
{
struct zoneref *z;
struct zone *zone;
for_each_zone_zonelist(zone, z, zonelist, high_zoneidx)
- wakeup_kswapd(zone, order);
+ wakeup_kswapd(zone, order, classzone_idx);
}
static inline int
gfp_to_alloc_flags(gfp_t gfp_mask)
{
- struct task_struct *p = current;
int alloc_flags = ALLOC_WMARK_MIN | ALLOC_CPUSET;
const gfp_t wait = gfp_mask & __GFP_WAIT;
@@ -1948,18 +1963,23 @@
alloc_flags |= (__force int) (gfp_mask & __GFP_HIGH);
if (!wait) {
- alloc_flags |= ALLOC_HARDER;
+ /*
+ * Not worth trying to allocate harder for
+ * __GFP_NOMEMALLOC even if it can't schedule.
+ */
+ if (!(gfp_mask & __GFP_NOMEMALLOC))
+ alloc_flags |= ALLOC_HARDER;
/*
* Ignore cpuset if GFP_ATOMIC (!wait) rather than fail alloc.
* See also cpuset_zone_allowed() comment in kernel/cpuset.c.
*/
alloc_flags &= ~ALLOC_CPUSET;
- } else if (unlikely(rt_task(p)) && !in_interrupt())
+ } else if (unlikely(rt_task(current)) && !in_interrupt())
alloc_flags |= ALLOC_HARDER;
if (likely(!(gfp_mask & __GFP_NOMEMALLOC))) {
if (!in_interrupt() &&
- ((p->flags & PF_MEMALLOC) ||
+ ((current->flags & PF_MEMALLOC) ||
unlikely(test_thread_flag(TIF_MEMDIE))))
alloc_flags |= ALLOC_NO_WATERMARKS;
}
@@ -1978,7 +1998,7 @@
int alloc_flags;
unsigned long pages_reclaimed = 0;
unsigned long did_some_progress;
- struct task_struct *p = current;
+ bool sync_migration = false;
/*
* In the slowpath, we sanity check order to avoid ever trying to
@@ -2003,7 +2023,9 @@
goto nopage;
restart:
- wake_all_kswapd(order, zonelist, high_zoneidx);
+ if (!(gfp_mask & __GFP_NO_KSWAPD))
+ wake_all_kswapd(order, zonelist, high_zoneidx,
+ zone_idx(preferred_zone));
/*
* OK, we're below the kswapd watermark and have kicked background
@@ -2034,21 +2056,26 @@
goto nopage;
/* Avoid recursion of direct reclaim */
- if (p->flags & PF_MEMALLOC)
+ if (current->flags & PF_MEMALLOC)
goto nopage;
/* Avoid allocations with no watermarks from looping endlessly */
if (test_thread_flag(TIF_MEMDIE) && !(gfp_mask & __GFP_NOFAIL))
goto nopage;
- /* Try direct compaction */
+ /*
+ * Try direct compaction. The first pass is asynchronous. Subsequent
+ * attempts after direct reclaim are synchronous
+ */
page = __alloc_pages_direct_compact(gfp_mask, order,
zonelist, high_zoneidx,
nodemask,
alloc_flags, preferred_zone,
- migratetype, &did_some_progress);
+ migratetype, &did_some_progress,
+ sync_migration);
if (page)
goto got_pg;
+ sync_migration = true;
/* Try direct reclaim and then allocating */
page = __alloc_pages_direct_reclaim(gfp_mask, order,
@@ -2102,13 +2129,27 @@
/* Wait for some write requests to complete then retry */
wait_iff_congested(preferred_zone, BLK_RW_ASYNC, HZ/50);
goto rebalance;
+ } else {
+ /*
+ * High-order allocations do not necessarily loop after
+ * direct reclaim and reclaim/compaction depends on compaction
+ * being called after reclaim so call directly if necessary
+ */
+ page = __alloc_pages_direct_compact(gfp_mask, order,
+ zonelist, high_zoneidx,
+ nodemask,
+ alloc_flags, preferred_zone,
+ migratetype, &did_some_progress,
+ sync_migration);
+ if (page)
+ goto got_pg;
}
nopage:
if (!(gfp_mask & __GFP_NOWARN) && printk_ratelimit()) {
printk(KERN_WARNING "%s: page allocation failure."
" order:%d, mode:0x%x\n",
- p->comm, order, gfp_mask);
+ current->comm, order, gfp_mask);
dump_stack();
show_mem();
}
@@ -2442,7 +2483,7 @@
" all_unreclaimable? %s"
"\n",
zone->name,
- K(zone_nr_free_pages(zone)),
+ K(zone_page_state(zone, NR_FREE_PAGES)),
K(min_wmark_pages(zone)),
K(low_wmark_pages(zone)),
K(high_wmark_pages(zone)),
@@ -2585,9 +2626,16 @@
static __init int setup_numa_zonelist_order(char *s)
{
- if (s)
- return __parse_numa_zonelist_order(s);
- return 0;
+ int ret;
+
+ if (!s)
+ return 0;
+
+ ret = __parse_numa_zonelist_order(s);
+ if (ret == 0)
+ strlcpy(numa_zonelist_order, s, NUMA_ZONELIST_ORDER_LEN);
+
+ return ret;
}
early_param("numa_zonelist_order", setup_numa_zonelist_order);
@@ -5517,7 +5565,6 @@
{1UL << PG_swapcache, "swapcache" },
{1UL << PG_mappedtodisk, "mappedtodisk" },
{1UL << PG_reclaim, "reclaim" },
- {1UL << PG_buddy, "buddy" },
{1UL << PG_swapbacked, "swapbacked" },
{1UL << PG_unevictable, "unevictable" },
#ifdef CONFIG_MMU
@@ -5565,7 +5612,7 @@
{
printk(KERN_ALERT
"page:%p count:%d mapcount:%d mapping:%p index:%#lx\n",
- page, page_count(page), page_mapcount(page),
+ page, atomic_read(&page->_count), page_mapcount(page),
page->mapping, page->index);
dump_page_flags(page->flags);
}
diff --git a/mm/pagewalk.c b/mm/pagewalk.c
index 38cc58b..7cfa6ae 100644
--- a/mm/pagewalk.c
+++ b/mm/pagewalk.c
@@ -34,6 +34,7 @@
pmd = pmd_offset(pud, addr);
do {
next = pmd_addr_end(addr, end);
+ split_huge_page_pmd(walk->mm, pmd);
if (pmd_none_or_clear_bad(pmd)) {
if (walk->pte_hole)
err = walk->pte_hole(addr, next, walk);
diff --git a/mm/percpu-vm.c b/mm/percpu-vm.c
index 7d9c1d0..ea53496 100644
--- a/mm/percpu-vm.c
+++ b/mm/percpu-vm.c
@@ -421,7 +421,7 @@
return NULL;
vms = pcpu_get_vm_areas(pcpu_group_offsets, pcpu_group_sizes,
- pcpu_nr_groups, pcpu_atom_size, GFP_KERNEL);
+ pcpu_nr_groups, pcpu_atom_size);
if (!vms) {
pcpu_free_chunk(chunk);
return NULL;
diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
new file mode 100644
index 0000000..d030548
--- /dev/null
+++ b/mm/pgtable-generic.c
@@ -0,0 +1,123 @@
+/*
+ * mm/pgtable-generic.c
+ *
+ * Generic pgtable methods declared in asm-generic/pgtable.h
+ *
+ * Copyright (C) 2010 Linus Torvalds
+ */
+
+#include <asm/tlb.h>
+#include <asm-generic/pgtable.h>
+
+#ifndef __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
+/*
+ * Only sets the access flags (dirty, accessed, and
+ * writable). Furthermore, we know it always gets set to a "more
+ * permissive" setting, which allows most architectures to optimize
+ * this. We return whether the PTE actually changed, which in turn
+ * instructs the caller to do things like update__mmu_cache. This
+ * used to be done in the caller, but sparc needs minor faults to
+ * force that call on sun4c so we changed this macro slightly
+ */
+int ptep_set_access_flags(struct vm_area_struct *vma,
+ unsigned long address, pte_t *ptep,
+ pte_t entry, int dirty)
+{
+ int changed = !pte_same(*ptep, entry);
+ if (changed) {
+ set_pte_at(vma->vm_mm, address, ptep, entry);
+ flush_tlb_page(vma, address);
+ }
+ return changed;
+}
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS
+int pmdp_set_access_flags(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp,
+ pmd_t entry, int dirty)
+{
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ int changed = !pmd_same(*pmdp, entry);
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+ if (changed) {
+ set_pmd_at(vma->vm_mm, address, pmdp, entry);
+ flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+ }
+ return changed;
+#else /* CONFIG_TRANSPARENT_HUGEPAGE */
+ BUG();
+ return 0;
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+}
+#endif
+
+#ifndef __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH
+int ptep_clear_flush_young(struct vm_area_struct *vma,
+ unsigned long address, pte_t *ptep)
+{
+ int young;
+ young = ptep_test_and_clear_young(vma, address, ptep);
+ if (young)
+ flush_tlb_page(vma, address);
+ return young;
+}
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH
+int pmdp_clear_flush_young(struct vm_area_struct *vma,
+ unsigned long address, pmd_t *pmdp)
+{
+ int young;
+#ifndef CONFIG_TRANSPARENT_HUGEPAGE
+ BUG();
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+ young = pmdp_test_and_clear_young(vma, address, pmdp);
+ if (young)
+ flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+ return young;
+}
+#endif
+
+#ifndef __HAVE_ARCH_PTEP_CLEAR_FLUSH
+pte_t ptep_clear_flush(struct vm_area_struct *vma, unsigned long address,
+ pte_t *ptep)
+{
+ pte_t pte;
+ pte = ptep_get_and_clear((vma)->vm_mm, address, ptep);
+ flush_tlb_page(vma, address);
+ return pte;
+}
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_CLEAR_FLUSH
+pmd_t pmdp_clear_flush(struct vm_area_struct *vma, unsigned long address,
+ pmd_t *pmdp)
+{
+ pmd_t pmd;
+#ifndef CONFIG_TRANSPARENT_HUGEPAGE
+ BUG();
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+ pmd = pmdp_get_and_clear(vma->vm_mm, address, pmdp);
+ flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+ return pmd;
+}
+#endif
+
+#ifndef __HAVE_ARCH_PMDP_SPLITTING_FLUSH
+pmd_t pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address,
+ pmd_t *pmdp)
+{
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+ pmd_t pmd = pmd_mksplitting(*pmdp);
+ VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+ set_pmd_at(vma->vm_mm, address, pmdp, pmd);
+ /* tlb flush only to serialize against gup-fast */
+ flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+#else /* CONFIG_TRANSPARENT_HUGEPAGE */
+ BUG();
+#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+}
+#endif
diff --git a/mm/rmap.c b/mm/rmap.c
index c95d2ba..f21f4a1 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -177,6 +177,10 @@
list_add(&avc->same_vma, &vma->anon_vma_chain);
anon_vma_lock(anon_vma);
+ /*
+ * It's critical to add new vmas to the tail of the anon_vma,
+ * see comment in huge_memory.c:__split_huge_page().
+ */
list_add_tail(&avc->same_anon_vma, &anon_vma->head);
anon_vma_unlock(anon_vma);
}
@@ -360,7 +364,7 @@
* Returns virtual address or -EFAULT if page's index/offset is not
* within the range mapped the @vma.
*/
-static inline unsigned long
+inline unsigned long
vma_address(struct page *page, struct vm_area_struct *vma)
{
pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
@@ -435,6 +439,8 @@
pmd = pmd_offset(pud, address);
if (!pmd_present(*pmd))
return NULL;
+ if (pmd_trans_huge(*pmd))
+ return NULL;
pte = pte_offset_map(pmd, address);
/* Make a quick check before getting the lock */
@@ -489,35 +495,17 @@
unsigned long *vm_flags)
{
struct mm_struct *mm = vma->vm_mm;
- pte_t *pte;
- spinlock_t *ptl;
int referenced = 0;
- pte = page_check_address(page, mm, address, &ptl, 0);
- if (!pte)
- goto out;
-
/*
* Don't want to elevate referenced for mlocked page that gets this far,
* in order that it progresses to try_to_unmap and is moved to the
* unevictable list.
*/
if (vma->vm_flags & VM_LOCKED) {
- *mapcount = 1; /* break early from loop */
+ *mapcount = 0; /* break early from loop */
*vm_flags |= VM_LOCKED;
- goto out_unmap;
- }
-
- if (ptep_clear_flush_young_notify(vma, address, pte)) {
- /*
- * Don't treat a reference through a sequentially read
- * mapping as such. If the page has been used in
- * another mapping, we will catch it; if this other
- * mapping is already gone, the unmap path will have
- * set PG_referenced or activated the page.
- */
- if (likely(!VM_SequentialReadHint(vma)))
- referenced++;
+ goto out;
}
/* Pretend the page is referenced if the task has the
@@ -526,9 +514,39 @@
rwsem_is_locked(&mm->mmap_sem))
referenced++;
-out_unmap:
+ if (unlikely(PageTransHuge(page))) {
+ pmd_t *pmd;
+
+ spin_lock(&mm->page_table_lock);
+ pmd = page_check_address_pmd(page, mm, address,
+ PAGE_CHECK_ADDRESS_PMD_FLAG);
+ if (pmd && !pmd_trans_splitting(*pmd) &&
+ pmdp_clear_flush_young_notify(vma, address, pmd))
+ referenced++;
+ spin_unlock(&mm->page_table_lock);
+ } else {
+ pte_t *pte;
+ spinlock_t *ptl;
+
+ pte = page_check_address(page, mm, address, &ptl, 0);
+ if (!pte)
+ goto out;
+
+ if (ptep_clear_flush_young_notify(vma, address, pte)) {
+ /*
+ * Don't treat a reference through a sequentially read
+ * mapping as such. If the page has been used in
+ * another mapping, we will catch it; if this other
+ * mapping is already gone, the unmap path will have
+ * set PG_referenced or activated the page.
