Commit graph

533,096 commits

Author SHA1 Message Date
Matt Fleming
59bf7fd45c perf/x86/intel: Enable conflicting event scheduling for CQM
We can leverage the workqueue that we use for RMID rotation to support
scheduling of conflicting monitoring events. Allowing events that
monitor conflicting things is done at various other places in the perf
subsystem, so there's precedent there.

An example of two conflicting events would be monitoring a cgroup and
simultaneously monitoring a task within that cgroup.

This uses the cache_groups list as a queuing mechanism, where every
event that reaches the front of the list gets the chance to be scheduled
in, possibly descheduling any conflicting events that are running.

Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Link: http://lkml.kernel.org/r/1422038748-21397-10-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:36 +01:00
Matt Fleming
bff671dba7 perf/x86/intel: Perform rotation on Intel CQM RMIDs
There are many use cases where people will want to monitor more tasks
than there exist RMIDs in the hardware, meaning that we have to perform
some kind of multiplexing.

We do this by "rotating" the RMIDs in a workqueue, and assigning an RMID
to a waiting event when the RMID becomes unused.

This scheme reserves one RMID at all times for rotation. When we need to
schedule a new event we give it the reserved RMID, pick a victim event
from the front of the global CQM list and wait for the victim's RMID to
drop to zero occupancy, before it becomes the new reserved RMID.

We put the victim's RMID onto the limbo list, where it resides for a
"minimum queue time", which is intended to save ourselves an expensive
smp IPI when the RMID is unlikely to have a occupancy value below
__intel_cqm_threshold.

If we fail to recycle an RMID, even after waiting the minimum queue time
then we need to increment __intel_cqm_threshold. There is an upper bound
on this threshold, __intel_cqm_max_threshold, which is programmable from
userland as /sys/devices/intel_cqm/max_recycling_threshold.

The comments above __intel_cqm_rmid_rotate() have more details.

Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Link: http://lkml.kernel.org/r/1422038748-21397-9-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:35 +01:00
Matt Fleming
bfe1fcd268 perf/x86/intel: Support task events with Intel CQM
Add support for task events as well as system-wide events. This change
has a big impact on the way that we gather LLC occupancy values in
intel_cqm_event_read().

Currently, for system-wide (per-cpu) events we defer processing to
userspace which knows how to discard all but one cpu result per package.

Things aren't so simple for task events because we need to do the value
aggregation ourselves. To do this, we defer updating the LLC occupancy
value in event->count from intel_cqm_event_read() and do an SMP
cross-call to read values for all packages in intel_cqm_event_count().
We need to ensure that we only do this for one task event per cache
group, otherwise we'll report duplicate values.

If we're a system-wide event we want to fallback to the default
perf_event_count() implementation. Refactor this into a common function
so that we don't duplicate the code.

Also, introduce PERF_TYPE_INTEL_CQM, since we need a way to track an
event's task (if the event isn't per-cpu) inside of the Intel CQM PMU
driver.  This task information is only availble in the upper layers of
the perf infrastructure.

Other perf backends stash the target task in event->hw.*target so we
need to do something similar. The task is used to determine whether
events should share a cache group and an RMID.

Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Cc: linux-api@vger.kernel.org
Link: http://lkml.kernel.org/r/1422038748-21397-8-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:34 +01:00
Matt Fleming
35298e554c perf/x86/intel: Implement LRU monitoring ID allocation for CQM
It's possible to run into issues with re-using unused monitoring IDs
because there may be stale cachelines associated with that ID from a
previous allocation. This can cause the LLC occupancy values to be
inaccurate.

To attempt to mitigate this problem we place the IDs on a least recently
used list, essentially a FIFO. The basic idea is that the longer the
time period between ID re-use the lower the probability that stale
cachelines exist in the cache.

Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Link: http://lkml.kernel.org/r/1422038748-21397-7-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:33 +01:00
Matt Fleming
4afbb24ce5 perf/x86/intel: Add Intel Cache QoS Monitoring support
Future Intel Xeon processors support a Cache QoS Monitoring feature that
allows tracking of the LLC occupancy for a task or task group, i.e. the
amount of data in pulled into the LLC for the task (group).

Currently the PMU only supports per-cpu events. We create an event for
each cpu and read out all the LLC occupancy values.

Because this results in duplicate values being written out to userspace,
we also export a .per-pkg event file so that the perf tools only
accumulate values for one cpu per package.

Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Link: http://lkml.kernel.org/r/1422038748-21397-6-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:32 +01:00
Peter P Waskiewicz Jr
cbc82b1726 x86: Add support for Intel Cache QoS Monitoring (CQM) detection
This patch adds support for the new Cache QoS Monitoring (CQM)
feature found in future Intel Xeon processors.  It includes the
new values to track CQM resources to the cpuinfo_x86 structure,
plus the CPUID detection routines for CQM.

CQM allows a process, or set of processes, to be tracked by the CPU
to determine the cache usage of that task group.  Using this data
from the CPU, software can be written to extract this data and
report cache usage and occupancy for a particular process, or
group of processes.

More information about Cache QoS Monitoring can be found in the
Intel (R) x86 Architecture Software Developer Manual, section 17.14.

Signed-off-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Borislav Petkov <bp@suse.de>
Cc: Chris Webb <chris@arachsys.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Jacob Shin <jacob.w.shin@gmail.com>
Cc: Jan Beulich <JBeulich@suse.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Steven Honeyman <stevenhoneyman@gmail.com>
Cc: Steven Rostedt <srostedt@redhat.com>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Link: http://lkml.kernel.org/r/1422038748-21397-5-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:31 +01:00
Matt Fleming
79dff51e90 perf: Move cgroup init before PMU ->event_init()
The Intel QoS PMU needs to know whether an event is part of a cgroup
during ->event_init(), because tasks in the same cgroup share a
monitoring ID.

Move the cgroup initialisation before calling into the PMU driver.

Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Link: http://lkml.kernel.org/r/1422038748-21397-4-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:30 +01:00
Matt Fleming
eacd3ecc34 perf: Add ->count() function to read per-package counters
For PMU drivers that record per-package counters, the ->count variable
cannot be used to record an accurate aggregated value, since it's not
possible to perform SMP cross-calls to cpus on other packages from the
context in which we update ->count.

Introduce a new optional ->count() accessor function that can be used to
customize how values are collected. If a PMU driver doesn't provide a
->count() function, we fallback to the existing code.

There is necessarily a window of staleness with this approach because
the task that generated the counter value may not have been scheduled by
the cpu recently.

An alternative and more complex approach would be to use a hrtimer to
periodically refresh the values from a more permissive scheduling
context. So, we're trading off complexity for accuracy.

Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Link: http://lkml.kernel.org/r/1422038748-21397-3-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:29 +01:00
Matt Fleming
39bed6cbb8 perf: Make perf_cgroup_from_task() global
Move perf_cgroup_from_task() from kernel/events/ to include/linux/ along
with the necessary struct definitions, so that it can be used by the PMU
code.

When the upcoming Intel Cache Monitoring PMU driver assigns monitoring
IDs to perf events, it needs to be able to check whether any two
monitoring events overlap (say, a cgroup and task event), which means we
need to be able to lookup the cgroup associated with a task (if any).

Signed-off-by: Matt Fleming <matt.fleming@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kanaka Juvva <kanaka.d.juvva@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Vikas Shivappa <vikas.shivappa@linux.intel.com>
Link: http://lkml.kernel.org/r/1422038748-21397-2-git-send-email-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 13:53:28 +01:00
Thierry Reding
a3f447a4f1 iommu/msm: Mark driver BROKEN
The MSM IOMMU driver unconditionally calls bus_set_iommu(), which is a
very stupid thing to do on multi-platform kernels. While marking the
driver BROKEN may seem a little extreme, there is no other way to make
the driver skip initialization. One of the problems is that it doesn't
have devicetree binding documentation and the driver doesn't contain a
struct of_device_id table either, so no way to check that it is indeed
valid to set up the IOMMU operations for this driver.

