Merge branch 'for-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth
Johan Hedberg says:
====================
pull request: bluetooth 2014-12-19
Here's one more pull request for 3.19. It contains the socket type
verification fixes from Al Viro as well as an skb double-free fix for
6lowpan from Jukka Rissanen.
Please let me know if there are any issues pulling. Thanks.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/.mailmap b/.mailmap
index 1ad6873..ada8ad6 100644
--- a/.mailmap
+++ b/.mailmap
@@ -17,7 +17,7 @@
Al Viro <viro@ftp.linux.org.uk>
Al Viro <viro@zenIV.linux.org.uk>
Andreas Herrmann <aherrman@de.ibm.com>
-Andrew Morton <akpm@osdl.org>
+Andrew Morton <akpm@linux-foundation.org>
Andrew Vasquez <andrew.vasquez@qlogic.com>
Andy Adamson <andros@citi.umich.edu>
Archit Taneja <archit@ti.com>
@@ -102,6 +102,8 @@
Rui Saraiva <rmps@joel.ist.utl.pt>
Sachin P Sant <ssant@in.ibm.com>
Sam Ravnborg <sam@mars.ravnborg.org>
+Santosh Shilimkar <ssantosh@kernel.org>
+Santosh Shilimkar <santosh.shilimkar@oracle.org>
Sascha Hauer <s.hauer@pengutronix.de>
S.Çağlar Onur <caglar@pardus.org.tr>
Shiraz Hashim <shiraz.linux.kernel@gmail.com> <shiraz.hashim@st.com>
diff --git a/Documentation/ABI/testing/sysfs-platform-dell-laptop b/Documentation/ABI/testing/sysfs-platform-dell-laptop
new file mode 100644
index 0000000..7969443
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-platform-dell-laptop
@@ -0,0 +1,60 @@
+What: /sys/class/leds/dell::kbd_backlight/als_setting
+Date: December 2014
+KernelVersion: 3.19
+Contact: Gabriele Mazzotta <gabriele.mzt@gmail.com>,
+ Pali Rohár <pali.rohar@gmail.com>
+Description:
+ This file allows to control the automatic keyboard
+ illumination mode on some systems that have an ambient
+ light sensor. Write 1 to this file to enable the auto
+ mode, 0 to disable it.
+
+What: /sys/class/leds/dell::kbd_backlight/start_triggers
+Date: December 2014
+KernelVersion: 3.19
+Contact: Gabriele Mazzotta <gabriele.mzt@gmail.com>,
+ Pali Rohár <pali.rohar@gmail.com>
+Description:
+ This file allows to control the input triggers that
+ turn on the keyboard backlight illumination that is
+ disabled because of inactivity.
+ Read the file to see the triggers available. The ones
+ enabled are preceded by '+', those disabled by '-'.
+
+ To enable a trigger, write its name preceded by '+' to
+ this file. To disable a trigger, write its name preceded
+ by '-' instead.
+
+ For example, to enable the keyboard as trigger run:
+ echo +keyboard > /sys/class/leds/dell::kbd_backlight/start_triggers
+ To disable it:
+ echo -keyboard > /sys/class/leds/dell::kbd_backlight/start_triggers
+
+ Note that not all the available triggers can be configured.
+
+What: /sys/class/leds/dell::kbd_backlight/stop_timeout
+Date: December 2014
+KernelVersion: 3.19
+Contact: Gabriele Mazzotta <gabriele.mzt@gmail.com>,
+ Pali Rohár <pali.rohar@gmail.com>
+Description:
+ This file allows to specify the interval after which the
+ keyboard illumination is disabled because of inactivity.
+ The timeouts are expressed in seconds, minutes, hours and
+ days, for which the symbols are 's', 'm', 'h' and 'd'
+ respectively.
+
+ To configure the timeout, write to this file a value along
+ with any the above units. If no unit is specified, the value
+ is assumed to be expressed in seconds.
+
+ For example, to set the timeout to 10 minutes run:
+ echo 10m > /sys/class/leds/dell::kbd_backlight/stop_timeout
+
+ Note that when this file is read, the returned value might be
+ expressed in a different unit than the one used when the timeout
+ was set.
+
+ Also note that only some timeouts are supported and that
+ some systems might fall back to a specific timeout in case
+ an invalid timeout is written to this file.
diff --git a/Documentation/DocBook/media/v4l/compat.xml b/Documentation/DocBook/media/v4l/compat.xml
index 0a2debf..350dfb3 100644
--- a/Documentation/DocBook/media/v4l/compat.xml
+++ b/Documentation/DocBook/media/v4l/compat.xml
@@ -2579,6 +2579,18 @@
</orderedlist>
</section>
+ <section>
+ <title>V4L2 in Linux 3.19</title>
+ <orderedlist>
+ <listitem>
+ <para>Rewrote Colorspace chapter, added new &v4l2-ycbcr-encoding;
+and &v4l2-quantization; fields to &v4l2-pix-format;, &v4l2-pix-format-mplane;
+and &v4l2-mbus-framefmt;.
+ </para>
+ </listitem>
+ </orderedlist>
+ </section>
+
<section id="other">
<title>Relation of V4L2 to other Linux multimedia APIs</title>
diff --git a/Documentation/DocBook/media/v4l/pixfmt.xml b/Documentation/DocBook/media/v4l/pixfmt.xml
index ccf6053..d5eca4b 100644
--- a/Documentation/DocBook/media/v4l/pixfmt.xml
+++ b/Documentation/DocBook/media/v4l/pixfmt.xml
@@ -138,9 +138,25 @@
<row>
<entry>__u32</entry>
<entry><structfield>flags</structfield></entry>
- <entry>Flags set by the application or driver, see <xref
+ <entry>Flags set by the application or driver, see <xref
linkend="format-flags" />.</entry>
</row>
+ <row>
+ <entry>&v4l2-ycbcr-encoding;</entry>
+ <entry><structfield>ycbcr_enc</structfield></entry>
+ <entry>This information supplements the
+<structfield>colorspace</structfield> and must be set by the driver for
+capture streams and by the application for output streams,
+see <xref linkend="colorspaces" />.</entry>
+ </row>
+ <row>
+ <entry>&v4l2-quantization;</entry>
+ <entry><structfield>quantization</structfield></entry>
+ <entry>This information supplements the
+<structfield>colorspace</structfield> and must be set by the driver for
+capture streams and by the application for output streams,
+see <xref linkend="colorspaces" />.</entry>
+ </row>
</tbody>
</tgroup>
</table>
@@ -232,9 +248,25 @@
<entry>Flags set by the application or driver, see <xref
linkend="format-flags" />.</entry>
</row>
+ <row>
+ <entry>&v4l2-ycbcr-encoding;</entry>
+ <entry><structfield>ycbcr_enc</structfield></entry>
+ <entry>This information supplements the
+<structfield>colorspace</structfield> and must be set by the driver for
+capture streams and by the application for output streams,
+see <xref linkend="colorspaces" />.</entry>
+ </row>
+ <row>
+ <entry>&v4l2-quantization;</entry>
+ <entry><structfield>quantization</structfield></entry>
+ <entry>This information supplements the
+<structfield>colorspace</structfield> and must be set by the driver for
+capture streams and by the application for output streams,
+see <xref linkend="colorspaces" />.</entry>
+ </row>
<row>
<entry>__u8</entry>
- <entry><structfield>reserved[10]</structfield></entry>
+ <entry><structfield>reserved[8]</structfield></entry>
<entry>Reserved for future extensions. Should be zeroed by the
application.</entry>
</row>
diff --git a/Documentation/DocBook/media/v4l/subdev-formats.xml b/Documentation/DocBook/media/v4l/subdev-formats.xml
index 18730b9..c5ea868 100644
--- a/Documentation/DocBook/media/v4l/subdev-formats.xml
+++ b/Documentation/DocBook/media/v4l/subdev-formats.xml
@@ -34,8 +34,24 @@
<xref linkend="colorspaces" /> for details.</entry>
</row>
<row>
+ <entry>&v4l2-ycbcr-encoding;</entry>
+ <entry><structfield>ycbcr_enc</structfield></entry>
+ <entry>This information supplements the
+<structfield>colorspace</structfield> and must be set by the driver for
+capture streams and by the application for output streams,
+see <xref linkend="colorspaces" />.</entry>
+ </row>
+ <row>
+ <entry>&v4l2-quantization;</entry>
+ <entry><structfield>quantization</structfield></entry>
+ <entry>This information supplements the
+<structfield>colorspace</structfield> and must be set by the driver for
+capture streams and by the application for output streams,
+see <xref linkend="colorspaces" />.</entry>
+ </row>
+ <row>
<entry>__u32</entry>
- <entry><structfield>reserved</structfield>[7]</entry>
+ <entry><structfield>reserved</structfield>[6]</entry>
<entry>Reserved for future extensions. Applications and drivers must
set the array to zero.</entry>
</row>
diff --git a/Documentation/DocBook/media/v4l/v4l2.xml b/Documentation/DocBook/media/v4l/v4l2.xml
index 7cfe618..ac0f8d9 100644
--- a/Documentation/DocBook/media/v4l/v4l2.xml
+++ b/Documentation/DocBook/media/v4l/v4l2.xml
@@ -152,6 +152,15 @@
applications. -->
<revision>
+ <revnumber>3.19</revnumber>
+ <date>2014-12-05</date>
+ <authorinitials>hv</authorinitials>
+ <revremark>Rewrote Colorspace chapter, added new &v4l2-ycbcr-encoding; and &v4l2-quantization; fields
+to &v4l2-pix-format;, &v4l2-pix-format-mplane; and &v4l2-mbus-framefmt;.
+ </revremark>
+ </revision>
+
+ <revision>
<revnumber>3.17</revnumber>
<date>2014-08-04</date>
<authorinitials>lp, hv</authorinitials>
@@ -539,7 +548,7 @@
</partinfo>
<title>Video for Linux Two API Specification</title>
- <subtitle>Revision 3.17</subtitle>
+ <subtitle>Revision 3.19</subtitle>
<chapter id="common">
&sub-common;
diff --git a/Documentation/devicetree/bindings/media/rcar_vin.txt b/Documentation/devicetree/bindings/media/rcar_vin.txt
index ba61782..9dafe6b 100644
--- a/Documentation/devicetree/bindings/media/rcar_vin.txt
+++ b/Documentation/devicetree/bindings/media/rcar_vin.txt
@@ -6,6 +6,8 @@
channel which can be either RGB, YUYV or BT656.
- compatible: Must be one of the following
+ - "renesas,vin-r8a7794" for the R8A7794 device
+ - "renesas,vin-r8a7793" for the R8A7793 device
- "renesas,vin-r8a7791" for the R8A7791 device
- "renesas,vin-r8a7790" for the R8A7790 device
- "renesas,vin-r8a7779" for the R8A7779 device
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index bda85f1..4df73da 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1457,6 +1457,15 @@
disable
Do not enable intel_pstate as the default
scaling driver for the supported processors
+ force
+ Enable intel_pstate on systems that prohibit it by default
+ in favor of acpi-cpufreq. Forcing the intel_pstate driver
+ instead of acpi-cpufreq may disable platform features, such
+ as thermal controls and power capping, that rely on ACPI
+ P-States information being indicated to OSPM and therefore
+ should be used with caution. This option does not work with
+ processors that aren't supported by the intel_pstate driver
+ or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
no_hwp
Do not enable hardware P state control (HWP)
if available.
diff --git a/Documentation/video4linux/vivid.txt b/Documentation/video4linux/vivid.txt
index e5a940e..6cfc854 100644
--- a/Documentation/video4linux/vivid.txt
+++ b/Documentation/video4linux/vivid.txt
@@ -640,6 +640,21 @@
Changing the colorspace will result in the V4L2_EVENT_SOURCE_CHANGE
to be sent since it emulates a detected colorspace change.
+Y'CbCr Encoding: selects which Y'CbCr encoding should be used when generating
+ a Y'CbCr image. This only applies if the CSC Colorbar test pattern is
+ selected, and if the format is set to a Y'CbCr format as opposed to an
+ RGB format.
+
+ Changing the Y'CbCr encoding will result in the V4L2_EVENT_SOURCE_CHANGE
+ to be sent since it emulates a detected colorspace change.
+
+Quantization: selects which quantization should be used for the RGB or Y'CbCr
+ encoding when generating the test pattern. This only applies if the CSC
+ Colorbar test pattern is selected.
+
+ Changing the quantization will result in the V4L2_EVENT_SOURCE_CHANGE
+ to be sent since it emulates a detected colorspace change.
+
Limited RGB Range (16-235): selects if the RGB range of the HDMI source should
be limited or full range. This combines with the Digital Video 'Rx RGB
Quantization Range' control and can be used to test what happens if
diff --git a/MAINTAINERS b/MAINTAINERS
index bae54c7..08f671d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4963,6 +4963,12 @@
S: Supported
F: drivers/idle/intel_idle.c
+INTEL PSTATE DRIVER
+M: Kristen Carlson Accardi <kristen@linux.intel.com>
+L: linux-pm@vger.kernel.org
+S: Supported
+F: drivers/cpufreq/intel_pstate.c
+
INTEL FRAMEBUFFER DRIVER (excluding 810 and 815)
M: Maik Broemme <mbroemme@plusserver.de>
L: linux-fbdev@vger.kernel.org
@@ -10237,13 +10243,13 @@
S: Maintained
F: drivers/net/ethernet/via/via-velocity.*
-VIVI VIRTUAL VIDEO DRIVER
+VIVID VIRTUAL VIDEO DRIVER
M: Hans Verkuil <hverkuil@xs4all.nl>
L: linux-media@vger.kernel.org
T: git git://linuxtv.org/media_tree.git
W: http://linuxtv.org
S: Maintained
-F: drivers/media/platform/vivi*
+F: drivers/media/platform/vivid/*
VLAN (802.1Q)
M: Patrick McHardy <kaber@trash.net>
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index e34934f..f7c65ad 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -484,7 +484,7 @@
armpmu->stop(armpmu);
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int armpmu_runtime_resume(struct device *dev)
{
struct arm_pmu_platdata *plat = dev_get_platdata(dev);
diff --git a/arch/arm/mach-davinci/pm_domain.c b/arch/arm/mach-davinci/pm_domain.c
index 6b98413..641edc3 100644
--- a/arch/arm/mach-davinci/pm_domain.c
+++ b/arch/arm/mach-davinci/pm_domain.c
@@ -14,7 +14,7 @@
#include <linux/pm_clock.h>
#include <linux/platform_device.h>
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int davinci_pm_runtime_suspend(struct device *dev)
{
int ret;
diff --git a/arch/arm/mach-exynos/Kconfig b/arch/arm/mach-exynos/Kconfig
index e4a00ba..603820e 100644
--- a/arch/arm/mach-exynos/Kconfig
+++ b/arch/arm/mach-exynos/Kconfig
@@ -21,7 +21,7 @@
select HAVE_S3C_RTC if RTC_CLASS
select PINCTRL
select PINCTRL_EXYNOS
- select PM_GENERIC_DOMAINS if PM_RUNTIME
+ select PM_GENERIC_DOMAINS if PM
select S5P_DEV_MFC
select SRAM
select MFD_SYSCON
diff --git a/arch/arm/mach-keystone/pm_domain.c b/arch/arm/mach-keystone/pm_domain.c
index ca79dda..ef6041e 100644
--- a/arch/arm/mach-keystone/pm_domain.c
+++ b/arch/arm/mach-keystone/pm_domain.c
@@ -19,7 +19,7 @@
#include <linux/clk-provider.h>
#include <linux/of.h>
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int keystone_pm_runtime_suspend(struct device *dev)
{
int ret;
diff --git a/arch/arm/mach-omap1/pm_bus.c b/arch/arm/mach-omap1/pm_bus.c
index 3f2d396..c40e209 100644
--- a/arch/arm/mach-omap1/pm_bus.c
+++ b/arch/arm/mach-omap1/pm_bus.c
@@ -21,7 +21,7 @@
#include "soc.h"
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int omap1_pm_runtime_suspend(struct device *dev)
{
int ret;
@@ -59,7 +59,7 @@
#define OMAP1_PM_DOMAIN (&default_pm_domain)
#else
#define OMAP1_PM_DOMAIN NULL
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
static struct pm_clk_notifier_block platform_bus_notifier = {
.pm_domain = OMAP1_PM_DOMAIN,
diff --git a/arch/arm/mach-omap2/io.c b/arch/arm/mach-omap2/io.c
index 4fc8383..a1bd6af 100644
--- a/arch/arm/mach-omap2/io.c
+++ b/arch/arm/mach-omap2/io.c
@@ -361,7 +361,7 @@
u8 postsetup_state;
/* Set the default postsetup state for all hwmods */
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
postsetup_state = _HWMOD_STATE_IDLE;
#else
postsetup_state = _HWMOD_STATE_ENABLED;
diff --git a/arch/arm/mach-omap2/omap_device.c b/arch/arm/mach-omap2/omap_device.c
index 8c58b71..be9541e 100644
--- a/arch/arm/mach-omap2/omap_device.c
+++ b/arch/arm/mach-omap2/omap_device.c
@@ -588,7 +588,7 @@
return ERR_PTR(ret);
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int _od_runtime_suspend(struct device *dev)
{
struct platform_device *pdev = to_platform_device(dev);
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 371b55b..074e52b 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -231,7 +231,7 @@
config IA64_HP_SIM
bool "Ski-simulator"
select SWIOTLB
- depends on !PM_RUNTIME
+ depends on !PM
endchoice
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 7ffe0a2..a6745e7 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -3968,7 +3968,7 @@
{
if (dev->power.is_prepared)
return true;
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
if (dev->power.runtime_status == RPM_SUSPENDING)
return true;
#endif
diff --git a/drivers/acpi/blacklist.c b/drivers/acpi/blacklist.c
index 7556e7c..9b693d5 100644
--- a/drivers/acpi/blacklist.c
+++ b/drivers/acpi/blacklist.c
@@ -305,60 +305,6 @@
*/
/*
- * Lenovo has a mix of systems OSI(Linux) situations
- * and thus we can not wildcard the vendor.
- *
- * _OSI(Linux) helps sound
- * DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad R61"),
- * DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad T61"),
- * T400, T500
- * _OSI(Linux) has Linux specific hooks
- * DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad X61"),
- * _OSI(Linux) is a NOP:
- * DMI_MATCH(DMI_PRODUCT_VERSION, "3000 N100"),
- * DMI_MATCH(DMI_PRODUCT_VERSION, "LENOVO3000 V100"),
- */
- {
- .callback = dmi_enable_osi_linux,
- .ident = "Lenovo ThinkPad R61",
- .matches = {
- DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
- DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad R61"),
- },
- },
- {
- .callback = dmi_enable_osi_linux,
- .ident = "Lenovo ThinkPad T61",
- .matches = {
- DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
- DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad T61"),
- },
- },
- {
- .callback = dmi_enable_osi_linux,
- .ident = "Lenovo ThinkPad X61",
- .matches = {
- DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
- DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad X61"),
- },
- },
- {
- .callback = dmi_enable_osi_linux,
- .ident = "Lenovo ThinkPad T400",
- .matches = {
- DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
- DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad T400"),
- },
- },
- {
- .callback = dmi_enable_osi_linux,
- .ident = "Lenovo ThinkPad T500",
- .matches = {
- DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
- DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad T500"),
- },
- },
- /*
* Without this this EEEpc exports a non working WMI interface, with
* this it exports a working "good old" eeepc_laptop interface, fixing
* both brightness control, and rfkill not working.
diff --git a/drivers/acpi/device_pm.c b/drivers/acpi/device_pm.c
index 8976401..c2daa85 100644
--- a/drivers/acpi/device_pm.c
+++ b/drivers/acpi/device_pm.c
@@ -680,13 +680,21 @@
if (error)
return error;
+ if (adev->wakeup.flags.enabled)
+ return 0;
+
res = acpi_enable_gpe(wakeup->gpe_device, wakeup->gpe_number);
- if (ACPI_FAILURE(res)) {
+ if (ACPI_SUCCESS(res)) {
+ adev->wakeup.flags.enabled = 1;
+ } else {
acpi_disable_wakeup_device_power(adev);
return -EIO;
}
} else {
- acpi_disable_gpe(wakeup->gpe_device, wakeup->gpe_number);
+ if (adev->wakeup.flags.enabled) {
+ acpi_disable_gpe(wakeup->gpe_device, wakeup->gpe_number);
+ adev->wakeup.flags.enabled = 0;
+ }
acpi_disable_wakeup_device_power(adev);
}
return 0;
diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c
index 5f9b74b..1b5853f 100644
--- a/drivers/acpi/ec.c
+++ b/drivers/acpi/ec.c
@@ -844,6 +844,8 @@
static void ec_remove_handlers(struct acpi_ec *ec)
{
+ if (!test_bit(EC_FLAGS_HANDLERS_INSTALLED, &ec->flags))
+ return;
acpi_disable_gpe(NULL, ec->gpe);
if (ACPI_FAILURE(acpi_remove_address_space_handler(ec->handle,
ACPI_ADR_SPACE_EC, &acpi_ec_space_handler)))
diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
index caf9b76..7a36f02 100644
--- a/drivers/acpi/fan.c
+++ b/drivers/acpi/fan.c
@@ -325,6 +325,7 @@
struct thermal_cooling_device *cdev;
struct acpi_fan *fan;
struct acpi_device *device = ACPI_COMPANION(&pdev->dev);
+ char *name;
fan = devm_kzalloc(&pdev->dev, sizeof(*fan), GFP_KERNEL);
if (!fan) {
@@ -346,7 +347,12 @@
}
}
- cdev = thermal_cooling_device_register("Fan", device,
+ if (!strncmp(pdev->name, "PNP0C0B", strlen("PNP0C0B")))
+ name = "Fan";
+ else
+ name = acpi_device_bid(device);
+
+ cdev = thermal_cooling_device_register(name, device,
&fan_cooling_ops);
if (IS_ERR(cdev)) {
result = PTR_ERR(cdev);
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 1b1cf55..16914cc 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -2214,7 +2214,7 @@
status = acpi_evaluate_reference(adev->handle, "_DEP", NULL,
&dep_devices);
if (ACPI_FAILURE(status)) {
- dev_err(&adev->dev, "Failed to evaluate _DEP.\n");
+ dev_dbg(&adev->dev, "Failed to evaluate _DEP.\n");
return;
}
@@ -2224,7 +2224,7 @@
status = acpi_get_object_info(dep_devices.handles[i], &info);
if (ACPI_FAILURE(status)) {
- dev_err(&adev->dev, "Error reading device info\n");
+ dev_dbg(&adev->dev, "Error reading _DEP device info\n");
continue;
}
diff --git a/drivers/acpi/utils.c b/drivers/acpi/utils.c
index dd8ff63..cd49a39 100644
--- a/drivers/acpi/utils.c
+++ b/drivers/acpi/utils.c
@@ -346,22 +346,16 @@
package = buffer.pointer;
if ((buffer.length == 0) || !package) {
- printk(KERN_ERR PREFIX "No return object (len %X ptr %p)\n",
- (unsigned)buffer.length, package);
status = AE_BAD_DATA;
acpi_util_eval_error(handle, pathname, status);
goto end;
}
if (package->type != ACPI_TYPE_PACKAGE) {
- printk(KERN_ERR PREFIX "Expecting a [Package], found type %X\n",
- package->type);
status = AE_BAD_DATA;
acpi_util_eval_error(handle, pathname, status);
goto end;
}
if (!package->package.count) {
- printk(KERN_ERR PREFIX "[Package] has zero elements (%p)\n",
- package);
status = AE_BAD_DATA;
acpi_util_eval_error(handle, pathname, status);
goto end;
@@ -380,17 +374,13 @@
if (element->type != ACPI_TYPE_LOCAL_REFERENCE) {
status = AE_BAD_DATA;
- printk(KERN_ERR PREFIX
- "Expecting a [Reference] package element, found type %X\n",
- element->type);
acpi_util_eval_error(handle, pathname, status);
break;
}
if (!element->reference.handle) {
- printk(KERN_WARNING PREFIX "Invalid reference in"
- " package %s\n", pathname);
status = AE_NULL_ENTRY;
+ acpi_util_eval_error(handle, pathname, status);
break;
}
/* Get the acpi_handle. */
diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c
index 185a57d..1eaadff 100644
--- a/drivers/acpi/video.c
+++ b/drivers/acpi/video.c
@@ -155,6 +155,7 @@
u8 dos_setting;
struct acpi_video_enumerated_device *attached_array;
u8 attached_count;
+ u8 child_count;
struct acpi_video_bus_cap cap;
struct acpi_video_bus_flags flags;
struct list_head video_device_list;
@@ -1159,8 +1160,12 @@
struct acpi_video_bus *video = device->video;
int i;
- /* If we have a broken _DOD, no need to test */
- if (!video->attached_count)
+ /*
+ * If we have a broken _DOD or we have more than 8 output devices
+ * under the graphics controller node that we can't proper deal with
+ * in the operation region code currently, no need to test.
+ */
+ if (!video->attached_count || video->child_count > 8)
return true;
for (i = 0; i < video->attached_count; i++) {
@@ -1413,6 +1418,7 @@
dev_err(&dev->dev, "Can't attach device\n");
break;
}
+ video->child_count++;
}
return status;
}
diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index cd4cccb..a3a1360 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -61,7 +61,7 @@
config SATA_ZPODD
bool "SATA Zero Power Optical Disc Drive (ZPODD) support"
- depends on ATA_ACPI && PM_RUNTIME
+ depends on ATA_ACPI && PM
default n
help
This option adds support for SATA Zero Power Optical Disc
diff --git a/drivers/base/power/opp.c b/drivers/base/power/opp.c
index 2d195f3..d24dd614a 100644
--- a/drivers/base/power/opp.c
+++ b/drivers/base/power/opp.c
@@ -84,7 +84,11 @@
*
* This is an internal data structure maintaining the link to opps attached to
* a device. This structure is not meant to be shared to users as it is
- * meant for book keeping and private to OPP library
+ * meant for book keeping and private to OPP library.
+ *
+ * Because the opp structures can be used from both rcu and srcu readers, we
+ * need to wait for the grace period of both of them before freeing any
+ * resources. And so we have used kfree_rcu() from within call_srcu() handlers.
*/
struct device_opp {
struct list_head node;
@@ -382,12 +386,34 @@
}
EXPORT_SYMBOL_GPL(dev_pm_opp_find_freq_floor);
+static struct device_opp *add_device_opp(struct device *dev)
+{
+ struct device_opp *dev_opp;
+
+ /*
+ * Allocate a new device OPP table. In the infrequent case where a new
+ * device is needed to be added, we pay this penalty.
+ */
+ dev_opp = kzalloc(sizeof(*dev_opp), GFP_KERNEL);
+ if (!dev_opp)
+ return NULL;
+
+ dev_opp->dev = dev;
+ srcu_init_notifier_head(&dev_opp->srcu_head);
+ INIT_LIST_HEAD(&dev_opp->opp_list);
+
+ /* Secure the device list modification */
+ list_add_rcu(&dev_opp->node, &dev_opp_list);
+ return dev_opp;
+}
+
static int dev_pm_opp_add_dynamic(struct device *dev, unsigned long freq,
unsigned long u_volt, bool dynamic)
{
struct device_opp *dev_opp = NULL;
struct dev_pm_opp *opp, *new_opp;
struct list_head *head;
+ int ret;
/* allocate new OPP node */
new_opp = kzalloc(sizeof(*new_opp), GFP_KERNEL);
@@ -400,7 +426,6 @@
mutex_lock(&dev_opp_list_lock);
/* populate the opp table */
- new_opp->dev_opp = dev_opp;
new_opp->rate = freq;
new_opp->u_volt = u_volt;
new_opp->available = true;
@@ -409,27 +434,12 @@
/* Check for existing list for 'dev' */
dev_opp = find_device_opp(dev);
if (IS_ERR(dev_opp)) {
- /*
- * Allocate a new device OPP table. In the infrequent case
- * where a new device is needed to be added, we pay this
- * penalty.
- */
- dev_opp = kzalloc(sizeof(struct device_opp), GFP_KERNEL);
+ dev_opp = add_device_opp(dev);
if (!dev_opp) {
- mutex_unlock(&dev_opp_list_lock);
- kfree(new_opp);
- dev_warn(dev,
- "%s: Unable to create device OPP structure\n",
- __func__);
- return -ENOMEM;
+ ret = -ENOMEM;
+ goto free_opp;
}
- dev_opp->dev = dev;
- srcu_init_notifier_head(&dev_opp->srcu_head);
- INIT_LIST_HEAD(&dev_opp->opp_list);
-
- /* Secure the device list modification */
- list_add_rcu(&dev_opp->node, &dev_opp_list);
head = &dev_opp->opp_list;
goto list_add;
}
@@ -448,18 +458,17 @@
/* Duplicate OPPs ? */
if (new_opp->rate == opp->rate) {
- int ret = opp->available && new_opp->u_volt == opp->u_volt ?
+ ret = opp->available && new_opp->u_volt == opp->u_volt ?
0 : -EEXIST;
dev_warn(dev, "%s: duplicate OPPs detected. Existing: freq: %lu, volt: %lu, enabled: %d. New: freq: %lu, volt: %lu, enabled: %d\n",
__func__, opp->rate, opp->u_volt, opp->available,
new_opp->rate, new_opp->u_volt, new_opp->available);
- mutex_unlock(&dev_opp_list_lock);
- kfree(new_opp);
- return ret;
+ goto free_opp;
}
list_add:
+ new_opp->dev_opp = dev_opp;
list_add_rcu(&new_opp->node, head);
mutex_unlock(&dev_opp_list_lock);
@@ -469,6 +478,11 @@
*/
srcu_notifier_call_chain(&dev_opp->srcu_head, OPP_EVENT_ADD, new_opp);
return 0;
+
+free_opp:
+ mutex_unlock(&dev_opp_list_lock);
+ kfree(new_opp);
+ return ret;
}
/**
@@ -511,10 +525,11 @@
{
struct device_opp *device_opp = container_of(head, struct device_opp, rcu_head);
- kfree(device_opp);
+ kfree_rcu(device_opp, rcu_head);
}
-void __dev_pm_opp_remove(struct device_opp *dev_opp, struct dev_pm_opp *opp)
+static void __dev_pm_opp_remove(struct device_opp *dev_opp,
+ struct dev_pm_opp *opp)
{
/*
* Notify the changes in the availability of the operable
@@ -592,7 +607,7 @@
static int opp_set_availability(struct device *dev, unsigned long freq,
bool availability_req)
{
- struct device_opp *tmp_dev_opp, *dev_opp = ERR_PTR(-ENODEV);
+ struct device_opp *dev_opp;
struct dev_pm_opp *new_opp, *tmp_opp, *opp = ERR_PTR(-ENODEV);
int r = 0;
@@ -606,12 +621,7 @@
mutex_lock(&dev_opp_list_lock);
/* Find the device_opp */
- list_for_each_entry(tmp_dev_opp, &dev_opp_list, node) {
- if (dev == tmp_dev_opp->dev) {
- dev_opp = tmp_dev_opp;
- break;
- }
- }
+ dev_opp = find_device_opp(dev);
if (IS_ERR(dev_opp)) {
r = PTR_ERR(dev_opp);
dev_warn(dev, "%s: Device OPP not found (%d)\n", __func__, r);
@@ -768,7 +778,7 @@
*/
void of_free_opp_table(struct device *dev)
{
- struct device_opp *dev_opp = find_device_opp(dev);
+ struct device_opp *dev_opp;
struct dev_pm_opp *opp, *tmp;
/* Check for existing list for 'dev' */
diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
index 1405b39..742eefb 100644
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -199,7 +199,14 @@
pid->integral += fp_error;
- /* limit the integral term */
+ /*
+ * We limit the integral here so that it will never
+ * get higher than 30. This prevents it from becoming
+ * too large an input over long periods of time and allows
+ * it to get factored out sooner.
+ *
+ * The value of 30 was chosen through experimentation.
+ */
integral_limit = int_tofp(30);
if (pid->integral > integral_limit)
pid->integral = integral_limit;
@@ -616,6 +623,11 @@
if (limits.no_turbo || limits.turbo_disabled)
max_perf = cpu->pstate.max_pstate;
+ /*
+ * performance can be limited by user through sysfs, by cpufreq
+ * policy, or by cpu specific default values determined through
+ * experimentation.
+ */
max_perf_adj = fp_toint(mul_fp(int_tofp(max_perf), limits.max_perf));
*max = clamp_t(int, max_perf_adj,
cpu->pstate.min_pstate, cpu->pstate.turbo_pstate);
@@ -717,11 +729,29 @@
u32 duration_us;
u32 sample_time;
+ /*
+ * core_busy is the ratio of actual performance to max
+ * max_pstate is the max non turbo pstate available
+ * current_pstate was the pstate that was requested during
+ * the last sample period.
+ *
+ * We normalize core_busy, which was our actual percent
+ * performance to what we requested during the last sample
+ * period. The result will be a percentage of busy at a
+ * specified pstate.
+ */
core_busy = cpu->sample.core_pct_busy;
max_pstate = int_tofp(cpu->pstate.max_pstate);
current_pstate = int_tofp(cpu->pstate.current_pstate);
core_busy = mul_fp(core_busy, div_fp(max_pstate, current_pstate));
+ /*
+ * Since we have a deferred timer, it will not fire unless
+ * we are in C0. So, determine if the actual elapsed time
+ * is significantly greater (3x) than our sample interval. If it
+ * is, then we were idle for a long enough period of time
+ * to adjust our busyness.
+ */
sample_time = pid_params.sample_rate_ms * USEC_PER_MSEC;
duration_us = (u32) ktime_us_delta(cpu->sample.time,
cpu->last_sample_time);
@@ -948,6 +978,7 @@
static int __initdata no_load;
static int __initdata no_hwp;
+static unsigned int force_load;
static int intel_pstate_msrs_not_valid(void)
{
@@ -1094,7 +1125,8 @@
case PSS:
return intel_pstate_no_acpi_pss();
case PPC:
- return intel_pstate_has_acpi_ppc();
+ return intel_pstate_has_acpi_ppc() &&
+ (!force_load);
}
}
@@ -1175,6 +1207,8 @@
no_load = 1;
if (!strcmp(str, "no_hwp"))
no_hwp = 1;
+ if (!strcmp(str, "force"))
+ force_load = 1;
return 0;
}
early_param("intel_pstate", intel_pstate_setup);
diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig
index 7708939..b899531 100644
--- a/drivers/infiniband/Kconfig
+++ b/drivers/infiniband/Kconfig
@@ -38,6 +38,17 @@
depends on INFINIBAND_USER_ACCESS != n
default y
+config INFINIBAND_ON_DEMAND_PAGING
+ bool "InfiniBand on-demand paging support"
+ depends on INFINIBAND_USER_MEM
+ select MMU_NOTIFIER
+ default y
+ ---help---
+ On demand paging support for the InfiniBand subsystem.
+ Together with driver support this allows registration of
+ memory regions without pinning their pages, fetching the
+ pages on demand instead.
+
config INFINIBAND_ADDR_TRANS
bool
depends on INFINIBAND
diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile
index ffd0af6..acf7367 100644
--- a/drivers/infiniband/core/Makefile
+++ b/drivers/infiniband/core/Makefile
@@ -11,6 +11,7 @@
ib_core-y := packer.o ud_header.o verbs.o sysfs.o \
device.o fmr_pool.o cache.o netlink.o
ib_core-$(CONFIG_INFINIBAND_USER_MEM) += umem.o
+ib_core-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += umem_odp.o umem_rbtree.o
ib_mad-y := mad.o smi.o agent.o mad_rmpp.o
diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index 8172d37..f80da50 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -176,8 +176,8 @@
unsigned long delay;
delay = time - jiffies;
- if ((long)delay <= 0)
- delay = 1;
+ if ((long)delay < 0)
+ delay = 0;
mod_delayed_work(addr_wq, &work, delay);
}
diff --git a/drivers/infiniband/core/multicast.c b/drivers/infiniband/core/multicast.c
index d2360a8..fa17b55 100644
--- a/drivers/infiniband/core/multicast.c
+++ b/drivers/infiniband/core/multicast.c
@@ -525,17 +525,22 @@
if (status)
process_join_error(group, status);
else {
+ int mgids_changed, is_mgid0;
ib_find_pkey(group->port->dev->device, group->port->port_num,
be16_to_cpu(rec->pkey), &pkey_index);
spin_lock_irq(&group->port->lock);
- group->rec = *rec;
if (group->state == MCAST_BUSY &&
group->pkey_index == MCAST_INVALID_PKEY_INDEX)
group->pkey_index = pkey_index;
- if (!memcmp(&mgid0, &group->rec.mgid, sizeof mgid0)) {
+ mgids_changed = memcmp(&rec->mgid, &group->rec.mgid,
+ sizeof(group->rec.mgid));
+ group->rec = *rec;
+ if (mgids_changed) {
rb_erase(&group->node, &group->port->table);
- mcast_insert(group->port, group, 1);
+ is_mgid0 = !memcmp(&mgid0, &group->rec.mgid,
+ sizeof(mgid0));
+ mcast_insert(group->port, group, is_mgid0);
}
spin_unlock_irq(&group->port->lock);
}
diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c
index df0c4f6..aec7a6a 100644
--- a/drivers/infiniband/core/umem.c
+++ b/drivers/infiniband/core/umem.c
@@ -39,6 +39,7 @@
#include <linux/hugetlb.h>
#include <linux/dma-attrs.h>
#include <linux/slab.h>
+#include <rdma/ib_umem_odp.h>
#include "uverbs.h"
@@ -69,6 +70,10 @@
/**
* ib_umem_get - Pin and DMA map userspace memory.
+ *
+ * If access flags indicate ODP memory, avoid pinning. Instead, stores
+ * the mm for future page fault handling in conjunction with MMU notifiers.
+ *
* @context: userspace context to pin memory for
* @addr: userspace virtual address to start at
* @size: length of region to pin
@@ -103,17 +108,30 @@
umem->context = context;
umem->length = size;
- umem->offset = addr & ~PAGE_MASK;
+ umem->address = addr;
umem->page_size = PAGE_SIZE;
umem->pid = get_task_pid(current, PIDTYPE_PID);
/*
- * We ask for writable memory if any access flags other than
- * "remote read" are set. "Local write" and "remote write"
+ * We ask for writable memory if any of the following
+ * access flags are set. "Local write" and "remote write"
* obviously require write access. "Remote atomic" can do
* things like fetch and add, which will modify memory, and
* "MW bind" can change permissions by binding a window.
*/
- umem->writable = !!(access & ~IB_ACCESS_REMOTE_READ);
+ umem->writable = !!(access &
+ (IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_WRITE |
+ IB_ACCESS_REMOTE_ATOMIC | IB_ACCESS_MW_BIND));
+
+ if (access & IB_ACCESS_ON_DEMAND) {
+ ret = ib_umem_odp_get(context, umem);
+ if (ret) {
+ kfree(umem);
+ return ERR_PTR(ret);
+ }
+ return umem;
+ }
+
+ umem->odp_data = NULL;
/* We assume the memory is from hugetlb until proved otherwise */
umem->hugetlb = 1;
@@ -132,7 +150,7 @@
if (!vma_list)
umem->hugetlb = 0;
- npages = PAGE_ALIGN(size + umem->offset) >> PAGE_SHIFT;
+ npages = ib_umem_num_pages(umem);
down_write(¤t->mm->mmap_sem);
@@ -235,6 +253,11 @@
struct task_struct *task;
unsigned long diff;
+ if (umem->odp_data) {
+ ib_umem_odp_release(umem);
+ return;
+ }
+
__ib_umem_release(umem->context->device, umem, 1);
task = get_pid_task(umem->pid, PIDTYPE_PID);
@@ -246,7 +269,7 @@
if (!mm)
goto out;
- diff = PAGE_ALIGN(umem->length + umem->offset) >> PAGE_SHIFT;
+ diff = ib_umem_num_pages(umem);
/*
* We may be called with the mm's mmap_sem already held. This
@@ -283,6 +306,9 @@
int n;
struct scatterlist *sg;
+ if (umem->odp_data)
+ return ib_umem_num_pages(umem);
+
shift = ilog2(umem->page_size);
n = 0;
@@ -292,3 +318,37 @@
return n;
}
EXPORT_SYMBOL(ib_umem_page_count);
+
+/*
+ * Copy from the given ib_umem's pages to the given buffer.
+ *
+ * umem - the umem to copy from
+ * offset - offset to start copying from
+ * dst - destination buffer
+ * length - buffer length
+ *
+ * Returns 0 on success, or an error code.
+ */
+int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset,
+ size_t length)
+{
+ size_t end = offset + length;
+ int ret;
+
+ if (offset > umem->length || length > umem->length - offset) {
+ pr_err("ib_umem_copy_from not in range. offset: %zd umem length: %zd end: %zd\n",
+ offset, umem->length, end);
+ return -EINVAL;
+ }
+
+ ret = sg_pcopy_to_buffer(umem->sg_head.sgl, umem->nmap, dst, length,
+ offset + ib_umem_offset(umem));
+
+ if (ret < 0)
+ return ret;
+ else if (ret != length)
+ return -EINVAL;
+ else
+ return 0;
+}
+EXPORT_SYMBOL(ib_umem_copy_from);
diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
new file mode 100644
index 0000000..6095872
--- /dev/null
+++ b/drivers/infiniband/core/umem_odp.c
@@ -0,0 +1,668 @@
+/*
+ * Copyright (c) 2014 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * 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.
+ */
+
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/pid.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+#include <linux/vmalloc.h>
+
+#include <rdma/ib_verbs.h>
+#include <rdma/ib_umem.h>
+#include <rdma/ib_umem_odp.h>
+
+static void ib_umem_notifier_start_account(struct ib_umem *item)
+{
+ mutex_lock(&item->odp_data->umem_mutex);
+
+ /* Only update private counters for this umem if it has them.
+ * Otherwise skip it. All page faults will be delayed for this umem. */
+ if (item->odp_data->mn_counters_active) {
+ int notifiers_count = item->odp_data->notifiers_count++;
+
+ if (notifiers_count == 0)
+ /* Initialize the completion object for waiting on
+ * notifiers. Since notifier_count is zero, no one
+ * should be waiting right now. */
+ reinit_completion(&item->odp_data->notifier_completion);
+ }
+ mutex_unlock(&item->odp_data->umem_mutex);
+}
+
+static void ib_umem_notifier_end_account(struct ib_umem *item)
+{
+ mutex_lock(&item->odp_data->umem_mutex);
+
+ /* Only update private counters for this umem if it has them.
+ * Otherwise skip it. All page faults will be delayed for this umem. */
+ if (item->odp_data->mn_counters_active) {
+ /*
+ * This sequence increase will notify the QP page fault that
+ * the page that is going to be mapped in the spte could have
+ * been freed.
+ */
+ ++item->odp_data->notifiers_seq;
+ if (--item->odp_data->notifiers_count == 0)
+ complete_all(&item->odp_data->notifier_completion);
+ }
+ mutex_unlock(&item->odp_data->umem_mutex);
+}
+
+/* Account for a new mmu notifier in an ib_ucontext. */
+static void ib_ucontext_notifier_start_account(struct ib_ucontext *context)
+{
+ atomic_inc(&context->notifier_count);
+}
+
+/* Account for a terminating mmu notifier in an ib_ucontext.
+ *
+ * Must be called with the ib_ucontext->umem_rwsem semaphore unlocked, since
+ * the function takes the semaphore itself. */
+static void ib_ucontext_notifier_end_account(struct ib_ucontext *context)
+{
+ int zero_notifiers = atomic_dec_and_test(&context->notifier_count);
+
+ if (zero_notifiers &&
+ !list_empty(&context->no_private_counters)) {
+ /* No currently running mmu notifiers. Now is the chance to
+ * add private accounting to all previously added umems. */
+ struct ib_umem_odp *odp_data, *next;
+
+ /* Prevent concurrent mmu notifiers from working on the
+ * no_private_counters list. */
+ down_write(&context->umem_rwsem);
+
+ /* Read the notifier_count again, with the umem_rwsem
+ * semaphore taken for write. */
+ if (!atomic_read(&context->notifier_count)) {
+ list_for_each_entry_safe(odp_data, next,
+ &context->no_private_counters,
+ no_private_counters) {
+ mutex_lock(&odp_data->umem_mutex);
+ odp_data->mn_counters_active = true;
+ list_del(&odp_data->no_private_counters);
+ complete_all(&odp_data->notifier_completion);
+ mutex_unlock(&odp_data->umem_mutex);
+ }
+ }
+
+ up_write(&context->umem_rwsem);
+ }
+}
+
+static int ib_umem_notifier_release_trampoline(struct ib_umem *item, u64 start,
+ u64 end, void *cookie) {
+ /*
+ * Increase the number of notifiers running, to
+ * prevent any further fault handling on this MR.
+ */
+ ib_umem_notifier_start_account(item);
+ item->odp_data->dying = 1;
+ /* Make sure that the fact the umem is dying is out before we release
+ * all pending page faults. */
+ smp_wmb();
+ complete_all(&item->odp_data->notifier_completion);
+ item->context->invalidate_range(item, ib_umem_start(item),
+ ib_umem_end(item));
+ return 0;
+}
+
+static void ib_umem_notifier_release(struct mmu_notifier *mn,
+ struct mm_struct *mm)
+{
+ struct ib_ucontext *context = container_of(mn, struct ib_ucontext, mn);
+
+ if (!context->invalidate_range)
+ return;
+
+ ib_ucontext_notifier_start_account(context);
+ down_read(&context->umem_rwsem);
+ rbt_ib_umem_for_each_in_range(&context->umem_tree, 0,
+ ULLONG_MAX,
+ ib_umem_notifier_release_trampoline,
+ NULL);
+ up_read(&context->umem_rwsem);
+}
+
+static int invalidate_page_trampoline(struct ib_umem *item, u64 start,
+ u64 end, void *cookie)
+{
+ ib_umem_notifier_start_account(item);
+ item->context->invalidate_range(item, start, start + PAGE_SIZE);
+ ib_umem_notifier_end_account(item);
+ return 0;
+}
+
+static void ib_umem_notifier_invalidate_page(struct mmu_notifier *mn,
+ struct mm_struct *mm,
+ unsigned long address)
+{
+ struct ib_ucontext *context = container_of(mn, struct ib_ucontext, mn);
+
+ if (!context->invalidate_range)
+ return;
+
+ ib_ucontext_notifier_start_account(context);
+ down_read(&context->umem_rwsem);
+ rbt_ib_umem_for_each_in_range(&context->umem_tree, address,
+ address + PAGE_SIZE,
+ invalidate_page_trampoline, NULL);
+ up_read(&context->umem_rwsem);
+ ib_ucontext_notifier_end_account(context);
+}
+
+static int invalidate_range_start_trampoline(struct ib_umem *item, u64 start,
+ u64 end, void *cookie)
+{
+ ib_umem_notifier_start_account(item);
+ item->context->invalidate_range(item, start, end);
+ return 0;
+}
+
+static void ib_umem_notifier_invalidate_range_start(struct mmu_notifier *mn,
+ struct mm_struct *mm,
+ unsigned long start,
+ unsigned long end)
+{
+ struct ib_ucontext *context = container_of(mn, struct ib_ucontext, mn);
+
+ if (!context->invalidate_range)
+ return;
+
+ ib_ucontext_notifier_start_account(context);
+ down_read(&context->umem_rwsem);
+ rbt_ib_umem_for_each_in_range(&context->umem_tree, start,
+ end,
+ invalidate_range_start_trampoline, NULL);
+ up_read(&context->umem_rwsem);
+}
+
+static int invalidate_range_end_trampoline(struct ib_umem *item, u64 start,
+ u64 end, void *cookie)
+{
+ ib_umem_notifier_end_account(item);
+ return 0;
+}
+
+static void ib_umem_notifier_invalidate_range_end(struct mmu_notifier *mn,
+ struct mm_struct *mm,
+ unsigned long start,
+ unsigned long end)
+{
+ struct ib_ucontext *context = container_of(mn, struct ib_ucontext, mn);
+
+ if (!context->invalidate_range)
+ return;
+
+ down_read(&context->umem_rwsem);
+ rbt_ib_umem_for_each_in_range(&context->umem_tree, start,
+ end,
+ invalidate_range_end_trampoline, NULL);
+ up_read(&context->umem_rwsem);
+ ib_ucontext_notifier_end_account(context);
+}
+
+static struct mmu_notifier_ops ib_umem_notifiers = {
+ .release = ib_umem_notifier_release,
+ .invalidate_page = ib_umem_notifier_invalidate_page,
+ .invalidate_range_start = ib_umem_notifier_invalidate_range_start,
+ .invalidate_range_end = ib_umem_notifier_invalidate_range_end,
+};
+
+int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem)
+{
+ int ret_val;
+ struct pid *our_pid;
+ struct mm_struct *mm = get_task_mm(current);
+
+ if (!mm)
+ return -EINVAL;
+
+ /* Prevent creating ODP MRs in child processes */
+ rcu_read_lock();
+ our_pid = get_task_pid(current->group_leader, PIDTYPE_PID);
+ rcu_read_unlock();
+ put_pid(our_pid);
+ if (context->tgid != our_pid) {
+ ret_val = -EINVAL;
+ goto out_mm;
+ }
+
+ umem->hugetlb = 0;
+ umem->odp_data = kzalloc(sizeof(*umem->odp_data), GFP_KERNEL);
+ if (!umem->odp_data) {
+ ret_val = -ENOMEM;
+ goto out_mm;
+ }
+ umem->odp_data->umem = umem;
+
+ mutex_init(&umem->odp_data->umem_mutex);
+
+ init_completion(&umem->odp_data->notifier_completion);
+
+ umem->odp_data->page_list = vzalloc(ib_umem_num_pages(umem) *
+ sizeof(*umem->odp_data->page_list));
+ if (!umem->odp_data->page_list) {
+ ret_val = -ENOMEM;
+ goto out_odp_data;
+ }
+
+ umem->odp_data->dma_list = vzalloc(ib_umem_num_pages(umem) *
+ sizeof(*umem->odp_data->dma_list));
+ if (!umem->odp_data->dma_list) {
+ ret_val = -ENOMEM;
+ goto out_page_list;
+ }
+
+ /*
+ * When using MMU notifiers, we will get a
+ * notification before the "current" task (and MM) is
+ * destroyed. We use the umem_rwsem semaphore to synchronize.
+ */
+ down_write(&context->umem_rwsem);
+ context->odp_mrs_count++;
+ if (likely(ib_umem_start(umem) != ib_umem_end(umem)))
+ rbt_ib_umem_insert(&umem->odp_data->interval_tree,
+ &context->umem_tree);
+ if (likely(!atomic_read(&context->notifier_count)))
+ umem->odp_data->mn_counters_active = true;
+ else
+ list_add(&umem->odp_data->no_private_counters,
+ &context->no_private_counters);
+ downgrade_write(&context->umem_rwsem);
+
+ if (context->odp_mrs_count == 1) {
+ /*
+ * Note that at this point, no MMU notifier is running
+ * for this context!
+ */
+ atomic_set(&context->notifier_count, 0);
+ INIT_HLIST_NODE(&context->mn.hlist);
+ context->mn.ops = &ib_umem_notifiers;
+ /*
+ * Lock-dep detects a false positive for mmap_sem vs.
+ * umem_rwsem, due to not grasping downgrade_write correctly.
+ */
+ lockdep_off();
+ ret_val = mmu_notifier_register(&context->mn, mm);
+ lockdep_on();
+ if (ret_val) {
+ pr_err("Failed to register mmu_notifier %d\n", ret_val);
+ ret_val = -EBUSY;
+ goto out_mutex;
+ }
+ }
+
+ up_read(&context->umem_rwsem);
+
+ /*
+ * Note that doing an mmput can cause a notifier for the relevant mm.
+ * If the notifier is called while we hold the umem_rwsem, this will
+ * cause a deadlock. Therefore, we release the reference only after we
+ * released the semaphore.
+ */
+ mmput(mm);
+ return 0;
+
+out_mutex:
+ up_read(&context->umem_rwsem);
+ vfree(umem->odp_data->dma_list);
+out_page_list:
+ vfree(umem->odp_data->page_list);
+out_odp_data:
+ kfree(umem->odp_data);
+out_mm:
+ mmput(mm);
+ return ret_val;
+}
+
+void ib_umem_odp_release(struct ib_umem *umem)
+{
+ struct ib_ucontext *context = umem->context;
+
+ /*
+ * Ensure that no more pages are mapped in the umem.
+ *
+ * It is the driver's responsibility to ensure, before calling us,
+ * that the hardware will not attempt to access the MR any more.
+ */
+ ib_umem_odp_unmap_dma_pages(umem, ib_umem_start(umem),
+ ib_umem_end(umem));
+
+ down_write(&context->umem_rwsem);
+ if (likely(ib_umem_start(umem) != ib_umem_end(umem)))
+ rbt_ib_umem_remove(&umem->odp_data->interval_tree,
+ &context->umem_tree);
+ context->odp_mrs_count--;
+ if (!umem->odp_data->mn_counters_active) {
+ list_del(&umem->odp_data->no_private_counters);
+ complete_all(&umem->odp_data->notifier_completion);
+ }
+
+ /*
+ * Downgrade the lock to a read lock. This ensures that the notifiers
+ * (who lock the mutex for reading) will be able to finish, and we
+ * will be able to enventually obtain the mmu notifiers SRCU. Note
+ * that since we are doing it atomically, no other user could register
+ * and unregister while we do the check.
+ */
+ downgrade_write(&context->umem_rwsem);
+ if (!context->odp_mrs_count) {
+ struct task_struct *owning_process = NULL;
+ struct mm_struct *owning_mm = NULL;
+
+ owning_process = get_pid_task(context->tgid,
+ PIDTYPE_PID);
+ if (owning_process == NULL)
+ /*
+ * The process is already dead, notifier were removed
+ * already.
+ */
+ goto out;
+
+ owning_mm = get_task_mm(owning_process);
+ if (owning_mm == NULL)
+ /*
+ * The process' mm is already dead, notifier were
+ * removed already.
+ */
+ goto out_put_task;
+ mmu_notifier_unregister(&context->mn, owning_mm);
+
+ mmput(owning_mm);
+
+out_put_task:
+ put_task_struct(owning_process);
+ }
+out:
+ up_read(&context->umem_rwsem);
+
+ vfree(umem->odp_data->dma_list);
+ vfree(umem->odp_data->page_list);
+ kfree(umem->odp_data);
+ kfree(umem);
+}
+
+/*
+ * Map for DMA and insert a single page into the on-demand paging page tables.
+ *
+ * @umem: the umem to insert the page to.
+ * @page_index: index in the umem to add the page to.
+ * @page: the page struct to map and add.
+ * @access_mask: access permissions needed for this page.
+ * @current_seq: sequence number for synchronization with invalidations.
+ * the sequence number is taken from
+ * umem->odp_data->notifiers_seq.
+ *
+ * The function returns -EFAULT if the DMA mapping operation fails. It returns
+ * -EAGAIN if a concurrent invalidation prevents us from updating the page.
+ *
+ * The page is released via put_page even if the operation failed. For
+ * on-demand pinning, the page is released whenever it isn't stored in the
+ * umem.
+ */
+static int ib_umem_odp_map_dma_single_page(
+ struct ib_umem *umem,
+ int page_index,
+ u64 base_virt_addr,
+ struct page *page,
+ u64 access_mask,
+ unsigned long current_seq)
+{
+ struct ib_device *dev = umem->context->device;
+ dma_addr_t dma_addr;
+ int stored_page = 0;
+ int remove_existing_mapping = 0;
+ int ret = 0;
+
+ mutex_lock(&umem->odp_data->umem_mutex);
+ /*
+ * Note: we avoid writing if seq is different from the initial seq, to
+ * handle case of a racing notifier. This check also allows us to bail
+ * early if we have a notifier running in parallel with us.
+ */
+ if (ib_umem_mmu_notifier_retry(umem, current_seq)) {
+ ret = -EAGAIN;
+ goto out;
+ }
+ if (!(umem->odp_data->dma_list[page_index])) {
+ dma_addr = ib_dma_map_page(dev,
+ page,
+ 0, PAGE_SIZE,
+ DMA_BIDIRECTIONAL);
+ if (ib_dma_mapping_error(dev, dma_addr)) {
+ ret = -EFAULT;
+ goto out;
+ }
+ umem->odp_data->dma_list[page_index] = dma_addr | access_mask;
+ umem->odp_data->page_list[page_index] = page;
+ stored_page = 1;
+ } else if (umem->odp_data->page_list[page_index] == page) {
+ umem->odp_data->dma_list[page_index] |= access_mask;
+ } else {
+ pr_err("error: got different pages in IB device and from get_user_pages. IB device page: %p, gup page: %p\n",
+ umem->odp_data->page_list[page_index], page);
+ /* Better remove the mapping now, to prevent any further
+ * damage. */
+ remove_existing_mapping = 1;
+ }
+
+out:
+ mutex_unlock(&umem->odp_data->umem_mutex);
+
+ /* On Demand Paging - avoid pinning the page */
+ if (umem->context->invalidate_range || !stored_page)
+ put_page(page);
+
+ if (remove_existing_mapping && umem->context->invalidate_range) {
+ invalidate_page_trampoline(
+ umem,
+ base_virt_addr + (page_index * PAGE_SIZE),
+ base_virt_addr + ((page_index+1)*PAGE_SIZE),
+ NULL);
+ ret = -EAGAIN;
+ }
+
+ return ret;
+}
+
+/**
+ * ib_umem_odp_map_dma_pages - Pin and DMA map userspace memory in an ODP MR.
+ *
+ * Pins the range of pages passed in the argument, and maps them to
+ * DMA addresses. The DMA addresses of the mapped pages is updated in
+ * umem->odp_data->dma_list.
+ *
+ * Returns the number of pages mapped in success, negative error code
+ * for failure.
+ * An -EAGAIN error code is returned when a concurrent mmu notifier prevents
+ * the function from completing its task.
+ *
+ * @umem: the umem to map and pin
+ * @user_virt: the address from which we need to map.
+ * @bcnt: the minimal number of bytes to pin and map. The mapping might be
+ * bigger due to alignment, and may also be smaller in case of an error
+ * pinning or mapping a page. The actual pages mapped is returned in
+ * the return value.
+ * @access_mask: bit mask of the requested access permissions for the given
+ * range.
+ * @current_seq: the MMU notifiers sequance value for synchronization with
+ * invalidations. the sequance number is read from
+ * umem->odp_data->notifiers_seq before calling this function
+ */
+int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt,
+ u64 access_mask, unsigned long current_seq)
+{
+ struct task_struct *owning_process = NULL;
+ struct mm_struct *owning_mm = NULL;
+ struct page **local_page_list = NULL;
+ u64 off;
+ int j, k, ret = 0, start_idx, npages = 0;
+ u64 base_virt_addr;
+
+ if (access_mask == 0)
+ return -EINVAL;
+
+ if (user_virt < ib_umem_start(umem) ||
+ user_virt + bcnt > ib_umem_end(umem))
+ return -EFAULT;
+
+ local_page_list = (struct page **)__get_free_page(GFP_KERNEL);
+ if (!local_page_list)
+ return -ENOMEM;
+
+ off = user_virt & (~PAGE_MASK);
+ user_virt = user_virt & PAGE_MASK;
+ base_virt_addr = user_virt;
+ bcnt += off; /* Charge for the first page offset as well. */
+
+ owning_process = get_pid_task(umem->context->tgid, PIDTYPE_PID);
+ if (owning_process == NULL) {
+ ret = -EINVAL;
+ goto out_no_task;
+ }
+
+ owning_mm = get_task_mm(owning_process);
+ if (owning_mm == NULL) {
+ ret = -EINVAL;
+ goto out_put_task;
+ }
+
+ start_idx = (user_virt - ib_umem_start(umem)) >> PAGE_SHIFT;
+ k = start_idx;
+
+ while (bcnt > 0) {
+ const size_t gup_num_pages =
+ min_t(size_t, ALIGN(bcnt, PAGE_SIZE) / PAGE_SIZE,
+ PAGE_SIZE / sizeof(struct page *));
+
+ down_read(&owning_mm->mmap_sem);
+ /*
+ * Note: this might result in redundent page getting. We can
+ * avoid this by checking dma_list to be 0 before calling
+ * get_user_pages. However, this make the code much more
+ * complex (and doesn't gain us much performance in most use
+ * cases).
+ */
+ npages = get_user_pages(owning_process, owning_mm, user_virt,
+ gup_num_pages,
+ access_mask & ODP_WRITE_ALLOWED_BIT, 0,
+ local_page_list, NULL);
+ up_read(&owning_mm->mmap_sem);
+
+ if (npages < 0)
+ break;
+
+ bcnt -= min_t(size_t, npages << PAGE_SHIFT, bcnt);
+ user_virt += npages << PAGE_SHIFT;
+ for (j = 0; j < npages; ++j) {
+ ret = ib_umem_odp_map_dma_single_page(
+ umem, k, base_virt_addr, local_page_list[j],
+ access_mask, current_seq);
+ if (ret < 0)
+ break;
+ k++;
+ }
+
+ if (ret < 0) {
+ /* Release left over pages when handling errors. */
+ for (++j; j < npages; ++j)
+ put_page(local_page_list[j]);
+ break;
+ }
+ }
+
+ if (ret >= 0) {
+ if (npages < 0 && k == start_idx)
+ ret = npages;
+ else
+ ret = k - start_idx;
+ }
+
+ mmput(owning_mm);
+out_put_task:
+ put_task_struct(owning_process);
+out_no_task:
+ free_page((unsigned long)local_page_list);
+ return ret;
+}
+EXPORT_SYMBOL(ib_umem_odp_map_dma_pages);
+
+void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 virt,
+ u64 bound)
+{
+ int idx;
+ u64 addr;
+ struct ib_device *dev = umem->context->device;
+
+ virt = max_t(u64, virt, ib_umem_start(umem));
+ bound = min_t(u64, bound, ib_umem_end(umem));
+ /* Note that during the run of this function, the
+ * notifiers_count of the MR is > 0, preventing any racing
+ * faults from completion. We might be racing with other
+ * invalidations, so we must make sure we free each page only
+ * once. */
+ for (addr = virt; addr < bound; addr += (u64)umem->page_size) {
+ idx = (addr - ib_umem_start(umem)) / PAGE_SIZE;
+ mutex_lock(&umem->odp_data->umem_mutex);
+ if (umem->odp_data->page_list[idx]) {
+ struct page *page = umem->odp_data->page_list[idx];
+ struct page *head_page = compound_head(page);
+ dma_addr_t dma = umem->odp_data->dma_list[idx];
+ dma_addr_t dma_addr = dma & ODP_DMA_ADDR_MASK;
+
+ WARN_ON(!dma_addr);
+
+ ib_dma_unmap_page(dev, dma_addr, PAGE_SIZE,
+ DMA_BIDIRECTIONAL);
+ if (dma & ODP_WRITE_ALLOWED_BIT)
+ /*
+ * set_page_dirty prefers being called with
+ * the page lock. However, MMU notifiers are
+ * called sometimes with and sometimes without
+ * the lock. We rely on the umem_mutex instead
+ * to prevent other mmu notifiers from
+ * continuing and allowing the page mapping to
+ * be removed.
+ */
+ set_page_dirty(head_page);
+ /* on demand pinning support */
+ if (!umem->context->invalidate_range)
+ put_page(page);
+ umem->odp_data->page_list[idx] = NULL;
+ umem->odp_data->dma_list[idx] = 0;
+ }
+ mutex_unlock(&umem->odp_data->umem_mutex);
+ }
+}
+EXPORT_SYMBOL(ib_umem_odp_unmap_dma_pages);
diff --git a/drivers/infiniband/core/umem_rbtree.c b/drivers/infiniband/core/umem_rbtree.c
new file mode 100644
index 0000000..727d788
--- /dev/null
+++ b/drivers/infiniband/core/umem_rbtree.c
@@ -0,0 +1,94 @@
+/*
+ * Copyright (c) 2014 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * 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.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/interval_tree_generic.h>
+#include <linux/sched.h>
+#include <linux/gfp.h>
+#include <rdma/ib_umem_odp.h>
+
+/*
+ * The ib_umem list keeps track of memory regions for which the HW
+ * device request to receive notification when the related memory
+ * mapping is changed.
+ *
+ * ib_umem_lock protects the list.
+ */
+
+static inline u64 node_start(struct umem_odp_node *n)
+{
+ struct ib_umem_odp *umem_odp =
+ container_of(n, struct ib_umem_odp, interval_tree);
+
+ return ib_umem_start(umem_odp->umem);
+}
+
+/* Note that the representation of the intervals in the interval tree
+ * considers the ending point as contained in the interval, while the
+ * function ib_umem_end returns the first address which is not contained
+ * in the umem.
+ */
+static inline u64 node_last(struct umem_odp_node *n)
+{
+ struct ib_umem_odp *umem_odp =
+ container_of(n, struct ib_umem_odp, interval_tree);
+
+ return ib_umem_end(umem_odp->umem) - 1;
+}
+
+INTERVAL_TREE_DEFINE(struct umem_odp_node, rb, u64, __subtree_last,
+ node_start, node_last, , rbt_ib_umem)
+
+/* @last is not a part of the interval. See comment for function
+ * node_last.