+ */
+ if (likely(!VM_SequentialReadHint(vma)))
+ referenced++;
+ }
+ pte_unmap_unlock(pte, ptl);
+ }
+
(*mapcount)--;
- pte_unmap_unlock(pte, ptl);
if (referenced)
*vm_flags |= vma->vm_flags;
@@ -864,8 +882,13 @@
struct vm_area_struct *vma, unsigned long address, int exclusive)
{
int first = atomic_inc_and_test(&page->_mapcount);
- if (first)
- __inc_zone_page_state(page, NR_ANON_PAGES);
+ if (first) {
+ if (!PageTransHuge(page))
+ __inc_zone_page_state(page, NR_ANON_PAGES);
+ else
+ __inc_zone_page_state(page,
+ NR_ANON_TRANSPARENT_HUGEPAGES);
+ }
if (unlikely(PageKsm(page)))
return;
@@ -893,7 +916,10 @@
VM_BUG_ON(address < vma->vm_start || address >= vma->vm_end);
SetPageSwapBacked(page);
atomic_set(&page->_mapcount, 0); /* increment count (starts at -1) */
- __inc_zone_page_state(page, NR_ANON_PAGES);
+ if (!PageTransHuge(page))
+ __inc_zone_page_state(page, NR_ANON_PAGES);
+ else
+ __inc_zone_page_state(page, NR_ANON_TRANSPARENT_HUGEPAGES);
__page_set_anon_rmap(page, vma, address, 1);
if (page_evictable(page, vma))
lru_cache_add_lru(page, LRU_ACTIVE_ANON);
@@ -911,7 +937,7 @@
{
if (atomic_inc_and_test(&page->_mapcount)) {
__inc_zone_page_state(page, NR_FILE_MAPPED);
- mem_cgroup_update_file_mapped(page, 1);
+ mem_cgroup_inc_page_stat(page, MEMCG_NR_FILE_MAPPED);
}
}
@@ -946,10 +972,14 @@
return;
if (PageAnon(page)) {
mem_cgroup_uncharge_page(page);
- __dec_zone_page_state(page, NR_ANON_PAGES);
+ if (!PageTransHuge(page))
+ __dec_zone_page_state(page, NR_ANON_PAGES);
+ else
+ __dec_zone_page_state(page,
+ NR_ANON_TRANSPARENT_HUGEPAGES);
} else {
__dec_zone_page_state(page, NR_FILE_MAPPED);
- mem_cgroup_update_file_mapped(page, -1);
+ mem_cgroup_dec_page_stat(page, MEMCG_NR_FILE_MAPPED);
}
/*
* It would be tidy to reset the PageAnon mapping here,
@@ -1202,7 +1232,7 @@
return ret;
}
-static bool is_vma_temporary_stack(struct vm_area_struct *vma)
+bool is_vma_temporary_stack(struct vm_area_struct *vma)
{
int maybe_stack = vma->vm_flags & (VM_GROWSDOWN | VM_GROWSUP);
@@ -1400,6 +1430,7 @@
int ret;
BUG_ON(!PageLocked(page));
+ VM_BUG_ON(!PageHuge(page) && PageTransHuge(page));
if (unlikely(PageKsm(page)))
ret = try_to_unmap_ksm(page, flags);
diff --git a/mm/slab.c b/mm/slab.c
index 2640374..37961d1 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -284,7 +284,7 @@
* Need this for bootstrapping a per node allocator.
*/
#define NUM_INIT_LISTS (3 * MAX_NUMNODES)
-struct kmem_list3 __initdata initkmem_list3[NUM_INIT_LISTS];
+static struct kmem_list3 __initdata initkmem_list3[NUM_INIT_LISTS];
#define CACHE_CACHE 0
#define SIZE_AC MAX_NUMNODES
#define SIZE_L3 (2 * MAX_NUMNODES)
@@ -4053,7 +4053,7 @@
* necessary. Note that the l3 listlock also protects the array_cache
* if drain_array() is used on the shared array.
*/
-void drain_array(struct kmem_cache *cachep, struct kmem_list3 *l3,
+static void drain_array(struct kmem_cache *cachep, struct kmem_list3 *l3,
struct array_cache *ac, int force, int node)
{
int tofree;
@@ -4317,7 +4317,7 @@
* @count: data length
* @ppos: unused
*/
-ssize_t slabinfo_write(struct file *file, const char __user * buffer,
+static ssize_t slabinfo_write(struct file *file, const char __user *buffer,
size_t count, loff_t *ppos)
{
char kbuf[MAX_SLABINFO_WRITE + 1], *tmp;
diff --git a/mm/slub.c b/mm/slub.c
index 008cd74..e15aa7f 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -3636,7 +3636,7 @@
len += sprintf(buf + len, "%7ld ", l->count);
if (l->addr)
- len += sprint_symbol(buf + len, (unsigned long)l->addr);
+ len += sprintf(buf + len, "%pS", (void *)l->addr);
else
len += sprintf(buf + len, "<not-available>");
@@ -3797,7 +3797,7 @@
}
}
- down_read(&slub_lock);
+ lock_memory_hotplug();
#ifdef CONFIG_SLUB_DEBUG
if (flags & SO_ALL) {
for_each_node_state(node, N_NORMAL_MEMORY) {
@@ -3838,7 +3838,7 @@
x += sprintf(buf + x, " N%d=%lu",
node, nodes[node]);
#endif
- up_read(&slub_lock);
+ unlock_memory_hotplug();
kfree(nodes);
return x + sprintf(buf + x, "\n");
}
@@ -3946,12 +3946,9 @@
static ssize_t ctor_show(struct kmem_cache *s, char *buf)
{
- if (s->ctor) {
- int n = sprint_symbol(buf, (unsigned long)s->ctor);
-
- return n + sprintf(buf + n, "\n");
- }
- return 0;
+ if (!s->ctor)
+ return 0;
+ return sprintf(buf, "%pS\n", s->ctor);
}
SLAB_ATTR_RO(ctor);
diff --git a/mm/sparse.c b/mm/sparse.c
index 95ac219..9325020 100644
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -671,10 +671,10 @@
static void free_map_bootmem(struct page *page, unsigned long nr_pages)
{
unsigned long maps_section_nr, removing_section_nr, i;
- int magic;
+ unsigned long magic;
for (i = 0; i < nr_pages; i++, page++) {
- magic = atomic_read(&page->_mapcount);
+ magic = (unsigned long) page->lru.next;
BUG_ON(magic == NODE_INFO);
diff --git a/mm/swap.c b/mm/swap.c
index 3f48542..bbc1ce9 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -56,17 +56,97 @@
del_page_from_lru(zone, page);
spin_unlock_irqrestore(&zone->lru_lock, flags);
}
+}
+
+static void __put_single_page(struct page *page)
+{
+ __page_cache_release(page);
free_hot_cold_page(page, 0);
}
+static void __put_compound_page(struct page *page)
+{
+ compound_page_dtor *dtor;
+
+ __page_cache_release(page);
+ dtor = get_compound_page_dtor(page);
+ (*dtor)(page);
+}
+
static void put_compound_page(struct page *page)
{
- page = compound_head(page);
- if (put_page_testzero(page)) {
- compound_page_dtor *dtor;
-
- dtor = get_compound_page_dtor(page);
- (*dtor)(page);
+ if (unlikely(PageTail(page))) {
+ /* __split_huge_page_refcount can run under us */
+ struct page *page_head = page->first_page;
+ smp_rmb();
+ /*
+ * If PageTail is still set after smp_rmb() we can be sure
+ * that the page->first_page we read wasn't a dangling pointer.
+ * See __split_huge_page_refcount() smp_wmb().
+ */
+ if (likely(PageTail(page) && get_page_unless_zero(page_head))) {
+ unsigned long flags;
+ /*
+ * Verify that our page_head wasn't converted
+ * to a a regular page before we got a
+ * reference on it.
+ */
+ if (unlikely(!PageHead(page_head))) {
+ /* PageHead is cleared after PageTail */
+ smp_rmb();
+ VM_BUG_ON(PageTail(page));
+ goto out_put_head;
+ }
+ /*
+ * Only run compound_lock on a valid PageHead,
+ * after having it pinned with
+ * get_page_unless_zero() above.
+ */
+ smp_mb();
+ /* page_head wasn't a dangling pointer */
+ flags = compound_lock_irqsave(page_head);
+ if (unlikely(!PageTail(page))) {
+ /* __split_huge_page_refcount run before us */
+ compound_unlock_irqrestore(page_head, flags);
+ VM_BUG_ON(PageHead(page_head));
+ out_put_head:
+ if (put_page_testzero(page_head))
+ __put_single_page(page_head);
+ out_put_single:
+ if (put_page_testzero(page))
+ __put_single_page(page);
+ return;
+ }
+ VM_BUG_ON(page_head != page->first_page);
+ /*
+ * We can release the refcount taken by
+ * get_page_unless_zero now that
+ * split_huge_page_refcount is blocked on the
+ * compound_lock.
+ */
+ if (put_page_testzero(page_head))
+ VM_BUG_ON(1);
+ /* __split_huge_page_refcount will wait now */
+ VM_BUG_ON(atomic_read(&page->_count) <= 0);
+ atomic_dec(&page->_count);
+ VM_BUG_ON(atomic_read(&page_head->_count) <= 0);
+ compound_unlock_irqrestore(page_head, flags);
+ if (put_page_testzero(page_head)) {
+ if (PageHead(page_head))
+ __put_compound_page(page_head);
+ else
+ __put_single_page(page_head);
+ }
+ } else {
+ /* page_head is a dangling pointer */
+ VM_BUG_ON(PageTail(page));
+ goto out_put_single;
+ }
+ } else if (put_page_testzero(page)) {
+ if (PageHead(page))
+ __put_compound_page(page);
+ else
+ __put_single_page(page);
}
}
@@ -75,7 +155,7 @@
if (unlikely(PageCompound(page)))
put_compound_page(page);
else if (put_page_testzero(page))
- __page_cache_release(page);
+ __put_single_page(page);
}
EXPORT_SYMBOL(put_page);
@@ -98,15 +178,13 @@
}
EXPORT_SYMBOL(put_pages_list);
-/*
- * pagevec_move_tail() must be called with IRQ disabled.
- * Otherwise this may cause nasty races.
- */
-static void pagevec_move_tail(struct pagevec *pvec)
+static void pagevec_lru_move_fn(struct pagevec *pvec,
+ void (*move_fn)(struct page *page, void *arg),
+ void *arg)
{
int i;
- int pgmoved = 0;
struct zone *zone = NULL;
+ unsigned long flags = 0;
for (i = 0; i < pagevec_count(pvec); i++) {
struct page *page = pvec->pages[i];
@@ -114,29 +192,49 @@
if (pagezone != zone) {
if (zone)
- spin_unlock(&zone->lru_lock);
+ spin_unlock_irqrestore(&zone->lru_lock, flags);
zone = pagezone;
- spin_lock(&zone->lru_lock);
+ spin_lock_irqsave(&zone->lru_lock, flags);
}
- if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
- int lru = page_lru_base_type(page);
- list_move_tail(&page->lru, &zone->lru[lru].list);
- pgmoved++;
- }
+
+ (*move_fn)(page, arg);
}
if (zone)
- spin_unlock(&zone->lru_lock);
- __count_vm_events(PGROTATED, pgmoved);
- release_pages(pvec->pages, pvec->nr, pvec->cold);
+ spin_unlock_irqrestore(&zone->lru_lock, flags);
+ release_pages(pvec->pages, pagevec_count(pvec), pvec->cold);
pagevec_reinit(pvec);
}
+static void pagevec_move_tail_fn(struct page *page, void *arg)
+{
+ int *pgmoved = arg;
+ struct zone *zone = page_zone(page);
+
+ if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
+ int lru = page_lru_base_type(page);
+ list_move_tail(&page->lru, &zone->lru[lru].list);
+ (*pgmoved)++;
+ }
+}
+
+/*
+ * pagevec_move_tail() must be called with IRQ disabled.
+ * Otherwise this may cause nasty races.
+ */
+static void pagevec_move_tail(struct pagevec *pvec)
+{
+ int pgmoved = 0;
+
+ pagevec_lru_move_fn(pvec, pagevec_move_tail_fn, &pgmoved);
+ __count_vm_events(PGROTATED, pgmoved);
+}
+
/*
* Writeback is about to end against a page which has been marked for immediate
* reclaim. If it still appears to be reclaimable, move it to the tail of the
* inactive list.
*/
-void rotate_reclaimable_page(struct page *page)
+void rotate_reclaimable_page(struct page *page)
{
if (!PageLocked(page) && !PageDirty(page) && !PageActive(page) &&
!PageUnevictable(page) && PageLRU(page)) {
@@ -173,27 +271,94 @@
}
/*
- * FIXME: speed this up?
+ * A page will go to active list either by activate_page or putback_lru_page.
+ * In the activate_page case, the page hasn't active bit set. The page might
+ * not in LRU list because it's isolated before it gets a chance to be moved to
+ * active list. The window is small because pagevec just stores several pages.
+ * For such case, we do nothing for such page.