This fixes a problem on Tegra20 where the DRM driver will try to use the
obviously non-existent MSM IOMMU.

Marking the driver BROKEN shouldn't do any harm, since there aren't any
users currently. There is no struct of_device_id table, so the device
can't be instantiated from device tree, and I couldn't find any code
that would instantiate a matching platform_device either, so the driver
is effectively unused.

Reported-by: Nicolas Chauvet <kwizart@gmail.com>
Cc: David Brown <davidb@codeaurora.org>
Cc: Daniel Walker <dwalker@fifo99.com>
Cc: Bryan Huntsman <bryanh@codeaurora.org>
Cc: Olav Haugan <ohaugan@codeaurora.org>
Acked-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-02-25 13:42:26 +01:00
Thierry Reding
425061b0f5 iommu/rockchip: Play nice in multi-platform builds
The Rockchip IOMMU driver unconditionally executes code and registers a
struct iommu_ops with the platform bus irrespective of whether it runs
on a Rockchip SoC or not. This causes problems in multi-platform kernels
where drivers for other SoCs will no longer be able to register their
own struct iommu_ops or even try to use a struct iommu_ops for an IOMMU
that obviously isn't there.

The smallest fix I could think of is to check for the existence of any
Rockchip IOMMU devices in the device tree and skip initialization
otherwise.

This fixes a problem on Tegra20 where the DRM driver will try to use the
obviously non-existent Rockchip IOMMU.

Reported-by: Nicolas Chauvet <kwizart@gmail.com>
Cc: Heiko Stuebner <heiko@sntech.de>
Cc: Daniel Kurtz <djkurtz@chromium.org>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-02-25 13:42:25 +01:00
Thierry Reding
f938aab2c4 iommu/omap: Play nice in multi-platform builds
The OMAP IOMMU driver unconditionally executes code and registers a
struct iommu_ops with the platform bus irrespective of whether it runs
on an OMAP SoC or not. This causes problems in multi-platform kernels
where drivers for other SoCs will no longer be able to register their
own struct iommu_ops or even try to use a struct iommu_ops for an IOMMU
that obviously isn't there.

The smallest fix I could think of is to check for the existence of any
OMAP IOMMU devices in the device tree and skip initialization otherwise.

This fixes a problem on Tegra20 where the DRM driver will try to use the
obviously non-existent OMAP IOMMU.

Reported-by: Nicolas Chauvet <kwizart@gmail.com>
Cc: Tony Lindgren <tony@atomide.com>
Cc: Suman Anna <s-anna@ti.com>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-02-25 13:41:33 +01:00
Thierry Reding
a7b67cd5d9 iommu/exynos: Play nice in multi-platform builds
The Exynos System MMU driver unconditionally executes code and registers
a struct iommu_ops with the platform bus irrespective of whether it runs
on an Exynos SoC or not. This causes problems in multi-platform kernels
where drivers for other SoCs will no longer be able to register their
own struct iommu_ops or even try to use a struct iommu_ops for an IOMMU
that obviously isn't there.

The smallest fix I could think of is to check for the existence of any
Exynos System MMU devices in the device tree and skip initialization
otherwise.

This fixes a problem on Tegra20 where the DRM driver will try to use the
obviously non-existent Exynos System MMU.

Reported-by: Nicolas Chauvet <kwizart@gmail.com>
Cc: Kukjin Kim <kgene@kernel.org>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-02-25 13:40:52 +01:00
Will Deacon
367bd978b8 iommu/io-pgtable-arm: Fix self-test WARNs on i386
Various build/boot bots have reported WARNs being triggered by the ARM
iopgtable LPAE self-tests on i386 machines.