+ */
+int rbt_ib_umem_for_each_in_range(struct rb_root *root,
+ u64 start, u64 last,
+ umem_call_back cb,
+ void *cookie)
+{
+ int ret_val = 0;
+ struct umem_odp_node *node;
+ struct ib_umem_odp *umem;
+
+ if (unlikely(start == last))
+ return ret_val;
+
+ for (node = rbt_ib_umem_iter_first(root, start, last - 1); node;
+ node = rbt_ib_umem_iter_next(node, start, last - 1)) {
+ umem = container_of(node, struct ib_umem_odp, interval_tree);
+ ret_val = cb(umem->umem, start, last, cookie) || ret_val;
+ }
+
+ return ret_val;
+}
diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h
index 643c08a..b716b08 100644
--- a/drivers/infiniband/core/uverbs.h
+++ b/drivers/infiniband/core/uverbs.h
@@ -258,5 +258,6 @@
IB_UVERBS_DECLARE_EX_CMD(create_flow);
IB_UVERBS_DECLARE_EX_CMD(destroy_flow);
+IB_UVERBS_DECLARE_EX_CMD(query_device);
#endif /* UVERBS_H */
diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index 5ba2a86..532d8eba8 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -36,6 +36,7 @@
#include <linux/file.h>
#include <linux/fs.h>
#include <linux/slab.h>
+#include <linux/sched.h>
#include <asm/uaccess.h>
@@ -288,6 +289,9 @@
struct ib_uverbs_get_context_resp resp;
struct ib_udata udata;
struct ib_device *ibdev = file->device->ib_dev;
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ struct ib_device_attr dev_attr;
+#endif
struct ib_ucontext *ucontext;
struct file *filp;
int ret;
@@ -325,8 +329,25 @@
INIT_LIST_HEAD(&ucontext->ah_list);
INIT_LIST_HEAD(&ucontext->xrcd_list);
INIT_LIST_HEAD(&ucontext->rule_list);
+ rcu_read_lock();
+ ucontext->tgid = get_task_pid(current->group_leader, PIDTYPE_PID);
+ rcu_read_unlock();
ucontext->closing = 0;
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ ucontext->umem_tree = RB_ROOT;
+ init_rwsem(&ucontext->umem_rwsem);
+ ucontext->odp_mrs_count = 0;
+ INIT_LIST_HEAD(&ucontext->no_private_counters);
+
+ ret = ib_query_device(ibdev, &dev_attr);
+ if (ret)
+ goto err_free;
+ if (!(dev_attr.device_cap_flags & IB_DEVICE_ON_DEMAND_PAGING))
+ ucontext->invalidate_range = NULL;
+
+#endif
+
resp.num_comp_vectors = file->device->num_comp_vectors;
ret = get_unused_fd_flags(O_CLOEXEC);
@@ -371,6 +392,7 @@
put_unused_fd(resp.async_fd);
err_free:
+ put_pid(ucontext->tgid);
ibdev->dealloc_ucontext(ucontext);
err:
@@ -378,6 +400,52 @@
return ret;
}
+static void copy_query_dev_fields(struct ib_uverbs_file *file,
+ struct ib_uverbs_query_device_resp *resp,
+ struct ib_device_attr *attr)
+{
+ resp->fw_ver = attr->fw_ver;
+ resp->node_guid = file->device->ib_dev->node_guid;
+ resp->sys_image_guid = attr->sys_image_guid;
+ resp->max_mr_size = attr->max_mr_size;
+ resp->page_size_cap = attr->page_size_cap;
+ resp->vendor_id = attr->vendor_id;
+ resp->vendor_part_id = attr->vendor_part_id;
+ resp->hw_ver = attr->hw_ver;
+ resp->max_qp = attr->max_qp;
+ resp->max_qp_wr = attr->max_qp_wr;
+ resp->device_cap_flags = attr->device_cap_flags;
+ resp->max_sge = attr->max_sge;
+ resp->max_sge_rd = attr->max_sge_rd;
+ resp->max_cq = attr->max_cq;
+ resp->max_cqe = attr->max_cqe;
+ resp->max_mr = attr->max_mr;
+ resp->max_pd = attr->max_pd;
+ resp->max_qp_rd_atom = attr->max_qp_rd_atom;
+ resp->max_ee_rd_atom = attr->max_ee_rd_atom;
+ resp->max_res_rd_atom = attr->max_res_rd_atom;
+ resp->max_qp_init_rd_atom = attr->max_qp_init_rd_atom;
+ resp->max_ee_init_rd_atom = attr->max_ee_init_rd_atom;
+ resp->atomic_cap = attr->atomic_cap;
+ resp->max_ee = attr->max_ee;
+ resp->max_rdd = attr->max_rdd;
+ resp->max_mw = attr->max_mw;
+ resp->max_raw_ipv6_qp = attr->max_raw_ipv6_qp;
+ resp->max_raw_ethy_qp = attr->max_raw_ethy_qp;
+ resp->max_mcast_grp = attr->max_mcast_grp;
+ resp->max_mcast_qp_attach = attr->max_mcast_qp_attach;
+ resp->max_total_mcast_qp_attach = attr->max_total_mcast_qp_attach;
+ resp->max_ah = attr->max_ah;
+ resp->max_fmr = attr->max_fmr;
+ resp->max_map_per_fmr = attr->max_map_per_fmr;
+ resp->max_srq = attr->max_srq;
+ resp->max_srq_wr = attr->max_srq_wr;
+ resp->max_srq_sge = attr->max_srq_sge;
+ resp->max_pkeys = attr->max_pkeys;
+ resp->local_ca_ack_delay = attr->local_ca_ack_delay;
+ resp->phys_port_cnt = file->device->ib_dev->phys_port_cnt;
+}
+
ssize_t ib_uverbs_query_device(struct ib_uverbs_file *file,
const char __user *buf,
int in_len, int out_len)
@@ -398,47 +466,7 @@
return ret;
memset(&resp, 0, sizeof resp);
-
- resp.fw_ver = attr.fw_ver;
- resp.node_guid = file->device->ib_dev->node_guid;
- resp.sys_image_guid = attr.sys_image_guid;
- resp.max_mr_size = attr.max_mr_size;
- resp.page_size_cap = attr.page_size_cap;
- resp.vendor_id = attr.vendor_id;
- resp.vendor_part_id = attr.vendor_part_id;
- resp.hw_ver = attr.hw_ver;
- resp.max_qp = attr.max_qp;
- resp.max_qp_wr = attr.max_qp_wr;
- resp.device_cap_flags = attr.device_cap_flags;
- resp.max_sge = attr.max_sge;
- resp.max_sge_rd = attr.max_sge_rd;
- resp.max_cq = attr.max_cq;
- resp.max_cqe = attr.max_cqe;
- resp.max_mr = attr.max_mr;
- resp.max_pd = attr.max_pd;
- resp.max_qp_rd_atom = attr.max_qp_rd_atom;
- resp.max_ee_rd_atom = attr.max_ee_rd_atom;
- resp.max_res_rd_atom = attr.max_res_rd_atom;
- resp.max_qp_init_rd_atom = attr.max_qp_init_rd_atom;
- resp.max_ee_init_rd_atom = attr.max_ee_init_rd_atom;
- resp.atomic_cap = attr.atomic_cap;
- resp.max_ee = attr.max_ee;
- resp.max_rdd = attr.max_rdd;
- resp.max_mw = attr.max_mw;
- resp.max_raw_ipv6_qp = attr.max_raw_ipv6_qp;
- resp.max_raw_ethy_qp = attr.max_raw_ethy_qp;
- resp.max_mcast_grp = attr.max_mcast_grp;
- resp.max_mcast_qp_attach = attr.max_mcast_qp_attach;
- resp.max_total_mcast_qp_attach = attr.max_total_mcast_qp_attach;
- resp.max_ah = attr.max_ah;
- resp.max_fmr = attr.max_fmr;
- resp.max_map_per_fmr = attr.max_map_per_fmr;
- resp.max_srq = attr.max_srq;
- resp.max_srq_wr = attr.max_srq_wr;
- resp.max_srq_sge = attr.max_srq_sge;
- resp.max_pkeys = attr.max_pkeys;
- resp.local_ca_ack_delay = attr.local_ca_ack_delay;
- resp.phys_port_cnt = file->device->ib_dev->phys_port_cnt;
+ copy_query_dev_fields(file, &resp, &attr);
if (copy_to_user((void __user *) (unsigned long) cmd.response,
&resp, sizeof resp))
@@ -947,6 +975,18 @@
goto err_free;
}
+ if (cmd.access_flags & IB_ACCESS_ON_DEMAND) {
+ struct ib_device_attr attr;
+
+ ret = ib_query_device(pd->device, &attr);
+ if (ret || !(attr.device_cap_flags &
+ IB_DEVICE_ON_DEMAND_PAGING)) {
+ pr_debug("ODP support not available\n");
+ ret = -EINVAL;
+ goto err_put;
+ }
+ }
+
mr = pd->device->reg_user_mr(pd, cmd.start, cmd.length, cmd.hca_va,
cmd.access_flags, &udata);
if (IS_ERR(mr)) {
@@ -3253,3 +3293,52 @@
return ret ? ret : in_len;
}
+
+int ib_uverbs_ex_query_device(struct ib_uverbs_file *file,
+ struct ib_udata *ucore,
+ struct ib_udata *uhw)
+{
+ struct ib_uverbs_ex_query_device_resp resp;
+ struct ib_uverbs_ex_query_device cmd;
+ struct ib_device_attr attr;
+ struct ib_device *device;
+ int err;
+
+ device = file->device->ib_dev;
+ if (ucore->inlen < sizeof(cmd))
+ return -EINVAL;
+
+ err = ib_copy_from_udata(&cmd, ucore, sizeof(cmd));
+ if (err)
+ return err;
+
+ if (cmd.reserved)
+ return -EINVAL;
+
+ err = device->query_device(device, &attr);
+ if (err)
+ return err;
+
+ memset(&resp, 0, sizeof(resp));
+ copy_query_dev_fields(file, &resp.base, &attr);
+ resp.comp_mask = 0;
+
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ if (cmd.comp_mask & IB_USER_VERBS_EX_QUERY_DEVICE_ODP) {
+ resp.odp_caps.general_caps = attr.odp_caps.general_caps;
+ resp.odp_caps.per_transport_caps.rc_odp_caps =
+ attr.odp_caps.per_transport_caps.rc_odp_caps;
+ resp.odp_caps.per_transport_caps.uc_odp_caps =
+ attr.odp_caps.per_transport_caps.uc_odp_caps;
+ resp.odp_caps.per_transport_caps.ud_odp_caps =
+ attr.odp_caps.per_transport_caps.ud_odp_caps;
+ resp.comp_mask |= IB_USER_VERBS_EX_QUERY_DEVICE_ODP;
+ }
+#endif
+
+ err = ib_copy_to_udata(ucore, &resp, sizeof(resp));
+ if (err)
+ return err;
+
+ return 0;
+}
diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c
index 71ab83f..e6c23b9 100644
--- a/drivers/infiniband/core/uverbs_main.c
+++ b/drivers/infiniband/core/uverbs_main.c
@@ -122,7 +122,8 @@
struct ib_udata *ucore,
struct ib_udata *uhw) = {
[IB_USER_VERBS_EX_CMD_CREATE_FLOW] = ib_uverbs_ex_create_flow,
- [IB_USER_VERBS_EX_CMD_DESTROY_FLOW] = ib_uverbs_ex_destroy_flow
+ [IB_USER_VERBS_EX_CMD_DESTROY_FLOW] = ib_uverbs_ex_destroy_flow,
+ [IB_USER_VERBS_EX_CMD_QUERY_DEVICE] = ib_uverbs_ex_query_device
};
static void ib_uverbs_add_one(struct ib_device *device);
@@ -296,6 +297,8 @@
kfree(uobj);
}
+ put_pid(context->tgid);
+
return context->device->dealloc_ucontext(context);
}
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
index c2b89cc..f93eb8d 100644
--- a/drivers/infiniband/core/verbs.c
+++ b/drivers/infiniband/core/verbs.c
@@ -879,7 +879,8 @@
if (rdma_link_local_addr((struct in6_addr *)qp_attr->ah_attr.grh.dgid.raw)) {
rdma_get_ll_mac((struct in6_addr *)qp_attr->ah_attr.grh.dgid.raw, qp_attr->ah_attr.dmac);
rdma_get_ll_mac((struct in6_addr *)sgid.raw, qp_attr->smac);
- qp_attr->vlan_id = rdma_get_vlan_id(&sgid);
+ if (!(*qp_attr_mask & IB_QP_VID))
+ qp_attr->vlan_id = rdma_get_vlan_id(&sgid);
} else {
ret = rdma_addr_find_dmac_by_grh(&sgid, &qp_attr->ah_attr.grh.dgid,
qp_attr->ah_attr.dmac, &qp_attr->vlan_id);
diff --git a/drivers/infiniband/hw/amso1100/c2_provider.c b/drivers/infiniband/hw/amso1100/c2_provider.c
index 2d5cbf4..bdf3507 100644
--- a/drivers/infiniband/hw/amso1100/c2_provider.c
+++ b/drivers/infiniband/hw/amso1100/c2_provider.c
@@ -476,7 +476,7 @@
c2mr->umem->page_size,
i,
length,
- c2mr->umem->offset,
+ ib_umem_offset(c2mr->umem),
&kva,
c2_convert_access(acc),
c2mr);
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 4b8c611..9edc200 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1640,7 +1640,8 @@
__state_set(&ep->com, MPA_REQ_RCVD);
/* drive upcall */
- mutex_lock(&ep->parent_ep->com.mutex);
+ mutex_lock_nested(&ep->parent_ep->com.mutex,
+ SINGLE_DEPTH_NESTING);
if (ep->parent_ep->com.state != DEAD) {
if (connect_request_upcall(ep))
abort_connection(ep, skb, GFP_KERNEL);
@@ -3126,6 +3127,8 @@
err = c4iw_wait_for_reply(&ep->com.dev->rdev,
&ep->com.wr_wait,
0, 0, __func__);
+ else if (err > 0)
+ err = net_xmit_errno(err);
if (err)
pr_err("cxgb4_create_server6/filter failed err %d stid %d laddr %pI6 lport %d\n",
err, ep->stid,
@@ -3159,6 +3162,8 @@
err = c4iw_wait_for_reply(&ep->com.dev->rdev,
&ep->com.wr_wait,
0, 0, __func__);
+ else if (err > 0)
+ err = net_xmit_errno(err);
}
if (err)
pr_err("cxgb4_create_server/filter failed err %d stid %d laddr %pI4 lport %d\n"
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index 72f1f05..eb5df4e 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -670,7 +670,7 @@
idr_for_each(&epd->devp->stid_idr, count_idrs, &count);
spin_unlock_irq(&epd->devp->lock);
- epd->bufsize = count * 160;
+ epd->bufsize = count * 240;
epd->buf = vmalloc(epd->bufsize);
if (!epd->buf) {
ret = -ENOMEM;
diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c
index 0744455..cb43c22 100644
--- a/drivers/infiniband/hw/cxgb4/mem.c
+++ b/drivers/infiniband/hw/cxgb4/mem.c
@@ -50,6 +50,13 @@
module_param(inline_threshold, int, 0644);
MODULE_PARM_DESC(inline_threshold, "inline vs dsgl threshold (default=128)");
+static int mr_exceeds_hw_limits(struct c4iw_dev *dev, u64 length)
+{
+ return (is_t4(dev->rdev.lldi.adapter_type) ||
+ is_t5(dev->rdev.lldi.adapter_type)) &&
+ length >= 8*1024*1024*1024ULL;
+}
+
static int _c4iw_write_mem_dma_aligned(struct c4iw_rdev *rdev, u32 addr,
u32 len, dma_addr_t data, int wait)
{
@@ -369,9 +376,11 @@
int ret;
ret = write_tpt_entry(&rhp->rdev, 0, &stag, 1, mhp->attr.pdid,
- FW_RI_STAG_NSMR, mhp->attr.perms,
+ FW_RI_STAG_NSMR, mhp->attr.len ?
+ mhp->attr.perms : 0,
mhp->attr.mw_bind_enable, mhp->attr.zbva,
- mhp->attr.va_fbo, mhp->attr.len, shift - 12,
+ mhp->attr.va_fbo, mhp->attr.len ?
+ mhp->attr.len : -1, shift - 12,
mhp->attr.pbl_size, mhp->attr.pbl_addr);
if (ret)
return ret;
@@ -536,6 +545,11 @@
return ret;
}
+ if (mr_exceeds_hw_limits(rhp, total_size)) {
+ kfree(page_list);
+ return -EINVAL;
+ }
+
ret = reregister_mem(rhp, php, &mh, shift, npages);
kfree(page_list);
if (ret)
@@ -596,6 +610,12 @@
if (ret)
goto err;
+ if (mr_exceeds_hw_limits(rhp, total_size)) {
+ kfree(page_list);
+ ret = -EINVAL;
+ goto err;
+ }
+
ret = alloc_pbl(mhp, npages);
if (ret) {
kfree(page_list);
@@ -699,6 +719,10 @@
php = to_c4iw_pd(pd);
rhp = php->rhp;
+
+ if (mr_exceeds_hw_limits(rhp, length))
+ return ERR_PTR(-EINVAL);
+
mhp = kzalloc(sizeof(*mhp), GFP_KERNEL);
if (!mhp)
return ERR_PTR(-ENOMEM);
diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c
index 2ed3ece..bb85d47 100644
--- a/drivers/infiniband/hw/cxgb4/qp.c
+++ b/drivers/infiniband/hw/cxgb4/qp.c
@@ -1538,9 +1538,9 @@
set_state(qhp, C4IW_QP_STATE_ERROR);
free = 1;
abort = 1;
- wake_up(&qhp->wait);
BUG_ON(!ep);
flush_qp(qhp);
+ wake_up(&qhp->wait);
out:
mutex_unlock(&qhp->mutex);
diff --git a/drivers/infiniband/hw/ehca/ehca_mrmw.c b/drivers/infiniband/hw/ehca/ehca_mrmw.c
index 3488e8c..f914b30 100644
--- a/drivers/infiniband/hw/ehca/ehca_mrmw.c
+++ b/drivers/infiniband/hw/ehca/ehca_mrmw.c
@@ -399,7 +399,7 @@
pginfo.num_kpages = num_kpages;
pginfo.num_hwpages = num_hwpages;
pginfo.u.usr.region = e_mr->umem;
- pginfo.next_hwpage = e_mr->umem->offset / hwpage_size;
+ pginfo.next_hwpage = ib_umem_offset(e_mr->umem) / hwpage_size;
pginfo.u.usr.next_sg = pginfo.u.usr.region->sg_head.sgl;
ret = ehca_reg_mr(shca, e_mr, (u64 *)virt, length, mr_access_flags,
e_pd, &pginfo, &e_mr->ib.ib_mr.lkey,
diff --git a/drivers/infiniband/hw/ipath/ipath_mr.c b/drivers/infiniband/hw/ipath/ipath_mr.c
index 5e61e9b..c7278f6 100644
--- a/drivers/infiniband/hw/ipath/ipath_mr.c
+++ b/drivers/infiniband/hw/ipath/ipath_mr.c
@@ -214,7 +214,7 @@
mr->mr.user_base = start;
mr->mr.iova = virt_addr;
mr->mr.length = length;
- mr->mr.offset = umem->offset;
+ mr->mr.offset = ib_umem_offset(umem);
mr->mr.access_flags = mr_access_flags;
mr->mr.max_segs = n;
mr->umem = umem;
diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c
index 8f9325c..c36ccbd 100644
--- a/drivers/infiniband/hw/mlx4/mr.c
+++ b/drivers/infiniband/hw/mlx4/mr.c
@@ -223,7 +223,6 @@
if (flags & IB_MR_REREG_TRANS) {
int shift;
- int err;
int n;
mlx4_mr_rereg_mem_cleanup(dev->dev, &mmr->mmr);
diff --git a/drivers/infiniband/hw/mlx5/Makefile b/drivers/infiniband/hw/mlx5/Makefile
index 4ea0135..27a7015 100644
--- a/drivers/infiniband/hw/mlx5/Makefile
+++ b/drivers/infiniband/hw/mlx5/Makefile
@@ -1,3 +1,4 @@
obj-$(CONFIG_MLX5_INFINIBAND) += mlx5_ib.o
mlx5_ib-y := main.o cq.o doorbell.o qp.o mem.o srq.o mr.o ah.o mad.o
+mlx5_ib-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += odp.o
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 1ba6c42..8a87404 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -244,6 +244,12 @@
props->max_mcast_grp;
props->max_map_per_fmr = INT_MAX; /* no limit in ConnectIB */
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ if (dev->mdev->caps.gen.flags & MLX5_DEV_CAP_FLAG_ON_DMND_PG)
+ props->device_cap_flags |= IB_DEVICE_ON_DEMAND_PAGING;
+ props->odp_caps = dev->odp_caps;
+#endif
+
out:
kfree(in_mad);
kfree(out_mad);
@@ -568,6 +574,10 @@
goto out_count;
}
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ context->ibucontext.invalidate_range = &mlx5_ib_invalidate_range;
+#endif
+
INIT_LIST_HEAD(&context->db_page_list);
mutex_init(&context->db_page_mutex);
@@ -858,7 +868,7 @@
struct mlx5_ib_dev *dev =
container_of(device, struct mlx5_ib_dev, ib_dev.dev);
- return sprintf(buf, "%d\n", dev->mdev->priv.reg_pages);
+ return sprintf(buf, "%d\n", atomic_read(&dev->mdev->priv.reg_pages));
}
static ssize_t show_hca(struct device *device, struct device_attribute *attr,
@@ -1321,6 +1331,8 @@
(1ull << IB_USER_VERBS_CMD_DESTROY_SRQ) |
(1ull << IB_USER_VERBS_CMD_CREATE_XSRQ) |
(1ull << IB_USER_VERBS_CMD_OPEN_QP);
+ dev->ib_dev.uverbs_ex_cmd_mask =
+ (1ull << IB_USER_VERBS_EX_CMD_QUERY_DEVICE);
dev->ib_dev.query_device = mlx5_ib_query_device;
dev->ib_dev.query_port = mlx5_ib_query_port;
@@ -1366,6 +1378,8 @@
dev->ib_dev.free_fast_reg_page_list = mlx5_ib_free_fast_reg_page_list;
dev->ib_dev.check_mr_status = mlx5_ib_check_mr_status;
+ mlx5_ib_internal_query_odp_caps(dev);
+
if (mdev->caps.gen.flags & MLX5_DEV_CAP_FLAG_XRC) {
dev->ib_dev.alloc_xrcd = mlx5_ib_alloc_xrcd;
dev->ib_dev.dealloc_xrcd = mlx5_ib_dealloc_xrcd;
@@ -1379,16 +1393,19 @@
goto err_eqs;
mutex_init(&dev->cap_mask_mutex);
- spin_lock_init(&dev->mr_lock);
err = create_dev_resources(&dev->devr);
if (err)
goto err_eqs;
- err = ib_register_device(&dev->ib_dev, NULL);
+ err = mlx5_ib_odp_init_one(dev);
if (err)
goto err_rsrc;
+ err = ib_register_device(&dev->ib_dev, NULL);
+ if (err)
+ goto err_odp;
+
err = create_umr_res(dev);
if (err)
goto err_dev;
@@ -1410,6 +1427,9 @@
err_dev:
ib_unregister_device(&dev->ib_dev);
+err_odp:
+ mlx5_ib_odp_remove_one(dev);
+
err_rsrc:
destroy_dev_resources(&dev->devr);
@@ -1425,8 +1445,10 @@
static void mlx5_ib_remove(struct mlx5_core_dev *mdev, void *context)
{
struct mlx5_ib_dev *dev = context;
+
ib_unregister_device(&dev->ib_dev);
destroy_umrc_res(dev);
+ mlx5_ib_odp_remove_one(dev);
destroy_dev_resources(&dev->devr);
free_comp_eqs(dev);
ib_dealloc_device(&dev->ib_dev);
@@ -1440,15 +1462,30 @@
static int __init mlx5_ib_init(void)
{
+ int err;
+
if (deprecated_prof_sel != 2)
pr_warn("prof_sel is deprecated for mlx5_ib, set it for mlx5_core\n");
- return mlx5_register_interface(&mlx5_ib_interface);
+ err = mlx5_ib_odp_init();
+ if (err)
+ return err;
+
+ err = mlx5_register_interface(&mlx5_ib_interface);
+ if (err)
+ goto clean_odp;
+
+ return err;
+
+clean_odp:
+ mlx5_ib_odp_cleanup();
+ return err;
}
static void __exit mlx5_ib_cleanup(void)
{
mlx5_unregister_interface(&mlx5_ib_interface);
+ mlx5_ib_odp_cleanup();
}
module_init(mlx5_ib_init);
diff --git a/drivers/infiniband/hw/mlx5/mem.c b/drivers/infiniband/hw/mlx5/mem.c
index dae07ea..b56e4c5 100644
--- a/drivers/infiniband/hw/mlx5/mem.c
+++ b/drivers/infiniband/hw/mlx5/mem.c
@@ -32,6 +32,7 @@
#include <linux/module.h>
#include <rdma/ib_umem.h>
+#include <rdma/ib_umem_odp.h>
#include "mlx5_ib.h"
/* @umem: umem object to scan
@@ -57,6 +58,17 @@
int entry;
unsigned long page_shift = ilog2(umem->page_size);
+ /* With ODP we must always match OS page size. */
+ if (umem->odp_data) {
+ *count = ib_umem_page_count(umem);
+ *shift = PAGE_SHIFT;
+ *ncont = *count;
+ if (order)
+ *order = ilog2(roundup_pow_of_two(*count));
+
+ return;
+ }
+
addr = addr >> page_shift;
tmp = (unsigned long)addr;
m = find_first_bit(&tmp, sizeof(tmp));
@@ -108,8 +120,36 @@
*count = i;
}
-void mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem,
- int page_shift, __be64 *pas, int umr)
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+static u64 umem_dma_to_mtt(dma_addr_t umem_dma)
+{
+ u64 mtt_entry = umem_dma & ODP_DMA_ADDR_MASK;
+
+ if (umem_dma & ODP_READ_ALLOWED_BIT)
+ mtt_entry |= MLX5_IB_MTT_READ;
+ if (umem_dma & ODP_WRITE_ALLOWED_BIT)
+ mtt_entry |= MLX5_IB_MTT_WRITE;
+
+ return mtt_entry;
+}
+#endif
+
+/*
+ * Populate the given array with bus addresses from the umem.
+ *
+ * dev - mlx5_ib device
+ * umem - umem to use to fill the pages
+ * page_shift - determines the page size used in the resulting array
+ * offset - offset into the umem to start from,
+ * only implemented for ODP umems
+ * num_pages - total number of pages to fill
+ * pas - bus addresses array to fill
+ * access_flags - access flags to set on all present pages.
+ use enum mlx5_ib_mtt_access_flags for this.
+ */
+void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem,
+ int page_shift, size_t offset, size_t num_pages,
+ __be64 *pas, int access_flags)
{
unsigned long umem_page_shift = ilog2(umem->page_size);
int shift = page_shift - umem_page_shift;
@@ -120,6 +160,21 @@
int len;
struct scatterlist *sg;
int entry;
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ const bool odp = umem->odp_data != NULL;
+
+ if (odp) {
+ WARN_ON(shift != 0);
+ WARN_ON(access_flags != (MLX5_IB_MTT_READ | MLX5_IB_MTT_WRITE));
+
+ for (i = 0; i < num_pages; ++i) {
+ dma_addr_t pa = umem->odp_data->dma_list[offset + i];
+
+ pas[i] = cpu_to_be64(umem_dma_to_mtt(pa));
+ }
+ return;
+ }
+#endif
i = 0;
for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) {
@@ -128,8 +183,7 @@
for (k = 0; k < len; k++) {
if (!(i & mask)) {
cur = base + (k << umem_page_shift);
- if (umr)
- cur |= 3;
+ cur |= access_flags;
pas[i >> shift] = cpu_to_be64(cur);
mlx5_ib_dbg(dev, "pas[%d] 0x%llx\n",
@@ -142,6 +196,13 @@
}
}
+void mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem,
+ int page_shift, __be64 *pas, int access_flags)
+{
+ return __mlx5_ib_populate_pas(dev, umem, page_shift, 0,
+ ib_umem_num_pages(umem), pas,
+ access_flags);
+}
int mlx5_ib_get_buf_offset(u64 addr, int page_shift, u32 *offset)
{
u64 page_size;
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index 386780f..83f22fe 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -111,6 +111,8 @@
*/
#define MLX5_IB_SEND_UMR_UNREG IB_SEND_RESERVED_START
+#define MLX5_IB_SEND_UMR_FAIL_IF_FREE (IB_SEND_RESERVED_START << 1)
+#define MLX5_IB_SEND_UMR_UPDATE_MTT (IB_SEND_RESERVED_START << 2)
#define MLX5_IB_QPT_REG_UMR IB_QPT_RESERVED1
#define MLX5_IB_WR_UMR IB_WR_RESERVED1
@@ -147,6 +149,29 @@
MLX5_QP_EMPTY
};
+/*
+ * Connect-IB can trigger up to four concurrent pagefaults
+ * per-QP.
+ */
+enum mlx5_ib_pagefault_context {
+ MLX5_IB_PAGEFAULT_RESPONDER_READ,
+ MLX5_IB_PAGEFAULT_REQUESTOR_READ,
+ MLX5_IB_PAGEFAULT_RESPONDER_WRITE,
+ MLX5_IB_PAGEFAULT_REQUESTOR_WRITE,
+ MLX5_IB_PAGEFAULT_CONTEXTS
+};
+
+static inline enum mlx5_ib_pagefault_context
+ mlx5_ib_get_pagefault_context(struct mlx5_pagefault *pagefault)
+{
+ return pagefault->flags & (MLX5_PFAULT_REQUESTOR | MLX5_PFAULT_WRITE);
+}
+
+struct mlx5_ib_pfault {
+ struct work_struct work;
+ struct mlx5_pagefault mpfault;
+};
+
struct mlx5_ib_qp {
struct ib_qp ibqp;
struct mlx5_core_qp mqp;
@@ -192,6 +217,21 @@
/* Store signature errors */
bool signature_en;
+
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ /*
+ * A flag that is true for QP's that are in a state that doesn't
+ * allow page faults, and shouldn't schedule any more faults.
+ */
+ int disable_page_faults;
+ /*
+ * The disable_page_faults_lock protects a QP's disable_page_faults
+ * field, allowing for a thread to atomically check whether the QP
+ * allows page faults, and if so schedule a page fault.
+ */
+ spinlock_t disable_page_faults_lock;
+ struct mlx5_ib_pfault pagefaults[MLX5_IB_PAGEFAULT_CONTEXTS];
+#endif
};
struct mlx5_ib_cq_buf {
@@ -206,6 +246,19 @@
MLX5_IB_QP_SIGNATURE_HANDLING = 1 << 1,
};
+struct mlx5_umr_wr {
+ union {
+ u64 virt_addr;
+ u64 offset;
+ } target;
+ struct ib_pd *pd;
+ unsigned int page_shift;
+ unsigned int npages;
+ u32 length;
+ int access_flags;
+ u32 mkey;
+};
+
struct mlx5_shared_mr_info {
int mr_id;
struct ib_umem *umem;
@@ -253,6 +306,13 @@
u32 xrcdn;
};
+enum mlx5_ib_mtt_access_flags {
+ MLX5_IB_MTT_READ = (1 << 0),
+ MLX5_IB_MTT_WRITE = (1 << 1),
+};
+
+#define MLX5_IB_MTT_PRESENT (MLX5_IB_MTT_READ | MLX5_IB_MTT_WRITE)
+
struct mlx5_ib_mr {
struct ib_mr ibmr;
struct mlx5_core_mr mmr;
@@ -261,12 +321,11 @@
struct list_head list;
int order;
int umred;
- __be64 *pas;
- dma_addr_t dma;
int npages;
struct mlx5_ib_dev *dev;
struct mlx5_create_mkey_mbox_out out;
struct mlx5_core_sig_ctx *sig;
+ int live;
};
struct mlx5_ib_fast_reg_page_list {
@@ -372,11 +431,18 @@
struct umr_common umrc;
/* sync used page count stats
*/
- spinlock_t mr_lock;
struct mlx5_ib_resources devr;
struct mlx5_mr_cache cache;
struct timer_list delay_timer;
int fill_delay;
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ struct ib_odp_caps odp_caps;
+ /*
+ * Sleepable RCU that prevents destruction of MRs while they are still
+ * being used by a page fault handler.
+ */
+ struct srcu_struct mr_srcu;
+#endif
};
static inline struct mlx5_ib_cq *to_mibcq(struct mlx5_core_cq *mcq)
@@ -490,6 +556,8 @@
int mlx5_ib_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr,
struct ib_recv_wr **bad_wr);
void *mlx5_get_send_wqe(struct mlx5_ib_qp *qp, int n);
+int mlx5_ib_read_user_wqe(struct mlx5_ib_qp *qp, int send, int wqe_index,
+ void *buffer, u32 length);
struct ib_cq *mlx5_ib_create_cq(struct ib_device *ibdev, int entries,
int vector, struct ib_ucontext *context,
struct ib_udata *udata);
@@ -502,6 +570,8 @@
struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
u64 virt_addr, int access_flags,
struct ib_udata *udata);
+int mlx5_ib_update_mtt(struct mlx5_ib_mr *mr, u64 start_page_index,
+ int npages, int zap);
int mlx5_ib_dereg_mr(struct ib_mr *ibmr);
int mlx5_ib_destroy_mr(struct ib_mr *ibmr);
struct ib_mr *mlx5_ib_create_mr(struct ib_pd *pd,
@@ -533,8 +603,11 @@
void mlx5_ib_cleanup_fmr(struct mlx5_ib_dev *dev);
void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, int *count, int *shift,
int *ncont, int *order);
+void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem,
+ int page_shift, size_t offset, size_t num_pages,
+ __be64 *pas, int access_flags);
void mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem,
- int page_shift, __be64 *pas, int umr);
+ int page_shift, __be64 *pas, int access_flags);
void mlx5_ib_copy_pas(u64 *old, u64 *new, int step, int num);
int mlx5_ib_get_cqe_size(struct mlx5_ib_dev *dev, struct ib_cq *ibcq);
int mlx5_mr_cache_init(struct mlx5_ib_dev *dev);
@@ -544,6 +617,38 @@
int mlx5_ib_check_mr_status(struct ib_mr *ibmr, u32 check_mask,
struct ib_mr_status *mr_status);
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+extern struct workqueue_struct *mlx5_ib_page_fault_wq;
+
+int mlx5_ib_internal_query_odp_caps(struct mlx5_ib_dev *dev);
+void mlx5_ib_mr_pfault_handler(struct mlx5_ib_qp *qp,
+ struct mlx5_ib_pfault *pfault);
+void mlx5_ib_odp_create_qp(struct mlx5_ib_qp *qp);
+int mlx5_ib_odp_init_one(struct mlx5_ib_dev *ibdev);
+void mlx5_ib_odp_remove_one(struct mlx5_ib_dev *ibdev);
+int __init mlx5_ib_odp_init(void);
+void mlx5_ib_odp_cleanup(void);
+void mlx5_ib_qp_disable_pagefaults(struct mlx5_ib_qp *qp);
+void mlx5_ib_qp_enable_pagefaults(struct mlx5_ib_qp *qp);
+void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start,
+ unsigned long end);
+
+#else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */
+static inline int mlx5_ib_internal_query_odp_caps(struct mlx5_ib_dev *dev)
+{
+ return 0;
+}
+
+static inline void mlx5_ib_odp_create_qp(struct mlx5_ib_qp *qp) {}
+static inline int mlx5_ib_odp_init_one(struct mlx5_ib_dev *ibdev) { return 0; }
+static inline void mlx5_ib_odp_remove_one(struct mlx5_ib_dev *ibdev) {}
+static inline int mlx5_ib_odp_init(void) { return 0; }
+static inline void mlx5_ib_odp_cleanup(void) {}
+static inline void mlx5_ib_qp_disable_pagefaults(struct mlx5_ib_qp *qp) {}
+static inline void mlx5_ib_qp_enable_pagefaults(struct mlx5_ib_qp *qp) {}
+
+#endif /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */
+
static inline void init_query_mad(struct ib_smp *mad)
{
mad->base_version = 1;
@@ -561,4 +666,7 @@
MLX5_PERM_LOCAL_READ;
}
+#define MLX5_MAX_UMR_SHIFT 16
+#define MLX5_MAX_UMR_PAGES (1 << MLX5_MAX_UMR_SHIFT)
+
#endif /* MLX5_IB_H */
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 5a80dd9..32a28bd 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -37,21 +37,34 @@
#include <linux/export.h>
#include <linux/delay.h>
#include <rdma/ib_umem.h>
+#include <rdma/ib_umem_odp.h>
+#include <rdma/ib_verbs.h>
#include "mlx5_ib.h"
enum {
MAX_PENDING_REG_MR = 8,
};
-enum {
- MLX5_UMR_ALIGN = 2048
-};
+#define MLX5_UMR_ALIGN 2048
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+static __be64 mlx5_ib_update_mtt_emergency_buffer[
+ MLX5_UMR_MTT_MIN_CHUNK_SIZE/sizeof(__be64)]
+ __aligned(MLX5_UMR_ALIGN);
+static DEFINE_MUTEX(mlx5_ib_update_mtt_emergency_buffer_mutex);
+#endif
-static __be64 *mr_align(__be64 *ptr, int align)
+static int clean_mr(struct mlx5_ib_mr *mr);
+
+static int destroy_mkey(struct mlx5_ib_dev *dev, struct mlx5_ib_mr *mr)
{
- unsigned long mask = align - 1;
+ int err = mlx5_core_destroy_mkey(dev->mdev, &mr->mmr);
- return (__be64 *)(((unsigned long)ptr + mask) & ~mask);
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ /* Wait until all page fault handlers using the mr complete. */
+ synchronize_srcu(&dev->mr_srcu);
+#endif
+
+ return err;
}
static int order2idx(struct mlx5_ib_dev *dev, int order)
@@ -146,7 +159,7 @@
mr->order = ent->order;
mr->umred = 1;
mr->dev = dev;
- in->seg.status = 1 << 6;
+ in->seg.status = MLX5_MKEY_STATUS_FREE;
in->seg.xlt_oct_size = cpu_to_be32((npages + 1) / 2);
in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8);
in->seg.flags = MLX5_ACCESS_MODE_MTT | MLX5_PERM_UMR_EN;
@@ -191,7 +204,7 @@
ent->cur--;
ent->size--;
spin_unlock_irq(&ent->lock);
- err = mlx5_core_destroy_mkey(dev->mdev, &mr->mmr);
+ err = destroy_mkey(dev, mr);
if (err)
mlx5_ib_warn(dev, "failed destroy mkey\n");
else
@@ -482,7 +495,7 @@
ent->cur--;
ent->size--;
spin_unlock_irq(&ent->lock);
- err = mlx5_core_destroy_mkey(dev->mdev, &mr->mmr);
+ err = destroy_mkey(dev, mr);
if (err)
mlx5_ib_warn(dev, "failed destroy mkey\n");
else
@@ -668,7 +681,7 @@
static int use_umr(int order)
{
- return order <= 17;
+ return order <= MLX5_MAX_UMR_SHIFT;
}
static void prep_umr_reg_wqe(struct ib_pd *pd, struct ib_send_wr *wr,
@@ -678,6 +691,7 @@
{
struct mlx5_ib_dev *dev = to_mdev(pd->device);
struct ib_mr *mr = dev->umrc.mr;
+ struct mlx5_umr_wr *umrwr = (struct mlx5_umr_wr *)&wr->wr.fast_reg;
sg->addr = dma;
sg->length = ALIGN(sizeof(u64) * n, 64);
@@ -692,21 +706,24 @@
wr->num_sge = 0;
wr->opcode = MLX5_IB_WR_UMR;
- wr->wr.fast_reg.page_list_len = n;
- wr->wr.fast_reg.page_shift = page_shift;
- wr->wr.fast_reg.rkey = key;
- wr->wr.fast_reg.iova_start = virt_addr;
- wr->wr.fast_reg.length = len;
- wr->wr.fast_reg.access_flags = access_flags;
- wr->wr.fast_reg.page_list = (struct ib_fast_reg_page_list *)pd;
+
+ umrwr->npages = n;
+ umrwr->page_shift = page_shift;
+ umrwr->mkey = key;
+ umrwr->target.virt_addr = virt_addr;
+ umrwr->length = len;
+ umrwr->access_flags = access_flags;
+ umrwr->pd = pd;
}
static void prep_umr_unreg_wqe(struct mlx5_ib_dev *dev,
struct ib_send_wr *wr, u32 key)
{
- wr->send_flags = MLX5_IB_SEND_UMR_UNREG;
+ struct mlx5_umr_wr *umrwr = (struct mlx5_umr_wr *)&wr->wr.fast_reg;
+
+ wr->send_flags = MLX5_IB_SEND_UMR_UNREG | MLX5_IB_SEND_UMR_FAIL_IF_FREE;
wr->opcode = MLX5_IB_WR_UMR;
- wr->wr.fast_reg.rkey = key;
+ umrwr->mkey = key;
}
void mlx5_umr_cq_handler(struct ib_cq *cq, void *cq_context)
@@ -742,7 +759,10 @@
struct ib_send_wr wr, *bad;
struct mlx5_ib_mr *mr;
struct ib_sge sg;
- int size = sizeof(u64) * npages;
+ int size;
+ __be64 *mr_pas;
+ __be64 *pas;
+ dma_addr_t dma;
int err = 0;
int i;
@@ -761,25 +781,31 @@
if (!mr)
return ERR_PTR(-EAGAIN);
- mr->pas = kmalloc(size + MLX5_UMR_ALIGN - 1, GFP_KERNEL);
- if (!mr->pas) {
+ /* UMR copies MTTs in units of MLX5_UMR_MTT_ALIGNMENT bytes.
+ * To avoid copying garbage after the pas array, we allocate
+ * a little more. */
+ size = ALIGN(sizeof(u64) * npages, MLX5_UMR_MTT_ALIGNMENT);
+ mr_pas = kmalloc(size + MLX5_UMR_ALIGN - 1, GFP_KERNEL);
+ if (!mr_pas) {
err = -ENOMEM;
goto free_mr;
}
- mlx5_ib_populate_pas(dev, umem, page_shift,
- mr_align(mr->pas, MLX5_UMR_ALIGN), 1);
+ pas = PTR_ALIGN(mr_pas, MLX5_UMR_ALIGN);
+ mlx5_ib_populate_pas(dev, umem, page_shift, pas, MLX5_IB_MTT_PRESENT);
+ /* Clear padding after the actual pages. */
+ memset(pas + npages, 0, size - npages * sizeof(u64));
- mr->dma = dma_map_single(ddev, mr_align(mr->pas, MLX5_UMR_ALIGN), size,
- DMA_TO_DEVICE);
- if (dma_mapping_error(ddev, mr->dma)) {
+ dma = dma_map_single(ddev, pas, size, DMA_TO_DEVICE);
+ if (dma_mapping_error(ddev, dma)) {
err = -ENOMEM;
goto free_pas;
}
memset(&wr, 0, sizeof(wr));
wr.wr_id = (u64)(unsigned long)&umr_context;
- prep_umr_reg_wqe(pd, &wr, &sg, mr->dma, npages, mr->mmr.key, page_shift, virt_addr, len, access_flags);
+ prep_umr_reg_wqe(pd, &wr, &sg, dma, npages, mr->mmr.key, page_shift,
+ virt_addr, len, access_flags);
mlx5_ib_init_umr_context(&umr_context);
down(&umrc->sem);
@@ -799,12 +825,14 @@
mr->mmr.size = len;
mr->mmr.pd = to_mpd(pd)->pdn;
+ mr->live = 1;
+
unmap_dma:
up(&umrc->sem);
- dma_unmap_single(ddev, mr->dma, size, DMA_TO_DEVICE);
+ dma_unmap_single(ddev, dma, size, DMA_TO_DEVICE);
free_pas:
- kfree(mr->pas);
+ kfree(mr_pas);
free_mr:
if (err) {
@@ -815,6 +843,128 @@
return mr;
}
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+int mlx5_ib_update_mtt(struct mlx5_ib_mr *mr, u64 start_page_index, int npages,
+ int zap)
+{
+ struct mlx5_ib_dev *dev = mr->dev;
+ struct device *ddev = dev->ib_dev.dma_device;
+ struct umr_common *umrc = &dev->umrc;
+ struct mlx5_ib_umr_context umr_context;
+ struct ib_umem *umem = mr->umem;
+ int size;
+ __be64 *pas;
+ dma_addr_t dma;
+ struct ib_send_wr wr, *bad;
+ struct mlx5_umr_wr *umrwr = (struct mlx5_umr_wr *)&wr.wr.fast_reg;
+ struct ib_sge sg;
+ int err = 0;
+ const int page_index_alignment = MLX5_UMR_MTT_ALIGNMENT / sizeof(u64);
+ const int page_index_mask = page_index_alignment - 1;
+ size_t pages_mapped = 0;
+ size_t pages_to_map = 0;
+ size_t pages_iter = 0;
+ int use_emergency_buf = 0;
+
+ /* UMR copies MTTs in units of MLX5_UMR_MTT_ALIGNMENT bytes,
+ * so we need to align the offset and length accordingly */
+ if (start_page_index & page_index_mask) {
+ npages += start_page_index & page_index_mask;
+ start_page_index &= ~page_index_mask;
+ }
+
+ pages_to_map = ALIGN(npages, page_index_alignment);
+
+ if (start_page_index + pages_to_map > MLX5_MAX_UMR_PAGES)
+ return -EINVAL;
+
+ size = sizeof(u64) * pages_to_map;
+ size = min_t(int, PAGE_SIZE, size);
+ /* We allocate with GFP_ATOMIC to avoid recursion into page-reclaim
+ * code, when we are called from an invalidation. The pas buffer must
+ * be 2k-aligned for Connect-IB. */
+ pas = (__be64 *)get_zeroed_page(GFP_ATOMIC);
+ if (!pas) {
+ mlx5_ib_warn(dev, "unable to allocate memory during MTT update, falling back to slower chunked mechanism.\n");
+ pas = mlx5_ib_update_mtt_emergency_buffer;
+ size = MLX5_UMR_MTT_MIN_CHUNK_SIZE;
+ use_emergency_buf = 1;
+ mutex_lock(&mlx5_ib_update_mtt_emergency_buffer_mutex);
+ memset(pas, 0, size);
+ }
+ pages_iter = size / sizeof(u64);
+ dma = dma_map_single(ddev, pas, size, DMA_TO_DEVICE);
+ if (dma_mapping_error(ddev, dma)) {
+ mlx5_ib_err(dev, "unable to map DMA during MTT update.\n");
+ err = -ENOMEM;
+ goto free_pas;
+ }
+
+ for (pages_mapped = 0;
+ pages_mapped < pages_to_map && !err;
+ pages_mapped += pages_iter, start_page_index += pages_iter) {
+ dma_sync_single_for_cpu(ddev, dma, size, DMA_TO_DEVICE);
+
+ npages = min_t(size_t,
+ pages_iter,
+ ib_umem_num_pages(umem) - start_page_index);
+
+ if (!zap) {
+ __mlx5_ib_populate_pas(dev, umem, PAGE_SHIFT,
+ start_page_index, npages, pas,
+ MLX5_IB_MTT_PRESENT);
+ /* Clear padding after the pages brought from the
+ * umem. */
+ memset(pas + npages, 0, size - npages * sizeof(u64));
+ }
+
+ dma_sync_single_for_device(ddev, dma, size, DMA_TO_DEVICE);
+
+ memset(&wr, 0, sizeof(wr));
+ wr.wr_id = (u64)(unsigned long)&umr_context;
+
+ sg.addr = dma;
+ sg.length = ALIGN(npages * sizeof(u64),
+ MLX5_UMR_MTT_ALIGNMENT);
+ sg.lkey = dev->umrc.mr->lkey;
+
+ wr.send_flags = MLX5_IB_SEND_UMR_FAIL_IF_FREE |
+ MLX5_IB_SEND_UMR_UPDATE_MTT;
+ wr.sg_list = &sg;
+ wr.num_sge = 1;
+ wr.opcode = MLX5_IB_WR_UMR;
+ umrwr->npages = sg.length / sizeof(u64);
+ umrwr->page_shift = PAGE_SHIFT;
+ umrwr->mkey = mr->mmr.key;
+ umrwr->target.offset = start_page_index;
+
+ mlx5_ib_init_umr_context(&umr_context);
+ down(&umrc->sem);
+ err = ib_post_send(umrc->qp, &wr, &bad);
+ if (err) {
+ mlx5_ib_err(dev, "UMR post send failed, err %d\n", err);
+ } else {
+ wait_for_completion(&umr_context.done);
+ if (umr_context.status != IB_WC_SUCCESS) {
+ mlx5_ib_err(dev, "UMR completion failed, code %d\n",
+ umr_context.status);
+ err = -EFAULT;
+ }
+ }
+ up(&umrc->sem);
+ }
+ dma_unmap_single(ddev, dma, size, DMA_TO_DEVICE);
+
+free_pas:
+ if (!use_emergency_buf)
+ free_page((unsigned long)pas);
+ else
+ mutex_unlock(&mlx5_ib_update_mtt_emergency_buffer_mutex);
+
+ return err;
+}
+#endif
+
static struct mlx5_ib_mr *reg_create(struct ib_pd *pd, u64 virt_addr,
u64 length, struct ib_umem *umem,
int npages, int page_shift,
@@ -825,6 +975,8 @@
struct mlx5_ib_mr *mr;
int inlen;
int err;
+ bool pg_cap = !!(dev->mdev->caps.gen.flags &
+ MLX5_DEV_CAP_FLAG_ON_DMND_PG);
mr = kzalloc(sizeof(*mr), GFP_KERNEL);
if (!mr)
@@ -836,8 +988,12 @@
err = -ENOMEM;
goto err_1;
}
- mlx5_ib_populate_pas(dev, umem, page_shift, in->pas, 0);
+ mlx5_ib_populate_pas(dev, umem, page_shift, in->pas,
+ pg_cap ? MLX5_IB_MTT_PRESENT : 0);
+ /* The MLX5_MKEY_INBOX_PG_ACCESS bit allows setting the access flags
+ * in the page list submitted with the command. */
+ in->flags = pg_cap ? cpu_to_be32(MLX5_MKEY_INBOX_PG_ACCESS) : 0;
in->seg.flags = convert_access(access_flags) |
MLX5_ACCESS_MODE_MTT;
in->seg.flags_pd = cpu_to_be32(to_mpd(pd)->pdn);
@@ -856,6 +1012,7 @@
goto err_2;
}
mr->umem = umem;
+ mr->live = 1;
kvfree(in);
mlx5_ib_dbg(dev, "mkey = 0x%x\n", mr->mmr.key);
@@ -910,6 +1067,10 @@
mlx5_ib_dbg(dev, "cache empty for order %d", order);
mr = NULL;
}
+ } else if (access_flags & IB_ACCESS_ON_DEMAND) {
+ err = -EINVAL;
+ pr_err("Got MR registration for ODP MR > 512MB, not supported for Connect-IB");
+ goto error;
}
if (!mr)
@@ -925,16 +1086,51 @@
mr->umem = umem;
mr->npages = npages;
- spin_lock(&dev->mr_lock);
- dev->mdev->priv.reg_pages += npages;
- spin_unlock(&dev->mr_lock);
+ atomic_add(npages, &dev->mdev->priv.reg_pages);
mr->ibmr.lkey = mr->mmr.key;
mr->ibmr.rkey = mr->mmr.key;
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ if (umem->odp_data) {
+ /*
+ * This barrier prevents the compiler from moving the
+ * setting of umem->odp_data->private to point to our
+ * MR, before reg_umr finished, to ensure that the MR
+ * initialization have finished before starting to
+ * handle invalidations.
+ */
+ smp_wmb();
+ mr->umem->odp_data->private = mr;
+ /*
+ * Make sure we will see the new
+ * umem->odp_data->private value in the invalidation
+ * routines, before we can get page faults on the
+ * MR. Page faults can happen once we put the MR in
+ * the tree, below this line. Without the barrier,
+ * there can be a fault handling and an invalidation
+ * before umem->odp_data->private == mr is visible to
+ * the invalidation handler.
+ */
+ smp_wmb();
+ }
+#endif
+
return &mr->ibmr;
error:
+ /*
+ * Destroy the umem *before* destroying the MR, to ensure we
+ * will not have any in-flight notifiers when destroying the
+ * MR.
+ *
+ * As the MR is completely invalid to begin with, and this
+ * error path is only taken if we can't push the mr entry into
+ * the pagefault tree, this is safe.
+ */
+
ib_umem_release(umem);
+ /* Kill the MR, and return an error code. */
+ clean_mr(mr);
return ERR_PTR(err);
}
@@ -971,17 +1167,14 @@
return err;
}
-int mlx5_ib_dereg_mr(struct ib_mr *ibmr)
+static int clean_mr(struct mlx5_ib_mr *mr)
{
- struct mlx5_ib_dev *dev = to_mdev(ibmr->device);
- struct mlx5_ib_mr *mr = to_mmr(ibmr);
- struct ib_umem *umem = mr->umem;
- int npages = mr->npages;
+ struct mlx5_ib_dev *dev = to_mdev(mr->ibmr.device);
int umred = mr->umred;
int err;
if (!umred) {
- err = mlx5_core_destroy_mkey(dev->mdev, &mr->mmr);
+ err = destroy_mkey(dev, mr);
if (err) {
mlx5_ib_warn(dev, "failed to destroy mkey 0x%x (%d)\n",
mr->mmr.key, err);
@@ -996,19 +1189,51 @@
free_cached_mr(dev, mr);
}
- if (umem) {
- ib_umem_release(umem);
- spin_lock(&dev->mr_lock);
- dev->mdev->priv.reg_pages -= npages;
- spin_unlock(&dev->mr_lock);
- }
-
if (!umred)
kfree(mr);
return 0;
}
+int mlx5_ib_dereg_mr(struct ib_mr *ibmr)
+{
+ struct mlx5_ib_dev *dev = to_mdev(ibmr->device);
+ struct mlx5_ib_mr *mr = to_mmr(ibmr);
+ int npages = mr->npages;
+ struct ib_umem *umem = mr->umem;
+
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ if (umem && umem->odp_data) {
+ /* Prevent new page faults from succeeding */
+ mr->live = 0;
+ /* Wait for all running page-fault handlers to finish. */
+ synchronize_srcu(&dev->mr_srcu);
+ /* Destroy all page mappings */
+ mlx5_ib_invalidate_range(umem, ib_umem_start(umem),
+ ib_umem_end(umem));
+ /*
+ * We kill the umem before the MR for ODP,
+ * so that there will not be any invalidations in
+ * flight, looking at the *mr struct.
+ */
+ ib_umem_release(umem);
+ atomic_sub(npages, &dev->mdev->priv.reg_pages);
+
+ /* Avoid double-freeing the umem. */
+ umem = NULL;
+ }
+#endif
+
+ clean_mr(mr);
+
+ if (umem) {
+ ib_umem_release(umem);
+ atomic_sub(npages, &dev->mdev->priv.reg_pages);
+ }
+
+ return 0;
+}
+
struct ib_mr *mlx5_ib_create_mr(struct ib_pd *pd,
struct ib_mr_init_attr *mr_init_attr)
{
@@ -1028,7 +1253,7 @@
goto err_free;
}
- in->seg.status = 1 << 6; /* free */
+ in->seg.status = MLX5_MKEY_STATUS_FREE;
in->seg.xlt_oct_size = cpu_to_be32(ndescs);
in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8);
in->seg.flags_pd = cpu_to_be32(to_mpd(pd)->pdn);
@@ -1113,7 +1338,7 @@
kfree(mr->sig);
}
- err = mlx5_core_destroy_mkey(dev->mdev, &mr->mmr);
+ err = destroy_mkey(dev, mr);
if (err) {
mlx5_ib_warn(dev, "failed to destroy mkey 0x%x (%d)\n",
mr->mmr.key, err);
@@ -1143,7 +1368,7 @@
goto err_free;
}
- in->seg.status = 1 << 6; /* free */
+ in->seg.status = MLX5_MKEY_STATUS_FREE;
in->seg.xlt_oct_size = cpu_to_be32((max_page_list_len + 1) / 2);
in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8);
in->seg.flags = MLX5_PERM_UMR_EN | MLX5_ACCESS_MODE_MTT;
diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c
new file mode 100644
index 0000000..a2c541c
--- /dev/null
+++ b/drivers/infiniband/hw/mlx5/odp.c
@@ -0,0 +1,798 @@
+/*
+ * Copyright (c) 2014 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * 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.
+ */
+
+#include <rdma/ib_umem.h>
+#include <rdma/ib_umem_odp.h>
+
+#include "mlx5_ib.h"
+
+#define MAX_PREFETCH_LEN (4*1024*1024U)
+
+/* Timeout in ms to wait for an active mmu notifier to complete when handling
+ * a pagefault. */
+#define MMU_NOTIFIER_TIMEOUT 1000
+
+struct workqueue_struct *mlx5_ib_page_fault_wq;
+
+void mlx5_ib_invalidate_range(struct ib_umem *umem, unsigned long start,
+ unsigned long end)
+{
+ struct mlx5_ib_mr *mr;
+ const u64 umr_block_mask = (MLX5_UMR_MTT_ALIGNMENT / sizeof(u64)) - 1;
+ u64 idx = 0, blk_start_idx = 0;
+ int in_block = 0;
+ u64 addr;
+
+ if (!umem || !umem->odp_data) {
+ pr_err("invalidation called on NULL umem or non-ODP umem\n");
+ return;
+ }
+
+ mr = umem->odp_data->private;
+
+ if (!mr || !mr->ibmr.pd)
+ return;
+
+ start = max_t(u64, ib_umem_start(umem), start);
+ end = min_t(u64, ib_umem_end(umem), end);
+
+ /*
+ * Iteration one - zap the HW's MTTs. The notifiers_count ensures that
+ * while we are doing the invalidation, no page fault will attempt to
+ * overwrite the same MTTs. Concurent invalidations might race us,
+ * but they will write 0s as well, so no difference in the end result.
+ */
+
+ for (addr = start; addr < end; addr += (u64)umem->page_size) {
+ idx = (addr - ib_umem_start(umem)) / PAGE_SIZE;
+ /*
+ * Strive to write the MTTs in chunks, but avoid overwriting
+ * non-existing MTTs. The huristic here can be improved to
+ * estimate the cost of another UMR vs. the cost of bigger
+ * UMR.
+ */
+ if (umem->odp_data->dma_list[idx] &
+ (ODP_READ_ALLOWED_BIT | ODP_WRITE_ALLOWED_BIT)) {
+ if (!in_block) {
+ blk_start_idx = idx;
+ in_block = 1;
+ }
+ } else {
+ u64 umr_offset = idx & umr_block_mask;
+
+ if (in_block && umr_offset == 0) {
+ mlx5_ib_update_mtt(mr, blk_start_idx,
+ idx - blk_start_idx, 1);
+ in_block = 0;
+ }
+ }
+ }
+ if (in_block)
+ mlx5_ib_update_mtt(mr, blk_start_idx, idx - blk_start_idx + 1,
+ 1);
+
+ /*
+ * We are now sure that the device will not access the
+ * memory. We can safely unmap it, and mark it as dirty if
+ * needed.
+ */
+
+ ib_umem_odp_unmap_dma_pages(umem, start, end);
+}
+
+#define COPY_ODP_BIT_MLX_TO_IB(reg, ib_caps, field_name, bit_name) do { \
+ if (be32_to_cpu(reg.field_name) & MLX5_ODP_SUPPORT_##bit_name) \
+ ib_caps->field_name |= IB_ODP_SUPPORT_##bit_name; \
+} while (0)
+
+int mlx5_ib_internal_query_odp_caps(struct mlx5_ib_dev *dev)
+{
+ int err;
+ struct mlx5_odp_caps hw_caps;
+ struct ib_odp_caps *caps = &dev->odp_caps;
+
+ memset(caps, 0, sizeof(*caps));
+
+ if (!(dev->mdev->caps.gen.flags & MLX5_DEV_CAP_FLAG_ON_DMND_PG))
+ return 0;
+
+ err = mlx5_query_odp_caps(dev->mdev, &hw_caps);
+ if (err)
+ goto out;
+
+ caps->general_caps = IB_ODP_SUPPORT;
+ COPY_ODP_BIT_MLX_TO_IB(hw_caps, caps, per_transport_caps.ud_odp_caps,
+ SEND);
+ COPY_ODP_BIT_MLX_TO_IB(hw_caps, caps, per_transport_caps.rc_odp_caps,
+ SEND);
+ COPY_ODP_BIT_MLX_TO_IB(hw_caps, caps, per_transport_caps.rc_odp_caps,
+ RECV);
+ COPY_ODP_BIT_MLX_TO_IB(hw_caps, caps, per_transport_caps.rc_odp_caps,
+ WRITE);
+ COPY_ODP_BIT_MLX_TO_IB(hw_caps, caps, per_transport_caps.rc_odp_caps,
+ READ);
+
+out:
+ return err;
+}
+
+static struct mlx5_ib_mr *mlx5_ib_odp_find_mr_lkey(struct mlx5_ib_dev *dev,
+ u32 key)
+{
+ u32 base_key = mlx5_base_mkey(key);
+ struct mlx5_core_mr *mmr = __mlx5_mr_lookup(dev->mdev, base_key);
+ struct mlx5_ib_mr *mr = container_of(mmr, struct mlx5_ib_mr, mmr);
+
+ if (!mmr || mmr->key != key || !mr->live)
+ return NULL;
+
+ return container_of(mmr, struct mlx5_ib_mr, mmr);
+}
+
+static void mlx5_ib_page_fault_resume(struct mlx5_ib_qp *qp,
+ struct mlx5_ib_pfault *pfault,
+ int error) {
+ struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.pd->device);
+ int ret = mlx5_core_page_fault_resume(dev->mdev, qp->mqp.qpn,
+ pfault->mpfault.flags,
+ error);
+ if (ret)
+ pr_err("Failed to resolve the page fault on QP 0x%x\n",
+ qp->mqp.qpn);
+}
+
+/*
+ * Handle a single data segment in a page-fault WQE.
+ *
+ * Returns number of pages retrieved on success. The caller will continue to
+ * the next data segment.
+ * Can return the following error codes:
+ * -EAGAIN to designate a temporary error. The caller will abort handling the
+ * page fault and resolve it.
+ * -EFAULT when there's an error mapping the requested pages. The caller will
+ * abort the page fault handling and possibly move the QP to an error state.
+ * On other errors the QP should also be closed with an error.
+ */
+static int pagefault_single_data_segment(struct mlx5_ib_qp *qp,
+ struct mlx5_ib_pfault *pfault,
+ u32 key, u64 io_virt, size_t bcnt,
+ u32 *bytes_mapped)
+{
+ struct mlx5_ib_dev *mib_dev = to_mdev(qp->ibqp.pd->device);
+ int srcu_key;
+ unsigned int current_seq;
+ u64 start_idx;
+ int npages = 0, ret = 0;
+ struct mlx5_ib_mr *mr;
+ u64 access_mask = ODP_READ_ALLOWED_BIT;
+
+ srcu_key = srcu_read_lock(&mib_dev->mr_srcu);
+ mr = mlx5_ib_odp_find_mr_lkey(mib_dev, key);
+ /*
+ * If we didn't find the MR, it means the MR was closed while we were
+ * handling the ODP event. In this case we return -EFAULT so that the
+ * QP will be closed.
+ */
+ if (!mr || !mr->ibmr.pd) {
+ pr_err("Failed to find relevant mr for lkey=0x%06x, probably the MR was destroyed\n",
+ key);
+ ret = -EFAULT;
+ goto srcu_unlock;
+ }
+ if (!mr->umem->odp_data) {
+ pr_debug("skipping non ODP MR (lkey=0x%06x) in page fault handler.\n",
+ key);
+ if (bytes_mapped)
+ *bytes_mapped +=
+ (bcnt - pfault->mpfault.bytes_committed);
+ goto srcu_unlock;
+ }
+ if (mr->ibmr.pd != qp->ibqp.pd) {
+ pr_err("Page-fault with different PDs for QP and MR.\n");
+ ret = -EFAULT;
+ goto srcu_unlock;
+ }
+
+ current_seq = ACCESS_ONCE(mr->umem->odp_data->notifiers_seq);
+ /*
+ * Ensure the sequence number is valid for some time before we call
+ * gup.
+ */
+ smp_rmb();
+
+ /*
+ * Avoid branches - this code will perform correctly
+ * in all iterations (in iteration 2 and above,
+ * bytes_committed == 0).
+ */
+ io_virt += pfault->mpfault.bytes_committed;
+ bcnt -= pfault->mpfault.bytes_committed;
+
+ start_idx = (io_virt - (mr->mmr.iova & PAGE_MASK)) >> PAGE_SHIFT;
+
+ if (mr->umem->writable)
+ access_mask |= ODP_WRITE_ALLOWED_BIT;
+ npages = ib_umem_odp_map_dma_pages(mr->umem, io_virt, bcnt,
+ access_mask, current_seq);
+ if (npages < 0) {
+ ret = npages;
+ goto srcu_unlock;
+ }
+
+ if (npages > 0) {
+ mutex_lock(&mr->umem->odp_data->umem_mutex);
+ if (!ib_umem_mmu_notifier_retry(mr->umem, current_seq)) {
+ /*
+ * No need to check whether the MTTs really belong to
+ * this MR, since ib_umem_odp_map_dma_pages already
+ * checks this.
+ */
+ ret = mlx5_ib_update_mtt(mr, start_idx, npages, 0);
+ } else {
+ ret = -EAGAIN;
+ }
+ mutex_unlock(&mr->umem->odp_data->umem_mutex);
+ if (ret < 0) {
+ if (ret != -EAGAIN)
+ pr_err("Failed to update mkey page tables\n");
+ goto srcu_unlock;
+ }
+
+ if (bytes_mapped) {
+ u32 new_mappings = npages * PAGE_SIZE -
+ (io_virt - round_down(io_virt, PAGE_SIZE));
+ *bytes_mapped += min_t(u32, new_mappings, bcnt);
+ }
+ }
+
+srcu_unlock:
+ if (ret == -EAGAIN) {
+ if (!mr->umem->odp_data->dying) {
+ struct ib_umem_odp *odp_data = mr->umem->odp_data;
+ unsigned long timeout =
+ msecs_to_jiffies(MMU_NOTIFIER_TIMEOUT);
+
+ if (!wait_for_completion_timeout(
+ &odp_data->notifier_completion,
+ timeout)) {
+ pr_warn("timeout waiting for mmu notifier completion\n");
+ }
+ } else {
+ /* The MR is being killed, kill the QP as well. */
+ ret = -EFAULT;
+ }
+ }
+ srcu_read_unlock(&mib_dev->mr_srcu, srcu_key);
+ pfault->mpfault.bytes_committed = 0;
+ return ret ? ret : npages;
+}
+
+/**
+ * Parse a series of data segments for page fault handling.
+ *
+ * @qp the QP on which the fault occurred.
+ * @pfault contains page fault information.
+ * @wqe points at the first data segment in the WQE.
+ * @wqe_end points after the end of the WQE.
+ * @bytes_mapped receives the number of bytes that the function was able to
+ * map. This allows the caller to decide intelligently whether
+ * enough memory was mapped to resolve the page fault
+ * successfully (e.g. enough for the next MTU, or the entire
+ * WQE).
+ * @total_wqe_bytes receives the total data size of this WQE in bytes (minus
+ * the committed bytes).
+ *
+ * Returns the number of pages loaded if positive, zero for an empty WQE, or a
+ * negative error code.
+ */
+static int pagefault_data_segments(struct mlx5_ib_qp *qp,
+ struct mlx5_ib_pfault *pfault, void *wqe,
+ void *wqe_end, u32 *bytes_mapped,
+ u32 *total_wqe_bytes, int receive_queue)
+{
+ int ret = 0, npages = 0;
+ u64 io_virt;
+ u32 key;
+ u32 byte_count;
+ size_t bcnt;
+ int inline_segment;
+
+ /* Skip SRQ next-WQE segment. */
+ if (receive_queue && qp->ibqp.srq)
+ wqe += sizeof(struct mlx5_wqe_srq_next_seg);
+
+ if (bytes_mapped)
+ *bytes_mapped = 0;
+ if (total_wqe_bytes)
+ *total_wqe_bytes = 0;
+
+ while (wqe < wqe_end) {
+ struct mlx5_wqe_data_seg *dseg = wqe;
+
+ io_virt = be64_to_cpu(dseg->addr);
+ key = be32_to_cpu(dseg->lkey);
+ byte_count = be32_to_cpu(dseg->byte_count);
+ inline_segment = !!(byte_count & MLX5_INLINE_SEG);
+ bcnt = byte_count & ~MLX5_INLINE_SEG;
+
+ if (inline_segment) {
+ bcnt = bcnt & MLX5_WQE_INLINE_SEG_BYTE_COUNT_MASK;
+ wqe += ALIGN(sizeof(struct mlx5_wqe_inline_seg) + bcnt,
+ 16);
+ } else {
+ wqe += sizeof(*dseg);
+ }
+
+ /* receive WQE end of sg list. */
+ if (receive_queue && bcnt == 0 && key == MLX5_INVALID_LKEY &&
+ io_virt == 0)
+ break;
+
+ if (!inline_segment && total_wqe_bytes) {
+ *total_wqe_bytes += bcnt - min_t(size_t, bcnt,
+ pfault->mpfault.bytes_committed);
+ }
+
+ /* A zero length data segment designates a length of 2GB. */
+ if (bcnt == 0)
+ bcnt = 1U << 31;
+
+ if (inline_segment || bcnt <= pfault->mpfault.bytes_committed) {
+ pfault->mpfault.bytes_committed -=
+ min_t(size_t, bcnt,
+ pfault->mpfault.bytes_committed);
+ continue;
+ }
+
+ ret = pagefault_single_data_segment(qp, pfault, key, io_virt,
+ bcnt, bytes_mapped);
+ if (ret < 0)
+ break;
+ npages += ret;
+ }
+
+ return ret < 0 ? ret : npages;
+}
+
+/*
+ * Parse initiator WQE. Advances the wqe pointer to point at the
+ * scatter-gather list, and set wqe_end to the end of the WQE.
+ */
+static int mlx5_ib_mr_initiator_pfault_handler(
+ struct mlx5_ib_qp *qp, struct mlx5_ib_pfault *pfault,
+ void **wqe, void **wqe_end, int wqe_length)
+{
+ struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.pd->device);
+ struct mlx5_wqe_ctrl_seg *ctrl = *wqe;
+ u16 wqe_index = pfault->mpfault.wqe.wqe_index;
+ unsigned ds, opcode;
+#if defined(DEBUG)
+ u32 ctrl_wqe_index, ctrl_qpn;
+#endif
+
+ ds = be32_to_cpu(ctrl->qpn_ds) & MLX5_WQE_CTRL_DS_MASK;
+ if (ds * MLX5_WQE_DS_UNITS > wqe_length) {
+ mlx5_ib_err(dev, "Unable to read the complete WQE. ds = 0x%x, ret = 0x%x\n",
+ ds, wqe_length);
+ return -EFAULT;
+ }
+
+ if (ds == 0) {
+ mlx5_ib_err(dev, "Got WQE with zero DS. wqe_index=%x, qpn=%x\n",
+ wqe_index, qp->mqp.qpn);
+ return -EFAULT;
+ }
+
+#if defined(DEBUG)
+ ctrl_wqe_index = (be32_to_cpu(ctrl->opmod_idx_opcode) &
+ MLX5_WQE_CTRL_WQE_INDEX_MASK) >>
+ MLX5_WQE_CTRL_WQE_INDEX_SHIFT;
+ if (wqe_index != ctrl_wqe_index) {
+ mlx5_ib_err(dev, "Got WQE with invalid wqe_index. wqe_index=0x%x, qpn=0x%x ctrl->wqe_index=0x%x\n",
+ wqe_index, qp->mqp.qpn,
+ ctrl_wqe_index);
+ return -EFAULT;
+ }
+
+ ctrl_qpn = (be32_to_cpu(ctrl->qpn_ds) & MLX5_WQE_CTRL_QPN_MASK) >>
+ MLX5_WQE_CTRL_QPN_SHIFT;
+ if (qp->mqp.qpn != ctrl_qpn) {
+ mlx5_ib_err(dev, "Got WQE with incorrect QP number. wqe_index=0x%x, qpn=0x%x ctrl->qpn=0x%x\n",
+ wqe_index, qp->mqp.qpn,
+ ctrl_qpn);
+ return -EFAULT;
+ }
+#endif /* DEBUG */
+
+ *wqe_end = *wqe + ds * MLX5_WQE_DS_UNITS;
+ *wqe += sizeof(*ctrl);
+
+ opcode = be32_to_cpu(ctrl->opmod_idx_opcode) &
+ MLX5_WQE_CTRL_OPCODE_MASK;
+ switch (qp->ibqp.qp_type) {
+ case IB_QPT_RC:
+ switch (opcode) {
+ case MLX5_OPCODE_SEND:
+ case MLX5_OPCODE_SEND_IMM:
+ case MLX5_OPCODE_SEND_INVAL:
+ if (!(dev->odp_caps.per_transport_caps.rc_odp_caps &
+ IB_ODP_SUPPORT_SEND))
+ goto invalid_transport_or_opcode;
+ break;
+ case MLX5_OPCODE_RDMA_WRITE:
+ case MLX5_OPCODE_RDMA_WRITE_IMM:
+ if (!(dev->odp_caps.per_transport_caps.rc_odp_caps &
+ IB_ODP_SUPPORT_WRITE))
+ goto invalid_transport_or_opcode;
+ *wqe += sizeof(struct mlx5_wqe_raddr_seg);
+ break;
+ case MLX5_OPCODE_RDMA_READ:
+ if (!(dev->odp_caps.per_transport_caps.rc_odp_caps &
+ IB_ODP_SUPPORT_READ))
+ goto invalid_transport_or_opcode;
+ *wqe += sizeof(struct mlx5_wqe_raddr_seg);
+ break;
+ default:
+ goto invalid_transport_or_opcode;
+ }
+ break;
+ case IB_QPT_UD:
+ switch (opcode) {
+ case MLX5_OPCODE_SEND:
+ case MLX5_OPCODE_SEND_IMM:
+ if (!(dev->odp_caps.per_transport_caps.ud_odp_caps &
+ IB_ODP_SUPPORT_SEND))
+ goto invalid_transport_or_opcode;
+ *wqe += sizeof(struct mlx5_wqe_datagram_seg);
+ break;
+ default:
+ goto invalid_transport_or_opcode;
+ }
+ break;
+ default:
+invalid_transport_or_opcode:
+ mlx5_ib_err(dev, "ODP fault on QP of an unsupported opcode or transport. transport: 0x%x opcode: 0x%x.\n",
+ qp->ibqp.qp_type, opcode);
+ return -EFAULT;
+ }
+
+ return 0;
+}
+
+/*
+ * Parse responder WQE. Advances the wqe pointer to point at the
+ * scatter-gather list, and set wqe_end to the end of the WQE.