+ * In the putback_lru_page case, the page isn't in lru list but has active
+ * bit set
*/
+static void __activate_page(struct page *page, void *arg)
+{
+ struct zone *zone = page_zone(page);
+ int file = page_is_file_cache(page);
+ int lru = page_lru_base_type(page);
+ bool putback = !PageLRU(page);
+
+ /* The page is isolated before it's moved to active list */
+ if (!PageLRU(page) && !PageActive(page))
+ return;
+ if ((PageLRU(page) && PageActive(page)) || PageUnevictable(page))
+ return;
+
+ if (!putback)
+ del_page_from_lru_list(zone, page, lru);
+ else
+ SetPageLRU(page);
+
+ SetPageActive(page);
+ lru += LRU_ACTIVE;
+ add_page_to_lru_list(zone, page, lru);
+
+ if (putback)
+ return;
+ __count_vm_event(PGACTIVATE);
+ update_page_reclaim_stat(zone, page, file, 1);
+}
+
+#ifdef CONFIG_SMP
+static DEFINE_PER_CPU(struct pagevec, activate_page_pvecs);
+
+static void activate_page_drain(int cpu)
+{
+ struct pagevec *pvec = &per_cpu(activate_page_pvecs, cpu);
+
+ if (pagevec_count(pvec))
+ pagevec_lru_move_fn(pvec, __activate_page, NULL);
+}
+
+void activate_page(struct page *page)
+{
+ if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
+ struct pagevec *pvec = &get_cpu_var(activate_page_pvecs);
+
+ page_cache_get(page);
+ if (!pagevec_add(pvec, page))
+ pagevec_lru_move_fn(pvec, __activate_page, NULL);
+ put_cpu_var(activate_page_pvecs);
+ }
+}
+
+/* Caller should hold zone->lru_lock */
+int putback_active_lru_page(struct zone *zone, struct page *page)
+{
+ struct pagevec *pvec = &get_cpu_var(activate_page_pvecs);
+
+ if (!pagevec_add(pvec, page)) {
+ spin_unlock_irq(&zone->lru_lock);
+ pagevec_lru_move_fn(pvec, __activate_page, NULL);
+ spin_lock_irq(&zone->lru_lock);
+ }
+ put_cpu_var(activate_page_pvecs);
+ return 1;
+}
+
+#else
+static inline void activate_page_drain(int cpu)
+{
+}
+
void activate_page(struct page *page)
{
struct zone *zone = page_zone(page);
spin_lock_irq(&zone->lru_lock);
- if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
- int file = page_is_file_cache(page);
- int lru = page_lru_base_type(page);
- del_page_from_lru_list(zone, page, lru);
-
- SetPageActive(page);
- lru += LRU_ACTIVE;
- add_page_to_lru_list(zone, page, lru);
- __count_vm_event(PGACTIVATE);
-
- update_page_reclaim_stat(zone, page, file, 1);
- }
+ if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page))
+ __activate_page(page, NULL);
spin_unlock_irq(&zone->lru_lock);
}
+#endif
/*
* Mark a page as having seen activity.
@@ -292,6 +457,7 @@
pagevec_move_tail(pvec);
local_irq_restore(flags);
}
+ activate_page_drain(cpu);
}
void lru_add_drain(void)
@@ -399,44 +565,70 @@
EXPORT_SYMBOL(__pagevec_release);
+/* used by __split_huge_page_refcount() */
+void lru_add_page_tail(struct zone* zone,
+ struct page *page, struct page *page_tail)
+{
+ int active;
+ enum lru_list lru;
+ const int file = 0;
+ struct list_head *head;
+
+ VM_BUG_ON(!PageHead(page));
+ VM_BUG_ON(PageCompound(page_tail));
+ VM_BUG_ON(PageLRU(page_tail));
+ VM_BUG_ON(!spin_is_locked(&zone->lru_lock));
+
+ SetPageLRU(page_tail);
+
+ if (page_evictable(page_tail, NULL)) {
+ if (PageActive(page)) {
+ SetPageActive(page_tail);
+ active = 1;
+ lru = LRU_ACTIVE_ANON;
+ } else {
+ active = 0;
+ lru = LRU_INACTIVE_ANON;
+ }
+ update_page_reclaim_stat(zone, page_tail, file, active);
+ if (likely(PageLRU(page)))
+ head = page->lru.prev;
+ else
+ head = &zone->lru[lru].list;
+ __add_page_to_lru_list(zone, page_tail, lru, head);
+ } else {
+ SetPageUnevictable(page_tail);
+ add_page_to_lru_list(zone, page_tail, LRU_UNEVICTABLE);
+ }
+}
+
+static void ____pagevec_lru_add_fn(struct page *page, void *arg)
+{
+ enum lru_list lru = (enum lru_list)arg;
+ struct zone *zone = page_zone(page);
+ int file = is_file_lru(lru);
+ int active = is_active_lru(lru);
+
+ VM_BUG_ON(PageActive(page));
+ VM_BUG_ON(PageUnevictable(page));
+ VM_BUG_ON(PageLRU(page));
+
+ SetPageLRU(page);
+ if (active)
+ SetPageActive(page);
+ update_page_reclaim_stat(zone, page, file, active);
+ add_page_to_lru_list(zone, page, lru);
+}
+
/*
* Add the passed pages to the LRU, then drop the caller's refcount
* on them. Reinitialises the caller's pagevec.
*/
void ____pagevec_lru_add(struct pagevec *pvec, enum lru_list lru)
{
- int i;
- struct zone *zone = NULL;
-
VM_BUG_ON(is_unevictable_lru(lru));
- for (i = 0; i < pagevec_count(pvec); i++) {
- struct page *page = pvec->pages[i];
- struct zone *pagezone = page_zone(page);
- int file;
- int active;
-
- if (pagezone != zone) {
- if (zone)
- spin_unlock_irq(&zone->lru_lock);
- zone = pagezone;
- spin_lock_irq(&zone->lru_lock);
- }
- VM_BUG_ON(PageActive(page));
- VM_BUG_ON(PageUnevictable(page));
- VM_BUG_ON(PageLRU(page));
- SetPageLRU(page);
- active = is_active_lru(lru);
- file = is_file_lru(lru);
- if (active)
- SetPageActive(page);
- update_page_reclaim_stat(zone, page, file, active);
- add_page_to_lru_list(zone, page, lru);
- }
- if (zone)
- spin_unlock_irq(&zone->lru_lock);
- release_pages(pvec->pages, pvec->nr, pvec->cold);
- pagevec_reinit(pvec);
+ pagevec_lru_move_fn(pvec, ____pagevec_lru_add_fn, (void *)lru);
}
EXPORT_SYMBOL(____pagevec_lru_add);
diff --git a/mm/swap_state.c b/mm/swap_state.c
index e10f583..5c8cfab 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -157,6 +157,12 @@
if (!entry.val)
return 0;
+ if (unlikely(PageTransHuge(page)))
+ if (unlikely(split_huge_page(page))) {
+ swapcache_free(entry, NULL);
+ return 0;
+ }
+
/*
* Radix-tree node allocations from PF_MEMALLOC contexts could
* completely exhaust the page allocator. __GFP_NOMEMALLOC
diff --git a/mm/swapfile.c b/mm/swapfile.c
index b6adcfb..07a458d 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -964,6 +964,8 @@
pmd = pmd_offset(pud, addr);
do {
next = pmd_addr_end(addr, end);
+ if (unlikely(pmd_trans_huge(*pmd)))
+ continue;
if (pmd_none_or_clear_bad(pmd))
continue;
ret = unuse_pte_range(vma, pmd, addr, next, entry, page);
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index eb5cc7d..f9b1667 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -748,7 +748,7 @@
va = alloc_vmap_area(VMAP_BLOCK_SIZE, VMAP_BLOCK_SIZE,
VMALLOC_START, VMALLOC_END,
node, gfp_mask);
- if (unlikely(IS_ERR(va))) {
+ if (IS_ERR(va)) {
kfree(vb);
return ERR_CAST(va);
}
@@ -1175,6 +1175,7 @@
{
vunmap_page_range(addr, addr + size);
}
+EXPORT_SYMBOL_GPL(unmap_kernel_range_noflush);
/**
* unmap_kernel_range - unmap kernel VM area and flush cache and TLB
@@ -1315,13 +1316,6 @@
-1, GFP_KERNEL, caller);
}
-struct vm_struct *get_vm_area_node(unsigned long size, unsigned long flags,
- int node, gfp_t gfp_mask)
-{
- return __get_vm_area_node(size, 1, flags, VMALLOC_START, VMALLOC_END,
- node, gfp_mask, __builtin_return_address(0));
-}
-
static struct vm_struct *find_vm_area(const void *addr)
{
struct vmap_area *va;
@@ -1537,17 +1531,47 @@
return NULL;
}
-void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot)
+/**
+ * __vmalloc_node_range - allocate virtually contiguous memory
+ * @size: allocation size
+ * @align: desired alignment
+ * @start: vm area range start
+ * @end: vm area range end
+ * @gfp_mask: flags for the page level allocator
+ * @prot: protection mask for the allocated pages
+ * @node: node to use for allocation or -1
+ * @caller: caller's return address
+ *
+ * Allocate enough pages to cover @size from the page level
+ * allocator with @gfp_mask flags. Map them into contiguous
+ * kernel virtual space, using a pagetable protection of @prot.
+ */
+void *__vmalloc_node_range(unsigned long size, unsigned long align,
+ unsigned long start, unsigned long end, gfp_t gfp_mask,
+ pgprot_t prot, int node, void *caller)
{
- void *addr = __vmalloc_area_node(area, gfp_mask, prot, -1,
- __builtin_return_address(0));
+ struct vm_struct *area;
+ void *addr;
+ unsigned long real_size = size;
+
+ size = PAGE_ALIGN(size);
+ if (!size || (size >> PAGE_SHIFT) > totalram_pages)
+ return NULL;
+
+ area = __get_vm_area_node(size, align, VM_ALLOC, start, end, node,
+ gfp_mask, caller);
+
+ if (!area)
+ return NULL;
+
+ addr = __vmalloc_area_node(area, gfp_mask, prot, node, caller);
/*
* A ref_count = 3 is needed because the vm_struct and vmap_area
* structures allocated in the __get_vm_area_node() function contain
* references to the virtual address of the vmalloc'ed block.
*/
- kmemleak_alloc(addr, area->size - PAGE_SIZE, 3, gfp_mask);
+ kmemleak_alloc(addr, real_size, 3, gfp_mask);
return addr;
}
@@ -1569,30 +1593,8 @@
gfp_t gfp_mask, pgprot_t prot,
int node, void *caller)
{
- struct vm_struct *area;
- void *addr;
- unsigned long real_size = size;
-
- size = PAGE_ALIGN(size);
- if (!size || (size >> PAGE_SHIFT) > totalram_pages)
- return NULL;
-
- area = __get_vm_area_node(size, align, VM_ALLOC, VMALLOC_START,
- VMALLOC_END, node, gfp_mask, caller);
-
- if (!area)
- return NULL;
-
- addr = __vmalloc_area_node(area, gfp_mask, prot, node, caller);
-
- /*
- * A ref_count = 3 is needed because the vm_struct and vmap_area
- * structures allocated in the __get_vm_area_node() function contain
- * references to the virtual address of the vmalloc'ed block.
- */
- kmemleak_alloc(addr, real_size, 3, gfp_mask);
-
- return addr;
+ return __vmalloc_node_range(size, align, VMALLOC_START, VMALLOC_END,
+ gfp_mask, prot, node, caller);
}
void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
@@ -2203,17 +2205,16 @@
* @sizes: array containing size of each area
* @nr_vms: the number of areas to allocate
* @align: alignment, all entries in @offsets and @sizes must be aligned to this
- * @gfp_mask: allocation mask
*
* Returns: kmalloc'd vm_struct pointer array pointing to allocated
* vm_structs on success, %NULL on failure
*
* Percpu allocator wants to use congruent vm areas so that it can
* maintain the offsets among percpu areas. This function allocates
- * congruent vmalloc areas for it. These areas tend to be scattered
- * pretty far, distance between two areas easily going up to
- * gigabytes. To avoid interacting with regular vmallocs, these areas
- * are allocated from top.
+ * congruent vmalloc areas for it with GFP_KERNEL. These areas tend to
+ * be scattered pretty far, distance between two areas easily going up
+ * to gigabytes. To avoid interacting with regular vmallocs, these
+ * areas are allocated from top.
*
* Despite its complicated look, this allocator is rather simple. It
* does everything top-down and scans areas from the end looking for
@@ -2224,7 +2225,7 @@
*/
struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
const size_t *sizes, int nr_vms,
- size_t align, gfp_t gfp_mask)
+ size_t align)
{
const unsigned long vmalloc_start = ALIGN(VMALLOC_START, align);
const unsigned long vmalloc_end = VMALLOC_END & ~(align - 1);
@@ -2234,8 +2235,6 @@
unsigned long base, start, end, last_end;
bool purged = false;
- gfp_mask &= GFP_RECLAIM_MASK;
-
/* verify parameters and allocate data structures */
BUG_ON(align & ~PAGE_MASK || !is_power_of_2(align));
for (last_area = 0, area = 0; area < nr_vms; area++) {
@@ -2268,14 +2267,14 @@
return NULL;
}
- vms = kzalloc(sizeof(vms[0]) * nr_vms, gfp_mask);
- vas = kzalloc(sizeof(vas[0]) * nr_vms, gfp_mask);
+ vms = kzalloc(sizeof(vms[0]) * nr_vms, GFP_KERNEL);
+ vas = kzalloc(sizeof(vas[0]) * nr_vms, GFP_KERNEL);
if (!vas || !vms)
goto err_free;
for (area = 0; area < nr_vms; area++) {
- vas[area] = kzalloc(sizeof(struct vmap_area), gfp_mask);
- vms[area] = kzalloc(sizeof(struct vm_struct), gfp_mask);
+ vas[area] = kzalloc(sizeof(struct vmap_area), GFP_KERNEL);
+ vms[area] = kzalloc(sizeof(struct vm_struct), GFP_KERNEL);
if (!vas[area] || !vms[area])
goto err_free;
}
@@ -2456,13 +2455,8 @@
seq_printf(m, "0x%p-0x%p %7ld",
v->addr, v->addr + v->size, v->size);
- if (v->caller) {
- char buff[KSYM_SYMBOL_LEN];
-
- seq_putc(m, ' ');
- sprint_symbol(buff, (unsigned long)v->caller);
- seq_puts(m, buff);
- }
+ if (v->caller)
+ seq_printf(m, " %pS", v->caller);
if (v->nr_pages)
seq_printf(m, " pages=%d", v->nr_pages);
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 9ca587c..99999a9 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -32,6 +32,7 @@
#include <linux/topology.h>
#include <linux/cpu.h>
#include <linux/cpuset.h>
+#include <linux/compaction.h>
#include <linux/notifier.h>
#include <linux/rwsem.h>
#include <linux/delay.h>
@@ -40,6 +41,7 @@
#include <linux/memcontrol.h>
#include <linux/delayacct.h>
#include <linux/sysctl.h>
+#include <linux/compaction.h>
#include <asm/tlbflush.h>
#include <asm/div64.h>
@@ -51,11 +53,23 @@
#define CREATE_TRACE_POINTS
#include <trace/events/vmscan.h>
-enum lumpy_mode {
- LUMPY_MODE_NONE,
- LUMPY_MODE_ASYNC,
- LUMPY_MODE_SYNC,
-};
+/*
+ * reclaim_mode determines how the inactive list is shrunk
+ * RECLAIM_MODE_SINGLE: Reclaim only order-0 pages
+ * RECLAIM_MODE_ASYNC: Do not block
+ * RECLAIM_MODE_SYNC: Allow blocking e.g. call wait_on_page_writeback
+ * RECLAIM_MODE_LUMPYRECLAIM: For high-order allocations, take a reference
+ * page from the LRU and reclaim all pages within a
+ * naturally aligned range
+ * RECLAIM_MODE_COMPACTION: For high-order allocations, reclaim a number of
+ * order-0 pages and then compact the zone
+ */
+typedef unsigned __bitwise__ reclaim_mode_t;
+#define RECLAIM_MODE_SINGLE ((__force reclaim_mode_t)0x01u)
+#define RECLAIM_MODE_ASYNC ((__force reclaim_mode_t)0x02u)
+#define RECLAIM_MODE_SYNC ((__force reclaim_mode_t)0x04u)
+#define RECLAIM_MODE_LUMPYRECLAIM ((__force reclaim_mode_t)0x08u)
+#define RECLAIM_MODE_COMPACTION ((__force reclaim_mode_t)0x10u)
struct scan_control {
/* Incremented by the number of inactive pages that were scanned */
@@ -88,7 +102,7 @@
* Intend to reclaim enough continuous memory rather than reclaim
* enough amount of memory. i.e, mode for high order allocation.