This boils down to two instances of right-shifting a 32-bit unsigned
long (i.e. an iova) by more than the size of the type. On 32-bit ARM,
this happens to give us zero, hence my testing didn't catch this
earlier.

This patch fixes the issue by using DIV_ROUND_UP and explicit case to
to avoid the erroneous shifts.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Reported-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
2015-02-25 13:37:32 +01:00
Srinivas Pandruvada
a35c5d1aa9 iio: imu: inv_mpu6050: Create mux clients for ACPI
This is a follow up patches after adding i2c mux adapter for bypass
mode. Potentially many different types of sensor can be attached to
INVMPU6XXX device, which can be connected to main cpu i2c bus in
bypass mode.
Why do we need this?
The system ACPI table entry will consist of only one device for
INV6XXX, assuming that this driver will handle all connected sensors.
That is not true for the Linux driver. There are bunch of IIO drivers
for each sensors, hence we created a mux on this device. So to load
these additional drivers, we need to create i2c devices for them
in this driver using this mux adapter.

There are multiple options:
1. Use the auto detect feature, this needs a new i2c class for the
adapter as the existing HWMON class is not acceptable. Also the
autodetect has overhead of executing detect method for each
matching class of adapters.
This is a simple implementation. This option was previously submitted
with not a happy feedback.

2. Option is use ACPI magic and parse the configuration data. What
we need to create a i2c device at a minimum is address and a name.
Address can be obtained for secondary device in more or less in a
standard way from using _CRS element. But there is no name. To get
name we need to process proprietary vendor data. Not having name is
not fun, as you have to create device using the device name of
INVN6XXXX, respecting driver duplicate name space restriction.
Also each client driver needs to have this name in the id table.
Since multiple driver can be loaded, the driver should be able to
detect its presence and gracefully exit for the other client driver
to take it over.
So we use two step process:
- Use DMI to id platform and parse propritery data. This is not uncommon
for many x86 platform specific driver. We will get both name and address.
The change created necessary infrastructure to add more properitery vendor
data parsing.
- If DMI match fails, then create device on INV6XXX-client (we can't
create with same name as INV6XXX as it will cause duplicate name and driver
model will reject.) With this each client sensor driver which needs to get
attached via INV6XXXX, need this name in the id table and detect the
physical presence of sensor in probe and exit if not found.

Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
2015-02-25 12:02:59 +00:00
Krzysztof Kozlowski
9820d88332 iio: jsa1212: Constify struct regmap_config
The regmap_config struct may be const because it is not modified by the
driver and regmap_init() accepts pointer to const.

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
2015-02-25 11:37:53 +00:00
Krzysztof Kozlowski
15313309cb staging:iio:hmc5843: Constify register tables and struct regmap_config
The regmap_access_table and regmap_config structures may be const
because they are not modified by the driver and regmap_init() accepts
pointer to const.

Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
2015-02-25 11:36:44 +00:00
Semen Protsenko
2e25b0ecf6 ARM: OMAP2+: gpmc: make gpmc_cs_get_name() static
Fix sparse warning:
  warning: symbol 'gpmc_cs_get_name' was not declared. Should it be static?

Signed-off-by: Semen Protsenko <semen.protsenko@globallogic.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Roger Quadros <rogerq@ti.com>
2015-02-25 12:47:32 +02:00
Semen Protsenko
9c4f757ee9 ARM: OMAP2+: gpmc: Fix writing in gpmc_cs_set_memconf
Some GPMC_CONFIG7 register bits marked as "RESERVED", means they
shouldn't be overwritten. A typical approach to handle such bits called
"Read-Modify-Write". Writing procedure used in gpmc_cs_set_memconf()
utilizes RMW technique, but implemented incorrectly. Due to obvious typo
in code read register value is being rewritten by another value, which
leads to loss of read RESERVED bits. This patch fixes this.