+ */
+static int mlx5_ib_mr_responder_pfault_handler(
+ struct mlx5_ib_qp *qp, struct mlx5_ib_pfault *pfault,
+ void **wqe, void **wqe_end, int wqe_length)
+{
+ struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.pd->device);
+ struct mlx5_ib_wq *wq = &qp->rq;
+ int wqe_size = 1 << wq->wqe_shift;
+
+ if (qp->ibqp.srq) {
+ mlx5_ib_err(dev, "ODP fault on SRQ is not supported\n");
+ return -EFAULT;
+ }
+
+ if (qp->wq_sig) {
+ mlx5_ib_err(dev, "ODP fault with WQE signatures is not supported\n");
+ return -EFAULT;
+ }
+
+ if (wqe_size > wqe_length) {
+ mlx5_ib_err(dev, "Couldn't read all of the receive WQE's content\n");
+ return -EFAULT;
+ }
+
+ switch (qp->ibqp.qp_type) {
+ case IB_QPT_RC:
+ if (!(dev->odp_caps.per_transport_caps.rc_odp_caps &
+ IB_ODP_SUPPORT_RECV))
+ goto invalid_transport_or_opcode;
+ break;
+ default:
+invalid_transport_or_opcode:
+ mlx5_ib_err(dev, "ODP fault on QP of an unsupported transport. transport: 0x%x\n",
+ qp->ibqp.qp_type);
+ return -EFAULT;
+ }
+
+ *wqe_end = *wqe + wqe_size;
+
+ return 0;
+}
+
+static void mlx5_ib_mr_wqe_pfault_handler(struct mlx5_ib_qp *qp,
+ struct mlx5_ib_pfault *pfault)
+{
+ struct mlx5_ib_dev *dev = to_mdev(qp->ibqp.pd->device);
+ int ret;
+ void *wqe, *wqe_end;
+ u32 bytes_mapped, total_wqe_bytes;
+ char *buffer = NULL;
+ int resume_with_error = 0;
+ u16 wqe_index = pfault->mpfault.wqe.wqe_index;
+ int requestor = pfault->mpfault.flags & MLX5_PFAULT_REQUESTOR;
+
+ buffer = (char *)__get_free_page(GFP_KERNEL);
+ if (!buffer) {
+ mlx5_ib_err(dev, "Error allocating memory for IO page fault handling.\n");
+ resume_with_error = 1;
+ goto resolve_page_fault;
+ }
+
+ ret = mlx5_ib_read_user_wqe(qp, requestor, wqe_index, buffer,
+ PAGE_SIZE);
+ if (ret < 0) {
+ mlx5_ib_err(dev, "Failed reading a WQE following page fault, error=%x, wqe_index=%x, qpn=%x\n",
+ -ret, wqe_index, qp->mqp.qpn);
+ resume_with_error = 1;
+ goto resolve_page_fault;
+ }
+
+ wqe = buffer;
+ if (requestor)
+ ret = mlx5_ib_mr_initiator_pfault_handler(qp, pfault, &wqe,
+ &wqe_end, ret);
+ else
+ ret = mlx5_ib_mr_responder_pfault_handler(qp, pfault, &wqe,
+ &wqe_end, ret);
+ if (ret < 0) {
+ resume_with_error = 1;
+ goto resolve_page_fault;
+ }
+
+ if (wqe >= wqe_end) {
+ mlx5_ib_err(dev, "ODP fault on invalid WQE.\n");
+ resume_with_error = 1;
+ goto resolve_page_fault;
+ }
+
+ ret = pagefault_data_segments(qp, pfault, wqe, wqe_end, &bytes_mapped,
+ &total_wqe_bytes, !requestor);
+ if (ret == -EAGAIN) {
+ goto resolve_page_fault;
+ } else if (ret < 0 || total_wqe_bytes > bytes_mapped) {
+ mlx5_ib_err(dev, "Error getting user pages for page fault. Error: 0x%x\n",
+ -ret);
+ resume_with_error = 1;
+ goto resolve_page_fault;
+ }
+
+resolve_page_fault:
+ mlx5_ib_page_fault_resume(qp, pfault, resume_with_error);
+ mlx5_ib_dbg(dev, "PAGE FAULT completed. QP 0x%x resume_with_error=%d, flags: 0x%x\n",
+ qp->mqp.qpn, resume_with_error, pfault->mpfault.flags);
+
+ free_page((unsigned long)buffer);
+}
+
+static int pages_in_range(u64 address, u32 length)
+{
+ return (ALIGN(address + length, PAGE_SIZE) -
+ (address & PAGE_MASK)) >> PAGE_SHIFT;
+}
+
+static void mlx5_ib_mr_rdma_pfault_handler(struct mlx5_ib_qp *qp,
+ struct mlx5_ib_pfault *pfault)
+{
+ struct mlx5_pagefault *mpfault = &pfault->mpfault;
+ u64 address;
+ u32 length;
+ u32 prefetch_len = mpfault->bytes_committed;
+ int prefetch_activated = 0;
+ u32 rkey = mpfault->rdma.r_key;
+ int ret;
+
+ /* The RDMA responder handler handles the page fault in two parts.
+ * First it brings the necessary pages for the current packet
+ * (and uses the pfault context), and then (after resuming the QP)
+ * prefetches more pages. The second operation cannot use the pfault
+ * context and therefore uses the dummy_pfault context allocated on
+ * the stack */
+ struct mlx5_ib_pfault dummy_pfault = {};
+
+ dummy_pfault.mpfault.bytes_committed = 0;
+
+ mpfault->rdma.rdma_va += mpfault->bytes_committed;
+ mpfault->rdma.rdma_op_len -= min(mpfault->bytes_committed,
+ mpfault->rdma.rdma_op_len);
+ mpfault->bytes_committed = 0;
+
+ address = mpfault->rdma.rdma_va;
+ length = mpfault->rdma.rdma_op_len;
+
+ /* For some operations, the hardware cannot tell the exact message
+ * length, and in those cases it reports zero. Use prefetch
+ * logic. */
+ if (length == 0) {
+ prefetch_activated = 1;
+ length = mpfault->rdma.packet_size;
+ prefetch_len = min(MAX_PREFETCH_LEN, prefetch_len);
+ }
+
+ ret = pagefault_single_data_segment(qp, pfault, rkey, address, length,
+ NULL);
+ if (ret == -EAGAIN) {
+ /* We're racing with an invalidation, don't prefetch */
+ prefetch_activated = 0;
+ } else if (ret < 0 || pages_in_range(address, length) > ret) {
+ mlx5_ib_page_fault_resume(qp, pfault, 1);
+ return;
+ }
+
+ mlx5_ib_page_fault_resume(qp, pfault, 0);
+
+ /* At this point, there might be a new pagefault already arriving in
+ * the eq, switch to the dummy pagefault for the rest of the
+ * processing. We're still OK with the objects being alive as the
+ * work-queue is being fenced. */
+
+ if (prefetch_activated) {
+ ret = pagefault_single_data_segment(qp, &dummy_pfault, rkey,
+ address,
+ prefetch_len,
+ NULL);
+ if (ret < 0) {
+ pr_warn("Prefetch failed (ret = %d, prefetch_activated = %d) for QPN %d, address: 0x%.16llx, length = 0x%.16x\n",
+ ret, prefetch_activated,
+ qp->ibqp.qp_num, address, prefetch_len);
+ }
+ }
+}
+
+void mlx5_ib_mr_pfault_handler(struct mlx5_ib_qp *qp,
+ struct mlx5_ib_pfault *pfault)
+{
+ u8 event_subtype = pfault->mpfault.event_subtype;
+
+ switch (event_subtype) {
+ case MLX5_PFAULT_SUBTYPE_WQE:
+ mlx5_ib_mr_wqe_pfault_handler(qp, pfault);
+ break;
+ case MLX5_PFAULT_SUBTYPE_RDMA:
+ mlx5_ib_mr_rdma_pfault_handler(qp, pfault);
+ break;
+ default:
+ pr_warn("Invalid page fault event subtype: 0x%x\n",
+ event_subtype);
+ mlx5_ib_page_fault_resume(qp, pfault, 1);
+ break;
+ }
+}
+
+static void mlx5_ib_qp_pfault_action(struct work_struct *work)
+{
+ struct mlx5_ib_pfault *pfault = container_of(work,
+ struct mlx5_ib_pfault,
+ work);
+ enum mlx5_ib_pagefault_context context =
+ mlx5_ib_get_pagefault_context(&pfault->mpfault);
+ struct mlx5_ib_qp *qp = container_of(pfault, struct mlx5_ib_qp,
+ pagefaults[context]);
+ mlx5_ib_mr_pfault_handler(qp, pfault);
+}
+
+void mlx5_ib_qp_disable_pagefaults(struct mlx5_ib_qp *qp)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&qp->disable_page_faults_lock, flags);
+ qp->disable_page_faults = 1;
+ spin_unlock_irqrestore(&qp->disable_page_faults_lock, flags);
+
+ /*
+ * Note that at this point, we are guarenteed that no more
+ * work queue elements will be posted to the work queue with
+ * the QP we are closing.
+ */
+ flush_workqueue(mlx5_ib_page_fault_wq);
+}
+
+void mlx5_ib_qp_enable_pagefaults(struct mlx5_ib_qp *qp)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&qp->disable_page_faults_lock, flags);
+ qp->disable_page_faults = 0;
+ spin_unlock_irqrestore(&qp->disable_page_faults_lock, flags);
+}
+
+static void mlx5_ib_pfault_handler(struct mlx5_core_qp *qp,
+ struct mlx5_pagefault *pfault)
+{
+ /*
+ * Note that we will only get one fault event per QP per context
+ * (responder/initiator, read/write), until we resolve the page fault
+ * with the mlx5_ib_page_fault_resume command. Since this function is
+ * called from within the work element, there is no risk of missing
+ * events.
+ */
+ struct mlx5_ib_qp *mibqp = to_mibqp(qp);
+ enum mlx5_ib_pagefault_context context =
+ mlx5_ib_get_pagefault_context(pfault);
+ struct mlx5_ib_pfault *qp_pfault = &mibqp->pagefaults[context];
+
+ qp_pfault->mpfault = *pfault;
+
+ /* No need to stop interrupts here since we are in an interrupt */
+ spin_lock(&mibqp->disable_page_faults_lock);
+ if (!mibqp->disable_page_faults)
+ queue_work(mlx5_ib_page_fault_wq, &qp_pfault->work);
+ spin_unlock(&mibqp->disable_page_faults_lock);
+}
+
+void mlx5_ib_odp_create_qp(struct mlx5_ib_qp *qp)
+{
+ int i;
+
+ qp->disable_page_faults = 1;
+ spin_lock_init(&qp->disable_page_faults_lock);
+
+ qp->mqp.pfault_handler = mlx5_ib_pfault_handler;
+
+ for (i = 0; i < MLX5_IB_PAGEFAULT_CONTEXTS; ++i)
+ INIT_WORK(&qp->pagefaults[i].work, mlx5_ib_qp_pfault_action);
+}
+
+int mlx5_ib_odp_init_one(struct mlx5_ib_dev *ibdev)
+{
+ int ret;
+
+ ret = init_srcu_struct(&ibdev->mr_srcu);
+ if (ret)
+ return ret;
+
+ return 0;
+}
+
+void mlx5_ib_odp_remove_one(struct mlx5_ib_dev *ibdev)
+{
+ cleanup_srcu_struct(&ibdev->mr_srcu);
+}
+
+int __init mlx5_ib_odp_init(void)
+{
+ mlx5_ib_page_fault_wq =
+ create_singlethread_workqueue("mlx5_ib_page_faults");
+ if (!mlx5_ib_page_fault_wq)
+ return -ENOMEM;
+
+ return 0;
+}
+
+void mlx5_ib_odp_cleanup(void)
+{
+ destroy_workqueue(mlx5_ib_page_fault_wq);
+}
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 1cae1c7..be0cd35 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -70,15 +70,6 @@
[MLX5_IB_WR_UMR] = MLX5_OPCODE_UMR,
};
-struct umr_wr {
- u64 virt_addr;
- struct ib_pd *pd;
- unsigned int page_shift;
- unsigned int npages;
- u32 length;
- int access_flags;
- u32 mkey;
-};
static int is_qp0(enum ib_qp_type qp_type)
{
@@ -110,6 +101,77 @@
return get_wqe(qp, qp->sq.offset + (n << MLX5_IB_SQ_STRIDE));
}
+/**
+ * mlx5_ib_read_user_wqe() - Copy a user-space WQE to kernel space.
+ *
+ * @qp: QP to copy from.
+ * @send: copy from the send queue when non-zero, use the receive queue
+ * otherwise.
+ * @wqe_index: index to start copying from. For send work queues, the
+ * wqe_index is in units of MLX5_SEND_WQE_BB.
+ * For receive work queue, it is the number of work queue
+ * element in the queue.
+ * @buffer: destination buffer.
+ * @length: maximum number of bytes to copy.
+ *
+ * Copies at least a single WQE, but may copy more data.
+ *
+ * Return: the number of bytes copied, or an error code.
+ */
+int mlx5_ib_read_user_wqe(struct mlx5_ib_qp *qp, int send, int wqe_index,
+ void *buffer, u32 length)
+{
+ struct ib_device *ibdev = qp->ibqp.device;
+ struct mlx5_ib_dev *dev = to_mdev(ibdev);
+ struct mlx5_ib_wq *wq = send ? &qp->sq : &qp->rq;
+ size_t offset;
+ size_t wq_end;
+ struct ib_umem *umem = qp->umem;
+ u32 first_copy_length;
+ int wqe_length;
+ int ret;
+
+ if (wq->wqe_cnt == 0) {
+ mlx5_ib_dbg(dev, "mlx5_ib_read_user_wqe for a QP with wqe_cnt == 0. qp_type: 0x%x\n",
+ qp->ibqp.qp_type);
+ return -EINVAL;
+ }
+
+ offset = wq->offset + ((wqe_index % wq->wqe_cnt) << wq->wqe_shift);
+ wq_end = wq->offset + (wq->wqe_cnt << wq->wqe_shift);
+
+ if (send && length < sizeof(struct mlx5_wqe_ctrl_seg))
+ return -EINVAL;
+
+ if (offset > umem->length ||
+ (send && offset + sizeof(struct mlx5_wqe_ctrl_seg) > umem->length))
+ return -EINVAL;
+
+ first_copy_length = min_t(u32, offset + length, wq_end) - offset;
+ ret = ib_umem_copy_from(buffer, umem, offset, first_copy_length);
+ if (ret)
+ return ret;
+
+ if (send) {
+ struct mlx5_wqe_ctrl_seg *ctrl = buffer;
+ int ds = be32_to_cpu(ctrl->qpn_ds) & MLX5_WQE_CTRL_DS_MASK;
+
+ wqe_length = ds * MLX5_WQE_DS_UNITS;
+ } else {
+ wqe_length = 1 << wq->wqe_shift;
+ }
+
+ if (wqe_length <= first_copy_length)
+ return first_copy_length;
+
+ ret = ib_umem_copy_from(buffer + first_copy_length, umem, wq->offset,
+ wqe_length - first_copy_length);
+ if (ret)
+ return ret;
+
+ return wqe_length;
+}
+
static void mlx5_ib_qp_event(struct mlx5_core_qp *qp, int type)
{
struct ib_qp *ibqp = &to_mibqp(qp)->ibqp;
@@ -814,6 +876,8 @@
int inlen = sizeof(*in);
int err;
+ mlx5_ib_odp_create_qp(qp);
+
gen = &dev->mdev->caps.gen;
mutex_init(&qp->mutex);
spin_lock_init(&qp->sq.lock);
@@ -1098,11 +1162,13 @@
in = kzalloc(sizeof(*in), GFP_KERNEL);
if (!in)
return;
- if (qp->state != IB_QPS_RESET)
+ if (qp->state != IB_QPS_RESET) {
+ mlx5_ib_qp_disable_pagefaults(qp);
if (mlx5_core_qp_modify(dev->mdev, to_mlx5_state(qp->state),
MLX5_QP_STATE_RST, in, sizeof(*in), &qp->mqp))
mlx5_ib_warn(dev, "mlx5_ib: modify QP %06x to RESET failed\n",
qp->mqp.qpn);
+ }
get_cqs(qp, &send_cq, &recv_cq);
@@ -1650,6 +1716,15 @@
if (mlx5_st < 0)
goto out;
+ /* If moving to a reset or error state, we must disable page faults on
+ * this QP and flush all current page faults. Otherwise a stale page
+ * fault may attempt to work on this QP after it is reset and moved
+ * again to RTS, and may cause the driver and the device to get out of
+ * sync. */
+ if (cur_state != IB_QPS_RESET && cur_state != IB_QPS_ERR &&
+ (new_state == IB_QPS_RESET || new_state == IB_QPS_ERR))
+ mlx5_ib_qp_disable_pagefaults(qp);
+
optpar = ib_mask_to_mlx5_opt(attr_mask);
optpar &= opt_mask[mlx5_cur][mlx5_new][mlx5_st];
in->optparam = cpu_to_be32(optpar);
@@ -1659,6 +1734,9 @@
if (err)
goto out;
+ if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT)
+ mlx5_ib_qp_enable_pagefaults(qp);
+
qp->state = new_state;
if (attr_mask & IB_QP_ACCESS_FLAGS)
@@ -1848,37 +1926,70 @@
umr->mkey_mask = frwr_mkey_mask();
}
+static __be64 get_umr_reg_mr_mask(void)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_LEN |
+ MLX5_MKEY_MASK_PAGE_SIZE |
+ MLX5_MKEY_MASK_START_ADDR |
+ MLX5_MKEY_MASK_PD |
+ MLX5_MKEY_MASK_LR |
+ MLX5_MKEY_MASK_LW |
+ MLX5_MKEY_MASK_KEY |
+ MLX5_MKEY_MASK_RR |
+ MLX5_MKEY_MASK_RW |
+ MLX5_MKEY_MASK_A |
+ MLX5_MKEY_MASK_FREE;
+
+ return cpu_to_be64(result);
+}
+
+static __be64 get_umr_unreg_mr_mask(void)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_FREE;
+
+ return cpu_to_be64(result);
+}
+
+static __be64 get_umr_update_mtt_mask(void)
+{
+ u64 result;
+
+ result = MLX5_MKEY_MASK_FREE;
+
+ return cpu_to_be64(result);
+}
+
static void set_reg_umr_segment(struct mlx5_wqe_umr_ctrl_seg *umr,
struct ib_send_wr *wr)
{
- struct umr_wr *umrwr = (struct umr_wr *)&wr->wr.fast_reg;
- u64 mask;
+ struct mlx5_umr_wr *umrwr = (struct mlx5_umr_wr *)&wr->wr.fast_reg;
memset(umr, 0, sizeof(*umr));
+ if (wr->send_flags & MLX5_IB_SEND_UMR_FAIL_IF_FREE)
+ umr->flags = MLX5_UMR_CHECK_FREE; /* fail if free */
+ else
+ umr->flags = MLX5_UMR_CHECK_NOT_FREE; /* fail if not free */
+
if (!(wr->send_flags & MLX5_IB_SEND_UMR_UNREG)) {
- umr->flags = 1 << 5; /* fail if not free */
umr->klm_octowords = get_klm_octo(umrwr->npages);
- mask = MLX5_MKEY_MASK_LEN |
- MLX5_MKEY_MASK_PAGE_SIZE |
- MLX5_MKEY_MASK_START_ADDR |
- MLX5_MKEY_MASK_PD |
- MLX5_MKEY_MASK_LR |
- MLX5_MKEY_MASK_LW |
- MLX5_MKEY_MASK_KEY |
- MLX5_MKEY_MASK_RR |
- MLX5_MKEY_MASK_RW |
- MLX5_MKEY_MASK_A |
- MLX5_MKEY_MASK_FREE;
- umr->mkey_mask = cpu_to_be64(mask);
+ if (wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_MTT) {
+ umr->mkey_mask = get_umr_update_mtt_mask();
+ umr->bsf_octowords = get_klm_octo(umrwr->target.offset);
+ umr->flags |= MLX5_UMR_TRANSLATION_OFFSET_EN;
+ } else {
+ umr->mkey_mask = get_umr_reg_mr_mask();
+ }
} else {
- umr->flags = 2 << 5; /* fail if free */
- mask = MLX5_MKEY_MASK_FREE;
- umr->mkey_mask = cpu_to_be64(mask);
+ umr->mkey_mask = get_umr_unreg_mr_mask();
}
if (!wr->num_sge)
- umr->flags |= (1 << 7); /* inline */
+ umr->flags |= MLX5_UMR_INLINE;
}
static u8 get_umr_flags(int acc)
@@ -1895,7 +2006,7 @@
{
memset(seg, 0, sizeof(*seg));
if (li) {
- seg->status = 1 << 6;
+ seg->status = MLX5_MKEY_STATUS_FREE;
return;
}
@@ -1912,19 +2023,23 @@
static void set_reg_mkey_segment(struct mlx5_mkey_seg *seg, struct ib_send_wr *wr)
{
+ struct mlx5_umr_wr *umrwr = (struct mlx5_umr_wr *)&wr->wr.fast_reg;
+
memset(seg, 0, sizeof(*seg));
if (wr->send_flags & MLX5_IB_SEND_UMR_UNREG) {
- seg->status = 1 << 6;
+ seg->status = MLX5_MKEY_STATUS_FREE;
return;
}
- seg->flags = convert_access(wr->wr.fast_reg.access_flags);
- seg->flags_pd = cpu_to_be32(to_mpd((struct ib_pd *)wr->wr.fast_reg.page_list)->pdn);
- seg->start_addr = cpu_to_be64(wr->wr.fast_reg.iova_start);
- seg->len = cpu_to_be64(wr->wr.fast_reg.length);
- seg->log2_page_size = wr->wr.fast_reg.page_shift;
+ seg->flags = convert_access(umrwr->access_flags);
+ if (!(wr->send_flags & MLX5_IB_SEND_UMR_UPDATE_MTT)) {
+ seg->flags_pd = cpu_to_be32(to_mpd(umrwr->pd)->pdn);
+ seg->start_addr = cpu_to_be64(umrwr->target.virt_addr);
+ }
+ seg->len = cpu_to_be64(umrwr->length);
+ seg->log2_page_size = umrwr->page_shift;
seg->qpn_mkey7_0 = cpu_to_be32(0xffffff00 |
- mlx5_mkey_variant(wr->wr.fast_reg.rkey));
+ mlx5_mkey_variant(umrwr->mkey));
}
static void set_frwr_pages(struct mlx5_wqe_data_seg *dseg,
@@ -2927,6 +3042,14 @@
int mlx5_state;
int err = 0;
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ /*
+ * Wait for any outstanding page faults, in case the user frees memory
+ * based upon this query's result.
+ */
+ flush_workqueue(mlx5_ib_page_fault_wq);
+#endif
+
mutex_lock(&qp->mutex);
outb = kzalloc(sizeof(*outb), GFP_KERNEL);
if (!outb) {
diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c
index fef067c..c0d0296 100644
--- a/drivers/infiniband/hw/nes/nes_verbs.c
+++ b/drivers/infiniband/hw/nes/nes_verbs.c
@@ -2341,9 +2341,9 @@
nes_debug(NES_DBG_MR, "User base = 0x%lX, Virt base = 0x%lX, length = %u,"
" offset = %u, page size = %u.\n",
(unsigned long int)start, (unsigned long int)virt, (u32)length,
- region->offset, region->page_size);
+ ib_umem_offset(region), region->page_size);
- skip_pages = ((u32)region->offset) >> 12;
+ skip_pages = ((u32)ib_umem_offset(region)) >> 12;
if (ib_copy_from_udata(&req, udata, sizeof(req))) {
ib_umem_release(region);
@@ -2408,7 +2408,7 @@
region_length -= skip_pages << 12;
for (page_index = skip_pages; page_index < chunk_pages; page_index++) {
skip_pages = 0;
- if ((page_count != 0) && (page_count<<12)-(region->offset&(4096-1)) >= region->length)
+ if ((page_count != 0) && (page_count << 12) - (ib_umem_offset(region) & (4096 - 1)) >= region->length)
goto enough_pages;
if ((page_count&0x01FF) == 0) {
if (page_count >= 1024 * 512) {
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c
index ac02ce4..f3cc8c9 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_ah.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_ah.c
@@ -96,7 +96,6 @@
struct ocrdma_pd *pd = get_ocrdma_pd(ibpd);
struct ocrdma_dev *dev = get_ocrdma_dev(ibpd->device);
union ib_gid sgid;
- u8 zmac[ETH_ALEN];
if (!(attr->ah_flags & IB_AH_GRH))
return ERR_PTR(-EINVAL);
@@ -118,9 +117,7 @@
goto av_conf_err;
}
- memset(&zmac, 0, ETH_ALEN);
- if (pd->uctx &&
- memcmp(attr->dmac, &zmac, ETH_ALEN)) {
+ if (pd->uctx) {
status = rdma_addr_find_dmac_by_grh(&sgid, &attr->grh.dgid,
attr->dmac, &attr->vlan_id);
if (status) {
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
index 4c68305..fb8d8c4 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
@@ -805,7 +805,7 @@
goto umem_err;
mr->hwmr.pbe_size = mr->umem->page_size;
- mr->hwmr.fbo = mr->umem->offset;
+ mr->hwmr.fbo = ib_umem_offset(mr->umem);
mr->hwmr.va = usr_addr;
mr->hwmr.len = len;
mr->hwmr.remote_wr = (acc & IB_ACCESS_REMOTE_WRITE) ? 1 : 0;
@@ -1410,6 +1410,8 @@
mutex_unlock(&dev->dev_lock);
if (status)
goto mbx_err;
+ if (qp->qp_type == IB_QPT_UD)
+ qp_attr->qkey = params.qkey;
qp_attr->qp_state = get_ibqp_state(IB_QPS_INIT);
qp_attr->cur_qp_state = get_ibqp_state(IB_QPS_INIT);
qp_attr->path_mtu =
diff --git a/drivers/infiniband/hw/qib/qib_mr.c b/drivers/infiniband/hw/qib/qib_mr.c
index 9bbb553..a77fb4f 100644
--- a/drivers/infiniband/hw/qib/qib_mr.c
+++ b/drivers/infiniband/hw/qib/qib_mr.c
@@ -258,7 +258,7 @@
mr->mr.user_base = start;
mr->mr.iova = virt_addr;
mr->mr.length = length;
- mr->mr.offset = umem->offset;
+ mr->mr.offset = ib_umem_offset(umem);
mr->mr.access_flags = mr_access_flags;
mr->umem = umem;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index d7562be..8ba80a6 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -98,9 +98,15 @@
IPOIB_MCAST_FLAG_FOUND = 0, /* used in set_multicast_list */
IPOIB_MCAST_FLAG_SENDONLY = 1,
- IPOIB_MCAST_FLAG_BUSY = 2, /* joining or already joined */
+ /*
+ * For IPOIB_MCAST_FLAG_BUSY
+ * When set, in flight join and mcast->mc is unreliable
+ * When clear and mcast->mc IS_ERR_OR_NULL, need to restart or
+ * haven't started yet
+ * When clear and mcast->mc is valid pointer, join was successful
+ */
+ IPOIB_MCAST_FLAG_BUSY = 2,
IPOIB_MCAST_FLAG_ATTACHED = 3,
- IPOIB_MCAST_JOIN_STARTED = 4,
MAX_SEND_CQE = 16,
IPOIB_CM_COPYBREAK = 256,
@@ -317,6 +323,7 @@
struct list_head multicast_list;
struct rb_root multicast_tree;
+ struct workqueue_struct *wq;
struct delayed_work mcast_task;
struct work_struct carrier_on_task;
struct work_struct flush_light;
@@ -477,10 +484,10 @@
void ipoib_pkey_event(struct work_struct *work);
void ipoib_ib_dev_cleanup(struct net_device *dev);
-int ipoib_ib_dev_open(struct net_device *dev, int flush);
+int ipoib_ib_dev_open(struct net_device *dev);
int ipoib_ib_dev_up(struct net_device *dev);
-int ipoib_ib_dev_down(struct net_device *dev, int flush);
-int ipoib_ib_dev_stop(struct net_device *dev, int flush);
+int ipoib_ib_dev_down(struct net_device *dev);
+int ipoib_ib_dev_stop(struct net_device *dev);
void ipoib_pkey_dev_check_presence(struct net_device *dev);
int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port);
@@ -492,7 +499,7 @@
void ipoib_mcast_restart_task(struct work_struct *work);
int ipoib_mcast_start_thread(struct net_device *dev);
-int ipoib_mcast_stop_thread(struct net_device *dev, int flush);
+int ipoib_mcast_stop_thread(struct net_device *dev);
void ipoib_mcast_dev_down(struct net_device *dev);
void ipoib_mcast_dev_flush(struct net_device *dev);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
index 933efce..56959ad 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
@@ -474,7 +474,7 @@
}
spin_lock_irq(&priv->lock);
- queue_delayed_work(ipoib_workqueue,
+ queue_delayed_work(priv->wq,
&priv->cm.stale_task, IPOIB_CM_RX_DELAY);
/* Add this entry to passive ids list head, but do not re-add it
* if IB_EVENT_QP_LAST_WQE_REACHED has moved it to flush list. */
@@ -576,7 +576,7 @@
spin_lock_irqsave(&priv->lock, flags);
list_splice_init(&priv->cm.rx_drain_list, &priv->cm.rx_reap_list);
ipoib_cm_start_rx_drain(priv);
- queue_work(ipoib_workqueue, &priv->cm.rx_reap_task);
+ queue_work(priv->wq, &priv->cm.rx_reap_task);
spin_unlock_irqrestore(&priv->lock, flags);
} else
ipoib_warn(priv, "cm recv completion event with wrid %d (> %d)\n",
@@ -603,7 +603,7 @@
spin_lock_irqsave(&priv->lock, flags);
list_move(&p->list, &priv->cm.rx_reap_list);
spin_unlock_irqrestore(&priv->lock, flags);
- queue_work(ipoib_workqueue, &priv->cm.rx_reap_task);
+ queue_work(priv->wq, &priv->cm.rx_reap_task);
}
return;
}
@@ -827,7 +827,7 @@
if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &tx->flags)) {
list_move(&tx->list, &priv->cm.reap_list);
- queue_work(ipoib_workqueue, &priv->cm.reap_task);
+ queue_work(priv->wq, &priv->cm.reap_task);
}
clear_bit(IPOIB_FLAG_OPER_UP, &tx->flags);
@@ -1255,7 +1255,7 @@
if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &tx->flags)) {
list_move(&tx->list, &priv->cm.reap_list);
- queue_work(ipoib_workqueue, &priv->cm.reap_task);
+ queue_work(priv->wq, &priv->cm.reap_task);
}
spin_unlock_irqrestore(&priv->lock, flags);
@@ -1284,7 +1284,7 @@
tx->dev = dev;
list_add(&tx->list, &priv->cm.start_list);
set_bit(IPOIB_FLAG_INITIALIZED, &tx->flags);
- queue_work(ipoib_workqueue, &priv->cm.start_task);
+ queue_work(priv->wq, &priv->cm.start_task);
return tx;
}
@@ -1295,7 +1295,7 @@
if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &tx->flags)) {
spin_lock_irqsave(&priv->lock, flags);
list_move(&tx->list, &priv->cm.reap_list);
- queue_work(ipoib_workqueue, &priv->cm.reap_task);
+ queue_work(priv->wq, &priv->cm.reap_task);
ipoib_dbg(priv, "Reap connection for gid %pI6\n",
tx->neigh->daddr + 4);
tx->neigh = NULL;
@@ -1417,7 +1417,7 @@
skb_queue_tail(&priv->cm.skb_queue, skb);
if (e)
- queue_work(ipoib_workqueue, &priv->cm.skb_task);
+ queue_work(priv->wq, &priv->cm.skb_task);
}
static void ipoib_cm_rx_reap(struct work_struct *work)
@@ -1450,7 +1450,7 @@
}
if (!list_empty(&priv->cm.passive_ids))
- queue_delayed_work(ipoib_workqueue,
+ queue_delayed_work(priv->wq,
&priv->cm.stale_task, IPOIB_CM_RX_DELAY);
spin_unlock_irq(&priv->lock);
}
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 72626c34..fe65abb 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -655,7 +655,7 @@
__ipoib_reap_ah(dev);
if (!test_bit(IPOIB_STOP_REAPER, &priv->flags))
- queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task,
+ queue_delayed_work(priv->wq, &priv->ah_reap_task,
round_jiffies_relative(HZ));
}
@@ -664,7 +664,7 @@
drain_tx_cq((struct net_device *)ctx);
}
-int ipoib_ib_dev_open(struct net_device *dev, int flush)
+int ipoib_ib_dev_open(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
int ret;
@@ -696,7 +696,7 @@
}
clear_bit(IPOIB_STOP_REAPER, &priv->flags);
- queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task,
+ queue_delayed_work(priv->wq, &priv->ah_reap_task,
round_jiffies_relative(HZ));
if (!test_and_set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
@@ -706,7 +706,7 @@
dev_stop:
if (!test_and_set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
napi_enable(&priv->napi);
- ipoib_ib_dev_stop(dev, flush);
+ ipoib_ib_dev_stop(dev);
return -1;
}
@@ -738,7 +738,7 @@
return ipoib_mcast_start_thread(dev);
}
-int ipoib_ib_dev_down(struct net_device *dev, int flush)
+int ipoib_ib_dev_down(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
@@ -747,7 +747,7 @@
clear_bit(IPOIB_FLAG_OPER_UP, &priv->flags);
netif_carrier_off(dev);
- ipoib_mcast_stop_thread(dev, flush);
+ ipoib_mcast_stop_thread(dev);
ipoib_mcast_dev_flush(dev);
ipoib_flush_paths(dev);
@@ -807,7 +807,7 @@
local_bh_enable();
}
-int ipoib_ib_dev_stop(struct net_device *dev, int flush)
+int ipoib_ib_dev_stop(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct ib_qp_attr qp_attr;
@@ -880,8 +880,7 @@
/* Wait for all AHs to be reaped */
set_bit(IPOIB_STOP_REAPER, &priv->flags);
cancel_delayed_work(&priv->ah_reap_task);
- if (flush)
- flush_workqueue(ipoib_workqueue);
+ flush_workqueue(priv->wq);
begin = jiffies;
@@ -918,7 +917,7 @@
(unsigned long) dev);
if (dev->flags & IFF_UP) {
- if (ipoib_ib_dev_open(dev, 1)) {
+ if (ipoib_ib_dev_open(dev)) {
ipoib_transport_dev_cleanup(dev);
return -ENODEV;
}
@@ -1040,12 +1039,12 @@
}
if (level >= IPOIB_FLUSH_NORMAL)
- ipoib_ib_dev_down(dev, 0);
+ ipoib_ib_dev_down(dev);
if (level == IPOIB_FLUSH_HEAVY) {
if (test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
- ipoib_ib_dev_stop(dev, 0);
- if (ipoib_ib_dev_open(dev, 0) != 0)
+ ipoib_ib_dev_stop(dev);
+ if (ipoib_ib_dev_open(dev) != 0)
return;
if (netif_queue_stopped(dev))
netif_start_queue(dev);
@@ -1097,7 +1096,7 @@
*/
ipoib_flush_paths(dev);
- ipoib_mcast_stop_thread(dev, 1);
+ ipoib_mcast_stop_thread(dev);
ipoib_mcast_dev_flush(dev);
ipoib_transport_dev_cleanup(dev);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 58b5aa3..6bad17d 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -108,7 +108,7 @@
set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
- if (ipoib_ib_dev_open(dev, 1)) {
+ if (ipoib_ib_dev_open(dev)) {
if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags))
return 0;
goto err_disable;
@@ -139,7 +139,7 @@
return 0;
err_stop:
- ipoib_ib_dev_stop(dev, 1);
+ ipoib_ib_dev_stop(dev);
err_disable:
clear_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
@@ -157,8 +157,8 @@
netif_stop_queue(dev);
- ipoib_ib_dev_down(dev, 1);
- ipoib_ib_dev_stop(dev, 0);
+ ipoib_ib_dev_down(dev);
+ ipoib_ib_dev_stop(dev);
if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) {
struct ipoib_dev_priv *cpriv;
@@ -839,7 +839,7 @@
return;
}
- queue_work(ipoib_workqueue, &priv->restart_task);
+ queue_work(priv->wq, &priv->restart_task);
}
static u32 ipoib_addr_hash(struct ipoib_neigh_hash *htbl, u8 *daddr)
@@ -954,7 +954,7 @@
__ipoib_reap_neigh(priv);
if (!test_bit(IPOIB_STOP_NEIGH_GC, &priv->flags))
- queue_delayed_work(ipoib_workqueue, &priv->neigh_reap_task,
+ queue_delayed_work(priv->wq, &priv->neigh_reap_task,
arp_tbl.gc_interval);
}
@@ -1133,7 +1133,7 @@
/* start garbage collection */
clear_bit(IPOIB_STOP_NEIGH_GC, &priv->flags);
- queue_delayed_work(ipoib_workqueue, &priv->neigh_reap_task,
+ queue_delayed_work(priv->wq, &priv->neigh_reap_task,
arp_tbl.gc_interval);
return 0;
@@ -1262,15 +1262,13 @@
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
- if (ipoib_neigh_hash_init(priv) < 0)
- goto out;
/* Allocate RX/TX "rings" to hold queued skbs */
priv->rx_ring = kzalloc(ipoib_recvq_size * sizeof *priv->rx_ring,
GFP_KERNEL);
if (!priv->rx_ring) {
printk(KERN_WARNING "%s: failed to allocate RX ring (%d entries)\n",
ca->name, ipoib_recvq_size);
- goto out_neigh_hash_cleanup;
+ goto out;
}
priv->tx_ring = vzalloc(ipoib_sendq_size * sizeof *priv->tx_ring);
@@ -1285,16 +1283,24 @@
if (ipoib_ib_dev_init(dev, ca, port))
goto out_tx_ring_cleanup;
+ /*
+ * Must be after ipoib_ib_dev_init so we can allocate a per
+ * device wq there and use it here
+ */
+ if (ipoib_neigh_hash_init(priv) < 0)
+ goto out_dev_uninit;
+
return 0;
+out_dev_uninit:
+ ipoib_ib_dev_cleanup(dev);
+
out_tx_ring_cleanup:
vfree(priv->tx_ring);
out_rx_ring_cleanup:
kfree(priv->rx_ring);
-out_neigh_hash_cleanup:
- ipoib_neigh_hash_uninit(dev);
out:
return -ENOMEM;
}
@@ -1317,6 +1323,12 @@
}
unregister_netdevice_many(&head);
+ /*
+ * Must be before ipoib_ib_dev_cleanup or we delete an in use
+ * work queue
+ */
+ ipoib_neigh_hash_uninit(dev);
+
ipoib_ib_dev_cleanup(dev);
kfree(priv->rx_ring);
@@ -1324,8 +1336,6 @@
priv->rx_ring = NULL;
priv->tx_ring = NULL;
-
- ipoib_neigh_hash_uninit(dev);
}
static const struct header_ops ipoib_header_ops = {
@@ -1636,7 +1646,7 @@
/* Stop GC if started before flush */
set_bit(IPOIB_STOP_NEIGH_GC, &priv->flags);
cancel_delayed_work(&priv->neigh_reap_task);
- flush_workqueue(ipoib_workqueue);
+ flush_workqueue(priv->wq);
event_failed:
ipoib_dev_cleanup(priv->dev);
@@ -1707,7 +1717,7 @@
/* Stop GC */
set_bit(IPOIB_STOP_NEIGH_GC, &priv->flags);
cancel_delayed_work(&priv->neigh_reap_task);
- flush_workqueue(ipoib_workqueue);
+ flush_workqueue(priv->wq);
unregister_netdev(priv->dev);
free_netdev(priv->dev);
@@ -1748,8 +1758,13 @@
* unregister_netdev() and linkwatch_event take the rtnl lock,
* so flush_scheduled_work() can deadlock during device
* removal.
+ *
+ * In addition, bringing one device up and another down at the
+ * same time can deadlock a single workqueue, so we have this
+ * global fallback workqueue, but we also attempt to open a
+ * per device workqueue each time we bring an interface up
*/
- ipoib_workqueue = create_singlethread_workqueue("ipoib");
+ ipoib_workqueue = create_singlethread_workqueue("ipoib_flush");
if (!ipoib_workqueue) {
ret = -ENOMEM;
goto err_fs;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index ffb83b5..bc50dd0 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -190,12 +190,6 @@
spin_unlock_irq(&priv->lock);
priv->tx_wr.wr.ud.remote_qkey = priv->qkey;
set_qkey = 1;
-
- if (!ipoib_cm_admin_enabled(dev)) {
- rtnl_lock();
- dev_set_mtu(dev, min(priv->mcast_mtu, priv->admin_mtu));
- rtnl_unlock();
- }
}
if (!test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) {
@@ -277,16 +271,27 @@
struct ipoib_mcast *mcast = multicast->context;
struct net_device *dev = mcast->dev;
+ /*
+ * We have to take the mutex to force mcast_sendonly_join to
+ * return from ib_sa_multicast_join and set mcast->mc to a
+ * valid value. Otherwise we were racing with ourselves in
+ * that we might fail here, but get a valid return from
+ * ib_sa_multicast_join after we had cleared mcast->mc here,
+ * resulting in mis-matched joins and leaves and a deadlock
+ */
+ mutex_lock(&mcast_mutex);
+
/* We trap for port events ourselves. */
if (status == -ENETRESET)
- return 0;
+ goto out;
if (!status)
status = ipoib_mcast_join_finish(mcast, &multicast->rec);
if (status) {
if (mcast->logcount++ < 20)
- ipoib_dbg_mcast(netdev_priv(dev), "multicast join failed for %pI6, status %d\n",
+ ipoib_dbg_mcast(netdev_priv(dev), "sendonly multicast "
+ "join failed for %pI6, status %d\n",
mcast->mcmember.mgid.raw, status);
/* Flush out any queued packets */
@@ -296,11 +301,15 @@
dev_kfree_skb_any(skb_dequeue(&mcast->pkt_queue));
}
netif_tx_unlock_bh(dev);
-
- /* Clear the busy flag so we try again */
- status = test_and_clear_bit(IPOIB_MCAST_FLAG_BUSY,
- &mcast->flags);
}
+out:
+ clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
+ if (status)
+ mcast->mc = NULL;
+ complete(&mcast->done);
+ if (status == -ENETRESET)
+ status = 0;
+ mutex_unlock(&mcast_mutex);
return status;
}
@@ -318,12 +327,14 @@
int ret = 0;
if (!test_bit(IPOIB_FLAG_OPER_UP, &priv->flags)) {
- ipoib_dbg_mcast(priv, "device shutting down, no multicast joins\n");
+ ipoib_dbg_mcast(priv, "device shutting down, no sendonly "
+ "multicast joins\n");
return -ENODEV;
}
- if (test_and_set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags)) {
- ipoib_dbg_mcast(priv, "multicast entry busy, skipping\n");
+ if (test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags)) {
+ ipoib_dbg_mcast(priv, "multicast entry busy, skipping "
+ "sendonly join\n");
return -EBUSY;
}
@@ -331,6 +342,9 @@
rec.port_gid = priv->local_gid;
rec.pkey = cpu_to_be16(priv->pkey);
+ mutex_lock(&mcast_mutex);
+ init_completion(&mcast->done);
+ set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
mcast->mc = ib_sa_join_multicast(&ipoib_sa_client, priv->ca,
priv->port, &rec,
IB_SA_MCMEMBER_REC_MGID |
@@ -343,12 +357,14 @@
if (IS_ERR(mcast->mc)) {
ret = PTR_ERR(mcast->mc);
clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
- ipoib_warn(priv, "ib_sa_join_multicast failed (ret = %d)\n",
- ret);
+ complete(&mcast->done);
+ ipoib_warn(priv, "ib_sa_join_multicast for sendonly join "
+ "failed (ret = %d)\n", ret);
} else {
- ipoib_dbg_mcast(priv, "no multicast record for %pI6, starting join\n",
- mcast->mcmember.mgid.raw);
+ ipoib_dbg_mcast(priv, "no multicast record for %pI6, starting "
+ "sendonly join\n", mcast->mcmember.mgid.raw);
}
+ mutex_unlock(&mcast_mutex);
return ret;
}
@@ -359,18 +375,29 @@
carrier_on_task);
struct ib_port_attr attr;
- /*
- * Take rtnl_lock to avoid racing with ipoib_stop() and
- * turning the carrier back on while a device is being
- * removed.
- */
if (ib_query_port(priv->ca, priv->port, &attr) ||
attr.state != IB_PORT_ACTIVE) {
ipoib_dbg(priv, "Keeping carrier off until IB port is active\n");
return;
}
- rtnl_lock();
+ /*
+ * Take rtnl_lock to avoid racing with ipoib_stop() and
+ * turning the carrier back on while a device is being
+ * removed. However, ipoib_stop() will attempt to flush
+ * the workqueue while holding the rtnl lock, so loop
+ * on trylock until either we get the lock or we see
+ * FLAG_ADMIN_UP go away as that signals that we are bailing
+ * and can safely ignore the carrier on work.
+ */
+ while (!rtnl_trylock()) {
+ if (!test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
+ return;
+ else
+ msleep(20);
+ }
+ if (!ipoib_cm_admin_enabled(priv->dev))
+ dev_set_mtu(priv->dev, min(priv->mcast_mtu, priv->admin_mtu));
netif_carrier_on(priv->dev);
rtnl_unlock();
}
@@ -385,60 +412,63 @@
ipoib_dbg_mcast(priv, "join completion for %pI6 (status %d)\n",
mcast->mcmember.mgid.raw, status);
+ /*
+ * We have to take the mutex to force mcast_join to
+ * return from ib_sa_multicast_join and set mcast->mc to a
+ * valid value. Otherwise we were racing with ourselves in
+ * that we might fail here, but get a valid return from
+ * ib_sa_multicast_join after we had cleared mcast->mc here,
+ * resulting in mis-matched joins and leaves and a deadlock
+ */
+ mutex_lock(&mcast_mutex);
+
/* We trap for port events ourselves. */
- if (status == -ENETRESET) {
- status = 0;
+ if (status == -ENETRESET)
goto out;
- }
if (!status)
status = ipoib_mcast_join_finish(mcast, &multicast->rec);
if (!status) {
mcast->backoff = 1;
- mutex_lock(&mcast_mutex);
if (test_bit(IPOIB_MCAST_RUN, &priv->flags))
- queue_delayed_work(ipoib_workqueue,
- &priv->mcast_task, 0);
- mutex_unlock(&mcast_mutex);
+ queue_delayed_work(priv->wq, &priv->mcast_task, 0);
/*
- * Defer carrier on work to ipoib_workqueue to avoid a
+ * Defer carrier on work to priv->wq to avoid a
* deadlock on rtnl_lock here.
*/
if (mcast == priv->broadcast)
- queue_work(ipoib_workqueue, &priv->carrier_on_task);
-
- status = 0;
- goto out;
- }
-
- if (mcast->logcount++ < 20) {
- if (status == -ETIMEDOUT || status == -EAGAIN) {
- ipoib_dbg_mcast(priv, "multicast join failed for %pI6, status %d\n",
- mcast->mcmember.mgid.raw, status);
- } else {
- ipoib_warn(priv, "multicast join failed for %pI6, status %d\n",
- mcast->mcmember.mgid.raw, status);
+ queue_work(priv->wq, &priv->carrier_on_task);
+ } else {
+ if (mcast->logcount++ < 20) {
+ if (status == -ETIMEDOUT || status == -EAGAIN) {
+ ipoib_dbg_mcast(priv, "multicast join failed for %pI6, status %d\n",
+ mcast->mcmember.mgid.raw, status);
+ } else {
+ ipoib_warn(priv, "multicast join failed for %pI6, status %d\n",
+ mcast->mcmember.mgid.raw, status);
+ }
}
+
+ mcast->backoff *= 2;
+ if (mcast->backoff > IPOIB_MAX_BACKOFF_SECONDS)
+ mcast->backoff = IPOIB_MAX_BACKOFF_SECONDS;
}
-
- mcast->backoff *= 2;
- if (mcast->backoff > IPOIB_MAX_BACKOFF_SECONDS)
- mcast->backoff = IPOIB_MAX_BACKOFF_SECONDS;
-
- /* Clear the busy flag so we try again */
- status = test_and_clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
-
- mutex_lock(&mcast_mutex);
+out:
spin_lock_irq(&priv->lock);
- if (test_bit(IPOIB_MCAST_RUN, &priv->flags))
- queue_delayed_work(ipoib_workqueue, &priv->mcast_task,
+ clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
+ if (status)
+ mcast->mc = NULL;
+ complete(&mcast->done);
+ if (status == -ENETRESET)
+ status = 0;
+ if (status && test_bit(IPOIB_MCAST_RUN, &priv->flags))
+ queue_delayed_work(priv->wq, &priv->mcast_task,
mcast->backoff * HZ);
spin_unlock_irq(&priv->lock);
mutex_unlock(&mcast_mutex);
-out:
- complete(&mcast->done);
+
return status;
}
@@ -487,10 +517,9 @@
rec.hop_limit = priv->broadcast->mcmember.hop_limit;
}
- set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
+ mutex_lock(&mcast_mutex);
init_completion(&mcast->done);
- set_bit(IPOIB_MCAST_JOIN_STARTED, &mcast->flags);
-
+ set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
mcast->mc = ib_sa_join_multicast(&ipoib_sa_client, priv->ca, priv->port,
&rec, comp_mask, GFP_KERNEL,
ipoib_mcast_join_complete, mcast);
@@ -504,13 +533,11 @@
if (mcast->backoff > IPOIB_MAX_BACKOFF_SECONDS)
mcast->backoff = IPOIB_MAX_BACKOFF_SECONDS;
- mutex_lock(&mcast_mutex);
if (test_bit(IPOIB_MCAST_RUN, &priv->flags))
- queue_delayed_work(ipoib_workqueue,
- &priv->mcast_task,
+ queue_delayed_work(priv->wq, &priv->mcast_task,
mcast->backoff * HZ);
- mutex_unlock(&mcast_mutex);
}
+ mutex_unlock(&mcast_mutex);
}
void ipoib_mcast_join_task(struct work_struct *work)
@@ -547,8 +574,8 @@
ipoib_warn(priv, "failed to allocate broadcast group\n");
mutex_lock(&mcast_mutex);
if (test_bit(IPOIB_MCAST_RUN, &priv->flags))
- queue_delayed_work(ipoib_workqueue,
- &priv->mcast_task, HZ);
+ queue_delayed_work(priv->wq, &priv->mcast_task,
+ HZ);
mutex_unlock(&mcast_mutex);
return;
}
@@ -563,7 +590,8 @@
}
if (!test_bit(IPOIB_MCAST_FLAG_ATTACHED, &priv->broadcast->flags)) {
- if (!test_bit(IPOIB_MCAST_FLAG_BUSY, &priv->broadcast->flags))
+ if (IS_ERR_OR_NULL(priv->broadcast->mc) &&
+ !test_bit(IPOIB_MCAST_FLAG_BUSY, &priv->broadcast->flags))
ipoib_mcast_join(dev, priv->broadcast, 0);
return;
}
@@ -571,23 +599,33 @@
while (1) {
struct ipoib_mcast *mcast = NULL;
+ /*
+ * Need the mutex so our flags are consistent, need the
+ * priv->lock so we don't race with list removals in either
+ * mcast_dev_flush or mcast_restart_task
+ */
+ mutex_lock(&mcast_mutex);
spin_lock_irq(&priv->lock);
list_for_each_entry(mcast, &priv->multicast_list, list) {
- if (!test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)
- && !test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags)
- && !test_bit(IPOIB_MCAST_FLAG_ATTACHED, &mcast->flags)) {
+ if (IS_ERR_OR_NULL(mcast->mc) &&
+ !test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags) &&
+ !test_bit(IPOIB_MCAST_FLAG_ATTACHED, &mcast->flags)) {
/* Found the next unjoined group */
break;
}
}
spin_unlock_irq(&priv->lock);
+ mutex_unlock(&mcast_mutex);
if (&mcast->list == &priv->multicast_list) {
/* All done */
break;
}
- ipoib_mcast_join(dev, mcast, 1);
+ if (test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags))
+ ipoib_mcast_sendonly_join(mcast);
+ else
+ ipoib_mcast_join(dev, mcast, 1);
return;
}
@@ -604,13 +642,13 @@
mutex_lock(&mcast_mutex);
if (!test_and_set_bit(IPOIB_MCAST_RUN, &priv->flags))
- queue_delayed_work(ipoib_workqueue, &priv->mcast_task, 0);
+ queue_delayed_work(priv->wq, &priv->mcast_task, 0);
mutex_unlock(&mcast_mutex);
return 0;
}
-int ipoib_mcast_stop_thread(struct net_device *dev, int flush)
+int ipoib_mcast_stop_thread(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
@@ -621,8 +659,7 @@
cancel_delayed_work(&priv->mcast_task);
mutex_unlock(&mcast_mutex);
- if (flush)
- flush_workqueue(ipoib_workqueue);
+ flush_workqueue(priv->wq);
return 0;
}
@@ -633,6 +670,9 @@
int ret = 0;
if (test_and_clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags))
+ ipoib_warn(priv, "ipoib_mcast_leave on an in-flight join\n");
+
+ if (!IS_ERR_OR_NULL(mcast->mc))
ib_sa_free_multicast(mcast->mc);
if (test_and_clear_bit(IPOIB_MCAST_FLAG_ATTACHED, &mcast->flags)) {
@@ -685,6 +725,8 @@
memcpy(mcast->mcmember.mgid.raw, mgid, sizeof (union ib_gid));
__ipoib_mcast_add(dev, mcast);
list_add_tail(&mcast->list, &priv->multicast_list);
+ if (!test_and_set_bit(IPOIB_MCAST_RUN, &priv->flags))
+ queue_delayed_work(priv->wq, &priv->mcast_task, 0);
}
if (!mcast->ah) {
@@ -698,8 +740,6 @@
if (test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags))
ipoib_dbg_mcast(priv, "no address vector, "
"but multicast join already started\n");
- else if (test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags))
- ipoib_mcast_sendonly_join(mcast);
/*
* If lookup completes between here and out:, don't
@@ -759,9 +799,12 @@
spin_unlock_irqrestore(&priv->lock, flags);
- /* seperate between the wait to the leave*/
+ /*
+ * make sure the in-flight joins have finished before we attempt
+ * to leave
+ */
list_for_each_entry_safe(mcast, tmcast, &remove_list, list)
- if (test_bit(IPOIB_MCAST_JOIN_STARTED, &mcast->flags))
+ if (test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags))
wait_for_completion(&mcast->done);
list_for_each_entry_safe(mcast, tmcast, &remove_list, list) {
@@ -794,8 +837,6 @@
ipoib_dbg_mcast(priv, "restarting multicast task\n");
- ipoib_mcast_stop_thread(dev, 0);
-
local_irq_save(flags);
netif_addr_lock(dev);
spin_lock(&priv->lock);
@@ -880,14 +921,38 @@
netif_addr_unlock(dev);
local_irq_restore(flags);
- /* We have to cancel outside of the spinlock */
+ /*
+ * make sure the in-flight joins have finished before we attempt
+ * to leave
+ */
+ list_for_each_entry_safe(mcast, tmcast, &remove_list, list)
+ if (test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags))
+ wait_for_completion(&mcast->done);
+
+ /*
+ * We have to cancel outside of the spinlock, but we have to
+ * take the rtnl lock or else we race with the removal of
+ * entries from the remove list in mcast_dev_flush as part
+ * of ipoib_stop(). We detect the drop of the ADMIN_UP flag
+ * to signal that we have hit this particular race, and we
+ * return since we know we don't need to do anything else
+ * anyway.
+ */
+ while (!rtnl_trylock()) {
+ if (!test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
+ return;
+ else
+ msleep(20);
+ }
list_for_each_entry_safe(mcast, tmcast, &remove_list, list) {
ipoib_mcast_leave(mcast->dev, mcast);
ipoib_mcast_free(mcast);
}
-
- if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
- ipoib_mcast_start_thread(dev);
+ /*
+ * Restart our join task if needed
+ */
+ ipoib_mcast_start_thread(dev);
+ rtnl_unlock();
}
#ifdef CONFIG_INFINIBAND_IPOIB_DEBUG
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
index c56d5d4..b72a753 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
@@ -145,10 +145,20 @@
int ret, size;
int i;
+ /*
+ * the various IPoIB tasks assume they will never race against
+ * themselves, so always use a single thread workqueue
+ */
+ priv->wq = create_singlethread_workqueue("ipoib_wq");
+ if (!priv->wq) {
+ printk(KERN_WARNING "ipoib: failed to allocate device WQ\n");
+ return -ENODEV;
+ }
+
priv->pd = ib_alloc_pd(priv->ca);
if (IS_ERR(priv->pd)) {
printk(KERN_WARNING "%s: failed to allocate PD\n", ca->name);
- return -ENODEV;
+ goto out_free_wq;
}
priv->mr = ib_get_dma_mr(priv->pd, IB_ACCESS_LOCAL_WRITE);
@@ -242,6 +252,10 @@
out_free_pd:
ib_dealloc_pd(priv->pd);
+
+out_free_wq:
+ destroy_workqueue(priv->wq);
+ priv->wq = NULL;
return -ENODEV;
}
@@ -270,6 +284,12 @@
if (ib_dealloc_pd(priv->pd))
ipoib_warn(priv, "ib_dealloc_pd failed\n");
+
+ if (priv->wq) {
+ flush_workqueue(priv->wq);
+ destroy_workqueue(priv->wq);
+ priv->wq = NULL;
+ }
}
void ipoib_event(struct ib_event_handler *handler,
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
index 20ca6a6..6a594aa 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
@@ -97,7 +97,7 @@
MODULE_PARM_DESC(pi_enable, "Enable T10-PI offload support (default:disabled)");
module_param_named(pi_guard, iser_pi_guard, int, 0644);
-MODULE_PARM_DESC(pi_guard, "T10-PI guard_type, 0:CRC|1:IP_CSUM (default:IP_CSUM)");
+MODULE_PARM_DESC(pi_guard, "T10-PI guard_type [deprecated]");
static struct workqueue_struct *release_wq;
struct iser_global ig;
@@ -164,18 +164,42 @@
return 0;
}
-int iser_initialize_task_headers(struct iscsi_task *task,
- struct iser_tx_desc *tx_desc)
+/**
+ * iser_initialize_task_headers() - Initialize task headers
+ * @task: iscsi task
+ * @tx_desc: iser tx descriptor
+ *
+ * Notes:
+ * This routine may race with iser teardown flow for scsi
+ * error handling TMFs. So for TMF we should acquire the
+ * state mutex to avoid dereferencing the IB device which
+ * may have already been terminated.
+ */
+int
+iser_initialize_task_headers(struct iscsi_task *task,
+ struct iser_tx_desc *tx_desc)
{
- struct iser_conn *iser_conn = task->conn->dd_data;
+ struct iser_conn *iser_conn = task->conn->dd_data;
struct iser_device *device = iser_conn->ib_conn.device;
struct iscsi_iser_task *iser_task = task->dd_data;
u64 dma_addr;
+ const bool mgmt_task = !task->sc && !in_interrupt();
+ int ret = 0;
+
+ if (unlikely(mgmt_task))
+ mutex_lock(&iser_conn->state_mutex);
+
+ if (unlikely(iser_conn->state != ISER_CONN_UP)) {
+ ret = -ENODEV;
+ goto out;
+ }
dma_addr = ib_dma_map_single(device->ib_device, (void *)tx_desc,
ISER_HEADERS_LEN, DMA_TO_DEVICE);
- if (ib_dma_mapping_error(device->ib_device, dma_addr))
- return -ENOMEM;
+ if (ib_dma_mapping_error(device->ib_device, dma_addr)) {
+ ret = -ENOMEM;
+ goto out;
+ }
tx_desc->dma_addr = dma_addr;
tx_desc->tx_sg[0].addr = tx_desc->dma_addr;
@@ -183,7 +207,11 @@
tx_desc->tx_sg[0].lkey = device->mr->lkey;
iser_task->iser_conn = iser_conn;
- return 0;
+out:
+ if (unlikely(mgmt_task))
+ mutex_unlock(&iser_conn->state_mutex);
+
+ return ret;
}
/**
@@ -199,9 +227,14 @@
iscsi_iser_task_init(struct iscsi_task *task)
{
struct iscsi_iser_task *iser_task = task->dd_data;
+ int ret;
- if (iser_initialize_task_headers(task, &iser_task->desc))
- return -ENOMEM;
+ ret = iser_initialize_task_headers(task, &iser_task->desc);
+ if (ret) {
+ iser_err("Failed to init task %p, err = %d\n",
+ iser_task, ret);
+ return ret;
+ }
/* mgmt task */
if (!task->sc)
@@ -508,8 +541,8 @@
*/
if (iser_conn) {
mutex_lock(&iser_conn->state_mutex);
- iscsi_conn_stop(cls_conn, flag);
iser_conn_terminate(iser_conn);
+ iscsi_conn_stop(cls_conn, flag);
/* unbind */
iser_conn->iscsi_conn = NULL;
@@ -541,12 +574,13 @@
static inline unsigned int
iser_dif_prot_caps(int prot_caps)
{
- return ((prot_caps & IB_PROT_T10DIF_TYPE_1) ? SHOST_DIF_TYPE1_PROTECTION |
- SHOST_DIX_TYPE1_PROTECTION : 0) |
- ((prot_caps & IB_PROT_T10DIF_TYPE_2) ? SHOST_DIF_TYPE2_PROTECTION |
- SHOST_DIX_TYPE2_PROTECTION : 0) |
- ((prot_caps & IB_PROT_T10DIF_TYPE_3) ? SHOST_DIF_TYPE3_PROTECTION |
- SHOST_DIX_TYPE3_PROTECTION : 0);
+ return ((prot_caps & IB_PROT_T10DIF_TYPE_1) ?
+ SHOST_DIF_TYPE1_PROTECTION | SHOST_DIX_TYPE0_PROTECTION |
+ SHOST_DIX_TYPE1_PROTECTION : 0) |
+ ((prot_caps & IB_PROT_T10DIF_TYPE_2) ?
+ SHOST_DIF_TYPE2_PROTECTION | SHOST_DIX_TYPE2_PROTECTION : 0) |
+ ((prot_caps & IB_PROT_T10DIF_TYPE_3) ?
+ SHOST_DIF_TYPE3_PROTECTION | SHOST_DIX_TYPE3_PROTECTION : 0);
}
/**
@@ -569,6 +603,7 @@
struct Scsi_Host *shost;
struct iser_conn *iser_conn = NULL;
struct ib_conn *ib_conn;
+ u16 max_cmds;
shost = iscsi_host_alloc(&iscsi_iser_sht, 0, 0);
if (!shost)
@@ -586,26 +621,41 @@
*/
if (ep) {
iser_conn = ep->dd_data;
+ max_cmds = iser_conn->max_cmds;
+
+ mutex_lock(&iser_conn->state_mutex);
+ if (iser_conn->state != ISER_CONN_UP) {
+ iser_err("iser conn %p already started teardown\n",
+ iser_conn);
+ mutex_unlock(&iser_conn->state_mutex);
+ goto free_host;
+ }
+
ib_conn = &iser_conn->ib_conn;
if (ib_conn->pi_support) {
u32 sig_caps = ib_conn->device->dev_attr.sig_prot_cap;
scsi_host_set_prot(shost, iser_dif_prot_caps(sig_caps));
- if (iser_pi_guard)
- scsi_host_set_guard(shost, SHOST_DIX_GUARD_IP);
- else
- scsi_host_set_guard(shost, SHOST_DIX_GUARD_CRC);
+ scsi_host_set_guard(shost, SHOST_DIX_GUARD_IP |
+ SHOST_DIX_GUARD_CRC);
}
+
+ if (iscsi_host_add(shost,
+ ib_conn->device->ib_device->dma_device)) {
+ mutex_unlock(&iser_conn->state_mutex);
+ goto free_host;
+ }
+ mutex_unlock(&iser_conn->state_mutex);
+ } else {
+ max_cmds = ISER_DEF_XMIT_CMDS_MAX;
+ if (iscsi_host_add(shost, NULL))
+ goto free_host;
}
- if (iscsi_host_add(shost, ep ?