*/
- enum lumpy_mode lumpy_reclaim_mode;
+ reclaim_mode_t reclaim_mode;
/* Which cgroup do we reclaim from */
struct mem_cgroup *mem_cgroup;
@@ -271,34 +285,37 @@
return ret;
}
-static void set_lumpy_reclaim_mode(int priority, struct scan_control *sc,
+static void set_reclaim_mode(int priority, struct scan_control *sc,
bool sync)
{
- enum lumpy_mode mode = sync ? LUMPY_MODE_SYNC : LUMPY_MODE_ASYNC;
+ reclaim_mode_t syncmode = sync ? RECLAIM_MODE_SYNC : RECLAIM_MODE_ASYNC;
/*
- * Some reclaim have alredy been failed. No worth to try synchronous
- * lumpy reclaim.
+ * Initially assume we are entering either lumpy reclaim or
+ * reclaim/compaction.Depending on the order, we will either set the
+ * sync mode or just reclaim order-0 pages later.
*/
- if (sync && sc->lumpy_reclaim_mode == LUMPY_MODE_NONE)
- return;
+ if (COMPACTION_BUILD)
+ sc->reclaim_mode = RECLAIM_MODE_COMPACTION;
+ else
+ sc->reclaim_mode = RECLAIM_MODE_LUMPYRECLAIM;
/*
- * If we need a large contiguous chunk of memory, or have
- * trouble getting a small set of contiguous pages, we
- * will reclaim both active and inactive pages.
+ * Avoid using lumpy reclaim or reclaim/compaction if possible by
+ * restricting when its set to either costly allocations or when
+ * under memory pressure
*/
if (sc->order > PAGE_ALLOC_COSTLY_ORDER)
- sc->lumpy_reclaim_mode = mode;
+ sc->reclaim_mode |= syncmode;
else if (sc->order && priority < DEF_PRIORITY - 2)
- sc->lumpy_reclaim_mode = mode;
+ sc->reclaim_mode |= syncmode;
else
- sc->lumpy_reclaim_mode = LUMPY_MODE_NONE;
+ sc->reclaim_mode = RECLAIM_MODE_SINGLE | RECLAIM_MODE_ASYNC;
}
-static void disable_lumpy_reclaim_mode(struct scan_control *sc)
+static void reset_reclaim_mode(struct scan_control *sc)
{
- sc->lumpy_reclaim_mode = LUMPY_MODE_NONE;
+ sc->reclaim_mode = RECLAIM_MODE_SINGLE | RECLAIM_MODE_ASYNC;
}
static inline int is_page_cache_freeable(struct page *page)
@@ -429,7 +446,7 @@
* first attempt to free a range of pages fails.
*/
if (PageWriteback(page) &&
- sc->lumpy_reclaim_mode == LUMPY_MODE_SYNC)
+ (sc->reclaim_mode & RECLAIM_MODE_SYNC))
wait_on_page_writeback(page);
if (!PageWriteback(page)) {
@@ -437,7 +454,7 @@
ClearPageReclaim(page);
}
trace_mm_vmscan_writepage(page,
- trace_reclaim_flags(page, sc->lumpy_reclaim_mode));
+ trace_reclaim_flags(page, sc->reclaim_mode));
inc_zone_page_state(page, NR_VMSCAN_WRITE);
return PAGE_SUCCESS;
}
@@ -622,7 +639,7 @@
referenced_page = TestClearPageReferenced(page);
/* Lumpy reclaim - ignore references */
- if (sc->lumpy_reclaim_mode != LUMPY_MODE_NONE)
+ if (sc->reclaim_mode & RECLAIM_MODE_LUMPYRECLAIM)
return PAGEREF_RECLAIM;
/*
@@ -739,7 +756,7 @@
* for any page for which writeback has already
* started.
*/
- if (sc->lumpy_reclaim_mode == LUMPY_MODE_SYNC &&
+ if ((sc->reclaim_mode & RECLAIM_MODE_SYNC) &&
may_enter_fs)
wait_on_page_writeback(page);
else {
@@ -895,7 +912,7 @@
try_to_free_swap(page);
unlock_page(page);
putback_lru_page(page);
- disable_lumpy_reclaim_mode(sc);
+ reset_reclaim_mode(sc);
continue;
activate_locked:
@@ -908,7 +925,7 @@
keep_locked:
unlock_page(page);
keep:
- disable_lumpy_reclaim_mode(sc);
+ reset_reclaim_mode(sc);
keep_lumpy:
list_add(&page->lru, &ret_pages);
VM_BUG_ON(PageLRU(page) || PageUnevictable(page));
@@ -1028,7 +1045,7 @@
case 0:
list_move(&page->lru, dst);
mem_cgroup_del_lru(page);
- nr_taken++;
+ nr_taken += hpage_nr_pages(page);
break;
case -EBUSY:
@@ -1086,7 +1103,7 @@
if (__isolate_lru_page(cursor_page, mode, file) == 0) {
list_move(&cursor_page->lru, dst);
mem_cgroup_del_lru(cursor_page);
- nr_taken++;
+ nr_taken += hpage_nr_pages(page);
nr_lumpy_taken++;
if (PageDirty(cursor_page))
nr_lumpy_dirty++;
@@ -1141,14 +1158,15 @@
struct page *page;
list_for_each_entry(page, page_list, lru) {
+ int numpages = hpage_nr_pages(page);
lru = page_lru_base_type(page);
if (PageActive(page)) {
lru += LRU_ACTIVE;
ClearPageActive(page);
- nr_active++;
+ nr_active += numpages;
}
if (count)
- count[lru]++;
+ count[lru] += numpages;
}
return nr_active;
@@ -1253,13 +1271,16 @@
spin_lock_irq(&zone->lru_lock);
continue;
}
- SetPageLRU(page);
lru = page_lru(page);
- add_page_to_lru_list(zone, page, lru);
if (is_active_lru(lru)) {
int file = is_file_lru(lru);
- reclaim_stat->recent_rotated[file]++;
+ int numpages = hpage_nr_pages(page);
+ reclaim_stat->recent_rotated[file] += numpages;
+ if (putback_active_lru_page(zone, page))
+ continue;
}
+ SetPageLRU(page);
+ add_page_to_lru_list(zone, page, lru);
if (!pagevec_add(&pvec, page)) {
spin_unlock_irq(&zone->lru_lock);
__pagevec_release(&pvec);
@@ -1324,7 +1345,7 @@
return false;
/* Only stall on lumpy reclaim */
- if (sc->lumpy_reclaim_mode == LUMPY_MODE_NONE)
+ if (sc->reclaim_mode & RECLAIM_MODE_SINGLE)
return false;
/* If we have relaimed everything on the isolated list, no stall */
@@ -1368,15 +1389,15 @@
return SWAP_CLUSTER_MAX;
}
- set_lumpy_reclaim_mode(priority, sc, false);
+ set_reclaim_mode(priority, sc, false);
lru_add_drain();
spin_lock_irq(&zone->lru_lock);
if (scanning_global_lru(sc)) {
nr_taken = isolate_pages_global(nr_to_scan,
&page_list, &nr_scanned, sc->order,
- sc->lumpy_reclaim_mode == LUMPY_MODE_NONE ?
- ISOLATE_INACTIVE : ISOLATE_BOTH,
+ sc->reclaim_mode & RECLAIM_MODE_LUMPYRECLAIM ?
+ ISOLATE_BOTH : ISOLATE_INACTIVE,
zone, 0, file);
zone->pages_scanned += nr_scanned;
if (current_is_kswapd())
@@ -1388,8 +1409,8 @@
} else {
nr_taken = mem_cgroup_isolate_pages(nr_to_scan,
&page_list, &nr_scanned, sc->order,
- sc->lumpy_reclaim_mode == LUMPY_MODE_NONE ?
- ISOLATE_INACTIVE : ISOLATE_BOTH,
+ sc->reclaim_mode & RECLAIM_MODE_LUMPYRECLAIM ?
+ ISOLATE_BOTH : ISOLATE_INACTIVE,
zone, sc->mem_cgroup,
0, file);
/*
@@ -1411,7 +1432,7 @@
/* Check if we should syncronously wait for writeback */
if (should_reclaim_stall(nr_taken, nr_reclaimed, priority, sc)) {
- set_lumpy_reclaim_mode(priority, sc, true);
+ set_reclaim_mode(priority, sc, true);
nr_reclaimed += shrink_page_list(&page_list, zone, sc);
}
@@ -1426,7 +1447,7 @@
zone_idx(zone),
nr_scanned, nr_reclaimed,
priority,
- trace_shrink_flags(file, sc->lumpy_reclaim_mode));
+ trace_shrink_flags(file, sc->reclaim_mode));
return nr_reclaimed;
}
@@ -1466,7 +1487,7 @@
list_move(&page->lru, &zone->lru[lru].list);
mem_cgroup_add_lru_list(page, lru);
- pgmoved++;
+ pgmoved += hpage_nr_pages(page);
if (!pagevec_add(&pvec, page) || list_empty(list)) {
spin_unlock_irq(&zone->lru_lock);
@@ -1534,7 +1555,7 @@
}
if (page_referenced(page, 0, sc->mem_cgroup, &vm_flags)) {
- nr_rotated++;
+ nr_rotated += hpage_nr_pages(page);
/*
* Identify referenced, file-backed active pages and
* give them one more trip around the active list. So
@@ -1805,6 +1826,57 @@
}
/*
+ * Reclaim/compaction depends on a number of pages being freed. To avoid
+ * disruption to the system, a small number of order-0 pages continue to be
+ * rotated and reclaimed in the normal fashion. However, by the time we get
+ * back to the allocator and call try_to_compact_zone(), we ensure that
+ * there are enough free pages for it to be likely successful
+ */
+static inline bool should_continue_reclaim(struct zone *zone,
+ unsigned long nr_reclaimed,
+ unsigned long nr_scanned,
+ struct scan_control *sc)
+{
+ unsigned long pages_for_compaction;
+ unsigned long inactive_lru_pages;
+
+ /* If not in reclaim/compaction mode, stop */
+ if (!(sc->reclaim_mode & RECLAIM_MODE_COMPACTION))
+ return false;
+
+ /*
+ * If we failed to reclaim and have scanned the full list, stop.
+ * NOTE: Checking just nr_reclaimed would exit reclaim/compaction far
+ * faster but obviously would be less likely to succeed
+ * allocation. If this is desirable, use GFP_REPEAT to decide
+ * if both reclaimed and scanned should be checked or just
+ * reclaimed
+ */
+ if (!nr_reclaimed && !nr_scanned)
+ return false;
+
+ /*
+ * If we have not reclaimed enough pages for compaction and the
+ * inactive lists are large enough, continue reclaiming
+ */
+ pages_for_compaction = (2UL << sc->order);
+ inactive_lru_pages = zone_nr_lru_pages(zone, sc, LRU_INACTIVE_ANON) +
+ zone_nr_lru_pages(zone, sc, LRU_INACTIVE_FILE);
+ if (sc->nr_reclaimed < pages_for_compaction &&
+ inactive_lru_pages > pages_for_compaction)
+ return true;
+
+ /* If compaction would go ahead or the allocation would succeed, stop */
+ switch (compaction_suitable(zone, sc->order)) {
+ case COMPACT_PARTIAL:
+ case COMPACT_CONTINUE:
+ return false;
+ default:
+ return true;
+ }
+}
+
+/*
* This is a basic per-zone page freer. Used by both kswapd and direct reclaim.