While at it, replace magic numbers with named constants to improve code
readability.

Signed-off-by: Semen Protsenko <semen.protsenko@globallogic.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Roger Quadros <rogerq@ti.com>
2015-02-25 12:47:32 +02:00
Robert Jarzmik
6f2116ebe2 clocksource: pxa: Fix section mismatch
As pxa_timer_common_init() is only called in init context, mark it as
such, and quiesce the compiler warnings :
WARNING: vmlinux.o(.text.unlikely+0x45d4): Section mismatch in reference
from the function pxa_timer_common_init() to the function
.init.text:sched_clock_register()

WARNING: vmlinux.o(.text.unlikely+0x4610): Section mismatch in reference
from the function pxa_timer_common_init() to the function
.init.text:clocksource_mmio_init()

Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
2015-02-25 10:28:55 +01:00
Matthias Brugger
d4a19eb3b1 clocksource: mtk: Fix race conditions in probe code
We have two race conditions in the probe code which could lead to a null
pointer dereference in the interrupt handler.

The interrupt handler accesses the clockevent device, which may not yet be
registered.

First race condition happens when the interrupt handler gets registered before
the interrupts get disabled. The second race condition happens when the
interrupts get enabled, but the clockevent device is not yet registered.

Fix that by disabling the interrupts before we register the interrupt and enable
the interrupts after the clockevent device got registered.

Reported-by: Gongbae Park <yongbae2@gmail.com>
Signed-off-by: Matthias Brugger <matthias.bgg@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
2015-02-25 10:28:49 +01:00
Daniel Lezcano
a8b1b9fc92 clockevents: asm9260: Fix compilation error with sparc/sparc64 allyesconfig
The Kconfig options for the asm9260 timer is wrong as it can be selected by
another platform with allyes config and thus leading to a compilation failure
as some non arch related code is pulled by the compilation.

Fix this by having the platform Kconfig to select the timer as it is done for
the others drivers.

Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Guenter Roeck <linux@roeck-us.net>
Acked-by: Oleksij Rempel <linux@rempel-privat.de>

Conflicts:
	drivers/clocksource/Kconfig
2015-02-25 10:28:22 +01:00
Emmanuel Grumbach
540623caa6 iwlwifi: fix max_ht_ampdu_exponent for older devices
The commit below didn't update the max_ht_ampdu_exponent
for the devices listed in iwl-[1-6]000.c which, in result,
became 0 instead of 8K. This reduced the size of the Rx
AMPDU from 64K to 8K which had an impact in the Rx
throughput. One user reported that because of this, his
downstream throughput droppped by a half.

CC: <stable@vger.kernel.org> [3.19]
Fixes: c064ddf318 ("iwlwifi: change max HT and VHT A-MPDU exponent")
Reported-and-tested-by: Valentin Manea <linux-wireless@mrs.ro>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
2015-02-25 10:39:19 +02:00
Rodrigo Vivi
62e537f8d5 drm/i915: Fix frontbuffer false positve.
This return 0 without setting atomic bits on fb == crtc->cursor->fb
where causing frontbuffer false positives.

According to Daniel:

The original regression seems to have been introduced in the original
check/commit split:

commit 757f9a3e5b
Author: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Date:   Wed Sep 24 14:20:24 2014 -0300

    drm/i915: move check of intel_crtc_cursor_set_obj() out

Which already cause other trouble, resulting in the check getting moved in

commit e391ea882b
Author: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Date:   Wed Sep 24 14:20:25 2014 -0300

    drm/i915: Fix not checking cursor and object sizes

The frontbuffer tracking itself only was broken when we shifted it into
the check/commit logic with:

commit 32b7eeec4d
Author: Matt Roper <matthew.d.roper@intel.com>
Date:   Wed Dec 24 07:59:06 2014 -0800

    drm/i915: Refactor work that can sleep out of commit (v7)

v2: When putting more debug prints I notice the solution was simpler
than I thought. AMS design is solid, just this return was wrong.
Sorry for the noise.

v3: Remove the entire chunck that would probably
    be removed by gcc anyway. (by Daniel)

Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2015-02-25 10:08:15 +02:00
Sravanthi Tangeda
5b8eb1766f i40e/i40evf: Update driver versions
Bump i40e to 1.2.9 and i40evf 1.2.3

Also update the copyright year.