- ib_conn->device->ib_device->dma_device : NULL))
- goto free_host;
-
- if (cmds_max > ISER_DEF_XMIT_CMDS_MAX) {
+ if (cmds_max > max_cmds) {
iser_info("cmds_max changed from %u to %u\n",
- cmds_max, ISER_DEF_XMIT_CMDS_MAX);
- cmds_max = ISER_DEF_XMIT_CMDS_MAX;
+ cmds_max, max_cmds);
+ cmds_max = max_cmds;
}
cls_session = iscsi_session_setup(&iscsi_iser_transport, shost,
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h
index cd4174c..5ce2681 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.h
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.h
@@ -69,34 +69,31 @@
#define DRV_NAME "iser"
#define PFX DRV_NAME ": "
-#define DRV_VER "1.4.8"
+#define DRV_VER "1.5"
#define iser_dbg(fmt, arg...) \
do { \
- if (iser_debug_level > 2) \
+ if (unlikely(iser_debug_level > 2)) \
printk(KERN_DEBUG PFX "%s: " fmt,\
__func__ , ## arg); \
} while (0)
#define iser_warn(fmt, arg...) \
do { \
- if (iser_debug_level > 0) \
+ if (unlikely(iser_debug_level > 0)) \
pr_warn(PFX "%s: " fmt, \
__func__ , ## arg); \
} while (0)
#define iser_info(fmt, arg...) \
do { \
- if (iser_debug_level > 1) \
+ if (unlikely(iser_debug_level > 1)) \
pr_info(PFX "%s: " fmt, \
__func__ , ## arg); \
} while (0)
-#define iser_err(fmt, arg...) \
- do { \
- printk(KERN_ERR PFX "%s: " fmt, \
- __func__ , ## arg); \
- } while (0)
+#define iser_err(fmt, arg...) \
+ pr_err(PFX "%s: " fmt, __func__ , ## arg)
#define SHIFT_4K 12
#define SIZE_4K (1ULL << SHIFT_4K)
@@ -144,6 +141,11 @@
ISER_MAX_TX_MISC_PDUS + \
ISER_MAX_RX_MISC_PDUS)
+#define ISER_GET_MAX_XMIT_CMDS(send_wr) ((send_wr \
+ - ISER_MAX_TX_MISC_PDUS \
+ - ISER_MAX_RX_MISC_PDUS) / \
+ (1 + ISER_INFLIGHT_DATAOUTS))
+
#define ISER_WC_BATCH_COUNT 16
#define ISER_SIGNAL_CMD_COUNT 32
@@ -247,7 +249,6 @@
* @va: MR start address (buffer va)
* @len: MR length
* @mem_h: pointer to registration context (FMR/Fastreg)
- * @is_mr: indicates weather we registered the buffer
*/
struct iser_mem_reg {
u32 lkey;
@@ -255,7 +256,6 @@
u64 va;
u64 len;
void *mem_h;
- int is_mr;
};
/**
@@ -323,8 +323,6 @@
char pad[ISER_RX_PAD_SIZE];
} __attribute__((packed));
-#define ISER_MAX_CQ 4
-
struct iser_conn;
struct ib_conn;
struct iscsi_iser_task;
@@ -375,7 +373,7 @@
struct list_head ig_list;
int refcount;
int comps_used;
- struct iser_comp comps[ISER_MAX_CQ];
+ struct iser_comp *comps;
int (*iser_alloc_rdma_reg_res)(struct ib_conn *ib_conn,
unsigned cmds_max);
void (*iser_free_rdma_reg_res)(struct ib_conn *ib_conn);
@@ -432,6 +430,7 @@
* @cma_id: rdma_cm connection maneger handle
* @qp: Connection Queue-pair
* @post_recv_buf_count: post receive counter
+ * @sig_count: send work request signal count
* @rx_wr: receive work request for batch posts
* @device: reference to iser device
* @comp: iser completion context
@@ -452,6 +451,7 @@
struct rdma_cm_id *cma_id;
struct ib_qp *qp;
int post_recv_buf_count;
+ u8 sig_count;
struct ib_recv_wr rx_wr[ISER_MIN_POSTED_RX];
struct iser_device *device;
struct iser_comp *comp;
@@ -482,6 +482,7 @@
* to max number of post recvs
* @qp_max_recv_dtos_mask: (qp_max_recv_dtos - 1)
* @min_posted_rx: (qp_max_recv_dtos >> 2)
+ * @max_cmds: maximum cmds allowed for this connection
* @name: connection peer portal
* @release_work: deffered work for release job
* @state_mutex: protects iser onnection state
@@ -507,6 +508,7 @@
unsigned qp_max_recv_dtos;
unsigned qp_max_recv_dtos_mask;
unsigned min_posted_rx;
+ u16 max_cmds;
char name[ISER_OBJECT_NAME_SIZE];
struct work_struct release_work;
struct mutex state_mutex;
diff --git a/drivers/infiniband/ulp/iser/iser_initiator.c b/drivers/infiniband/ulp/iser/iser_initiator.c
index 5a489ea..3821633 100644
--- a/drivers/infiniband/ulp/iser/iser_initiator.c
+++ b/drivers/infiniband/ulp/iser/iser_initiator.c
@@ -369,7 +369,7 @@
return 0;
}
-static inline bool iser_signal_comp(int sig_count)
+static inline bool iser_signal_comp(u8 sig_count)
{
return ((sig_count % ISER_SIGNAL_CMD_COUNT) == 0);
}
@@ -388,7 +388,7 @@
struct iscsi_scsi_req *hdr = (struct iscsi_scsi_req *)task->hdr;
struct scsi_cmnd *sc = task->sc;
struct iser_tx_desc *tx_desc = &iser_task->desc;
- static unsigned sig_count;
+ u8 sig_count = ++iser_conn->ib_conn.sig_count;
edtl = ntohl(hdr->data_length);
@@ -435,7 +435,7 @@
iser_task->status = ISER_TASK_STATUS_STARTED;
err = iser_post_send(&iser_conn->ib_conn, tx_desc,
- iser_signal_comp(++sig_count));
+ iser_signal_comp(sig_count));
if (!err)
return 0;
diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c
index 6c5ce35..abce933 100644
--- a/drivers/infiniband/ulp/iser/iser_memory.c
+++ b/drivers/infiniband/ulp/iser/iser_memory.c
@@ -73,7 +73,6 @@
if (cmd_dir == ISER_DIR_OUT) {
/* copy the unaligned sg the buffer which is used for RDMA */
- int i;
char *p, *from;
sgl = (struct scatterlist *)data->buf;
@@ -409,7 +408,6 @@
regd_buf->reg.rkey = device->mr->rkey;
regd_buf->reg.len = ib_sg_dma_len(ibdev, &sg[0]);
regd_buf->reg.va = ib_sg_dma_address(ibdev, &sg[0]);
- regd_buf->reg.is_mr = 0;
iser_dbg("PHYSICAL Mem.register: lkey: 0x%08X rkey: 0x%08X "
"va: 0x%08lX sz: %ld]\n",
@@ -440,13 +438,13 @@
return 0;
}
-static inline void
+static void
iser_set_dif_domain(struct scsi_cmnd *sc, struct ib_sig_attrs *sig_attrs,
struct ib_sig_domain *domain)
{
domain->sig_type = IB_SIG_TYPE_T10_DIF;
- domain->sig.dif.pi_interval = sc->device->sector_size;
- domain->sig.dif.ref_tag = scsi_get_lba(sc) & 0xffffffff;
+ domain->sig.dif.pi_interval = scsi_prot_interval(sc);
+ domain->sig.dif.ref_tag = scsi_prot_ref_tag(sc);
/*
* At the moment we hard code those, but in the future
* we will take them from sc.
@@ -454,8 +452,7 @@
domain->sig.dif.apptag_check_mask = 0xffff;
domain->sig.dif.app_escape = true;
domain->sig.dif.ref_escape = true;
- if (scsi_get_prot_type(sc) == SCSI_PROT_DIF_TYPE1 ||
- scsi_get_prot_type(sc) == SCSI_PROT_DIF_TYPE2)
+ if (sc->prot_flags & SCSI_PROT_REF_INCREMENT)
domain->sig.dif.ref_remap = true;
};
@@ -473,26 +470,16 @@
case SCSI_PROT_WRITE_STRIP:
sig_attrs->wire.sig_type = IB_SIG_TYPE_NONE;
iser_set_dif_domain(sc, sig_attrs, &sig_attrs->mem);
- /*
- * At the moment we use this modparam to tell what is
- * the memory bg_type, in the future we will take it
- * from sc.
- */
- sig_attrs->mem.sig.dif.bg_type = iser_pi_guard ? IB_T10DIF_CSUM :
- IB_T10DIF_CRC;
+ sig_attrs->mem.sig.dif.bg_type = sc->prot_flags & SCSI_PROT_IP_CHECKSUM ?
+ IB_T10DIF_CSUM : IB_T10DIF_CRC;
break;
case SCSI_PROT_READ_PASS:
case SCSI_PROT_WRITE_PASS:
iser_set_dif_domain(sc, sig_attrs, &sig_attrs->wire);
sig_attrs->wire.sig.dif.bg_type = IB_T10DIF_CRC;
iser_set_dif_domain(sc, sig_attrs, &sig_attrs->mem);
- /*
- * At the moment we use this modparam to tell what is
- * the memory bg_type, in the future we will take it
- * from sc.
- */
- sig_attrs->mem.sig.dif.bg_type = iser_pi_guard ? IB_T10DIF_CSUM :
- IB_T10DIF_CRC;
+ sig_attrs->mem.sig.dif.bg_type = sc->prot_flags & SCSI_PROT_IP_CHECKSUM ?
+ IB_T10DIF_CSUM : IB_T10DIF_CRC;
break;
default:
iser_err("Unsupported PI operation %d\n",
@@ -503,26 +490,28 @@
return 0;
}
-static int
+static inline void
iser_set_prot_checks(struct scsi_cmnd *sc, u8 *mask)
{
- switch (scsi_get_prot_type(sc)) {
- case SCSI_PROT_DIF_TYPE0:
- break;
- case SCSI_PROT_DIF_TYPE1:
- case SCSI_PROT_DIF_TYPE2:
- *mask = ISER_CHECK_GUARD | ISER_CHECK_REFTAG;
- break;
- case SCSI_PROT_DIF_TYPE3:
- *mask = ISER_CHECK_GUARD;
- break;
- default:
- iser_err("Unsupported protection type %d\n",
- scsi_get_prot_type(sc));
- return -EINVAL;
- }
+ *mask = 0;
+ if (sc->prot_flags & SCSI_PROT_REF_CHECK)
+ *mask |= ISER_CHECK_REFTAG;
+ if (sc->prot_flags & SCSI_PROT_GUARD_CHECK)
+ *mask |= ISER_CHECK_GUARD;
+}
- return 0;
+static void
+iser_inv_rkey(struct ib_send_wr *inv_wr, struct ib_mr *mr)
+{
+ u32 rkey;
+
+ memset(inv_wr, 0, sizeof(*inv_wr));
+ inv_wr->opcode = IB_WR_LOCAL_INV;
+ inv_wr->wr_id = ISER_FASTREG_LI_WRID;
+ inv_wr->ex.invalidate_rkey = mr->rkey;
+
+ rkey = ib_inc_rkey(mr->rkey);
+ ib_update_fast_reg_key(mr, rkey);
}
static int
@@ -536,26 +525,17 @@
struct ib_send_wr *bad_wr, *wr = NULL;
struct ib_sig_attrs sig_attrs;
int ret;
- u32 key;
memset(&sig_attrs, 0, sizeof(sig_attrs));
ret = iser_set_sig_attrs(iser_task->sc, &sig_attrs);
if (ret)
goto err;
- ret = iser_set_prot_checks(iser_task->sc, &sig_attrs.check_mask);
- if (ret)
- goto err;
+ iser_set_prot_checks(iser_task->sc, &sig_attrs.check_mask);
if (!(desc->reg_indicators & ISER_SIG_KEY_VALID)) {
- memset(&inv_wr, 0, sizeof(inv_wr));
- inv_wr.opcode = IB_WR_LOCAL_INV;
- inv_wr.wr_id = ISER_FASTREG_LI_WRID;
- inv_wr.ex.invalidate_rkey = pi_ctx->sig_mr->rkey;
+ iser_inv_rkey(&inv_wr, pi_ctx->sig_mr);
wr = &inv_wr;
- /* Bump the key */
- key = (u8)(pi_ctx->sig_mr->rkey & 0x000000FF);
- ib_update_fast_reg_key(pi_ctx->sig_mr, ++key);
}
memset(&sig_wr, 0, sizeof(sig_wr));
@@ -585,12 +565,7 @@
sig_sge->lkey = pi_ctx->sig_mr->lkey;
sig_sge->addr = 0;
- sig_sge->length = data_sge->length + prot_sge->length;
- if (scsi_get_prot_op(iser_task->sc) == SCSI_PROT_WRITE_INSERT ||
- scsi_get_prot_op(iser_task->sc) == SCSI_PROT_READ_STRIP) {
- sig_sge->length += (data_sge->length /
- iser_task->sc->device->sector_size) * 8;
- }
+ sig_sge->length = scsi_transfer_length(iser_task->sc);
iser_dbg("sig_sge: addr: 0x%llx length: %u lkey: 0x%x\n",
sig_sge->addr, sig_sge->length,
@@ -613,7 +588,6 @@
struct ib_fast_reg_page_list *frpl;
struct ib_send_wr fastreg_wr, inv_wr;
struct ib_send_wr *bad_wr, *wr = NULL;
- u8 key;
int ret, offset, size, plen;
/* if there a single dma entry, dma mr suffices */
@@ -645,14 +619,8 @@
}
if (!(desc->reg_indicators & ind)) {
- memset(&inv_wr, 0, sizeof(inv_wr));
- inv_wr.wr_id = ISER_FASTREG_LI_WRID;
- inv_wr.opcode = IB_WR_LOCAL_INV;
- inv_wr.ex.invalidate_rkey = mr->rkey;
+ iser_inv_rkey(&inv_wr, mr);
wr = &inv_wr;
- /* Bump the key */
- key = (u8)(mr->rkey & 0x000000FF);
- ib_update_fast_reg_key(mr, ++key);
}
/* Prepare FASTREG WR */
@@ -770,15 +738,11 @@
regd_buf->reg.rkey = desc->pi_ctx->sig_mr->rkey;
regd_buf->reg.va = sig_sge.addr;
regd_buf->reg.len = sig_sge.length;
- regd_buf->reg.is_mr = 1;
} else {
- if (desc) {
+ if (desc)
regd_buf->reg.rkey = desc->data_mr->rkey;
- regd_buf->reg.is_mr = 1;
- } else {
+ else
regd_buf->reg.rkey = device->mr->rkey;
- regd_buf->reg.is_mr = 0;
- }
regd_buf->reg.lkey = data_sge.lkey;
regd_buf->reg.va = data_sge.addr;
diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c
index 67225bb..695a270 100644
--- a/drivers/infiniband/ulp/iser/iser_verbs.c
+++ b/drivers/infiniband/ulp/iser/iser_verbs.c
@@ -76,7 +76,7 @@
static int iser_create_device_ib_res(struct iser_device *device)
{
struct ib_device_attr *dev_attr = &device->dev_attr;
- int ret, i;
+ int ret, i, max_cqe;
ret = ib_query_device(device->ib_device, dev_attr);
if (ret) {
@@ -104,11 +104,19 @@
return -1;
}
- device->comps_used = min(ISER_MAX_CQ,
+ device->comps_used = min_t(int, num_online_cpus(),
device->ib_device->num_comp_vectors);
- iser_info("using %d CQs, device %s supports %d vectors\n",
+
+ device->comps = kcalloc(device->comps_used, sizeof(*device->comps),
+ GFP_KERNEL);
+ if (!device->comps)
+ goto comps_err;
+
+ max_cqe = min(ISER_MAX_CQ_LEN, dev_attr->max_cqe);
+
+ iser_info("using %d CQs, device %s supports %d vectors max_cqe %d\n",
device->comps_used, device->ib_device->name,
- device->ib_device->num_comp_vectors);
+ device->ib_device->num_comp_vectors, max_cqe);
device->pd = ib_alloc_pd(device->ib_device);
if (IS_ERR(device->pd))
@@ -122,7 +130,7 @@
iser_cq_callback,
iser_cq_event_callback,
(void *)comp,
- ISER_MAX_CQ_LEN, i);
+ max_cqe, i);
if (IS_ERR(comp->cq)) {
comp->cq = NULL;
goto cq_err;
@@ -162,6 +170,8 @@
}
ib_dealloc_pd(device->pd);
pd_err:
+ kfree(device->comps);
+comps_err:
iser_err("failed to allocate an IB resource\n");
return -1;
}
@@ -187,6 +197,9 @@
(void)ib_dereg_mr(device->mr);
(void)ib_dealloc_pd(device->pd);
+ kfree(device->comps);
+ device->comps = NULL;
+
device->mr = NULL;
device->pd = NULL;
}
@@ -425,7 +438,10 @@
*/
static int iser_create_ib_conn_res(struct ib_conn *ib_conn)
{
+ struct iser_conn *iser_conn = container_of(ib_conn, struct iser_conn,
+ ib_conn);
struct iser_device *device;
+ struct ib_device_attr *dev_attr;
struct ib_qp_init_attr init_attr;
int ret = -ENOMEM;
int index, min_index = 0;
@@ -433,6 +449,7 @@
BUG_ON(ib_conn->device == NULL);
device = ib_conn->device;
+ dev_attr = &device->dev_attr;
memset(&init_attr, 0, sizeof init_attr);
@@ -460,8 +477,20 @@
if (ib_conn->pi_support) {
init_attr.cap.max_send_wr = ISER_QP_SIG_MAX_REQ_DTOS + 1;
init_attr.create_flags |= IB_QP_CREATE_SIGNATURE_EN;
+ iser_conn->max_cmds =
+ ISER_GET_MAX_XMIT_CMDS(ISER_QP_SIG_MAX_REQ_DTOS);
} else {
- init_attr.cap.max_send_wr = ISER_QP_MAX_REQ_DTOS + 1;
+ if (dev_attr->max_qp_wr > ISER_QP_MAX_REQ_DTOS) {
+ init_attr.cap.max_send_wr = ISER_QP_MAX_REQ_DTOS + 1;
+ iser_conn->max_cmds =
+ ISER_GET_MAX_XMIT_CMDS(ISER_QP_MAX_REQ_DTOS);
+ } else {
+ init_attr.cap.max_send_wr = dev_attr->max_qp_wr;
+ iser_conn->max_cmds =
+ ISER_GET_MAX_XMIT_CMDS(dev_attr->max_qp_wr);
+ iser_dbg("device %s supports max_send_wr %d\n",
+ device->ib_device->name, dev_attr->max_qp_wr);
+ }
}
ret = rdma_create_qp(ib_conn->cma_id, device->pd, &init_attr);
@@ -475,7 +504,11 @@
return ret;
out_err:
+ mutex_lock(&ig.connlist_mutex);
+ ib_conn->comp->active_qps--;
+ mutex_unlock(&ig.connlist_mutex);
iser_err("unable to alloc mem or create resource, err %d\n", ret);
+
return ret;
}
@@ -610,9 +643,11 @@
mutex_unlock(&ig.connlist_mutex);
mutex_lock(&iser_conn->state_mutex);
- if (iser_conn->state != ISER_CONN_DOWN)
+ if (iser_conn->state != ISER_CONN_DOWN) {
iser_warn("iser conn %p state %d, expected state down.\n",
iser_conn, iser_conn->state);
+ iser_conn->state = ISER_CONN_DOWN;
+ }
/*
* In case we never got to bind stage, we still need to
* release IB resources (which is safe to call more than once).
@@ -662,8 +697,10 @@
/* post an indication that all flush errors were consumed */
err = ib_post_send(ib_conn->qp, &ib_conn->beacon, &bad_wr);
- if (err)
+ if (err) {
iser_err("conn %p failed to post beacon", ib_conn);
+ return 1;
+ }
wait_for_completion(&ib_conn->flush_comp);
}
@@ -846,20 +883,21 @@
break;
case RDMA_CM_EVENT_DISCONNECTED:
case RDMA_CM_EVENT_ADDR_CHANGE:
- iser_disconnected_handler(cma_id);
+ case RDMA_CM_EVENT_TIMEWAIT_EXIT:
+ iser_cleanup_handler(cma_id, false);
break;
case RDMA_CM_EVENT_DEVICE_REMOVAL:
/*
* we *must* destroy the device as we cannot rely
* on iscsid to be around to initiate error handling.
- * also implicitly destroy the cma_id.
+ * also if we are not in state DOWN implicitly destroy
+ * the cma_id.
*/
iser_cleanup_handler(cma_id, true);
- iser_conn->ib_conn.cma_id = NULL;
- ret = 1;
- break;
- case RDMA_CM_EVENT_TIMEWAIT_EXIT:
- iser_cleanup_handler(cma_id, false);
+ if (iser_conn->state != ISER_CONN_DOWN) {
+ iser_conn->ib_conn.cma_id = NULL;
+ ret = 1;
+ }
break;
default:
iser_err("Unexpected RDMA CM event (%d)\n", event->event);
@@ -981,7 +1019,6 @@
mem_reg->rkey = mem->fmr->rkey;
mem_reg->len = page_vec->length * SIZE_4K;
mem_reg->va = io_addr;
- mem_reg->is_mr = 1;
mem_reg->mem_h = (void *)mem;
mem_reg->va += page_vec->offset;
@@ -1008,7 +1045,7 @@
struct iser_mem_reg *reg = &iser_task->rdma_regd[cmd_dir].reg;
int ret;
- if (!reg->is_mr)
+ if (!reg->mem_h)
return;
iser_dbg("PHYSICAL Mem.Unregister mem_h %p\n",reg->mem_h);
@@ -1028,11 +1065,10 @@
struct ib_conn *ib_conn = &iser_conn->ib_conn;
struct fast_reg_descriptor *desc = reg->mem_h;
- if (!reg->is_mr)
+ if (!desc)
return;
reg->mem_h = NULL;
- reg->is_mr = 0;
spin_lock_bh(&ib_conn->lock);
list_add_tail(&desc->list, &ib_conn->fastreg.pool);
spin_unlock_bh(&ib_conn->lock);
@@ -1049,7 +1085,7 @@
sge.length = ISER_RX_LOGIN_SIZE;
sge.lkey = ib_conn->device->mr->lkey;
- rx_wr.wr_id = (unsigned long)iser_conn->login_resp_buf;
+ rx_wr.wr_id = (uintptr_t)iser_conn->login_resp_buf;
rx_wr.sg_list = &sge;
rx_wr.num_sge = 1;
rx_wr.next = NULL;
@@ -1073,7 +1109,7 @@
for (rx_wr = ib_conn->rx_wr, i = 0; i < count; i++, rx_wr++) {
rx_desc = &iser_conn->rx_descs[my_rx_head];
- rx_wr->wr_id = (unsigned long)rx_desc;
+ rx_wr->wr_id = (uintptr_t)rx_desc;
rx_wr->sg_list = &rx_desc->rx_sg;
rx_wr->num_sge = 1;
rx_wr->next = rx_wr + 1;
@@ -1110,7 +1146,7 @@
DMA_TO_DEVICE);
send_wr.next = NULL;
- send_wr.wr_id = (unsigned long)tx_desc;
+ send_wr.wr_id = (uintptr_t)tx_desc;
send_wr.sg_list = tx_desc->tx_sg;
send_wr.num_sge = tx_desc->num_sge;
send_wr.opcode = IB_WR_SEND;
@@ -1160,6 +1196,7 @@
iser_handle_comp_error(struct ib_conn *ib_conn,
struct ib_wc *wc)
{
+ void *wr_id = (void *)(uintptr_t)wc->wr_id;
struct iser_conn *iser_conn = container_of(ib_conn, struct iser_conn,
ib_conn);
@@ -1168,8 +1205,8 @@
iscsi_conn_failure(iser_conn->iscsi_conn,
ISCSI_ERR_CONN_FAILED);
- if (is_iser_tx_desc(iser_conn, (void *)wc->wr_id)) {
- struct iser_tx_desc *desc = (struct iser_tx_desc *)wc->wr_id;
+ if (is_iser_tx_desc(iser_conn, wr_id)) {
+ struct iser_tx_desc *desc = wr_id;
if (desc->type == ISCSI_TX_DATAOUT)
kmem_cache_free(ig.desc_cache, desc);
@@ -1193,14 +1230,14 @@
struct iser_rx_desc *rx_desc;
ib_conn = wc->qp->qp_context;
- if (wc->status == IB_WC_SUCCESS) {
+ if (likely(wc->status == IB_WC_SUCCESS)) {
if (wc->opcode == IB_WC_RECV) {
- rx_desc = (struct iser_rx_desc *)wc->wr_id;
+ rx_desc = (struct iser_rx_desc *)(uintptr_t)wc->wr_id;
iser_rcv_completion(rx_desc, wc->byte_len,
ib_conn);
} else
if (wc->opcode == IB_WC_SEND) {
- tx_desc = (struct iser_tx_desc *)wc->wr_id;
+ tx_desc = (struct iser_tx_desc *)(uintptr_t)wc->wr_id;
iser_snd_completion(tx_desc, ib_conn);
} else {
iser_err("Unknown wc opcode %d\n", wc->opcode);
diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c
index 5461924..db3c8c8 100644
--- a/drivers/infiniband/ulp/srp/ib_srp.c
+++ b/drivers/infiniband/ulp/srp/ib_srp.c
@@ -2929,7 +2929,7 @@
return -ENOMEM;
sep_opt = options;
- while ((p = strsep(&sep_opt, ",")) != NULL) {
+ while ((p = strsep(&sep_opt, ",\n")) != NULL) {
if (!*p)
continue;
diff --git a/drivers/leds/leds-gpio.c b/drivers/leds/leds-gpio.c
index 8a8ba11..7ea1ea42 100644
--- a/drivers/leds/leds-gpio.c
+++ b/drivers/leds/leds-gpio.c
@@ -203,7 +203,7 @@
fwnode_property_read_string(child, "linux,default-trigger",
&led.default_trigger);
- if (!fwnode_property_read_string(child, "linux,default_state",
+ if (!fwnode_property_read_string(child, "default-state",
&state)) {
if (!strcmp(state, "keep"))
led.default_state = LEDS_GPIO_DEFSTATE_KEEP;
diff --git a/drivers/media/Kconfig b/drivers/media/Kconfig
index 3c89fcb..49cd308 100644
--- a/drivers/media/Kconfig
+++ b/drivers/media/Kconfig
@@ -160,7 +160,6 @@
source "drivers/media/pci/Kconfig"
source "drivers/media/platform/Kconfig"
source "drivers/media/mmc/Kconfig"
-source "drivers/media/parport/Kconfig"
source "drivers/media/radio/Kconfig"
comment "Supported FireWire (IEEE 1394) Adapters"
diff --git a/drivers/media/Makefile b/drivers/media/Makefile
index 620f275..e608bbc 100644
--- a/drivers/media/Makefile
+++ b/drivers/media/Makefile
@@ -28,6 +28,6 @@
# Finally, merge the drivers that require the core
#
-obj-y += common/ platform/ pci/ usb/ mmc/ firewire/ parport/
+obj-y += common/ platform/ pci/ usb/ mmc/ firewire/
obj-$(CONFIG_VIDEO_DEV) += radio/
diff --git a/drivers/media/i2c/Kconfig b/drivers/media/i2c/Kconfig
index f40b4cf..205d713 100644
--- a/drivers/media/i2c/Kconfig
+++ b/drivers/media/i2c/Kconfig
@@ -284,15 +284,6 @@
To compile this driver as a module, choose M here: the
module will be called saa7115.
-config VIDEO_SAA7191
- tristate "Philips SAA7191 video decoder"
- depends on VIDEO_V4L2 && I2C
- ---help---
- Support for the Philips SAA7191 video decoder.
-
- To compile this driver as a module, choose M here: the
- module will be called saa7191.
-
config VIDEO_TVP514X
tristate "Texas Instruments TVP514x video decoder"
depends on VIDEO_V4L2 && I2C
diff --git a/drivers/media/i2c/Makefile b/drivers/media/i2c/Makefile
index 01ae932..98589001 100644
--- a/drivers/media/i2c/Makefile
+++ b/drivers/media/i2c/Makefile
@@ -18,7 +18,6 @@
obj-$(CONFIG_VIDEO_SAA717X) += saa717x.o
obj-$(CONFIG_VIDEO_SAA7127) += saa7127.o
obj-$(CONFIG_VIDEO_SAA7185) += saa7185.o
-obj-$(CONFIG_VIDEO_SAA7191) += saa7191.o
obj-$(CONFIG_VIDEO_SAA6752HS) += saa6752hs.o
obj-$(CONFIG_VIDEO_ADV7170) += adv7170.o
obj-$(CONFIG_VIDEO_ADV7175) += adv7175.o
diff --git a/drivers/media/pci/cx88/cx88-blackbird.c b/drivers/media/pci/cx88/cx88-blackbird.c
index 4160ca4..d3c79d9 100644
--- a/drivers/media/pci/cx88/cx88-blackbird.c
+++ b/drivers/media/pci/cx88/cx88-blackbird.c
@@ -647,6 +647,7 @@
dev->ts_packet_size = 188 * 4;
dev->ts_packet_count = 32;
sizes[0] = dev->ts_packet_size * dev->ts_packet_count;
+ alloc_ctxs[0] = dev->alloc_ctx;
return 0;
}
@@ -662,14 +663,11 @@
{
struct cx8802_dev *dev = vb->vb2_queue->drv_priv;
struct cx88_buffer *buf = container_of(vb, struct cx88_buffer, vb);
- struct sg_table *sgt = vb2_dma_sg_plane_desc(vb, 0);
struct cx88_riscmem *risc = &buf->risc;
if (risc->cpu)
pci_free_consistent(dev->pci, risc->size, risc->cpu, risc->dma);
memset(risc, 0, sizeof(*risc));
-
- dma_unmap_sg(&dev->pci->dev, sgt->sgl, sgt->nents, DMA_FROM_DEVICE);
}
static void buffer_queue(struct vb2_buffer *vb)
diff --git a/drivers/media/pci/cx88/cx88-dvb.c b/drivers/media/pci/cx88/cx88-dvb.c
index c344bfd..5780e2f 100644
--- a/drivers/media/pci/cx88/cx88-dvb.c
+++ b/drivers/media/pci/cx88/cx88-dvb.c
@@ -92,6 +92,7 @@
dev->ts_packet_size = 188 * 4;
dev->ts_packet_count = dvb_buf_tscnt;
sizes[0] = dev->ts_packet_size * dev->ts_packet_count;
+ alloc_ctxs[0] = dev->alloc_ctx;
*num_buffers = dvb_buf_tscnt;
return 0;
}
@@ -108,14 +109,11 @@
{
struct cx8802_dev *dev = vb->vb2_queue->drv_priv;
struct cx88_buffer *buf = container_of(vb, struct cx88_buffer, vb);
- struct sg_table *sgt = vb2_dma_sg_plane_desc(vb, 0);
struct cx88_riscmem *risc = &buf->risc;
if (risc->cpu)
pci_free_consistent(dev->pci, risc->size, risc->cpu, risc->dma);
memset(risc, 0, sizeof(*risc));
-
- dma_unmap_sg(&dev->pci->dev, sgt->sgl, sgt->nents, DMA_FROM_DEVICE);
}
static void buffer_queue(struct vb2_buffer *vb)
diff --git a/drivers/media/pci/cx88/cx88-mpeg.c b/drivers/media/pci/cx88/cx88-mpeg.c
index f181a3a..1c1f69e 100644
--- a/drivers/media/pci/cx88/cx88-mpeg.c
+++ b/drivers/media/pci/cx88/cx88-mpeg.c
@@ -235,10 +235,6 @@
return -EINVAL;
vb2_set_plane_payload(&buf->vb, 0, size);
- rc = dma_map_sg(&dev->pci->dev, sgt->sgl, sgt->nents, DMA_FROM_DEVICE);
- if (!rc)
- return -EIO;
-
rc = cx88_risc_databuffer(dev->pci, risc, sgt->sgl,
dev->ts_packet_size, dev->ts_packet_count, 0);
if (rc) {
@@ -733,6 +729,11 @@
if (NULL == dev)
goto fail_core;
dev->pci = pci_dev;
+ dev->alloc_ctx = vb2_dma_sg_init_ctx(&pci_dev->dev);
+ if (IS_ERR(dev->alloc_ctx)) {
+ err = PTR_ERR(dev->alloc_ctx);
+ goto fail_core;
+ }
dev->core = core;
/* Maintain a reference so cx88-video can query the 8802 device. */
@@ -752,6 +753,7 @@
return 0;
fail_free:
+ vb2_dma_sg_cleanup_ctx(dev->alloc_ctx);
kfree(dev);
fail_core:
core->dvbdev = NULL;
@@ -798,6 +800,7 @@
/* common */
cx8802_fini_common(dev);
cx88_core_put(dev->core,dev->pci);
+ vb2_dma_sg_cleanup_ctx(dev->alloc_ctx);
kfree(dev);
}
diff --git a/drivers/media/pci/cx88/cx88-vbi.c b/drivers/media/pci/cx88/cx88-vbi.c
index 6ab6e27..32eb7fd 100644
--- a/drivers/media/pci/cx88/cx88-vbi.c
+++ b/drivers/media/pci/cx88/cx88-vbi.c
@@ -120,6 +120,7 @@
sizes[0] = VBI_LINE_NTSC_COUNT * VBI_LINE_LENGTH * 2;
else
sizes[0] = VBI_LINE_PAL_COUNT * VBI_LINE_LENGTH * 2;
+ alloc_ctxs[0] = dev->alloc_ctx;
return 0;
}
@@ -131,7 +132,6 @@
struct sg_table *sgt = vb2_dma_sg_plane_desc(vb, 0);
unsigned int lines;
unsigned int size;
- int rc;
if (dev->core->tvnorm & V4L2_STD_525_60)
lines = VBI_LINE_NTSC_COUNT;
@@ -142,10 +142,6 @@
return -EINVAL;
vb2_set_plane_payload(vb, 0, size);
- rc = dma_map_sg(&dev->pci->dev, sgt->sgl, sgt->nents, DMA_FROM_DEVICE);
- if (!rc)
- return -EIO;
-
cx88_risc_buffer(dev->pci, &buf->risc, sgt->sgl,
0, VBI_LINE_LENGTH * lines,
VBI_LINE_LENGTH, 0,
@@ -157,14 +153,11 @@
{
struct cx8800_dev *dev = vb->vb2_queue->drv_priv;
struct cx88_buffer *buf = container_of(vb, struct cx88_buffer, vb);
- struct sg_table *sgt = vb2_dma_sg_plane_desc(vb, 0);
struct cx88_riscmem *risc = &buf->risc;
if (risc->cpu)
pci_free_consistent(dev->pci, risc->size, risc->cpu, risc->dma);
memset(risc, 0, sizeof(*risc));
-
- dma_unmap_sg(&dev->pci->dev, sgt->sgl, sgt->nents, DMA_FROM_DEVICE);
}
static void buffer_queue(struct vb2_buffer *vb)
diff --git a/drivers/media/pci/cx88/cx88-video.c b/drivers/media/pci/cx88/cx88-video.c
index a64ae31..860c98fc 100644
--- a/drivers/media/pci/cx88/cx88-video.c
+++ b/drivers/media/pci/cx88/cx88-video.c
@@ -440,6 +440,7 @@
*num_planes = 1;
sizes[0] = (dev->fmt->depth * core->width * core->height) >> 3;
+ alloc_ctxs[0] = dev->alloc_ctx;
return 0;
}
@@ -449,7 +450,6 @@
struct cx88_core *core = dev->core;
struct cx88_buffer *buf = container_of(vb, struct cx88_buffer, vb);
struct sg_table *sgt = vb2_dma_sg_plane_desc(vb, 0);
- int rc;
buf->bpl = core->width * dev->fmt->depth >> 3;
@@ -457,10 +457,6 @@
return -EINVAL;
vb2_set_plane_payload(vb, 0, core->height * buf->bpl);
- rc = dma_map_sg(&dev->pci->dev, sgt->sgl, sgt->nents, DMA_FROM_DEVICE);
- if (!rc)
- return -EIO;
-
switch (core->field) {
case V4L2_FIELD_TOP:
cx88_risc_buffer(dev->pci, &buf->risc,
@@ -505,14 +501,11 @@
{
struct cx8800_dev *dev = vb->vb2_queue->drv_priv;
struct cx88_buffer *buf = container_of(vb, struct cx88_buffer, vb);
- struct sg_table *sgt = vb2_dma_sg_plane_desc(vb, 0);
struct cx88_riscmem *risc = &buf->risc;
if (risc->cpu)
pci_free_consistent(dev->pci, risc->size, risc->cpu, risc->dma);
memset(risc, 0, sizeof(*risc));
-
- dma_unmap_sg(&dev->pci->dev, sgt->sgl, sgt->nents, DMA_FROM_DEVICE);
}
static void buffer_queue(struct vb2_buffer *vb)
@@ -530,7 +523,6 @@
if (list_empty(&q->active)) {
list_add_tail(&buf->list, &q->active);
- start_video_dma(dev, q, buf);
buf->count = q->count++;
dprintk(2,"[%p/%d] buffer_queue - first active\n",
buf, buf->vb.v4l2_buf.index);
@@ -1345,6 +1337,12 @@
err = -EIO;
goto fail_core;
}
+ dev->alloc_ctx = vb2_dma_sg_init_ctx(&pci_dev->dev);
+ if (IS_ERR(dev->alloc_ctx)) {
+ err = PTR_ERR(dev->alloc_ctx);
+ goto fail_core;
+ }
+
/* initialize driver struct */
spin_lock_init(&dev->slock);
@@ -1549,6 +1547,7 @@
free_irq(pci_dev->irq, dev);
mutex_unlock(&core->lock);
fail_core:
+ vb2_dma_sg_cleanup_ctx(dev->alloc_ctx);
core->v4ldev = NULL;
cx88_core_put(core,dev->pci);
fail_free:
@@ -1582,6 +1581,7 @@
/* free memory */
cx88_core_put(core,dev->pci);
+ vb2_dma_sg_cleanup_ctx(dev->alloc_ctx);
kfree(dev);
}
diff --git a/drivers/media/pci/cx88/cx88.h b/drivers/media/pci/cx88/cx88.h
index 3b0ae75..7748ca9 100644
--- a/drivers/media/pci/cx88/cx88.h
+++ b/drivers/media/pci/cx88/cx88.h
@@ -485,6 +485,7 @@
/* pci i/o */
struct pci_dev *pci;
unsigned char pci_rev,pci_lat;
+ void *alloc_ctx;
const struct cx8800_fmt *fmt;
@@ -548,6 +549,7 @@
/* pci i/o */
struct pci_dev *pci;
unsigned char pci_rev,pci_lat;
+ void *alloc_ctx;
/* dma queues */
struct cx88_dmaqueue mpegq;
diff --git a/drivers/media/platform/Kconfig b/drivers/media/platform/Kconfig
index 0c61155..765bffb 100644
--- a/drivers/media/platform/Kconfig
+++ b/drivers/media/platform/Kconfig
@@ -65,14 +65,6 @@
---help---
Add support for the Video In peripherial of the timberdale FPGA.
-config VIDEO_VINO
- tristate "SGI Vino Video For Linux"
- depends on I2C && SGI_IP22 && VIDEO_V4L2
- select VIDEO_SAA7191 if MEDIA_SUBDRV_AUTOSELECT
- help
- Say Y here to build in support for the Vino video input system found
- on SGI Indy machines.
-
config VIDEO_M32R_AR
tristate "AR devices"
depends on VIDEO_V4L2
@@ -112,7 +104,7 @@
config VIDEO_S3C_CAMIF
tristate "Samsung S3C24XX/S3C64XX SoC Camera Interface driver"
depends on VIDEO_V4L2 && I2C && VIDEO_V4L2_SUBDEV_API
- depends on PM_RUNTIME
+ depends on PM
depends on ARCH_S3C64XX || PLAT_S3C24XX || COMPILE_TEST
depends on HAS_DMA
select VIDEOBUF2_DMA_CONTIG
diff --git a/drivers/media/platform/Makefile b/drivers/media/platform/Makefile
index b818afb..a49936b 100644
--- a/drivers/media/platform/Makefile
+++ b/drivers/media/platform/Makefile
@@ -2,9 +2,6 @@
# Makefile for the video capture/playback device drivers.
#
-obj-$(CONFIG_VIDEO_VINO) += indycam.o
-obj-$(CONFIG_VIDEO_VINO) += vino.o
-
obj-$(CONFIG_VIDEO_TIMBERDALE) += timblogiw.o
obj-$(CONFIG_VIDEO_M32R_AR_M64278) += arv.o
diff --git a/drivers/media/platform/s5p-tv/Kconfig b/drivers/media/platform/s5p-tv/Kconfig
index beb180e..5a1835d 100644
--- a/drivers/media/platform/s5p-tv/Kconfig
+++ b/drivers/media/platform/s5p-tv/Kconfig
@@ -8,7 +8,7 @@
config VIDEO_SAMSUNG_S5P_TV
bool "Samsung TV driver for S5P platform"
- depends on PM_RUNTIME
+ depends on PM
depends on ARCH_S5PV210 || ARCH_EXYNOS || COMPILE_TEST
default n
---help---
diff --git a/drivers/media/platform/soc_camera/rcar_vin.c b/drivers/media/platform/soc_camera/rcar_vin.c
index 126ac7c..0c1f556 100644
--- a/drivers/media/platform/soc_camera/rcar_vin.c
+++ b/drivers/media/platform/soc_camera/rcar_vin.c
@@ -64,6 +64,30 @@
#define VNDMR_REG 0x58 /* Video n Data Mode Register */
#define VNDMR2_REG 0x5C /* Video n Data Mode Register 2 */
#define VNUVAOF_REG 0x60 /* Video n UV Address Offset Register */
+#define VNC1A_REG 0x80 /* Video n Coefficient Set C1A Register */
+#define VNC1B_REG 0x84 /* Video n Coefficient Set C1B Register */
+#define VNC1C_REG 0x88 /* Video n Coefficient Set C1C Register */
+#define VNC2A_REG 0x90 /* Video n Coefficient Set C2A Register */
+#define VNC2B_REG 0x94 /* Video n Coefficient Set C2B Register */
+#define VNC2C_REG 0x98 /* Video n Coefficient Set C2C Register */
+#define VNC3A_REG 0xA0 /* Video n Coefficient Set C3A Register */
+#define VNC3B_REG 0xA4 /* Video n Coefficient Set C3B Register */
+#define VNC3C_REG 0xA8 /* Video n Coefficient Set C3C Register */
+#define VNC4A_REG 0xB0 /* Video n Coefficient Set C4A Register */
+#define VNC4B_REG 0xB4 /* Video n Coefficient Set C4B Register */
+#define VNC4C_REG 0xB8 /* Video n Coefficient Set C4C Register */
+#define VNC5A_REG 0xC0 /* Video n Coefficient Set C5A Register */
+#define VNC5B_REG 0xC4 /* Video n Coefficient Set C5B Register */
+#define VNC5C_REG 0xC8 /* Video n Coefficient Set C5C Register */
+#define VNC6A_REG 0xD0 /* Video n Coefficient Set C6A Register */
+#define VNC6B_REG 0xD4 /* Video n Coefficient Set C6B Register */
+#define VNC6C_REG 0xD8 /* Video n Coefficient Set C6C Register */
+#define VNC7A_REG 0xE0 /* Video n Coefficient Set C7A Register */
+#define VNC7B_REG 0xE4 /* Video n Coefficient Set C7B Register */
+#define VNC7C_REG 0xE8 /* Video n Coefficient Set C7C Register */
+#define VNC8A_REG 0xF0 /* Video n Coefficient Set C8A Register */
+#define VNC8B_REG 0xF4 /* Video n Coefficient Set C8B Register */
+#define VNC8C_REG 0xF8 /* Video n Coefficient Set C8C Register */
/* Register bit fields for R-Car VIN */
/* Video n Main Control Register bits */
@@ -106,6 +130,7 @@
#define VNDMR2_VPS (1 << 30)
#define VNDMR2_HPS (1 << 29)
#define VNDMR2_FTEV (1 << 17)
+#define VNDMR2_VLV(n) ((n & 0xf) << 12)
#define VIN_MAX_WIDTH 2048
#define VIN_MAX_HEIGHT 2048
@@ -117,6 +142,324 @@
RCAR_E1,
};
+struct vin_coeff {
+ unsigned short xs_value;
+ u32 coeff_set[24];
+};
+
+static const struct vin_coeff vin_coeff_set[] = {
+ { 0x0000, {
+ 0x00000000, 0x00000000, 0x00000000,
+ 0x00000000, 0x00000000, 0x00000000,
+ 0x00000000, 0x00000000, 0x00000000,
+ 0x00000000, 0x00000000, 0x00000000,
+ 0x00000000, 0x00000000, 0x00000000,
+ 0x00000000, 0x00000000, 0x00000000,
+ 0x00000000, 0x00000000, 0x00000000,
+ 0x00000000, 0x00000000, 0x00000000 },
+ },
+ { 0x1000, {
+ 0x000fa400, 0x000fa400, 0x09625902,
+ 0x000003f8, 0x00000403, 0x3de0d9f0,
+ 0x001fffed, 0x00000804, 0x3cc1f9c3,
+ 0x001003de, 0x00000c01, 0x3cb34d7f,
+ 0x002003d2, 0x00000c00, 0x3d24a92d,
+ 0x00200bca, 0x00000bff, 0x3df600d2,
+ 0x002013cc, 0x000007ff, 0x3ed70c7e,
+ 0x00100fde, 0x00000000, 0x3f87c036 },
+ },
+ { 0x1200, {
+ 0x002ffff1, 0x002ffff1, 0x02a0a9c8,
+ 0x002003e7, 0x001ffffa, 0x000185bc,
+ 0x002007dc, 0x000003ff, 0x3e52859c,
+ 0x00200bd4, 0x00000002, 0x3d53996b,
+ 0x00100fd0, 0x00000403, 0x3d04ad2d,
+ 0x00000bd5, 0x00000403, 0x3d35ace7,
+ 0x3ff003e4, 0x00000801, 0x3dc674a1,
+ 0x3fffe800, 0x00000800, 0x3e76f461 },
+ },
+ { 0x1400, {
+ 0x00100be3, 0x00100be3, 0x04d1359a,
+ 0x00000fdb, 0x002003ed, 0x0211fd93,
+ 0x00000fd6, 0x002003f4, 0x0002d97b,
+ 0x000007d6, 0x002ffffb, 0x3e93b956,
+ 0x3ff003da, 0x001003ff, 0x3db49926,
+ 0x3fffefe9, 0x00100001, 0x3d655cee,
+ 0x3fffd400, 0x00000003, 0x3d65f4b6,
+ 0x000fb421, 0x00000402, 0x3dc6547e },
+ },
+ { 0x1600, {
+ 0x00000bdd, 0x00000bdd, 0x06519578,
+ 0x3ff007da, 0x00000be3, 0x03c24973,
+ 0x3ff003d9, 0x00000be9, 0x01b30d5f,
+ 0x3ffff7df, 0x001003f1, 0x0003c542,
+ 0x000fdfec, 0x001003f7, 0x3ec4711d,
+ 0x000fc400, 0x002ffffd, 0x3df504f1,
+ 0x001fa81a, 0x002ffc00, 0x3d957cc2,
+ 0x002f8c3c, 0x00100000, 0x3db5c891 },
+ },
+ { 0x1800, {
+ 0x3ff003dc, 0x3ff003dc, 0x0791e558,
+ 0x000ff7dd, 0x3ff007de, 0x05328554,
+ 0x000fe7e3, 0x3ff00be2, 0x03232546,
+ 0x000fd7ee, 0x000007e9, 0x0143bd30,
+ 0x001fb800, 0x000007ee, 0x00044511,
+ 0x002fa015, 0x000007f4, 0x3ef4bcee,
+ 0x002f8832, 0x001003f9, 0x3e4514c7,
+ 0x001f7853, 0x001003fd, 0x3de54c9f },
+ },
+ { 0x1a00, {
+ 0x000fefe0, 0x000fefe0, 0x08721d3c,
+ 0x001fdbe7, 0x000ffbde, 0x0652a139,
+ 0x001fcbf0, 0x000003df, 0x0463292e,
+ 0x002fb3ff, 0x3ff007e3, 0x0293a91d,
+ 0x002f9c12, 0x3ff00be7, 0x01241905,
+ 0x001f8c29, 0x000007ed, 0x3fe470eb,
+ 0x000f7c46, 0x000007f2, 0x3f04b8ca,
+ 0x3fef7865, 0x000007f6, 0x3e74e4a8 },
+ },
+ { 0x1c00, {
+ 0x001fd3e9, 0x001fd3e9, 0x08f23d26,
+ 0x002fbff3, 0x001fe3e4, 0x0712ad23,
+ 0x002fa800, 0x000ff3e0, 0x05631d1b,
+ 0x001f9810, 0x000ffbe1, 0x03b3890d,
+ 0x000f8c23, 0x000003e3, 0x0233e8fa,
+ 0x3fef843b, 0x000003e7, 0x00f430e4,
+ 0x3fbf8456, 0x3ff00bea, 0x00046cc8,
+ 0x3f8f8c72, 0x3ff00bef, 0x3f3490ac },
+ },
+ { 0x1e00, {
+ 0x001fbbf4, 0x001fbbf4, 0x09425112,
+ 0x001fa800, 0x002fc7ed, 0x0792b110,
+ 0x000f980e, 0x001fdbe6, 0x0613110a,
+ 0x3fff8c20, 0x001fe7e3, 0x04a368fd,
+ 0x3fcf8c33, 0x000ff7e2, 0x0343b8ed,
+ 0x3f9f8c4a, 0x000fffe3, 0x0203f8da,
+ 0x3f5f9c61, 0x000003e6, 0x00e428c5,
+ 0x3f1fb07b, 0x000003eb, 0x3fe440af },
+ },
+ { 0x2000, {
+ 0x000fa400, 0x000fa400, 0x09625902,
+ 0x3fff980c, 0x001fb7f5, 0x0812b0ff,
+ 0x3fdf901c, 0x001fc7ed, 0x06b2fcfa,
+ 0x3faf902d, 0x001fd3e8, 0x055348f1,
+ 0x3f7f983f, 0x001fe3e5, 0x04038ce3,
+ 0x3f3fa454, 0x001fefe3, 0x02e3c8d1,
+ 0x3f0fb86a, 0x001ff7e4, 0x01c3e8c0,
+ 0x3ecfd880, 0x000fffe6, 0x00c404ac },
+ },
+ { 0x2200, {
+ 0x3fdf9c0b, 0x3fdf9c0b, 0x09725cf4,
+ 0x3fbf9818, 0x3fffa400, 0x0842a8f1,
+ 0x3f8f9827, 0x000fb3f7, 0x0702f0ec,
+ 0x3f5fa037, 0x000fc3ef, 0x05d330e4,
+ 0x3f2fac49, 0x001fcfea, 0x04a364d9,
+ 0x3effc05c, 0x001fdbe7, 0x038394ca,
+ 0x3ecfdc6f, 0x001fe7e6, 0x0273b0bb,
+ 0x3ea00083, 0x001fefe6, 0x0183c0a9 },
+ },
+ { 0x2400, {
+ 0x3f9fa014, 0x3f9fa014, 0x098260e6,
+ 0x3f7f9c23, 0x3fcf9c0a, 0x08629ce5,
+ 0x3f4fa431, 0x3fefa400, 0x0742d8e1,
+ 0x3f1fb440, 0x3fffb3f8, 0x062310d9,
+ 0x3eefc850, 0x000fbbf2, 0x050340d0,
+ 0x3ecfe062, 0x000fcbec, 0x041364c2,
+ 0x3ea00073, 0x001fd3ea, 0x03037cb5,
+ 0x3e902086, 0x001fdfe8, 0x022388a5 },
+ },
+ { 0x2600, {
+ 0x3f5fa81e, 0x3f5fa81e, 0x096258da,
+ 0x3f3fac2b, 0x3f8fa412, 0x088290d8,
+ 0x3f0fbc38, 0x3fafa408, 0x0772c8d5,
+ 0x3eefcc47, 0x3fcfa800, 0x0672f4ce,
+ 0x3ecfe456, 0x3fefaffa, 0x05531cc6,
+ 0x3eb00066, 0x3fffbbf3, 0x047334bb,
+ 0x3ea01c77, 0x000fc7ee, 0x039348ae,
+ 0x3ea04486, 0x000fd3eb, 0x02b350a1 },
+ },
+ { 0x2800, {
+ 0x3f2fb426, 0x3f2fb426, 0x094250ce,
+ 0x3f0fc032, 0x3f4fac1b, 0x086284cd,
+ 0x3eefd040, 0x3f7fa811, 0x0782acc9,
+ 0x3ecfe84c, 0x3f9fa807, 0x06a2d8c4,
+ 0x3eb0005b, 0x3fbfac00, 0x05b2f4bc,
+ 0x3eb0186a, 0x3fdfb3fa, 0x04c308b4,
+ 0x3eb04077, 0x3fefbbf4, 0x03f31ca8,
+ 0x3ec06884, 0x000fbff2, 0x03031c9e },
+ },
+ { 0x2a00, {
+ 0x3f0fc42d, 0x3f0fc42d, 0x090240c4,
+ 0x3eefd439, 0x3f2fb822, 0x08526cc2,
+ 0x3edfe845, 0x3f4fb018, 0x078294bf,
+ 0x3ec00051, 0x3f6fac0f, 0x06b2b4bb,
+ 0x3ec0185f, 0x3f8fac07, 0x05e2ccb4,
+ 0x3ec0386b, 0x3fafac00, 0x0502e8ac,
+ 0x3ed05c77, 0x3fcfb3fb, 0x0432f0a3,
+ 0x3ef08482, 0x3fdfbbf6, 0x0372f898 },
+ },
+ { 0x2c00, {
+ 0x3eefdc31, 0x3eefdc31, 0x08e238b8,
+ 0x3edfec3d, 0x3f0fc828, 0x082258b9,
+ 0x3ed00049, 0x3f1fc01e, 0x077278b6,
+ 0x3ed01455, 0x3f3fb815, 0x06c294b2,
+ 0x3ed03460, 0x3f5fb40d, 0x0602acac,
+ 0x3ef0506c, 0x3f7fb006, 0x0542c0a4,
+ 0x3f107476, 0x3f9fb400, 0x0472c89d,
+ 0x3f309c80, 0x3fbfb7fc, 0x03b2cc94 },
+ },
+ { 0x2e00, {
+ 0x3eefec37, 0x3eefec37, 0x088220b0,
+ 0x3ee00041, 0x3effdc2d, 0x07f244ae,
+ 0x3ee0144c, 0x3f0fd023, 0x07625cad,
+ 0x3ef02c57, 0x3f1fc81a, 0x06c274a9,
+ 0x3f004861, 0x3f3fbc13, 0x060288a6,
+ 0x3f20686b, 0x3f5fb80c, 0x05529c9e,
+ 0x3f408c74, 0x3f6fb805, 0x04b2ac96,
+ 0x3f80ac7e, 0x3f8fb800, 0x0402ac8e },
+ },
+ { 0x3000, {
+ 0x3ef0003a, 0x3ef0003a, 0x084210a6,
+ 0x3ef01045, 0x3effec32, 0x07b228a7,
+ 0x3f00284e, 0x3f0fdc29, 0x073244a4,
+ 0x3f104058, 0x3f0fd420, 0x06a258a2,
+ 0x3f305c62, 0x3f2fc818, 0x0612689d,
+ 0x3f508069, 0x3f3fc011, 0x05728496,
+ 0x3f80a072, 0x3f4fc00a, 0x04d28c90,
+ 0x3fc0c07b, 0x3f6fbc04, 0x04429088 },
+ },
+ { 0x3200, {
+ 0x3f00103e, 0x3f00103e, 0x07f1fc9e,
+ 0x3f102447, 0x3f000035, 0x0782149d,
+ 0x3f203c4f, 0x3f0ff02c, 0x07122c9c,
+ 0x3f405458, 0x3f0fe424, 0x06924099,
+ 0x3f607061, 0x3f1fd41d, 0x06024c97,
+ 0x3f909068, 0x3f2fcc16, 0x05726490,
+ 0x3fc0b070, 0x3f3fc80f, 0x04f26c8a,
+ 0x0000d077, 0x3f4fc409, 0x04627484 },
+ },
+ { 0x3400, {
+ 0x3f202040, 0x3f202040, 0x07a1e898,
+ 0x3f303449, 0x3f100c38, 0x0741fc98,
+ 0x3f504c50, 0x3f10002f, 0x06e21495,
+ 0x3f706459, 0x3f1ff028, 0x06722492,
+ 0x3fa08060, 0x3f1fe421, 0x05f2348f,
+ 0x3fd09c67, 0x3f1fdc19, 0x05824c89,
+ 0x0000bc6e, 0x3f2fd014, 0x04f25086,
+ 0x0040dc74, 0x3f3fcc0d, 0x04825c7f },
+ },
+ { 0x3600, {
+ 0x3f403042, 0x3f403042, 0x0761d890,
+ 0x3f504848, 0x3f301c3b, 0x0701f090,
+ 0x3f805c50, 0x3f200c33, 0x06a2008f,
+ 0x3fa07458, 0x3f10002b, 0x06520c8d,
+ 0x3fd0905e, 0x3f1ff424, 0x05e22089,
+ 0x0000ac65, 0x3f1fe81d, 0x05823483,
+ 0x0030cc6a, 0x3f2fdc18, 0x04f23c81,
+ 0x0080e871, 0x3f2fd412, 0x0482407c },
+ },
+ { 0x3800, {
+ 0x3f604043, 0x3f604043, 0x0721c88a,
+ 0x3f80544a, 0x3f502c3c, 0x06d1d88a,
+ 0x3fb06851, 0x3f301c35, 0x0681e889,
+ 0x3fd08456, 0x3f30082f, 0x0611fc88,
+ 0x00009c5d, 0x3f200027, 0x05d20884,
+ 0x0030b863, 0x3f2ff421, 0x05621880,
+ 0x0070d468, 0x3f2fe81b, 0x0502247c,
+ 0x00c0ec6f, 0x3f2fe015, 0x04a22877 },
+ },
+ { 0x3a00, {
+ 0x3f904c44, 0x3f904c44, 0x06e1b884,
+ 0x3fb0604a, 0x3f70383e, 0x0691c885,
+ 0x3fe07451, 0x3f502c36, 0x0661d483,
+ 0x00009055, 0x3f401831, 0x0601ec81,
+ 0x0030a85b, 0x3f300c2a, 0x05b1f480,
+ 0x0070c061, 0x3f300024, 0x0562047a,
+ 0x00b0d867, 0x3f3ff41e, 0x05020c77,
+ 0x00f0f46b, 0x3f2fec19, 0x04a21474 },
+ },
+ { 0x3c00, {
+ 0x3fb05c43, 0x3fb05c43, 0x06c1b07e,
+ 0x3fe06c4b, 0x3f902c3f, 0x0681c081,
+ 0x0000844f, 0x3f703838, 0x0631cc7d,
+ 0x00309855, 0x3f602433, 0x05d1d47e,
+ 0x0060b459, 0x3f50142e, 0x0581e47b,
+ 0x00a0c85f, 0x3f400828, 0x0531f078,
+ 0x00e0e064, 0x3f300021, 0x0501fc73,
+ 0x00b0fc6a, 0x3f3ff41d, 0x04a20873 },
+ },
+ { 0x3e00, {
+ 0x3fe06444, 0x3fe06444, 0x0681a07a,
+ 0x00007849, 0x3fc0503f, 0x0641b07a,
+ 0x0020904d, 0x3fa0403a, 0x05f1c07a,
+ 0x0060a453, 0x3f803034, 0x05c1c878,
+ 0x0090b858, 0x3f70202f, 0x0571d477,
+ 0x00d0d05d, 0x3f501829, 0x0531e073,
+ 0x0110e462, 0x3f500825, 0x04e1e471,
+ 0x01510065, 0x3f40001f, 0x04a1f06d },
+ },
+ { 0x4000, {
+ 0x00007044, 0x00007044, 0x06519476,
+ 0x00208448, 0x3fe05c3f, 0x0621a476,
+ 0x0050984d, 0x3fc04c3a, 0x05e1b075,
+ 0x0080ac52, 0x3fa03c35, 0x05a1b875,
+ 0x00c0c056, 0x3f803030, 0x0561c473,
+ 0x0100d45b, 0x3f70202b, 0x0521d46f,
+ 0x0140e860, 0x3f601427, 0x04d1d46e,
+ 0x01810064, 0x3f500822, 0x0491dc6b },
+ },
+ { 0x5000, {
+ 0x0110a442, 0x0110a442, 0x0551545e,
+ 0x0140b045, 0x00e0983f, 0x0531585f,
+ 0x0160c047, 0x00c08c3c, 0x0511645e,
+ 0x0190cc4a, 0x00908039, 0x04f1685f,
+ 0x01c0dc4c, 0x00707436, 0x04d1705e,
+ 0x0200e850, 0x00506833, 0x04b1785b,
+ 0x0230f453, 0x00305c30, 0x0491805a,
+ 0x02710056, 0x0010542d, 0x04718059 },
+ },
+ { 0x6000, {
+ 0x01c0bc40, 0x01c0bc40, 0x04c13052,
+ 0x01e0c841, 0x01a0b43d, 0x04c13851,
+ 0x0210cc44, 0x0180a83c, 0x04a13453,
+ 0x0230d845, 0x0160a03a, 0x04913c52,
+ 0x0260e047, 0x01409838, 0x04714052,
+ 0x0280ec49, 0x01208c37, 0x04514c50,
+ 0x02b0f44b, 0x01008435, 0x04414c50,
+ 0x02d1004c, 0x00e07c33, 0x0431544f },
+ },
+ { 0x7000, {
+ 0x0230c83e, 0x0230c83e, 0x04711c4c,
+ 0x0250d03f, 0x0210c43c, 0x0471204b,
+ 0x0270d840, 0x0200b83c, 0x0451244b,
+ 0x0290dc42, 0x01e0b43a, 0x0441244c,
+ 0x02b0e443, 0x01c0b038, 0x0441284b,
+ 0x02d0ec44, 0x01b0a438, 0x0421304a,
+ 0x02f0f445, 0x0190a036, 0x04213449,
+ 0x0310f847, 0x01709c34, 0x04213848 },
+ },
+ { 0x8000, {
+ 0x0280d03d, 0x0280d03d, 0x04310c48,
+ 0x02a0d43e, 0x0270c83c, 0x04311047,
+ 0x02b0dc3e, 0x0250c83a, 0x04311447,
+ 0x02d0e040, 0x0240c03a, 0x04211446,
+ 0x02e0e840, 0x0220bc39, 0x04111847,
+ 0x0300e842, 0x0210b438, 0x04012445,
+ 0x0310f043, 0x0200b037, 0x04012045,
+ 0x0330f444, 0x01e0ac36, 0x03f12445 },
+ },
+ { 0xefff, {
+ 0x0340dc3a, 0x0340dc3a, 0x03b0ec40,
+ 0x0340e03a, 0x0330e039, 0x03c0f03e,
+ 0x0350e03b, 0x0330dc39, 0x03c0ec3e,
+ 0x0350e43a, 0x0320dc38, 0x03c0f43e,
+ 0x0360e43b, 0x0320d839, 0x03b0f03e,
+ 0x0360e83b, 0x0310d838, 0x03c0fc3b,
+ 0x0370e83b, 0x0310d439, 0x03a0f83d,
+ 0x0370e83c, 0x0300d438, 0x03b0fc3c },
+ }
+};
+
enum rcar_vin_state {
STOPPED = 0,
RUNNING,
@@ -161,6 +504,9 @@
/* Client output, as seen by the VIN */
unsigned int width;
unsigned int height;
+ /* User window from S_FMT */
+ unsigned int out_width;
+ unsigned int out_height;
/*
* User window from S_CROP / G_CROP, produced by client cropping and
* scaling, VIN scaling and VIN cropping, mapped back onto the client
@@ -332,7 +678,7 @@
vnmc |= VNMC_BPS;
/* progressive or interlaced mode */
- interrupts = progressive ? VNIE_FIE | VNIE_EFE : VNIE_EFE;
+ interrupts = progressive ? VNIE_FIE : VNIE_EFE;
/* ack interrupts */
iowrite32(interrupts, priv->base + VNINTS_REG);
@@ -667,6 +1013,60 @@
/* VIN does not have "mclk" */
}
+static void set_coeff(struct rcar_vin_priv *priv, unsigned short xs)
+{
+ int i;
+ const struct vin_coeff *p_prev_set = NULL;
+ const struct vin_coeff *p_set = NULL;
+
+ /* Look for suitable coefficient values */
+ for (i = 0; i < ARRAY_SIZE(vin_coeff_set); i++) {
+ p_prev_set = p_set;
+ p_set = &vin_coeff_set[i];
+
+ if (xs < p_set->xs_value)
+ break;
+ }
+
+ /* Use previous value if its XS value is closer */
+ if (p_prev_set && p_set &&
+ xs - p_prev_set->xs_value < p_set->xs_value - xs)
+ p_set = p_prev_set;
+
+ /* Set coefficient registers */
+ iowrite32(p_set->coeff_set[0], priv->base + VNC1A_REG);
+ iowrite32(p_set->coeff_set[1], priv->base + VNC1B_REG);
+ iowrite32(p_set->coeff_set[2], priv->base + VNC1C_REG);
+
+ iowrite32(p_set->coeff_set[3], priv->base + VNC2A_REG);
+ iowrite32(p_set->coeff_set[4], priv->base + VNC2B_REG);
+ iowrite32(p_set->coeff_set[5], priv->base + VNC2C_REG);
+
+ iowrite32(p_set->coeff_set[6], priv->base + VNC3A_REG);
+ iowrite32(p_set->coeff_set[7], priv->base + VNC3B_REG);
+ iowrite32(p_set->coeff_set[8], priv->base + VNC3C_REG);
+
+ iowrite32(p_set->coeff_set[9], priv->base + VNC4A_REG);
+ iowrite32(p_set->coeff_set[10], priv->base + VNC4B_REG);
+ iowrite32(p_set->coeff_set[11], priv->base + VNC4C_REG);
+
+ iowrite32(p_set->coeff_set[12], priv->base + VNC5A_REG);
+ iowrite32(p_set->coeff_set[13], priv->base + VNC5B_REG);
+ iowrite32(p_set->coeff_set[14], priv->base + VNC5C_REG);
+
+ iowrite32(p_set->coeff_set[15], priv->base + VNC6A_REG);
+ iowrite32(p_set->coeff_set[16], priv->base + VNC6B_REG);
+ iowrite32(p_set->coeff_set[17], priv->base + VNC6C_REG);
+
+ iowrite32(p_set->coeff_set[18], priv->base + VNC7A_REG);
+ iowrite32(p_set->coeff_set[19], priv->base + VNC7B_REG);
+ iowrite32(p_set->coeff_set[20], priv->base + VNC7C_REG);
+
+ iowrite32(p_set->coeff_set[21], priv->base + VNC8A_REG);
+ iowrite32(p_set->coeff_set[22], priv->base + VNC8B_REG);
+ iowrite32(p_set->coeff_set[23], priv->base + VNC8C_REG);
+}
+
/* rect is guaranteed to not exceed the scaled camera rectangle */
static int rcar_vin_set_rect(struct soc_camera_device *icd)
{
@@ -676,6 +1076,7 @@
unsigned int left_offset, top_offset;
unsigned char dsize = 0;
struct v4l2_rect *cam_subrect = &cam->subrect;
+ u32 value;
dev_dbg(icd->parent, "Crop %ux%u@%u:%u\n",
icd->user_width, icd->user_height, cam->vin_left, cam->vin_top);
@@ -695,40 +1096,64 @@
/* Set Start/End Pixel/Line Pre-Clip */
iowrite32(left_offset << dsize, priv->base + VNSPPRC_REG);
- iowrite32((left_offset + cam->width - 1) << dsize,
+ iowrite32((left_offset + cam_subrect->width - 1) << dsize,
priv->base + VNEPPRC_REG);
switch (priv->field) {
case V4L2_FIELD_INTERLACED:
case V4L2_FIELD_INTERLACED_TB:
case V4L2_FIELD_INTERLACED_BT:
iowrite32(top_offset / 2, priv->base + VNSLPRC_REG);
- iowrite32((top_offset + cam->height) / 2 - 1,
+ iowrite32((top_offset + cam_subrect->height) / 2 - 1,
priv->base + VNELPRC_REG);
break;
default:
iowrite32(top_offset, priv->base + VNSLPRC_REG);
- iowrite32(top_offset + cam->height - 1,
+ iowrite32(top_offset + cam_subrect->height - 1,
priv->base + VNELPRC_REG);
break;
}
+ /* Set scaling coefficient */
+ value = 0;
+ if (cam_subrect->height != cam->out_height)
+ value = (4096 * cam_subrect->height) / cam->out_height;
+ dev_dbg(icd->parent, "YS Value: %x\n", value);
+ iowrite32(value, priv->base + VNYS_REG);
+
+ value = 0;
+ if (cam_subrect->width != cam->out_width)
+ value = (4096 * cam_subrect->width) / cam->out_width;
+
+ /* Horizontal upscaling is up to double size */
+ if (0 < value && value < 2048)
+ value = 2048;
+
+ dev_dbg(icd->parent, "XS Value: %x\n", value);
+ iowrite32(value, priv->base + VNXS_REG);
+
+ /* Horizontal upscaling is carried out by scaling down from double size */
+ if (value < 4096)
+ value *= 2;
+
+ set_coeff(priv, value);
+
/* Set Start/End Pixel/Line Post-Clip */
iowrite32(0, priv->base + VNSPPOC_REG);
iowrite32(0, priv->base + VNSLPOC_REG);
- iowrite32((cam_subrect->width - 1) << dsize, priv->base + VNEPPOC_REG);
+ iowrite32((cam->out_width - 1) << dsize, priv->base + VNEPPOC_REG);
switch (priv->field) {
case V4L2_FIELD_INTERLACED:
case V4L2_FIELD_INTERLACED_TB:
case V4L2_FIELD_INTERLACED_BT:
- iowrite32(cam_subrect->height / 2 - 1,
+ iowrite32(cam->out_height / 2 - 1,
priv->base + VNELPOC_REG);
break;
default:
- iowrite32(cam_subrect->height - 1, priv->base + VNELPOC_REG);
+ iowrite32(cam->out_height - 1, priv->base + VNELPOC_REG);
break;
}
- iowrite32(ALIGN(cam->width, 0x10), priv->base + VNIS_REG);
+ iowrite32(ALIGN(cam->out_width, 0x10), priv->base + VNIS_REG);
return 0;
}
@@ -819,7 +1244,7 @@
if (ret < 0 && ret != -ENOIOCTLCMD)
return ret;
- val = priv->field == V4L2_FIELD_NONE ? VNDMR2_FTEV : 0;
+ val = VNDMR2_FTEV | VNDMR2_VLV(1);
if (!(common_flags & V4L2_MBUS_VSYNC_ACTIVE_LOW))
val |= VNDMR2_VPS;
if (!(common_flags & V4L2_MBUS_HSYNC_ACTIVE_LOW))
@@ -880,6 +1305,14 @@
.layout = SOC_MBUS_LAYOUT_PLANAR_Y_C,
},
{
+ .fourcc = V4L2_PIX_FMT_YUYV,
+ .name = "YUYV",
+ .bits_per_sample = 16,
+ .packing = SOC_MBUS_PACKING_NONE,
+ .order = SOC_MBUS_ORDER_LE,
+ .layout = SOC_MBUS_LAYOUT_PACKED,
+ },
+ {
.fourcc = V4L2_PIX_FMT_UYVY,
.name = "UYVY",
.bits_per_sample = 16,
@@ -999,6 +1432,8 @@
cam->subrect = rect;
cam->width = mf.width;
cam->height = mf.height;
+ cam->out_width = mf.width;
+ cam->out_height = mf.height;
icd->host_priv = cam;
} else {
@@ -1259,6 +1694,9 @@
dev_dbg(dev, "W: %u : %u, H: %u : %u\n",
vin_sub_width, pix->width, vin_sub_height, pix->height);
+ cam->out_width = pix->width;
+ cam->out_height = pix->height;
+
icd->current_fmt = xlate;
priv->field = field;
@@ -1310,8 +1748,12 @@
if (ret < 0)
return ret;
- pix->width = mf.width;
- pix->height = mf.height;
+ /* Adjust only if VIN cannot scale */
+ if (pix->width > mf.width * 2)
+ pix->width = mf.width * 2;
+ if (pix->height > mf.height * 3)
+ pix->height = mf.height * 3;
+
pix->field = mf.field;
pix->colorspace = mf.colorspace;
@@ -1395,6 +1837,8 @@
#ifdef CONFIG_OF
static struct of_device_id rcar_vin_of_table[] = {
+ { .compatible = "renesas,vin-r8a7794", .data = (void *)RCAR_GEN2 },
+ { .compatible = "renesas,vin-r8a7793", .data = (void *)RCAR_GEN2 },
{ .compatible = "renesas,vin-r8a7791", .data = (void *)RCAR_GEN2 },
{ .compatible = "renesas,vin-r8a7790", .data = (void *)RCAR_GEN2 },
{ .compatible = "renesas,vin-r8a7779", .data = (void *)RCAR_H1 },
diff --git a/drivers/media/platform/vivid/vivid-vid-out.c b/drivers/media/platform/vivid/vivid-vid-out.c
index ee5c399..39ff79f 100644
--- a/drivers/media/platform/vivid/vivid-vid-out.c
+++ b/drivers/media/platform/vivid/vivid-vid-out.c
@@ -625,7 +625,7 @@
sel->r = dev->fmt_out_rect;
break;
case V4L2_SEL_TGT_CROP_BOUNDS:
- if (!dev->has_compose_out)
+ if (!dev->has_crop_out)
return -EINVAL;
sel->r = vivid_max_rect;
break;
diff --git a/drivers/media/usb/Kconfig b/drivers/media/usb/Kconfig
index 056181f..7496f33 100644
--- a/drivers/media/usb/Kconfig
+++ b/drivers/media/usb/Kconfig
@@ -24,7 +24,6 @@
comment "Analog TV USB devices"
source "drivers/media/usb/pvrusb2/Kconfig"
source "drivers/media/usb/hdpvr/Kconfig"
-source "drivers/media/usb/tlg2300/Kconfig"
source "drivers/media/usb/usbvision/Kconfig"
source "drivers/media/usb/stk1160/Kconfig"
source "drivers/media/usb/go7007/Kconfig"
diff --git a/drivers/media/usb/Makefile b/drivers/media/usb/Makefile
index 6f2eb7c..8874ba7 100644
--- a/drivers/media/usb/Makefile
+++ b/drivers/media/usb/Makefile
@@ -16,7 +16,6 @@
obj-$(CONFIG_VIDEO_AU0828) += au0828/
obj-$(CONFIG_VIDEO_HDPVR) += hdpvr/
obj-$(CONFIG_VIDEO_PVRUSB2) += pvrusb2/
-obj-$(CONFIG_VIDEO_TLG2300) += tlg2300/
obj-$(CONFIG_VIDEO_USBVISION) += usbvision/
obj-$(CONFIG_VIDEO_STK1160) += stk1160/
obj-$(CONFIG_VIDEO_CX231XX) += cx231xx/
diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
index 7565871..faac2f4 100644
--- a/drivers/media/v4l2-core/v4l2-ioctl.c
+++ b/drivers/media/v4l2-core/v4l2-ioctl.c
@@ -1017,6 +1017,12 @@
ret = ops->vidioc_querycap(file, fh, cap);
cap->capabilities |= V4L2_CAP_EXT_PIX_FORMAT;
+ /*
+ * Drivers MUST fill in device_caps, so check for this and
+ * warn if it was forgotten.