*/
static void shrink_zone(int priority, struct zone *zone,
@@ -1813,9 +1885,12 @@
unsigned long nr[NR_LRU_LISTS];
unsigned long nr_to_scan;
enum lru_list l;
- unsigned long nr_reclaimed = sc->nr_reclaimed;
+ unsigned long nr_reclaimed;
unsigned long nr_to_reclaim = sc->nr_to_reclaim;
+ unsigned long nr_scanned = sc->nr_scanned;
+restart:
+ nr_reclaimed = 0;
get_scan_count(zone, sc, nr, priority);
while (nr[LRU_INACTIVE_ANON] || nr[LRU_ACTIVE_FILE] ||
@@ -1841,8 +1916,7 @@
if (nr_reclaimed >= nr_to_reclaim && priority < DEF_PRIORITY)
break;
}
-
- sc->nr_reclaimed = nr_reclaimed;
+ sc->nr_reclaimed += nr_reclaimed;
/*
* Even if we did not try to evict anon pages at all, we want to
@@ -1851,6 +1925,11 @@
if (inactive_anon_is_low(zone, sc))
shrink_active_list(SWAP_CLUSTER_MAX, zone, sc, priority, 0);
+ /* reclaim/compaction might need reclaim to continue */
+ if (should_continue_reclaim(zone, nr_reclaimed,
+ sc->nr_scanned - nr_scanned, sc))
+ goto restart;
+
throttle_vm_writeout(sc->gfp_mask);
}
@@ -2124,38 +2203,87 @@
}
#endif
+/*
+ * pgdat_balanced is used when checking if a node is balanced for high-order
+ * allocations. Only zones that meet watermarks and are in a zone allowed
+ * by the callers classzone_idx are added to balanced_pages. The total of
+ * balanced pages must be at least 25% of the zones allowed by classzone_idx
+ * for the node to be considered balanced. Forcing all zones to be balanced
+ * for high orders can cause excessive reclaim when there are imbalanced zones.
+ * The choice of 25% is due to
+ * o a 16M DMA zone that is balanced will not balance a zone on any
+ * reasonable sized machine
+ * o On all other machines, the top zone must be at least a reasonable
+ * precentage of the middle zones. For example, on 32-bit x86, highmem
+ * would need to be at least 256M for it to be balance a whole node.
+ * Similarly, on x86-64 the Normal zone would need to be at least 1G
+ * to balance a node on its own. These seemed like reasonable ratios.
+ */
+static bool pgdat_balanced(pg_data_t *pgdat, unsigned long balanced_pages,
+ int classzone_idx)
+{
+ unsigned long present_pages = 0;
+ int i;
+
+ for (i = 0; i <= classzone_idx; i++)
+ present_pages += pgdat->node_zones[i].present_pages;
+
+ return balanced_pages > (present_pages >> 2);
+}
+
/* is kswapd sleeping prematurely? */
-static int sleeping_prematurely(pg_data_t *pgdat, int order, long remaining)
+static bool sleeping_prematurely(pg_data_t *pgdat, int order, long remaining,
+ int classzone_idx)
{
int i;
+ unsigned long balanced = 0;
+ bool all_zones_ok = true;
/* If a direct reclaimer woke kswapd within HZ/10, it's premature */
if (remaining)
- return 1;
+ return true;
- /* If after HZ/10, a zone is below the high mark, it's premature */
+ /* Check the watermark levels */
for (i = 0; i < pgdat->nr_zones; i++) {
struct zone *zone = pgdat->node_zones + i;
if (!populated_zone(zone))
continue;
- if (zone->all_unreclaimable)
+ /*
+ * balance_pgdat() skips over all_unreclaimable after
+ * DEF_PRIORITY. Effectively, it considers them balanced so
+ * they must be considered balanced here as well if kswapd
+ * is to sleep
+ */
+ if (zone->all_unreclaimable) {
+ balanced += zone->present_pages;
continue;
+ }
- if (!zone_watermark_ok(zone, order, high_wmark_pages(zone),
- 0, 0))
- return 1;
+ if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone),
+ classzone_idx, 0))
+ all_zones_ok = false;
+ else
+ balanced += zone->present_pages;
}
- return 0;
+ /*
+ * For high-order requests, the balanced zones must contain at least
+ * 25% of the nodes pages for kswapd to sleep. For order-0, all zones
+ * must be balanced
+ */
+ if (order)
+ return pgdat_balanced(pgdat, balanced, classzone_idx);
+ else
+ return !all_zones_ok;
}
/*
* For kswapd, balance_pgdat() will work across all this node's zones until
* they are all at high_wmark_pages(zone).
*
- * Returns the number of pages which were actually freed.
+ * Returns the final order kswapd was reclaiming at
*
* There is special handling here for zones which are full of pinned pages.
* This can happen if the pages are all mlocked, or if they are all used by
@@ -2172,11 +2300,14 @@
* interoperates with the page allocator fallback scheme to ensure that aging
* of pages is balanced across the zones.
*/
-static unsigned long balance_pgdat(pg_data_t *pgdat, int order)
+static unsigned long balance_pgdat(pg_data_t *pgdat, int order,
+ int *classzone_idx)
{
int all_zones_ok;
+ unsigned long balanced;
int priority;
int i;
+ int end_zone = 0; /* Inclusive. 0 = ZONE_DMA */
unsigned long total_scanned;
struct reclaim_state *reclaim_state = current->reclaim_state;
struct scan_control sc = {
@@ -2199,7 +2330,6 @@
count_vm_event(PAGEOUTRUN);
for (priority = DEF_PRIORITY; priority >= 0; priority--) {
- int end_zone = 0; /* Inclusive. 0 = ZONE_DMA */
unsigned long lru_pages = 0;
int has_under_min_watermark_zone = 0;
@@ -2208,6 +2338,7 @@
disable_swap_token();
all_zones_ok = 1;
+ balanced = 0;
/*
* Scan in the highmem->dma direction for the highest
@@ -2230,9 +2361,10 @@
shrink_active_list(SWAP_CLUSTER_MAX, zone,
&sc, priority, 0);
- if (!zone_watermark_ok(zone, order,
+ if (!zone_watermark_ok_safe(zone, order,
high_wmark_pages(zone), 0, 0)) {
end_zone = i;
+ *classzone_idx = i;
break;
}
}
@@ -2255,6 +2387,7 @@
* cause too much scanning of the lower zones.
*/
for (i = 0; i <= end_zone; i++) {
+ int compaction;
struct zone *zone = pgdat->node_zones + i;
int nr_slab;
@@ -2276,7 +2409,7 @@
* We put equal pressure on every zone, unless one
* zone has way too many pages free already.
*/
- if (!zone_watermark_ok(zone, order,
+ if (!zone_watermark_ok_safe(zone, order,
8*high_wmark_pages(zone), end_zone, 0))
shrink_zone(priority, zone, &sc);
reclaim_state->reclaimed_slab = 0;
@@ -2284,9 +2417,26 @@
lru_pages);
sc.nr_reclaimed += reclaim_state->reclaimed_slab;
total_scanned += sc.nr_scanned;
+
+ compaction = 0;
+ if (order &&
+ zone_watermark_ok(zone, 0,
+ high_wmark_pages(zone),
+ end_zone, 0) &&
+ !zone_watermark_ok(zone, order,
+ high_wmark_pages(zone),
+ end_zone, 0)) {
+ compact_zone_order(zone,
+ order,
+ sc.gfp_mask, false,
+ COMPACT_MODE_KSWAPD);
+ compaction = 1;
+ }
+
if (zone->all_unreclaimable)
continue;
- if (nr_slab == 0 && !zone_reclaimable(zone))
+ if (!compaction && nr_slab == 0 &&
+ !zone_reclaimable(zone))
zone->all_unreclaimable = 1;
/*
* If we've done a decent amount of scanning and
@@ -2297,7 +2447,7 @@
total_scanned > sc.nr_reclaimed + sc.nr_reclaimed / 2)
sc.may_writepage = 1;
- if (!zone_watermark_ok(zone, order,
+ if (!zone_watermark_ok_safe(zone, order,
high_wmark_pages(zone), end_zone, 0)) {
all_zones_ok = 0;
/*
@@ -2305,7 +2455,7 @@
* means that we have a GFP_ATOMIC allocation
* failure risk. Hurry up!
*/
- if (!zone_watermark_ok(zone, order,
+ if (!zone_watermark_ok_safe(zone, order,
min_wmark_pages(zone), end_zone, 0))
has_under_min_watermark_zone = 1;
} else {
@@ -2317,10 +2467,12 @@
* spectulatively avoid congestion waits
*/
zone_clear_flag(zone, ZONE_CONGESTED);
+ if (i <= *classzone_idx)
+ balanced += zone->present_pages;
}
}
- if (all_zones_ok)
+ if (all_zones_ok || (order && pgdat_balanced(pgdat, balanced, *classzone_idx)))
break; /* kswapd: all done */
/*
* OK, kswapd is getting into trouble. Take a nap, then take
@@ -2343,7 +2495,13 @@
break;
}
out:
- if (!all_zones_ok) {
+
+ /*
+ * order-0: All zones must meet high watermark for a balanced node
+ * high-order: Balanced zones must make up at least 25% of the node
+ * for the node to be balanced
+ */
+ if (!(all_zones_ok || (order && pgdat_balanced(pgdat, balanced, *classzone_idx)))) {
cond_resched();
try_to_freeze();
@@ -2368,7 +2526,88 @@
goto loop_again;
}
- return sc.nr_reclaimed;
+ /*
+ * If kswapd was reclaiming at a higher order, it has the option of
+ * sleeping without all zones being balanced. Before it does, it must
+ * ensure that the watermarks for order-0 on *all* zones are met and
+ * that the congestion flags are cleared. The congestion flag must
+ * be cleared as kswapd is the only mechanism that clears the flag
+ * and it is potentially going to sleep here.
+ */
+ if (order) {
+ for (i = 0; i <= end_zone; i++) {
+ struct zone *zone = pgdat->node_zones + i;
+
+ if (!populated_zone(zone))
+ continue;
+
+ if (zone->all_unreclaimable && priority != DEF_PRIORITY)
+ continue;
+
+ /* Confirm the zone is balanced for order-0 */
+ if (!zone_watermark_ok(zone, 0,
+ high_wmark_pages(zone), 0, 0)) {
+ order = sc.order = 0;
+ goto loop_again;
+ }
+
+ /* If balanced, clear the congested flag */
+ zone_clear_flag(zone, ZONE_CONGESTED);
+ }
+ }
+
+ /*
+ * Return the order we were reclaiming at so sleeping_prematurely()
+ * makes a decision on the order we were last reclaiming at. However,
+ * if another caller entered the allocator slow path while kswapd
+ * was awake, order will remain at the higher level
+ */
+ *classzone_idx = end_zone;
+ return order;
+}
+
+static void kswapd_try_to_sleep(pg_data_t *pgdat, int order, int classzone_idx)
+{
+ long remaining = 0;
+ DEFINE_WAIT(wait);
+
+ if (freezing(current) || kthread_should_stop())
+ return;
+
+ prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
+
+ /* Try to sleep for a short interval */
+ if (!sleeping_prematurely(pgdat, order, remaining, classzone_idx)) {
+ remaining = schedule_timeout(HZ/10);
+ finish_wait(&pgdat->kswapd_wait, &wait);
+ prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
+ }
+
+ /*
+ * After a short sleep, check if it was a premature sleep. If not, then
+ * go fully to sleep until explicitly woken up.
+ */
+ if (!sleeping_prematurely(pgdat, order, remaining, classzone_idx)) {
+ trace_mm_vmscan_kswapd_sleep(pgdat->node_id);
+
+ /*
+ * vmstat counters are not perfectly accurate and the estimated
+ * value for counters such as NR_FREE_PAGES can deviate from the
+ * true value by nr_online_cpus * threshold. To avoid the zone
+ * watermarks being breached while under pressure, we reduce the
+ * per-cpu vmstat threshold while kswapd is awake and restore
+ * them before going back to sleep.
+ */
+ set_pgdat_percpu_threshold(pgdat, calculate_normal_threshold);
+ schedule();
+ set_pgdat_percpu_threshold(pgdat, calculate_pressure_threshold);
+ } else {
+ if (remaining)
+ count_vm_event(KSWAPD_LOW_WMARK_HIT_QUICKLY);
+ else
+ count_vm_event(KSWAPD_HIGH_WMARK_HIT_QUICKLY);
+ }
+ finish_wait(&pgdat->kswapd_wait, &wait);
}
/*
@@ -2387,9 +2626,10 @@
static int kswapd(void *p)
{
unsigned long order;
+ int classzone_idx;
pg_data_t *pgdat = (pg_data_t*)p;
struct task_struct *tsk = current;
- DEFINE_WAIT(wait);
+
struct reclaim_state reclaim_state = {
.reclaimed_slab = 0,
};
@@ -2417,49 +2657,30 @@
set_freezable();
order = 0;
+ classzone_idx = MAX_NR_ZONES - 1;
for ( ; ; ) {
unsigned long new_order;
+ int new_classzone_idx;
int ret;
- prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
new_order = pgdat->kswapd_max_order;
+ new_classzone_idx = pgdat->classzone_idx;
pgdat->kswapd_max_order = 0;
- if (order < new_order) {
+ pgdat->classzone_idx = MAX_NR_ZONES - 1;
+ if (order < new_order || classzone_idx > new_classzone_idx) {
/*
* Don't sleep if someone wants a larger 'order'
- * allocation
+ * allocation or has tigher zone constraints
*/
order = new_order;
+ classzone_idx = new_classzone_idx;
} else {
- if (!freezing(current) && !kthread_should_stop()) {
- long remaining = 0;
-
- /* Try to sleep for a short interval */
- if (!sleeping_prematurely(pgdat, order, remaining)) {
- remaining = schedule_timeout(HZ/10);
- finish_wait(&pgdat->kswapd_wait, &wait);
- prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
- }
-
- /*
- * After a short sleep, check if it was a
- * premature sleep. If not, then go fully
- * to sleep until explicitly woken up
- */
- if (!sleeping_prematurely(pgdat, order, remaining)) {
- trace_mm_vmscan_kswapd_sleep(pgdat->node_id);
- schedule();
- } else {
- if (remaining)
- count_vm_event(KSWAPD_LOW_WMARK_HIT_QUICKLY);
- else
- count_vm_event(KSWAPD_HIGH_WMARK_HIT_QUICKLY);
- }
- }
-
+ kswapd_try_to_sleep(pgdat, order, classzone_idx);
order = pgdat->kswapd_max_order;
+ classzone_idx = pgdat->classzone_idx;
+ pgdat->kswapd_max_order = 0;
+ pgdat->classzone_idx = MAX_NR_ZONES - 1;
}
- finish_wait(&pgdat->kswapd_wait, &wait);
ret = try_to_freeze();
if (kthread_should_stop())
@@ -2471,7 +2692,7 @@
*/
if (!ret) {
trace_mm_vmscan_kswapd_wake(pgdat->node_id, order);
- balance_pgdat(pgdat, order);
+ order = balance_pgdat(pgdat, order, &classzone_idx);
}
}
return 0;
@@ -2480,23 +2701,26 @@
/*
* A zone is low on free memory, so wake its kswapd task to service it.