Change-ID: I345d777e94abd0acffe6a28793f675d251a86299
Signed-off-by: Sravanthi Tangeda <sravanthi.tangeda@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 23:38:40 -08:00
Andy Lutomirski
72c6fb4f74 x86/ia32-compat: Fix CLONE_SETTLS bitness of copy_thread()
CLONE_SETTLS is expected to write a TLS entry in the GDT for
32-bit callers and to set FSBASE for 64-bit callers.

The correct check is is_ia32_task(), which returns true in the
context of a 32-bit syscall.  TIF_IA32 is set if the task itself
has a 32-bit personality, which is not the same thing.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Link: http://lkml.kernel.org/r/45e2d0d695393d76406a0c7225b82c76223e0cc5.1424822291.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 08:27:50 +01:00
Andy Lutomirski
08571f1ae3 x86/ptrace: Remove checks for TIF_IA32 when changing CS and SS
The ability for modified CS and/or SS to be useful has nothing
to do with TIF_IA32.  Similarly, if there's an exploit involving
changing CS or SS, it's exploitable with or without a TIF_IA32
check.

So just delete the check.

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Link: http://lkml.kernel.org/r/71c7ab36456855d11ae07edd4945a7dfe80f9915.1424822291.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-02-25 08:27:49 +01:00
Carolyn Wyborny
9a21a007b6 i40evf: Add more info to interrupt vector names
This patch adds the netdev name to the VF misc vector name.  Without
this patch, all the interrupts show the same info, so it difficult to
distinguish them.

Change-ID: I247828697e1373ecfb5f8dc1bc9618e98a7f4942
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 23:24:12 -08:00
Kazuya Mizuguchi
118e4e6a71 ARM: shmobile: r8a7791: Fix HSUSB clock to hp_clk from mp_clk
HSUSB uses hp_clk rather than mp_clk for H/W register access.

Signed-off-by: Kazuya Mizuguchi <kazuya.mizuguchi.ks@renesas.com>
Signed-off-by: Yoshihiro Kaneko <ykaneko0929@gmail.com>
[horms: updated changelog]
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 16:16:56 +09:00
Kazuya Mizuguchi
b621f6d458 ARM: shmobile: r8a7790: Fix HSUSB clock to hp_clk from mp_clk
HSUSB uses hp_clk rather than mp_clk for H/W register access.

Signed-off-by: Kazuya Mizuguchi <kazuya.mizuguchi.ks@renesas.com>
Signed-off-by: Yoshihiro Kaneko <ykaneko0929@gmail.com>
[horms: updated changelog]
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 16:16:47 +09:00
Kuninori Morimoto
e849b0650d ARM: shmobile: r8a7791: tidyup SDHI register size on DTSI
r8a7791 SDHI ch0/ch1 has SD_DMACR which is located in 0x324.
This patch updates register size

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 16:16:29 +09:00
Kuninori Morimoto
66f47ed0e8 ARM: shmobile: r8a7790: tidyup SDHI register size on DTSI
r8a7790 SDHI ch0/ch1 has SD_DMACR which is located in 0x324.
This patch updates register size

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 16:16:08 +09:00
Laurent Pinchart
ae67fa2f8b ARM: shmobile: r8a7791: Reference DMA channels in SDHI DT nodes
Add references to the transmit and receive DMA channels in the three
SDHI nodes.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 16:15:06 +09:00
Laurent Pinchart
941fe36b40 ARM: shmobile: r8a7790: Reference DMA channels in SDHI DT nodes
Add references to the transmit and receive DMA channels in the four
SDHI nodes.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 16:14:52 +09:00
Laurent Pinchart
914d7d1484 ARM: shmobile: r8a73a4: Remove legacy code
All r8a73a4 boards are now used with multiplatform kernels only. We can
remove all the unused r8a73a4 legacy device and clock registration code.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 16:14:02 +09:00
Greg Rose
7e45ab4408 i40e: Use ethtool private flags to display NPAR status
Allow an application to query the i40e driver's private flags to get the
status of NPAR enablement.  This will be used by applications to determine
if there are NPAR specific features available.