+ */
+ WARN_ON(!(cap->capabilities & V4L2_CAP_DEVICE_CAPS) ||
+ !cap->device_caps);
cap->device_caps |= V4L2_CAP_EXT_PIX_FORMAT;
return ret;
diff --git a/drivers/misc/mic/host/mic_debugfs.c b/drivers/misc/mic/host/mic_debugfs.c
index 028ba5d6..687e9aa 100644
--- a/drivers/misc/mic/host/mic_debugfs.c
+++ b/drivers/misc/mic/host/mic_debugfs.c
@@ -326,21 +326,27 @@
}
avail = vrh->vring.avail;
seq_printf(s, "avail flags 0x%x idx %d\n",
- avail->flags, avail->idx & (num - 1));
+ vringh16_to_cpu(vrh, avail->flags),
+ vringh16_to_cpu(vrh, avail->idx) & (num - 1));
seq_printf(s, "avail flags 0x%x idx %d\n",
- avail->flags, avail->idx);
+ vringh16_to_cpu(vrh, avail->flags),
+ vringh16_to_cpu(vrh, avail->idx));
for (j = 0; j < num; j++)
seq_printf(s, "avail ring[%d] %d\n",
j, avail->ring[j]);
used = vrh->vring.used;
seq_printf(s, "used flags 0x%x idx %d\n",
- used->flags, used->idx & (num - 1));
+ vringh16_to_cpu(vrh, used->flags),
+ vringh16_to_cpu(vrh, used->idx) & (num - 1));
seq_printf(s, "used flags 0x%x idx %d\n",
- used->flags, used->idx);
+ vringh16_to_cpu(vrh, used->flags),
+ vringh16_to_cpu(vrh, used->idx));
for (j = 0; j < num; j++)
seq_printf(s, "used ring[%d] id %d len %d\n",
- j, used->ring[j].id,
- used->ring[j].len);
+ j, vringh32_to_cpu(vrh,
+ used->ring[j].id),
+ vringh32_to_cpu(vrh,
+ used->ring[j].len));
}
}
mutex_unlock(&mdev->mic_mutex);
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 62aba9a..03d7c75 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -2561,7 +2561,7 @@
static const struct dev_pm_ops atmci_dev_pm_ops = {
SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend,
pm_runtime_force_resume)
- SET_PM_RUNTIME_PM_OPS(atmci_runtime_suspend, atmci_runtime_resume, NULL)
+ SET_RUNTIME_PM_OPS(atmci_runtime_suspend, atmci_runtime_resume, NULL)
};
static struct platform_driver atmci_driver = {
diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c
index 868d0f6..705f334 100644
--- a/drivers/net/ethernet/cisco/enic/enic_main.c
+++ b/drivers/net/ethernet/cisco/enic/enic_main.c
@@ -1060,10 +1060,14 @@
PKT_HASH_TYPE_L4 : PKT_HASH_TYPE_L3);
}
- if ((netdev->features & NETIF_F_RXCSUM) && !csum_not_calc) {
- skb->csum = htons(checksum);
- skb->ip_summed = CHECKSUM_COMPLETE;
- }
+ /* Hardware does not provide whole packet checksum. It only
+ * provides pseudo checksum. Since hw validates the packet
+ * checksum but not provide us the checksum value. use
+ * CHECSUM_UNNECESSARY.
+ */
+ if ((netdev->features & NETIF_F_RXCSUM) && tcp_udp_csum_ok &&
+ ipv4_csum_ok)
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
if (vlan_stripped)
__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tci);
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index b935bf3..943cbd4 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -171,9 +171,9 @@
{
int i;
- for (i = 0; i < dev->caps.num_ports - 1; i++) {
- if (port_type[i] != port_type[i + 1]) {
- if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_DPDP)) {
+ if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_DPDP)) {
+ for (i = 0; i < dev->caps.num_ports - 1; i++) {
+ if (port_type[i] != port_type[i + 1]) {
mlx4_err(dev, "Only same port types supported on this HCA, aborting\n");
return -EINVAL;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index ab68446..da82991 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -157,6 +157,8 @@
return "MLX5_EVENT_TYPE_CMD";
case MLX5_EVENT_TYPE_PAGE_REQUEST:
return "MLX5_EVENT_TYPE_PAGE_REQUEST";
+ case MLX5_EVENT_TYPE_PAGE_FAULT:
+ return "MLX5_EVENT_TYPE_PAGE_FAULT";
default:
return "Unrecognized event";
}
@@ -279,6 +281,11 @@
}
break;
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ case MLX5_EVENT_TYPE_PAGE_FAULT:
+ mlx5_eq_pagefault(dev, eqe);
+ break;
+#endif
default:
mlx5_core_warn(dev, "Unhandled event 0x%x on EQ 0x%x\n",
@@ -446,8 +453,12 @@
int mlx5_start_eqs(struct mlx5_core_dev *dev)
{
struct mlx5_eq_table *table = &dev->priv.eq_table;
+ u32 async_event_mask = MLX5_ASYNC_EVENT_MASK;
int err;
+ if (dev->caps.gen.flags & MLX5_DEV_CAP_FLAG_ON_DMND_PG)
+ async_event_mask |= (1ull << MLX5_EVENT_TYPE_PAGE_FAULT);
+
err = mlx5_create_map_eq(dev, &table->cmd_eq, MLX5_EQ_VEC_CMD,
MLX5_NUM_CMD_EQE, 1ull << MLX5_EVENT_TYPE_CMD,
"mlx5_cmd_eq", &dev->priv.uuari.uars[0]);
@@ -459,7 +470,7 @@
mlx5_cmd_use_events(dev);
err = mlx5_create_map_eq(dev, &table->async_eq, MLX5_EQ_VEC_ASYNC,
- MLX5_NUM_ASYNC_EQE, MLX5_ASYNC_EVENT_MASK,
+ MLX5_NUM_ASYNC_EQE, async_event_mask,
"mlx5_async_eq", &dev->priv.uuari.uars[0]);
if (err) {
mlx5_core_warn(dev, "failed to create async EQ %d\n", err);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
index 087c4c7..06f9036 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
@@ -69,6 +69,46 @@
return mlx5_core_get_caps(dev, caps, HCA_CAP_OPMOD_GET_CUR);
}
+int mlx5_query_odp_caps(struct mlx5_core_dev *dev, struct mlx5_odp_caps *caps)
+{
+ u8 in[MLX5_ST_SZ_BYTES(query_hca_cap_in)];
+ int out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out);
+ void *out;
+ int err;
+
+ if (!(dev->caps.gen.flags & MLX5_DEV_CAP_FLAG_ON_DMND_PG))
+ return -ENOTSUPP;
+
+ memset(in, 0, sizeof(in));
+ out = kzalloc(out_sz, GFP_KERNEL);
+ if (!out)
+ return -ENOMEM;
+ MLX5_SET(query_hca_cap_in, in, opcode, MLX5_CMD_OP_QUERY_HCA_CAP);
+ MLX5_SET(query_hca_cap_in, in, op_mod, HCA_CAP_OPMOD_GET_ODP_CUR);
+ err = mlx5_cmd_exec(dev, in, sizeof(in), out, out_sz);
+ if (err)
+ goto out;
+
+ err = mlx5_cmd_status_to_err_v2(out);
+ if (err) {
+ mlx5_core_warn(dev, "query cur hca ODP caps failed, %d\n", err);
+ goto out;
+ }
+
+ memcpy(caps, MLX5_ADDR_OF(query_hca_cap_out, out, capability_struct),
+ sizeof(*caps));
+
+ mlx5_core_dbg(dev, "on-demand paging capabilities:\nrc: %08x\nuc: %08x\nud: %08x\n",
+ be32_to_cpu(caps->per_transport_caps.rc_odp_caps),
+ be32_to_cpu(caps->per_transport_caps.uc_odp_caps),
+ be32_to_cpu(caps->per_transport_caps.ud_odp_caps));
+
+out:
+ kfree(out);
+ return err;
+}
+EXPORT_SYMBOL(mlx5_query_odp_caps);
+
int mlx5_cmd_init_hca(struct mlx5_core_dev *dev)
{
struct mlx5_cmd_init_hca_mbox_in in;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/qp.c b/drivers/net/ethernet/mellanox/mlx5/core/qp.c
index 5261a2b..575d853 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/qp.c
@@ -88,6 +88,95 @@
mlx5_core_put_rsc(common);
}
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+void mlx5_eq_pagefault(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe)
+{
+ struct mlx5_eqe_page_fault *pf_eqe = &eqe->data.page_fault;
+ int qpn = be32_to_cpu(pf_eqe->flags_qpn) & MLX5_QPN_MASK;
+ struct mlx5_core_rsc_common *common = mlx5_get_rsc(dev, qpn);
+ struct mlx5_core_qp *qp =
+ container_of(common, struct mlx5_core_qp, common);
+ struct mlx5_pagefault pfault;
+
+ if (!qp) {
+ mlx5_core_warn(dev, "ODP event for non-existent QP %06x\n",
+ qpn);
+ return;
+ }
+
+ pfault.event_subtype = eqe->sub_type;
+ pfault.flags = (be32_to_cpu(pf_eqe->flags_qpn) >> MLX5_QPN_BITS) &
+ (MLX5_PFAULT_REQUESTOR | MLX5_PFAULT_WRITE | MLX5_PFAULT_RDMA);
+ pfault.bytes_committed = be32_to_cpu(
+ pf_eqe->bytes_committed);
+
+ mlx5_core_dbg(dev,
+ "PAGE_FAULT: subtype: 0x%02x, flags: 0x%02x,\n",
+ eqe->sub_type, pfault.flags);
+
+ switch (eqe->sub_type) {
+ case MLX5_PFAULT_SUBTYPE_RDMA:
+ /* RDMA based event */
+ pfault.rdma.r_key =
+ be32_to_cpu(pf_eqe->rdma.r_key);
+ pfault.rdma.packet_size =
+ be16_to_cpu(pf_eqe->rdma.packet_length);
+ pfault.rdma.rdma_op_len =
+ be32_to_cpu(pf_eqe->rdma.rdma_op_len);
+ pfault.rdma.rdma_va =
+ be64_to_cpu(pf_eqe->rdma.rdma_va);
+ mlx5_core_dbg(dev,
+ "PAGE_FAULT: qpn: 0x%06x, r_key: 0x%08x,\n",
+ qpn, pfault.rdma.r_key);
+ mlx5_core_dbg(dev,
+ "PAGE_FAULT: rdma_op_len: 0x%08x,\n",
+ pfault.rdma.rdma_op_len);
+ mlx5_core_dbg(dev,
+ "PAGE_FAULT: rdma_va: 0x%016llx,\n",
+ pfault.rdma.rdma_va);
+ mlx5_core_dbg(dev,
+ "PAGE_FAULT: bytes_committed: 0x%06x\n",
+ pfault.bytes_committed);
+ break;
+
+ case MLX5_PFAULT_SUBTYPE_WQE:
+ /* WQE based event */
+ pfault.wqe.wqe_index =
+ be16_to_cpu(pf_eqe->wqe.wqe_index);
+ pfault.wqe.packet_size =
+ be16_to_cpu(pf_eqe->wqe.packet_length);
+ mlx5_core_dbg(dev,
+ "PAGE_FAULT: qpn: 0x%06x, wqe_index: 0x%04x,\n",
+ qpn, pfault.wqe.wqe_index);
+ mlx5_core_dbg(dev,
+ "PAGE_FAULT: bytes_committed: 0x%06x\n",
+ pfault.bytes_committed);
+ break;
+
+ default:
+ mlx5_core_warn(dev,
+ "Unsupported page fault event sub-type: 0x%02hhx, QP %06x\n",
+ eqe->sub_type, qpn);
+ /* Unsupported page faults should still be resolved by the
+ * page fault handler
+ */
+ }
+
+ if (qp->pfault_handler) {
+ qp->pfault_handler(qp, &pfault);
+ } else {
+ mlx5_core_err(dev,
+ "ODP event for QP %08x, without a fault handler in QP\n",
+ qpn);
+ /* Page fault will remain unresolved. QP will hang until it is
+ * destroyed
+ */
+ }
+
+ mlx5_core_put_rsc(common);
+}
+#endif
+
int mlx5_core_create_qp(struct mlx5_core_dev *dev,
struct mlx5_core_qp *qp,
struct mlx5_create_qp_mbox_in *in,
@@ -322,3 +411,33 @@
return err;
}
EXPORT_SYMBOL_GPL(mlx5_core_xrcd_dealloc);
+
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+int mlx5_core_page_fault_resume(struct mlx5_core_dev *dev, u32 qpn,
+ u8 flags, int error)
+{
+ struct mlx5_page_fault_resume_mbox_in in;
+ struct mlx5_page_fault_resume_mbox_out out;
+ int err;
+
+ memset(&in, 0, sizeof(in));
+ memset(&out, 0, sizeof(out));
+ in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_PAGE_FAULT_RESUME);
+ in.hdr.opmod = 0;
+ flags &= (MLX5_PAGE_FAULT_RESUME_REQUESTOR |
+ MLX5_PAGE_FAULT_RESUME_WRITE |
+ MLX5_PAGE_FAULT_RESUME_RDMA);
+ flags |= (error ? MLX5_PAGE_FAULT_RESUME_ERROR : 0);
+ in.flags_qpn = cpu_to_be32((qpn & MLX5_QPN_MASK) |
+ (flags << MLX5_QPN_BITS));
+ err = mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out));
+ if (err)
+ return err;
+
+ if (out.hdr.status)
+ err = mlx5_cmd_status_to_err(&out.hdr);
+
+ return err;
+}
+EXPORT_SYMBOL_GPL(mlx5_core_page_fault_resume);
+#endif
diff --git a/drivers/net/ethernet/sun/sunvnet.c b/drivers/net/ethernet/sun/sunvnet.c
index 45c408e..d2835bf 100644
--- a/drivers/net/ethernet/sun/sunvnet.c
+++ b/drivers/net/ethernet/sun/sunvnet.c
@@ -1201,6 +1201,7 @@
segs = skb_gso_segment(skb, dev->features & ~NETIF_F_TSO);
if (IS_ERR(segs)) {
dev->stats.tx_dropped++;
+ dev_kfree_skb_any(skb);
return NETDEV_TX_OK;
}
diff --git a/drivers/nfc/trf7970a.c b/drivers/nfc/trf7970a.c
index d2ccd28..aa6a333 100644
--- a/drivers/nfc/trf7970a.c
+++ b/drivers/nfc/trf7970a.c
@@ -2154,7 +2154,7 @@
}
#endif
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int trf7970a_pm_runtime_suspend(struct device *dev)
{
struct spi_device *spi = container_of(dev, struct spi_device, dev);
diff --git a/drivers/phy/phy-omap-usb2.c b/drivers/phy/phy-omap-usb2.c
index 4e489a8..6f4aef3 100644
--- a/drivers/phy/phy-omap-usb2.c
+++ b/drivers/phy/phy-omap-usb2.c
@@ -318,7 +318,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int omap_usb2_runtime_suspend(struct device *dev)
{
diff --git a/drivers/phy/phy-ti-pipe3.c b/drivers/phy/phy-ti-pipe3.c
index c297b7a..1387b4d 100644
--- a/drivers/phy/phy-ti-pipe3.c
+++ b/drivers/phy/phy-ti-pipe3.c
@@ -423,7 +423,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int ti_pipe3_runtime_suspend(struct device *dev)
{
diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index a2eabe6..638e7970 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -38,7 +38,8 @@
config ACERHDF
tristate "Acer Aspire One temperature and fan driver"
- depends on THERMAL && ACPI
+ depends on ACPI && THERMAL
+ select THERMAL_GOV_BANG_BANG
---help---
This is a driver for Acer Aspire One netbooks. It allows to access
the temperature sensor and to control the fan.
@@ -128,10 +129,10 @@
be called dell-wmi-aio.
config DELL_SMO8800
- tristate "Dell Latitude freefall driver (ACPI SMO8800/SMO8810)"
+ tristate "Dell Latitude freefall driver (ACPI SMO88XX)"
depends on ACPI
---help---
- Say Y here if you want to support SMO8800/SMO8810 freefall device
+ Say Y here if you want to support SMO88XX freefall devices
on Dell Latitude laptops.
To compile this driver as a module, choose M here: the module will
diff --git a/drivers/platform/x86/acerhdf.c b/drivers/platform/x86/acerhdf.c
index aaf37c5..594c918 100644
--- a/drivers/platform/x86/acerhdf.c
+++ b/drivers/platform/x86/acerhdf.c
@@ -50,7 +50,7 @@
*/
#undef START_IN_KERNEL_MODE
-#define DRV_VER "0.5.26"
+#define DRV_VER "0.7.0"
/*
* According to the Atom N270 datasheet,
@@ -119,116 +119,152 @@
u8 cmd_auto;
};
+struct manualcmd {
+ u8 mreg;
+ u8 moff;
+};
+
+/* default register and command to disable fan in manual mode */
+static const struct manualcmd mcmd = {
+ .mreg = 0x94,
+ .moff = 0xff,
+};
+
/* BIOS settings */
-struct bios_settings_t {
+struct bios_settings {
const char *vendor;
const char *product;
const char *version;
- unsigned char fanreg;
- unsigned char tempreg;
+ u8 fanreg;
+ u8 tempreg;
struct fancmd cmd;
+ int mcmd_enable;
};
/* Register addresses and values for different BIOS versions */
-static const struct bios_settings_t bios_tbl[] = {
+static const struct bios_settings bios_tbl[] = {
/* AOA110 */
- {"Acer", "AOA110", "v0.3109", 0x55, 0x58, {0x1f, 0x00} },
- {"Acer", "AOA110", "v0.3114", 0x55, 0x58, {0x1f, 0x00} },
- {"Acer", "AOA110", "v0.3301", 0x55, 0x58, {0xaf, 0x00} },
- {"Acer", "AOA110", "v0.3304", 0x55, 0x58, {0xaf, 0x00} },
- {"Acer", "AOA110", "v0.3305", 0x55, 0x58, {0xaf, 0x00} },
- {"Acer", "AOA110", "v0.3307", 0x55, 0x58, {0xaf, 0x00} },
- {"Acer", "AOA110", "v0.3308", 0x55, 0x58, {0x21, 0x00} },
- {"Acer", "AOA110", "v0.3309", 0x55, 0x58, {0x21, 0x00} },
- {"Acer", "AOA110", "v0.3310", 0x55, 0x58, {0x21, 0x00} },
+ {"Acer", "AOA110", "v0.3109", 0x55, 0x58, {0x1f, 0x00}, 0},
+ {"Acer", "AOA110", "v0.3114", 0x55, 0x58, {0x1f, 0x00}, 0},
+ {"Acer", "AOA110", "v0.3301", 0x55, 0x58, {0xaf, 0x00}, 0},
+ {"Acer", "AOA110", "v0.3304", 0x55, 0x58, {0xaf, 0x00}, 0},
+ {"Acer", "AOA110", "v0.3305", 0x55, 0x58, {0xaf, 0x00}, 0},
+ {"Acer", "AOA110", "v0.3307", 0x55, 0x58, {0xaf, 0x00}, 0},
+ {"Acer", "AOA110", "v0.3308", 0x55, 0x58, {0x21, 0x00}, 0},
+ {"Acer", "AOA110", "v0.3309", 0x55, 0x58, {0x21, 0x00}, 0},
+ {"Acer", "AOA110", "v0.3310", 0x55, 0x58, {0x21, 0x00}, 0},
/* AOA150 */
- {"Acer", "AOA150", "v0.3114", 0x55, 0x58, {0x1f, 0x00} },
- {"Acer", "AOA150", "v0.3301", 0x55, 0x58, {0x20, 0x00} },
- {"Acer", "AOA150", "v0.3304", 0x55, 0x58, {0x20, 0x00} },
- {"Acer", "AOA150", "v0.3305", 0x55, 0x58, {0x20, 0x00} },
- {"Acer", "AOA150", "v0.3307", 0x55, 0x58, {0x20, 0x00} },
- {"Acer", "AOA150", "v0.3308", 0x55, 0x58, {0x20, 0x00} },
- {"Acer", "AOA150", "v0.3309", 0x55, 0x58, {0x20, 0x00} },
- {"Acer", "AOA150", "v0.3310", 0x55, 0x58, {0x20, 0x00} },
+ {"Acer", "AOA150", "v0.3114", 0x55, 0x58, {0x1f, 0x00}, 0},
+ {"Acer", "AOA150", "v0.3301", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Acer", "AOA150", "v0.3304", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Acer", "AOA150", "v0.3305", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Acer", "AOA150", "v0.3307", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Acer", "AOA150", "v0.3308", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Acer", "AOA150", "v0.3309", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Acer", "AOA150", "v0.3310", 0x55, 0x58, {0x20, 0x00}, 0},
/* LT1005u */
- {"Acer", "LT-10Q", "v0.3310", 0x55, 0x58, {0x20, 0x00} },
+ {"Acer", "LT-10Q", "v0.3310", 0x55, 0x58, {0x20, 0x00}, 0},
/* Acer 1410 */
- {"Acer", "Aspire 1410", "v0.3108", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v0.3113", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v0.3115", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v0.3117", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v0.3119", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v0.3120", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v1.3204", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v1.3303", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v1.3308", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v1.3310", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1410", "v1.3314", 0x55, 0x58, {0x9e, 0x00} },
+ {"Acer", "Aspire 1410", "v0.3108", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v0.3113", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v0.3115", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v0.3117", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v0.3119", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v0.3120", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v1.3204", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v1.3303", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v1.3308", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v1.3310", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1410", "v1.3314", 0x55, 0x58, {0x9e, 0x00}, 0},
/* Acer 1810xx */
- {"Acer", "Aspire 1810TZ", "v0.3108", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v0.3108", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v0.3113", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v0.3113", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v0.3115", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v0.3115", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v0.3117", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v0.3117", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v0.3119", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v0.3119", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v0.3120", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v0.3120", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v1.3204", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v1.3204", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v1.3303", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v1.3303", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v1.3308", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v1.3308", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v1.3310", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v1.3310", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810TZ", "v1.3314", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1810T", "v1.3314", 0x55, 0x58, {0x9e, 0x00} },
+ {"Acer", "Aspire 1810TZ", "v0.3108", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v0.3108", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v0.3113", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v0.3113", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v0.3115", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v0.3115", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v0.3117", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v0.3117", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v0.3119", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v0.3119", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v0.3120", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v0.3120", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v1.3204", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v1.3204", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v1.3303", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v1.3303", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v1.3308", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v1.3308", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v1.3310", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v1.3310", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810TZ", "v1.3314", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1810T", "v1.3314", 0x55, 0x58, {0x9e, 0x00}, 0},
+ /* Acer 5755G */
+ {"Acer", "Aspire 5755G", "V1.20", 0xab, 0xb4, {0x00, 0x08}, 0},
+ {"Acer", "Aspire 5755G", "V1.21", 0xab, 0xb3, {0x00, 0x08}, 0},
+ /* Acer 521 */
+ {"Acer", "AO521", "V1.11", 0x55, 0x58, {0x1f, 0x00}, 0},
/* Acer 531 */
- {"Acer", "AO531h", "v0.3104", 0x55, 0x58, {0x20, 0x00} },
- {"Acer", "AO531h", "v0.3201", 0x55, 0x58, {0x20, 0x00} },
- {"Acer", "AO531h", "v0.3304", 0x55, 0x58, {0x20, 0x00} },
+ {"Acer", "AO531h", "v0.3104", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Acer", "AO531h", "v0.3201", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Acer", "AO531h", "v0.3304", 0x55, 0x58, {0x20, 0x00}, 0},
/* Acer 751 */
- {"Acer", "AO751h", "V0.3212", 0x55, 0x58, {0x21, 0x00} },
+ {"Acer", "AO751h", "V0.3206", 0x55, 0x58, {0x21, 0x00}, 0},
+ {"Acer", "AO751h", "V0.3212", 0x55, 0x58, {0x21, 0x00}, 0},
+ /* Acer 753 */
+ {"Acer", "Aspire One 753", "V1.24", 0x93, 0xac, {0x14, 0x04}, 1},
/* Acer 1825 */
- {"Acer", "Aspire 1825PTZ", "V1.3118", 0x55, 0x58, {0x9e, 0x00} },
- {"Acer", "Aspire 1825PTZ", "V1.3127", 0x55, 0x58, {0x9e, 0x00} },
+ {"Acer", "Aspire 1825PTZ", "V1.3118", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Acer", "Aspire 1825PTZ", "V1.3127", 0x55, 0x58, {0x9e, 0x00}, 0},
+ /* Acer Extensa 5420 */
+ {"Acer", "Extensa 5420", "V1.17", 0x93, 0xac, {0x14, 0x04}, 1},
+ /* Acer Aspire 5315 */
+ {"Acer", "Aspire 5315", "V1.19", 0x93, 0xac, {0x14, 0x04}, 1},
+ /* Acer Aspire 5739 */
+ {"Acer", "Aspire 5739G", "V1.3311", 0x55, 0x58, {0x20, 0x00}, 0},
/* Acer TravelMate 7730 */
- {"Acer", "TravelMate 7730G", "v0.3509", 0x55, 0x58, {0xaf, 0x00} },
+ {"Acer", "TravelMate 7730G", "v0.3509", 0x55, 0x58, {0xaf, 0x00}, 0},
+ /* Acer TravelMate TM8573T */
+ {"Acer", "TM8573T", "V1.13", 0x93, 0xa8, {0x14, 0x04}, 1},
/* Gateway */
- {"Gateway", "AOA110", "v0.3103", 0x55, 0x58, {0x21, 0x00} },
- {"Gateway", "AOA150", "v0.3103", 0x55, 0x58, {0x20, 0x00} },
- {"Gateway", "LT31", "v1.3103", 0x55, 0x58, {0x9e, 0x00} },
- {"Gateway", "LT31", "v1.3201", 0x55, 0x58, {0x9e, 0x00} },
- {"Gateway", "LT31", "v1.3302", 0x55, 0x58, {0x9e, 0x00} },
- {"Gateway", "LT31", "v1.3303t", 0x55, 0x58, {0x9e, 0x00} },
+ {"Gateway", "AOA110", "v0.3103", 0x55, 0x58, {0x21, 0x00}, 0},
+ {"Gateway", "AOA150", "v0.3103", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Gateway", "LT31", "v1.3103", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Gateway", "LT31", "v1.3201", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Gateway", "LT31", "v1.3302", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Gateway", "LT31", "v1.3303t", 0x55, 0x58, {0x9e, 0x00}, 0},
/* Packard Bell */
- {"Packard Bell", "DOA150", "v0.3104", 0x55, 0x58, {0x21, 0x00} },
- {"Packard Bell", "DOA150", "v0.3105", 0x55, 0x58, {0x20, 0x00} },
- {"Packard Bell", "AOA110", "v0.3105", 0x55, 0x58, {0x21, 0x00} },
- {"Packard Bell", "AOA150", "v0.3105", 0x55, 0x58, {0x20, 0x00} },
- {"Packard Bell", "ENBFT", "V1.3118", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "ENBFT", "V1.3127", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMU", "v1.3303", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMU", "v0.3120", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMU", "v0.3108", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMU", "v0.3113", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMU", "v0.3115", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMU", "v0.3117", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMU", "v0.3119", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMU", "v1.3204", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMA", "v1.3201", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMA", "v1.3302", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTMA", "v1.3303t", 0x55, 0x58, {0x9e, 0x00} },
- {"Packard Bell", "DOTVR46", "v1.3308", 0x55, 0x58, {0x9e, 0x00} },
+ {"Packard Bell", "DOA150", "v0.3104", 0x55, 0x58, {0x21, 0x00}, 0},
+ {"Packard Bell", "DOA150", "v0.3105", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Packard Bell", "AOA110", "v0.3105", 0x55, 0x58, {0x21, 0x00}, 0},
+ {"Packard Bell", "AOA150", "v0.3105", 0x55, 0x58, {0x20, 0x00}, 0},
+ {"Packard Bell", "ENBFT", "V1.3118", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "ENBFT", "V1.3127", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMU", "v1.3303", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMU", "v0.3120", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMU", "v0.3108", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMU", "v0.3113", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMU", "v0.3115", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMU", "v0.3117", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMU", "v0.3119", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMU", "v1.3204", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMA", "v1.3201", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMA", "v1.3302", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTMA", "v1.3303t", 0x55, 0x58, {0x9e, 0x00}, 0},
+ {"Packard Bell", "DOTVR46", "v1.3308", 0x55, 0x58, {0x9e, 0x00}, 0},
/* pewpew-terminator */
- {"", "", "", 0, 0, {0, 0} }
+ {"", "", "", 0, 0, {0, 0}, 0}
};
-static const struct bios_settings_t *bios_cfg __read_mostly;
+static const struct bios_settings *bios_cfg __read_mostly;
+
+/*
+ * this struct is used to instruct thermal layer to use bang_bang instead of
+ * default governor for acerhdf
+ */
+static struct thermal_zone_params acerhdf_zone_params = {
+ .governor_name = "bang_bang",
+};
static int acerhdf_get_temp(int *temp)
{
@@ -275,6 +311,12 @@
fanstate = state;
ec_write(bios_cfg->fanreg, cmd);
+
+ if (bios_cfg->mcmd_enable && state == ACERHDF_FAN_OFF) {
+ if (verbose)
+ pr_notice("turning off fan manually\n");
+ ec_write(mcmd.mreg, mcmd.moff);
+ }
}
static void acerhdf_check_param(struct thermal_zone_device *thermal)
@@ -401,6 +443,21 @@
{
if (trip == 0)
*type = THERMAL_TRIP_ACTIVE;
+ else if (trip == 1)
+ *type = THERMAL_TRIP_CRITICAL;
+ else
+ return -EINVAL;
+
+ return 0;
+}
+
+static int acerhdf_get_trip_hyst(struct thermal_zone_device *thermal, int trip,
+ unsigned long *temp)
+{
+ if (trip != 0)
+ return -EINVAL;
+
+ *temp = fanon - fanoff;
return 0;
}
@@ -410,6 +467,10 @@
{
if (trip == 0)
*temp = fanon;
+ else if (trip == 1)
+ *temp = ACERHDF_TEMP_CRIT;
+ else
+ return -EINVAL;
return 0;
}
@@ -429,6 +490,7 @@
.get_mode = acerhdf_get_mode,
.set_mode = acerhdf_set_mode,
.get_trip_type = acerhdf_get_trip_type,
+ .get_trip_hyst = acerhdf_get_trip_hyst,
.get_trip_temp = acerhdf_get_trip_temp,
.get_crit_temp = acerhdf_get_crit_temp,
};
@@ -481,9 +543,7 @@
}
if (state == 0) {
- /* turn fan off only if below fanoff temperature */
- if ((cur_state == ACERHDF_FAN_AUTO) &&
- (cur_temp < fanoff))
+ if (cur_state == ACERHDF_FAN_AUTO)
acerhdf_change_fanstate(ACERHDF_FAN_OFF);
} else {
if (cur_state == ACERHDF_FAN_OFF)
@@ -558,7 +618,7 @@
static int acerhdf_check_hardware(void)
{
char const *vendor, *version, *product;
- const struct bios_settings_t *bt = NULL;
+ const struct bios_settings *bt = NULL;
/* get BIOS data */
vendor = dmi_get_system_info(DMI_SYS_VENDOR);
@@ -660,12 +720,20 @@
if (IS_ERR(cl_dev))
return -EINVAL;
- thz_dev = thermal_zone_device_register("acerhdf", 1, 0, NULL,
- &acerhdf_dev_ops, NULL, 0,
+ thz_dev = thermal_zone_device_register("acerhdf", 2, 0, NULL,
+ &acerhdf_dev_ops,
+ &acerhdf_zone_params, 0,
(kernelmode) ? interval*1000 : 0);
if (IS_ERR(thz_dev))
return -EINVAL;
+ if (strcmp(thz_dev->governor->name,
+ acerhdf_zone_params.governor_name)) {
+ pr_err("Didn't get thermal governor %s, perhaps not compiled into thermal subsystem.\n",
+ acerhdf_zone_params.governor_name);
+ return -EINVAL;
+ }
+
return 0;
}
@@ -722,9 +790,15 @@
MODULE_ALIAS("dmi:*:*Acer*:pnAO751h*:");
MODULE_ALIAS("dmi:*:*Acer*:pnAspire*1410*:");
MODULE_ALIAS("dmi:*:*Acer*:pnAspire*1810*:");
+MODULE_ALIAS("dmi:*:*Acer*:pnAspire*5755G:");
MODULE_ALIAS("dmi:*:*Acer*:pnAspire*1825PTZ:");
+MODULE_ALIAS("dmi:*:*Acer*:pnAO521*:");
MODULE_ALIAS("dmi:*:*Acer*:pnAO531*:");
+MODULE_ALIAS("dmi:*:*Acer*:pnAspire*5739G:");
+MODULE_ALIAS("dmi:*:*Acer*:pnAspire*One*753:");
+MODULE_ALIAS("dmi:*:*Acer*:pnAspire*5315:");
MODULE_ALIAS("dmi:*:*Acer*:TravelMate*7730G:");
+MODULE_ALIAS("dmi:*:*Acer*:TM8573T:");
MODULE_ALIAS("dmi:*:*Gateway*:pnAOA*:");
MODULE_ALIAS("dmi:*:*Gateway*:pnLT31*:");
MODULE_ALIAS("dmi:*:*Packard*Bell*:pnAOA*:");
@@ -733,6 +807,7 @@
MODULE_ALIAS("dmi:*:*Packard*Bell*:pnENBFT*:");
MODULE_ALIAS("dmi:*:*Packard*Bell*:pnDOTMA*:");
MODULE_ALIAS("dmi:*:*Packard*Bell*:pnDOTVR46*:");
+MODULE_ALIAS("dmi:*:*Acer*:pnExtensa 5420*:");
module_init(acerhdf_init);
module_exit(acerhdf_exit);
diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
index 05647f1..f71700e 100644
--- a/drivers/platform/x86/asus-laptop.c
+++ b/drivers/platform/x86/asus-laptop.c
@@ -843,8 +843,7 @@
static void asus_backlight_exit(struct asus_laptop *asus)
{
- if (asus->backlight_device)
- backlight_device_unregister(asus->backlight_device);
+ backlight_device_unregister(asus->backlight_device);
asus->backlight_device = NULL;
}
diff --git a/drivers/platform/x86/asus-nb-wmi.c b/drivers/platform/x86/asus-nb-wmi.c
index c1a6cd6..abdaed3 100644
--- a/drivers/platform/x86/asus-nb-wmi.c
+++ b/drivers/platform/x86/asus-nb-wmi.c
@@ -191,6 +191,15 @@
},
{
.callback = dmi_matched,
+ .ident = "ASUSTeK COMPUTER INC. X551CA",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "X551CA"),
+ },
+ .driver_data = &quirk_asus_wapf4,
+ },
+ {
+ .callback = dmi_matched,
.ident = "ASUSTeK COMPUTER INC. X55A",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c
index 21fc932..7543a56 100644
--- a/drivers/platform/x86/asus-wmi.c
+++ b/drivers/platform/x86/asus-wmi.c
@@ -1308,8 +1308,7 @@
static void asus_wmi_backlight_exit(struct asus_wmi *asus)
{
- if (asus->backlight_device)
- backlight_device_unregister(asus->backlight_device);
+ backlight_device_unregister(asus->backlight_device);
asus->backlight_device = NULL;
}
diff --git a/drivers/platform/x86/dell-laptop.c b/drivers/platform/x86/dell-laptop.c
index f6a28d7..9411eae 100644
--- a/drivers/platform/x86/dell-laptop.c
+++ b/drivers/platform/x86/dell-laptop.c
@@ -2,9 +2,11 @@
* Driver for Dell laptop extras
*
* Copyright (c) Red Hat <mjg@redhat.com>
+ * Copyright (c) 2014 Gabriele Mazzotta <gabriele.mzt@gmail.com>
+ * Copyright (c) 2014 Pali Rohár <pali.rohar@gmail.com>
*
- * Based on documentation in the libsmbios package, Copyright (C) 2005 Dell
- * Inc.
+ * Based on documentation in the libsmbios package:
+ * Copyright (C) 2005-2014 Dell Inc.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
@@ -32,6 +34,13 @@
#include "../../firmware/dcdbas.h"
#define BRIGHTNESS_TOKEN 0x7d
+#define KBD_LED_OFF_TOKEN 0x01E1
+#define KBD_LED_ON_TOKEN 0x01E2
+#define KBD_LED_AUTO_TOKEN 0x01E3
+#define KBD_LED_AUTO_25_TOKEN 0x02EA
+#define KBD_LED_AUTO_50_TOKEN 0x02EB
+#define KBD_LED_AUTO_75_TOKEN 0x02EC
+#define KBD_LED_AUTO_100_TOKEN 0x02F6
/* This structure will be modified by the firmware when we enter
* system management mode, hence the volatiles */
@@ -62,6 +71,13 @@
struct quirk_entry {
u8 touchpad_led;
+
+ int needs_kbd_timeouts;
+ /*
+ * Ordered list of timeouts expressed in seconds.
+ * The list must end with -1
+ */
+ int kbd_timeouts[];
};
static struct quirk_entry *quirks;
@@ -76,6 +92,15 @@
return 1;
}
+/*
+ * These values come from Windows utility provided by Dell. If any other value
+ * is used then BIOS silently set timeout to 0 without any error message.
+ */
+static struct quirk_entry quirk_dell_xps13_9333 = {
+ .needs_kbd_timeouts = 1,
+ .kbd_timeouts = { 0, 5, 15, 60, 5 * 60, 15 * 60, -1 },
+};
+
static int da_command_address;
static int da_command_code;
static int da_num_tokens;
@@ -267,6 +292,15 @@
},
.driver_data = &quirk_dell_vostro_v130,
},
+ {
+ .callback = dmi_matched,
+ .ident = "Dell XPS13 9333",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "XPS13 9333"),
+ },
+ .driver_data = &quirk_dell_xps13_9333,
+ },
{ }
};
@@ -331,17 +365,29 @@
}
}
-static int find_token_location(int tokenid)
+static int find_token_id(int tokenid)
{
int i;
+
for (i = 0; i < da_num_tokens; i++) {
if (da_tokens[i].tokenID == tokenid)
- return da_tokens[i].location;
+ return i;
}
return -1;
}
+static int find_token_location(int tokenid)
+{
+ int id;
+
+ id = find_token_id(tokenid);
+ if (id == -1)
+ return -1;
+
+ return da_tokens[id].location;
+}
+
static struct calling_interface_buffer *
dell_send_request(struct calling_interface_buffer *buffer, int class,
int select)
@@ -362,6 +408,20 @@
return buffer;
}
+static inline int dell_smi_error(int value)
+{
+ switch (value) {
+ case 0: /* Completed successfully */
+ return 0;
+ case -1: /* Completed with error */
+ return -EIO;
+ case -2: /* Function not supported */
+ return -ENXIO;
+ default: /* Unknown error */
+ return -EINVAL;
+ }
+}
+
/* Derived from information in DellWirelessCtl.cpp:
Class 17, select 11 is radio control. It returns an array of 32-bit values.
@@ -563,7 +623,7 @@
{
static bool extended;
- if (str & 0x20)
+ if (str & I8042_STR_AUXDATA)
return false;
if (unlikely(data == 0xe0)) {
@@ -716,7 +776,7 @@
else
dell_send_request(buffer, 1, 1);
-out:
+ out:
release_buffer();
return ret;
}
@@ -740,7 +800,7 @@
ret = buffer->output[1];
-out:
+ out:
release_buffer();
return ret;
}
@@ -789,6 +849,984 @@
led_classdev_unregister(&touchpad_led);
}
+/*
+ * Derived from information in smbios-keyboard-ctl:
+ *
+ * cbClass 4
+ * cbSelect 11
+ * Keyboard illumination
+ * cbArg1 determines the function to be performed
+ *
+ * cbArg1 0x0 = Get Feature Information
+ * cbRES1 Standard return codes (0, -1, -2)
+ * cbRES2, word0 Bitmap of user-selectable modes
+ * bit 0 Always off (All systems)
+ * bit 1 Always on (Travis ATG, Siberia)
+ * bit 2 Auto: ALS-based On; ALS-based Off (Travis ATG)
+ * bit 3 Auto: ALS- and input-activity-based On; input-activity based Off
+ * bit 4 Auto: Input-activity-based On; input-activity based Off
+ * bit 5 Auto: Input-activity-based On (illumination level 25%); input-activity based Off
+ * bit 6 Auto: Input-activity-based On (illumination level 50%); input-activity based Off
+ * bit 7 Auto: Input-activity-based On (illumination level 75%); input-activity based Off
+ * bit 8 Auto: Input-activity-based On (illumination level 100%); input-activity based Off
+ * bits 9-15 Reserved for future use
+ * cbRES2, byte2 Reserved for future use
+ * cbRES2, byte3 Keyboard illumination type
+ * 0 Reserved
+ * 1 Tasklight
+ * 2 Backlight
+ * 3-255 Reserved for future use
+ * cbRES3, byte0 Supported auto keyboard illumination trigger bitmap.
+ * bit 0 Any keystroke
+ * bit 1 Touchpad activity
+ * bit 2 Pointing stick
+ * bit 3 Any mouse
+ * bits 4-7 Reserved for future use
+ * cbRES3, byte1 Supported timeout unit bitmap
+ * bit 0 Seconds
+ * bit 1 Minutes
+ * bit 2 Hours
+ * bit 3 Days
+ * bits 4-7 Reserved for future use
+ * cbRES3, byte2 Number of keyboard light brightness levels
+ * cbRES4, byte0 Maximum acceptable seconds value (0 if seconds not supported).
+ * cbRES4, byte1 Maximum acceptable minutes value (0 if minutes not supported).
+ * cbRES4, byte2 Maximum acceptable hours value (0 if hours not supported).
+ * cbRES4, byte3 Maximum acceptable days value (0 if days not supported)
+ *
+ * cbArg1 0x1 = Get Current State
+ * cbRES1 Standard return codes (0, -1, -2)
+ * cbRES2, word0 Bitmap of current mode state
+ * bit 0 Always off (All systems)
+ * bit 1 Always on (Travis ATG, Siberia)
+ * bit 2 Auto: ALS-based On; ALS-based Off (Travis ATG)
+ * bit 3 Auto: ALS- and input-activity-based On; input-activity based Off
+ * bit 4 Auto: Input-activity-based On; input-activity based Off
+ * bit 5 Auto: Input-activity-based On (illumination level 25%); input-activity based Off
+ * bit 6 Auto: Input-activity-based On (illumination level 50%); input-activity based Off
+ * bit 7 Auto: Input-activity-based On (illumination level 75%); input-activity based Off
+ * bit 8 Auto: Input-activity-based On (illumination level 100%); input-activity based Off
+ * bits 9-15 Reserved for future use
+ * Note: Only One bit can be set
+ * cbRES2, byte2 Currently active auto keyboard illumination triggers.
+ * bit 0 Any keystroke
+ * bit 1 Touchpad activity
+ * bit 2 Pointing stick
+ * bit 3 Any mouse
+ * bits 4-7 Reserved for future use
+ * cbRES2, byte3 Current Timeout
+ * bits 7:6 Timeout units indicator:
+ * 00b Seconds
+ * 01b Minutes
+ * 10b Hours
+ * 11b Days
+ * bits 5:0 Timeout value (0-63) in sec/min/hr/day
+ * NOTE: A value of 0 means always on (no timeout) if any bits of RES3 byte
+ * are set upon return from the [Get feature information] call.
+ * cbRES3, byte0 Current setting of ALS value that turns the light on or off.
+ * cbRES3, byte1 Current ALS reading
+ * cbRES3, byte2 Current keyboard light level.
+ *
+ * cbArg1 0x2 = Set New State
+ * cbRES1 Standard return codes (0, -1, -2)
+ * cbArg2, word0 Bitmap of current mode state
+ * bit 0 Always off (All systems)
+ * bit 1 Always on (Travis ATG, Siberia)
+ * bit 2 Auto: ALS-based On; ALS-based Off (Travis ATG)
+ * bit 3 Auto: ALS- and input-activity-based On; input-activity based Off
+ * bit 4 Auto: Input-activity-based On; input-activity based Off
+ * bit 5 Auto: Input-activity-based On (illumination level 25%); input-activity based Off
+ * bit 6 Auto: Input-activity-based On (illumination level 50%); input-activity based Off
+ * bit 7 Auto: Input-activity-based On (illumination level 75%); input-activity based Off
+ * bit 8 Auto: Input-activity-based On (illumination level 100%); input-activity based Off
+ * bits 9-15 Reserved for future use
+ * Note: Only One bit can be set
+ * cbArg2, byte2 Desired auto keyboard illumination triggers. Must remain inactive to allow
+ * keyboard to turn off automatically.
+ * bit 0 Any keystroke
+ * bit 1 Touchpad activity
+ * bit 2 Pointing stick
+ * bit 3 Any mouse
+ * bits 4-7 Reserved for future use
+ * cbArg2, byte3 Desired Timeout
+ * bits 7:6 Timeout units indicator:
+ * 00b Seconds
+ * 01b Minutes
+ * 10b Hours
+ * 11b Days
+ * bits 5:0 Timeout value (0-63) in sec/min/hr/day
+ * cbArg3, byte0 Desired setting of ALS value that turns the light on or off.
+ * cbArg3, byte2 Desired keyboard light level.
+ */
+
+
+enum kbd_timeout_unit {
+ KBD_TIMEOUT_SECONDS = 0,
+ KBD_TIMEOUT_MINUTES,
+ KBD_TIMEOUT_HOURS,
+ KBD_TIMEOUT_DAYS,
+};
+
+enum kbd_mode_bit {
+ KBD_MODE_BIT_OFF = 0,
+ KBD_MODE_BIT_ON,
+ KBD_MODE_BIT_ALS,
+ KBD_MODE_BIT_TRIGGER_ALS,
+ KBD_MODE_BIT_TRIGGER,
+ KBD_MODE_BIT_TRIGGER_25,
+ KBD_MODE_BIT_TRIGGER_50,
+ KBD_MODE_BIT_TRIGGER_75,
+ KBD_MODE_BIT_TRIGGER_100,
+};
+
+#define kbd_is_als_mode_bit(bit) \
+ ((bit) == KBD_MODE_BIT_ALS || (bit) == KBD_MODE_BIT_TRIGGER_ALS)
+#define kbd_is_trigger_mode_bit(bit) \
+ ((bit) >= KBD_MODE_BIT_TRIGGER_ALS && (bit) <= KBD_MODE_BIT_TRIGGER_100)
+#define kbd_is_level_mode_bit(bit) \
+ ((bit) >= KBD_MODE_BIT_TRIGGER_25 && (bit) <= KBD_MODE_BIT_TRIGGER_100)
+
+struct kbd_info {
+ u16 modes;
+ u8 type;
+ u8 triggers;
+ u8 levels;
+ u8 seconds;
+ u8 minutes;
+ u8 hours;
+ u8 days;
+};
+
+struct kbd_state {
+ u8 mode_bit;
+ u8 triggers;
+ u8 timeout_value;
+ u8 timeout_unit;
+ u8 als_setting;
+ u8 als_value;
+ u8 level;
+};
+
+static const int kbd_tokens[] = {
+ KBD_LED_OFF_TOKEN,
+ KBD_LED_AUTO_25_TOKEN,
+ KBD_LED_AUTO_50_TOKEN,
+ KBD_LED_AUTO_75_TOKEN,
+ KBD_LED_AUTO_100_TOKEN,
+ KBD_LED_ON_TOKEN,
+};
+
+static u16 kbd_token_bits;
+
+static struct kbd_info kbd_info;
+static bool kbd_als_supported;
+static bool kbd_triggers_supported;
+
+static u8 kbd_mode_levels[16];
+static int kbd_mode_levels_count;
+
+static u8 kbd_previous_level;
+static u8 kbd_previous_mode_bit;
+
+static bool kbd_led_present;
+
+/*
+ * NOTE: there are three ways to set the keyboard backlight level.
+ * First, via kbd_state.mode_bit (assigning KBD_MODE_BIT_TRIGGER_* value).
+ * Second, via kbd_state.level (assigning numerical value <= kbd_info.levels).
+ * Third, via SMBIOS tokens (KBD_LED_* in kbd_tokens)
+ *
+ * There are laptops which support only one of these methods. If we want to
+ * support as many machines as possible we need to implement all three methods.
+ * The first two methods use the kbd_state structure. The third uses SMBIOS
+ * tokens. If kbd_info.levels == 0, the machine does not support setting the
+ * keyboard backlight level via kbd_state.level.
+ */
+
+static int kbd_get_info(struct kbd_info *info)
+{
+ u8 units;
+ int ret;
+
+ get_buffer();
+
+ buffer->input[0] = 0x0;
+ dell_send_request(buffer, 4, 11);
+ ret = buffer->output[0];
+
+ if (ret) {
+ ret = dell_smi_error(ret);
+ goto out;
+ }
+
+ info->modes = buffer->output[1] & 0xFFFF;
+ info->type = (buffer->output[1] >> 24) & 0xFF;
+ info->triggers = buffer->output[2] & 0xFF;
+ units = (buffer->output[2] >> 8) & 0xFF;
+ info->levels = (buffer->output[2] >> 16) & 0xFF;
+
+ if (units & BIT(0))
+ info->seconds = (buffer->output[3] >> 0) & 0xFF;
+ if (units & BIT(1))
+ info->minutes = (buffer->output[3] >> 8) & 0xFF;
+ if (units & BIT(2))
+ info->hours = (buffer->output[3] >> 16) & 0xFF;
+ if (units & BIT(3))
+ info->days = (buffer->output[3] >> 24) & 0xFF;
+
+ out:
+ release_buffer();
+ return ret;
+}
+
+static unsigned int kbd_get_max_level(void)
+{
+ if (kbd_info.levels != 0)
+ return kbd_info.levels;
+ if (kbd_mode_levels_count > 0)
+ return kbd_mode_levels_count - 1;
+ return 0;
+}
+
+static int kbd_get_level(struct kbd_state *state)
+{
+ int i;
+
+ if (kbd_info.levels != 0)
+ return state->level;
+
+ if (kbd_mode_levels_count > 0) {
+ for (i = 0; i < kbd_mode_levels_count; ++i)
+ if (kbd_mode_levels[i] == state->mode_bit)
+ return i;
+ return 0;
+ }
+
+ return -EINVAL;
+}
+
+static int kbd_set_level(struct kbd_state *state, u8 level)
+{
+ if (kbd_info.levels != 0) {
+ if (level != 0)
+ kbd_previous_level = level;
+ if (state->level == level)
+ return 0;
+ state->level = level;
+ if (level != 0 && state->mode_bit == KBD_MODE_BIT_OFF)
+ state->mode_bit = kbd_previous_mode_bit;
+ else if (level == 0 && state->mode_bit != KBD_MODE_BIT_OFF) {
+ kbd_previous_mode_bit = state->mode_bit;
+ state->mode_bit = KBD_MODE_BIT_OFF;
+ }
+ return 0;
+ }
+
+ if (kbd_mode_levels_count > 0 && level < kbd_mode_levels_count) {
+ if (level != 0)
+ kbd_previous_level = level;
+ state->mode_bit = kbd_mode_levels[level];
+ return 0;
+ }
+
+ return -EINVAL;
+}
+
+static int kbd_get_state(struct kbd_state *state)
+{
+ int ret;
+
+ get_buffer();
+
+ buffer->input[0] = 0x1;
+ dell_send_request(buffer, 4, 11);
+ ret = buffer->output[0];
+
+ if (ret) {
+ ret = dell_smi_error(ret);
+ goto out;
+ }
+
+ state->mode_bit = ffs(buffer->output[1] & 0xFFFF);
+ if (state->mode_bit != 0)
+ state->mode_bit--;
+
+ state->triggers = (buffer->output[1] >> 16) & 0xFF;
+ state->timeout_value = (buffer->output[1] >> 24) & 0x3F;
+ state->timeout_unit = (buffer->output[1] >> 30) & 0x3;
+ state->als_setting = buffer->output[2] & 0xFF;
+ state->als_value = (buffer->output[2] >> 8) & 0xFF;
+ state->level = (buffer->output[2] >> 16) & 0xFF;
+
+ out:
+ release_buffer();
+ return ret;
+}
+
+static int kbd_set_state(struct kbd_state *state)
+{
+ int ret;
+
+ get_buffer();
+ buffer->input[0] = 0x2;
+ buffer->input[1] = BIT(state->mode_bit) & 0xFFFF;
+ buffer->input[1] |= (state->triggers & 0xFF) << 16;
+ buffer->input[1] |= (state->timeout_value & 0x3F) << 24;
+ buffer->input[1] |= (state->timeout_unit & 0x3) << 30;
+ buffer->input[2] = state->als_setting & 0xFF;
+ buffer->input[2] |= (state->level & 0xFF) << 16;
+ dell_send_request(buffer, 4, 11);
+ ret = buffer->output[0];
+ release_buffer();
+
+ return dell_smi_error(ret);
+}
+
+static int kbd_set_state_safe(struct kbd_state *state, struct kbd_state *old)
+{
+ int ret;
+
+ ret = kbd_set_state(state);
+ if (ret == 0)
+ return 0;
+
+ /*
+ * When setting the new state fails,try to restore the previous one.
+ * This is needed on some machines where BIOS sets a default state when
+ * setting a new state fails. This default state could be all off.
+ */
+
+ if (kbd_set_state(old))
+ pr_err("Setting old previous keyboard state failed\n");
+
+ return ret;
+}
+
+static int kbd_set_token_bit(u8 bit)
+{
+ int id;
+ int ret;
+
+ if (bit >= ARRAY_SIZE(kbd_tokens))
+ return -EINVAL;
+
+ id = find_token_id(kbd_tokens[bit]);
+ if (id == -1)
+ return -EINVAL;
+
+ get_buffer();
+ buffer->input[0] = da_tokens[id].location;
+ buffer->input[1] = da_tokens[id].value;
+ dell_send_request(buffer, 1, 0);
+ ret = buffer->output[0];
+ release_buffer();
+
+ return dell_smi_error(ret);
+}
+
+static int kbd_get_token_bit(u8 bit)
+{
+ int id;
+ int ret;
+ int val;
+
+ if (bit >= ARRAY_SIZE(kbd_tokens))
+ return -EINVAL;
+
+ id = find_token_id(kbd_tokens[bit]);
+ if (id == -1)
+ return -EINVAL;
+
+ get_buffer();
+ buffer->input[0] = da_tokens[id].location;
+ dell_send_request(buffer, 0, 0);
+ ret = buffer->output[0];
+ val = buffer->output[1];
+ release_buffer();
+
+ if (ret)
+ return dell_smi_error(ret);
+
+ return (val == da_tokens[id].value);
+}
+
+static int kbd_get_first_active_token_bit(void)
+{
+ int i;
+ int ret;
+
+ for (i = 0; i < ARRAY_SIZE(kbd_tokens); ++i) {
+ ret = kbd_get_token_bit(i);
+ if (ret == 1)
+ return i;
+ }
+
+ return ret;
+}
+
+static int kbd_get_valid_token_counts(void)
+{
+ return hweight16(kbd_token_bits);
+}
+
+static inline int kbd_init_info(void)
+{
+ struct kbd_state state;
+ int ret;
+ int i;
+
+ ret = kbd_get_info(&kbd_info);
+ if (ret)
+ return ret;
+
+ kbd_get_state(&state);
+
+ /* NOTE: timeout value is stored in 6 bits so max value is 63 */
+ if (kbd_info.seconds > 63)
+ kbd_info.seconds = 63;
+ if (kbd_info.minutes > 63)
+ kbd_info.minutes = 63;
+ if (kbd_info.hours > 63)
+ kbd_info.hours = 63;
+ if (kbd_info.days > 63)
+ kbd_info.days = 63;
+
+ /* NOTE: On tested machines ON mode did not work and caused
+ * problems (turned backlight off) so do not use it
+ */
+ kbd_info.modes &= ~BIT(KBD_MODE_BIT_ON);
+
+ kbd_previous_level = kbd_get_level(&state);
+ kbd_previous_mode_bit = state.mode_bit;
+
+ if (kbd_previous_level == 0 && kbd_get_max_level() != 0)
+ kbd_previous_level = 1;
+
+ if (kbd_previous_mode_bit == KBD_MODE_BIT_OFF) {
+ kbd_previous_mode_bit =
+ ffs(kbd_info.modes & ~BIT(KBD_MODE_BIT_OFF));
+ if (kbd_previous_mode_bit != 0)
+ kbd_previous_mode_bit--;
+ }
+
+ if (kbd_info.modes & (BIT(KBD_MODE_BIT_ALS) |
+ BIT(KBD_MODE_BIT_TRIGGER_ALS)))
+ kbd_als_supported = true;
+
+ if (kbd_info.modes & (
+ BIT(KBD_MODE_BIT_TRIGGER_ALS) | BIT(KBD_MODE_BIT_TRIGGER) |
+ BIT(KBD_MODE_BIT_TRIGGER_25) | BIT(KBD_MODE_BIT_TRIGGER_50) |
+ BIT(KBD_MODE_BIT_TRIGGER_75) | BIT(KBD_MODE_BIT_TRIGGER_100)
+ ))
+ kbd_triggers_supported = true;
+
+ /* kbd_mode_levels[0] is reserved, see below */
+ for (i = 0; i < 16; ++i)
+ if (kbd_is_level_mode_bit(i) && (BIT(i) & kbd_info.modes))
+ kbd_mode_levels[1 + kbd_mode_levels_count++] = i;
+
+ /*
+ * Find the first supported mode and assign to kbd_mode_levels[0].
+ * This should be 0 (off), but we cannot depend on the BIOS to
+ * support 0.