*/
-void wakeup_kswapd(struct zone *zone, int order)
+void wakeup_kswapd(struct zone *zone, int order, enum zone_type classzone_idx)
{
pg_data_t *pgdat;
if (!populated_zone(zone))
return;
- pgdat = zone->zone_pgdat;
- if (zone_watermark_ok(zone, order, low_wmark_pages(zone), 0, 0))
- return;
- if (pgdat->kswapd_max_order < order)
- pgdat->kswapd_max_order = order;
- trace_mm_vmscan_wakeup_kswapd(pgdat->node_id, zone_idx(zone), order);
if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL))
return;
+ pgdat = zone->zone_pgdat;
+ if (pgdat->kswapd_max_order < order) {
+ pgdat->kswapd_max_order = order;
+ pgdat->classzone_idx = min(pgdat->classzone_idx, classzone_idx);
+ }
if (!waitqueue_active(&pgdat->kswapd_wait))
return;
+ if (zone_watermark_ok_safe(zone, order, low_wmark_pages(zone), 0, 0))
+ return;
+
+ trace_mm_vmscan_wakeup_kswapd(pgdat->node_id, zone_idx(zone), order);
wake_up_interruptible(&pgdat->kswapd_wait);
}
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 312d728..0c3b504 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -83,7 +83,31 @@
#ifdef CONFIG_SMP
-static int calculate_threshold(struct zone *zone)
+int calculate_pressure_threshold(struct zone *zone)
+{
+ int threshold;
+ int watermark_distance;
+
+ /*
+ * As vmstats are not up to date, there is drift between the estimated
+ * and real values. For high thresholds and a high number of CPUs, it
+ * is possible for the min watermark to be breached while the estimated
+ * value looks fine. The pressure threshold is a reduced value such
+ * that even the maximum amount of drift will not accidentally breach
+ * the min watermark
+ */
+ watermark_distance = low_wmark_pages(zone) - min_wmark_pages(zone);
+ threshold = max(1, (int)(watermark_distance / num_online_cpus()));
+
+ /*
+ * Maximum threshold is 125
+ */
+ threshold = min(125, threshold);
+
+ return threshold;
+}
+
+int calculate_normal_threshold(struct zone *zone)
{
int threshold;
int mem; /* memory in 128 MB units */
@@ -142,7 +166,7 @@
for_each_populated_zone(zone) {
unsigned long max_drift, tolerate_drift;
- threshold = calculate_threshold(zone);
+ threshold = calculate_normal_threshold(zone);
for_each_online_cpu(cpu)
per_cpu_ptr(zone->pageset, cpu)->stat_threshold
@@ -161,6 +185,26 @@
}
}
+void set_pgdat_percpu_threshold(pg_data_t *pgdat,
+ int (*calculate_pressure)(struct zone *))
+{
+ struct zone *zone;
+ int cpu;
+ int threshold;
+ int i;
+
+ for (i = 0; i < pgdat->nr_zones; i++) {
+ zone = &pgdat->node_zones[i];
+ if (!zone->percpu_drift_mark)
+ continue;
+
+ threshold = (*calculate_pressure)(zone);
+ for_each_possible_cpu(cpu)
+ per_cpu_ptr(zone->pageset, cpu)->stat_threshold
+ = threshold;
+ }
+}
+
/*
* For use when we know that interrupts are disabled.
*/
@@ -836,6 +880,7 @@
"numa_local",
"numa_other",
#endif
+ "nr_anon_transparent_hugepages",
"nr_dirty_threshold",
"nr_dirty_background_threshold",
@@ -911,7 +956,7 @@
"\n scanned %lu"
"\n spanned %lu"
"\n present %lu",
- zone_nr_free_pages(zone),
+ zone_page_state(zone, NR_FREE_PAGES),
min_wmark_pages(zone),
low_wmark_pages(zone),
high_wmark_pages(zone),
diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c
index bb86d29..6da5dae 100644
--- a/net/ax25/af_ax25.c
+++ b/net/ax25/af_ax25.c
@@ -1392,7 +1392,7 @@
ax25_cb *ax25;
int err = 0;
- memset(fsa, 0, sizeof(fsa));
+ memset(fsa, 0, sizeof(*fsa));
lock_sock(sk);
ax25 = ax25_sk(sk);
diff --git a/net/core/dev.c b/net/core/dev.c
index 06d0e7b..54277df 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5523,34 +5523,6 @@
}
}
-/**
- * dev_txq_stats_fold - fold tx_queues stats
- * @dev: device to get statistics from
- * @stats: struct rtnl_link_stats64 to hold results
- */
-void dev_txq_stats_fold(const struct net_device *dev,
- struct rtnl_link_stats64 *stats)
-{
- u64 tx_bytes = 0, tx_packets = 0, tx_dropped = 0;
- unsigned int i;
- struct netdev_queue *txq;
-
- for (i = 0; i < dev->num_tx_queues; i++) {
- txq = netdev_get_tx_queue(dev, i);
- spin_lock_bh(&txq->_xmit_lock);
- tx_bytes += txq->tx_bytes;
- tx_packets += txq->tx_packets;
- tx_dropped += txq->tx_dropped;
- spin_unlock_bh(&txq->_xmit_lock);
- }
- if (tx_bytes || tx_packets || tx_dropped) {
- stats->tx_bytes = tx_bytes;
- stats->tx_packets = tx_packets;
- stats->tx_dropped = tx_dropped;
- }
-}
-EXPORT_SYMBOL(dev_txq_stats_fold);
-
/* Convert net_device_stats to rtnl_link_stats64. They have the same
* fields in the same order, with only the type differing.
*/
@@ -5594,7 +5566,6 @@
netdev_stats_to_stats64(storage, ops->ndo_get_stats(dev));
} else {
netdev_stats_to_stats64(storage, &dev->stats);
- dev_txq_stats_fold(dev, storage);
}
storage->rx_dropped += atomic_long_read(&dev->rx_dropped);
return storage;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 19d6c21..d31bb36 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -380,6 +380,8 @@
}
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
nf_conntrack_put(skb->nfct);
+#endif
+#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED
nf_conntrack_put_reasm(skb->nfct_reasm);
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c
index f9d7ac9..44d2b42 100644
--- a/net/ethernet/eth.c
+++ b/net/ethernet/eth.c
@@ -351,7 +351,7 @@
* @sizeof_priv: Size of additional driver-private structure to be allocated
* for this Ethernet device
* @txqs: The number of TX queues this device has.
- * @txqs: The number of RX queues this device has.
+ * @rxqs: The number of RX queues this device has.
*
* Fill in the fields of the device structure with Ethernet-generic
* values. Basically does everything except registering the device.
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 94b5bf1..5f8d242 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -401,6 +401,9 @@
goto drop;
}
+ if (skb->pkt_type != PACKET_HOST)
+ goto drop;
+
skb_forward_csum(skb);
/*
diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c
index 99abfb5..97c5b21 100644
--- a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c
+++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c
@@ -19,13 +19,15 @@
#include <linux/netfilter_ipv6.h>
#include <linux/netfilter_bridge.h>
+#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
#include <net/netfilter/nf_conntrack.h>
#include <net/netfilter/nf_conntrack_helper.h>
#include <net/netfilter/nf_conntrack_l4proto.h>
#include <net/netfilter/nf_conntrack_l3proto.h>
#include <net/netfilter/nf_conntrack_core.h>
-#include <net/netfilter/nf_conntrack_zones.h>
#include <net/netfilter/ipv6/nf_conntrack_ipv6.h>
+#endif
+#include <net/netfilter/nf_conntrack_zones.h>
#include <net/netfilter/ipv6/nf_defrag_ipv6.h>
static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum,
@@ -33,8 +35,10 @@
{
u16 zone = NF_CT_DEFAULT_ZONE;
+#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
if (skb->nfct)
zone = nf_ct_zone((struct nf_conn *)skb->nfct);
+#endif
#ifdef CONFIG_BRIDGE_NETFILTER
if (skb->nf_bridge &&
@@ -56,9 +60,11 @@
{
struct sk_buff *reasm;
+#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
/* Previously seen (loopback)? */
if (skb->nfct && !nf_ct_is_template((struct nf_conn *)skb->nfct))
return NF_ACCEPT;
+#endif
reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(hooknum, skb));
/* queued */
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index 5cb8d30..2b7eef3 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -972,7 +972,8 @@
free:
kfree_skb(skb2);
out:
- return err;
+ /* this avoids a loop in nfnetlink. */
+ return err == -EAGAIN ? -ENOBUFS : err;
}
#ifdef CONFIG_NF_NAT_NEEDED
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 0b9bb20..74c064c 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -808,7 +808,7 @@
goto error_call_jar;
}
- rxrpc_workqueue = create_workqueue("krxrpcd");
+ rxrpc_workqueue = alloc_workqueue("krxrpcd", 0, 1);
if (!rxrpc_workqueue) {
printk(KERN_NOTICE "RxRPC: Failed to allocate work queue\n");
goto error_work_queue;
diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c
index af9360d..84ce48e 100644
--- a/net/sched/sch_teql.c
+++ b/net/sched/sch_teql.c
@@ -59,6 +59,10 @@
struct net_device *dev;
struct Qdisc *slaves;
struct list_head master_list;
+ unsigned long tx_bytes;
+ unsigned long tx_packets;
+ unsigned long tx_errors;
+ unsigned long tx_dropped;
};
struct teql_sched_data
@@ -274,7 +278,6 @@
static netdev_tx_t teql_master_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct teql_master *master = netdev_priv(dev);
- struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
struct Qdisc *start, *q;
int busy;
int nores;
@@ -314,8 +317,8 @@
__netif_tx_unlock(slave_txq);
master->slaves = NEXT_SLAVE(q);
netif_wake_queue(dev);
- txq->tx_packets++;
- txq->tx_bytes += length;
+ master->tx_packets++;
+ master->tx_bytes += length;
return NETDEV_TX_OK;
}
__netif_tx_unlock(slave_txq);
@@ -342,10 +345,10 @@
netif_stop_queue(dev);
return NETDEV_TX_BUSY;
}
- dev->stats.tx_errors++;
+ master->tx_errors++;
drop:
- txq->tx_dropped++;
+ master->tx_dropped++;
dev_kfree_skb(skb);
return NETDEV_TX_OK;
}
@@ -398,6 +401,18 @@
return 0;
}
+static struct rtnl_link_stats64 *teql_master_stats64(struct net_device *dev,
+ struct rtnl_link_stats64 *stats)
+{
+ struct teql_master *m = netdev_priv(dev);
+
+ stats->tx_packets = m->tx_packets;
+ stats->tx_bytes = m->tx_bytes;
+ stats->tx_errors = m->tx_errors;
+ stats->tx_dropped = m->tx_dropped;
+ return stats;
+}
+
static int teql_master_mtu(struct net_device *dev, int new_mtu)
{
struct teql_master *m = netdev_priv(dev);
@@ -422,6 +437,7 @@
.ndo_open = teql_master_open,
.ndo_stop = teql_master_close,
.ndo_start_xmit = teql_master_xmit,
+ .ndo_get_stats64 = teql_master_stats64,
.ndo_change_mtu = teql_master_mtu,
};
diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c
index 75ee993..9576f35 100644
--- a/net/sunrpc/auth_gss/gss_krb5_crypto.c
+++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c
@@ -137,7 +137,7 @@
ms_usage = 13;
break;
default:
- return EINVAL;;
+ return -EINVAL;
}
salt[0] = (ms_usage >> 0) & 0xff;
salt[1] = (ms_usage >> 8) & 0xff;
diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c
index dec2a6f..bcdae78 100644
--- a/net/sunrpc/auth_gss/svcauth_gss.c
+++ b/net/sunrpc/auth_gss/svcauth_gss.c
@@ -67,7 +67,6 @@
#define RSI_HASHBITS 6
#define RSI_HASHMAX (1<<RSI_HASHBITS)
-#define RSI_HASHMASK (RSI_HASHMAX-1)
struct rsi {
struct cache_head h;
@@ -319,7 +318,6 @@
#define RSC_HASHBITS 10
#define RSC_HASHMAX (1<<RSC_HASHBITS)
-#define RSC_HASHMASK (RSC_HASHMAX-1)
#define GSS_SEQ_WIN 128
diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
index e433e75..72ad836 100644
--- a/net/sunrpc/cache.c
+++ b/net/sunrpc/cache.c
@@ -37,7 +37,7 @@
#define RPCDBG_FACILITY RPCDBG_CACHE
-static void cache_defer_req(struct cache_req *req, struct cache_head *item);
+static bool cache_defer_req(struct cache_req *req, struct cache_head *item);
static void cache_revisit_request(struct cache_head *item);
static void cache_init(struct cache_head *h)
@@ -128,6 +128,7 @@
{
head->expiry_time = expiry;
head->last_refresh = seconds_since_boot();
+ smp_wmb(); /* paired with smp_rmb() in cache_is_valid() */
set_bit(CACHE_VALID, &head->flags);
}
@@ -208,11 +209,36 @@
/* entry is valid */
if (test_bit(CACHE_NEGATIVE, &h->flags))
return -ENOENT;
- else
+ else {
+ /*
+ * In combination with write barrier in
+ * sunrpc_cache_update, ensures that anyone
+ * using the cache entry after this sees the
+ * updated contents:
+ */
+ smp_rmb();
return 0;
+ }
}
}
+static int try_to_negate_entry(struct cache_detail *detail, struct cache_head *h)
+{
+ int rv;
+
+ write_lock(&detail->hash_lock);
+ rv = cache_is_valid(detail, h);
+ if (rv != -EAGAIN) {
+ write_unlock(&detail->hash_lock);
+ return rv;
+ }
+ set_bit(CACHE_NEGATIVE, &h->flags);
+ cache_fresh_locked(h, seconds_since_boot()+CACHE_NEW_EXPIRY);
+ write_unlock(&detail->hash_lock);
+ cache_fresh_unlocked(h, detail);
+ return -ENOENT;
+}
+
/*
* This is the generic cache management routine for all
* the authentication caches.