Change-ID: Ia6d9477a48f9c4cb41ca022bd433f77da3f2146c
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 23:09:32 -08:00
Geert Uytterhoeven
7b9ad9a0ab ARM: shmobile: r8a73a4 dtsi: Add PM domain support
Add a device node for the System Controller, with subnodes that
represent the hardware power area hierarchy.
Hook up all devices to their respective PM domains.

Add a minimal device node for the Coresight-ETM hardware block, and
hook it up to the D4 PM domain, so the R-Mobile System Controller
driver can keep the domain powered, until the new Coresight code
handles runtime PM.

The System Controller is also used by the R-Mobile Reset driver, which
can now restart the system.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 16:02:22 +09:00
Geert Uytterhoeven
212c23924c ARM: shmobile: defconfig: Drop useless SERIAL_8250_EXTENDED=y
SERIAL_8250_EXTENDED only serves as a guard for other Kconfig options,
none of which is enabled in shmobile_defconfig.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
2015-02-25 15:55:51 +09:00
Kevin Scott
3b38cd17f8 i40e: Set FLAG_RD when sending buffer FW must read
Set FLAG_RD for send_driver_version AQ command.

Change-ID: I8253051eff85a1d4b5a4e12ce0395b65ceb91e62
Signed-off-by: Kevin Scott <kevin.c.scott@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 22:55:02 -08:00
Takashi Iwai
de5d0ad506 ALSA: hda - Disable runtime PM for Panther Point again
This is essentially a partial revert of the commit [b1920c2110:
'ALSA: hda - Enable runtime PM on Panther Point'].  There was a bug
report showing the HD-audio bus hang during runtime PM on HP Spectre
XT.

Reported-by: Dang Sananikone <dang.sananikone@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2015-02-25 07:53:31 +01:00
Mitch Williams
aba237d124 i40e: print Rx packet split status
Add the RX routine in use to the features log message.

Change-ID: Ifbbf28fb7f42b9a3d2828586488e9e6331107dd5
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 22:40:24 -08:00
Vasu Dev
ccafbce41b i40e: setup FCoE device type
Setup FCoE netdev device type as "fcoe", so that it shows up in
sysfs as FCoE device.

Change-ID: Ie13a1a332dba4d5802586926104ee01ef20da44f
Signed-off-by: Vasu Dev <vasu.dev@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 22:26:00 -08:00
Kevin Scott
a6db5a4d01 i40e: Set BUF flag for Set Version AQ command
BUF flag must be set for indirect AQ command.

Change-ID: I6819718a47baf69d1a91ebaed89f735ed6e86025
Signed-off-by: Kevin Scott <kevin.c.scott@intel.com>
Acked-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 22:11:33 -08:00
Neerav Parikh
51616018dd i40e: Add support for getlink, setlink ndo ops
Add support for bridge offload ndo_ops getlink and setlink to
enable bridge hardware mode as per the mode set via IFLA_BRIDGE_MODE.
The support is only enabled in case of a PF VSI and not available for
any other VSI type.

By default the i40e driver inserts a bridge as part of the bring-up
when a FDIR type VSI and/or a FCoE VSI is created. This bridge is
created in VEB mode by default i.e. after creating the bridge using
"Add VEB" AQ command the loopback for the PF's default VSI is enabled.