+ */
+ if (kbd_mode_levels_count > 0) {
+ for (i = 0; i < 16; ++i) {
+ if (BIT(i) & kbd_info.modes) {
+ kbd_mode_levels[0] = i;
+ break;
+ }
+ }
+ kbd_mode_levels_count++;
+ }
+
+ return 0;
+
+}
+
+static inline void kbd_init_tokens(void)
+{
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(kbd_tokens); ++i)
+ if (find_token_id(kbd_tokens[i]) != -1)
+ kbd_token_bits |= BIT(i);
+}
+
+static void kbd_init(void)
+{
+ int ret;
+
+ ret = kbd_init_info();
+ kbd_init_tokens();
+
+ if (kbd_token_bits != 0 || ret == 0)
+ kbd_led_present = true;
+}
+
+static ssize_t kbd_led_timeout_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct kbd_state new_state;
+ struct kbd_state state;
+ bool convert;
+ int value;
+ int ret;
+ char ch;
+ u8 unit;
+ int i;
+
+ ret = sscanf(buf, "%d %c", &value, &ch);
+ if (ret < 1)
+ return -EINVAL;
+ else if (ret == 1)
+ ch = 's';
+
+ if (value < 0)
+ return -EINVAL;
+
+ convert = false;
+
+ switch (ch) {
+ case 's':
+ if (value > kbd_info.seconds)
+ convert = true;
+ unit = KBD_TIMEOUT_SECONDS;
+ break;
+ case 'm':
+ if (value > kbd_info.minutes)
+ convert = true;
+ unit = KBD_TIMEOUT_MINUTES;
+ break;
+ case 'h':
+ if (value > kbd_info.hours)
+ convert = true;
+ unit = KBD_TIMEOUT_HOURS;
+ break;
+ case 'd':
+ if (value > kbd_info.days)
+ convert = true;
+ unit = KBD_TIMEOUT_DAYS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ if (quirks && quirks->needs_kbd_timeouts)
+ convert = true;
+
+ if (convert) {
+ /* Convert value from current units to seconds */
+ switch (unit) {
+ case KBD_TIMEOUT_DAYS:
+ value *= 24;
+ case KBD_TIMEOUT_HOURS:
+ value *= 60;
+ case KBD_TIMEOUT_MINUTES:
+ value *= 60;
+ unit = KBD_TIMEOUT_SECONDS;
+ }
+
+ if (quirks && quirks->needs_kbd_timeouts) {
+ for (i = 0; quirks->kbd_timeouts[i] != -1; i++) {
+ if (value <= quirks->kbd_timeouts[i]) {
+ value = quirks->kbd_timeouts[i];
+ break;
+ }
+ }
+ }
+
+ if (value <= kbd_info.seconds && kbd_info.seconds) {
+ unit = KBD_TIMEOUT_SECONDS;
+ } else if (value / 60 <= kbd_info.minutes && kbd_info.minutes) {
+ value /= 60;
+ unit = KBD_TIMEOUT_MINUTES;
+ } else if (value / (60 * 60) <= kbd_info.hours && kbd_info.hours) {
+ value /= (60 * 60);
+ unit = KBD_TIMEOUT_HOURS;
+ } else if (value / (60 * 60 * 24) <= kbd_info.days && kbd_info.days) {
+ value /= (60 * 60 * 24);
+ unit = KBD_TIMEOUT_DAYS;
+ } else {
+ return -EINVAL;
+ }
+ }
+
+ ret = kbd_get_state(&state);
+ if (ret)
+ return ret;
+
+ new_state = state;
+ new_state.timeout_value = value;
+ new_state.timeout_unit = unit;
+
+ ret = kbd_set_state_safe(&new_state, &state);
+ if (ret)
+ return ret;
+
+ return count;
+}
+
+static ssize_t kbd_led_timeout_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct kbd_state state;
+ int ret;
+ int len;
+
+ ret = kbd_get_state(&state);
+ if (ret)
+ return ret;
+
+ len = sprintf(buf, "%d", state.timeout_value);
+
+ switch (state.timeout_unit) {
+ case KBD_TIMEOUT_SECONDS:
+ return len + sprintf(buf+len, "s\n");
+ case KBD_TIMEOUT_MINUTES:
+ return len + sprintf(buf+len, "m\n");
+ case KBD_TIMEOUT_HOURS:
+ return len + sprintf(buf+len, "h\n");
+ case KBD_TIMEOUT_DAYS:
+ return len + sprintf(buf+len, "d\n");
+ default:
+ return -EINVAL;
+ }
+
+ return len;
+}
+
+static DEVICE_ATTR(stop_timeout, S_IRUGO | S_IWUSR,
+ kbd_led_timeout_show, kbd_led_timeout_store);
+
+static const char * const kbd_led_triggers[] = {
+ "keyboard",
+ "touchpad",
+ /*"trackstick"*/ NULL, /* NOTE: trackstick is just alias for touchpad */
+ "mouse",
+};
+
+static ssize_t kbd_led_triggers_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct kbd_state new_state;
+ struct kbd_state state;
+ bool triggers_enabled = false;
+ bool als_enabled = false;
+ bool disable_als = false;
+ bool enable_als = false;
+ int trigger_bit = -1;
+ char trigger[21];
+ int i, ret;
+
+ ret = sscanf(buf, "%20s", trigger);
+ if (ret != 1)
+ return -EINVAL;
+
+ if (trigger[0] != '+' && trigger[0] != '-')
+ return -EINVAL;
+
+ ret = kbd_get_state(&state);
+ if (ret)
+ return ret;
+
+ if (kbd_als_supported)
+ als_enabled = kbd_is_als_mode_bit(state.mode_bit);
+
+ if (kbd_triggers_supported)
+ triggers_enabled = kbd_is_trigger_mode_bit(state.mode_bit);
+
+ if (kbd_als_supported) {
+ if (strcmp(trigger, "+als") == 0) {
+ if (als_enabled)
+ return count;
+ enable_als = true;
+ } else if (strcmp(trigger, "-als") == 0) {
+ if (!als_enabled)
+ return count;
+ disable_als = true;
+ }
+ }
+
+ if (enable_als || disable_als) {
+ new_state = state;
+ if (enable_als) {
+ if (triggers_enabled)
+ new_state.mode_bit = KBD_MODE_BIT_TRIGGER_ALS;
+ else
+ new_state.mode_bit = KBD_MODE_BIT_ALS;
+ } else {
+ if (triggers_enabled) {
+ new_state.mode_bit = KBD_MODE_BIT_TRIGGER;
+ kbd_set_level(&new_state, kbd_previous_level);
+ } else {
+ new_state.mode_bit = KBD_MODE_BIT_ON;
+ }
+ }
+ if (!(kbd_info.modes & BIT(new_state.mode_bit)))
+ return -EINVAL;
+ ret = kbd_set_state_safe(&new_state, &state);
+ if (ret)
+ return ret;
+ kbd_previous_mode_bit = new_state.mode_bit;
+ return count;
+ }
+
+ if (kbd_triggers_supported) {
+ for (i = 0; i < ARRAY_SIZE(kbd_led_triggers); ++i) {
+ if (!(kbd_info.triggers & BIT(i)))
+ continue;
+ if (!kbd_led_triggers[i])
+ continue;
+ if (strcmp(trigger+1, kbd_led_triggers[i]) != 0)
+ continue;
+ if (trigger[0] == '+' &&
+ triggers_enabled && (state.triggers & BIT(i)))
+ return count;
+ if (trigger[0] == '-' &&
+ (!triggers_enabled || !(state.triggers & BIT(i))))
+ return count;
+ trigger_bit = i;
+ break;
+ }
+ }
+
+ if (trigger_bit != -1) {
+ new_state = state;
+ if (trigger[0] == '+')
+ new_state.triggers |= BIT(trigger_bit);
+ else {
+ new_state.triggers &= ~BIT(trigger_bit);
+ /* NOTE: trackstick bit (2) must be disabled when
+ * disabling touchpad bit (1), otherwise touchpad
+ * bit (1) will not be disabled */
+ if (trigger_bit == 1)
+ new_state.triggers &= ~BIT(2);
+ }
+ if ((kbd_info.triggers & new_state.triggers) !=
+ new_state.triggers)
+ return -EINVAL;
+ if (new_state.triggers && !triggers_enabled) {
+ if (als_enabled)
+ new_state.mode_bit = KBD_MODE_BIT_TRIGGER_ALS;
+ else {
+ new_state.mode_bit = KBD_MODE_BIT_TRIGGER;
+ kbd_set_level(&new_state, kbd_previous_level);
+ }
+ } else if (new_state.triggers == 0) {
+ if (als_enabled)
+ new_state.mode_bit = KBD_MODE_BIT_ALS;
+ else
+ kbd_set_level(&new_state, 0);
+ }
+ if (!(kbd_info.modes & BIT(new_state.mode_bit)))
+ return -EINVAL;
+ ret = kbd_set_state_safe(&new_state, &state);
+ if (ret)
+ return ret;
+ if (new_state.mode_bit != KBD_MODE_BIT_OFF)
+ kbd_previous_mode_bit = new_state.mode_bit;
+ return count;
+ }
+
+ return -EINVAL;
+}
+
+static ssize_t kbd_led_triggers_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct kbd_state state;
+ bool triggers_enabled;
+ int level, i, ret;
+ int len = 0;
+
+ ret = kbd_get_state(&state);
+ if (ret)
+ return ret;
+
+ len = 0;
+
+ if (kbd_triggers_supported) {
+ triggers_enabled = kbd_is_trigger_mode_bit(state.mode_bit);
+ level = kbd_get_level(&state);
+ for (i = 0; i < ARRAY_SIZE(kbd_led_triggers); ++i) {
+ if (!(kbd_info.triggers & BIT(i)))
+ continue;
+ if (!kbd_led_triggers[i])
+ continue;
+ if ((triggers_enabled || level <= 0) &&
+ (state.triggers & BIT(i)))
+ buf[len++] = '+';
+ else
+ buf[len++] = '-';
+ len += sprintf(buf+len, "%s ", kbd_led_triggers[i]);
+ }
+ }
+
+ if (kbd_als_supported) {
+ if (kbd_is_als_mode_bit(state.mode_bit))
+ len += sprintf(buf+len, "+als ");
+ else
+ len += sprintf(buf+len, "-als ");
+ }
+
+ if (len)
+ buf[len - 1] = '\n';
+
+ return len;
+}
+
+static DEVICE_ATTR(start_triggers, S_IRUGO | S_IWUSR,
+ kbd_led_triggers_show, kbd_led_triggers_store);
+
+static ssize_t kbd_led_als_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct kbd_state state;
+ struct kbd_state new_state;
+ u8 setting;
+ int ret;
+
+ ret = kstrtou8(buf, 10, &setting);
+ if (ret)
+ return ret;
+
+ ret = kbd_get_state(&state);
+ if (ret)
+ return ret;
+
+ new_state = state;
+ new_state.als_setting = setting;
+
+ ret = kbd_set_state_safe(&new_state, &state);
+ if (ret)
+ return ret;
+
+ return count;
+}
+
+static ssize_t kbd_led_als_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct kbd_state state;
+ int ret;
+
+ ret = kbd_get_state(&state);
+ if (ret)
+ return ret;
+
+ return sprintf(buf, "%d\n", state.als_setting);
+}
+
+static DEVICE_ATTR(als_setting, S_IRUGO | S_IWUSR,
+ kbd_led_als_show, kbd_led_als_store);
+
+static struct attribute *kbd_led_attrs[] = {
+ &dev_attr_stop_timeout.attr,
+ &dev_attr_start_triggers.attr,
+ &dev_attr_als_setting.attr,
+ NULL,
+};
+ATTRIBUTE_GROUPS(kbd_led);
+
+static enum led_brightness kbd_led_level_get(struct led_classdev *led_cdev)
+{
+ int ret;
+ u16 num;
+ struct kbd_state state;
+
+ if (kbd_get_max_level()) {
+ ret = kbd_get_state(&state);
+ if (ret)
+ return 0;
+ ret = kbd_get_level(&state);
+ if (ret < 0)
+ return 0;
+ return ret;
+ }
+
+ if (kbd_get_valid_token_counts()) {
+ ret = kbd_get_first_active_token_bit();
+ if (ret < 0)
+ return 0;
+ for (num = kbd_token_bits; num != 0 && ret > 0; --ret)
+ num &= num - 1; /* clear the first bit set */
+ if (num == 0)
+ return 0;
+ return ffs(num) - 1;
+ }
+
+ pr_warn("Keyboard brightness level control not supported\n");
+ return 0;
+}
+
+static void kbd_led_level_set(struct led_classdev *led_cdev,
+ enum led_brightness value)
+{
+ struct kbd_state state;
+ struct kbd_state new_state;
+ u16 num;
+
+ if (kbd_get_max_level()) {
+ if (kbd_get_state(&state))
+ return;
+ new_state = state;
+ if (kbd_set_level(&new_state, value))
+ return;
+ kbd_set_state_safe(&new_state, &state);
+ return;
+ }
+
+ if (kbd_get_valid_token_counts()) {
+ for (num = kbd_token_bits; num != 0 && value > 0; --value)
+ num &= num - 1; /* clear the first bit set */
+ if (num == 0)
+ return;
+ kbd_set_token_bit(ffs(num) - 1);
+ return;
+ }
+
+ pr_warn("Keyboard brightness level control not supported\n");
+}
+
+static struct led_classdev kbd_led = {
+ .name = "dell::kbd_backlight",
+ .brightness_set = kbd_led_level_set,
+ .brightness_get = kbd_led_level_get,
+ .groups = kbd_led_groups,
+};
+
+static int __init kbd_led_init(struct device *dev)
+{
+ kbd_init();
+ if (!kbd_led_present)
+ return -ENODEV;
+ kbd_led.max_brightness = kbd_get_max_level();
+ if (!kbd_led.max_brightness) {
+ kbd_led.max_brightness = kbd_get_valid_token_counts();
+ if (kbd_led.max_brightness)
+ kbd_led.max_brightness--;
+ }
+ return led_classdev_register(dev, &kbd_led);
+}
+
+static void brightness_set_exit(struct led_classdev *led_cdev,
+ enum led_brightness value)
+{
+ /* Don't change backlight level on exit */
+};
+
+static void kbd_led_exit(void)
+{
+ if (!kbd_led_present)
+ return;
+ kbd_led.brightness_set = brightness_set_exit;
+ led_classdev_unregister(&kbd_led);
+}
+
static int __init dell_init(void)
{
int max_intensity = 0;
@@ -841,6 +1879,8 @@
if (quirks && quirks->touchpad_led)
touchpad_led_init(&platform_device->dev);
+ kbd_led_init(&platform_device->dev);
+
dell_laptop_dir = debugfs_create_dir("dell_laptop", NULL);
if (dell_laptop_dir != NULL)
debugfs_create_file("rfkill", 0444, dell_laptop_dir, NULL,
@@ -908,6 +1948,7 @@
debugfs_remove_recursive(dell_laptop_dir);
if (quirks && quirks->touchpad_led)
touchpad_led_exit();
+ kbd_led_exit();
i8042_remove_filter(dell_laptop_i8042_filter);
cancel_delayed_work_sync(&dell_rfkill_work);
backlight_device_unregister(dell_backlight_device);
@@ -924,5 +1965,7 @@
module_exit(dell_exit);
MODULE_AUTHOR("Matthew Garrett <mjg@redhat.com>");
+MODULE_AUTHOR("Gabriele Mazzotta <gabriele.mzt@gmail.com>");
+MODULE_AUTHOR("Pali Rohár <pali.rohar@gmail.com>");
MODULE_DESCRIPTION("Dell laptop driver");
MODULE_LICENSE("GPL");
diff --git a/drivers/platform/x86/dell-smo8800.c b/drivers/platform/x86/dell-smo8800.c
index a653716..0aec4fd 100644
--- a/drivers/platform/x86/dell-smo8800.c
+++ b/drivers/platform/x86/dell-smo8800.c
@@ -1,5 +1,5 @@
/*
- * dell-smo8800.c - Dell Latitude ACPI SMO8800/SMO8810 freefall sensor driver
+ * dell-smo8800.c - Dell Latitude ACPI SMO88XX freefall sensor driver
*
* Copyright (C) 2012 Sonal Santan <sonal.santan@gmail.com>
* Copyright (C) 2014 Pali Rohár <pali.rohar@gmail.com>
@@ -209,7 +209,13 @@
static const struct acpi_device_id smo8800_ids[] = {
{ "SMO8800", 0 },
+ { "SMO8801", 0 },
{ "SMO8810", 0 },
+ { "SMO8811", 0 },
+ { "SMO8820", 0 },
+ { "SMO8821", 0 },
+ { "SMO8830", 0 },
+ { "SMO8831", 0 },
{ "", 0 },
};
@@ -228,6 +234,6 @@
module_acpi_driver(smo8800_driver);
-MODULE_DESCRIPTION("Dell Latitude freefall driver (ACPI SMO8800/SMO8810)");
+MODULE_DESCRIPTION("Dell Latitude freefall driver (ACPI SMO88XX)");
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Sonal Santan, Pali Rohár");
diff --git a/drivers/platform/x86/dell-wmi.c b/drivers/platform/x86/dell-wmi.c
index 25721bf..6512a06 100644
--- a/drivers/platform/x86/dell-wmi.c
+++ b/drivers/platform/x86/dell-wmi.c
@@ -65,10 +65,8 @@
/* Battery health status button */
{ KE_KEY, 0xe007, { KEY_BATTERY } },
- /* This is actually for all radios. Although physically a
- * switch, the notification does not provide an indication of
- * state and so it should be reported as a key */
- { KE_KEY, 0xe008, { KEY_WLAN } },
+ /* Radio devices state change */
+ { KE_IGNORE, 0xe008, { KEY_RFKILL } },
/* The next device is at offset 6, the active devices are at
offset 8 and the attached devices at offset 10 */
@@ -145,57 +143,154 @@
static struct input_dev *dell_wmi_input_dev;
+static void dell_wmi_process_key(int reported_key)
+{
+ const struct key_entry *key;
+
+ key = sparse_keymap_entry_from_scancode(dell_wmi_input_dev,
+ reported_key);
+ if (!key) {
+ pr_info("Unknown key %x pressed\n", reported_key);
+ return;
+ }
+
+ pr_debug("Key %x pressed\n", reported_key);
+
+ /* Don't report brightness notifications that will also come via ACPI */
+ if ((key->keycode == KEY_BRIGHTNESSUP ||
+ key->keycode == KEY_BRIGHTNESSDOWN) && acpi_video)
+ return;
+
+ sparse_keymap_report_entry(dell_wmi_input_dev, key, 1, true);
+}
+
static void dell_wmi_notify(u32 value, void *context)
{
struct acpi_buffer response = { ACPI_ALLOCATE_BUFFER, NULL };
union acpi_object *obj;
acpi_status status;
+ acpi_size buffer_size;
+ u16 *buffer_entry, *buffer_end;
+ int len, i;
status = wmi_get_event_data(value, &response);
if (status != AE_OK) {
- pr_info("bad event status 0x%x\n", status);
+ pr_warn("bad event status 0x%x\n", status);
return;
}
obj = (union acpi_object *)response.pointer;
-
- if (obj && obj->type == ACPI_TYPE_BUFFER) {
- const struct key_entry *key;
- int reported_key;
- u16 *buffer_entry = (u16 *)obj->buffer.pointer;
- int buffer_size = obj->buffer.length/2;
-
- if (buffer_size >= 2 && dell_new_hk_type && buffer_entry[1] != 0x10) {
- pr_info("Received unknown WMI event (0x%x)\n",
- buffer_entry[1]);
- kfree(obj);
- return;
- }
-
- if (buffer_size >= 3 && (dell_new_hk_type || buffer_entry[1] == 0x0))
- reported_key = (int)buffer_entry[2];
- else if (buffer_size >= 2)
- reported_key = (int)buffer_entry[1] & 0xffff;
- else {
- pr_info("Received unknown WMI event\n");
- kfree(obj);
- return;
- }
-
- key = sparse_keymap_entry_from_scancode(dell_wmi_input_dev,
- reported_key);
- if (!key) {
- pr_info("Unknown key %x pressed\n", reported_key);
- } else if ((key->keycode == KEY_BRIGHTNESSUP ||
- key->keycode == KEY_BRIGHTNESSDOWN) && acpi_video) {
- /* Don't report brightness notifications that will also
- * come via ACPI */
- ;
- } else {
- sparse_keymap_report_entry(dell_wmi_input_dev, key,
- 1, true);
- }
+ if (!obj) {
+ pr_warn("no response\n");
+ return;
}
+
+ if (obj->type != ACPI_TYPE_BUFFER) {
+ pr_warn("bad response type %x\n", obj->type);
+ kfree(obj);
+ return;
+ }
+
+ pr_debug("Received WMI event (%*ph)\n",
+ obj->buffer.length, obj->buffer.pointer);
+
+ buffer_entry = (u16 *)obj->buffer.pointer;
+ buffer_size = obj->buffer.length/2;
+
+ if (!dell_new_hk_type) {
+ if (buffer_size >= 3 && buffer_entry[1] == 0x0)
+ dell_wmi_process_key(buffer_entry[2]);
+ else if (buffer_size >= 2)
+ dell_wmi_process_key(buffer_entry[1]);
+ else
+ pr_info("Received unknown WMI event\n");
+ kfree(obj);
+ return;
+ }
+
+ buffer_end = buffer_entry + buffer_size;
+
+ while (buffer_entry < buffer_end) {
+
+ len = buffer_entry[0];
+ if (len == 0)
+ break;
+
+ len++;
+
+ if (buffer_entry + len > buffer_end) {
+ pr_warn("Invalid length of WMI event\n");
+ break;
+ }
+
+ pr_debug("Process buffer (%*ph)\n", len*2, buffer_entry);
+
+ switch (buffer_entry[1]) {
+ case 0x00:
+ for (i = 2; i < len; ++i) {
+ switch (buffer_entry[i]) {
+ case 0xe043:
+ /* NIC Link is Up */
+ pr_debug("NIC Link is Up\n");
+ break;
+ case 0xe044:
+ /* NIC Link is Down */
+ pr_debug("NIC Link is Down\n");
+ break;
+ case 0xe045:
+ /* Unknown event but defined in DSDT */
+ default:
+ /* Unknown event */
+ pr_info("Unknown WMI event type 0x00: "
+ "0x%x\n", (int)buffer_entry[i]);
+ break;
+ }
+ }
+ break;
+ case 0x10:
+ /* Keys pressed */
+ for (i = 2; i < len; ++i)
+ dell_wmi_process_key(buffer_entry[i]);
+ break;
+ case 0x11:
+ for (i = 2; i < len; ++i) {
+ switch (buffer_entry[i]) {
+ case 0xfff0:
+ /* Battery unplugged */
+ pr_debug("Battery unplugged\n");
+ break;
+ case 0xfff1:
+ /* Battery inserted */
+ pr_debug("Battery inserted\n");
+ break;
+ case 0x01e1:
+ case 0x02ea:
+ case 0x02eb:
+ case 0x02ec:
+ case 0x02f6:
+ /* Keyboard backlight level changed */
+ pr_debug("Keyboard backlight level "
+ "changed\n");
+ break;
+ default:
+ /* Unknown event */
+ pr_info("Unknown WMI event type 0x11: "
+ "0x%x\n", (int)buffer_entry[i]);
+ break;
+ }
+ }
+ break;
+ default:
+ /* Unknown event */
+ pr_info("Unknown WMI event type 0x%x\n",
+ (int)buffer_entry[1]);
+ break;
+ }
+
+ buffer_entry += len;
+
+ }
+
kfree(obj);
}
@@ -213,11 +308,16 @@
for (i = 0; i < hotkey_num; i++) {
const struct dell_bios_keymap_entry *bios_entry =
&dell_bios_hotkey_table->keymap[i];
- keymap[i].type = KE_KEY;
- keymap[i].code = bios_entry->scancode;
- keymap[i].keycode = bios_entry->keycode < 256 ?
+ u16 keycode = bios_entry->keycode < 256 ?
bios_to_linux_keycode[bios_entry->keycode] :
KEY_RESERVED;
+
+ if (keycode == KEY_KBDILLUMTOGGLE)
+ keymap[i].type = KE_IGNORE;
+ else
+ keymap[i].type = KE_KEY;
+ keymap[i].code = bios_entry->scancode;
+ keymap[i].keycode = keycode;
}
keymap[hotkey_num].type = KE_END;
diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 5a54d35..844c209 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -417,8 +417,7 @@
switch (value) {
case 0:
if (eeepc->cpufv_disabled)
- pr_warn("cpufv enabled (not officially supported "
- "on this model)\n");
+ pr_warn("cpufv enabled (not officially supported on this model)\n");
eeepc->cpufv_disabled = false;
return count;
case 1:
@@ -580,60 +579,59 @@
mutex_lock(&eeepc->hotplug_lock);
pci_lock_rescan_remove();
- if (eeepc->hotplug_slot) {
- port = acpi_get_pci_dev(handle);
- if (!port) {
- pr_warning("Unable to find port\n");
- goto out_unlock;
- }
+ if (!eeepc->hotplug_slot)
+ goto out_unlock;
- bus = port->subordinate;
-
- if (!bus) {
- pr_warn("Unable to find PCI bus 1?\n");
- goto out_put_dev;
- }
-
- if (pci_bus_read_config_dword(bus, 0, PCI_VENDOR_ID, &l)) {
- pr_err("Unable to read PCI config space?\n");
- goto out_put_dev;
- }
-
- absent = (l == 0xffffffff);
-
- if (blocked != absent) {
- pr_warn("BIOS says wireless lan is %s, "
- "but the pci device is %s\n",
- blocked ? "blocked" : "unblocked",
- absent ? "absent" : "present");
- pr_warn("skipped wireless hotplug as probably "
- "inappropriate for this model\n");
- goto out_put_dev;
- }
-
- if (!blocked) {
- dev = pci_get_slot(bus, 0);
- if (dev) {
- /* Device already present */
- pci_dev_put(dev);
- goto out_put_dev;
- }
- dev = pci_scan_single_device(bus, 0);
- if (dev) {
- pci_bus_assign_resources(bus);
- pci_bus_add_device(dev);
- }
- } else {
- dev = pci_get_slot(bus, 0);
- if (dev) {
- pci_stop_and_remove_bus_device(dev);
- pci_dev_put(dev);
- }
- }
-out_put_dev:
- pci_dev_put(port);
+ port = acpi_get_pci_dev(handle);
+ if (!port) {
+ pr_warning("Unable to find port\n");
+ goto out_unlock;
}
+ bus = port->subordinate;
+
+ if (!bus) {
+ pr_warn("Unable to find PCI bus 1?\n");
+ goto out_put_dev;
+ }
+
+ if (pci_bus_read_config_dword(bus, 0, PCI_VENDOR_ID, &l)) {
+ pr_err("Unable to read PCI config space?\n");
+ goto out_put_dev;
+ }
+
+ absent = (l == 0xffffffff);
+
+ if (blocked != absent) {
+ pr_warn("BIOS says wireless lan is %s, but the pci device is %s\n",
+ blocked ? "blocked" : "unblocked",
+ absent ? "absent" : "present");
+ pr_warn("skipped wireless hotplug as probably inappropriate for this model\n");
+ goto out_put_dev;
+ }
+
+ if (!blocked) {
+ dev = pci_get_slot(bus, 0);
+ if (dev) {
+ /* Device already present */
+ pci_dev_put(dev);
+ goto out_put_dev;
+ }
+ dev = pci_scan_single_device(bus, 0);
+ if (dev) {
+ pci_bus_assign_resources(bus);
+ pci_bus_add_device(dev);
+ }
+ } else {
+ dev = pci_get_slot(bus, 0);
+ if (dev) {
+ pci_stop_and_remove_bus_device(dev);
+ pci_dev_put(dev);
+ }
+ }
+out_put_dev:
+ pci_dev_put(port);
+
out_unlock:
pci_unlock_rescan_remove();
mutex_unlock(&eeepc->hotplug_lock);
@@ -821,11 +819,15 @@
return 0;
}
+static char EEEPC_RFKILL_NODE_1[] = "\\_SB.PCI0.P0P5";
+static char EEEPC_RFKILL_NODE_2[] = "\\_SB.PCI0.P0P6";
+static char EEEPC_RFKILL_NODE_3[] = "\\_SB.PCI0.P0P7";
+
static void eeepc_rfkill_exit(struct eeepc_laptop *eeepc)
{
- eeepc_unregister_rfkill_notifier(eeepc, "\\_SB.PCI0.P0P5");
- eeepc_unregister_rfkill_notifier(eeepc, "\\_SB.PCI0.P0P6");
- eeepc_unregister_rfkill_notifier(eeepc, "\\_SB.PCI0.P0P7");
+ eeepc_unregister_rfkill_notifier(eeepc, EEEPC_RFKILL_NODE_1);
+ eeepc_unregister_rfkill_notifier(eeepc, EEEPC_RFKILL_NODE_2);
+ eeepc_unregister_rfkill_notifier(eeepc, EEEPC_RFKILL_NODE_3);
if (eeepc->wlan_rfkill) {
rfkill_unregister(eeepc->wlan_rfkill);
rfkill_destroy(eeepc->wlan_rfkill);
@@ -897,9 +899,9 @@
if (result == -EBUSY)
result = 0;
- eeepc_register_rfkill_notifier(eeepc, "\\_SB.PCI0.P0P5");
- eeepc_register_rfkill_notifier(eeepc, "\\_SB.PCI0.P0P6");
- eeepc_register_rfkill_notifier(eeepc, "\\_SB.PCI0.P0P7");
+ eeepc_register_rfkill_notifier(eeepc, EEEPC_RFKILL_NODE_1);
+ eeepc_register_rfkill_notifier(eeepc, EEEPC_RFKILL_NODE_2);
+ eeepc_register_rfkill_notifier(eeepc, EEEPC_RFKILL_NODE_3);
exit:
if (result && result != -ENODEV)
@@ -915,7 +917,7 @@
struct eeepc_laptop *eeepc = dev_get_drvdata(device);
if (eeepc->wlan_rfkill) {
- bool wlan;
+ int wlan;
/*
* Work around bios bug - acpi _PTS turns off the wireless led
@@ -923,7 +925,8 @@
* we should kick it ourselves in case hibernation is aborted.
*/
wlan = get_acpi(eeepc, CM_ASL_WLAN);
- set_acpi(eeepc, CM_ASL_WLAN, wlan);
+ if (wlan >= 0)
+ set_acpi(eeepc, CM_ASL_WLAN, wlan);
}
return 0;
@@ -935,9 +938,9 @@
/* Refresh both wlan rfkill state and pci hotplug */
if (eeepc->wlan_rfkill) {
- eeepc_rfkill_hotplug_update(eeepc, "\\_SB.PCI0.P0P5");
- eeepc_rfkill_hotplug_update(eeepc, "\\_SB.PCI0.P0P6");
- eeepc_rfkill_hotplug_update(eeepc, "\\_SB.PCI0.P0P7");
+ eeepc_rfkill_hotplug_update(eeepc, EEEPC_RFKILL_NODE_1);
+ eeepc_rfkill_hotplug_update(eeepc, EEEPC_RFKILL_NODE_2);
+ eeepc_rfkill_hotplug_update(eeepc, EEEPC_RFKILL_NODE_3);
}
if (eeepc->bluetooth_rfkill)
@@ -977,18 +980,28 @@
#define EEEPC_EC_SFB0 0xD0
#define EEEPC_EC_FAN_CTRL (EEEPC_EC_SFB0 + 3) /* Byte containing SF25 */
+static inline int eeepc_pwm_to_lmsensors(int value)
+{
+ return value * 255 / 100;
+}
+
+static inline int eeepc_lmsensors_to_pwm(int value)
+{
+ value = clamp_val(value, 0, 255);
+ return value * 100 / 255;
+}
+
static int eeepc_get_fan_pwm(void)
{
u8 value = 0;
ec_read(EEEPC_EC_FAN_PWM, &value);
- return value * 255 / 100;
+ return eeepc_pwm_to_lmsensors(value);
}
static void eeepc_set_fan_pwm(int value)
{
- value = clamp_val(value, 0, 255);
- value = value * 100 / 255;
+ value = eeepc_lmsensors_to_pwm(value);
ec_write(EEEPC_EC_FAN_PWM, value);
}
@@ -1002,15 +1015,19 @@
return high << 8 | low;
}
+#define EEEPC_EC_FAN_CTRL_BIT 0x02
+#define EEEPC_FAN_CTRL_MANUAL 1
+#define EEEPC_FAN_CTRL_AUTO 2
+
static int eeepc_get_fan_ctrl(void)
{
u8 value = 0;
ec_read(EEEPC_EC_FAN_CTRL, &value);
- if (value & 0x02)
- return 1; /* manual */
+ if (value & EEEPC_EC_FAN_CTRL_BIT)
+ return EEEPC_FAN_CTRL_MANUAL;
else
- return 2; /* automatic */
+ return EEEPC_FAN_CTRL_AUTO;
}
static void eeepc_set_fan_ctrl(int manual)
@@ -1018,10 +1035,10 @@
u8 value = 0;
ec_read(EEEPC_EC_FAN_CTRL, &value);
- if (manual == 1)
- value |= 0x02;
+ if (manual == EEEPC_FAN_CTRL_MANUAL)
+ value |= EEEPC_EC_FAN_CTRL_BIT;
else
- value &= ~0x02;
+ value &= ~EEEPC_EC_FAN_CTRL_BIT;
ec_write(EEEPC_EC_FAN_CTRL, value);
}
@@ -1156,8 +1173,7 @@
static void eeepc_backlight_exit(struct eeepc_laptop *eeepc)
{
- if (eeepc->backlight_device)
- backlight_device_unregister(eeepc->backlight_device);
+ backlight_device_unregister(eeepc->backlight_device);
eeepc->backlight_device = NULL;
}
@@ -1216,7 +1232,7 @@
static void eeepc_input_notify(struct eeepc_laptop *eeepc, int event)
{
if (!eeepc->inputdev)
- return ;
+ return;
if (!sparse_keymap_report_event(eeepc->inputdev, event, 1, true))
pr_info("Unknown key %x pressed\n", event);
}
@@ -1224,6 +1240,7 @@
static void eeepc_acpi_notify(struct acpi_device *device, u32 event)
{
struct eeepc_laptop *eeepc = acpi_driver_data(device);
+ int old_brightness, new_brightness;
u16 count;
if (event > ACPI_MAX_SYS_NOTIFY)
@@ -1234,34 +1251,32 @@
count);
/* Brightness events are special */
- if (event >= NOTIFY_BRN_MIN && event <= NOTIFY_BRN_MAX) {
-
- /* Ignore them completely if the acpi video driver is used */
- if (eeepc->backlight_device != NULL) {
- int old_brightness, new_brightness;
-
- /* Update the backlight device. */
- old_brightness = eeepc_backlight_notify(eeepc);
-
- /* Convert event to keypress (obsolescent hack) */
- new_brightness = event - NOTIFY_BRN_MIN;
-
- if (new_brightness < old_brightness) {
- event = NOTIFY_BRN_MIN; /* brightness down */
- } else if (new_brightness > old_brightness) {
- event = NOTIFY_BRN_MAX; /* brightness up */
- } else {
- /*
- * no change in brightness - already at min/max,
- * event will be desired value (or else ignored)
- */
- }
- eeepc_input_notify(eeepc, event);
- }
- } else {
- /* Everything else is a bona-fide keypress event */
+ if (event < NOTIFY_BRN_MIN || event > NOTIFY_BRN_MAX) {
eeepc_input_notify(eeepc, event);
+ return;
}
+
+ /* Ignore them completely if the acpi video driver is used */
+ if (!eeepc->backlight_device)
+ return;
+
+ /* Update the backlight device. */
+ old_brightness = eeepc_backlight_notify(eeepc);
+
+ /* Convert event to keypress (obsolescent hack) */
+ new_brightness = event - NOTIFY_BRN_MIN;
+
+ if (new_brightness < old_brightness) {
+ event = NOTIFY_BRN_MIN; /* brightness down */
+ } else if (new_brightness > old_brightness) {
+ event = NOTIFY_BRN_MAX; /* brightness up */
+ } else {
+ /*
+ * no change in brightness - already at min/max,
+ * event will be desired value (or else ignored)
+ */
+ }
+ eeepc_input_notify(eeepc, event);
}
static void eeepc_dmi_check(struct eeepc_laptop *eeepc)
@@ -1293,8 +1308,8 @@
*/
if (strcmp(model, "701") == 0 || strcmp(model, "702") == 0) {
eeepc->cpufv_disabled = true;
- pr_info("model %s does not officially support setting cpu "
- "speed\n", model);
+ pr_info("model %s does not officially support setting cpu speed\n",
+ model);
pr_info("cpufv disabled to avoid instability\n");
}
@@ -1320,8 +1335,8 @@
Check if cm_getv[cm] works and, if yes, assume cm should be set. */
if (!(eeepc->cm_supported & (1 << cm))
&& !read_acpi_int(eeepc->handle, cm_getv[cm], &dummy)) {
- pr_info("%s (%x) not reported by BIOS,"
- " enabling anyway\n", name, 1 << cm);
+ pr_info("%s (%x) not reported by BIOS, enabling anyway\n",
+ name, 1 << cm);
eeepc->cm_supported |= 1 << cm;
}
}
diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c
index be55bd7..7c21c1c4 100644
--- a/drivers/platform/x86/fujitsu-laptop.c
+++ b/drivers/platform/x86/fujitsu-laptop.c
@@ -1153,8 +1153,7 @@
fail_hotkey:
platform_driver_unregister(&fujitsupf_driver);
fail_backlight:
- if (fujitsu->bl_device)
- backlight_device_unregister(fujitsu->bl_device);
+ backlight_device_unregister(fujitsu->bl_device);
fail_sysfs_group:
sysfs_remove_group(&fujitsu->pf_device->dev.kobj,
&fujitsupf_attribute_group);
@@ -1178,8 +1177,7 @@
platform_driver_unregister(&fujitsupf_driver);
- if (fujitsu->bl_device)
- backlight_device_unregister(fujitsu->bl_device);
+ backlight_device_unregister(fujitsu->bl_device);
sysfs_remove_group(&fujitsu->pf_device->dev.kobj,
&fujitsupf_attribute_group);
diff --git a/drivers/platform/x86/hp-wireless.c b/drivers/platform/x86/hp-wireless.c
index 415348f..4e4cc8b 100644
--- a/drivers/platform/x86/hp-wireless.c
+++ b/drivers/platform/x86/hp-wireless.c
@@ -85,6 +85,9 @@
int err;
err = hp_wireless_input_setup();
+ if (err)
+ pr_err("Failed to setup hp wireless hotkeys\n");
+
return err;
}
diff --git a/drivers/platform/x86/hp_accel.c b/drivers/platform/x86/hp_accel.c
index 6bec745..10ce6cb 100644
--- a/drivers/platform/x86/hp_accel.c
+++ b/drivers/platform/x86/hp_accel.c
@@ -246,6 +246,7 @@
AXIS_DMI_MATCH("HPB64xx", "HP ProBook 64", xy_swap),
AXIS_DMI_MATCH("HPB64xx", "HP EliteBook 84", xy_swap),
AXIS_DMI_MATCH("HPB65xx", "HP ProBook 65", x_inverted),
+ AXIS_DMI_MATCH("HPZBook15", "HP ZBook 15", x_inverted),
{ NULL, }
/* Laptop models without axis info (yet):
* "NC6910" "HP Compaq 6910"
diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c
index c860eac..b3d419a 100644
--- a/drivers/platform/x86/ideapad-laptop.c
+++ b/drivers/platform/x86/ideapad-laptop.c
@@ -729,8 +729,7 @@
static void ideapad_backlight_exit(struct ideapad_private *priv)
{
- if (priv->blightdev)
- backlight_device_unregister(priv->blightdev);
+ backlight_device_unregister(priv->blightdev);
priv->blightdev = NULL;
}
diff --git a/drivers/platform/x86/intel_ips.c b/drivers/platform/x86/intel_ips.c
index ecd36e3..e2065e0 100644
--- a/drivers/platform/x86/intel_ips.c
+++ b/drivers/platform/x86/intel_ips.c
@@ -33,7 +33,7 @@
* performance by allocating more power or thermal budget to the CPU or GPU
* based on available headroom and activity.
*
- * The basic algorithm is driven by a 5s moving average of tempurature. If
+ * The basic algorithm is driven by a 5s moving average of temperature. If
* thermal headroom is available, the CPU and/or GPU power clamps may be
* adjusted upwards. If we hit the thermal ceiling or a thermal trigger,
* we scale back the clamp. Aside from trigger events (when we're critically
diff --git a/drivers/platform/x86/intel_oaktrail.c b/drivers/platform/x86/intel_oaktrail.c
index 0afaaef..a4a4258 100644
--- a/drivers/platform/x86/intel_oaktrail.c
+++ b/drivers/platform/x86/intel_oaktrail.c
@@ -271,8 +271,7 @@
static void oaktrail_backlight_exit(void)
{
- if (oaktrail_bl_device)
- backlight_device_unregister(oaktrail_bl_device);
+ backlight_device_unregister(oaktrail_bl_device);
}
static int oaktrail_probe(struct platform_device *pdev)
diff --git a/drivers/platform/x86/msi-laptop.c b/drivers/platform/x86/msi-laptop.c
index a3f06cb..0859877 100644
--- a/drivers/platform/x86/msi-laptop.c
+++ b/drivers/platform/x86/msi-laptop.c
@@ -820,7 +820,7 @@
{
static bool extended;
- if (str & 0x20)
+ if (str & I8042_STR_AUXDATA)
return false;
/* 0x54 wwan, 0x62 bluetooth, 0x76 wlan, 0xE4 touchpad toggle*/
diff --git a/drivers/platform/x86/msi-wmi.c b/drivers/platform/x86/msi-wmi.c
index 70222f2..6d2bac0 100644
--- a/drivers/platform/x86/msi-wmi.c
+++ b/drivers/platform/x86/msi-wmi.c
@@ -354,8 +354,7 @@
sparse_keymap_free(msi_wmi_input_dev);
input_unregister_device(msi_wmi_input_dev);
}
- if (backlight)
- backlight_device_unregister(backlight);
+ backlight_device_unregister(backlight);
}
module_init(msi_wmi_init);
diff --git a/drivers/platform/x86/sony-laptop.c b/drivers/platform/x86/sony-laptop.c
index a1a0fd7..6dd1c0e 100644
--- a/drivers/platform/x86/sony-laptop.c
+++ b/drivers/platform/x86/sony-laptop.c
@@ -3140,8 +3140,7 @@
static void sony_nc_backlight_cleanup(void)
{
- if (sony_bl_props.dev)
- backlight_device_unregister(sony_bl_props.dev);
+ backlight_device_unregister(sony_bl_props.dev);
}
static int sony_nc_add(struct acpi_device *device)
@@ -3716,8 +3715,7 @@
dev->event_types = type2_events;
out:
- if (pcidev)
- pci_dev_put(pcidev);
+ pci_dev_put(pcidev);
pr_info("detected Type%d model\n",
dev->model == SONYPI_DEVICE_TYPE1 ? 1 :
diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index 6414cfe..c3d11fa 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -6557,6 +6557,17 @@
* bits 3-0 (volume). Other bits in NVRAM may have other functions,
* such as bit 7 which is used to detect repeated presses of MUTE,
* and we leave them unchanged.
+ *
+ * On newer Lenovo ThinkPads, the EC can automatically change the volume
+ * in response to user input. Unfortunately, this rarely works well.
+ * The laptop changes the state of its internal MUTE gate and, on some
+ * models, sends KEY_MUTE, causing any user code that responds to the
+ * mute button to get confused. The hardware MUTE gate is also
+ * unnecessary, since user code can handle the mute button without
+ * kernel or EC help.
+ *
+ * To avoid confusing userspace, we simply disable all EC-based mute
+ * and volume controls when possible.
*/
#ifdef CONFIG_THINKPAD_ACPI_ALSA_SUPPORT
@@ -6611,11 +6622,21 @@
TPACPI_VOL_CAP_MAX
};
+enum tpacpi_mute_btn_mode {
+ TP_EC_MUTE_BTN_LATCH = 0, /* Mute mutes; up/down unmutes */
+ /* We don't know what mode 1 is. */
+ TP_EC_MUTE_BTN_NONE = 2, /* Mute and up/down are just keys */
+ TP_EC_MUTE_BTN_TOGGLE = 3, /* Mute toggles; up/down unmutes */
+};
+
static enum tpacpi_volume_access_mode volume_mode =
TPACPI_VOL_MODE_MAX;
static enum tpacpi_volume_capabilities volume_capabilities;
static bool volume_control_allowed;
+static bool software_mute_requested = true;
+static bool software_mute_active;
+static int software_mute_orig_mode;
/*
* Used to syncronize writers to TP_EC_AUDIO and
@@ -6633,6 +6654,8 @@
return;
if (!volume_control_allowed)
return;
+ if (software_mute_active)
+ return;
vdbg_printk(TPACPI_DBG_MIXER,
"trying to checkpoint mixer state to NVRAM...\n");
@@ -6694,6 +6717,12 @@
dbg_printk(TPACPI_DBG_MIXER, "set EC mixer to 0x%02x\n", status);
+ /*
+ * On X200s, and possibly on others, it can take a while for
+ * reads to become correct.
+ */
+ msleep(1);
+
return 0;
}
@@ -6776,6 +6805,57 @@
return rc;
}
+static int volume_set_software_mute(bool startup)
+{
+ int result;
+
+ if (!tpacpi_is_lenovo())
+ return -ENODEV;
+
+ if (startup) {
+ if (!acpi_evalf(ec_handle, &software_mute_orig_mode,
+ "HAUM", "qd"))
+ return -EIO;
+
+ dbg_printk(TPACPI_DBG_INIT | TPACPI_DBG_MIXER,
+ "Initial HAUM setting was %d\n",
+ software_mute_orig_mode);
+ }
+
+ if (!acpi_evalf(ec_handle, &result, "SAUM", "qdd",
+ (int)TP_EC_MUTE_BTN_NONE))
+ return -EIO;
+
+ if (result != TP_EC_MUTE_BTN_NONE)
+ pr_warn("Unexpected SAUM result %d\n",
+ result);
+
+ /*
+ * In software mute mode, the standard codec controls take
+ * precendence, so we unmute the ThinkPad HW switch at
+ * startup. Just on case there are SAUM-capable ThinkPads
+ * with level controls, set max HW volume as well.
+ */
+ if (tp_features.mixer_no_level_control)
+ result = volume_set_mute(false);
+ else
+ result = volume_set_status(TP_EC_VOLUME_MAX);
+
+ if (result != 0)
+ pr_warn("Failed to unmute the HW mute switch\n");
+
+ return 0;
+}
+
+static void volume_exit_software_mute(void)
+{
+ int r;
+
+ if (!acpi_evalf(ec_handle, &r, "SAUM", "qdd", software_mute_orig_mode)
+ || r != software_mute_orig_mode)
+ pr_warn("Failed to restore mute mode\n");
+}
+
static int volume_alsa_set_volume(const u8 vol)
{
dbg_printk(TPACPI_DBG_MIXER,
@@ -6883,7 +6963,12 @@
static void volume_resume(void)
{
- volume_alsa_notify_change();
+ if (software_mute_active) {
+ if (volume_set_software_mute(false) < 0)
+ pr_warn("Failed to restore software mute\n");
+ } else {
+ volume_alsa_notify_change();
+ }
}
static void volume_shutdown(void)
@@ -6899,6 +6984,9 @@
}
tpacpi_volume_checkpoint_nvram();
+
+ if (software_mute_active)
+ volume_exit_software_mute();
}
static int __init volume_create_alsa_mixer(void)
@@ -7083,16 +7171,20 @@
"mute is supported, volume control is %s\n",
str_supported(!tp_features.mixer_no_level_control));
- rc = volume_create_alsa_mixer();
- if (rc) {
- pr_err("Could not create the ALSA mixer interface\n");
- return rc;
- }
+ if (software_mute_requested && volume_set_software_mute(true) == 0) {
+ software_mute_active = true;
+ } else {
+ rc = volume_create_alsa_mixer();
+ if (rc) {
+ pr_err("Could not create the ALSA mixer interface\n");
+ return rc;
+ }
- pr_info("Console audio control enabled, mode: %s\n",
- (volume_control_allowed) ?
- "override (read/write)" :
- "monitor (read only)");
+ pr_info("Console audio control enabled, mode: %s\n",
+ (volume_control_allowed) ?
+ "override (read/write)" :
+ "monitor (read only)");
+ }
vdbg_printk(TPACPI_DBG_INIT | TPACPI_DBG_MIXER,
"registering volume hotkeys as change notification\n");
@@ -9089,6 +9181,10 @@
"Enables software override for the console audio "
"control when true");
+module_param_named(software_mute, software_mute_requested, bool, 0444);
+MODULE_PARM_DESC(software_mute,
+ "Request full software mute control");
+
/* ALSA module API parameters */
module_param_named(index, alsa_index, int, 0444);
MODULE_PARM_DESC(index, "ALSA index for the ACPI EC Mixer");
diff --git a/drivers/platform/x86/toshiba_acpi.c b/drivers/platform/x86/toshiba_acpi.c
index ab6151f..fc34a71 100644
--- a/drivers/platform/x86/toshiba_acpi.c
+++ b/drivers/platform/x86/toshiba_acpi.c
@@ -186,6 +186,7 @@
static const struct acpi_device_id toshiba_device_ids[] = {
{"TOS6200", 0},
+ {"TOS6207", 0},
{"TOS6208", 0},
{"TOS1900", 0},
{"", 0},
@@ -928,9 +929,7 @@
static int set_lcd_brightness(struct toshiba_acpi_dev *dev, int value)
{
- u32 in[TCI_WORDS] = { HCI_SET, HCI_LCD_BRIGHTNESS, 0, 0, 0, 0 };
- u32 out[TCI_WORDS];
- acpi_status status;
+ u32 hci_result;
if (dev->tr_backlight_supported) {
bool enable = !value;
@@ -941,20 +940,9 @@
value--;
}
- in[2] = value << HCI_LCD_BRIGHTNESS_SHIFT;
- status = tci_raw(dev, in, out);
- if (ACPI_FAILURE(status) || out[0] == TOS_FAILURE) {
- pr_err("ACPI call to set brightness failed");
- return -EIO;
- }
- /* Extra check for "incomplete" backlight method, where the AML code
- * doesn't check for HCI_SET or HCI_GET and returns TOS_SUCCESS,
- * the actual brightness, and in some cases the max brightness.
- */
- if (out[2] > 0 || out[3] == 0xE000)
- return -ENODEV;
-
- return out[0] == TOS_SUCCESS ? 0 : -EIO;
+ value = value << HCI_LCD_BRIGHTNESS_SHIFT;
+ hci_result = hci_write1(dev, HCI_LCD_BRIGHTNESS, value);
+ return hci_result == TOS_SUCCESS ? 0 : -EIO;
}
static int set_lcd_status(struct backlight_device *bd)
@@ -1406,12 +1394,6 @@
if (ret)
return ret;
- /* Update sysfs entries on successful mode change*/
- ret = sysfs_update_group(&toshiba->acpi_dev->dev.kobj,
- &toshiba_attr_group);
- if (ret)
- return ret;
-
toshiba->kbd_mode = mode;
}
@@ -1586,10 +1568,32 @@
return exists ? attr->mode : 0;
}
+/*
+ * Hotkeys
+ */
+static int toshiba_acpi_enable_hotkeys(struct toshiba_acpi_dev *dev)
+{
+ acpi_status status;
+ u32 result;
+
+ status = acpi_evaluate_object(dev->acpi_dev->handle,
+ "ENAB", NULL, NULL);
+ if (ACPI_FAILURE(status))
+ return -ENODEV;
+
+ result = hci_write1(dev, HCI_HOTKEY_EVENT, HCI_HOTKEY_ENABLE);
+ if (result == TOS_FAILURE)
+ return -EIO;
+ else if (result == TOS_NOT_SUPPORTED)
+ return -ENODEV;
+
+ return 0;
+}
+
static bool toshiba_acpi_i8042_filter(unsigned char data, unsigned char str,
struct serio *port)
{
- if (str & 0x20)
+ if (str & I8042_STR_AUXDATA)
return false;
if (unlikely(data == 0xe0))
@@ -1648,9 +1652,45 @@
pr_info("Unknown key %x\n", scancode);
}
+static void toshiba_acpi_process_hotkeys(struct toshiba_acpi_dev *dev)
+{
+ u32 hci_result, value;
+ int retries = 3;
+ int scancode;
+
+ if (dev->info_supported) {
+ scancode = toshiba_acpi_query_hotkey(dev);
+ if (scancode < 0)
+ pr_err("Failed to query hotkey event\n");
+ else if (scancode != 0)
+ toshiba_acpi_report_hotkey(dev, scancode);
+ } else if (dev->system_event_supported) {
+ do {
+ hci_result = hci_read1(dev, HCI_SYSTEM_EVENT, &value);
+ switch (hci_result) {
+ case TOS_SUCCESS:
+ toshiba_acpi_report_hotkey(dev, (int)value);
+ break;
+ case TOS_NOT_SUPPORTED:
+ /*
+ * This is a workaround for an unresolved
+ * issue on some machines where system events
+ * sporadically become disabled.
+ */
+ hci_result =
+ hci_write1(dev, HCI_SYSTEM_EVENT, 1);
+ pr_notice("Re-enabled hotkeys\n");
+ /* fall through */
+ default:
+ retries--;
+ break;
+ }
+ } while (retries && hci_result != TOS_FIFO_EMPTY);
+ }
+}
+
static int toshiba_acpi_setup_keyboard(struct toshiba_acpi_dev *dev)
{
- acpi_status status;
acpi_handle ec_handle;
int error;
u32 hci_result;
@@ -1677,7 +1717,6 @@
* supported, so if it's present set up an i8042 key filter
* for this purpose.
*/
- status = AE_ERROR;
ec_handle = ec_get_handle();
if (ec_handle && acpi_has_method(ec_handle, "NTFY")) {
INIT_WORK(&dev->hotkey_work, toshiba_acpi_hotkey_work);
@@ -1708,10 +1747,9 @@
goto err_remove_filter;
}
- status = acpi_evaluate_object(dev->acpi_dev->handle, "ENAB", NULL, NULL);
- if (ACPI_FAILURE(status)) {
+ error = toshiba_acpi_enable_hotkeys(dev);
+ if (error) {
pr_info("Unable to enable hotkeys\n");
- error = -ENODEV;
goto err_remove_filter;
}
@@ -1721,7 +1759,6 @@
goto err_remove_filter;
}
- hci_result = hci_write1(dev, HCI_HOTKEY_EVENT, HCI_HOTKEY_ENABLE);
return 0;
err_remove_filter:
@@ -1810,8 +1847,7 @@
rfkill_destroy(dev->bt_rfk);
}
- if (dev->backlight_dev)
- backlight_device_unregister(dev->backlight_dev);
+ backlight_device_unregister(dev->backlight_dev);
if (dev->illumination_supported)
led_classdev_unregister(&dev->led_dev);
@@ -1967,41 +2003,29 @@
static void toshiba_acpi_notify(struct acpi_device *acpi_dev, u32 event)
{
struct toshiba_acpi_dev *dev = acpi_driver_data(acpi_dev);
- u32 hci_result, value;
- int retries = 3;
- int scancode;
+ int ret;
- if (event != 0x80)
- return;
-
- if (dev->info_supported) {
- scancode = toshiba_acpi_query_hotkey(dev);
- if (scancode < 0)
- pr_err("Failed to query hotkey event\n");
- else if (scancode != 0)
- toshiba_acpi_report_hotkey(dev, scancode);
- } else if (dev->system_event_supported) {
- do {
- hci_result = hci_read1(dev, HCI_SYSTEM_EVENT, &value);
- switch (hci_result) {
- case TOS_SUCCESS:
- toshiba_acpi_report_hotkey(dev, (int)value);
- break;
- case TOS_NOT_SUPPORTED:
- /*
- * This is a workaround for an unresolved
- * issue on some machines where system events
- * sporadically become disabled.
- */
- hci_result =
- hci_write1(dev, HCI_SYSTEM_EVENT, 1);
- pr_notice("Re-enabled hotkeys\n");
- /* fall through */
- default:
- retries--;
- break;
- }
- } while (retries && hci_result != TOS_FIFO_EMPTY);
+ switch (event) {
+ case 0x80: /* Hotkeys and some system events */
+ toshiba_acpi_process_hotkeys(dev);
+ break;
+ case 0x92: /* Keyboard backlight mode changed */
+ /* Update sysfs entries */
+ ret = sysfs_update_group(&acpi_dev->dev.kobj,
+ &toshiba_attr_group);
+ if (ret)
+ pr_err("Unable to update sysfs entries\n");
+ break;
+ case 0x81: /* Unknown */
+ case 0x82: /* Unknown */
+ case 0x83: /* Unknown */
+ case 0x8c: /* Unknown */
+ case 0x8e: /* Unknown */
+ case 0x8f: /* Unknown */
+ case 0x90: /* Unknown */
+ default:
+ pr_info("Unknown event received %x\n", event);
+ break;
}
}
@@ -2020,16 +2044,12 @@
static int toshiba_acpi_resume(struct device *device)
{
struct toshiba_acpi_dev *dev = acpi_driver_data(to_acpi_device(device));
- u32 result;
- acpi_status status;
+ int error;
if (dev->hotkey_dev) {
- status = acpi_evaluate_object(dev->acpi_dev->handle, "ENAB",
- NULL, NULL);
- if (ACPI_FAILURE(status))
+ error = toshiba_acpi_enable_hotkeys(dev);
+ if (error)
pr_info("Unable to re-enable hotkeys\n");
-
- result = hci_write1(dev, HCI_HOTKEY_EVENT, HCI_HOTKEY_ENABLE);
}
return 0;
diff --git a/drivers/power/pm2301_charger.c b/drivers/power/pm2301_charger.c
index 62c15af..7773249 100644
--- a/drivers/power/pm2301_charger.c
+++ b/drivers/power/pm2301_charger.c
@@ -951,8 +951,6 @@
#endif
-#ifdef CONFIG_PM_RUNTIME
-
static int pm2xxx_runtime_suspend(struct device *dev)
{
struct i2c_client *pm2xxx_i2c_client = to_i2c_client(dev);
@@ -977,8 +975,6 @@
return 0;
}
-#endif
-
static const struct dev_pm_ops pm2xxx_pm_ops = {
SET_SYSTEM_SLEEP_PM_OPS(pm2xxx_wall_charger_suspend,
pm2xxx_wall_charger_resume)
diff --git a/drivers/scsi/scsi_pm.c b/drivers/scsi/scsi_pm.c
index 7454498..9e43ae1 100644
--- a/drivers/scsi/scsi_pm.c
+++ b/drivers/scsi/scsi_pm.c
@@ -213,8 +213,6 @@
#endif /* CONFIG_PM_SLEEP */
-#ifdef CONFIG_PM_RUNTIME
-
static int sdev_runtime_suspend(struct device *dev)
{
const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL;
@@ -332,14 +330,6 @@
pm_runtime_put_sync(&shost->shost_gendev);
}
-#else
-
-#define scsi_runtime_suspend NULL
-#define scsi_runtime_resume NULL
-#define scsi_runtime_idle NULL
-
-#endif /* CONFIG_PM_RUNTIME */
-
const struct dev_pm_ops scsi_bus_pm_ops = {
.prepare = scsi_bus_prepare,
.suspend = scsi_bus_suspend,
diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index 2dc4a83..e3902fc 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -155,8 +155,7 @@
/* scsi_pm.c */
#ifdef CONFIG_PM
extern const struct dev_pm_ops scsi_bus_pm_ops;
-#endif
-#ifdef CONFIG_PM_RUNTIME
+
extern void scsi_autopm_get_target(struct scsi_target *);
extern void scsi_autopm_put_target(struct scsi_target *);
extern int scsi_autopm_get_host(struct Scsi_Host *);
@@ -166,7 +165,7 @@
static inline void scsi_autopm_put_target(struct scsi_target *t) {}
static inline int scsi_autopm_get_host(struct Scsi_Host *h) { return 0; }
static inline void scsi_autopm_put_host(struct Scsi_Host *h) {}
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
extern struct async_domain scsi_sd_pm_domain;
extern struct async_domain scsi_sd_probe_domain;
diff --git a/drivers/scsi/ufs/ufshcd-pci.c b/drivers/scsi/ufs/ufshcd-pci.c
index 955ed55..d15eaa4 100644
--- a/drivers/scsi/ufs/ufshcd-pci.c
+++ b/drivers/scsi/ufs/ufshcd-pci.c
@@ -62,12 +62,7 @@
{
return ufshcd_system_resume(dev_get_drvdata(dev));
}
-#else
-#define ufshcd_pci_suspend NULL
-#define ufshcd_pci_resume NULL
-#endif /* CONFIG_PM */
-#ifdef CONFIG_PM_RUNTIME
static int ufshcd_pci_runtime_suspend(struct device *dev)
{
return ufshcd_runtime_suspend(dev_get_drvdata(dev));
@@ -80,11 +75,13 @@
{
return ufshcd_runtime_idle(dev_get_drvdata(dev));
}
-#else /* !CONFIG_PM_RUNTIME */
+#else /* !CONFIG_PM */
+#define ufshcd_pci_suspend NULL
+#define ufshcd_pci_resume NULL
#define ufshcd_pci_runtime_suspend NULL
#define ufshcd_pci_runtime_resume NULL
#define ufshcd_pci_runtime_idle NULL
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
/**
* ufshcd_pci_shutdown - main function to put the controller in reset state
diff --git a/drivers/scsi/ufs/ufshcd-pltfrm.c b/drivers/scsi/ufs/ufshcd-pltfrm.c
index 0c030ad..7db9564 100644
--- a/drivers/scsi/ufs/ufshcd-pltfrm.c
+++ b/drivers/scsi/ufs/ufshcd-pltfrm.c
@@ -261,12 +261,7 @@
{
return ufshcd_system_resume(dev_get_drvdata(dev));
}
-#else
-#define ufshcd_pltfrm_suspend NULL
-#define ufshcd_pltfrm_resume NULL
-#endif
-#ifdef CONFIG_PM_RUNTIME
static int ufshcd_pltfrm_runtime_suspend(struct device *dev)
{
return ufshcd_runtime_suspend(dev_get_drvdata(dev));
@@ -279,11 +274,13 @@
{
return ufshcd_runtime_idle(dev_get_drvdata(dev));
}
-#else /* !CONFIG_PM_RUNTIME */
+#else /* !CONFIG_PM */
+#define ufshcd_pltfrm_suspend NULL
+#define ufshcd_pltfrm_resume NULL
#define ufshcd_pltfrm_runtime_suspend NULL
#define ufshcd_pltfrm_runtime_resume NULL
#define ufshcd_pltfrm_runtime_idle NULL
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
static void ufshcd_pltfrm_shutdown(struct platform_device *pdev)
{
diff --git a/drivers/spi/spi-coldfire-qspi.c b/drivers/spi/spi-coldfire-qspi.c
index e2fa628..41b5dc4 100644
--- a/drivers/spi/spi-coldfire-qspi.c
+++ b/drivers/spi/spi-coldfire-qspi.c
@@ -491,7 +491,7 @@
}
#endif
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int mcfqspi_runtime_suspend(struct device *dev)
{
struct spi_master *master = dev_get_drvdata(dev);
diff --git a/drivers/spi/spi-orion.c b/drivers/spi/spi-orion.c
index 932da48..3dec9e0 100644
--- a/drivers/spi/spi-orion.c
+++ b/drivers/spi/spi-orion.c
@@ -523,7 +523,7 @@
MODULE_ALIAS("platform:" DRIVER_NAME);
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int orion_spi_runtime_suspend(struct device *dev)
{
struct spi_master *master = dev_get_drvdata(dev);
diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c
index 2a41b2d..05c623c 100644
--- a/drivers/spi/spi-pxa2xx.c
+++ b/drivers/spi/spi-pxa2xx.c
@@ -1531,7 +1531,7 @@
}
#endif
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int pxa2xx_spi_runtime_suspend(struct device *dev)
{
struct driver_data *drv_data = dev_get_drvdata(dev);
diff --git a/drivers/spi/spi-qup.c b/drivers/spi/spi-qup.c
index 390ed71..e7fb5a0 100644
--- a/drivers/spi/spi-qup.c
+++ b/drivers/spi/spi-qup.c
@@ -646,7 +646,7 @@
return ret;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int spi_qup_pm_suspend_runtime(struct device *device)
{
struct spi_master *master = dev_get_drvdata(device);
@@ -672,7 +672,7 @@
writel_relaxed(config, controller->base + QUP_CONFIG);
return 0;
}
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
#ifdef CONFIG_PM_SLEEP
static int spi_qup_suspend(struct device *device)
diff --git a/drivers/spi/spi-rockchip.c b/drivers/spi/spi-rockchip.c
index 44c1225..daabbab 100644
--- a/drivers/spi/spi-rockchip.c
+++ b/drivers/spi/spi-rockchip.c
@@ -799,7 +799,7 @@
}
#endif /* CONFIG_PM_SLEEP */
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int rockchip_spi_runtime_suspend(struct device *dev)
{
struct spi_master *master = dev_get_drvdata(dev);
@@ -827,7 +827,7 @@
return ret;
}
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
static const struct dev_pm_ops rockchip_spi_pm = {
SET_SYSTEM_SLEEP_PM_OPS(rockchip_spi_suspend, rockchip_spi_resume)
diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c
index 197bcf0..37b1983 100644
--- a/drivers/spi/spi-s3c64xx.c
+++ b/drivers/spi/spi-s3c64xx.c
@@ -1267,7 +1267,7 @@
}
#endif /* CONFIG_PM_SLEEP */
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int s3c64xx_spi_runtime_suspend(struct device *dev)
{
struct spi_master *master = dev_get_drvdata(dev);
@@ -1297,7 +1297,7 @@
return 0;
}
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
static const struct dev_pm_ops s3c64xx_spi_pm = {
SET_SYSTEM_SLEEP_PM_OPS(s3c64xx_spi_suspend, s3c64xx_spi_resume)
diff --git a/drivers/staging/gdm72xx/Kconfig b/drivers/staging/gdm72xx/Kconfig
index 5836503..bf11a7f 100644
--- a/drivers/staging/gdm72xx/Kconfig
+++ b/drivers/staging/gdm72xx/Kconfig
@@ -53,7 +53,7 @@
config WIMAX_GDM72XX_USB_PM
bool "Enable power management support"
- depends on PM_RUNTIME
+ depends on PM
help
Enable USB power management in order to reduce power consumption
while the interface is not in use.
diff --git a/drivers/staging/media/Kconfig b/drivers/staging/media/Kconfig
index 96498b7..2a054a9 100644
--- a/drivers/staging/media/Kconfig
+++ b/drivers/staging/media/Kconfig
@@ -27,12 +27,18 @@
source "drivers/staging/media/dt3155v4l/Kconfig"
+source "drivers/staging/media/tlg2300/Kconfig"
+
source "drivers/staging/media/mn88472/Kconfig"
source "drivers/staging/media/mn88473/Kconfig"
source "drivers/staging/media/omap4iss/Kconfig"
+source "drivers/staging/media/parport/Kconfig"
+
+source "drivers/staging/media/vino/Kconfig"
+
# Keep LIRC at the end, as it has sub-menus
source "drivers/staging/media/lirc/Kconfig"
diff --git a/drivers/staging/media/Makefile b/drivers/staging/media/Makefile
index 30fb352..412b284 100644
--- a/drivers/staging/media/Makefile
+++ b/drivers/staging/media/Makefile
@@ -6,4 +6,7 @@
obj-$(CONFIG_VIDEO_OMAP4) += omap4iss/
obj-$(CONFIG_DVB_MN88472) += mn88472/
obj-$(CONFIG_DVB_MN88473) += mn88473/
+obj-y += parport/
+obj-$(CONFIG_VIDEO_TLG2300) += tlg2300/
+obj-y += vino/
diff --git a/drivers/media/parport/Kconfig b/drivers/staging/media/parport/Kconfig
similarity index 65%
rename from drivers/media/parport/Kconfig
rename to drivers/staging/media/parport/Kconfig
index 948c981..15974ef 100644
--- a/drivers/media/parport/Kconfig
+++ b/drivers/staging/media/parport/Kconfig
@@ -7,18 +7,22 @@
if MEDIA_PARPORT_SUPPORT
config VIDEO_BWQCAM
- tristate "Quickcam BW Video For Linux"
+ tristate "Quickcam BW Video For Linux (Deprecated)"
depends on PARPORT && VIDEO_V4L2
select VIDEOBUF2_VMALLOC
help
Say Y have if you the black and white version of the QuickCam
camera. See the next option for the color version.
+ This driver is deprecated and will be removed soon. If you have
+ hardware for this and you want to work on this driver, then contact
+ the linux-media mailinglist.
+
To compile this driver as a module, choose M here: the
module will be called bw-qcam.
config VIDEO_CQCAM
- tristate "QuickCam Colour Video For Linux"
+ tristate "QuickCam Colour Video For Linux (Deprecated)"
depends on PARPORT && VIDEO_V4L2
help
This is the video4linux driver for the colour version of the
@@ -28,18 +32,26 @@
as a module (c-qcam).
Read <file:Documentation/video4linux/CQcam.txt> for more information.
+ This driver is deprecated and will be removed soon. If you have
+ hardware for this and you want to work on this driver, then contact
+ the linux-media mailinglist.
+
config VIDEO_PMS
- tristate "Mediavision Pro Movie Studio Video For Linux"
+ tristate "Mediavision Pro Movie Studio Video For Linux (Deprecated)"
depends on ISA && VIDEO_V4L2
help
Say Y if you have the ISA Mediavision Pro Movie Studio
capture card.
+ This driver is deprecated and will be removed soon. If you have
+ hardware for this and you want to work on this driver, then contact
+ the linux-media mailinglist.
+
To compile this driver as a module, choose M here: the
module will be called pms.
config VIDEO_W9966
- tristate "W9966CF Webcam (FlyCam Supra and others) Video For Linux"
+ tristate "W9966CF Webcam (FlyCam Supra and others) Video For Linux (Deprecated)"
depends on PARPORT_1284 && PARPORT && VIDEO_V4L2
help
Video4linux driver for Winbond's w9966 based Webcams.
@@ -50,4 +62,8 @@
Check out <file:Documentation/video4linux/w9966.txt> for more
information.
+
+ This driver is deprecated and will be removed soon. If you have
+ hardware for this and you want to work on this driver, then contact
+ the linux-media mailinglist.
endif
diff --git a/drivers/media/parport/Makefile b/drivers/staging/media/parport/Makefile
similarity index 100%
rename from drivers/media/parport/Makefile
rename to drivers/staging/media/parport/Makefile
diff --git a/drivers/media/parport/bw-qcam.c b/drivers/staging/media/parport/bw-qcam.c
similarity index 100%
rename from drivers/media/parport/bw-qcam.c
rename to drivers/staging/media/parport/bw-qcam.c
diff --git a/drivers/media/parport/c-qcam.c b/drivers/staging/media/parport/c-qcam.c
similarity index 100%
rename from drivers/media/parport/c-qcam.c
rename to drivers/staging/media/parport/c-qcam.c
diff --git a/drivers/media/parport/pms.c b/drivers/staging/media/parport/pms.c
similarity index 100%
rename from drivers/media/parport/pms.c
rename to drivers/staging/media/parport/pms.c
diff --git a/drivers/media/parport/w9966.c b/drivers/staging/media/parport/w9966.c
similarity index 100%
rename from drivers/media/parport/w9966.c
rename to drivers/staging/media/parport/w9966.c
diff --git a/drivers/media/usb/tlg2300/Kconfig b/drivers/staging/media/tlg2300/Kconfig
similarity index 63%
rename from drivers/media/usb/tlg2300/Kconfig
rename to drivers/staging/media/tlg2300/Kconfig
index 645d915..81784c6 100644
--- a/drivers/media/usb/tlg2300/Kconfig
+++ b/drivers/staging/media/tlg2300/Kconfig
@@ -1,5 +1,5 @@
config VIDEO_TLG2300
- tristate "Telegent TLG2300 USB video capture support"
+ tristate "Telegent TLG2300 USB video capture support (Deprecated)"
depends on VIDEO_DEV && I2C && SND && DVB_CORE
select VIDEO_TUNER
select VIDEO_TVEEPROM
@@ -12,5 +12,9 @@
This is a video4linux driver for Telegent tlg2300 based TV cards.
The driver supports V4L2, DVB-T and radio.
+ This driver is deprecated and will be removed soon. If you have
+ hardware for this and you want to work on this driver, then contact
+ the linux-media mailinglist.
+
To compile this driver as a module, choose M here: the
module will be called poseidon
diff --git a/drivers/media/usb/tlg2300/Makefile b/drivers/staging/media/tlg2300/Makefile
similarity index 100%
rename from drivers/media/usb/tlg2300/Makefile
rename to drivers/staging/media/tlg2300/Makefile
diff --git a/drivers/media/usb/tlg2300/pd-alsa.c b/drivers/staging/media/tlg2300/pd-alsa.c
similarity index 100%
rename from drivers/media/usb/tlg2300/pd-alsa.c
rename to drivers/staging/media/tlg2300/pd-alsa.c
diff --git a/drivers/media/usb/tlg2300/pd-common.h b/drivers/staging/media/tlg2300/pd-common.h
similarity index 100%
rename from drivers/media/usb/tlg2300/pd-common.h
rename to drivers/staging/media/tlg2300/pd-common.h
diff --git a/drivers/media/usb/tlg2300/pd-dvb.c b/drivers/staging/media/tlg2300/pd-dvb.c
similarity index 100%
rename from drivers/media/usb/tlg2300/pd-dvb.c
rename to drivers/staging/media/tlg2300/pd-dvb.c
diff --git a/drivers/media/usb/tlg2300/pd-main.c b/drivers/staging/media/tlg2300/pd-main.c
similarity index 100%
rename from drivers/media/usb/tlg2300/pd-main.c
rename to drivers/staging/media/tlg2300/pd-main.c
diff --git a/drivers/media/usb/tlg2300/pd-radio.c b/drivers/staging/media/tlg2300/pd-radio.c
similarity index 100%
rename from drivers/media/usb/tlg2300/pd-radio.c
rename to drivers/staging/media/tlg2300/pd-radio.c
diff --git a/drivers/media/usb/tlg2300/pd-video.c b/drivers/staging/media/tlg2300/pd-video.c
similarity index 100%
rename from drivers/media/usb/tlg2300/pd-video.c
rename to drivers/staging/media/tlg2300/pd-video.c
diff --git a/drivers/media/usb/tlg2300/vendorcmds.h b/drivers/staging/media/tlg2300/vendorcmds.h
similarity index 100%
rename from drivers/media/usb/tlg2300/vendorcmds.h
rename to drivers/staging/media/tlg2300/vendorcmds.h
diff --git a/drivers/staging/media/vino/Kconfig b/drivers/staging/media/vino/Kconfig
new file mode 100644
index 0000000..03700da
--- /dev/null
+++ b/drivers/staging/media/vino/Kconfig
@@ -0,0 +1,24 @@
+config VIDEO_VINO
+ tristate "SGI Vino Video For Linux (Deprecated)"
+ depends on I2C && SGI_IP22 && VIDEO_V4L2
+ select VIDEO_SAA7191 if MEDIA_SUBDRV_AUTOSELECT
+ help
+ Say Y here to build in support for the Vino video input system found
+ on SGI Indy machines.
+
+ This driver is deprecated and will be removed soon. If you have
+ hardware for this and you want to work on this driver, then contact
+ the linux-media mailinglist.
+
+config VIDEO_SAA7191
+ tristate "Philips SAA7191 video decoder (Deprecated)"
+ depends on VIDEO_V4L2 && I2C
+ ---help---
+ Support for the Philips SAA7191 video decoder.
+
+ This driver is deprecated and will be removed soon. If you have
+ hardware for this and you want to work on this driver, then contact
+ the linux-media mailinglist.