@@ -251,14 +277,8 @@
case -EINVAL:
clear_bit(CACHE_PENDING, &h->flags);
cache_revisit_request(h);
- if (rv == -EAGAIN) {
- set_bit(CACHE_NEGATIVE, &h->flags);
- cache_fresh_locked(h, seconds_since_boot()+CACHE_NEW_EXPIRY);
- cache_fresh_unlocked(h, detail);
- rv = -ENOENT;
- }
+ rv = try_to_negate_entry(detail, h);
break;
-
case -EAGAIN:
clear_bit(CACHE_PENDING, &h->flags);
cache_revisit_request(h);
@@ -268,9 +288,11 @@
}
if (rv == -EAGAIN) {
- cache_defer_req(rqstp, h);
- if (!test_bit(CACHE_PENDING, &h->flags)) {
- /* Request is not deferred */
+ if (!cache_defer_req(rqstp, h)) {
+ /*
+ * Request was not deferred; handle it as best
+ * we can ourselves:
+ */
rv = cache_is_valid(detail, h);
if (rv == -EAGAIN)
rv = -ETIMEDOUT;
@@ -618,18 +640,19 @@
discard->revisit(discard, 1);
}
-static void cache_defer_req(struct cache_req *req, struct cache_head *item)
+/* Return true if and only if a deferred request is queued. */
+static bool cache_defer_req(struct cache_req *req, struct cache_head *item)
{
struct cache_deferred_req *dreq;
if (req->thread_wait) {
cache_wait_req(req, item);
if (!test_bit(CACHE_PENDING, &item->flags))
- return;
+ return false;
}
dreq = req->defer(req);
if (dreq == NULL)
- return;
+ return false;
setup_deferral(dreq, item, 1);
if (!test_bit(CACHE_PENDING, &item->flags))
/* Bit could have been cleared before we managed to
@@ -638,6 +661,7 @@
cache_revisit_request(item);
cache_limit_defers();
+ return true;
}
static void cache_revisit_request(struct cache_head *item)
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 0e659c6..08e05a8 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -1001,6 +1001,7 @@
rqstp->rq_splice_ok = 1;
/* Will be turned off only when NFSv4 Sessions are used */
rqstp->rq_usedeferral = 1;
+ rqstp->rq_dropme = false;
/* Setup reply header */
rqstp->rq_xprt->xpt_ops->xpo_prep_reply_hdr(rqstp);
@@ -1102,7 +1103,7 @@
*statp = procp->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp);
/* Encode reply */
- if (*statp == rpc_drop_reply) {
+ if (rqstp->rq_dropme) {
if (procp->pc_release)
procp->pc_release(rqstp, NULL, rqstp->rq_resp);
goto dropit;
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index 3f2c555..ab86b79 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -13,6 +13,7 @@
#include <linux/sunrpc/stats.h>
#include <linux/sunrpc/svc_xprt.h>
#include <linux/sunrpc/svcsock.h>
+#include <linux/sunrpc/xprt.h>
#define RPCDBG_FACILITY RPCDBG_SVCXPRT
@@ -128,6 +129,9 @@
if (test_bit(XPT_CACHE_AUTH, &xprt->xpt_flags))
svcauth_unix_info_release(xprt);
put_net(xprt->xpt_net);
+ /* See comment on corresponding get in xs_setup_bc_tcp(): */
+ if (xprt->xpt_bc_xprt)
+ xprt_put(xprt->xpt_bc_xprt);
xprt->xpt_ops->xpo_free(xprt);
module_put(owner);
}
@@ -303,6 +307,15 @@
list_del(&rqstp->rq_list);
}
+static bool svc_xprt_has_something_to_do(struct svc_xprt *xprt)
+{
+ if (xprt->xpt_flags & ((1<<XPT_CONN)|(1<<XPT_CLOSE)))
+ return true;
+ if (xprt->xpt_flags & ((1<<XPT_DATA)|(1<<XPT_DEFERRED)))
+ return xprt->xpt_ops->xpo_has_wspace(xprt);
+ return false;
+}
+
/*
* Queue up a transport with data pending. If there are idle nfsd
* processes, wake 'em up.
@@ -315,8 +328,7 @@
struct svc_rqst *rqstp;
int cpu;
- if (!(xprt->xpt_flags &
- ((1<<XPT_CONN)|(1<<XPT_DATA)|(1<<XPT_CLOSE)|(1<<XPT_DEFERRED))))
+ if (!svc_xprt_has_something_to_do(xprt))
return;
cpu = get_cpu();
@@ -343,28 +355,7 @@
dprintk("svc: transport %p busy, not enqueued\n", xprt);
goto out_unlock;
}
- BUG_ON(xprt->xpt_pool != NULL);
- xprt->xpt_pool = pool;
- /* Handle pending connection */
- if (test_bit(XPT_CONN, &xprt->xpt_flags))
- goto process;
-
- /* Handle close in-progress */
- if (test_bit(XPT_CLOSE, &xprt->xpt_flags))
- goto process;
-
- /* Check if we have space to reply to a request */
- if (!xprt->xpt_ops->xpo_has_wspace(xprt)) {
- /* Don't enqueue while not enough space for reply */
- dprintk("svc: no write space, transport %p not enqueued\n",
- xprt);
- xprt->xpt_pool = NULL;
- clear_bit(XPT_BUSY, &xprt->xpt_flags);
- goto out_unlock;
- }
-
- process:
if (!list_empty(&pool->sp_threads)) {
rqstp = list_entry(pool->sp_threads.next,
struct svc_rqst,
@@ -381,13 +372,11 @@
rqstp->rq_reserved = serv->sv_max_mesg;
atomic_add(rqstp->rq_reserved, &xprt->xpt_reserved);
pool->sp_stats.threads_woken++;
- BUG_ON(xprt->xpt_pool != pool);
wake_up(&rqstp->rq_wait);
} else {
dprintk("svc: transport %p put into queue\n", xprt);
list_add_tail(&xprt->xpt_ready, &pool->sp_sockets);
pool->sp_stats.sockets_queued++;
- BUG_ON(xprt->xpt_pool != pool);
}
out_unlock:
@@ -426,7 +415,6 @@
void svc_xprt_received(struct svc_xprt *xprt)
{
BUG_ON(!test_bit(XPT_BUSY, &xprt->xpt_flags));
- xprt->xpt_pool = NULL;
/* As soon as we clear busy, the xprt could be closed and
* 'put', so we need a reference to call svc_xprt_enqueue with:
*/
@@ -722,7 +710,10 @@
if (test_bit(XPT_CLOSE, &xprt->xpt_flags)) {
dprintk("svc_recv: found XPT_CLOSE\n");
svc_delete_xprt(xprt);
- } else if (test_bit(XPT_LISTENER, &xprt->xpt_flags)) {
+ /* Leave XPT_BUSY set on the dead xprt: */
+ goto out;
+ }
+ if (test_bit(XPT_LISTENER, &xprt->xpt_flags)) {
struct svc_xprt *newxpt;
newxpt = xprt->xpt_ops->xpo_accept(xprt);
if (newxpt) {
@@ -747,28 +738,23 @@
spin_unlock_bh(&serv->sv_lock);
svc_xprt_received(newxpt);
}
- svc_xprt_received(xprt);
- } else {
+ } else if (xprt->xpt_ops->xpo_has_wspace(xprt)) {
dprintk("svc: server %p, pool %u, transport %p, inuse=%d\n",
rqstp, pool->sp_id, xprt,
atomic_read(&xprt->xpt_ref.refcount));
rqstp->rq_deferred = svc_deferred_dequeue(xprt);
- if (rqstp->rq_deferred) {
- svc_xprt_received(xprt);
+ if (rqstp->rq_deferred)
len = svc_deferred_recv(rqstp);
- } else {
+ else
len = xprt->xpt_ops->xpo_recvfrom(rqstp);
- svc_xprt_received(xprt);
- }
dprintk("svc: got len=%d\n", len);
}
+ svc_xprt_received(xprt);
/* No data, incomplete (TCP) read, or accept() */
- if (len == 0 || len == -EAGAIN) {
- rqstp->rq_res.len = 0;
- svc_xprt_release(rqstp);
- return -EAGAIN;
- }
+ if (len == 0 || len == -EAGAIN)
+ goto out;
+
clear_bit(XPT_OLD, &xprt->xpt_flags);
rqstp->rq_secure = svc_port_is_privileged(svc_addr(rqstp));
@@ -777,6 +763,10 @@
if (serv->sv_stats)
serv->sv_stats->netcnt++;
return len;
+out:
+ rqstp->rq_res.len = 0;
+ svc_xprt_release(rqstp);
+ return -EAGAIN;
}
EXPORT_SYMBOL_GPL(svc_recv);
@@ -935,7 +925,12 @@
if (test_and_set_bit(XPT_BUSY, &xprt->xpt_flags))
/* someone else will have to effect the close */
return;
-
+ /*
+ * We expect svc_close_xprt() to work even when no threads are
+ * running (e.g., while configuring the server before starting
+ * any threads), so if the transport isn't busy, we delete
+ * it ourself:
+ */
svc_delete_xprt(xprt);
}
EXPORT_SYMBOL_GPL(svc_close_xprt);
@@ -945,16 +940,16 @@
struct svc_xprt *xprt;
struct svc_xprt *tmp;
+ /*
+ * The server is shutting down, and no more threads are running.
+ * svc_xprt_enqueue() might still be running, but at worst it
+ * will re-add the xprt to sp_sockets, which will soon get
+ * freed. So we don't bother with any more locking, and don't
+ * leave the close to the (nonexistent) server threads:
+ */
list_for_each_entry_safe(xprt, tmp, xprt_list, xpt_list) {
set_bit(XPT_CLOSE, &xprt->xpt_flags);
- if (test_bit(XPT_BUSY, &xprt->xpt_flags)) {
- /* Waiting to be processed, but no threads left,
- * So just remove it from the waiting list
- */
- list_del_init(&xprt->xpt_ready);
- clear_bit(XPT_BUSY, &xprt->xpt_flags);
- }
- svc_close_xprt(xprt);
+ svc_delete_xprt(xprt);
}
}
@@ -1028,6 +1023,7 @@
}
svc_xprt_get(rqstp->rq_xprt);
dr->xprt = rqstp->rq_xprt;
+ rqstp->rq_dropme = true;
dr->handle.revisit = svc_revisit;
return &dr->handle;
@@ -1065,14 +1061,13 @@
if (!test_bit(XPT_DEFERRED, &xprt->xpt_flags))
return NULL;
spin_lock(&xprt->xpt_lock);
- clear_bit(XPT_DEFERRED, &xprt->xpt_flags);
if (!list_empty(&xprt->xpt_deferred)) {
dr = list_entry(xprt->xpt_deferred.next,
struct svc_deferred_req,
handle.recent);
list_del_init(&dr->handle.recent);
- set_bit(XPT_DEFERRED, &xprt->xpt_flags);
- }
+ } else
+ clear_bit(XPT_DEFERRED, &xprt->xpt_flags);
spin_unlock(&xprt->xpt_lock);
return dr;
}
diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c
index 4e9393c..7963569 100644
--- a/net/sunrpc/svcauth.c
+++ b/net/sunrpc/svcauth.c
@@ -118,7 +118,6 @@
#define DN_HASHBITS 6
#define DN_HASHMAX (1<<DN_HASHBITS)
-#define DN_HASHMASK (DN_HASHMAX-1)
static struct hlist_head auth_domain_table[DN_HASHMAX];
static spinlock_t auth_domain_lock =
diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c
index 560677d..30916b0 100644
--- a/net/sunrpc/svcauth_unix.c
+++ b/net/sunrpc/svcauth_unix.c
@@ -30,7 +30,9 @@
struct unix_domain {
struct auth_domain h;
+#ifdef CONFIG_NFSD_DEPRECATED
int addr_changes;
+#endif /* CONFIG_NFSD_DEPRECATED */
/* other stuff later */
};
@@ -64,7 +66,9 @@
return NULL;
}
new->h.flavour = &svcauth_unix;
+#ifdef CONFIG_NFSD_DEPRECATED
new->addr_changes = 0;
+#endif /* CONFIG_NFSD_DEPRECATED */
rv = auth_domain_lookup(name, &new->h);
}
}
@@ -85,14 +89,15 @@
*/
#define IP_HASHBITS 8
#define IP_HASHMAX (1<<IP_HASHBITS)
-#define IP_HASHMASK (IP_HASHMAX-1)
struct ip_map {
struct cache_head h;
char m_class[8]; /* e.g. "nfsd" */
struct in6_addr m_addr;
struct unix_domain *m_client;
+#ifdef CONFIG_NFSD_DEPRECATED
int m_add_change;
+#endif /* CONFIG_NFSD_DEPRECATED */
};
static void ip_map_put(struct kref *kref)
@@ -146,7 +151,9 @@
kref_get(&item->m_client->h.ref);
new->m_client = item->m_client;
+#ifdef CONFIG_NFSD_DEPRECATED
new->m_add_change = item->m_add_change;
+#endif /* CONFIG_NFSD_DEPRECATED */
}
static struct cache_head *ip_map_alloc(void)
{
@@ -331,6 +338,7 @@
ip.h.flags = 0;
if (!udom)
set_bit(CACHE_NEGATIVE, &ip.h.flags);
+#ifdef CONFIG_NFSD_DEPRECATED
else {
ip.m_add_change = udom->addr_changes;
/* if this is from the legacy set_client system call,
@@ -339,6 +347,7 @@
if (expiry == NEVER)
ip.m_add_change++;
}
+#endif /* CONFIG_NFSD_DEPRECATED */
ip.h.expiry_time = expiry;
ch = sunrpc_cache_update(cd, &ip.h, &ipm->h,
hash_str(ipm->m_class, IP_HASHBITS) ^
@@ -358,6 +367,7 @@
return __ip_map_update(sn->ip_map_cache, ipm, udom, expiry);
}
+#ifdef CONFIG_NFSD_DEPRECATED
int auth_unix_add_addr(struct net *net, struct in6_addr *addr, struct auth_domain *dom)
{
struct unix_domain *udom;
@@ -402,8 +412,7 @@
return NULL;
if ((ipm->m_client->addr_changes - ipm->m_add_change) >0) {
- if (test_and_set_bit(CACHE_NEGATIVE, &ipm->h.flags) == 0)
- auth_domain_put(&ipm->m_client->h);
+ sunrpc_invalidate(&ipm->h, sn->ip_map_cache);
rv = NULL;
} else {
rv = &ipm->m_client->h;
@@ -413,6 +422,7 @@
return rv;
}
EXPORT_SYMBOL_GPL(auth_unix_lookup);
+#endif /* CONFIG_NFSD_DEPRECATED */
void svcauth_unix_purge(void)
{
@@ -497,7 +507,6 @@
*/
#define GID_HASHBITS 8
#define GID_HASHMAX (1<<GID_HASHBITS)
-#define GID_HASHMASK (GID_HASHMAX - 1)
struct unix_gid {
struct cache_head h;
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index d265aa7..7bd3bbb 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -331,19 +331,21 @@
len = onelen;
break;
}
- if (toclose && strcmp(toclose, buf + len) == 0)
+ if (toclose && strcmp(toclose, buf + len) == 0) {
closesk = svsk;
- else
+ svc_xprt_get(&closesk->sk_xprt);
+ } else
len += onelen;
}
spin_unlock_bh(&serv->sv_lock);
- if (closesk)
+ if (closesk) {
/* Should unregister with portmap, but you cannot
* unregister just one protocol...