The patch adds capability where all the VSIs created as downlink to
the bridge inherits the loopback property and enables loopback only
if the uplink bridge is operating in VEB mode.
Hence, there is no need to explicitly enable loopback as part of
allocating resources for SR-IOV VFs and call to do that has been
removed.

In case a user-request is made either via "bridge" utility or using
the bridge netlink interface that requires to change the hardware
bridge mode then that would require a PF reset and rebuild of the
switch hierarchy.

Also update the copyright year.

Change-ID: I4d78fc1c83158efda29ba7be92239b74f75d6d25
Signed-off-by: Neerav Parikh <neerav.parikh@intel.com>
Tested-By: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 21:57:05 -08:00
Greg Rose
96664483a3 i40e: Implement configfs for NPAR BW configuration
Add configfs controls to get, set and commit NPAR BW configurations.

We export three controls:
	min_bw - Can take a value from 0 to 100 inclusive
	max_bw - Can take a value from 1 to 100 inclusive
	commit - A write-only control that accepts only a value of 1 and will
		cause the BW settings to be permanently committed to NVM so
		that they are persistent across power cycles and system
		resets

The BW values are relative and are expressed as percentages.  For more
information on the interpretation of the BW settings see the Dell
specifications for NPAR.

Also update the copyright year.

Change-ID: Id7496ca65630b5037e32ba6a5a748fbc1632881b
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-By: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 21:42:40 -08:00
Greg Rose
f4492db16d i40e: Add NPAR BW get and set functions
We need to be able to get, set and commit permanently the NPAR
partition BW configuration through configfs.  These are necessary
precursor functions for that feature.

Also update the copyright year.

Change-ID: I9d5ca160a9288145f1dd2042994028679fff55f3
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 21:28:11 -08:00
Mitch Williams
2bc7ee8ac5 i40e: enable packet split only when IOMMU present
When an IOMMU is in use, the packet split receive path shows a distinct
advantage over the single-buffer path because it minimizes DMA mapping
and unmapping. However, this is not an advantage for systems with no
IOMMU. At init time, check to see if an IOMMU is enabled and enable
packet split receives.

Change-ID: I4f70d2e9c31bbea3dc8fd0c5734959a6e6602210
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 21:13:35 -08:00
Ashish Shah
0e888980a3 i40evf: allow enabling of debug prints via ethtool
Copy setting from ethtool to the HW specific struct to actually
enable prints.  Change print from i40e to i40evf to differentiate
drivers in bare metal scenarios.

Also update the copyright year.

Change-ID: I06fee26247299a08f2e1c70fc811a9ea0931c4dd
Signed-off-by: Ashish Shah <ashish.n.shah@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 20:59:12 -08:00
Marcelo Ricardo Leitner
d752c36457 ipvs: allow rescheduling of new connections when port reuse is detected
Currently, when TCP/SCTP port reusing happens, IPVS will find the old
entry and use it for the new one, behaving like a forced persistence.
But if you consider a cluster with a heavy load of small connections,
such reuse will happen often and may lead to a not optimal load
balancing and might prevent a new node from getting a fair load.

This patch introduces a new sysctl, conn_reuse_mode, that allows
controlling how to proceed when port reuse is detected. The default
value will allow rescheduling of new connections only if the old entry
was in TIME_WAIT state for TCP or CLOSED for SCTP.

Signed-off-by: Marcelo Ricardo Leitner <mleitner@redhat.com>
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2015-02-25 13:46:35 +09:00
Carolyn Wyborny
88eee9bc54 i40e: Add method to keep track of current rxnfc settings
This patch adds a struct to the VSI struct to keep track of rxnfc
settings done via ethtool.  Without this patch, the device can only
list the options available, not the current settings and this is not
clear to the user.  Without current settings, the available settings
never changing looks like a bug.

Also update the copyright year.

Change-ID: I087bbfdb33b330496a671630a7586773e3b3e589
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2015-02-24 20:44:40 -08:00