+
+ To compile this driver as a module, choose M here: the
+ module will be called saa7191.
diff --git a/drivers/staging/media/vino/Makefile b/drivers/staging/media/vino/Makefile
new file mode 100644
index 0000000..914c251
--- /dev/null
+++ b/drivers/staging/media/vino/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_VIDEO_VINO) += indycam.o
+obj-$(CONFIG_VIDEO_VINO) += vino.o
+obj-$(CONFIG_VIDEO_SAA7191) += saa7191.o
diff --git a/drivers/media/platform/indycam.c b/drivers/staging/media/vino/indycam.c
similarity index 100%
rename from drivers/media/platform/indycam.c
rename to drivers/staging/media/vino/indycam.c
diff --git a/drivers/media/platform/indycam.h b/drivers/staging/media/vino/indycam.h
similarity index 100%
rename from drivers/media/platform/indycam.h
rename to drivers/staging/media/vino/indycam.h
diff --git a/drivers/media/i2c/saa7191.c b/drivers/staging/media/vino/saa7191.c
similarity index 100%
rename from drivers/media/i2c/saa7191.c
rename to drivers/staging/media/vino/saa7191.c
diff --git a/drivers/media/i2c/saa7191.h b/drivers/staging/media/vino/saa7191.h
similarity index 100%
rename from drivers/media/i2c/saa7191.h
rename to drivers/staging/media/vino/saa7191.h
diff --git a/drivers/media/platform/vino.c b/drivers/staging/media/vino/vino.c
similarity index 100%
rename from drivers/media/platform/vino.c
rename to drivers/staging/media/vino/vino.c
diff --git a/drivers/media/platform/vino.h b/drivers/staging/media/vino/vino.h
similarity index 100%
rename from drivers/media/platform/vino.h
rename to drivers/staging/media/vino/vino.h
diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c
index b4b58ae..555de07 100644
--- a/drivers/tty/serial/8250/8250_dw.c
+++ b/drivers/tty/serial/8250/8250_dw.c
@@ -530,7 +530,7 @@
}
#endif /* CONFIG_PM_SLEEP */
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int dw8250_runtime_suspend(struct device *dev)
{
struct dw8250_data *data = dev_get_drvdata(dev);
diff --git a/drivers/tty/serial/8250/8250_mtk.c b/drivers/tty/serial/8250/8250_mtk.c
index 6f93123..7a11fac 100644
--- a/drivers/tty/serial/8250/8250_mtk.c
+++ b/drivers/tty/serial/8250/8250_mtk.c
@@ -244,7 +244,7 @@
}
#endif /* CONFIG_PM_SLEEP */
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int mtk8250_runtime_suspend(struct device *dev)
{
struct mtk8250_data *data = dev_get_drvdata(dev);
diff --git a/drivers/tty/serial/mfd.c b/drivers/tty/serial/mfd.c
index e1f4fda..8fe4501 100644
--- a/drivers/tty/serial/mfd.c
+++ b/drivers/tty/serial/mfd.c
@@ -1252,12 +1252,7 @@
}
return 0;
}
-#else
-#define serial_hsu_suspend NULL
-#define serial_hsu_resume NULL
-#endif
-#ifdef CONFIG_PM_RUNTIME
static int serial_hsu_runtime_idle(struct device *dev)
{
pm_schedule_suspend(dev, 500);
@@ -1274,6 +1269,8 @@
return 0;
}
#else
+#define serial_hsu_suspend NULL
+#define serial_hsu_resume NULL
#define serial_hsu_runtime_idle NULL
#define serial_hsu_runtime_suspend NULL
#define serial_hsu_runtime_resume NULL
diff --git a/drivers/tty/serial/msm_serial_hs.c b/drivers/tty/serial/msm_serial_hs.c
index 8abe8ea..62da853 100644
--- a/drivers/tty/serial/msm_serial_hs.c
+++ b/drivers/tty/serial/msm_serial_hs.c
@@ -1792,7 +1792,7 @@
}
module_exit(msm_serial_hs_exit);
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int msm_hs_runtime_idle(struct device *dev)
{
/*
diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c
index 435478a..2e1073d 100644
--- a/drivers/tty/serial/omap-serial.c
+++ b/drivers/tty/serial/omap-serial.c
@@ -1776,7 +1776,7 @@
}
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static void serial_omap_restore_context(struct uart_omap_port *up)
{
if (up->errata & UART_ERRATA_i202_MDR1_ACCESS)
diff --git a/drivers/usb/core/Kconfig b/drivers/usb/core/Kconfig
index 9cfda6a..cc0ced0 100644
--- a/drivers/usb/core/Kconfig
+++ b/drivers/usb/core/Kconfig
@@ -43,7 +43,7 @@
config USB_OTG
bool "OTG support"
- depends on PM_RUNTIME
+ depends on PM
default n
help
The most notable feature of USB OTG is support for a
diff --git a/drivers/usb/phy/Kconfig b/drivers/usb/phy/Kconfig
index 0cd1f44..c6d0c8e74 100644
--- a/drivers/usb/phy/Kconfig
+++ b/drivers/usb/phy/Kconfig
@@ -20,7 +20,7 @@
config FSL_USB2_OTG
bool "Freescale USB OTG Transceiver Driver"
- depends on USB_EHCI_FSL && USB_FSL_USB2 && USB_OTG_FSM && PM_RUNTIME
+ depends on USB_EHCI_FSL && USB_FSL_USB2 && USB_OTG_FSM && PM
select USB_OTG
select USB_PHY
help
@@ -153,7 +153,7 @@
config USB_MV_OTG
tristate "Marvell USB OTG support"
- depends on USB_EHCI_MV && USB_MV_UDC && PM_RUNTIME
+ depends on USB_EHCI_MV && USB_MV_UDC && PM
select USB_OTG
select USB_PHY
help
diff --git a/drivers/usb/storage/Kconfig b/drivers/usb/storage/Kconfig
index 715f299..ec84758 100644
--- a/drivers/usb/storage/Kconfig
+++ b/drivers/usb/storage/Kconfig
@@ -41,7 +41,7 @@
config REALTEK_AUTOPM
bool "Realtek Card Reader autosuspend support"
- depends on USB_STORAGE_REALTEK && PM_RUNTIME
+ depends on USB_STORAGE_REALTEK && PM
default y
config USB_STORAGE_DATAFAB
diff --git a/drivers/vhost/vringh.c b/drivers/vhost/vringh.c
index 5174eba..3bb02c6 100644
--- a/drivers/vhost/vringh.c
+++ b/drivers/vhost/vringh.c
@@ -11,6 +11,7 @@
#include <linux/uaccess.h>
#include <linux/slab.h>
#include <linux/export.h>
+#include <uapi/linux/virtio_config.h>
static __printf(1,2) __cold void vringh_bad(const char *fmt, ...)
{
@@ -28,13 +29,14 @@
/* Returns vring->num if empty, -ve on error. */
static inline int __vringh_get_head(const struct vringh *vrh,
- int (*getu16)(u16 *val, const u16 *p),
+ int (*getu16)(const struct vringh *vrh,
+ u16 *val, const __virtio16 *p),
u16 *last_avail_idx)
{
u16 avail_idx, i, head;
int err;
- err = getu16(&avail_idx, &vrh->vring.avail->idx);
+ err = getu16(vrh, &avail_idx, &vrh->vring.avail->idx);
if (err) {
vringh_bad("Failed to access avail idx at %p",
&vrh->vring.avail->idx);
@@ -49,7 +51,7 @@
i = *last_avail_idx & (vrh->vring.num - 1);
- err = getu16(&head, &vrh->vring.avail->ring[i]);
+ err = getu16(vrh, &head, &vrh->vring.avail->ring[i]);
if (err) {
vringh_bad("Failed to read head: idx %d address %p",
*last_avail_idx, &vrh->vring.avail->ring[i]);
@@ -144,28 +146,32 @@
}
/* No reason for this code to be inline. */
-static int move_to_indirect(int *up_next, u16 *i, void *addr,
+static int move_to_indirect(const struct vringh *vrh,
+ int *up_next, u16 *i, void *addr,
const struct vring_desc *desc,
struct vring_desc **descs, int *desc_max)
{
+ u32 len;
+
/* Indirect tables can't have indirect. */
if (*up_next != -1) {
vringh_bad("Multilevel indirect %u->%u", *up_next, *i);
return -EINVAL;
}
- if (unlikely(desc->len % sizeof(struct vring_desc))) {
+ len = vringh32_to_cpu(vrh, desc->len);
+ if (unlikely(len % sizeof(struct vring_desc))) {
vringh_bad("Strange indirect len %u", desc->len);
return -EINVAL;
}
/* We will check this when we follow it! */
- if (desc->flags & VRING_DESC_F_NEXT)
- *up_next = desc->next;
+ if (desc->flags & cpu_to_vringh16(vrh, VRING_DESC_F_NEXT))
+ *up_next = vringh16_to_cpu(vrh, desc->next);
else
*up_next = -2;
*descs = addr;
- *desc_max = desc->len / sizeof(struct vring_desc);
+ *desc_max = len / sizeof(struct vring_desc);
/* Now, start at the first indirect. */
*i = 0;
@@ -287,22 +293,25 @@
if (unlikely(err))
goto fail;
- if (unlikely(desc.flags & VRING_DESC_F_INDIRECT)) {
+ if (unlikely(desc.flags &
+ cpu_to_vringh16(vrh, VRING_DESC_F_INDIRECT))) {
+ u64 a = vringh64_to_cpu(vrh, desc.addr);
+
/* Make sure it's OK, and get offset. */
- len = desc.len;
- if (!rcheck(vrh, desc.addr, &len, &range, getrange)) {
+ len = vringh32_to_cpu(vrh, desc.len);
+ if (!rcheck(vrh, a, &len, &range, getrange)) {
err = -EINVAL;
goto fail;
}
- if (unlikely(len != desc.len)) {
+ if (unlikely(len != vringh32_to_cpu(vrh, desc.len))) {
slow = true;
/* We need to save this range to use offset */
slowrange = range;
}
- addr = (void *)(long)(desc.addr + range.offset);
- err = move_to_indirect(&up_next, &i, addr, &desc,
+ addr = (void *)(long)(a + range.offset);
+ err = move_to_indirect(vrh, &up_next, &i, addr, &desc,
&descs, &desc_max);
if (err)
goto fail;
@@ -315,7 +324,7 @@
goto fail;
}
- if (desc.flags & VRING_DESC_F_WRITE)
+ if (desc.flags & cpu_to_vringh16(vrh, VRING_DESC_F_WRITE))
iov = wiov;
else {
iov = riov;
@@ -336,12 +345,14 @@
again:
/* Make sure it's OK, and get offset. */
- len = desc.len;
- if (!rcheck(vrh, desc.addr, &len, &range, getrange)) {
+ len = vringh32_to_cpu(vrh, desc.len);
+ if (!rcheck(vrh, vringh64_to_cpu(vrh, desc.addr), &len, &range,
+ getrange)) {
err = -EINVAL;
goto fail;
}
- addr = (void *)(unsigned long)(desc.addr + range.offset);
+ addr = (void *)(unsigned long)(vringh64_to_cpu(vrh, desc.addr) +
+ range.offset);
if (unlikely(iov->used == (iov->max_num & ~VRINGH_IOV_ALLOCATED))) {
err = resize_iovec(iov, gfp);
@@ -353,14 +364,16 @@
iov->iov[iov->used].iov_len = len;
iov->used++;
- if (unlikely(len != desc.len)) {
- desc.len -= len;
- desc.addr += len;
+ if (unlikely(len != vringh32_to_cpu(vrh, desc.len))) {
+ desc.len = cpu_to_vringh32(vrh,
+ vringh32_to_cpu(vrh, desc.len) - len);
+ desc.addr = cpu_to_vringh64(vrh,
+ vringh64_to_cpu(vrh, desc.addr) + len);
goto again;
}
- if (desc.flags & VRING_DESC_F_NEXT) {
- i = desc.next;
+ if (desc.flags & cpu_to_vringh16(vrh, VRING_DESC_F_NEXT)) {
+ i = vringh16_to_cpu(vrh, desc.next);
} else {
/* Just in case we need to finish traversing above. */
if (unlikely(up_next > 0)) {
@@ -387,7 +400,8 @@
static inline int __vringh_complete(struct vringh *vrh,
const struct vring_used_elem *used,
unsigned int num_used,
- int (*putu16)(u16 *p, u16 val),
+ int (*putu16)(const struct vringh *vrh,
+ __virtio16 *p, u16 val),
int (*putused)(struct vring_used_elem *dst,
const struct vring_used_elem
*src, unsigned num))
@@ -420,7 +434,7 @@
/* Make sure buffer is written before we update index. */
virtio_wmb(vrh->weak_barriers);
- err = putu16(&vrh->vring.used->idx, used_idx + num_used);
+ err = putu16(vrh, &vrh->vring.used->idx, used_idx + num_used);
if (err) {
vringh_bad("Failed to update used index at %p",
&vrh->vring.used->idx);
@@ -433,7 +447,9 @@
static inline int __vringh_need_notify(struct vringh *vrh,
- int (*getu16)(u16 *val, const u16 *p))
+ int (*getu16)(const struct vringh *vrh,
+ u16 *val,
+ const __virtio16 *p))
{
bool notify;
u16 used_event;
@@ -447,7 +463,7 @@
/* Old-style, without event indices. */
if (!vrh->event_indices) {
u16 flags;
- err = getu16(&flags, &vrh->vring.avail->flags);
+ err = getu16(vrh, &flags, &vrh->vring.avail->flags);
if (err) {
vringh_bad("Failed to get flags at %p",
&vrh->vring.avail->flags);
@@ -457,7 +473,7 @@
}
/* Modern: we know when other side wants to know. */
- err = getu16(&used_event, &vring_used_event(&vrh->vring));
+ err = getu16(vrh, &used_event, &vring_used_event(&vrh->vring));
if (err) {
vringh_bad("Failed to get used event idx at %p",
&vring_used_event(&vrh->vring));
@@ -478,20 +494,22 @@
}
static inline bool __vringh_notify_enable(struct vringh *vrh,
- int (*getu16)(u16 *val, const u16 *p),
- int (*putu16)(u16 *p, u16 val))
+ int (*getu16)(const struct vringh *vrh,
+ u16 *val, const __virtio16 *p),
+ int (*putu16)(const struct vringh *vrh,
+ __virtio16 *p, u16 val))
{
u16 avail;
if (!vrh->event_indices) {
/* Old-school; update flags. */
- if (putu16(&vrh->vring.used->flags, 0) != 0) {
+ if (putu16(vrh, &vrh->vring.used->flags, 0) != 0) {
vringh_bad("Clearing used flags %p",
&vrh->vring.used->flags);
return true;
}
} else {
- if (putu16(&vring_avail_event(&vrh->vring),
+ if (putu16(vrh, &vring_avail_event(&vrh->vring),
vrh->last_avail_idx) != 0) {
vringh_bad("Updating avail event index %p",
&vring_avail_event(&vrh->vring));
@@ -503,7 +521,7 @@
* sure it's written, then check again. */
virtio_mb(vrh->weak_barriers);
- if (getu16(&avail, &vrh->vring.avail->idx) != 0) {
+ if (getu16(vrh, &avail, &vrh->vring.avail->idx) != 0) {
vringh_bad("Failed to check avail idx at %p",
&vrh->vring.avail->idx);
return true;
@@ -516,11 +534,13 @@
}
static inline void __vringh_notify_disable(struct vringh *vrh,
- int (*putu16)(u16 *p, u16 val))
+ int (*putu16)(const struct vringh *vrh,
+ __virtio16 *p, u16 val))
{
if (!vrh->event_indices) {
/* Old-school; update flags. */
- if (putu16(&vrh->vring.used->flags, VRING_USED_F_NO_NOTIFY)) {
+ if (putu16(vrh, &vrh->vring.used->flags,
+ VRING_USED_F_NO_NOTIFY)) {
vringh_bad("Setting used flags %p",
&vrh->vring.used->flags);
}
@@ -528,14 +548,18 @@
}
/* Userspace access helpers: in this case, addresses are really userspace. */
-static inline int getu16_user(u16 *val, const u16 *p)
+static inline int getu16_user(const struct vringh *vrh, u16 *val, const __virtio16 *p)
{
- return get_user(*val, (__force u16 __user *)p);
+ __virtio16 v = 0;
+ int rc = get_user(v, (__force __virtio16 __user *)p);
+ *val = vringh16_to_cpu(vrh, v);
+ return rc;
}
-static inline int putu16_user(u16 *p, u16 val)
+static inline int putu16_user(const struct vringh *vrh, __virtio16 *p, u16 val)
{
- return put_user(val, (__force u16 __user *)p);
+ __virtio16 v = cpu_to_vringh16(vrh, val);
+ return put_user(v, (__force __virtio16 __user *)p);
}
static inline int copydesc_user(void *dst, const void *src, size_t len)
@@ -577,7 +601,7 @@
* Returns an error if num is invalid: you should check pointers
* yourself!
*/
-int vringh_init_user(struct vringh *vrh, u32 features,
+int vringh_init_user(struct vringh *vrh, u64 features,
unsigned int num, bool weak_barriers,
struct vring_desc __user *desc,
struct vring_avail __user *avail,
@@ -589,6 +613,7 @@
return -EINVAL;
}
+ vrh->little_endian = (features & (1ULL << VIRTIO_F_VERSION_1));
vrh->event_indices = (features & (1 << VIRTIO_RING_F_EVENT_IDX));
vrh->weak_barriers = weak_barriers;
vrh->completed = 0;
@@ -729,8 +754,8 @@
{
struct vring_used_elem used;
- used.id = head;
- used.len = len;
+ used.id = cpu_to_vringh32(vrh, head);
+ used.len = cpu_to_vringh32(vrh, len);
return __vringh_complete(vrh, &used, 1, putu16_user, putused_user);
}
EXPORT_SYMBOL(vringh_complete_user);
@@ -792,15 +817,16 @@
EXPORT_SYMBOL(vringh_need_notify_user);
/* Kernelspace access helpers. */
-static inline int getu16_kern(u16 *val, const u16 *p)
+static inline int getu16_kern(const struct vringh *vrh,
+ u16 *val, const __virtio16 *p)
{
- *val = ACCESS_ONCE(*p);
+ *val = vringh16_to_cpu(vrh, ACCESS_ONCE(*p));
return 0;
}
-static inline int putu16_kern(u16 *p, u16 val)
+static inline int putu16_kern(const struct vringh *vrh, __virtio16 *p, u16 val)
{
- ACCESS_ONCE(*p) = val;
+ ACCESS_ONCE(*p) = cpu_to_vringh16(vrh, val);
return 0;
}
@@ -836,7 +862,7 @@
*
* Returns an error if num is invalid.
*/
-int vringh_init_kern(struct vringh *vrh, u32 features,
+int vringh_init_kern(struct vringh *vrh, u64 features,
unsigned int num, bool weak_barriers,
struct vring_desc *desc,
struct vring_avail *avail,
@@ -848,6 +874,7 @@
return -EINVAL;
}
+ vrh->little_endian = (features & (1ULL << VIRTIO_F_VERSION_1));
vrh->event_indices = (features & (1 << VIRTIO_RING_F_EVENT_IDX));
vrh->weak_barriers = weak_barriers;
vrh->completed = 0;
@@ -962,8 +989,8 @@
{
struct vring_used_elem used;
- used.id = head;
- used.len = len;
+ used.id = cpu_to_vringh32(vrh, head);
+ used.len = cpu_to_vringh32(vrh, len);
return __vringh_complete(vrh, &used, 1, putu16_kern, putused_kern);
}
diff --git a/drivers/video/fbdev/s3c-fb.c b/drivers/video/fbdev/s3c-fb.c
index a623a4d..7e3a05f 100644
--- a/drivers/video/fbdev/s3c-fb.c
+++ b/drivers/video/fbdev/s3c-fb.c
@@ -1630,7 +1630,7 @@
}
#endif
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int s3c_fb_runtime_suspend(struct device *dev)
{
struct s3c_fb *sfb = dev_get_drvdata(dev);
diff --git a/drivers/video/fbdev/sh_mobile_meram.c b/drivers/video/fbdev/sh_mobile_meram.c
index 1d56108..baadfb2 100644
--- a/drivers/video/fbdev/sh_mobile_meram.c
+++ b/drivers/video/fbdev/sh_mobile_meram.c
@@ -569,7 +569,7 @@
* Power management
*/
-#if defined(CONFIG_PM_SLEEP) || defined(CONFIG_PM_RUNTIME)
+#ifdef CONFIG_PM
static int sh_mobile_meram_suspend(struct device *dev)
{
struct platform_device *pdev = to_platform_device(dev);
@@ -612,7 +612,7 @@
meram_write_reg(priv->base, common_regs[i], priv->regs[i]);
return 0;
}
-#endif /* CONFIG_PM_SLEEP || CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
static UNIVERSAL_DEV_PM_OPS(sh_mobile_meram_dev_pm_ops,
sh_mobile_meram_suspend,
diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c
index f226658..b9f70df 100644
--- a/drivers/virtio/virtio.c
+++ b/drivers/virtio/virtio.c
@@ -162,6 +162,27 @@
spin_unlock_irq(&dev->config_lock);
}
+static int virtio_finalize_features(struct virtio_device *dev)
+{
+ int ret = dev->config->finalize_features(dev);
+ unsigned status;
+
+ if (ret)
+ return ret;
+
+ if (!virtio_has_feature(dev, VIRTIO_F_VERSION_1))
+ return 0;
+
+ add_status(dev, VIRTIO_CONFIG_S_FEATURES_OK);
+ status = dev->config->get_status(dev);
+ if (!(status & VIRTIO_CONFIG_S_FEATURES_OK)) {
+ dev_err(&dev->dev, "virtio: device refuses features: %x\n",
+ status);
+ return -ENODEV;
+ }
+ return 0;
+}
+
static int virtio_dev_probe(struct device *_d)
{
int err, i;
@@ -170,7 +191,6 @@
u64 device_features;
u64 driver_features;
u64 driver_features_legacy;
- unsigned status;
/* We have a driver! */
add_status(dev, VIRTIO_CONFIG_S_DRIVER);
@@ -208,21 +228,10 @@
if (device_features & (1ULL << i))
__virtio_set_bit(dev, i);
- err = dev->config->finalize_features(dev);
+ err = virtio_finalize_features(dev);
if (err)
goto err;
- if (virtio_has_feature(dev, VIRTIO_F_VERSION_1)) {
- add_status(dev, VIRTIO_CONFIG_S_FEATURES_OK);
- status = dev->config->get_status(dev);
- if (!(status & VIRTIO_CONFIG_S_FEATURES_OK)) {
- dev_err(_d, "virtio: device refuses features: %x\n",
- status);
- err = -ENODEV;
- goto err;
- }
- }
-
err = drv->probe(dev);
if (err)
goto err;
@@ -372,7 +381,7 @@
/* We have a driver! */
add_status(dev, VIRTIO_CONFIG_S_DRIVER);
- ret = dev->config->finalize_features(dev);
+ ret = virtio_finalize_features(dev);
if (ret)
goto err;
diff --git a/drivers/virtio/virtio_pci_common.c b/drivers/virtio/virtio_pci_common.c
index 953057d..2ef9529 100644
--- a/drivers/virtio/virtio_pci_common.c
+++ b/drivers/virtio/virtio_pci_common.c
@@ -458,7 +458,44 @@
return virtio_device_restore(&vp_dev->vdev);
}
-const struct dev_pm_ops virtio_pci_pm_ops = {
+static const struct dev_pm_ops virtio_pci_pm_ops = {
SET_SYSTEM_SLEEP_PM_OPS(virtio_pci_freeze, virtio_pci_restore)
};
#endif
+
+
+/* Qumranet donated their vendor ID for devices 0x1000 thru 0x10FF. */
+static const struct pci_device_id virtio_pci_id_table[] = {
+ { PCI_DEVICE(0x1af4, PCI_ANY_ID) },
+ { 0 }
+};
+
+MODULE_DEVICE_TABLE(pci, virtio_pci_id_table);
+
+static int virtio_pci_probe(struct pci_dev *pci_dev,
+ const struct pci_device_id *id)
+{
+ return virtio_pci_legacy_probe(pci_dev, id);
+}
+
+static void virtio_pci_remove(struct pci_dev *pci_dev)
+{
+ virtio_pci_legacy_remove(pci_dev);
+}
+
+static struct pci_driver virtio_pci_driver = {
+ .name = "virtio-pci",
+ .id_table = virtio_pci_id_table,
+ .probe = virtio_pci_probe,
+ .remove = virtio_pci_remove,
+#ifdef CONFIG_PM_SLEEP
+ .driver.pm = &virtio_pci_pm_ops,
+#endif
+};
+
+module_pci_driver(virtio_pci_driver);
+
+MODULE_AUTHOR("Anthony Liguori <aliguori@us.ibm.com>");
+MODULE_DESCRIPTION("virtio-pci");
+MODULE_LICENSE("GPL");
+MODULE_VERSION("1");
diff --git a/drivers/virtio/virtio_pci_common.h b/drivers/virtio/virtio_pci_common.h
index d840dad..adddb64 100644
--- a/drivers/virtio/virtio_pci_common.h
+++ b/drivers/virtio/virtio_pci_common.h
@@ -27,7 +27,6 @@
#include <linux/virtio.h>
#include <linux/virtio_config.h>
#include <linux/virtio_ring.h>
-#define VIRTIO_PCI_NO_LEGACY
#include <linux/virtio_pci.h>
#include <linux/highmem.h>
#include <linux/spinlock.h>
@@ -129,8 +128,8 @@
int vp_set_vq_affinity(struct virtqueue *vq, int cpu);
void virtio_pci_release_dev(struct device *);
-#ifdef CONFIG_PM_SLEEP
-extern const struct dev_pm_ops virtio_pci_pm_ops;
-#endif
+int virtio_pci_legacy_probe(struct pci_dev *pci_dev,
+ const struct pci_device_id *id);
+void virtio_pci_legacy_remove(struct pci_dev *pci_dev);
#endif
diff --git a/drivers/virtio/virtio_pci_legacy.c b/drivers/virtio/virtio_pci_legacy.c
index 2588252..6c76f0f 100644
--- a/drivers/virtio/virtio_pci_legacy.c
+++ b/drivers/virtio/virtio_pci_legacy.c
@@ -19,14 +19,6 @@
#include "virtio_pci_common.h"
-/* Qumranet donated their vendor ID for devices 0x1000 thru 0x10FF. */
-static const struct pci_device_id virtio_pci_id_table[] = {
- { PCI_DEVICE(0x1af4, PCI_ANY_ID) },
- { 0 }
-};
-
-MODULE_DEVICE_TABLE(pci, virtio_pci_id_table);
-
/* virtio config->get_features() implementation */
static u64 vp_get_features(struct virtio_device *vdev)
{
@@ -220,7 +212,7 @@
};
/* the PCI probing function */
-static int virtio_pci_probe(struct pci_dev *pci_dev,
+int virtio_pci_legacy_probe(struct pci_dev *pci_dev,
const struct pci_device_id *id)
{
struct virtio_pci_device *vp_dev;
@@ -300,7 +292,7 @@
return err;
}
-static void virtio_pci_remove(struct pci_dev *pci_dev)
+void virtio_pci_legacy_remove(struct pci_dev *pci_dev)
{
struct virtio_pci_device *vp_dev = pci_get_drvdata(pci_dev);
@@ -312,15 +304,3 @@
pci_disable_device(pci_dev);
kfree(vp_dev);
}
-
-static struct pci_driver virtio_pci_driver = {
- .name = "virtio-pci",
- .id_table = virtio_pci_id_table,
- .probe = virtio_pci_probe,
- .remove = virtio_pci_remove,
-#ifdef CONFIG_PM_SLEEP
- .driver.pm = &virtio_pci_pm_ops,
-#endif
-};
-
-module_pci_driver(virtio_pci_driver);
diff --git a/drivers/watchdog/imx2_wdt.c b/drivers/watchdog/imx2_wdt.c
index 65b84d8..d6add51 100644
--- a/drivers/watchdog/imx2_wdt.c
+++ b/drivers/watchdog/imx2_wdt.c
@@ -326,6 +326,52 @@
}
}
+#ifdef CONFIG_PM_SLEEP
+/* Disable watchdog if it is active during suspend */
+static int imx2_wdt_suspend(struct device *dev)
+{
+ struct watchdog_device *wdog = dev_get_drvdata(dev);
+ struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
+
+ imx2_wdt_set_timeout(wdog, IMX2_WDT_MAX_TIME);
+ imx2_wdt_ping(wdog);
+
+ /* Watchdog has been stopped but IP block is still running */
+ if (!watchdog_active(wdog) && imx2_wdt_is_running(wdev))
+ del_timer_sync(&wdev->timer);
+
+ clk_disable_unprepare(wdev->clk);
+
+ return 0;
+}
+
+/* Enable watchdog and configure it if necessary */
+static int imx2_wdt_resume(struct device *dev)
+{
+ struct watchdog_device *wdog = dev_get_drvdata(dev);
+ struct imx2_wdt_device *wdev = watchdog_get_drvdata(wdog);
+
+ clk_prepare_enable(wdev->clk);
+
+ if (watchdog_active(wdog) && !imx2_wdt_is_running(wdev)) {
+ /* Resumes from deep sleep we need restart
+ * the watchdog again.
+ */
+ imx2_wdt_setup(wdog);
+ imx2_wdt_set_timeout(wdog, wdog->timeout);
+ imx2_wdt_ping(wdog);
+ } else if (imx2_wdt_is_running(wdev)) {
+ imx2_wdt_ping(wdog);
+ mod_timer(&wdev->timer, jiffies + wdog->timeout * HZ / 2);
+ }
+
+ return 0;
+}
+#endif
+
+static SIMPLE_DEV_PM_OPS(imx2_wdt_pm_ops, imx2_wdt_suspend,
+ imx2_wdt_resume);
+
static const struct of_device_id imx2_wdt_dt_ids[] = {
{ .compatible = "fsl,imx21-wdt", },
{ /* sentinel */ }
@@ -337,6 +383,7 @@
.shutdown = imx2_wdt_shutdown,
.driver = {
.name = DRIVER_NAME,
+ .pm = &imx2_wdt_pm_ops,
.of_match_table = imx2_wdt_dt_ids,
},
};
diff --git a/fs/hfsplus/catalog.c b/fs/hfsplus/catalog.c
index 32602c6..7892e6f 100644
--- a/fs/hfsplus/catalog.c
+++ b/fs/hfsplus/catalog.c
@@ -38,21 +38,30 @@
return hfsplus_strcmp(&k1->cat.name, &k2->cat.name);
}
-void hfsplus_cat_build_key(struct super_block *sb, hfsplus_btree_key *key,
- u32 parent, struct qstr *str)
+/* Generates key for catalog file/folders record. */
+int hfsplus_cat_build_key(struct super_block *sb,
+ hfsplus_btree_key *key, u32 parent, struct qstr *str)
{
- int len;
+ int len, err;
key->cat.parent = cpu_to_be32(parent);
- if (str) {
- hfsplus_asc2uni(sb, &key->cat.name, HFSPLUS_MAX_STRLEN,
- str->name, str->len);
- len = be16_to_cpu(key->cat.name.length);
- } else {
- key->cat.name.length = 0;
- len = 0;
- }
+ err = hfsplus_asc2uni(sb, &key->cat.name, HFSPLUS_MAX_STRLEN,
+ str->name, str->len);
+ if (unlikely(err < 0))
+ return err;
+
+ len = be16_to_cpu(key->cat.name.length);
key->key_len = cpu_to_be16(6 + 2 * len);
+ return 0;
+}
+
+/* Generates key for catalog thread record. */
+void hfsplus_cat_build_key_with_cnid(struct super_block *sb,
+ hfsplus_btree_key *key, u32 parent)
+{
+ key->cat.parent = cpu_to_be32(parent);
+ key->cat.name.length = 0;
+ key->key_len = cpu_to_be16(6);
}
static void hfsplus_cat_build_key_uni(hfsplus_btree_key *key, u32 parent,
@@ -167,11 +176,16 @@
hfsplus_cat_entry *entry, int type,
u32 parentid, struct qstr *str)
{
+ int err;
+
entry->type = cpu_to_be16(type);
entry->thread.reserved = 0;
entry->thread.parentID = cpu_to_be32(parentid);
- hfsplus_asc2uni(sb, &entry->thread.nodeName, HFSPLUS_MAX_STRLEN,
+ err = hfsplus_asc2uni(sb, &entry->thread.nodeName, HFSPLUS_MAX_STRLEN,
str->name, str->len);
+ if (unlikely(err < 0))
+ return err;
+
return 10 + be16_to_cpu(entry->thread.nodeName.length) * 2;
}
@@ -183,7 +197,7 @@
int err;
u16 type;
- hfsplus_cat_build_key(sb, fd->search_key, cnid, NULL);
+ hfsplus_cat_build_key_with_cnid(sb, fd->search_key, cnid);
err = hfs_brec_read(fd, &tmp, sizeof(hfsplus_cat_entry));
if (err)
return err;
@@ -250,11 +264,16 @@
if (err)
return err;
- hfsplus_cat_build_key(sb, fd.search_key, cnid, NULL);
+ hfsplus_cat_build_key_with_cnid(sb, fd.search_key, cnid);
entry_size = hfsplus_fill_cat_thread(sb, &entry,
S_ISDIR(inode->i_mode) ?
HFSPLUS_FOLDER_THREAD : HFSPLUS_FILE_THREAD,
dir->i_ino, str);
+ if (unlikely(entry_size < 0)) {
+ err = entry_size;
+ goto err2;
+ }
+
err = hfs_brec_find(&fd, hfs_find_rec_by_key);
if (err != -ENOENT) {
if (!err)
@@ -265,7 +284,10 @@
if (err)
goto err2;
- hfsplus_cat_build_key(sb, fd.search_key, dir->i_ino, str);
+ err = hfsplus_cat_build_key(sb, fd.search_key, dir->i_ino, str);
+ if (unlikely(err))
+ goto err1;
+
entry_size = hfsplus_cat_build_record(&entry, cnid, inode);
err = hfs_brec_find(&fd, hfs_find_rec_by_key);
if (err != -ENOENT) {
@@ -288,7 +310,7 @@
return 0;
err1:
- hfsplus_cat_build_key(sb, fd.search_key, cnid, NULL);
+ hfsplus_cat_build_key_with_cnid(sb, fd.search_key, cnid);
if (!hfs_brec_find(&fd, hfs_find_rec_by_key))
hfs_brec_remove(&fd);
err2:
@@ -313,7 +335,7 @@
if (!str) {
int len;
- hfsplus_cat_build_key(sb, fd.search_key, cnid, NULL);
+ hfsplus_cat_build_key_with_cnid(sb, fd.search_key, cnid);
err = hfs_brec_find(&fd, hfs_find_rec_by_key);
if (err)
goto out;
@@ -329,7 +351,9 @@
off + 2, len);
fd.search_key->key_len = cpu_to_be16(6 + len);
} else
- hfsplus_cat_build_key(sb, fd.search_key, dir->i_ino, str);
+ err = hfsplus_cat_build_key(sb, fd.search_key, dir->i_ino, str);
+ if (unlikely(err))
+ goto out;
err = hfs_brec_find(&fd, hfs_find_rec_by_key);
if (err)
@@ -360,7 +384,7 @@
if (err)
goto out;
- hfsplus_cat_build_key(sb, fd.search_key, cnid, NULL);
+ hfsplus_cat_build_key_with_cnid(sb, fd.search_key, cnid);
err = hfs_brec_find(&fd, hfs_find_rec_by_key);
if (err)
goto out;
@@ -405,7 +429,11 @@
dst_fd = src_fd;
/* find the old dir entry and read the data */
- hfsplus_cat_build_key(sb, src_fd.search_key, src_dir->i_ino, src_name);
+ err = hfsplus_cat_build_key(sb, src_fd.search_key,
+ src_dir->i_ino, src_name);
+ if (unlikely(err))
+ goto out;
+
err = hfs_brec_find(&src_fd, hfs_find_rec_by_key);
if (err)
goto out;
@@ -419,7 +447,11 @@
type = be16_to_cpu(entry.type);
/* create new dir entry with the data from the old entry */
- hfsplus_cat_build_key(sb, dst_fd.search_key, dst_dir->i_ino, dst_name);
+ err = hfsplus_cat_build_key(sb, dst_fd.search_key,
+ dst_dir->i_ino, dst_name);
+ if (unlikely(err))
+ goto out;
+
err = hfs_brec_find(&dst_fd, hfs_find_rec_by_key);
if (err != -ENOENT) {
if (!err)
@@ -436,7 +468,11 @@
dst_dir->i_mtime = dst_dir->i_ctime = CURRENT_TIME_SEC;
/* finally remove the old entry */
- hfsplus_cat_build_key(sb, src_fd.search_key, src_dir->i_ino, src_name);
+ err = hfsplus_cat_build_key(sb, src_fd.search_key,
+ src_dir->i_ino, src_name);
+ if (unlikely(err))
+ goto out;
+
err = hfs_brec_find(&src_fd, hfs_find_rec_by_key);
if (err)
goto out;
@@ -449,7 +485,7 @@
src_dir->i_mtime = src_dir->i_ctime = CURRENT_TIME_SEC;
/* remove old thread entry */
- hfsplus_cat_build_key(sb, src_fd.search_key, cnid, NULL);
+ hfsplus_cat_build_key_with_cnid(sb, src_fd.search_key, cnid);
err = hfs_brec_find(&src_fd, hfs_find_rec_by_key);
if (err)
goto out;
@@ -459,9 +495,14 @@
goto out;
/* create new thread entry */
- hfsplus_cat_build_key(sb, dst_fd.search_key, cnid, NULL);
+ hfsplus_cat_build_key_with_cnid(sb, dst_fd.search_key, cnid);
entry_size = hfsplus_fill_cat_thread(sb, &entry, type,
dst_dir->i_ino, dst_name);
+ if (unlikely(entry_size < 0)) {
+ err = entry_size;
+ goto out;
+ }
+
err = hfs_brec_find(&dst_fd, hfs_find_rec_by_key);
if (err != -ENOENT) {
if (!err)
diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c
index 610a326..435bea2 100644
--- a/fs/hfsplus/dir.c
+++ b/fs/hfsplus/dir.c
@@ -44,7 +44,10 @@
err = hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &fd);
if (err)
return ERR_PTR(err);
- hfsplus_cat_build_key(sb, fd.search_key, dir->i_ino, &dentry->d_name);
+ err = hfsplus_cat_build_key(sb, fd.search_key, dir->i_ino,
+ &dentry->d_name);
+ if (unlikely(err < 0))
+ goto fail;
again:
err = hfs_brec_read(&fd, &entry, sizeof(entry));
if (err) {
@@ -97,9 +100,11 @@
be32_to_cpu(entry.file.permissions.dev);
str.len = sprintf(name, "iNode%d", linkid);
str.name = name;
- hfsplus_cat_build_key(sb, fd.search_key,
+ err = hfsplus_cat_build_key(sb, fd.search_key,
HFSPLUS_SB(sb)->hidden_dir->i_ino,
&str);
+ if (unlikely(err < 0))
+ goto fail;
goto again;
}
} else if (!dentry->d_fsdata)
@@ -145,7 +150,7 @@
err = -ENOMEM;
goto out;
}
- hfsplus_cat_build_key(sb, fd.search_key, inode->i_ino, NULL);
+ hfsplus_cat_build_key_with_cnid(sb, fd.search_key, inode->i_ino);
err = hfs_brec_find(&fd, hfs_find_rec_by_key);
if (err)
goto out;
diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h
index eb5e059..b0441d6 100644
--- a/fs/hfsplus/hfsplus_fs.h
+++ b/fs/hfsplus/hfsplus_fs.h
@@ -443,8 +443,10 @@
const hfsplus_btree_key *k2);
int hfsplus_cat_bin_cmp_key(const hfsplus_btree_key *k1,
const hfsplus_btree_key *k2);
-void hfsplus_cat_build_key(struct super_block *sb, hfsplus_btree_key *key,
+int hfsplus_cat_build_key(struct super_block *sb, hfsplus_btree_key *key,
u32 parent, struct qstr *str);
+void hfsplus_cat_build_key_with_cnid(struct super_block *sb,
+ hfsplus_btree_key *key, u32 parent);
void hfsplus_cat_set_perms(struct inode *inode, struct hfsplus_perm *perms);
int hfsplus_find_cat(struct super_block *sb, u32 cnid,
struct hfs_find_data *fd);
diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index 4cf2024..593af2f 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -515,7 +515,9 @@
err = hfs_find_init(sbi->cat_tree, &fd);
if (err)
goto out_put_root;
- hfsplus_cat_build_key(sb, fd.search_key, HFSPLUS_ROOT_CNID, &str);
+ err = hfsplus_cat_build_key(sb, fd.search_key, HFSPLUS_ROOT_CNID, &str);
+ if (unlikely(err < 0))
+ goto out_put_root;
if (!hfs_brec_read(&fd, &entry, sizeof(entry))) {
hfs_find_exit(&fd);
if (entry.type != cpu_to_be16(HFSPLUS_FOLDER))
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index a93bf98..fcae9ef 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -5662,7 +5662,7 @@
struct ocfs2_extent_tree *et,
u32 cpos, u32 phys_cpos, u32 len, int flags,
struct ocfs2_cached_dealloc_ctxt *dealloc,
- u64 refcount_loc)
+ u64 refcount_loc, bool refcount_tree_locked)
{
int ret, credits = 0, extra_blocks = 0;
u64 phys_blkno = ocfs2_clusters_to_blocks(inode->i_sb, phys_cpos);
@@ -5676,11 +5676,13 @@
BUG_ON(!(OCFS2_I(inode)->ip_dyn_features &
OCFS2_HAS_REFCOUNT_FL));
- ret = ocfs2_lock_refcount_tree(osb, refcount_loc, 1,
- &ref_tree, NULL);
- if (ret) {
- mlog_errno(ret);
- goto bail;
+ if (!refcount_tree_locked) {
+ ret = ocfs2_lock_refcount_tree(osb, refcount_loc, 1,
+ &ref_tree, NULL);
+ if (ret) {
+ mlog_errno(ret);
+ goto bail;
+ }
}
ret = ocfs2_prepare_refcount_change_for_del(inode,
@@ -7021,6 +7023,7 @@
u64 refcount_loc = le64_to_cpu(di->i_refcount_loc);
struct ocfs2_extent_tree et;
struct ocfs2_cached_dealloc_ctxt dealloc;
+ struct ocfs2_refcount_tree *ref_tree = NULL;
ocfs2_init_dinode_extent_tree(&et, INODE_CACHE(inode), di_bh);
ocfs2_init_dealloc_ctxt(&dealloc);
@@ -7130,9 +7133,18 @@
phys_cpos = ocfs2_blocks_to_clusters(inode->i_sb, blkno);
+ if ((flags & OCFS2_EXT_REFCOUNTED) && trunc_len && !ref_tree) {
+ status = ocfs2_lock_refcount_tree(osb, refcount_loc, 1,
+ &ref_tree, NULL);
+ if (status) {
+ mlog_errno(status);
+ goto bail;
+ }
+ }
+
status = ocfs2_remove_btree_range(inode, &et, trunc_cpos,
phys_cpos, trunc_len, flags, &dealloc,
- refcount_loc);
+ refcount_loc, true);
if (status < 0) {
mlog_errno(status);
goto bail;
@@ -7147,6 +7159,8 @@
goto start;
bail:
+ if (ref_tree)
+ ocfs2_unlock_refcount_tree(osb, ref_tree, 1);
ocfs2_schedule_truncate_log_flush(osb, 1);
diff --git a/fs/ocfs2/alloc.h b/fs/ocfs2/alloc.h
index ca381c5..fb09b97 100644
--- a/fs/ocfs2/alloc.h
+++ b/fs/ocfs2/alloc.h
@@ -142,7 +142,7 @@
struct ocfs2_extent_tree *et,
u32 cpos, u32 phys_cpos, u32 len, int flags,
struct ocfs2_cached_dealloc_ctxt *dealloc,
- u64 refcount_loc);
+ u64 refcount_loc, bool refcount_tree_locked);
int ocfs2_num_free_extents(struct ocfs2_super *osb,
struct ocfs2_extent_tree *et);
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index d9f2229..46d93e9 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -894,7 +894,7 @@
}
}
-static void ocfs2_free_write_ctxt(struct ocfs2_write_ctxt *wc)
+static void ocfs2_unlock_pages(struct ocfs2_write_ctxt *wc)
{
int i;
@@ -915,7 +915,11 @@
page_cache_release(wc->w_target_page);
}
ocfs2_unlock_and_free_pages(wc->w_pages, wc->w_num_pages);
+}
+static void ocfs2_free_write_ctxt(struct ocfs2_write_ctxt *wc)
+{
+ ocfs2_unlock_pages(wc);
brelse(wc->w_di_bh);
kfree(wc);
}
@@ -2042,11 +2046,19 @@
ocfs2_update_inode_fsync_trans(handle, inode, 1);
ocfs2_journal_dirty(handle, wc->w_di_bh);
+ /* unlock pages before dealloc since it needs acquiring j_trans_barrier
+ * lock, or it will cause a deadlock since journal commit threads holds
+ * this lock and will ask for the page lock when flushing the data.
+ * put it here to preserve the unlock order.
+ */
+ ocfs2_unlock_pages(wc);
+
ocfs2_commit_trans(osb, handle);
ocfs2_run_deallocs(osb, &wc->w_dealloc);
- ocfs2_free_write_ctxt(wc);
+ brelse(wc->w_di_bh);
+ kfree(wc);
return copied;
}
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index 79d56dc..319e786 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -4479,7 +4479,7 @@
p_cpos = ocfs2_blocks_to_clusters(dir->i_sb, blkno);
ret = ocfs2_remove_btree_range(dir, &et, cpos, p_cpos, clen, 0,
- &dealloc, 0);
+ &dealloc, 0, false);
if (ret) {
mlog_errno(ret);
goto out;
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index 3689b35..a6944b2 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -695,14 +695,6 @@
res->inflight_assert_workers);
}
-static void dlm_lockres_grab_inflight_worker(struct dlm_ctxt *dlm,
- struct dlm_lock_resource *res)
-{
- spin_lock(&res->spinlock);
- __dlm_lockres_grab_inflight_worker(dlm, res);
- spin_unlock(&res->spinlock);
-}
-
static void __dlm_lockres_drop_inflight_worker(struct dlm_ctxt *dlm,
struct dlm_lock_resource *res)
{
@@ -1646,6 +1638,7 @@
}
mlog(0, "%u is the owner of %.*s, cleaning everyone else\n",
dlm->node_num, res->lockname.len, res->lockname.name);
+ spin_lock(&res->spinlock);
ret = dlm_dispatch_assert_master(dlm, res, 0, request->node_idx,
DLM_ASSERT_MASTER_MLE_CLEANUP);
if (ret < 0) {
@@ -1653,7 +1646,8 @@
response = DLM_MASTER_RESP_ERROR;
dlm_lockres_put(res);
} else
- dlm_lockres_grab_inflight_worker(dlm, res);
+ __dlm_lockres_grab_inflight_worker(dlm, res);
+ spin_unlock(&res->spinlock);
} else {
if (res)
dlm_lockres_put(res);
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 69fb9f7..3950693 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1803,7 +1803,7 @@
ret = ocfs2_remove_btree_range(inode, &et, trunc_cpos,
phys_cpos, trunc_len, flags,
- &dealloc, refcount_loc);
+ &dealloc, refcount_loc, false);
if (ret < 0) {
mlog_errno(ret);
goto out;
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index aa1eee0..d3ebf2e 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -12,6 +12,9 @@
#include <linux/vmstat.h>
#include <linux/atomic.h>
#include <linux/vmalloc.h>
+#ifdef CONFIG_CMA
+#include <linux/cma.h>
+#endif
#include <asm/page.h>
#include <asm/pgtable.h>
#include "internal.h"
@@ -138,6 +141,10 @@
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
"AnonHugePages: %8lu kB\n"
#endif
+#ifdef CONFIG_CMA
+ "CmaTotal: %8lu kB\n"
+ "CmaFree: %8lu kB\n"
+#endif
,
K(i.totalram),
K(i.freeram),
@@ -187,12 +194,16 @@
vmi.used >> 10,
vmi.largest_chunk >> 10
#ifdef CONFIG_MEMORY_FAILURE
- ,atomic_long_read(&num_poisoned_pages) << (PAGE_SHIFT - 10)
+ , atomic_long_read(&num_poisoned_pages) << (PAGE_SHIFT - 10)
#endif
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
- ,K(global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) *
+ , K(global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) *
HPAGE_PMD_NR)
#endif
+#ifdef CONFIG_CMA
+ , K(totalcma_pages)
+ , K(global_page_state(NR_FREE_CMA_PAGES))
+#endif
);
hugetlb_report_meminfo(m);
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index 7581518e3..61e32ec 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -313,6 +313,7 @@
u8 valid:1; /* Can successfully enable wakeup? */
u8 run_wake:1; /* Run-Wake GPE devices */
u8 notifier_present:1; /* Wake-up notify handler has been installed */
+ u8 enabled:1; /* Enabled for wakeup */
};
struct acpi_device_wakeup_context {
diff --git a/include/linux/cma.h b/include/linux/cma.h
index a93438b..9384ba6 100644
--- a/include/linux/cma.h
+++ b/include/linux/cma.h
@@ -15,6 +15,7 @@
struct cma;
+extern unsigned long totalcma_pages;
extern phys_addr_t cma_get_base(struct cma *cma);
extern unsigned long cma_get_size(struct cma *cma);
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index ea4f1c4..4e5bd81 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -120,6 +120,15 @@
};
enum {
+ MLX5_MKEY_INBOX_PG_ACCESS = 1 << 31
+};
+
+enum {
+ MLX5_PFAULT_SUBTYPE_WQE = 0,
+ MLX5_PFAULT_SUBTYPE_RDMA = 1,
+};
+
+enum {
MLX5_PERM_LOCAL_READ = 1 << 2,
MLX5_PERM_LOCAL_WRITE = 1 << 3,
MLX5_PERM_REMOTE_READ = 1 << 4,
@@ -180,6 +189,19 @@
MLX5_MKEY_MASK_FREE = 1ull << 29,
};
+enum {
+ MLX5_UMR_TRANSLATION_OFFSET_EN = (1 << 4),
+
+ MLX5_UMR_CHECK_NOT_FREE = (1 << 5),
+ MLX5_UMR_CHECK_FREE = (2 << 5),
+
+ MLX5_UMR_INLINE = (1 << 7),
+};
+
+#define MLX5_UMR_MTT_ALIGNMENT 0x40
+#define MLX5_UMR_MTT_MASK (MLX5_UMR_MTT_ALIGNMENT - 1)
+#define MLX5_UMR_MTT_MIN_CHUNK_SIZE MLX5_UMR_MTT_ALIGNMENT
+
enum mlx5_event {
MLX5_EVENT_TYPE_COMP = 0x0,
@@ -206,6 +228,8 @@
MLX5_EVENT_TYPE_CMD = 0x0a,
MLX5_EVENT_TYPE_PAGE_REQUEST = 0xb,
+
+ MLX5_EVENT_TYPE_PAGE_FAULT = 0xc,
};
enum {
@@ -225,6 +249,7 @@
MLX5_DEV_CAP_FLAG_APM = 1LL << 17,
MLX5_DEV_CAP_FLAG_ATOMIC = 1LL << 18,
MLX5_DEV_CAP_FLAG_BLOCK_MCAST = 1LL << 23,
+ MLX5_DEV_CAP_FLAG_ON_DMND_PG = 1LL << 24,
MLX5_DEV_CAP_FLAG_CQ_MODER = 1LL << 29,
MLX5_DEV_CAP_FLAG_RESIZE_CQ = 1LL << 30,
MLX5_DEV_CAP_FLAG_DCT = 1LL << 37,
@@ -290,6 +315,8 @@
enum {
HCA_CAP_OPMOD_GET_MAX = 0,
HCA_CAP_OPMOD_GET_CUR = 1,
+ HCA_CAP_OPMOD_GET_ODP_MAX = 4,
+ HCA_CAP_OPMOD_GET_ODP_CUR = 5
};
struct mlx5_inbox_hdr {
@@ -319,6 +346,23 @@
u8 vsd_psid[16];
};
+enum mlx5_odp_transport_cap_bits {
+ MLX5_ODP_SUPPORT_SEND = 1 << 31,
+ MLX5_ODP_SUPPORT_RECV = 1 << 30,
+ MLX5_ODP_SUPPORT_WRITE = 1 << 29,
+ MLX5_ODP_SUPPORT_READ = 1 << 28,
+};
+
+struct mlx5_odp_caps {
+ char reserved[0x10];
+ struct {
+ __be32 rc_odp_caps;
+ __be32 uc_odp_caps;
+ __be32 ud_odp_caps;
+ } per_transport_caps;
+ char reserved2[0xe4];
+};
+
struct mlx5_cmd_init_hca_mbox_in {
struct mlx5_inbox_hdr hdr;
u8 rsvd0[2];
@@ -439,6 +483,27 @@
__be32 rsvd1[5];
};
+struct mlx5_eqe_page_fault {
+ __be32 bytes_committed;
+ union {
+ struct {
+ u16 reserved1;
+ __be16 wqe_index;
+ u16 reserved2;
+ __be16 packet_length;
+ u8 reserved3[12];
+ } __packed wqe;
+ struct {
+ __be32 r_key;
+ u16 reserved1;
+ __be16 packet_length;
+ __be32 rdma_op_len;
+ __be64 rdma_va;
+ } __packed rdma;
+ } __packed;
+ __be32 flags_qpn;
+} __packed;
+
union ev_data {
__be32 raw[7];
struct mlx5_eqe_cmd cmd;
@@ -450,6 +515,7 @@
struct mlx5_eqe_congestion cong;
struct mlx5_eqe_stall_vl stall_vl;
struct mlx5_eqe_page_req req_pages;
+ struct mlx5_eqe_page_fault page_fault;
} __packed;
struct mlx5_eqe {
@@ -776,6 +842,10 @@
struct mlx5_eq_context ctx;
};
+enum {
+ MLX5_MKEY_STATUS_FREE = 1 << 6,
+};
+
struct mlx5_mkey_seg {
/* This is a two bit field occupying bits 31-30.
* bit 31 is always 0,
@@ -812,7 +882,7 @@
struct mlx5_create_mkey_mbox_in {
struct mlx5_inbox_hdr hdr;
__be32 input_mkey_index;
- u8 rsvd0[4];
+ __be32 flags;
struct mlx5_mkey_seg seg;
u8 rsvd1[16];
__be32 xlat_oct_act_size;
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index b1bf415..166d931 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -113,6 +113,13 @@
MLX5_REG_HOST_ENDIANNESS = 0x7004,
};
+enum mlx5_page_fault_resume_flags {
+ MLX5_PAGE_FAULT_RESUME_REQUESTOR = 1 << 0,
+ MLX5_PAGE_FAULT_RESUME_WRITE = 1 << 1,
+ MLX5_PAGE_FAULT_RESUME_RDMA = 1 << 2,
+ MLX5_PAGE_FAULT_RESUME_ERROR = 1 << 7,
+};
+
enum dbg_rsc_type {
MLX5_DBG_RSC_QP,
MLX5_DBG_RSC_EQ,
@@ -467,7 +474,7 @@
struct workqueue_struct *pg_wq;
struct rb_root page_root;
int fw_pages;
- int reg_pages;
+ atomic_t reg_pages;
struct list_head free_list;
struct mlx5_core_health health;
@@ -703,6 +710,9 @@
void mlx5_fill_page_array(struct mlx5_buf *buf, __be64 *pas);
void mlx5_cq_completion(struct mlx5_core_dev *dev, u32 cqn);
void mlx5_rsc_event(struct mlx5_core_dev *dev, u32 rsn, int event_type);
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+void mlx5_eq_pagefault(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe);
+#endif
void mlx5_srq_event(struct mlx5_core_dev *dev, u32 srqn, int event_type);
struct mlx5_core_srq *mlx5_core_get_srq(struct mlx5_core_dev *dev, u32 srqn);
void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector);
@@ -740,6 +750,8 @@
int npsvs, u32 *sig_index);
int mlx5_core_destroy_psv(struct mlx5_core_dev *dev, int psv_num);
void mlx5_core_put_rsc(struct mlx5_core_rsc_common *common);
+int mlx5_query_odp_caps(struct mlx5_core_dev *dev,
+ struct mlx5_odp_caps *odp_caps);
static inline u32 mlx5_mkey_to_idx(u32 mkey)
{
diff --git a/include/linux/mlx5/qp.h b/include/linux/mlx5/qp.h
index 3fa075d..61f7a34 100644
--- a/include/linux/mlx5/qp.h
+++ b/include/linux/mlx5/qp.h
@@ -50,6 +50,9 @@
#define MLX5_BSF_APPTAG_ESCAPE 0x1
#define MLX5_BSF_APPREF_ESCAPE 0x2
+#define MLX5_QPN_BITS 24
+#define MLX5_QPN_MASK ((1 << MLX5_QPN_BITS) - 1)
+
enum mlx5_qp_optpar {
MLX5_QP_OPTPAR_ALT_ADDR_PATH = 1 << 0,
MLX5_QP_OPTPAR_RRE = 1 << 1,
@@ -189,6 +192,14 @@
__be32 imm;
};
+#define MLX5_WQE_CTRL_DS_MASK 0x3f
+#define MLX5_WQE_CTRL_QPN_MASK 0xffffff00
+#define MLX5_WQE_CTRL_QPN_SHIFT 8
+#define MLX5_WQE_DS_UNITS 16
+#define MLX5_WQE_CTRL_OPCODE_MASK 0xff
+#define MLX5_WQE_CTRL_WQE_INDEX_MASK 0x00ffff00
+#define MLX5_WQE_CTRL_WQE_INDEX_SHIFT 8
+
struct mlx5_wqe_xrc_seg {
__be32 xrc_srqn;
u8 rsvd[12];
@@ -292,6 +303,8 @@
u8 rsvd1[11];
};
+#define MLX5_WQE_INLINE_SEG_BYTE_COUNT_MASK 0x3ff
+
struct mlx5_wqe_inline_seg {
__be32 byte_count;
};
@@ -360,9 +373,46 @@
__be16 num_entries;
};
+enum mlx5_pagefault_flags {
+ MLX5_PFAULT_REQUESTOR = 1 << 0,
+ MLX5_PFAULT_WRITE = 1 << 1,
+ MLX5_PFAULT_RDMA = 1 << 2,
+};
+
+/* Contains the details of a pagefault. */
+struct mlx5_pagefault {
+ u32 bytes_committed;
+ u8 event_subtype;
+ enum mlx5_pagefault_flags flags;
+ union {
+ /* Initiator or send message responder pagefault details. */
+ struct {
+ /* Received packet size, only valid for responders. */
+ u32 packet_size;
+ /*
+ * WQE index. Refers to either the send queue or
+ * receive queue, according to event_subtype.
+ */
+ u16 wqe_index;
+ } wqe;
+ /* RDMA responder pagefault details */
+ struct {
+ u32 r_key;
+ /*
+ * Received packet size, minimal size page fault
+ * resolution required for forward progress.
+ */
+ u32 packet_size;
+ u32 rdma_op_len;
+ u64 rdma_va;
+ } rdma;
+ };
+};
+
struct mlx5_core_qp {
struct mlx5_core_rsc_common common; /* must be first */
void (*event) (struct mlx5_core_qp *, int);
+ void (*pfault_handler)(struct mlx5_core_qp *, struct mlx5_pagefault *);
int qpn;
struct mlx5_rsc_debug *dbg;
int pid;
@@ -530,6 +580,17 @@
return radix_tree_lookup(&dev->priv.mr_table.tree, key);
}
+struct mlx5_page_fault_resume_mbox_in {
+ struct mlx5_inbox_hdr hdr;
+ __be32 flags_qpn;
+ u8 reserved[4];
+};
+
+struct mlx5_page_fault_resume_mbox_out {
+ struct mlx5_outbox_hdr hdr;
+ u8 rsvd[8];
+};
+
int mlx5_core_create_qp(struct mlx5_core_dev *dev,
struct mlx5_core_qp *qp,
struct mlx5_create_qp_mbox_in *in,
@@ -549,6 +610,10 @@
void mlx5_cleanup_qp_table(struct mlx5_core_dev *dev);
int mlx5_debug_qp_add(struct mlx5_core_dev *dev, struct mlx5_core_qp *qp);
void mlx5_debug_qp_remove(struct mlx5_core_dev *dev, struct mlx5_core_qp *qp);
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+int mlx5_core_page_fault_resume(struct mlx5_core_dev *dev, u32 qpn,
+ u8 context, int error);
+#endif
static inline const char *mlx5_qp_type_str(int type)
{
diff --git a/include/linux/module.h b/include/linux/module.h
index 71f282a..ebfb0e1 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -210,20 +210,6 @@
MODULE_STATE_UNFORMED, /* Still setting it up. */
};
-/**
- * struct module_ref - per cpu module reference counts
- * @incs: number of module get on this cpu
- * @decs: number of module put on this cpu
- *
- * We force an alignment on 8 or 16 bytes, so that alloc_percpu()
- * put @incs/@decs in same cache line, with no extra memory cost,
- * since alloc_percpu() is fine grained.
- */
-struct module_ref {
- unsigned long incs;
- unsigned long decs;
-} __attribute((aligned(2 * sizeof(unsigned long))));
-
struct module {
enum module_state state;
@@ -367,7 +353,7 @@
/* Destruction function. */
void (*exit)(void);
- struct module_ref __percpu *refptr;
+ atomic_t refcnt;
#endif
#ifdef CONFIG_CONSTRUCTORS
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 66a656e..8b59763 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -351,8 +351,6 @@
#define SET_RUNTIME_PM_OPS(suspend_fn, resume_fn, idle_fn)
#endif
-#define SET_PM_RUNTIME_PM_OPS SET_RUNTIME_PM_OPS
-
/*
* Use this if you want to use the same suspend and resume callbacks for suspend
* to RAM and hibernation.
diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h
index 7979f85..ca3ed78 100644
--- a/include/linux/virtio_config.h
+++ b/include/linux/virtio_config.h
@@ -19,6 +19,9 @@
* offset: the offset of the configuration field
* buf: the buffer to read the field value from.
* len: the length of the buffer
+ * @generation: config generation counter
+ * vdev: the virtio_device
+ * Returns the config generation counter
* @get_status: read the status byte
* vdev: the virtio_device
* Returns the status byte
@@ -60,6 +63,7 @@
void *buf, unsigned len);
void (*set)(struct virtio_device *vdev, unsigned offset,
const void *buf, unsigned len);
+ u32 (*generation)(struct virtio_device *vdev);
u8 (*get_status)(struct virtio_device *vdev);
void (*set_status)(struct virtio_device *vdev, u8 status);
void (*reset)(struct virtio_device *vdev);
@@ -301,11 +305,33 @@
return ret;
}
+/* Read @count fields, @bytes each. */
+static inline void __virtio_cread_many(struct virtio_device *vdev,
+ unsigned int offset,
+ void *buf, size_t count, size_t bytes)
+{
+ u32 old, gen = vdev->config->generation ?
+ vdev->config->generation(vdev) : 0;
+ int i;
+
+ do {
+ old = gen;
+
+ for (i = 0; i < count; i++)
+ vdev->config->get(vdev, offset + bytes * i,
+ buf + i * bytes, bytes);
+
+ gen = vdev->config->generation ?
+ vdev->config->generation(vdev) : 0;
+ } while (gen != old);
+}
+
+
static inline void virtio_cread_bytes(struct virtio_device *vdev,
unsigned int offset,
void *buf, size_t len)
{
- vdev->config->get(vdev, offset, buf, len);
+ __virtio_cread_many(vdev, offset, buf, len, 1);
}
static inline void virtio_cwrite8(struct virtio_device *vdev,
@@ -349,6 +375,7 @@
{
u64 ret;
vdev->config->get(vdev, offset, &ret, sizeof(ret));
+ __virtio_cread_many(vdev, offset, &ret, 1, sizeof(ret));
return virtio64_to_cpu(vdev, (__force __virtio64)ret);
}
diff --git a/include/linux/vringh.h b/include/linux/vringh.h
index 749cde2..a3fa537 100644
--- a/include/linux/vringh.h
+++ b/include/linux/vringh.h
@@ -24,12 +24,16 @@
#ifndef _LINUX_VRINGH_H
#define _LINUX_VRINGH_H
#include <uapi/linux/virtio_ring.h>
+#include <linux/virtio_byteorder.h>
#include <linux/uio.h>
#include <linux/slab.h>
#include <asm/barrier.h>
/* virtio_ring with information needed for host access. */
struct vringh {
+ /* Everything is little endian */
+ bool little_endian;
+
/* Guest publishes used event idx (note: we always do). */
bool event_indices;
@@ -105,7 +109,7 @@
#define VRINGH_IOV_ALLOCATED 0x8000000
/* Helpers for userspace vrings. */
-int vringh_init_user(struct vringh *vrh, u32 features,
+int vringh_init_user(struct vringh *vrh, u64 features,
unsigned int num, bool weak_barriers,
struct vring_desc __user *desc,
struct vring_avail __user *avail,
@@ -167,7 +171,7 @@
void vringh_notify_disable_user(struct vringh *vrh);
/* Helpers for kernelspace vrings. */
-int vringh_init_kern(struct vringh *vrh, u32 features,
+int vringh_init_kern(struct vringh *vrh, u64 features,
unsigned int num, bool weak_barriers,
struct vring_desc *desc,
struct vring_avail *avail,
@@ -222,4 +226,33 @@
vrh->notify(vrh);
}
+static inline u16 vringh16_to_cpu(const struct vringh *vrh, __virtio16 val)
+{
+ return __virtio16_to_cpu(vrh->little_endian, val);
+}
+
+static inline __virtio16 cpu_to_vringh16(const struct vringh *vrh, u16 val)
+{
+ return __cpu_to_virtio16(vrh->little_endian, val);
+}
+
+static inline u32 vringh32_to_cpu(const struct vringh *vrh, __virtio32 val)
+{
+ return __virtio32_to_cpu(vrh->little_endian, val);
+}
+
+static inline __virtio32 cpu_to_vringh32(const struct vringh *vrh, u32 val)
+{
+ return __cpu_to_virtio32(vrh->little_endian, val);
+}
+
+static inline u64 vringh64_to_cpu(const struct vringh *vrh, __virtio64 val)
+{
+ return __virtio64_to_cpu(vrh->little_endian, val);
+}
+
+static inline __virtio64 cpu_to_vringh64(const struct vringh *vrh, u64 val)
+{
+ return __cpu_to_virtio64(vrh->little_endian, val);
+}
#endif /* _LINUX_VRINGH_H */
diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h
index a2bf41e..2d83cfd 100644
--- a/include/rdma/ib_umem.h
+++ b/include/rdma/ib_umem.h
@@ -38,11 +38,12 @@
#include <linux/workqueue.h>
struct ib_ucontext;
+struct ib_umem_odp;
struct ib_umem {
struct ib_ucontext *context;
size_t length;
- int offset;
+ unsigned long address;
int page_size;
int writable;
int hugetlb;
@@ -50,17 +51,43 @@
struct pid *pid;
struct mm_struct *mm;
unsigned long diff;
+ struct ib_umem_odp *odp_data;
struct sg_table sg_head;
int nmap;
int npages;
};
+/* Returns the offset of the umem start relative to the first page. */
+static inline int ib_umem_offset(struct ib_umem *umem)
+{
+ return umem->address & ((unsigned long)umem->page_size - 1);
+}
+
+/* Returns the first page of an ODP umem. */
+static inline unsigned long ib_umem_start(struct ib_umem *umem)
+{
+ return umem->address - ib_umem_offset(umem);
+}
+
+/* Returns the address of the page after the last one of an ODP umem. */
+static inline unsigned long ib_umem_end(struct ib_umem *umem)
+{
+ return PAGE_ALIGN(umem->address + umem->length);
+}
+
+static inline size_t ib_umem_num_pages(struct ib_umem *umem)
+{
+ return (ib_umem_end(umem) - ib_umem_start(umem)) >> PAGE_SHIFT;
+}
+
#ifdef CONFIG_INFINIBAND_USER_MEM
struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
size_t size, int access, int dmasync);
void ib_umem_release(struct ib_umem *umem);
int ib_umem_page_count(struct ib_umem *umem);
+int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset,
+ size_t length);
#else /* CONFIG_INFINIBAND_USER_MEM */
@@ -73,7 +100,10 @@
}
static inline void ib_umem_release(struct ib_umem *umem) { }
static inline int ib_umem_page_count(struct ib_umem *umem) { return 0; }
-
+static inline int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset,
+ size_t length) {
+ return -EINVAL;
+}
#endif /* CONFIG_INFINIBAND_USER_MEM */
#endif /* IB_UMEM_H */
diff --git a/include/rdma/ib_umem_odp.h b/include/rdma/ib_umem_odp.h
new file mode 100644
index 0000000..3da0b16
--- /dev/null
+++ b/include/rdma/ib_umem_odp.h
@@ -0,0 +1,160 @@
+/*
+ * Copyright (c) 2014 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * 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 IB_UMEM_ODP_H
+#define IB_UMEM_ODP_H
+
+#include <rdma/ib_umem.h>
+#include <rdma/ib_verbs.h>
+#include <linux/interval_tree.h>
+
+struct umem_odp_node {
+ u64 __subtree_last;
+ struct rb_node rb;
+};
+
+struct ib_umem_odp {
+ /*
+ * An array of the pages included in the on-demand paging umem.