*/
svc_close_xprt(&closesk->sk_xprt);
- else if (toclose)
+ svc_xprt_put(&closesk->sk_xprt);
+ } else if (toclose)
return -ENOENT;
return len;
}
@@ -992,15 +994,17 @@
vec[0] = rqstp->rq_arg.head[0];
} else {
/* REPLY */
- if (svsk->sk_bc_xprt)
- req = xprt_lookup_rqst(svsk->sk_bc_xprt, xid);
+ struct rpc_xprt *bc_xprt = svsk->sk_xprt.xpt_bc_xprt;
+
+ if (bc_xprt)
+ req = xprt_lookup_rqst(bc_xprt, xid);
if (!req) {
printk(KERN_NOTICE
"%s: Got unrecognized reply: "
- "calldir 0x%x sk_bc_xprt %p xid %08x\n",
+ "calldir 0x%x xpt_bc_xprt %p xid %08x\n",
__func__, ntohl(calldir),
- svsk->sk_bc_xprt, xid);
+ bc_xprt, xid);
vec[0] = rqstp->rq_arg.head[0];
goto out;
}
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 4c8f18a..856274d 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -965,6 +965,7 @@
xprt = kzalloc(size, GFP_KERNEL);
if (xprt == NULL)
goto out;
+ kref_init(&xprt->kref);
xprt->max_reqs = max_req;
xprt->slot = kcalloc(max_req, sizeof(struct rpc_rqst), GFP_KERNEL);
@@ -1101,8 +1102,10 @@
-PTR_ERR(xprt));
return xprt;
}
+ if (test_and_set_bit(XPRT_INITIALIZED, &xprt->state))
+ /* ->setup returned a pre-initialized xprt: */
+ return xprt;
- kref_init(&xprt->kref);
spin_lock_init(&xprt->transport_lock);
spin_lock_init(&xprt->reserve_lock);
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 96549df..c431f5a 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -2359,6 +2359,15 @@
struct svc_sock *bc_sock;
struct rpc_xprt *ret;
+ if (args->bc_xprt->xpt_bc_xprt) {
+ /*
+ * This server connection already has a backchannel
+ * export; we can't create a new one, as we wouldn't be
+ * able to match replies based on xid any more. So,
+ * reuse the already-existing one:
+ */
+ return args->bc_xprt->xpt_bc_xprt;
+ }
xprt = xs_setup_xprt(args, xprt_tcp_slot_table_entries);
if (IS_ERR(xprt))
return xprt;
@@ -2375,16 +2384,6 @@
xprt->reestablish_timeout = 0;
xprt->idle_timeout = 0;
- /*
- * The backchannel uses the same socket connection as the
- * forechannel
- */
- xprt->bc_xprt = args->bc_xprt;
- bc_sock = container_of(args->bc_xprt, struct svc_sock, sk_xprt);
- bc_sock->sk_bc_xprt = xprt;
- transport->sock = bc_sock->sk_sock;
- transport->inet = bc_sock->sk_sk;
-
xprt->ops = &bc_tcp_ops;
switch (addr->sa_family) {
@@ -2407,6 +2406,20 @@
xprt->address_strings[RPC_DISPLAY_PROTO]);
/*
+ * Once we've associated a backchannel xprt with a connection,
+ * we want to keep it around as long as long as the connection
+ * lasts, in case we need to start using it for a backchannel
+ * again; this reference won't be dropped until bc_xprt is
+ * destroyed.
+ */
+ xprt_get(xprt);
+ args->bc_xprt->xpt_bc_xprt = xprt;
+ xprt->bc_xprt = args->bc_xprt;
+ bc_sock = container_of(args->bc_xprt, struct svc_sock, sk_xprt);
+ transport->sock = bc_sock->sk_sock;
+ transport->inet = bc_sock->sk_sk;
+
+ /*
* Since we don't want connections for the backchannel, we set
* the xprt status to connected
*/
@@ -2415,6 +2428,7 @@
if (try_module_get(THIS_MODULE))
return xprt;
+ xprt_put(xprt);
ret = ERR_PTR(-EINVAL);
out_err:
xprt_free(xprt);
diff --git a/sound/pci/cs5535audio/cs5535audio_pm.c b/sound/pci/cs5535audio/cs5535audio_pm.c
index a3301cc..185b000 100644
--- a/sound/pci/cs5535audio/cs5535audio_pm.c
+++ b/sound/pci/cs5535audio/cs5535audio_pm.c
@@ -90,12 +90,7 @@
int i;
pci_set_power_state(pci, PCI_D0);
- if (pci_restore_state(pci) < 0) {
- printk(KERN_ERR "cs5535audio: pci_restore_state failed, "
- "disabling device\n");
- snd_card_disconnect(card);
- return -EIO;
- }
+ pci_restore_state(pci);
if (pci_enable_device(pci) < 0) {
printk(KERN_ERR "cs5535audio: pci_enable_device failed, "
"disabling device\n");
diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index 52462ae..e032716 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -61,6 +61,9 @@
-r::
--realtime=::
Collect data with this RT SCHED_FIFO priority.
+-D::
+--no-delay::
+ Collect data without buffering.
-A::
--append::
Append to the output file to do incremental profiling.
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 7069bd3..df6064a 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -49,6 +49,7 @@
static const char *output_name = "perf.data";
static int group = 0;
static int realtime_prio = 0;
+static bool nodelay = false;
static bool raw_samples = false;
static bool sample_id_all_avail = true;
static bool system_wide = false;
@@ -307,6 +308,11 @@
attr->sample_type |= PERF_SAMPLE_CPU;
}
+ if (nodelay) {
+ attr->watermark = 0;
+ attr->wakeup_events = 1;
+ }
+
attr->mmap = track;
attr->comm = track;
attr->inherit = !no_inherit;
@@ -331,9 +337,6 @@
else if (err == ENODEV && cpu_list) {
die("No such device - did you specify"
" an out-of-range profile CPU?\n");
- } else if (err == ENOENT) {
- die("%s event is not supported. ",
- event_name(evsel));
} else if (err == EINVAL && sample_id_all_avail) {
/*
* Old kernel, no attr->sample_id_type_all field
@@ -480,6 +483,7 @@
process_buildids();
perf_header__write(&session->header, output, true);
perf_session__delete(session);
+ perf_evsel_list__delete();
symbol__exit();
}
}
@@ -845,6 +849,8 @@
"record events on existing thread id"),
OPT_INTEGER('r', "realtime", &realtime_prio,
"collect data with this RT SCHED_FIFO priority"),
+ OPT_BOOLEAN('D', "no-delay", &nodelay,
+ "collect data without buffering"),
OPT_BOOLEAN('R', "raw-samples", &raw_samples,
"collect raw sample records from all opened counters"),
OPT_BOOLEAN('a', "all-cpus", &system_wide,
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index abd4b84..29e7ffd 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -1843,15 +1843,15 @@
"-f",
"-m", "1024",
"-c", "1",
- "-e", "sched:sched_switch:r",
- "-e", "sched:sched_stat_wait:r",
- "-e", "sched:sched_stat_sleep:r",
- "-e", "sched:sched_stat_iowait:r",
- "-e", "sched:sched_stat_runtime:r",
- "-e", "sched:sched_process_exit:r",
- "-e", "sched:sched_process_fork:r",
- "-e", "sched:sched_wakeup:r",
- "-e", "sched:sched_migrate_task:r",
+ "-e", "sched:sched_switch",
+ "-e", "sched:sched_stat_wait",
+ "-e", "sched:sched_stat_sleep",
+ "-e", "sched:sched_stat_iowait",
+ "-e", "sched:sched_stat_runtime",
+ "-e", "sched:sched_process_exit",
+ "-e", "sched:sched_process_fork",
+ "-e", "sched:sched_wakeup",
+ "-e", "sched:sched_migrate_task",
};
static int __cmd_record(int argc, const char **argv)
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index c385a63..0ff11d9 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -743,6 +743,7 @@
out_free_fd:
list_for_each_entry(pos, &evsel_list, node)
perf_evsel__free_stat_priv(pos);
+ perf_evsel_list__delete();
out:
thread_map__delete(threads);
threads = NULL;
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 6ce4042..05344c6 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -1247,8 +1247,6 @@
die("Permission error - are you root?\n"
"\t Consider tweaking"
" /proc/sys/kernel/perf_event_paranoid.\n");
- if (err == ENOENT)
- die("%s event is not supported. ", event_name(evsel));
/*
* If it's cycles then fall back to hrtimer
* based cpu-clock-tick sw counter, which
@@ -1473,6 +1471,8 @@
pos->attr.sample_period = default_interval;
}
+ sym_evsel = list_entry(evsel_list.next, struct perf_evsel, node);
+
symbol_conf.priv_size = (sizeof(struct sym_entry) +
(nr_counters + 1) * sizeof(unsigned long));
@@ -1490,6 +1490,7 @@
out_free_fd:
list_for_each_entry(pos, &evsel_list, node)
perf_evsel__free_mmap(pos);
+ perf_evsel_list__delete();
return status;
}
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index 5b1ecd6..595d0f4 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -286,8 +286,6 @@
status = p->fn(argc, argv, prefix);
exit_browser(status);
- perf_evsel_list__delete();
-
if (status)
return status & 0xff;
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 7f68625..f29abeb 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -104,8 +104,26 @@
inline int kvm_is_mmio_pfn(pfn_t pfn)
{
if (pfn_valid(pfn)) {
- struct page *page = compound_head(pfn_to_page(pfn));
- return PageReserved(page);
+ int reserved;
+ struct page *tail = pfn_to_page(pfn);
+ struct page *head = compound_trans_head(tail);
+ reserved = PageReserved(head);
+ if (head != tail) {
+ /*
+ * "head" is not a dangling pointer
+ * (compound_trans_head takes care of that)
+ * but the hugepage may have been splitted
+ * from under us (and we may not hold a
+ * reference count on the head page so it can
+ * be reused before we run PageReferenced), so
+ * we've to check PageTail before returning
+ * what we just read.
+ */
+ smp_rmb();
+ if (PageTail(tail))
+ return reserved;
+ }
+ return PageReserved(tail);
}
return true;
@@ -352,6 +370,22 @@
return young;
}
+static int kvm_mmu_notifier_test_young(struct mmu_notifier *mn,
+ struct mm_struct *mm,
+ unsigned long address)
+{
+ struct kvm *kvm = mmu_notifier_to_kvm(mn);
+ int young, idx;
+
+ idx = srcu_read_lock(&kvm->srcu);
+ spin_lock(&kvm->mmu_lock);
+ young = kvm_test_age_hva(kvm, address);
+ spin_unlock(&kvm->mmu_lock);
+ srcu_read_unlock(&kvm->srcu, idx);
+
+ return young;
+}
+
static void kvm_mmu_notifier_release(struct mmu_notifier *mn,
struct mm_struct *mm)
{
@@ -368,6 +402,7 @@
.invalidate_range_start = kvm_mmu_notifier_invalidate_range_start,
.invalidate_range_end = kvm_mmu_notifier_invalidate_range_end,
.clear_flush_young = kvm_mmu_notifier_clear_flush_young,
+ .test_young = kvm_mmu_notifier_test_young,
.change_pte = kvm_mmu_notifier_change_pte,
.release = kvm_mmu_notifier_release,
};