+ * Indices of pages that are currently not mapped into the device will
+ * contain NULL.
+ */
+ struct page **page_list;
+ /*
+ * An array of the same size as page_list, with DMA addresses mapped
+ * for pages the pages in page_list. The lower two bits designate
+ * access permissions. See ODP_READ_ALLOWED_BIT and
+ * ODP_WRITE_ALLOWED_BIT.
+ */
+ dma_addr_t *dma_list;
+ /*
+ * The umem_mutex protects the page_list and dma_list fields of an ODP
+ * umem, allowing only a single thread to map/unmap pages. The mutex
+ * also protects access to the mmu notifier counters.
+ */
+ struct mutex umem_mutex;
+ void *private; /* for the HW driver to use. */
+
+ /* When false, use the notifier counter in the ucontext struct. */
+ bool mn_counters_active;
+ int notifiers_seq;
+ int notifiers_count;
+
+ /* A linked list of umems that don't have private mmu notifier
+ * counters yet. */
+ struct list_head no_private_counters;
+ struct ib_umem *umem;
+
+ /* Tree tracking */
+ struct umem_odp_node interval_tree;
+
+ struct completion notifier_completion;
+ int dying;
+};
+
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+
+int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem);
+
+void ib_umem_odp_release(struct ib_umem *umem);
+
+/*
+ * The lower 2 bits of the DMA address signal the R/W permissions for
+ * the entry. To upgrade the permissions, provide the appropriate
+ * bitmask to the map_dma_pages function.
+ *
+ * Be aware that upgrading a mapped address might result in change of
+ * the DMA address for the page.
+ */
+#define ODP_READ_ALLOWED_BIT (1<<0ULL)
+#define ODP_WRITE_ALLOWED_BIT (1<<1ULL)
+
+#define ODP_DMA_ADDR_MASK (~(ODP_READ_ALLOWED_BIT | ODP_WRITE_ALLOWED_BIT))
+
+int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 start_offset, u64 bcnt,
+ u64 access_mask, unsigned long current_seq);
+
+void ib_umem_odp_unmap_dma_pages(struct ib_umem *umem, u64 start_offset,
+ u64 bound);
+
+void rbt_ib_umem_insert(struct umem_odp_node *node, struct rb_root *root);
+void rbt_ib_umem_remove(struct umem_odp_node *node, struct rb_root *root);
+typedef int (*umem_call_back)(struct ib_umem *item, u64 start, u64 end,
+ void *cookie);
+/*
+ * Call the callback on each ib_umem in the range. Returns the logical or of
+ * the return values of the functions called.
+ */
+int rbt_ib_umem_for_each_in_range(struct rb_root *root, u64 start, u64 end,
+ umem_call_back cb, void *cookie);
+
+struct umem_odp_node *rbt_ib_umem_iter_first(struct rb_root *root,
+ u64 start, u64 last);
+struct umem_odp_node *rbt_ib_umem_iter_next(struct umem_odp_node *node,
+ u64 start, u64 last);
+
+static inline int ib_umem_mmu_notifier_retry(struct ib_umem *item,
+ unsigned long mmu_seq)
+{
+ /*
+ * This code is strongly based on the KVM code from
+ * mmu_notifier_retry. Should be called with
+ * the relevant locks taken (item->odp_data->umem_mutex
+ * and the ucontext umem_mutex semaphore locked for read).
+ */
+
+ /* Do not allow page faults while the new ib_umem hasn't seen a state
+ * with zero notifiers yet, and doesn't have its own valid set of
+ * private counters. */
+ if (!item->odp_data->mn_counters_active)
+ return 1;
+
+ if (unlikely(item->odp_data->notifiers_count))
+ return 1;
+ if (item->odp_data->notifiers_seq != mmu_seq)
+ return 1;
+ return 0;
+}
+
+#else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */
+
+static inline int ib_umem_odp_get(struct ib_ucontext *context,
+ struct ib_umem *umem)
+{
+ return -EINVAL;
+}
+
+static inline void ib_umem_odp_release(struct ib_umem *umem) {}
+
+#endif /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */
+
+#endif /* IB_UMEM_ODP_H */
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 470a011..0d74f1d 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -51,6 +51,7 @@
#include <uapi/linux/if_ether.h>
#include <linux/atomic.h>
+#include <linux/mmu_notifier.h>
#include <asm/uaccess.h>
extern struct workqueue_struct *ib_wq;
@@ -123,7 +124,8 @@
IB_DEVICE_MEM_WINDOW_TYPE_2A = (1<<23),
IB_DEVICE_MEM_WINDOW_TYPE_2B = (1<<24),
IB_DEVICE_MANAGED_FLOW_STEERING = (1<<29),
- IB_DEVICE_SIGNATURE_HANDOVER = (1<<30)
+ IB_DEVICE_SIGNATURE_HANDOVER = (1<<30),
+ IB_DEVICE_ON_DEMAND_PAGING = (1<<31),
};
enum ib_signature_prot_cap {
@@ -143,6 +145,27 @@
IB_ATOMIC_GLOB
};
+enum ib_odp_general_cap_bits {
+ IB_ODP_SUPPORT = 1 << 0,
+};
+
+enum ib_odp_transport_cap_bits {
+ IB_ODP_SUPPORT_SEND = 1 << 0,
+ IB_ODP_SUPPORT_RECV = 1 << 1,
+ IB_ODP_SUPPORT_WRITE = 1 << 2,
+ IB_ODP_SUPPORT_READ = 1 << 3,
+ IB_ODP_SUPPORT_ATOMIC = 1 << 4,
+};
+
+struct ib_odp_caps {
+ uint64_t general_caps;
+ struct {
+ uint32_t rc_odp_caps;
+ uint32_t uc_odp_caps;
+ uint32_t ud_odp_caps;
+ } per_transport_caps;
+};
+
struct ib_device_attr {
u64 fw_ver;
__be64 sys_image_guid;
@@ -186,6 +209,7 @@
u8 local_ca_ack_delay;
int sig_prot_cap;
int sig_guard_cap;
+ struct ib_odp_caps odp_caps;
};
enum ib_mtu {
@@ -1073,7 +1097,8 @@
IB_ACCESS_REMOTE_READ = (1<<2),
IB_ACCESS_REMOTE_ATOMIC = (1<<3),
IB_ACCESS_MW_BIND = (1<<4),
- IB_ZERO_BASED = (1<<5)
+ IB_ZERO_BASED = (1<<5),
+ IB_ACCESS_ON_DEMAND = (1<<6),
};
struct ib_phys_buf {
@@ -1115,6 +1140,8 @@
u8 page_shift;
};
+struct ib_umem;
+
struct ib_ucontext {
struct ib_device *device;
struct list_head pd_list;
@@ -1127,6 +1154,24 @@
struct list_head xrcd_list;
struct list_head rule_list;
int closing;
+
+ struct pid *tgid;
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ struct rb_root umem_tree;
+ /*
+ * Protects .umem_rbroot and tree, as well as odp_mrs_count and
+ * mmu notifiers registration.
+ */
+ struct rw_semaphore umem_rwsem;
+ void (*invalidate_range)(struct ib_umem *umem,
+ unsigned long start, unsigned long end);
+
+ struct mmu_notifier mn;
+ atomic_t notifier_count;
+ /* A list of umems that don't have private mmu notifier counters yet. */
+ struct list_head no_private_counters;
+ int odp_mrs_count;
+#endif
};
struct ib_uobject {
@@ -1662,7 +1707,10 @@
static inline int ib_copy_to_udata(struct ib_udata *udata, void *src, size_t len)
{
- return copy_to_user(udata->outbuf, src, len) ? -EFAULT : 0;
+ size_t copy_sz;
+
+ copy_sz = min_t(size_t, len, udata->outlen);
+ return copy_to_user(udata->outbuf, src, copy_sz) ? -EFAULT : 0;
}
/**
diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h
index 6364e23..3a4edd1 100644
--- a/include/scsi/scsi_device.h
+++ b/include/scsi/scsi_device.h
@@ -441,13 +441,13 @@
extern void sdev_disable_disk_events(struct scsi_device *sdev);
extern void sdev_enable_disk_events(struct scsi_device *sdev);
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
extern int scsi_autopm_get_device(struct scsi_device *);
extern void scsi_autopm_put_device(struct scsi_device *);
#else
static inline int scsi_autopm_get_device(struct scsi_device *d) { return 0; }
static inline void scsi_autopm_put_device(struct scsi_device *d) {}
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
static inline int __must_check scsi_device_reprobe(struct scsi_device *sdev)
{
diff --git a/include/trace/events/module.h b/include/trace/events/module.h
index 7c5cbfe..81c4c18 100644
--- a/include/trace/events/module.h
+++ b/include/trace/events/module.h
@@ -80,7 +80,7 @@
TP_fast_assign(
__entry->ip = ip;
- __entry->refcnt = __this_cpu_read(mod->refptr->incs) - __this_cpu_read(mod->refptr->decs);
+ __entry->refcnt = atomic_read(&mod->refcnt);
__assign_str(name, mod->name);
),
diff --git a/include/uapi/linux/v4l2-mediabus.h b/include/uapi/linux/v4l2-mediabus.h
index 5a86d8e..26db206 100644
--- a/include/uapi/linux/v4l2-mediabus.h
+++ b/include/uapi/linux/v4l2-mediabus.h
@@ -31,9 +31,9 @@
__u32 code;
__u32 field;
__u32 colorspace;
- __u32 ycbcr_enc;
- __u32 quantization;
- __u32 reserved[5];
+ __u16 ycbcr_enc;
+ __u16 quantization;
+ __u32 reserved[6];
};
#ifndef __KERNEL__
diff --git a/include/uapi/linux/virtio_pci.h b/include/uapi/linux/virtio_pci.h
index e5ec1ca..35b552c7 100644
--- a/include/uapi/linux/virtio_pci.h
+++ b/include/uapi/linux/virtio_pci.h
@@ -41,6 +41,8 @@
#include <linux/virtio_config.h>
+#ifndef VIRTIO_PCI_NO_LEGACY
+
/* A 32-bit r/o bitmask of the features supported by the host */
#define VIRTIO_PCI_HOST_FEATURES 0
@@ -67,16 +69,11 @@
* a read-and-acknowledge. */
#define VIRTIO_PCI_ISR 19
-/* The bit of the ISR which indicates a device configuration change. */
-#define VIRTIO_PCI_ISR_CONFIG 0x2
-
/* MSI-X registers: only enabled if MSI-X is enabled. */
/* A 16-bit vector for configuration changes. */
#define VIRTIO_MSI_CONFIG_VECTOR 20
/* A 16-bit vector for selected queue notifications. */
#define VIRTIO_MSI_QUEUE_VECTOR 22
-/* Vector value used to disable MSI for queue */
-#define VIRTIO_MSI_NO_VECTOR 0xffff
/* The remaining space is defined by each driver as the per-driver
* configuration space */
@@ -94,4 +91,12 @@
/* The alignment to use between consumer and producer parts of vring.
* x86 pagesize again. */
#define VIRTIO_PCI_VRING_ALIGN 4096
+
+#endif /* VIRTIO_PCI_NO_LEGACY */
+
+/* The bit of the ISR which indicates a device configuration change. */
+#define VIRTIO_PCI_ISR_CONFIG 0x2
+/* Vector value used to disable MSI for queue */
+#define VIRTIO_MSI_NO_VECTOR 0xffff
+
#endif
diff --git a/include/uapi/rdma/ib_user_verbs.h b/include/uapi/rdma/ib_user_verbs.h
index 26daf55..4275b96 100644
--- a/include/uapi/rdma/ib_user_verbs.h
+++ b/include/uapi/rdma/ib_user_verbs.h
@@ -90,8 +90,9 @@
};
enum {
+ IB_USER_VERBS_EX_CMD_QUERY_DEVICE = IB_USER_VERBS_CMD_QUERY_DEVICE,
IB_USER_VERBS_EX_CMD_CREATE_FLOW = IB_USER_VERBS_CMD_THRESHOLD,
- IB_USER_VERBS_EX_CMD_DESTROY_FLOW
+ IB_USER_VERBS_EX_CMD_DESTROY_FLOW,
};
/*
@@ -201,6 +202,32 @@
__u8 reserved[4];
};
+enum {
+ IB_USER_VERBS_EX_QUERY_DEVICE_ODP = 1ULL << 0,
+};
+
+struct ib_uverbs_ex_query_device {
+ __u32 comp_mask;
+ __u32 reserved;
+};
+
+struct ib_uverbs_odp_caps {
+ __u64 general_caps;
+ struct {
+ __u32 rc_odp_caps;
+ __u32 uc_odp_caps;
+ __u32 ud_odp_caps;
+ } per_transport_caps;
+ __u32 reserved;
+};
+
+struct ib_uverbs_ex_query_device_resp {
+ struct ib_uverbs_query_device_resp base;
+ __u32 comp_mask;
+ __u32 reserved;
+ struct ib_uverbs_odp_caps odp_caps;
+};
+
struct ib_uverbs_query_port {
__u64 response;
__u8 port_num;
diff --git a/kernel/module.c b/kernel/module.c
index e52a873..3965511 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -42,7 +42,6 @@
#include <linux/vermagic.h>
#include <linux/notifier.h>
#include <linux/sched.h>
-#include <linux/stop_machine.h>
#include <linux/device.h>
#include <linux/string.h>
#include <linux/mutex.h>
@@ -98,7 +97,7 @@
* 1) List of modules (also safely readable with preempt_disable),
* 2) module_use links,
* 3) module_addr_min/module_addr_max.
- * (delete uses stop_machine/add uses RCU list operations). */
+ * (delete and add uses RCU list operations). */
DEFINE_MUTEX(module_mutex);
EXPORT_SYMBOL_GPL(module_mutex);
static LIST_HEAD(modules);
@@ -158,13 +157,13 @@
* Protected by module_mutex. */
static unsigned long module_addr_min = -1UL, module_addr_max = 0;
-int register_module_notifier(struct notifier_block * nb)
+int register_module_notifier(struct notifier_block *nb)
{
return blocking_notifier_chain_register(&module_notify_list, nb);
}
EXPORT_SYMBOL(register_module_notifier);
-int unregister_module_notifier(struct notifier_block * nb)
+int unregister_module_notifier(struct notifier_block *nb)
{
return blocking_notifier_chain_unregister(&module_notify_list, nb);
}
@@ -628,18 +627,23 @@
EXPORT_TRACEPOINT_SYMBOL(module_get);
+/* MODULE_REF_BASE is the base reference count by kmodule loader. */
+#define MODULE_REF_BASE 1
+
/* Init the unload section of the module. */
static int module_unload_init(struct module *mod)
{
- mod->refptr = alloc_percpu(struct module_ref);
- if (!mod->refptr)
- return -ENOMEM;
+ /*
+ * Initialize reference counter to MODULE_REF_BASE.
+ * refcnt == 0 means module is going.
+ */
+ atomic_set(&mod->refcnt, MODULE_REF_BASE);
INIT_LIST_HEAD(&mod->source_list);
INIT_LIST_HEAD(&mod->target_list);
/* Hold reference count during initialization. */
- raw_cpu_write(mod->refptr->incs, 1);
+ atomic_inc(&mod->refcnt);
return 0;
}
@@ -721,8 +725,6 @@
kfree(use);
}
mutex_unlock(&module_mutex);
-
- free_percpu(mod->refptr);
}
#ifdef CONFIG_MODULE_FORCE_UNLOAD
@@ -740,60 +742,39 @@
}
#endif /* CONFIG_MODULE_FORCE_UNLOAD */
-struct stopref
+/* Try to release refcount of module, 0 means success. */
+static int try_release_module_ref(struct module *mod)
{
- struct module *mod;
- int flags;
- int *forced;
-};
+ int ret;
-/* Whole machine is stopped with interrupts off when this runs. */
-static int __try_stop_module(void *_sref)
-{
- struct stopref *sref = _sref;
+ /* Try to decrement refcnt which we set at loading */
+ ret = atomic_sub_return(MODULE_REF_BASE, &mod->refcnt);
+ BUG_ON(ret < 0);
+ if (ret)
+ /* Someone can put this right now, recover with checking */
+ ret = atomic_add_unless(&mod->refcnt, MODULE_REF_BASE, 0);
- /* If it's not unused, quit unless we're forcing. */
- if (module_refcount(sref->mod) != 0) {
- if (!(*sref->forced = try_force_unload(sref->flags)))
- return -EWOULDBLOCK;
- }
-
- /* Mark it as dying. */
- sref->mod->state = MODULE_STATE_GOING;
- return 0;
+ return ret;
}
static int try_stop_module(struct module *mod, int flags, int *forced)
{
- struct stopref sref = { mod, flags, forced };
+ /* If it's not unused, quit unless we're forcing. */
+ if (try_release_module_ref(mod) != 0) {
+ *forced = try_force_unload(flags);
+ if (!(*forced))
+ return -EWOULDBLOCK;
+ }
- return stop_machine(__try_stop_module, &sref, NULL);
+ /* Mark it as dying. */
+ mod->state = MODULE_STATE_GOING;
+
+ return 0;
}
unsigned long module_refcount(struct module *mod)
{
- unsigned long incs = 0, decs = 0;
- int cpu;
-
- for_each_possible_cpu(cpu)
- decs += per_cpu_ptr(mod->refptr, cpu)->decs;
- /*
- * ensure the incs are added up after the decs.
- * module_put ensures incs are visible before decs with smp_wmb.
- *
- * This 2-count scheme avoids the situation where the refcount
- * for CPU0 is read, then CPU0 increments the module refcount,
- * then CPU1 drops that refcount, then the refcount for CPU1 is
- * read. We would record a decrement but not its corresponding
- * increment so we would see a low count (disaster).
- *
- * Rare situation? But module_refcount can be preempted, and we
- * might be tallying up 4096+ CPUs. So it is not impossible.
- */
- smp_rmb();
- for_each_possible_cpu(cpu)
- incs += per_cpu_ptr(mod->refptr, cpu)->incs;
- return incs - decs;
+ return (unsigned long)atomic_read(&mod->refcnt) - MODULE_REF_BASE;
}
EXPORT_SYMBOL(module_refcount);
@@ -877,8 +858,10 @@
seq_printf(m, " %lu ", module_refcount(mod));
- /* Always include a trailing , so userspace can differentiate
- between this and the old multi-field proc format. */
+ /*
+ * Always include a trailing , so userspace can differentiate
+ * between this and the old multi-field proc format.
+ */
list_for_each_entry(use, &mod->source_list, source_list) {
printed_something = 1;
seq_printf(m, "%s,", use->source->name);
@@ -886,11 +869,11 @@
if (mod->init != NULL && mod->exit == NULL) {
printed_something = 1;
- seq_printf(m, "[permanent],");
+ seq_puts(m, "[permanent],");
}
if (!printed_something)
- seq_printf(m, "-");
+ seq_puts(m, "-");
}
void __symbol_put(const char *symbol)
@@ -935,7 +918,7 @@
{
if (module) {
preempt_disable();
- __this_cpu_inc(module->refptr->incs);
+ atomic_inc(&module->refcnt);
trace_module_get(module, _RET_IP_);
preempt_enable();
}
@@ -948,11 +931,11 @@
if (module) {
preempt_disable();
-
- if (likely(module_is_live(module))) {
- __this_cpu_inc(module->refptr->incs);
+ /* Note: here, we can fail to get a reference */
+ if (likely(module_is_live(module) &&
+ atomic_inc_not_zero(&module->refcnt) != 0))
trace_module_get(module, _RET_IP_);
- } else
+ else
ret = false;
preempt_enable();
@@ -963,11 +946,12 @@
void module_put(struct module *module)
{
+ int ret;
+
if (module) {
preempt_disable();
- smp_wmb(); /* see comment in module_refcount */
- __this_cpu_inc(module->refptr->decs);
-
+ ret = atomic_dec_if_positive(&module->refcnt);
+ WARN_ON(ret < 0); /* Failed to put refcount */
trace_module_put(module, _RET_IP_);
preempt_enable();
}
@@ -978,7 +962,7 @@
static inline void print_unload_info(struct seq_file *m, struct module *mod)
{
/* We don't know the usage count, or what modules are using. */
- seq_printf(m, " - -");
+ seq_puts(m, " - -");
}
static inline void module_unload_free(struct module *mod)
@@ -1131,7 +1115,7 @@
static int check_version(Elf_Shdr *sechdrs,
unsigned int versindex,
const char *symname,
- struct module *mod,
+ struct module *mod,
const unsigned long *crc,
const struct module *crc_owner)
{
@@ -1165,7 +1149,7 @@
return 0;
bad_version:
- printk("%s: disagrees about version of symbol %s\n",
+ pr_warn("%s: disagrees about version of symbol %s\n",
mod->name, symname);
return 0;
}
@@ -1200,7 +1184,7 @@
static inline int check_version(Elf_Shdr *sechdrs,
unsigned int versindex,
const char *symname,
- struct module *mod,
+ struct module *mod,
const unsigned long *crc,
const struct module *crc_owner)
{
@@ -1288,15 +1272,13 @@
return !(sect->sh_flags & SHF_ALLOC) || sect->sh_size == 0;
}
-struct module_sect_attr
-{
+struct module_sect_attr {
struct module_attribute mattr;
char *name;
unsigned long address;
};
-struct module_sect_attrs
-{
+struct module_sect_attrs {
struct attribute_group grp;
unsigned int nsections;
struct module_sect_attr attrs[0];
@@ -1550,7 +1532,8 @@
(attr->test && attr->test(mod))) {
memcpy(temp_attr, attr, sizeof(*temp_attr));
sysfs_attr_init(&temp_attr->attr);
- error = sysfs_create_file(&mod->mkobj.kobj,&temp_attr->attr);
+ error = sysfs_create_file(&mod->mkobj.kobj,
+ &temp_attr->attr);
++temp_attr;
}
}
@@ -1566,7 +1549,7 @@
/* pick a field to test for end of list */
if (!attr->attr.name)
break;
- sysfs_remove_file(&mod->mkobj.kobj,&attr->attr);
+ sysfs_remove_file(&mod->mkobj.kobj, &attr->attr);
if (attr->free)
attr->free(mod);
}
@@ -1697,18 +1680,6 @@
mod_sysfs_fini(mod);
}
-/*
- * unlink the module with the whole machine is stopped with interrupts off
- * - this defends against kallsyms not taking locks
- */
-static int __unlink_module(void *_mod)
-{
- struct module *mod = _mod;
- list_del(&mod->list);
- module_bug_cleanup(mod);
- return 0;
-}
-
#ifdef CONFIG_DEBUG_SET_MODULE_RONX
/*
* LKM RO/NX protection: protect module's text/ro-data
@@ -1860,7 +1831,12 @@
/* Now we can delete it from the lists */
mutex_lock(&module_mutex);
- stop_machine(__unlink_module, mod, NULL);
+ /* Unlink carefully: kallsyms could be walking list. */
+ list_del_rcu(&mod->list);
+ /* Remove this module from bug list, this uses list_del_rcu */
+ module_bug_cleanup(mod);
+ /* Wait for RCU synchronizing before releasing mod->list and buglist. */
+ synchronize_rcu();
mutex_unlock(&module_mutex);
/* This may be NULL, but that's OK */
@@ -1955,7 +1931,7 @@
/* We compiled with -fno-common. These are not
supposed to happen. */
pr_debug("Common symbol: %s\n", name);
- printk("%s: please compile with -fno-common\n",
+ pr_warn("%s: please compile with -fno-common\n",
mod->name);
ret = -ENOEXEC;
break;
@@ -2259,7 +2235,7 @@
}
static bool is_core_symbol(const Elf_Sym *src, const Elf_Shdr *sechdrs,
- unsigned int shnum)
+ unsigned int shnum)
{
const Elf_Shdr *sec;
@@ -2735,7 +2711,7 @@
* This shouldn't happen with same compiler and binutils
* building all parts of the module.
*/
- printk(KERN_WARNING "%s: has both .ctors and .init_array.\n",
+ pr_warn("%s: has both .ctors and .init_array.\n",
mod->name);
return -EINVAL;
}
@@ -3023,8 +2999,10 @@
if (mod->init != NULL)
ret = do_one_initcall(mod->init);
if (ret < 0) {
- /* Init routine failed: abort. Try to protect us from
- buggy refcounters. */
+ /*
+ * Init routine failed: abort. Try to protect us from
+ * buggy refcounters.
+ */
mod->state = MODULE_STATE_GOING;
synchronize_sched();
module_put(mod);
@@ -3202,7 +3180,7 @@
static int unknown_module_param_cb(char *param, char *val, const char *modname)
{
- /* Check for magic 'dyndbg' arg */
+ /* Check for magic 'dyndbg' arg */
int ret = ddebug_dyndbg_module_param_cb(param, val, modname);
if (ret != 0)
pr_warn("%s: unknown parameter '%s' ignored\n", modname, param);
@@ -3352,6 +3330,8 @@
/* Unlink carefully: kallsyms could be walking list. */
list_del_rcu(&mod->list);
wake_up_all(&module_wq);
+ /* Wait for RCU synchronizing before releasing mod->list. */
+ synchronize_rcu();
mutex_unlock(&module_mutex);
free_module:
module_deallocate(mod, info);
@@ -3685,8 +3665,8 @@
/* Informative for users. */
seq_printf(m, " %s",
- mod->state == MODULE_STATE_GOING ? "Unloading":
- mod->state == MODULE_STATE_COMING ? "Loading":
+ mod->state == MODULE_STATE_GOING ? "Unloading" :
+ mod->state == MODULE_STATE_COMING ? "Loading" :
"Live");
/* Used by oprofile and other similar tools. */
seq_printf(m, " 0x%pK", mod->module_core);
@@ -3695,7 +3675,7 @@
if (mod->taints)
seq_printf(m, " %s", module_flags(mod, buf));
- seq_printf(m, "\n");
+ seq_puts(m, "\n");
return 0;
}
diff --git a/kernel/params.c b/kernel/params.c
index db97b79..0af9b2c 100644
--- a/kernel/params.c
+++ b/kernel/params.c
@@ -603,74 +603,67 @@
const struct kernel_param *kp,
const char *name)
{
- struct module_param_attrs *new;
- struct attribute **attrs;
- int err, num;
+ struct module_param_attrs *new_mp;
+ struct attribute **new_attrs;
+ unsigned int i;
/* We don't bother calling this with invisible parameters. */
BUG_ON(!kp->perm);
if (!mk->mp) {
- num = 0;
- attrs = NULL;
- } else {
- num = mk->mp->num;
- attrs = mk->mp->grp.attrs;
+ /* First allocation. */
+ mk->mp = kzalloc(sizeof(*mk->mp), GFP_KERNEL);
+ if (!mk->mp)
+ return -ENOMEM;
+ mk->mp->grp.name = "parameters";
+ /* NULL-terminated attribute array. */
+ mk->mp->grp.attrs = kzalloc(sizeof(mk->mp->grp.attrs[0]),
+ GFP_KERNEL);
+ /* Caller will cleanup via free_module_param_attrs */
+ if (!mk->mp->grp.attrs)
+ return -ENOMEM;
}
- /* Enlarge. */
- new = krealloc(mk->mp,
- sizeof(*mk->mp) + sizeof(mk->mp->attrs[0]) * (num+1),
- GFP_KERNEL);
- if (!new) {
- kfree(attrs);
- err = -ENOMEM;
- goto fail;
- }
- /* Despite looking like the typical realloc() bug, this is safe.
- * We *want* the old 'attrs' to be freed either way, and we'll store
- * the new one in the success case. */
- attrs = krealloc(attrs, sizeof(new->grp.attrs[0])*(num+2), GFP_KERNEL);
- if (!attrs) {
- err = -ENOMEM;
- goto fail_free_new;
- }
+ /* Enlarge allocations. */
+ new_mp = krealloc(mk->mp,
+ sizeof(*mk->mp) +
+ sizeof(mk->mp->attrs[0]) * (mk->mp->num + 1),
+ GFP_KERNEL);
+ if (!new_mp)
+ return -ENOMEM;
+ mk->mp = new_mp;
- /* Sysfs wants everything zeroed. */
- memset(new, 0, sizeof(*new));
- memset(&new->attrs[num], 0, sizeof(new->attrs[num]));
- memset(&attrs[num], 0, sizeof(attrs[num]));
- new->grp.name = "parameters";
- new->grp.attrs = attrs;
+ /* Extra pointer for NULL terminator */
+ new_attrs = krealloc(mk->mp->grp.attrs,
+ sizeof(mk->mp->grp.attrs[0]) * (mk->mp->num + 2),
+ GFP_KERNEL);
+ if (!new_attrs)
+ return -ENOMEM;
+ mk->mp->grp.attrs = new_attrs;
/* Tack new one on the end. */
- sysfs_attr_init(&new->attrs[num].mattr.attr);
- new->attrs[num].param = kp;
- new->attrs[num].mattr.show = param_attr_show;
- new->attrs[num].mattr.store = param_attr_store;
- new->attrs[num].mattr.attr.name = (char *)name;
- new->attrs[num].mattr.attr.mode = kp->perm;
- new->num = num+1;
+ sysfs_attr_init(&mk->mp->attrs[mk->mp->num].mattr.attr);
+ mk->mp->attrs[mk->mp->num].param = kp;
+ mk->mp->attrs[mk->mp->num].mattr.show = param_attr_show;
+ /* Do not allow runtime DAC changes to make param writable. */
+ if ((kp->perm & (S_IWUSR | S_IWGRP | S_IWOTH)) != 0)
+ mk->mp->attrs[mk->mp->num].mattr.store = param_attr_store;
+ mk->mp->attrs[mk->mp->num].mattr.attr.name = (char *)name;
+ mk->mp->attrs[mk->mp->num].mattr.attr.mode = kp->perm;
+ mk->mp->num++;
/* Fix up all the pointers, since krealloc can move us */
- for (num = 0; num < new->num; num++)
- new->grp.attrs[num] = &new->attrs[num].mattr.attr;
- new->grp.attrs[num] = NULL;
-
- mk->mp = new;
+ for (i = 0; i < mk->mp->num; i++)
+ mk->mp->grp.attrs[i] = &mk->mp->attrs[i].mattr.attr;
+ mk->mp->grp.attrs[mk->mp->num] = NULL;
return 0;
-
-fail_free_new:
- kfree(new);
-fail:
- mk->mp = NULL;
- return err;
}
#ifdef CONFIG_MODULES
static void free_module_param_attrs(struct module_kobject *mk)
{
- kfree(mk->mp->grp.attrs);
+ if (mk->mp)
+ kfree(mk->mp->grp.attrs);
kfree(mk->mp);
mk->mp = NULL;
}
@@ -695,8 +688,10 @@
if (kparam[i].perm == 0)
continue;
err = add_sysfs_param(&mod->mkobj, &kparam[i], kparam[i].name);
- if (err)
+ if (err) {
+ free_module_param_attrs(&mod->mkobj);
return err;
+ }
params = true;
}
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index 67d6369..979ccde 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -55,7 +55,7 @@
obj-$(CONFIG_EVENT_TRACING) += trace_events_trigger.o
obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o
obj-$(CONFIG_TRACEPOINTS) += power-traces.o
-ifeq ($(CONFIG_PM_RUNTIME),y)
+ifeq ($(CONFIG_PM),y)
obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o
endif
ifeq ($(CONFIG_TRACING),y)
diff --git a/lib/bug.c b/lib/bug.c
index d1d7c78..0c3bd95 100644
--- a/lib/bug.c
+++ b/lib/bug.c
@@ -64,16 +64,22 @@
static const struct bug_entry *module_find_bug(unsigned long bugaddr)
{
struct module *mod;
+ const struct bug_entry *bug = NULL;
- list_for_each_entry(mod, &module_bug_list, bug_list) {
- const struct bug_entry *bug = mod->bug_table;
+ rcu_read_lock();
+ list_for_each_entry_rcu(mod, &module_bug_list, bug_list) {
unsigned i;
+ bug = mod->bug_table;
for (i = 0; i < mod->num_bugs; ++i, ++bug)
if (bugaddr == bug_addr(bug))
- return bug;
+ goto out;
}
- return NULL;
+ bug = NULL;
+out:
+ rcu_read_unlock();
+
+ return bug;
}
void module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs,
@@ -99,13 +105,15 @@
* Strictly speaking this should have a spinlock to protect against
* traversals, but since we only traverse on BUG()s, a spinlock
* could potentially lead to deadlock and thus be counter-productive.
+ * Thus, this uses RCU to safely manipulate the bug list, since BUG
+ * must run in non-interruptive state.
*/
- list_add(&mod->bug_list, &module_bug_list);
+ list_add_rcu(&mod->bug_list, &module_bug_list);
}
void module_bug_cleanup(struct module *mod)
{
- list_del(&mod->bug_list);
+ list_del_rcu(&mod->bug_list);
}
#else
diff --git a/lib/show_mem.c b/lib/show_mem.c
index 5e25627..7de89f4 100644
--- a/lib/show_mem.c
+++ b/lib/show_mem.c
@@ -8,6 +8,7 @@
#include <linux/mm.h>
#include <linux/nmi.h>
#include <linux/quicklist.h>
+#include <linux/cma.h>
void show_mem(unsigned int filter)
{
@@ -38,7 +39,12 @@
printk("%lu pages RAM\n", total);
printk("%lu pages HighMem/MovableOnly\n", highmem);
+#ifdef CONFIG_CMA
+ printk("%lu pages reserved\n", (reserved - totalcma_pages));
+ printk("%lu pages cma reserved\n", totalcma_pages);
+#else
printk("%lu pages reserved\n", reserved);
+#endif
#ifdef CONFIG_QUICKLIST
printk("%lu pages in pagetable cache\n",
quicklist_total_size());
diff --git a/mm/cma.c b/mm/cma.c
index f891762..a85ae28 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -337,6 +337,7 @@
if (ret)
goto err;
+ totalcma_pages += (size / PAGE_SIZE);
pr_info("Reserved %ld MiB at %pa\n", (unsigned long)size / SZ_1M,
&base);
return 0;
diff --git a/mm/memory.c b/mm/memory.c
index 6efe36a..d8aebc5 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2996,6 +2996,12 @@
if (set_page_dirty(fault_page))
dirtied = 1;
+ /*
+ * Take a local copy of the address_space - page.mapping may be zeroed
+ * by truncate after unlock_page(). The address_space itself remains
+ * pinned by vma->vm_file's reference. We rely on unlock_page()'s
+ * release semantics to prevent the compiler from undoing this copying.
+ */
mapping = fault_page->mapping;
unlock_page(fault_page);
if ((dirtied || vma->vm_ops->page_mkwrite) && mapping) {
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index e58725a..f22c559 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -162,12 +162,6 @@
enum mpol_rebind_step step);
} mpol_ops[MPOL_MAX];
-/* Check that the nodemask contains at least one populated zone */
-static int is_valid_nodemask(const nodemask_t *nodemask)
-{
- return nodes_intersects(*nodemask, node_states[N_MEMORY]);
-}
-
static inline int mpol_store_user_nodemask(const struct mempolicy *pol)
{
return pol->flags & MPOL_MODE_FLAGS;
@@ -202,7 +196,7 @@
static int mpol_new_bind(struct mempolicy *pol, const nodemask_t *nodes)
{
- if (!is_valid_nodemask(nodes))
+ if (nodes_empty(*nodes))
return -EINVAL;
pol->v.nodes = *nodes;
return 0;
@@ -234,7 +228,7 @@
nodes = NULL; /* explicit local allocation */
else {
if (pol->flags & MPOL_F_RELATIVE_NODES)
- mpol_relative_nodemask(&nsc->mask2, nodes,&nsc->mask1);
+ mpol_relative_nodemask(&nsc->mask2, nodes, &nsc->mask1);
else
nodes_and(nsc->mask2, *nodes, nsc->mask1);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index fa974d8..7633c50 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -111,6 +111,7 @@
unsigned long totalram_pages __read_mostly;
unsigned long totalreserve_pages __read_mostly;
+unsigned long totalcma_pages __read_mostly;
/*
* When calculating the number of globally allowed dirty pages, there
* is a certain number of per-zone reserves that should not be
@@ -5586,7 +5587,7 @@
pr_info("Memory: %luK/%luK available "
"(%luK kernel code, %luK rwdata, %luK rodata, "
- "%luK init, %luK bss, %luK reserved"
+ "%luK init, %luK bss, %luK reserved, %luK cma-reserved"
#ifdef CONFIG_HIGHMEM
", %luK highmem"
#endif
@@ -5594,7 +5595,8 @@
nr_free_pages() << (PAGE_SHIFT-10), physpages << (PAGE_SHIFT-10),
codesize >> 10, datasize >> 10, rosize >> 10,
(init_data_size + init_code_size) >> 10, bss_size >> 10,
- (physpages - totalram_pages) << (PAGE_SHIFT-10),
+ (physpages - totalram_pages - totalcma_pages) << (PAGE_SHIFT-10),
+ totalcma_pages << (PAGE_SHIFT-10),
#ifdef CONFIG_HIGHMEM
totalhigh_pages << (PAGE_SHIFT-10),
#endif
diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
index 4d0a063..b724039 100644
--- a/mm/zsmalloc.c
+++ b/mm/zsmalloc.c
@@ -884,19 +884,6 @@
.notifier_call = zs_cpu_notifier
};
-static void zs_unregister_cpu_notifier(void)
-{
- int cpu;
-
- cpu_notifier_register_begin();
-
- for_each_online_cpu(cpu)
- zs_cpu_notifier(NULL, CPU_DEAD, (void *)(long)cpu);
- __unregister_cpu_notifier(&zs_cpu_nb);
-
- cpu_notifier_register_done();
-}
-
static int zs_register_cpu_notifier(void)
{
int cpu, uninitialized_var(ret);
@@ -914,6 +901,19 @@
return notifier_to_errno(ret);
}
+static void zs_unregister_cpu_notifier(void)
+{
+ int cpu;
+
+ cpu_notifier_register_begin();
+
+ for_each_online_cpu(cpu)
+ zs_cpu_notifier(NULL, CPU_DEAD, (void *)(long)cpu);
+ __unregister_cpu_notifier(&zs_cpu_nb);
+
+ cpu_notifier_register_done();
+}
+
static void init_zs_size_classes(void)
{
int nr;
@@ -925,31 +925,6 @@
zs_size_classes = nr;
}
-static void __exit zs_exit(void)
-{
-#ifdef CONFIG_ZPOOL
- zpool_unregister_driver(&zs_zpool_driver);
-#endif
- zs_unregister_cpu_notifier();
-}
-
-static int __init zs_init(void)
-{
- int ret = zs_register_cpu_notifier();
-
- if (ret) {
- zs_unregister_cpu_notifier();
- return ret;
- }
-
- init_zs_size_classes();
-
-#ifdef CONFIG_ZPOOL
- zpool_register_driver(&zs_zpool_driver);
-#endif
- return 0;
-}
-
static unsigned int get_maxobj_per_zspage(int size, int pages_per_zspage)
{
return pages_per_zspage * PAGE_SIZE / size;
@@ -967,6 +942,202 @@
return true;
}
+unsigned long zs_get_total_pages(struct zs_pool *pool)
+{
+ return atomic_long_read(&pool->pages_allocated);
+}
+EXPORT_SYMBOL_GPL(zs_get_total_pages);
+
+/**
+ * zs_map_object - get address of allocated object from handle.
+ * @pool: pool from which the object was allocated
+ * @handle: handle returned from zs_malloc
+ *
+ * Before using an object allocated from zs_malloc, it must be mapped using
+ * this function. When done with the object, it must be unmapped using
+ * zs_unmap_object.
+ *
+ * Only one object can be mapped per cpu at a time. There is no protection
+ * against nested mappings.
+ *
+ * This function returns with preemption and page faults disabled.
+ */
+void *zs_map_object(struct zs_pool *pool, unsigned long handle,
+ enum zs_mapmode mm)
+{
+ struct page *page;
+ unsigned long obj_idx, off;
+
+ unsigned int class_idx;
+ enum fullness_group fg;
+ struct size_class *class;
+ struct mapping_area *area;
+ struct page *pages[2];
+
+ BUG_ON(!handle);
+
+ /*
+ * Because we use per-cpu mapping areas shared among the
+ * pools/users, we can't allow mapping in interrupt context
+ * because it can corrupt another users mappings.
+ */
+ BUG_ON(in_interrupt());
+
+ obj_handle_to_location(handle, &page, &obj_idx);
+ get_zspage_mapping(get_first_page(page), &class_idx, &fg);
+ class = pool->size_class[class_idx];
+ off = obj_idx_to_offset(page, obj_idx, class->size);
+
+ area = &get_cpu_var(zs_map_area);
+ area->vm_mm = mm;
+ if (off + class->size <= PAGE_SIZE) {
+ /* this object is contained entirely within a page */
+ area->vm_addr = kmap_atomic(page);
+ return area->vm_addr + off;
+ }
+
+ /* this object spans two pages */
+ pages[0] = page;
+ pages[1] = get_next_page(page);
+ BUG_ON(!pages[1]);
+
+ return __zs_map_object(area, pages, off, class->size);
+}
+EXPORT_SYMBOL_GPL(zs_map_object);
+
+void zs_unmap_object(struct zs_pool *pool, unsigned long handle)
+{
+ struct page *page;
+ unsigned long obj_idx, off;
+
+ unsigned int class_idx;
+ enum fullness_group fg;
+ struct size_class *class;
+ struct mapping_area *area;
+
+ BUG_ON(!handle);
+
+ obj_handle_to_location(handle, &page, &obj_idx);
+ get_zspage_mapping(get_first_page(page), &class_idx, &fg);
+ class = pool->size_class[class_idx];
+ off = obj_idx_to_offset(page, obj_idx, class->size);
+
+ area = this_cpu_ptr(&zs_map_area);
+ if (off + class->size <= PAGE_SIZE)
+ kunmap_atomic(area->vm_addr);
+ else {
+ struct page *pages[2];
+
+ pages[0] = page;
+ pages[1] = get_next_page(page);
+ BUG_ON(!pages[1]);
+
+ __zs_unmap_object(area, pages, off, class->size);
+ }
+ put_cpu_var(zs_map_area);
+}
+EXPORT_SYMBOL_GPL(zs_unmap_object);
+
+/**
+ * zs_malloc - Allocate block of given size from pool.
+ * @pool: pool to allocate from
+ * @size: size of block to allocate
+ *
+ * On success, handle to the allocated object is returned,
+ * otherwise 0.
+ * Allocation requests with size > ZS_MAX_ALLOC_SIZE will fail.
+ */
+unsigned long zs_malloc(struct zs_pool *pool, size_t size)
+{
+ unsigned long obj;
+ struct link_free *link;
+ struct size_class *class;
+ void *vaddr;
+
+ struct page *first_page, *m_page;
+ unsigned long m_objidx, m_offset;
+
+ if (unlikely(!size || size > ZS_MAX_ALLOC_SIZE))
+ return 0;
+
+ class = pool->size_class[get_size_class_index(size)];
+
+ spin_lock(&class->lock);
+ first_page = find_get_zspage(class);
+
+ if (!first_page) {
+ spin_unlock(&class->lock);
+ first_page = alloc_zspage(class, pool->flags);
+ if (unlikely(!first_page))
+ return 0;
+
+ set_zspage_mapping(first_page, class->index, ZS_EMPTY);
+ atomic_long_add(class->pages_per_zspage,
+ &pool->pages_allocated);
+ spin_lock(&class->lock);
+ }
+
+ obj = (unsigned long)first_page->freelist;
+ obj_handle_to_location(obj, &m_page, &m_objidx);
+ m_offset = obj_idx_to_offset(m_page, m_objidx, class->size);
+
+ vaddr = kmap_atomic(m_page);
+ link = (struct link_free *)vaddr + m_offset / sizeof(*link);
+ first_page->freelist = link->next;
+ memset(link, POISON_INUSE, sizeof(*link));
+ kunmap_atomic(vaddr);
+
+ first_page->inuse++;
+ /* Now move the zspage to another fullness group, if required */
+ fix_fullness_group(pool, first_page);
+ spin_unlock(&class->lock);
+
+ return obj;
+}
+EXPORT_SYMBOL_GPL(zs_malloc);
+
+void zs_free(struct zs_pool *pool, unsigned long obj)
+{
+ struct link_free *link;
+ struct page *first_page, *f_page;
+ unsigned long f_objidx, f_offset;
+ void *vaddr;
+
+ int class_idx;
+ struct size_class *class;
+ enum fullness_group fullness;
+
+ if (unlikely(!obj))
+ return;
+
+ obj_handle_to_location(obj, &f_page, &f_objidx);
+ first_page = get_first_page(f_page);
+
+ get_zspage_mapping(first_page, &class_idx, &fullness);
+ class = pool->size_class[class_idx];
+ f_offset = obj_idx_to_offset(f_page, f_objidx, class->size);
+
+ spin_lock(&class->lock);
+
+ /* Insert this object in containing zspage's freelist */
+ vaddr = kmap_atomic(f_page);
+ link = (struct link_free *)(vaddr + f_offset);
+ link->next = first_page->freelist;
+ kunmap_atomic(vaddr);
+ first_page->freelist = (void *)obj;
+
+ first_page->inuse--;
+ fullness = fix_fullness_group(pool, first_page);
+ spin_unlock(&class->lock);
+
+ if (fullness == ZS_EMPTY) {
+ atomic_long_sub(class->pages_per_zspage,
+ &pool->pages_allocated);
+ free_zspage(first_page);
+ }
+}
+EXPORT_SYMBOL_GPL(zs_free);
+
/**
* zs_create_pool - Creates an allocation pool to work from.
* @flags: allocation flags used to allocate pool metadata
@@ -1075,201 +1246,30 @@
}
EXPORT_SYMBOL_GPL(zs_destroy_pool);
-/**
- * zs_malloc - Allocate block of given size from pool.
- * @pool: pool to allocate from
- * @size: size of block to allocate
- *
- * On success, handle to the allocated object is returned,
- * otherwise 0.
- * Allocation requests with size > ZS_MAX_ALLOC_SIZE will fail.
- */
-unsigned long zs_malloc(struct zs_pool *pool, size_t size)
+static int __init zs_init(void)
{
- unsigned long obj;
- struct link_free *link;
- struct size_class *class;
- void *vaddr;
+ int ret = zs_register_cpu_notifier();
- struct page *first_page, *m_page;
- unsigned long m_objidx, m_offset;
-
- if (unlikely(!size || size > ZS_MAX_ALLOC_SIZE))
- return 0;
-
- class = pool->size_class[get_size_class_index(size)];
-
- spin_lock(&class->lock);
- first_page = find_get_zspage(class);
-
- if (!first_page) {
- spin_unlock(&class->lock);
- first_page = alloc_zspage(class, pool->flags);
- if (unlikely(!first_page))
- return 0;
-
- set_zspage_mapping(first_page, class->index, ZS_EMPTY);
- atomic_long_add(class->pages_per_zspage,
- &pool->pages_allocated);
- spin_lock(&class->lock);
+ if (ret) {
+ zs_unregister_cpu_notifier();
+ return ret;
}
- obj = (unsigned long)first_page->freelist;
- obj_handle_to_location(obj, &m_page, &m_objidx);
- m_offset = obj_idx_to_offset(m_page, m_objidx, class->size);
+ init_zs_size_classes();
- vaddr = kmap_atomic(m_page);
- link = (struct link_free *)vaddr + m_offset / sizeof(*link);
- first_page->freelist = link->next;
- memset(link, POISON_INUSE, sizeof(*link));
- kunmap_atomic(vaddr);
-
- first_page->inuse++;
- /* Now move the zspage to another fullness group, if required */
- fix_fullness_group(pool, first_page);
- spin_unlock(&class->lock);
-
- return obj;
+#ifdef CONFIG_ZPOOL
+ zpool_register_driver(&zs_zpool_driver);
+#endif
+ return 0;
}
-EXPORT_SYMBOL_GPL(zs_malloc);
-void zs_free(struct zs_pool *pool, unsigned long obj)
+static void __exit zs_exit(void)
{
- struct link_free *link;
- struct page *first_page, *f_page;
- unsigned long f_objidx, f_offset;
- void *vaddr;
-
- int class_idx;
- struct size_class *class;
- enum fullness_group fullness;
-
- if (unlikely(!obj))
- return;
-
- obj_handle_to_location(obj, &f_page, &f_objidx);
- first_page = get_first_page(f_page);
-
- get_zspage_mapping(first_page, &class_idx, &fullness);
- class = pool->size_class[class_idx];
- f_offset = obj_idx_to_offset(f_page, f_objidx, class->size);
-
- spin_lock(&class->lock);
-
- /* Insert this object in containing zspage's freelist */
- vaddr = kmap_atomic(f_page);
- link = (struct link_free *)(vaddr + f_offset);
- link->next = first_page->freelist;
- kunmap_atomic(vaddr);
- first_page->freelist = (void *)obj;
-
- first_page->inuse--;
- fullness = fix_fullness_group(pool, first_page);
- spin_unlock(&class->lock);
-
- if (fullness == ZS_EMPTY) {
- atomic_long_sub(class->pages_per_zspage,
- &pool->pages_allocated);
- free_zspage(first_page);
- }
+#ifdef CONFIG_ZPOOL
+ zpool_unregister_driver(&zs_zpool_driver);
+#endif
+ zs_unregister_cpu_notifier();
}
-EXPORT_SYMBOL_GPL(zs_free);
-
-/**
- * zs_map_object - get address of allocated object from handle.
- * @pool: pool from which the object was allocated
- * @handle: handle returned from zs_malloc
- *
- * Before using an object allocated from zs_malloc, it must be mapped using
- * this function. When done with the object, it must be unmapped using
- * zs_unmap_object.
- *
- * Only one object can be mapped per cpu at a time. There is no protection
- * against nested mappings.
- *
- * This function returns with preemption and page faults disabled.
- */
-void *zs_map_object(struct zs_pool *pool, unsigned long handle,
- enum zs_mapmode mm)
-{
- struct page *page;
- unsigned long obj_idx, off;
-
- unsigned int class_idx;
- enum fullness_group fg;
- struct size_class *class;
- struct mapping_area *area;
- struct page *pages[2];
-
- BUG_ON(!handle);
-
- /*
- * Because we use per-cpu mapping areas shared among the
- * pools/users, we can't allow mapping in interrupt context
- * because it can corrupt another users mappings.
- */
- BUG_ON(in_interrupt());
-
- obj_handle_to_location(handle, &page, &obj_idx);
- get_zspage_mapping(get_first_page(page), &class_idx, &fg);
- class = pool->size_class[class_idx];
- off = obj_idx_to_offset(page, obj_idx, class->size);
-
- area = &get_cpu_var(zs_map_area);
- area->vm_mm = mm;
- if (off + class->size <= PAGE_SIZE) {
- /* this object is contained entirely within a page */
- area->vm_addr = kmap_atomic(page);
- return area->vm_addr + off;
- }
-
- /* this object spans two pages */
- pages[0] = page;
- pages[1] = get_next_page(page);
- BUG_ON(!pages[1]);
-
- return __zs_map_object(area, pages, off, class->size);
-}
-EXPORT_SYMBOL_GPL(zs_map_object);
-
-void zs_unmap_object(struct zs_pool *pool, unsigned long handle)
-{
- struct page *page;
- unsigned long obj_idx, off;
-
- unsigned int class_idx;
- enum fullness_group fg;
- struct size_class *class;
- struct mapping_area *area;
-
- BUG_ON(!handle);
-
- obj_handle_to_location(handle, &page, &obj_idx);
- get_zspage_mapping(get_first_page(page), &class_idx, &fg);
- class = pool->size_class[class_idx];
- off = obj_idx_to_offset(page, obj_idx, class->size);
-
- area = this_cpu_ptr(&zs_map_area);
- if (off + class->size <= PAGE_SIZE)
- kunmap_atomic(area->vm_addr);
- else {
- struct page *pages[2];
-
- pages[0] = page;
- pages[1] = get_next_page(page);
- BUG_ON(!pages[1]);
-
- __zs_unmap_object(area, pages, off, class->size);
- }
- put_cpu_var(zs_map_area);
-}
-EXPORT_SYMBOL_GPL(zs_unmap_object);
-
-unsigned long zs_get_total_pages(struct zs_pool *pool)
-{
- return atomic_long_read(&pool->pages_allocated);
-}
-EXPORT_SYMBOL_GPL(zs_get_total_pages);
module_init(zs_init);
module_exit(zs_exit);
diff --git a/sound/pci/hda/hda_controller.c b/sound/pci/hda/hda_controller.c
index 8337645..8276a74 100644
--- a/sound/pci/hda/hda_controller.c
+++ b/sound/pci/hda/hda_controller.c
@@ -1676,7 +1676,7 @@
u8 sd_status;
int i;
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
if (chip->driver_caps & AZX_DCAPS_PM_RUNTIME)
if (!pm_runtime_active(chip->card->dev))
return IRQ_NONE;
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index 5ac0d39..2bf0b56 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -872,7 +872,7 @@
}
#endif /* CONFIG_PM_SLEEP || SUPPORT_VGA_SWITCHEROO */
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int azx_runtime_suspend(struct device *dev)
{
struct snd_card *card = dev_get_drvdata(dev);
@@ -970,9 +970,6 @@
return 0;
}
-#endif /* CONFIG_PM_RUNTIME */
-
-#ifdef CONFIG_PM
static const struct dev_pm_ops azx_pm = {
SET_SYSTEM_SLEEP_PM_OPS(azx_suspend, azx_resume)
SET_RUNTIME_PM_OPS(azx_runtime_suspend, azx_runtime_resume, azx_runtime_idle)
diff --git a/sound/soc/codecs/cs35l32.c b/sound/soc/codecs/cs35l32.c
index c125925..ec55c590 100644
--- a/sound/soc/codecs/cs35l32.c
+++ b/sound/soc/codecs/cs35l32.c
@@ -550,7 +550,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int cs35l32_runtime_suspend(struct device *dev)
{
struct cs35l32_private *cs35l32 = dev_get_drvdata(dev);
diff --git a/sound/soc/codecs/cs42xx8.c b/sound/soc/codecs/cs42xx8.c
index 02b1520..670ebfe 100644
--- a/sound/soc/codecs/cs42xx8.c
+++ b/sound/soc/codecs/cs42xx8.c
@@ -537,7 +537,7 @@
}
EXPORT_SYMBOL_GPL(cs42xx8_probe);
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int cs42xx8_runtime_resume(struct device *dev)
{
struct cs42xx8_priv *cs42xx8 = dev_get_drvdata(dev);
diff --git a/sound/soc/codecs/max98090.c b/sound/soc/codecs/max98090.c
index 151f718..b112b1c 100644
--- a/sound/soc/codecs/max98090.c
+++ b/sound/soc/codecs/max98090.c
@@ -2611,7 +2611,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int max98090_runtime_resume(struct device *dev)
{
struct max98090_priv *max98090 = dev_get_drvdata(dev);
diff --git a/sound/soc/codecs/pcm512x.c b/sound/soc/codecs/pcm512x.c
index 0c8aefa..e5f2fb8 100644
--- a/sound/soc/codecs/pcm512x.c
+++ b/sound/soc/codecs/pcm512x.c
@@ -517,7 +517,7 @@
}
EXPORT_SYMBOL_GPL(pcm512x_remove);
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int pcm512x_suspend(struct device *dev)
{
struct pcm512x_priv *pcm512x = dev_get_drvdata(dev);
diff --git a/sound/soc/codecs/tas2552.c b/sound/soc/codecs/tas2552.c
index b505212..ae23acd 100644
--- a/sound/soc/codecs/tas2552.c
+++ b/sound/soc/codecs/tas2552.c
@@ -115,7 +115,7 @@
{"ClassD", NULL, "PLL"},
};
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static void tas2552_sw_shutdown(struct tas2552_data *tas_data, int sw_shutdown)
{
u8 cfg1_reg;
@@ -264,7 +264,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int tas2552_runtime_suspend(struct device *dev)
{
struct tas2552_data *tas2552 = dev_get_drvdata(dev);
diff --git a/sound/soc/codecs/wm2200.c b/sound/soc/codecs/wm2200.c
index cdea9d9..1559984 100644
--- a/sound/soc/codecs/wm2200.c
+++ b/sound/soc/codecs/wm2200.c
@@ -2440,7 +2440,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int wm2200_runtime_suspend(struct device *dev)
{
struct wm2200_priv *wm2200 = dev_get_drvdata(dev);
diff --git a/sound/soc/codecs/wm5100.c b/sound/soc/codecs/wm5100.c
index a01ad62..b80970d 100644
--- a/sound/soc/codecs/wm5100.c
+++ b/sound/soc/codecs/wm5100.c
@@ -2664,7 +2664,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int wm5100_runtime_suspend(struct device *dev)
{
struct wm5100_priv *wm5100 = dev_get_drvdata(dev);
diff --git a/sound/soc/codecs/wm8962.c b/sound/soc/codecs/wm8962.c
index 1534d88..d32d554 100644
--- a/sound/soc/codecs/wm8962.c
+++ b/sound/soc/codecs/wm8962.c
@@ -3785,7 +3785,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int wm8962_runtime_resume(struct device *dev)
{
struct wm8962_priv *wm8962 = dev_get_drvdata(dev);
diff --git a/sound/soc/fsl/fsl_asrc.c b/sound/soc/fsl/fsl_asrc.c
index 9deabdd..026a801 100644
--- a/sound/soc/fsl/fsl_asrc.c
+++ b/sound/soc/fsl/fsl_asrc.c
@@ -928,7 +928,7 @@
return 0;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int fsl_asrc_runtime_resume(struct device *dev)
{
struct fsl_asrc *asrc_priv = dev_get_drvdata(dev);
@@ -954,7 +954,7 @@
return 0;
}
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
#ifdef CONFIG_PM_SLEEP
static int fsl_asrc_suspend(struct device *dev)
diff --git a/sound/soc/samsung/i2s.c b/sound/soc/samsung/i2s.c
index 95340ba..b1a7c5b 100644
--- a/sound/soc/samsung/i2s.c
+++ b/sound/soc/samsung/i2s.c
@@ -1135,7 +1135,7 @@
platform_get_device_id(pdev)->driver_data;
}
-#ifdef CONFIG_PM_RUNTIME
+#ifdef CONFIG_PM
static int i2s_runtime_suspend(struct device *dev)
{
struct i2s_dai *i2s = dev_get_drvdata(dev);
@@ -1153,7 +1153,7 @@
return 0;
}
-#endif /* CONFIG_PM_RUNTIME */
+#endif /* CONFIG_PM */
static int samsung_i2s_probe(struct platform_device *pdev)
{
diff --git a/tools/power/cpupower/utils/cpuidle-info.c b/tools/power/cpupower/utils/cpuidle-info.c
index 458d69b..75e66de 100644
--- a/tools/power/cpupower/utils/cpuidle-info.c
+++ b/tools/power/cpupower/utils/cpuidle-info.c
@@ -22,13 +22,13 @@
static void cpuidle_cpu_output(unsigned int cpu, int verbose)
{
- int idlestates, idlestate;
+ unsigned int idlestates, idlestate;
char *tmp;
printf(_ ("Analyzing CPU %d:\n"), cpu);
idlestates = sysfs_get_idlestate_count(cpu);
- if (idlestates < 1) {
+ if (idlestates == 0) {
printf(_("CPU %u: No idle states\n"), cpu);
return;
}
@@ -100,10 +100,10 @@
static void proc_cpuidle_cpu_output(unsigned int cpu)
{
long max_allowed_cstate = 2000000000;
- int cstate, cstates;
+ unsigned int cstate, cstates;
cstates = sysfs_get_idlestate_count(cpu);
- if (cstates < 1) {
+ if (cstates == 0) {
printf(_("CPU %u: No C-states info\n"), cpu);
return;
}
diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Makefile
index b3831f4..4e51122 100644
--- a/tools/testing/selftests/Makefile
+++ b/tools/testing/selftests/Makefile
@@ -1,22 +1,23 @@
TARGETS = breakpoints
TARGETS += cpu-hotplug
TARGETS += efivarfs
+TARGETS += exec
+TARGETS += firmware
+TARGETS += ftrace
TARGETS += kcmp
TARGETS += memfd
TARGETS += memory-hotplug
-TARGETS += mqueue
TARGETS += mount
+TARGETS += mqueue
TARGETS += net
-TARGETS += ptrace
-TARGETS += timers
-TARGETS += vm
TARGETS += powerpc
-TARGETS += user
-TARGETS += sysctl
-TARGETS += firmware
-TARGETS += ftrace
-TARGETS += exec
+TARGETS += ptrace
TARGETS += size
+TARGETS += sysctl
+TARGETS += timers
+TARGETS += user
+TARGETS += vm
+#Please keep the TARGETS list alphabetically sorted
TARGETS_HOTPLUG = cpu-hotplug
TARGETS_HOTPLUG += memory-hotplug
diff --git a/tools/virtio/Makefile b/tools/virtio/Makefile
index 9325f46..505ad51 100644
--- a/tools/virtio/Makefile
+++ b/tools/virtio/Makefile
@@ -3,7 +3,7 @@
virtio_test: virtio_ring.o virtio_test.o
vringh_test: vringh_test.o vringh.o virtio_ring.o
-CFLAGS += -g -O2 -Wall -I. -I../include/ -I ../../usr/include/ -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE
+CFLAGS += -g -O2 -Werror -Wall -I. -I../include/ -I ../../usr/include/ -Wno-pointer-sign -fno-strict-overflow -fno-strict-aliasing -fno-common -MMD -U_FORTIFY_SOURCE
vpath %.c ../../drivers/virtio ../../drivers/vhost
mod:
${MAKE} -C `pwd`/../.. M=`pwd`/vhost_test
diff --git a/tools/virtio/linux/virtio.h b/tools/virtio/linux/virtio.h
index 8eb6421..a3e0701 100644
--- a/tools/virtio/linux/virtio.h
+++ b/tools/virtio/linux/virtio.h
@@ -6,6 +6,7 @@
/* TODO: empty stubs for now. Broken but enough for virtio_ring.c */
#define list_add_tail(a, b) do {} while (0)
#define list_del(a) do {} while (0)
+#define list_for_each_entry(a, b, c) while (0)
/* end of stubs */
struct virtio_device {
diff --git a/tools/virtio/linux/virtio_byteorder.h b/tools/virtio/linux/virtio_byteorder.h
new file mode 100644
index 0000000..9de9e6a
--- /dev/null
+++ b/tools/virtio/linux/virtio_byteorder.h
@@ -0,0 +1,8 @@
+#ifndef _LINUX_VIRTIO_BYTEORDER_STUB_H
+#define _LINUX_VIRTIO_BYTEORDER_STUB_H
+
+#include <asm/byteorder.h>
+#include "../../include/linux/byteorder/generic.h"
+#include "../../include/linux/virtio_byteorder.h"
+
+#endif
diff --git a/tools/virtio/linux/virtio_config.h b/tools/virtio/linux/virtio_config.h
index 83b27e8..806d683 100644
--- a/tools/virtio/linux/virtio_config.h
+++ b/tools/virtio/linux/virtio_config.h
@@ -1,6 +1,72 @@
-#define VIRTIO_TRANSPORT_F_START 28
-#define VIRTIO_TRANSPORT_F_END 32
+#include <linux/virtio_byteorder.h>
+#include <linux/virtio.h>
+#include <uapi/linux/virtio_config.h>
+
+/*
+ * __virtio_test_bit - helper to test feature bits. For use by transports.
+ * Devices should normally use virtio_has_feature,
+ * which includes more checks.
+ * @vdev: the device
+ * @fbit: the feature bit
+ */
+static inline bool __virtio_test_bit(const struct virtio_device *vdev,
+ unsigned int fbit)
+{
+ return vdev->features & (1ULL << fbit);
+}
+
+/**
+ * __virtio_set_bit - helper to set feature bits. For use by transports.
+ * @vdev: the device
+ * @fbit: the feature bit
+ */
+static inline void __virtio_set_bit(struct virtio_device *vdev,
+ unsigned int fbit)
+{
+ vdev->features |= (1ULL << fbit);
+}
+
+/**
+ * __virtio_clear_bit - helper to clear feature bits. For use by transports.
+ * @vdev: the device
+ * @fbit: the feature bit
+ */
+static inline void __virtio_clear_bit(struct virtio_device *vdev,
+ unsigned int fbit)
+{
+ vdev->features &= ~(1ULL << fbit);
+}
#define virtio_has_feature(dev, feature) \
(__virtio_test_bit((dev), feature))
+static inline u16 virtio16_to_cpu(struct virtio_device *vdev, __virtio16 val)
+{
+ return __virtio16_to_cpu(virtio_has_feature(vdev, VIRTIO_F_VERSION_1), val);
+}
+
+static inline __virtio16 cpu_to_virtio16(struct virtio_device *vdev, u16 val)
+{
+ return __cpu_to_virtio16(virtio_has_feature(vdev, VIRTIO_F_VERSION_1), val);
+}
+
+static inline u32 virtio32_to_cpu(struct virtio_device *vdev, __virtio32 val)
+{
+ return __virtio32_to_cpu(virtio_has_feature(vdev, VIRTIO_F_VERSION_1), val);
+}
+
+static inline __virtio32 cpu_to_virtio32(struct virtio_device *vdev, u32 val)
+{
+ return __cpu_to_virtio32(virtio_has_feature(vdev, VIRTIO_F_VERSION_1), val);
+}
+
+static inline u64 virtio64_to_cpu(struct virtio_device *vdev, __virtio64 val)
+{
+ return __virtio64_to_cpu(virtio_has_feature(vdev, VIRTIO_F_VERSION_1), val);
+}
+
+static inline __virtio64 cpu_to_virtio64(struct virtio_device *vdev, u64 val)
+{
+ return __cpu_to_virtio64(virtio_has_feature(vdev, VIRTIO_F_VERSION_1), val);
+}
+
diff --git a/tools/virtio/uapi/linux/virtio_types.h b/tools/virtio/uapi/linux/virtio_types.h
new file mode 100644
index 0000000..e7a1096
--- /dev/null
+++ b/tools/virtio/uapi/linux/virtio_types.h
@@ -0,0 +1 @@
+#include "../../include/uapi/linux/virtio_types.h"
diff --git a/tools/virtio/virtio_test.c b/tools/virtio/virtio_test.c
index db3437c..e044589 100644
--- a/tools/virtio/virtio_test.c
+++ b/tools/virtio/virtio_test.c
@@ -11,6 +11,7 @@
#include <sys/types.h>
#include <fcntl.h>
#include <stdbool.h>
+#include <linux/virtio_types.h>
#include <linux/vhost.h>
#include <linux/virtio.h>
#include <linux/virtio_ring.h>
@@ -227,6 +228,14 @@
.val = 'i',
},
{
+ .name = "virtio-1",
+ .val = '1',
+ },
+ {
+ .name = "no-virtio-1",
+ .val = '0',
+ },
+ {
.name = "delayed-interrupt",
.val = 'D',
},
@@ -243,6 +252,7 @@
fprintf(stderr, "Usage: virtio_test [--help]"
" [--no-indirect]"
" [--no-event-idx]"
+ " [--no-virtio-1]"
" [--delayed-interrupt]"
"\n");
}
@@ -251,7 +261,7 @@
{
struct vdev_info dev;
unsigned long long features = (1ULL << VIRTIO_RING_F_INDIRECT_DESC) |
- (1ULL << VIRTIO_RING_F_EVENT_IDX);
+ (1ULL << VIRTIO_RING_F_EVENT_IDX) | (1ULL << VIRTIO_F_VERSION_1);
int o;
bool delayed = false;
@@ -272,6 +282,9 @@
case 'i':
features &= ~(1ULL << VIRTIO_RING_F_INDIRECT_DESC);
break;
+ case '0':
+ features &= ~(1ULL << VIRTIO_F_VERSION_1);
+ break;
case 'D':
delayed = true;
break;
diff --git a/tools/virtio/vringh_test.c b/tools/virtio/vringh_test.c
index 9d4b1bc..5f94f51 100644
--- a/tools/virtio/vringh_test.c
+++ b/tools/virtio/vringh_test.c
@@ -7,6 +7,7 @@
#include <linux/virtio.h>
#include <linux/vringh.h>
#include <linux/virtio_ring.h>
+#include <linux/virtio_config.h>
#include <linux/uaccess.h>
#include <sys/types.h>
#include <sys/stat.h>
@@ -131,7 +132,7 @@
return 1;
}
-static int parallel_test(unsigned long features,
+static int parallel_test(u64 features,
bool (*getrange)(struct vringh *vrh,
u64 addr, struct vringh_range *r),
bool fast_vringh)
@@ -456,6 +457,8 @@
__virtio_set_bit(&vdev, VIRTIO_RING_F_INDIRECT_DESC);
else if (strcmp(argv[1], "--eventidx") == 0)
__virtio_set_bit(&vdev, VIRTIO_RING_F_EVENT_IDX);
+ else if (strcmp(argv[1], "--virtio-1") == 0)
+ __virtio_set_bit(&vdev, VIRTIO_F_VERSION_1);
else if (strcmp(argv[1], "--slow-range") == 0)
getrange = getrange_slow;
else if (strcmp(argv[1], "--fast-vringh") == 0)